Language Test Analysis- Review and Evaluation of a Language Assessment
· The purpose of this assignment is to measure your understanding and application of assessment theories and dimensions in relation to language learning based on our readings and discussions from Sessions 1-
· You will evaluate a language assessment instrument in use in your classroom and or school based on your learning.The instrument selected must be used to measure language proficiency, language use, or language acquisition.
*Permission and additional guidelines must be obtained before choosing this option.
Include a blank copy of the assessment as ‘Appendix 1’ to your paper.Be sure that it is not a secure test. (Refer to Baker, Ch. 4 and Herrera, Ch. 5 for more examples).
1) Description of Test:
· Briefly discuss the purpose, practicality, & psychometrics of the test.
· What methods are used?
· What does it intend to measure?
· How does it take into account home language use?
· Discuss wash back may occur and how it may affect the test. (Refer to our text and relevant Brown chapters)
2) Application to Emergent Bilinguals:
· Explain how the assessment views Emergent Bilinguals along the continuum of promise or deficit.
· What are the test’s limitations and relevant applications for this student group?
· For Emergent Bilinguals, does the test accurately measure what it is meant to measure?
3) Use:
· Briefly describe the psychometrics of the assessment. According to the test designers, what are the intended uses of the assessment?
· How is the test scored, and how should those scores be interpreted and used?
· Does provide a holistic picture of student capacities and needs?
· How can it be adjusted to better suit its purpose for Emergent Bilinguals?
4) Discussion of Validity:
· Discuss the reliability and construct irrelevant variance of the assessment’s use for micro- and macro-level decision-making in relation to Emergent Bilinguals.
· Based on your readings and interpretation, how should this test be used to ensure equity for Emergent Bilinguals?Refer to the relevant Abedi and Mahoney readings.
All ideas must be cited using APA format.A Works Cited page must all be attached to your paper following Appendix 1.
Formatting Instructions
All parts of your paper should double-spaced, typed in Times-Roman 12 point font, with 1-in. margins, and with page numbering on the bottom of the page alongside your last name (i.e., Suriel, pg. 1). PLEASE use the spell and grammar checks BEFORE you hand it in. Proofread each section carefully.Quality of writing is a part of the grade on the project.
Length of Assignment
Considering the demands of the task, this paper should be 6 complete pages in length.However, I am looking for quality instead of quantity.As long as you have fulfilled the demands of the task.Please refer to the rubric to guide your analysis and evaluation.
Page 2 of 2


5 points

4 Points

3 Points

2 Points

1 Point

Description & Use

Demonstrates a clear under-standing of test purpose and use by applying ideas beyond the text and including implications for students beyond the current grade classroom.

Demonstrates a clear under-standing of test purpose and use by applying ideas beyond the text and into the classroom experience.

Demonstrates a clear under-standing of test purpose and use by applying ideas within the text to interpreting test.

Demonstrates understanding of test use and purpose by including information clearly stated in the course material.

Demonstrates understanding of test purpose and use but does not use ideas in course material to evaluate the test.

Validity & Emergent Bilinguals

Demonstrates a clear under-standing of validity by critiquing specific constructs and implications for EBs.May suggest alter-natives for increasing equitable test use.

Demonstrates a clear under-standing of validity by critiquing specific constructs and implications for Emergent Bilinguals.

Demonstrates understanding of validity by dis-cussing general construct implications for Emergent Bilinguals.

Demonstrates understanding of validity by discussing general construct implications.Does not discuss Emergent Bilinguals.

Refers to validity and application but does not apply to test nor to Emergent Bilinguals.


Cites at least 5 ideas from the course.
Citations come from 4 or more course texts.

 Cites at least 4 key ideas from the course
 Citations must come from at least 3 course texts

 Cites at least 3 key ideas from the course
 Citations come from at least 2 course texts

 Cites at least 2 key ideas from the course
 Citations may come from one course text.

 Cites at least 1 key idea from the course
 The citation must come from 1 course text

Reference to Test Items

Specifically refers to and explains test items in relation to validity, descriptions and use.

Specifically refers to test items in relation so validity, descriptions, and use

Generally refers to test items in relations to validity, descriptions or use.

Vaguely refers to test in relation to validity, descriptions or use.

Does not refer to test in relation to validity, descriptions or use.

Language Command

 _Paragraph & sentence structure, word choice, and other English Language mechanics are superior and effective for the writing task.
 _There are very few or no errors in grammar and mechanics.

-Paragraph structure, sentence structure, word choice, and other English Language mechanics are superior and effective for the writing task.
 _There are a few errors in grammar and mechanics that do not interfere with meaning.

 _Paragraph structure sentence structure, word choice, and other English Language mechanics are good and effective for the writing task.
 _There are some errors in grammar and mechanics that may interfere with under-standing.

_Paragraph structure, sentence structure, word choice, and other English Language mechanics are adequate for the writing task.
 _There are several errors in grammar and mechanics that interfere with understanding.

-Paragraph structure, sentence structure, word choice, and other English Language mechanics are poor and ineffective for the writing task.
 _There are numerous errors in grammar and mechanics that interfere with understanding.
New York State English As A
Second Language Achievement Test
Test Booklet
Developed and published under contract with the New York State Education Department by MetriTech, Inc. Copyright © 2019 by the New York
State Education Department. Permission is hereby granted for school administrators and educators to reproduce these materials, located online
at http://www.p12.nysed.gov/assessment/nyseslat, in the quantities necessary for their school’s use, but not for sale, provided copyright notices are
retained as they appear in these publications. This permission does not apply to distribution of these materials, electronically or by other means,
other than for school use.
Grades 9–12
. .
. .
Page 1NYSESLAT Test Sampler • Grades 9–12 • Speaking
I am going to ask you some questions. Listen and then answer. Be sure to answer in
English using your own words.
Scientists study many different things.
1 2
3 4
1 Tell me what other scientists do.
Go On
Grades 9–12
For hundreds of years, scientists have used different tools to look at things that are
very small. In the past, they used things like special glasses and microscopes. More
advanced and powerful versions of these tools have been developed over time. Some
scientists today use computer technology.
1. 2.
Page 2 NYSESLAT Test Sampler • Grades 9–12 • Speaking
Tell me how scientific technology and tools are different today than in the past.
Now let’s talk about a scientist named Mary Anning.
When an animal dies, its body may become covered by dirt, and over time, that
dirt can turn into rock. This rock, shaped like the animal’s body, is a fossil. Mary
Anning has been called a great fossil hunter. In the early 1800s, she and her brother
discovered the fossil of an entire dinosaur on the cliffs near their home in England.
This led 10-year-old Mary Anning on a lifelong search for fossils. She eventually
became a respected scientist known for her discoveries of many fossils. These fossils
can be seen in various museums around the world today.
Before her time, Earth science was not considered to be as important as physics or
chemistry. Thanks to the work of people like Mary Anning, scientists realized that
the study of rocks and fossils helps us better understand Earth and its history. Today,
Earth science is an important branch of science.
3 Tell me about Mary Anning’s work as a scientist.
Go On
Page 3NYSESLAT Test Sampler • Grades 9–12 • Speaking
Grades 9–12
You just learned about a scientist named Mary Anning.
When an animal dies, its body may become covered by dirt, and over time, that
dirt can turn into rock. This rock, shaped like the animal’s body, is a fossil. Mary
Anning has been called a great fossil hunter. In the early 1800s, she and her brother
discovered the fossil of an entire dinosaur on the cliffs near their home in England.
This led 10-year-old Mary Anning on a lifelong search for fossils. She eventually
became a respected scientist known for her discoveries of many fossils. These fossils
can be seen in various museums around the world today.
Before her time, Earth science was not considered to be as important as physics or
chemistry. Thanks to the work of people like Mary Anning, scientists realized that
the study of rocks and fossils helps us better understand Earth and its history. Today,
Earth science is an important branch of science.
Page 4 NYSESLAT Test Sampler • Grades 9–12 • Speaking
Tell me why some people would say that Mary Anning made an important
contribution to science.
Now let’s talk about Roman concrete.
The ancient Roman Empire is famous today for many of its buildings. Today we
can still see the Coliseum, an arena that was used for sporting events, as well as the
Pantheon, which is still used today as a church. We can also still see aqueducts, which
were structures used for the transportation of water across the empire. All of these
were built using concrete.
Today people are very interested in this concrete because it lasted for such a long
time. These buildings were built nearly 2,000 years ago. In some ways, this concrete
is not as strong as modern concrete; it cannot hold as much weight. However,
Roman concrete lasts much longer. Scientists think that this durability comes from
the ingredients used to make the concrete. One of the ingredients was ash from the
eruption of a volcano. This ash is not used in modern concrete, but may be helpful in
improving modern engineering.
Go On
Page 5NYSESLAT Test Sampler • Grades 9–12 • Speaking
Tell me how ancient Romans used concrete.
Grades 9–12
You just learned about Roman concrete.
Tell me how the concrete used in ancient Rome is different from
modern concrete.
The ancient Roman Empire is famous today for many of its buildings. Today we
can still see the Coliseum, an arena that was used for sporting events, as well as the
Pantheon, which is still used today as a church. We can also still see aqueducts, which
were structures used for the transportation of water across the empire. All of these
were built using concrete.
Today people are very interested in this concrete because it lasted for such a long
time. These buildings were built nearly 2,000 years ago. In some ways, this concrete
is not as strong as modern concrete; it cannot hold as much weight. However,
Roman concrete lasts much longer. Scientists think that this durability comes from
the ingredients used to make the concrete. One of the ingredients was ash from the
eruption of a volcano. This ash is not used in modern concrete, but may be helpful in
improving modern engineering.
Page 6 NYSESLAT Test Sampler • Grades 9–12 • Speaking
Page intentionally
left blank
Grades 10–12
Listen to the passage. Then answer Questions X through X. Fill in the correct circle on your
answer sheet.
Page 9NYSESLAT Test Sampler • Grades 9–12 • Listening
Listen to the passage. Then answer Questions 1 through 6. Fill in the correct circle on your
answer sheet.
Bringing Silk to the West
Go On

Look at Question 1. Listen to this paragraph from the passage again. Then I will ask
you, “Which phrase explains where the Roman Empire obtained new items?”
“At its peak, the Roman Empire had grown very large and wealthy. Because of this,
the Romans obtained new, unfamiliar, and exotic items from around the world. When
traders began selling a luxurious, soft material that was available only from China,
wealthier Romans began to seek large amounts of this product through trade.”
Which phrase explains where the Roman Empire obtained new items?
A At its peak
B Around the world
C Luxurious, soft material
D Large amounts
Pause for about 15 seconds.
Grades 9–12

Which phrase explains where the Roman Empire obtained new items?
A At its peak
B Around the world
C Luxurious, soft material
D Large amounts
SAY Look at Question X.
Which phrase from the passage indicates that silk is a popular clothing material,
even today?
A Originally believed
B Collected these threads
C Still frequently used
D In some areas
Pause for about 15 seconds.
Grades 9–12

Look at Question 2. Listen to these sentences from the passage again. Then I will ask
you, “Which phrase helps explain the meaning of traders?”
“Because of this, the Romans obtained new, unfamiliar, and exotic items from around
the world. When traders began selling a luxurious, soft material that was available
only from China, wealthier Romans began to seek large amounts of this product
through trade.”
Which phrase helps explain the meaning of traders?
A The world
B Began selling
C Was available
D Began to seek
Pause for about 15 seconds.
Grades 9–12
Page 10 NYSESLAT Test Sampler • Grades 9–12 • Listening
Which phrase helps explain the meaning of traders?
A The world
B Began selling
C Was available
D Began to seek
3 Which phrase from the passage indicates that silk is a popular clothing material,
even today?
A Originally believed
B Collected these threads
C Still frequently used
D In some areas

Look at Question 5.
“The passage says that wealthier Romans began to seek large amounts of silk to buy.”
Which pair of phrases from the passage supports the idea that the Romans were eager
to buy silk?
A The Roman Empire had grown
Between the Roman Empire and China
B Spent large sums of money
Traded gold and other goods to obtain it
C Grew on Chinese trees
Collected these threads
D Still frequently used around the world
Continues to symbolize wealth
Pause for about 15 seconds.
Grades 9–12

SAY Look at Question X. Listen to these sentences from the passage again.
“At its peak, the Roman Empire had grown very large and wealthy. Because of this, the
Romans obtained new, unfamiliar, and exotic items from around the world.”
Which word describes the items the Romans obtained?
A Grown
B Wealthy
C Exotic
D World
Pause for about 15 seconds.
Grades 9–12

Look at Question 4. Listen to the first paragraph of the passage again. Then I will ask
you, “Which phrase helps explain the meaning of exotic?”
“At its peak, the Roman Empire had grown very large and wealthy. Because of this,
the Romans obtained new, unfamiliar, and exotic items from around the world. When
traders began selling a luxurious, soft material that was available only from China,
wealthier Romans began to seek large amounts of this product through trade.”
Which phrase helps explain the meaning of exotic?
A Large and wealthy
B New, unfamiliar
C Luxurious, soft material
D Began to seek
Pause for about 15 seconds.
Grades 9–12
Which word describes the items the Romans obtained?
A Grown
B Wealthy
C Exotic
D World
Page 11NYSESLAT Test Sampler • Grades 9–12 • Listening
Which phrase helps explain the meaning of exotic?
A Large and wealthy
B New, unfamiliar
C Luxurious, soft material
D Began to seek
6 Which pair of phrases from the passage supports the idea that the Romans were eager to
buy silk?
A The Roman Empire had grown
Between the Roman Empire and China
B Spent large sums of money
Traded gold and other goods to obtain it
C Grew on Chinese trees
Collected these threads
D Still frequently used around the world
Page intentionally
left blank
Grades 10–12
Read the passage. Then answer Questions X through X. Fill in the correct circle on your
answer sheet.

Page 13NYSESLAT Test Sampler • Grades 9–12 • Writing
Page 13NYSESLAT Test Sampler • Grades 9–12 • Reading
Read the passage. Then answer Questions 7 through 12. Fill in the correct circle on your
answer sheet.
Oral Histories: New and Old
An oral history involves obtaining, recording, and preserving people’s memories. It is a
kind of interview in which one person recalls events or circumstances from the past and
another person records the interview in order to preserve it. The means of recording can vary
from taking notes by hand to using audio and video technologies.
Since an oral history is based on an individual’s personal memories and experiences, it
must take place during the lifetime of the person being interviewed. For this reason, there is
often a sense of urgency about oral histories. For example, the men and women who served
in World War II during the 1940s are now in their eighties and nineties. Soon there will be no
survivors still living. There is an urgency to collect any stories the survivors have not yet told.
No one else can recall what happened in the same way that the survivors can.
Go On
Oral histories focus on the lives of ordinary people and also enhance what we know about
public figures. Shortly after Abraham Lincoln’s death in 1865, his secretary and law partner,
William Herndon, gathered recollections of Lincoln from people who had known and worked
with him. Similarly, soon after Bill Clinton left the presidency, former officials from his
administration began to record their memories of what they had experienced while working
with Clinton. These oral histories greatly improve our understanding about these presidencies.
The story of the Civil Rights Movement in this country also has been told through oral
histories. Many projects begun over the last 50 years have captured the voices of men, women,
and children who participated in the national effort by black people and others in the 1950s
and 1960s to eliminate segregation and gain equal rights.
Still other oral history projects have been conducted within communities to document the
lives of long-time residents or local leaders. Students in middle and high schools frequently
conduct oral history projects as part of their education. Although these projects are of a
smaller scale, they serve to record people’s experiences that would otherwise be lost.
Oral history is not new. Although technologies may have changed the methods, it remains
the oldest type of historical documentation. Over the years, the goal has remained the same—
to tell history in the voices of those who experienced it.
Grades 9–12

Page 14 NYSESLAT Test Sampler • Grades 9–12 • Writing
Page 14 NYSESLAT Test Sampler • Grades 9–12 • Reading
Read these sentences again.
“Shortly after Abraham Lincoln’s death in 1865, his secretary and law partner, William
Herndon, gathered recollections of Lincoln from people who had known and worked
with him. Similarly, soon after Bill Clinton left the presidency, former officials from
his administration began to record their memories of what they had experienced while
working with Clinton.”
Which word shows that the two oral history projects were alike in some way?
A Shortly
B After
C Similarly
D Soon
Grades 9–12

Grades 9–12

Go On
Page 15NYSESLAT Test Sampler • Grades 9–12 • Writing
Page 15NYSESLAT Test Sampler • Grades 9–12 • Reading
Read this paragraph again.
“Oral histories focus on the lives of ordinary people and also enhance what we know
about public figures. Shortly after Abraham Lincoln’s death in 1865, his secretary and
law partner, William Herndon, gathered recollections of Lincoln from people who had
known and worked with him. Similarly, soon after Bill Clinton left the presidency,
former officials from his administration began to record their memories of what they
had experienced while working with Clinton. These oral histories greatly improve our
understanding about these presidencies.”
Which phrase explains why oral histories about Abraham Lincoln and Bill Clinton
are important?
A “. . . William Herndon, gathered recollections of Lincoln . . . .”
B “. . . improve our understanding about these presidencies.”
C “. . . focus on the lives of ordinary people . . . .”
D “. . . former officials from his administration began to record . . . .”
9 Read this paragraph again.
“Still other oral history projects have been conducted within communities to
document the lives of long-time residents or local leaders. Students in middle and high
schools frequently conduct oral history projects as part of their education. Although
these projects are on a smaller scale, they serve to record people’s experiences that would
otherwise be lost.”
Which phrase helps explain the meaning of document?
A Have been conducted
B Record people’s experiences
C On a smaller scale
D Part of their education
Grades 9–12
Grades 9–12
Page 16 NYSESLAT Test Sampler • Grades 9–12 • Writing
Page 16 NYSESLAT Test Sampler • Grades 9–12 • Reading
Read this paragraph again.
“Oral histories focus on the lives of ordinary people and also enhance what we know
about public figures. Shortly after Abraham Lincoln’s death in 1865, his secretary and
law partner, William Herndon, gathered recollections of Lincoln from people who had
known and worked with him. Similarly, soon after Bill Clinton left the presidency,
former officials from his administration began to record their memories of what they
had experienced while working with Clinton. These oral histories greatly improve our
understanding about these presidencies.”
Which phrase helps explain the meaning of improve?
A Focus on the lives
B Gathered recollections
C Began to record
D Enhance what we know
11 Read the last paragraph of the passage again.
“Oral history is not new. Although technologies may have changed the methods,
it remains the oldest type of historical documentation. Over the years, the goal has
remained the same—to tell history in the voices of those who experienced it.”
Which phrase shows the reason for this type of documentation?
A Tell history
B Not new
C The oldest type
D Remained the same
Grades 9–12

Page 17NYSESLAT Test Sampler • Grades 9–12 • Writing
Page 17NYSESLAT Test Sampler • Grades 9–12 • Reading
Read these phrases again.
“. . . obtaining, recording, and preserving people’s memories.”
“. . . collect any stories the survivors have not yet told.”
“. . . tell history in the voices of those who experienced it.”
Which idea about oral histories do these phrases support?
A Oral histories teach us how to repeat the activities that people experienced during
the past.
B Oral histories help us avoid repeating the mistakes that occurred in the past.
C Oral histories help us preserve thoughts and activities of the past.
D Oral histories teach us how people used technology in the past.
Page intentionally
left blank
Grades 10–12
Read this passage again. Then you will be asked to write one paragraph based on
the passage.
Page 19NYSESLAT Test Sampler • Grades 9–12 • Writing
Read this passage. Then you will be asked to write one paragraph based on the passage.
Oral Histories: New and Old
An oral history involves obtaining, recording, and preserving people’s memories. It is a
kind of interview in which one person recalls events or circumstances from the past and
another person records the interview in order to preserve it. The means of recording can vary
from taking notes by hand to using audio and video technologies.
Since an oral history is based on an individual’s personal memories and experiences, it
must take place during the lifetime of the person being interviewed. For this reason, there is
often a sense of urgency about oral histories. For example, the men and women who served
in World War II during the 1940s are now in their eighties and nineties. Soon there will be no
survivors still living. There is an urgency to collect any stories the survivors have not yet told.
No one else can recall what happened in the same way that the survivors can.
Go On
Oral histories focus on the lives of ordinary people and also enhance what we know about
public figures. Shortly after Abraham Lincoln’s death in 1865, his secretary and law partner,
William Herndon, gathered recollections of Lincoln from people who had known and worked
with him. Similarly, soon after Bill Clinton left the presidency, former officials from his
administration began to record their memories of what they had experienced while working
with Clinton. These oral histories greatly improve our understanding about these presidencies.
The story of the Civil Rights Movement in this country also has been told through oral
histories. Many projects begun over the last 50 years have captured the voices of men, women,
and children who participated in the national effort by black people and others in the 1950s
and 1960s to eliminate segregation and gain equal rights.
Still other oral history projects have been conducted within communities to document the
lives of long-time residents or local leaders. Students in middle and high schools frequently
conduct oral history projects as part of their education. Although these projects are of a
smaller scale, they serve to record people’s experiences that would otherwise be lost.
Oral history is not new. Although technologies may have changed the methods, it remains
the oldest type of historical documentation. Over the years, the goal has remained the same—
to tell history in the voices of those who experienced it.
Grades 9–12
Page 20 NYSESLAT Test Sampler • Grades 9–12 • Writing
Now read the directions below.
The passage “Oral Histories: New and Old” mentions several oral history
projects. Who would you like to interview for an oral history project and why? Write
one paragraph to tell who you would interview for an oral history project and why
you would choose that person. Use your own ideas and ideas from the passage to
help you write.
Grades 9–12
Checklist ☑
☐ Write about the topic.
Plan your writing from beginning to end.
Use your own ideas and ideas from the passage.
Support your answer with details.
Use complete sentences.
Check your writing for grammar, capitalization, punctuation, and spelling.

On the lines below, tell who you would interview for an oral history project and why you
would choose that person. Remember to use your own ideas and ideas from the passage to
help you write.
Page 21NYSESLAT Test Sampler • Grades 9–12 • Writing
Page intentionally
left blank
Read this passage. Then you will be asked to write at least two paragraphs based on the passage.
Grades 9–12
Read the passage. Then answer Questions X through X. Fill in the correct circle on your
answer sheet.
Protecting Our Night Skies
Page 23NYSESLAT Test Sampler • Grades 9–12 • Writing
Many people around the world are concerned about the impact humans have on our
environment. While this may seem like a modern problem, over one hundred years ago the
U.S. government began to recognize the need to protect its country’s unique places. In 1872,
the U.S. Congress took a historic step toward protecting and preserving natural spaces with
the establishment of its first national park—Yellowstone National Park. In the years following
Yellowstone’s creation, national park land was set aside across the United States. The National
Park Service was founded in 1916 in order to manage and promote these precious sites.
Go On
Grades 9–12
Millions of people visit national parks in the United States each year to enjoy gorgeous
scenery, stunning natural wonders, and the opportunity to see wildlife in its natural habitat.
What many don’t realize is that, in addition to protecting landscapes and wildlife, the National
Park Service also assists in the preservation of another rapidly disappearing resource—views of
the night sky. Since national parks have been maintained in their natural state, their skies have
less light pollution than the skies over areas with houses, cities, and other sources of artificial,
or man-made, light.
Light pollution, also referred to as “photopollution,” was not a problem when the first
national parks were established. Over the years, however, it has increased significantly. Today,
light from man-made sources affects the ability of nearly all people in the United States to
see the stars, planets, and other features of the night sky. This increase in artificial light is
problematic for many reasons. The feeding and mating patterns of nocturnal animals, which
are active in the dark, can be negatively influenced if the night sky is too bright. In addition,
light pollution can be confusing to migrating birds and sea turtles, which use natural light
sources to guide them as they travel.
Humans, too, can be affected by photopollution. When we are exposed to too much light
at night, our mood and sleeping patterns can be disturbed or disrupted. In addition, night-sky
darkness has scientific and educational value—looking up at the stars encourages us to learn
about the universe and our place in it.
Educating the public is one of the ways in which the National Park Service works to protect
the night skies. Many national parks offer nighttime tours or stargazing opportunities, where
guides teach visitors about the stars. The National Park Service developed the Night Skies
Program in 1999. This program is led by a team of researchers that evaluates and monitors the
parks’ night skies. By using advanced light-sensing equipment, these scientists take detailed
measurements of the light in the sky at night. In this way, they can track light pollution levels
and the different kinds of light pollution in the parks.
The Night Skies Program team uses all of this data to find new and better ways to
decrease existing light pollution. Through its efforts to solve the problem of photopollution,
the National Park Service shows that it is able to face modern challenges while honoring its
commitment to protect its unique environment.
Page 24 NYSESLAT Test Sampler • Grades 9–12 • Writing
Go On
Grades 5–6
Planning Page
Page 26 NYSESLAT Test Sampler • Grades 9–12 • Writing
Now read the directions below.
The passage “Protecting Our Night Skies” describes how the National Park
Service works to control light pollution. How does light pollution affect humans
and the environment? Write at least two paragraphs to describe how light pollution
affects humans and the environment. Use your own ideas and ideas from the passage
to help you write.
You may plan your writing for Question 14 here, if you wish. Use the space below to organize
your ideas about what to write. Your writing on this planning page will NOT count toward
your final score.
Write your final answer on Pages 27 and 28.
Grades 3–4 MT2274
Checklist ☑
☐ Write about the topic.
Plan your writing from beginning to end.
Use your own ideas and ideas from the passage.
Support your answer with details.
Use complete sentences.
Check your writing for grammar, capitalization, punctuation, and spelling.

Page 27NYSESLAT Test Sampler • Grades 9–12 • Writing
On the lines below, describe how light pollution affects humans and the environment.
Remember to use your own ideas and ideas from the passage to help you write.
Go On
Writing, continued
Page 28 NYSESLAT Test Sampler • Grades 9–12 • Writing
2019 NYSESLAT Grades 9–12 Sampler Test Booklet
Structure Bookmarks
New York State English As ASecond Language Achievement Test
© Stanford University


Issues and Opportunities in Improving the Quality of Large Scale
Assessment Systems for English Language Learners

Jamal Abedi, University of California, Davis
Robert Linquanti, WestEd

Large-scale academic content assessments primarily developed for and field tested on native
speakers of English and those proficient in academic English may not produce reliable and valid
outcomes for English Learner (ELL) students due to several extraneous factors. Key among
these factors are ELL students’ current level of English language proficiency (ELP); the
unnecessary linguistic complexity of assessment items relative to the construct(s) being
measured; and the validity of accommodations provided to improve accessibility of content-
based assessments for ELLs. Next-generation assessment systems aligned to Common Core
State Standards (CCSS) currently being developed by the two multistate Race-To-the-Top
assessment consortia must engage these challenges and develop assessment systems that are
accessible to ELLs.

This paper briefly summarizes some fundamental concepts in assessing ELLs, reviews issues
that threaten the validity of interpretation of academic content assessments for ELL students,
and provides recommendations on how to address such threats. It also highlights ELL-relevant
assessment innovations on the horizon, and briefly discusses their promise and potential
pitfalls. Finally, it suggests ways to strengthen connections between the academic assessment
system development work of the PARCC and SBAC assessment consortia, and the work of
next-generation ELP assessment developers and consortia, all with an eye to building a more
coherent overall assessment system
for ELLs.

Fundamental Considerations
Unlike all other subgroup memberships for students, English Learner as a status is meant to be
temporary, and ELLs are expected to leave the category as a result of effective, specialized
language instruction and academic support services that they are legally required to receive.i
ELL status is operationalized typically using both linguistic and academic performance
standards, so the most linguistically and academically accomplished students exit the ELL
category over time, while those not making sufficient progress remain and are joined by newly-
entering ELLs, who are by definition at low ELP levels (Kim-Wolf et al., 2008; National Research
Council, 2011; Working Group on ELL Policy, 2010). While assessment and accountability
systems usually treat the ELL category as binary (a student is ELL or not), ELLs are very
diverse and exhibit a wide range of language and academic competencies, both in English and
their primary language (Capps et al., 2005; Taylor, Stecher, O’Day, Naftel, & LeFlock, 2010).

An ELL’s ELP level clearly affects her ability to learn academic content in English and to
demonstrate academic knowledge and skills on assessment events carried out in English. For
most English Learners to learn academic English skills they need to effectively handle grade-
level content demands, it takes 4 to 7 years depending upon several factors including their initial
English language proficiency, age/grade on entry, and prior educational experiences (Cook,
© Stanford University


Linquanti, Chinen & Jung, 2012; Hakuta, Butler, & Witt, 2000; Linquanti and George, 2007).
Therefore academic assessments that fail to take account of ELLs’ English language proficiency
will likely inadequately measure their content area knowledge and skills.ii If an ELL student
performs poorly on a content assessment, educators and policymakers need to better
understand whether this is due to: 1) insufficient English language proficiency to demonstrate
content knowledge; 2) a lack of content knowledge or opportunity to learn content; 3) construct-
irrelevant interference (e.g., unnecessarily complex language in the assessment); or 4) other
sources of bias or error (e.g., cultural distance, rater misinterpretation).

Language Demands of CCSS and the Role of Comprehensive
Assessment Systems

The CCSS specify to an unprecedented degree the kinds of academic language competencies
that students need in order to perform content area tasks and demonstrate subject matter
mastery. In addition to explicitly defining K-12 listening, speaking, reading, and writing standards
in English Language Arts (ELA), the CCSS in ELA also define literacy standards for
history/social studies, science, and technical subjects at the secondary level. Across these
different content areas and including mathematics, all students will now be expected to engage
with more complex texts and to carry out more language-rich tasks (e.g., obtaining information,
demonstrating understanding, constructing explanations, engaging in arguments, etc.) in
discipline-appropriate ways during both learning and assessment situations.

As states adopt and implement the CCSS, many are also collaborating in ELP assessment
consortia to revise their existing ELP standards to better correspond to the academic language
demands reflected in the CCSS. The greater language-explicitness of the CCSS creates
opportunities to better signal both general and discipline-specific academic language uses that
all teachers need to foster and all students master within given content areas (Wong-Fillmore
and Snow, 2002). Indeed, language is integral to these next-generation content standards and
some content standards may need to be assessed in part by measuring such language uses
within the content assessment. Nevertheless, assessment developers still need to carefully
distinguish what language is content-related (construct-relevant) in order to ensure that
language that is unrelated to the focal construct (construct-irrelevant) is not confounded with the
content being measured.

The comprehensive systems of both assessment consortia (SBAC and PARCC) need to strike a
judicious balance among the three key dimensions of assessment: summative assessments of
cognitively complex knowledge and behaviors for program review and accountability purposes;
interim benchmark assessments at key intervals during the school year to predict outcomes and
guide interventions; and formative assessment practices, processes and tools to directly inform,
support, and enhance teacher pedagogy and student learning. While assessment developers
usually focus the least attention and resources on the last of these, this form of assessment (for
and as learning) is critically important to get right for ELLs because it is the most instructionally
relevant. Indeed, formative assessment processes, when seen within a teaching and learning
paradigm (versus a measurement paradigm), can be used productively by teachers with all
students (Heritage, 2010). The inequitable distribution of instructional resources to appropriately
support ELLs’ learning and the substantial need for better preparation, coaching, and ongoing
professional development of all teachers of ELLs, make it all the more important to develop
ELL-relevant formative assessment processes and practices that can provide feedback and
© Stanford University


guide next steps in teaching and learning for linguistic and academic growth (Gandara,
Rumberger, Maxwell-Jolly, & Callahan, 2003; Taylor et al., 2010).iii

Formative assessment tools and practices can also be particularly useful in measuring progress
in ELL students’ academic English development. As suggested above, ELL students who are
instructed and assessed in English need to advance toward a level of proficiency in English that
allows them to increasingly participate in academic activities using English. Formative
assessment outcomes, or other interim assessments of students’ English proficiency used
formatively, can also help teachers make more informed decisions about their ELL students’
degree of readiness to participate in assessment events delivered in English, and about what
accommodations may be appropriate to facilitate participation (see below).

A key concern about interim/benchmark assessments meriting note – especially if these are
used summatively – involves the language demands that correspond to particular content and
the timing of interim assessments measuring such content. Since ELLs’ language competencies
develop throughout the school year, differential opportunities to learn and demonstrate subject
matter knowledge may occur within the school year.iv In particular, the outcomes of interim
assessments administered earlier in the academic year may unfairly represent ELL student
results if the language competencies needed to display such academic knowledge are targeted
and learned later in the school year. It is therefore critical to determine key target language uses
corresponding to the curricular material to be taught, and to ensure that ELLs receive language
instruction that addresses these target uses in the time period covered by the interim academic
assessment. This implies the need for careful articulation of language instruction and content
instruction goals.

Issues in Assessing College and Career Readiness

Another issue challenging next-generation assessments is the conceptualization and
measurement of “college and career readiness.” While the CCSS’s goal of identifying and
fostering skills needed for success in college and career is laudable, creating tests to measure
such skills poses serious content and psychometric challenges. Under current practice, college
entrance examinations such as the SAT and ACT are often used as external validity criteria for
predicting students’ success in college from their academic performance in high school. Yet
such assessments suffer from the same construct-irrelevant language complexity (discussed
below) that clearly threatens their validity in predicting ELLs’ college and career readiness.
Unless such issues are systematically addressed through careful attention to the language used
in new academic consortia assessments, biases against ELLs in the interpretation of their
college and career readiness will likely be perpetuated.

The Impact of Construct-Irrelevant Factors on Assessment Outcomes
for ELL Students

Assessment content and questions should address only the focal construct, or the construct the
assessment claims to measure. However, many factors unrelated to the focal construct impact
assessment outcomes. Some factors are considered random measurement error (e.g., error
due to inconsistencies in scoring open-ended questions).v Other extraneous variables
systematically affect measurement outcomes and their effects cannot be removed from
assessment outcomes even with unlimited repeated measurements. Linguistic complexity that is
© Stanford University


unrelated to the focal construct represents such an unnecessary, “construct-irrelevant” source of
systematic measurement error and may seriously affect ELL measurement outcomes.vi

Research has clearly demonstrated the impact of language factors on the assessment of ELL
students both on Title I and Title III assessments (Kopriva, 2008; Solano-Flores, 2006).
Language that is unrelated, unnecessary, or irrelevant to the construct can cause ELLs difficulty
in understanding and responding to assessment questions. Examples of language difficulty
include unfamiliar vocabulary, complex grammatical structures, nominalization, multiple
embedded clauses, and passive voice constructions. In reading/language arts and ELP
assessments, language is so inherent to the focal construct that the concept of unnecessary
linguistic complexity may not apply. (Even in these areas, excessive linguistic complexity can
still be avoided.) However, in mathematics and science, test items may have complex linguistic
structures unrelated to the focal construct that unnecessarily add to cognitive load and slow the
reader down (Abedi, Lord, & Plummer, 1997; Abedi, 2006; Abedi, et al (in press); Kopriva, 2008;
Shaftel, Belton-Kocher, Glasnapp, & Poggio, 2006; Solano-Flores, 2006).

Distinguishing Language that is Relevant and Irrelevant to the Focal Construct: A Simple

Distinguishing language that is related versus unrelated to the construct poses serious
challenges and requires both content and language experts to carefully develop and review test
items and tasks and to determine such distinctions. Even in content areas that are not
commonly understood as language-heavy in assessment events (e.g., mathematics and
science), language plays an essential role both to set the context and to define the content.
Based on this premise, decisions are made (explicitly or implicitly) about what language is
necessary in assessing content and how to help students understand the content knowledge
and skills being elicited, versus what language is irrelevant and causes the student unjustified
burden and confusion.

In a study on the impact of language factors on the assessment of ELL students, Abedi and
Lord (2001) compared the performance of Grade 8 students who received an original NAEP
mathematics version of the test items with the performance of those students who received a
linguistically-revised version of the items. The authors found significant improvements on the
performance of ELL students taking the mathematics test version which was linguistically
modified to reduce the level of unnecessary linguistic complexity. The revised version was
prepared by a team of experts in such a way as to not alter or modify any content-related
language. Before administration of the two test forms, two mathematics content experts
independently compared the original and the linguistically modified versions to make sure the
math content was not altered. Subsequent studies (e.g., Abedi, 2006) have confirmed these
findings and suggest that reducing unnecessary linguistic complexity improves the validity of
content-based assessments for ELL students.

To illustrate the process of linguistic modification that addresses language irrelevant to the focal
construct, we present a grade 4 released mathematics test item prompt in its original form and a
linguistically-revised form, and then elaborate on what linguistic modifications were conducted
on the item.1

1 Retrieved from the California Department of Education website at
http://www.cde.ca.gov/ta/tg/sr/documents/cstrtqmath4.pdf.The authors thank Nancy Ewers for providing
this example.

© Stanford University


Original Prompt: A cookie factory can bake 62 trays of cookies in the morning and 53 trays of
cookies in the afternoon. If each tray holds 12 cookies, how many cookies can be baked in 1

Revised Prompt with reduced linguistic complexity: A bakery bakes 62 trays of bread in the
morning and 53 trays in the afternoon. Each tray holds 12 loaves of bread. How many loaves
did they bake in one day?

The linguistic complexity is reduced by making the following changes:

1. Replace cookie factory with the accurate word for such a facility, bakery.
2. Replace cookies with the more cross-culturally familiar word bread.
3. Replace the modal verb phrase can bake with simple present tense.
4. Replace the subordinate clause if each tray holds 12 cookies, with a simple declarative
5. Eliminate the modal with passive voice can be baked with a past tense interrogative

A comparison of the linguistic structure of the two versions of this item reveals that the
modifications target unnecessary linguistic complexity and do not alter any language related to
the content. This makes the reduced linguistic complexity math item potentially more accessible
to ELL students.

Leveraging Accommodations for Incremental Improvement:
What Have We Learned?

Federal law requires that ELLs be provided with accommodations to make assessments more
accessible to them. However, accommodations used by ELLs must be effective in improving
accessibility and must also be valid (demonstrate an absence of advantage for non-ELLs).
Decisions on the number and type of accommodations to be used with ELLs are left to each
state (Kieffer, Lesaux, Rivera & Francis, 2009; Shafer Willner, Rivera & Acosta, 2008; Solano-
Flores, 2006; Young et al., 2008).vii

Among the most important criteria for selecting appropriate accommodations for ELL students is
relevance to specific need. Unlike students with disabilities with different needs, ELLs share a
common need for assistance with the language of assessments. Research evidence supports
the value of accommodations offering direct and indirect linguistic support when appropriately
tailored to ELL characteristics and testing conditions (Pennock-Roman & Rivera, 2011; Shafer
Willner et al., 2008). For example, English customized dictionaries/ glossaries that clarify key
vocabulary not directly tied to the construct being measured are effective in paper-and-pencil
versions when ELLs are given extra time. Pop-up English glossaries (via computer-based
delivery) are more helpful under restricted time constraints. Students instructed bilingually may
need to be accommodated based on their ELP level as well as on the goals of instructional
program. For example, primary-language (L1) versions of test items are promising for L1
speakers at low-ELP levels and those being instructed in L1 while learning English. This
accommodation is less effective for ELLs at intermediate ELP levels and those receiving content
© Stanford University


instruction in English. Dual-language formats (parallel bilingual versions and bilingual
glossaries) also show promise if there are generous time limits to help students with differing
capacities in both languages. Finally, plain-English accommodations, though promising, have
yielded mixed results with effectiveness, though not with validity when used in rigorous
experimental research designs (Duran, 2008; Kieffer et al., 2009).viii Since the CCSS calls forth
many academic language skills that are inextricably related to more complex content knowledge
and are central to many focal constructs, such necessary language complexity may be ineligible
for simplification.

Clearly, not all accommodations are appropriate for all ELLs since ELLs are heterogeneous in
ways that can measurably influence the effectiveness of particular accommodations. Using a
decision algorithm to assign configurations of accommodations tailored to ELLs’ linguistic and
socio-cultural characteristics shows promise in yielding better performance outcomes than
providing all available accommodations or no accommodations (Kopriva, Emick, Hipolito-
Delgado & Cameron, 2007). While such an algorithm depends on the capacity of states and
districts to collect relevant data at the student level, new data systems coming online in states
and districts may soon make this a
feasible option.

Computer Adaptive Tests, Automated Scoring, & Language-
Minimizing Accommodations

Several innovations on the horizon hold promise for improving the assessment system’s
responsiveness to ELLs but these must be pursued carefully. Computer adaptive testing, or
online testing formats presenting students with test questions of a level of difficulty continually
adjusted based on how the student has answered previous questions, may more accurately
estimate ELLs’ content knowledge while also increasing testing efficiency and reducing stigma
and demoralization (see Reckase, 2011). In order for CAT to work for ELLs, the optimization
algorithms that assign test items must be sensitive to an ELL’s ELP level (particularly in literacy
domains) so that items of equivalent construct difficulty, but with differing levels of linguistic
complexity, can be assigned to ELLs of different ELP levels. Given that both consortia are
expected to develop tens of thousands of items, it will be crucial to categorize the language
demands inherent in test items and tasks by ELP performance level descriptors, even if such
benchmarking and anchoring must be done in terms broad enough to be comparable across
different ELP assessments (e.g., beginning, intermediate, advanced). Likewise, automated
scoring routines that enable computerized scoring of short essays and constructed responses
may need to be specially programmed to recognize common “trans-language” features of ELL
writing. This would require training such artificial intelligence engines with exemplars that
include grammatical, vocabulary, and discourse features of ELLs at various stages of second
language acquisition, in order to provide a more careful analysis and meaningful judgment of
performance by ELP level.

Finally, substantial work is being done using online formats that increase access by conveying
information to and receiving information from ELLs at the lowest ELP levels (Kopriva, 2011).
Multi-semiotic approaches in particular appear promising in accessing conceptual and
procedural knowledge in science and math of ELLs at the lowest ELP levels.ix Such methods
can provide more accurate and valid information on these ELLs and also signal to educators
that ELLs at all stages of language development can learn and be assessed in the academic
content. However, these “language-minimizing” accommodations must be understood and
© Stanford University


utilized as temporary strategies to measure students’ knowledge while students develop
language competencies required by CCSS. If not, such efforts could unintentionally suggest to
teachers that ELLs’ language development is not essential to learning and demonstrating
academic content knowledge, and contribute to their instructional marginalization.

Implications for Moving Forward

As the nation implements more rigorous, language-rich academic content standards in English
language arts, mathematics, and science and moves towards more comprehensive academic
and ELP assessment systems, educators and assessment developers have a clear opportunity
and obligation to ensure that the growing ELL student population has meaningful access to
rigorous instruction and valid, useful assessments to measure language and content
knowledge, skills, and abilities.

There have been intriguing suggestions that, just as ELL students reaching a “threshold” level of
English language proficiency must be effectively supported in developing their interpersonal,
presentational, and interpretive uses of language by content area teachers (Walqui & Heritage,
2012), so too such discipline-specific social and academic language competencies delineated in
next-generation standards might be called out, measured and reported for all students – English
learners, standard English learners, and monolingual standard English speakers – as part of the
Race To the Top academic content assessments. While it would respond to a steadily growing
call to operationalize and measure academic language uses for all students, this approach also
raises several significant challenges. These include ensuring the validity of the language
competencies postulated within the content standards as integral to demonstrating mastery of
subject matter content; avoiding unnecessary linguistic complexity in assessing necessary
linguistic competencies; and aligning the assessment infrastructure to clearly and coherently
articulate where the “threshold” is crossed from ESL/ELD precursor language progressions to
language constructs found in the content standards. (See Bailey & Wolf [2012] for further
discussion of ELP standards.)

Given the demonstrated impact of English language proficiency on ELLs’ opportunity to learn
and on their assessment outcomes, states and consortia first need to develop a coherent
framework to ensure that the breadth, depth, and complexity of academic language uses
reflected in CCSS are adequately captured in new ELP standards. Academic assessment
consortia and ELP assessment developers should also strengthen communication, data
collection and analysis, experimentation, and prototyping of ELP and academic assessment
tasks to yield more aligned and coherent information across assessment systems. They can
also invest heavily in formative assessment processes and practices, tools and tasks that
carefully map key academic language competencies and target language uses, and ensure
these language competencies are articulated in language learning progressions reflected in ELP
standards and ELP assessment specifications. Moreover, academic assessment developers
can use these ELP performance standards to evaluate the language demands of different
content assessment items and tasks, which will be critical in the adaptive assignment of the tens
of thousands of test items to ELLs of different ELP levels.

The PARCC and SBAC assessment consortia can also incorporate lessons from research on
ELL assessment and accommodations into their test development processes. For example,
they can: 1) Examine different interpretations of test scores by subgroups of students, including
ELLs at different ELP levels, to identify possible threats to valid interpretation of assessment
© Stanford University


outcomes; 2) Identify possible construct-irrelevant sources in assessment items and tasks by
conducting cognitive labs and think-aloud procedures on ELLs at different ELP levels; 3) Have
content, language, and assessment experts identify unnecessary linguistic complexity; 4)
Distinguish linguistic structures that are related and unrelated to the focal construct; even
construct-relevant language can be made more accessible for ELLs by, for example, avoiding
long and complex reading comprehension passages when such complexities are not required
by CCSS standards being measured; 5) Specify accommodations for ELLs based on student
characteristics, testing conditions, and instructional services provided; avoiding
accommodations that are ineffective or irrelevant for ELLs, or that alter the focal construct; and
6) Provide evidence to substantiate selection and delivery of accommodations for ELLs.

The two academic assessment consortia have access to the insights of rigorous, current
research on ELL academic assessment. They also have colleagues working in parallel on next-
generation ELP standards and assessments articulating the language demands of CCSS.
Where gaps in knowledge exist, enormous opportunities also exist to collaboratively prototype,
field test, study, and advance understanding of ELL-responsive assessment tasks and
strategies. Such resources and opportunities can and must inform the development of next-
generation assessment systems that are more accessible, valid, and useful to ELLs.

i English learners are language minority students not sufficiently proficient in English to be able to benefit adequately
from regular mainstream instruction and demonstrate their knowledge and abilities using English.
ii Note that issues of measurement are distinct from issues of accountability. From a measurement perspective,
knowing an ELL’s ELP level (particularly with respect to literacy) is essential to judging the validity of the inferences
from the assessment. With respect to educator accountability however, there may still be a rationale for including
such results to determine school or district effectiveness, particularly if ELLs have not been supported to progress in
their English-language proficiency over time (see Cook et al., 2012).
iii ELL-focused formative approaches are slowly evolving and include pilot academic content learning progressions
and associated language learning targets; prototyped performance tasks and instructional supports linked to those
tasks; and professional development models that systematically build teachers’ capacities to evaluate ELLs’ access
to and accomplishment of language and content objectives that indicate progress toward larger instructional goals.
See FLARE; WestEd’s Quality Teaching for English Learners Program; Bailey & Heritage, 2010; Heritage, 2008.
iv See Wise (2011) for discussion of issues related to different aggregation methods of interim/benchmark
assessment results.
v Random measurement error affects the observed score (X) but its impact reduces by averaging over repeated
observation. The observed score (X) becomes closer to the true score (T) as the number of measurements
increases. In classical test theory (Thorndike, 2005), the correlation between the true score and error score is
assumed to be zero. For example, in scoring open-ended questions, some judges might be lenient and provide
higher scores for everyone whereas some other raters may be less generous in their rating and assign lower ratings
to everyone. Averaging over a number of ratings from the two groups of raters should control for this source of
measurement error. Accordingly, in a generalizability approach (Shavelson & Webb, 1991), the number of levels
within each facet of measurement is increased in order to reduce measurement error and improve score
vi Test items that are culturally biased may also pose difficulty for test takers with different cultural backgrounds. Such
difficulties could systematically distort the assessment outcomes and reduce students’ test scores significantly.
vii For a more elaborated summary of this and the following section, see Linquanti (2011).
viii Possible explanations include heightened sensitivity of test developers in recent years to avoid unnecessarily
linguistically-complex items, which may reduce gains produced by plain-English versions; and the presence of
comparison groups of non-ELLs that include former ELLs with ongoing linguistic needs who also benefit from the
accommodation, which may affect comparison statistics.
ix See www.onpar.us for examples of this strategy.

© Stanford University



Abedi, J. (2006). Language Issues in Item-Development. In Downing, S. M. and Haladyna, T. M. (Ed.), Handbook of
Test Development (pp. 377-398). New Jersey: Lawrence Erlbaum Associates, Publishers.

Abedi, J.; Bayley, R.; Ewers, N.; Mundhenk, K. Leon, S. & Kao, J. (in press). Accessible Reading Assessments for
Students with Disabilities. International Journal of Disability, Development and Education, 59(1).

Abedi, J., & Lord, C. (2001). The language factor in mathematics tests. Applied Measurement in Education, 14(3),

Abedi, J., Lord, C., & Plummer, J. (1997). Language background as a variable in NAEP mathematics performance
(CSE Tech. Rep. No. 429). Los Angeles: University of California, National Center for Research on
Evaluation, Standards, and Student Testing.

Bailey, A. & Heritage, M. (2010). English-language proficiency assessment foundations: External judgments of
adequacy. Evaluating the validity of English-language proficiency assessments. (An enhanced assessment
grant). Washington DC: EVEA Project.

Capps, R., Fix, M., Murray, J., Ost, J., Passel J., & Herwantoro, S. (2005). The new demography of America’s
schools: Immigration and the No Child Left Behind Act. Washington DC: Urban Institute.

Cook, G., Linquanti, R., Chinen, M., & Jung, H. (2012). National evaluation of Title III implementation supplemental
report: Exploring approaches to setting English language proficiency performance criteria and monitoring
English learner progress. Washington DC: US Department of Education, Office of Planning, Evaluation and
Policy Development.

Duran, R. (2008). Assessing English-language learners’ achievement. Review of Research in Education, 32, 292-327

FLARE (Formative Language Assessment Records for English-Language Learners) project.

Gándara, P., Rumberger, R., Maxwell-Jolly, J., & Callahan, R. (2003). English learners in California Schools: Unequal
Resources; Unequal Outcomes. Educational Policy Analysis Archives.http://.epaa.asu.edu/epaa/v11n36/

Hakuta, K., Butler, Y., & Witt, D. (2000). How long does it take English learners to attain proficiency? University of
California Linguistic Minority Research Institute Policy Report 2000-1. Santa Barbara: UC-LMRI.

Heritage, M. (2010). Formative assessment and next-generation assessment systems: Are we losing an opportunity?
(p. 9). Washington DC: Council of Chief State School Officers.

Heritage, M. (2008). Learning progressions: Supporting instruction and formative assessment. Washington DC:
Council of Chief State School Officers.

Kieffer, M., Lesaux, N., Rivera, M. & Francis, D. (2009). Accommodations for English-language learners taking large-
scale assessments: A meta-analyses on effectiveness and validity. Review of Educational Research, 79(3),

Kopriva, R., (2011, April 9). The promise of demonstration-based interactive test task environments for struggling
readers and English learners. Paper presented at the annual meeting of the National Council on
Measurement in Education, New Orleans, LA.

Kopriva R. (Ed.). (2008). Improving testing for English Language learners. New York: Routledge Press.

Kopriva, R., Emick, J., Hipolito-Delgado, C., & Cameron, C. (2007). Do proper accommodation assignments make a
difference? Examining the impact of improved decision making on scores for English language learners.
Educational Measurement: Issues and Practice, 26(3), 11-20.
© Stanford University


Linquanti, R. (2011). Strengthening assessment for English learner success: How can the promise of the common
core state standards and innovative assessment systems be realized?In D. Plank and J. Norton (Eds.),
The road ahead for state assessments (pp. 13-25).Palo Alto & Cambridge: Policy Analysis for California
Education and Rennie Center for Education Research & Policy.

Linquanti, R. & George, C. (2007). Establishing and utilizing an NCLB Title III accountability system: California’s
approach and findings to date. In J. Abedi (Ed.), English language proficiency assessment in the nation:
Current status and future practice (pp. 105-118). Davis: University of California Press.

National Research Council (2011). Allocating federal funds for state programs for English-language learners. Panel to
review alternative sources for the limited English proficiency allocation formula under Title III, Part A.,
Elementary and Secondary Education Act, Committee on National Statistics and Board Testing and
Assessment. Washington DC: National Academies Press.

Pennock-Roman, M. and Rivera, C. (2011), Mean effects of test accommodations for ELLs and non-ELLs: A meta-
analysis of experimental studies. Educational Measurement: Issues and Practice, 30: 10–28.

Reckase, M.D. (2011). Computerized adaptive assessment (CAA): The way forward. InD. Plank and J. Norton
(Eds.), The road ahead for state assessments (pp. 1-11). Palo Alto & Cambridge: Policy Analysis for
California Education and Rennie Center for Education Research & Policy.

Shafer Willner, L., Rivera, C., & Acosta, B. (2008). Descriptive study of state assessment policies for accommodating
English language learners. Prepared for the LEP Partnership, U.S. Department of Education. Arlington, VA:
The George Washington University Center for Equity and Excellence in Education.

Shaftel, J., Belton-Kocher, E.,Glasnapp, D., & Poggio, J. ( 2006). The impact of language characteristics in
mathematics test items on the performance of English language learners and students with disabilities.
Educational Assessment, 11(2), 105-126.

Shavelson & Webb (1991). Generalizability theory, a primer. NewBury Park: Sage Publications.

Solano-Flores, G. (2006). Language, dialect, and register: Sociolinguistics and the estimation of measurement error
in the testing of English language learners. Teachers College Record (108)11, pp. 2354-2379.

Taylor, J., Stecher, R., O’Day, J., Naftel, S. & LeFloch, K.C. (2010).State and Local Implementation of the No Child
Left Behind Act, Volume IX – Accountability under NCLB: Final Report. Washington, DC: U.S. Department of

Thorndike, R. M. (2005). Measurement and Evaluation in Psychology and Education. Upper Saddle River, NJ:
Pearson, Merrill.

WestEd (2011). Quality teaching for English learners program.http://www.wested.org/cs/tqip/print/docs/qt/home.htm

Wise, L. (2011). Picking up the pieces: Aggregating results from through-course assessments. Princeton: Center for
K-12 Assessment & Performance Management at ETS.

Wolf, M. K., Herman, J. L., Kim, J., Abedi, J., Leon, S., Griffin, N., Shin, H. (2008). Providing Validity Evidence to
Improve the Assessment of English Language Learners. (CSE Tech. Report No. 738). Los Angeles:
University of California, National Center for Research on Evaluation, Standards, and Student Testing.

Wong-Fillmore, L., & Snow, C. E. (2002). What teachers need to know about language. In C. T. Adger, C. E. Snow, &
D. Christian (Eds.), What teachers need to know about language (pp. 7-54). Washington DC: Center for
Applied Linguistics.

Working Group on ELL Policy (2010). Improving educational outcomes for English language learners:
Recommendations for ESEA reauthorization. Palo Alto: Author. Available at: http://ellpolicy.org

© Stanford University


Young, J. W., & King, T. C. (2008). Testing accommodations for English language learners: A review of state and
district policies: New York: The College Board. Report # 2008-6.

The Understanding Language Initiative would like to thank the Carnegie Corporation of New
York and the Bill and Melinda Gates Foundation for making this work possible. For more
information about this paper, please contact UnderstandingLanguage@stanford.edu

Understanding Language
Stanford University School of Education
485 Lasuen Mall
Stanford, CA 94305-3096
cover next page >

title: Foundations of Bilingual Education and Bilingualism
Bilingual Education and Bilingualism ; 1
author: Baker, Colin.
publisher: Multilingual Matters
isbn10 | asin: 1853593583
print isbn13: 9781853593581
ebook isbn13: 9780585195773
language: English
subject Education, Bilingual, Education, Bilingual–Great Britain,
Bilingualism, Bilingualism–Great Britain.
publication date: 1996
lcc: LC3715.B35 1996eb
ddc: 370.117
subject: Education, Bilingual, Education, Bilingual–Great Britain,
Bilingualism, Bilingualism–Great Britain.
cover next page >
< previous page page_i next page >
Page i
Foundations of Bilingual Education and Bilingualism

< previous page page_i next page >
< previous page page_ii next page >
Page ii
Series Editors
Professor Colin Baker, University of Wales, Bangor
and Nancy Hornberger, University of Pennsylvania
Other Books in the Series
Becoming Bilingual: Language Acquisition in a Bilingual Community
Building Bridges: Multilingual Resources for Children
Curriculum Related Assessment, Cummins and Bilingual Children
Language Minority Students in the Mainstream Classroom
A Parents’ and Teachers’ Guide to Bilingualism
Policy and Practice in Bilingual Education
O. GARCÍA and C. BAKER (eds)
Multicultural Child Care
Teaching Science to Language Minority Students
Working with Bilingual Children
Please contact us for the latest book information:
Mulitlingual Matters Ltd, Frankfurt Lodge, Clevedon Hall,
Victoria Road, Clevedon, Avon BS21 7SJ, England

< previous page page_ii next page >
< previous page page_iii next page >
Page iii
Foundations of Bilingual Education and Bilingualism
Second Edition
Bilingual Education and Bilingualism 1
Series Editors: Colin Baker and Nancy Hornberger
Colin Baker
Academic Consultant:
School of Education, City College of New York

< previous page page_iii next page >
< previous page page_iv next page >
Page iv
Library of Congress Cataloging in Publication Data
Baker, Colin, 1949-
Foundations of Bilingual Education and Bilingualism/Colin Baker, 2nd edn.
Bilingual Education and Bilingualism: 1.
Includes bibliographical references.
1. Education, Bilingual. 2. Education, Bilingual-Great Britain. 3. Bilingualism.
4. Bilingualism-Great Britain. I. Title. II. Series.
371.97-dc20 95-26506
British Library Cataloguing in Publication Data
A CIP catalogue record for this book is available from the British Library.
ISBN 1-85359-358-3 (hbk)
ISBN 1-85359-357-5 (pbk)
Multilingual Matters Ltd
UK: Frankfurt Lodge, Clevedon Hall, Victoria Road, Clevedon, Avon BS21 7SJ.
USA: 1900 Frost Road, Suite 101, Bristol, PA 19007, USA.
Australia: P.O. Box 6025, 83 Gilles Street, Adelaide, SA 5000, Australia.
Copyright ã 1996 Colin Baker
All rights reserved. No part of this work may be reproduced in any form or by any means without permission in
writing from the publisher.
Printed and bound in Great Britain by the Bath Press.

< previous page page_iv next page >
< previous page page_v next page >
Page v
Foreword by
Professor Ofelia García vii
Acknowledgements x
Note on the Second Edition xi
Introduction xiv
Section A: The Individual and Social Nature of Bilingualism
1. Bilingualism: Definitions and Distinctions 3
2. The Measurement of Bilingualism 17
3. Languages in Society 34
4. Language Revival and Reversal 51
5. The Development of Bilingualism 75
6. Second Language Acquisition 94
7. Bilingualism and Intelligence 116
8. Bilingualism and Thinking 128
9. Cognitive Theories of Bilingualism and the Curriculum 144
Section B: Bilingual Education Policies and Classroom Practices
10. An Introduction to Bilingual Education 164
11. The Effectiveness of Bilingual Education 199
12. Language Development and Language Allocation in Bilingual
Education Settings 225
13. Bilingual Schooling Issues 250
14. Second Language Learning 274

< previous page page_v next page >
< previous page page_vi next page >
Page vi
15. Literacy in a Multicultural Society 292
16. Literacy and Biliteracy in the Classroom 314
17. Canadian Immersion Classrooms 330
18. A Model and a Framework of Bilingual Education 340
19. The Politics of Bilingualism 351
20. Multiculturalism and Anti-Racism 373
Recommended Further Reading 389
Appendices 393
Bibliography 401
Author Index 437
Subject Index 444

< previous page page_vi next page >
< previous page page_vii next page >
Page vii
Through time, the task given to most teachers has been to educate children from their own communities, children
with whom they were deeply familiar as they shared a common language and ethnicity. Decisions about the
education of these children centered mostly on socio-pedagogical issues, having to do with the content of the
curriculum and the appropriate educational methodology used to communicate that content.
But in the last thirty years, as the world has become increasingly inter-dependent, as ethnolinguistic diversity has
become more recognized, and as the right to an education has been extended to more minority groups, teachers
have been sternly challenged. The challenge comes from the new ethnolinguistic complexity that is increasingly
reflected in their classrooms.
For some teachers, the change has to do with students who are now different linguistically, culturally, and perhaps
racially from the children of their families and immediate communities. For others, the change hinges on their need
to develop in children the ability to communicate with, and be mindful of, those who are different. For yet others,
the change involves helping children whose ethnolinguistic identity is itself being transformed. In some cases, the
teacher is faced with majority children for whom bilingualism is becoming a defining feature of their culture; in
other cases, with children for whom the minority language has become the principal focus of cultural identity.
The greatest failure of contemporary education has been precisely its inability to help teachers understand the
ethnolinguistic complexity of children, classrooms, speech communities, and society, in such a way as to enable
them to make informed decisions about language and culture in the classroom. The success of Colin Baker’s first
edition of Foundations of Bilingual Education and Bilingualism confirms the need in all contemporary educational
circles for information about the sociolinguistic and psycholinguistic issues that surround bilingualism and
bilingual education.
This second edition keeps the features that made Baker’s text so valuable; yet, it makes significant changes and
additions. Baker has spent the last two years talking to students, educators, and scholars who have used the first
edition. This second edition has benefited

< previous page page_vii next page >
< previous page page_viii next page >
Page viii
from the advice derived from these conversations with the international academic community.
As in the first edition, the book continues to offer a treatment of psychological and sociological questions that
enables educators to take a more active role in defining what it is that they want to do in a classroom. The book
seldom presents educators with a closed position. Rather, it gives theoretical and practical information from a
variety of societal contexts, making it possible for teachers to see possibilities beyond the constraints of their
particular classroom. Baker’s is a liberating view of bilingual education because it follows neither a simple
governmental definition nor a single societal perspective; and because it empowers teachers to make decisions
about their chosen role in either developing or destroying their children’s bilingualism. Thus the book performs the
valuable function of raising the consciousness of the bilingual educator in her role as language planner.
Another aspect that has made Baker’s book extremely popular, and that is preserved in this second edition, is its
inter-disciplinary approach. Linguistic aspects are intertwined throughout the discussion of psychological,
sociological, pedagogical and political issues. But at all times the teacher remains the focus, presenting the
knowledge necessary to observe children, individuals, and speech communities; the knowledge necessary to decide
on the socio-educational goals that would meet the needs observed; the knowledge necessary to design and
develop programs, models and practices that would fit those goals.
While taking into account the plurality of societal contexts and of bilingual education models and strategies that
characterizes the field, this second edition responds to many of the particular needs of bilingual educators,
especially those in the United States. For example, although this edition continues to make no mention of the
familiar labels used in the United States to limit the possibilities of bilingual education, it does present a history of
language policy in the United States. This makes it easier for bilingual educators, whether in the United States or
elsewhere, to understand their present role within a historical and sociological perspective.
Other important additions are the sections on Biliteracy and on Language Use in the bilingual classroom.
Biliteracy is, after all, one of the most distinguishing characteristics of bilingual schooling, and one that teachers in
particular are responsible for. A most important issue of bilingual schooling is how the two languages are allocated
in the curriculum. Baker gives a full treatment of the topic, again offering a range of possibilities according to the
different socio-educational needs of the many societies and language groups herein included.
As a keen observer of the European Union, Baker has included in this second edition a section on Bilingualism and
the Economy, a topic that Baker himself feels has been neglected, and that is becoming extremely important in
today’s interdependent world.
Although the first edition of Baker’s book covered the entire globe, it did not include a very important language
minority population, the Deaf Community. This edition gives the Deaf the appropriate place they deserve within
the field of bilingualism and bilingual education.

< previous page page_viii next page >
< previous page page_ix next page >
Page ix
Finally, another topic in bilingual education that has become very important is the education of language
minorities with special needs. This second edition also includes a section on Bilingual Special Education.
Baker introduces bilingualism on page one of his text by reminding us that whereas binoculars are for two eyes,
bilingualism is simply not about two languages. Indeed, the lenses through which bilingualism and bilingual
education are examined in this book extend and, at the same time, deepen our definitions and knowledge. Baker
provides us not simply with binoculars, but with powerful lenses through which the multiple visions of
bilingualism and bilingual education are reflected.
As a US bilingual educator, first a classroom practitioner and then an instructor of other bilingual teachers, I am
poignantly aware of how important it is to bring Baker’s multiple vision to classroom teachers; for teachers are
often caught between administrative policies, parental requests, children’s needs, and their own abilities and
attitudes. The image of their classroom then becomes the only reality they see through their limited lens. Baker’s
book allows them to see their reality reflected and altered in different societal contexts, with other bilingual
individuals, and other educational programs. It allows them to find the cracks in their static vision, and to see the
possibility of transformation and change in their own educational practices and in their society.
My own vision has been expanded through the constant interaction I have had with Colin Baker while he prepared
the first and second editions of this book. Our almost daily e-mail exchanges have become a most important way
to check positions derived from seeing only the local bilingual reality. I contributed to the multiplicity of the
vision contained in the first edition only because Baker is a good listener. But this second edition has gone beyond
our own vision, including now all the many comments, suggestions, reflections and criticisms that have been
passed on by students and educators. In this second edition, Baker’s talent becomes apparent in his ability to
connect more closely with the multiple readers of the first edition, bringing the focus closer to home, while at the
same time pushing the readers to see realities beyond the ones that confront them.

< previous page page_ix next page >
< previous page page_x next page >
Page x
The idea of this introductory text derived from a fellow Essex gentleman, Mike Grover. He wrote one simple
sentence that has affected my academic life ever since: `Consider writing THE textbook on Bilingual Education’.
My thanks to Mike of Multilingual Matters for not only risking me with this responsibility, but also for his
continual encouragement and friendly, facilitative style. No publisher can be so gratifying to work with than Mike
In the first edition, I recorded my grateful thanks to those who helped in the construction of that edition. I wish to
repeat my sincere gratitude to: Iolo Williams, Andrew Cohen, Edward David, Viv Edwards, Peter Garrett, Sharon
Lapkin, Hilaire Lemoine, Bernard Spolsky, Merrill Swain, Ann Illsley and Dilys Parry. Choosing to write on
bilingualism has always been inspired by my Welsh wife, Anwen, and three bilingual children who daily teach me
the beauty of bilingualism. Their loving support is eternally appreciated. Diolch yn fawr iawn am bopeth.
Multilingual Matters perceptively appointed Ofelia García of the City College of New York as Academic
Consultant for the first and second edition. I have always received detailed, sensitive, wise and judicious advice on
each draft chapter. Achilles’ heels have been quickly spotted, detailed polishing recommended, and cultural
assumptions gently revealed. A great debt is owed for her generous amounts of time and patience, both in scores of
e-mail responses and when visiting New York. The ideas shared, and her deep empathy for language minorities
has been a true learning experience.
Pocos lectores hay que sean tan comprensivos, entusiastas, perspicaces, y conocedores tanto de la escuela como
de la calle, y tan capaces de saber criticar y al mismo tiempo saber apoyar. Generosa y entregada a sus
estudiantes tanto como a su investigación, la Profesora Ofelia García supo enseñarme, de hecho y de palabra, por
correo electrónico y por su ejemplo personal, como se puede combinar en una misma persona el elevado logro
académico y el alto altruísmo personal.
Professor García was first introduced to me in a letter from the publishers. The letter simply said: ‘I have been in
contact with one of the present top-rate people in the States’. So have I, and I totally agree.
The help and support given me has been extremely generous and far more than is deserved. However, the
responsibility for all that is not as fair or just as it should be, is totally mine.

< previous page page_x next page >
< previous page page_xi next page >
Page xi
When the first edition of Foundations was written and published, a substantial sigh of relief reverberated around.
Like a ferret, I had collected the available literature on bilingualism and bilingual education. Like a squirrel, I had
carefully accumulated, then read and reviewed it all. The eighteen original chapters were written, rewritten,
repetitively edited, passed for initial comment, then sent to learned experts in Europe and the United States. The
book was revamped and revised. By September 30th 1992, the task allocated to me by the Multilingual Matters
was complete. And forever. I could revert to what comes more naturally: researching and writing on specialist
academic topics with increasing myopicity.
The only predictable part of life is unpredictability. I certainly didn’t predict the first edition of this book would sell
so many copies in so many different areas of the world. The reviews of the first edition were kind and understood
the need for a book that attempted to be comprehensive and readable. However, the continuing sales of the book
made me feel increasingly uncomfortable. Written between 1990 and 1992, the datedness of some sections, the
movement of new ideas in bilingual education, the thrill of reading submitted manuscripts and proposals as a
Series Editor for Multilingual Matters, and obtaining some important feedback from lectures in the United States
and Britain, all made it important to up-date the book. The ferret and squirrel habits remained, even though
administration and committee obligations tried, and failed, to change those habits.
These were the initial stimuli for producing a second edition. There were also critical bull’s-eyes in reviews of the
first edition that required a rethinking of the contents. Professors Nancy Dorian, Viv Edwards and John Eggleston
were particularly instrumental here. Each made highly perceptive remarks in their reviews of the first edition that
became embedded in my own critical view of the first edition. I am particularly grateful to Viv Edwards for an
important observation about literacy that led to major changes in the second edition. The second edition will
hopefully reveal that the subject matter of bilingualism and particularly bilingual education is not static. It is still
developing and

< previous page page_xi next page >
< previous page page_xii next page >
Page xii
evolving. Such movements are represented in the new edition, particularly in chapters four, and eleven to twenty.
Instructors, in particular, will wish to know what are the specific changes in the second edition. First, there are
many minor changes. For example, references have been updated, research findings added, study questions added,
corrections made and more commentary added. Second, and importantly, there are new sections and chapters. The
additions include: Literacy and Biliteracy (new chapters); the History of Bilingual Education in the United States;
Assessment and Bilingualism; Bilingual Special Education; the Dual Coding Model of the Brain; Economics and
Bilingualism; the Deaf and Bilingualism, and Language Boundaries in Classrooms.
The second edition does not go into great detail on classroom methodology. There are important reasons why this
is so. Professor Ofelia García, City College New York, and the present author compiled a Reader to accompany
this book. Areas that are important, but were only very briefly covered in the first edition of Foundations, are a
major part of this reader entitled: Policy and Practice in Bilingual Education: Extending the Foundations (1995)
and published by Multilingual Matters Ltd. This Reader contains a selection of the most important and influential
recent contributions on bilingualism and bilingual education. Each has become a ‘classic’ or a pivotal paper. The
Reader provides an introduction to recent fundamental and formative ideas on Bilingualism and Bilingual
Education in four sections: Section 1, Policy and Legislation on Bilingualism in Schools and Bilingual Education;
Section 2, Implementation of Bilingual Policy in Schools: Structuring Schools; Section 3, Using Bilingualism in
Instruction: Structuring Classrooms; Section 4, Using the Bilingualism of the School Community: Teachers and
Discussions with students and instructors about the first edition led to attempting the following: to balance a North
American and European perspective while increasingly revealing the global interest in bilingualism and bilingual
education; to attempt to marry language minority perspectives with those of relatively advantaged bilinguals (e.g.
speakers of French and English) and by doing so, to celebrate both for a common good; to write in straightforward
English while doing justice to the depth of ideas; to write for undergraduates, postgraduates and in-service and
pre-service teachers; to provide a balanced, fair and constructive introduction while adding analysis, interpretation
and discussion; and to provide a synthesis of relevant material on bilingualism and bilingual education while
adding a critical commentary.
Some instructors wanted a version that was only for Europe or only for the US. Other instructors indicated that the
subject is international and transcontinental and wanted to retain a book that was universal. This is my bias; that
any geographical limiting would do a disservice to one basis of the book: language minorities gain strength from
association, and lose if isolated. Also, there are so many common features of bilingualism and bilingual education
throughout the world that a universal perspective is liberating and empowering. Any nationally insular or
introspective ‘Foundations of Bilingualism’ would be a self-contradiction and parochialize an authentically global

< previous page page_xii next page >
< previous page page_xiii next page >
Page xiii
In compiling the second edition, I have been superbly supported by Professor Ofelia García (City College of New
York), Dr Sylvia Prys Jones (University of Wales, Bangor), Mike and Marjukka Grover (Multilingual Matters
Ltd), Jim Kyle (Centre for Deaf Studies, University of Bristol), Professor Viv Edwards (University of Reading)
and Professor Nancy Hornberger, University of Pennsylvania. In producing a Japanese translation of Foundations,
Professor Hideo Oka’s meticulousness and concern for accuracy of detail enabled some rough edges to be made
smooth. To each, I offer my great gratitude for their time, generosity, patience and care. I am solely responsible for
imperfections that remain in the text.
The hope is that the second edition will have retained the strengths of the first edition, eradicated perceived
weaknesses of the first edition, and added new text to provide a comprehensive, international introduction to this
increasingly important area of study.

< previous page page_xiii next page >
< previous page page_xiv next page >
Page xiv
This book is intended as an introduction to bilingual education and bilingualism. Written from a cross-disciplinary
perspective, the book covers a wide range of topics: individual and societal concepts in minority and majority
languages; childhood developmental perspectives; general bilingual education issues, bilingual and second
language classrooms, and political and multicultural perspectives.
In writing this introductory textbook, tough decisions had to be made as to what to include and exclude, what to
present in detail and what to summarize, what assumptions to explore and what to take ‘as read’. There was a
debate as to what level to pitch the analysis and prose to avoid being simplistic and patronizing but also to avoid
mystifying complexity. Another compromise surrounded developing a clear structure when there are myriads of
inter-connections between the parts. I have often been asked about a variety of chapters: ‘Why don’t you put
Chapter X earlier?’ I agree, everything should be earlier. The introductions to the two sections explain the order of
the book.
An attempt is made to balance the psychological and the sociological, the macro education issues and the micro
classroom issues, the linguistic and the sociopolitical, and to balance discussion at individual and societal levels.
The book is designed for a new generation of bilingual students. Therefore, an attempt is made to be inclusive of
the major concerns in bilingualism and bilingual education with a future perspective. Faced with the social and
political problems that surround bilinguals, students will find in this book an attempt to face constructively those
problems and recognize the positive values and virtues in a future bilingual and multicultural world.
In a wide ranging, foundational textbook, the order and structure of the contents is important. For the novice, the
structure may influence the accommodation of future reading and experience. This book starts with wide, macro
issues, and then funnels down to specific, micro issues. Definitional, sociological and psychological issues provide
the foundation. Discussions of bilingual education and bilingual classrooms are built on that essential foundation.
However, the book is more than a cross-disciplinary foundation with a series of education layers built on top.
Within the boundaries of clarity in writing style and readable structuring, explicit inter-connections are made
between chapters. A simple

< previous page page_xiv next page >
< previous page page_xv next page >
Page xv
map of these inter-connections is given in Appendix 4. This conceptual map will, for most, only make sense after
reading the whole book. The map also provides the instructor with a simplified but integrated plan of the book. It
is an attempt to avoid the compartmentalization that sometimes results from reading discrete chapters, to show
how themes link together into an integrated whole.
In writing the book, a constant challenge has been ‘From whose perspective?’ There are majority mainstream
viewpoints, relatively advantaged minority language viewpoints and various disadvantaged minority language
viewpoints. There are left wing and right wing politics, activist and construed vist ideas. The book attempts to
represent a variety of viewpoints and beliefs. Where possible, multiple perspectives are shared. Readers and
reviewers have kindly pointed out some of the hidden and implicit assumptions made, and kindly provided
alternative viewpoints that I have tried to represent faithfully in the text. Where there are conclusions and
dominating perspectives, I alone stand responsible.
Another problem has concerned generalization and contextualization. The book was written for an international
audience to reflect ideas that transcend national boundaries. The book attempts to locate issues of international
generalizability. Unfortunately, space limits discussion of a variety of regional and national language situations.
There are other writings that will provide necessary contextualization. A list of these is presented after the last
chapter. Where particular situations have been discussed (e.g. Canadian Immersion or US Language Minority
debates), it is usually because of the thoroughness of documentation and the depth of analysis in the surrounding
literature. Where bilingual situations are discussed, by inference this will usually include multilingual situations. In
this book, multilinguals are implied in the use of the term ‘bilinguals’.
In an attempt to make the contents of this book relevant to a variety of contexts and regions, various chapters focus
on integrating theories. From one individual research study, it is usually impossible to generalize. A study from
Canada may say little about Catalans. Results on six year olds may say nothing about sixteen or sixty year olds.
Research on middle-class children in a French and English, dual majority language environment (e.g. Canada) may
say little or nothing about children from a lower social class in a ‘subtractive’ bilingual environment (where the
second language is likely to replace the first language). Therefore, an overview and integration of major areas of
bilingual research is required. From a gray mass of research (e.g. on second language acquisition) and sometimes
from a paucity of research (e.g. on cognition and bilingualism), a theoretical framework will attempt to outline the
crucial parameters and processes. Thus a theoretical framework on a particular area of bilingualism may attempt to
do one or more of the following: attempt to explain phenomena; integrate a diversity of (apparently contradictory)
findings; locate the key parameters and interactions operating; be able to predict outcomes and patterns of
bilingual behavior; be capable of testing for falsification or refinement; express the various conditions that will
allow the theory to be appropriate in a variety of contexts.

< previous page page_xv next page >
< previous page page_xvi next page >
Page xvi
Organization of the Book
The book is divided into two sections: The Individual and Social Nature of Bilingualism, and Bilingual Education
Policies and Classroom Practices.
Section A:
The Individual and Social Nature of Bilingualism
The starting point of the book is an introduction to the language used in discussing bilingual education and
bilingualism. Not only are important terms introduced, but also key concepts, distinctions and debates which
underpin later chapters are presented. There are important dualisms and paradoxes throughout the study of
bilingualism: the individual bilingual person as different from groups and societies where bilinguals live; the
linguistic view compared with the sociocultural view; preservation and revivalist viewpoints; language skills and
language competences; subtractive and additive forms of bilingualism. These are introduced in the first section
which presents foundational issues that precede and influence discussions about bilingual education. Bilingual
education is a component in a wider whole and directly relates to underlying macro issues. Before we can sensibly
talk about bilingual education we need to tackle questions such as:
Who are bilinguals?
How does bilingual education fit into minority language maintenance, language decay and language revival?
How does a child become bilingual?
What effect does the home and the neighborhood play in developing bilingualism?
Does bilingualism have a positive or negative effect on thinking?
Section B:
Bilingual Education Policies and Classroom Practices
Having contextualized bilingual education inside more universal issues, the second section focuses on the many
aspects of bilingual education. The section commences with a broad discussion of different types of bilingual
education, followed by an examination of the effectiveness of those types. After a focus on systems of bilingual
education, the section proceeds to examine bilingual classrooms. Situations where bilingualism is fostered through
preserving the minority language, and situations involving the learning or acquisition of a second language are
discussed. The underlying questions are:
What forms of bilingual education are more successful?
What are the aims and outcomes of different types of bilingual education?
What are the essential features and approaches of a classroom fostering bilingualism?
What are the key problems and issues of bilingual classrooms?
How is a second language best learnt in a classroom setting?
The second section continues by looking at the political and cultural dimensions that surround bilingualism in
society (and bilingual education in particular). Different views of the overall value and purpose of bilingualism
join together many of the threads of the book, and culminate in a consideration of the nature of multiculturalism in
society, in school and the classroom. The finale of the book responds to integrating and concluding issues such as:

< previous page page_xvi next page >
< previous page page_xvii next page >
Page xvii
Why are there different viewpoints about language minorities and bilingual education?
Why do some people prefer the assimilation of language minorities and others prefer linguistic diversity?
Can schools play a role in a more multicultural and less racist society?
Study Questions and Study Activities
Study questions and study activities are placed at the end of each chapter. These are designed for revision purposes
and for students wishing to extend their learning by engaging in various practical activities. Such activities are
flexible and adaptable. Instructors and students will be able to vary them according to local circumstances. Many
more study activities are found in the Reader that accompanies this text.
To end the beginning. The motivating force behind the book is to introduce students to a positive world of
bilingualism and bilingual education, The book has been written for minority language students seeking to
understand and preserve, and for majority language students seeking to become more sensitized. The book is an
attempt to contribute to the conservation of a colorful world in which bilinguals and multiculturals are increasingly
important language preservationists and diplomats between cultures. Bilinguals help preserve the variety and the
beauty of the language garden of the world. Bilinguals are environmentally friendly people.

< previous page page_xvii next page >
< previous page page_1 next page >
Page 1
Section A of the book engages four foundational areas of bilingualism. Each area aims to provide a vocabulary and
a framework for understanding bilingual education and bilingual classrooms which are considered in Chapters 10
and 20. These four areas are:
The Terminology of Bilingualism
Most social science topics have a particular vocabulary used to refine discussion. Bilingualism and bilingual
education is no exception. Along with the introduction of vocabulary go important definitions and distinctions
which are examined mainly in Chapter 1, but also in Chapters 2, 3 and 4. The aim is to introduce not just
terminology but concepts and constructs needed to understand more deeply bilingual people and situations.
Terms and concepts raise issues about ‘measurement’. For example, ‘to what extent’ someone is bilingual suggests
that measurement (quantitative and qualitative) may sometimes be required. Chapter 2 examines the measurement
of bilingualism. Such measurement is not important in itself but aids conceptual clarification and is shown in
Chapter 2 to have direct curriculum relevance.
The Social Nature of Bilingualism
Chapters 3 and 4 move on from an ‘individual person’ consideration (which underlies Chapters 1 and 2) to a focus
on groups and communities of language speakers. Language speakers live in language communities. Such
language communities often change in their use of two languages. Therefore Chapters 3 and 4 consider key
societal topics such as language planning, language shift, language maintenance, language death, language vitality,
language revival and language reversal. The presupposition is that bilingual education is part of a wider social and
political movement. Education is used to effect

< previous page page_1 next page >
< previous page page_2 next page >
Page 2
language change; it is also affected by language shift. The social and political nature of two languages in contact is
returned to as a finale to the book.
The Individual Development of Bilingualism
Chapters 5 and 6 consider various routes in infancy, childhood and adulthood to achieving bilingualism. The role
of parents, community and society as well as individual differences in development are discussed. Through
presenting whole models and theories of bilingual development, the accent is on showing how social factors join
individual factors in dual language acquisition. The developmental path to bilingualism is not simple. Single
factors such as age of learning and aptitude in languages cannot be considered in isolation. Thus Chapter 6
provides holistic, interactive and integrated frameworks from different authors. The aim is to portray simply the
complexity of bilingual development through comprehensive major models and theories. Instead of integrating and
collapsing these theories, their original integrity as ‘stand-alone’ theories has been maintained.
Thinking and Bilingualism
Section A of the book concludes by tackling the question of whether bilingualism affects thinking. Are there
advantages, disadvantages or no effects for being bilingual when considering thinking skills and processes? Does
bilingualism affect a person’s ‘intelligence’? Do bilinguals think more creatively if they have dual vocabularies and
grammars? Are there differences between bilinguals and monolinguals inside the operating system of the brain?
These questions are tackled in Chapter 7 (‘intelligence’) and 8 (thinking products and processes). Chapter 9 draws
together parts of Chapters 5 to 8 by examining a theory that has evolved to explain research on the development of
languages and thinking. This theory is shown to have direct relevance to the classroom.

< previous page page_2 next page >
< previous page page_3 next page >
Page 3
Chapter 1
Definitions and Distinctions
Bilingual Ability
The Four Language Abilities
A Fifth Language Competence
Minimal and Maximal Bilingualism
Balanced Bilinguals
‘Semilingualism’/’Double Semilingualism’
Conversational Fluency and Academic Language
An Individual’s Use of Bilingualism

< previous page page_3 next page >
< previous page page_4 next page >
Page 4
Since a bicycle has two wheels and binoculars are for two eyes, it would seem that bilingualism is simply about
two languages. The aim of this chapter is to show that the ownership of two languages is not so simple as having
two wheels or two eyes. Ask someone if they are bilingual. While some may answer ‘yes’ or ‘no’, others would
wish to qualify their answer. Is someone bilingual if they are fluent in one language but less than fluent in their
other language? Is someone bilingual if they rarely or never use one of their languages? Such basic questions need
addressing before other topics in this book can be sensibly discussed.
Before examining these questions, it is important to make an initial distinction between bilingualism as an
individual phenomenon and bilingualism as a group or societal possession. Bilingualism and multilingualism can
be examined as the possession of the individual. Various themes in this book start with bilingualism as experienced
by individual people. For example, a discussion of whether or not bilingualism affects thinking requires research
on individual monolinguals and bilinguals. From sociology, sociolinguistics, politics, geography, education and
social psychology comes a different perspective. Bilinguals and multilinguals are usually found in groups. Such
groups may be located in a particular region (e.g. Catalans in Spain), or may be scattered across communities (e.g.
Chinese in the US). Bilinguals may form a distinct language group as a majority or a minority. Bilinguals and
multilinguals within a country may be analyzed as a distinct group. For example, linguists study how the
vocabulary of bilingual groups change across time. Geographers plot the density of bilinguals in a country.
Educationists examine bilingual educational policy and provision for minority language groups.
The first distinction is therefore between bilingualism as an individual possession and as a group possession. This
is usually termed individual bilingualism and societal

< previous page page_4 next page >
< previous page page_5 next page >
Page 5
bilingualism. Like most distinctions, there are important links between the two parts. For example, the attitudes of
individuals towards a particular language may affect language maintenance, language restoration, language shift or
language death in society. In order to understand the term ‘bilingualism’, some important further distinctions at the
individual level are discussed in this chapter. An introduction to bilingualism as a group possession (societal
bilingualism) is provided in Chapter 3.
If a person is asked whether he or she speaks two languages, the question is ambiguous. A person may be able to
speak two languages, but tends to speak only one language in practice. Alternatively, the individual may regularly
speak two languages, but competence in one language may be limited. Another person will use one language for
conversation and another for writing and reading. The essential distinction is therefore between ability and use.
This is sometimes referred to as the difference between degree and function. This chapter continues by examining
bilinguals’ language abilities. Language use is discussed later.
Before discussing the nature of language abilities, a note about terminology. Entry into the many areas of
bilingualism and bilingual education is helped by understanding often used terms and distinctions. There exists a
range of terms in this area: language ability, language achievement, language competence, language performance,
language proficiency and language skills. Do they all refer to the same entity, or are there subtle distinctions
between the terms? To add to the problem, different authors and researchers sometimes tend to adopt their own
specific meanings and distinctions. There is no standardized use of these terms (Stern, 1992).
Language skills tend to refer to highly specific, observable, clearly definable components such as handwriting. In
contrast, language competence is a broad and general term, used particularly to describe an inner, mental
representation of language, something latent rather than overt. Such competence refers usually to an underlying
system inferred from language performance. Language performance hence becomes the outward evidence for
language competence. By observing general language comprehension and production, language competence may
be presumed. Language ability and language proficiency tend to be used more as ‘umbrella’ terms and therefore
used somewhat ambiguously. For some, language ability is a general, latent disposition, a determinant of eventual
language success. For others, it tends to be used as an outcome, similar but less specific than language skills,
providing an indication of current language level. Similarly, language proficiency is sometimes used
synonymously with language competence (e.g. Ellis, 1985); other times as a specific, measurable outcome from
language testing. However, both language proficiency and language ability are distinct from language achievement
(attainment). Language achievement is usually seen as the outcome of formal instruction. Language proficiency
and language ability are, in contrast, viewed as the product of a variety of mechanisms: formal learning, informal
uncontrived language acquisition (e.g. on the street) and of individual characteristics such as ‘intelligence’.

