Introduction To Language Development: July 2013

See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/234038955
Introduction to Language Development
Book · July 2013
CITATIONS READS
87 19,656
1 author:
Shelia Kennison
Oklahoma State University - Stillwater
572 PUBLICATIONS 2,091 CITATIONS
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
Parents and Children View project
Tolerance View project
All content following this page was uploaded by Shelia Kennison on 29 March 2015.
The user has requested enhancement of the downloaded file.

C H A P T E R 7
LIFE WITH MORE THAN

ONE LANGUAGE
BILINGUAL CHILDREN LEARNING SECOND LANGUAGES

How Do Bilingual Children Differ From What Teaching Methods Are Used in Language
Monolingual Children? Classes?
Language Milestones Do All Countries Have Bilingual Education?
Perceptual Abilities How Does First Language Influence Second
Vocabulary Size Language Acquisition?
Language Mixing Text Box 7.2. Extraordinary Individuals:
Code-Switching Emil Krebs: An Extraordinary Polyglot
Text Box 7.1. Diversity of Human Languages: LIFE WITH TWO OR MORE LANGUAGES
Pidgins and Creoles What Factors Predict Proficiency?
Are There Cognitive Benefits When Children Text Box 7.3. Research Discovery: Bilingual
Are Bilingual? Problem Solving
Attentional Processing How Are Multiple Languages Stored in
Language Skills Memory?
What Brain Changes Are Associated With Can One Forget One’s First Language?
Bilingualism? SUMMARY AND THEORETICAL IMPLICATIONS
In some parts of the world, such as the United States, there is the misperception that knowing
only one language, or monolingualism, is the norm. In fact, it is common to know more than
one language. Approximately half of the world’s population use more than one language in their
daily life. Even in the United States, 55% of those surveyed for the 2000 census reported being
proficient in English and using a language other than English at home (U.S. Census Bureau,
177
178 INTRODUCTION TO LANGUAGE DEVELOPMENT
2000). In the United States, it is expected

that the number of bilinguals will con-
tinue to rise in the coming decades.
The term bilingualism is used to
describe cases in which one knows
more than one language. The term mul-
tilingualism is used to describe cases in
which one knows more than two lan-
guages. In numerous regions of the
world, residents must be trilingual to
be successful, mastering the language
of the home, which may be a tribal or
regional language, and also mastering
official languages of the local and
 Photo 7.1 A bilingual infant lies between Mom and Dad. Would national government. Table 7.1 displays
learning two languages in childhood be appealing to you? a list of countries with populations over
30 million in which there is more than
one official language. Many countries, including those in the table, recognize regional lan-
guages. The table includes only those languages that have been recognized at the national level.
There are many more countries in the world where communities, cities, and provinces are
unofficially multilingual. In this chapter, you will learn about how bilingual children differ from
monolingual children in terms of their language development as well as how they differ in
terms of other aspects of their cognitive processing. Regardless of the social or political environ-
ment where children are born, they will learn whatever languages are spoken regularly in their
environments. This chapter will also review the different ways in which children and adults
acquire second and third languages and what is known about how bilinguals store knowledge
of multiple languages in memory and access the knowledge during language processing.
Table 7.1 Countries With Populations Over Thirty Million With More Than One Official
Language
Country Population Official Languages

India 1.2 billion 2 (Hindi and English)
Pakistan 197 million 2 (Urdu and English)
Philippines 94 million 2 (Filipino and English)
Democratic Republic of the Congo 66 million 5 (French and four others)
South Africa 50 million 11 (Afrikaans, English, and nine others)
Spain 46 million 5 (Spanish and four others)
Tanzania 43 million 2 (Swahili and English)
Kenya 39 million 2 (Swahili and English)
CHAPTER 7 Life With More Than One Language 179
Country Population Official Languages

Algeria 36 million 2 (Arabic and Berber)
Canada 35 million 2 (English and French)
Morocco 32 million 2 (Arabic and Amazigh)
Uganda 32 million 2 (Swahili and English)
Iraq 32 million 2 (Arabic and Kurdish)
Afghanistan 31 million 2 (Dari and Pashto)
Sudan 31 million 2 (Arabic and English)
Source: Adapted from The World Factbook (2009).
Bilingual Children
How Do Bilingual Children Differ From Monolingual Children?