< previous page page_5 next page >
< previous page page_6 next page >
Page 6
Bilingual Ability
The Four Language Abilities
If we confine the question ‘Are you bilingual?’ to ability in two languages, the issue becomes ‘what ability’? There
are four basic language abilities: listening, speaking, reading and writing. These four abilities fit into two
dimensions: receptive and productive skills; oracy and literacy. The following table illustrates:
Oracy Literacy
Receptive skills Listening Reading
Productive skills Speaking Writing
The table suggests avoiding a simple classification of who is, or is not, bilingual. Some speak a language, but do
not read or write in a language. Some listen with understanding and read a language (passive bilingualism) but do
not speak or write that language. Some understand a spoken language but do not themselves speak that language.
To classify people as either bilinguals or monolinguals is too simplistic. Or, to return to the opening analogies, the
two wheels of bilingualism exist in different sizes and styles. The two lenses of bilingualism will vary in strength
and size.
The four basic language abilities do not exist in black and white terms. Between black and white are not only
many shades of gray; there also exist a wide variety of colors. The multi-colored landscape of bilingual abilities is
now portrayed. Each language ability can be more or less developed. Reading ability can be simple and basic to
fluent and accomplished. Someone may listen with understanding in one context (e.g. shops) but not in another
context (e.g. an academic lecture). This suggests that the four basic abilities can be further refined into sub-scales
and dimensions. There are skills within skills. The main abilities have traditionally been listed as: pronunciation,
extent of vocabulary, correctness of grammar, the ability to convey exact meanings in different situations and
variations in style.
The range and type of sub-skills that can be measured is large and debated (Lado, 1961; Mackey, 1965;
Macnamara, 1969; Oller, 1979; Carroll, 1980; Baetens Beardsmore, 1986). Language abilities such as speaking or
reading can be divided into increasingly microscopic parts. Numerous colors are needed to paint an accurate
picture. What in practice is tested and measured to portray an individual’s bilingual performance is considered later
in the book. What has emerged so far is that a person’s ability in two languages will tend to evade simple
categorization. Language abilities are multidimensional.
If language abilities are multicolored, and if bilinguals have a range of colors in both languages, then positive
terms are needed to portray the variety. Calling bilinguals ESL (English Second Language) students in Canada and
Britain, LEP (Limited English Proficiency) students in the US is negative and pejorative. Such labels stress
children’s perceived deficiency rather than their proficiencies, children’s perceived ‘deprivation’

< previous page page_6 next page >
< previous page page_7 next page >
Page 7
rather than their accomplishments, their lower, marginalized, minority status through majority eyes rather than
their bilingual potentiality. Such labels stress past and present performance rather than potentialities and the
possibility of full bilingualism.
A Fifth Language Competence
The four basic language abilities are commonly regarded as speaking, listening, reading and writing. However,
there are times when a person is not speaking, listening, reading or writing but still using language. As Skutnabb-
Kangas (1981) proposes, the language used for thinking may be a fifth area of language competence. This may be
simply termed inner speech and placed under the umbrella title of ‘speaking’. It may alternatively be worth
differentiating from actual speaking as it raises the dimension of the ability of bilinguals to use both languages as
thinking tools. Cummins (1984b) expresses this notion as cognitive competence in a language. That is, the ability
to use one or both languages for reasoning and deliberation.
Minimal and Maximal Bilingualism
So far, it has been argued that deciding who is or is not bilingual is difficult. Simple categorization is arbitrary and
requires a value judgment about the minimal competence needed to achieve a label of ‘bilingual’. Therefore, a
classic definition of bilingualism such as ‘the native-like control of two or more languages’ (Bloomfield, 1933)
appears too extreme and maximalist (‘native like’). The definition is also ambiguous ( what is meant by ‘control’
and who forms the ‘native’ reference group?). At the other end is a minimalist definition, as in Diebold’s (1964)
concept of incipient bilingualism. (See the end of chapter for further terms and distinctions in bilingualism.) The
term incipient bilingualism allows people with minimal competence in a second language to squeeze into the
bilingual category. Tourists with a few phrases and business people with a few greetings in a second language
would be incipient bilinguals. The danger of being too exclusive is not overcome by being too inclusive. Trawling
with too wide a fishing net will catch too much variety and therefore make discussion about bilinguals ambiguous
and imprecise. Trawling with narrow criteria may be too insensitive and restrictive.
Who is categorized as a bilingual or not will depend on the purpose of the categorization. At different times,
governments, for example, may wish to include or exclude language minorities. Where a single indigenous
language exists (e.g. in Ireland and Wales), a government may wish to maximize its count of bilinguals. A high
count may indicate government success in its indigenous language policy. In comparison, in a suppressive,
assimilationist approach, minority languages and bilinguals may be minimized (e.g. England).
Is there a middle ground in-between maximal and minimal definitions? The danger is in making arbitrary cut-off
points about who is bilingual or not along the competence dimensions. Differences in classification will exist in
terms of what language abilities make someone bilingual, and in terms of how much ability someone must possess
to be labeled as bilingual.

< previous page page_7 next page >
< previous page page_8 next page >
Page 8
One alternative is to move away from the multi-colored canvas of proficiency levels to a portrait of everyday use
of the two languages. Categorization of bilinguals by their use of language is considered later in the chapter.
Before such consideration, important labels and distinctions in terms of language ability are discussed.
Balanced Bilinguals
The literature on bilingualism frequently spotlights one particular group of bilinguals whose competences in both
languages are well developed. Someone who is approximately equally fluent in two languages across various
contexts may be termed an equilingual or ambilingual or, more commonly, a balanced bilingual. As will be
considered in Chapter 8, balanced bilinguals are important when discussing the possible cognitive advantages of
Balanced bilingualism is sometimes used as an idealized concept. Fishman (1971) has argued that rarely will
anyone be equally competent across all situations. Most bilinguals will use their two languages for different
purposes and functions. For example, a person may use one language at work; the other language at home and in
the local community.
Balanced bilingualism is also a problematic concept for other reasons. The balance may exist at a low level of
competence in the two languages. Someone may have two relatively undeveloped languages which are
nevertheless approximately equal in proficiency. While this is within the literal interpretation of ‘balanced’
bilingual, it is not the sense employed by many researchers on bilingualism. The implicit idea of balanced
bilingualism has often been of ‘reasonable’ or ‘good’ competence in both languages. A child who can understand
the delivery of the curriculum in school in either language, and operate in classroom activity in either language
would be an example of a balanced bilingual.
Should we conclude that balanced bilingualism is of no use as a term? While it has limitations of definition and
measurement, it has proved to be of value in research and discussion (see Chapter 8). However, categorizing
individuals into such groups raises the issue of comparisons.
What is judged normal, proficient, skilled, fluent or competent? Who judges? The danger may be in using
monolinguals as the point of reference. Grosjean (1985) argues that comparing bilinguals and monolinguals does
not compare like with like. Can we fairly judge a 100 meter sprinter against a decathlete? Can we justly compare
someone who only uses the breast stroke with a medley swimmer? Is it inappropriate to compare monolingual
proficiency and dual language proficiency? Should bilinguals only be measured and categorized by reference to
other bilinguals? When, for example, someone learns English as a second language, should that competency in
English be measured against monolingual English speakers or against ‘balanced’ bilinguals?
‘Semilingualism’/’Double Semilingualism’
Bilinguals will tend to be dominant in one of their languages in all or some of their language abilities. This may
vary with context and change over time. Dominance in one

< previous page page_8 next page >
< previous page page_9 next page >
Page 9
language may change over time with geographical or social mobility. For others, the dominance may be relatively
stable across time and place. The topic of dominance will be considered later (Chapter 2) when tests are discussed.
For the present, there is a group that has been proposed as distinct from balanced and dominant bilinguals.
Sometimes termed pejoratively as semilinguals or double semilinguals, the group is regarded as not having
‘sufficient’ competence in either language.
Hansegård (1975; see Skutnabb-Kangas, 1981) described semilingualism in terms of deficits in six language
Size of vocabulary
Correctness of language
Unconscious processing of language (automatism)
Language creation (neologization)
Mastery of the functions of language (e.g. emotive, cognitive)
Meanings and imagery
Thus a semilingual is seen as someone with quantitative and qualitative deficiencies in both their languages when
compared with monolinguals.
A ‘semilingual’ is considered to exhibit the following profile in both their languages: displays a small vocabulary
and incorrect grammar, consciously thinks about language production, is stilted and uncreative with each language,
and finds it difficult to think and express emotions in either language.
The notion of semilingualism, or double semilingualism, has received criticism (e.g. Skutnabb-Kangas, 1981).
There are six major problems. First, the term has taken on disparaging and belittling overtones, particularly in
Scandinavia and with in-migrant groups in the US. [The term in-migrant is used throughout the book to avoid the
negative connotations of the term ‘immigrant’ and to avoid the imprecise and loaded distinctions between migrant
workers, guest workers, short stay, long stay and relatively permanent in-migrants]. Semilingualism may be used
as a negative label which invokes expectations of underachievement which may evoke a self-fulfilling prophecy.
Second, if languages are relatively undeveloped, the origins may not be in bilingualism per se, but in the
economic, political and social conditions that evoke under-development. This is a theme considered in detail in
later chapters. The danger of the term semilingualism is that it locates the origins of under-development in the
internal, individual possession of bilingualism, rather than in external, societal factors that co-exist with
bilingualism. Thus the term may be a political rather than a linguistic concept. Third, many bilinguals use their two
languages for different purposes and events. Language may be specific to a context. A person may be competent in
some contexts but not in others.
Fourth, the educational tests that are most often used to measure language proficiencies and differentiate between
people may be insensitive to the qualitative aspects of languages and to the great range of language competences.
Language tests may measure a small, unrepresentative sample of a person’s total language behavior. This is
considered in detail in the next chapter.

< previous page page_9 next page >
< previous page page_10 next page >
Page 10
Fifth, there is dispute regarding the frequency of double semilingualism, for example among Finnish-Swedish
speakers. How many or how few fit clearly into a semilingual category will be disputed. Establishing a cut-off
point for who is and is not a double semilingual will be arbitrary and value laden. Is the term an empty concept
lacking in meaning? There is a lack of sound objective empirical evidence on such a categorization.
Sixth, the comparison with monolinguals may not be fair. It is important to distinguish if bilinguals are ‘naturally’
qualitatively and quantitatively different from monolinguals in their use of their two languages (as a function of
being bilingual). An apparent deficiency may be due to unfair comparisons with monolinguals.
The criticisms raise serious doubts about the value of the term ‘semilingualism’. However, this does not detract
from there being many language abilities on which people do differ, with some people at the earlier stages of
development. Being at an early stage may not be a result of being bilingual. Economic and social factors or
educational provision may, for example, be the cause of under-development in language. Rather than highlight the
apparent ‘deficit’ in language development, the more positive approach is to emphasize that, when suitable
conditions are provided, languages are easily capable of evolution beyond the ‘semi’ state.
Conversational Fluency and Academic Language Competence
So far, the chapter has centered on the variety of language abilities and the danger of categorization using a small
or biased selection of language sub-skills. The question is whether the variety of sub-skills can be reduced to a
small number of important dimensions. Hernández-Chávez et al. (1978), for example, suggest there are 64 separate
components to language proficiency. In comparison, tests abound which purport to measure reading ability as a
single entity. Many reading tests tacitly assume that reading can be reduced to one dimension (Levy & Goldstein,
Is it the case that children who perform well on a spelling test also do well on an oral comprehension test? Oller’s
(1982) research suggested that different language skills do tend to correlate moderately well. If a variety of
measures of speaking, reading and writing are analyzed, the claim was of one underlying dimension. The
individual test scores could be collapsed into one global score. The overlap was sufficient for Oller & Perkins
(1980) to suggest that there exists a single factor of global language proficiency. This global factor is seen as co-
existing with other specific language factors. However, such specific factors are seen as measuring relatively
minor language skills.
The idea of a global language factor is contentious as Oller (1982) admits. ‘The evidence now shows that there are
both global and componential aspects of language proficiencies. The perfect theory of the right mix of general and
specific components, however, has not been foundand probably will never be agreed on’ (Oller, 1982: 710). Oller’s
(1982) idea of a global language factor is based on quantitative testing. As will be considered later in the book,
such tests leave qualitative differences between people unexplored.

< previous page page_10 next page >
< previous page page_11 next page >
Page 11
Oller’s (1982) much disputed claim for one global language factor provides a starting point for a distinction
between two different language abilities. Oller’s (1982) language proficiency factor has been allied to the language
abilities needed to cope in the classroom. Most (but not all) language tests are closely linked to the cognitive,
academic language skills of the classroom. Reading and writing tests are obvious examples. The notion of a
curriculum based language competence led various authors to make an important distinction. Apart from
academically related language competence, it has been proposed that there is a conceptually distinct category of
conversational competence. Skutnabb-Kangas & Toukomaa (1976) proposed a difference between surface fluency
and academically related aspects of language competence. Surface fluency would include the ability to hold a
simple conversation in the shop or street and may be acquired fairly quickly (e.g. in two or three years) by second
language learning. To cope in the curriculum, conversational language competence may not be enough.
Academically related language competence in a second language may take from five to seven years or longer to
acquire. This theme is considered in detail later in the book when a distinction is made between basic interpersonal
communicative skills and cognitive/academic language proficiency. Such a distinction between two levels of
language competence is important as it involves disputing Oller’s (1982) ‘single factor’ language skill.
The chapter now moves to focusing on language use rather than language ability.
An Individual’s Use of Bilingualism
In discussing an individual’s language competence, it has become evident that language cannot be divorced from
context. Language is not produced in a vacuum; it is enacted in changing dramas. As props and scenery, co-actors
and actresses, the play and the part played change, so does language. A pure linguistic or psychological approach
to two language competences is not sufficient. Communication includes not only the structure of language (e.g.
grammar, vocabulary) but also who is saying what, to whom, in which circumstances. One person may have
limited linguistic skills but, in certain situations, be successful in communication. Another person may have
relative linguistic mastery, but through undeveloped social interaction skills, be relatively unsuccessful in
communication. The social environment where the two languages function is crucial to understanding bilingual
usage. This section considers the use and function of an individual’s two languages.
An individual’s use of their bilingual ability (functional bilingualism) moves away from the complex, unresolvable
arguments about language proficiency that tend to be based around school success and academic performance.
Functional bilingualism moves into language production across an encyclopedia of everyday events. Functional
bilingualism concerns when, where, and with whom people use their two languages (Fishman, 1965).
To categorize when, where, and with whom a person uses either language will vary from culture to culture. The
nature and range of social events vary from region to region,

< previous page page_11 next page >
< previous page page_12 next page >
Page 12
sub-culture to sub-culture. The table below provides examples of the different targets (people) and contexts
(domains) where functional bilingualism is enacted.
Examples of
Language Targets
Examples of Language Contexts (Domains)
1. Nuclear Family 1. Shopping
2. Extendend Family
2. Visual and Auditory Media (e.g. TV,
Radio, Records, Cassettes, CDs,Video)
3. Work Colleagues 3. Printed Media (e.g. Newspapers, Books)
4. Friends 4. Cinema/Discos/Theater/Concerts
5. Neighbors 5. Work
6. Religious Leaders
6. Correspondence/Telephone/Official
7. Teachers
7. Clubs, Societies, Organizations, Sporting
8. Presidents,
Principals, Other
8. Leisure & Hobbies
9. Bureaucrats 9. Religious Meetings
10. Local
Community 10. Information Technology (e.g.
A distinction needs to be made between functional bilingualism and language background. (Baker & Hinde, 1984;
Baker, 1985). Language background is a wider concept, referring to both participative and non-participative
experience of language. Non-participative language background is indirect, bystander experience, and measured by
questions such as ‘What language does your mother speak to your father when you are present?’. Functional
bilingualism is a narrower concept, concerning direct involvement in a language domain. Functional bilingualism
is therefore restricted to the personal production and reception of language (i.e. speaking, writing, reading and
direct listening in various domains). Functional bilingualism requires the study of five actions.
(1) Who is the subject? (i.e. who is the speaker?)
(2) Who is the language target ? (i.e. who is the listener(s)?)
(3) What is the situation? (e.g. in the factory, classroom, mosque)
(4) What is the topic of conversation? (e.g. sport, work, food)
(5) For what purpose? To what effect?
As one or more of these five factors changes, so may the language used. This suggests that language choice can be
the result of a large and complex set of factors. In trying to predict ‘who will speak what language, when and to
whom’ (Fishman, 1965), an individual’s decision may be intricate. Sankoff(1972) uses a decision tree to give a
taste of the complexity of choice.

< previous page page_12 next page >
< previous page page_13 next page >
Page 13
The illustration below (adapted from Appel & Muysken, 1987) illustrates the choice of an adult Berber-speaking
Moroccan in the Netherlands. The table commences with a Moroccan speaker and ends with a decision about
which of five languages to speak. The decision will be based on who is the target person in the conversation and
whether the situation is formal or informal. Romaine (1995: 32) gives a more complex example.
Inevitably, a decision tree oversimplifies and abbreviates a complex choice. It suggests the decision is logical,
consistent across time and place, and is predictable. The reality is often less simply patterned and predetermined.
Psychological factors are important. For example, someone may decide to use the language perceived as being
more socially desirable or prestigious, or as more accommodating to the listener.
Sometimes a speaker may mix and switch between languages, (e.g. to explain an idea more exactly, to include
other listeners, to tell a story). Often, bilinguals use their two languages in different domains (e.g. work compared
with home; grandparents compared with offspring).
A consideration of distinctions and classifications about bilingual ability and function often moves to
measurement. To what extent can we measure someone’s performance in their two languages? How can we portray
when, where and with whom people use their two languages? What are the problems and dangers in measuring
bilinguals? These questions provide the themes for the next chapter.
Defining who is or is not bilingual is essentially elusive and ultimately impossible. Some categorization is often
necessary and helpful to make sense of the world. Therefore categorizations and approximations may be required.
Definitions in a few sentences (e.g.

< previous page page_13 next page >
< previous page page_14 next page >
Page 14
Bloomfield’s (1933) ‘native-like control of two languages’ are of little help. Intrinsically arbitrary and ambiguous in
nature, they can be easily criticized and are difficult to defend.
A more helpful approach may be to locate important distinctions and dimensions surrounding the term
‘bilingualism’. These should help to refine thinking about bilingualism. The foundational distinction is between
bilingual ability and bilingual usage. Some bilinguals may be fluent in two languages but rarely use both. Others
may be much less fluent but use their two languages regularly in different contexts. Many other patterns are
possible. This distinction leads naturally into dimensions. In terms of proficiency in two languages, the four basic
dimensions are listening, speaking, reading and writing. Thinking in those languages may be a fifth language
proficiency. With each of these proficiency dimensions, it is possible to fragment into more and more microscopic
and detailed dimensions (e.g. pronunciation, vocabulary, grammar, meaning and style). Those subdimensions can
subsequently be further dissected and divided.
Creating a multidimensional, elaborate structure of bilingual proficiency may make for sensitivity and precision.
However, ease of conceptualization requires simplicity rather than complexity. Therefore simple categorization is
the paradoxical partner of complex amplification. This chapter has focused on the categories of balanced
bilingualism, semilingualism, and one-factor ideas of language ability. These categories have received some depth
of discussion and critical response. As will be revealed in later chapters, these categories also relate to central
research on bilingualism and bilingual education.
Separate from bilingual ability is a person’s use of their two languages. That is, what use is made of two languages
by the individual; when, where and with whom? This highlights the importance of considering domain or context.
As a bilingual moves from one situation to another, so may the language being used in terms of type (e.g. Spanish
or English), content (e.g. vocabulary) and style. Over time and place, an individual’s two languages are never static
but ever changing and evolving.
Suggested Further Reading
BAETENS BEARDSMORE, H., 1986, Bilingualism: Basic Principles. Clevedon: Multilingual Matters.
HOFFMANN, C., 1991, An Introduction to Bilingualism. London: Longman.
SKUTNABB-KANGAS, T., 1981, Bilingualism or Not: The Education of Minorities. Clevedon: Multilingual
Review and Study Questions
(1) Write short revision notes on the following:
(i) The differences and overlaps in the following terms: language skills, language competence, language
performance, language ability, language proficiency and language achievement.
(ii) Balanced bilinguals and the notion and criticisms of ‘double semilingualism’.

< previous page page_14 next page >
< previous page page_15 next page >
Page 15
(iii) The difference between conversational fluency and academic language competence.
(iv) The different domains/contexts where languages are used.
(2) What are the main differences between language ability and language use? In words or in a table, locate
differences between these terms. Provide examples to illustrate the major differences.
(3) In your language context, how valid and how useful do you find the terms ‘balanced bilingual’ and
‘semilingualism’? What problems do you see in using these terms to describe children and adults?
(4) What labels are used in your region to describe bilinguals? What problems and potential do you find in the use
of these terms? Are there preferable terms?
Study Activities
(1) Do you consider yourself and/or people known to you as ‘bilingual’? Would you describe yourself, or someone
known to you, as ‘balanced’ in their languages? Which language or languages do you think in? Does this change in
different contexts? In which language or languages do you dream?
(2) This activity can be based on self-reflection or you may wish to interview someone who is bilingual. Make a
table or diagram to illustrate how one person’s dual or multilingual ability and language usage has changed and
developed since birth. Write down how different contexts have affected that change and development. The
diagram or table should illustrate the life history of someone’s bilingualism indicating changes over time and over
different contexts.
Further Distinctions & Descriptions
Achieved Bilingualism: acquisition of bilingualism later than childhood. Also termed Successive Bilingualism. See
Chapter 6.
Ascribed Bilingualism: acquisition of bilingualism early in childhood. Related terms are Infant Bilingualism,
Consecutive Bilingualism and Simultaneous Bilingualism. See Chapter 6
Biliteracy: reading and writing in two languages. Biliteracy is less common than bilingualism due to the increased
complexity of skills (Hornberger, 1989). See Chapters 15 and 16.
Compound Bilingualism: one language learnt distinctly later than the other often in separate contexts (Weinreich,
1970). See Chapter 6.
Co-ordinate Bilingualism: two languages learnt during childhood in a fused context (Weinreich, 1970). See
Chapter 6.
Diagonal Bilingualism: situations where a ‘non standard’ language or a dialect co-exists with an unrelated
‘standard’ language (Pohl, 1965).
Dominant Language: the language with the greater proficiency and/or usage.
First Language: sometimes used to refer to the first language learnt; sometimes to the language most used;
sometimes to the stronger language. Second Language is similarly ambiguous, being used variously to indicate a
weaker language, a learnt

< previous page page_15 next page >
< previous page page_16 next page >
Page 16
language (as in secondary bilingualism), the chronology in acquisition, or the ‘less-used’ language.
Horizontal Bilingualism: situations where two languages have similar or equal status (Pohl, 1965).
Incipient Bilingualism: the early stages of bilingualism where one language is not strongly developed (Diebold,
Mother Tongue: the term is used ambiguously. It variously means (a) the language learnt from the mother (b) the
first language learnt, irrespective of ‘from whom’ (c) the stronger language at any time of life (d) the ‘mother
tongue’ of the area or country (e.g. Irish in Ireland) (e) the language most used by a person (f) the language to
which a person has the more positive attitude and affection (see Skutnabb-Kangas & Phillipson, 1989).
Preferred Language: self assessment of the more proficient language (Dodson, 1981).
Primary Bilingualism: where two languages have been learnt ‘naturally’ (not via school teaching) (Houston, 1972).
See Chapter 6.
Productive Bilingualism: speaking and writing in the second language as well as listening and reading.
Receptive Bilingualism: understanding and reading a second language without speaking or writing in that second
Secondary Bilingualism: where the second language has been formally learnt. (Houston, 1972). See Chapter 14.
Vertical Bilingualism: situations where two related languages (or a language and a dialect) co-exist, particularly
within the individual (Pohl, 1965).
See also the Glossary in GARCÍA, O. and BAKER, C., 1995, Policy and Practice in Bilingual Education.
Clevedon: Multilingual Matters.

< previous page page_16 next page >
< previous page page_17 next page >
Page 17
Chapter 2
The Measurement of Bilingualism
The Pruposes of the Measurement of Bilinguals
Examples of the Measurement of Bilinguals
Language Background Scales
Self Rating on Proficieny
Language Balance and Dominance Measures
Communicative Language Testing
Criterion Referenced Language Tests
The Structure of Language Competence
Canale and Swains’s of Language Competence
Bachman’s Model of Language Competence

< previous page page_17 next page >
< previous page page_18 next page >
Page 18
Having discussed definitions, dimensions and distinctions, the topic of measuring bilinguals both elaborates and
illuminates that discussion. Problems of categorizing ‘bilinguals’ makes further sense when we attempt to measure
and categorize. We start by returning to the paradox of the last chapter. It is natural and meaningful to try to
categorize the complexity of individual differences in bilingualism. We make sense of our world by continual
classification. People are constantly compared and contrasted. Yet the simplification of categorization often hides
the subtle complexity of reality. Sorting often simplifies unsympathetically. Individual differences are reduced to
similarities. Yet over-complexity can be frustrating and confusing. Complications can confound those needing
order and pattern. The measurement of bilinguals attempts to locate similarities, order and pattern.
The Purposes of the Measurement of Bilinguals
Measurement of bilinguals can take place for a variety of purposes, and it is important initially to differentiate
between some of these overlapping aims.
An example of the measurement of bilinguals is found in Census questions, requesting information about ability or
usage in two or more languages (e.g. in US, Canada, Ireland, Israel). Such Census data allows a researcher to
estimate the size and distribution of bilinguals in a particular area. For example, geographers map the proportion
and location of minority language groups within a state or region.

< previous page page_18 next page >
< previous page page_19 next page >
Page 19
Bilinguals may be distinguished as a ‘separate’ group for selection purposes. For example, a school may wish to
allocate children to classes or groups based on their degree of bilingual proficiency or language background. A
different example is measuring bilinguals at the outset of research. An investigation may require the initial
formation of two or more groups (e.g. ‘balanced’ bilinguals, ‘partial’ bilinguals and monolinguals).
When measuring the current performance level of a person, a wide variety of language proficiency and
achievement tests are available (e.g. reading comprehension, reading vocabulary, spelling, grammar). Such tests
may be used in schools to measure the four basic language abilities. In a minority language context, emphasis is
often on measuring proficiency in both the minority language and the majority language. In the United States,
emphasis on minority language groups becoming proficient in English has been a dominant issue in the testing of
bilinguals. Summing up someone’s language proficiency may occur at the end of a semester or a school year.
With proficiency testing, the measurement of bilinguals becomes fused with second language testing and the
general area of language testing. Apart from general language tests, there are measures of the relative dominance
of a person’s two languages and the mixing of a person’s two languages (sometimes pejoratively termed
‘interference’ see Chapters 5 & 6). Such tests spotlight the particular characteristics of bilinguals as different from
second language learners and first language development. Examples are provided later in this chapter.
Language proficiency tests are often used for summative judgments. In comparison, a test or assessment device
that gives feedback during learning, and to aid further development is formative assessment. A student may be
profiled on a precise breakdown of language skills to provide facilitative feedback to the teacher that directly leads
to action. This is the notion of formative testing. If the test reveals areas where a child’s language needs
developing, there can be immediate intervention and remedial help. A diagnosis of a problem in language may lead
to the formation of a plan to effect a remedy. The assessment of bilinguals is considered in Chapter 13.
Examples of the Measurement of Bilinguals
A full inventory of bilingual measurement devices would be immense and is not provided. The examples given
below help to make essential points.
Language Background Scales
Language background or functional bilingualism scales are self rating scales. They endeavor to measure actual use
of two languages as opposed to proficiency.
An example for schoolchildren is now presented (adapted from Baker, 1992):

< previous page page_19 next page >
< previous page page_20 next page >
Page 20
Here are some questions about the language in which you talk to different
people, and the language in which certain people speak to you. Please
answer as honestly as possible. There are no right or wrong answers.
Leave an empty space if a question does not fit your position.
In which language do YOU speak to the following people? Choose one of
these answers
In Spanish
more often
than English
In Spanish
and English
In English
more often
than Spanish
Friends in the
Friends on the
Other relatives
Friends outside
In which language do the following people speak TO YOU?
In Spanish
more often
than English
In Spanish
and English
In English
more often
than Spanish
Friends in the
Friends on the
Other relatives
Friends outside

< previous page page_20 next page >
< previous page page_21 next page >
Page 21
Which language do YOU use with the following?
In Spanish
more often
In Spanish
In English
more often
Watching TV/Videos
Listening to Radio
Playing Sport
On the Telephone
Reading Books
Earning Money
Other Leisure
This scale has limitations besides the problems of ambiguity and ‘social desirability’ considered later. It is not
exhaustive of targets (people) or of domains (contexts). Language activity with uncles and aunts, discos,
correspondence, organizations, clubs, societies, leisure, hobbies and travel are not included, for example. The
choice of items included in such a scale is somewhere between an all-inclusive scale and a more narrow sample of
major domains. At first glance, it may appear the more inclusive a scale the better. There is a problem, illustrated
by Baker & Hinde (1984: 46):
‘A person who says she speaks Welsh to her father (mostly away at sea), her grandparents (seen once a
year), her friends (but tends to be an isolate), reads Welsh books and newspapers (only occasionally), attends
Welsh Chapel (marriages and funerals only) but spends most of her time with an English speaking mother
and in an English speaking school might gain a fairly high ”Welsh” score.’
The example suggests that the ‘to whom’ question is insufficient. Frequency of usage in such contexts and with
certain targets needs adding. Besides ‘to whom and where’, a ‘how often’ question is necessary. Further problems of
language background and functional bilingualism scales are discussed by Baker & Hinde (1984) and Baker (1985).

< previous page page_21 next page >
< previous page page_22 next page >
Page 22
Self Rating on Proficiency
Two examples of self rating on language proficiency are provided: the first from Census questions; the second
from survey research. In a Census, questions are sometimes included about speaking a minority language. For
example, a Census may ambiguously ask ‘Does the person speak Welsh’ (Baker, 1985). This question reveals a
failure to discriminate between language ability and language use. Some can but don’t. Others do but with
difficulty. In the US, Census questions have been phrased as ‘Does this person speak a language other than English
at home? [If yes] What is the language? How well does this person speak English? Very well/Well/Not well/Not at
all. This US Census question concerns non-English language usage rather than ability through specifying
production at home. However, restricting the context to home ignores the wide variety of other contexts where
languages may be used. The US Census question on bilingualism then switches to ability in English. Bilingualism
is thus implicitly portrayed as use of a language other than English, and (in contrast) ability in English.
In the 1986 Canadian Census, one of the three language questions was ‘Can you speak English or French well
enough to conduct a conversation? This again reveals the ambiguity of Census questions. The terms ‘speak’, ‘well
enough’ and ‘conversation’ may be interpreted in different ways by different people. What one person considers
‘well enough’ may be at a different level of fluency to another. A brief conversation in the shop can be at a
different level from a conversation in the classroom or in the President’s office.
A current movement in education is to involve students in their own assessment. While this will only form a small
part of the assessment of language in school, it may help students to understand ‘why’ and ‘how’ their work is
assessed, aiding motivation and interest. Students may assess their language strengths and weaknesses, review
progress made in a semester, and set future learning targets. This may give students an increased awareness of the
context, structure and evolution of classroom language activity as it affects them personally. While open to abuse
and inaccuracy, self assessment in partnership with teacher assessment can be a powerful tool of teacher-student
collaboration, and of increased student responsibility for their work.
The second example of self rating on language proficiency comes from a survey by the Linguistic Minorities
Project (1985: 349). Children in London were asked to rate themselves on the four basic dimensions of language
competency (see opposite).
Limitations in Measurement
The self rating covers the basic four language abilities across two languages (e.g. Spanish and English). The
answers are possibly too broad (e.g. there are many graduations possible in between ‘yes’, ‘quite well’ and ‘only a
little’. Apart from this problem of scaling, there are other problems frequently encountered with measuring
language competence. These may be listed as:
(1) Ambiguity. Words such as ‘speak’, ‘understand’, ‘read’ and ‘write’ include a wide variety of levels of proficiency.
The range is from those with minimal proficiency to Bloomfield’s (1933) maximum notion of ‘native-like control
of two languages’.

< previous page page_22 next page >
< previous page page_23 next page >
Page 23
Tick one box for each question for each language
Can you understand this language if it is
spoken to you now? __
Yes, quite
well __
Only a
little __
No, not
now __
Can you speak this language now? __
Yes, quite
well __
Only a
little __
No, not
now __
Can you read this language now? __
Yes, quite
well __
Only a
little __
No, not
now __
could __
Can you write this language now? __
Yes, quite
well __
Only a
little __
No, not
now __
could __
Allowing a range from learner to expert casts a very wide net without separating the catch into different varieties.
(2) Context. A bilingual may be able to understand a language in one context (e.g. a shop) and not in another
context (e.g. academic lecture). Another bilingual may be able to read newspapers but not textbooks. Proficiency
and usage will vary with changing environments. A response summated across contexts is required. This may not
be sensitive to different levels of proficiency across different contexts.
(3) Social desirability. Respondents may consciously or unconsciously give a ‘halo’ version of themselves. Self
ratings are vulnerable to exaggeration or understatement. People may say they are fluent in a second language
(when in reality they are not) for self esteem or status reasons. Others may indicate they do not speak a language
when they can. This may occur, for example, in a low prestige, ‘subtractive’ minority language environment where
the introduction of the second language may replace the first language. Questions about proficiency can be
interpreted as political referendum or attitudinal questions (Baker, 1985; Baetens Beardsmore, 1986).

< previous page page_23 next page >
< previous page page_24 next page >
Page 24
(4) Acquiescent response. There is a slight tendency of respondents to answer ‘yes’ rather than ‘no’ in self-rating
questions. It appears preferable to be positive rather than negative (Kline, 1983). This also tends to hold with a
preference for ‘Agree’ rather than ‘Disagree’ and ‘Like Me’ rather than ‘Not Like Me’.
(5) Self awareness. A self rating depends on accuracy of knowledge about oneself. This entails a knowledgeable
frame of reference. For one person, the frame of reference may be other neighborhood children who are not so
fluent. When compared to children in another community, apparent fluency may be less. What is competent in one
environment may seem less competent in another. The age, nature and location of the reference group may cause
self assessment not to be strictly comparable across a representative sample of people. A child may also self rate
on surface fluency and not be aware of much less fluency in cognitively demanding language tasks (Skutnabb-
Kangas, 1981).
(6) Point of reference. There is a danger of using monolingual proficiency and performance as the point of
(7) Test aura. Another danger is of raising language measurement to the level of scientific measurement with an
accompanying exaggerated mystique. More ‘natural’ forms of language sampling may be given lower status (e.g.
recording natural conversation) as they rarely carry the mystique of educational and psychological (psychometric)
(8) Narrow sampling of dimensions of language. Language measurement may unwittingly be perceived as
something tangible and concrete (as when measuring height and weight). Rather, language tests mostly contain a
specification of language skills that is hypothetical and debated. Such tests often contain only a small and
unrepresentative sample of the totality of language proficiencies.
(9) Insensitivity to change. It is customary and seen as good practice to produce measurement which is reliable
over time and across occasions (give consistent scores for the same individual over weeks or months). However
the paradox is that such measurement may be insensitive to change within individuals. Test scores need an expiry
date (e.g. not valid after two years).
(10) Labeling. Test scores are apt to create labels for individuals (e.g. someone is seen as having low performance)
which create expectations (e.g. of further underachievement) that may lead to a self fulfilling prophecy.
Language Balance and Dominance Measures
Various tests have been devised to gauge the relative dominance or balance of a bilingual’s two languages. While
such measures have been used in research, they have also been important in US education. Because in many US
states instruction for language minority children was mandated to take place in the child’s dominant language,
some measure of language dominance was needed. This may be through, for example, English language and
Spanish language proficiency tests.
Five examples of psychometric tests are given below:

< previous page page_24 next page >
< previous page page_25 next page >
Page 25
Speed of reaction in a word association task. This seeks to measure whether a bilingual can give an association to
stimulus words more quickly in one language than the other. No particular difference would seem to indicate a
balanced bilingual. An example is presenting a word such as ‘house’, then measuring the time taken to produce an
association (e.g. window). When a person is consistently quicker in giving associations in one language than
another, the likelihood is that one language is dominant. However, dominance is different from competence. A
person may be competent in two or more languages while being dominant in one. Similarly, there could be equal
dominance and a low level of competence in both languages.
Quantity of reactions to a word association task. Bilinguals are measured for the number of associations given
within one minute when a stimulus word (e.g. ‘color’) is presented. An approximately equal number of responses
might indicate a balance between the two languages.
Detection of words using both languages. Words in both languages are to be extracted from a nonsense word such
as DANSONODEND. The letters in the nonsense words must be equally representative of both languages. This is
not easily achievable, and depends on relatively similar alphabets and scripts.
Time taken to read a set of words in the respondent’s two languages.
Amount of mixing the two languages, the borrowing (‘interference’) and switching from one language to another.
A major problem with such balance and dominance tests lies in the representativeness of the measure of language
proficiency and performance. In this respect, such tests would appear to tap only a small part of a much larger and
more complex whole (language ability or language use). The tests cover a small sample of language sub-skills that
might be tested. What can be tested is also only a small sample of a wide definition of language attributes of
individuals. Dominance will vary by domain and across time, being a constantly shifting personal characteristic. It
is possible to be approximately equally proficient in two languages, yet one may be dominant. Speed of processing
may provide evidence about balance but not about dominance in actual language use, in different sociocultural
contexts and over time (Valdés & Figueroa, 1994).
Communicative Language Testing
In attempting to assess a bilingual’s competence in two languages, there is a danger of using a simple paper and
pencil test believing the test will provide a faithful estimation of everyday language life. Multiple choice language
tests, dictation, reading comprehension tests and spelling tests are all well worn paths in the testing of language
skills. Reducing everyday language competence to tests of specific skills is like measuring Michelangelo’s art
solely by its range of colors. A radical alternative is seeing how bilinguals perform in both languages in a range of
real communicative situations. Observing a bilingual in a shop, at home, at work and during leisure activity might
seem the ideal way of measuring bilingual competence. This idea is impractical in terms of time and may be
biased by who is the tester. Such an observation situation is unnatural because of the presence of the tester and
being a ‘test’ situation, intrusive into individual

< previous page page_25 next page >
< previous page page_26 next page >
Page 26
privacy and unrepresentative across time and place. In short, real life observation is likely to be imperfect in terms
of reliability and validity.
In order to collect data that is realistic and representative, we need to know how situations (domains) relate to one
another. We also need to know the sample of language performance that relates adequately to all round language
competence and what examples of test performance relate strongly to language competence. This demands an
overall model or theory of language competence, a subject considered later in the chapter.
A particular emphasis in language testing is on communicative competence (Hart, Lapkin & Swain, 1987). While
tests of spelling, grammar, written comprehension and reading abound, the importance of using languages in
realistic, everyday settings is reflected in current testing movements. The ideal is expressed by Skehan (1988).
‘Genuine communication is interaction-based, with more than one participant; unpredictable and creative, i.e.
genuine communication may take the participants in unforeseen directions; is situated in a context which is
both linguistic/discoursal and also sociocultural; has a purpose, in that participants will be trying to achieve
something by use of language, e.g. to persuade, to deceive, etc.; uses authentic stimulus materials, and
avoids contrived, specially produced materials; is based on real psychological conditions, such as time
pressure; and is outcome evaluated, in that successful performance is judged in terms of whether
communicative purposes have been achieved.’ (p. 215)
A test of language proficiency which meets Skehan’s (1988) criteria is probably impossible to achieve. A test that
truly measures purposeful communication across sufficient contexts without tester effects is improbable. For some,
the answer is simply not to test. For others, a best approximation is accepted. A test may therefore be used that
measures the more limited notion of performance rather than the wider idea of competence.
A test that attempts to approximate the conditions outlined by Skehan (1988) is the oral interview. An example is
the US Foreign Service oral interview which is in four stages (Lowe, 1983; Shohamy, 1983). Following a warm-
up period, there is a check on the level of language proficiency, a deeper probe of that level, finishing with a wind-
down period. The interview takes about half an hour. The two middle stages check that a person can perform
consistently at a level across varying themes and in various language functions. The session is jointly conducted by
two interviewers.
By training, the interviewers avoid narrow, predetermined checklists and attempt to make the interview sensitive to
the candidate. The interviewers judge and score using prescribed criterion. A person is assigned to a level ranging
from 0 (no competence) to 5 (Educated Native Speaker competence). This assumes a distinction between educated
native speakers and other native speakers in their language competence. The levels 1 to 5 also have the possibility
of a ‘+’ rating, thus giving an eleven point rating scale.

< previous page page_26 next page >
< previous page page_27 next page >
Page 27
Such interview procedures may not reflect reality. Does genuine communication take place between strangers, in a
contrived, artificial context? Is the language repertoire of a person truly elicited? Is ‘interview language’
representative of a person’s everyday language functioning? Can we generalize from oral communicative tests
based on a single type of test, given on a single occasion, based on a test interview which is not a typical event in
real life? There are doubts about whether such interview procedures can validly imitate and investigate real
communicative competence. At the same time, they are a compromise between artificial pencil and paper tests and
the impracticality of the detailed observation of individuals.
Criterion Referenced Language Tests:
A Curriculum Approach to Language Testing
A recent shift in testing has been away from norm referenced tests to criterion referenced tests (UNESCO, World
Education Report, 1991). This is partly due to the movement in language education towards communicative skills,
curriculum objectives and mastery learning. What is the distinction between norm referenced and criterion
referenced tests?
Language proficiency tests are usually classified into Norm Referenced and Criterion Referenced tests; the former
usually being summative tests, the latter mostly being formative tests. Standardized norm referenced tests
essentially compare one individual with others, as in an IQ test. A norm referenced test of reading ability, for
example, may enable the teacher to compare one student with a national or regional average (norm). The student
can then be exactly placed in an ordered list (e.g. in the top 16%). A criterion referenced test moves away from
comparing one person with another. Instead it profiles an individual child on a particular language skill. The
profile will test what a child can and cannot do on a precise breakdown of language skills. The parallel is with a
car driving test. There are a variety of components to driving (e.g. backing around a corner, three-point turn,
starting on a steep hill). Proficiency in driving often requires being able to satisfy an examiner on these sub-skills.
Comparisons with other drivers are unimportant. An individual’s mastery of specific tasks is the criterion for
passing or failing.
Specifying driving criteria is easier than identifying language criteria, hence the analogy is not exact. The sub-
components of language proficiency will be contested, and may not be easily definable or measurable. Apart from
language skills, there are the qualitative aspects of language which are not simply reducible for testing (e.g. the
emotive and poetic functions of languages).
One advantage for bilinguals of criterion referenced testing over norm referenced testing is the point of
comparison. Norm referenced testing may compare bilinguals with monolinguals. Various authors (e.g. Grosjean,
1985) regard such comparison as unfair and invalid. Can the sprinter be fairly compared with the decathlete? In
criterion referenced testing the bilingual will be profiled on specific language skills. In theory, unfair comparisons
between bilingual and monolingual may then be avoided. In practice, however, criterion referenced tests can be
used to create comparisons between children, between groups of children and between schools. An advantage of
criterion referenced language tests is that they may facilitate feedback to the teacher that directly

< previous page page_27 next page >
< previous page page_28 next page >
Page 28
Listening Objectives Speaking Objectives
1. Sound discrimination 1. Pronunciation
2. Listening vocabulary 2. Speaking vocabulary
3. General comprehension of
3. Fluency of oral
4. Understanding morphology
of verbs
4. Control of morphology
of verbs
5. Understanding morphology
of prepositions
5. Control of morphology
of prepositions
6. Understanding morphology
of qualifiers
6. Control of morphology
of qualifiers
7. Understanding morphology
of nouns
7. Control of morphology
of nouns
8. Understanding syntax of
8. Control of syntax of
9. Understanding syntax of
9. Control of syntax of
leads to action. This is the notion of formative testing. If the test reveals areas where a child’s language needs
developing, further action can be taken.
One illustration of testing language proficiency by criterion referenced tests is given by Harris (1984). Focusing on
the attainment of Irish listening and speaking objectives, the above table specifies the test items.
The number of items to measure each objective ranged from three (control of syntax of questions) to 25 items
(general comprehension of speech). Fourth grade children were assessed for their mastery of 140 separate items.
Harris’ (1984) language tests are based on linguistic theory. The next example is based on a communicative
approach to language testing.
A broad set of language goals, for which assessment tasks were developed, was part of the National Curriculum in
England and Wales from 1990 to 1995. The basic premise was that teaching and testing should not be separate but
integrated. The National Curriculum specified the curriculum areas on which students will be assessed. Examples
of requirements (language goals) for speaking in Welsh is given below as an illustration of language testing
directly related to the curriculum (Department of Education and Science and The Welsh Office, 1990). An
example indicates the general kind of criterion referenced tasks that could be created to indicate an individual’s
level of attainment. This approach to language minority teaching and learning is explored in detail in Chapter 12.
Such test tasks are important because they tend to accent what a child can do, rather than typical classroom tests
which focus on what cannot be done. Communicative skills, knowledge and understanding are profiled rather than
marks, percentages or grades. Competence is highlighted rather than errors and deficiencies. Such tasks compare
the child with a scheme of language development, rather than comparing the child with other

< previous page page_28 next page >
< previous page page_29 next page >
Page 29
LEVEL ONE (Approximate Age 5)
Example of
a Language
Participate as a speaker and listener in group
activities including imaginative play.
Example of
Play the role of shopkeeper or a customer in a
classroom shop. The teacher observes and records
speaking proficiency with provided guidelines.
LEVEL TWO (Approximate Age 7)
Example of
a Language
Respond appropriately to complex instructions by
the teacher, and give simple instructions.
Example of
Follow three consecutive actions such as: list three
places where a flower will grow best in the
classroom; find out the views of others in the
classroom, and reach a consensus viewpoint.
LEVEL THREE (Approximate Age 9)
Example of
a Language
Give, receive and follow accurately precise
instructions when pursuing a task as an individual or
as a member of a group.
Example of
Plan a wall display or arrange an outing together in a
LEVEL FOUR (Approximate Age 11)
Example of
a Language
Give a detailed oral account of an event or explain
with reasons why a particulars cours of action has
been taken.
Example of
Report orally on a scientific investigation.
children. Such tests are formative, giving direct feedback and ‘feedforward’ to enable curriculum decisions about
individual children.
Criterion referenced language tests should provide direct feedback into (Baker, 1995b):
teaching decisions (e.g. diagnosis of curriculum areas not mastered by an individual student);
reporting to, and discussing achievement with parents;
locating children needing special support and the type of curriculum support they need;
identifying children for accelerated learning;
informing about standards in the class in terms of curriculum development through a subject.

< previous page page_29 next page >
< previous page page_30 next page >
Page 30
The chapter now continues by looking at various language structure theories. Theories about the structure of
language competence provide an integrating consideration of the themes of the definition and measurement of
The Structure of Language Competence
The language theories of the 1960s (e.g. Lado, 1961; Carroll, 1968) tended to center on language skills and
components. The skills comprise listening, speaking, reading and writing and the components of knowledge
comprise grammar, vocabulary, phonology and graphology. These earlier models did not indicate how skills and
knowledge were integrated (Bachman, 1990). For example, how does listening differ from speaking? How does
reading differ from writing? It has also been suggested that such skill and knowledge models tend to ignore the
sociocultural and sociolinguistic context of language (Hymes, 1972a, 1972b). Earlier models fail to probe the
competence of ‘other’ people in a conversation. In a conversation, there is negotiation of meaning between two or
more people. Real communication involves anticipating a listener’s response, understandings and
misunderstandings, sometimes clarifying one’s own language to ensure joint understanding (Paulston, 1992a).
Following the critique of Oller’s (1979) proposal of one underlying competence for language behavior, various
descriptive and empirical models of language competence have been developed. Two major examples will be
briefly outlined; both have implications for the definition and measurement of bilingual proficiency, and both
illustrate relatively comprehensive considerations of language competence.
Canale & Swain’s Model of Language Competence
Canale & Swain (1980) and Canale (1983, 1984) suggest that language competence has four components: a
linguistic component (e.g. syntax and vocabulary); a sociolinguistic component (e.g. use of appropriate language in
different situations); a discourse component (e.g. ability to initiate and participate in sustained conversations and
read sizable written texts), and a strategic component (e.g. improvisation with language when there is difficulty in
Just as Oller’s (1979) ‘one factor’ proposal has been criticized for not recognizing specific language areas, so
Canale & Swain (1980) have been criticized for not showing whether or how the four components are inter-
related, even if they can be subsumed into one overall global factor. Also, the description of a structure for
language competence does not always relate easily to tests. Testing sociolinguistic and strategic competence is, of
essence, likely to be difficult, contested and elusive, although not impossible, as Bachman (1990) reveals.
Bachman’s Model of Language Competence
A second major model of language competence has been proposed by Bachman (1990). Bachman’s model is
valuable in that it considers both language competence and language performance. The model includes not only
grammatical knowledge but also knowledge of how to use language in a particular communicative context.

< previous page page_30 next page >
< previous page page_31 next page >
Page 31
tive Language Ability can be described as consisting of both knowledge, or competence, and the capacity for
implementing, or executing that competence in appropriate, contextualized communicative language use’
(Bachman, 1990: 84).
To fully define, refine and enable the testing of communicative competence, Bachman (1990) has proposed a
detailed model that is summarized in the following table.
Language Competence
(1)Organizational Competence
(i) Grammatical (e.g. Syntax, Vocabulary)
(ii)Textual (e.g. Written and oral cohesion)
(2)Pragmatic Competence
Illocutionary Competence (e.g. speech strategies, language
Sociolinguistic Competence (e.g. sensitivity to register,
dialect, cultural figures of speech)
To explain the table: for Bachman (1990), communicative competence is composed of two major components:
organizational competence and pragmatic competence. Organizational competence is broken down into two parts,
grammatical competence and textual competence. Grammatical competence comprises knowledge of vocabulary,
syntax, morphology and phonology/graphology. For example, a person needs to arrange words in a correct order in
a sentence with appropriate endings (e.g. high, higher, highest).
Textual competence involves ‘the knowledge of the conventions for joining utterances together to form a text,
which is essentially a unit of languagespoken or writtenconsisting of two or more utterances or sentences’
(Bachman, 1990: 88).
Pragmatic competence is composed of two sub-parts: illocutionary competence and sociolinguistic competence.
Following Halliday (1973), Bachman (1990) lists four language functions as part of illocutionary competence:
ideational (the way we convey meanings and experiences), manipulative (using language in an instrumental way to
achieve ends), heuristic (the use of language to discover new things about our world and solving problems), and
the imaginative function (using language beyond the ‘here and now’ (e.g. for humor or fantasy). The second part of
pragmatic competence is sociolinguistic competence. Sociolinguistic competence is sensitivity to the context where
language is used, ensuring that language is appropriate to the person or the situation. This may entail sensitivity to
differences in local geographical dialect, sensitivity to differences in register (e.g. the register of boardroom,
baseball, bar and bedroom). Sociolinguistic competence also refers to sensitivity to speaking in a native-like or
natural way. This will include cultural variations in grammar and vocabulary (e.g. Black English). Another part of
sociolinguistic competence is the ability to interpret cultural references and figures of speech. Sometimes, to
understand a particular conversation, one needs inner cultural understanding of a specific language. A Welsh
figure of speech such as ‘to go round the Orme’ (meaning ‘to be long-winded’) is only fully understandable within
local northern Welsh cultural idioms.

< previous page page_31 next page >
< previous page page_32 next page >
Page 32
In order to represent language as a dynamic process, Bachman (1990) argues that the listed components given in
the table above interact with each other. He therefore adds to the model the notion of strategic competence where
individuals constantly plan, execute and assess their communication strategies and delivery.
Since competence in a language is viewed as an integral part of language performance and not abstracted from it,
tests of language competence cannot just use pencil and paper tests, but also need to investigate the language of
genuine communication. Instead of tests that are artificial and stilted (e.g. language dictation tests), communicative
performance testing involves creative, unpredictable, contextualized conversation.
This suggests that it will be difficult to measure communicative proficiency in an unbiased, comprehensive, valid
and reliable way. Simple classroom tests are likely to be but a partial measure of the bilinguals’ everyday
performance. In testing communicative competence, Bachman (1990) adds a list of features that influence
measured communicative competence. Such external influences are likely to affect the language competence
profile of the person tested, often in a negative way. This list includes:
(1) Test Environment (e.g. effect of the tester; the scheduling and place of the test).
(2) Test Rubric (e.g. lay-out of the test, time allocation, language of the instructions, details as to what constitutes
an appropriate response).
(3) Format of the Test (e.g. written, ‘taped’, ‘live’ testing).
(4) Nature of the Test Language (e.g. chosen topic, context embedded or context reduced communication).
(5) Nature of the Test Response (e.g. multiple choice or self created; written or spoken; concrete or abstract;
allowing non-verbal communication or not).
The topic of measurement often follows from a discussion of definitions, distinctions, dimensions and
categorizations. Just as dimensions and categorizations can never capture the full flavor of the global nature of
bilingualism, so measurement usually fails to capture fully various conceptual dimensions and categorizations. Just
as the statistics of a football or an ice hockey game do not convey the richness of the event, so language tests and
measurements are unlikely to fully represent an idea or theoretical concept. Complex and rich descriptions are the
indispensable partner of measurement and testing. The stark statistics of the football or ice hockey game and the
colorful commentary are complementary, not incompatible.
Language background scales to measure language usage and a plethora of tests to measure language proficiency
exist. The latter includes norm and criterion referenced language tests, self rating scales and language dominance
tests. Particular attention in this chapter has been given to criterion referenced tests measuring mastery of specific
language objectives. Suitable for both first and second languages, sometimes based on theoretical principles,
sometimes eclectic, such tests tend to relate directly to the process of teaching and learning.
The chapter considered theories of the structure of language competence. Such theories join the first two chapters,
integrating issues of definition with ideas about measurement.

< previous page page_32 next page >
< previous page page_33 next page >
Page 33
In particular, the focus has been on linking a linguistic view of language competence with a communicative view.
Language can be decomposed into its linguistic constituents (e.g. grammar, vocabulary). It is also important to
consider language as a means of making relationships and communicating information. This important dualism
will follow us through the book: ability and use; the linguistic and the social; competence and communication.
Suggested Further Reading
BACHMAN, L.F., 1990, Fundamental Considerations in Language Testing. Oxford: Oxford University Press.
BAETENS BEARDSMORE, H., 1986, Bilingualism: Basic Principles. Clevedon: Multilingual Matters.
SKUTNABB-KANGAS, T., 1981, Bilingualism or Not: The Education of Minorities. Clevedon: Multilingual
VALDÉS, G. and FIGUEROA, R.A., 1994, Bilingualism and Testing: A Special Case of Bias. Norwood, NJ:
Review and Study Questions
(1) Write short revision notes on the following:
(i) The different purposes of measuring bilinguals.
(ii) Styles of self rating on language proficiency and its problems.
(iii) The nature of communicative language testing, criterion referenced tests.
(iv) The structure of language competence.
(2) Consider the different reasons why minority and majority languages are ‘measured’? What problems may be
met in such measurement?
(3) Find out more about the distinction between norm referenced and criterion referenced testing. (Also read the
section on assessment in Chapter 13.) What outcomes may derive from these two different forms of testing on
language minority children? Does criterion referenced testing have similar or different potential outcomes to norm
referenced testing for language minority children?
Study Activities
(1) Use the Language Background Scale (modified to suit your context) on yourself or someone you know. This
may also be used with a class of students. Examine the answers given and sum up in words or in numbers the
balance and use of languages. If used with a group of students, are there groups or clusters of students? Are there
those dominant in one language, those dominant in another language and those with a balance between the two?
(2) Using a local school(s), find out what tests are used to measure language achievement in the classroom. These
may be listening, speaking, reading, writing or language development tests. Find out whether these are norm
referenced or criterion referenced tests. For what purposes are these tests being used? How fair are these to
bilingual children?

< previous page page_33 next page >
< previous page page_34 next page >
Page 34
Chapter 3
Languages in Society
The Language Garden: Language Planning
Language Shift and Language Maintenance
Language Decline and Death

< previous page page_34 next page >
< previous page page_35 next page >
Page 35
Bilingual individuals do not exist as separated islands. Rather, people who speak two or more languages usually
exist in groups, in communities and sometimes in regions. People who speak a minority language within a majority
language context may be said to form a speech community or language community. Bilingualism at the individual
level is half the story. The other essential half is to analyze how groups of language speakers behave and change.
Such an examination particularly focuses on the movement and change of a language across decades. A language
minority is rarely stable in its size, strength or safety. It also entails examining the politics and power situation in
which minority languages are situated.
This chapter focuses on the idea that there is no language without a language community. Since language
communities do not usually exist in isolation of other communities, it becomes important to examine the contact
between different language communities. In a world of mass communication, easy travel across continents and the
movement towards a global village, language communities are rarely isolated from other language communities.
The rapid growth of information and inter-continental travel has meant that language communities are rarely if
ever stable. With every minority language and majority language there is constant change and movement. Some
languages become stronger; other languages tending to decline, even die. Some languages thought to be dead, may
occasionally be revived (e.g. Cornish). This chapter therefore seeks to examine language communities, language
contact, language change and language conflict. It will be shown how decisions about bilingual education are a
part of a much wider whole. That is, bilingual education can only be properly understood by examining the
circumstances of language communities in which such education is placed.
This chapter takes a sociolinguistic perspective of language in society, examining central concepts such as
diglossia, language shift, language maintenance, language death and language spread. There is also a linguistic
view of these topics (e.g. change in syntax, semantics and lexicon) that is not covered in this or the following
chapter on language

< previous page page_35 next page >
< previous page page_36 next page >
Page 36
revival (see Aitchison, 1991; McMahon, 1994 and Romaine, 1995 for a linguistic discussion of these topics). The
sociolinguistic perspective begins by examining a central idea in sociolinguistics, that of diglossia.
The term bilingualism is typically used to describe the two languages of an individual. When the focus changes to
two languages in society, the term often used is diglossia (Ferguson, 1959; Fishman, 1972, 1980). While the term
diglossia has in the last twenty years become broadened and more refined, it was originally a Greek word for two
languages. In practice, a language community is unlikely to use both languages for the same purpose. A language
community is more likely to use one language in certain situations and for certain functions, the other language in
different circumstances and for different functions. For example, a language community may use its heritage,
minority language in the home, for religious purposes and in social activity. This language community may use the
majority language at work, in education and when experiencing the mass media.
Ferguson (1959) first described diglossia in terms of two varieties of the same language (dialects). Fishman
(1972,1980) extended the idea of diglossia to two languages existing side by side within a geographical area.
Ferguson’s (1959) original description distinguishes between a high language variety (called H) and a low variety
(called L). This distinction can also be between a majority (H) and minority (L) language within a country, which
is a rather non-neutral and discriminatory distinction. In both situations, different languages or varieties may be
used for different purposes as the table below illustrates.
Context Majority
Language (H)
Language (L)
1. The home and family
2. Schooling
3. Mass media
4. Business and commerce
5. Social & cultural activity in the
6. Correspondence with relations
and friends
7. Correspondence with government
The example shows that languages may be used in different situations, with the low variety more likely to be used
in informal, personal situations; the high or majority language being more used in formal, official communication
contexts. To use the low variety of language in a situation where the high variety is expected, is likely to be
embarrassing, even ridiculed.

< previous page page_36 next page >
< previous page page_37 next page >
Page 37
The table suggests that the different language contexts usually make one language more prestigious than the other.
The majority language may sometimes be perceived as more superior, more elegant and educative a language. The
high variety may be seen as the door to both educational and economic success.
The concept of diglossia can be usefully examined alongside the concept of bilingualism. Bilingualism, argued
Fishman (1972), is the subject for psychologists and linguists. Bilingualism refers to an individual’s ability to use
more than one language. Diglossia, he argued, was a concept for sociologists and sociolinguists to study. Fishman
(1980) combines the terms bilingualism and diglossia to portray four language situations where bilingualism and
diglossia may exist with or without each other. The following table, based on Fishman (1980; see Glyn Williams,
1992, for a critique) portrays this relationship between bilingualism and diglossia.
+ –
1. Diglossia and
Bilingualism together
3. Bilingualism
without Diglossia

2. Diglossia without
4. Neither
Bilingualism nor
The first situation is a language community containing both individual bilingualism and diglossia. In such a
community, almost everyone will be able to use both the high language (or variety) and the low language (or
variety). The high language is used for one set of functions, the low language for a separate set of functions.
Fishman (1972) cites Paraguay as the example. Guaraní and Spanish are spoken by almost all inhabitants. The
former is the Low variety, Spanish is the High language.
The second situation outlined by Fishman (1972, 1980) is diglossia without bilingualism. In such a context there
will be two languages within a particular geographical area. One group of inhabitants will speak one language,
another group a different language. One example is Switzerland where, to a large extent, different language groups
(German, French, Italian, Romansch) are located in different areas. The official status of the different languages
may be theoretically equal. Fluent bilingual speakers of both languages may be the exception rather than the rule
(Andres, 1990).
In some cases, the ruling power group will typically speak the high language, with the larger less powerful group
speaking only the low language. For example, in a colonial situation, English or French was spoken by the ruling
elite, with the indigenous language spoken by the masses.
The third situation is bilingualism without diglossia. In this situation, most people will be bilingual and will not
restrict one language to a specific set of purposes. Either language may be used for almost any function. Fishman
(1972, 1980) regards such

< previous page page_37 next page >
< previous page page_38 next page >
Page 38
communities as unstable and in a state of change. Where bilingualism exists without diglossia, the expectation may
be that one language will, in the future, become more powerful and have more purposes. The other language may
decrease in its functions and decay in status and usage.
The fourth situation is where there is neither bilingualism nor diglossia. One example is where a linguistically
diverse society has been forcibly changed to a relatively monolingual society. In Cuba and the Dominican
Republic, the native languages have been exterminated. A different example would be a small speech community
using its minority language for all functions and insisting on having no relationship with the neighboring majority
Fishman (1980) argues that diglossia with and without bilingualism tends to provide a relatively stable, enduring
language arrangement. Yet such stability may be increasingly rare. With increasing ease of travel and
communication, increased social and vocational mobility, a more global economy and more urbanization, there
tends to be more contact between language communities. As we shall see later in this chapter, language shift tends
to be more typical than language stability. Changes in the fate and fortune of a minority language occur because
the separate purposes of the two languages tend to change across generations. The boundaries that separate one
language from another are never permanent. Neither a minority language community nor the uses that community
makes of its low/minority language can be permanently compartmentalized. Even with the territorial principle (one
language being given official status in one geographical area, the second language being given status within a
separate geographical area), the political and power base of the two languages changes over time. However,
keeping boundaries between the languages and compartmentalizing their use in society may be necessary for the
weaker or lower variety to survive.
Where the territorial principle exists (e.g. in Wales, Switzerland), geography is used to define language
boundaries, with inhabitants of a region classified as a distinct language group. The argument for the survival,
maintenance and spread of the language is based on its historic existence within a defined boundary. As the
indigenous language of the region, language rights may be enshrined in law. Welsh speakers have certain language
rights in Wales (e.g. using Welsh in courts of law) but not when they cross the border into England. The territorial
principle benefits the Welsh but has unfortunate implications for other ‘in-migrant’ language minorities in Britain
(Stubbs, 1991). The danger and discriminatory nature of the territorial principle is revealed in a set of questions. If
Welsh is the language of Wales, is English to be seen as the only rightful language of England? Do languages
belong to regions and territories and not to the speakers of those languages or to groups of those languages
wherever they may be found? Do Panjabi, Urdu, Bengali, Hindi, Greek and Turkish only belong in the home
country? Do such languages have no home in Britain?
Under the territorial principle, should language minorities either speak the language of the territory or return to the
home country? The territorial principle thus has benefits for some (e.g. the Welsh). For others, it is unacceptable
and unfair. In Europe, there are

< previous page page_38 next page >
< previous page page_39 next page >
Page 39
many indigenous (or autochthonous) languages that are seeking preservation status in the European Community.
But what status is accorded the in-migrant languages of Europe (e.g. the various Asian languages such as Panjabi,
Urdu, Bengali, Hindi and Gujerati)?
The term personality principle is particularly helpful to describe minority groups who cannot claim a language
territory principle. The ‘personality’ of each language is the sum of its more or less distinctive attributes (e.g. as
revealed in its uses and functions, customs and rituals, culture and meanings, communication styles and literature;
see Allardt, 1979). One example of the use of the personality principle is the Pennsylvania Amish who decided to
ensure the continuity of their heritage language by reserving an exclusive place for that language at home, and
reserving English for school and for contacts with the outside secular world. One language is reserved for
particular societal functions; another language for distinctly separate functions. This compartmentalization exists in
a relatively stable arrangement. Such separation of identity for each language exists within the psyche of each
group member. There cannot be diglossia without bilingualism if diglossia is based on the personality principle. In
contrast, there can be diglossia without bilingualism where the territorial principle exists.
The personality principle can be an attribute of those claiming territorial principle. However, the personality
principle is an especially supportive concept for in-migrant groups (e.g. Asian languages in Europe and the many
in-migrant language minorities in Canada and the US). Those who cannot claim territorial rights for their language
can assert that their heritage language has a ‘personality’ and is used for particular functions that need safeguarding
and separating from uses of the majority language. Such language groups can argue that their right to use their
minority language is fully portable across national borders and boundaries because it has its own ‘personality’ of
An attempted merging of the territorial and personality principles when applied to language rights has been termed
the ‘asymmetrical principle’ or ‘asymmetrical bilingualism’ (Reid, 1993). As conceived by its Canadian advocates
(e.g. in Quebec), the principle gives full rights to minority language speakers and less rights to the speakers of a
majority language. This is a form of positive discrimination, seeking to discriminate in favor of those who are
usually discriminated against. Some of the functions of a minority language may be legislated for, so as to preserve
that language in a proactive manner. The question of language rights is returned to in Chapter 19.
In language communities, the functions and boundaries of the two languages will both affect and be reflected in
bilingual education policy and practice. In a diglossic situation, is the high or low variety of language used in the
different stages of schooling, from kindergarten to University? If the low variety is used in the school, in which
curriculum areas does it function? Is the low variety just used for oral communication or is biliteracy encouraged
in the school? Are science, technology and computing taught in the high or low variety? Is the low variety just
allowed for a year or two in the elementary school with the higher variety taking over thereafter? Does the school
deliberately exclude the low variety as a medium for classroom learning? The purposes and functions of each

< previous page page_39 next page >
< previous page page_40 next page >
Page 40
language in a diglossic situation are both symbolized and enacted in the school situation. This links with Chapter
10 where different forms of bilingual education are examined.
The Language Garden:
Language Planning
Ofelia García (1992b), uses a powerful analogy to portray language planning. Her Language Garden Analogy
commences with the idea that if we traveled through the countries of the world and found field after field, garden
after garden of the same, one-color flower, how dull and boring our world would be. If a single color flower was
found throughout the world without variety of shape, size or color, how tedious and impoverished the world would
Fortunately, there is a wide variety of flowers throughout the world of all shapes and sizes, all tints and textures, all
hues and shades. A garden full of different colored flowers enhances the beauty of that garden and enriches our
visual and aesthetic experience.
The same argument can be made about the language garden of the world. If there were just one language in the
garden, it would be easy to administer, easy to tend. If one of the majority languages of the world (e.g. English)
solely existed, how dull and uninteresting it would be. Rather, we have a language garden full of variety and color.
For example, in London alone, there are said to be around 184 different languages being spoken. The initial
conclusion is simply that language diversity in the garden of the world makes for a richer, more interesting and
more colorful world.
However, language diversity makes the garden more difficult to tend. In a garden, some flowers and shrubs spread
alarmingly quickly. Some majority languages, particularly English, have expanded considerably during this
century. When the garden is neglected, one species of flower may take over, and small minority flowers may be in
danger of extinction. Therefore some flowers need extra care and protection. This leads to the second part of
Garcà’s (1992b) analogy. A free language economy allowing some flowers to dominate gardens is less preferred
than careful language planning (Rubin, 1977; Eastman, 1983; Cooper, 1989). When a gardener wishes to create a
beautiful garden, there will be both careful planning and continued care and protection. Sometimes radical action
may be taken to preserve and protect. The analogy suggests that language diversity requires planning and care.
Four examples follow:
(1) Adding flowers to the garden. The analogy suggests that, where the majority language is a person’s first
language, it may be enriching to add a second, even a third and fourth language. For example, in Canada, English
speakers learning French may ensure that the colorful diversity of the Canadian multilingual situation is
maintained. In mainland Europe, speakers of French, German, Spanish or Italian, for example, often learn a second
or third language. In much of the United States, Australasia and Britain, a monochrome language garden seems
relatively more common.
(2) Protecting rare flowers. In the diglossic situation of many countries throughout the world, the minority or
indigenous language may be under threat from the quickly

< previous page page_40 next page >
< previous page page_41 next page >
Page 41
spreading majority languages. Just as environmentalists in the twentieth century have awoken to the need to
preserve the variety of flora and fauna, so in the language garden, it is environmentally friendly to protect rare
language flowers. Through education and legislation, through pressure groups and planning, protection of language
species in danger of extinction may be attempted.
(3) Augmenting the numbers of those flowers that are in danger of extinction. Where a language species is in
danger of extinction, stronger action may be needed than protection. Special efforts to revive a threatened language
may be needed through intervention. For example, positive economic discrimination towards the Irish language in
certain defined heartland areas has attempted to preserve the indigenous language in its traditional strongholds.
(4) Controlling flowers that spread quickly and naturally. Flowers that spread rapidly and take the space of other
flowers also need supervision and planning. While majority languages for international communication are an
irreplaceable part of the Information Society, language planning may seek to allow spread without replacing and
killing endangered species.
It is clear from using the language garden analogy that a laissez-faire situation is less desirable than deliberate,
rational language planning. Gardeners are needed (e.g. teachers in schools) to plant, water, fertilize and reseed the
different minority language flowers in the garden to ensure an enriching world language garden. While there are
gardeners (e.g. teachers) tending the language garden, there are landscape engineers who tend to plan and control
the overall shape of the language garden. The view of language landscape engineers (e.g. politicians, policy
makers) is often to regard the language garden as just one part of a wider control of the environment. The
dominant power groups who determine the social, economic and cultural environment may see language as just
one element in an overall landscape design. For example, the type of bilingual education program that is allowed
in a region (submersion, transitional, immersion or supportive of the minority community languagesee Chapter 10)
is but part of a design for the total landscape in which the languages are located.
A language landscape engineer who is concerned only for majority language flowers, will regard protecting rare
flowers to be expensive and unnecessary, and will wish to standardize the variety of language in the country. A
landscape engineer who wishes to protect rare flowers and increase flowers in danger of extinction may encourage
the growth of such flowers alongside majority language flowers within bilingual education. In the US, for
example, many politicians prefer monolingualism rather than bilingualism. The preference is for the assimilation
of minority language communities into a more standardized, monochrome garden. The dominant ruling group in
US society is monolingual, with little perceived need to know or speak the minority languages existing in the
country. Thus their view of bilingual education is determined by their wider ideology.
It is important to note variations in attitude to the language environment. Williams (1991a) sums up differing
‘environmental’ attitudes to the survival and spread of minority languages. First, the evolutionist will tend to follow
Darwin’s idea of the survival of the

< previous page page_41 next page >
< previous page page_42 next page >
Page 42
fittest. Those languages that are strong will survive. The weaker languages will either have to adapt themselves to
their environment, or die. A different way of expressing this is in terms of a free, laissez-faire language economy.
Languages must survive on their own merits without the support of language planning.
However, survival of the fittest is too simplistic a view of evolution. It only accents the negative side of evolution:
killing, exploitation and suppression. A more positive view of evolution is interdependence rather than constant
competition. Cooperation for mutually beneficial outcomes can be as possible as exploitation (Williams, 1991a).
An evolutionist argument about language shift also fails to realize that it is not a simple, spontaneous or impulsive
process. Rather, the fate of languages is often related to the manipulated politics and power bases of different
groups in society. Language shift (in terms of numbers of speakers and uses) occurs through deliberate decisions
that directly or indirectly affect languages and reflects economic, political, cultural, social and technological
change. It is therefore possible to analyze and determine what causes language shift rather than simply believing
language shift occurs by accident. Thus, those who support an evolutionary perspective on languages may be
supporting the spread of majority languages and the replacement of minority languages. Evolutionists who argue
for an economic, cost-benefit approach to languages, with the domination of a few majority languages for
international communication, hold a myopic view of the function of languages. Languages are not purely for
economic communication. They are also concerned with human culture, human heritage, the value of a garden full
of different colored flowers rather than the one variety.
The second approach to languages is that of conservationists (Williams, 1991a). Conservationists will argue for the
maintenance of variety in the language garden. For conservationists, language planning must care for and cherish
minority languages.
Just as certain animal species are now deliberately preserved within particular territorial areas, so conservationists
will argue that threatened languages should receive special status in heartland regions of that language. Catalan in
Spain, native Indian languages in North America, the Celtic languages in Britain and France, have invoked the
conservationist argument. In Ireland, certain areas called Gaeltachta are officially designated for Irish
The third attitude to languages is that of preservationists (Williams, 1991a). Preservationists are different from
conservationists by being more conservative and seeking to maintain the status quo rather than develop the
language. Preservationists are concerned that any change, not just language change, will damage the chances of
survival of their language. Such a group are therefore traditionalists, anti-modern in outlook. Whereas
conservationists may think global and act local, preservationists will tend to think local and act local.
One example of language preservation may be when language is closely tied in with religion. The historical
survival of Pennsylvania German within the Amish community in the US has been a classic illustration of a
preservationist approach to language. The

< previous page page_42 next page >
< previous page page_43 next page >
Page 43
Pennsylvania Germans, sometimes called the Pennsylvania Dutch, came to the US from Germany in the early
eighteenth century. They originally settled in farm communities in southeastern and central Pennsylvania. The
language is a German dialect related to that spoken in the German Palatinate along the river Rhine. Distinctive in
dress, these Protestant Old Order Amish and Old Order Mennonite sectarians have spoken Pennsylvania German (a
German dialect) at home and in the community. English is learnt at school since English is the language of
instruction. English is also spoken with outsidersincreasingly so (Huffines, 1991). Archaic forms of German are
used in Protestant religious worship. The language of the community has thus been preserved within the
established boundaries of that community. However, particularly among the non-sectarian Pennsylvania Germans,
the language is dying (Huffines, 1991). As English replaces the High German used in religious worship, the raison
d’être for the use of Pennsylvania German in the home and community disappears. Religion has preserved the
language. As religious practices change, preservation changes to transformation.
Language Shift and Language Maintenance
No garden of flowers is stable and unchanging. With changes of season and weather comes growth and death,
blossoming and weakening. Minority language communities are similarly in a constant state of change. Such
language shift may be fast or slow, upwards or downwards, but shift is as likely as is garden growth.
Generally, language shift is used in the literature to refer to a downwards language movement. That is, there is a
lessening of the number of speakers of a language, a decreasing saturation of language speakers in the population,
a loss in language proficiency, or a decreasing use of that language in different domains. The last stages of
language shift are called language death. Language maintenance usually refers to relative language stability in its
number and distribution of speakers, its proficient usage in children and adults, and to retaining the use of the
language in specific domains (e.g. home, school, religion). Language spread concerns the increase, numerically,
geographically or functionally in language users, networks and use (Cooper, 1989).
However, there is a danger in the ways these terms are used. First, the terms are ambiguous and may refer to the
numerical size of the language minority, their saturation in a region, their proficiency in the language, or the use of
the language in different domains. Second, these are predominantly sociolinguistic concepts. Linguists have their
own use of these terms (e.g. referring to changes in grammar and vocabulary over timesee Aitchison, 1991;
McMahon, 1994; Romaine, 1995). Third, languages do not lose or acquire speakers. Rather, speakers acquire or
lose languages.
A variety of factors create language shift. For example, out-migration from a region may be vital to secure
employment, a higher salary or promotion. In-migration can be forced (e.g. the capture of slaves) or may be more
of free will (e.g. guest workers). Sometimes there is also forced or voluntary movement of minority language
groups within a particular geographical area. Within a country, marriage may also cause shifting bilingualism. For
example, a bilingual person from a minority language community may

< previous page page_43 next page >
< previous page page_44 next page >
Page 44
marry a majority language monolingual. The result may be majority language monolingual children. Increasing
industrialization and urbanization in the twentieth century has led to increased movement of labor. With the growth
of mass communications, information technology, tourism, road, sea and air links, minority languages seem more
at risk. Bilingual education, or its absence, will also be a factor in the ebb and flow of minority and majority
A relatively comprehensive list of factors that may create language maintenance and shift is given by Conklin &
Lourie (1983) (see also Gaarder, 1977: 141 and following). This list essentially refers to in-migrants rather than
indigenous minorities, but many factors are common to both groups. What is missing from this list is the power
dimension (such as being in subordinate statuse.g. the Puerto Ricans in New York Citysee Zentella, 1988).
A. Political, Social and Demographic Factors
Large number of speakers living
closely together.
Small number of speakers well
Recent and/or continuing in-
migration. Long and stable residence.
Close proximity to the homeland
and ease of travel to homeland. Homeland remote or inaccessible.
Preference to return to homeland
with many actually returning.
Low rate of return to homeland
and/or little intention to return
and/or impossible to return.
Homeland language community
Homeland language community
decaying in vitality.
6. Stability in occupation.
Occupational shift, especially
from rural to urban areas.
Employment available where
home language is spoken daily.
Employment requires use of the
majority language.
Low social and economic
mobility in main occupations.
High social and economic
mobility in main occupations.
Low level of education to
restrict social and economic
mobility, but educated and
articulate community leaders
loyal to their language
High levels of education giving
social and economic mobility.
Potential community leaders are
alienated from their language
community by education.
Ethnic group identity rather than
identity with majority language
community via nativism, racism
and ethnic discrimination.
Ethnic identity is denied to
achieve social and vocational
mobility; this is forced by
nativism, racism and ethnic

< previous page page_44 next page >
< previous page page_45 next page >
Page 45
B. Cultural Factors
Mother-tongue institutions (e.g.
schools, community organizations,
mass media, leisure activities). Lack of mother-tongue institutions.
Cultural and religious ceremonies in
the home language.
Cultural and religious activity in the
majority language.
Ethnic identity strongly tied to home
Ethnic identity defined by factors
other than language.
Nationalistic aspirations as a
language group. Few nationalistic aspirations.
Mother tongue the homeland
national language.
Mother tongue not the only
homeland national language, or
mother tongue spans several nations
Emotional attachment to mother
tongue giving self-identity and
Self-identity derived from factors
other than shared home language.
Emphasis on family ties and
community cohesion.
Low emphasis on family and
community ties. High emphasis on
individual achievement.
Emphasis on education in mother
tongue schools to enhance ethnic
Emphasis on education in mother
Low emphasis on eudcation if in
majority language.
Acceptance of majority language
Culture unlike majority language
Culture and religion similar to that
of the majority language.
This concludes the initial consideration of the factors in language shift. It has been shown that such shift is
particularly related to economic and social change, to politics and power, to the availability of local social
networks of communication between minority language speakers and to the legislative and institutional support
supplied for the conservation of a minority language. While such factors help clarify what affects language shift,
the relative importance of factors is debated and unclear. There are various levels of establishing causes of
language shift, levels such as the political, the economic, the psychological (e.g. at the individual or home level)
and at the sociolinguistic level. A list of the relative importance of these factors is simplistic because the factors
interact and intermingle in a complicated equation. Such a list does not prioritize more or less

< previous page page_45 next page >
< previous page page_46 next page >
Page 46
C. Linguistic Factors
Mother tongue is standardized
and exists in a written form.
Mother tongue is non-standard
and/or not in written form.
Use of an alphabet which
makes printing and literacy
relatively easy.
Use of writing system which is
expensive to reproduce and relatively
difficult to learn.
Home language has
international status.
Home language of little or no
international importance.
Home language literacy used
in community and with
Illiteracy (or aliteracy) in the home
Flexibility in the development
of the home language (e.g.
limited use of new terms from
the majority language).
No tolerance of new terms from
majority language; or too much
tolerance of loan words leading to
mixing and eventual language loss.
(Adapted from Conklin & Lourie, 1983).
important factors in language shift. Nor does it reveal the processes and mechanisms of language shift.
It is difficult to predict which minority languages are in more or less danger of declining, and which languages are
more or less likely to be revived. A frequent, if generalized, scenario for in-migrants is given by García & Diaz
‘Most US immigrant groups have experienced a language shift to English as a consequence of assimilation
into American life. The first generation immigrants sustain their native or first language while learning
English. The second generation, intent upon assimilation into a largely English-speaking community, begin
the shift towards English by using the native language with first generation speakers (parents, grandparents,
others) and English in more formal settings. By slow degrees, English is used in contexts once reserved for
the first language. Encroachment of English into the domain of the first language serves to destabilize the
native language.
Eventually, third generation speakers discontinue the use of the native language entirely. The shift completes
when most of the third generation are monolingual English speakers’ (p. 14).
However, a ‘three generation shift’ is not the only possible pattern. Paulston (1994) cites the Greeks in Pittsburgh
as experiencing a four generation shift. She attributes this slower shift to: use of a standardized, prestigious written
language; access to an institution teaching Greek language and literacy (i.e. Greek churches in Pittsburgh), and

< previous page page_46 next page >
< previous page page_47 next page >
Page 47
marriages with one partner being a monolingual Greek speaker from Greece. In contrast, the three generation shift
among Italians in Pittsburgh is attributed to their speaking a non-standard, non-written dialect of Italian with no
prestige; no religious institutional support as they shared English language Roman Catholic services with, for
example, Irish priests, nuns and laity, and marriage to Roman Catholics with religious compatibility more
important than language compatibility.
Amongst Panjabi, Italian, Gaelic and Welsh communities in Britain, there are occasional ‘fourth generation’
individuals who sometimes wish to revive the language of their ethnic origins. For some, assimilation into the
majority language and culture does not give self-fulfillment. Rather, such revivalists seek a return to their roots by
recovering the language and culture of their ethnic heritage. In Europe, with a pressure towards a European
identity rapidly increasing, language minority members seem increasingly aware of the benefits of a more
distinctive and intimate local identity. The pressure to become part of a larger whole seems to result in a counter-
balancing need to have secure roots within a smaller and more domestic community. A local language is valuable
in this more particular identity. Bilingualism provides the means to be international and local.
Language Decline and Death
Another way of identifying the causes of language shift is to examine a dying language within a particular region
(e.g. Dorian, 1989).
Susan Gal (1979) studied in detail the replacement of Hungarian by German in the town of Oberwart in eastern
Austria. After 400 years of relatively stable Hungarian-German bilingualism, economic, social and family life
became more German language based. Using an anthropological style, Gal (1979) studied the process of language
decline. For Gal (1979) the issue was not the correlates of language shift, but the process. For example, while
industrialization was related to language decline in Oberwart, the crucial question became:
‘By what intervening processes does industrialization, or any other social change, effect changes in the uses
to which speakers put their languages in everyday interactions?’ (p. 3)
Gal (1979) showed how social changes (e.g. through industrialization and urbanization) change social networks,
relationships between people, and patterns in language use in communities. As new environments arise with new
speakers, languages take on new forms, new meanings and create new patterns of social interaction.
Another celebrated study is by Nancy Dorian (1981). Dorian carried out a detailed case study of the decline of
Gaelic in east Sutherland, a region in the north-east Highlands of Scotland. In the history of the region, English
and Gaelic co-existed with English generally being the ruling language and the ‘civilized’ language. Gaelic was
regarded as more of the ‘savage’ language of lower prestige. In this region of east Sutherland, the last two groups
to speak Gaelic were the ‘crofters’ (farmers of a small amount of land) and the fishing community. Dorian (1981)
studied the fishing community who had become a

< previous page page_47 next page >
< previous page page_48 next page >
Page 48
separate and distinct group of people in a small geographical area. Surrounded by English speaking communities,
these fisher-people originally spoke only Gaelic and later became bilingual in English and Gaelic. The fisher-folk
thought of themselves, and were thought of by their neighbors, as of lower social status. They tended to marry
within their own group. When the fishing industry began to decline, the Gaelic speaking fishing-folk began to find
other jobs. The boundaries between the Gaelic speakers and the English speakers began to crumble. Inter-marriage
replaced in-group marriage, and ‘outside’ people migrated to the east Sutherland area. Over time, the community
gave up its fisher identity and the Gaelic language tended to decline with it:
‘Since Gaelic had become one of the behaviors which allowed the labeling of individuals as fishers there was
a tendency to abandon Gaelic along with other ”fisher” behaviors.’ (Dorian, 1981: 67)
Across generations in the twentieth century, Gaelic in east Sutherland declined. Whereas grandparents talked, and
were talked to, only in Gaelic, parents would speak Gaelic to other people but use English with their children and
expect their children to speak English in reply. The children were able to understand Gaelic from hearing their
parents speak it, but were not used to speaking it themselves:
‘The home is the last bastion of a subordinate language in competition with a dominant official language of
wider currency . . . speakers have failed to transmit the language to their children so that no replacement
generation is available when the parent generation dies away.’ (Dorian, 1981: 105)
A different and controversial perspective is given by John Edwards (1985). When languages die, Edwards asks, are
they murdered or do they commit suicide? In the histories of the native Indian languages of Canada and the United
States, and particularly in the histories of the African languages of those who became slaves, there is evidence of
murder. In histories of the Irish, Gaelic and the Welsh language it is typically argued that the language has been
murdered by English and England’s dominant rule over the peripheries of Britain. Are monolingual English
schools part of a language murder machine? It is debated whether the ‘destruction’ of Celtic languages has been
deliberate and conscious, or unconscious and through indifference.
When minority language speakers become bilingual and prefer the majority language, the penalty for the minority
language may be decline, even death. Yet, where people are determined to keep a language alive, it may be
impossible to destroy a language. Language activists, pressure groups, affirmative action and language
conservationists may fight for the survival of the threatened language. In Puerto Rico, the government has
introduced English into schools to attempt bilingualization of the island. Over three-quarters of the population
remain functionally monolingual in Spanish. Resnick (1993) has shown that nationalism, political uncertainty and
the relationship between language and identity have made many Puerto Ricans resistant to language change and
the use of English.

< previous page page_48 next page >
< previous page page_49 next page >
Page 49
For Edwards (1985), language shift often reflects a pragmatic desire for social and vocational mobility, an
improved standard of living. This provides a different slant on the language garden analogy. The answer to the
environmentalist who wishes to preserve a garden of great beauty is that, when the priority is food in the stomach
and clothes upon the back, ‘you can’t eat the view’. Sometimes, there may be a gap between the rhetoric of
language preservation and harsh reality. This is illustrated in a story from Bernard Spolsky (1989b):
‘A Navajo student of mine once put the problem quite starkly: if I have to choose, she said, between living in
a hogan a mile from the nearest water where my son will grow up speaking Navajo or moving to a house in
the city with indoor plumbing where he will speak English with the neighbors, I’ll pick English and a
bathroom!’ (p.451)
However, where there are oppressed language minorities who are forced to live in segregated societies, there is
often little choice of where to live and work. In the quote, the Navajo may have had the choice. In actuality, many
language minorities have little or no real choice. Thus, electing for language suicide is misleading. Attribution of
suicide is a way of ‘blaming the victim’ and diverting a focus on the determination of real causes of language shift.
Freedom of choice is more apparent than real. There is often no viable choice among language minority speakers.
This chapter has focused on languages at the group, social and community level. Two languages within a region is
termed diglossia. Majority and minority languages are frequently in contact, sometimes in conflict. The
relationship between the two languages tends to shift constantly as a consequence of a variety of changeable
cultural, economic, linguistic, social, demographic and political factors.
The argument for preserving language minority communities sometimes relates to existing as an indigenous
language in a defined territory. Also, language minorities may lay claim to the personality principle; their having
unifying ethnic characteristics and identity. One ‘preservationist’ argument uses a gardening analogy. There is a
need to maintain the colorful diversity in the language garden of the world. Language planning by adding and
protecting, controlling and propagating is required to avoid monochrome gardens of majority-only languages.
Conscious language planning to conserve and preserve may be preferred to the narrow view of evolutionists.
Language planning may seek to create language vitality by attending to the economic, social and symbolic status,
geographical distribution and institutional support accorded to a minority language. Not to plan for language
maintenance and spread may be to court language death. Preserving a multicolored language garden requires the
optimism of resurrection rather than the pessimism of decline and death. It is to a positive accent of construction
rather than destruction that we turn in the next chapter.

< previous page page_49 next page >
< previous page page_50 next page >
Page 50
Suggested Further Reading
EDWARDS, J., 1985, Language, Society and Identity. Oxford: Blackwell.
EDWARDS, J., 1994, Multilingualism. London: Routlege.
FISHMAN, J.A., 1989, Language and Ethnicity in Minority Sociolinguistic Perspective. Clevedon: Multilingual
PAULSTON, C.B., 1994, Linguistic Minorities in Multilingual Settings. Amsterdam/Philadelphia: John
PAULSTON, C.B. (ed.) 1988, International Handbook of Bilingualism and Bilingual Education. New York:
SKUTNABB-KANGAS, T., 1981, Bilingualism or Not: The Education of Minorities. Clevedon: Multilingual
Review and Study Questions
(1) Write short revision notes on the following:
(i) The meaning of the following terms: diglossia, the territorial principle, the personality principle, language
shift, language maintenance, language loss, language death.
(ii) The differences between an evolutionist, conservationist and preservationist attitude to languages.
(2) How does diglossia differ from bilingualism? Make a list of the differences.
(3) What is the distinction between the territorial principle and the personality principle? Contrast two societal
contexts known to you (or from your reading) and explain how they are examples of the territorial and/or
personality principle.
(4) What are the major factors in language retention and language loss? What factors are important in language
minority maintenance and spread?
Study Activities
(1) As a group activity or with a partner, make an abbreviated list of the factors of language shift from Conklin &
Lourie (1983). On a 5 point scale (5 = very important, 4 = fairly important, 3 = neither important or unimportant, 2
= fairly unimportant, 1 = unimportant), rate your perception of the strength of these factors in language shift.
(2) Using the list of factors in language shift from Conklin & Lourie (1983), compare the prospects for
maintenance or shift in two different language communities.
(3) Create a poster or a chart using the language garden analogy. This may be a chart to convey a large amount of
information about the language garden. Alternatively it may be a simple poster advertising the importance of
language variety in the world.
(4) Find out how many different languages are spoken in your area (local or regional). List the names of these
languages and in which regions they are spoken as an indigenous language. Through interviews with language
minority members, or through a library search, provide brief example of that language (on paper or on tape).
Are the languages in your list ‘native’ or ‘in-migrant’ languages in your area? Are they spoken by different
generations or more by older generations? Is there preservation and maintenance of the language, or is the
language declining?

< previous page page_50 next page >
< previous page page_51 next page >
Page 51
Chapter 4
Language Revival and Reversal
A Model of Language Shift and Vitality
Status Factors
Demographic Factors
Institutional Support Factors
Bilingualism and the Economy
Bilingualism as an Economic Advantage
Minority Languages
Language Revival and Reversal
Steps in Reversing Language Shift
Limits and Critics

< previous page page_51 next page >
< previous page page_52 next page >
Page 52
The recent history of Hebrew and Welsh, of Catalan and Bahasa Melayu (the indigenous language of Malaysia)
provide a more optimistic lyric than the lament of language decline and death. Just as rain forests and whales have
become the subject of environmental protection, so languages may be protected and promoted to effect revival and
reversal. This chapter focuses on the possibilities of language maintenance, and on optimistic attempts to
encourage the growth of a minority language. To achieve this, major theoretical contributions are considered in
detail. The first contribution seeks to answer the question ‘How is language vitality achieved? Building on one part
of this theory, the chapter examines the importance of economics in the maintenance and revitalization of a
minority language. The next contribution considers ‘How can a language be revived and reversed? We start with
Giles, Bourhis & Taylor’s (1977) consideration of language vitality.
A Model of Language Shift and Vitality
In an attempt to create a model rather than a list of the many factors involved in language vitality, Giles, Bourhis
& Taylor (1977) propose a three factor model: status factors, demographic factors and institutional support factors,
which combine to give more or less minority language vitality. Consideration of this model gives some
explanatory flesh to the framework of Conklin & Lourie (1983), presented in the last chapter.
Status Factors
A key issue in language status is whether the language minority is in the ascendancy (superordinate) or is
subordinate. The survival of Swedish in Finland is partly explained by the superior status of Swedish speakers in
Finland. Language minorities are more often in subordinate status to a language majority (e.g. Asian languages in
Canada and England; Spanish in the US). However, a more refined analysis requires a breakdown into different
types of language status.

< previous page page_52 next page >
< previous page page_53 next page >
Page 53
The economic status of a minority language is likely to be a key element in language vitality. Where, for example,
a minority language community experiences considerable unemployment or widespread low income, the pressure
may be to shift towards the majority language. A minority language may be sacrificed on the altar of economic
progress. A language of the poor and of the peasant is not the language of prosperity and power. Guest workers,
in-migrants and refugees looking for social and vocational mobility may place a high value on education in the
majority language. After a consideration of Giles, Bourhis & Taylor’s (1977) three factor model, the importance of
an economic dimension to a minority language will be explored in detail as it is such a key element in language
vitality and revival.
The social status of a languageits prestige valuewill be closely related to economic status of a language and is also
a powerful factor in language vitality. When a majority language is seen as giving higher social status and more
political power, a shift towards the majority language may occur. Where language minorities exist, there is often a
preponderance of lower socio-economic class families. Where a minority language is seen to co-exist with
unemployment, poverty, social deprivation and few amenities, the social status of the language may be negatively
A language’s symbolic status is also important in language vitality. A heritage language may be an important
symbol of ethnic identity, of roots in ‘the glorious past’. The symbolic importance of a language is exemplified in
the Celtic countries. In Ireland, for example, the Irish language is sometimes regarded as a mark of nationhood, a
symbol of cultural heritage and identity. There tend to exist positive public attitudes to Irish, Scottish Gaelic and
Welsh, but private skepticism. There exists interest in the survival of the minority language, but not in personal
involvement in that language. As a symbol of ethnic history, of heritage and national culture, the Celtic languages
are sometimes valued by the public. As a tool of widespread personal communication, as a medium of mass
education, the languages are less valued. When the personal balance sheet includes employment, educational and
vocational success and interpersonal communication, the credit of positive attitudes towards language as a cultural
and ethnic symbol is diminished by the costs of perceived prior needs and motives. Goodwill towards the language
stops when the personal pay-off is not great.
Is there a paradox beginning to emerge in the relationship between majority and minority languages? Majority
languages such as English have high status as languages of international communication. At the same time,
internationalism (e.g. the increasing Europeanization of the late 1990s) appears to awaken a basic need for
rootedness, for an anchor in a local language and a local cultural community. Becoming European can revive and
reawaken the need to belong to one’s local heritage and historical groups. In becoming part of a larger whole, a
local identity is essential and foundational. The push to become a member of the global village seems to lead to a
strong pull towards primary roots.
Demographic Factors
The second of Giles, Bourhis & Taylor’s (1977) factors concerns the geographical distribution of a language
minority group. One part of this is the territorial principle;

< previous page page_53 next page >
< previous page page_54 next page >
Page 54
two languages having their own rights in different areas within a country. Or, in Ireland, there being designated
heartland areas where some protection and maintenance of the Irish language is encouraged (the Gaeltachta). A
second part of this factor is the number of speakers of a certain language and their saturation within a particular
Research in Wales, using Census data, indicates that saturation of speakers within a particular area is important in
language maintenance (Baker, 1985). For example, in communities where over 70% of people speak Welsh, there
appears to be more chance of the language surviving. Also important in a language maintenance equation is the
demographics of biliteracy. Analysis of the Welsh language Census data shows that where bilingualism exists
without biliteracy, there is an increased likelihood of language decay. When someone can speak a minority
language and not write in that language, the number of functions and uses of that language is diminished.
Bilingualism without biliteracy also means a decrease in the status of that language, and less chances of a
linguistically stable language.
At the same time, it is possible for a small language minority to survive when surrounded by the majority
language. Three examples will illustrate that small numbers of minority language speakers can still provide a lively
language community, even when surrounded by a majority language community. First, in a large city or in border
areas, a small number of minority language speakers may be socially and culturally active in their minority
language. Such speakers may interact regularly and create a strong language cell. The language of children in the
home and street may be important for continuation over time. Second, when some language groups have strong
religious beliefs, they may prefer not to interact with majority language speakers. Such is the case of the Old Order
Amish, Pennsylvania Germans. They continued to speak Pennsylvania German at home and in the community.
Such a minority has historically created strong boundaries in their language usage. Third, when minority language
speakers can travel easily between the homeland and their current area of residence, the minority language may be
invigorated and strengthened (e.g. Puerto Ricans in New York and Mexicans in Texas and California).
The idea of demographic factors relates to mixed, inter-language marriages. In such marriages, the higher status
language will usually have the best chance of survival as the home language. With inter-language marriages
specifically, and with language minority communities in general, there is likely to be movement across
generations. For example, in-migrants may lose their heritage language by the second or third or fourth generation.
This highlights the key importance of languages in the home as a major direct cause in the decline, revival or
maintenance of a minority language. Language reproduction in the young is an important part of language vitality.
As a generalization, a minority language is more likely to be preserved in a rural than an urban area. Once
migration of rural people to urban areas occurs, there is an increased chance of the minority language losing its
work function. In the office and in the factory, the dominant language is likely to be the majority language with a
minority language depreciated. In rural areas, the language of work and of cultural activity is relatively more

< previous page page_54 next page >
< previous page page_55 next page >
Page 55
likely to be the historical language of that area. The language of the farm or of the fishing boat, of religion and of
rustic culture is more likely to be the minority language. There will be important exceptions (e.g. where an urban
ethnic group generates its own industries, or is sited in a particular part of a city, or congregates regularly for
religious purposes).
Institutional Support Factors
Language vitality is affected by the extent and nature of a minority language’s use in a wide variety of institutions
in a region. Such institutions will include national, regional and local government, religious and cultural
organizations, mass media, commerce and industry, and not least education. The absence or presence of a minority
language in the mass media (television, radio, newspapers, magazines, tapes and computer software) at the very
least affects the prestige of a language. The use of a minority language in books and magazines, for example, is
also important for biliteracy. Language may be given the status of a modern, twentieth century language when it is
used in the media. The perceived quality of television programs (compared to majority language programs) will be
important (Baetens Beardsmore & van Beeck, 1984). However, it is possible to exaggerate the importance of
television and radio in its effect on the communicative use of a minority language. Television and radio provide
only a passive medium for language. Research from Wales (see Baker, 1985,1992) suggests that it is majority
language mass media that is the destroyer of a minority language and culture, rather than minority language
television and radio being the salvation of the language. The glossy, high quality of TV English language
programs provides fierce competition for minority languages. Media attractiveness has resulted in the invasion of
Hollywood productions in countries where English is not prevalent. Active participation in a minority language is
required for language survival; an event that mass media by its receptive nature does not provide.
Religion can be a strong and important vehicle for the maintenance of a majority and a minority language. The use
of classical Arabic in Islam, Hebrew in Judaism, and German among the Protestant Old Order Amish in
Pennsylvania each illustrates that religion has been a preserver of language. It is said in Wales that the language
would not have survived into the twentieth century had it not been for its dominant position in Welsh chapels and
in Welsh religious life inside the home (e.g. the family reading the Bible in Welsh). Religion may help standardize
a language. Through its holy books, tracts and pamphlets, traveling missionaries and teachers, a relatively standard
form of a language may evolve.
Providing administrative services in a minority language also serves to give status to that language. It also
increases the usefulness of that minority language for communication. The use of languages within educational
institutions is probably an essential but not sufficient condition for language maintenance. Where schooling in a
minority language does not exist, the chances of the long-term survival of that language in a modern society may
be severely diminished. Where the minority language is used in the school situation, the chances of survival are
greatly increased but not guaranteed. Nancy Hornberger’s (1988) acclaimed anthropological study of language
planning, language shift and bilingual education in the Quechua speaking highlands of southern Peru suggested

< previous page page_55 next page >
< previous page page_56 next page >
Page 56
‘Schools cannot be agents for language maintenance if their communities, for whatever reason, do not want
them to be.’ (p. 229)
Community support for bilingual education in the minority language and culture outside school are important.
Education by itself cannot enable a minority language to survive. Other supports are essential, particularly
regeneration of languages within the family and an economic basis for a language. The importance of the family in
language maintenance is considered in the next chapter. After a brief critical comment on the concept of
ethnolinguistic vitality, this chapter continues by a deeper consideration of the major importance of the economy in
language reversal and maintenance.
Giles, Bourhis & Taylor’s (1977) theory of language vitality has been criticized by Husband & Khan (1982). First,
they suggest that the dimensions and factors are not separate and independent of each other. In reality, these
elements interact, inter-relate and are often mutually dependent on each other. While presented as comprehensive
and wide-ranging, a critique that the factors are ‘something of an untheorised pot pourri’ (p. 195), relates to a
consensus rather than a conflict perspective (see Chapter 19), with no means of ‘sifting and weighing the many
variables that are listed’ (p. 195). These authors doubt whether the listing of variables that constitute language
vitality is any greater than the sum of its parts or creates an integrated whole.
Second, different contexts have widely differing processes and recipes that lead to more or less language vitality.
The historical, social, economic, cultural and political processes that operate in any language community vary
considerably. We need a historical and sociological perspective, for example, to help explain patterns of interaction
that relate to language vitality. Change and variability across time is seemingly not catered for in this theory.
Third, the theory is not easy to operationalize in research. Many of the factors may not be easily or unambiguously
measured. Thus the theory is difficult to test.
This completes the consideration of Giles, Bourhis & Taylor’s (1977) theory of language vitality. When the
economic status of a language was introduced earlier in this chapter, it was suggested that economics is a key
element in language loss, shift, vitality and revival. Therefore, this chapter now proceeds to give a more detailed
examination of the importance of the economic dimension to language shift and vitality. However, it needs
remembering that (1) the economic dimension is only one dimension in the status of a language there are social
and symbolic status factors as well; (2) the economic dimension interacts with other vitality factors, and is not a
separate entity.
Bilingualism and the Economy
There is sometimes a danger that writing and speeches about minority languages become romantic and cloistered.
The love of a language and the need to preserve all that is valuable from the past is important. But life is often
about finding a job, getting enough money to buy food, find shelter and own a TV, avoiding unemployment and
poverty. For others, life is much about increasing salaries, affording a bigger house in a more pleasant

< previous page page_56 next page >
< previous page page_57 next page >
Page 57
neighborhood, affording vacations and cars, extras for the children and economic advantages over others.
Bilingual education is devalued unless it connects with the economy. Students want jobs, money and increasing
affluence after leaving school. When a school fosters a form of bilingualism that connects with a local, national or
international economy, the higher the value of that bilingual education. Thus, one premise of this section is that the
future of language minorities is centrally related to individual, home and community economies. Because of this
central relationship, the section will widen the discussion beyond language vitality.
This section explores a paradox among bilinguals. Among language minority groups, there is often unemployment,
poverty and low wages. Yet bilinguals have linguistic capital. Bilinguals often have marketable language skills and
intercultural knowledge. Bilinguals are often economically impoverished yet linguistically accomplished.
Bilingualism as an Economic Advantage
In an increasingly bilingual and multilingual world, with trade barriers being broken, with single markets in areas
such as Europe growing, and with economic competition rapidly developing on a global scale, competence in
languages is increasingly important. Those who have multilinguistic capital may be in a position to increase their
economic capital.
This immediately raises the question of which languages may be useful for economic advancement? In many
countries of the world, it is English as a second or foreign language that has visible economic value. As Coulmas
(1992) suggests,
‘no Japanese businessman ever tries to operate on the American market without a sufficient command of
English, whereas the reverse case, of American business people who expect to be able to do business in
Japan without being proficient in Japanese, is not at all rare. On the one hand, this is a reflection of the
arrogance of power, but on the other hand, it testifies to the fact that the opportunities for realizing the
functional potential of English on the Japanese market are far better than those of realizing the functional
potential of Japanese on the American market.’ (pp. 66 & 67)
Nevertheless Japanese has increasingly become a desirable modern foreign language, as people realize that to sell
to the Japanese requires a knowledge of the Japanese language and culture. As Helmut Schmidt, former chancellor
of the old Federal Republic of Germany once said: ‘If you wish to buy fromus, you can talk any language you like,
for we shall try to understand you. If you want to sell to us, then you must speak our language’.
Alongside the English language, the Japanese, German, French, Spanish and Portuguese languages have
historically been regarded as important trading languages. However, in the future, this list of modern languages for
marketing and trading purposes is likely to grow significantly. For example, Bahasa Melayu, Mandarin and
Cantonese, Panjabi and Hindi, Arabic and Korean may each become increasingly valuable. This

< previous page page_57 next page >
< previous page page_58 next page >
Page 58
reflects the important half-truth of Coulmas (1992) who suggests that ‘in spite of the non-economic values attached
to language, what prevails in matters of language is often that which is profitable’ (p. 152). The other half of the
half-truth is that language is also related to less tangible, measurable and affective characteristics such as the
social, cultural and particularly religious value in some language communities for a particular language. The half-
truth also hides the tension that can often exist between economic development and cultural reproduction. The
economic value of a language may at times clash with preserving that language to reflect heritage, home values
and historical traditions. Nevertheless, the economic value of language relates strongly to the prestige of that
language and hence its chances of maintenance and reproduction.
If languages are increasingly important in international trading, are they equally important in small and medium-
sized businesses as in multinational corporations? In large businesses, as García & Otheguy (1994) reveal, we
should be cautious in assuming that a modern foreign language is needed in top executives and top managers. In
multinational corporations, for example, the tendency may not be to send executives with modern language
qualifications abroad. Rather there will be local nationals within a country, working for that corporation, who are
bilingual in both the regional or national language and an international language such as English. United State
corporations who run their operations abroad, and multi-national organizations, may prefer locally based
executives because of their native-like fluency in the national languages, their thorough understanding of the
cultures of that region, and their ability to communicate with customers and the work force as well as with
international colleagues. Local nationals are often paid a lower salary than executives sent from, for example, the
United States.
This indicates that large corporations and multinational companies are increasingly aware that international
business requires the use of regional and national languages. The imposition, for example of the English language,
will not help to sell products abroad. To sell in Germany, Latin America, Japan, the Pacific Basin increasingly
requires business people whose language and culture is congruent with that of the buyer. A congruent language is
valuable for communication. Sometimes, there may be no other way to conduct trade (e.g. if the buyer does not
speak English). Speaking the buyer’s language also signals favorable attitudes to the buyer’s ethnicity and culture.
A competent understanding of the buyer’s culture also signals respect for the buyer and their way of doing
García & Otheguy (1994) found in their research, smaller and medium sized businesses often require executives
and managers who are capable of speaking in languages other than English or another majority language. To have
the edge over competitors, the language of business is important, and where it is congruent with the language of
the area for business, a competitive edge is added. Those who insist that English should be the international
medium of business may be selling themselves short, economically, socially and personally.
This reflects a paradox. While English is often accepted as the international language in multinational companies
and international trade, the massive spread of English during

< previous page page_58 next page >
< previous page page_59 next page >
Page 59
this century has led to a reaction, rather like adjusting a balance, with national and regional groups wishing to
assert their local identity through a retention of their heritage language and culture. This has occurred in France
and England, for example, and in many other European countries. An increased accent on national and regional
languages in the face of international trade, has led to buying from those who are willing to speak the language
and culture of the region rather than imposing their own language on the buyer. Nevertheless, the competitive edge
has often more to do with prices, lowest quotations, and the lowest bid rather than language preference.
Minority Languages
In suggesting that bilingualism has economic advantages, the only languages to be highlighted so far in this section
have been majority languages. English, German, French and Arabic, for example, are relatively prestigious
majority languages. What is the place of minority languages in the economy? Will bilinguals from language
minorities have no economic advantage, no valuable trading language in their minority language, no chance due to
their bilingualism of getting out of the poverty trap that many experience?
In Europe, for example, the move from regional and state economies to a single European market policy,
interlocking business structures in different European countries, encouraging mobility in businesses across
European countries, may leave a distinction between core and periphery, between those in important urban
business areas and those in rural peripheries (e.g. rural areas in Ireland, Wales and Scotland). Since many
indigenous minority languages in Europe are found in regions that are relatively sparsely populated, economically
under-developed, with poorer rural road and transport systems, there is a danger that there will growing inequality
between core and periphery.
In the economic restructuring that has occurred in the last 60 years, increased competition has led to the need for
more efficiency to maintain profit. Industries and services have frequently had to ‘automate, emigrate or evaporate’.
Emigration of industries has been to countries such as Taiwan, Mexico, Brazil and Singapore, where wages, and
therefore production costs, are relatively low. Such outside investment may sometimes offer work and wages to
language minority members. Such outside investment also has negative consequences for language minorities.
Such consequences now need to be discussed.
One negative consequence is that the investment may not reach a language minority. For example, where such
minorities live in rural areas (e.g. Irish and Welsh speakers), economic growth may be in the urban ‘core’ rather
than the rural periphery. Alternatively, the higher grade jobs may be in relatively affluent city areas, and the lower
grade, poorly paid work in the more remote areas.
A different scenario is when a peripheral area attracts inward investment (e.g. factories are located in remoter
areas). The tendency is for the local language minorities to provide relatively cheap workers, while the better paid
(language majority) managers either to operate from their far-away city headquarters or move into the periphery

< previous page page_59 next page >
< previous page page_60 next page >
Page 60
community. In both cases (given below), the managers may have a negative effect on language maintenance:
(1) By working from city headquarters, there is a geographical separation of majority language manager and
minority language worker that represents a status, and a social, cultural, economic and power division. Such a
division has prestige consequences for the majority and minority language. Each language is identified with greater
or lesser affluence, higher and lower status, more or less power.
(2) By living in the language minority community, a manager who does not learn the local language evokes a class
distinction. The manager speaks the majority language; the workers the minority language. One is a higher socio-
economic class; the other a lower socio-economic class. There is a social class division, and a separation or
fracture within the social class structure of the community (Morris, 1992). Social tensions may result that lead to
divisions along both social class and language dimensions. As Morris (1992) found in her research, one solution is
language minority managers who are able to operate across social classes inside their language group. Being
bilingual, such managers can also operate (e.g. for export purposes) in language majority and core regions.
For language minorities, it seems essential that, through entrepreneurial activity, enterprise and risk-taking, they
develop networked small and medium sized businesses. Such businesses, often located in rural areas, sometimes
producing goods for export, other times using ease of communications (e.g. via computers) to deliver services, are
essential for local languages to be preserved. Language minority businesses need to be networked where possible,
within a region and increasingly internationally.
The absence of community based, ethnically based businesses, increases the risk of the emigration of more able,
more skilled and more entrepreneurial people away from the area, hence leaving the language itself in peril. A
language community without economic activity is in danger of starving the language of one essential support
mechanism. An economically wealthy language has a higher probability of being a healthy language. An
economically impoverished language places that language at risk.
While language communities are not necessarily internally integrated, (having their own inner power and status
struggles, differences and divisions), such language communities need to generate their own local economies and
to able to market their products and services beyond the region. In Gwynedd, North Wales, for example, there are
two examples of economic language planning. First, new businesses located in language minority communities are
initially subsidized, professionally advised and developed. Second, many jobs now demand bilingual applicants.
The ability to use Welsh and English is regarded as essential for employment in teaching, many administrative
jobs, local authority services, secretarial and clerical posts. Thus the Welsh language is given a vital economic
Some minority language businesses (e.g. local radio and television stations) create language-related economic
activity that, in itself, spawns secondary economic activity to

< previous page page_60 next page >
< previous page page_61 next page >
Page 61
support such businesses (e.g. catering services, translation services, music and drama activities).
In many countries of the world, there are examples of language communities who are immigrant rather than
indigenous who, through fast food (e.g. the Chinese restaurants) and the clothes trade (e.g. various Asian groups)
maintain their language via ethnic industries. The language is supported by businesses where factory workers, shop
workers and managers work partly or mainly through their heritage language.
This raises an issue about those language minorities who are often located in inner city, urban, in-migrant
communities. As already expressed at the start of this section, such families and language communities are often
surrounded by poverty, deprivation, unemployment and inequality of opportunity to acquire wealth. García &
Otheguy (1994) provide an analysis of the economic possibilities of such urban language communities, particularly
in the United States. The ability to speak a minority language can be crucial in gaining access to such an ethnic
business. If everyone else in the organization speaks Spanish or Cantonese, to gain employment requires
knowledge of that language.
A minority language may also be valuable at the customer interface. García & Otheguy (1994) relate an example:
‘As the burgers and fries popped over the counter at the usual speed on the line with the Spanish-speaking
employee, the line with the English-speaking employee was moving slowly, creating inefficiencies caused
by her difficulty in communicating with the Spanish-speaking customers.’ (p. 110)
The problem with many ethnic businesses is the level of status, salary and conditions that such businesses provide
for their employees. Often the jobs available are of a lower status with a relatively lower salary, and with
advancement not always easy or possible. García & Otheguy (1994) found that in United States businesses, the
demand for bilingual workers was not usually at the top executive levels. Rather, the need for bilinguals was for
the lower and middle ranges of occupation. For example, US businesses required bilingual secretaries, clerks, shop
floor assistants, rather than bilingual managers and executives. Waldman’s (1994) US research found that the five
most needed language abilities required were: telephone; customer contact or relations; translation and
interpretation; word processing and correspondence. Spanish was the language most required by companies,
followed by German, French and Japanese.
However, smaller companies indicated that language skills are of value in their higher positions. In small
companies, the language of the whole workforce may be the minority language. Thus cohesion and an integrated
purpose is attained when managers speak the same language as those whom they manage.
With the Spanish speaking population of the United States rapidly growing, and with Spanish an important trading
language in Latin America and elsewhere, there is a possibility that Spanish may be of increasing economic value
inside the United States and for marketing abroad. Therefore, it seems economically valuable in the next century to

< previous page page_61 next page >
< previous page page_62 next page >
Page 62
be bilingual. For some individuals, this is to gain employment and try to avoid poverty. For others, bilingualism
may be of value to work locally for international and multinational corporations. For yet others, who wish to travel
abroad to do their trade, languages also become increasingly important. The concept that speaking English is all
one needs, whether in Europe or the United States, seems to only contain a part of the truth.
There is no guarantee that those who become linguistically assimilated (e.g. speak English only) in countries like
the United States will gain employment. The evidence from García & Otheguy (1994) and Morales & Bonilla
(1993) confirms that the ability to speak English does not give equal access to jobs and wealth. Linguistic
assimilation does not mean incorporation into the economic structure of the country. If there is a growth of ethnic
businesses (e.g. in urban areas), and a development of language minority businesses in peripheral, rural areas, then
bilingualism rather than English monolingualism may become more economically valuable.
As important as is the economy of a language minority, this is only one factor in language maintenance and
language revival. Therefore, this chapter moves on to examine a more global consideration of the factors and
processes of language revival. Before examining the important contribution of Joshua Fishman, a short
consideration of terminology is required.
Paulston, Chen & Connerty (1993) offer a non-exclusive distinction between language revival, language
revitalization and language reversal which can be subsumed under the term ‘language regenesis’. Language revival
is reserved for the giving of new life to a dead language. An example is the revival of Hebrew bringing a dead
language alive. Maori and Welsh would not be included as there has been a continuous use of the language over
many centuries. Language revitalization is the imparting of new vigor to an existing language, often by increasing
the domains where the language is used (e.g. Finnish and Welsh). Language reversal refers to the turning around
of current (downward) trends in a language. For Paulston, Chen & Connerty (1993), language reversal often
focuses on a more prominent use of a language vis-à-vis its relation to other state language(s). This may be in legal
terms (e.g. the gaining of official status for Catalan in 1978); a renaissance in use (e.g. Maori); or non-native
language being accepted after a period of rejection for economic and international communication purposes (e.g.
English in Singapore and Malaysia).
Language Revival and Reversal
A major contribution to the theory of attempting the reversal of language shift is by Joshua Fishman (1991, 1993).
Fishman (1991) notes a changing perspective in the topic of language shift. The premise has been that minority
languages, like patients in a hospital or doctor’s surgery, will ultimately die. Therefore all one can do is to
understand the causes of death and illness, and attempt to overcome those causes for as long as possible. Instead,
Fishman (1991) argues that language shift needs to take the jump of modern medicine by attempting ‘not only to
combat illness, but to cultivate ”wellness”‘ (p. xii).

< previous page page_62 next page >
< previous page page_63 next page >
Page 63
Fishman (1991) seeks to answer the question ‘what are the priorities in planning language shift?’ For example,
what is the point of pouring money into minority language mass media and bilingual bureaucracy when home,
family, neighborhood and face-to-face community use of the minority language is lacking? It is like blowing air
into a punctured balloon. Blowing minority language air in through the mass media and legislation doesn’t make a
usable balloon, because of the unmended hole.
Fishman (1991) provides a list of priorities to halt language decline and attempt to reverse language shift. This
plan also shows why many efforts to reverse minority language situations have often resulted in failure rather than
success. Before we examine this plan, Fishman (1991) provides a basic philosophy and set of assumptions that are
required before establishing priorities in reversing language shift.
Assumptions of Reversing Language Shift
First, when a society or community is losing its language and culture, it is likely to feel pain. This hurt may be
symptomatic of the social injustice towards that community. Although it may not be a cancer but more like a
toothache, to those who experience it, there is real rather than imaginary suffering. Such suffering needs
remedying. Second, the basis of reversing language shift is that a more global village, a world more unified by
mass communication and speedy travel, a more integrated eastern and western Europe, does not bury the need for
local language and local culture. Indeed, a more centrally organized and uniform world may increase rather than
decrease the need for language and cultural identity at a local level. Having local cultural and linguistic roots may
be a necessary precondition before integrating into a global village. Third, and most importantly for Fishman
(1991), the political basis of the plan is to support cultural pluralism and cultural self-determination. The
destruction of minority languages is the destruction of intimacy, family and community, often involving oppression
of the weak by the strong, subjugating the unique and traditional by the uniform and central. Thus, Fishman (1991)
argues for ‘greater sociocultural self-sufficiency, self-help, self-regulation and initiative’ (p. 4) among linguistic
Fourth, an ethnic or culture group that has lost its language is different from that group with their minority
language. Fishman (1991) cites the case of Jews who, not speaking Hebrew, tend to have a different daily life-
pattern, a different kind of sub-culture. Language shift accompanies cultural change. This suggests that language
shift is not just about language; it is about the attendant culture as well. The argument for language restoration and
resurrection must therefore involve a call for cultural change and greater cultural self-determination. Fishman
(1991) warns of the danger of language as the sole focus for shift. The danger is that minority language activists
‘become so accustomed to speaking only to each other that they forget how to speak effectively to others.
What activists no longer need to explain to each other they often no longer know how to explain to others.
They often need to remind themselves and to make themselves more conscious of their own RLS [Reversing
Language Shift] ‘basic principles’, so that they can then make others more conscious of these principles as
well.’ (pp. 18/19)

< previous page page_63 next page >
< previous page page_64 next page >
Page 64
A different warning is given to those who believe that reversing language shift is purely about the accumulation of
power and money (e.g. as has been said about the use of Hebrew and Welsh). Believing that language minorities
who are attempting language reversal and resurrection are concerned with achieving power and increasing wealth,
is simplistic and misguided. Fishman (1991) argues that human values, feelings, loyalties and basic life-
philosophies are present in the complex reasons for language change. Language activists often have ideals,
commitments, even altruism that makes their motives more than just power and money. Minority languages and
cultures, in their desire for a healthy existence, may be sometimes irrational or super-rational. This is similar to
religion, love, art and music where there are personal elements that transcend conscious rationality, and transcend
self-interest in power and money.
Fifth, to help understand language shift, Fishman (1991) clarifies the relationship between language and culture in
terms of three links:
(1) A language indexes its culture. A language and its attendant culture will have grown up together over a long
period of history, and be in harmony with each other. Thus the language that has grown up round a culture best
expresses that culture. Its vocabulary, idioms, metaphors are the ones that best explain at a cognitive and emotive
level that culture.
(2) A language symbolizes its culture. To speak German in the US, during World War 1 and France and Britain
during World War 2 was not appropriate nor acceptable. Not that the allies were at war with the German language.
Rather, the German language symbolized the enemy. Therefore, that language was inappropriate in allied
countries. A language tends to symbolize the status of that language. For example, to speak English in Kuwait
following the victory against Saddam Hussein of Iraq was to be symbolically associated with status, power and
victory. Speaking English often symbolizes money and modernity, affluence and achievement. English may also
symbolize colonial subjugation. A language that is apparently dying may symbolize low status and low income. In
certain parts of Ireland and Wales, the indigenous language is sometimes perceived as a symbol of the past rather
than the present, of disadvantage rather than advantage.
(3) Culture is partly created from its language. Much of a culture is enacted and transmitted verbally. The songs,
hymns, prayers of a culture, its folk tales and shrewd sayings, its appropriate forms of greeting and leaving, its
history, wisdom and ideals are all wrapped up in its language. The taste and flavor of a culture is given through its
language; its memories and traditions are stored in its language. An example is a saying or a figure of speech in a
minority language that requires a long explanation in another language. Even then that pithy saying may sacrifice
some of its meaning and feeling in translation. At the same time, culture is derived from many more sources than
language. For example, there are many different cultures which all use the Spanish language.
Sixth, Fishman (1991) makes an argument for language planning. Just as there is economic planning, educational
planning and family planning, so there can and should

< previous page page_64 next page >
< previous page page_65 next page >
Page 65
be language planning. Such planning has as its base ‘re-establishing local options, local control, local hope, and
local meaning to life. It reveals a humanistic and positive outlook vis-à-vis intergroup life, rather than a
mechanistic and fatalistic one. It espouses the right and the ability of small cultures to live and to inform life for
their own members as well as to contribute thereby to the enrichment of human kind as a whole’ (p. 35).
Language planning, also called language engineering, refers to ‘deliberate efforts to influence the behavior of
others with respect to the acquisition, structure, or functional allocation of their language codes’ (Cooper, 1989:
45). Such language planning involves status planning (raising the status of a language within society), corpus
planning (concerning the vocabulary, spelling, grammar and standardization of the minority language) and
acquisition planning (creating language spread by increasing the number of speakers and uses (via empowerment)
by, for example, language teaching).
Cooper (1989: 98) provides a scheme for understanding language planning by asking a series of key questions:
Which actors (e.g. elites, influential people, counter-elites, non-
elite policy implementers)
attempt to influence which behaviors (e.g. the purposes or
functions for which the language is to be used)
of which people (e.g. of which individuals or organizations)
for what ends (e.g. overt (language-related behaviors) or latent
(non-language- related behaviors, the satisfaction of interests)
under what conditions (e.g. political, economic, social,
demographic, ecological, cultural)
by what means (e.g. authority, force, promotion, persuasion)
through what decision-making processes and means
with what effect or outcome?
This relatively comprehensive set of questions makes some important points. First, language planning may be
generated by different groups. For example, poets, linguists, lexicographers, missionaries, soldiers as well as
administrators, legislators and politicians may become involved in language planning. Cooper (1989) argues,
however, that language planning is more likely to succeed when it is embraced or promoted by elite groups or
counter-elites. Such elites tend to work primarily from their own self-interests. Language planning is motivated by
efforts to secure or reinforce the interests of particular people. However, language planning may positively affect
the masses by adding to their self identity, self-esteem, social connectedness and links with their heritage and
A consideration of language planning runs the danger of over-emphasizing its importance in overall political
planning. First, there is often piecemeal pragmatism rather than planning. The revival of Hebrew is often quoted as
one triumphant and successful example of language planning. Yet the rapid advance of Hebrew in Israel appears to
have occurred

< previous page page_65 next page >
< previous page page_66 next page >
Page 66
by improvisation and diverse ventures rather than by carefully structured, systematic and sequenced language
planning. Second, as Heath (1986) has shown, political and economic decisions usually govern language decisions.
Language decisions are subsidiary and minor concerns of those in power, whose pervading interests are more
frequently about power and purse. Language is usually an outcome from other decisions rather than a determinant
of social, political or economic policies. Yet, as the European Commission has increasingly stressed, language
minority issues and the need for bilingualism and multilingualism in a future Europe are relevant to European
cultural and economic development. Bilingualism is one part of an interconnected whole in Europeanization.
Language status and corpus planning require full consideration of bilingualism and not just planning for the
minority language. Where minority languages exist, there is usually the need to be bilingual if not multilingual.
Minority language monolingualism is usually impracticable. To create blockades and barricades between
languages is almost impossible in the twentieth century. Cross-cultural communication is the reality; a bunker
approach to a minority language and culture may be tantamount to language death. This returns us to the concept
of diglossia, highlighting the importance of different functions for the minority and majority language. Where
different languages have different functions, then an additive rather than a subtractive bilingual situation may exist.
An additive bilingual situation is where the addition of a second language and culture is unlikely to replace or
displace the first language and culture (Lambert, 1980). For example, English-speaking North Americans who
learn a second language (e.g. French, Spanish) will not lose their English but gain another language and parts of its
attendant culture. The ‘value added’ benefits may not only be linguistic and cultural, but social and economic as
well. Positive attitudes to diglossia and bilingualism may also result. In contrast, the learning of a majority second
language may undermine a person’s minority first language and culture, thus creating a subtractive situation. For
example, an in-migrant may find pressure to use the dominant language and feel embarrassment in using the home
When the second language is prestigious and powerful, exclusively used in education and in the jobs market, and
when the minority language is perceived as of low status and value, stable diglossia and bilingualism may be
threatened. Instead of addition, there is subtraction; division instead of multiplication. Additive and subtractive
bilingualism is further discussed in the next chapter.
Steps in Reversing Language Shift
Fishman’s (1990, 1991, 1993) Graded Intergenerational Disruption Scale (GIDS) is an aid to language planning
and attempted language reversal. Just as the Richter scale measures intensity of earthquakes, so Fishman’s scale
gives a guide to how far a minority language is threatened and disrupted. The higher the number on the scale, the
more a language is threatened. The idea of stages is that it is little good attempting later stages if earlier stages are
not at least partly achieved. Various foundations are needed before building the upper levels. The value of the
scale is not just in its eight sequenced steps or

< previous page page_66 next page >
< previous page page_67 next page >
Page 67
Fishman’s (1990, 1991) Graded Intergenerational Disruption Scale for
Threatened Languages
Social isolation of the few remaining speakers of the minority
language. Need to record the language for later possible
Minority language used by older and not younger generation.
Need to multiply the language in the younger generation.
Minority language is passed on from generation to generation
and used in the community. Need to support the family in
intergenerational continuity (e.g. provision of minority language
nursery schools).
Literacy in the minority language. Need to support literacy
movements in the minority language, particularly when there is
no government support.
Formal, compulsory education available in the minority
language. May need to be financially supported by the minority
language community.
Use of the minority language in less specialized work areas
involving interaction with majority language speakers.
Lower government services and mass media available in the
minority language.
Some use of minority language available in higher education,
central government and national media.
stages. Rather it provides a plan for action for reversing languages in decline and a set of priorities. The eight
stages are briefly summarized, and then considered one by one.
Stage 8
Stage 8 represents the ‘worst case’ for a language. A few of the older generation will still be able to speak the
language but probably not to each other because they are socially isolated. The few remaining speakers of a
language are so scattered that minority language interaction is rarely possible. At this stage it is seen as important
that folklorists and linguists collect as much information as they can from these few survivors of the language
community. The folk-tales and sayings, grammar and vocabulary, need to be collected on tape and paper as a
permanent record of that language. Since the language building is in ruins and the foundations have crumbled, can
anything be done to save the language? The one ray of hope is that the records of the language can be used by a
younger generation to revive the language. With Australian Aboriginal languages and Cornish in England, this has
been attempted. Thus the remnants of the foundations can be reused to start to reconstruct the language.
Stage 7
A language in Stage 7 will be used on a daily communication basis, but by more mature-age speakers beyond
child-bearing. A language used by the older rather than the younger generation is likely to die as that older
generation disappears. The language is unlikely to be reproduced in the younger generation as mothers and fathers
speak the majority language with their children. While the aim of Stage 8 is to reassemble the

< previous page page_67 next page >
< previous page page_68 next page >
Page 68
language, the aim of Stage 7 is to spread the language amongst the young. Mothers and fathers are encouraged to
bring up their children in the minority language. It is essential to reproduce that dying language in children who
may later bring up their children in that language to ensure language continuity.
Fishman (1991) is clear about well-intentioned but, in the long term future of a language, less important events.
The danger of Stage 7 includes positive attitudes towards the language without positive action: ‘the road to societal
language death is paved with the good intentions called “positive attitudes'” (p. 91). The danger also is in
exaggerating the value of symbolic events in the minority language, on the stage and the written page, in the
gathering of the clans and at ceremonies. These events are relatively unimportant in the long term salvation of the
language compared with the raising of children in that minority language. However, this is not to argue that such
language events are valueless. Their value ultimately lies in indirectly encouraging language life at the daily
participative level.
When a language is passed on from generation to generation through child rearing practices, then a language has
some chance of long term success. In child development and particularly in the teenage years, using rather losing a
minority language becomes crucial. When the popular majority language culture becomes attractive to teenagers,
minority language youth activities with youth participating and interacting in their minority language becomes
critical (Baker, 1992).
The importance of ensuring that a minority language is passed from one generation to another is summed up by
Fishman (1991):
‘The road to societal death is paved by language activity that is not focused on intergenerational continuity,
i.e. that is diverted into efforts that do not involve and influence the socialization behaviors of families of
child-bearing age.’ (p. 91)
Performances and publications, ceremonies and cultural meetings need to be seen as a means to an end, and not as
an end in themselves. Such formal language events, however carefully arranged, do not intrinsically lead to the
passing of a language from generation to generation. Such events are valuable to the extent they foster the passing
of a language across generations. Such events must not just be for the elite, the firm disciples or the converts, but
have missionary aims to secure the language in the younger generation. Such events are important to the extent to
which they encourage everyday activity, at family and community level, in that language. One-off events (e.g.
eisteddfodau in Wales), may produce the feeling of a strong language and an emotional lift for the individual.
Fishman’s (1991) argument is that unless such events link into usual, daily, family socialization, the language in
the long term is not being promoted.
Stage 6
Stage 6 is seen as the crucial, pivotal stage for the survival of a language. In Stage 6, a language will be passed to
the next generation. At this sixth stage, the minority language

< previous page page_68 next page >
< previous page page_69 next page >
Page 69
will be used between grandparents, parents and children. This is regarded as quintessential in the fate of a
language. If the minority language is used in the family, it may also be used in the street, in shops, with neighbors
and friends. The language will more probably be spoken in the neighborhood and in community life, in religious
and cultural events, in leisure and local informal commercial activity. At Stage 6, there will be a language
community of a greater or lesser proportion.
This stage essentially concerns the informal use of a language in the home and the community. As such, the
language may be supported and encouraged, but it may be outside the realms of formal language planning. The
focus of this stage is the family, and the family within its community. As an institution, the family creates and
maintains boundaries from the outside that may prevent the majority language from over-intrusion. However,
Fishman (1991) points out that the more urbanized twentieth century family does not live in a strong enough castle
to prevent considerable outside influence from other languages. With the increase in one parent families and both
parents going out to work, the idea of mother and father as sole transmitters of the minority language is less strong.
The alternative is that agencies and institutions supporting the home also create a minority language and minority
cultural environment (e.g. the availability of minority language nursery schools). What is crucial is that early
childhood and teenage socialization is enacted in the minority language and culture. The danger is in believing that
other institutions such as the school, the mass media, the economy and government legislation will reverse
language shift in themselves. Rather, language reversal is pivotal on Stage 6. Fishman (1991) argues that Stage 6
cannot be dispensed with.
Unless a language is transmitted across the generations, other activity may have short term success and long term
‘If this stage is not satisfied, all else can amount to little more than biding time . . . Attaining Stage 6 is a
necessary, even if not a sufficient, desideratum of RLS [Reversing Language Shift].’ (p. 399)
Stage 5
Stage 5 occurs when a minority language in the home, school and community goes beyond oracy to literacy. First,
literacy in the minority language is seen to be important because it facilitates alternative means of communication,
especially across distance and time. Second, the image and status of a minority language is raised when it is
present in print. Such status is not merely symbolic. Literacy in the minority language means that a person is not
just subject to majority language print media. Majority language media will contain dominant and powerful
viewpoints, the attitudes of the center, the values and the beliefs of the majority language. Minority language
literacy allows the possibility of minority language culture, political and ideological viewpoints to be presented.
Third, literacy ensures a much wider variety of functions for a language. Such literacy may open more doors to
employment, increase the chance of social and vocational mobility. However, at Stage 5, literacy in the majority
language or biliteracy may be more important than literacy purely in the minority language if such mobility is

< previous page page_69 next page >
< previous page page_70 next page >
Page 70
Where education is through the majority language, then literacy in the minority language may be achieved through
local community effort. Via Saturday and Sunday schools, in evening classes and through religious institutions, a
literacy program in the minority language may be established. While such local efforts may be costly in money,
they have the advantage of giving control in such literacy education to the community rather than to the central
majority language government. With such control, local institutions can determine the appropriate means of
literacy and cultural acquisition. While creating a financial burden, such activities may give a language community
a focus, a shared commitment and a further raison d’être. While self-sacrifice is often hard and unfair, it can result
in vitality, commitment, enthusiasm and a sense of unifying purpose.
For Fishman (1991), Stages 8 to 5 constitute the minimum basis of reversing language shift. The activities at these
stages rely solely on the efforts of the language community itself. Such stages reflect a diglossic situation where
the minority language has separate functions from the majority language. Given that bilingualism rather than
minority language monolingualism will exist, such a minority language group may not be disadvantaged. They
will usually have access to mainstream education from elementary to University levels. The full range of
government services also will be available, including educational and welfare services. After Stage 5 comes the
attempt to capture the formal functions so far reserved for the majority language (e.g. mass media, compulsory
education, political self-determination). It is at this point that the minority language may come out of its enclave
and seek to challenge majority language castles.
Stage 4
One of the first approaches into the majority language castle may be through education. Schools may be created
and supported by the minority language community itself, not funded by the central purse. Such private schools
may be outside the budgets of many relatively poor minority language communities. Therefore, minority language
medium education paid for by central government may be sought. Central government will often still require some
control over the curriculum of such schools. That is, minority language medium education may only partly be
under the control of the local minority language community. Such schools may need to prove that they are as
effective and as successful as neighboring majority language schools; those attending them not being at a
disadvantage but an advantage.
Stage 3
In previous stages, some people will use their minority language in a place of work at the ‘lower work’ (local and
less specialized work) level. In previous stages, a few such workers will function economically in a relatively
isolated way in the local neighborhood or as a self-contained group. Few bridges will exist with the majority
language community. At Stage 3, creating a wider economic base for the minority language becomes important.
Such economic activity will involve the establishment of minority language staffed enterprises and services, not
just for the local market but for national and international

< previous page page_70 next page >
< previous page page_71 next page >
Page 71
markets as well. Such minority language enterprise will, at times, require communication in the majority language.
However, the on-the-floor activity and in-house communication may all be through the minority language.
However, success by language minority enterprise in international and national economic activity may not be
without danger to the minority language. The economic value of majority languages, and the arrival of in-migrants
should there be economic success, are just two possible after-effects. As economic independence and self-
regulation evolve, the temptation may be to increase profit by switching to international majority languages.
Such minority language economic activity nevertheless opens doors to increasing affluence, vocational mobility
and social status. Young minority language speakers may be more willing to stay in an area rather than migrate to
majority language pastures when there is economic opportunity. When there are jobs and money in the local
community, the minority language has firmer roots and resources.
Stage 2
The penultimate stage is extending lower government services and mass media into the minority language. Local
government institutions may offer their services through the minority language; health and postal services, courts
and the police service for example. Telephone and banking services, energy providing bodies and supermarkets
may also become willing to use the minority language in their service and communication with the public.
Central government may wish to control its services to language minority regions to influence attitudes and
opinions, information and ideology. The more the decision making processes (educational, economic and political)
are released to such regions, the more power that local language has to capture the minds of the minority. In Stage
2, national radio and television may be asked to provide a set number of hours in the minority language.
Alternatively, as in Wales, a particular channel may be mostly dedicated to minority language use. Not only does
the use of minority language television and radio help to disseminate the attendant culture. It will also provide
primary and secondary employment for minority language speakers.
The particular dangers of this stage may be in creaming off the most able and ambitious professionals into
majority language positions. Minority language media also face severe viewing competition from Anglo-
American, high-budget programs. The mass media provide status for a minority language, a channel for the
minority cultural message, but no salvation for the language in and by itself.
Stage 1
In this final stage, the pinnacle of achievement in language shift, the minority language will be used at University
level and will be strongly represented in mass media output, governmental services and throughout occupations.
Alongside some degree of economic

< previous page page_71 next page >
< previous page page_72 next page >
Page 72
autonomy will also be cultural autonomy. At this stage the language will be officially recognized in central
government legislation or a Language Act.
Throughout the four latter stages, Fishman (1991) is keen to point out that Stage 6 is still pre-eminent. When mass
communications, economic rewards and vocational opportunities exist through the minority language, it is still the
family, the neighborhood and community language life that is vital in the long term success of the language. All
the trappings of language status (e.g. mass media), all the power of legislation, all the success in economic self-
determination does not assure the future of a language. Important as are Stages 4 to 1, a language is ultimately lost
and won inside the minds and hearts of individuals. While such individuals will doubtless be affected by economic,
political and media factors, there is a personal cost-benefit analysis that ultimately determines whether one
language is passed on to the next generation or not.
Limits and Critics
While Fishman (1991) is careful to point out that one stage is not necessarily dependent on a previous stage, there
are priorities. The more advanced stages cannot usually be secured unless more foundational stages are either first
built or repaired. The danger increases in advancing on all fronts. Attempting to win individual battles without
having a strategy for the whole war does not champion success. There is also a danger in working solely for
tangible, newsworthy, easily recognized victories (Fishman, 1993). Changing the language of road signs, tax forms
and gaining minority language presence on television are battles that have been fought and won in some minority
language regions. It is more difficult, but more important, to support and encourage the minority language for
communication in daily family and community life. For Fishman (1991, 1993), it is the informal and intimate
spoken language reproduced across generations that is the ultimate pivot of language shift.
Initial activity to reverse language shift will usually derive solely from the minority language community. The
language community needs to be awoken and mobilized to support its language, especially at a family and
community participative level. However, there may come a time when the majority language government will
support that community’s efforts to survive. Through the provision of bilingual education, government services and
a minority language television service, the central government may come to support its minority languages.
Fishman (1991) is particularly guarded about how much bilingual education can achieve in reversing language
shift. There is sometimes the belief that, where families do not transmit the minority language, the school is there
to do it instead. Where parents do not bring up their children in the minority language, the school is expected to be
the substitute minority language parent. The school may initiate second language acquisition in the minority
language. But few rather than many may use the school-learnt language throughout life, particularly in parenting
their children. Even when a child successfully learn minority language oracy and literacy skills in school, unless
there is considerable support in the community and the economy outside school, that language may wither and die.
A classroom-learnt second language may become a school-only language.

< previous page page_72 next page >
< previous page page_73 next page >
Page 73
For that language to survive inside the individual, a person needs to become bonded in the language community
while at school, and particularly after leaving school. There needs to be preschool, out-of-school and after-school
support and reward systems for using the minority language. The minority language needs to be embedded in the
family-neighborhood-community experience and in the economics of the family. Unless this happens, it is
unlikely that bilingually educated children will pass on the minority language to the next generation. Thus, for
Fishman (1991, 1993), each stage needs examining for how it can be used to feed into Stage 6the inter-generation
transmission of the minority language.
Fishman’s (1991) eight stages must be seen as overlapping and interacting. In language revival, it is not the case of
going one step or stage at a time. The myriad of factors in language reversal link together in complex patterns. A
language at Stage 2 may still be securing elements of previous stages. A language at Stage 6 may be engaged in
long-term planning to secure higher stages. Also, different communities and different geographical areas may be at
different stages within the same nation. One area may have secured bilingual education provision, a neighboring
area may have undeveloped bilingual education provision. Minority language literacy may be strong in some
communities, weak in others. The use of the minority language in business and the local economy may vary
considerably from rural to urban areas, and according to closeness of access to airports, roads, railways and sea
links. In some villages, language death may be close. In other villages within the same region, most community
life may be in the minority language.
Glyn Williams (1992), in a critique of Fishman, has argued that the presupposition is that change is gradual,
mechanical, evolutionary and cumulative. Williams (1992) suggests that the viewpoint of Fishman tends to be of a
consensus nature, concerned with integration, equilibrium, order and cohesion. Williams (1992) regards the work
of Fishman as politically conservative with a consequent limited discussion of deviance, power, struggle and
conflict. The preference is to play down the conflict while ignoring power, thereby not expressing the anger,
discrimination and frustration felt by language minority groups and their members. This theme is returned to when
considering Paulston’s (1980) equilibrium and conflict paradigm’s in Chapter 19.
The theories of Fishman may be isolated from the world as viewed by language minority actors and actresses
themselves. The explanations ethnic minority participants themselves provide can be an alternative basis for the
explanation and understanding of language revival and reversal (a phenomenological perspective).
Typologies and graded scales are helpful in organizing thinking and create a general guide rather than a
comprehensive map. They represent, in outline form, a highly complex interaction of a variety of phenomena. The
GIDS scale is a valuable attempt at sequencing and prioritizing.

< previous page page_73 next page >
< previous page page_74 next page >
Page 74
This chapter has focused on language vitality and language reversal. Languages do not just exist inside individuals;
they exist inside groups varying in size and strength. Sometimes such language groups shift to insignificance even
death. Other times they attempt to spread and not just survive. One argument for the survival of languages has
been that as languages die, so does part of the totality of human history and culture. Just as species of animals and
plants are threatened, so are many minority languages. In the twentieth century, preservation of the environment
has become a dominant issue. Maintaining the diversity of nature has developed as part of twentieth century
thinking. Preserving language diversity may also be environmentally friendly.
Suggested Further Reading
FISHMAN, J.A., 1991, Reversing Language Shift. Clevedon: Multilingual Matters.
GILES, H. and COUPLAND, N., 1991, Language: Contexts and Consequences. Milton Keynes: Open University
PAULSTON, C.B., 1994, Linguistic Minorities in Multilingual Settings. Philadelphia: John Benjamins.
Review and Study Questions
(1) Write short revision notes on the following:
(i) The components of Language Vitality (economic status, social status, symbolic status, demographic and
institutional support factors).
(ii) The ‘assumptions of reversing language shift’.
(iii) The difference between status planning and corpus planning.
(iv) The difference between additive and subtractive bilingual situations.
(2) What are the economic advantages of bilingualism? Discuss the place of minority languages in the economy.
(3) List the eight stages of language shift by Fishman. Which stages seem more crucial in language death and in
language reversal? What overlaps might exist between the stages? What criticisms are there of the GIDS scale.
Study Activities
(1) Draw a diagram, chart or poster of Fishman’s eight RLS (Reversing Language Shift) stages. This may be a
simple poster with an immediate effect or a chart giving the crucial factors in each stage.
(2) Using language minority groups or communities known to you, where do you think they may be placed on
Fishman’s RLS scale? Do you find any difficulties in placement of groups on that scale?

< previous page page_74 next page >
< previous page page_75 next page >
Page 75
Chapter 5
The Development of Bilingualism
Types of Childhood Bilingualism
The Simultaneous Acquisition of Bilingualism
Case Studies of Simultaneous Childhood Bilingual
Language Loss in Children
The Sequential Acquisition of Bilingualism
Background Issues
Informal Second Language Learning
Formal Second Language Learning
The Age Factor
Children as Interpreters

< previous page page_75 next page >
< previous page page_76 next page >
Page 76
This chapter looks at the various ways in which children and adults become bilingual. There are various routes to
bilingualism. Such routes include: learning two languages early on in the home; acquiring a second language in the
street, in the wider community, in the nursery school, elementary or high school; and, after childhood, learning a
second or foreign language by adult language classes and courses. This chapter outlines different major routes to
becoming bilingual and examines some of the central issues involved in traveling along these routes.
As the previous chapters of this book illustrated, a discussion of bilingualism has to take in psychological,
linguistic, social and educational factors. While psychologists and linguists have studied the development of
children’s two languages, it is important to examine simultaneously the social context in which children acquire
their languages. Macro social contexts such as being a member of an in-migrant community, an elite group, a
counter-elite or a majority or minority language group are important influences in the acquisition or attrition of
bilingualism. There are also micro social contexts of the street, the nursery, the school, and the local community
that similarly foster functional bilingualism. Such contexts tend to make bilingualism a constantly shifting, rather
than a stable phenomenon.
The variety of individual differences and social contexts makes simple generalizations about the development of
bilingualism difficult and dangerous. The chapter therefore commences with a simple typology of the development
of childhood bilingualism.
Types of Childhood Bilingualism
An important initial distinction is between simultaneous and sequential childhood bilingualism. Simultaneous
childhood bilingualism refers to a child acquiring two languages at the same time early in life. For example, where
one parent speaks one language to the child, and the other parent speaks a different language, the child may learn
both languages simultaneously. Sequential childhood bilingualism is when the child

< previous page page_76 next page >
< previous page page_77 next page >
Page 77
learns one language first, and then a second language later in life. An example would be where a child learns the
language of the home, then goes to a nursery or elementary school and learns a second language. An approximate
boundary between simultaneous and sequential childhood bilingualism is the age of three (McLaughlin,
1984,1985). When a child acquires two languages’. Before the age of three, Swain (1972) calls it ‘bilingualism as
the first language’. Before the age of three, the acquisition of two languages is likely to be natural, informal and
untutored. After the age of three, there is an increased likelihood of the second language being acquired by formal
Some types of nursery provision and bilingual education encourage a child to acquire a second language without
formal instruction in that language (Sharpe, 1994). In contrast, second language classes for children and adults
foster bilingualism through direct instruction. This provides a distinction between informal language acquisition
and more formal language learning (Krashen, 1977, 1981, 1982, 1985). However, the boundary between
acquisition and learning is not distinct and separate (e.g. informal language acquisition can occur in a second
language class). As has been indicated in Chapter 1, there is a movement towards making second language
acquisition more naturalistic in an educational setting, developing communicative competence in a less formal
way. Thus, the distinction between naturally becoming bilingual and being taught to become bilingual may have
cloudy borders.
The Simultaneous Acquisition of Bilingualism
There are four basic dimensions along which the simultaneous acquisition of bilingualism in childhood varies.
These four dimensions may be translated into four questions.
(1) What language(s) is each parent ABLE to speak? In some family situations, the parents or guardians may both
be bilingual. That is, both parents may be able to speak both the languages of the particular society. For example in
the US, the parents may both be able to speak English and Spanish fluently. Alternatively, both parents may be
monolingual with the child acquiring the second language from relations, neighbors and the local community. In
other families, one parent may be more or less bilingual, the other monolingual. It is important when asking the
question of what language or languages each parent is able to speak, that consideration is made of whether those
languages are minority or majority languages. Does the context concern additive or subtractive bilingualism?
(2) What language(s) does each parent speak to the child IN PRACTICE? While parents have the ability to speak
both languages to their children, there is often a conscious decision or a latent understanding about which language
to use with the child from birth upwards. A bilingual parent may choose to use both the languages with the child.
A mother, for example, may use both English and Spanish with the child. A different situation is when one parent
speaks one language to the child, the other parent speaks a different language. For example, the mother may speak
Spanish to the child and the father will speak English only. A third circumstance is when

< previous page page_77 next page >
< previous page page_78 next page >
Page 78
bilingual parents both speak the minority language to their children, leaving the child to learn the majority
language outside the home.
(3) What language(s) do other family members speak to the child? There are families where both parents use the
same language in speaking to their children, but where the children speak to each other in the ‘outside’ language.
For example, with ‘later generation’ in-migrants, the parents speak the heritage language; the children speak to
each other in the language of the street, school and television. Playing with neighborhood children, making friends
in and out of school with majority language speakers and use of mass-media may create bilingualism in the child.
An alternative scenario is when the grandparents and other relations use a different language with the child than
the home language. For example, Chinese children in the US may speak English at home and at school, but
acquire at least a passive understanding of Cantonese Chinese through regular visits to extended family members.
(4) What language(s) does the child experience in the community? Even before the age of three, the language
experience with neighbors, local community and the nursery school may be a particularly important part of
becoming bilingual (Cummins, 1991b). Sometimes a child may experience both the languages of home in the
outside world. Alternatively, the child raised monolingually might pick up a second language outside the home.
For example, children whose parents speak Spanish to them in the home, may attend an English medium nursery
school and become bilingual in that way. The chapter continues by focusing on the more typical and better
documented routes to childhood bilingualism.
Case Studies of Simultaneous Childhood Bilingual Acquisition
Some of the earliest research on bilingualism concerns detailed case studies of children becoming bilingual. For
example, Ronjat (1913) described a case of the mother speaking German and the father speaking French in a
French community. Ronjat’s (1913) case study introduced the principle of ‘one person, one language’. That is, the
case study announced the idea that a most effective method of raising children bilingually was for each parent to
speak a separate language to the child.
While there have been a number of case studies of children growing up bilingually since Ronjat’s first study, the
most detailed case study is still Leopold’s (1939 to 1949). Leopold’s classic study of his daughter Hildegard was
based on the father speaking German only in the home and the mother speaking only English.
Leopold was a phonetician by training and made a comprehensive record of the development of Hildegard’s
speech. The results were published in four books which Hakuta (1986) describes as ‘a masterpiece of devotion to
detailed description’ (p. 48). Leopold analyzed the development of vocabulary, the sound system, word
combinations and sentences. In the first two years of her life, Hildegard did not separate out her languages. During
the first two years, German and English vocabulary was often mixed. It was not until she was three that Hildegard
started to use the two languages separately. At the age of three she spoke German to her father, English to her
mother and engaged in simple translations between the two languages. A few words of German were still used

< previous page page_78 next page >
< previous page page_79 next page >
Page 79
in English sentences, and vice versa. However Leopold suggested that this was not due to interference between the
languages but to the child systematically simplifying the production of language. Efficient communication from
the child’s perspective naturally involved some language mixing.
One interesting aspect of Leopold’s studies is the shifting balance of the two languages in childhood. When
Hildegard went to Germany, her German became stronger. When back in the United States and attending school,
Hildegard’s English became the dominant language. This seems to reflect the reality of many bilingual situations
where, at an individual level (and not just at a societal level), the languages shift in dominance. Hildegard, for
example was reluctant to speak German during her mid-teens, with German becoming the weaker language.
Leopold’s second daughter understood German but spoke very little German to her father. In childhood, the second
daughter, Karla, was a passive bilingual. Yet at the age of 19, Karla visited Germany where she was able to change
from receptive German to productive German, managing to converse relatively fluently in German. Another
example of shifting bilingualism in childhood is given by Fantini (1985) who details a child’s shift between
English, Italian and Spanish from birth to the age of 10.
The problem with case studies such as Leopold’s is that the cases may be atypical. The children studied often tend
to be relatively precocious children or the offspring of the linguists who are conducting the research. That is, the
case studies may be of exceptional rather than ordinary children.
Apart from the ‘one person, one language’ method of raising bilingually, there are other case studies of different
patterns (see Schinke-Llano, 1989; Romaine, 1995). Romaine (1995) provides a six-fold typology of early
childhood bilingualism in terms of the language spoken by mother and father, the language of the community and
the strategy of the parents. Each of the six patterns reveal different approaches that can be successful in bringing
up children bilingually. We have mentioned two of these approaches: the parents speaking a different language to
the child; and parents speaking a minority language to the child who acquires a second language in the community
or extended family. Two further approaches must also be mentioned.
First, there is the case where both parents (and the community) are bilingual and constantly mix their languages.
Romaine (1995) considers this ‘a more common category than it might seem on the basis of its representation in
the literature’ (p. 186). For example, Maltese and English may be mixed together by both parents with little
demarcation. The child grows up mirroring the language mixing of the parents. This, at times, may be both stable,
frequent and totally acceptable within the language community. The Puerto-Ricans of New York mixing Spanish
and English is a particularly strong example of this.
Second, a notable case study of simultaneous bilingualism is given by Saunders (1988) who raised his sons in
English and German in Australia. Even though Saunders was not a native speaker of German, and there was little
support for German speakers in the community, he raised his sons Frank and Thomas bilingually. From birth, the

< previous page page_79 next page >
< previous page page_80 next page >
Page 80
spoke to the children in English while the father used German. The parents spoke English to each other as did the
two boys when speaking to each other.
Saunders (1988) observed that his children went through a three stage developmental sequence in becoming
bilingual. While the notion of discrete stages (and allocation of ages) is contested and debated, the general
developmental pattern follows. The first stage lasts until the child is approximately two years of age. During this
time the child does not differentiate between the two languages. Vocabulary, for example, is treated as part of one
global language system. During the second stage, the child may still be mixing the two languages on occasions.
However there will be increasing differentiation between the two languages. For example, the vocabulary will be
separated, the child will know what language to speak to which person, and what language to speak in which
situations. Saunders (1988) also suggests that during this second stage, the child sometimes says both words for
the same thing.
This may be to ensure the meaning is clear, or it may ensure that the parents attend to the child. The age at which a
child maximally differentiates the two language systems and rarely mixes the two (Stage 3) will differ
considerably from child to child. A variety of factors may affect the point at which a child separates the two
languages: exposure to the two languages in different domains, the attitudes of parents to the two languages and to
mixing the languages, the language abilities of the child, personality, peers and exposure to different forms of
language education.
Studies of the acquisition of simultaneous bilingualism in childhood make two other key points. First, Swain
(1972) found that the simultaneous acquisition of two languages did not differ in development order or process
from the acquisition of one language. Children appear to learn two languages as if they were learning one. Second,
and related, it seems possible to conclude that bilingualism can be the first language. This includes the idea that
there may be a single language system underlying both the languages of the bilingual child. The two languages
may be fused and integrated inside the linguisticcognitive system of the bilingual. This is a theme examined in
Chapter 9.
Language Loss in Children
Before considering the sequential acquisition of bilingualism, it is important to mention language loss in bilingual
children. In particular, children from language minorities (indigenous and particularly in-migrant children) are
often at risk of losing their minority language. With a higher status majority language ever present on the screen,
in the street, at school and in shops, children quickly learn which language has prestige, power and preference.
They quickly learn that they are different in language, behavior, ethnicity and culture, and may come to perceive
their language and cultural differences as undesirable. Young children quickly perceive what helps them belong
and get accepted in mainstream society. As Wong Fillmore (1991a) relates about US language minority children,
they soon ‘discover that one of the things that stands between them and easy participation in their world is
language. They can tell by the way people interact with them that the only language that counts for much is
English . . . If they want to be accepted, they have to learn English, because the others are not going to learn their
language’ (p. 342).

< previous page page_80 next page >
< previous page page_81 next page >
Page 81
Research by Hakuta & D’Andrea (1992) and Wong Fillmore (1991a) suggests that the younger the US language
minority child is in acquiring English, the greater chance that the minority language will be replaced. The later the
age of learning to speak English, the greater the chance of retaining the minority language and becoming bilingual.
Hakuta & D’Andrea (1992) propose that early exposure to English (e.g. in the home) not only may mean a shift
from Spanish to English. It may also decrease the chances of placement in a bilingual education program where
Spanish is used and possibly maintained.
When English is introduced early into a US language minority child’s life, the minority language may be
insufficiently stable and developed not to be affected by the majority language. A loss of the minority language
may have consequences for the child: social, emotional, cognitive and educational, as later chapters will examine.
As Wong Fillmore (1991a: 343) argues: ‘What is lost is no less than the means by which parents socialize their
children: When parents are unable to talk to their children, they cannot easily convey to them their values, beliefs,
understandings, or wisdom about how to cope with their experiences’.
Thus minority language loss is an issue not just of regions and language communities but also of individual
children, particularly when they are very young. The problem is not the majority language, but the timing of the
introduction of the minority language. Young children need to build competence in the minority language before
the powerful and almost inevitable encounter with the majority language begins.
The Sequential Acquisition of Bilingualism
Sequential acquisition of bilingualism refers to the situation where a child acquires a first language, and later
becomes proficient in the second language. McLaughlin (1984) recommends that the age of three is used as an
arbitrary demarcation line between simultaneous and sequential acquisition of bilingualism. Sequential acquisition
of bilingualism takes us into the field of second language acquisition. Such acquisition may be through formal or
informal means; informally through street, nursery school and community, or formally through school, adult
classes and language courses. There is no single ‘best’ route by which learners above the age of three become
competent in a second language. There are a variety of informal and formal educational means of acquiring
competency in a second language.
Background Issues
The evidence of the case studies is that children can become competent bilinguals through the process of
simultaneous bilingualism. The evidence of bilingualism through second language acquisition or learning is not
always so positive. In the US and Britain, despite the extensive second language and foreign language learning in
school (and the extensive research on second language acquisition), only a small proportion of children learning a
second language become functionally and fluently bilingual. In the US, less than 1 in 20 children become bilingual
following second language instruction. There are several popular reasons for such failure: the emphasis on reading
and writing rather than on authentic communication; having low ability or aptitude to learn a second language; a

< previous page page_81 next page >
< previous page page_82 next page >
Page 82
lack of motivation and interest, and a lack of opportunity to practice second language skills. Another popular
explanation is attempting to learn a language too late; that is, believing that it is easier to learn a language when
someone is younger rather than older. The issue of age in learning a language is considered later.
In certain European countries (e.g. Netherlands, Finland, Sweden, Belgium) and eastern countries (e.g. Israel,
Singapore), second language learning has been relatively more successful. Such international comparisons
highlight the need to bring political, cultural and economic factors into second language learning discussions. No
language learner is an island. Surrounding the shores of the individual psychology of effective second language
acquisition, lie the seas of social, cultural and political context. Any map of sequential bilingualism needs to
include all these features.
Informal Second Language Learning
Bilingualism is often achieved through the informal acquisition processes of the street and screen, neighborhood
and newspaper. A child sometimes rapidly acquires a language in addition to that of the home without planning or
intent by parents. Peers in the street, cartoons and shows on television are two examples of language influences
that may informally lead to bilingualism in the child and teenager. Little researched, the almost incidental addition
of a second or third language via the street and screen may be as influential as formal education, and sometimes
more potent than language classes. This particularly tends to be the case with acquiring the majority language of
the neighborhood or nation.
Formal Second Language Learning
Where a second language cannot be acquired in the community (natural second language acquisition), the school
has been the major institution expected to produce second language learning. Through second language and
foreign language lessons, via language laboratories and computer aided language learning, drill and practice
routines, immersion classes, drama and dance, the initial stages of monolingualism to bilingualism may occur. The
role of the school in formal second language learning is considered in detail in the second half of this book.
The routes to bilingualism are not solely in early childhood and in formal education. Community provided,
voluntary language classes sometimes exist for school-age children. In England and Canada, for example, evening
classes, Saturday schools and Sunday schools are organized by various communities for children to learn the
heritage language of their parents and grandparents. For example, children of second, third or fourth generation in-
migrants may have learnt English as their first or dominant language. If parents have chosen to speak English to
their children, even if their own first language is not English, the heritage language may be learnt in voluntary
classes. Alternatively, where English is the dominant language of the community and the only language of the
school, such voluntary classes may be important in attaining bilingualism rather than moving children towards
majority language monolingualism (Linguistic Minorities Project, 1985; Ghuman, 1993, 1994).

< previous page page_82 next page >
< previous page page_83 next page >
Page 83
Such voluntary provision may be for religious, cultural, social, integrative and ethnic minority vitality reasons.
Thus the providers are often religious groups such as Synagogues, Mosques, Temples and Orthodox churches. In
other communities, the providers are groups of enthusiastic parents and local community organizations who rent
premises such as schools and halls to teach a heritage language. Tansley & Craft (1984) located at least 28
different languages being taught in over 500 community and supplementary schools in England. Asian languages
such as Urdu, Panjabi, Bengali, Hindi and Gujerati were frequently taught. Also, European languages such as
Greek, Italian, Spanish and Portuguese were supported by such voluntary provision.
Apart from voluntary classes for children, another well traveled route to second language acquisition is adult
provision (see Baker & Prys Jones, 1997). Such provision takes varying forms in different geographical areas:
Evening classes. Sometimes called night schools or classes, a second or foreign language is taught on a once or
twice a week basis for several months to several years. Such classes have often traditionally aimed at securing
formal qualifications in the language (e.g. passing exams in a second majority language), or at gaining proficiency
in the majority language One example is ‘English as a Second Language’ classes established for in-migrants into
the US and England. Recently, the growth has also been in acquiring communicative competence in a heritage
language (e.g. Hebrew, Welsh).
Ulpan courses. Ulpan courses originated in Israel in the 1970s and are more intensive than evening classes. Such
courses are often for three to five mornings or three to five evenings every week, lasting from several months to a
year or more. Intensive and saturated, there is evidence from Israel and Wales of their success. This success is
partly due to the warm and encouraging environment created in an Ulpan as well as the language teaching.
Short term and long term residential courses. Weekend, week-long and up to three month residential courses for
minority language learning (taken in vacation time or via sponsorship from an organization) are other forms of
intensive language learning. For example, in North Wales, a deserted village by the sea (Nant Gwrtheyrn) has been
converted into an ‘isolated’ language learning center. The whole day’s activity is varied, but is centrally focused on
learning Welsh as a second language.
Distance learning methods. A variety of media-based courses for learning a second language are often available to
adults. Radio and television series, cassette tapes, records, compact discs, videos, video discs magazines and self-
teach books, computer programs (Computer Assisted Language Learning) and correspondence courses are all well
tried approaches in second language acquisition. Evaluation studies of the relative effectiveness of these different
approaches tend to be lacking.
In early childhood, becoming bilingual is often an unconscious event, as natural as learning to walk and to ride a
bicycle. In a school situation, a child is not usually the one who has made a decision about the language(s) of the
classroom. Second language acquisition at school is often imposed by parents, teachers and a regional educational
policy. For migrant workers, refugees and in-migrants, adult language learning also may

< previous page page_83 next page >
< previous page page_84 next page >
Page 84
be far less voluntary. In adulthood, second language acquisition sometimes becomes more voluntary, more open to
The Age Factor
A much debated theme in second language acquisition is the relationship of age in learning a second language and
success in gaining language proficiency (Harley, 1986). One set of proponents suggest that the younger the age a
second language is learnt, the greater the long term proficiency gained in that language. According to this
viewpoint, young children learn a language more easily and successfully. Others tend to argue that older children
and young adults learn a language more efficiently and quickly than do young children. For example, a 14 year old
learning Spanish as a second language has superior intellectual processing skills than the five year old learning
Spanish. Therefore, less time is required in the teenage years to learn a second language than in the younger years.
A comprehensive and balanced review of this area is provided by Singleton (1989). Singleton’s careful analysis
may be briefly summarized as follows:
(1) Younger second language learners are neither globally more nor less efficient and successful than older
learners in second language acquisition. There are many factors that intervene and make simple statements about
age and language learning simplistic and untenable.
(2) Children who learn a second language in childhood do tend to achieve higher levels of proficiency than those
who begin after childhood. Such a finding does not contradict the idea that someone can become proficient in
learning a second language after childhood. This tendency may be related to social contexts in which language is
acquired and maintained or lost, as well as to the psychology of individual learning. Younger children appear to
pick up the sound systems and grammar of a new language more easily than adults.
(3) Broadly speaking, there are no age related differences in the process of language learning. Younger and older
second language learners tend to show a similar developmental sequence and order.
(4) In a formal classroom language learning situation, older learners tend initially to learn quicker than younger
learners. However, the length of exposure (e.g. the number of years of second language instruction) is an
important factor in second language success. Those children who begin to learn a second language in the
elementary school and continue throughout schooling, tend to show higher proficiency than those who start to
learn the second language later in their schooling. In absolute rather than comparative terms, this still includes the
possibility of late learners becoming highly proficient, particularly when they are highly motivated.
(5) Where the second language is used in schools as the medium of instruction, and where that second language is
a majority language replacing the home minority language, early use of the second language may have negative
educational and linguistic effects.

< previous page page_84 next page >
< previous page page_85 next page >
Page 85
(6) Support for second language instrnction at an early age in school needs to find its rationale and support from
areas other than second language research. For example, teaching a second language early in the elementary school
needs to be defended in terms of general intellectual stimulation, the general curriculum value of teaching a
modern language, the benefits of biculturalism and the benefits of learning a language for as long as possible
rather than as quickly and as efficiently as possible. Second language instruction in the elementary school rests on
the suitable provision of language teachers, suitable materials and resources, favorable attitudes of the teachers and
parents, and the need to make the learning experience enjoyable for such children.
(7) There are no critical periods in a child’s development in childhood or adolescence when a second language
should or should not be introduced in the school. As Singleton (1989) notes, the whole issue of age and second
language learning is ‘that the various age-related phenomena isolated by language acquisition research probably
result from the interaction of a multiplicity of causes and that different phenomena may have different
combinations of causes’ (p. 266).
How successful have adults been in becoming bilingual? There is distinction between answering this question in an
absolute and a relative manner. The ‘absolute’ answer simply is that adults do learn a second language to varying
degrees of fluency. Some fall by the wayside, others reach a basic, simple level of communication, yet others
become functionally bilingual. In Israel and Wales, New Zealand and Canada, the US and many parts of Asia, the
adult route to bilingualism has many success stories.
The ‘relative’ answer involves comparing children and adults of varying ages. In this sense, the question becomes
‘Who is more likely to master a second language, children or adults?’ A specific example of adult success provides
an illustration of a typical pattern. This example focuses on language usage rather than acquisition.
In-migrants into Israel during this century have often learnt Hebrew in a central attempt to revive the ethnic,
religious language. From Israeli Census data, it is possible to examine whether older or younger adults become
functional in Hebrew. For example, do young in-migrants become more or less functional in Hebrew as a second
language compared with older in-migrants? The results follow a clear pattern (Bachi, 1956; Braine, 1987). As the
figure on the following page illustrates (based on Bachi, 1956), the extent of the everyday use of Hebrew varies
with age of in-migration. The younger the child, the more likely he or she will be to use Hebrew. Between 30 and
40 years of age, a notable drop occurs. Is this due to a loss of learning ability, less exposure to Hebrew, less
motivation or decreasing social pressure? From age 40 onwards, the likelihood of being functional in Hebrew falls
One issue that frequently occurs for parents and teachers of children of differing ages is when one language is
mixed with another. Terms such as Spanglish (mixed Spanish and English), Wenglish (mixed Welsh and English)
are invented to describe such mixing. This will now be considered from a sociolinguistic perspective. Those
interested in a purely linguistic perspective can consult Romaine (1995).

< previous page page_85 next page >
< previous page page_86 next page >
Page 86
One issue in language processing in bilingual children has been the extent to which such children mix and switch
their languages. Code mixing has tended to be the term used to describe changes at the word level (e.g. a few
words in a sentence). Codeswitching is when an individual (more or less deliberately) alternates between two or
more languages. Such alternation can range from one word mixing, to switching in mid-sentence, to switching in
larger speech blocks (Hoffmann, 1991). Language borrowing is the inclusion of foreign loan words into a language
(e.g. ‘le weekend’ in French).
Codeswitching tends to be used to describe changes which are relatively deliberate and with a purpose. This may
be a switch in a particular phrase within a sentence or in different sentences. ‘Leo un magazine’ (I read a magazine)
would be mixing; ‘Come to the table. Bwydyn barod’ (Food is ready) would be switching. However the distinction
easily becomes blurred. For example is ‘I wanted to fight her con los punos, OK! Is this a mix or a switch?
As Eastman (1992) suggests: ‘efforts to distinguish codeswitching, codemixing and borrowing are doomed’ (p. 1).
Particularly in research in multilingual urban settings, where people of diverse linguistic backgrounds regularly
interact, ‘in normal everyday conversation, material from many languages may be embedded in a matrix language
regularly and unremarkably’ Eastman (1992: 1). Where codeswitching is regularly used, this is called an unmarked
choice of language. In contrast, where people more deliberately use codeswitching for social, political, economic
purposes, this is termed a marked choice of language (Myers Scotton, 1983, 1991).