A common way in which infants become bilingual is that they are born to parents who
speak different languages. When both languages are used regularly in the home, the infant
is usually able to acquire both languages simultaneously. The term simultaneous bilingual-
ism (or simultaneous multilingualism) is used to describe circumstances in which multiple
languages are learned from birth. In contrast, when a second language (L2) or third
language (L3) is learned after a first language (L1), then the term sequential bilingualism
(or sequential multilingualism) applies.
A long-standing empirical question has been whether bilingual infants experience
delays in language development as compared with infants who are learning just one
language. Delays for bilingual infants might be expected because they are doing twice
the work of the monolingual infant. Often, there is the concern among monolingual
family members or family friends that the infant will be confused by being exposed to
two languages. Some may have been so concerned about the possibility that a child
might be adversely affected by learning two languages that the infant was prevented
from being exposed to more than one language. The remainder of this section will
review the differences between bilingual and monolingual children in terms of the
reaching of language milestones, the development of perceptual abilities, and vocabu-
lary development as well as the research on the extent to which bilingual children inap-
propriately mix their languages.
Language Milestones
Decades of research on childhood bilingualism indicates that there is no need for concern.
Bilingual and monolingual children reach the language development milestones at roughly
the same times. Research suggests that the age at which first words are produced (i.e.,
around 12 months) does not differ for bilingual and monolingual children (Genesee, 2003;
Patterson & Pearson, 2004). A study by Petitto and colleagues (2001) compared the
language development of bilingual children raised in a home in which both French and
English were used and children raised in a home in which French and a signed language
used in Quebec were used. They found that both groups of bilingual children began speak-
ing/signing around the same time as is normally observed for monolingual children (i.e.,
12 months). The bilingual children began using two-word utterances around the age
observed for monolinguals (i.e., 18 months).
Perceptual Abilities
Between birth and 12 months, bilingual and monolingual children may differ in their
ability to distinguish phonemes; however, studies have produced conflicting results on
this issue. Research has shown that there are differences in the perceptual abilities of
bilingual and monolingual infants. Bosch and Sebastián-Gallés (2003) compared the
perceptual abilities of monolingual and bilingual infants. One group of monolingual
infants was reared in Spanish-speaking homes. Another group of monolingual infants
was reared in homes speaking Catalan, a Romance language that is distinct from
Spanish. A third group of bilingual infants was reared in homes in which both Spanish
and Catalan were spoken. They tested how well infants perceived a vowel contrast that
involved two distinct vowels in Catalan but involved a single phoneme in Spanish. They
found that all infants could perceive the vowel contrast at 4 months. Infants reared in
homes where Catalan was spoken were able to perceive the contrast when they were
retested at 8 and 12 months. They found that monolingual infants reared in Spanish-
speaking homes became less and less able to perceive the contrast when they were
retested at 8 and 12 months. The bilingual infants showed a U-shaped pattern of per-
formance. Their ability to perceive the contrast declined at 8 months as compared with
their original performance, but when tested at 12 months, their performance increased
back to the level observed at 4 months. The results suggested that the perceptual devel-
opment of bilingual and monolingual infants may proceed somewhat differently.
More recent studies have found similar perceptual development for bilingual and mono-
lingual infants (Burns, Yoshida, Hill, & Werker, 2007; Sundara, Polka, & Molnar, 2008). They
compared monolingual English, monolingual French, and bilingual (English⁄French)
infants’ ability to perceive a consonant contrast (i.e., the French /d/, which differs in terms
of place of articulation from the English /d/). The phonemes are not perceived as distinct
by either English or French speakers but are perceived as distinct by speakers of other
languages. When they tested infants who were between 6 and 8 months old, both
the monolingual and bilingual infants were able to distinguish the sounds. When they were
retested at 10 to 12 months, the English monolingual infants and the bilingual infants were
able to distinguish the sounds, but French monolingual infants could not. The authors
pointed out that French-speaking adults do not reliably distinguish the two sounds
(Sundara & Polka, 2008).
In a similar study, Burns and colleagues (2007) compared the perceptual abilities of
monolingual infants reared in English-speaking homes with those of bilingual infants reared
in homes where both English and French were spoken. The contrast involved a pair of con-
sonants that differed in voice-onset time. As you learned in Chapter 3, some consonants vary
only in terms of the duration of silence between the beginning of the sound and when voicing
begins (e.g., /p/ and /b/ in English). Both groups of infants distinguished English/French when
tested at 6 to 8 months. When retested at 10 to 12 months and also at 14 to 20 months, bilin-
gual infants performed as well as they had when tested at 6 to 8 months, but monolingual
infants were able to perceive only the consonant that was a phoneme in English.
Werker, Weikum, and Yoshida (2006) suggested that bilingual infants may vary in terms
of whether the two languages spoken in the home are equally dominant or whether one
language is used more often or viewed as more important than the other. In their study,
they tested infants reared in homes where both French and English were spoken. They
tested their ability to perceive sounds found only in French and sounds found only in Eng-
lish. When they tested infants between 14 and 17 months, they found that some of the
bilingual infants could discriminate both types of sounds, while other bilingual infants
could discriminate just the sounds found in English or could discriminate just the sounds
found in French.
Vocabulary Size
Studies focusing on vocabulary size for bilingual and monolingual children have observed
differences (Pearson & Fernández, 1994; Pearson, Fernández, & Oller, 1993; Rescorla &
Achenbach, 2002). Bilinguals appear to have slightly smaller vocabularies than monolin-
guals. In an early study, Pearson and colleagues (Pearson & Fernández, 1994; Pearson et al.,
1993) compared the vocabulary knowledge of Spanish–English bilingual children with that
of monolingual children. They found that the vocabulary knowledge and the occurrence
of the word spurts were similar for both groups of children. Rescorla and Achenbach (2002)
analyzed data from a national language development survey (Rescorla, 1989) and found
that bilingual children had smaller vocabularies than monolingual children. The authors
pointed out that the methodology might have contributed to the different results for bilin-
gual and monolingual children, because vocabulary size was reported by parents. The task
for parents of bilingual children was likely more complex and more prone to error than for
parents of monolingual children.
Critics point out methodological weaknesses of the study, such as a smaller sample size
for the bilingual group than the monolingual group (Patterson, 2004). A point that was not
made by Rescorla and Achenbach (2002) or Patterson (2004) was that a comparison of a
bilingual child’s vocabulary in one of that child’s languages with a monolingual child’s
vocabulary may be the wrong comparison. The bilingual child is likely to know two words
for each concept, one word from each language. Assessing the bilingual child’s vocabulary
size in both languages and comparing it with the monolingual child’s vocabulary in one lan-
guage is likely to reveal that the bilingual child knows more words overall—even that child
knows fewer words than the monolingual children in L2. However, recent research has found
that bilingual school-age children have smaller productive vocabularies when vocabulary in
both languages is taken into account (Yan & Nicoladis, 2009). Similar results have been
obtained in studies in which the productive vocabularies of children under 3 years of age
were compared (Junker & Stockman, 2002; Oller & Eilers, 2002; Pearson et al., 1993, 1995;
Petitto & Kovelman, 2003).
Studies of children’s receptive vocabu-