< previous page page_86 next page >
< previous page page_87 next page >
Page 87
Codeswitches have a variety of purposes and aims. Codeswitching will vary according to who is in the
conversation, what is the topic, and in what kind of context the conversation occurs. The languages used may be
negotiated and may change with the topic of conversation. Thirteen over-lapping purposes of codeswitching will
now be considered.
(1) Codeswitches may be used to emphasize a particular point in a conversation. If one word needs stressing or is
central in part of a conversation, a switch may be made.
(2) If a person does not know a word or a phrase in a language, they may substitute a word in another language.
This often happens because bilinguals use different languages in different domains. A young person may, for
instance, switch from the home language to the language used in school to talk about a subject such as
mathematics or computers. For example, a university student in Kenya may switch from Kikuyu to English to
discuss Geometry with his younger brother. ‘Atiriri angle niati has ina degree eighty; nayo this one ina mirongo
itatu’. Similarly, an adult may codeswitch when talking about work, because the technical terms associated with
work are only known in that language.
(3) Words or phrases in two languages may not correspond exactly and the bilingual may switch to one language
to express a concept that has no equivalent in the culture of the other language. For example, a French-English
bilingual living in Britain may use words like ‘pub’, ‘bingo hall’ ‘underground’, when speaking French, because
there are no French equivalents for these words. When such words or phrases from one language become
established and in frequent use in the conversation of speakers of the other language, they are often called ‘loans’
or ‘borrowings’. However, there is no clear line of demarcation between a codeswitch and a borrowing.
(4) Code switching may be used to reinforce a request. For example, a teacher may repeat a command to accent
and underline it. (e.g. Taisez-vous les enfants! Be quiet, children!) In a majority/minority language situation, the
majority language may be used to underline authority. Hoffmann(1981) found that Puerto Rican mothers in Jersey
City tended to use English with their children for short commands like ‘Stop it! Don’t do that!’ and then switch
back into Spanish. Research in Wales found that language minority nurses repeat or amplify commands to Welsh
speaking patients in English to emphasize their authority. Roberts (1994: 66) relates the following observation of a
patient in hospital who is quickly becoming unpopular with nurses for continuously ringing the call button. The
patient, on this occasion, rings twice. The nurse arrives at the bed:
Ganoch chi’r gloch DWY WAITH! Peidiwch byth a’i ganu e dwywaithemergency ydy hynny. Dim on
UNWAITH sydd angen. Only ring it ONCE!’ [You rang the bell TWICE! Don’t ever ring it twicethat’s an
emergency. ONCE is enough. Only ring it ONCE!]
(5) Repetition may also be used to clarify a point. Some teachers in classrooms explain a concept in one language,
and then explain it again in another language believing that repetition adds reinforcement and completeness of

< previous page page_87 next page >
< previous page page_88 next page >
Page 88
(6) Codeswitching may be used to communicate friendship. For example, moving from the common majority
language to the home language or minority language both the listener and speaker understand well, may
communicate friendship and common identity. Similarly, a person may deliberately use codeswitching to indicate
the need to be accepted by a peer group. Someone with a rudimentary knowledge of a language may inject words
of that new language into sentences to indicate a desire to identify and affiliate. The use of the listener’s stronger
language in part of the conversation may indicate deference, wanting to belong or to be accepted.
(7) In relating a conversation held previously, the person may report the conversation in the language or languages
used. For example, two people may be speaking Spanish together. When one reports a previous conversation with
an English monolingual, that conversation is reported authenticallyin Englishas it occurred.
(8) Codeswitching is sometimes used as a way of interjecting into a conversation. A person attempting to break
into a conversation politely may introduce a different language. Interrupting a conversation may be signaled by
changing language. The message to the speakers from the listener is that ‘I would like to become involved in this
(9) Codeswitching may be used to ease tension and inject humor into a conversation. If discussions are becoming
tense in a committee, the use of a second language may signal a change in the tune being played. Just as in an
orchestra, different instruments may be brought in during a composition to signal a change of mood and pace, so a
switch in language may indicate a need to change mood within the conversation.
(10) Codeswitching often relates to social distance. For example, when two people meet, they may use the
common majority language (e.g. Swahili or English in Kenya). As the conversation proceeds and roles, status and
tribal identity are revealed, a change to a regional language may indicate that boundaries are being broken down.
A codeswitch signals there is less social distance, with expressions of solidarity and growing rapport indicated by
the switch. A study of Italian in-migrants into the United States at the turn of the century (Di Pietro, 1977) showed
that the in-migrants would tell a joke in English and give the punch line in Italian, not only because it was better
expressed in that language, but also to emphasize the shared values and experiences of the minority group.
Conversely, a change from a minority language or dialect to a majority language may indicate the speaker’s wish
to elevate their own status, or create a distance between themselves and the listener. Myers Scotton & Ury (1977)
describe a conversation between a Kenyan shop keeper and his sister, who had come in to buy some salt. After
exchanging greetings in their own Luyia dialect, the brother switches to Swahili in front of the other customers
and says: Dada, sasa leo unahitaji nini? (Sister, what do you need today?) For the rest of the conversation the
brother speaks in Swahili and the sister in the Luyia dialect. The brother’s code switch to Swahili, the business
language of Kenya, indicates to his sister that she cannot expect to receive anything for free.
(11) Codeswitching can also be used to exclude people from a conversation. For example, when traveling on the
Metro (subway, underground), two people having been speaking English may switch to their minority language to
talk about private matters,

< previous page page_88 next page >
< previous page page_89 next page >
Page 89
thus excluding others from the conversation. Bilingual parents may use one language together to exclude their
monolingual children from a private discussion. A doctor at a hospital may make a brief aside to a colleague in a
language not understood by the patient.
(12) Codeswitching may be used to indicate a change of attitude during the conversation. For example, greetings
may be expressed in the home, minority language. But when one person asks to borrow money or asks a favor of
the other, the moneylender may change to the majority language. To indicate a change of relationship and a
change of attitude, codeswitching occurs. Greetings in the native language; business in the majority language. At
the end of the business conversation, farewells may revert to the native language. A change in relationship, on a
temporary basis, is indicated by codeswitching.
(13) In some bilingual situations, codeswitching occurs regularly when certain topics are introduced (e.g. money).
Spanish-English bilinguals in the South West United States regularly switch to English to discuss money. For
example, a person may say ‘La consulta era (the visit costs) ten dollars and 50 cents’. This reflects that English is
the language of commerce, and also the language of education.
Familiarity, projected status, the ethos of the context and the perceived linguistic skills of the listeners affect the
nature and process of code switching. Thus codeswitching has more than linguistic properties (Hoffmann, 1991).
These thirteen purposes of codeswitching demonstrate that there are important social and power aspects of
switching between languages, as there are between switching between dialects and registers.
A variety of factors may affect the extent to which children and adults switch between their languages (Romaine,
1995). The perceived status of the listeners, familiarity with those persons, atmosphere of the setting and perceived
linguistic skills of the listeners are examples of variables that may foster or prevent code-switching. Such factors
operate as young as two years of age. Whereas two-year-old’s mixing of language has tended to be seen as
‘interference’ or a lack of differentiation between languages, Lanza (1992) has shown from one case study that
codeswitching by two-year-olds can be context-sensitive (e.g. according to which parent is being addressed).
One situation where bilinguals mix the two language systems is sometimes found where both parents speak both
languages to the children. Various investigators have argued that the ‘one person, one language’ or ‘one language,
one context’ is the optimal method of rearing a child bilingually. This may help to avoid mixing two languages. For
example, where both the father and mother speak French and German to the child, the child may have temporary
difficulty in separating out the two languages. This used to be called ‘interference’. This term reveals a negative
stance from the listener’s viewpoint. In contrast, a child mixing two languages may be conveying thoughts and
ideas in the most personally efficient manner, and in a way that is understood by listeners. Such ‘interference’ may
be termed ‘transfer’ and is thus part of a natural bilingual developmental sequence. That is, mixing words often
changes as children grow older. With increasing years there usually develops increasing separation of languages
(Arnberg &

< previous page page_89 next page >
< previous page page_90 next page >
Page 90
Arnberg, 1992). Such ‘interference’ is often reduced as language experience increases. Thus Fantini (1985) argues
for a gradual evolutionary process of separation of, and transfer between, a child’s two languages rather than the
more static idea of ‘interference’. However, some bilingual communities (e.g. Puerto Ricans bom in New York) do
not show increased separation as their children grow older. There can be continuous mixing throughout life, and
this need not create a transitional state but a relatively stable language community (Zentella, 1988; Eastman, 1992).
The chapter concludes by examining a topic related to codeswitching: children acting as interpreters for their
parents and others. Bilingual children (and adults) are frequently expected to act as go-betweens by interpreting
from one language to another. Such an interpreter’s role also illustrates how bilingual development impacts on
other aspects of child development such as personality and family relationships.
Children As Interpreters
In language minority families, children often act as interpreters for their parents. For example, in first and second
generation in-migrant families, parents may have a little or no competency in the majority language. Therefore,
their children act as interpreters in a variety of contexts. When there are visitors to the house, such as sellers and
traders, religious persuasionists and local officials, a parent may call a child to the door to help translate what is
being said. The child interprets for both parties (e.g. the parent and the caller). Similarly, at hospitals, the doctor’s,
dentist’s, optician’s, school and many other places where parents visit, the child may be taken to help interpret.
Interpretation may be needed in more informal places: on the street, when a parent is watching the television or
listening to the radio, reading a local newspaper or working on the computer.
Pressure is placed on children in interpreting: linguistic, emotional, social and attitudinal pressure. First, children
may find an exact translation difficult to achieve as their language is still developing. Second, children may be
hearing information (e.g. medical troubles, financial problems, arguments and conflicts) that is the preserve of
adults rather than children. Third, children may be expected to be adult-like when interpreting and child-like at all
other times; to mix with adults when interpreting and ‘be seen and not heard’ with adults on other occasions.
Fourth, seeing their parents in an inferior position may lead to children despising their minority language. Children
may quickly realize when interpreting that the language of power, prestige and purse is the majority language.
Negative attitudes to the minority language may result. Fifth, not all bilinguals are good interpreters (Malakoff,
1992). Proficiency in two or more languages is not enough. Some reflection on language (in Chapter 8 discussed
as metalinguistic awareness), an awareness of the linguistic nature of the message, reflecting on the structural
features of languages is also required (Malakoff, 1992). However, many bilinguals appear to be ahead of
monolinguals on such metalinguistic awareness (see Chapter 8 for a review).
Interpreting also has many positive outcomes. First, it can bring parental praise, reward and status within the
family for playing a valuable role. The research of Malakoff (1992) found that this ability is widely distributed
among bilingual children who are quite

< previous page page_90 next page >
< previous page page_91 next page >
Page 91
expert as early as the third or fourth grade. Such ability may gain both esteem from others and raise self-esteem.
Second, the child learns adult information quickly and learns to act with some authority and trust. Early maturity
has its own rewards in the teenage peer group. Third, Kaur & Mills (1993) found that children accustomed to
acting as interpreters learned to take the initiative. For example, a child may give the answer to a question rather
than relaying the question to the parent. This puts children in a position of some power, even of language
Fourth, when parents become dependent on their children for interpretation, it may make the family more close
and integrated. Such interpretation is a lifeline for the many parents who have to hand over much power to their
children. Yet it may make parents aware of their own language inadequacies, feeling frustration and resentment,
particularly in language minority cultures where children are expected to stay in a subordinate position for a long
Fifth, the cognitive outcomes for child interpreters may be valuable. Children who are regular interpreters for their
parents may realize early on the problems and possibilities of translation of words, figures of speech and ideas. For
example, such children may learn early on that one language never fully parallels another, and that it is hard to
translate exactly the inner meaning of words and metaphors. This may lead such children to be more introspective
about their languages. This is termed metalinguistic awareness and is considered in Chapter 8. Thus, interpretation
may both require and stimulate metalinguistic awareness.
Sixth, another advantage for the child interpreter is in character formation, for example, possibly gaining more
empathy. The children are negotiating between two different social and cultural worlds, trying to understand both,
and provide bridges between these two worlds. This handling of dialogue may lead to increased maturity,
astuteness, independence and higher self-esteem. Being expected to carry an adult role early on may lead to a
positive self-concept, and feeling responsible like an adult.
This chapter has discussed two routes to bilingualism: simultaneous and sequential. The former route occurs up to
the age of three when two languages are acquired consecutively. The latter route, sequential bilingualism, may be
through formal educational avenues or informal paths (e.g. street, neighborhood and nursery schools). Language
learning in classes and courses and ‘street’ acquisition allow individuals of all ages to become bilingual and
In the home, in the street, at a nursery school, fast and efficient progress is made in an unconscious, natural
manner. A particularly efficient route is the ‘one parent, one language’ approach. Here, each parent speaks a
different language to the child. In simultaneous bilingual acquisition, the ‘one person, one language’ principle and
the ‘one context, one language’ idea help encourage language boundaries to be set. Mixing languages is often an
initial stage, with increasing separation with age and experience.

< previous page page_91 next page >
< previous page page_92 next page >
Page 92
However, in some communities, mixing is normal and relatively stable. Codeswitching is a frequent behavior
among bilinguals with a variety of valuable purposes and benefits. Interpreting is a similarly frequent expectation
of bilingualsincluding young children in in-migrant families.
Another route to bilingualism is via formal second language learning in school, or using a second language in
school as the medium of learning. Bilingualism can also be reached by avenues outside of school. The vehicles of
voluntary classes and adult courses provide the opportunity for a second or foreign language to be learnt and
developed. The street and screen can also be vehicles to childhood bilingualism. Having introduced simultaneous
and sequential bilingualism, the next chapter takes a global look at some of the key issues and current ideas in
second language acquisition.
Suggested Further Reading
BAKER, C., 1995, A Parents’ and Teachers’ Guide to Bilingualism. Clevedon: Multilingual Matters.
HAKUTA, K., 1986, Mirror of Language. The Debate on Bilingualism. New York: Basic Books.
ROMAINE, S., 1995, Bilingualism (2nd edn). Oxford: Basil Blackwell.
Review and Study Questions
(1) Write short revision notes on the following:
(i) The difference between language acquisition and language learning.
(ii) Different methods of simultaneous childhood bilingual acquisition.
(iii) The advantages and problems of children acting as interpreters.
(2) What are the main differences between simultaneous and sequential bilingualism?
(3) How influential do you think age is in learning a second language?
(4) Give examples of the ten purposes of codeswitching. Are there any other purposes you can think of or have
(5) Examine between three and six of the 117 ‘questions about bilingualism most asked by parents’ in Baker
(1995b). Does the advice given in the answers (a) match your experience and (b) seem relevant. Write your own
advice to any one question selected from that book.
Study Activities
(1) Create a case study of one person’s bilingual development. This may be yourself or someone you know. By
interviewing that person, or self reflection, make a cassette tape or a video or a written case study of the factors
which seem personally important to that person in their bilingual development. This may be developed as a
project. A project may include the following stages. First, look in the library for case studies of bilinguals. Books
by Arnberg (1987), Fantini (1985), Harding & Riley (1987), de Jong (1986), Saunders (1988), Lyon (1996) and
Döpke (1992) provide examples. Second, prepare an interview guide. Write down (as an aide memoire) the topics

< previous page page_92 next page >
< previous page page_93 next page >
Page 93
kinds of questions you would like to ask in an interview. Third, try to use a tape recorder, or write down when the
respondent is talking to record the interview. If you make a tape, transcribe the key quotes or all of the interview.
Fourth, write out a case study of the bilingual development of that person. Are there particular stages or periods in
the development? Or was the development more smooth with gradual changes?
(2) Interview a mother or father who is learning a second language at the same time as their child. Ask about the
progress each is making. What differences are there? Ask about why there are qualitative and quantitative
differences in progress? What attitudes and motivations do the language learners have? If there are differences of
attitude, try to work out an explanation.
(3) By observation or tape-recording, gather samples of codeswitching. Try to find illustrations of different
purposes of codeswitching. Ask the people in your sample (after completing the observation/recording) how
conscious they are of codeswitching. What are their explanations for codeswitching? What particular purposes for
codeswitching were most found in your samples? How regular was codeswitching in the conversation? What was
the personal history of those you observed or recorded that helps explain such codeswitching?

< previous page page_93 next page >
< previous page page_94 next page >
Page 94
Chapter 6
Second Language Acquisition
The Central Question
Distinctions and Definitions
Ellis’ Framework
Situational Factors
Linguistic Input
Individual Learner Differences
Learner Processes
Second Language Outputs
Socio-Psychological Theories
Lambert’s Model
Gardner’s Socio-Educational Model
Theories of Second Language Acquisition and
The Acculturation Model
Accommodation Theory
Krashen’s Monitor Model
Krashen’s Theory Applied to the Classroom

< previous page page_94 next page >
< previous page page_95 next page >
Page 95
This chapter examines the foundations of second language acquisition research and theory. The topic is a broad
one, and is an important part of the question ‘How do you become bilingual?’ This chapter on second language
acquisition focuses on key summarizing frameworks and theories. These include theories of a psychological, socio-
psychological and a pedagogic nature that attempt to describe holistically major features of language learning in an
interactive and inter-connected model. Such theories provide an overview of a wide variety of research from
different perspectives, and make a relatively comprehensive statement of the current state of understanding of
second language acquisition.
The Central Question
The important issues surrounding second language acquisition can be summarized in one question: ‘Who learns
how much of what language under what conditions?’ Spolsky (1989c: 3). Answers to this question form the
substance of this chapter.
The ‘who learns’ question raises a debate about individual differences. Who learns a second language more easily,
more quickly and with better retention? What part do general ability, aptitude for learning a language, attitudes to
a language, motives and personality play in second language learning? What is the size of each factor’s
contribution? How do these factors inter-relate in a ‘successful’ second language acquisition equation? The word
learns accents the idea of a process, constantly changing. The factors listed above need to be seen as parts in a
moving film and not a static snapshot.
The ‘how much of what language’ part of the question focuses on what is being learnt: oral skills, written skills,
fluency for everyday communication, grammar for test and examination purposes? What is the yardstick for
successful learning? What dialect of language has been learnt? Has another culture been acquired as part of
language acquisition?

< previous page page_95 next page >
< previous page page_96 next page >
Page 96
The ‘under what conditions’ phrase in the question highlights situation and context. What effect do different
learning environments have on the acquisition of a second language: a formal teacher directed classroom, flexible
individual learning, adult classes (e.g. an Ulpan), acquisition in the street and community, correspondence courses,
different forms of bilingual education (e.g. immersion and submersionsee Chapters 10 and 11)? What teaching
strategies are more or less effective? The topic of second language learning in school is considered in Chapter 14
of this book.
Simple, clear guidelines for effective second language acquisition that suits all learners would be much welcomed.
In the second language shop window, there are many alternative theories, many different teaching methods.
Choosing the ‘best’ theory or the ‘best’ teaching method is as dangerous as choosing one garment for all occasions.
The complexity of language acquisition makes the wearing of the same cloth unwise. Different learners, different
environments, different teachers, different props (e.g. computer assisted language learning) make theories and
teaching methods contain ‘part-truths’.
Such ‘part-truths’ will be a resource for teachers, and not a source that solely determines teaching; having
implications rather than direct application; providing propositions rather than prescriptions; insights rather than
edicts. In this spirit, the chapter proceeds by highlighting some important ideas in second language acquisition.
Distinctions and Definitions
Ellis (1985) makes a distinction between three parts to the development of a second language. First, there is the
sequence in second language learning. This refers to the general stages through which children and adults move in
learning a second language. Ellis (1985) argues that, irrespective of the language and irrespective of whether that
language is acquired naturally or formally in the classroom, there is a natural and almost invariant sequence of
development. Moving from simple vocabulary to basic syntax, to the structure and shape of simple sentences, to
complex sentences is a fairly universal sequence in language acquisition. Second, the order in which a language is
learnt may be different from the sequence. The term ‘order’ in this respect refers to specific, detailed features of a
language. For example, the order in which specific grammatical features or situation-specific vocabularies of a
language are acquired may differ from person to person, classroom to classroom. Third, there is the rate of
development of the second language and the level of proficiency achieved. While the sequence of second language
development may be invariant, and while minor variations in the order of development may occur, there may be
major variations in the speed in which a second language is acquired and in the level of final proficiency achieved.
Ellis (1985) suggests that situational factors (who is talking to whom, about what, where and when) considerably
affect the rate of development of the second language. However, situational or contextual factors ‘do not influence
the sequence of development, and affect the order of development only in minor and temporary ways’ (p. 278).
Similarly differences in attitude, motivation, learning strategy and personality may affect the

< previous page page_96 next page >
< previous page page_97 next page >
Page 97
rate at which the second language is acquired and the level of final proficiency, but do not influence sequence nor
the order of the development of the second language. The sequence of development is also not affected by the
learner’s first language. The first language, particularly concerning the degree to which it has developed (see
Cummins’ interdependence theory in Chapter 9) is likely to affect the order of development, the rate of
development and the level of final proficiency.
In becoming functionally bilingual in the classroom and in the community, Cummins’ interdependence theory
suggests that second language acquisition is influenced considerably by the extent to which the first language has
developed. When a first language has developed sufficiently well to cope with decontextualized classroom
learning (see Chapter 9), a second language may be relatively easily acquired. When the first language is less well
developed, or where there is attempted replacement of the first language by the second language (e.g. in the
classroom), the development of the second language may be relatively impeded
Ellis’ (1985) Framework
Ellis’ (1985) summarizing framework of second language acquisition provides an overview of research and theory.
Ellis (1985) suggests that there are five inter-related factors that govern the acquisition of a second language:
situational factors, input, learner differences, learner processes and linguistic output. The relationship between
these is illustrated in the diagram below (adapted from Ellis, 1985):
Situational Factors
Situational factors have been mentioned already in consideration of sequence, order and rate of second language
acquisition. Situational factors will be considered in more detail later in the chapter when examining the theories
of Krashen, Byrne & Giles and Gardner. Situational factors refer to who is talking to whom, the environment of
the interaction,

< previous page page_97 next page >
< previous page page_98 next page >
Page 98
whether it is in a classroom, formal situation or in a naturalistic setting (e.g. a basketball game) and the topic of the
conversation. Previous chapters of this book have suggested that situation has an important effect on language
Linguistic Input
Linguistic input concerns the type of second language input received when listening or reading in a second
language. For example, how do teachers or native speakers adjust their language to the level of second language
learners to make it comprehensible? What kind of differences are there in the input from natural settings compared
with formal classroom settings? In a behaviorist theory of language learning, precise and tight control of input
from the teacher is regarded as very important. The second language has to be presented in small, highly
sequenced doses with plenty of practice and reinforcement. Individual bricks need to be carefully laid in a precise
sequence to build second language skills and habits. In contrast, Chomsky’s mentalist view of language acquisition
regards input as merely activating the learner’s internal language acquisition device. Input from a teacher sets the
wheels in motion rather than creating the wheels of language.
Current research and theory is between the behaviorist and the Chomskyian view point. Learning a language is not
simply putting bricks in place nor pressing the button to start the machine. Efficient and effective second language
learning does not occur purely by the building of stimulus-response links. Nor does second language learning
occur by merely exposing a child or adult to the second language. Providing input which suits the stage of
development of the second language learner becomes important. Discourse analysis has shown that a second
language learner and a native speaker work together to produce purposeful and efficient communication. That is,
one needs to understand the interaction, particularly the negotiation of meaning to understand how input and
output interact. There are strategies and tactics to make conversation appropriate and meaningful. For example,
finding topics of conversation that can be mutually understood, speaking at a slow pace, repeating important
phrases, stressing the key words in a sentence will help the input factor in second language acquisition. A learner
will similarly give signals by verbal and non-verbal communication to indicate understanding, lack of
understanding or to indicate the need to switch topics or level of language.
The input of language learning classrooms varies according to the type of second language learning occurring.
While it is dangerous to over-generalize, the foreign language and the second language classroom may sometimes
focus more on the form of the language (e.g. grammar) rather than on meaning. In contrast, in genuinely bilingual
classrooms, where the second language may be a medium of teaching in the curriculum, the focus may be more on
meaning than on form. While the aim in both situations is to ensure the comprehensible input of the second
language, input is different from intake. The learner receives input of the second language from ‘outside’. Intake
refers to the inner assimilation of that second language. Input does not always result in intake; only when there is
intake does second language acquisition occur.
In terms of what might comprise optimal input in the classroom situation, Wong Fillmore (1982) compared second
language acquisition in United States kindergarten

< previous page page_98 next page >
< previous page page_99 next page >
Page 99
classrooms. She found that effective input differed according to the language composition of different classrooms.
In classrooms where there were large numbers of second language learners, effective input comprised a teacher
directed rather than an open, informal classroom organization. In contrast, where the classes comprised second
language learners and native speaking children, open classroom organization rather than teacher direction seemed
to constitute the optimal learning environment. This may be explained as follows:
(1) In classes where there were large numbers of second language learners, the teacher was most effective by
herself controlling the input. In such classes where there was more open organization, students tended to talk to
each other in their first language, thus not obtaining practice in the second language.
(2) In classes of mixed second language learners and native speaking children, the optimal environment was a
more open organization where second language learners received input from the teacher and from native speaking
children. In such mixed classes, where the teacher tended to control the input, this tended to be at the level of
native speakers and did not necessarily provide comprehensible input for second language learners.
Wong Fillmore (1982) thus shows the importance in second language input of the interaction between teaching
style and the peer composition of different classes.
Individual Learner Differences
An important part of Ellis’ (1985) framework is individual learner differences. It is popularly regarded that the
level of proficiency a child attains in the second language is not only a factor of exposure to various contexts and
to classroom teaching methodology, it is also due to individual differences. For example, the age at which
somebody learns a second language, their aptitude for learning languages, cognitive style, motivation, attitude,
previous knowledge, learning style, learning strategies and personality variables such as anxiety have variously
been thought to influence second language acquisition.
A distinction is important between variables on which there are individual differences (e.g. anxiety level) and
universal capabilities. Universal capabilities are basic, shared features of human beings. An example is Chomsky’s
(1965) idea of an innate, endowed capability for developing grammar. Available to all learners, universal
capabilities are a necessary condition for learning. However, they need to be viewed as a ‘prior assumption’ that
will not explain variations in second language learning among learners.
Variations among individual language learners creates two different questions (Ellis, 1985). First, do individual
differences in age and learning style, for example, result in children and adults following different routes in second
language acquisition? Second, do individual differences affect the speed or rate at which second language
acquisition occurs and the level of final proficiency achieved? People who research on individual differences in
second language acquisition tend to emphasize the importance of individual differences (Wong Fillmore, 1979).
Inbuilt into research designs is often the likelihood of finding significant differences between learners. On the
other hand, second language

< previous page page_99 next page >
< previous page page_100 next page >
Page 100
acquisition theory and research that concentrate on situation, input and process tend to de-emphasize the role of
individual differences.
While it is possible to list the factors which research has connected with more or less effective second language
acquisition, what is unclear is the extent to which those factors affect both the route and the rate of second
language acquisition. For example, there is some evidence to suggest that extroversion and reduced inhibition may
be connected to second language acquisition. In both these cases, the research is not only methodologically weak
(see Ellis, 1985) but also fails to examine, in an overall model, the relative influence of these factors against other
individual differences, situational factors and language input variables.
It is possible to specify a list of factors that appear to be related to second language acquisition. Anxiety, self-
esteem and self-concept, competitiveness in the classroom, anxiety that may be facilitative or interruptive, field
independence as a cognitive style and social skills have each and all been related by research to second language
acquisition. On the other hand, the separate and interacting size of influence of each of these ingredients in the
overall recipe is not clear.
Learner Processes
Another part of Ellis’ (1985) framework is learner processes. It is clearly insufficient to consider second language
acquisition by external input and by second language output. The input that second language learners receive is
sifted, processed and organized. Such processing cannot be easily observed; rather it has to be inferred mostly from
cognitive strategies in language learning. Clarification of the processing strategies of the learner is important for
the teacher to decide what comprises comprehensible input and facilitative situations. One three-fold typology of
learner strategy is by Tarone (1980). First, there are learning strategies, that is ways in which the learner
consciously and subconsciously processes second language input (e.g. memorization). Second, there are production
strategies that comprise attempts to use second language knowledge in an efficient way. Third, a learner has
communication strategies or the means of communicating with others in using the second language when there is a
lack of linguistic proficiency available.
An alternative way of peering into the black box of the mind is that of Chomsky (1965). Chomsky tends to depart
from positing general cognitive strategy devices, claiming instead that there are mental mechanisms that are
specifically linguistic. Chomsky describes this as the language acquisition device that contains an innate blueprint
for a person to acquire a language. Chomsky thus proposed that in-between language input and language
production is a linguistic process that involves the activation of universal principles of grammar with which the
learner is endowed.
Second Language Outputs
Another part of Ellis’ (1985) framework is second language outputs. The language proficiency of any learner at
any one point of time is best seen as:

< previous page page_100 next page >
< previous page page_101 next page >
Page 101
(1) Evolutionary and not fixed. A language competence test as a measure of current language output should ideally
reveal not just the current ceiling, but also the fittings and floors that need to be added and developed.
(2) Variable according to the context where the learner is placed. A learner may appear relatively fluent in a
restaurant or shop situation, yet much less fluent in a business or religious context.
An important contribution to the idea of language output is by Swain (1985, 1986). Swain argues that the
opportunity to engage in meaningful oral exchanges (in the classroom or in the community) is a necessary
component in second language acquisition. In conveying meaning, a person learns about the structure and form of
a language.
A person may understand a language (passive, receptive skills) but, through lack of meaningful practice, speak that
second language less than fluently. People learn to read by reading, and learn to write by writing. To speak, and to
be understood when speaking, requires meaningful and realistic conversations. We learn to speak a second
language when given the opportunity to speak it. Such opportunities may be too infrequent in language
The danger of the classroom is that students may learn to understand a second language (comprehensible input),
but not to produce (comprehensible output). The classroom emphasis has traditionally been on written correctness
and not on spoken language skills. When a student has opportunities to use their spoken language outside the
classroom (e.g. in the street), language skills (e.g. grammar, syntax and communication of meaning) may be
considerably enhanced (Housen & Baetens Beardsmore, 1987; Baetens Beardsmore & Swain, 1985).
Socio-Psychological Theories
Lambert’s (1974) Model
Lambert’s (1974) model is valuable because it combines both the individual and societal elements of bilingualism
and is presented in a diagram below. It is the important societal element of the model that is emphasized in the
following discussion.

< previous page page_101 next page >
< previous page page_102 next page >
Page 102
The model starts with an individual’s attitudes and aptitude towards a language. As will be considered later,
aptitude and attitude are regarded as two major and relatively separate influences on becoming bilingual. For
example, aptitude in learning a second language may be an important factor in second language learning (Skehan,
1986). Similarly, the attitudes of a person towards a language may be important not only in learning that language
but also in maintaining or restoring the language and avoiding attrition. The next part of Lambert’s (1974) model is
motivationthe readiness to engage in language learning or language activity. The third part of the model is a
person’s bilingual proficiency. The factor that comes after bilingual proficiency is self concept.
For Lambert (1974), becoming bilingual or being bilingual has effects on the self esteem and the ego. Having
mastered a second language and being able to interact with a different language group may change one’s self
concept and self esteem. An English monolingual who has learnt Spanish may develop new reference groups and
engage in new cultural activities that affect the self concept. This suggests that bilingualism usually involves
enculturation. Someone who is bicultural or multicultural may have different aspirations, world views, values and
beliefs because of being bilingual or multilingual.
Lambert’s (1974) model finishes with an alternative outcome: additive or subtractive bilingualism. This outcome
can be interpreted both in personal and societal ways. When a second language and culture have been acquired
with little or no pressure to replace or reduce the first language, an additive form of bilingualism may occur.
Positive self concept is likely to relate to additive bilingualism. When the second language and culture are acquired
(e.g. in-migrants) with pressure to replace or demote the first language, a subtractive form of bilingualism may
occur. This may relate to a less positive self concept, loss of cultural identity, with possible alienation and
Additive and subtractive bilingualism have become important concepts in the explanation of research. Lambert’s
(1974) distinction between additive and subtractive bilingualism has been used in two different ways. First,
additive bilingualism is used to refer to positive cognitive outcomes from being bilingual (see Chapter 8).
Subtractive bilingualism hence refers to the negative affective and cognitive effects of bilingualism (e.g. where
both languages are ‘under developed’). Landry, Allard & Théberge (1991) suggest this first use is too narrow, with
a second use of additive and subtractive bilingualism being more appropriate. This wider use of additive and
subtractive bilingualism relates to the enrichment or loss of minority language, culture and ethnolinguistic identity
at a societal level. In additive bilingualism, language minority members are proficient (or becoming proficient) in
both languages, have positive attitudes to the first and second language, with ethnolinguistic vitality in the
language community (Landry, Allard & Thacute;eberge, 1991).
Lambert’s (1974) model contains the basic ingredients that help make up an explanation of individual and societal
bilingualism. It suggests that both individual and sociocultural factors are important in the possession and passage
of bilingualism. Like most models, it is static rather than dynamic. It tends to suggest that there is an easy,
functional flow in relationships between the factors. What it may fail to do is to represent

< previous page page_102 next page >
< previous page page_103 next page >
Page 103
the dynamic, ever changing, often conflicting and politicized path of bilingualism at an individual and at a societal
Gardner’s Socio-Educational Model
Gardner and his colleagues have presented a model that has been well researched and tested. Some of its key parts
will first discussed before the model is presented as a whole. We begin with ability and aptitude for languages, two
concepts that are part of the popular ‘common sense’ explanation of why some learn languages quickly, some not.
Ability and Aptitude
Second language learning in the classroom has often been connected to the general ability of a child (‘intelligence’)
and to a specific language ability usually termed language aptitude. While the idea of a general academic ability or
‘intelligence’ has been criticized (see Baker, 1988), authors such as Oller & Perkins (1978) have argued that the
general factor of intelligence is allied to a general factor of language ability. At its simplest, this means that a more
‘intelligent’ a person is likely to learn a second language more easily. An overview would suggest that general
academic ability can be substantially related to the acquisition of second language in a formal classroom setting.
General ability may positively correlate with test scores on the formal aspects of language learning, (e.g. grammar,
translation, parsing verbs).
However, as Cummins (1984b) has discussed, basic everyday language skills (see Chapter 9) may not be so
related to general academic ability. That is, the skills required for ‘street’ conversation may be less dependent on
general academic ability than the language required to operate in an academic environmentsee Chapter 9). Genesee
(1976) found that IQ was related to second language French reading, grammar and vocabulary but was relatively
unconnected to oral skills. Naturalistic second language acquisition may be less connected to IQ than formal
classroom acquisition. Also, as Ellis (1985) notes, general academic ability may affect the rate and success of
classroom second language acquisition. There is ‘no evidence that intelligence affects the route of acquisition’ (p.
In a similar way, tests of language aptitude have been connected with second language learning. This connection
may be largely due to the similarity of aptitude test and language proficiency test items. Aptitude tests have been
related to second language learning in the classroom rather than second language proficiency in naturalistic,
communicative contexts. Krashen (1981), for example, argues that aptitude relates to formal language learning but
not to the subconscious internal acquisition of language that occurs naturally and spontaneously. Aptitude tests
measure the ability of children to discriminate sounds in a language, to connect sounds with written symbols and
the rote memorization of words of an artificial language. Such items tend to relate to formal, traditional approaches
to language teaching. The items tend to relate less to the modern communicative approach in language teaching.
The concept of aptitude tends to be a popular explanation for failing to acquire a second language. An adult
finding difficulty in learning a specific second language may

< previous page page_103 next page >
< previous page page_104 next page >
Page 104
place the blame on a lack of aptitude for language learning in general. This tends to indicate a belief that there is
something in an individual’s nature that cannot be nurtured. However, the concept of aptitude has recently come
under attack. It is unclear how aptitude is different from general academic ability. If there is a difference between
aptitude and general ability, it is unclear as to its constituent features. If we are unsure of its definition and
structure, it is difficult to know precisely what is being tested in modern language aptitude tests.
While aptitude may affect the speed of second language acquisition in the formal classroom environment, it would
not seem to affect the sequence or order of second language acquisition. There is also no evidence to show that
aptitude affects the route which people take in second language acquisition.
Attitudes and Motivation
Another popular explanation of failure to learn a second language (or of success in learning) is attitudes and
motivation (Baker, 1992). What are the motives for learning a second language? Are the motives economic,
cultural, social, vocational, integrative or for self esteem and self actualization? Reasons for learning a second
(minority or majority) language tend to fall into two major groups:
Group 1:
A Wish to Identify with or Join Another Language Group
Learners sometimes want to affiliate with a different language community. Such learners wished to join in with the
minority or majority language’s cultural activities, find their roots or form friendships. This is termed integrative
Group 2:
Learning a Language for Useful Purposes
The second reason is utilitarian in nature. Learners may acquire a second language to find a job and earn money,
further career prospects, pass exams, help fulfill the demands of their job, or assist their children in bilingual
schooling. This is termed instrumental motivation.
Considerable research on this area has been conducted by Gardner and associates (see Gardner, 1985). Gardner
argues that integrative and instrumental attitudes are independent of ‘intelligence’ and aptitude. Integrative
motivation may be particularly strong in an additive bilingual environment. This is more fully considered in the
next chapter.
Much of the research in this area, but not all, links integrative motivation rather than instrumental motivation with
the greater likelihood of achieving proficiency in the second language. Gardner & Lambert (1972) originally
considered that integrative motivation was more powerful in language learning than instrumental motivation. The
reason was that integrative motivation concerns personal relationships that may be long lasting. On the other hand,
instrumental motivation may be purely self-oriented and short term. When employment has been obtained or
financial gain has accrued, instrumental motivation may wane. An integrative motive was thought to be a more
sustained motive than an instrumental motive due to the relative endurance of personal relationships.

< previous page page_104 next page >
< previous page page_105 next page >
Page 105
Research has subsequently suggested that there may be occasions when the instrumental motive is stronger than
the integrative motive in learning a language. Lukmani (1972) found that Bombay female school students gave
instrumental rather than integrative reasons for learning English. In the research of Yatim (1988), the language
motivations of student teachers in Malaysia appeared to combine instrumental and integrative motives into an
integrated entity. A person’s motives may be a subtle mix of instrumental and integrative motives, without clear
discrimination between the two.
The research on language attitudes and motivations is summarized by Gardner (1985) and Baker (1988, 1992).
Research relates such motivation not only to the desire to learn a language but also to predicting language
retention and language loss in individuals over time. Another important strand in this research examines
instrumental and integrative motivation within the classroom. One study is by Gliksman (1976, 1981). He
classified 1416 year olds by their level of integrative motivation. Gliksman (1976) also systematically observed the
number of times the students:
(1) volunteered information by raising a hand;
(2) were asked by teachers without volunteering;
(3) answered correctly or incorrectly;
(4) asked questions;
(5) received positive, negative or no feedback from the teacher.
Gliksman (1976, 1981) found that students with a higher integrative motivation volunteered information more
frequently, gave correct answers and received more positive feedback from the teacher than did less integratively
motivated students. The two groups did not differ significantly on the number of questions they asked in class.
Integratively motivated students were rated as more interested in their lessons. Gliksman found that such
differences were consistent across a whole term and were not sporadic nor temporary.
The explanation lies in the functions that a second language plays in a particular society. Factors such as
employment and career development can be stronger than integrative motivation. The social context will be one
determinant of which kind of motivation is more powerful, and this does not preclude both motivations being
equally and strongly operative in a particular context. It is clear that motivation is an important factor in second
language acquisition, affecting the speed and final proficiency of the second language. It is unlikely to affect the
sequence or order of acquisition.
In summary: the research of Gardner and colleagues since the early 1970s suggests that attitudes to the second
language and motivation to learn a second language are crucial additional ingredients in the language learning
recipe. Having the ability and aptitude without the motivation and favorable attitude would tend to result in lower
achievement than having both aptitude and motivation.
To represent in an integrated fashion his own and others’ research on second language learning, Gardner (1979,
1983, 1985) offers a four stage model. This is presented in the diagram on the following page.

< previous page page_105 next page >
< previous page page_106 next page >
Page 106
There are four stages to Gardner’s (1979, 1983, 1985) model. First, the model starts with the social and cultural
background of language learning. In this sense, Gardner starts where Lambert finishes. Children may be influenced
by the beliefs, values and culture of the community in which they are placed. For many people living in England,
for example, the belief is that the ‘universal’ English language is all that is required; bilingualism is unnecessary. In
other communities of Europe, bilingualism and biculturalism reflect the values of the community. In Gardner’s
model, social and cultural background refers not only to the wider community but also to the influence of the
home, neighbors and friends. This influence is further explored in the model of Hamers & Blanc (1982, 1983) and
Siguan & Mackey (1987).
The second stage of Gardner’s (1979, 1983, 1985) model is termed individual differences. This comprises four
major variables: intelligence, language aptitude, motivation and situational anxiety. Attitudes and personality are
taken to be subsumed within this section. Thus Gardner suggests that the degree of intelligence of an individual,
their aptitude or talent for language learning, their instrumental and integrative motives and the anxiety they feel in
language learning will all affect the outcomes of language learning.
The third stage of Gardner’s (1979, 1983, 1985) model concerns the context or environment where language is
acquired. He makes a distinction between formal and informal environments in language learning. An example of
a formal context is the classroom that explicitly aims to teach a child a second language by a defined teaching
method and various classroom materials and resources. A language laboratory, drill and practice, computer
assisted language learning, audio visual methods and translations and grammar exercises are all examples of a
formal and directed approach to language learning. An informal language learning context or experience is when
language learning is more incidental, accidental or uncontrived. For example, a person watches a Spanish film not
primarily to widen vocabulary, but for the entertainment value of that film. In

< previous page page_106 next page >
< previous page page_107 next page >
Page 107
this example, a person’s Spanish vocabulary may be extended but it is an unintended outcome. Or, talking to a
friend of relative in a second language may not intentionally be to practice the second language but to foster good
relationships. While the person is practicing their second language skills this is an incidental outcome, not the
prime reason for such conversation. There are also examples when the formal and informal experience merge. In
the classroom there will be episodes where the teacher is having a friendly chat or giving simple instructions that
are not primarily for language learning but have that effect.
The fourth and final stage of Gardner’s model (1979, 1983, 1985) has two outcomes. One outcome refers to
bilingual proficiency. The second outcome refers to non-linguistic outcomes such as change in attitudes, self
concept, cultural values and beliefs. The placing of attitudes in the second and the fourth stage suggests that the
model should be conceived as something that is not static but cyclical. Attitudes are not only ingredients in the
language learning situation. Attitudes are also products or outcomes of language learning. That is, the learning of a
second language and the act of becoming bilingual, may change attitudes. As the diagram above indicates by the
dotted arrow from bilingual proficiency to the second stage, competence in a second language may in turn affect
intelligence, motivation and situational anxiety levels.
The value of Gardner’s model is that it is not only a summary of existing research in the social psychology of
language learning. It also has been directly and formally tested as a complete model (Gardner, 1983; Gardner,
Lalonde & Pierson, 1983; Lalonde, 1982) and found to fit collected data. This is different from most theories of
bilingualism where the theory attempts to summarize previous research but is not subsequently directly and
rigorously tested.
One limitation of Gardner’s model is that it does not include the sociopolitical dimension that often surrounds
routes to bilingual proficiency. As will be considered in Chapter 19, a full understanding of language acquisition,
language change and bilingual education (at individual and societal levels) requires a political dimension. For
example, the preference in a region for assimilation of minorities, integration or cultural pluralism, influences
language policies, provision and practices surrounding bilinguals.
There are other limitations of models. Models are valuable for summarizing and providing clarity but tend to imply
descriptions rather than deep explanations. They suggest a straightforward, well functioning, clearly ordered
system of actions and reactions, ingredients and events. An individual’s internal conflicts, the complexity of
competing multiple pressures and motives, the many and varied individual recipes of success and failure, the many
individual exceptions, the complexity of contexts and effects of immediate and long-standing situations are not
represented in a model.

< previous page page_107 next page >
< previous page page_108 next page >
Page 108
Theories of Second Language Acquisition and Learning
The topics of bilingualism and second language acquisition are like brother and sister. While both have separate
sets of friends and mother and father figures, they come from the same family. Becoming bilingual often involves
second language acquisition, either achieved formally (e.g. in the classroom) or informally (naturally, for example,
in the street and playground, via television and radio). At the same time, research into bilingualism feeds into the
wide topic of second language acquisition (e.g. the Development of Bilingual Proficiency project; see Harley et
al., 1987, 1990).
There are multiple theories of second language acquisition. In major reviews, McLaughlin (1987) examines five
important second language theories while Ellis (1985) considers seven second language theories. The essence of
second language theories is to describe the individual and contextual conditions for efficient second language
learning to occur.
The Acculturation Model of Second Language Learning
John Schumann (1978) proposes an acculturation model of second language acquisition whose essential element is
the second language learner adapting to a new culture. The model starts with the idea that language is one aspect
of culture, and the relationship between the language community of the learner and the second language
community is important in second language acquisition. The basic premise of the model is ‘the degree to which a
learner acculturates to the target language group will control the degree to which he acquires the second language’
(Schumann, 1978: 34). Schumann (1978) portrays the various factors that are important in ‘good’ language
learning. Expressed in group rather than individual terms, these facilitating social factors comprise:
(1) The target language group and the second language learner group are self-perceived as relatively socially
equal. The greater the equality distance (e.g. domination,subordination) the less the chance of language learning.
(2) The target language and second language learner groups both desire assimilation of the learner’s social group.
(3) Both groups expect the second language group to share social facilities as operated by the target language
(4) The second language learner group is small, not very cohesive and can be assimilated into the target language
(5) The extent to which the second language learner group’s culture is congruent and similar to that of the target
language group thus assisting assimilation.
(6) The extent to which both groups have positive attitudes and expectations of each other; and
(7) The extent to which the second language learner group expects to stay with the target language group for a
longer rather than a short period (e.g. in-migrants).
These are the social factors that for Schumann (1978) determine the probability of a second language individual or
group receiving the target language group. Schumann

< previous page page_108 next page >
< previous page page_109 next page >
Page 109
(1978) also lists psychological factors that are important in second language learning. These include: possible
language confusion when using the second language (language shock); the feeling of stress, anxiety or
disorientation because of the differences between the learner’s culture and the target language culture (culture
shock); the degree of motivation in learning a language, and the degree of inhibition or self consciousness
adolescent learners particularly may have in language learning. Schumann’s (1978) factors provide dimensions that
may determine the amount of contact a language learner will have with the target language. The amount of contact
is defined both by social, external factors and by individual factors. When social and/or psychological distances
are large, the learner may fail to progress very far in learning a second language. When the social and
psychological factors are positive, second language acquisition may occur relatively painlessly. When
psychological and social distances are great, then, at an individual and societal level, pidginization (the
development of a simplified form of a language) may occur. For Schumann (1978), pidginization is the
characteristic of early second language acquisition. Also, when conditions of social and psychological distance are
great, pidginization will occur at a societal level. When pidginization persists over a period, that language may
decline, even die.
For Schumann (1978), language has three broad functions: a communication function, an integrative function and
an expressive function. That is, language aids the transmission of information, aids affiliation and belonging to a
particular social group, and allows the display of individual feelings, ideas and personality. Schumann (1978)
argues that second language learners will initially use their second language for communication. Second language
learners who develop in that language will then seek the use that language to affiliate to a social group. Some
learners, but not all may achieve the expressive use of that second language.
Schumann’s (1978) Acculturation model, extended by Andersen’s (1983) Nativization model, is one way of
understanding the politics and power of language in its societal contexts. It provides a valuable explanation of why
some children with aptitude and ability fail to learn or use a second language. It does not specify how such
language acquisition occurs via internalized learning processes. The information processing approach to second
language learning is not included in this model. Absent from this account of second language learning is: the
interaction between a particular defined context and the learner; the changing, variable nature of attitudes and
motivation; whether attitude is a cause of language learning or an outcome of language learning, or is both a cause
and an effect (Gardner, 1985); the shifting power relations and distance between groups; the role of individual and
group negotiation over crucial influencing factors; and the difficulty of testing the theory. The model may also be
more appropriate to learning a second language naturally rather than to formal language learning within a
Accommodation Theory
Accommodation theory derives from Giles and colleagues (e.g. Giles & Byrne, 1982). Like Schumann (1978),
Giles & Byrne’s theory seeks to explain second language acquisition in a group or intergroup situation. For Giles
& Byrne (1982), the important

< previous page page_109 next page >
< previous page page_110 next page >
Page 110
factor is the perceived social difference between the ingroup (the language learner’s social group) and the outgroup
(the target language community). The relationships between the ingroup and outgroup are seen as both fluid and
constantly negotiated. There is a tendency in Schumann’s Acculturation model for social and psychological
distances to be seen as static or changing relatively slowly over time. For Giles & Byrne (1982), relationships
between ingroup and outgroup are dynamic and constantly changing. One way of portraying Giles & Byrne’s
(1982) model is to profile a person from a subordinate group who is likely to acquire the language of the dominant
The learner is likely to show the following characteristics (simplified from Giles & Coupland, 1991):
(1) Have a relatively weak identification with their own ethnic group. That is, such learners do not see themselves
as purely a member of their minority language group separate from the dominant language group. Alternatively,
their first language is not important to membership of their ethnic group.
(2) Do not regard their ethnic group as inferior to the dominant group. A good language learner makes ‘quiescent’
comparisons between their ethnic group and the dominant group, or is not concerned about a difference of status.
(3) Perceive their ethnic group as having low vitality compared with the dominant group. Giles & Byrne (1982)
talk of a perception of ethnolinguistic vitality which includes (see Chapter 4): (1) the economic, historic, social,
political and language status of the ethnic group; (2) size and distribution of an ethnic group, mixed marriages,
amount of in-migration and out-migration; and (3) institutional support for the ethnic group (e.g. mass-media,
education, religion, industry, services, culture and government).
(4) See their ethnic group boundaries as ‘soft and open’ and not ‘hard and closed’.
(5) Hold adequate status within their ethnic group (e.g. in terms of employment, gender, power and religion).
Thus for Giles & Byrne (1982), a person less likely to acquire a second language may have: a strong identification
with their own group, makes ‘insecure’ comparisons with the outgroup; regard their own language community as
having high vitality, good institutional support, being sizable and stable and of high status; perceive the boundaries
between their own and the second language group as separate and rigid, and have inadequate status within their
first language group. Further factors are considered in Giles & Coupland (1991).
Like the Acculturation model, Accommodation Theory does not explain the internal mechanisms of how a child
acquires a second language. It is essentially a socio-psychological model rather than a cognitive-processing model
of second language acquisition. Tollefson (1991) also criticizes Accommodation Theory for its ahistorical analysis
and failure to account for domination and coercion in language shift.
Glyn Williams (1992) is also critical in that Giles & Byrne may wrongly assume that language development is
gradual and cumulative, of a consensus nature, concerned with order and cohesion with a consequent limited
discussion of power, struggle and conflict.

< previous page page_110 next page >
< previous page page_111 next page >
Page 111
Williams (1992: 224) argues that ‘ethnolinguistic vitality is more of a typological construct than a refined concept.
It suffers from a lack of integration of its constituent parts, as a consequence of which it is difficult to understand
how the subjective and objective dimensions relate’. And of inter-group relationships, Williams (1992: 224)
critically comments that the argument is merely ‘couched in terms of individuals, rationally conditioned to behave
in terms of optimization, with the quest for status and the associated positive identity being the motivating force of
behavior’. This tends to play down conflict and power, thereby not expressing the anger, discrimination and
frustration felt by language minority groups and their members.
One strength of the theory is that is takes into account ethnic identity in language learning, an important
determining factor for many children and adults in second language acquisition.
Krashen’s Monitor Model
The Monitor model of Stephen Krashen (1977, 1981, 1982, 1985) is probably the most widely cited of theories of
second language acquisition. While there are other important theories not considered in this chapter (e.g. Discourse
Theory proposed by Hatch (1978), Ellis’s (1984) Variable Competence model, Chomsky’s (1965) Universal
Hypothesis, Wong Fillmore’s (1991b) Socio-Linguistic Cognitive Theory, and the Neurofunctional Theory (e.g.
Lamendella, 1979)), it is Krashen’s theory that has often dominated education research and education debate in
second language acquisition. Krashen’s Monitor model comprises five central hypothesis plus other variables that
need considering in second language acquisition. The five hypothesis are as follows.
The Acquisition Learning Hypothesis
An important initial distinction is between second language that is acquired naturally and second language that is
acquired in formal situations. Thus Krashen distinguishes between acquisition and learning. Acquisition is a
subconscious process that results from informal, natural communication between people where language is a
means and not a focus nor an end in itself. Learning occurs in a more formal situation where the overt properties of
a language are taught. Language learning has traditionally involved grammar, vocabulary learning and the teaching
of other formal linguistic properties. Learning is a conscious process that enables a learner to ‘know about’ the
second language. With second language learning, the analysis and correction of errors is formally and explicitly
Acquisition and learning are not defined by ‘where’ a second language occurs. Formal learning can occur in the
street when a person asks questions about correct grammar, mistakes and difficulties. The key factor for Krashen is
a distinction between language judgments based on rules and based on feelings. Conscious thinking about
language rules is said to occur in second language learning; unconscious feelings about what is correct and
appropriate occurs in language acquisition.
The distinction between acquisition and learning is a central idea in education theory. It is a distinction between
deductive and inductive approaches, classroom and naturalistic learning, formal and informal language learning.
Advocates of the commu-

< previous page page_111 next page >
< previous page page_112 next page >
Page 112
nicative method in language teaching emphasize the importance of informal acquisition of authentic language
inside the classroom. In educational terms, this is also referred to as a distinction between a formal training
approach and apprenticeship. Formal training attempts to provide learning experiences in which languages will be
learnt efficiently. Apprenticeship involves acquiring skills naturally while working on a particular task.
The Natural Order Hypothesis
This hypothesis suggests that grammatical structures are acquired in a predictable order for both children and
adults, irrespective of the language being learnt. When a learner engages in natural communication, then the
standard order will occur. This hypothesis has been criticized (e.g. McLaughlin, 1987). Research on morphemes
and on the development of specific grammatical forms does not support a ‘strong’ version of the hypothesis.
Variations between different people and a lack of supportive evidence suggest that only a ‘weak’ version of the
natural order hypothesis is tenable.
The Monitor Hypothesis
The Monitor is an editing device that may operate before language performance. Utterances may be modified by
being acted upon by the Monitor of learnt knowledge. Such editing may occur before the natural output of speech;
it may occur after the output via a correcting device. Krashen suggests that monitoring occurs when there is
sufficient time, when there is pressure to communicate correctly and not just convey meaning, and when the
appropriate rules of speech are known. Examples include knowing the correct tense to use, when to use the third or
first person and rules about plurals. This hypothesis has been criticized for being untestable and for a lack of
supportive research evidence.
The Input Hypothesis
To explain how language acquisition occurs, Krashen proposes that when learners are exposed to grammatical
features a little beyond their current level, those features are ‘acquired’. Krashen emphasizes that ‘acquisition’ is the
result of comprehensible language input and not of language production. Input is made comprehensible because of
the help provided by the context. If the language student receives understandable input, language structures will
be, according to Krashen, naturally acquired. For Krashen, the ability to communicate in a second language
’emerges’ rather than is directly put in place by teaching. Second language is said to be caused by the process of
understanding second language input.
The Affective Filter Hypothesis
An Affective Filter was proposed by Dulay & Burt (1977) with the idea that there is a filter that determines how
much a person learns in a formal or informal language setting. The filter comprises affective factors such as
attitudes to language, motivation, self-confidence and anxiety. Thus learners with favorable attitudes and self-
confidence may have ‘a low filter’ with consequent efficient second language learning. Those with unfavorable
attitudes and/or high anxiety have ‘high filters’ and so the input of second language learning may be blocked or
impeded. The affective filter proposed by Krashen influences

< previous page page_112 next page >
< previous page page_113 next page >
Page 113
the rate of development in second language learning and the level of success in becoming bilingual.
Before considering how Krashen’s Monitor model relates to classroom language learning and teaching, it is
important to note various criticisms that have been made of this model. Apart from the possibility of the informal
acquisition of language occurring in a classroom and language being learnt in informal discussion, the criticism
has been that the acquisition-learning hypothesis cannot be tested empirically. Because acquisition is more
subconscious and learning is relatively more conscious, it is difficult even impossible to test the hypothesis
empirically and comprehensively. It may also be the case that acquisition and learning are not separate. It may be
that acquired knowledge can become learnt knowledge. Once learnt knowledge is practiced it may reach a level of
automatization that equates to acquired knowledge, being available in spontaneous ‘unconscious’ conversation. As
Larsen-Freeman (1983) comments, Krashen does not explain the cognitive processes that underlie acquisition and
learning. Thus the Monitor model is a ‘black box’ theory of language acquisition. It does not specify what goes on
in cognitive processes to explain second language acquisition.
Another line of criticism concerns the monitoring hypothesis. In reality, is there a distinction between rule
application (as in the Monitoring device) and having a subconscious feel for what is right and wrong in a
communication situation? Is there an underlying critical faculty (Morrison & Low, 1983) that makes people aware
of the correctness or incorrectness of their language communication? Krashen’s theory also tends to fail to explain
variability between individuals in language learning. Further criticisms may be found in McLaughlin (1978, 1987),
Morrison & Low (1983) and Ellis (1985).
Krashen’s Theory Applied to the Classroom
Krashen (1982) and Krashen & Terrell (1983) have applied the Monitor model to language teaching and classroom
language learning. The classroom applications that follow from the Monitor model are as follows. First, Krashen &
Terrell (1983) argue that the goal of language teaching must be to supply understandable input in order for the
child or adult to acquire language easily. A good teacher therefore is someone who continuously delivers at a level
understandable by the second language speaker. Just as father and mother talk (motherese) help the young child to
acquire the first language by a simplified and comprehensible language (and non verbal language), so an effective
teacher is said to facilitate second language learning by ensuring a close match between the level of delivery and
the level that is understandable. Second, teaching must prepare the learner for real life communication situations.
The classroom needs to provide conversational confidence so that, when in the outside world, the student can both
linguistically cope and continue language learning.
Conversational competence also means learning strategies to get native speakers to explain their meaning when it
is not initially apparent, devices for changing topics and for facilitating understandable communication with the
native speaker.

< previous page page_113 next page >
< previous page page_114 next page >
Page 114
Third, Krashen & Terrell (1983) suggest that teachers must ensure that learners do not become anxious or
defensive in language learning. This relates to the Affective Filter hypothesis. The confidence of a language
learner must be encouraged in a language acquisition process. When a learner is relaxed, confident and not
anxious, then the input of the classroom situation will be more efficient and effective. If teachers insist on children
conversing before they feel comfortable in doing so, or a teacher constantly corrects errors and makes negative
remarks, the learner may feel inhibited in learning. Fourth, formal grammar teaching is of limited value because it
contributes to learning rather than acquisition. Only simple rules should be learnt. Complex rules will not be used
consciously or unconsciously by the language learner. Therefore, there appears little to be gained from formally
teaching the rules of a second language. Fifth, errors should not be corrected when acquisition is occurring. They
may be corrected when the goal is formal learning. Error correction is valuable when learning simple rules but
may have negative effects in terms of anxiety and inhibition.
For Krashen & Terrell (1983), a ‘Natural Approach’ is required in language teaching. The Natural Approach is very
different from traditional grammar teaching and language laboratory types of approach. Its main tenets are as
follows: communicative skills should be the aim of the good language classroom; comprehension of language
should precede production (listening should precede speaking); speaking and then writing will emerge when the
language learner is ready and should not be forced; acquisition rather than formal learning is central in good
language learning; and the affective filter needs to be kept low (that is, favorable attitudes, positive motivation and
low anxiety are important in language learning).
This chapter has discussed central ideas in the acquisition of a second language. The overarching question is ‘Who
learns how much of what language under what conditions?’ This brings into play individual differences (e.g. in
ability and attitude), the skills and competences being learnt (e.g. grammatical accuracy, conversational fluency
and literacy), contextual and situational factors, linguistic inputs and learner processes (e.g. code switching).
To the highways and byways of bilingualism needs to be added the geography of the journey. The journey is
affected by the psychology of the individual, the environment and conditions of language travel, the political and
cultural weather surrounding acquisition, fellow travelers and map makers. Becoming bilingual is a linguistic,
social and a psychological event. The chapter has sought to portray the main features of different routes, while
indicating that no two journeys will be the same.
A variety of theories of second language acquisition have been presented. Each provides some clues as to how a
second language is more efficiently and effectively acquired or learnt. The individual and social aspects are
necessarily fused; the linguistic, sociological and psychological cannot be realistically separated. General theories
have been considered with Krashen’s theory being directly related to classroom practice. While

< previous page page_114 next page >
< previous page page_115 next page >
Page 115
Krashen’s model has been criticized, it has provided teachers with a set of general guidelines grounded in second
language research and theory. Whether such theories derived from psychology or linguistics have sufficient
understanding of complex classrooms and sufficient power to provide a technology of teaching is doubtful. Such
theories give ideas and insights, they do not always provide answers that are translatable into comprehensive
procedures or recipes for classroom practice. ‘Best’ or ‘perfect’ recipes are as unlikely in teaching as they are in
cordon bleu cooking.
Suggested Further Reading
ELLIS, R., 1985, Understanding Second Language Acquisition. Oxford: Oxford University Press.
GARDNER, R.C., 1985, Social Psychology and Second Language Learning. London: Edward Arnold.
McLAUGHLIN, B., 1987, Theories of Second-Language Learning. London: Edward Arnold.
SPOLSKY, B., 1989, Conditions for Second Language Learning. Oxford: Oxford University Press.
Review and Study Questions
(1) Write short revision notes on the following:
(i) The difference between the sequence, order and rate of language acquisition.
(ii) The nature of language aptitude and language attitude.
(2) Make a list of the factors that research and theory has located as influential in second language acquisition.
(3) Describe an idealized version of a ‘good’ and ‘poor’ second language learner from the contents of this chapter.
(4) From the various theories you have read in this chapter, make a list of those elements relevant to particular
schools and classrooms known to you.
(5) Take one of the theories presented in this chapter and, by further reading, write about why you think that theory
is important and valuable.
Study Activities
(1) Locate a teacher or a parent or a friend who has learnt a second language in school or in adult life. Ask them
about the importance of school, classroom and learning factors in their second language acquisition. Discuss with
them how they see their current ability, attitudes and usage of their second language.
(2) Take Gardner’s or Lambert’s model and show how the model relates to one person well known to you. Show
how the elements of the model are translatable into the language life of an individual of your choice.

< previous page page_115 next page >
< previous page page_116 next page >
Page 116
Chapter 7
Bilingualism and Intelligence
Bilingualism and ‘Intelligence’
The Period of Detrimental Effects
The Period of Neutral Effects
The Period of Additive Effects
Bilingualism and the Brain

< previous page page_116 next page >
< previous page page_117 next page >
Page 117
There is one piece of advice that parents sometimes receive from well-meaning teachers, doctors, speech
therapists, school psychologists and other professionals. Don’t raise your child bilingually or problems will result.
Predicted problems range from bilingualism as a burden on the brain, mental confusion, inhibition of the
acquisition of the majority language, even split personality. Parents and teachers are sometimes advised to use only
one language with individual children. When children persist in speaking two languages in school, having their
mouths washed with soap and water (Isaacs, 1976) and being beaten with a cane for speaking Welsh (the Welsh
‘Not’) have been offered as a remedy.
A quotation from a Professor at Cambridge University portrays this deficit viewpoint:
‘If it were possible for a child to live in two languages at once equally well, so much the worse. His
intellectual and spiritual growth would not thereby be doubled, but halved. Unity of mind and character
would have great difficulty in asserting itself in such circumstances’. (Laurie, 1890: 15).
The anxiety that two languages may have a negative effect on an individual’s thinking skills tends to be expressed
in two different ways. First, some tend to believe that the more someone learns and uses a second language, the
less skill a person will have in their first language. Rather like weighing scales or a balance, the more one
increases, the more the other decreases. This issue is addressed in Chapters 8 and 9 of this book. Second, concern
is sometimes expressed that the ability to speak two languages may be at the cost of efficiency in thinking. The
intuitive belief is sometimes that two languages residing inside the thinking quarters will mean less room to store
other areas of learning. In comparison, the monolingual is pictured as having one language in residence and
therefore maximal storage space for other information.
Does the ownership of two languages interfere with efficient thinking? Do monolinguals have more effective
thinking quarters? Is a bilingual less intelligent than a monolingual due to a dual language system? This chapter
and the next examines these

< previous page page_117 next page >
< previous page page_118 next page >
Page 118
typically negatively phrased questions and evaluates the evidence on bilingualism and thinking. We start by
considering the relationship between intelligence and bilingualism. ‘Intelligence’ has been a major concept in
psychology and often related to bilingualism. It is also a term used by members of the public in phrasing questions
about bilingualism.
Bilingualism and ‘Intelligence’
The Period of Detrimental Effects
From the early nineteenth century to approximately the 1960s, the dominant belief amongst academics was that
bilingualism had a detrimental effect on thinking. For example, the quote from Professor Laurie (1890) suggested
that a bilingual’s intellectual growth would not be doubled by being bilingual. Rather both intellectual and spiritual
growth would be halved. This view of Laurie (1890) tends to parallel a view commonly held amongst the British
and US populace right through the twentieth century: that bilingualism has disadvantages rather than advantages in
terms of thinking.
The early research on bilingualism and cognition tended to confirm this negative viewpoint, finding that
monolinguals were superior to bilinguals on mental tests (Darcy, 1953; Náñez, Padilla & Máez, 1992). Research
up to the 1960s looked at this issue through one concept’intelligence’. A typical piece of research gave bilinguals
and monolinguals an ‘intelligence’ test. When bilinguals and monolinguals were compared on their IQ scores,
particularly on verbal IQ, the usual result was that bilinguals were behind monolinguals. An example of this early
research is by a Welsh researcher, D.J. Saer (1923). He gathered a sample of 1,400 children aged seven to fourteen
from bilingual and monolingual backgrounds. A 10 point difference in IQ was found between bilinguals and
monolingual English speakers from the rural areas of Wales.
Saer (1923) concluded that bilinguals were mentally confused and at a disadvantage in thinking compared with
monolinguals. Further research by Saer, Smith & Hughes (1924) suggested that University student monolinguals
were superior to bilinguals: ‘the difference in mental ability as revealed by intelligence tests is of a permanent
nature since it persists in students throughout their University career’ (p. 53).
While it is possible that situations exist where bilinguals will perform on such tests at a lower level than
monolinguals (this is considered in Chapters 8 and 9), the early research that pointed to detrimental effects has a
series of weaknesses. Such weaknesses tend to invalidate the research in terms of individual studies and
cumulatively across studies. These limitations may be listed as follows.
The concept of ‘intelligence’ and the use of intelligence tests is controversial and hotly debated. One part of the
controversy lies in the problems of defining and measuring intelligence. The underlying questions are: what is
intelligence and who is intelligent? A thief who cracks a bank vault? A famous football coach? Someone poor who
becomes a billionaire? Don Juan? A chairperson who manipulates the members of a board? Is there social
intelligence, musical intelligence, military intelligence, marketing intelligence,

< previous page page_118 next page >
< previous page page_119 next page >
Page 119
motoring intelligence, political intelligence? Are all, or indeed any of these forms of intelligence measured by a
simple pencil and paper IQ test which requires a single, acceptable, correct solution to each question? What is
intelligent behavior or not requires a subjective value judgment as to the kind of behavior and the kind of person
regarded as of more worth.
There are three related controversies to that of the definition of ‘intelligence’:
There is a fierce debate about the relative effects of heredity and environment on the development of intelligence.
A strong hereditarian viewpoint tends to argue that intelligence is relatively fixed and unlikely to be affected by
becoming bilingual. An environmental view of the origins of intelligence may be more appealing to supporters of
bilingualism. The environmental view holds that intelligence is not fixed or static, but modifiable by experience
(e.g. family, education, culture and sub-culture). The ‘extra’ experience of two languages may thus contribute to
the nature and growth of intelligence.
Does intelligence comprise one unitary factor, or can intelligence be divided into a wide variety of factors or
components? Is there one all-embracing general factor of intelligence (labeled ‘g’), or is Guilford’s (1982) 150
factor model of intelligence more valid? A multi-factor view of intelligence is more likely to reveal differences
between monolinguals and bilinguals.
IQ tests tend to relate to a middle class, white, Western view of intelligence. The cultural boundedness or relativity
of IQ tests suggests that cross-cultural generalizations are dangerous and limited. The relationship between
bilingualism and a non-Western view of intelligence may be fruitful and revealing.
To summarize: in the relationship between intelligence and bilingualism, the first problem is that IQ tests only
measure a minute sample of everyday ‘intelligence’. Therefore, whatever pattern is found between IQ tests and
bilingualism refers to ‘pencil and paper’ intelligence. Research does not investigate the relationship between dual
language ownership and all the components that might go under the wide heading of ‘intelligence’. To use an
analogy: we cannot fully portray a football or basketball match simply from the number of passes. Similarly, with
bilingualism and intelligence, the whole game has not been studied, just one small statistic.
Language of Testing
The second problem is the language of the IQ test given to bilinguals. It is preferable to test the IQ of bilinguals in
their stronger language or in both languages. In the early research, many verbal IQ tests were administered in
English only. This tended to be to the disadvantage of bilinguals in that they were tested in their weaker language
and thus under-performed in the IQ test.
The early research tended to use simple averages when comparing monolingual and bilinguals groups. Statistical
tests were often not performed to see whether the difference between the average scores was real or due to chance
factors. Thus, for example, when

< previous page page_119 next page >
< previous page page_120 next page >
Page 120
W.R. Jones (1966) re-analyzed Saer’s (1923) research, he found that there was no statistically significant
difference between the monolingual and bilingual groups.
As has been shown in Chapter 1, the classification of people into bilingual and monolingual groups is fraught with
difficulty. It is too simplistic to place people into a monolingual or a bilingual group. We need to ask what
language competences are being used for classification? Are all four basic language abilities being used? What is
the degree of fluency in each language? Were bilinguals classified by their use of languages (functional
bilingualism) or by their ability in language? As Chapter 1 revealed, who is or who is not bilingual is a complex
issue. The earlier research on bilingualism and cognition tended to regard classification as non-problematic. This
means that the research results are simplistic and ambiguous, having classified bilinguals in an insensitive and
imprecise manner.
A fifth problem concerns sampling and the generalization of research results to the population of bilinguals. With
all research, the findings should be restricted to the population that the sample exactly represents. In particular,
research using a non-random sample of a population, merely a convenience sample, should theoretically have no
generalization beyond that sample. Much of the research on bilingualism and cognition is based on convenience
samples. Thus research on 11 year olds cannot be generalized to other age groups. Findings in the US cannot be
generalized to bilinguals in the rest of the world. In much of the early research on bilingualism and cognition, the
sampling is both small and inadequate making generalization dangerous.
The language environment of the research sample needs to be considered. This relates to the notion of subtractive
and additive environments. Negative, detrimental cognitive findings may be more associated with minority
language groups in subtractive environments. Subtractive environments are where the child’s first language is in
danger of being replaced by a more prestigious second language. Where bilingualism has high prestige in an
additive environment, a different pattern of results may be more likely.
Matched Groups
The final problem is particularly important. To compare a group of bilingual children with monolinguals on IQ, or
on any other measure of cognitive ability, requires that the two groups be equal in all other respects. The only
difference between the two groups should be in their bilingualism and monolingualism. If such control does not
occur, then the results of the research may be due to the other factor or factors on which the groups differ (rather
than their monolingualism or bilingualism). Take the example of a monolingual group being mostly of higher
socioeconomic status, and the bilingual group being mostly of a lower socioeconomic status. A result (e.g.
showing monolinguals to be ahead of bilinguals) may be due to social class rather than, or as well as, bilingualism.
The great majority of researches on bilingualism and ‘intelligence’ failed to match the groups on

< previous page page_120 next page >
< previous page page_121 next page >
Page 121
other factors that might explain the results. It is necessary to match the groups on variables such as sociocultural
class, gender, age, type of school attended and urban/rural and subtractive/additive environments.
The period of detrimental effects research lasted from approximately the 1920s to the 1960s. While the dominant
result was that bilinguals were inferior to monolinguals, particularly on verbal IQ, these early researches share
many serious methodological weaknesses. Singly and cumulatively, the early research on bilingualism and IQ has
so many limitations and methodological flaws that its conclusion of detrimental effects cannot be accepted. While
it is possible that, in some contexts, bilinguals may have cognitive disadvantages (see Chapter 9), the early
research cannot be used to support this claim. Indeed, as will be seen later in this chapter, different conclusions
may better reflect the current state of research.
The Period of Neutral Effects
There are a series of researches that reported no difference between bilinguals and monolinguals in IQ. For
example, research in the United States by Pintner & Arsenian (1937) found a zero correlation (no relationship)
between verbal (and non-verbal) IQ and Yiddish-English bilingualism/monolingualism. While the number of
researches with a ‘no difference’ conclusion is small in number, the period of neutral effects is important because it
highlighted the inadequacies of the early detrimental effects research. An example is the research by W.R. Jones
(1959) in Wales. Using 2,500 children aged 10 and 11, Jones (1959) initially found that bilinguals were inferior to
monolinguals on IQ. A re-analysis showed that this conclusion was invalid. After taking into account the varying
socioeconomic class of bilinguals and monolinguals, Jones (1959) concluded that monolinguals and bilinguals did
not differ significantly in non-verbal IQ so long as parental occupation was taken into account. He also concluded
that socioeconomic class largely accounts for previous research that reported the inferiority of bilinguals on non-
verbal IQ. Therefore, his conclusion was that bilingualism is not necessarily a source of intellectual disadvantage.
While the period of neutral effects overlaps chronologically with the detrimental and additive periods, there was a
period when (in Wales, for example) such neutral effects were taught and publicized. Such a ‘neutral’ conclusion
was historically important as it gave a boost to parents who wished to support bilingualism in the home and in the
school. As a transitional period, it both helped to question a fashionable belief of bilingualism as cerebral
confusion, and became a herald for the additive effects period.
The Period of Additive Effects
A major turning point in the history of the relationship between bilingualism and cognition was reached in
Canadian research by Peal & Lambert (1962). It is this piece of research that heralded in the modern approach to
bilingualism and cognitive functioning. This research broke new territory in three respects, each setting the pattern
for future research.

< previous page page_121 next page >
< previous page page_122 next page >
Page 122
First, the research overcame many of the methodological deficiencies of the period of detrimental effects. Second,
the research found evidence that bilingualism need not have detrimental or even neutral consequences. Rather,
there is the possibility that bilingualism leads to cognitive advantages over monolingualism. Peal & Lambert’s
(1962) finding has been widely quoted to support a variety of bilingual policies in various educational contexts.
The political implication of the study was that bilingualism within a country was not a source of national
intellectual inferiority (Reynolds, 1991). Third, the research by Peal & Lambert (1962), while using IQ tests,
moved research to a broader look at cognition. Other areas of mental activity apart from IQ were placed firmly on
the agenda for research into bilingualism and cognitive functioning.
Peal & Lambert (1962) commenced with a sample of 364 children aged 10 years old drawn from middle-class
French schools in Montreal, Canada. The original sample of 364 children was reduced to 110 children for two
reasons. First, to create a group of balanced bilinguals (see Chapter 1) and a group of monolinguals. Second, to
ensure that the bilingual and monolingual groups were matched on socioeconomic class.
Bilinguals performed significantly higher on 15 out of the 18 variables measuring IQ. On the other three variables,
there was no difference between balanced bilinguals and monolinguals. Peal & Lambert (1962) concluded that
bilingualism provides: greater mental flexibility; the ability to think more abstractly, more independently of words,
providing superiority in concept formation; that a more enriched bilingual and bicultural environment benefits the
development of IQ; and that there is a positive transfer between a bilingual’s two languages facilitating the
development of verbal IQ.
These conclusions are historically more important than the specific results concerning IQ. That is, it is analysis of
the results rather than the details of the results that provided the stimulus for further research and debate.
The study of Peal & Lambert (1962), while being pivotal in research on bilingualism and cognitive functioning,
has four basic methodological weaknesses that need to be briefly considered before accepting the research at its
face value. First, the results concern 110 children of 10 years of age and of middle-class, Montreal extraction. This
is not a sample that can be generalized to the population of bilinguals either in Canada or throughout the world.
This is particularly so since the results concern 110 children selected from the original sample of 364. An
unanswered question is how the other 254 children performed across the broad range of tests given by Peal &
Lambert (1962).
Second, children in the bilingual group were ‘balanced’ bilinguals (see Chapter 1). While the term ‘bilinguals’
includes balanced bilinguals, there are many other groups of children ‘less balanced’. We cannot assume that the
results from this study apply to such ‘less balanced’ bilinguals. Are balanced bilinguals a special group with their
own characteristics in terms of their motivation, aptitude for languages, cognitive abilities and attitudes? Are
balanced bilinguals a special group of children who have a higher IQ that is due not only to owning two languages,
but due to other factors as well (e.g. parental values and expectations)?

< previous page page_122 next page >
< previous page page_123 next page >
Page 123
The third problem with Peal & Lambert’s (1962) research is the chicken and the egg problem which comes first?
What is the cause and what is the effect? Is it bilingualism that enhances IQ? Or does a higher IQ increase the
chances of becoming bilingual? When research suggests that IQ and bilingualism are positively related, we cannot
conclude the order of cause and effect. It may be that bilingualism enhances IQ. It may be that those with a higher
IQ are more likely to become bilingual. The relationship may also be such that one is both the cause and the effect
of the other. Research by Diaz (1985) suggests that, if there is a particular direction in the relationship, it is more
likely to be bilingualism positively affecting ‘intelligence’, rather than ‘intelligence’ affecting bilingualism.
The fourth problem concerns socioeconomic status. While Peal & Lambert (1962) tried to equate their bilingual
and monolingual groups for socioeconomic class by exclusion of some children, there are residual problems. As
Cummins (1976, 1984a) and MacNab (1979) have suggested, equating socioeconomic class does not control for all
the differences in a child’s home environment. Socioeconomic class is only a rough, simple and very partial
measure of a child’s home and environmental background. This is true of monolingual children. It is even more so
with children who are bilingual and bicultural where there may be an even more complicated home and family
background regarding sociocultural factors. Parental occupation of bilingual children is likely to summarize
differences between children very inadequately.
In the following example, notice how the sociocultural element is very different, yet the socioeconomic class is the
same. Take two Latino children of the same age and gender living in the same street in New York. Their fathers
both have the same jobtaxi drivers. One family regularly attends church services in Spanish and belongs to a
Hispanic organization with cultural activities in Spanish. This taxi driver and his wife send their children to a
Spanish-English dual language school. The child is bilingual. In the second family, the child speaks English only.
There is no interest in sending their children to a dual-language school. Neither does the family attend a church or
another organization where Spanish is spoken and valued. The Latin-American roots are neither discussed nor
appreciated. While the families are matched on socioeconomic status, the sociocultural differences between them
are considerable. In this example, the first child is bilingual and the second child is monolingual, with the bilingual
child having a higher IQ. The child’s bilingualism may not be the only explanation of a higher IQ. Rather the
alternative or additional explanation may be in the different social and cultural environment of these children.
Thus, with Peal & Lambert’s (1962) study, socioeconomic class may have been controlled, but not sociocultural
This completes the examination of Peal & Lambert’s (1962) important and pivotal study. Since their research, the
dominant approach to bilingualism and cognitive functioning has moved away from IQ testing to a multi-
component view of intelligence and cognition. Although there are researches after Peal & Lambert (1962) that
examine IQ and bilingualism, most recent studies look at bilingualism in terms of a range of thinking styles,
strategies and skills.

< previous page page_123 next page >
< previous page page_124 next page >
Page 124
While studies since Peal & Lambert (1962) are notable for their lack of integration, their non-theoretical rationale
and their lack of reference to advances in cognitive psychology, the studies mostly confirm Peal & Lambert’s
(1962) positive findings. These research studies are reviewed in the next chapter. Before such a review, a concept
allied to IQ is considered. Just as members of the public ask basic questions about ‘intelligence’ and bilingualism,
so questions often arise about bilinguals’ brains. For example, is the storage of information different for bilinguals’
and monolinguals’ brains? Evidence on this topic will now be briefly considered.
Bilingualism and the Brain
A frequently asked question is whether a bilinguals’ brain functions differently compared with that of a
monolingual’s brain? The issue becomes whether language is differently organized and processed in the brain of a
bilingual compared with the monolingual (e.g. Gomeztortosa et al., 1995). There are various researches on
neurolinguistics and bilingualism. A review by Obler (1983) admits that knowledge of cellular brain structure and
neurophysiological behavior as it relates to language processing is at a very rudimentary stage. The number of
studies looking at bilingual aphasia, laterality and split-brain patients is not great, and Obler (1983) finds it
difficult to make clear cut conclusions. An example is the research of Fromm (1970) who conducted a case study
of a man who was regressed by hypnosis to the age of seven. Under hypnosis, the man spoke fluent Japanese.
When he returned to his adult self, he could not speak Japanese. Yet other studies have not replicated this finding.
This symbolizes the necessity of great caution, as yet, in assertions about bilinguals and the brain.
A dominant topic in the study of bilingualism and the brain is lateralization. In the majority of right-handed adults,
the left hemisphere of the brain is dominant for language processing. The question has naturally arisen as to
whether bilinguals are different from monolinguals in this left lateralization? Vaid & Hall (1991) provide a review
of this topic in terms of five propositions derived from existing research:
(1) Balanced bilinguals will use the right hemisphere more than monolinguals for first and second language
(2) Second language acquisition will involve the right hemisphere in language processing more than first language
(3) As proficiency in a second language grows, right hemisphere involvement will decrease and left hemisphere
involvement will increase. This assumes that the right hemisphere is concerned with the more immediate,
pragmatic and emotive aspects of language; the left hemisphere with the more analytic aspects of language (e.g.
syntax). That is, the core aspects of language processing are assumed to reside in the left hemisphere.
(4) Those who acquire a second language naturally (e.g. on the street) will use their right hemisphere more for
language processing than those who learn a second language formally (e.g. in the language classroom). Learning
rules about grammar, spelling and irregular verbs will result in more left hemisphere involvement in second

< previous page page_124 next page >
< previous page page_125 next page >
Page 125
language learning. Picking up a language in a natural manner and using it for straightforward communication will
involve more right hemisphere involvement.
(5) Late bilinguals will be more likely to use the right hemisphere than early bilinguals. This proposition states that
there might be a ‘predominance of a left-hemisphere ‘semantic-type’ strategy in early bilinguals and for a right
hemisphere ‘acoustic-type’ strategy in late bilinguals’ (Vaid & Hall, 1991: 90).
Using a quantitative procedure called meta-analysis to review previous research in this area, Vaid & Hall (1991)
found that the left hemisphere strongly dominated language processing for both monolinguals and bilinguals.
However, differences between monolinguals and bilinguals were the exception rather than the rule. Bilinguals did
not seem to vary from monolinguals in neurological processes; the lateralization of language of the two groups
being relatively similar:
‘The largely negative findings from the meta-analysis must be taken seriously as reflecting a general lack of
support for the five hypotheses as they have been addressed in the literature to date’ (Vaid & Hall, 1991:
While the relationship between the brain and bilingualism is an important area, the present state of knowledge
makes a simple generalization unsafe.
A related area concerns the mental representation of a bilingual’s two languages and the processing emanating
from such representation. A principal issue has been the extent to which a bilingual’s two languages function
independently or interdependently. The early research attempted to show that early bilinguals (compound
bilinguals) were more likely to show interconnections and interrelatedness in their two languages than late
(coordinate) bilinguals. In the 1960s, Kolers (1963) re-defined the issue in terms of memory storage. A separate
storage hypothesis stated that bilinguals have two independent language storage and retrieval systems with the only
channel of communication being a translation process between the two separate systems. A shared storage
hypothesis stated that the two languages are kept in a single memory store with two different language input
channels and two different language output channels. Evidence exists for both independence and interdependence
requiring both hypotheses to be rejected. Rather, an integrated model involving both independence and
interdependence needs to be produced (Paivio, 1986, 1991; Heredia & McLaughlin, 1992; Keatley, 1992;
Hummel, 1993).
Recent theories and research have therefore emphasized both the separate and connected aspects of bilingual’s
mental representations by integrating the topic with general cognitive processing theories (e.g. Matsumi, 1994).
For example, the bilingual dual coding systems model of Paivio & Desrochers (1980) contains:
Two separate verbal language systems, one for each of a bilingual’s two languages.
A separate non-verbal imagery system, independent from the two language systems.
The non-verbal imagery system functioning as a shared conceptual system for the two languages.
Strong, direct interconnecting channels between each of these three separate systems.

< previous page page_125 next page >
< previous page page_126 next page >
Page 126
The interconnections between the two languages comprising association and translation systems; common images
also being mediators. The model is given in the diagram below (adapted and simplified from Paivio & Desrochers,
1980, 1991):
Bilingual Dual Coding Model

< previous page page_126 next page >
< previous page page_127 next page >
Page 127
This chapter has reviewed the strong, traditional and popular expectation that bilingualism and intelligence are
linked negatively. The conception is often that bilingualism leads to lower intelligence. Research from the 1920s to
the 1960s supported that conception. Recent research currently regards a simple negative relationship as a
misconception. The narrow view of intelligence contained in IQ tests and severe flaws in the design of early
research combine with other limitations to cast doubts on this negative link.
Rather, the need is to specify the language ability levels of bilinguals (see Chapter 1) and to ensure like is
compared with like. Since 1960, the indication has been that a more positive relationship between bilingualism and
cognitive functioning can be expected, particularly in ‘balanced’ bilinguals. The next chapter examines recent
research on the cognitive advantages and disadvantages of bilingualism.
Suggested Further Reading
BAKER, C. 1988, Key Issues in Bilingualism and Bilingual Education. Clevedon: Multilingual Matters.
HAKUTA, K. 1986, Mirror of Language. The Debate on Bilingualism. New York: Basic Books.
HARRIS, R.J. (ed.) 1992, Cognitive Processing in Bilinguals. Amsterdam: North-Holland.
Review and Study Questions
(1) Write short revision notes on the following:
(i) The meaning of the three periods: detrimental, neutral and additive.
(ii) The findings and criticisms of Peal & Lambert’s (1962) research.
(iii) The difference between the separate storage and shared storage hypotheses.
(2) What were the limitations of the research in the detrimental effects period?
(3) Why is the concept of IQ a controversial one, and how does this controversy relate to bilingualism?
Study Activities
(1) In a school of your choice, interview one or more teachers. Find out whether IQ tests have been used in the
past and/or are currently being used. For what purposes were IQ tests used? Do teachers expect bilingual children
to show a lower IQ than monolinguals? Do teachers think that IQ tests in the majority language are fair to
bilingual children?
(2) Find one or more examples of an IQ test. Examine the content of the test and locate any items which you think
will be unfair to bilinguals. Examine both the language and the cultural content of the IQ test. What kinds of IQ
tests seem to be more or less appropriate for bilinguals? Are culture-free tests and non-verbal IQ tests more fair?

< previous page page_127 next page >
< previous page page_128 next page >
Page 128
Chapter 8
Bilingualism and Thinking
Bilingualism and Divergent and Creative Thinking
Bilingualism and Metalinguistic Awareness: Initial
Bilingualism and Metalinguistic Awareness: Recent
Bilingualism and Communicative Sensitivity
Field Dependency and Independency and
Explanations of Findings
Limitations of the Findings

< previous page page_128 next page >
< previous page page_129 next page >
Page 129
Recent research tends not to compare monolingual and bilingual groups on an IQ test. Rather, the modern
approach is to focus on a wider sample of products and processes of a bilingual’s cognition. Do bilinguals and
monolinguals differ in thinking styles? Are there differences in the processing of information? Does owning two
languages create differences in thinking about language? These type of questions are examined in this chapter.
Bilingualism and Divergent and Creative Thinking
One problem with IQ tests is that they restrict children to finding the one correct answer to each question. This is
often termed convergent thinking. Children have to converge onto the sole acceptable answer. An alternative style
is called divergent thinking. A child regarded as a diverger is more creative, imaginative, elastic, open ended and
free in thinking. Instead of finding the one correct answer, divergent thinkers prefer to provide a variety of
answers, all of which can be valid.
Divergent thinking is investigated by asking questions such as: ‘How many uses can you think of for a brick?’;
‘How many interesting and unusual uses can you think of for tin cans?’; ‘How many different uses can you think of
for car tires?’. On this kind of question, the student has to diverge and find as many answers as possible. For
example, on the ‘uses of a brick’ question, a convergent thinker would tend to produce a few rather obvious
answers to the question: to build a house, to build a barbecue, to build a wall. The divergent thinker will tend to
produce not only many different answers, but also some may be fairly original: for blocking up a rabbit hole, for
propping up a wobbly table, as a foot wiper, breaking a window, making a bird bath.
In the British tradition, the term used for this area is divergent thinking (Hudson, 1966, 1968). In the North
American tradition, it is more usual to talk about Creative Thinking. In the North American tradition, Torrance
(1974a, 1974b) analyzes answers to the ‘Uses of an Object’ test (e.g. unusual uses of cardboard boxes, unusual uses
of tin cans) by four

< previous page page_129 next page >
< previous page page_130 next page >
Page 130
categories. This test may be adapted into any language, with a culturally appropriate use of objects.
A person’s fluency score in creative thinking is the number of different acceptable answers that are given. A
flexibility score is the number of different categories (listed in the Test Manual) into which answers can be placed.
Originality is measured by reference to the Test Manual that gives scores of 0,1, or 2 for the originality (statistical
infrequency) of each response. Elaboration refers to the extent of the extra detail that a person gives beyond the
basic use of an object. A similar scoring system is used with divergent and creativity tests containing non-verbal
questions. Sometimes called Figural Tests, a person is given a sheet of 40 circles or 40 squares and asked to draw
pictures using these individual circles or squares, and subsequently place a label underneath.
The underlying hypothesis concerning creative thinking and bilingualism is that the ownership of two or more
languages may increase fluency, flexibility, originality and elaboration in thinking. Bilinguals will have two or
more words for a single object or idea. For example, in Welsh, the word ysgol not only means a school but also a
ladder. Thus having the word ‘ysgol’ in Welsh and ‘school’ in English may provide the bilingual with an added
dimensionthe idea of the school as a ladder. Similarly, having two or more words for folk dancing or square
dancing in different languages may give a wider variety of associations than having a single word in one language.
Does having two or more words for the one object or idea allow a person more freedom and richness of thought?
Research has compared bilinguals and monolinguals on a variety of measures of divergent thinking (see Baker
(1988) and Ricciardelli (1992) for reviews). The research is international and cross cultural: from Ireland,
Malaysia, Eastern Europe, Canada, Singapore, Mexico and the US, sampling bilinguals using English plus
Chinese, Bahasa Melayu, Tamil, Polish, German, Greek, Spanish, French, Ukrainian, Yorubo, Welsh, Italian or
Kannada. As Laurén (1991) notes, such research has mostly occurred in additive bilingual contexts. The research
findings largely suggest that bilinguals are superior to monolinguals on divergent thinking tests. An example will
Cummins (1975, 1977) found that balanced bilinguals were superior to ‘matched’ non-balanced bilinguals on the
fluency and flexibility scales of verbal divergence, and marginally on originality. The ‘matched’ monolingual group
obtained similar scores to the balanced bilingual group on verbal fluency and flexibility but scored substantially
higher than the non-balanced group. On originality, monolinguals scored at a similar level to the non-balanced
bilinguals and substantially lower than the balanced group. Probably due to the small numbers involved, the results
did not quite attain customary levels of statistical significance. That there are differences between matched groups
of balanced bilinguals and non-balanced bilinguals suggests that bilingualism and superior divergent thinking skills
are not simply related. Cummins (1977) proposed that:
‘there may be a threshold level of linguistic competence which a bilingual child must attain both in order to
avoid cognitive deficits and allow the potentially beneficial aspects of becoming bilingual to influence his
cognitive growth’. (p. 10)

< previous page page_130 next page >
< previous page page_131 next page >
Page 131
The difference between balanced and non-balanced bilinguals is thus explained by a threshold. Once children have
obtained a certain level of competence in their second language, positive cognitive consequences can result.
However, competence in a second language below a certain threshold level may fail to give any cognitive benefits.
Cognitive benefits may accrue when a child’s second language competence is fairly similar to first language skills.
This is the basic notion of the threshold theory which is examined further in Chapter 9.
The weight of the evidence suggests that balanced bilinguals have superior divergent thinking skills compared with
unbalanced bilinguals and monolinguals. In a review of 24 studies, Ricciardelli (1992) found that on 20 of these
bilinguals performed higher than monolinguals. Studies not supporting bilingual superiority sampled less proficient
bilinguals; a result consistent with the Thresholds Theory. However, some care must be taken in too firm a
conclusion. The researches on this topic have methodological limitations and deficiencies. To ensure a fair
judgment, these problems need to be borne in mind.
The five major problems may be listed as follows: (1) Some studies fail to control adequately for differences
between bilingual and monolingual groups (e.g. age and socioeconomic differences). (2) Some studies have small
samples such that generalization must be very restricted or non-existent. (3) Some studies fail to define or describe
the level or degree of bilingualism in their sample. As Cummins (1977) has shown, the extent of such competence
is an important intermediate variable. (4) Not all studies find a positive relationship between bilingualism and
divergent thinking. Cummins & Gulutsan (1974), for example, found differences on only one of their five
measures of divergent thinking. Laurén (1991) found that linguistic creativity differences were found in grades
three, six and nine. Older bilinguals were ahead on only one of the four measures (use of compound nouns). (5)
The term ‘creativity’ is defined in different ways. As Laurén (1991) reveals, there is an interpretation by
psychologists (e.g. cognitive flexibility, fluency, originality and elaboration; tolerance of ambiguity); by linguists
(e.g. the ability to create new meanings in different contexts, complexity of sentences); child development
researchers (e.g. transforming the language input of parents and teachers); and creative writing proponents. Do
bilinguals show advantages on psychological rather than the purely linguistic creativity measures? (6) Human
beings are ever likely to change and develop, so is it the case that bilingualism gives cognitive advantages of a
permanent nature? If there are positive cognitive advantages linked with bilingualism, it is important to ask
whether these are temporary, or cumulative and everlasting? Research studies tend to use children aged 4 to 17.
What happens into the twenties or middle age or older age? Does bilingualism accelerate cognitive growth in the
early years, with monolinguals catching up in later years? Balkan (1970) and Ben-Zeev (1977a) have suggested
that such cognitive advantages are predominant in younger rather than older children. Arnberg (1981) disagrees
and argues that cognitive development is additive and cumulative, and that the further the child moves towards
balanced bilingualism, the more there are cognitive advantages to be gained. With cognitive style (e.g. divergent
thinking) may go relatively stable and lasting effects.

< previous page page_131 next page >
< previous page page_132 next page >
Page 132
Bilingualism and Metalinguistic Awareness:
Initial Research
The research on bilingualism and divergent thinking tentatively suggests that bilinguals may have some advantage
over matched monolinguals. The ownership of two languages may provide some advantages in the way language
relates to thinking. For many bilingual children, the size of their total vocabulary in both languages is likely to be
greater than that of a monolingual child (Swain, 1972). Does a larger overall vocabulary allow a bilingual to be
more free and open, more flexible and original particularly in meanings attached to words? Is a bilingual person
therefore less bound by words, more elastic in thinking due to owning two languages? For example, Doyle et al.
(1978) found that bilinguals tend to be superior in their ability to relate stories and to express concepts within those
stories when compared with monolinguals.
Leopold’s (19391949) famous case study (see Chapter 5) of the German-English development of his daughter,
Hildegard, noted the looseness of the link between word and meaningan effect apparently due to bilingualism.
Favorite stories were not repeated with stereotyped wording; vocabulary substitutions were made freely in
memorized songs and rhymes. Word sound and word meaning were separated. The name of an object or concept
was separated from the object or concept itself.
To illustrate: imagine a kindergarten monolingual and bilingual child are taught the same nursery rhyme:
Jack and Jill went up the hill
To fetch a pail of water.
Jack fell down and broke his crown
And Jill came tumbling after.
The monolingual child may be relatively more likely to repeat the verse with little or no alteration. The focus is on
the sound and the rhyme. The bilingual child may be more likely to center on the meaning than repeating the
words parrot fashion. For example, the bilingual may offer the following version; the word substitutions
suggesting that the nursery rhyme has been processed for meaning:
Jack and Jill climbed a hill
To fetch a bucket of water.
Jack fell down and banged his head
And Jill came falling after.
Hildegard is a single case and the nursery rhyme is merely an illustration based on one child. What has research
revealed about samples of bilinguals?
Ianco-Worrall (1972) tested the sound and meaning separation idea on 30 Afrikaans-English bilinguals aged four
to nine. The bilingual group was matched with monolinguals on IQ, age, sex, school grade and social class. In the
first experiment, a typical question was: ‘I have three words: CAP, CAN and HAT. Which is more like CAP, CAN
or HAT?’ A child who says that CAN is more like CAP would appear to be making a choice

< previous page page_132 next page >
< previous page page_133 next page >
Page 133
determined by the sound of the word. That is, CAP and CAN have two out of three letters in common. A child who
chooses HAT would appear to be making a choice based on the meaning of the word.
That is, HAT and CAP refer to similar objects. Ianco-Worrall (1972) showed that, by seven years of age, there was
no difference between bilinguals and monolinguals in their choices. Both groups chose HAT, their answer being
governed by the meaning of the word. However, with 4 to 6 year olds, she found that bilinguals tended to respond
to word meaning, monolinguals more to the sound of the word. This led Ianco-Worrall (1972) to conclude that
‘reach a stage of semantic development, as measured by our test, some 23 years earlier than their
monolingual peers’. (p. 1398).
In a further experiment, Ianco-Worrall (1972) asked the following type of question: ‘Suppose you were making up
names for things, could you call a cow ”dog” and a dog “cow'”? Bilinguals mostly felt that names could be
interchangeable. Monolinguals, in comparison, more often said that names for objects such as cow and dog could
not be interchanged. Another way of describing this is to say that monolinguals tend to be bound by words,
bilinguals tend to believe that language is more arbitrary. For bilinguals, names and objects are separate. This
seems to be a result of owning two languages, giving the bilingual child awareness of the free, non-fixed
relationship between objects and their labels.
Other research in this area, for example, by Ben-Zeev (1977a, 1977b) suggests that the ability of bilinguals to
analyze and inspect their languages, stems from the necessity of avoiding ‘interference’ between the two languages.
That is, the process of separating two languages and avoiding code-mixing, may give bilinguals superiority over
monolinguals through an increased analytical orientation to language. One of Ben-Zeev’s (1977a) tests, called the
Symbol Substitution Test, asked children to substitute one word for another in a sentence. For example, they had to
use the word ‘macaroni’ instead of ‘I’ in a sentence. The sentence to say becomes ‘Macaroni am warm’, thus
avoiding saying ‘I am warm’. Respondents have to ignore word meaning, avoid framing a correct sentence and
evade the interference of word substitution in order to respond to the task correctly. Bilinguals were found by Ben-
Zeev (1977a) to be superior on this kind of test, not only with regard to meaning, but also with regard to sentence
Ben-Zeev (1977a), using Hebrew-English bilinguals from Israel and the United States aged 58, argues that
bilinguals have advantages because they experience two language systems with two different sets of construction
rules. Therefore bilinguals appear to be more flexible and analytical in language skills.
Bilingualism and Metalinguistic Awareness:
Recent Trends
Much of the research on bilingualism and cognitive functioning has concentrated on cognitive style (e.g. divergent
and creative thinking). The focus of research tends to be

< previous page page_133 next page >
< previous page page_134 next page >
Page 134
on the person and on the product. Such research has tended to focus on whether bilinguals are superior or inferior
to monolinguals as people. It has attempted to locate dimensions of thinking where bilinguals perform better than
monolinguals. The recent trend has been to look at the process of thinking rather than the products of thinking,
working within the information processing, memorization and language processing approaches in psychology (e.g.
Kardash et al., 1988; Ransdell & Fischler, 1987,1989; Bialystok, 1991; Harris, R., 1992; Keatley, 1992; Padilla &
Sung, 1992; Hummel, 1993). Not all such studies are ‘favorable to bilinguals’ (e.g. Ellis, N. (1992) and Geary et
al., (1993) on the processing of numbers). However, research on memorization, language recall, reaction times and
processing times tends to use bilinguals to help describe and explain cognitive processing and language processing.
Comparisons with monolinguals aim to aid psychological understanding of cognitive processes rather than to focus
on bilinguals per se.
Process research which builds cumulatively on previous research on bilingualism and cognitive functioning has
focused particularly on the metalinguistic awareness of bilingual children (Galambos & Hakuta, 1988; Bialystok &
Ryan, 1985; Bialystok, 1987a, 1987b; Ricciardelli, 1993; Campbell & Sais, 1995). While the topic of
metalinguistic awareness is wide and ambiguous (Tunmer, Pratt & Herriman, 1984), it may loosely be defined as
the ability to think about and reflect upon the nature and functions of language:
‘As a first approximation, metalinguistic awareness may be defined as the ability to reflect upon and
manipulate the structural features of spoken language, treating language itself as an object of thought, as
opposed to simply using the language system to comprehend and produce sentences. To be
metalinguistically aware is to begin to appreciate that the stream of speech, beginning with the acoustic
signal and ending with the speaker’s intended meaning, can be looked at with the mind’s eye and taken apart’
(Tunmer & Herriman, 1984: 12)
Such metalinguistic awareness is regarded by Donaldson (1978) as a key factor in the development of reading in
young children. This hints that bilinguals may be ready slightly earlier than monolinguals in learning to read. Early
research suggested a relationship favoring bilinguals in terms of increased metalinguistic awareness (e.g. Ianco-
Worrall, 1972; Ben-Zeev, 1977a, 1977b; see Tunmer & Myhill, 1984 for a review). It appears that bilingual
children develop a more analytical orientation to language through organizing their two language systems.
In research that directly engages bilingualism and metalinguistic awareness, Bialystok (1987a, 1987b) found that
bilingual children were superior to monolingual children on measures of the cognitive control of linguistic
processes. Bialystok (1987a) conducted three studies each involving around 120 children age 59. In the
experiments, children were asked to judge or correct sentences for their syntactic acceptability irrespective of
meaningfulness. Sentences could be meaningfully grammatical (e.g. why is the dog barking so loudly?);
meaningful but not grammatical (e.g. why the dog is barking so loudly?); anomalous and grammatical (e.g. why is
the cat barking so loudly?), or anomalous and ungrammatical (e.g. why the cat is barking so loudly?). These

< previous page page_134 next page >
< previous page page_135 next page >
Page 135
test the level of analysis of a child’s linguistic knowledge. The instructions requested that the children focus on
whether a given sentence was grammatically correct or not. It did not matter that the sentence was silly or
anomalous. Bialystok (1987a) found that bilingual children in all three studies consistently judged grammaticality
more accurately than did monolingual children at all the ages tested.
Bialystok (1987b) also examined the difference between bilinguals and monolinguals in their processing of words
and the development of the concept of a word. She found in three studies that bilingual children showed more
advanced understanding of some aspects of the idea of words than did monolingual children. A procedure for
testing children’s awareness of ‘What is a word?’ is to ask children to determine the number of words in a sentence.
It can be surprisingly difficult for young children to count how many words there are in a sentence. Until children
are about 67 years of age and learning to read, they do not appear to have this processing ability.
To be able to count how many words there are in a sentence depends on two things: first, a knowledge of the
boundaries of words; second, a knowledge of the relationship between word meaning and sentence meaning. At
around 7 years of age, children learn that words can be isolated from the sentences in which they are contained,
having their own individual meaning. Bialystok (1987b) found that bilingual children were ahead of monolingual
children on counting words in sentences because: (1) they were more clear about the criteria that determined the
identity of words and, (2) they were more capable of attending to the units of speech they considered relevant.
‘Bilingual children were most notably advanced when required to separate out individual words from
meaningful sentences, focus on only the form or meaning of a word under highly distracting conditions, and
re-assign a familiar name to a different object.’ (Bialystok, 1987b: 138).
The conclusion can be summarized as follows. Fully fluent bilinguals have increased metalinguistic abilities. This
relates to two components (Bialystok, 1988; Ricciardelli, 1993): bilinguals’ enhanced analyzing of their knowledge
of language; and their greater control of internal language processing. In turn, this may facilitate earlier reading
acquisition that, in turn, can lead to higher levels of academic achievement.
The research of Galambos & Hakuta (1988) provides further refinement of the reasons for differences between
bilinguals and monolinguals on cognitive processes. In two studies with low income Spanish-English bilingual
children in the US, Galambos & Hakuta (1988) used a series of tests that examined children’s ability to spot
various errors in Spanish sentences. Such errors might be in terms of gender, word order, singular and plural, verb
tense and time. For example, on a grammatically oriented test item, a child had to correct the following: ‘La perro
es grande’. The correction would be ‘El perro es grande’. In a content oriented test item, ‘La perro es grande’
would become ‘El perro es pequeno’. In the experiment, children were read the sentences, asked to judge whether
it had been said in the right way, and then correct the error.

< previous page page_135 next page >
< previous page page_136 next page >
Page 136
The effect of bilingualism on the processing of the test items was found to vary depending on the level of
bilingualism and the difficulty level of the items. The more bilingual the child, that is where both languages were
relatively well developed, the better the performance on the test items:
‘The information-processing approach successfully accounts for our findings that bilingualism by and large
enhances the metalinguistic abilities to note errors and correct errors . . . The bilingual experience requires
that the form of the two languages being learned be attended to on a routine basis. Experience at attending to
form would be predicted to facilitate any task that required the child to focus on form upon demand.’
(Galambos & Hakuta, 1988: 153)
Galambos & Hakuta (1988) concluded that metalinguistic awareness is most developed when a child’s two
languages are developed to their highest level. The development of two languages to their fullest, particularly the
minority language, encourages metalinguistic awareness. Galambos & Hakuta (1988) agree with Cummins (1976)
that a certain level of proficiency in both languages must be attained before the positive effects of bilingualism on
metalinguistic awareness can occur. This is usually termed the thresholds theory and is considered in Chapter 9.
The evidence of advantages for bilinguals in terms of metalinguistic awareness seems fairly strong. What is not
apparent, as yet, is the extent of these effects. For example, are these effects temporary and located with younger
children? Are the effects in any way permanent? Do they give a child an initial advantage that soon disappears
with growing cognitive competence? Are these early benefits for bilinguals cumulative and additive? Apart from
‘balanced bilinguals’, which groups of bilinguals gain such advantages?
Bilingualism and Communicative Sensitivity
In Ben-Zeev’s research (1977b) on the comparative performance of bilingual and monolingual children on
Piagetian tests, she found that bilinguals were more responsive to hints and clues given in the experimental
situation. That is, bilinguals seem more sensitive in an experimental situation, and corrected their errors faster
compared to monolinguals. While Piagetian research and bilingualism is not considered in this chapter (see Baker
(1988) for a review), Ben-Zeev’s (1977b) research gave the first clue that bilinguals may have cognitive
advantages regarding ‘communicative sensitivity’.
What is ‘communicative sensitivity?’ Bilinguals need to be aware of which language to speak in which situation.
They need constantly to monitor what is the appropriate language in which to respond or when initiating a
conversation (e.g. on the telephone, in a shop, speaking to a superior). Not only do bilinguals often attempt to
avoid ‘interference’ between their two languages, they also have to pick up clues and cues when to switch
languages. The literature suggests that this may give a bilingual increased sensitivity to the social nature and
communicative functions of language.
An interesting experiment on sensitivity to communication is by Genesee, Tucker & Lambert (1975). They
compared children in bilingual and monolingual education on their

< previous page page_136 next page >
< previous page page_137 next page >
Page 137
performance on a game. In this simple but ingenious research, students aged 58 were asked to explain a board and
dice game to two listeners. One listener was blindfolded, the other not. The listeners were classmates and not
allowed to ask any questions after the explanation. The classmates then attempted to play the game with a person
giving the explanation. It was found that children in a bilingual education program (total immersionsee Chapter
10) were more sensitive to the needs of listeners. This bilingual education group gave more information to the
blindfolded children than to the sighted listener compared with children in the monolingual education control
group. The authors concluded that children in bilingual education ‘may have been better able than the control
children to take the role of others experiencing communicational difficulties, to perceive their needs, and
consequently to respond appropriately to these needs’ (p. 1013).
In conclusion, this implies that bilingual children may be more sensitive than monolingual children in a social
situation that requires careful communication. A bilingual child may be more aware of the needs of the listener.
More research is needed to define precisely the characteristics and the extent of the sensitivity to communication
that bilinguals may share. Research in this area is important because it connects cognition with interpersonal
relationships. It moves from questions about skills of the bilingual mind to a bilingual’s social skills.
Field Dependency and Independency and Bilingualism
One well researched dimension of cognitive style along which people vary is field dependencefield independence
(Witkin et al., 1971). An example from one of Witkin et al.’ s (1971) tests helps illustrate this dimension. In the
Embedded Figures Test, a child has to draw in the left hand figure over the right hand figure. Can the child
recognize the cuboid in the right hand figure?
Children who are field independent will tend to see the object separate from the other lines in the right hand figure.
Those field dependent will tend not to separate out the cuboid figure from the background. The simple shape is not
identified in the more complex shape. Simply stated, some people tend to see in wholes, others in parts. Witkin et
al. (1962) found that as children grow to maturity they become more field independent. While field

< previous page page_137 next page >
< previous page page_138 next page >
Page 138
dependence-independence may appear as a perceptual ability, Witkin and his co-workers regard it as a general
ability to be aware of visual contexts, and relates particularly to problem solving ability and ease of cognitive
restructuring. Field independent individuals tend to achieve higher academically than those field dependent.
One early study which examined whether bilinguals have a particular cognitive style compared with monolinguals
was conducted in Switzerland by Balkan (1970). FrenchEnglish balanced bilinguals aged 11 to 17 were matched
with monolinguals on IQ, and compared on an adaptation of the Embedded Figures Test. Balkan (1970) found that
bilinguals were more field independent and that those who learnt their second language before the age of four
tended to be more field independent than late learners (who had learnt their second language between 4 and 8
years). In Canada, students age 12 to 13 years in a bilingual education program were superior to the control groups
on the Embedded Figures Test (Bruck, Lambert & Tucker, quoted in Cummins, 1978). Although the comparison
groups were different from Balkan (1970), the Canadian research provided further evidence for bilingualism being
linked to greater field independence. However, Genesee & Hamayan (1980) found that those more field
independent learnt a second language better. This highlights the cause-effect relationship. Does field independence
cause bilingualism to occur more easily? Does bilingualism cause field independence? Or is one both the cause
and the effect of the other?
Apart from the Swiss and Canadian research, one further confirmatory piece of evidence comes from
English/Spanish bilinguals in the USA (Duncan & De Avila, 1979). Using the Children’s Embedded Figures Test,
the descending order of scores on field independence was:
(1) proficient bilinguals;
(2) partial bilinguals, monolinguals and limited bilinguals;
(3) late language learners/
The authors conclude that proficient bilinguals may have advantages in cognitive clarity and in analytical
functioning. This theme, of more proficient bilinguals having cognitive advantages over monolinguals and less
proficient bilinguals is returned to when considering the Thresholds Theory in the next chapter.
The Embedded Figures Test focuses on spatial ability. In contrast, metalinguistic awareness tests focus on
linguistic domains. On the surface, they have little in common. Yet bilinguals do well on both tests compared with
monolinguals. Are there any features in common between the tests that explain bilinguals cognitive advantages.
Bialystok (1992) argues that there is a commonality: selective attention that transfers across the spatial and
linguistic domains. Bilingual children can reconstruct a situation (perceptual or linguistic), focus on the key parts
of a problem and select the crucial parts of a solution. They can escape from perceptual seduction and overcome
cues that are irrelevant. Their ‘early experience with two languages may lead them to develop more sensitive
means for controlling attention to linguistic input. They are used to hearing to things referred to in two different
ways, and this can alert them earlier to the arbitrariness of referential forms’

< previous page page_138 next page >
< previous page page_139 next page >
Page 139
(Bialystok, 1992: 510). Bilinguals select which language to use with whom and avoid inappropriate language
forms. They may need to attend to social conventions that occur in each language (e.g. different forms of
politeness and deference to adults, different ways of greeting in Arabic and Spanish). Such selective attention may
transfer across spatial, cognitive and linguistic domains. For Bialystok (1992), it is selective attention that is the
explanation of bilinguals’ advantages on divergent thinking, creative thinking, metalinguistic awareness,
communicative sensitivity, Piagetian tests and on the Embedded Figures Test.
Explanations of Findings
Cummins (1976) has suggested that there are three different ways one might explain a relationship between
bilingualism and cognitive advantages. The first explanation is that bilinguals may have a wider and more varied
range of experiences than monolinguals due to their operating in two languages and probably two or more cultures.
The extra range of meanings that two languages provide, the added stream of experience that two languages may
give, and the extra cultural values and modes of thinking embedded in two languages may be increased for the
bilingual. Cummins (1976) believed that this was a hypothesis yet to be confirmed or rejected. This remains true at
the present.
The second explanation of the cognitive advantages of bilingualism concerns a switching mechanism. Because
bilingual children switch between their two languages, they may be more flexible in their thinking. Neufeld (1974)
was critical of this hypothesis, suggesting that monolinguals also switch from one register to another. That is, in
different situations and contexts, monolinguals and bilinguals have to know when and how to switch between
different modes of speech. This second proposition may be a part explanation onlya small component in a wider
and larger whole.
The third explanation is termed the process of objectification (Imedadze, 1960; Cummins & Gulutsan, 1974). A
bilingual may consciously and subconsciously compare and contrast their two languages. Comparing nuances of
meaning, different grammatical forms, being constantly vigilant over their languages may be an intrinsic process
due to being bilingual. Inspecting the two languages, resolving interference between languages may provide the
bilingual with metalinguistic skills. A famous Russian psychologist, Vygotsky (1962), suggested that bilingualism
enables a child ‘to see his language as one particular system among many, to view its phenomena under more
general categories, and this leads to awareness of his linguistic operations’ (p. 110). The suggestion of Vygotsky
(1962) seems the most commonly accepted single explanation for the current belief that bilinguals have cognitive
advantages over monolinguals. This third explanation is backed by the recent research on the metalinguistic
advantages of bilingualism (Diaz & Klingler, 1991).
While objectification may be a central component in explaining the cognitive advantages of bilingualism, it is
unlikely to be a sufficient explanation. Nor does objectification relate to a complete theory of intellectual
functioning. A creative attempt to fit the disparate research on bilingualism and cognition into an overarching
explanation is

< previous page page_139 next page >
< previous page page_140 next page >
Page 140
provided by Reynolds (1991). Reynolds (1991) commences with Sternberg’s (1985, 1988) three part model of
intelligence. This model of intelligence has three subtheories: contextual, experiential and componential.
In the Contextual Subtheory, intelligent behavior is regarded as adapting to the environment in which one is
placed. Going to a party and meeting new people on Saturday, next day attending church and a family get-
together, then attending school and leisure activities from Monday to Friday, each and all require a person to
constantly adapt to changing contexts, sometimes selecting contexts, even altering the environment to suit oneself.
Reynolds (1991) argues that bilinguals may be more capable at adapting to changing environments because of
their experience of separate linguistic environments and their (sometimes) wider social and cultural environments.
Evidence to support this comes from studies of bilinguals’ increased communicative sensitivity.
In the Experiential Subtheory, intelligent behavior is seen as variable, depending on how much experience a
person has of a situation. Acting intelligently is through effectively adapting to new situations. Effectiveness is
when behaviors become automated and habitual earlier rather than later. ‘Automatization’ allows cognitive
resources to be more sensibly allocated to the processing of new situations and challenges. According to Reynolds
(1991), bilinguals’ experience of two languages and switching between languages early in life allows easier
automatization in dealing with language tasks and frees resources for less familiar linguistic demands.
The Componential Subtheory concerns the processes that underlie intelligent behavior. Such processes firstly
involve Metacomponents which execute, control and monitor the processing of information. Because a bilingual
has to control and monitor two language systems, the metacomponent system may be more evolved and efficient.
Secondly, there are Performance components which administer the schemes devised by the Metacomponents.
Bilinguals may have advantages with performance components:
‘Having command of two languages leads to greater use of verbal mediation and increased use of language as
a cognitive regulatory tool. Having two interlocking performance systems for linguistic codes gives double
the resources for executing verbal tasks . . . Also there is greater use of learning strategies when learning two
languages.’ (Reynolds, 1991: 167)
Thirdly, Sternberg’s (1985, 1988) Componential Subtheory contains Knowledge-Acquisition components. Such
components encode new information and match old, memorized information with incoming new information to
enable intelligent functioning. Reynolds (1991) considers that a bilingual’s dual language system may make the
reception of new information more easy and fluent. New information can be assimilated into either or both the
language systems of the bilingual. A bilingual may sometimes have a double chance of acquiring new information
with two vocabularies and two verbal memories available.

< previous page page_140 next page >
< previous page page_141 next page >
Page 141
Limitations of the Findings
Research in the area of bilingualism and cognitive functioning often shares methodological problems. Before
conclusions are presented, it is important to bear in mind the limitations of research in this area. First, research
needs to match monolingual and bilingual groups on all variables other than language. This is in order for any
difference between such groups to be explained by bilingualism rather than by any other factors. While some
studies do attempt to control for alternative explanations by matching students in pairs or by the statistical
technique of analysis of co-variance, there are factors that may provide alternative explanations that are missing
from such research. These factors are the motivation of the children, parental attitude, school experience and the
culture of the home and community. These may be alternative explanations of many of the findings rather than, or
as well as bilingualism.
A second criticism is in the nature of the bilinguals that are used in such research. Researchers who find cognitive
advantages mostly focus on balanced bilinguals. Do balanced bilinguals represent all bilinguals? MacNab (1979)
argues that bilinguals are a special, idiosyncratic group in society. Because they have learnt a second language and
are often bicultural, bilinguals are different in major ways from monolinguals. For example, parents who want their
children to be bicultural and bilingual may emphasize divergent thinking skills, encourage creative thinking in
their children and foster metalinguistic skills. The parents of bilingual children may be the ones who want to
accelerate their children’s language skills. Such parents may give high priority to the development of languages
within their children compared with monolingual parents. While this does not detract from the possibility that
bilinguals do share some cognitive advantages, it does suggest a need to take care about a decision as to what are
the determining factors. It may be that it is not only language that is important. Other non-language factors may be
influential as well.
Third, the chicken and egg, cause and effect relationship must again be highlighted. What comes first? Most
research in this area assumes that bilingualism comes first and causes cognitive benefits. It is not impossible that
the causal link may run from cognitive abilities to enhanced language learning. Or it may be that language learning
and cognitive development work hand in hand. One both promotes and stimulates the other. It is unlikely that a
simple cause effect pattern exists. A more likely situation is the continuous process of interaction between
language and cognition. However Diaz (1985), using sophisticated statistical techniques (structural equation
modeling), suggests that bilingualism is more likely to be the cause of increased cognitive abilities than the
reverse. From current studies, this seems a fair conclusion.
Fourth, we need to ask which types of children have the benefits? This concerns whether children of all abilities
share the cognitive advantages of bilingualism. Do children below average in cognitive abilities also gain the
advantages of bilingualism? There is certainly a tendency in research to use children from the middle classes,
particularly those of above average ability. However Rueda (1983), using analytical orientation to language tests
(see earlier in this chapter), found that bilingual, less able

< previous page page_141 next page >
< previous page page_142 next page >
Page 142
children (5169 IQ level) tended to have cognitive advantages over ‘matched’ monolinguals. Rueda’s (1983)
research, which needs thorough replication, hints that cognitive advantages may be shared by below average ability
children and not just the average and above average ability children.
Fifth, when reviewing research we need to consider the hopes and the ideologies of the researcher. Rosenthal
(1966) has shown that experimenters’ expectations can affect the outcomes and results of human and animal
studies. As Hakuta (1986) suggests:
‘a full account of the relationship between bilingualism and intelligence, of why negative effects suddenly
turn in to positive effects, will have to examine the motivations of the researcher as well as more traditional
considerations at the level of methodology.’ (p. 43)
Have the assumptions and preferences of authors crept unintentionally into their research and affected both the
results and the interpretations of the results? In the choice of psychological tests and the choice of a sample, has
there been a built-in bias towards finding positive results on bilingualism and cognitive functioning?
Finally, we must ask whether the positive benefits from bilingualism in terms of thinking are temporary or
permanent? Most research studies tend to use children of school age. There is almost no research on the cognitive
functioning of bilinguals and monolinguals after the age of 17. Does bilingualism accelerate cognitive growth in
the early years of childhood with monolinguals catching up in later years? Are cognitive advantages predominant
in younger rather than older children? Or are the advantages additive, cumulative and long lasting? It is possible
that certain advantages (e.g. sensitivity to language, separation of word sound from word meaning) may be
temporary. Age and experience may eventually give bilingual and monolingual similar cognitive skills. However,
with the ideas of cognitive style (e.g. divergent and creative thinking) and communicative sensitivity, may go
relatively more stable and lasting effects.
A review of research on cognitive functioning and bilingualism suggests that two extreme conclusions may both be
untenable. To conclude that bilingualism gives undoubted cognitive advantage fails to consider the various
criticisms and limitations of research in this area. It also fails to recognize that there are studies (e.g. on memory),
where bilinguals may sometimes be at a disadvantage compared with monolinguals (e.g. Ransdell & Fischler,
1987). However, to conclude that all the research is invalid, fails to acknowledge that the judgment of the clear
majority of researchers tends to be that there are positive links between bilingualism and cognitive functioning.
While there is insufficient evidence to satisfy the skeptic, the evidence that currently exists does lead in the
direction of bilinguals having some cognitive advantages over monolinguals.

< previous page page_142 next page >
< previous page page_143 next page >
Page 143
Suggested Further Reading
BAKER, C. 1988, Key Issues in Bilingualism and Bilingual Education. Clevedon: Multilingual Matters.
BIALYSTOK, E. (ed.) 1991, Language Processing in Bilingual Children. Cambridge: Cambridge University
HARRIS, R.J. (ed.) 1992, Cognitive Processing in Bilinguals. Amsterdam: North-Holland.
PADILLA, R.V. and BENAVIDES, A.H. (eds.) 1992, Critical Perspectives on Bilingual Education Research.
Tempe, AZ: Bilingual Press.
Review and Study Questions
(1) Write short revision notes on the following:
(i) The relationship between bilingualism and (a) divergent and creative thinking, (b) metalinguistic
awareness, (c) communicative sensitivity and (d) field dependency and field independency.
(ii) The limitations of findings that find a positive link between bilingualism and thinking.
(2) What is meant by metalinguistic awareness? Try to find in the library some recent references to metalinguistic
awareness. Write down details of the variety of tests and definitions of metalinguistic awareness.
(3) What are the limitations of a conclusion that there are cognitive advantages to being bilingual? How fair do you
think such a conclusion is?
Study Activities
(1) Find a student or a teacher who you consider to be bilingual. Ask them to talk about the relationship between
their bilingualism and thinking. Ask them if they feel it gives them any advantages and any disadvantages. Collect
from them examples and illustrations.
(2) Using one of the tests or experiments mentioned in this chapter, select a student (or a group of students) and
give them that test. For example, ask them how many uses they can think of for a brick or for a cardboard box.
Compare the answers of those who are more and less bilingual and see if there are differences in quality and
quantity of answers.

< previous page page_143 next page >
< previous page page_144 next page >
Page 144
Chapter 9
Cognitive Theories of Bilingualism and the Curriculum
The Balance Theory
The Iceberg Analogy
The Thresholds Theory
The Development of a Theory
Curriculum Relevance

< previous page page_144 next page >
< previous page page_145 next page >
Page 145
The previous two chapters examined the relationship between bilingualism and IQ (Chapter 7) and bilingualism
and cognition (Chapter 8). These chapters were primarily based on research findings and culminated in
explanations of the likely positive relationship between bilingualism and thinking processes and products. This
chapter extends that discussion of explanations by firstly considering a ‘naive’ theory of language and cognitive
functioning; then, secondly, examining the development of a major and dominating theory of bilingualism and
cognition. The culmination of the chapter is a discussion of how this evolved theory has direct curriculum
The Balance Theory
Previous chapters noted that initial research into bilingualism and cognitive functioning and into bilingualism and
educational attainment often found bilinguals to be inferior to monolinguals. This connects with a naive theory of
bilingualism which represents the two languages as existing together in a balance. The picture is of weighing
scales, with a second language increasing at the expense of the first language. An alternative naive picture-theory
attached to the early research is of two language balloons inside the head. The picture portrays the monolingual as
having one well filled balloon. The bilingual is pictured as having two less filled or half filled balloons. As the
second language balloon is pumped higher (e.g. English in the US), so the first language balloon (e.g. Spanish)
diminishes in size. As one language balloon increases, the other decreases.
The balance and balloon picture theories of bilingualism and cognition appear to be held intuitively by many
people. Many parents and teachers, politicians and large sections of the public appear to latently, subconsciously
take the balloon picture as the one that best represents bilingual functioning. Cummins (1980a) refers to this as the

< previous page page_145 next page >
< previous page page_146 next page >
Page 146
Underlying Proficiency Model of Bilingualism.
This model conceives of the two languages operating separately without transfer and with a restricted amount of
‘room’ for languages.
What appears logical is not always psychologically valid. While both the balance or balloon ideas are plausible,
neither fits the evidence. As Chapter 8 concluded, when children become balanced bilinguals, the evidence
suggests that there are cognitive advantages rather than disadvantages for being bilingual. Similarly, Chapter 11
will show that certain types of bilingual education (e.g. early total immersion and heritage language, bilingual
education) appear to result in performance advantages (e.g. in two languages and in general curriculum
performance) compared with submersion or monolingual education.
Research has also suggested that it is wrong to assume that the brain has only a limited amount of room for
language skills, such that monolingualism is preferable. Evidence (see Chapters 5 and 8) suggests that there are
enough cerebral living quarters not only for two languages, but for other languages as well. The picture of the
weighing scales, of one language increasing at the expense the second language, does not fit the data. Other
pictures, provided later in this chapter, better encapsulate research findings.
There is another fallacy with the balance or balloon theory. The assumption of the theory is that the first and
second language are kept apart in two ‘balloons’ inside the head. The evidence suggests to the contrary, that
language attributes are not apart in the cognitive system, but transfer readily and are interactive. For example,
when school lessons are through the medium of Spanish, they do not solely feed a Spanish part of the brain. Or
when other lessons are in English, they do not only feed the English part of the

< previous page page_146 next page >
< previous page page_147 next page >
Page 147
brain. Rather lessons learnt in one language can readily transfer into the other language. Teaching a child to
multiply numbers in Spanish or use a dictionary in English easily transfers to multiplication or dictionary use in
the other language. A child does not have to be re-taught to multiply numbers in English. A mathematical concept
can be easily and immediately used in English or Spanish if those languages are sufficiently well developed. Such
easy exchange leads to an alternative idea called Common Underlying Proficiency (Cummins, 1980a, 1981a).
The Iceberg Analogy
Cummins’ (1980a, 1981a) Common Underlying Proficiency model of bilingualism can be pictorially represented in
the form of two icebergs (see below). The two icebergs are separate above the surface. That is, two languages are
visibly different in outward conversation. Underneath the surface, the two icebergs are fused so that the two
languages do not function separately. Both languages operate through the same central processing system.
A distinction can thus been made between the Separate Underlying Proficiency (SUP) and Common Underlying
Proficiency models of bilingualism (CUP). The former (SUP) relates to the ‘two balloon’ idea presented earlier in
this chapter; the latter (CUP) relates to the iceberg idea. The Common Underlying Proficiency model of
bilingualism may be summarized in six parts:
(1) Irrespective of the language in which a person is operating, the thoughts that accompany talking, reading,
writing and listening come from the same central engine. When a person owns two or more languages, there is one
integrated source of thought.

< previous page page_147 next page >
< previous page page_148 next page >
Page 148
(2) Bilingualism and multilingualism are possible because people have the capacity to store easily two or more
languages. People can also function in two or more languages with relative ease.
(3) Information processing skills and educational attainment may be developed through two languages as well as
through one language. Cognitive functioning and school achievement may be fed through one monolingual
channel or equally successfully through two well developed language channels. Both channels feed the same
central processor.
(4) The language the child is using in the classroom needs to be sufficiently well developed to be able to process
the cognitive challenges of the classroom.
(5) Speaking, listening, reading or writing in the first or the second language helps the whole cognitive system to
develop. However, if children are made to operate in an insufficiently developed second language (e.g. in a
‘submersion’ classroom), the system will not function at its best. If children are made to operate in the classroom in
a poorly developed second language, the quality and quantity of what they learn from complex curriculum
materials and produce in oral and written form may be relatively weak and impoverished. This has been the
experience of some Finns in Swedish schools who were forced to operate in Swedish (Skutnabb-Kangas &
Toukomaa, 1976). Such children tended to perform poorly in the curriculum in both Finnish and Swedish because
both languages were insufficiently developed to cope with given curriculum material.
(6) When one or both languages are not functioning fully (e.g. because of an unfavorable attitude to learning
through the second language, pressure to replace the home language with the majority language) cognitive
functioning and academic performance may be negatively affected.
The distinction between Separate Underlying Proficiency (SUP) and Common Underlying Proficiency models of
bilingualism (CUP) does not fully sum up the findings from research on cognitive functioning and bilingualism.
Therefore this chapter moves on to examining other more sophisticated theories.
The Thresholds Theory
Several studies have suggested that the further the child moves towards balanced bilingualism, the greater the
likelihood of cognitive advantages (e.g. Cummins & Mulcahy, 1978; Duncan & de Avila, 1979; Kessler & Quinn,
1982; Dawe, 1982, 1983; Clarkson, 1992). Thus the question has become ‘Under what conditions does
bilingualism have positive, neutral and negative effects on cognition?’ How far does someone have to travel up the
two language ladders to obtain cognitive advantages from bilingualism?
One theory that partially summarizes the relationship between cognition and degree of bilingualism is called the
Thresholds Theory. This was first postulated by Toukomaa & Skutnabb-Kangas (1977) and by Cummins (1976).
They suggest that the research on cognition and bilingualism is best explained by the idea of two thresholds. Each
threshold is a level of language competence that has consequences for a child. The first threshold

< previous page page_148 next page >
< previous page page_149 next page >
Page 149
is a level for a child to reach to avoid the negative consequences of bilingualism. The second threshold is a level
required to experience the possible positive benefits of bilingualism. Such a theory therefore limits which children
will be likely to obtain cognitive benefits from bilingualism. It also suggests that there are children who may
derive detrimental consequences from their bilingualism.
The Thresholds theory may be portrayed in terms of a house with three floors (see below). Up the sides of the
house are placed two language ladders, indicating that a bilingual child will usually be moving upward and is not
stationary on a floor. On the bottom floor of the house will be those whose current competence in both their
languages is insufficiently or relatively inadequately developed, especially compared with their age group. When
there is a low level of competence in both languages, there may be negative

< previous page page_149 next page >
< previous page page_150 next page >
Page 150
or detrimental cognitive effects. For example, a child who is unable to cope in the classroom in either language
may suffer when processing information. At the middle level, the second floor of the house, will be those with
age-appropriate competence in one of their languages but not in both. For example, children who can operate in
the classroom in one of their languages but not in their second language may reside in this second level. At this
level, a partly-bilingual child will be little different in cognition from the monolingual child and is unlikely to have
any significant positive or negative cognitive differences compared with a monolingual. At the top of the house,
the third floor, there resides children who approximate ‘balanced’ bilinguals. At this level, children will have age-
appropriate competence in two or more languages. For example, they can cope with curriculum material in either
of their languages. It is at this level that the positive cognitive advantages of bilingualism may appear. When a
child has age-appropriate ability in both their languages, they may have cognitive advantages over monolinguals.
Research support for the Thresholds Theory comes, for example, from Bialystok (1988), Clarkson & Galbraith
(1992), Clarkson (1992) and Dawe (1983). Dawe’s (1983) study examined bilingual Panjabi, Mirpuri and Jamaican
children age 11 to 13. On tests of deductive mathematical reasoning, Dawe (1983) found evidence for both the
lower and the higher threshold. As competency in two languages increased, so did deductive reasoning skills in
mathematics. Limited competence in both languages appear to result in negative cognitive outcomes. Bialystok
(1988) examined two parts to metalinguistic awareness (analysis of linguistic knowledge and control of linguistic
processing) in six- to seven-year-old monolingual, partial bilingual and fluently French-English children. She
found that ‘the level of bilingualism is decisive in determining the effect it will have on development’ (p. 567).
The Thresholds theory relates not only to cognition but also to education. With children in Immersion Education in
Canada (see Chapters 10 and 11), there is usually a temporary lag in achievement when the curriculum is taught
through the second language. Until the second language (French) has developed well enough to cope with
curriculum material, a temporary delay may be expected. Once French is developed sufficiently to cope with the
conceptual tasks of the classroom, Immersion Education is unlikely to have detrimental achievement consequences
for children. Indeed, such an immersion experience seems to enable children to reach the third floor of the house,
with resulting positive cognitive advantages.
The Thresholds theory also helps to summarize why minority language children taught through a second language
(e.g. in-migrants in the US) sometimes fail to develop sufficient competency in their second language (e.g.
English) and fail to benefit from ‘weak’ forms of bilingual education. Their low level of proficiency in English, for
example, limits their ability to cope in the curriculum. Therefore Heritage Language Programs, that allow a child to
operate in their more developed home language, can result in superior performance compared with submersion and
transitional bilingual education.
The problem with the Thresholds theory is in precisely defining the level of language proficiency a child must
obtain in order, firstly to avoid negative effects of bilingualism, and secondly, to obtain the positive advantages of
bilingualism. At what language ‘height’

< previous page page_150 next page >
< previous page page_151 next page >
Page 151
the ceilings become floors is not clear. Indeed, the danger may be in constructing artificial ‘critical stages’ or
levels, when transition is gradual and smooth. This point is returned to in the following section.
The Development of a Theory
From out of the Thresholds theory developed a succession of more refined theories of bilingualism. The first
evolution of the Thresholds theory considered the relationship between a bilingual’s two languages. To this end,
Cummins (1978) outlined the Developmental Interdependence hypothesis.
This hypothesis suggested that a child’s second language competence is partly dependent on the level of
competence already achieved in the first language. The more developed the first language, the easier it will be to
develop the second language. When the first language is at a low stage of evolution, the more difficult the
achievement of bilingualism will be.
Alongside this, in the 1970s, there developed a distinction between surface fluency and the more evolved language
skills required to benefit from the education process. Simple communication skills (e.g. being able to hold a simple
conversation with a shopkeeper) may hide a child’s relative inadequacy in the language proficiency necessary to
meet the cognitive and academic demands of the classroom. Cummins (1984a, 1984b) expressed this distinction in
terms of basic interpersonal communicative skills (BICS) and cognitive/academic language proficiency (CALP).
BICS is said to occur when there are contextual supports and props for language delivery. Face to face ‘context
embedded’ situations provide, for example, non-verbal support to secure understanding. Actions with eyes and
hands, instant feedback, cues and clues support verbal language. CALP, on the other hand, is said to occur in
context reduced academic situations.
Where higher order thinking skills (e.g. analysis, synthesis, evaluation) are required in the curriculum, language is
‘disembedded’ from a meaningful, supportive context. Where language is ‘disembedded’, the situation is often
referred to as ‘context reduced’. The following example illustrates the difference at classroom level between BICS
and CALP. A child is given a mathematical question such as: ‘You have 20 dollars. You have 6 dollars more than
me. How many dollars do I have?’ At the higher CALP level, the child will conceptualize the problem correctly as
20 minus 6 equals 14. At the BICS level, the word ‘more’ may be taken to mean ‘add-up’ with the child getting the
wrong answer of 26. The BICS child may think of ‘more’ as used in basic conversation. However, in the
mathematics classroom, this illustration requires ‘more’ to be understood by the mathematical phrasing of the
The distinction between BICS and CALP has been portrayed in the image of an iceberg (see Cummins, 1984b).
Above the surface are language skills such as comprehension and speaking. Underneath the surface are the skills
of analysis and synthesis. Above the surface are the language skills of pronunciation, vocabulary and grammar.
Below the surface are the deeper, subtle language skills of meanings and creative composition. This is illustrated
on the next page.

< previous page page_151 next page >
< previous page page_152 next page >
Page 152
Before leaving this BICS/CALP distinction, it is important to declare its limitations.
(1) The distinction between BICS and CALP has intuitive appeal and does appear to fit the case of children who
are seemingly fluent in their second language, yet cannot cope in the curriculum in that language. However, it only
paints a two stage idea. The idea of a larger number of dimensions of language competences may be more exact.
Children and adults may move forward on language dimensions in terms of sliding scales rather than in big jumps.
Such development is like gradually increasing in language competence analogous to increasing gradually the
volume on a television set. Harley et al.’s (1987, 1990) research suggests that a bilingual’s language competencies
are evolving, dynamic, interacting and intricate. They are not simple dichotomies, easily compartmentalized and
(2) The distinction enabled an understanding and explanation of previous research (e.g. Wong Fillmore, 1979;
Snow & Hoefnagel-Höhle, 1978; Cummins, 1984b). However, the distinction between BICS and CALP lacks
direct empirical support. Martin-Jones & Romaine (1986) express doubts about it being possible to test the
distinction. The distinction between BICS and CALP does not indicate how the two ideas may be precisely
defined and accurately tested. Thus the distinction becomes difficult to operationalize in research. For example, the
abilities referred to in CALP concern culture-specific types of literacy. This may be difficult to measure validly
and fully.
(3) Terms such as BICS and CALP tend to be imprecise, value-laden and become over-compartmentalized,
simplified and misused. These hypothetical terms may unwittingly be regarded as real entities. The terms may
over-simplify reality. Such terms may be used to label and stereotype students.
(4) The relationship between language development and cognitive development is not unequivocal or simple. It is
not simply a case of one growing as a direct result of the other. Cognitive and linguistic acquisition exist in a
relationship that is influenced by various other factors (e.g. motivation, school, home and community effects).
Language proficiency relates to an individual’s total environment, not just to cognitive skills.

< previous page page_152 next page >
< previous page page_153 next page >
Page 153
The distinction between BICS and CALP helps explain the relative failure within the educational system of many
minority language children. For example, in the United States, transitional bilingual education programs aim to
give students English language skills sufficient for them to be able to converse with peers and teachers in
mainstream education and to operate in the curriculum. Having achieved surface fluency, they may be transferred
to mainstream education. The transfer is enacted because children appear to have sufficient language competence
(BICS) to cope in mainstream education. Cummins’ (1984a) distinction between BICS and CALP explains why
such children tend to fail when mainstreamed. Their cognitive academic language proficiency is not developed
enough to cope with the demands of the curriculum. What Cummins (1984a) regards as essential in the bilingual
education of children is that the ‘common underlying proficiency’ is well developed. That is, a child’s language-
cognitive abilities need to be sufficiently well developed to cope with the curriculum processes of the classroom.
This underlying ability could be developed in the first or the second language, but also in both languages
A further development of this theory proposes two dimensions (Cummins, 1981b, 1983b, 1984b). This theory is
represented in the diagram below:
Both dimensions concern communicative proficiency. The first dimension refers to the amount of contextual
support available to a student. Context embedded communication exists when there is a good degree of support in
communication, particularly via body language (Argyle, 1975). For example, by pointing to objects, using the
eyes, head nods, hand gestures and intonation, people give and receive plenty of clues and cues to help understand
the content of the message.

< previous page page_153 next page >
< previous page page_154 next page >
Page 154
An example of context embedded communication would be when two children who are hardly able to use each
other’s languages seem able to communicate quite well by gestures, non-verbal reinforcements and bodily
movements. It is not infrequent to see two young children of different languages playing together without
difficulty. In context reduced communication there will be very few cues to the meaning that is being transmitted.
The words of the sentence exist almost alone in conveying the meaning. An example of context reduced
communication is often the classroom where the meaning is restricted to words, with a subtlety and precision of
meanings in the vocabulary of the teacher or the book.
The second dimension is the level of cognitive demands required in communication. Cognitively demanding
communication may occur in a classroom where much information at a challenging level needs processing quickly.
Cognitively undemanding communication is where a person has the mastery of language skills sufficient to enable
easy communication. An example would be having a conversation in the street, shop, or stadium, where the
processing of information is relatively simple and straightforward.
Surface fluency or basic interpersonal communication skills will fit into the first quadrant (see diagram). That is,
BICS (basic interpersonal communication skills) is context embedded, cognitively undemanding use of a language.
Language that is cognitively and academically more advanced (CALP) fits into the fourth quadrant (context
reduced and cognitively demanding). Cummins (1981b) theory suggests that second language competency in the
first quadrant (surface fluency) develops relatively independently of first language surface fluency. In comparison,
context reduced, cognitively demanding communication develops inter-dependently and can be promoted by either
language or by both languages in an interactive way. Thus, the theory suggests that bilingual education will be
successful when children have enough first or second language proficiency to work in the context reduced,
cognitively demanding situation of the classroom. For Cummins (1981b) it often takes one or two years for a child
to acquire context-embedded second language fluency, but five to seven years or more to acquire context-reduced
fluency. This is illustrated in the graphs below. Research by Hakuta & D’Andrea (1992) with Mexican-Americans
found that ‘English proficiency reaches asymptotic performance after about eight years. This corresponds quite
well with the figures of five to seven years required for attainment of the full range of second language acquisition
as estimated by Cummins (1984) based on a heterogeneous L1 population in Canada'(p.96).
Children with some conversational ability in their second language may falsely appear ready to be taught through
their second language in a classroom. Cummins (1981b) theory suggests that children operating at the context
embedded level in the language of the classroom may fail to understand the content of the curriculum and fail to
engage in the higher order cognitive processes of the classroom, such as synthesis, discussion, analysis, evaluation
and interpretation.
This two dimensional model helps explain various research findings:

< previous page page_154 next page >
< previous page page_155 next page >
Page 155
(1) In the United States, minority language children may be transferred from transitional bilingual programs into
English-only schooling when their conversational ability in English seems sufficient. Such students then frequently
perform poorly in mainstream schooling. The theory suggests that this is due to their not having the developed
ability in English (or their home language) to operate in an environment that is more cognitively and academically
(2) Immersion students in Canada tend to lag behind their monolingual peers for a short period. Once they acquire
second language proficiency sufficient to operate in a cognitively demanding and context reduced environment,
they usually catch up with their peers.
(3) Experiments in the United States, Canada and Europe with minority language children who are allowed to use
their minority language for part or much of their elementary schooling show that such children do not experience
retardation in school achievement or in majority language proficiency. Through their minority language, they
develop the ability to be relatively successful in the cognitively demanding and context reduced classroom
environment (Secada, 1991). This ability then transfers to the majority language when that language is well
enough developed. Children learning to read in their home language, be it Welsh, Gaelic, Irish, Spanish, Frisian or
whatever, are not just developing home language skills. They are also developing higher order cognitive and
linguistic skills that will help with the future development of reading in the majority language as well as with
general intellectual development. As Cummins (1984a) notes, ‘transfer is much more likely to occur from minority

< previous page page_155 next page >
< previous page page_156 next page >
Page 156
majority language because of the greater exposure to literacy in the majority language and the strong social
pressure to learn it’ (p. 143).
Curriculum Relevance
What a child brings to the classroom in terms of previous learning is a crucial starting point for the teacher. A
child’s reservoir of knowledge, understanding and experience can provide a meaningful context from which the
teacher can build (Robson, 1995). For example, there will be occasions when a child will learn more from a story
read by the teacher than listening to a language tape. When the teacher dramatizes a story by adding gestures,
pictures, facial expressions and other acting skills, the story becomes more context-embedded than listening to a
tape cassette. Getting a child to talk about something familiar will be cognitively less demanding than talking
about something culturally or academically unfamiliar. This means that any curriculum task presented to the
bilingual child needs considering for the:
what the task requires of the child; the cognitive demands inherent in the task (as found by an individual child);
the ‘entry skills’ that a task necessitates. This is illustrated in the table below:
The Cognitive Demands of the Classroom and
Contextual Supports

< previous page page_156 next page >
< previous page page_157 next page >
Page 157
form of presentation to the child (degree of context embeddedness or context reduction); what form of presentation
will be meaningful to the child; use of visual aids, demonstration, modeling, computers, oral and written
instructions; amount of teacher assistance;
child’s language proficiencies;
child’s previous cultural and educational experience and knowledge, individual learning style and learning
strategies; expectations and attitudes, confidence and initiative; the child’s familiarity with the type of task;
what is acceptable as evidence that learning has successfully occurred; what constitutes mastery or a sufficient
approximation; an appropriate form of ‘formative’ and ‘summative’ assessment (see Chapter 13) that may be
gestural, action (e.g. building a model), drawing, oral or written (Robson, 1995);
A simple example of using the two dimensions to produce an appropriate teaching strategy is now presented (see
Frederickson & Cline, 1990).
A teacher wants a group to learn how to measure height and to understand the concept of height. Listed below are
a few of the teaching strategies for teaching about height. Following the list is a diagram placing the four strategies
on the two dimensions:
One to one, individual teaching using various objects to measure height (1).
A demonstration from the front of the room by the teacher using various objects (2).
Teacher giving oral instructions without objects (3).
Reading instructions from a work card without pictures (4).

< previous page page_157 next page >
< previous page page_158 next page >
Page 158
As the above diagram indicates, the example of teaching height can be analyzed in terms of the two dimensions.
One to one individual teaching will fit somewhere in the context embedded, cognitively undemanding quadrant.
Using work cards may be closer to the context reduced, cognitively demanding area. Demonstrations and oral
explanations appear on the diagonal from ‘top left’ to ‘bottom right’, in-between individual teaching and work
cards. The exact location of teaching approaches on the graph will vary according to teacher, topic, learner and
lesson. The example illustrates that the two dimensions can be a valuable way of examining teaching approaches
with bilingual children. The dimensions are also useful for analyzing appropriate methods of classroom
assessment. The dimensions may help focus on task-related curriculum assessment that is more fair and
appropriate to bilingual children than norm referenced testing. A teacher wanting to check progress on measuring
height has a choice, for example:
observing a child measure the height of a new object (1);
asking the child to give a commentary while measuring a new object (2);
asking the child to provide a write-up of the process (3);
discussing in an abstract way the concept of height (4).
In plotting these four methods of assessment (see below), placement on the graph will vary with different kinds of
tasks and testing procedures. All four quadrants can be ‘filled’ depending on the student, teacher, topic and test.
There is also value in comparing the two graphs presented above. The teaching and learning approach taken may
well influence the form of assessment. That is, if a context embedded, cognitively undemanding

< previous page page_158 next page >
< previous page page_159 next page >
Page 159
learning strategy is used with a child, assessment may be on similar lines (e.g. observation of child activity).
Equally, a context reduced, cognitively demanding learning strategy suggests a ‘matched’ method of assessment
(e.g. discussion).
There are criticisms of Cummins’ (1981b) theory of the relationship between language and cognition. The main
criticisms by Edelsky et al. (1983), Martin-Jones & Romaine (1986), Rivera (1984), Frederickson & Cline (1990)
and Robson (1995) can be briefly summarized as follows:
(1) Cummins’ (1981b) early theory may artificially isolate certain ingredients in a bilingual’s cognitive or
classroom experience. The attainment of bilingualism or the relationship between bilingual education and school
achievement rests on many other factors than are presented in this theory. The early theory was essentially
individual and psychological. Bilingualism and bilingual education need to consider other variables; cultural,
social, political, community, teacher expectations and home factors. Each and all of these variables help explain
bilingualism as an individual and societal phenomenon.
(2) Cummins (1981b) criterion of educational success tended to center on dominant, middle-class indices of
achievement. Thus language skills, literacy and formal educational achievement are highlighted. Alternative
outcomes of schooling such as self-esteem, social and emotional development, divergent and creative thinking,
long-term attitude to learning, employment and moral development were not initially considered.
(3) The theory has been produced as a post hoc explanation of a variety of research findings. The theoretical
framework requires direct empirical investigation and confirmation with replication across culture and country,
time and educational tradition.
(4) Terms such as BICS and CALP tend to be vague, value-laden and in danger of creating an over-simplification
and a stereotyping of individual functioning and classroom processes. Essentially hypothetical and abstract, the
terms may be adopted as concrete and real (see earlier in this chapter).
(5) The two dimensions are not necessarily distinct, and may not best be represented by two maximally separated
(90 degrees apart) axes. When applying Cummins’ two dimensions to curriculum tasks, Frederickson & Cline
(1990) found it . . . ‘difficult to disentangle the ‘cognitive’ from the ‘contextual’. In some cases, movement along the
contextual dimensions has actually been represented on the model as a diagonal shift [on the diagram from top left
to bottom right], as it was found in practice that making tasks or instructions more context embedded also made
them somewhat less cognitively demanding. Similarly, changes in cognitive demand may result in tasks actually
being presented with greater context embeddedness’ (p. 26).
(6) When bilingual children appear to have learning difficulties, a teacher may decide to simplify tasks into smaller
and more isolated steps. Such a strategy is part of the behavioral objectives approach or a task analysis approach to
the curriculum. It may

< previous page page_159 next page >
< previous page page_160 next page >
Page 160
sometimes result in a non-meaningful context to a curriculum task. By making the task context-reduced, the
learning may become more difficult rather than easier, as intended.
(7) Attempting to achieve context embeddedness in any curriculum situation requires empathic understanding of a
child’s cultural background which itself is dynamic and ever evolving. A danger lies in the teacher developing self-
defeating stereotyped assumptions about a child’s ethnic experience which may transmit low expectations.
(8) The theory does not make allowances for a child’s cognitive strategies in learning, nor their learning style.
Naive theories of two languages within an individual are represented by two pictures. First, two languages as a
balance. Second, two languages operating as two separate balloons in the head. Such misconceptions can be
replaced by pictures such as the dual iceberg and the three tiered house. Depending on language development in
both languages, the cognitive functioning of an individual can be viewed as integrated, with easy transfer of
concepts and knowledge between languages. Understanding and thinking will be affected by the contextual support
that exists and the degree of cognitive demands in a task. Successful cognitive operations in the classroom will
depend on matching curriculum tasks with language competences. Sensitivity to the need for contextual support
and the cognitive demands of a classroom are important if an individual is to maximize learning in the curriculum.
Suggested Further Reading
CUMMINS, J. 1984, Bilingualism and Special Education: Issues in Assessment and Pedagogy. Clevedon:
Multilingual Matters.
CUMMINS, J. and SWAIN, M. 1986, Bilingualism in Education. New York: Longman.
RIVERA, C. (ed.) 1984, Language Proficiency and Academic Achievement. Clevedon: Multilingual Matters.
Review and Study Questions
(1) Write short revision notes on the following:
(i) The central idea(s) of the balance theory, the iceberg analogy and the thresholds theory.
(ii) The distinction between BICS and CALP, the value and use of these terms and criticisms of the
(iii) The meaning of: context embedded communication, context reduced communication, cognitively
demanding communication and cognitively undemanding communication.
(2) What is the curriculum relevance of theories presented in this chapter? Provide examples of relevance as in the
diagrams at the end of this chapter.

< previous page page_160 next page >
< previous page page_161 next page >
Page 161
(3) Define the terms used in this chapter in your own words and discuss their relevance to your own experience.
Study Activity
(1) Observe a classroom with bilingual children. Make a 10 minute cassette tape of the discourse between the
teacher and various students, and/or between students themselves. Using the framework on pp. 143 and 144,
describe and discuss the language used.

< previous page page_161 next page >
< previous page page_162 next page >
Page 162
Chapters 1 to 9 provided the definitional, psychological and sociological foundation upon which bilingual
education is built. The remaining chapters focus on bilingual education and bilingual classrooms. With basic
questions about language within individuals and within society having been considered in Section A, questions
about bilingual education will now have more meaning and fit into a wider societal framework. Bilingual
education cannot be viewed in isolation of its social, economic, cultural and political context.
This section is based around various questions about bilingual education:
What forms of bilingual education are more successful?
What are the aims and outcomes of different types of bilingual education?
How important is literacy and biliteracy in bilingual education?
What kinds of literacy are needed by language minority children?
What are the essential features and approaches of a classroom fostering bilingualism?
How should two languages be allocated in bilingual schooling?
How should deaf bilinguals be educated?
How should we approach bilingual children with special needs?
How is a second language best learnt in a classroom setting?
Why are there different viewpoints about language minorities and bilingual education?
Why do some people prefer the assimilation of language minorities and others prefer linguistic diversity?
Can schools play a role in a more multicultural and less racist society?
These questions are now examined in Section B. Chapter 10 provides a wide ranging discussion of different types
of education in which bilinguals are placed. Some types of bilingual education support a child’s bilingualism;
others do not. Such a discussion looks at a variety of international examples. Having described types of bilingual
education in

< previous page page_162 next page >
< previous page page_163 next page >
Page 163
Chapter 10, the next chapter analyzes the effectiveness of different forms of such education. Research evidence is
considered and differing viewpoints are encountered. These two chapters (10 and 11) may best be read as a pair.
Chapters 12 to 17 focus on different situations in bilingual education. Each moves from bilingual education
systems to a consideration of classroom related issues. While Chapters 10 and 11 discuss broad bilingual education
issues, Chapters 12 to 17 narrow the focus to major, practical classroom issues. Chapter 12 concerns the case of
classrooms where the minority language is a major medium of instruction. It discusses the language curriculum
and the way two languages need to be separated and allocated in bilingual classrooms. Chapter 13 considers the
issues of the assessment of bilinguals and particularly those from language minorities; the topic of bilingual special
education and analyzes causes of perceived underachievement in language minority children. Chapter 14 examines
how bilingualism can be attempted through second language lessons, second language teaching and foreign
language instruction. Here part of the focus is on majority language children learning a second (minority or
majority) language. The chapter is also relevant to minority language children formally learning a majority
language. Chapters 15 and 16 center on the important topic of literacy and biliteracy in bilingual children and
multicultural societies. What kind of literacy should be transmitted in bilingual and multicultural classrooms is
In Chapter 17, the well established Canadian Immersion classroom methodology is discussed. With the focus on
classroom strategies, immersion classrooms provide an illustration of successful education intervention that
promotes bilingualism. This area has been carefully researched and has become an internationally influential form
of bilingual education
Chapter 18 is an attempt to provide two summarizing thinking tools to encapsulate Chapters 10 to 16. A model of
bilingual education and a framework for language minority intervention are considered so as to gain an overall
conceptualization of these chapters.
The two final chapters (19 and 20) are closely linked and form an essential conclusion to the book. Issues about
bilingual education philosophy, policy, provision and practice are constantly infused with political considerations.
Politics runs through much of the discourse of the book, particularly Chapters 3 and 4, and Chapters 10, 11 13, 15
and 18. The crucial sociopolitical aspect of bilingual education is addressed as a culmination of the book. Various
threads from different chapters appear in these final two chapters, with an assumption that political issues are best
understood once concrete examples of diglossia and bilingual education are in place. Chapters 19 and 20 discuss
orientations and ideologies that connect together, and give a wide perspective and overview of the previous
chapters. The key debate about assimilation and pluralism runs through these final two chapters and shows how
sociological, psychological and education issues about bilingualism are fundamentally related to wider issues of
personal prejudices, public perceptions and pervading politics.

< previous page page_163 next page >
< previous page page_164 next page >
Page 164
Chapter 10
An Introduction to Bilingual Education
Historical Introduction
A Short History of Bilingual Education in the United
Varieties of Bilingual Education
Submersion Education
Submersion with Pull-Out Classes
Segregationist Education
Transitional Bilingual Education
Mainstream Education (with Foreign Language
Separatist Education
Immersion Bilingual Education
Developmental Maintenance and Heritage Language
Bilingual Education
Two Way/Dual Language Bilingual Education
Bilingual Education in Majority Languages

< previous page page_164 next page >
< previous page page_165 next page >
Page 165
Historical Introduction
One of the illusions about bilingual education is that it is a twentieth century phenomenon. In the US it may
appear that bilingual education was born in the 1960s. In Ireland, bilingual education is sometimes presented as a
child of the Irish Free State of 1921. The story of bilingual education in Wales often starts in 1939 with the
establishment of the first Welsh-medium primary school. The Canadian bilingual education movement is often
charted from an experimental kindergarten class set up in St Lambert, Montreal, in 1965. Despite these twentieth
century events, the historical origins of bilingual education lie well before this century.
The illusion of bilingual education as a modern phenomenon is dangerous on two counts. First, it fails to recognize
that bilingual education has existed in one form or another for 5000 years or more (Mackey, 1978). Bilingualism
and multilingualism are ‘a very early characteristic of human societies, and monolingualism a limitation induced by
some forms of social change, cultural and ethnocentric developments’ (E.G. Lewis, 1977: 22). Lewis (1977,1981)
sketches the history of bilingualism and bilingual education from the Ancient World through the Renaissance to
the modern world.
Second, there is a danger in isolating current bilingualism and bilingual education from their historical roots.
Bilingual education in the US, England and Sweden, for example, needs to be understood within the historical
context of in-migration as well as political movements such as civil rights, equality of educational opportunity and
melting pot (integrationist, assimilationist) policies. Bilingual education in Ireland and Wales can only be properly
understood by the rise of nationalism and language rights movements, for example. Bilingual education, while
isolated as a concept in this chapter, is one component inside a wider social, economic, cultural and political
framework. As Paulston (1992b: 80) observes: ‘unless we try in some way to account for the socio-historical,

< previous page page_165 next page >
< previous page page_166 next page >
Page 166
cultural, and economic-political factors which lead to certain forms of bilingual education, we will never
understand the consequences of that education’. The political context of bilingual education is considered in
Chapters 19 and 20. The history of bilingual education in the United States is now considered, with a particular
emphasis on the dynamic and ever developing nature of bilingual education policy.
A Short History of Bilingual Education in the United States
Bilingual education in the United States has moved through four overlapping periods: permissive, restrictive,
opportunist and dismissive. These periods are not neat in their divisions and, in each period, there are variations in
different States in policies and practices. There are exceptions inside these broad historical periods. However,
generally, there have been changes in the perspectives of politicians, administrators, educationalists and in school
practice that indicate that discernible shifts in ideology, preference and practice have occurred (Andersson &
Boyer, 1971; Casanova, 1992; Casanova & Arias, 1993; Crawford, 1991; Fitzgerald, 1993; Kloss, 1977; Lyons,
1990; Malakoff & Hakuta, 1990; Perlmann, 1990; Schlossman, 1983).
The Permissive Period
Long before European in-migrants arrived in the United States, the land contained a variety of native (indigenous)
languages. When the Italian, German, Dutch, French, Polish, Czech, Irish, Welsh and other in-migrant groups
arrived, there were already around two hundred, mostly mutually unintelligible languages in the United States. The
in-migrants brought with them a wide variety of languages. In the eighteenth and nineteenth centuries in the
United States, up until the first World War, linguistic diversity was generally accepted and the presence of
different languages was encouraged. Language variety was accepted as the norm and encouraged through religion,
newspapers in different languages, and in both private and public schools.
Examples of this permissive period in bilingual education in the United States are found in German-English
schools. Set up by German communities in Ohio, Pennsylvania, Missouri, Minnesota, North and South Dakota and
Wisconsin, bilingual as well as monolingual German education was accepted. Norwegian and Dutch were also
languages of instruction within ethnic based schools. This openness to in-migrant languages in the latter half of the
nineteenth century was partly motivated by competition for students between public and private schools. Other
factors such as benevolent (or uninterested) school administrators, the isolation of schools in rural areas, and ethnic
homogeneity within an area also enabled a permissive attitude to mother tongue and bilingual education during
this period. In most large cities in the latter half of the nineteenth century, English monolingual education was the
pattern. However, in cities such as Cincinnati, Baltimore, Denver and San Francisco, dual language education was
present. In some schools in Cincinnati for example, half the day was spent learning through German and the other
half of the curriculum was delivered through English.

< previous page page_166 next page >
< previous page page_167 next page >
Page 167
At the turn of the twentieth century, Italian and Jewish in-migrants were mostly placed in mainstream schools.
However, examples of bilingual education existed and were permitted. For example, some Polish in-migrants in
Chicago attended Catholic schools where a small amount of teaching was through the mother tongue. So long as
policy was within the jurisdiction of local towns and districts, the language of instruction did not become an issue
in educational provision.
The Restrictive Period
In the first two decades of the twentieth century, a change in attitude to bilingualism and bilingual education
occurred in the United States. A variety of factors are linked to this change and the subsequent restriction of
bilingual education.
The number of in-migrants increased dramatically around the turn of the century. Public schools filled their
classrooms with in-migrants. This gave rise to fears of foreigners, and the call for integration, harmonization and
assimilation of in-migrants. In-migrants lack of English language and English literacy was seen as a source of
social, political and economic concern. The call for Americanization was launched, with competence in English
becoming associated with loyalty to the United States. The Nationality Act (1906) required in-migrants to speak
English to become naturalized Americans. The call for child literacy in English rather than child labor,
socialization into a unified America rather than ethnic separation, along with increased centralized control, led to a
belief in a common language for compulsory schooling. By the turn of the century, California and New Mexico
had ‘English only’ instruction laws.
In 1919, the Americanization Department of the United States Bureau of Education adopted a resolution
recommending ‘all states to prescribe that all schools, private and public, be conducted in the English language and
that instruction in the elementary classes of all schools be in English’ (quoted in García, 1992). By 1923, thirty-
four states had decreed that English must be the sole language of instruction in all elementary schools, public and
A major influence in the movement from permissiveness to repression in bilingualism and bilingual education in
the United States came with the entry of the United States into the first World War in 1917. Anti-German feeling
in the United States spread, with a consequent extra pressure for English monolingualism and a melting pot policy
achieved through monolingual education. The German language was portrayed as posing a threat to the unity of
Americanization. Linguistic diversity was replaced by linguistic intolerance. Schools became the tool for the
assimilation and integration of diverse languages and cultures. Socialization into being American meant the
elimination of languages and cultures other than English from schools. The interest in learning foreign languages
Mandatory attendance laws in public schools were established. Such legislation made the continuation of private
schools difficult. Public funding for church-affiliated schools was eliminated.

< previous page page_167 next page >
< previous page page_168 next page >
Page 168
This period was not totally restrictive as García (1992c) reveals. She speaks of a ‘tolerant parenthesis’ between
1923 and 1942. In 1923, the US Supreme Court declared that a Nebraska state law prohibiting the teaching of a
foreign language to elementary school students was unconstitutional under the Fourteenth Amendment. This case,
known as Meyer v. Nebraska concerned a case against a teacher convicted of teaching German literacy to a 10 year
old child. The original Nebraska ruling was that such mother-tongue teaching cultivated ideas and affections
foreign to the best interests of the country. The Supreme Court, in overturning the Nebraska ruling, found that
proficiency in a foreign language was ‘not injurious to the health, morals, or understanding of the ordinary child’.
Although a ‘tolerant parenthesis’, this Supreme Court finding did not support bilingualism or bilingual education.
This is revealed in three ways: (1) the ruling included a requirement for English language instruction in all schools
(2) it commented that the desire of a state legislature to foster a homogeneous people was ‘easy to appreciate’ and
(3) the finding is symbolic of a shift in attitude to non-English languages (García, 1992c). No longer were such
languages ‘ethnic’ but ‘foreign’. Such languages were no longer seen as spoken by ethnolinguistic minorities in the
US, but by ‘foreigners’ (i.e. non-US or unnaturalized citizens).
This restrictive movement continued at least until the 1960s, and has carried on being one major theme in US
politics until the present. This theme is considered further in Chapter 19.
The Period of Opportunity
In 1957, the Russians launched their Sputnik into space. For United States politicians and public, a period of soul-
searching led to debates about the quality of US education, US scientific creativity and US competence to compete
in an increasingly international world. Doubts arose about the hitherto over-riding concern with English as the
melting-pot language, and a new consciousness was aroused about the need for foreign language instruction. In
1958, the National Defense and Education Act was passed, promoting foreign language learning in elementary
schools, high schools and universities. This in turn helped to create a slightly more tolerant attitude to languages
other than English spoken among ethnic groups in the US.
In the United States in the 1960s, various other factors allowed a few opportunities to bring back bilingual
education, albeit in a disparate, semi-isolated manner. The ‘opportunity’ movement for bilingual schools to be re-
established in the United States needs to be understood in the wider perspective of the Civil Rights movement, the
concern for the rights of African-Americans, and the need to establish equality of opportunity and equality of
educational opportunity for all people, irrespective of race, color or creed. The 1964 Civil Rights Act prohibited
discrimination on the basis of color, race or national origin, and led to the establishment of the Office of Civil
Rights. This Act is an important marker that symbolizes the beginning of a change in a less negative attitude to
ethnic groups, and possibilities for increased integration and tolerance of ethnic languages. However, the
restoration of the practice of bilingual education in the US is often regarded as starting in 1963, in one school in

< previous page page_168 next page >
< previous page page_169 next page >
Page 169
In 1963, Cuban exiles established a dual language school (Coral Way Elementary School) in Dade County in
Florida. Believing they were only in exile for a short period, the educated, middle class Cubans set up this dual
language (Spanish-English) bilingual school. The need to maintain their mother tongue of Spanish was aided by
having highly trained professional teachers ready to service such schools, the Cubans plight as victims of a harsh
Communist state, and their expected temporary stay in the United States. Their unquestioned loyalty to United
States policies and democratic politics gained sympathy for the Cubans. Bilingual education in Dade County
received both political support and funding. More details of this movement are found in the section on dual
language schools.
While the re-establishment of bilingual schools in the US has benefited from the example and success of Coral
Way Elementary School, an understanding of bilingual education in the United States requires a comprehension of
legislation and lawsuits.
In 1967, a Texas Senator, Ralph Yarborough, introduced a Bilingual Education Act as an amendment of the 1965
Elementary and Secondary Education Act. The legislation was designed to help mother tongue Spanish speakers
who were seen as failing in the school system. Enacted in 1968 as Title VII of the Elementary and Secondary
Education Act, the Bilingual Education Act indicated that bilingual education programs were to be seen as part of
federal educational policy. It authorized the use of federal funds for the education of speakers of languages other
than English. It also undermined the English-only legislation still lawful in many states. However, the 1968
Bilingual Education Act allocated funds for such minority language speakers while they shifted to working
through English in the classroom. The underlying aim was a transition from a minority language (e.g. Spanish) to
English, rather than support for the mother tongue.
A landmark in United States’ bilingual education was a lawsuit. A court-case was brought on behalf of Chinese
students against the San Francisco School District in 1970. The case concerned whether or not non-English
speaking students received equal educational opportunities when instructed in a language they could not
understand. The failure to provide bilingual education was alleged to violate both the equal protection clause of
the 14th Amendment and Title VI of the Civil Rights Act of 1964. The case, known as Lau versus Nichols, was
rejected by the federal district court and a court of appeals, but was accepted by the Supreme Court in 1974. The
verdict outlawed English submersion programs for language minority children and resulted in nationwide ‘Lau
remedies’. The Lau remedies acknowledged that students not proficient in English needed help. Such remedies
included English as a Second Language classes, English tutoring and some form of bilingual education. The Lau
remedies created some expansion in the temporary use of minority languages in schools. They rarely resulted in
heritage language, enrichment or maintenance programs as the accent was still on a transitional use of the home
language for English language learners. The Lau court case is symbolic of the dynamic and continuing contest to
establish language rights in the US particularly through testing the law in the courtroom (Casanova, 1991; Hakuta,
1986; Lyons, 1990).
In this period of opportunity, the kind of bilingual education needed to achieve equality of educational opportunity
for language minority children was not defined. However, the

< previous page page_169 next page >
< previous page page_170 next page >
Page 170
right to equal opportunity for language minorities was asserted. The means of achieving that right was not
declared. During this period, there was a modest growth in developmental maintenance bilingual education and
ethnic community mother tongue schools. These schools are considered in separate sections on heritage language
bilingual education and dual language schools.
The Dismissive Period
While bilingual education has not disappeared in the United States in the last two decades, since the 1980s there
have been both central moves against an emergence of a strong version of bilingual education, a preference for
submersion and transitional bilingual education (see later in this chapter), and the rise of pressure groups such as
English First and US English that seek to establish English monolingualism and cultural assimilation.
Consideration of such political movements is found in Chapter 19.
To understand the current dismissive or repudiation period, it is helpful to examine the legislative changes with
respect to bilingual education in the United States. We return to the 1968 Bilingual Education Act (Title VII and
part of the Elementary and Secondary Education Act). This provided a ‘poverty program’ for the educationally
disadvantaged among language minorities. It did not require schools to use a child’s home language other than
English. However, it did allow a few opportunists to bring ‘home languages’ into the classroom rather than exclude
them. The 1974 amendments to this Bilingual Education Act (see the chronology table) required schools receiving
grants to include teaching in a student’s home language and culture so as to allow the child to progress effectively
through the educational system. Effective progress in student achievement could occur via the home language or
via English. However, this gave rise to fierce debates about how much a student’s native language should be used
in school. Some argued that it was essential to develop a child’s speaking and literacy skills in their native
language before English was introduced in a major way. Others argued that educational equality of opportunity
could best be realized by teaching English as early as possible and assimilating language minority children into
mainstream culture.
In 1978, the United States Congress re-authorized Transitional Bilingual Education, allowing the native language
to be used only to the extent necessary for a child to achieve competence in the English language. Title VII funds
could not be used for Maintenance Bilingual Education programs. The 1984 and 1988 amendments allowed
increasing percentages of the funds available to be allocated to programs where a students’ first language was not
The Reagan administration was particularly hostile to bilingual education. In the New York Times on the 3rd
March 1981, President Reagan is quoted as saying that ‘It is absolutely wrong and against the American concept to
have a bilingual education program that is now openly, admittedly, dedicated to preserving their native language
and never getting them adequate in English so they can go out into the job market’. Reagan believed that
preservation of the native language meant neglect of English language acquisition. Bilingual education programs
were seen as serving to neglect English language compe-

< previous page page_170 next page >
< previous page page_171 next page >
Page 171
tence in students. Reagan dismissed bilingual education in favor of submersion and transitional programs.
In 1985, William Bennett, the new Secretary of Education, suggested that there was no evidence that children from
language minorities (whom the Bilingual Education Act had sought to help), had benefited from this Act. Some
25% of funds were made available for English monolingual, alternative instructional programs (e.g. Structured
English programs; Sheltered English programs). This represented a further political dismissal of education through
the minority language and a dismissal of ‘strong’ forms of bilingual education.
The Lau remedies were withdrawn by the Reagan government and do not have the force of law. The federal
government left local politicians to create their own policies. Further changes in the rights to United States
bilingual education are given in the table below. Legislation and litigation has often led to ‘weak’ forms of
bilingual education (e.g. transitional bilingual education). Also, recent legislation has not tended to increase rights
to a bilingual education. In the Reagan and Bush years in the United States Presidency, the accent was more on
submersion and transitional bilingual education. The right to early education through a minority language failed to
blossom in those years.
This section has suggested that bilingual education in the United States has moved through four overlapping
periods: permissive, restrictive, opportunist and dismissive. These broad historical trends contain plenty of
exceptions and variations within each period.
Two conclusions. First, there is common perception that educational policy is often static, always conservative and
very slow to change. The history of bilingual education in the United States tends to falsify and contradict such
beliefs. Such history shows that there is constant change, a constant movement in ideas, ideology and impetus.
There is action and reaction, movement and contra-movement, assertion and response. One conclusion is that
change will always occur in bilingual education policy and provision. Nothing is static. While there will be periods
when bilingual education is criticized, forbidden and rejected, there will be reactions, with the possibility of more
positive, accepting periods ahead. There is no certainty in the future history of bilingual education, only
uncertainty and change. Yet uncertainty and change provide occasional opportunities for bilingual education to
Second, the conclusion must not be that the four overlapping periods (permissive, restrictive, opportunist and
dismissive) will always operate in that sequence. The history of bilingual education in Wales follows a different
sequence. In Wales, the last 200 years of bilingual education have witnessed a movement through the following
five approximate periods: permissive; dismissive; suppressive; opportunist and progressive. From a time when
Welsh was banned in the classroom (suppressive period), there is currently a widespread acceptance, and provision
of bilingual education in Wales (progressive period). No universal patterns of change can or should be deduced
from either the United States or the Welsh experience. Such unpredictability provides a challenge to bilingual

< previous page page_171 next page >
< previous page page_172 next page >
Page 172
A summary of major events affecting the history of United States bilingual education this century is given in the
table below.
YearUnited States,
affecting Bilingual
1906Nationality Act passed First legislation requiring in-migrants to
speak English to become naturalized.
1923Supreme Court in
Meyer v. Nebraska
Rules that state provision requiring
English language instruction is permitted
by the Constitution, but proficiency in a
foreign language was also constitutional.
1950Amendments to the
Nationality Act
English literacy required for
1954Brown v. Board of
Segregated education based on race made
1958National Defense and
Education Act
Promotion of foreign language learning.
1965Elementary and
Secondary Education
Act (ESEA)
Funds granted to meet the needs of
‘educationally deprived children’.
1968Elementary and
Secondary Education
Act (ESEA)
amendment: The
Bilingual Education
Act, Title VII
Provided funding to establish bilingual
programs for students who did not speak
English and who were economically
1974Lau v. Nichols Established that language programs for
language minorities not proficient in
English were necessary to provide equal
educational opportunities.
1974Reauthorization of
Bilingual Education Act
Title VII of ESEA
Grants available to include native
language instruction in schools to ‘effect
progress’ of students not proficient in
English. Bilingual Education was defined
as transitional (TBE).
1976Keyes v. School District
no. 1,
Denver, Colorado
Established bilingual education as
compatible with desegregation.
1977National Clearinghouse
for Bilingual Education
is established
To provide a national information center
for bilingual education.
1978Reauthorization of
Bilingual Education Act
Title VII of ESEA
Reauthorized funding for bilingual
programs (transitional). Poverty criterion
for eligibility was removed. The term
‘Limited English Proficient’ (LEP)
1983US English Movement
Debates about the dominant place of
English in law, society and education
became more prominent.
Varieties of Bilingual Education
So far in this chapter, the term bilingual education has been used as if its meaning is unambiguous and self-
evident. The opposite is the case. Bilingual education is ‘a simple label for a complex phenomenon’ (Cazden &
Snow, 1990a). At the outset, a distinction needs making between education that uses and promotes two languages
and education

< previous page page_172 next page >
< previous page page_173 next page >
Page 173
for language minority children. This is a difference between a classroom where formal instruction is to foster
bilingualism and a classroom where bilingual children are present, but bilingualism is not fostered in the
curriculum. The umbrella term, bilingual education, refers to both situations leaving the term ambiguous and
imprecise. Precision can be attempted by specifying the major types of bilingual education.
One early and detailed classification of bilingual education is by Mackey (1970). This account of 90 different
patterns of bilingual schooling considers: the languages of the home; the languages of the curriculum; the
languages of the community in which the school is located and the international and regional status of the
languages. A different approach to categorizing types of bilingual education is to examine the aims of such
education. A frequent distinction in aims is between transitional and maintenance Bilingual Education (Fishman,
1976; Hornberger, 1991).
Transitional Bilingual Education aims to shift the child from the home, minority language to the dominant,
majority language. Social and cultural assimilation into the language majority is the underlying aim. Maintenance
Bilingual Education attempts to foster the minority language in the child, strengthening the child’s sense of cultural
identity and affirming the rights of an ethnic minority group in a nation. Otheguy & Otto (1980) make the
distinction between the different aims of static maintenance and developmental maintenance. Static maintenance
aims to maintain language skills at the level of the child entering a school. Developmental maintenance seeks to
develop a student’s home language skills to full proficiency and full biliteracy or literacy. This is sometimes
referred to as Enrichment Bilingual Education for language minority children. (The term ‘Enrichment Bilingual
Education’ is also used for language majority children who are adding a second language in school). Static
maintenance attempts to prevent home language loss but not increase skills in that first language. Developmental
maintenance has a ‘goal of proficiency and literacy in the home language equal to English’ (Otheguy & Otto, 1980:
351). Enrichment Bilingual Education aims to go beyond static maintenance to extending the individual and group
use of minority languages, leading to cultural pluralism (see Chapter 20) and to the social autonomy of an ethnic
Ferguson, Houghton & Wells (1977) widen these distinctions and provide ten examples of varying aims of
bilingual education:
(1) To assimilate individuals or groups into the mainstream of society; to socialize people for full participation in
the community.
(2) To unify a multilingual society; to bring unity to a multi-ethnic, multi-tribal, or multi-national linguistically
diverse state.
(3) To enable people to communicate with the outside world.
(4) To provide language skills which are marketable, aiding employment and status.
(5) To preserve ethnic and religious identity.
(6) To reconcile and mediate between different linguistic and political communities.
(7) To spread the use of a colonializing language, socializing an entire population to a colonial existence.
(8) To strengthen elite groups and preserve their position in society.

< previous page page_173 next page >
< previous page page_174 next page >
Page 174
(9) To give equal status in law to languages of unequal status in daily life.
(10) To deepen understanding of language and culture.
This list shows that bilingual education does not necessarily concern the balanced use of two languages in the
classroom. Behind bilingual education are varying and conflicting philosophies of what education is for.
Sociocultural, political and economic issues are ever present in the debate over the provision of bilingual
education. This will be addressed in Chapter 19.
The typology of education that is adopted in this chapter is presented in the table on the following page. Ten types
of language education are portrayed.
The ten different types of program have multitudinous sub-varieties, as Mackey’s (1970) 90 varieties of bilingual
education indicate. One of the intrinsic limitations of typologies is that not all real-life examples will fit easily into
the classification. For example, elite ‘finishing schools’ in Switzerland, and classrooms in Wales where first
language Welsh speakers are taught alongside ‘immersion’ English first language speakers make classification
essentially simplistic, although necessary for discussion and understanding. Thus typologies have value for
conceptual clarity but they have limitations: (1) models suggest static systems whereas bilingual schools and
classrooms constantly develop and evolve (as the short history of bilingual education in the US revealed); (2):
there are wide and many variations within a model; (3) models address ‘inputs’ and ‘outputs’ of the education
system, but rarely address the process (see Chapter 18); (4) models do not explain the success or failure or the
relative effectiveness of bilingual education.
Each of ten broad types of program will now be briefly considered.
Submersion Education
A water metaphor is present in the idea of submersion education. The analogy is of a swimming pool. Rather than
a quick dip into a second language in mainstream education, submersion contains the idea of a student thrown into
the deep end and expected to learn to swim as quickly as possible without the help of floats or special swimming
lessons. The language of the pool will be the majority language (e.g. English in the US) and not the home
language of the child (e.g. Spanish). The language minority student will be taught all day in the majority language
alongside fluent speakers of the majority language. Both teachers and students will be expected to use only the
majority language in the classroom, not the home language. Students may either sink, struggle or swim.
Submersion Education is the label to describe education for language minority children who are placed in
mainstream education. In the US, such a submersion experience is also found in ‘Structured Immersion’ programs.
Structured Immersion programs contain only language minority children and no language majority children. As
will be apparent later in this chapter, the language experience in a Structured Immersion program is submersion
rather than immersion. Structured Immersion programs are thus for minority language speakers conducted in the
majority language. The first language is not developed but is replaced by the majority language. Different from
Submersion Education, the

< previous page page_174 next page >
< previous page page_175 next page >
Page 175
Type of Program Typical
Type of
Language of
Societal and
Aim in
Assimilation Monolingualism
Withdrawal Classes
with ‘Pull-
out’ L2
Assimilation Monolingualism
(forced, no
Apartheid Monolingualism
Moves from
Minority to
Assimilation Relative
with Foreign
Language Teaching
with L2/FL
(out of
Type of Program Typical
Type of
Language of
Societal and
Aim in
with Initial
on L2
Pluralism and
Bilingualism &
on L1
Pluralism and
Bilingualism &
Pluralism and
Bilingualism &
Pluralism and
Bilingualism &
Notes: (1) L2 = Second Language; L1 = First Language; FL = Foreign
(2) Formulation of this table owes much to discussions with Professor
Ofelia García.

< previous page page_175 next page >
< previous page page_176 next page >
Page 176
Structured Immersion teacher will use a simplified form of the majority language, and may initially accept
contributions from children in their home language (Hornberger, 1991). In the US, there are also ‘Sheltered
English’ programs, an alternative to ESL (English as a Second Language) programs. However, since such
programs can be ‘pull-out’ with special content and curriculum materials, they are discussed in the next section.
There are various criticisms of a submersion form of education, including its variants. In terms of the language
garden analogy (see Chapter 3), one blossoming flower is displaced. The garden is re-seeded after the initial
language plants have shown growth. One tender but healthy plant is simply replaced by the majority flower of the
garden. The basic aim of submersion education is thus assimilation of language minority speakers, particularly
where there has been in-migration (e.g. US, England). Also, where indigenous language minorities are perceived
as working against the common good, submersion education becomes a tool of integration. The school becomes a
melting pot to help create common social, political and economic ideals. As Theodore Roosevelt urged in 1917:
‘We must have but one flag. We must have but one language. That must be the language of the Declaration
of Independence, of Washington’s Farewell Address, of Lincoln’s Gettysburg Speech and Second Inaugural.
We cannot tolerate any attempt to oppose or supplant the language and culture that has come down to us
from the builders of the republic with the language and culture of any European country. The greatness of
this nation depends on the swift assimilation of the aliens she welcomes to her shores. Any force which
attempts to retard that assimilative process is a force hostile to the highest interests of our country’ (quoted in
Wagner, 1980:32).
Language diversity has often been discouraged in the US so, it was argued, that harmony would hasten a healthy
and homogeneous nation. A common language would provide, it was thought, common attitudes, aims and values.
A common language and culture would cement society. A God-blest-English speaking America was preferable to
the threat of Babel.
Considerable variations of language skill in a classroom may often create problems in teaching and class
management for the teacher. With students who range from fluent majority language speakers to those who can
understand little classroom talk, the burden on the teacher may be great. In such formal ‘context-reduced’
classrooms, there is no reason to assume that children will quickly and effortlessly acquire the majority language
skills necessary to cope in the curriculum. Alongside problems of language, there are likely to be problems of
social and emotional adjustment for language minority children which tend to have connections with later drop-out
rates from high school. The child, the parents, the home language and culture appear to be disparaged. McKay
(1988: 341) quotes from a student in a Submersion classroom:
‘School was a nightmare. I dreaded going to school and facing my classmates and teacher. Every activity the
class engaged in meant another exhibition of my

< previous page page_176 next page >
< previous page page_177 next page >
Page 177
incompetence. Each activity was another incidence for my peers to laugh and ridicule me with and for my
teacher to stare hopelessly disappointed at me. My self-image was a serious inferiority complex. I became
frustrated at not being able to do anything right. I felt like giving up the entire mess.’
Skutnabb-Kangas (1981) indicates the stresses of learning through an undeveloped language as in submersion
education. Listening to a new language demands high concentration, it is tiring, with a constant pressure to think
about the form of the language and less time to think about curriculum content. A child has to take in information
from different curriculum areas and learn a language at the same time. Stress, lack of self confidence, ‘opting-out’,
disaffection and alienation may occur.
Submersion with Pull-Out Classes
Submersion Education may occur with or without the addition of withdrawal classes or ‘pull-out’ classes to teach
the majority language. Language minority children in mainstream schools may be withdrawn for ‘compensatory’
lessons in the majority language (e.g. English as a second language (ESL) pull-out programs in the US and
England). Such Withdrawal Classes are provided as a way of keeping language minority children in mainstream
schooling. However, ‘withdrawn’ children may fall behind on curriculum content delivered to others not in
withdrawal classes. There may also be a stigma for absence. A withdrawal child may be seen by peers as
‘remedial’, ‘disabled’ or ‘limited in English’. Other Submersion schools will have no withdrawal classes to help
students develop skills in the school language. Withdrawal classes are administratively simple and require little or
no additional expense. For many administrators and budget managers, Submersion creates ease of supervision and
financial management.
A variation under this heading is Sheltered English or Sheltered Content Teaching, where minority language
students are taught the curriculum with a simplified vocabulary, but also, purpose-made materials and methods
(such as cooperative learning) are usedbut in English only (Faltis, 1993b). In Sheltered English or Sheltered
Content Teaching, content and curriculum materials are developed and pitched to match the English proficiency of
the student (Faltis, 1993b). Sheltered English or Sheltered Content English is distinct from ESL (English as a
Second Language), as in ESL, English is taught as a language, with a focus on learning that language.
Sheltered Content Teaching may involve temporary segregation from first language English speakers. Such
segregation may produce: (1) greater opportunity for participation among students (they may be less inhibited due
to no competition or comparisons with first language speakers of English); (2) greater sensitivity among teachers
to the linguistic, cultural and educational needs of a homogeneous group of students; and (3) a collective identity
among students in a similar situation (Faltis, 1993b). However, such language segregation removes first language
role models; may produce a social isolation with overtones of stigmatization and reinforce negative stereotypes;
encouraging the labeling of segregated students as linguistically and educationally inferior, deprived and in need of
remedial attention; and may generate inequality in treatment (e.g. in curriculum materials and training of teachers).

< previous page page_177 next page >
< previous page page_178 next page >
Page 178
Segregationist Education
A form of ‘minority language only’ education is segregationist language education (Skutnabb-Kangas, 1981).
Monolingual education through the medium of the minority language can be for apartheid (e.g. educating the
Bantus only in their native language). The ruling elite prescribes education solely in the minority language to
maintain subservience and segregation. Such a language minority ‘do not learn enough of the power language to be
able to influence the society or, especially, to acquire a common language with the other subordinated groups, a
shared medium of communication and analysis’ (Skutnabb-Kangas, 1981: 128). Segregationist education forces a
monolingual language policy on the relatively powerless.
Transitional Bilingual Education
Transitional bilingual education is the most common type of bilingual education in the United States and the form
most supported by Title VII funds. The aim of transitional bilingual education is assimilationist. It differs from
submersion education by language minority students temporarily being allowed to use their home language, and
often being taught through their home language, until they are thought to be proficient enough in the majority
language to cope in mainstream education. Thus, transitional education is a brief, temporary swim in one pool until
the child is perceived as capable of using the four language strokes in the mainstream pool. The aim is to increase
use of the majority language in the classroom while proportionately decreasing the use of the home language in
the classroom.
The educational rationale is based on a question of perceived priorities: children need to function in the majority
language in society. The argument used is that if competency in the majority language is not quickly established,
such children may fall behind their majority language peers. Thus, arguments about equality of opportunity and
maximizing student performance are used to justify such transitional programs. The extent to which such
justifications are valid is considered later.
Transitional Bilingual Education (TBE) can be split into two major types: early exit and late exit (Ramirez &
Merino, 1990). Early exit TBE refers to two years maximum help using the mother tongue. Late exit TBE often
allows around 40% of classroom teaching in the mother tongue until the 6th grade.
While majority language monolingualism is the aim of transitional bilingual education, teachers or their assistants
need to be bilingual. The temporary home language swim requires, for example, a Spanish speaking teacher who
may be more sensitive and successful in teaching English to Spanish speaking children than English-only teachers.
The former can switch from one language to another and be more sympathetic to the language of the children. For
the government and school administrator, such bilingual teachers may become valuable allies or Trojan horses.
A bilingual teacher can become the unwitting promoter of transition from one language to another, and
assimilation into the majority culture. Such teachers can promote the transition from home to school language from
within their own cultural group. An

< previous page page_178 next page >
< previous page page_179 next page >
Page 179
Anglo teacher is thus not imposed on the Hispanic language minority; the conversion job is achieved by a
bilingual teacher. Similarly, in US schools with large transitional programs, there is sometimes a hidden message
in the staffing. The people with power and prestige, such as principals and assistant principals, are often English
monolinguals (García, 1993). The people with the least power and status are cooks, lunch monitors, cleaners and
janitors, who, for example, speak a minority language such as Spanish. The language of formal announcements is
English; the language used by those who serve may be Spanish. The language to learn is the language of power
and prestige. The language to forget is the language of servitude, stigma and shame. This message is loudly
proclaimed without words.
However, such Hispanic teachers may alternatively recognize the needs and wishes of their own communities.
Such communities may desire their young to speak English early in schooling, but differ from administrators in
wanting to preserve Spanish (García, 1991a). Hispanic teachers may continue to teach English in transitional
bilingual education, but also try to preserve Spanish in the children, becoming allies of the community and not just
allies of politicians and bureaucrats.
Mainstream Education (with Foreign Language Teaching)
In the US, Australasia, Canada and much of Europe, most language majority schoolchildren take their education
through their home language. For example, a child whose parents are English speaking monolinguals attends
school where English is the teaching medium (although some second (foreign) language teaching may occur). In
Canada, this would be called a core program. In Wales and elsewhere, it is sometimes called a ‘drip-feed’ language
program. The term ‘drip-feed’ highlights the kind of language element in mainstream schooling. Second (foreign)
language lessons of half an hour per day may constitute the sole ‘other’ language diet. Drip-feeding French,
German, Japanese, Russian, Spanish makes the language a subject in the curriculum similar to History, Science,
Geography and Maths. This is distinct from teaching through the medium of a second language where curriculum
content is the main focus rather than language learning (sometimes called embedding).
The problem in some countries (e.g. US, England) is that only a few second language flowers blossom. Too often,
the tender second language shoot quickly withers and fades. Where children receive a half an hour second
language lesson per day for between five and twelve years, few students become functionally fluent in the second
language. LeBlanc (1992) poses the critical question. ‘We all know how much our country [Canada] invests in
second-language training. We are talking in terms of millions and millions of dollars . . . All these students are
taking second-language courses and, once they have finished, should normally be able to function in the second
language. But what happens in reality?’ (p. 35).
The Canadians found that after 12 years of French drip-feed language teaching, many English-background
students were not fluent enough to communicate in French with French Canadians. Similarly in Britain, five years
of French or German or Spanish in secondary school (age 11 to 16) results in only a few growing in a second
language. For

< previous page page_179 next page >
< previous page page_180 next page >
Page 180
the great majority, the second language quickly shrivels and dies. Mainstream education rarely produces fully
bilingual children. A very limited form of fluency in a foreign language tends to be the typical outcome for the
mass of the language majority.
This is not the only outcome of second and foreign language teaching. The learning of English in Scandinavia does
not fit this pattern, with many learners becoming fluent in English. When motivation is high, when economic
circumstances encourage the acquisition of a trading language, then foreign language teaching may be more
Separatist Education
A more narrow view of language minority education would be to choose to foster monolingualism in the minority
language. The aims are minority language monolingualism and monoculturalism in a context where such choice is
self-determined. Schermerhom (1970) called this a secessionist movement where a language minority aims to
detach itself from the language majority to pursue an independent existence. As a way of trying to protect a
minority language from being over-run by the language majority, or for political, religious or cultural reasons,
separatist minority language education may be promoted. This type of education may be organized by the language
community for its own survival and for self-protection.
It is unlikely that a school would formally state its aims in a linguistic separatist fashion. Rather, in the implicit
functioning of isolationist religious schools and the political rhetoric of extreme language activists, the ‘separatist’
idea of such schools exists. Small in number, the importance of this category is that it highlights that language
minority education is capable of moving from the goal of pluralism to separatism.
Immersion Bilingual Education
Submersion, Withdrawal Classes and Transitional approaches are often given the title of bilingual education. This
is because such schemes contain bilingual children. This counts as a ‘weak’ use of the term bilingual education
because bilingualism is not fostered in school. Such education does not by aim, content or structure have
bilingualism as a defined outcome. The next form, immersion education, has bilingualism as an intended outcome,
and therefore represents a ‘strong’ use of the term bilingual education.
Immersion bilingual education derives from Canadian educational experiments. The immersion movement started
in St Lambert, Montreal, in 1965 (Lambert & Tucker, 1972; Rebuffot, 1993). Some disgruntled English speaking,
middle-class parents persuaded school district administrators to set up an experimental kindergarten class of 26
children. The aims were for students (1) to become competent to speak, read and write in French; (2) to reach
normal achievement levels throughout the curriculum including the English language; (3) to appreciate the
traditions and culture of French speaking Canadians as well as English speaking Canadians. In short, the aims
were for children to become bilingual and bicultural without loss of achievement.

< previous page page_180 next page >
< previous page page_181 next page >
Page 181
Types of Immersion Bilingual Education
Immersion education is an umbrella term. Within the concept of immersion experience are various Canadian
programs differing in terms of the following:
age at which a child commences the experience. This may be at the kindergarten or infant stage (early immersion);
at nine to ten years old (delayed or middle immersion), or at secondary level (late immersion);
amount of time spent in immersion. Total immersion usually commences with 100% immersion (e.g. per week) in
the second language, after two or three years reducing to 80% per week for the next three or four years, finishing
junior schooling with approximately 50% immersion in French per week. Partial immersion provides close to 50%
immersion in the second language throughout infant and junior schooling.
Early Total Immersion is the most popular entry level program, followed by late and then middle immersion
(Canadian Education Association, 1992). The histograms on the following pages illustrate several possibilities,
with many other variations around these.
The St Lambert experiment suggested that the aims were met. Attitudes and achievement were not hindered by the
immersion experience. Tucker & d’Anglejan (1972: 19), summarized the outcomes as follows:
‘the experimental students appear to be able to read, write, speak, understand, and use English as well as
youngsters instructed in English in the conventional manner. In addition and at no cost they can also read,
write, speak and understand French in a way that English students who follow a traditional program of
French as a second language never do.’
Since 1965, immersion bilingual education has spread rapidly in Canada (Rebuffot, 1993). There are currently
around 300,000 English speaking Canadian children in some 2000 French immersion schools. This represents
some 6% of the total school population in Canada. What are the essential features of this speedy educational
growth? First, immersion in Canada aims at bilingualism in two prestigious, majority languages (French and
English). This relates to an additive bilingual situation. Such a situation is different from the wrongly termed
‘immersion’ or ‘structured immersion’ of children from language minority backgrounds in the majority language
(e.g. Spanish speakers in the US). Use of the term ‘immersion’ in such a subtractive, assimilationist situation is best
avoided. Submersion is a more appropriate term.
Second, immersion bilingual education in Canada has been optional not compulsory. Parents choose to send their
children to such schools. The convictions of parents plus the commitment of the teachers may aid the motivation
of students. Immersion appears to thrive on conviction not on conformity. Third, children in early immersion are
often allowed to use their home language for up to one and a half years for classroom communication. There is no
compulsion to speak French in the playground or dining hall. The child’s home language is appreciated and not
belittled. Fourth, the teachers are

< previous page page_181 next page >
< previous page page_182 next page >
Page 182
Early Total Immersion Bilingual Education
Early Partial Immersion Bilingual Education

< previous page page_182 next page >
< previous page page_183 next page >
Page 183
Delayed Immersion Bilingual Education
Late Immersion Bilingual Education

< previous page page_183 next page >
< previous page page_184 next page >
Page 184
competent bilinguals. However, they initially appear to the children as able to speak French but only understand
(and not speak) English.
Classroom language communication aims to be meaningful, authentic and relevant to the child’s needs; not
contrived, tightly controlled or repetitive. The content of the curriculum becomes the focus for the language.
Perpetual insistence on correct communication is avoided. Learning second language French in early immersion
becomes incidental and unconscious, similar to the way a first language is acquired. Emphasis is placed on
understanding French before speaking French. Later on, formal instruction may occur (see Chapter 17).
Fifth, the students start immersion education with a similar lack of experience of the second language. Most are
monolingual. Students commencing schooling with relatively homogeneous language skills not only simplifies the
teacher’s task, it also means that students’ self esteem and classroom motivation is not at risk due to some students
being linguistically more expert. Sixth, students in immersion education experience the same curriculum as
mainstream ‘core’ students.
With over a 1000 research studies, Canadian immersion bilingual education has been an educational experiment of
unusual success and growth. It has influenced bilingual education in Europe and beyond. For example, with
variations to suit regional and national contexts, the Catalans and Basques (Artigal, 1991, 1993, 1995; Bel, 1993)
Finnish (Laurén, 1994; Manzer, 1993), Japanese (Oka, 1994), Australians (Berthold, 1992, 1995; Caldwell &
Berthold, 1995), the Gaelic speakers in Scotland (MacNeil, 1994), Swiss (Stotz & Andres, 1990) and the Welsh
and Irish (Baker, 1988, 1993) have emulated the experiment with similar success. In Catalonia, research indicates
that Spanish speaking children who follow an immersion program not only become fluent in Catalan, but also their
Spanish does not suffer. Throughout the curriculum, such Catalan immersion children ‘perform as well and
sometimes better than their Hispanophone peers who do not’ [follow an immersion program] (Artigal, 1993: 4041).
Similarly, the EIFE studies in the Basque Country show that their Model B immersion program (50% Basque and
50% Spanish) has successful outcomes in bilingual proficiency (Sierra & Olaziregi, 1989).
Developmental Maintenance and Heritage Language Bilingual Education
Alongside immersion education, there is another form of bilingual education that merits a ‘strong’ use of the term.
This occurs where language minority children use their native, ethnic, home or heritage language in the school as a
medium of instruction and the goal is full bilingualism. Examples include education through, or more often partly
through, the medium of Navajo and Spanish in the US, Catalan in Spain, Ukrainian in Canada, Gaelic in Scotland,
Finnish in Sweden and Welsh in Wales. The child’s native language is protected and developed alongside
development in the majority language. In New Zealand, the Maori language has increasingly been promoted in
schools (Spolsky, 1989a) with some indications of positive outcomes in achievement (Jacques & Hamlin, 1992). A
similar promotion of Aboriginal languages is occurring in heritage language education in Australia (Caldwell &
Berthold, 1995). In Ireland, Irish medium education is often available for children from Irish language
backgrounds. While the overall language

< previous page page_184 next page >
< previous page page_185 next page >
Page 185
garden will contain English and Irish and possibly other European languages, the indigenous language flower of
Ireland is protected in school lest it wither and die amongst the pervading growth of English, and lately of other
European majority languages.
In the US, there are schools that may be termed ‘ethnic community mother tongue schools’ (Fishman, 1989).
Numbering over 5000 and located in every State of the US, the list includes schools using the mother tongue of the
following varied communities: Arabs, Africans, Asians, French, German, Greek, Haitian, Jewish, Russian, Polish,
Japanese, Latin American, Armenian, Dutch, Bulgarian, Irish, Rumanian, Serbian and Turkish. Maintained by
communities who have lost or are losing their ‘native’ language, the schools mostly teach that native language and
use it as a medium of instruction. Other schools are supported by foreign governments and religious organizations.
‘These schools must be recognized as filling an important identity-forming and identity-providing function for
millions of Americans’ (Fishman, 1989: 454).
Since they are fee paying, private schools, the students tend to come from middle class or the relatively more
affluent working class backgrounds and achieve biliteracy and biculturalism with some success (García, 1988). For
example, there are around 130 Hebrew day schools in New York alone, ‘successful in teaching children to read and
write English, Hebrew, and many times Yiddish’ (García, 1988: 22).
The ‘public school’ US example is called Maintenance Bilingual Education or Developmental Maintenance
Bilingual Education. These are relatively few in number with Navajo education being one major example (Holm &
Holm, 1990). In Canada, the term used to describe such education is Heritage Language Education. However, in
Canada, there is a distinction between Heritage Language lessons and Heritage Language Bilingual Education. (1)
Heritage Language Programs give around two and a half hours per week language teaching, currently in more than
60 languages to about 100,000 students. These lessons often occur during dinner-hours, after school and at
weekends. (2) In provinces such as Manitoba, British Columbia, Saskatchewan and Alberta, there are Heritage
Language Bilingual Education Programs (e.g. see Benyon & Toohey (1991) on programs in British Columbia).
The heritage language is the medium of instruction for about 50% of the day (e.g. Ukrainian, Italian, German,
Hebrew, Yiddish, Mandarin Chinese, Arabic and Polish; see Cummins, 1992).
In essence, heritage or maintenance language education refers to the education of language minority children
through their minority and majority language. In most cases, the majority language will be present in the
curriculum, ranging from second language lessons to a varying proportion of the curriculum being taught the
majority language.
The term ‘heritage language’ may also be called ‘native language’, ‘ethnic language’, ‘minority language’, ‘ancestral
language’, aboriginal language, or, in French, ‘langues d’origine’. The danger of the term ‘heritage’ is that it points
to the past and not the future, to traditions rather than the contemporary. Partly for this reason, the British term
tends to be ‘community language’ or ‘where English is an additional language’. The heritage language may or may
not be an indigenous language. Both Navajo and Spanish can be

< previous page page_185 next page >
< previous page page_186 next page >
Page 186
perceived as heritage languages in the US depending on an individual’s perception of what constitutes their
heritage language.
Developmental Maintenance programs in the US (also Heritage language programs in Canada and other countries)
vary in structure and content. Some of the likely features are as follows:
(1) Most, but not necessarily all of the children will come from language minority homes. At the same time, the
minority language may be the majority language of a local community. In certain areas of the US, Spanish
speakers are in a majority in their neighborhood or community. In Gwynedd, North Wales, where the minority
language (Welsh) is often the majority language of the community, heritage language programs are prevalent. The
children will be joined in most programs by a number of majority language children (e.g. English monolinguals in
Wales taking their education through Welsh).
(2) Parents will often have the choice of sending their children to mainstream schools or to heritage language
education. Ukrainian, Jewish and Mohawkian heritage language programs in Canada, for example, gave parents
freedom of choice in selecting schools.
(3) The language minority student’s home language will often be used for approximately half or more of
curriculum time. The Ukrainian programs in Alberta and Manitoba allotted half the time to Ukrainian, half to
English. Mathematics and science, for example, were taught in English; music, art and social studies in Ukrainian.
There is a tendency to teach technological, scientific studies through the majority language.
(4) Where a minority language is used for a majority of classroom time (e.g. 80% to almost 100% in Wales), the
justification is usually that children easily transfer ideas, concepts, skills and knowledge into the majority
language. Having taught a child multiplication in Spanish, this mathematical concept does not have to be re-taught
in English. Classroom teaching transfers relatively easily between languages when such languages are sufficiently
developed to cope with concepts, content and curriculum materials.
(5) The justification given for heritage language education is also that a minority language is easily lost, a majority
language is easily gained. Children tend to be surrounded by the majority language. Television and train
advertisements, shops and signs, videos and visits often provide or induce bilingual proficiency in an incidental
way by accenting a majority language. Thus bilingualism is achieved by a concentration on the minority language
at school.
(6) Heritage language schools are mostly elementary schools. This need not be the case. In Wales, for example,
such schools are available to the end of secondary education and the heritage language can be used as a medium of
study at College and University.
Two Way/Dual Language Bilingual Education
Two Way or Dual Language Bilingual Education occurs when approximately equal numbers of language minority
and language majority students are in the same classroom. Or sometimes in the US, the program will contain
approximately 50% of students who

< previous page page_186 next page >
< previous page page_187 next page >
Page 187
are English language learners (e.g. Haitian Creole speakers); the remaining students are language minority
students relatively fluent in English.
Mostly to be found in the US at elementary school level, two languages are used in the classroom. For example,
around half the children may be from Spanish speaking homes; the other half from English monolingual homes
and they work together in the classroom in harmony. Since both languages are used for instruction and learning,
the aim is to produce relatively balanced bilinguals (Lindholm, 1991; Morison, 1990). Biliteracy is as much an aim
as full bilingualism, with literacy being acquired in both languages either simultaneously or with an initial
emphasis on native language literacy (see Chapters 15 and 16).
There are a variety of terms used to describe such schools: Two Way Schools, Two Way Immersion, Two Way
Bilingual Education, Developmental Bilingual Education, Dual Language Education, Bilingual Immersion, Double
Immersion and Interlocking Education.
Christian (1994) and Lindholm (1987) note the growth of such programs since 1983 with the oldest dating back to
1963 in Dade County, Florida, and developed by a US Cuban community (see García & Otheguy, 1985, 1988). For
Lindholm (1987), Two Way programs have four characteristics:
(1) A non-English language (i.e. the minority language) is used for at least 50% of instruction.
(2) In each period of instruction, only one language is used.
(3) Both English and non-English speakers are present in preferably balanced numbers.
(4) The English and non-English speakers are integrated in all lessons.
Two Way bilingual classrooms contain a mixture of language majority and language minority students. For
example, half the children may come from Spanish speaking homes; the other half from English language
backgrounds. A language balance close to 50%-50% is attempted. If one language becomes dominant (e.g. due to
much larger numbers of one language group), the aim of bilingualism and biliteracy may be at risk. The reality of
such schools is often different, with an imbalance towards larger numbers of language minority students being
more common.
An imbalance in the two languages among students may result in one language being used to the exclusion of the
other (e.g. Spanish speaking children having to switch to English to work cooperatively). Alternatively, one
language group may become sidelined (e.g. Spanish speakers become excluded from English speaking groups).
Segregation rather than integration may occur. In the creation of a dual language school or classroom, careful
student selection decisions have to be made to ensure a language balance.
When an imbalance does exist, it may be preferable to have slightly more language minority children. Where there
is a preponderance of language majority children, the tendency is for language minority children to switch to the
higher status, majority language. In most (but not all) language contexts, the majority language is well represented
outside school (e.g. in the media and for employment). Therefore, the balance towards the majority language
outside school can be complemented by a slight balance

< previous page page_187 next page >
< previous page page_188 next page >
Page 188
towards the minority language in school (among student in-take and in curriculum delivery). However, if the
school contains a particularly high number of language minority children, the prestige of the school will
sometimes suffer (both among language majority and language minority parents).
In Two Way bilingual Magnet schools in the US, students may be drawn from a wide catchment area. For
example, if a Magnet school focuses on Environmental Arts and Sciences, parents throughout a district may be
eager to gain a place for their children. Ensuring a language balance in each classroom then becomes a key
selection feature.
There are situations where attracting language majority students to a Two Way bilingual school is difficult. Where
the monolingual mainstream school is as (or more) attractive to prospective parents, recruitment to Two Way
bilingual schools may be a challenge. For parents, allocation of their children to such Dual Language Bilingual
Programs will be voluntary and not enforced. Hence, the good reputation, perceived effectiveness and curriculum
success of such Two Way bilingual schools becomes crucial to their continuation. Evidence from the US suggests
that language minority parents may be supportive of such a program (e.g. Hornberger, 1991). Majority language
parents may need more persuading. Community backing and involvement in the school may also be important in
long-term success.
The development of Two Way bilingual schools often starts with the creation of a dual language kindergarten
class rather than an implementation throughout a school. As the kindergarten students move through the grades, a
new dual language class is created each year. Apart from Elementary Two Way bilingual schools, there is also
Dual Language secondary education in the US and, with different names, in many other countries of the world
(e.g. Wales, Spain, India). A Two Way bilingual school may be a whole school in itself. Also, there may be dual
language classrooms within a ‘mainstream’ school. For example, there may be one dual language classroom in each
Dolson & Meyer (1992) list seven major goals in Two Way bilingual programs: (1) students should develop high
levels of proficiency in their first language; (2) students should achieve high levels of proficiency in their second
language (e.g. English); (3) academic performance should eventually be at or above grade level in both languages
(e.g. by Grade 6); (4) students should demonstrate positive cross-cultural attitudes and behaviors; (5) students
should show evidence of high levels of personal and social competence; (6) schools should have programs of
academic excellence for language majority and language minority students, and (7) communities and society at
large should benefit from having citizens who are bilingual and biliterate, who are more positive towards people of
different cultural backgrounds, and, who can meet national needs for language competence and a more peaceful
coexistence with peoples of other nations.
The aim of a Two Way bilingual schools is thus not just to produce bilingual and biliterate children. To gain status
and to flourish, such a school needs to show success throughout the curriculum. On standardized tests, on
attainment compared with other schools in the locality, and in specialisms (e.g. music, sport, science), a Two Way

< previous page page_188 next page >
< previous page page_189 next page >
Page 189
bilingual school will strive to show relative success. A narrow focus on proficiency in two languages will be
insufficient in aim.
The mission of Two Way bilingual schools may be couched in terms such as ‘equality of educational opportunity
for children from different language backgrounds’, ‘child-centered education building on the child’s existing
language competence’, ‘a positive self-image for each child’, ‘a community dedicated to the integration of all its
children’, ‘enrichment not compensatory education’, ‘a family-like experience to produce multicultural children’,
and ‘supporting Bilingual Proficiency not Limited English Proficiency’.
One of the special aims of Two Way bilingual schools (e.g. compared with mainstream schools) is to produce
bilingual, biliterate and multicultural children. Language minority students are expected to become literate in their
native language as well as in the majority language. Language majority students are expected to develop language
and literacy skills in a second language. At the same time, majority language students must make normal progress
in their first language. To achieve these aims, a variety of practices are implemented in Two Way bilingual
(1) The two languages of the school (e.g. Spanish and English, Chinese and English, Haitian Creole and English)
have equal status. Both languages will be used as a medium of instruction. Math, Science and Social Studies, for
example, may be taught in both languages. However, care has to be taken not to be repetitive, not to teach the
same content in both languages. Strategies of separating, allocating and integrating two languages in the classroom
are considered in Chapter 12.
(2) The school ethos will be bilingual. Such an ethos is created by classroom and corridor displays, notice boards,
curriculum resources, cultural events, dinner-time and extra curricula activity using both languages in a relatively
balanced way. Announcements across the school address system will be bilingual. Letters to parents will also be in
two languages. While playground conversations and student to student talk in the classroom is difficult to influence
or manipulate, the school environment aims to be transparently bilingual.
(3) In some Two Way bilingual schools, the two languages are taught as languages (sometimes called Language
Arts instruction). Here, aspects of spelling, grammar, metaphors and communicative skills may be directly taught.
In other Two Way bilingual schools, use of both languages as media of instruction is regarded as sufficient to
ensure bilingual development. In such schools, children are expected to acquire proficiency in language informally
throughout the curriculum and through interaction with children who are effective first language role models. In
both cases, reading and writing in both languages are likely to receive direct attention in the curriculum. Biliteracy
is as much an aim as full bilingualism. Literacy will be acquired in both languages either simultaneously or with an
initial emphasis on native language literacy.
(4) Staff in the dual language classrooms are often bilingual. Such teachers use both languages on different
occasions with their students. Where this is difficult (e.g. due to teacher supply or selection), teachers may be
paired and work together closely as

< previous page page_189 next page >
< previous page page_190 next page >
Page 190
a team. A teacher’s aide, paraprofessionals, secretaries, custodial staff, parents offering or invited to help the
teacher may also be bilingual. Language minority parents can be valuable ‘teacher auxiliaries’ in the classroom. For
example, when a wide variety of Spanish cultures from many regions is brought to the classroom, parents and
grandparents may describe and provide the most authentic stories, dances, recipes, folklore and festivals. This
underlines the importance of the cultural heritage of language minorities being shared in the classroom to create an
additive bilingual and multicultural environment.
(5) The length of the Two Way bilingual program needs to be longer rather than shorter. Such a program for two
or three grades is insufficient. A minimum of four years extending through the grades as far as possible is more
defensible. Length of experience of a Two Way bilingual program is important to ensure a fuller and deeper
development of language skills, and biliteracy in particular. Where a US Two Way bilingual program exists across
more years, there is a tendency for the curriculum to be increasingly taught in the majority languageEnglish. In
some schools, the curriculum is initially taught for around 90% of time through the minority language. In later
grades in the elementary school, this percentage is decreased until it reaches approximately 50%. In these
programs, majority language students have an immersion experience in a second language (e.g. Spanish), while
minority language students initially receive most of their education in their home language
A central idea in Two Way bilingual schools is language separation and compartmentalization. In each period of
instruction, only one language is used. Language boundaries are established in terms of time, curriculum content
and teaching. These will each be briefly considered, and then extended in Chapter 12.
First, a decision is made about when to teach through each language. One frequent preference is for each language
to be used on alternate days. On the door of the classroom may be a message about which language is to be used
that day. For example, Spanish is used one day, English the next, in a strict sequence. Alternately, different lessons
may use different languages with a regular change over to ensure both languages are used in all curricula areas.
For example, Spanish may be used to teach Mathematics on Monday and Wednesday and Friday; English to teach
Mathematics on Tuesday and Thursday. During the next week, the languages are reversed, with Mathematics
taught in Spanish on Tuesday and Thursday. There are other possibilities. The division of time may be in half
days, alternate weeks, alternate half semesters. The essential element is the distribution of time to achieve bilingual
and biliterate students.
The amount of time spent learning through each language varies from school to school. Often, a 50%-50% balance
in use of languages is attempted in early and later grades. In other classrooms, the minority language will be given
more time (60%, 75%, 80% and 90% is not uncommon), especially in the first two or three years. In the middle
and secondary years of schooling, there is sometimes a preference for a 50%-50% balance, or more accent on the
majority language (e.g. 70% through English; 30% through Spanish).

< previous page page_190 next page >
< previous page page_191 next page >
Page 191
Whatever the division of time, instruction in a Two Way bilingual school will keep boundaries between the
languages. Switching languages within a lesson is not considered helpful. If language mixing by the teacher
occurs, students may wait until there is delivery in their stronger language, and become uninvolved at other times.
When there is clear separation, the Spanish speakers, for example, may help the English-speakers on Spanish days,
and the English speakers help the Spanish speakers on English days. Interdependence may stimulate cooperation
and friendship, as well as learning and achievement. The potential problems of segregation and racial hostility may
thus be considerably reduced.
However, the two languages will sometimes be switched or mixed in the classroom (e.g. in private conversations,
in further explanations by a teacher, and internal use of the dominant language). Use of languages by children,
especially when young, is not usually consciously controlled. Switching language can be as natural as smiling and
using a particular accent.
Second, bilingual teachers ensure they do not switch languages. Children hear them using one language (during a
lesson period or during a whole day) and are expected to respond in that same language. When, as in many forms
of ‘strong’ bilingual education, there is a shortage of bilingual teachers, a pairing of teachers may ensure language
separation. A teacher using Spanish only will work in close association with a teacher who only uses English with
the same class. Such teamwork requires teachers to be committed to bilingualism and multiculturalism as
important educational aims.
Third, language boundaries may be established in the curriculum. This may occur according to which ‘language
day’ it is. Alternatively, in some schools, different parts of the curriculum are taught in different languages. For
example, Social Studies and Environmental Studies may be taught in Spanish, Science and Math in English. Such a
policy establishes separate occasions where each language is to be used, and keeps the two languages apart.
However, one danger is that the majority language becomes aligned with modem technology and science, while the
minority language becomes associated with tradition and culture. This may affect the status of the language in the
eyes of the child, parents and society. The relationship of languages to employment prospects, economic advantage
and power thus need to be considered.
Such Two Way bilingual schools differ from Transitional Bilingual Education and Submersion/Mainstream/ESL
approaches. US Transitional Bilingual education essentially aims to move children into English-only instruction
within two or more years. In contrast, Two Way bilingual schools aim to enable the child to achieve increasing
proficiency in two languages (e.g. Spanish and English in the US). Two Way bilingual schools differ from
Canadian Immersion Bilingual Education in the language backgrounds of the students. Immersion schools usually
contain only language majority children learning much or part of the curriculum through a second language (e.g.
English speaking children learning through the medium of French in Canadian schools). Two Way bilingual
schools aim to contain a balanced mixture of children from two (or more) different language backgrounds (e.g.
from Spanish speaking and English speaking homes in the US). Two Way bilingual schools differ from
Developmental Maintenance

< previous page page_191 next page >
< previous page page_192 next page >
Page 192
(Heritage Language) schools in aiming for more of a balance of majority and minority language children, with
slightly more emphasis on bilingual and multicultural outcomes. Developmental Maintenance education is
comparatively more concerned with preservation of the ethnic language, ethnic culture, and in many cases, has a
preponderance of language minority children. The type of school possible in a neighborhood is often determined
by the sociolinguistic character of the school population (e.g. the size of one or more language minority groups,
the presence of recent or more established in-migrants, the numbers of majority language speakers).
Two Way bilingual schools in the US appear to date from 1963 in Dade County, Florida, and were developed by
the US Cuban community in that area (see Lindholm 1987; Garciía & Otheguy, 1985, 1988). In September 1963,
the Coral Way Elementary School started a bilingual program that embraced both Spanish and English speaking
students. During the 1960s, another 14 such bilingual schools were set up in Dade County. This is related to the
fact that many Cubans expected to return to Cuba, believing the Castro regime wouldn’t survive. Local people
supported the maintenance of Spanish among the soon-to-leave Cubans. Local English speaking children from
middle class families were enrolled in the school. This reflected a wish among parents for foreign language
instruction following Russia’s initial triumph over the US in the space race. (In 1958, the Russian Sputnik was
Since that era, there has been a steady rise in the number of Two Way bilingual schools in the US, particularly
since 1989as the graph illustrates using data in Donna Christian (1994). Approximately a third of such schools are
located in California and another third in New York. Of around 170 schools in the United States, sixteen other
states have Two Way bilingual schools (e.g. Arizona, Massachusetts, Texas and Virginia). The languages of
instruction in Two Way Bilingual Programs in the US are predominantly Spanish/ English (over 90% of such
schools) but with the following combinations also represented: Cantonese/English; Portuguese/English; Haitian
Creole/English; Korean/English;
The Rise of Dual Language Schools

< previous page page_192 next page >
< previous page page_193 next page >
Page 193
Russian/English; Japanese/English and French/English. Around 85% of all these schools operate from kindergarten
to the 6th Grade.
Evaluations of the effectiveness of Dual Language schools indicate relative success (Cazabon, Lambert & Hall,
1993; Lambert & Cazabon, 1994; Lindholm, 1991; Lindholm & Aclan, 1991; Lindholm, 1994; Lindholm, 1995;
Thomas, Collier & Abbott, 1993). As Christian (1994, overview page) suggests: ‘Emerging results of studies of
Two Way bilingual programs point to their effectiveness in educating non-native English speaking students, their
promise of expanding our nation’s language resources by conserving the native language skills of minority students
and developing second language skills in English speaking students, and their hope of improving relationships
between majority and minority groups by enhancing cross-cultural understanding and appreciation.’
One example of a Two Way bilingual school is the James F. Oyster Bilingual Elementary School in Washington
DC which has a two way program that commenced in 1971. The initiative was taken by the local community (by
parents and local politicians) to produce a school that crossed language, cultural, ethnic and social class lines.
Parents are very active in the running of the school. With students from kindergarten to 6th Grade, there is a ethnic
mix of 58% Hispanic, 26% white; 12% Black and 4% Asian (Freeman, 1995). Around two in every five children
come from low-income families (e.g. are awarded a free lunch).
The James F. Oyster Bilingual Elementary School program is distinctive because it has two teachers in each
classroom: one teacher speaks only Spanish to the students; the other teacher speaks only English. The students
experience Spanish and English medium instruction for approximately equal amounts of time. A strong
multicultural dimension pervades the curriculum, with the contributions of different children encouraged and
respected. The notion of equality permeates the ethos of the school (Freeman, 1995). This is reflected in the
school’s mission statement: ‘Oyster Bilingual School’s focus is on the development of bilingualism, biliteracy, and
biculturalism for every student through the mastery of academic skills, the acquisition of language and
communicative fluency, the appreciation of differences in racial and ethnic backgrounds, and the building of a
positive self-concept and pride in one’s heritage.’
Bilingual Education in Majority Languages
Bilingual Education in Majority Languages comprises the joint use of two (or more) majority languages in a
school. The aims of such schools usually include bilingualism or multilingualism, biliteracy and cultural pluralism.
Such schools are in societies where much of the population is already bilingual or multilingual (e.g. Singapore,
Luxembourg) or where there are significant numbers of natives or expatriates wanting to become bilingual (e.g.
learning through English and Japanese in Japan). Asian examples of Bilingual Education in Majority Languages
include Arabic-English, Bahasa Melayu-English, Mandarin Chinese-English and Japanese-English. In Africa and
India there are also schools where a ‘majority’ regional language and an international language coexist as teaching
medium in a school. Bilingualism in that regional language and an international language (e.g. French, English) is
the aim and outcome of formal education.

< previous page page_193 next page >
< previous page page_194 next page >
Page 194
Generally, such schools will contain majority language children, with variations in the language heterogeneity or
language homogeneity of the classes.
In the Asian examples, a country (e.g. Brunei, Taiwan) or a region may have one dominant indigenous language
with a desire to introduce a second international language such as English into the school. The international
language will be used as a medium of instruction alongside the native language. The aim is for fully bilingual and
biliterate students through an Enrichment bilingual education program. For example, the Dwibahasa (two
language) school system in Brunei operates through Malay (Bahasa Melayu) and English (Jones, Martin & Ozóg,
1993). In Nigeria, bilingual education is present, particularly at the secondary school level, in English plus one of
the national languages of Nigeria: Hausa, Ibo or Yoruba (Afolayan, 1995). In Singapore, English plus Mandarin,
Malay or Tamil (the four official languages of the country) create bilingual education (Pakir, 1994). In Germany,
German is paired with French, English, Spanish and Dutch to create a ‘German model’ of European multicultural
and bilingual education (Mäsch, 1994). Under the umbrella heading of Bilingual Education in Majority Languages
are different societal situations (e.g. a country already bilingual; a country wishing to become more bilingual) as
the above discussion has revealed.
Two special examples of Bilingual Education in Majority Languages will now be considered, commencing with
the International School Movement and continuing with the European School Movement.
International Schools
International Schools are a diverse collection of schools throughout the world. Numbering over 850 schools, they
are found in over 80 countries of the world, mostly in large cities. Mainly for the affluent, parents pay fees for
mostly private, selective, independent education. Children in these schools often have parents in the diplomatic
service, multinational organizations, or in international businesses and who are geographically and vocationally
mobile. Other children in an International school come from the locality, whose parents want their children to have
an internationally flavored education. One language of the school is ordinarily English. International schools that
have English as the sole medium of transmitting the curriculum could not be included under the heading of
Bilingual Education in Majority Languages. Such schools become bilingual when a national or international
language is incorporated in the curriculum. Sometimes the second language taught (for up to 12 years) is only
taught as a language. In other schools, the second language is used as a medium to teach part of the curriculum.
Some schools enable their students to acquire third and fourth languages. Generally, the languages of International
schools are majority languages with international prestige. Minority languages are rarely found in such schools.
The primary and secondary curriculum of International Schools tends to reflect United States, British as well as
the local curriculum tradition. The teachers are from various countries, usually with a plentiful supply of British
and American trained staff. Sometimes preparing children for the International Baccalaureate, United States tests
or British examinations, most prepare their clientele for Universities in Europe and North

< previous page page_194 next page >
< previous page page_195 next page >
Page 195
America. Baker & Prys Jones (1997) and Ochs (1993) provide illustrations and further discussion of the
International Schools Movement. Three illustrations of such schools follow.
One example is the Nishimachi International School in Tokyo. The age range of the students is from five to 15 and
the school caters for over 400 students from around 30 nationalities. The fifty or so teachers in the school come
from 12 nationalities. The school is a dedicated center for bilingualism in English and Japanese and for
multicultural studies. The student composition is 20% Japanese; 35% United States; 20% one parent Japanesethe
other parent from another nationality; and 25% other nationalities (ECIS, 1994). Another example is the John F.
Kennedy School in Berlin. With students aged five to 19, the school comprises around 1300 students from over 20
nationalities, with German and US students being in the majority. Fifty per cent of teachers are from the United
States; 50% from Germany. The school is bilingual, binational and bicultural in German and English, with
instruction in both languages. Tuition is free (ECIS, 1994). The Bayan Bilingual School in Kuwait enrolls students
aged three to 17 and contains around 900 students from 14 nationalities. Teachers come from 15 nationalities. It is
a bilingual school stressing Arab culture, traditions, heritage and identity while preparing its graduates for college
and university placement throughout the world. The languages of instruction are Arabic and English (ECIS, 1994).
European Schools Movement
Another European example of Bilingual Education in Majority Languages is the European Schools movement
(Baetens Beardsmore, 1993; Housen & Baetens Beardsmore, 1987; Baetens Beardsmore & Swain, 1985; Tosi,
1991; Baker & Prys Jones, 1997). Mostly for the relatively elite workers of the European Community (EC), such
schools are multilingual and cater for some 15,000 children of different EC nations. Started officially in 1958 in
Luxembourg, with schools sited in Belgium, Italy, Germany, the Netherlands and England, such European Schools
have up to 11 different language sections reflecting the first language of the students (and this may increase as
other countries join the European Community). Younger children use their native language as the medium of
learning but also receive second language instruction (English, French, or German) in the primary school years.
Older children take part of their schooling in their native language and part through the medium of a ‘vehicular’ or
‘working’ language. The ‘vehicular’ language will usually be a ‘majority’ second language for the child selected
from English, French or German. This language will be taught by native speakers. Native student speakers of that
language will also be present in the school as language models. The vehicular language is used to teach mixed
language groups of students history, geography and economics from the third year of secondary education. In
addition, students are taught a third language for a minimum of 360 hours.
The outcome of such schooling tends to be functionally bilingual and often multilingual students with a sense of
cultural pluralism, European multiculturalism and European identity. Integration and harmonization of students
from different nationalities is formally achieved in the ‘European Hours’ lessons using the vehicular language.
‘European Hours’ are an important curriculum component from Grade 3 in Primary education. In

< previous page page_195 next page >
< previous page page_196 next page >
Page 196
classes of 20 to 25 students for three lessons a week, children from different language backgrounds work
cooperatively together. A small group project with a realistic, attainable goal (e.g. making puppets) provides the
focus for a context embedded and cognitively undemanding ‘European Hour’. Deliberately and explicitly, students
are encouraged to respect each person’s native language. Games and physical education are also occasions for a
cooperative mixing of students from the different language sections. Students are linguistically mixed to avoid
stereotypes and prejudices, and to build a supranational European identity (Baetens Beardsmore, 1993).
A major difference between the European schools movement and the Canadian immersion programs is that the
second language is taught as a subject before being used as a medium of instruction. That second language also
continues to be taught as a subject, leading to a high level of grammatical accuracy (Baetens Beardsmore, 1993).
According to Housen & Baetens Beardsmore’s (1987) research in one European School: ‘This strong language
commitment has no detrimental effects on academic achievement as can be gauged from results on the final
European Baccalaureate examination, on which 90% of students have been successful’ (p. 85). However,
bilingualism, biliteracy and multiculturalism is not only due to the effects of schooling. The parents may also be
bilingual or multilingual, and the children are more likely to come from literacy-oriented, middle class bureaucrat
homes, with a positive view of bilingualism. Playgrounds are multilingual, satellite TV in Europe is multilingual
and the growing notion of Europeanization creates privileged European schoolchildren who are ‘educated
bilinguals, equally at ease with two languages, with their own national culture and the supranational European
identity’ (Tosi, 1991:33). The Stated Objectives of the European Schools Movement are (adapted from information
from the Central Office of the Representative of the Board of Governors of the European Schools):
to give students confidence in their own cultural identitythe bedrock for their development as European citizens;
to give a sound education, based on a broad range of subjects, from nursery level to university-entrance;
to develop high standards in speaking and writing both the mother tongue and two or more foreign languages;
to develop mathematical and scientific skills throughout the whole period of schooling;
to encourage a European and global perspective in the study of history and geography, rather than a narrower,
nationalistic one;
to encourage creativity in music and the plastic arts and an appreciation of all that is best in a common European
artistic heritage;
to develop physical skills and instill in students an appreciation of the need for healthy living through participation
in sporting and recreational activities;
to offer students professional guidance in their choice of subjects and in career/university decisions in the later
years of the secondary school;
to foster tolerance, co-operation, communication and concern for others throughout the school community;

< previous page page_196 next page >
< previous page page_197 next page >
Page 197
to provide high-quality teaching through the recruitment of well qualified and experienced staff by the respective
countries’ Ministries of Education.
In Europe, there are other schools (apart from the European School Movement) that use two or more prestigious
languages in the curriculum (Baetens Beardsmore, 1993). For example, in Luxembourg, children who speak
Luxemburger after birth become trilingual (Luxemburger, French and German) through education (Lebrun &
Baetens Beardsmore, 1993). Children start their education at age five through the medium of Luxemburger (a
variety of Low German). German is initially a subject in the curriculum, then introduced as the main teaching
medium. By the end of Grade 6, children function in much of the curriculum in German. French is introduced as a
subject in Grade 2, and is increasingly used as a teaching medium in secondary education. Most students have a
working knowledge of three languages by the conclusion of schooling (Lebrun & Baetens Beardsmore, 1993).
Through emphasis on the home tongue in the early years, emphasis on German in the primary school and emphasis
on French in the secondary school, children become trilingual and biliterate (French and German literacy).
Having considered ten types of bilingual education, the natural question to ask is whether one type is more
effective than another. For Spanish speaking children in the US, is it better for them to be placed in Submersion,
Transitional, Maintenance or Two Way schooling? For a monolingual English speaker, is it detrimental to enter
immersion schooling compared with mainstream schooling? Such questions will now be examined in the next
chapter by ‘effectiveness’ research.
Suggested Further Reading
BAETENS BEARDSMORE, H. (ed.) 1993, European Models of Bilingual Education. Clevedon: Multilingual
CAZDEN, C.B. and SNOW, C.E. 1990, English Plus: Issues in Bilingual Education. (The Annals of the American
Academy of Political and Social Science 508, March 1990). London: Sage.
CRAWFORD, J. 1991, Bilingual Education: History, Politics, Theory and Practice (2nd edn). Los Angeles:
Bilingual Education Services.
CUMMINS, J. and DANESI, M. 1990, Heritage Languages. The Development and Denial of Canada’s Linguistic
Resources. Toronto: Our Schools/Ourselves Education Foundation & Garamond Press.
DANESI, M., McLEOD, K. and MORRIS, S. 1993, Heritage Languages and Education: The Canadian
Experience. Oakville, Ontario: Mosaic Press.
GARCÍA, O. and BAKER, C. (eds) 1995, Policy and Practice in Bilingual Education: A Reader Extending the
Foundations. Clevedon: Multilingual Matters. (Sections 1 & 2).

< previous page page_197 next page >
< previous page page_198 next page >
Page 198
Review and Study Questions
(1) Write short revision notes on the following:
(i) List the important developments in the history of bilingual education in the US or any other country of
your choice.
(ii) What is the difference between ‘weak’ and ‘strong’ types of bilingual education.
(iii) What are the important characteristics of the ten types of bilingual education.
(iv) State the difference(s) between submersion and transitional bilingual education. Also, consider the
differences between variations within submersion and transitional bilingual education.
(v) State the difference(s) between two way bilingual education, heritage language/developmental
maintenance education, Canadian immersion and bilingual education in majority languages.
(2) Research the history of bilingual education in your country or area (e.g. at state level). What have been the
major changes? What seems to be the causes of such major changes?
(3) What is the distinction between ‘weak’ and ‘strong’ forms of bilingual education?
(4) In what types of bilingual education may the distinction between ‘strong’ and ‘weak’ forms of bilingual
education become blurred?
(5) Describe in detail one of the types of bilingual education with which you are most familiar. What are the
important characteristics of that form of bilingual education, particularly those characteristics not focused on in the
Study Activities
(1) Write a personal account of one type of bilingual education which you have experienced or with which you are
most familiar. Describe your experience of the language dimensions of this form of education. Present this in a
small group seminar to find out differences and similarities of experience.
(2) Visit a school, and by interview and observation decide the extent to which that school fits one or more of the
types of bilingual education in this chapter.
(3) Visit one or more schools and ask about the history of a bilingual education program or language program
within that school. What have been the aims of the school with regard to languages? Have these aims changed
over the last 10 or 20 years? How do the teachers perceive the first and second language of children being ignored
or used over the last decade or more?
(4) By using documents, interviews, visits to schools and visits to administrators, try to sketch the history of
language and bilingual education within a specific community. Also observe what signs and symbols there are of
language within the community. For example, on posters, in newspapers, mass media and community activity, is
there more than one language in use?
(5) Imagine you are a parent or teacher and were required to make a public speech about changing a school from a
‘weak’ form of bilingual education to a ‘strong’ form. Prepare, and then deliver in front of the class, a speech of
about five minutes to persuade the administration.

< previous page page_198 next page >
< previous page page_199 next page >
Page 199
Chapter 11
The Effectiveness of Bilingual Education
Research Studies
Reviews and Overviews of Research
Canadian Immersion Bilingual Education
Heritage Language Education/Developmental
Maintenance Education
The US Debate
Advancing the Effectiveness of Bilingual Education

< previous page page_199 next page >
< previous page page_200 next page >
Page 200
Having considered ten types of bilingual education, this chapter turns to considering research on the major types of
bilingual education. How effective are these major models for which types of children? What are the recorded
successes and limitations? What makes a bilingual school more or less effective? These are the issues considered
in this chapter.
Research Studies
From early research in the 1920s in Wales (Saer, 1922) and Malherbe’s (1946) evaluation of bilingual education in
South Africa, there has been a flood of evaluations of bilingual projects and experiments, programs and
experiences. The research is international; for example, Ireland (e.g. Harris, 1984) and England (e.g. Fitzpatrick,
1987); US (e.g. Danoff et al., 1977, 1978) and Canada (e.g. Swain & Lapkin, 1982); Peru (e.g. Hornberger, 1988,
1990a); Hong Kong (e.g. Boyle, 1990); Wales (e.g. Eurwen Price, 1985; Baker, 1995) and Spain (e.g. Sierra &
Olaziregi, EIFE2, 1989; Artigal, 1995; Bel, 1993). A report from the World Bank (1982) provided one of the few
comparative international studies, covering bilingual education in the Philippines, Ireland, Canada, Mexico,
Nigeria, Sweden and the United States.
It is possible to find support for most of the different forms of bilingual education by selecting and emphasizing a
particular study. Some examples will illustrate. Criticism of Irish immersion education for children from English
speaking homes was given by Macnamara (1966). He found such immersion children to be 11 months behind
mainstream children on mechanical arithmetic. He suggested that it is ‘probable that the use of Irish in teaching
problem arithmetic hinders the progress of English speaking children’ (MacNamara, 1966: 103). In comparison,
support for immersion education comes from the Canadian research studies (e.g. Swain & Lapkin, 1982). Danoff
et al. (1977, 1978)

< previous page page_200 next page >
< previous page page_201 next page >
Page 201
found submersion to be superior to transitional bilingual education with a large US sample of almost 9000
children. McConnell (1980) found US transitional bilingual education to be better than submersion, while
Matthews (1979), also in the US, found no difference between these two ‘weak’ forms of bilingual education.
Heritage language education has been evaluated as successful in Canada by Keyser & Brown (1981) and in
England by Fitzpatrick (1987). In contrast, early research in Wales (e.g. Smith, 1923) questioned educational
outcomes from heritage language programs for Welsh bilinguals.
Whether research finds a consensus in favor of one or some of the types of bilingual education is examined later in
this chapter. For the moment, it is important to outline some reasons why studies vary in their findings.
The Sample of Children
The results of one study are limited to that sample of children at the time of the study. If there is some form of
probability sampling (e.g. a random sample of a defined population is chosen), then these results may generalize to
that specific population. Such sampling rarely occurs in bilingual education evaluations where instead, many
studies have small and unrepresentative samples. It is usually ethically questionable and practically impossible to
allocate children randomly into experimental and control groups that contain perfect mirrors of a large population
of schoolchildren.
Given the wide variety of samples of children used in bilingual education effectiveness research, it is not
surprising that differences in findings emerge. Samples of children include urban and rural schools, various social
class backgrounds, different ages and varying levels of motivation. The international research includes a mixture
of bilingual groups: indigenous language minority groups, in-migrants, and majority language children in minority
language education. Generalization of results from one group to another is not valid. Such children may be in a
subtractive or additive environment at home, school, community and nation.
Unlike the physical world, simple laws of behavior that govern large groups of people are not likely. The immense
variety of individual differences and environment differences makes clear cut, simple research results intrinsically
elusive. Such results also say something about what has been, not what will always be. They do not guarantee that
results will be stable across time.
Interacting Factors
Various factors, other than the sample of children, may have a variable effect on bilingual education. Parental
interest, parental involvement in their children’s education and parental co-operation with teachers is one
intervening factor. Another factor is likely to be the enthusiasm and commitment of teachers to the education
program. With a novel experiment in bilingual education, there may be extra enthusiasm and interest. The level of
material support (e.g. books, curriculum guidelines, computers, science equipment) may also produce variable

< previous page page_201 next page >
< previous page page_202 next page >
Page 202
There is likely to be as much variation in outcomes (e.g. achievement in different curriculum areas) inside a
particular bilingual education program (e.g. transitional, immersion or heritage language) as between different
types of program. As Berliner (1988: 289) suggests of intervention programs, ‘One of the major findings from
studies of Follow-through programs, where competing theories and programs of early childhood education were
purposefully funded, was that the variation across sites within the same program (whether it was behavioral,
cognitive, requiring parent involvement, technology oriented, or whatever) was equal to the variation across the
sites between the different programs.’
The crucial point is this. The language policy and language practice in schooling are only one element amongst
many that make a school more or less successful. A recipe for success is unlikely to result from one ingredient
(e.g. the language of the classroom). A great variety of factors act and interact to determine whether bilingual
education is successful or not. As will be considered at the end of the chapter, it makes more sense to consider the
wide variety of conditions which make bilingual education more or less successful. We need to specify all the
ingredients in different recipes to fully understand the success or failure of forms of bilingual education. Bilingual
education, whatever type or model, is no guarantee of effective schooling.
Measures of Success
An important question is: ‘What tests or other sources of evidence are used to determine whether a form of
bilingual education is successful?’ Should the sole outcomes be competence in one or two languages? What aspects
of language should be assessed? Should science and social studies be included? Should the measure of success be
performance across the whole curriculum? Whose version of the curriculum should be used (e.g. computational
versus conceptual emphasis in Mathematics; religious knowledge versus moral development)? How important is it
to include non-cognitive outcomes such as self-esteem, moral development, school attendance, social and
emotional adjustment, integration into society and gaining employment? What are the long term effects of
bilingual schooling (e.g. students later becoming parents who raise their children in the minority language)? The
questions indicate there will be debates and disputes over what are the valuable outcomes of schooling. Research
on the effectiveness of bilingual education has varied in the choice of measures of outcome, as is illustrated later in
the chapter.
A particular problem is that measures of success tend to be restricted to what is measurable. Quantitative outcomes
(e.g. test scores) are used; qualitative evidence is rarely gathered. Can a play be judged only on an applause meter
reading? Do a drama critic’s notes add a vigorous, insightful interpretation to the performance? While critics will
differ in their evaluations, they may add flesh and life to the statistical skeleton of educational tests.

< previous page page_202 next page >
< previous page page_203 next page >
Page 203
The Researchers
As Fishman (1977) and Edwards (1981) have indicated, research on bilingual education is rarely neutral. Often the
researchers have hypotheses which hide their expectations. No educational research can be totally value-free,
neutral or objective. The questions asked, the methodological tools chosen, decisions in analysis and manner of
reporting usually reveal ideological preferences. Many researchers will be supporters of bilingual education, ethnic
diversity, minority language rights and cultural pluralism. This is not to argue that all evaluation research on
bilingual education is invalid. Rather, it cannot be assumed that results are not affected by researchers, their
beliefs, opinions and preferences.
Some of the research on bilingual education is committed, prescriptive in nature, with interests, idealism and
ideology mixed with investigation and intelligent discussion. ‘Bilingual education is not merely a disinterested
exercise in the application of theory and research to real-life situations. It is also an exercise in social policy and
ideology’ (Edwards, 1981: 27).
Reviews and Overviews of Research
After a substantial number of different studies on bilingual education had accumulated, various reviews and
overviews appeared. A reviewer will assemble as many individual studies as possible and attempt to find a
systematic pattern and an orderliness in the findings. Is there a consensus in the findings? Is it possible to make
some generalizations about the effectiveness of different forms of bilingual education? Rarely, if ever, will all the
evaluations agree. Therefore the reviewer’s task is to detect reasons for variations. For example, different age
groups, different social class backgrounds and varying types of measurement device may explain variations in
The early reviews of bilingual education effectiveness were published in the late 1970s. Zappert & Cruz (1977),
Troike (1978) and Dulay & Burt (1978, 1979) each concluded that bilingual education in the US effectively
promoted bilingualism with language minority children and was preferable to monolingual English programs. That
is, language minority students became skilled in both the majority and minority language. Since the late 1970s,
many individual studies have been added and more recent reviews have emerged (e.g. Collier, 1989; Collier, 1992;
Lam, 1992). This section now examines recent reviews in three parts: (1) reviews of Canadian immersion
education; (2) a review of Heritage Language education; (3) and major, influential and controversial reviews in the
US. Each has a relatively large collection of literature allowing some depth of treatment. (In comparison, there are
relatively few evaluations of other types of bilingual education (e.g. Mainstream Bilingual Education, Two Way
Bilingual Education) preventing consideration here.)
Canadian Immersion Bilingual Education
The various reviews of Canadian immersion tend to paint a relatively uniform picture. The overviews of Swain &
Lapkin (1982), the California State Department of Education

< previous page page_203 next page >
< previous page page_204 next page >
Page 204
(1984), and Genesee (1983, 1984, 1987) highlight four major outcomes of Canadian immersion bilingual
Second Language (French) Learning
It is easy to predict that immersion students will surpass those in mainstream (core) programs given ‘drip-feed’
French lessons for 30 minutes a day. Most students in early total immersion programs approach native-like
performance in French around 11 years old in receptive language skills (listening and reading). Such levels are not
so well attained in the productive skills of speaking and writing (Lapkin, Swain & Shapson, 1990).
The reviews confirm that one kind of bilingualism can be educationally engineered. Immersion students mostly
succeed in gaining competence in two languages. However, as Chapter 1 revealed, bilingual ability is not the same
as being functionally bilingual. One of the limitations of immersion bilingual education is that for many students,
French can become a school-only phenomenon. Outside the school walls, immersion students tend not to use
French more than ‘drip feed’ students (Genesee, 1978). Such students are competent in French, but tend not to
communicate in French in the community. Potential does not necessarily lead to production; skill does not ensure
street speech. Lack of spontaneous or contrived French language opportunity and a dearth of French cultural
occasions to actively and purposefully use the second language may partly be the explanation. Other explanations
will be considered later. Stern (1984) argued that the immersion programs were strong on language, but weak on
widening students’ cultural horizons and weak on sensitizing them to francophone culture and values.
First Language (English) Learning
If immersion education provides the route to near-native fluency in a second language, is it at the cost of
attainment in the first language? Does bilingualism result in lesser achievement in English compared with
‘mainstream’ students? Like a balance, as one goes up, does the other go down?
For the first four years of early total immersion, students tend not to progress in English as do monolingual
English students in mainstream classes. Reading, spelling and punctuation, for example, are not so developed.
Since such children are usually not given English language instruction for one, two or three years after starting
school, these results are to be expected. However, the initial pattern does not last. After approximately six years of
schooling, early total immersion children have caught up with their monolingual peers in English language skills.
By the end of elementary schooling, the early total immersion experience has generally not affected first language
speaking and writing development. Parents of these children believe the same as the attainment tests reveal.
Indeed, when differences in English language achievement between immersion and mainstream children have been
located by research, it is often in favor of immersion students (Swain & Lapkin, 1982, 1991a). This finding links
with Chapter 9 that discussed the possible cognitive advantages consequential from bilingualism. If bilingualism
permits increased linguistic awareness, more flexibility in thought, more internal inspection

< previous page page_204 next page >
< previous page page_205 next page >
Page 205
of language, such cognitive advantages may help to explain the favorable English progress of early immersion
Early partial immersion students also tend to lag behind for three or four years in their English language skills.
Their performance is little different from that of total early immersion students, which is surprising since early
partial immersion education has more English language content. By the end of elementary schooling, partial early
immersion children catch up with mainstream peers in English language attainment. Unlike early total immersion
students, partial immersion children do not tend to surpass mainstream comparison groups in English language
achievement. Similarly, late immersion has no detrimental effect on English language skills (Genesee, 1983).
The evidence suggests that immersion children learn French at no cost to their English. Indeed, not only is there
the gain of a second language, there is also evidence to suggest that immersion results in possible extra benefits in
English proficiency. Rather than acting like a weighing balance, early total immersion, in particular, seems more
analogous to cooking. The ingredients, when mixed and baked, react together in additive ways. The product
becomes more than the sum of its parts.
Other Curriculum Areas
If immersion education results in children becoming bilingual in French and English, the question is whether this
is at the cost of achievement in other curriculum areas. Compared with children in mainstream education how do
immersion children progress in curriculum areas such as mathematics and science, history and geography? The
reviews of research suggest that early total immersion students generally perform as well in these subjects as do
mainstream children. That is, achievement in the curriculum is typically not adversely affected by early total
immersion bilingual education.
The evaluations of early partial immersion education are not quite so positive. When children in early partial
immersion learn mathematics and science through the medium of French, they tend to lag behind comparable
mainstream children, at least initially. This may be because their French skills are insufficiently developed to be
able to think mathematically and scientifically in their second language.
The results for late immersion are similar. The important factor appears to be whether second language skills
(French) are sufficiently developed to cope with fairly complex curriculum material. Johnson & Swain (1994)
argue that there is a gap in second language proficiency that needs bridging when students move from learning a
language as a subject to learning through that second language. The more demanding the curriculum area, the
higher the level of learning expected, and the later the switch to learning through a second language, the more
important it is to provide ‘bridging’ programs. Such ‘bridging programs’ ease the discrepancy between second
language proficiency and the language proficiency required to understand the curriculum. A ‘bridging program’
may require a language teacher and a content teacher (e.g. of mathematics) to operate together.
The results overall suggest that bilingual education by an immersion experience need not have negative effects on
curriculum performance, particularly in early total immersion

< previous page page_205 next page >
< previous page page_206 next page >
Page 206
programs. Indeed, most children gain a second language without cost to their performance in the curriculum.
However, the key factor seems to be whether their language skills have evolved sufficiently in order to work in the
curriculum in their second language (see Chapter 9).
Attitudes and Social Adjustment
Apart from performance throughout the curriculum, evaluations of immersion education have examined whether
immersion has positive or negative effects on students’ motivation, attitude and study skills. The most positive
results in this area have been found with early total immersion students. Parents of such students tend to express
satisfaction with their offspring’s learning as well as their personal and social behavior. Early immersion students
also tend to have more positive attitudes towards themselves, their education and to French Canadians in
comparison, for example, with late immersion students. However the danger here is attributing the positive
attitudes to schooling. The cause may alternatively be parental values and beliefs, home culture and environment.
This is further discussed in the next section.
Problems and Limitations
Various authors have recently highlighted possible limitations in immersion education that were not present in the
early evaluations (e.g. Hammerly, 1988; reply by Allen et al., 1989). First, Selinker, Swain & Dumas (1975)
suggested that immersion students do not always become grammatically accurate in their French. Immersion
students also tend to lack the social and stylistic sense of appropriate language use which the native speaker
possesses. For example, the restricted use of verb tenses other than the present, and the sometimes incorrect use of
‘tu’ and ‘vous’, appear to be related to the functionally restricted language of the classroom. Certain forms of
language do not naturally nor regularly occur in the classroom (e.g. because of the ‘present’ focus of learning and
adult-student relationships). One solution is to investigate problematic areas of vocabulary and grammar, provide
increased opportunities for receptive and particularly productive language in such weak areas, integrate a focus on
the form of language used with meaningful content teaching, and give systematic and consistent feedback on
language development to the student. Increasing group and collaborative learning is also seen as important in
developing the productive language proficiency of students towards the standards of native French-Canadian
speakers (Swain, 1993).
Second, surveys of graduates of immersion programs tend to find that relatively few students make much use of
French after leaving school (Harley, 1994; Wesche, 1993). This partly reflects opportunity, partly a lack of
confidence in their competence in speaking French, and partly a preference for English use. Also, immersion
students do not tend to interact significantly more with francophones than students in mainstream (core) programs.
Ability in French is not frequently translated into use of French outside the school gates, except, for example,
where employment and personal economics become a focus.

< previous page page_206 next page >
< previous page page_207 next page >
Page 207
Third, there is difficulty in pinpointing the crucial interacting factors that create an effective immersion experience.
There are, for example, intervening variables such as teaching techniques that may change the pattern of results.
Genesee (1983) argued that individualized, activity based teaching techniques may be more effective than
traditional whole class techniques. Genesee (1983) also argued that the intensity of language learning, for example,
how many hours per day, is likely to be more important than the length of language learning (e.g. the number of
years of second language learning). This is connected with the finding that older students tend to learn a second
language more quickly than younger learners. Is it immersion as a system that leads to relatively successful
outcomes or, or as well as, factors such as ‘student motivation, teachers’ preparation, home culture, parental
attitude, ethnolinguistic vitality, amount of time studying different curricula’ (Carey, 1991: 953).
Fourth, immersion programs can have effects on mainstream schools. For example, effects may include: a
redistribution of classroom teachers and leaders, a change in the linguistic and ability profile of mainstream
classes, discrepancies in class size with increasing numbers of mixed aged classes.
Fifth, Heller (1994) has argued that immersion schools provide anglophones with the linguistic and cultural capital
for increased social and economic mobility and for political power. Immersion education is thus, in this
perspective, about ulterior motives and vested interests. Such education is about gaining advantages in Canadian
society: educational, cultural, linguistic, social, power, wealth and dominance advantages. Hence, immersion
education may produce conflict with the minority francophone community (e.g. in Ontario) rather than the
harmonious unity and ‘bridge building’ that bilingualism aims to achieve in Canadian society. This issue is
considered further in Chapter 19.
Sixth, there is a danger in generalizing from the Canadian experience to elsewhere in the world. In Canada,
immersion concerns two major high status international languages: French and English. In many countries where
bilingualism is present or fostered, the situation is different. Often the context is one of a majority and a minority
language (or languages) co-existing. This links with additive and subtractive bilingual situations. Canada is
regarded as an additive bilingual context. Many countries across the five continents contain subtractive bilingual
If immersion education is thought worthy of generalizing to other countries, there are certain conditions which
need to be kept in mind:
(1) Immersion bilingual education as practiced in Canada is optional not enforced. The convictions of teachers and
parents and of the children themselves effect the ethos of the school and the motivation and achievement of the
children. Immersion education will work best when there is conviction and not enforced conformity.
(2) Immersion education in Canada starts with children who are at a similar level in their language skills. Such a
homogeneous grouping of children may make the language classroom more efficient. Where there are wide
variations in ability in a second

< previous page page_207 next page >
< previous page page_208 next page >
Page 208
language, teachers may have problems in providing an efficient and well-structured curriculum with equality of
provision and opportunity.
(3) The Canadian immersion experience ensures there is respect for the child’s home language and culture. This
relates to the additive bilingual situation. Parents have generally been seen as partners in the immersion movement
and some dialogue has existed with administrators, teachers and researchers.
(4) Immersion teachers in Canada tend to be committed to such immersion education. Research in Wales has
pointed to the crucial importance of teacher commitment to bilingual education in effecting achievement in school
(Roberts, 1985).
(5) It is important not to view immersion education in Canada in purely educational terms. Behind immersion
education is political, social and cultural ideology. Immersion education is not just immersion in a second language
(French). Such bilingual education has aims and assumptions, beliefs and values that sometimes differ from, other
times are additional to, mainstream education. It is important to see immersion education not just as a means to
promote bilingualism, but also as a move to a different kind of society (see Chapter 19). By promoting
bilingualism in English speakers, immersion education in Canada may support French language communities,
increase the opportunities for francophones outside Quebec and help promote bilingualism in the public sector (and
debatably in the private sector). However, immersion education is seen as a Trojan horse of further English
assimilation by some francophones. ‘Francophones question whether an increase in bilingual anglophones will
simply act to deprive them of their historical advantage in occupying bilingual jobs’ (Lapkin, Swain & Shapson,
1990: 649). This is linked to the finding that children from higher socioeconomic backgrounds tend to be over-
represented in immersion programs. Thus immersion education may act to reproduce elite groups, giving
anglophone children with bilingual abilities an advantage in the jobs market (Heller, 1994).
Heritage Language Education/Developmental Maintenance Education
Major reviews of heritage language education are provided by Cummins (1983a, 1993) and Cummins & Danesi
(1990). Apart from looking at individual international educational interventions, the reviews also look at the
pattern that can be found in the results of evaluations of heritage language education, thus attempting to derive
international generalizations.
The results of such evaluations, particularly in Canada, suggest that heritage language programs can be effective in
four different ways. First, the students maintain their home language. This is especially in comparison with
language minority children who are placed in mainstream or transitional education. Such mainstreamed children
tend partly to lose and sometimes avoid using their heritage language. Second, such children tend to perform as
well as comparable mainstream children in curriculum areas such as mathematics, science, history and geography.
That is, there is no loss in curriculum performance for such children taking their education in their home language.
Indeed the evaluations suggest that they perform better than comparable children in mainstream education. To
illustrate: take two ‘equal’ children from a language minority background.

< previous page page_208 next page >
< previous page page_209 next page >
Page 209
One attends a mainstream program, the other attends heritage language education. The chances are that the child in
heritage language education will achieve more highly, all other factors being equal. One ‘cognitive’ explanation is
that heritage language education commences at the level of linguistic-cognitive competence reached on entry to
school. (The cognitive reasons for this increased performance are considered in Chapter 9). In comparison,
mainstreaming such language minority students has negative cognitive implications. It seemingly rejects a child’s
level of cognitive competence. It entails re-developing sufficient language capability in order for them to cope
with the curriculum. If the analogy will stand, it is like someone with a basic level of skill in salmon fishing (with
a fishing rod) who is made to learn big game sea fishing instead. The instructor ignores skills already attained with
a rod. The student is made to practice casting on dry land, instead of building on existing skills with the fishing
Third, studies suggest that children’s attitudes are positive when placed in heritage language education. When the
home language is used in school, there is the possibility that a child’s self-esteem and self concept will be
enhanced. The child may perceive that the home language, the home and community culture, parents and relations
are accepted by the school when the home language is used. In comparison, a language minority child who is
mainstreamed is vulnerable to a loss of self esteem and status. The home language and culture may seem
disparaged. The school system and the teachers may seem latently or manifestly to be rejecting the child’s home
language and values. This may affect the child’s motivation and interest in school work and thereby affect
performance. A student whose skills are recognized and encouraged may feel encouraged and motivated; a student
whose skills are ignored may feel discouraged and rejected.
The fourth finding of heritage language evaluations is perhaps the most unexpected. Indeed, it tends to go against
‘common sense’. When testing children’s English language performance (or whatever the second language is for
that child), performance is generally comparable with mainstreamed children. To explain this, take the previous
example of two children from identical heritage language backgrounds with the same ‘intelligence’, socioeconomic
class and age. One is placed in heritage language education, the other in mainstream schooling. It might be
expected that the child placed in mainstream English language education would perform far better in English
language tests than the child in a heritage language education program. The prediction might be that the greater the
exposure to English in mainstream education, the higher the English language test performance. Evaluations of
heritage language education suggest something different. The child in heritage language education is likely to
perform at least as well as the child in mainstream education. The explanation seems to lie in self esteem being
enhanced, and language and intellectual skills better promoted by education in the home language. Such skills
appear to transfer easily into second language (majority language) areas.
While evaluations of heritage language education are positive, the Canadian population are divided on the issue.
For some, empowerment of heritage language groups is perceived as a major societal challengea challenge to
existing power and political arrangements

< previous page page_209 next page >
< previous page page_210 next page >
Page 210
(Taylor, Crago & McAlpine, 1993). Official Canadian policy has been supportive of multiculturalism, especially of
the two ‘solitudes’French language and English language cultures. Extending multiculturalism to other ‘heritage’
languages has been more contentious (Cummins, 1992). Ethnocultural Communities (e.g. Ukrainian, German,
Hebrew, Yiddish, Mandarin Chinese, Arabic and Polish) tend to support heritage language education. Anglophone
and francophone Canadians tend to have a tolerance and goodwill towards such communities. Lukewarm support
for heritage language communities tends to stop short if public monies are to be used to support heritage language
education. The anxieties of sections of public opinion and of government include: the disruption of mainstream
schools (e.g. falling rolls), problems of staffing, minimal communication between heritage language teachers and
mainstream teachers, segregation of school communities, financial burdens of the absorption of in-migrants into
the education system, loss of time for core curriculum subjects, social tensions, and effects on the integration and
stability of Canadian society (Cummins & Danesi, 1990; Cummins, 1992; Edwards & Redfern, 1992).
Such anxieties have increased with the high levels of Canadian in-migrants since the mid 1980s onwards. Due to
low birth rates and a rapidly ageing population in Canada, the population has been increased by in-migration
policies. Hence language diversity in Canada has increased. In Toronto and Vancouver, for example, more than
half the school population comes from a non-English speaking background (Cummins, 1992). With increased in-
migration and hence increased linguistic and cultural diversity, how does a government respond? What is the
consensus of public opinion in Canada about fostering language diversity and multiculturalism?
Cummins (1992: 285) portrays the debate about the nature of Canadian identity, the debate about multiculturalism
and the self-interest of public opinion thus:
‘While the dominant anglophone and francophone groups generally are strongly in favor of learning the other
official language, they see few benefits to promoting heritage languages for themselves, for Canadian society
as a whole, or for children from ethnocultural backgrounds. The educational focus for such children should
be on acquiring English and becoming Canadian rather than on erecting linguistic and cultural barriers
between them and their Canadian peers. In short, whereas advocates of heritage language teaching stress the
value of bilingual and multilingual skills for the individual and society as a whole, opponents see heritage
languages as socially divisive, excessively costly, and educationally retrograde in view of minority children’s
need to succeed academically in the school language.’
If the focus switches from public political opinion to the educational opinion of teachers, parents and students,
there is general satisfaction with Canadian Heritage Language programs. While such programs may present
administrative challenges (e.g. shortage of teachers, availability of pre-service and in-service teacher education and
a lack of curriculum materials), the advantages may be summarized as follows (Canadian Education Association,

< previous page page_210 next page >
< previous page page_211 next page >
Page 211
positive self concept and pride in one’s background;
better integration of child into school and society;
more tolerance of other people and different cultures;
increased cognitive, social and emotional development;
ease in learning of new languages;
increased probability of employment;
fostering stronger relationships between home and school;
responding to the needs and wishes of community.
The overall conclusions from Cummins’ (1983a, 1993) and Cummins & Danesi’s (1990) review of heritage
language education is that such education is not likely to have detrimental effects on a child’s performance
throughout the curriculum. Indeed, the indication from research is that language minority children tend to prosper
more in such education than when placed in mainstream education. They maintain and enrich their home language
and culture. Their performance throughout the curriculum does not suffer. This notably includes performance in
the second language (majority language). Cognitive enhancement can also occur (Cummins, 1993; Danesi, 1991).
The US Debate
The Baker and de Kanter Review of Bilingual Education (1983)
At the beginning of the 1980s, the United States Federal Government commissioned a major review of transitional
bilingual education by Keith Baker and Adriana de Kanter (1983). While in the 1960s and 1970s bilingual
education slowly evolved in the United States, in the late 1970s, 1980s and 1990s public support for bilingual
education tended not to favor such evolution. One branch of public opinion in the United States saw bilingual
education as failing to foster integration. Rather, such opinion saw bilingual education as leading to social and
economic divisions in society along language grounds. Minority language groups were sometimes portrayed as
using bilingual education for political and economic self interest, even separatism. Thus the Baker & de Kanter
(1983) review needs viewing in its political context (see Chapter 19).
Baker & de Kanter (1983) posed two questions to focus their review. These two questions were:
(1) Does Transitional Bilingual Education lead to better performance in English?
(2) Does Transitional Bilingual Education lead to better performance in non-language subject areas?
The review looked at bilingual education through ‘transitional’ eyes. It did not start from a neutral, comprehensive
look at the various different forms of bilingual education. Notice also the narrow range of expected outcomes of
bilingual education in the questions. Only English language and non-language subject areas were regarded as the
desirable outcome of schooling. Other outcomes such as self esteem, employment, preservation of minority
languages, the value of different cultures were not considered. Nor were areas such as moral development, social
adjustment and personality development considered.

< previous page page_211 next page >
< previous page page_212 next page >
Page 212
At the outset of their investigation, Baker & de Kanter (1981,1983) located 300 pieces of research from North
America and the rest of the bilingual world. Of these 300, they rejected 261 studies. The 39 they considered in
their review of bilingual education had to conform to six criteria. These may be listed as:
(1) English and non-language subject area performance must have been measured in the research.
(2) Comparisons between bilingual education and, for example, mainstream children must have ensured the groups
were relatively matched at the commencement. This means that initial differences between the two comparison
groups must have been taken into account. If not, the results may be explained by such initial differences (e.g.
different socioeconomic grouping) rather than the form of education in which children were placed.
(3) Baker & de Kanter (1983) required the studies to be statistically valid. For example, appropriate statistical tests
needed to have been performed.
(4) Some studies were rejected because they compared the rate of progress of a bilingual education sample with
national averages for a particular subject area. Such a comparison is invalid as the comparison would be between
bilinguals and monolingual English speakers, rather than two different groups of bilinguals in different forms of
(5) It was insufficient for chosen studies to show that a group of students had progressed over the year. Rather
‘gain scores’ needed to involve comparisons between different forms of schooling. That is, relative gain (one form
of bilingual education program compared with another) rather than absolute gain (how much progress made in a
specified time) was required.
(6) Studies were rejected which solely used grade equivalent scores. There are problems of comparability and
compatibility between students, schools and states when US grade scores are used.
The conclusion of Baker & de Kanter’s (1983) review is that no particular education program should be legislated
for or preferred by the US Federal Government:
‘The common sense observation that children should be taught in a language they understand does not
necessarily lead to the conclusion they should be taught in their home language. They can be taught
successfully in a second language if the teaching is done right. The key to successful teaching in the second
language seems to be to ensure that the second language and subject matter are taught simultaneously so that
subject content never gets ahead of language. Given the American setting, where the language minority child
must ultimately function in an English speaking society, carefully conducted second language instruction in
all subjects may well be preferable to bilingual methods’. (p. 51).
The review therefore came out in support of the dominant government preference for English-only and transitional
bilingual education. Functioning in the English language rather than bilingually was preferred. Assimilation and
integration appear as the social and political preference that underlies the conclusions.

< previous page page_212 next page >
< previous page page_213 next page >
Page 213
There has been considerable criticism of the Baker & de Kanter (1983) review (e.g. Willig, 1981/82; American
Psychological Association, 1982). The main criticisms may be summarized as follows: a narrow range of outcome
measures was considered, although this is often the fault of the original research rather than the review; focusing
on transitional bilingual education implicitly valued assimilation and integration and devalued aims such as the
preservation of a child’s home language and culture; and the criteria used for selecting only 39 out of 300 studies
was narrow and rigid.
While the studies included may be relatively more sophisticated, this still left studies with technical deficiencies
(e.g. studies with small samples) included in the review. It also excluded well-known and oft-quoted studies such
as the Rock Point Navajo research (Rosier & Holm, 1980; see also Holm & Holm, 1990). The review therefore
concentrated on a selective sample of technically superior research. It failed to look at patterns across the broadest
range of research. This is an issue returned to later in this chapter.
A further criticism of the Baker & de Kanter (1983) study provides a bridge with the next section. We have
suggested that single studies seldom provide a definitive answer. Therefore an overview strategy attempts to
integrate the findings of a variety of studies. Baker & de Kanter’s (1983) approach is narrative integration. This is
essentially an intuitive process, and the methods of procedure tend to be variable from reviewer to reviewer. A
comparison of the reviews of Baker & de Kanter (1983) with the earlier reviews by Zappert & Cruz (1977); Troike
(1978) and Dulay & Burt (1978, 1979), shows that reviews of similar studies can result in differing conclusions.
That is, different reviewers use the same research reports to support contrary conclusions.
An alternative and more rule-bound strategy is to use meta-analysis. This is a methodological technique which
quantitatively integrates empirical research studies. The technique mathematically examines the amount of effect
or differences in the research studies (Glass, McGaw & Smith, 1981; Hunter, Schmidt & Jackson, 1982). For
example, how much difference is there in outcome between transitional and immersion bilingual education? There
is no need to exclude studies from the meta-analysis which the reviewer finds marginal or doubtful in terms of
methodology. The quality of the evaluations is something that is examined in the meta-analysis and can be allowed
for statistically. Meta-analysis may show consistency of finding where narrative reviewers tend to highlight
variation and disagreement in findings (McGaw, 1988).
Willig’s (1985) Meta-analysis
Willig’s (1985) adopted a statistical meta-analysis approach to reviewing bilingual education. She selected 23
studies from the Baker & de Kanter (1981, 1983) review. All of her 23 studies concerned United States bilingual
education evaluations and deliberately excluded Canadian immersion education evaluations. As a result of the
meta-analysis, Willig (1985) concluded that bilingual education programs that supported the minority language
were consistently superior in various outcomes. Bilingual education programs tend to produce higher performance
in tests of achievement throughout the curriculum. Small to moderate advantages were found for bilingual
education students in reading, language skills, mathematics and overall achievement when the tests were in the

< previous page page_213 next page >
< previous page page_214 next page >
Page 214
second language (English). Similar advantages were found for these curriculum areas and for writing, listening,
social studies and self concept when non-English language tests were used.
Willig’s (1985) analysis also portrays the variety of bilingual education programs in existence that makes simple
generalization difficult and dangerous. For example, the social and cultural ethos surrounding such programs is
one major variation. Another variation is the nature of students in such programs and the variety of language
intake within such programs. For example, some bilingual education programs start with children at a similar level
of language skills. In other classrooms, there are various language and second language abilities, rendering
classroom teaching more difficult.
A criticism of Willig’s (1985) meta-analysis is that it only included 23 studies. An international review of the
bilingual educational effectiveness studies could have included many more studies and provided more
generalizable conclusions. Further criticisms of Willig (1985) are given in a response by Keith Baker (1987); this
includes a discussion of the relationship of meta-analysis to government policy regarding bilingual education.
While there are disagreements about the various reviews of US evaluation studies of bilingual education, there is a
consensus that more technically sophisticated and policy related research is required to make more rational
decisions of educational policy, provision and practice for language minority children in the United States and
beyond. One example follows:
Recent Research
Some major US research is now described to exemplify the problems of limited focus evaluations of bilingual
education, as well as to exemplify some recent trends in research findings. An eight year, congressionally
mandated, four and a half million dollar longitudinal study of bilingual education in the US compared Structured
English ‘Immersion’, Early Exit and Late Exit Bilingual Education Programs (Ramirez, Yuen & Ramey, 1991;
Ramirez, 1992). (The term ‘Immersion’ is not used in the original Canadian senseEnglish Submersion is more
accurate.) Dual-Language or other forms of ‘strong’ bilingual education were not evaluated. The focus was only on
‘weak’ forms of bilingual education. The programs compared ‘have the same instructional goals, the acquisition of
English language skills so that the language-minority child can succeed in an English only mainstream classroom’
(Ramirez, Yuen & Ramey, 1991: 1).
Over 2300 Spanish speaking students from 554 kindergarten to sixth grade classrooms in New York, New Jersey,
Florida, Texas and California were studied. Ramirez & Merino (1990) examined the processes of bilingual
education classrooms. The language of the classrooms were radically different in Grades 1 and 2:
‘Structured Immersion’ (Submersion) contained almost 100% English language.
Early Exit Transitional Bilingual Education contained around two-thirds English and one third Spanish.
Late Exit Transitional Bilingual Education moved from three quarters Spanish in Grade 1 to a little over half
Spanish in Grade 2.

< previous page page_214 next page >
< previous page page_215 next page >
Page 215
As a generalization, the outcomes were different for the three types of bilingual education. By the end of the third
grade, Mathematics, Language and English reading skills were not particularly different between the three
programs. By the sixth grade, Late Exit Transitional Bilingual Education students were performing higher at
Mathematics, English Language and English reading than other programs. Parental involvement appears to be
greatest in the late exit transitional programs. Although Spanish language achievement was measured in the
research, these results were not included in the final statistical analyses.
One conclusion reached by Ramirez, Yuen & Ramey (1991) was that Spanish speaking students ‘can be provided
with substantial amounts of primary language instruction without impeding their acquisition of English language
and reading skills’ (p. 39). When language minority students are given instruction in their home language, this
‘does not interfere with or delay their acquisition of English language skills, but helps them to ‘catch-up’ to their
English speaking peers in English language arts, English reading and math. In contrast, providing LEP students
with almost exclusive instruction in English does not accelerate their acquisition of English language arts, reading
or math, i.e., they do not appear to be ‘catching-up.’ The data suggest that by Grade 6, students provided with
English-only instruction may actually fall further behind their English speaking peers. Data also document that
learning a second language will take six or more years’ (Ramirez, 1992: 1).
This is evidence to support ‘strong’ forms of bilingual education and support for the use of the native language as a
teaching medium. The results also showed little difference between early exit and the English Immersion
(Submersion) students. Opponents of bilingual education have used this result to argue for the relative
administrative ease and less expensive mainstreaming (Submersion) of language minority students and for the
over-riding desirability of Transitional Bilingual education (e.g. Baker, K., 1992). Cziko (1992: 12) neatly sums up
the ambiguity of the conclusions, suggesting that the research: ‘provides evidence both for and against bilingual
education, or rather, against what bilingual education normally is and for what it could be’.
A series of reviews and criticisms of the Ramirez, Yuen & Ramey (1991) research followed (e.g. Cazden, 1992;
Meyer & Fienberg, 1992; Thomas, 1992) with particular emphasis on the following:
The benefits of ‘strong’ forms of bilingual education programs are not considered (e.g. Two Way bilingual
education; Heritage Language education). This makes statements about bilingual education based on an incomplete
range of possibilities (Cummins, 1992b). Mainstream classrooms with English Second Language (ESL) pull-out
(withdrawal) classes widely implemented in the US were also not included in the study (Rossell, 1992).
The range of variables used to measure ‘success’ is narrow. For example, language minority parents may expect
attitudinal, self esteem, cultural and ethnic heritage goals to be examined as a measure of successful outcomes
(Dolson & Meyer, 1992).

< previous page page_215 next page >
< previous page page_216 next page >
Page 216
The considerable differences that exist within bilingual education programs (let alone comparing differing types of
bilingual education program) makes comparisons and conclusions most difficult (Meyer & Fienberg, 1992). Also,
the complexity of organization within a school, the ethos and varying classroom practices makes categorization of
schools into watertight bilingual education programs formidable (Willig & Ramírez, 1993).
The National Academy of Sciences reviewed this (and other) studies and concluded that the design of the study
was ill-suited to answer key policy questions (Meyer & Fienberg, 1992; US Department of Education, 1992).
More clarity in the aims and goals of bilingual education in the US is needed before research can be appropriately
focused. Currently, the goals for bilingual education in the US are implicit and not defined.
There is a lack of data to support the long term benefits of late exit transitional programs over other programs
(Baker, K., 1992).
Expert Overviews of the Effectiveness of Bilingual Education
One inadequacy in bilingual education research is the relative absence of public opinion surveys. We lack the
evidence for the amount of parental and public support that exists for different forms of bilingual education in
various countries where bilingual education is a political and educational issue. Parents and children are rarely
asked about their degree of satisfaction with bilingual education during or after the experience. There is a paucity
of information about the democratic wishes of language minorities in the US. What does the US public think about
bilingual education? What are the varieties of opinion and viewpoint expressed? How do different forms of
bilingual education in the eyes of clients and consumers satisfy their wants and wishes? Is the debate only between
politicians and liberal academics? Are there Hispanics who prefer assimilation via bilingual education and are
there US academics who have a reasoned argument against bilingual education? There is a danger that we deal
with bilingual education controversies without fairly or fully representing the varied viewpoints of public,
politicians and policy makers. Rarely have the present or future clientele or the general public been asked their
opinions on the aims and nature of bilingual education.
An exception is Huddy & Sears (1990) who ‘telephone’ interviewed a US national sample of 1170 in 1983. They
found that while the majority tended to be favorable towards bilingual education, a substantial minority (around a
quarter of respondents, depending on the specific question) who included well-informed respondents, opposed
bilingual education particularly on the integration issue.
While public opinion surveys are infrequent, expert opinion is more likely to be privately or publicly sought.
Following the various narrative reviews of research on bilingual education and Willig’s (1985) meta-analysis, there
followed an expert overview in the US. The United States Committee on Education and Labor asked the General
Accounting Office (1987) to conduct a study on whether or not the research evidence on bilingual education
supported the current government preference for assimilationist, transitional bilingual education. The General
Accounting Office (1987) therefore decided

< previous page page_216 next page >
< previous page page_217 next page >
Page 217
to conduct a survey of experts on bilingual education. Ten experts were assembled, mostly professors of education,
selected from prestigious institutions throughout the United States. Each expert was provided with a set of
questions to answer in written form. The experts were asked to compare research findings with central political
statements made about such research. The purpose was to verify the veracity of official statements.
In terms of learning English, eight out often experts favored using the native or heritage language in the
classroom. They believed that progress in the native language aided children in learning English because it
strengthened literacy skills which easily transferred to operating in the second language. With the learning of other
subjects in the curriculum, six experts supported the use of heritage languages in such teaching. However, it was
suggested that learning English is important in making academic progress (General Accounting Office, 1987).
That all the experts did not agree was to be expected. The group of ten had diverse research backgrounds and
perspectives. Eight were knowledgeable about language learning and schooling for language minority children.
Two were expert on social science accumulation and synthesis (meta-analysis). These ten people were also sent ten
literary reviews to examine with the detailed questionnaire. The experts worked individually; they were not
brought together for an overall discussion to produce a consensus opinion. When the experts had submitted their
written report, they were given an opportunity to clarify and correct the General Accounting Office’s Report to
Congress of their views. Also, an outside evaluation expert reviewed the written responses of the experts, as well
as the draft text, to check on the accuracy of the representation of the experts’ views.
The key question is whether a different group or groups of ten experts would produce different conclusions.
Experts tend to disagree amongst themselves. This tends to reflect the developing nature of research in this area
and the complexity and political nature of what makes a particular school or program work successfully or not.
One problem is the effect of interacting factors with different types of bilingual education. For example, the
characteristics of the student, their parents and the community each serve to make a program, school or child more
or less successful. The degree of parental interest and involvement in bilingual education is sometimes seen as an
important intervening variable. Also, the status of the heritage language in the community and the country may
affect the success of a bilingual education program. If the analogy may stand, there are multitudinous ingredients
that go in the educational recipe. Focusing purely on the language part of schooling (e.g. bilingual or monolingual
education) only examines a narrow range of ingredients in the recipe. Complex reactions between ingredients
mean that making simple statements about what works successfully or not is difficult.
One area on which seven out of the ten experts agreed was that evidence did not exist on the long term effects of
various forms of bilingual education. Seven out of the ten firmly rejected the idea that there was support for
connecting bilingual education, either positively or negatively, to long-term outcomes. This reveals that research
on the effectiveness of bilingual education is still low down in the evolutionary process. In the

< previous page page_217 next page >
< previous page page_218 next page >
Page 218
experts’ survey, four out of ten experts agreed that the literature on language learning did not allow generalizations
to be drawn at this stage.
Having considered overviews of research on bilingual education, it is important to note one basic factor. There is
likely to be a divergence of opinion about the aims of education and bilingual education. This difference of
viewpoint will exist in both the academic and the non-academic outcomes of schooling. Some may emphasize
English language skills, some attainment throughout the curriculum, some the importance of second and even third
language learning. Others may focus on the non-academic outcomes of education such as moral and social skills,
employment, drop-out rates, absenteeism and self-esteem. At societal level, there will also be a variety of aims. For
some pluralism, biculturalism and multilingualism are a desirable outcome. For others the assimilation of minority
languages, the integration of minorities within overall society are the important outcomes. This suggests that a
definitive statement that bilingual education is more or less successful than, for example, mainstream education is
impossible due to the variety of underlying values and beliefs that different interest groups have about education
and the kind of future society desired.
Trueba (1989: 104) sums up the use of effectiveness studies by different interest groups for their own ends. ‘It is
unfortunate that bilingual education and other educational programs for minority students have become part of a
political struggle between opposing groups. Educators and parents have been forced into political camps, and
campaigned for or against these programs, without a thorough understanding of their instructional attributes and
characteristics. Perhaps it would be easier to reach a consensus regarding the nature of sound pedagogical
principles and practices rather than to continue to debate such politically loaded issues’.
Advancing the Effectiveness of Bilingual Education
Articles by Carter & Chatfield (1986), Lucas, Henze & Donateo (1990), Baker (1990) and Cziko (1992) have
suggested that the effectiveness of bilingual education question can be addressed from a different perspective.
Bilingual education research can look at effectiveness at four different levels. First, there is the effectiveness at the
level of the individual child. Within the same classroom, children may respond and perform differently. Second,
there is effectiveness at the classroom level. Within the same school and type of bilingual education program,
classrooms may vary considerably. It is important to analyze the factors connected with varying effectiveness at
classroom level. Third, effectiveness is often analyzed at the school level. What makes some schools more
effective than others even within the same type of bilingual education program and with similar student
characteristics? Fourth, beyond the school level there can be aggregations of schools into different types of
programs (e.g. transitional compared with heritage language programs) or into different geographical regions.
It is possible to look at effective bilingual education at each and all of these levels, and at the inter-relationship
between these four levels. For example, at the individual level we need to know how bilingual education can best
be effective for different social classes,

< previous page page_218 next page >
< previous page page_219 next page >
Page 219
and for children of different levels of ‘intelligence’ or ability. How do children with learning difficulties and
specific language disorders fare in bilingual education (Cummins, 1984a)? At the classroom level, we need to
know what teaching methods and classroom characteristics create optimally effective bilingual education. At the
school level, the characteristics of staffing, the size of groups and the language composition of the school all need
to be taken into account to find out where and when bilingual education is more and less successful.
Apart from individual classroom and school characteristics, the effectiveness of bilingual education can take into
account the social, political and cultural context in which such education is placed. For example, the differences
between being in a subtractive or additive context may affect the outcomes of bilingual education. The willingness
of teachers to involve parents, and good relationships between the school and its community may be important in
effective bilingual education. These different facets are included in the input-context-process-output model of
bilingual education (see Chapter 18).
It is also important in bilingual education effectiveness research to examine a wide variety of outcomes from such
education. Such variety may include examination results, tests of basic skills (e.g. oracy, literacy, numeracy), the
broadest range of curriculum areas (e.g. science and technology, humanities, mathematics, languages, arts,
physical, practical and theoretical pursuits, skills as well as knowledge). Non-cognitive outcomes are also
important to include in an assessment of effectiveness. Such non-cognitive outcomes may include: attendance at
school, attitudes, self concept and self esteem, social and emotional adjustment, employment and moral
The point behind such a comprehensive consideration of bilingual education is that effective bilingual education is
not a simple or automatic consequence of using a child’s home language in school (as in heritage language
education) or a second language (as in immersion bilingual education). Various home and parental, community,
teacher, school and society effects may act and interact to make bilingual education more or less effective. The
relative importance of different ingredients and processes in various school and cultural contexts needs
investigating to build a comprehensive and wide ranging theory of when, where, how and why bilingual education
can be effective.
This approach to studying effectiveness of bilingual education not only considers the infrastructure of such
education. It can also use the important studies from Britain and North America on what makes a school effective
(Hallinger & Murphey, 1986; Mortimer et al., 1988; Purkey & Smith, 1983; Reynolds, 1985; Smith & Tomlinson,
1989). For example, Mortimer et al. (1988) found that 12 factors were important in making a school effective.
These may be listed as: purposeful leadership by the head teacher, involvement of the deputy head teacher, the
degree of involvement of the teachers, consistency amongst teachers, having structured classroom sessions,
providing intellectually challenging teaching, a work centered environment, a limited focus within sessions,
maximum communication between teachers and students, good record keeping, plenty of parental involvement and
a positive classroom atmosphere.

< previous page page_219 next page >
< previous page page_220 next page >
Page 220
When the focus changes from school to teacher effectiveness when dealing with language minority students,
certain elements appear important (Tikunoff, 1983; E. García, 1988, 1991). These include:
(1) Teachers having high expectations of their students. Teachers being committed to the educational success of
their students and serving as student advocates.
(2) Teachers displaying a sense of confidence in their ability to be successful with language minority students.
(3) Teachers communicating directions clearly, pacing lessons appropriately, involving students in decisions,
monitoring students’ progress and providing immediate feedback.
(4) Teachers using students’ native language for instruction; alternating between languages in a compartmentalized
way to ensure clarity and understanding but without translating.
(5) Teachers integrating aspects of a student’s home culture and values into classroom activity to build trust and
self-esteem as well as promoting cultural diversity and cultural pluralism. Teachers involving parents in their
student’s learning and activities in the classroom.
(6) Teachers promoting a curriculum that has coherence, balance, breadth, relevance, progression and continuity.
(7) Teachers organizing instruction to utilize collaborative instructional techniques. Students interacting with
students in clearly explained, focused and meaningful activities.
(8) Teachers placing no undue pressure on students to proceed from literacy in the home to the second language
(e.g. Spanish to English literacy).
An example of research into bilingual education effectiveness is a case study by Lucas, Henze & Donato (1990) in
six schools in California and Arizona. This research revealed eight features seemingly important in promoting the
success of language minority students.
(1) Value and status were given to the language minority students language and culture. While English literacy
was a major goal, native language skills were celebrated, encouraged inside and outside the formal curriculum and
flagged as an advantage rather than a liability.
(2) High expectations of language minority students were prevalent. Apart from strategies to motivate students and
recognize their achievement, individualized support of language minority students was available. The provision of
counseling, co-operation with parents and the hiring of language minority staff in leadership positions to act as
role models were some of the ploys to raise expectations of success at school.
(3) School leaders gave the education of language minority students a relatively high priority. This included good
awareness of curriculum approaches to language minority children and communicating this to the staff. Strong
leadership, the will-

< previous page page_220 next page >
< previous page page_221 next page >
Page 221
ingness to hire bilingual teachers and high expectations of such students were also part of the repertoire of such
(4) Staff development was designed to help all the staff effectively serve language minority students. For example,
the teachers were provided with staff development programs which sensitized them to students’ language and
cultural background, increased their knowledge of second language acquisition and of effective curriculum
approaches in teaching language minority students.
(5) A variety of courses for language minority students was offered. Such courses included English as a second
language and first language courses. Small class sizes (e.g. 2025) were created to maximize interaction.
(6) A Counseling program was available. Counselors were able to speak the students’ home language, could give
post-secondary opportunity advice and monitored the success of the language minority students.
(7) Parents of language minority children were encouraged to become involved in their children’s education. This
included parents’ meetings, contact with teachers and counselors, telephone contact and neighborhood meetings.
(8) School staff were committed to the empowerment of language minority students through education. Such
commitment was realized through extra curricula activities, participation in community activities, interest in
developing their pedagogic skills and interest in the political process of improving the lot of language minority
This chapter has examined the development of studies which have investigated whether bilingual education is
more or less effective than monolingual education. It has also examined studies which look at the relative
effectiveness of different forms of bilingual education. Having considered the historical origins and the nature of
different forms of bilingual education, the chapter has sought to portray how questions about the effectiveness of
bilingual education have evolved. The initial studies examined individual programs and schools. A wide variety of
different outcomes and conclusions resulted. Following this first stage, the second stage reviewed the voluminous
research. This stage continues to the present. Reviews of Canadian immersion education, heritage language
education throughout the world, reviews of United States research by government officials and by meta-analysis
produced differing conclusions.
If there is a tentative overall pattern to the research, it would seem to be supportive of early total immersion
education for children whose first language is a majority language. The tentative conclusion also is that
maintenance or heritage language education has advantages for language minority children. Public opinion on such
matters tends to be divided about the effectiveness of bilingual education. This reveals the political undertones of
discussion about different forms of bilingual education.
However, one of the conclusions that comes from this chapter may be that simple questions give simplistic
answers. We cannot expect a simple answer to the question of whether or not bilingual education is more or less
effective than mainstream education. The

< previous page page_221 next page >
< previous page page_222 next page >
Page 222
question itself needs to become more refined. It needs to look at the different conditions under which different
forms of bilingual education become more or less successful. This means departing from simple studies and simple
results to broad investigations that include a wide variety of different conditions and situations. The scales of
justice of bilingual education cannot give a simple verdict. The evidence is wide and complex; witnesses have
complex accounts and arguments. There is no simple right or wrong, good and bad; no simple orthodoxy of
approach that can be guaranteed to give success.
It has been suggested that the effectiveness of bilingual education needs to consider children, teachers, the
community, the school itself and type of program. One particular factor cannot be isolated from another. We need
to consider a whole variety of ingredients at the same time, all of which make for a more or less successful recipe.
Children have a wide variety of characteristics which need investigating. Children cannot be isolated from the
classroom characteristics within which they work. Within the classroom there are a variety of factors which may
make for a more or less effective education. Outside the classroom the different attributes of schools may in their
turn interact with children and their classrooms to make education for language minority children more or less
effective. Outside the school is the important effect of community. The social, cultural milieu and political
environment in which a school works will affect the education of language minority children at all levels.
The key issue becomes ‘what are the optimal conditions for children who are either bilingual, becoming bilingual
or wish to be bilingual?’ Answers about optimal condition questions may involve a complex set of conditions.
Rather than a simple black and white sketch, a complex multicolored canvas may need to be painted.
Suggested Further Reading
ARIAS, M. and CASANOVA, U. (eds) 1993, Bilingual Education: Politics, Practice, Research. Chicago:
National Society for the Study of Education/University of Chicago Press.
BAETENS BEARDSMORE, H. (ed.) 1993, European Models of Bilingual Education. Clevedon: Multilingual
CAZDEN,C.B. and SNOW, C.E. 1990, English Plus: Issues in Bilingual Education. (The Annals of the American
Academy of Political and Social Science 508, March 1990). London: Sage.
CUMMINS, J. and DANESI, M. 1990, Heritage Languages. The Development and Denial of Canada’s Linguistic
Resources. Toronto: Our Schools/Ourselves Education Foundation & Garamond Press.
HELLER, M. 1994, Crosswords: Language, Education and Ethnicity in French. Ontario & New York: Mouton de
GARCÍA, O. and BAKER, C. (eds) 1995, Policy and Practice in Bilingual Education: A Reader Extending the
Foundations. Clevedon: Multilingual Matters. (Sections 1 & 2).
SWAIN, M. and LAPKIN, S. 1982, Evaluating Bilingual Education: A Canadian Case Study. Clevedon:
Multilingual Matters.

< previous page page_222 next page >
< previous page page_223 next page >
Page 223
Review and Study Questions
(1) Write short revision notes on the following:
(i) The reasons why individual researches vary in their findings.
(ii) What are the main findings from (a) Canadian Immersion research studies, (b) Heritage
Language/Developmental Maintenance, (c) Baker and de Kanter’s review, and (d) Ramirez, Yuen & Ramey’s
(1991) research.
(2) What are the differences between Canadian and United States contexts in the study of bilingual education
effectiveness? Are there similarities in their research results? Explain why there are differences.
(3) List those factors which seem to make bilingual education more or less effective. Using a rating scale (e.g. very
important to very unimportant), rate the importance to unimportance of these different factors.
(4) How do different approaches to bilingual education reflect different values among various groups of people?
Study Activities
(1) Read the Lucas et al. (1990) article in the Harvard Educational Review. Write a summary in approximately
600 words of that article. List those effectiveness factors which you think are part of any form of education, and
list separately those which you think are solely concerned with the language part of bilingual education. Are there
other factors which you think are important in effectiveness not mentioned in this article? Discuss in a group what
are the priorities in producing bilingual children through formal education.
(2) Make a list of ‘effectiveness factors’ from your reading. Following observation in one or more classrooms,
consider the effectiveness of these classrooms against this list. What factors seem, as the result of your classroom
observation, to be more and less important?
(3 Using the same list of ‘effectiveness factors’, study one program in your area (e.g. transitional bilingual
education). What features of that program are effective and less than effective due to the aims and nature of that
(4) Collect together newspaper clippings about language in schools within your area. Try to find different attitudes
in these clippings. Different newspapers may report the same story in different ways. Collect together these
different interpretations and try to give an explanation of the variation.
(5) Sit with the group of children or in a formal class of children, and attempt to time how much children and the
teacher spend using each language in the classroom. Try to work out over two or three sessions how much time is
spent in either language. When recording, try to locate which language is being used for what purpose. For
example, what language is used for classroom instructions, classroom management, discipline, questions, on-task
and off-task talk among children, greetings, rewards and reinforcement.

< previous page page_223 next page >
< previous page page_224 next page >
Page 224
(6) With a partner or in a small group, prepare a two minute television interview sequence on bilingual education.
Locate questions which are important and likely to be of concern to the public. Prepare short punchy answers to
these questions.
(7) Arrange a debate among students where two people support ‘weak’ forms of bilingual education, two support
‘strong’ forms. Present the arguments for two major kinds of bilingual education, highlighting those which are most
important in your region. Allow other students to ask questions of the four presenters.

< previous page page_224 next page >
< previous page page_225 next page >
Page 225
Chapter 12
Language Development and Language Allocation in Bilingual Education Settings
Language Development at School
Language Allocation in Bilingual Classrooms
Bilingualism, Deaf and Hearing Impaired People

< previous page page_225 next page >
< previous page page_226 next page >
Page 226
This chapter begins by examining the aims and goals of schools in the first language development of bilinguals.
Schools often play a major role in the maintenance and development of home language oracy, literacy and cultural
awareness. When two languages are used in the school, boundaries between the majority and minority language
may be necessary. The chapter therefore examines strategies for the separation of two languages in a school. It
also discusses dual language integration in the curriculum.
The chapter finishes with a special group of bilinguals who form their own language minority: deaf people. Often
a neglected language minority, it will be demonstrated that many of the attributes of hearing bilinguals are shared
by deaf bilinguals. Different views about deaf bilinguals are presented, followed by the debate about the kind of
bilingual education valuable for deaf students.
Language Development at School
For bilingual children, the school is usually an essential agent in developing the home, native, heritage language.
When a child enters kindergarten or elementary school, first language development needs to be formally addressed,
irrespective of whether that child has age-appropriate competency or not in the home language. While first
language development throughout schooling is important for majority and minority language children, the minority
context places extra reasons for careful nurturance of a minority language.

< previous page page_226 next page >
< previous page page_227 next page >
Page 227
Two quotes from a historically influential and important report by UNESCO (1953) entitled ‘The Use of
Vernacular Languages in Education’ provide the basis for the use of the home language in school, where possible:
‘It is axiomatic that the best medium for teaching a child is his mother tongue. Psychologically, it is the
system of meaningful signs that in his mind works automatically for expression and understanding.
Sociologically, it is a means of identification among the members of the community to which he belongs.
Educationally, he learns more quickly through it than through an unfamiliar linguistic medium. But as was
said earlier, it is not always possible to use the mother tongue in school and, even when possible, some
factors may impede or condition its use.’ (p.11)
‘It is important that every effort should be made to provide education in the mother tongue . . . On
educational grounds we recommend that the use of the mother tongue be extended to as late a stage in
education as possible. In particular, pupils should begin their schooling through the medium of the mother
tongue, because they understand it best and because to begin their school life in the mother tongue will make
the break between home and school as small as possible.’ (pp. 4748)
The relative status, nationally and internationally, of the minority language, the Anglophone nature of most mass
media, and the dominance of ‘common denominator’ majority languages outside the school requires special
attention to the continual evolution and progression of the minority language in school. Language minority
children sometimes start school speaking a different form of their heritage language compared with monolinguals.
For example, Greek and Bengali speaking five year old children in London will not tend to speak the same form of
Greek and Bengali as those from Greece and Bangladesh. Nor do they usually speak English in the same way as
English monolingual Londoners. In the US, many Latino children speak Spanish unlike their monolingual
counterparts, and speak English unlike those from English monolingual homes.
To preserve and reproduce the minority language in the young, a ‘strong’ form of bilingual education (that supports
the heritage language) needs to include a ‘first language’ program with explicit language aims and goals. Such a
program may involve lessons devoted to that language (e.g. Spanish listening, speaking, writing and reading
development). It may also valuably involve a strategy for heritage language development across the curriculum
(Corson, 1990a). Where children take lessons (e.g. Social Studies, Science) through the medium of their home
language, language development in those curriculum areas can be overtly and consciously fostered. A child’s home
language develops when it is cultivated, encouraged and promoted in a purposeful way in all curriculum areas.
The benefits for a well developed home (minority) language spread to the learning of a second or third language.
As Swain & Lapkin (1991a) found in research, those students literate in their heritage language progressed
significantly more in written and oral French than those without such skills. First language heritage language
literacy, in

< previous page page_227 next page >
< previous page page_228 next page >
Page 228
particular, enables relative ease of learning (and learning through) a second language by the transfer of knowledge,
language abilities (e.g. literacy strategies, communication skills) and learning processes.
With respect to learning a third language, there is belief that bilinguals are relatively better at learning a new
language than monolinguals, and the few studies that exist tend (although not unequivocally) to support this
assertion (Bild & Swain, 1989; Cenoz & Valencia, 1994; Genesee & Lambert, 1983; Hurd, 1993; Mägiste, 1984;
Ringbom, 1985; Saif & Sheldon, 1969; Swain, Lapkin, Rowen & Hart, 1990; Thomas, 1988). Cenoz & Valencia’s
(1994) conducted a study in the Basque country of 320 seventeen to nineteen year old students. They found that
the English language achievement of bilingual students (Spanish and Basque) was higher than monolingual
Spanish speakers. This result was found when other influential factors (intelligence, motivation, age and length of
exposure to English) were taken into account. Bilingualism, in and by itself, appears to give an advantage in
learning a third language. One explanation for this result is the greater metalinguistic awareness of bilinguals and
their possible greater sensitivity to communication (see Chapter 8). Another explanation may be in the transfer
between languages of phonological (sound system) and pragmatic (communication) abilities (Verhoeven, 1994).
This is predicted in Cummins’ (1986b) interdependence hypothesis that suggests such a transfer between
Speaking and Listening
What areas of first language development need to be cultivated in language minority children? One set of teaching
and learning targets for language minority children (whose first language is Welsh) comes from the National
Curriculum in Wales (Department of Education and Science and the Welsh Office, 1990). These targets or
language goals are somewhere in between wide and general curriculum aims and not so specific and fragmented as
behavioral objectives. The attainment targets are illustrated in summarized form below.
The Attainment of Skills, Knowledge and Understanding in Speaking and Listening
LEVEL 1 (Approximate Age 4 & 5)
Participate as a speaker and listener in group activities including imaginative play, (e.g. play the role of
shopkeeper or a customer in a class shop).
Listen attentively and respond to stories and poems, (e.g. re-tell a story, enact a poem, or draw a picture to
illustrate a story or poem).
Respond appropriately to simple instructions given by a teacher (e.g. follow two consecutive instructions, or
choose two flowers from the tray and draw pictures of them).
LEVEL 2 (Approximate Age 6 & 7)
Participate as a speaker and listener in a group engaged in a given task, (e.g. compose a story in a group).
Describe a real or imagined event to the teacher or another pupil (e.g. tell the listener about something which
happened at home).

< previous page page_228 next page >
< previous page page_229 next page >
Page 229
Listen attentively to stories and poems, and talk about them (e.g. state likes and dislikes about a story or poem).
Talk with the teacher, listen, ask and answer questions (e.g. talk to the teacher about events in or out of school).
Respond appropriately to complex instructions by the teacher, and give simple instructions (e.g. follow three
consecutive actions such as: list three places where a flower will grow best in the classroom; find out the views of
others in the classroom, and reach a consensus viewpoint).
LEVEL 3 (Approximate Age 8 & 9)
Relate real or imaginary events in a connected narrative which convey meaning to a group of pupils, or the teacher
or another known adult (e.g. tell a story with a beginning, middle and an ending).
Convey accurately a simple message (e.g. relay a simple telephone message).
Listen with an increased span of concentration to other children and adults, asking and responding to questions and
commenting on what is being said (e.g. listen to a new topic and discuss).
Give, receive and follow accurately precise instructions when pursuing a task as an individual or as a member of a
group (e.g. plan a wall display or arrange an outing together).
LEVEL 4 (Approximate Age 10 & 11)
Give a detailed oral account of an event, or something that has been learned in the classroom, or explain with
reasons why a particular course of action has been taken (e.g. report orally on a scientific investigation).
Ask and respond to questions in a range of situations with increasing confidence (e.g. conduct an interview on
radio devised with other pupils).
Take part as a speaker and listener in a group discussion or activity, expressing a personal view and commenting
constructively on what is being discussed or experienced (e.g. contribute to the planning and implementation of a
group activity).
Participate in a presentation (e.g. co-operate in describing the outcome of a group activity).
Level 5 (approximate age 11 to 13), Level 6 (approximate age 12 to 14) and Levels 7 to 10 (approximate age 14 to
16) also contain attainment targets that highlight: participating in and listening attentively to group discussion;
arguing and persuading by reasoning; effective oral presentation; expressing both opinions and counter-opinions in
a well organized way; leading a group discussion; summarizing a discussion in a balanced manner; discussion of
vocabulary (e.g. borrowings from English); regional variations; personal research using the minority language (e.g.
conducting an interview survey on the street).
In addition to listening and speaking attainment targets, the National Curriculum specifies reading and writing
targets for Welsh first language speakers. However, the inter-relationship and merger between the four language
skills (listening, speaking,

< previous page page_229 next page >
< previous page page_230 next page >
Page 230
reading and writing) may be prominent in a successful ‘language’ classroom. Integration rather than separation of
the four language skills is more customary and advantageous. A flavor of these reading and writing targets is now
presented, illustrated from the Welsh (as a first language) National Curriculum (Department of Education and
Science and The Welsh Office, 1990).
At Level 1 (age approximately 4 & 5), students should be able to appreciate that print conveys meaning, begin to
recognize the alphabet and a few words (e.g. their own and friends’ names), enjoy picking up a book, looking at the
pictures and pretending to read a book sequentially.
At Level 2 (age approximately 6 & 7), students develop sight-reading skills, initially using pictures to help
understanding and anticipation, begin to read semi-independently, recognize words in different contexts, talk about
the stories and poems they have read (e.g. relate a story, guess what comes next in a story) to show understanding.
At Level 3 (age approximately 8 & 9), students are expected to read in an independent and self-motivated manner,
to read to a group of friends with suitable colorful expression, talk in detail about the plot, characters, context and
ideas of a book.
At Level 4 (age approximately 10 & 11), students should be reading a variety of books, including fiction, non-
fiction and poetry, being openly aware of their favorite type of books, find information in encyclopedias, indexes,
content lists, databases, catalogs, dictionaries, describing and analyzing a relatively complex plot and show
awareness of rhyme and rhythm.
Language targets at Levels 5 to 10 (age approximately 11 to 16) include differentiating between fact and opinion,
cause and effect in a novel; reading a variety of print (e.g. magazines, reference works, archive material) to form a
synthesis; responding personally and creatively to stories, drama and poetry; discuss the language register, dialect,
imagery, atmosphere and style of different texts; recognizing devices to indoctrinate and persuade (e.g.
advertising); analyzing the relationship between print and graphics, and evaluating critically.
At Level 1 (age approximately 4 & 5), students should start distinguishing between pictures and handwriting,
numbers and letters; start to associate written letters with sounds, and write their own names and simple words.
At Level 2 (age approximately 6 & 7), students will write short passages, often referring to personal experiences,
in their own words. Sentencing will begin to evolve as will sense of sequencing. Written vocabulary will develop
alongside spelling and punctuation.
At Level 3 (age approximately 8 & 9), students are expected to write passages that convey feelings, are more
expressive, use more characterization, punctuate correctly (e.g.

< previous page page_230 next page >
< previous page page_231 next page >
Page 231
capital letters, full stops, question marks), plus use a word-processor to draft and re-draft work to edit out mistakes
and improve content.
At Level 4 (age approximately 10 & 11), students should write in different forms (e.g. diary, letter, report, story,
poetry), express opinions and thoughts in print, begin to summarize information, write with different readers and
audiences in mind, paragraph correctly, discuss their writing with peers and re-draft to improve content and
communication, and use correct tenses consistently.
At Levels 5 to 10 (age approximately 11 to 16), the development in writing skills is expected to include: writing
small plays conveying tension and conflict; writing dramatic and descriptive poetry; developing a logical, well
reasoned sequence of points and defending a viewpoint; writing clear instructions and explaining a process in
detail; describing unfamiliar and imaginary situations in authentic detail; avoiding stilted, uniform writing, using
variety in simple, compound and complex sentences; using metaphors and similes; conveying opinions clearly on
controversial topics and writing persuasive prose; integration of narrative and conversation; awareness of writing
for different social contexts: writing announcements and questionnaires, writing for a community newspaper, a
report on a TV program, or a football or basketball game, writing up a project, reporting to the police on a road
accident; an awareness of regional language variations and dialects; developing a recognizable personal style of
Cultural Awareness
While the four basic language skills are important in first language development, a language taught without its
attendant culture is like presenting a body without a heart. Language and culture are entwined in the healthy
functioning of a body. Therefore, developing heritage cultural awareness and multiculturalism alongside first
language teaching is an important element in language minority education.
Classroom activities to foster minority culture awareness can include: performing social conventions; cultural
rituals and traditions using authentic visual and written materials; discussing cultural variations (e.g. the colorful
kaleidoscope of Latin American dances, festivals, customs and traditions); identifying the varying experiences and
perspectives of the particular language variety (e.g. of French Canadians, of the French majority in France, of
bilinguals in France (e.g. Bretons, Provencal); classroom visits by native speakers of the language for ‘question and
answer’ sessions.
In Wales, developing a cultural awareness about Wales is contained in the concept of the ‘Curriculum Cymreig’
that endeavors to reflect the whole range of historical, social, cultural, political and environmental influences that
have shaped contemporary Wales. This involves giving students a sense of place and heritage, of belonging to the
local and a wider community with its own traditions, access to the literature of Wales, differences and traditions in
use of the Welsh and English languages, the distinctive nature of Welsh music, arts, crafts, technology, religious
beliefs and practices (ACAC, 1993). There is a clear emphasis on developing heritage cultural awareness in
Mathematics, Science and Technology and not just in Humanities and Aesthetic areas. For example, students study

< previous page page_231 next page >
< previous page page_232 next page >
Page 232
William Jones (16751749), a mathematician from Llanfihangel Tre’r’ Beirdd who first used the ‘pi (p) symbol and
Robert Recorde (died 1558) a mathematician from Tenby who devised the ‘equals’ (=) symbol. In Science, children
discuss local soil samples and how they connect with local farming and agriculture, coal and slate mining that have
been important to the Welsh economy, local small industries, and working from home via computer and
communication networks. Some of this involves cross-currcular activity.
It is sometimes argued that a minority language must be fostered to preserve the attendant culture. The opposite is
also tenable. The attendant culture must be fostered in the classroom to preserve the minority language. While
separation of culture and language is false, minority language culture can be weakly or strongly represented in the
classroom and in the whole ethos of the school. Such culture may be incidentally taught with little intent or
rationale. Alternatively, such culture may be consciously included in language teaching and the overall physical
and psychological environment of the school. This is particularly valuable to encourage participation by children
in their heritage culture. Language skills in the minority language are no guarantee of continued use of that
language into teens and adulthood. Enculturation therefore becomes essential if that language is to be useful and
To foster a minority language in school without fostering its attendant culture may be to fund a costly life-support
machine attached to a dying organism. To promote the attendant culture alongside minority language teaching may
be to give a life-preserving injection to that language and culture.
When the minority language and culture is fostered in the school, the majority language is also usually developed.
This raises an important question. How should these two languages be allocated in the curriculum? Can both
languages be used to transmit content? What boundaries are needed between languages? These questions about
bilingual methodology are now considered.
Language Allocation in Bilingual Classrooms
Apart from teaching the curriculum in the first language or, as in submersion education, solely in the second
language, there are a wide variety of possibilities in using two languages in the curriculum. One example is the
immersion strategy, for example in Canada, where majority language children are educated through the medium of
their second language, but where their first language is gradually used in the curriculum. Immersion education is
considered in detail in Chapter 17.
The purpose of this section is to look at different dimensions of ‘how and when’ two languages can be (1) separated
and (2) integrated in bilingual classrooms. The initial focus will be the separation of languages within the
curriculum. For example, when two languages are assigned to different areas of the curriculum, the intention is
usually to establish clear boundaries between the use of those languages in the curriculum. In the second part of
this section, an examination will be made of the more integrated or

< previous page page_232 next page >
< previous page page_233 next page >
Page 233
concurrent use of two languages. For example, a teacher may repeat an explanation in a different language. The
concluding part of this section will examine the problems and limitations of the concurrent use of two languages in
the classroom.
There are different reasons for separating or sometimes integrating languages in bilingual classrooms. First, if the
purpose of bilingual education is to obtain language shift towards the majority language (e.g. transitional bilingual
education), integrating languages may be preferable to separation. When there is integration and no strong
boundaries between languages, the majority language is expected to develop as the minority language diminishes.
Second, if the purpose of bilingual education is to maintain and develop the minority language, separation may be
important. Classroom activity needs to create a distinct space and time for minority language usage across much of
the curriculum. Clear boundaries for using the minority language, and separately using the majority language, need
establishing. Without boundaries and separation, the majority language may be increasingly used to the cost of the
minority language.
However, thirdly, if a strategic integration of languages in a class (e.g. where students are competent in both
languages) gives higher student performance in science, mathematics, humanities and the arts than language
separation, a concurrent use of two languages may be valued. For example, if a strategy of the teacher transmitting
a concept in Spanish followed by a written activity in English reinforced and better assimilated that concept,
concurrent use of languages may be desirable.
An example of a school situation where a decision about language methodology is necessary or desirable is a
mixed language classroom. Where there is a mixture of, for example, Spanish first language speakers and English
first language speakers in two way education, a clear policy and practice in language allocation is required. Where
children in linguistically mixed classrooms have different levels of ability in two languages, as well as different
levels of ability in tackling the curriculum, a bilingual methodology may be essential. In Wales, a class may
contain children who are relative beginners in Welsh (e.g. recent in-migrants), those with some second language
Welsh fluency, and native Welsh speakers. Some native Welsh speakers will be fluent in English, others may have
more competence in Welsh than English. A school that aims to achieve a high standard of ability in both
languages will need to consider the use and allocation of languages in classrooms and across the curriculum.
Language Separation
In the allocation of two languages in the classroom and in the curriculum, the need for distinct separation and clear
boundaries between two languages has often been advocated. First, such separation has support from the
sociolinguists who argue that, for a minority language to survive, it must have separate and distinct uses in society.
In the discussion of diglossia, maintaining a relatively stable societal arrangement for the separate use of two
languages was presented as important in minority language survival. Therefore, the argument has been in schools
that, for a minority language to have purpose and strength, it must have a distinct language allocation in
transmitting the curriculum.

< previous page page_233 next page >
< previous page page_234 next page >
Page 234
Second, from child development research with bilinguals, it has often been argued that the ‘one parent, one
language’ situation is one of the most effective patterns enabling childhood bilingualism to occur. If the mother
speaks one language and the father another language, there is a distinct separation of languages. The child will
therefore learn to speak one language to one parent, another language to the other parent. Such separation tends to
result in the successful development of both languages within the very young child because there are distinct
boundaries between the languages and a separate reception and production of language. Thus, an argument is that
effective bilingual schools should find ways of mirroring language separation in such ‘one parent, one language’
homes by establishing boundaries between languages. What can constitute boundaries will be examined in detail
Third, a constant problem in many minority language situations is the development of an unstable codemixing
among adults as well as children. Derogatory terms such as ‘Spanglish’ (for a mixture of Spanish and English) and
‘Wenglish’ (for a mixture of Welsh and English) highlight the concern of minority language preservationists about
codemixing. The random mixing of languages within sentences and across sentences can be a half-way house,
indicating a movement away from a minority language towards the majority language. Therefore, educationists
have often argued for the importance of separation of languages within the school. This is particularly to
strengthen and maintain the purity and integrity of a minority language.
Having established three bases for language separation in the classroom and curriculum, we will now consider
how languages can be separated. There are eight non-independent dimensions along which the separation of
languages can occur in school settings. These will now be considered in turn:
Subject or Topic
In elementary schools and high schools, different curriculum areas may be taught in different languages. For
example, Language, Arts, Social Studies, Religious Education, Art, Music and Physical Education may be taught
through the minority language (e.g. Spanish in the US; Filipino in the Philippines). Math, Science, Technology
and Computer Studies may be taught through the majority language (e.g. English). Certain curriculum areas are
reserved for one language, other curriculum areas for the second language. In kindergartens and elementary
schools, the allocation may be by topic rather than subject. For example, a project on ‘Conservation’ may be
through the minority language, a project on ‘weather’ through the majority language.
The examples given so far reveal ‘content sensitive’ issues. That is, there is sometimes a call for arts and
humanities culture to be relayed through the minority language, with Science and Technology relayed in an
international majority language. Curriculum areas such as History and Geography are often regarded as best
reflecting the heritage and culture of a language group. In contrast, Science and Technology may be seen as
international, and dominated by the English language. The danger is that the minority language becomes
associated with tradition and history rather than with technology and science. The minority language is thereby
allocated a lower status and seen as much less

< previous page page_234 next page >
< previous page page_235 next page >
Page 235
relevant to modern existence. The danger may be that the minority language will be seen as irrelevant in computer
communications, with a view that Science is white, western and mostly English in language.
The use of two languages in a school may be separated according to person. Just as in some homes there is the
‘one parent, one language’ system, so different school staff may be identified with different languages. For
example, there may be two teachers working in a team teaching situation. One teacher communicates with the
children through the majority language, the other teacher through the children’s minority language. There is a clear
language boundary established by person. Alternatively, teachers’ assistants, parents helping in the classroom,
auxiliaries and paraprofessionals may function in the classroom as an alternative but separate language source for
the children.
A frequently used strategy in language allocation in schools is for classes to operate at different times in different
languages. For example, in some dual language schools, one day may be through Spanish, the next day through
English. Other schools alternate by half days. The morning may be in Spanish, the afternoon in English, the next
day reversing that situation. A third alternative is for different lessons to be allocated to different languages. This
overlaps with the idea of separation by subject and topic.
The separation of time need not be solely in terms of days, half-days or lessons. It may be also in terms of whole
weeks or whole months or whole semesters. It also may valuably include a policy that varies by grade and age. For
example, children may be taught through the minority language for the first two or three years of elementary
education for 100% of time. Over Grades 3 to 7, an increasing amount of time may be allocated to the majority
language inside the school.
A lesser used means of language separation in the classroom is via different physical locations for different
languages. For example, if there are two Science laboratories in the school, one may be labeled as a Spanish
speaking laboratory, the other the English speaking laboratory. In Canadian immersion schools, students may have
different classrooms for French and English lessons, imitating the principle of territorial diglossia. The children are
expected to speak the language of that laboratory when entering it. The assumption is that a physical location
provides enough cues and clues to prompt the child to adhere to different languages in different places. In reality,
the teacher and other students may be more crucial in influencing the choice of language. Location is however, a
valuable ‘extra’ in encouraging language segregation.
A consideration of ‘place’ in language allocation needs to include all areas of the school. During morning and
afternoon breaks, lunch-breaks, religious services and announcements, music and drama events, games and sports,
the language balance of the school is affected. Sometimes teachers can allocate languages (e.g. in drama); other
times, students have a dominant influence (e.g. in the playground). However, all informal events

< previous page page_235 next page >
< previous page page_236 next page >
Page 236
in the school create a language atmosphere, a language balance, and are summed to create the overall language
experience of the school.
Medium of Activity
Another form of separation focuses on distinguishing between listening, speaking, reading and writing in the
classroom. For example, the teacher may give an oral explanation of a concept in one language, with a follow-up
discussion with the class in that same language. Then the teacher may ask the children to complete their written
work in a second language. This sequence may be deliberately reversed in a second lesson. Another example
would be when children read material through one language, and are then invited to write about it in the second
The aim of such a teaching strategy is to reinforce and strengthen learning by children. What is initially
assimilated in one language is transferred and reinterpreted in a second language. By re-processing the information
in a different language, greater understanding may be achieved. However, this example shows that language
separation soon links with a strategy to use both languages in the learning process. This idea of a rational,
structured and sequenced use of two languages in learning is considered laterwhen the concurrent use of languages
is examined.
One danger of ‘different medium’ separation is that one language may be used for oracy and another language for
literacy. Where a minority language does not have a written script this may be a necessary boundary. Where a
minority language has written materials, the danger is that the minority language will be identified with oracy and
the majority language with literacy. This has a potential effect of giving higher status and more functions to the
majority language and its associated literacy. Hence, where this kind of ‘medium separation tactic’ is used, it is
important for reversal to occur. For example, if oral explanations are given in one language with reading/writing in
a different language, on the next day or next lesson, that situation can be reversed.
Curriculum Material
There are a variety of ways in which written, audio-visual and information technology curriculum material can be
delivered in two languages, ensuring clear separation and non-duplication. Textbooks and course materials may
only be allocated in one language with the oral teaching in another language. For example, in Wales, the
equivalent of some three million US dollars is allocated for producing curriculum material in Welsh for all ages
and all ability levels. This is separation ‘across’ curriculum material; there is also separation possible ‘within’ such
material. For example, in elementary schools, dual language books (e.g. with English one side of the page and
Spanish on the other, sharing the same pictures) may be useful in the early stages of biliteracy. Consideration of
dual language books is given later in this section.
In high schools, reading material in History or Science may be as allocated by the teacher in both languages.
However, duplication of information and ideas is not necessarily present. Rather, there can be a continuation and
progression in themes and content. For example, a new topic may be introduced in one language by the student

< previous page page_236 next page >
< previous page page_237 next page >
Page 237
selected material. When a major sub-topic occurs, the teacher may require students to conduct their reading in a
different language. Use of both languages helps understanding of the content while separation is occurring. The
child does not have the choice of reading the English or the Spanish text, but needs to read both to gain a full
comprehension of the topic. Again, it will be seen that language separation does not necessarily mean using two
languages in isolation (or repeating the same content). A concern for separation can involves policy and practice
that includes more or less use of both languages. The outcome is a classroom paradox; languages that are separated
by function but are connected in an overall, holistic language and educational development policy.
Such a ‘curriculum material separation’ strategy tends to be used when, and only when both the child’s languages
are relatively secure, competent and well-developed. When this has occurred, the argument is that a child has to
think more deeply about the material when moving between languages, comparing and contrasting, developing the
theme of the material, assimilating and accommodating, transferring and sometimes translating in order to secure a
concept and understanding.
In schools and classes where there are bilingual children, there is often teaching in the majority language while the
management of the classroom occurs in the minority language. For example, in US ‘transitional bilingual
education’ schools with many Spanish speakers, the teacher may transmit the ‘formal’ subject matter in English,
while conducting ‘informal’ episodes in Spanish. In this example, the stipulated curriculum is relayed in the
majority language. However, when the teacher is organizing students, disciplining, in informal talks with
individual students or small groups, in short ‘extraexplanation’ episodes, a switch to the minority language is made.
The previous seven dimensions have suggested that a school or a teacher may develop a policy about language
allocation in a bilingual classroom. However, there is the real situation where students themselves help define the
language that is being used in a classroom. For example, if a student addresses the teacher, it may not be in the
language that the teacher has used to deliver the curriculum. To explain something with clarity, and for ease of
communication, the student may switch to his or her preferred language. Students influence when and where
languages are used, and affect boundary making. Thus separate dual language use can occur in the classroom by
varying circumstance, student influence, and the formality or informality of particular events (e.g. students may
feel at ease talking to teachers in their minority language in private or in classroom conversations).
Concurrent Uses of Languages in a Lesson
In many bilingual classrooms, the frequent switching between two or more languages is customary. The concurrent
use of two languages in a bilingual classroom tends to be regular in practice, but rare as a predetermined teaching
and learning strategy. Jacobson (1990) has argued that, on occasions, the integrated use of both languages rather

< previous page page_237 next page >
< previous page page_238 next page >
Page 238
language separation can be of value in a lesson. Such a purposeful and structured use of both languages inside a
lesson has certain problems that will be considered later. Four concurrent uses will now be considered (Jacobson,
Randomly Switching Languages
Bilingual children may switch languages both within sentences and across sentences. In many minority language
groups, this is frequent both in the home and street as it is in the school. It is relatively rare for such switching to
be stable across time. More often it is a half-way house, a sign of movement towards the majority language. For
minority languages to survive as relatively distinct and standardized languages, few would argue for such a random
practice to be encouraged in a bilingual classroom.
In some bilingual classrooms, teachers will repeat what they have previously said in another language. For
example, the teacher may explain a concept in Spanish, and then repeat the same explanation in English.
Everything is said twice for the benefit of children who are dominant in different languages. The danger is that the
student will opt-out of listening when the teacher is transmitting in the weaker language. The student knows that
the same content will be given in their preferred language and waits for that to occur. Such duplication appears to
result in less efficiency, less language maintenance value and less likely achievement in the curriculum.
Previewing and Reviewing
One strategy in the concurrent use of languages in a classroom is to give the preview in the minority language and
then the fuller review in the majority language. That is, a topic is introduced in the child’s minority language, for
example, to give an initial understanding. Then, the subject matter is considered in depth in the majority language.
This may be reversed. While there is an extension and reinforcement of ideas occurs by moving from one language
to another, there is sometimes also unnecessary duplication and a slow momentum.
Purposeful Concurrent Usage
In a variety of publications, Jacobson proposed the purposeful concurrent use of two languages (see Jacobson,
1990). The properties of what Jacobson calls the ‘New Concurrent Approach’ is that equal amounts of time are
allocated to two languages, and teachers consciously initiate movement from one language to another. Such
language movement inside a lesson occurs where there are discrete events and episodes with distinct goals for each
language There must be a conscious and planned movement from one language to another in a regular and rational
manner. Jacobson (1990) proposes this to strengthen and develop both languages, and to reinforce taught concepts
by being considered and processed in both languages. A use of both languages, it is suggested, contributes to a
deeper understanding of the subject matter being studied.
Jacobson (1990) suggests a variety of cues that can trigger a switch from one language to another. Examples of
such cues include:

< previous page page_238 next page >
< previous page page_239 next page >
Page 239
reinforcement of concepts;
capturing (and re-capturing) student attention;
praising and reprimanding students;
change of topic;
change in the stimulus material;
change from formality to informality;
to gain rapport;
when there is fatigue.
The amount of time allocated to each language in the curriculum is important. Some kind of balance (e.g.
50%/50%; or 75%/25%) between languages is needed, but more important is the purpose, manner and method in
which the two languages are used. All four language abilities may need fostering in both languages. To emphasize
oracy in one language and literacy in another language may result in lopsided bilingualism.
As is discussed in the limitations given below, there are potential problems in the complexity of managing,
allocating and organizing such a purposeful use of two languages. However, the value of the idea is that the
teacher plans the strategic use of two languages, thinks consciously about use of two languages in the classroom,
reflects and reviews what is happening, and attempts to cognitively stimulate students by a ‘language provocative’
and ‘language diversified’ lesson.
Issues and Limitations
In decisions about language separation and language allocation in a bilingual classroom, there are other ingredients
and contexts that need taking into account before an effective dual language policy can be formulated.
First, the aims of the school in terms of language preservation and second language competence need to be
examined carefully. School teachers are language planners, even if sub-consciously. Where a minority language is
to be preserved in the children, then separation may be a central part of the policy. Where teachers are more
enthusiastic about competence in the majority language (a second language), then different practices and outcomes
in terms of language allocation will be desired. Less language boundaries may be desired. There may be a
deliberate loosening of boundaries to ensure the development of the majority language.
Second, the nature of the students must be taken into account in any policy with regard to language boundaries and
concurrent use. Different policies and practices may be required according to the age and grade level of the
children. If the children’s language development is still at an early evolutionary stage, boundary setting will be
more important. With older children, whose languages are relatively well developed, the concurrent use of two
languages may be more viable and desirable. Older children may have more stability and separation in their
language abilities.
Third, this suggests that a static policy with regard to boundaries and concurrent use within the school is less
justifiable than a progressive policy that examines different uses

< previous page page_239 next page >
< previous page page_240 next page >
Page 240
across years and across grade levels. Early separation may be very important, later on in high school more
concurrent use may be more defensible to enable conceptual clarity, depth of understanding and possibly to
accelerate cognitive development.
Fourth, different dimensions of the school need different discussions about the separation and integration of
languages (e.g. curriculum, whole school policies, classrooms, lessons). At what level of organization should
separation occur? The discussion of language separation by curriculum material and medium of curriculum
delivery showed how language separation is not a distinct issue from rational concurrent use. Language separation
merges into a consideration of integrated use.
Fifth, the language balance of the class is often an important factor in a language separation decision. If all the
children speak a minority or majority language, there may be ease in determining a language allocation. However,
classes may be mixed, with differing balances of majority and minority language children. Who dominates
numerically, linguistically and psychologically in the classroom? When the balance is tilted against language
minority children, a clear separation with the curriculum balance towards the minority language may be desirable.
Whether the children in a school are language minority children in a subtractive or additive situation will also
impact on language allocation policies.
Sixth, where language minority children are in the numerical majority, a slow change rather than a sharp shift from
language separation to concurrent use may be advisable.
Seventh, a school policy needs to take into account ‘out of school exposure’ to the first and second language.
Sometimes, an equal amount of time is advocated for two languages in the school, with half the curriculum in one
language, the other half in a different language. If the child is surrounded by the majority language outside the
school (e.g. street, screen and shop), then the balance may need to be more towards the minority language in
Eighth, the replication and duplication of content is a danger in bilingual classroom methodology. Where the same
subject matter is repeated in a different language, some students will not concentrate or go ‘off task’. Yet in many
situations, where there is a mixture of different home and preferred languages in the classroom, for the sake of
comprehension, a teacher may need to repeat. There are some multilingual classrooms (e.g. in New York, Toronto
and London), with a considerable variety of languages represented among the children. This makes debates about
concurrent use and language boundaries all the more difficult. One solution is the possibility of individualized
learning via small groups (as occurs in progressive British and North American primary school practice). This may
allow some language boundaries to be established, with different children addressed (e.g. by teachers aides) in
their preferred language.
Ninth, the use of bilingual materials, particularly when they are built in a non-repetitive, non-parallel, but
incremental and well sequenced and structured manner, can be valuable, particularly in high schools. However, the
production and availability of materials in

< previous page page_240 next page >
< previous page page_241 next page >
Page 241
many minority languages is difficult to achieve. Funding may be difficult to secure for such separate minority
language materials.
Tenth, in a concurrent and purposeful use of two languages in the classroom as advocated by Jacobson (1990),
there is a danger of a unnatural, artificial, and highly complex language situation demanded of teachers. Where
teachers are expected to manage the concurrent use of two languages in a classroom, it assumes a very high level
of management skill, monitoring and reflection by such teachers. In reality, classrooms are busy, very fluid, often
unpredictable places. Teachers have to react to the moment, to individual students who don’t understand, with
many situations being unplanable and unpredictable. Students themselves have an important influence on language
use in the classroom, needing to communicate their understanding or lack of understanding in the most appropriate
way. Classroom management of learning and behavior needs to be fluent, accepted by students and not abrupt and
too unpredictable. Thus, language allocation in the classroom must fit naturally, predictably, fluently and flexibly
into the complex management of the curriculum.
This completes a consideration of the concurrent use and separation of languages in the classroom. The chapter
now moves on to focus on bilingual deaf people and issues in their schooling.
Bilingualism, Deaf and Hearing Impaired People
There is a growing awareness that those who are deaf or have partial hearing are already, or can become,
bilinguals. There is a also a current campaign for the deaf to become bilingual through learning to sign first of all,
followed by literacy in the language of the non-deaf (e.g. English in the United States, French in France). The aim
of this section is to examine the relationship between bilingualism and the deaf and partial hearing (College for
Continuing Education, 1992; Miller-Nomeland, 1993). We will find that there are many similarities between
hearing bilinguals and current interests in the languages of the deaf.
Deaf Bilinguals
Deaf individuals can become bilingual through learning to sign first of all, followed by literacy in a language of
the non-deaf. There are other forms of deaf bilingualism: for example, those who learn to speak from hearing
parents, followed by learning to sign. Others learn to sign first and then learn an oral form of the hearing language.
Others learn to communicate by different means, by signs, speaking and writing a Total Communication approach.
Some deaf people sign when in face-to-face communication, and use a written form (e.g. English) to communicate
to members of the deaf community via fax, textphone and letters. The path from signing to majority language
literacy is discussed later in this section.
For the moment, it is important to recognize that the deaf (like many hearing bilinguals) form relatively
disadvantaged language minorities and have certain things in common

< previous page page_241 next page >
< previous page page_242 next page >
Page 242
with hearing, language minority individuals and groups. Deaf people often have their own deaf language
community. The deaf will usually wish to identify with the deaf cultural community. Sometimes, they have been
required to attempt to assimilate into the hearing community. Those with a hearing loss vary in their identification,
sometimes identifying with deaf people, other times with hearing people, with many possible variations. Like most
hearing bilinguals, the deaf may use their two languages for different functions and purposes. For example, signing
may be used to communicate with the deaf community, while a spoken language or literacy in a majority language
is used to communicate with the hearing community. The deaf have too often been placed in deficit types of
education that submerge them in the language and culture of hearers rather than an enrichment model where
signing is allowed as the primary language.
Two Viewpoints about the Deaf
The first viewpoint is the medical view of deafness. Here deafness is defined as a defect or a handicap that
distinguishes ‘abnormal’ deaf people from ‘normal’ hearing persons. Deafness is seen as a condition that needs to be
remedied or cured as much as possible. Viewed as a disability, hearing aids and other devices that enhance hearing
or the understanding of speech are recommended. Deaf people are expected to become as ‘normal’ as possible by
avoiding purely visual methods of communication such as sign language, and learning spoken language as much as
is viable.
Sign language is perceived as a primitive, rudimentary, simple picture language whereas the spoken language is
seen as the natural language for all in the community, including the deaf. A central aim of education for the deaf
deriving from this viewpoint therefore becomes a mastery of spoken and eventually written language. Such
education, and social pressures within the community, aim for the integration or assimilation of deaf people with
hearing people. Deaf people existing in separate communities, miming to each other, using sign-speech or having
a deaf culture is not entertained as wholesome or desirable. Thus such educators, professionals and policy makers
see themselves as helping the deaf to overcome their ‘handicap’ and to live in the hearing world.
A second viewpoint is more in line with that increasingly expressed about hearing bilinguals: one of bilingualism,
vitality as a linguistic and cultural minority community, and the need for an enriching dual language education.
Such a viewpoint commences with the assertion that deaf people can do everything except hear. While there are
differences between deaf and hearing people, these are natural cultural differences, not deviations from a hearing
norm. Thomas Gallaudet, the founder of the first deaf school in the United States, argued that deaf children cannot
learn to speak or ‘speech-read’ well enough to use it as their primary means of communication. Signing for the
deaf is natural. As Veditz expressed in 1913, sign language is God’s noblest gift to deaf people (quoted in Miller-
Nomeland, 1993).
Thus, in this second viewpoint, deafness is regarded as a difference, a characteristic which distinguishes ‘normal’
deaf people from ‘normal’ hearing people. Deaf people are regarded as owning sign language which is a full
language in itself, grammatically complex and capable of expressing as much as any spoken language. Deaf people

< previous page page_242 next page >
< previous page page_243 next page >
Page 243
regarded as a linguistic and cultural minority that needs preservation, enrichment and celebration.
Instead of emphasizing the deficiencies of deaf people, their abilities are emphasized. A strong emphasis is placed
on the use of vision as a positive, efficient and full communicative alternative to audible speaking and hearing.
Sign language is thus used not only as equal to spoken language, but also as the most natural language for people
who are born deaf.
In deaf education, the focus is increasingly on language development through signing, and later on bilingualism
through literacy in the majority language. It is felt important that maximal language development occurs early on
through signing. Early signing enables a focus on the subject matter of the curriculum (rather than learning to
speak a majority language). Early use of signing in the classroom can enable deaf people to perform relatively well
in the curriculum. While there are currently few teachers fluent in sign, ideas about the importance of signing in
education are gradually evolving.
In this viewpoint, the deaf should form, where possible, a cultural community of their own. The deaf have a
common language in signing, a culture and a set of needs that are distinct from the hearing community. The deaf
community is seen as an important vehicle for the socialization of deaf and partially hearing people. Deaf adults
can provide important role models for deaf children, either as teachers or in the community. Thus, this second
viewpoint supports bilingualism, a deaf culture and the bilingual education of deaf people.
Differences in Bilingual Goals
It is important to recognize that, among the deaf and the hearing, there are not only different sub-groupings but
also differences of opinion about the appropriate language of the deaf, the education of the deaf and their
integration into mainstream speaking society. Such differences will now be briefly represented.
Hagemeyer (1992) suggests that there are nine sub-populations among the deaf in the United States. Such sub-
dimensions refer to deaf culture and not to whether a person is also a member of Hispanic, native American, Asian
or any other ethnic group that exists alongside being deaf. The nine groupings are as follows:
(1) Those who use sign language (e.g. American Sign Languageoften represented as ASL) as their primary
(2) Those who can communicate both in ASL and English.
(3) Those mostly from the hearing impaired group who can communicate primarily through speech.
(4) Adults who became deaf later in life, who were not born deaf and may have acquired speech before deafness.
Such people have the experience of hearing normally for a shorter or a longer period and may have speech patterns
relatively well embedded before deafness occurred.
(5) The elderly who became hearing impaired or deaf later in life as the result of the aging process.

< previous page page_243 next page >
< previous page page_244 next page >
Page 244
(6) Those who do not know either ASL nor English, but communicate through gestures, mime and their own
signing system. Such people may have been denied access to a deaf culture, to ASL and to education at an early
(7) Those who have residual hearing, perhaps describing themselves as hard of hearing, and who can hear with the
use of various aids.
(8) Those people who are deaf and blind. An example is Helen Keller.
(9) Those people who have normal hearning, but because their parents, children, or other members of the family
are deaf, they understand signing, or are fully conversant with deaf culture and integrate with the deaf community.
Sign Language
Sign language is a fully developed, real language with a full ability to communicate the same, complete meaning
as in spoken language. Sign language is not gesturing. Gesturingis relatively unsytematic, and is used in an ad hoc
way to express a small number of basic expressions. We all use non-verbal communication to add emphasis to our
speech. In contrast, signing is very comprehensive, structurally complex, rule bound, full means of
communication. Sign language can perform the same range of functions as a spoken language.
There is a variety of sign languages in existence throughout the world. American sign language, British sign
language. Chines sign language, Danish sign language, French sign language and Russian sign language are few
The Education of Deaf Students
There are a wide variety of approaches to the education of deaf students and thos who are hearing impaired. These
approaches range from minimal help to specially designed programs. At its worst, in mainstream ‘hearing’
education, the deaf may be regarded as having a serious mental as well as an auditory ‘defect’ and classified as
remedial. In the history of schooling in most countries, there are plentiful examples of such an insensitive and
uncivilized treatment.
In contrast, there are Special Schools and units for the deaf where, for example, the children are taught to
thoroughly learn signing first of all, are given a full curriculum mostly through signing, and develop written and/or
oral skills in the majority spoken language. Strong (1995) provides a review of nine programs in North America
that use American Sign Language (ASL) and English in the classroom, and aim to promote a ‘strong’ version of
bilingualism and biculturalism among the deaf. English is usually taught as a second language (ESL) with varying
emphasis on oracy and literacy in English.
There is considerable discussion, debate and development occurring in deaf education. There is not total
agreement as to preferred methods of approach: for example, using signing and literacy in the spoken language, a
Total Communication approach, using Sign Supported English or Signed Exact English, the kind of sign language
to use, whether to develop or and/or written skills in the spoken majority language, policies of integration with
hearing children and the development of relatively segregated deaf communities.

< previous page page_244 next page >
< previous page page_245 next page >
Page 245
One traditional approach has been to develop any residual hearing with the assistance of hearing aids and to
develop speech reading skills and speech production among the deaf and the hearing impaired. For most of this
century until the 1970s, this was the approach that dominated the education of deaf in North America and Europe.
Such an approach was based on a belief:
that deaf children should integrate into mainstream society;
that the curriculum could not be taught through sign language but required majority language proficiency;
that signing as a language was insufficient for full intellectual development;
that sign language was only a temporary crutch for those for whom the majority spoken language was essential;
that achievement in the curriculum requires oracy and literacy in the majority language (e.g. English).
A second approach developed since the 1970s has been based on the philosophy of Total Communication. All
modes of communication are regarded as appropriate for those who are deaf or partially hearing. Simultaneous
communication is used that combines auditory input plus visual information, for example, via the use of signed
English. However, a Total Communication approach is often assimilationist, aiming to enable deaf people to
communicate with hearing people.
A third approach concerns bilingualism for deaf children through bilingual education. This recent initiative is
contained in twelve suggestions that are mirrored in spoken bilingual education:
(1) That sign language should be the first language of all deaf children and be regarded as their primary language.
(2) That sign language should be used to teach curriculum subjects such as science, humanities, social studies and
(3) Sign language can be used to teach English or another majority language as a second language. Usually this
will be to teach reading and writing skills in English rather than English oracy.
(4) The culture and language of the deaf community are recognized and validated, with children learning that they
belong to the culture of the deaf. This approach tends to be favored by most but not all the deaf community, but
has not been favored by many politicians and education professionals who formulate policy and provision.
(5) Such bilingual education for the deaf is partly based on the research and arguments for an enrichment form of
bilingual education for hearing children:
bilingual education builds on a child’s existing linguistic and intellectual resources;
concepts and knowledge developed in the first language transfer easily to the second language;
use of a children’s heritage language gives pride and confidence in their culture and community;

< previous page page_245 next page >
< previous page page_246 next page >
Page 246
a child’s self-esteem and self-identity are boosted and not threatened by use of their first language;
school performance and curriculum attainment is raised when the first language is celebrated rather than
the lower achievement of minority language students and deaf students needs to be addressed by enrichment
forms (or ‘strong forms’) of bilingual education.
(6) Deaf children cannot acquire a spoken language easily or quickly because they have limited hearing abilities. If
the curriculum is transmitted in the spoken language, they are being expected to learn the content of the curriculum
using a level of language not yet acquired. This is analogous to minority language children being expected to
operate in submersion education in the language of the majority they have yet to master.
(7) English or another majority language is developed through sign language. Often native sign language teachers
will be employed in schools, where possible, to teach and to act as role models.
(8) The acquisition of a sign language should begin as early as possible, ideally soon after birth. Since 90% of deaf
children are born to hearing parents, this is often difficult, but with such parents being increasingly willing to sign
and learn signing, the first language of such deaf children can be sign language. Current thinking among the deaf
tends to suggest that early signing is preferable in most cases. Parents of deaf children need to be aware of deaf
communities, of bilingual education for deaf children, and to enhance their child’s curriculum achievement, to
expect signing as the medium of curriculum delivery plus literacy in the majority language.
(9) It is important to avoid language-delay in deaf children, as has been found to occur when using auditory
approaches and sometimes the Total Communication approach. Curriculum achievement will suffer if there is
language delay.
(10) The supply of trained personnel in deaf bilingual education, staff pre-service education programs, in-service
education and certification and funding are often current challenges that are being faced by deaf educators. These
are practical problems to be overcome rather than problems of principles that are insurmountable.
(11) The parents of deaf children need considerable emotional support, information and guidance to help their
children become bilingual. In order for cognitive, linguistic, social and emotional development to occur among
deaf children, there needs to be a partnership between school and parents, and between school and community.
While there is a considerable debate about the integration of deaf children into a hearing society, (and their first
loyalty being to the deaf community), hearing parents of deaf children need considerable support and sensitivity.
(12) The teaching situation in a bilingual deaf education system may involve team teaching. The deaf teacher may
be a natural model for the acquisition of sign language with a hearing teacher acting as a model for the acquisition
of proficiency in a majority language such as English or French. Ideally, both teachers should be bilingual models,
being able to communicate in both sign language and the ‘hearing’ language. Also, both teachers in the team
should have a knowledge of deaf culture, deaf differences and all the possibilities for deaf children and adults.

< previous page page_246 next page >
< previous page page_247 next page >
Page 247
In conclusion, this section on deaf bilinguals has shown that there are many similarities between hearing bilinguals
and deaf bilinguals. Many of the justifications for retaining a minority language child’s first language and for a
‘strong’ form of bilingual education for such children also hold for deaf children. The argument that children from
language minorities should become bicultural and culturally pluralistic also tends to hold for deaf bilinguals.
Language minorities are often the poor, low status, low power relations of majority language speakers. Even more
so the deaf. Deaf bilinguals are often the poor relations of language minority bilinguals. A spoken minority
language is often disdained and derided. Even more so sign language. Achievement and status among deaf
children and adults are often only recognized and granted when speech monolingualism occurs.
When the deaf come from language minority communities they form a minority of a minority. The examples of a
Latino deaf person in the US, a deaf Turk in Germany and a deaf Bengali in England all represent individuals who
are a minority within a minority. They are often the doubly underprivileged and the doubly despised. Where being
a member of a language minority is joined by being deaf, disempowerment, low status, discrimination and low self
esteem may be compounded. If many groups of bilinguals are underprivileged, even more so are deaf bilinguals.
This chapter has highlighted some of the key elements in the schooling of bilinguals. The importance of
developing the home language in its oral and literate form was initially stressed. Bilingual schooling appears an
essential, but not by itself, a sufficient condition for the reproduction of the minority language in students. Such
language development needs to occur across the curriculum, in Science as well as Language Arts, in Math and
Technology as well as History and Geography. Language development needs to be joined by cultural development.
Throughout the curriculum, in all areas and cross-curricula topics, the culture surrounding the minority language
needs transmitting to give that language purpose, meaning and vitality.
The minority language is also secured in students if there are clear boundaries between it and the majority
language in school. Separate functions for the two languages, and a clear demarcation in the classroom and
curriculum are particularly important for younger children in the elementary school. Later on, when students’ two
languages are both well developed, the integration and concurrent use of two languages in the classroom may be
considered as a means of reinforcement and reconceptualizing curriculum material.
The chapter ended by portraying the many parallels between deaf bilinguals and hearing bilinguals. Views about
deaf people vary, with the medical view and a ‘competent bilingual’ view contrasted in this chapter. Both views
link to different beliefs about the education of the deaf. The discussion of different forms of bilingual education
for the deaf highlight this topic as an important area for discussion and debate.

< previous page page_247 next page >
< previous page page_248 next page >
Page 248
Suggested Further Reading
CUMMINS, J. and SWAIN, M. 1986, Bilingualism in Education. New York: Longman.
GARCÍA, O. and BAKER, C. (eds.) 1995, Policy and Practice in Bilingual Education: A Reader Extending the
Foundations. Clevedon: Multilingual Matters. (Section 3).
JACOBSON, R. 1990, Allocating two languages as a key feature of a bilingual methodology. In R. JACOBSON
and C. FALTIS Language Distribution Issues in Bilingual Schooling. Clevedon: Multilingual Matters.
KYLE, J.G. (ed.) 1987, Sign and School. Clevedon: Multilingual Matters.
KYLE, J.G. 1988, Sign Language: The Study of Deaf People and their Language. Cambridge: Cambridge
University Press.
MASHIE, S.N. 1995, Educating Deaf Children Bilingually. Washington, DC: Gallaudet University.
McCRACKEN, W. 1991, Deaf-Ability Not Disability. Clevedon: Multilingual Matters.
OVANDO, C.J. and COLLIER, V.P. 1987, Bilingual and ESL Classrooms: Teaching in Multicultural Contexts.
New York: McGraw Hill.
WONG FILLMORE, L. and VALADEZ, C. 1986, Teaching bilingual learners. In M.C. WITTROCK (ed.)
Handbook of Research on Teaching (3rd edn). New York: Macmillan.
Review and Study Questions
(1) Write short revision notes on the following:
(i) The eight different ways languages can be separated in a classroom.
(ii) Examples of the concurrent use of two languages in a classroom, especially purposeful concurrent use.
(iii) The ten factors that affect decisions and strategies of language separation and concurrent use.
(iv) Two viewpoints about deaf people.
(v) Different approaches to the education of deaf students.
(2) What goals appear particularly important in the development of native language skills in school for one age
group of your choice?
(3) What are the reasons for language separation in a class? On what occasions might a concurrent use of two
languages be justified in a class?
(4) In a table or essay, show the similarities and differences between hearing bilinguals and deaf bilinguals.
(5) Read the article by Michael Strong (1995) ‘A Review of Bilingual/Bicultural Programs for Deaf Children in
North America’ in American Annals of the Deaf 140 (2), 8494 (or a similar article for your region nominated by
your teacher). Compare and contrast the schools using the following headings: (1) Aims of the school and mission
(2) Use of languages (ASL, English) in the school (3) Staff (4) Parents (5) Other particular features.

< previous page page_248 next page >
< previous page page_249 next page >
Page 249
Study Activities
(1) Interview some parents about bilingual education (preferably gathering a variety of different viewpoints). Ask
them what their preferences are for the use of the home language in school. Ask the parents if they want their
children to be bilingual as a result of schooling. Ask parents what value and use they see in languages being taught
or two languages being used as a medium of learning in school.
(2) Interview an administrator, a parent, a teacher and a principal about their attitudes to children’s home languages
in school. Find out what differences and what similarities there are between these different people. Try to explain
the origins and the causes of the differences. Are there different majority and minority viewpoints? Can there be
any integration of these varying viewpoints?
(3) Use one or more of the attitude scales in Appendix 1 with a classroom of children. In words and in simple
percentages, show the differences between students in attitude. If there are differences between groups (e.g.
gender, age groups, different language backgrounds), try to provide interpretations and explanations.
(4) Visit a school where there is a policy about the allocation of languages in the classroom. Try to locate a written
school policy about language allocation and compare it with what you observe in one or more classrooms. Discuss
your findings with a teacher. Chart the results of your observation and discussion with a teacher on a poster to
display to a small group. Compare your results with other members of the group.
(5) Visit a family with a deaf student and/or a school where there are deaf students. What are the goals in language
development? What form of bilingualism exists? What are the attitudes and aspirations of the student and
parents/teachers regarding language.

< previous page page_249 next page >
< previous page page_250 next page >
Page 250
Chapter 13
Bilingual Schooling Issues
Explanations of Under-Achievement
Special Needs and Bilingual Education
Assessment and Bilingual Children

< previous page page_250 next page >
< previous page page_251 next page >
Page 251
This chapter begins by examining some key themes in the education of bilinguals. Under-achievement is often
found among language minorities, both in leaving school relatively early and in lower performance in the
curriculum. The chapter examines the varying explanations for this, particularly whether bilingualism is a cause of
such under-achievement. This discussion extends into children with special needs, and the purpose and
effectiveness of bilingual special education. The role of assessment in the identification, and particularly the mis-
identification of bilingual children follows.
Explanations of Under-Achievement
There are frequent occasions when language minorities are found to under-achieve. Sometimes this is reported in
research as differences between ethnic groups (e.g. Department of Education and Science, 1985; Figueroa, 1984;
Tomlinson, 1986). It is also discussed by teachers, educational psychologists, speech therapists, parents and
students themselves in ‘single individual’ terms.
When language minority children appear to exhibit under-achievement in the classroom, what may be the
explanation? When first, second or third generation in-migrant children appear to fail in the classroom, where is
the ‘blame’ popularly placed? When guest workers’ children, indigenous minorities and distinct ethnic groups are
shown statistically to leave school earlier, achieve less in examinations and tests or receive lower grades and
averages, what is the cause?
First, the attributed blame may be on the child being bilingual. Bilingualism itself is often popularly seen as
causing cognitive confusion. The explanation given is a picture of the bilingual brain with two engines working at
half throttle, while the monolingual has one well tuned engine at full throttle. As Chapter 9 revealed, such an
explanation is usually incorrect. Where two languages are well developed, then bilingualism is more likely to lead
to cognitive advantages than disadvantages (see the three level house diagram in Chapter 9). Only when a child’s
two languages are both poorly developed can

< previous page page_251 next page >
< previous page page_252 next page >
Page 252
‘blame’ be attributed to bilingualism itself. Even then, the blame should not go to the victim, but to the societal
circumstances that create under-developed languages.
Second, where under-achievement exists, the reason may be given as lack of exposure to the majority language. In
the US and England, a typical explanation for the under-achievement of certain language minorities is insufficient
exposure to English. Failure or below average performance is attributed to students having insufficiently
developed English language competence to cope with the curriculum. Those who use Spanish or Bengali at home
and in the neighborhood are perceived to struggle at school due to a lack of competence in the dominant,
mainstream language. Thus submersion and transitional forms of bilingual education attempt to ensure a fast
conversion to the majority language (Cummins, 1980b).
A fast conversion to the majority language stands the chance of doing more harm than good. It denies the child’s
skills in the home language, even denies the identity and self-respect of the child itself. Instead of using existing
language skills, the ‘sink or swim’ approach attempts to replace those skills. The level of English used in the
curriculum may also cause the child to show under-achievement, with consequent demands for more of the same
medicine (more English language lessons).
Under-achievement in majority language education (e.g. submersion and transitional bilingual education) may be
combated by providing education through the medium of the minority language (e.g. Two-Way, Developmental
Maintenance, Heritage Language Programs). When the language minority child is allowed to operate in their
heritage language in the curriculum, the evidence (see Chapter 11) suggests that success rather than failure results.
Such success includes becoming fluent in the majority language (e.g. English). Thus lack of exposure to English is
a popular but improper explanation of under-achievement. This explanation fails to note the advantages of
education in the minority language for achievement. It inappropriately seeks an answer in increased majority
language tuition rather than increased minority language education.
Third, when bilingual children exhibit under-achievement, the attributed reason is sometimes a mismatch between
home and school. Such a mismatch is seen as not just about language differences but also about dissimilarities in
culture, values and beliefs (Delgado-Gaitan & Trueba, 1991). As an extreme, this tends to reflect a majority
viewpoint that is assimilationist, imperialist and even oppressive. The child and family is expected to adjust to the
system, not the system to be pluralist and incorporate variety. For such an assimilationist viewpoint, the solution is
in the home adjusting to mainstream language and culture to prepare the child for school. Past advice by some
educational psychologists and speech therapists has been for language minority parents to raise their children in the
majority, school language.
The alternative view is that, where practicable, the school system should be flexible enough to incorporate the
home language and culture. A mismatch between home and school can be positively addressed by ‘strong’ forms of
bilingual education for language minorities. By Two-Way, Developmental Maintenance and Heritage Language

< previous page page_252 next page >
< previous page page_253 next page >
Page 253
grams, through the inclusion of parents in the running of the school, by involving parents as partners and
participants in their child’s education (e.g. paired reading schemes), the mismatch can become a merger.
Fourth, under-achievement may be attributed to socioeconomic factors that surround a language minority group.
Some typical circumstances are described by Trueba (1991). ‘Many immigrant and refugee children have a life of
poverty and rural isolation in crowded dwellings where they lack privacy, toilet and shower facilities, comfort, and
basic medical attention. In some cases migrant life for children means abuse, malnutrition, poor health, ignorance
and neglect. Uprooting a child from his/her land can lead to a life of stigma and low status.’ (Trueba, 1991: 53).
Socioeconomic status is a broad umbrella term that rightly points to a definite cause of language minority under-
achievement. It provides an example of the importance of not blaming the victim, but analyzing societal features
that contribute to under-achievement. Such features may be economic deprivation, material circumstances and
living conditions as well as psychological and social features as discrimination, racial prejudice, pessimism and
immobilizing inferiority.
While socioeconomic factors are a proper partial explanation of language minority underachievement, two cautions
must be sounded. Socioeconomic status does not explain why different language minorities of similar
socioeconomic status may perform differently at school. In Chapter 19 we discuss the different ideologies or
orientations that may vary between ethnic groups. Sociocultural factors within and between ethnic groups and not
simply socioeconomic status are needed to begin to work out the equation of language minority achievement and
This raises another issue. Under-achievement cannot be simply related to one of several causes. The equation of
under-achievement is going to be complex, involving a number of factors. Those factors will interact together and
not be simple ‘standalone’ effects. For example, umbrella labels such as socioeconomic status need decomposing
into more definable predictors of under-achievement (e.g. parents’ attitude to education). Home factors will then
interact with school factors providing an enormous number of different routes that may lead to varying school
success and failure. The recipes of success and failure are many, with varying ingredients that interact together in
complex ways. However, socioeconomic and sociocultural features are important ingredients in most equations of
Fifth, part of the language minority achievement and under-achievement equation is the type of school a child
attends. This chapter and Chapter 11 highlight the different outcomes for language minority children in ‘strong’
compared with ‘weak’ forms of bilingual education. The same child will tend to attain more if placed in programs
that use the heritage language as a medium of instruction than in programs which seek to replace the home
language as quickly as possible. Therefore, when under-achievement occurs in a language minority child or within
a language minority group, the system of schooling needs scrutiny. A system that suppresses the home language is
likely to be part

< previous page page_253 next page >
< previous page page_254 next page >
Page 254
of the explanation of individual and ethnic group under-achievement where such problems exist.
Sixth, type of school is a broad heading under which there can exist superior and inferior submersion schools,
outstanding and mediocre Dual Language and Heritage Language schools. Where under-achievement exists, it is
sometimes too simple to blame the type of school rather than digging deeper and locating more specific causes.
Baker (1988,1993), Cummins (1984a) and Hornberger (1991) have listed some of the attributes that need
examining to establish the quality of education for language minority children (e.g. the supply, ethnic origins and
bilingualism of teachers, balance of language minority and language majority students in the classroom, use and
sequencing of the two languages across the curriculum over different grades, reward systems for enriching the
minority language and culture).
Seventh, under-achievement may be due to real learning difficulties and the need for some form of special
education. It is important to make a distinction between real and apparent learning difficulties. Too often, bilingual
children are labeled as having learning difficulties which are attributed to their bilingualism. As we have discussed
in this section, the causes of apparent learning problems may be much less in the child and much more in the
school or in the education system. The child is perceived as having learning difficulties when the problem may lie
in the subtractive, assimilative system which itself creates negative attitudes and low motivation. In the ‘sink or
swim’ approach, ‘sinking’ can be attributed to an unsympathetic system and to insensitive teaching methods rather
than individual learning problems. Apart from system-generated and school-generated learning problems, there
will genuinely be those who are bilingual and have learning difficulties (Cummins, 1984a). The essential
beginning is to distinguish between real, genuine individual learning difficulties and problems which are caused by
factors outside the individual.
Such a distinction between the real and the apparent, the system-generated and the remediable problems of the
individual, highlights the alternatives. When under-achievement exists, do we blame the victim, blame the teacher
and the school, or blame the system? When assessment, tests and examinations occur and show relatively low
performance of language minority individuals and groups, will prejudices about bilingual children and ethnic
groups be confirmed? Or can we use such assessment to reveal deficiencies in the architecture of the school
system and the design of the curriculum rather than blame the child? As this section has revealed, under-
achievement tends to be blamed on the child and the language minority group. Often the explanation lies in factors
outside of the individual. This discussion is now extended by examining bilinguals with special needs and
Bilingual Special Education.
Special Needs and Bilingual Education
It is initially important to define who are the children with special needs who may need Special Education
provision. Categories of special need vary from country to country but are likely to include the following areas:
visual impairment, hearing impairment,

< previous page page_254 next page >
< previous page page_255 next page >
Page 255
communication disorders, learning disabilities (e.g. dyslexia and developmental aphasia), severe subnormality in
cognitive development, behavioral problems and physical handicaps. Special needs can also include those with
exceptional abilities (e.g. high IQ, outstanding musical or mathematical aptitude). Some bilingual children will
have special needs, and this includes children from ‘elite’ bilingual families (e.g. English-German) as well as from
language minorities. In the United States, the Office of Special Education has estimated that nearly one million
children from language minority backgrounds are in need of some form of special education. This immediately
raises questions as to why language minority children may need special education.
(1) Do bilingual children experience special needs more frequently than other children due to their bilingualism
and their communicative differences from monolingual children?
(2) Are bilingual children more likely to experience any specific ‘problems’ due to their bilingualism (e.g.
stammering, language delay)?
(3) Are bilingual children often wrongly assessed and categorized as having a ‘disability’ and, as such, incorrectly
placed in special education?
(4) Are children of language minorities regarded as having special needs due to problems resulting from their
ethnic culture, community or family character? What are alternative causes of special needs and learning
difficulties in children?
These four questions will now be considered in turn.
Do Bilingual Children Experience Special Needs More Frequently Than Other Children Due to Their Bilingualism
and Their Communicative Differences from Monolingual Children?
It seems very unlikely that bilingualism is a direct cause of, or linked to the following main categories of special
need, namely visual impairment, hearing impairment, learning disabilities (e.g. dyslexia and developmental
aphasia), severe subnormality in cognitive development, behavioral problems and physical handicaps. Being a
member of a language minority may co-exist with such conditions, but is not connected.
There is some evidence in the United States (on a state by state analysis) that bilingual children are over-
represented among those in need of special education (Harry, 1992; Gersten & Woodward, 1994). Harry (1992)
also suggests that the rate of placement of Latinos in Special Education rises as the number of these students in the
school system increases. Any over-representation in the categories listed above is very unlikely to be a result of
bilingualism by itself. The causes of over-representation are considered later.
But what about language and communication disorders (e.g. language delay and stammering)? Are these
connected to bilingualism? This raises the second question.
Are Bilingual Children More Likely to Experience Any Specific ‘Problems’ Due to Their Bilingualism (e.g.
Stammering, Language Delay)?
The communicative differences of bilingual children must be distinguished from communicative disorders. The
failure to make this important distinction partly occurs

< previous page page_255 next page >
< previous page page_256 next page >
Page 256
because basic mistakes in assessment and categorization are made. The child is often assessed in a weaker, second
language, thereby inaccurately measuring both language development and general cognitive development. For
example, in Britain and the United States, in-migrant children are sometimes still assessed through the medium of
English and on their English proficiency. Their level of language competence in Spanish, Bengali, Cantonese,
Turkish, Italian or Panjabi, for example, is ignored.
The result is that such children are sometimes classed as having a ‘language disability’ and perhaps a ‘learning
disability’. Instead of being seen as developing bilinguals (i.e. children with a good command of their first
language who are in the process of acquiring a second, majority language), they may be classed as of ‘limited
English proficiency’ (LEP in the United States), or even as having general difficulties with learning. Their below
average test scores in the second language (e.g. English) are wrongly defined as a ‘deficit’ or ‘disability’ that can be
remedied by some form of special education.
Two particular language conditions require specific discussion: language delay and stuttering (stammering).
Language Delay
A particular language condition in children that illustrates an often wrongly attributed link between bilingualism
and developmental problems is ‘language delay’. Language delay occurs when a child is very late in beginning to
talk, or lags well behind peers in language development. Estimates of young children experiencing language delay
vary from 1 in 20 to 1 in 5 of the child population. Such varying estimates partly reflect that some delays are brief
and hardly noticeable. Others are more severe.
Language delay has a variety of causes (e.g. autism, severe subnormality, cerebral palsy, physical problems (e.g.
cleft palate), psychological disturbance, emotional difficulties). However, in approximately two-thirds of all cases,
the precise reason for language delay is not known. Children who are medically normal, with no hearing loss, of
normal IQ and memory, who are not socially deprived or emotionally disturbed, can be delayed in starting to
speak, slow in development or have problems in expressing themselves well. In such cases, specialist, professional
help needs to be sought. Speech therapists, clinical psychologists, educational psychologists, counselors or doctors
may be able to give an expert diagnosis and suggest possible treatment of the problem. It is important that such
professionals have an understanding of the nature of bilingualism in the clients they advise and treat.
Occasionally, well-meaning professionals make the diagnosis that bilingualism is the cause of language delay. If
the causes are unknown, bilingualism might seem a likely cause. Raising children bilingually is widely believed to
produce language delayed children. The evidence does not tend to support this belief.
For the teacher, psychologist, speech therapist, counselor and parent, a practical decision has to be made in respect
of language delayed bilingual children. Will the removal of one language improve, worsen or have no affect on
such a child’s language

< previous page page_256 next page >
< previous page page_257 next page >
Page 257
development? Given that the cause of the problem may be partially unknown, intuition and guesswork rather than
‘science’ often occurs as research in this area is still developing.
Let us assume the professional advice is to move from bilingualism to monolingualism. One issue immediately
becomes which language to concentrate on if there is major diagnosed language delay. The danger is that parents,
teachers and other professionals will want to accent the perceived importance of the majority language. In the
United States, the advice is often that the child should have a solid diet of English (at home and particularly at
school). The perceived language of school and success, employment and opportunity is the majority language. The
advice often given is that the home, minority language should be replaced by the majority language.
Even when professionals accept that bilingualism is not the cause of a child’s problem, moving from bilingualism
to monolingualism is seen by some as a way to help improve the problem. The reasoning is usually that the ‘extra
demands’ of bilingualism, if removed, will lighten the burden for the child. For example, if the child has an
emotional problem or a language delay condition, for whatever cause, simplifying the language demands on the
child may be seen as one way of solving or reducing the problem. The apparent complexity of a two language life
is relieved by monolingualism. Is this the rational and suitable solution?
There are many occasions when changing from bilingualism to monolingualism will have no effect on language
delay. For example, if the child seems slow to speak without an obvious cause, or seems low in self esteem,
dropping one language is unlikely to have any effect. On the contrary, the sudden change in schooling or family
life may exacerbate the problem. The child may be further confused, even upset, if there is a dramatic change in
the language of the school or family. If someone who has loved, cared for, or educated the child in one language
(e.g. a minority language) suddenly only uses another language (e.g. the majority language), the emotional well-
being of the child may well be negatively affected. The language used to express love and caring disappears.
Simultaneously, and by association, the child may feel that the love and care is also not as before. Such an
overnight switch may well have painful outcomes for the language delayed child. The mother tongue is denied, the
language of the family is implicitly derided, and the communicative medium of the community is disparaged. The
child may feel as if thrown from a secure boat into strange waters. The solution in itself may exacerbate the
An alternative is that the anchor language is retained. The home language gives a sense of assurance and security
amid stormy seas. Even if the child is slow in sailing in that language, with progress delayed, it is the boat known
to the child. Being forced to switch to the majority language will not make the journey faster or less problematic.
It is more important to learn to sail in a familiar boat (the home language) in minority language situations. Thus, in
most cases, it is inappropriate to move from bilingualism to monolingualism.

< previous page page_257 next page >
< previous page page_258 next page >
Page 258
However, it is dangerous to make this suggestion absolute and unequivocal. When there is language delay, there
may be a few situations where maximal experience in one language is preferable. For example, where one
language of a child is much securer and more well developed than another, it may be sensible to concentrate on
developing the stronger language.
This does not mean that the chance of bilingualism is lost forever. If, or when, language delay disappears, the
other language can be reintroduced. If a child with emotional problems and language delay really detests using (or
even being spoken to) in a particular language, as part of a solution, the family may sensibly decide to accede to
the child’s preference. Again, once behavioral and language problems have been resolved, the ‘dropped’ language
can be reintroduced, so long as it is immediately associated with pleasurable experiences.
Any temporary move from bilingualism to monolingualism should not be seen as the only solution needed. A
focus on such a language change as the sole remedy to the child’s problem is naive and dangerous. For example,
emotional problems causing language delay may require other rearrangements in the school or family’s pattern of
behavior. Language delay may require visits to a speech therapist for advice about language interaction between
the child and significant adults. Temporary monolingualism should only be seen as one component in a package of
attempted changes to solve the child’s language problem. However, it is important to reiterate that, in the majority
of cases, language delay will not be affected by retaining a bilingual approach.
For a short period, many children stutter (stammer). This includes repeating sounds or words (e.g. p-p-p-play with
m-m-me; I’ve got to – got to – got to go now); lengthening sounds (e.g. ffffish); time gaps between words (e.g. I’m
going . . .. . . home now); and unfinished words or phrases. Around three and four years of age, such stuttering is
particularly common. Many older children and adults, when under pressure, lose fluency. In addressing a large
audience, a speaker may be so tense as to lose customary smoothness in presentation.
The causes of stuttering are still not well understood and seem varied. Some neurophysiological theories locate the
problem in brain activity, others in a feedback problem between the ear and the brain. Further theories attempt a
psychological explanation in terms of personality traits, and particularly anxiety (e.g. caused by impatient,
overcorrective parents). None of these theories can explain all cases of stuttering.
One linguistic theory sees the cause as a difference between the language that is available in the brain (potential)
and control over speech apparatus (production). Another linguistic theory sees ‘cognitive overload’ as the cause
(e.g. the child is experiencing a difficulty in producing required complex sentences; the child’s capacity for fluency
does not equal the demands placed on the child). Such linguistic theories may have implications for a few young

< previous page page_258 next page >
< previous page page_259 next page >
Page 259
It has been argued in one case study that coping with two languages around the age of two to four may cause a few
children ‘cognitive overload’ (Karniol, 1992). The resulting argument was that bilingualism causes a few children
to require additional processing time (i.e. increasing the difference between ‘potential’ and ‘production’). This case
study may be rather atypical and idiosyncratic, and requires replication many times before the argument is
supported. However, it does reflect the assumption made about bilinguals who stutter by some doctors, teachers,
and other professionals. Overall, the evidence available indicates that bilinguals appear no more likely to stutter
than monolinguals (Saunders, 1988). Bilingual and multilingual countries do not appear to have a greater incidence
of stuttering than monolingual countries. The current conclusion seems to be that stuttering is very rarely related to
The Israeli case study referred to above suggests that any problems due to ‘cognitive overload’ are temporary in
bilingual children. As language competence in the two languages increases, stammering usually disappears. It is
also not clear from such research whether the causes were purely cognitive. Emotional problems (e.g. anxiety)
may have been a prime cause; such anxiety derived from sources other than bilingualism.
Methods of treatment of stuttering also hint that bilingualism is rarely a cause. Developing a stutterer’s breath
control, learning to speak more slowly, tension reduction and relaxation techniques are examples of treatments. If
potential causes of stuttering are more about specific anxieties, nervousness, worries about speaking, or more
general worries and fears in the child’s life, the remedy appears to be in those areas and not in bilingualism. For
example, teachers who criticize a child for stuttering, or parents who exhibit anxiety when their child stutters, only
increase the child’s nervousness and frequency of stuttering. If the sources of tension and anxiety in the child can
be located, reduced or swept away, stuttering is quite likely to be a temporary problem. For a very small number of
children, however, stuttering will be a continuous characteristic, and parents or teachers can do little about
treatment (e.g. when stuttering is related to brain activity).
When stuttering first occurs, it may not be necessary to seek immediate psychological or clinical advice. The sheer
act of highlighting the problem may exacerbate the problem itself. Showing anxiety, indicating to the child that a
problem exists, correcting the child and being impatient may only make the condition worse. Stuttering is frequent
in many young children, both monolinguals and bilinguals. It is usually temporary, and reappears when a child (or
adult) is particularly agitated and excited. If stuttering continues for a longer period, it may be wise to seek advice
from experts in child development or clinical psychology.
Stopping children speaking one of their languages will usually be counter productive. It will not change the
stuttering. Rather such an act will focus on language. This may increase children’s anxiety about their language
production. If it is genuinely felt that stuttering is caused by the child’s ‘potential with language’ not matching
‘ability in production’ or by ‘cognitive overload’ (see the linguistic theories mentioned above), teachers and parents
can attempt to:

< previous page page_259 next page >
< previous page page_260 next page >
Page 260
(1) relax the level of their questions and language demands on the child;
(2) ensure that language expectations are not unrealisticin both languages;
(3) be extra encouraging and patient when listening to the child, and
(4) in extreme cases, if felt strongly desirable, move to monolingualism for a temporary period.
Are Bilingual Children Often Wrongly Assessed and Categorized as Having a ‘Disability’ and, as such, Incorrectly
Placed in Special Education?
When bilingual children are assessed, it is important to keep three different aspects of their development distinct:
(1) first language proficiency, (2) second language proficiency, and (3) the existence (or not) of a physical,
learning or behavioral difficulty. This three-fold distinction enables a more accurate and fair assessment to be
made with regard to special education. The child’s level of functioning in a second language must not be seen as
representing the child’s level of language development. The child’s development in the first language needs to be
assessed (e.g. by observation if psychological and educational tests are not available) so as to paint a picture of
proficiency rather than deficiency. The child’s language proficiency is different from potential problems in an
individual’s capacities that require specialist treatment (e.g. hearing impairment, severely subnormal in ‘IQ’).
Neither the language and culture of the home, nor socioeconomic and ethnic differences should be considered as
handicapping conditions in themselves. Social, cultural, family, educational and personal information needs to be
collected to make a valid and reliable assessment and to make an accurate placement of the child in mainstream or
special education. This is considered separately in the following section on the assessment of bilingual children.
The Example of the United States
In the United States, Public Law (94142) gives the right to assessment that is not culturally discriminatory, to tests
in the child’s native language, to multi-dimensional ‘all areas’ assessment for all ‘handicapped’ students. The
misdiagnosis of bilingual students for special education has led to court cases (e.g. Diana v. The California State
Board of Educationsee later in this chapter). Such court cases revealed how bilingual students were wrongly
assessed as in need of special education (Maldonado, 1994). In some cases, teachers were unsure how to cope with
a child whose English was relatively ‘weak’. On this basis only, the teacher wanted special education for the
‘Limited English Proficient’ child.
Such litigation has showed the importance of separating bilinguals with real learning difficulties from those
bilinguals whose second language (e.g. English) proficiency is below ‘native’ average. The latter group should not
be assessed as having learning difficulties and therefore in need of special education (Figueroa, 1989; Mercer,
1973; Gersten & Woodward, 1994). The litigation also showed the wrongs done to bilingual students:
misidentification, misplacement, misuse of tests and failure when allocated to special education.

< previous page page_260 next page >
< previous page page_261 next page >
Page 261
Unfortunately, the fear of litigation by school districts can lead to an over-referral of bilingual students with a real
need of special education. In the early 1980s, the trend was (in California, for example) to assume that too many
language minority students were in need of special education. When students did not appear to be benefiting from
instruction in ‘regular’ classrooms, special education classes became the easy answer. Or, if teachers were unsure
how to deal with a behavioral or learning problem, transfer to special education provision became an instant
Towards the end of the 1980s, this had been reversed. The tendency moved to underestimating the special needs of
language minority children (Gersten & Woodward, 1994). Wrongful placement of children in special education
(over-referral) made various administrators cautious of special education placement. A fear of legal action by
parents and a realization that assessment devices often had low validity led administrators to be hesitant to place
bilingual children in Special Education.
The see-saw between over-estimation and under-referral of special education need makes accurate assessment very
important. The section on the assessment of bilingual children raises the important issues about valid assessment of
special need. However, accurate assessment and placement in different schools is not enough. The development of
effective instruction strategies and an appropriate curriculum is crucial. So is the need to train teachers for
bilingual students in special education. Educating the parents of special needs children is also a high priority.
Assessment will sometimes locate those who are bilingual and have a physical, neurological, learning, emotional,
cognitive or behavioral difficulty. Such children may need some kind of special education or intervention. One
estimate in the United States is that 1 in 8 (approximately 12%) of language minority students will fit into this
category. Similar figures are quoted in other countries. What form of Special Education should such children
receive? Should such education be in their home language, where feasible, or in the majority language of the
region? Or should such children be provided with education that uses both home and majority languages?
Bilingual Special Education
Special Education bilingual children can be served by a variety of institutional arrangements (Cloud, 1994). These
include: Special Education schools (resident and non-resident), hospital-based education, residential homes,
Special Education Units attached to mainstream schools, specially resourced classes in mainstream schools,
withdrawal and pull-out programs (e.g. for extra speech and language help, behavioral management) and special
help given by teachers, paraprofessionals or support staff in ‘regular’ classes. The extent to which such provision
will be bilingual or monolingual will vary within and across such institutional arrangements. Such bilingual or
monolingual provision will also depend on availability of provision (material and human), the type and degree of
special education need or condition, proficiency in both languages, learning capacity, age, social and emotional
maturity, degree of success in any previous education placements, and not least, the wishes of the parents and

< previous page page_261 next page >
< previous page page_262 next page >
Page 262
When bilingual or language minority children have been assessed as having special needs, many educators will
argue that education solely in the dominant, majority language is needed. In the United States, the advice is
sometimes given that Latino and other language minority children with special needs should be educated in
monolingual, English language special schools. The argument is that such children are going to live in an English
speaking society. When there is serious mental retardation, it seems sensible that a child should be educated
monolingually, in the minority or majority language. Such a child develops very slowly in one language.
Many special needs children will benefit considerably from bilingual special education rather than monolingual
special education (Baca & Cervantes, 1989; Carrasquillo, 1990). One example is the recently arrived in-migrant,
special needs child. Placing such a child in a class where he or she doesn’t speak the language of the classroom
(e.g. English in the United States) will only increase failure and lower self-esteem. To be educated, the child
preferably needs initial instruction mostly in the first language, with the chance to become as bilingual as possible.
Most children with special needs are capable of developing in two languages. Many do not reach levels of
proficiency in either language compared with peers in mainstream classrooms. Nevertheless, they reach
satisfactory levels of proficiency in two languages according to their abilities. Becoming bilingual does not detract
from achievement in other areas of the curriculum (e.g. mathematics and the creative arts). Canadian research
tends to show that less able bilingual children share some of the cognitive advantages of bilingualism (Rueda,
1983). Just as their mathematical ability, literacy and scientific development may occur at a slower pace, so the
two languages will develop with less speed. The size of vocabulary and accuracy of grammar may be less in both
languages than the average bilingual child. Nevertheless, such children acquiring two languages early, will usually
be able to communicate in both languages, often as well as they would communicate in one language.
The movement of a bilingual student into special education often occurs after a conclusion is reached that the
child’s needs cannot be met by integration in a regular (i.e. non-special) school. This holds for those currently in
bilingual education who may need special bilingual education because generally, integration is regarded as
preferable to segregation. If children are placed in bilingual special education, it is important that they gain the
benefits of those in other forms of bilingual education: dual language competence, biculturalism and
multiculturalism, and other educational, cultural, self-identity and self-esteem benefits. These benefits have been
discussed previously in this book.
One example not discussed so far is when a child is failing in a mainstream school due to his or her language
proficiency not being sufficient to operate in the curriculum. For example, in the United States, some Spanish
speaking children are in mainstream schools (a ‘submersion’ experience) and, although of normal ability, fail in the
system (e.g. drop out of school, repeat grades, leave High school without a diploma) because their English
proficiency is insufficiently developed to comprehend the increasingly complex curriculum.
This situation creates an apparent dilemma. By being placed in some form of special education, the child is
possibly stigmatized as having a ‘deficiency’ and a ‘language

< previous page page_262 next page >
< previous page page_263 next page >
Page 263
deficit’. Such special education may be a separate school (or a special unit within a larger school) that provides
special (‘remedial’) education for bilingual children. Such schools and units may not foster bilingualism. Often,
they will emphasize children becoming competent in the majority language (e.g. English in the United States).
Such segregation may allow more attention to the second language but result in ghettoization of language
minorities. While giving some sanctuary from sinking in second language submersion in a mainstream school,
special education can be a retreat, marginalizing the child. Will children in such special education realize their
potential across the curriculum? Will they have increased access to employment? Will their apparent failure be
accepted and validated because they are associated with a remedial institution? Will there be decreased
opportunities for success in school achievement, employment and self-enhancement?
The ideal for children in this dilemma may be neither mainstreaming nor special education. It is education which
allows them to start and continue learning in their first language. The second language is nurtured as well to
ensure the development of bilinguals who can operate in mainstream society. In such schools, both languages are
developed and used in the curriculum. Such schools avoid the ‘remedial’ or ‘compensatory’ associations of special
education. Such schools celebrate the cultural and linguistic diversity of their students.
Yet such bilingual education is sometimes in danger of being seen as a form of special education. Even when the
‘language delayed’ are separated from those who are in the early stages of learning the majority language (e.g.
English in the United States), the danger is that the latter will still be assessed as in need of compensatory,
remedial special education, including being allocated to bilingual education.
Are Children of Language Minorities Regarded as Having Special Needs Due to Problems Resulting From Their
Ethnic Culture, Community or Family Character? What Are Alternative Causes of Special Needs and Learning
Difficulties in Children?
The discussion of Special Education for bilingual children has suggested that one unfair assumption made about
bilingual children who show learning difficulties is that bilingualism is the cause of those difficulties. Bilingualism
is rarely a cause of learning difficulties. Learning difficulties are caused by a variety of possibilities, almost none
of them aligned to bilingualism. Six examples of causes outside the child and his or her bilingualism follow. This
list, which is not exhaustive or comprehensive, will indicate that bilingualism has nothing directly to do with many
learning problems, either as a secondary or a primary cause.
(1) Poverty and deprivation, child neglect and abuse, feelings of helplessness and desperation in the home,
extended family and community may create personality, attitudinal and learning conditions that make assessment
of learning difficulties more probable. Sometimes, such assessment will reflect prejudice, misjudgements and
misperceptions about the child’s home experiences. The learning problem may thus be in a mismatch between the
culture, attitudes, expectations about education and values of the home and school. In Chapters 15 and 16 on
literacy and biliteracy, this

< previous page page_263 next page >
< previous page page_264 next page >
Page 264
variation is examined with regard to local literacy and variations in ethnic group and social class uses of literacy.
Different beliefs, culture, knowledge and cognitive approaches may be devalued with the child immediately
labeled as of inferior intelligence, academically incompetent and of low potential.
(2) The problem may be in the standard of education. A child may be struggling in the classroom due to poor
teaching methods, a non-motivating, even hostile, classroom environment, a dearth of suitable teaching materials,
or clashes with the teacher.
(3) The school may be inhibiting or obstructing learning progress. If a child is being taught in a second language
and the home language is ignored, then failure and perceived learning difficulties may result. One example is that
of some Spanish speaking children in the United States. Such children are often placed in English only classrooms
on entry to school. They must sink or swim in English. Some swim; others sink and may be deemed to have a
deficiency. By being assessed in their weaker second language (English) rather than in their stronger home
language (Spanish), such children are labeled as in need of special or remedial education. Thus the monolingual
school system is itself responsible for learning failure. A school that promoted bilingualism would probably ensure
learning success for the same child.
(4) Another set of causes of learning difficulties are a lack of self confidence, low self esteem, a fear of failure and
high anxiety in the classroom.
(5) A fifth possibility is failure caused partly by interactions among children in the classroom. For example, where
a group of children encourage each other to fool around, have a low motivation to succeed, or where there is
bullying, hostility, social division, rather than cohesion among children in a classroom, the learning ethos may
hinder the child’s development.
(6) Another case is where the child is a slow learner, where there is a mismatch between the gradient of learning
expected and the ability level of the child. Some children learn to read more slowly than others, still learning to
read well, but after a longer period of time. Less able children can learn two languages within the (unknowable)
limits of their ability. Other children experience specific learning difficulties (for example, dyslexia, neurological
dysfunction, ‘short-term memory’ problems, poor physical co-ordination, problems in attention span or
motivation). None of these specific learning difficulties or other language disorders are caused by bilingualism. At
the same time, bilingual children will not escape from being included in this group. Bilingual families are no less
likely to be affected than other families.
Almost the only occasion when a learning difficulty of a bilingual child is attached to bilingualism is when a child
enters the classroom with neither language sufficiently developed to cope with the higher order language skills
demanded by the curriculum. In the rare cases where a child has simple conversational skills in two languages but
cannot cope in the curriculum in either language, language may be related to learning difficulties. In this case, the
problem is not really with bilingualism but with insufficient language practice in the home, in the nursery school
or in the outside world. Here we are talking not about bilingual deprivation, but about deprivation in any language.

< previous page page_264 next page >
< previous page page_265 next page >
Page 265
Assessment and Bilingual Children
The allocation of bilinguals to Special Education and the attribution of learning difficulties usually depends on
some form of assessment. It is essential for any psychological and educational assessment of bilingual children to
be fair, accurate and broad. Too often, tests given to bilingual children only serve to suggest their ‘disabilities’,
supposed ‘deficits’ or lack of proficiency in a second language. Assessment can too easily legitimize the disabling
of language minority students. Such students may come to be stigmatized by such tests, for example, because tests
locate apparent weaknesses in the majority language and use monolingual scores as points of comparison. English
language tests and IQ tests administered in the second language to US children are particular examples.
In the United States, there is legislation to govern appropriate processes of assessment of those for whom English
is an additional language. For example, the 1973 Californian case of Diana v. The California State Board of
Education was based on nine Mexican-American parents who protested that their children (who were dominant in
Spanish) were given an English language IQ test. The IQ test revealed ‘normal’ non-verbal IQ scores, but very low
verbal IQ scores (as low as 30 for one child). As a result of this linguistically and culturally inappropriate IQ test,
the Mexican-American children were placed in classes for the ‘mentally retarded’. This case never went to court,
but it established that testing should be conducted in a child’s native language (and in English), and that non-verbal
IQ tests were usually a fairer measurement of IQ than verbal tests (Valdés & Figueroa, 1994). As a result of this
case, the collection of broader data on language minority children was required (rather than simple test data) to
justify placement of such children in Special Education.
In 1975, Public Law 94142, the Education for All Handicapped Children Act, federally mandated that all testing
and assessment procedures should be non-discriminatory. ‘Non-discriminatory’ means using tests that are culturally
and linguistically appropriate. Such testing procedures were only to be used by trained members of a
multidisciplinary team. Apart from tests, teacher recommendations, observations of a child and other relevant
information should create a multi-source file of evidence (Barona & Barona, 1992).
Yet for all the advice in legislation, all the discussions in academic literature and in research, bilinguals tend in
many countries to be discriminated against in testing and assessment. Therefore, it is important to review the
components of such bias, and more constructively, to suggest desirable practice. Ten overlapping and interacting
issues in the assessment of bilingual children are now considered.
(1) The temporary difficulties faced by bilinguals must be distinguished from relatively more permanent
difficulties that impede everyday functioning and learning. Brief language delays, temporary adjustment problems
of in-migrants and short-term stammering (stuttering) are examples of transient difficulties (Baker, 1995).
Dyslexia, hearing loss and neuroticism are examples where longer term problems may need treatment. This simple
distinction hides different and complex dimensions. The means of distinguishing temporary and longer term
problems follows below.

< previous page page_265 next page >
< previous page page_266 next page >
Page 266
(2) Diagnosis needs to go beyond a few, simple tests and engage a wide diversity of measurement and observation
devices. Diagnosis needs to be extended over a time period and to avoid an instant conclusion and instantaneous
remedy. Observing the child in different contexts (and not just in the classroom) will provide a more valid profile
(language and behavior) of that child. The family and educational history of the child needs assembling. Parents
and teachers need consulting, sometimes doctors, counselors, speech therapists and social workers as well.
Samples of a child’s natural communication need gathering, with the child in different roles and different situations
(Hernandez, 1994).
An awareness or basic knowledge of a child’s ethnic, cultural and linguistic background is important in fair
assessment, but it may not be enough. To interpret test scores and classifications meaningfully and wisely, and in
particular, to make decisions on the basis of assessment, requires a highly sensitive and sympathetic understanding
of a child’s community, culture, family life and individual characteristics. Children come to school with very
differing home experiences, where different kinds of ability are cultivated and stressed. Such abilities prized by
parents may be different from the abilities learned at school (Chamberlain & Medinos-Landurand, 1991).
For example, discovery learning and learning through play is not a part of all cultures. An investigative,
questioning mode of thinking may not be encouraged within a culture. As reflected in parent-child relationships,
adults provide authoritative knowledge that must be accepted and enacted by the child. Parents may teach literacy
in a similar style, where the child is expected to memorize much or all of the holy book and repeat it without
comprehension. Such family and community socialization practices have implications (1) for the type of
assessment that reveals the strengths and weaknesses of the child, (2) for the importance of acquiring evidence
about the culture of the child, and (3) for the process of assessment by the teacher, psychologist, speech therapist,
counselor or other professional.
Test scores (e.g. on educational and psychometric tests) tend to be decontextualized and attenuated (Resnick &
Resnick,1992). An analogy helps to illustrate this important point. Test scores are like latitude and longitude. They
provide points of reference on a map of human characteristics. As a standard measurement usable on all maps,
they provide initial, rapid and instantly comparable information. But imagine the most beautiful place you know
(e.g. a flower-enfolded, azure-colored lake set amid tall, green-sloped, ice-capped mountains). Does the expression
of the latitude and longitude of that scene do justice to characterizing the personality and distinctiveness of that
location? The attempted precision of the sextant needs to be joined by the full empathic exploration and evaluation
of the character and qualities of the child.
Authentic language and behavior needs assessing rather than decomposed and decontextualized language skills.
The levels of thinking and understanding of a child through both languages need assessing rather than superficial
comprehension gleaned from a superficially ‘scientific’ test (e.g. with high test-retest reliability, and high
correlations with similar tests). ‘Inauthentic language’ tests relate too often to

< previous page page_266 next page >
< previous page page_267 next page >
Page 267
a transmission style curriculum where obtaining the correct answer and ‘teaching to the test’ dominates teacher
thinking and classroom activity. Such tests do not capture the communication abilities of the child, as for example,
occur in the playground, on the street, at the family meal-table and in internal conversations when alone.
(3) The choice of assessors for the child will affect the assessment. Whether the assessors are perceived to be from
the same language group as the child will affect the child’s performance (and possibly the diagnosis). The
perceived age, social class, powerfulness and gender of the assessor(s) will affect how the child responds and
possibly the assessment outcomes. The assessment process is not neutral. Who assesses, using what devices, under
what conditions, all contribute to the judgment being made. An inappropriate assessor creates an increased risk of
making two common, opposite judgmental errors concerning bilingual children: (1) generating a ‘false positive’,
that is, diagnosing a problem when none is present; (2) generating a ‘false negative’, that is, failing to locate a
problem when one exists.
(4) Children need assessing in their stronger language. Ideally children need assessing as bilinguals in both their
languages. The tests and assessment devices applied, and the language of communication used in assessment,
should ideally be in the child’s stronger language. An assessment based on tests of (and in) the child’s weaker
language may lead to a misdiagnosis, a false impression of the abilities of the child and a very partial and biased
picture of the child. This sometimes occurs in the United Kingdom and the United States where bilingual children
are tested in English, partly because of the availability of well regarded psychometric tests.
(5) Parents and educators need to make sure the language used in the test is appropriate to the child. For example,
a translation of the test (e.g. from English to Spanish) may produce inappropriate, stilted language. Also, the
different varieties of Spanish, for example, may not that be used by the student. Chicano Spanish speaking parents
will want the tests in Chicano or at least Mexican Spanish rather than the Cuban, Puerto Rican or Castilian variety
of Spanish. Once Spanish speaking children have been in the United States for a time, their Spanish changes.
English influences their way of speaking Spanish. So a test in ‘standard’ Spanish is inappropriate. A Spanish test
may accept only one right answer, penalizing children for their bilingualism and their United States Spanish. A
monolingual standard of Spanish is inappropriate to such bilingual children.
Since there are language problems with tests (as outlined above and continued below), it is important to distinguish
between a child’s language profile and performance profile (Cummins, 1984a). The performance profile is more
important as it attempts to portray a child’s underlying cognitive abilities rather than just language abilities. A
performance profile seeks to understand the overall potential of the child, not just their language proficiency.
Cummins (1984a) has shown how this distinction between a language profile and a performance profile on one
frequently used IQ test in individual assessmentthe Wechsler Intelligence Scale for Children: Revised (WISC-R).
Bilingual children tend to score significantly higher on the Performance than the Verbal subtests. As Cummins
(1984a: 30) suggests: ”The analysis of student performance on this test

< previous page page_267 next page >
< previous page page_268 next page >
Page 268
suggested that the majority of Verbal subtests were tapping ESL students’ knowledge of the English language and
North American culture in addition to, or instead of, verbal cognitive/academic ability’.
(6) There are times when a test or assessment device cannot be given in the child’s stronger language. For
example, appropriate bilingual professionals may not be available to join the assessment team, tests may not be
available in the child’s home language, and translations of tests may make those tests invalid and unreliable.
Interpreters are sometimes necessary. Interpreters do have a valuable function. If trained in the linguistic,
professional and rapport-making competences needed, they can make assessment more fair and accurate.
Interpreters can also bring a possible bias into the assessment (i.e. ‘heightening’ or ‘lowering’ the assessment results
through the interpretation they provide).
(7) There is a danger of focusing the assessment solely on the child. If the child is tested, the assumption is that the
‘problem’ lies within the child. At the same time, and sometimes instead, the focus needs to shift to causes outside
the child. Is the problem in the school? Is the school failing the child by denying abilities in the first language and
focusing on failures in the second (school) language? Is the school system denying a child’s culture and ethnic
character, thereby affecting the child’s academic success and self-esteem? Is the curriculum delivered at a level that
is beyond the child’s comprehension or is culturally strange to the child? The remedy may be in a change to the
school system and not to the child.
(8) The danger of assessment is that it will lead to the disablement rather than the empowerment of bilingual
children (Cummins, 1984a, 1986b). If the assessment separates children from powerful, dominant, mainstream
groups in society, the child may become disabled. The assessment may lead to categorization in an inferior group
of society and marginalization. Instead, assessment should work in the best, longterm interests of the child. Best
interests doesn’t only mean short-term educational remedies, but also long term employment and wealth sharing
opportunities. The assessment should initiate advocating for the child, and not against the child.
(9) It is important to use the understanding of a child’s teachers who have observed that child in a variety of
learning environments over time. What do the child’s teachers think is the root problem? What solutions and
interventions do teachers suggest? Have the child’s teachers a plan of action? A team of teachers, meeting regularly
to discuss children with problems, is a valuable first attempt to assess and treat the child. Such a team can also be
the school decision-maker for referral to other professionals (e.g. speech therapist, psychologist, counselor).
(10) Norm referenced tests are often used to assess the child (e.g. for entry into, or exit from a bilingual program).
This means that the assessor can compare the child with other so called ‘normal’ children. The assessor can indicate
how different the child is from the average. Many such tests are based on scores from ‘native’ language majority
children. Thus, comparisons can be unfair for bilingual children. For example, tests of English language
proficiency may have ‘norms’ (averages, and scores around the average) based on native speakers of English. This
makes them

< previous page page_268 next page >
< previous page page_269 next page >
Page 269
biased against bilinguals and leads to the stereotyping of particular language and ethnic groups.
The testing of bilinguals has developed from the practice of testing monolinguals. Bilinguals are not the simple
sum of two monolinguals but are a unique combination and integration of languages. The language configuration
of bilinguals means that, for example, a bilingual’s English language performance should not be compared with a
monolingual’s English language competence. A decathlete should not be compared with a 100 metre sprinter solely
for speed of running. Monolingual norms are simply inappropriate for bilinguals (Grosjean, 1985). One example
helps illustrate this point. Bilinguals use their languages in different contexts (domains). Thus they may have
linguistic competence in varying curriculum areas, on different curriculum topics and on different language
functions. Equal language facility in both languages is rare. Comparison on monolingual norms assumes such
equal language facility across all domains, language functions and curriculum areas. This is unfair and inequitable.
Such norm referenced tests are often written by white, middle class Anglo test producers. The test items often
reflect their language style and culture. For example, the words used such as ‘tennis racquet’, ‘snowman’ and ‘credit
cards’ may be unfamiliar to some in-migrants who have never seen a snowman, played or watched tennis, or know
about the culture of plastic money. Assessment items that reflect the unique learning experiences of language
minority children will be excluded from a test for majority language or mainstream children. Such items will be
highlighted by item analysis of an early draft of the test as ‘unconnected’ with the majority of the test items.
Such norm referenced tests are often ‘pencil and paper’ tests, sometimes involving multiple choice answers (one
answer is chosen from a set of given answers). Such tests do not measure all the different aspects of language, of
‘intelligence’ or any curriculum subject. Spoken, conversational language and financial intelligence, for example,
cannot be adequately measured by a simple pencil and paper test.
Some norm referenced tests report the results in percentiles (especially in the United States). Percentiles refer to
the percentage of children below (and above) the child being tested. For example, being in the 40th percentile
means that 39% of children score lower than the child being tested. Sixty percent of children of an age group score
above the 40th percentile child. The child is in the 40th group from the bottom, all children being assumed to be
divided into 100 equally sized groups. Percentiles are often used for entry and exit of students into bilingual
programs in the US. For example, a student may need to reach the 40th percentile on an English language test to
be exited to a mainstream class. What ‘percentile threshold’ (e.g. 40th percentile) is used for entry or exit is
essentially arbitrary, and liable to political rather than educationally derived decisions.
Norm referenced tests essentially compare one person against others. Is this important? Isn’t the more important
measurement about what a bilingual child can and cannot do in each curriculum subject? To say a child is in the
40th percentile in English doesn’t tell the parent or the teacher what are a child’s strengths and weaknesses,
capabilities and needs

< previous page page_269 next page >
< previous page page_270 next page >
Page 270
in English. The alternative is assessment related directly to progress in each curriculum area.
Curriculum based assessment is called criterion referenced testing, and seeks to establish the relative mastery of a
child in a curriculum area (Baker, 1994b). It seeks to establish what a child can do, and what is the next area of a
curriculum (e.g. reading) where progress can be made. A diagnosis may also be made, not necessarily of a
fundamental psychological or learning problem, but of weaknesses that need remedial treatment to enable
conceptual understanding (e.g. if a child has a problem in reading certain consonantal blends). Such criterion
referenced assessment of bilingual students gives parents and teachers more usable and important information. It
profiles for parents and teachers what a child can do in a subject (e.g. mathematics) and where development, or
accelerated learning, should occur next. Such assessment enables an individualized program to be set for the child.
However, the sequences of learning that underpin a criterion referenced test may still reflect a cultural mismatch
for the child. Criterion referenced assessment (usually, but not necessarily) assumes there is a relatively linear,
step-by-step progression in a curriculum area (e.g. in learning to read, in number skills, in science). Such a
progression may be culturally relative and culturally determined. As Cline (1993: 63) noted ‘because of their
different prior experiences, their learning hierarchy for a particular task may follow different steps or a different
sequence (e.g. in relation to phonology and orthography in learning to read)’. Therefore, it is important to use
curriculum assessment contexts and processes that are appropriate, comprehensible and meaningful to the bilingual
Three solutions to the problems of testing bilinguals are provided by Valdés & Figueroa (1994). Their first
solution is to attempt to minimize the potential harm of existing tests when given to bilingual individuals, by
applying some of the guidelines given above. Their second solution is, temporarily, to ban all testing of bilinguals
until more valid tests can be produced for bilingual populations. Their third alternative is that alternative
approaches to testing and development be developed. This third option may be the one most favored by many
teachers, educationists and parents. It would, for example, bring in bilingual norms, more curriculum based
assessment and portfolio type assessment, and a greater cultural and linguistic awareness of bilinguals.
However, this third solution is probably a new beginning but is not enough. A more radical solution places change
in assessment within (and not separate from) a change in expectations about the nature and behavior of bilingual
students. This entails a shift in the politics and policy dimensions of the assessment of bilinguals. Merely changing
tests may alleviate the symptoms of a problem, but not change the root cause. The root cause tends to be a bias
against language minorities that is endemic in many societies and is substantiated by tests. By being biased against
bilinguals in a cultural and linguistic form, and by a failure to incorporate an understanding of the cognitive
constitution of bilinguals, assessment confirms and perpetuates various discriminatory perceptions about language
minority children.

< previous page page_270 next page >
< previous page page_271 next page >
Page 271
Assessment thus sometimes serves by its nature and purpose, its form, use and outcomes, to provide the evidence
for discrimination and prejudice against language minorities (e.g. in the US and Britain) to be perpetuated.
Assessment results serve to marginalize and demotivate, to reveal under-achievement and lower performance in
language minority children. These results are often a reflection of the low status, inequitable treatment and poverty
found among many language minorities. Assessment provides the data to make expectations about language
minority students self-fulfilling and self-perpetuating. Changing assessment does not necessarily break into that
cycle, unless it is a component of a wider reform movement.
Assessment must not in itself be blamed for bias against bilinguals. It is a conveyer and not a root cause of
language minority discrimination and bias. Until there is authentic and genuine acceptance of cultural pluralism in
a nation, a more widespread assent for multiculturalism, a minimizing of racism and prejudice against ethnic
minorities, minor modifications in the assessment of bilinguals stand the chance of merely confirming the lower
status and perceived ‘deficiencies’ of language minority children.
Too often the focus in assessing bilinguals is on their language competence (or on their ‘limited proficiencies’ as in
the test-designated ‘Limited English Proficient’ students in the United States via tests such as the English Language
Assessment Battery (LAB)). Rarely does the assessment of bilinguals in the education system focus on other
important competences of a bilingual.
In the US, the 1991 Secretary’s Commission on Achieving Necessary Skills (SCANS) recommended that attention
be given in high schools to work-related competences. For example, performance standards were called for in:
interpersonal skills (e.g. working in a team, negotiating, working well with people from culturally diverse
backgrounds), interpreting and communicating information, thinking creatively, taking responsibility, sociability,
self-management and integrity. These important ‘life-skills’ are regarded as ‘essential accomplishments’ for
employment, self-respect and building a better world.
On some of these attributes, there is evidence to suggest that bilinguals may have advantages over monolinguals
(e.g. negotiating, working well with people from culturally diverse backgrounds, interpreting and communicating
information, thinking creatively). If the assessment of bilinguals focused more on these ‘outside world’ attributes
and less on classroom linguistic skills in the majority language, a more affirmative and favorable, productive and
constructive view of bilinguals might be promoted.
The chapter has attempted to indicate that under-achievement and placement in special education is not a cause of
bilingualism. The reasons for under-achievement tend to reside in factors that surround bilinguals (e.g. their social
and economic conditions) and are not an inevitable component of bilingualism. Traditional assessment in education
often serves to misplace and misdiagnose bilinguals. Apparent rather than real difficulties are identified, with a
depreciation of bilinguals’ language and potential curriculum performance.

< previous page page_271 next page >
< previous page page_272 next page >
Page 272
Suggested Further Reading
BAKER, C. 1995, Bilingual education and assessment. In B.M. JONES and P. GHUMAN (eds) Bilingualism,
Education and Identity. Cardiff: University of Wales Press.
CLINE, T. and FREDERICKSON, N. 1995, Progress in Curriculum Related Assessment with Bilingual Pupils.
Clevedon: Multilingual Matters.
CUMMINS, J. 1984, Bilingualism and Special Education: Issues in Assessment and Pedagogy. Clevedon:
Multilingual Matters.
GARCÍA, O. and BAKER, C. (eds) 1995, Policy and Practice in Bilingual Education: A Reader Extending the
Foundations. Clevedon: Multilingual Matters.
HAMAYAN, E.V. and DAMICO, J.S. (eds) 1991, Limiting Bias in the Assessment of Bilingual Students. Austin,
TX: Pro-ed.
HARRY, B. 1992, Cultural Diversity, Families and the Special Education System. New York: Teachers College
KEEL, P. (ed.) 1994, Assessment in the Multi-Ethnic Primary Classroom. Stoke on Trent (UK): Trentham Books
VALDÉS, G. and FIGUEROA, R.A. 1994, Bilingualism and Testing: A Special Case of Bias. Norwood, NJ: Ablex
Publishing Corporation.
Review and Study Questions
(1) Write short revision notes on the following:
(i) The seven explanations of under-achievement in bilinguals.
(ii) Language delay, stuttering and bilingual special education.
(iii) The assessment of bilingual children with special needs.
(2) What explanations are given for under-achievement of language minority groups? In what circumstances can
such under-achievement be reversed?
(3) Why are children with special needs sometimes wrongly allocated to special education?
(4) Describe the nature of Bilingual Special Education? For what types of students is such education particularly
(5) Describe the kinds of assessment procedures that are potentially unfair to bilingual children? What may be the
most fair and appropriate methods of assessment?
Study Activities
(1) Interview some parents or teachers with a bilingual student who has special needs (preferably gathering a
variety of different viewpoints). Define the exact nature of those special needs. Find out the history of that
student’s education. Ask the parents or teachers what their preferences are for the use of languages in school and if
they want their children to be bilingual as a result of schooling. Inquire about the value and use they see in
languages for special needs children.
(2) Locate two students who are considered to have ‘special needs’. Write a case study of each student, with
contrasts and comparisons, to show the history of assessment

< previous page page_272 next page >
< previous page page_273 next page >
Page 273
and school placement. Evaluate the decisions made, indicating particular issues of practical importance.
(3) Find out about assessment practices in schools in your neighborhood. What practices exist at district and
school level? What tests are used? How do assessment practices impact on bilingual students?
(4) Researching on two or three bilingual students, administer one or more assessment tests, tasks or other devices
as allocated by your instructor. Analyze the results and profile the students (a) purely in terms of the test results,
then (b) relating the test results to other qualitative information available on the students (e.g. interview the

< previous page page_273 next page >
< previous page page_274 next page >
Page 274
Chapter 14
Second Language Learning
Introduction: Reasons for Second Language Learning
Ideological Reasons
International Reasons
Individual Reasons
Ten Dimensions of Classroom Language Learning
The Structural Approach
The Functional Approach
The Interactional Approach
The Multidimensional Language Curriculum

< previous page page_274 next page >
< previous page page_275 next page >
Page 275
Reasons for Second Language Learning
While there is much in common in second language learning for majority and minority language children, it is
important initially to locate the differences. This can be achieved by examining the different reasons for second
language learning.
An analysis of language teaching and learning commences from the implicit or explicit aims, assumptions and
‘philosophy of practice’ of the educationalist. Beneath observable second language teaching will be varying aims in
second language teaching, different views of the value of second languages and foreign languages in the local and
wider society, as well as varying ideas about learning and teaching in classrooms. The various overlapping reasons
why children or adults learn a second or a third language may be clustered under three headings: ideological,
international and individual.
Ideological Reasons
Examining the societal and political reasons for second language learning highlights the difference between
language majority and language minority children. For language minority children, the aim of second language
learning may be assimilationist. For example, the teaching of English as a second language in the United States
and in England often aims at rapidly integrating minority language groups into mainstream society. Assimilationist
ideology tends to work for the dominance of the second language, even the repression of the home, minority
language. In contrast, when children learn a minority language as their second language, maintenance and
preservation of that minority language may be the societal aim. For example, when English speaking children are
taught in school language lessons to speak Irish in Ireland or Maori in New Zealand, the aim is to preserve and
strengthen the indigenous minority language. Such maintenance may not only exist in indigenous language
‘territory’. Where first language English children in the US learn to speak Spanish as a second language, there may
be an attempt to preserve the language community within a particular area. This provides an additive situation: a
second language is added at no cost to the first language.

< previous page page_275 next page >
< previous page page_276 next page >
Page 276
A different societal reason for second language learning other than assimilationist or preservationist, is to obtain
increased harmony between language groups through bilingualism. In Canada, French speaking children learning
English, and English speaking children learning French seek to produce a dual language, integrated Canadian
society. For example, the promotional video by the Canadian Official Languages Board entitled ‘One Country Two
Languages’ enthusiastically conveys the idea that integration into Canadian society is best achieved through
widespread bilingualism and biculturalism.
The assimilationist, preservationist, additive and harmony viewpoints all argue for the importance of a second
language for careers, access to higher education, access to the information explosion and for travel. However, it is
important to distinguish whether the second language is to replace the first language or add to that first language.
Rather than multiplying experience, second language teaching may be for divisive reasons, impoverishing the
language minority child.
While teachers may be relatively powerless to change the basic aims and reasons in second language teaching,
understanding the role they play in such teaching is important. Second language teaching does not exist in a
political vacuum. Nor is language teaching a neutral, value-free activity. Therefore, second language teachers need
to be aware of their goals at a conscious level.
International Reasons
Apart from the political and social reasons for language learning, there are international reasons given by second
language educationists for second language learning. Second and third language learning is often encouraged or
enforced for economic and trade reasons. Given notions such as common markets, open access to trade, the free
market economy, the importance of international trade to developing nations, then facility with languages is seen
as opening doors to economic activity. Selling cars and computers to the Japanese, for example, may be almost
impossible through English or German. Speaking Japanese and having a sympathetic understanding of Japanese
culture, manners, values and thinking may be the essential foundation for economic activity. Helmut Schmidt,
former Chancellor of the old Federal Republic of Germany once said: ‘If you wish to buy from us, you can talk any
language you like, for we shall try to understand you. If you want to sell to us, then you must speak our language’.
There is a growing realization that speaking foreign languages is important in increasingly competitive
international trade. The Californian State Department Education’s (1989) ‘Foreign Language Framework’ suggests
that, ‘On a pragmatic level, schools in California as well as in other states need to develop more individuals with
strong skills in a second language as a matter of long-range economic self-interest’ (p. 4). The Report notes that
two-thirds of translating jobs in the US Department of State are filled by foreign-born individuals because there
are so few US students and adults who are proficient in the second languages required for such posts.
Second and third language learning is also encouraged for its value in travel across continents. For many mainland
Europeans, for example, to speak two or three or four

< previous page page_276 next page >
< previous page page_277 next page >
Page 277
languages is not uncommon. Such language facility enables holidays to be spent in neighboring European countries
or in North, Central or South America. In the attempted unification of Europe, traveling across frontiers is
becoming more common, encouraging a person to acquire a repertoire of languages.
Third, languages provide access to information. In the information society of the twentieth century, access to
information is often access to power. Whether the information is in technical journals, on large computer
databases, on satellite television or in international faxing, a repertoire of languages gives wider access to social,
cultural, political, economic and educational information. For the business person and the bureaucrat, for the
scholar and the sports person, access to multilingual international information opens doors to new knowledge, new
skills and new understanding.
Individual Reasons
There are four reasons frequently given why the individual child or adult should learn a second or third language.
One reason is for cultural awareness. To break down national ethnic and language stereotypes, one motive in
second language learning has become intercultural sensitivity and awareness. The Californian State Department
Education’s (1989) Foreign Language Framework regards one second language learning aim as being the civic and
cultural benefits of foreign language teaching. Second language learning is important because ‘the power language
has to foster improved understanding between peoples of various cultural backgrounds. Culture is embedded in
language’ (p. 5). Increasing cultural sensitivity is seen as important as the world becomes more international, more
of a global village, with more sharing of experience.
Cultural awareness in the classroom may be achieved by discussing ethnic variations in eating and drinking, rituals
of birth, death and marriage, comparing the greetings of New Zealand Maori, Arabs and Jews, or comparing
shopping rituals in Malaysian markets, San Francisco superstores or all-purpose village stores in Venezuela. Such
activity widens human understanding and attempts to encourage sensitivity towards other cultures and creeds.
While cultural awareness may be conveyed in the first language, the inseparability of culture and language means
that such awareness may best be achieved through simultaneous language learning.
The second ‘individual’ reason for second language learning has traditionally been for cognitive training. The
learning of foreign languages has been for general educational and academic value. Just as history and geography,
physics and chemistry, mathematics and music have traditionally been taught to increase intellectual fitness and
stamina, so modern language learning has been defended as a way of sharpening the mind and developing the
intellect. Given the memorization, analysis (e.g. of grammar and sentence structure) and the need to negotiate in
communication, language learning has been regarded as a valuable academic activity particularly after elementary
The third reason for an individual to learn a language is for social, emotional and moral development, self-
awareness, self-confidence, and social and ethical values (van Ek, 1986). Such affective goals include the
possibility of incipient bilinguals being able to

< previous page page_277 next page >
< previous page page_278 next page >
Page 278
create more, and more effective relationships with target language speakers. Bilinguals can potentially build social
bridges with those who speak the second language. Self-confidence and enhanced self-esteem may result from
being able to operate socially or vocationally with those who speak the second language. The addition of a second
language skill can boost an individual’s self-concept as a learner, a liaisor and a linguist.
The fourth ‘individual’ reason for learning a language is for careers and employment. For language minority and
language majority children, being able to speak a second or third or fourth language may mean escaping
unemployment, opening up possibilities of a wider variety of careers or gaining promotion in a career. Potential
careers include becoming translators and interpreters, buying and selling goods and services, exchanging
information with local, regional, national and international organizations, migrating across national frontiers to find
work, gaining promotion in neighboring countries, becoming part of an international team or company, as well as
working from home or from the local village and using multilingual telecommunications to spread a product.
Ten Dimensions of Classroom Language Learning
We continue by examining some of the basic variations in second language learning and teaching strategies. In
second language lessons, language is the subject matter being taught and is also the means of transmitting the
subject matter. In second language classrooms, the second language is often also used for classroom management.
This makes second language teaching different from other curriculum areas where the language of control is
separate from the subject matter taught. History and Science are different from second language lessons because
the content is separate from the language of transmission and organization. That apart, what are the major
variations in second language learning?
There are considerable variations in the amount of time in different courses and syllabuses for second language
learning. There are short-course introductions to second languages that last only two weeks of a total of thirty
hours duration. In stark comparison, school second language learning from kindergarten to high school may be
scheduled for as long as 5,000 hours. Separate from the total amount of time in second language lessons is the
distribution or concentration of that time. There are ‘drip feed’ lessons of about half an hour per day or per week
over elementary schooling that result in a low degree of concentration. At the other end is Canadian immersion
education with its high concentration, intensive second language learning.
There are a variety of labels for different second language learning and teaching methods and approaches:
grammar-translation method, the direct method, the reading method, the Audiolingual approach, the audiovisual
approach, situational language teaching, the total physical response method, the silent way, community language
learning, the natural approach, Suggestopedia, the functional-notional approach, the information communicative
approach and the social communicative approach (Stern, 1983a; Richards & Rodgers, 1986; Cook, 1991). Analysis
of these approaches suggests that there are ten basic issues around which the different language teaching methods

< previous page page_278 next page >
< previous page page_279 next page >
Page 279
approaches are based. These ten overlapping and interlinked issues are first reviewed, and then, for
exemplification, three major variations in approach will be outlined.
(1) Explicitly, or more likely implicitly, a language teacher will have a theory of what constitutes a second
language and what is its purpose. For some teachers, language is essentially about vocabulary, correct grammar
and correct sentence structure. Such a view of language is of a rule governed structure or a linguistic system which
needs to be conveyed to the student. For other teachers, language concerns communication. Such a utilitarian and
sociological view of language highlights that language is for functional communication, for communicating ideas,
meaning and information. For some teachers, the implicit theory or purpose of language is ‘to share information’.
For others, language is ultimately about creating personal relationships, about transactions between people and the
successful negotiation of meaning between those conversing. For such teachers, language needs to be learnt for its
ability to foster social interaction.
(2) Teachers are also likely, implicitly rather than explicitly, to hold a theory of how children and adults best learn
a language. Is a second language best learnt through memorizing vocabulary, constantly practicing correct
grammar and sentence structure, and forming correct habits? Is language best taught by constant drills and
practice, constant correcting of mistakes to achieve as perfect secondary language fluency as possible? Or is
language best taught as a means rather than as an end? Should the focus be on meaning and not on language
forms? Should language practice be on meaningful tasks involving real communication to acquire the skills of
effective communication? Many teachers believe that language has to be learnt, particularly vocabulary and
grammar. Gaarder (1977) gives the opposite view. ‘Never try to teach language per se; rather, teach life (joy,
sorrow, work, play, relationships, concepts, differentiation, self-awareness of others, etc.) by involving the children
in situations and activities that are highly significant to them’, (p. 78)
(3) From an implicit theory of language and a theory of language learning, there should naturally follow classroom
procedures and processes, roles and relationships. Thus, some teachers will define second language classroom
goals as the accurate control of the four language abilities, of comprehension, (e.g. listening) before production
(e.g. speaking), of conscious second language learning and successful academic examination outcomes. For such
teachers, the important goal is mastery over the second language to achieve skills that approach those of the native
speaker. For other teachers, language for communication purposes is acquired subconsciously. The goal becomes
social communication or communicating information, equipping the child with functional skills to communicate in
an uninhibited and as intelligible way as possible (but not necessarily correctly).
(4) Goals in language learning have implications for a language syllabus. Some teachers will move through a
tightly controlled and carefully sequenced set of lessons. New vocabulary will be introduced slowly in a carefully
sequenced manner. The difficulty level of grammar and sentence structure will be slowly increased on a precise
ladder of progress. Other teachers will have a list of functions and tasks

< previous page page_279 next page >
< previous page page_280 next page >
Page 280
which are introduced to guide second language acquisition (e.g. greetings, taking leave, thanking, apologizing,
giving instructions, relaying intentions).
(5) The preferred syllabus will lead to classroom activities that may vary from drill and practice, repetition and
substitution, translation and memorization, to activities such as working in a small group on a set task (e.g. moving
around a street map from shop to shop), roleplay (for example, going into a cafe ordering coffee and being
introduced to a new friend) and pair work.
(6) Such classroom activities have direct consequences for the role taken by the teacher. When classroom activities
center around drill and practice, the teacher will usually control and orchestrate the lesson. The teacher drills,
corrects and tests progress. In this situation, the teacher is authoritatively working from the front of the class and
directing operations. When students are working in small groups or in pairs, the teacher may take on a more
informal role, being both the facilitator and a participant. The different cogs in the classroom machine are
separately oiled and their motion separately monitored.
(7) The role taken by the teacher will determine the role taken by the learner. When teachers are controlling,
students will be dependent and reactive. The student answers, imitates and hopefully internalizes. When the
teacher is the facilitator rather than the director, then students may be more proactive, negotiating with teachers
and their peers the route through the task. Being more independent may encourage more language improvisation
by students and flexibility in the direction and format of the lesson.
(8) The role taken by the teacher and the student will relate to the materials and facilities that exist in and outside
the classroom. A more structured, linguistic approach will tend to co-exist with graded text books, visual aids,
graded exercises, language laboratories and computer assisted language learning programs. Curriculum materials
will be tightly organized and carefully structured. The language laboratory or the computer assisted language
learning program will give experience in carefully graded drills and practice. This may result in contrived patterns
such as:
Teacher : Do you like the weather today?
Student : No, I do not like the weather today. It is raining today.
Teacher : What kind of weather do you like?
Student : The kind of weather I like is sunshine.
This is termed non-authentic language. It is correct, exact language. However, it is not usually how people speak
in a conversation. A more communicative approach will not be so concerned with perfectly correct language, but
more on whether the meaning has been conveyed to the listener. Attempting authentic communication will tend to
connect with the use of small group work and pairwork, improvised acting and dialogue in the classroom, rather
than to tightly controlled, graded lessons. Under this broad heading may be placed out-of-classroom planned
experiences (e.g. short term visits to the target language community, study abroad, student exchanges, pen-pal
correspondence via letter, video or cassette tape and electronic mail, inten-

< previous page page_280 next page >
< previous page page_281 next page >
Page 281
sive language learning weekend ‘camps’, cultural visits and local contacts with ‘native’ speakers of the second
(9) Different methods and different approaches will relate to different forms of assessment of the outcomes of
second language learning. For some language teachers, the outcomes should be academic examination success,
often accenting literacy as much as, or more than, oracy. For example, the ability to translate from one language to
another, spot deliberate mistakes, fill in missing words or phrases in a text, write a free story in the second
language will often be judged as the appropriate outcomes of structured language teaching. For other second
language teachers, assessment should be the child’s ability to understand authentic materials (e.g. reading shop and
roadsigns to travel around a town or city). For others, assessment is best achieved through an oral examination
where the child’s ability to convey and negotiate meaning with the examiner is the important outcome. In such
assessment, interlanguage (the language abilities of a child in-between first lessons and fluency) and code
switching (e.g. using first language words when negotiating meaning) are both allowable. For some, where
language learning is self-controlled, self-planned and self-responsible, then assessment can be self-evaluatory.
(10) Another variation tends to be the contexts of second language learning. The first issue under the context
heading is the place of culture in language teaching. It is possible to teach a second or foreign language with little
or no reference to the attendant culture. A student may become fluent in the Chinese or German language without
being at all fluent in Chinese or German culture. The language learnt becomes the language of the classroom and
no more. No affiliation, understanding, empathy or identification is felt with the language community.
Alternatively, the language and culture may be taught in an integrated way. The way of life of different Chinese
groups may be presented to, and experienced by the language learner. The second issue concerning context is the
ethos of the classroom. Sometimes language learning goes on in classrooms without bright and colorful wall
displays that create a suitable atmosphere for learning French or learning Spanish. When the same classroom
serves as a science laboratory, a history room, a social studies area and a second language classroom, language
learning may be more removed from a cultural context and a cultural ethos. Other second language classrooms
tend to surround the student with appropriate language and cultural artifacts: posters and pictures, videos and
computer assisted language learning programs, objects (e.g. costumes and food packets) and tourist relics. The
student is provided with an appropriate atmosphere for learning the target language. When second language
learning occurs in the Mosque, Temple, Synagogue, Church Hall or Language Community Center, the context may
engender a feeling of becoming and belonging.
Having considered ten basic issues around which language teaching and language learning vary, we will now
consider three major, broadly different approaches. The Structural Approach is an umbrella title under which
comes the Grammar-Translation method, the traditional North American Audiolingual approach and the Direct
Method. Second, the Functional Approach, sometimes called the Notional-Functional approach,

< previous page page_281 next page >
< previous page page_282 next page >
Page 282
will be considered. This particularly has been the method encouraged by the Council of Europe in its ‘Language
and Teaching Modern Languages for Communication’ project and is sometimes termed the Information
Communicative Approach. The third method will be termed the Interactional Approach and is sometimes labeled
the Social Communicative approach.
The Structural Approach
The structural approach asserts that there is a language system to be mastered. The historically dominant method
of second language acquisition in school has been the grammar-translation approach. Students were expected to
memorize vocabulary lessons, learn verb declensions, learn rules of grammar and their exceptions, take dictation
and translate written passages. While tests and examinations were passed and paper qualifications issued, students
mostly did not become functionally bilingual. Second language learning started and stayed in school.
A new approach to language teaching, called audiolingualism developed, much allied to behaviorist psychology.
According to audiolingualism, second language learning is possible by the acquisition of a distinct set of speech
habits. A habit is the ability to make a sound or say a word or use grammar correctly in an automatic, unconscious
fashion. In behaviorist terms, language learning comprises the linking of a particular response to a particular
stimulus. The teacher provides specific, well defined stimuli. The learner responds and is reinforced or corrected.
Through repetition and drills it is hoped that the student will be able to use a second language automatically.
In the Audiolingual approach, the system to be memorized is structure-based dialogues. Language learning
becomes the learning of structures, sounds and words to attempt to achieve native speaker-like pronunciation. Such
a structural view sees language as a linguistic system, with the aim being to instill linguistic competence in the
student. For the grammar translation method, the accent is on literacy rather than oracy skills. The opposite is the
case for the Audiolingual approach where listening and speaking are given priority. Because language teaching is
based on linguistic structure, lessons should be carefully sequenced and graded. The sequencing is to be
determined wholly by linguistic complexity (e.g. moving from simple grammar rules to more complex rules with
their exceptions; practicing simple speaking patterns followed by ever increasing width of vocabulary and
difficulty in sentence structure).
In the Audiolingual method, the focus is constantly on correct language. To achieve this, there is repetition of
vocabulary phrases and sentences, imitative mimicry and memorization of short dialogues. The aim is for the
learner to avoid making linguistic mistakes in sentence construction and gain an automatic, accurate control of
basic sentence structures. Mastery of basic vocabulary and structures is required, by overlearning if necessary.
Part of the theory of audiolingualism is that the learner’s first language interferes with the acquisition of the second
language. Such interference occurs because the learner is

< previous page page_282 next page >
< previous page page_283 next page >
Page 283
ensnared by the habits of the first language. Therefore, teachers are encouraged to focus on areas of difficulty
posed by negative transfer from the first to the second language (e.g. Lado, 1964). A procedure called Contrastive
Analysis was developed to show the areas of difficulty in transfer from first language to second language. Such
analysis resulted in a list of features of the second language, different from the first language, which posed
potential problems for the teacher.
The origins of audiolingualism are found in the Army method of American Language Programs during World War
II. Just as drill and practice are the essentials in basic training on the parade square, so drilling is the central
technique in a structural approach. In the late 1960s and 1970s, the assumptions of the audiolingual method were
shown to be doubtful. When the Contrastive Analysis hypothesis was researched (e.g. Dulay & Burt, 1973, 1974),
grammatical errors were found that could not be explained by negative transfer from the first to the second
language. The Chomskyian revolution in linguistics also cast doubts on the behaviorist view of second language
learning. Chomsky (1965) suggested that human beings have an inbuilt cognitive readiness for language. Instead
of language being a series of surface patterns and habits, Chomsky emphasized the abstract, mentalist and
universal nature of rules that underlie an individual’s language competence. A child is endowed with a language-
acquisition device that comprises innate knowledge of grammatical principles.
Purity in second language is demanded in the structural approach. Therefore, students are not allowed to use their
first language in conveying meaning. The model for correct language is provided by the teacher, or via prepared
tapes in the language laboratory or more recently through computer assisted language learning programs. While
such structural viewpoints have tended to be criticized by language learning researchers and advisors, such a
formal, teacher orchestrated style is to be found as a major component in many teachers’ classroom repertoire.
The Functional Approach
In the 1970s, there developed a major alternative approach to language teaching. Instead of regarding language as
a linguistic system to be conveyed to the student, an alternative viewpoint was that language is essentially about
conveying meaning. The focus since the 1970s has tended to shift partly away from teaching the formal nature of a
language to socially appropriate forms of communication. For example, the sociolinguist Hymes (1972a) argued
that language was essentially about communication, essentially about being able to use a language for a purpose.
Language is a means rather than a structural end. Effective language does not mean grammatical accuracy nor
articulate fluency, but the competence to communicate meaning effectively.
Criticisms of the audiolingual method have led to a current emphasis on a more communicative approach in
second language learning. What is the language basis of the communicative approach?

< previous page page_283 next page >
< previous page page_284 next page >
Page 284
Second language learning is controlled by the learner rather than by the teacher. Whereas in audiolingualism the
teacher was regarded as crucial, the recent focus has changed to a selection and organization of second language
input partly by the learner.
Human beings possess an innate capacity and natural propensity for learning a second language. Whereas in
audiolingualism the first language was thought to interfere with the learning of a second language, current thinking
is that a second language can be acquired in a similar way to the first language.
It is not necessary to attend to the form of language to acquire a second language. The systematic learning of
grammar becomes of much less importance. Since first language learning occurs successfully without formal
grammar teaching, considerably less emphasis could be placed on grammatical form, more on informal acquisition.
Language errors and ‘interference’ or transfer are a natural part of the learning process. Children spontaneously
produce unique utterances and do not just imitate adult language as behaviorists would claim. For example, a child
who says ‘all finish food’ (I’ve completely finished my food) reflects the use of some underlying implicit rules. The
early incorrect production of sentences does not prevent a child from going on to produce correct, well formed
sentences later. Corder (1967) suggested that second language learners need to discover their own errors rather
than to be continuously corrected by the teacher. Therefore, the problem of interference becomes a problem of
ignorance. The avoidance of interference between languages cannot be taught in the classroom; rather the
avoidance of interference comes via observation of one’s own language. Language development can thus be seen
as the process of implicit rule formation rather than explicit habit formation.
Van Ek (1986, 1987) outlines six different forms of language competence to be acquired for communication
purposes: linguistic, sociolinguistic, discourse, strategic social-cultural and social competence. This essentially is a
belief that there is something more than linguistic competence as fostered by the structural approach that needs to
be acquired by the student.
There exists sociolinguistic competence in a language that concerns the ability to communicate accurately in
different contexts, with different people, and when there are different intentions in the communication.
Sociolinguistic competence is the awareness of the language form required in different situations. Such
competence arises when natural interaction occurs between students, without prompting by the teacher. Discourse
competence is the ability to use appropriate strategies in constructing and interpreting different texts, the ability to
contribute to the construction of a spoken discourse in communication. For example, discourse competence is
found in the ability to use pronouns, transition words and insert progression into communication.
Strategic competence is the ability to use verbal and nonverbal communication strategies to compensate for gaps in
the language user’s knowledge. Strategic competence concerns the ability to use body language (gesturing, head
nods, eye contact) to give meaning when verbal language is not at a level of competence to convey meaning.
Sociocultural competence is the ‘awareness of the sociocultural context in which the

< previous page page_284 next page >
< previous page page_285 next page >
Page 285
language concerned is used by native speakers and the ways in which this context affects the choice and the
communicative effect of particular language forms’ (van Ek, 1987: 8). Finally, social competence is the ability to
use particular social strategies to achieve communicative goals. For example, social competence includes the
ability to take the initiative in a conversation, to know when to interrupt someone else speaking without being ill-
mannered. Social competence involves understanding the social conventions that govern communication within a
Based on the work of Wilkins (1976), Widdowson (1978), the Council of Europe in van Ek (1986, 1987) and in
Girard (1988), the notional-functional syllabus developed for second language teaching. The notional categories to
be taught include time, quantity, space, motion, sequence and location, as well as the communicative functions
such as persuasion, inquiry, relaying emotions and establishing personal relationships. A Council of Europe
initiative resulted in a list of topic areas to be covered against a list of language functions (van Ek, 1987: 17). The
21 different topic areas can be considered in combination with language functions (e.g. forms of greeting and
taking leave need considering in the different contexts of: home, school, extended family, work, leisure, at a
supermarket check-out, when happy, angry, sad and when meeting someone for the first time):
Topic Areas Language Functions
Being polite and
Family and daily routine Attracting attention
School Introducing someone
Work Expressing good wishes
Leisure Thanking
Holiday and travel Apologizing
Environment, places and
Agreeing and
Food and drink Refusing and accepting
Goods and services
Approving and
Accidents and emergencies
Coping with language
Interesting events in the past,
present and future
Asking for information
Stating facts
Clothes and fashion Opinions and feelings
People Like and dislikes
Personal belongings, pets and
Needs, requests and
Immediate plans
Instructions and
Times and dates Intentions
The weather Permission
Emotional state (e.g. happy,
bored) Inviting

< previous page page_285 next page >
< previous page page_286 next page >
Page 286
Topic Areas Language Functions
Physical state (hungry, ill)
An example would be learning forms of greeting and taking leave. Appropriate vocabulary and examples of simple
structures is introduced by the teacher, or video or Computer Assisted Language Learning. For instance, in French,
vocabulary introduced might include: bonjour, bonsoir, salut, au revoir, à demain, bonne fête, bon voyage, bon
anniversaire, bonnes vacances, bon appétit, bonne année, Joyeux Noël, Madame, Mademoiselle and excusez-moi.
Cultural variations are discussed in the classroom (e.g. shaking hands, signs of peace, conventions about kissing,
use of body language such as distance from the other person, eye contact, smiles, head nods and posture). Different
situations will also be explored for necessary variations (e.g. greeting older compared with younger people,
different conventions in shops, with peers compared with their parents). After pair or group work to practice
greetings and ‘goodbyes’, appropriate linguistic and social-cultural behavior may have been learnt (Girard, 1988;
Sheils, 1988).
A functional view of language is language for real life activity. The language acquired in language lessons is for
use in the second language community or for travel abroad. Such language learning is based on an analysis of what
a student needs to communicate with the second language community. Thus, the emphasis is less on correct
grammar and perfect sentence structuring, more on the ability to communicate meaningful information. An
awareness of the context (e.g. supermarket, booking an air ticket) and of the comprehension level of the listener is
regarded as important. The sequencing of the lessons is determined by a careful consideration of an order of
difficulty in content and function rather than of the linguistic attributes of the second language. Using task based
materials (e.g. finding the way to the supermarket, obtaining a reservation on a plane), language is learnt through
trial and error, with accuracy judged in terms of context and meaningfulness rather than purely in linguistic terms.
Using small group work, pair work and task based, authentic individual activity, students are encouraged to gain
confidence in using the second language even in the most rudimentary form, to communicate information
successfully. Thus a second language may be learnt in an unconscious, informal way rather than explicitly and
The Interactional Approach
From the beginning of the 1970s, a slow shift towards teaching methods that emphasized communication has been
visible. The functional approach tended to redefine what the student had to learn to be competent in
communication rather than competent in linguistic structures. Knowing how to use language appropriately in
particular situations has been emphasized more than knowledge of grammatical rules only in second language
The accent recently has been slightly away from language functions and notions to communication for its social,
transactional, personal relationship goals. In the functional

< previous page page_286 next page >
< previous page page_287 next page >
Page 287
approach to second language teaching, it is possible that the teacher is controlling and the students responding.
Learning how to ask directions or how to greet someone can still be taught in a formal, drill and repetition style of
teaching. A task such as finding the way to the cafe and ordering a snack can still result in exercises that make the
student substitute words, fill in blanks in sentences and write down short phrases. While the activities of cafe
visiting and ticket booking may be authentic and meaningful, the language being used may be artificial, and
abstracted from the real social interaction of authentic dialogue.
In the interactional view of second language learning, the communicative approach requires a classroom where
real interpersonal communication regularly takes place and where social and personal transactions between people
actively occur. Moving away from a teacher based, teacher controlled, responsive, information communicative
style, the interactional view emphasizes maximal active social communication in the classroom, between students
in pairs and in groups. The teacher becomes a facilitator rather than a drill sergeant. In the interactional approach
to second language teaching, the following type of exercise may be given.
A pair of students work together. One person has a simple map of Barcelona, the other person has a list of
supermarkets with street names. Each only has partial information. Working together as best they can in the
second language, they ask each other questions to find a route to particular supermarkets. The students are given
basic vocabulary on the map with a sample question-and-answer sequence. The pair have to improvise a dialogue
to solve their communicative tasks. Another example is when students engage in roleplay. One student has to
engage in a telephone conversation booking a ticket for herself and her son on the plane from Barcelona to New
York, Tuesday morning next week. The conversation includes problems of diet and complications about the cost.
A cassette or video may be collaboratively created.
In these two examples, the essential nature of the tasks is social communication between people. Inter-
relationships and information combine rather than the processing of information being the sole focus. In such an
interactional approach to second language teaching, language is about methods of talking to people. Such a method
aims to give students both the competence and the confidence to engage in real conversations with people. Errors
in speech become relatively unimportant compared with finding strategies for successful communication. While
the teacher may provide some feedback and correction, the essential element is to encourage students to make
attempts at social communication, however grammatically flawed and limited in vocabulary. Such an approach
rests on intrinsic motivation among students and a willingness to forego authoritative control by teachers. Some
teachers and students find such roles difficult to accept. For some, such roles are culturally inappropriate. The
expectation is teaching by formal, traditional methods of second language teaching. The teacher is transmitting, the
student is receiving. The student centered rather than teacher centered approach can seem very foreign to foreign
language teaching.

< previous page page_287 next page >
< previous page page_288 next page >
Page 288
The Multidimensional Language Curriculum
Most language classrooms and language teachers tend to be eclectic in approach, combining in different ways and
in a different balance the structural, functional and interactional methods. In real classrooms, there is often a
combination of drills and pairwork, task based activity and teacher directed transmission, correct habit formation
and improvised dialogue between students with interlanguage allowed. At its worst, eclecticism is unselective,
directionless, with constantly shifting ground. At its best, eclecticism may lead to rational integration of
‘classroom-wise’ approaches.
An attempt at integrating the different aims and goals of a second or foreign language approach stems particularly
from the Boston Paper on Curriculum Materials (Lange, 1980). A group of second language learning experts
proposed the idea of a multidimensional curriculum. This would comprise four types of syllabuses: (1) Linguistic,
(2) Cultural, (3) Communicative, and (4) General Language Education. The Boston Paper argued that current
syllabuses were too ambitious in their linguistic features. Too much time was spent in language classrooms in
learning word lists and grammar. Too little time was spent on generating communicative competence in children.
The Boston Paper suggested that a linguistic syllabus was important but it should be joined by a cultural syllabus
that would accent the content of modern languages rather than just the form of the language. Children require an
understanding of the cultural attributes of the language being learnt and this should be thoroughly integrated with
the linguistic elements. Such a cultural syllabus needs to be assessed and examined in order to raise its status.
The Boston Paper (Lange, 1980) argued that a multidimensional curriculum would also involve a communicative
syllabus. Here the focus would be on efficient communication rather than on accurate language. Language errors
would be tolerated so that the accent could be on effective transmission of meaning. Authentic language needed to
be encouraged. While simulation and role playing may lead to a limited degree of communicative ability, contact
with foreign language speakers, preferably in natural situations, was much to be desired. Encouragement should be
given to students spending time in the homes and communities of second language speakers. Alongside immersion
experience in the second language community, language camps, festivals, visits to language communities and the
use of native speakers within the classroom is encouraged.
The final part of the Boston Paper (Lange, 1980) suggested that a multidimensional language curriculum include a
general language education syllabus. In the syllabus, students would come to understand more about the nature and
functions of language. Topics such as learning how to learn a language, language varieties, and language and
culture would be considered.
The theoretical basis of a multidimensional foreign language curriculum was subsequently expanded on and
refined by Stern (1983b, 1992), by trial materials (Tremblay et al., 1989) and was furthered by the Canadian
National Core French Study (LeBlanc, 1990). A national study in Canada has suggested major developments in the
teaching and learning of French as a second language in mainstream, core programs.

< previous page page_288 next page >
< previous page page_289 next page >
Page 289
Canadian Core French programs have hitherto taught French for between 20 and 50 minutes a day. Sometimes
starting at Grade 1, often at Grade 4, or later at Grade 7, a child may receive (e.g. in Ontario) around 1200 hours of
instruction in French, with a target active vocabulary of 3000 to 5000 words and around 100 basic sentence
patterns (Lapkin, Harley & Taylor, 1993). The suggested development of this Core program is through four
integrated syllabuses:
A language teaching syllabus which is structural and functional in approach. Both the form, the functions and the
context of language learning are addressed. Grammatical ‘consciousness raising’ and an analytical approach to
language learning is to be partnered by an experiential approach. Thus, communicative competence in a second
language needs to be integrated with grammatical knowledge.
A communicative activities/experiential syllabus is designed to give the second language learner authentic
experiences in communication and usable communication skills. Curriculum themes are to be motivating,
interesting, relevant, and to enrich experience as well as give communicative competence.
A cultural syllabus presents the French culture as something to be observed and analyzed as well as experienced. It
aims to develop sociocultural awareness and integrate French language learning with French culture. Focusing on
contemporary culture, the syllabus is principally concerned with enabling students to communicate effectively first
and foremost with French-Canadians, but with other francophones as well. Stern’s (1992) vision extends the
cultural syllabus to increased contact with the target language community, if possible. While cultural competence
is essential for successful communicative skills, the cultural syllabus underlines the essential relationship between
language and culture. The one stimulates and enriches the other in second language learning. The learning of a
language gathers meaning and purpose when there is simultaneous enculturation. For example, French lessons may
cover different areas of cultural experience: everyday activities, (e.g. food and drink), personal and social life (e.g.
relationships with friends and family), the world around (e.g. home region compared with region abroad), the
world of education, training and work (e.g. tourism), the world of communications (e.g. international electronic
mail and satellite TV), the international world (e.g. travel and holidays) and the world of imagination and
creativity (e.g. the making of a video or magazine).
A general language education syllabus explores languages and communication as intrinsically important topics.
This aims to increase students’ awareness of language and of the process of language learning via a three pronged
attack: language awareness, cultural awareness and strategic awareness. Topics might include: dialects and
registers, languages across the world, the origins of languages, the language development of children, language
prejudice, and correspondence (including across distance by electronic mail).
A mathematically imaginative answer is given to the question of how much time to allot to these four strands. ‘An
ideal distribution of time in the core French program might be: 75% communicative/experiential, 40% linguistic
syllabus, 15% general language, and 25% culture. While this may seem to add up to 155%, in reality it adds up to
an effective

< previous page page_289 next page >
< previous page page_290 next page >
Page 290
and efficient core French program. Providing that French is the language of communication in the classroom, and
providing that teachers and students focus on learning a living second-language for use in real life, each syllabus
will compliment the others and be realized simultaneously. That is, integration will occur almost naturally and the
time allocations will truly overlap’ (LeBlanc & Courtel, 1990:91). The integration of the four syllabuses is crucial.
Limited time for second language learning, and too demanding a schedule within available time makes integration
a challenge for language educators.
This chapter has considered the route to bilingualism via second language learning in school. While the route has a
long history in education, recent movements away from a structural to a communicative approach, highlight
classroom variations in aim, approach, strategy and style. Such variations indicate different kinds of bilingualism
that may be fostered in school. One variation is linguistic bilingualism, another is functional bilingualism, another,
bilingualism dedicated to human interaction. Variations in approach have recently been integrated into the concept
of a multidimensional language curriculum. Culture and communication, linguistics and general language
education may be combined in an attempt to harmonize varied aims and approaches, strategies and styles.
Suggested Further Reading
CALIFORNIAN STATE DEPARTMENT OF EDUCATION, 1989. Foreign Language Framework. Sacramento,
CA: California State Department of Education.
RICHARDS, J.C. and ROGERS, T.S. 1986, Approaches and Methods in Language Teaching. Cambridge:
Cambridge University Press.
STERN, H.H. 1992, Issues and Options in Language Teaching. Oxford: Oxford University Press.
Review and Study Questions
(1) Write short revision notes on the following:
(i) Ideological, international and individual reasons for second language learning.
(ii) The multidimensional language curriculum.
(2) What are the important dimensions of second language learning? How do approaches to second language
learning differ in their aims and differ in their classroom activities?
(3) With a form of language learning with which you are most familiar, use the 10 dimensions of classroom
language learning to expand on a particular style of teaching and learning.
Study Activities
(1) In a small group of students, discuss different experiences of learning a second language at school. Use the 10
dimensions of classroom language learning to record the differences between the group. From the experiences of
the group, which of these

< previous page page_290 next page >
< previous page page_291 next page >
Page 291
dimensions seem most important? Which approaches in the classroom seem from the personal experiences of the
group to be most effective?
(2) Visit a school where students are learning a foreign language. By interviewing the teachers and observing
classroom sessions, describe in terms of the 10 dimensions, the approach or approaches being used. Ask the
teachers and the students their purposes in learning a foreign language. If there are differences in aim between
teachers and students, examine whether you think these can be made compatible or are in conflict.
(3) Observe a classroom in which a foreign language is being taught to the majority, and one in which a second
language is being taught to a minority. Describe and explain the differences in approach.
(4) Imagine you are in a classroom 20 years hence. In that futuristic classroom, describe how language learning
might be taking place. What kind of technology might be in use? Will there be more or less emphasis on learning
minority and majority languages? What motivations might the students have in the futuristic classroom? For what
purposes are languages being learnt? What forms of assessment are being used?

< previous page page_291 next page >
< previous page page_292 next page >
Page 292
Chapter 15
Literacy in a Multicultural Society
The Uses of Literacy in Bilingual and
Multicultural Societies
Definitions of Literacy
Approaches to Literacy
The Skills Approach
The Whole Language Approach
The Construction of Meaning Approach
The Socio-Cultural Literacy Approach
The Critical Literacy Approach

< previous page page_292 next page >
< previous page page_293 next page >
Page 293
In education systems throughout the world, reading and writing are usually regarded as central in the curriculum.
In developed and developing countries, literacy is often associated with progress, civilization, social mobility and
economic advancement. A classic, if disputed claim, is by Anderson (1966) who suggests that any society requires
a 40% literacy rate for economic ‘take off’, a belief which is embedded in many literacy programs. But, as this
chapter will show, what precisely is meant by literacy is neither simple nor uncontroversial.
In today’s highly literate or less literate multilingual and multicultural societies, the ability to read and write is
often regarded as essential for personal survival, security and status. Literacy impacts on people’s daily lives in
innumerable ways. Where language minority members are relatively powerless and underprivileged, literacy is
often regarded as a major key to self advancement as well as empowerment. If this is so, it is important to consider
the needs or uses for literacy in students and adults in bilingual and multicultural societies.
In considering such needs and uses, it is important to note that literacy education for language minority students is
often in a majority language (e.g. English language literacy in the United Kingdom and the United States; English
in parts of Africa as an ‘official’ or international language). However, where language minorities have access to
bilingual education, literacy may be introduced in the home/minority language. In that case, literacy in the majority
language (and hence biliteracy) will be developed later in the elementary school. This issue is considered in the
next chapter. For the moment, the consideration of uses of literacy refers to literacy introduced in either the
minority or majority language (or both languages developed in parallel).

< previous page page_293 next page >
< previous page page_294 next page >
Page 294
The Uses of Literacy in Bilingual and Multicultural Societies
Literacy is Needed for Survival
There are basic day to day uses of literacy. For example, a motorist needs to read road signs, a shopper needs to
read instructions and labels on food packaging, the traveler needs to understand signs at bus stations, train stations
and airports.
Literacy is Needed for Learning
Students need to be able to read text books, story books, laboratory manuals, examination sheets and tests for
learning to take place.
Literacy is Needed for Citizenship and Political Empowerment
To complete bureaucratic forms, participate in local or regional government, read local and national newspapers,
read information that comes through the post or is on billboards, citizens need basic reading skills. To respond,
react, complain and assert democratic rights, reading and writing are often valuable allies to speaking skills.
Literacy is Needed for Personal Relationships
Writing letters to friends and relatives who are no longer local, and sending seasonal greeting cards, are each aided
by literacy.
Literacy is Needed for Personal Pleasure and Creativity
Being able to read gives students and adults access to magazines, novels, comics, story books and non-fiction to be
enjoyed in leisure time. Literacy enables the imagination to be stimulated, and creative instincts to be expressed.
Literacy is Needed for Employment
Many higher status and higher paid jobs require reading and writing skills.
Literacy is Needed to Empower the Mind
Wells & Chang-Wells (1992) suggest that literate thinking is ‘the building up, metaphorically speaking, of a set of
mental muscles that enable one effectively to tackle intellectual tasks that would otherwise be beyond one’s powers’
(p. 122). Literacy is seen as a mode of thinking, as a means of reasoning, reflecting and interacting with oneself.
This links with the idea of the empowerment of individuals, and having a public voice.
The importance and usefulness of literacy for language minorities is revealed in public perceptions of those who
are illiterate. Being unable to read and write is often regarded as shameful, embarrassing, a symbol of low or
marginal status, and in need of remediation in school or in adult classes. Reducing illiteracy is regarded as a key
priority in UNESCO’s aims, irrespective of country, continent, culture or caste.
Whether language minority children should first become literate in the majority language or in their minority
language will be discussed later in this chapter. Also considered later will be whether majority language
monoliteracy or biliteracy should be attempted. Before engaging such discussions, it is important to explore the
kind of literacy

< previous page page_294 next page >
< previous page page_295 next page >
Page 295
that language minority students and adults require. We start by considering three contrasting definitions of literacy.
This provides an instant flavor of the keen debate about the nature and value of different kinds of literacy. The
debate has important implications for bilingual students, as will be revealed.
Definitions of Literacy
The list of needs for literacy given above has already hinted that literacy plays different roles for different people.
Compare two definitions of literacy. The first is a functional definition of, and skills approach to, literacy by
UNESCO dating from 1962.
A literate person is someone who ‘has acquired the essential knowledge and skills which enable him to engage in
all those activities in which literacy is required for effective functioning in his group and community and whose
attainments in reading, writing and arithmetic make it possible for him to continue to use these skills towards his
own and the community’s developments’ (cited in Oxenham, 1980: 87).
In comparison, Hudelson (1994: 130) defines reading as ‘a language process in which an individual constructs
meaning through a transaction with written text that has been created by symbols that represent language. The
transaction involves the reader’s acting upon or interpreting the text, and the interpretation is influenced by the
reader’s past experiences, language background, and cultural framework, as well as the reader’s purpose for
reading’. This is a definition that concentrates on reading as the construction of meaning, rather than on the skills of
reading. While this definition only refers to reading, the idea extends to literacy in general.
The third definition derives from Wells & Chang-Wells (1992) who argue that ‘To be literate is to have the
disposition to engage appropriately with texts of different types in order to empower action, thinking, and feeling
in the context of purposeful social activity’ (p. 147). This definition deliberately allows for the possibility that
different language minority communities attach a different value to different types of literacy.
Cultures differ from each other in their uses and purposes for literacy. For some cultures literacy is about
promoting abstract thought, rationality, critical thinking, balanced and detached awareness, empathy and
sensitivity. For other cultures literacy is about memorization, transmission of life stories revealing their heritage,
values and morality, and centrally in certain religions, for the transmission of rules of religious and moral behavior.
A Moslem for example, will be expected to read aloud from the Qu’ran. Many Moslems are not expected to
understand what they read, this being provided in their mother tongue. In some cultures, the mother is expected to
read to her children and help them develop literacy skills, but is not expected to read national newspapers or
complete bureaucratic forms herself. The concept of literacy is therefore not single but plural, relative to a culture
and not universal.
These three definitions (skills, the construction of meaning, and sociocultural) will now be extended and explored
by examining different educational approaches to literacy. These approaches, like the above definitions, are not
necessarily discrete but overlap and

< previous page page_295 next page >
< previous page page_296 next page >
Page 296
can be combined in literacy learning. However, their emphases are different in ways that they impact on students.
Approaches to Literacy
The Skills Approach
The concept of functional literacy, as in the UNESCO definition, appears to assume that literacy is the simple act
of reading and writing. ‘Acquiring essential knowledge and skills’ refers to the ability to decode symbols on a page
into sounds, followed by making meaning from those sounds. Reading is about saying the words on the page and
writing about being able to spell correctly, and write in correct grammatical sentences. In both reading and writing,
a literate person is able to understand and comprehend the printed word.
The assumption of this approach is that literacy is primarily a technical skill, neutral in its aims and universal
across languages. The skills of rea