lary development (i.e., ability to compre-
hend words when spoken to them) have
also been conducted with bilingual and
monolingual children. Some studies
have shown that bilingual children have
smaller receptive vocabularies in each of
their languages than do monolingual
children (Bialystok, Barac, Blaye, & Poulin-
Dubois, 2010; Bialystok, Luk, Peets, &
Yang, 2010; Mahon & Crutchley, 2006).
Other studies have found that bilingual
and monolingual children have similar
 Photo 7.2 Two girls are talking on their way to school; one receptive vocabularies (Cromdal, 1999;
of the girls is bilingual. Do you know how many children in the Yan & Nicoladis, 2009). Recent research
United States speak a language other than English at home? suggests that the differences in vocabu-
lary size for bilinguals and monolinguals
is also observed for young adults (Portocarrero, Burright, & Donovick, 2007); however, other
studies have failed to observe differences (Bialystok, Craik, Klein, & Viswanathan, 2004).
When researchers investigated the composition of bilingual children’s vocabularies,
they found that many words were translation equivalents (i.e., words from the bilingual’s
two languages that mean the same thing, such as apple and manzana). Genesee and Nicoladis
(2007) studied the vocabularies of children between 8 and 30 months reared in homes in
which both Spanish and English were used. They found that 30% of the children’s vocabu-
lary involved translation equivalents (e.g., apple–manzana). Petitto and colleagues (2001)
observed similar results in a study of children who spoke French and used a signed lan-
guage. Between 36% and 50% of children’s vocabularies were translation equivalents.
There are likely many factors, in addition to whether a child is a bilingual or mono-
lingual, that can influence vocabulary size. Among those are being raised in a low-
income home (Hart & Risley, 1995) and speaking a minority language (i.e., not the
dominant language of a country or region) (August & Shanahan, 2006). A recent study
by Vagh, Pan, and Mancilla-Martinez (2009) investigated the vocabularies of children
between the ages of 24 and 36 months who were being raised in low-income homes.
They relied on both parent and teacher reports of children’s vocabulary knowledge.
They found that vocabulary size increased more slowly for Spanish–English bilingual
children than for monolingual children. Among bilingual children, those for whom
English was the dominant language had larger vocabularies than those for whom
Spanish was the dominant language.
Differences in bilingual and monolingual children’s vocabulary acquisition may be
related to their using different strategies in word learning. As you learned in Chapter 3,
research has shown that children exhibit the mutual exclusivity principle when learning
new words. Children tend to assume that if an object is called something then it is unlikely
to be called something else. Bilinguals generally know two words for each object—a word
in each language. Houston-Price, Caloghiris, and Raviglione (2010) compared how
monolingual and bilingual infants between the ages of 17 and 22 months interacted with
novel objects in a task using the preferential looking paradigm. Infants were shown two
objects—a novel object and one whose name was familiar. They were then asked to look at
the dax. The results showed that monolingual infants looked more often at the novel object,
demonstrating the mutual exclusivity principle. In contrast, bilingual infants did not. The
authors concluded that infants raised with multiple languages do not develop the mutual
exclusivity strategy. Older bilingual children appear to use the strategy. In a study with
children between the ages of 3 and 6, Davidson, Jergovic, Imami, and Theodos (1997) found
that bilingual children relied on the mutual exclusivity principle but to a lesser extent than
did monolingual children.
Language Mixing
Another common concern among parents raising bilingual children is that the children will
be confused learning two languages at the same time and will mix up the languages. Again,
the fear appears to be that initially mixing up the languages and later figuring out that the
two languages are separate entities could cause bilingual children to lag behind their
monolingual peers in the long term. Meisel (1989) has referred to this possibility as the
fusion hypothesis. Early in life, the bilingual children's two languages are fused. They do
not differentiate their languages early in life but only do so at some point later in child-
hood. Observations of language mixing have been reported by parents as well as research-
ers in children between the ages of 2 and 3 (Köppe, 1996). The fear that language mixing
may indicate that the child is experiencing a problematic form of language confusion
appears to be unfounded. After the age of 3, the amount of language mixing observed in
these children's speech decreases dramatically. Only about 2% of bilingual preschoolers
produced utterances in which languages were mixed (Lindholm & Padilla, 1978).
Bilingual children’s production of sentences containing words from both languages may
be not be due to their language systems being fused early in life. Critics of the fusion
hypothesis suggest that language mixing could occur because children are not pragmati-
cally competent (De Houwer, 2005; Genesee, 1989; Köppe, 1996). They may not yet under-
stand that the language used in an utterance must be tailored to the language(s) known by
listener(s). Another possibility is that children are purposely mixing the words of their
languages together as a form of play or linguistic creativity. As we learned in Chapter 3,
children create idiosyncratic words or idiomorphs when learning new words. It is possible
that they enjoy playing with sentences as well, creating novel combinations of words. Some
children spontaneously create their own language games (Cumming, 2007). For example,
every so often, my 3-year-old grandson Griffin finds it very amusing to produce a long
string of nonsense syllables. His father told the group that he had been doing that for a
while, and when someone asks him if that is his own language, he laughs and says, “Yes.
Griffin’s language!”
Code-Switching
The view that language mixing by bilingual children may reflect pragmatic development
is supported by the observation that language mixing occurs in the speech of adult bilinguals.
As you learned in Chapter 4, adult bilinguals often will produce utterances containing
words from both of the languages that they know when speaking with other bilinguals.
This form of language use has been called code-switching. Poplack (1980) titled her
article about code-switching with an example of code-switching: “Sometimes I’ll start a
sentence in Spanish y termino en Espanol.” Examples of code-switching by a Spanish–
English bilingual from Poplack (1980) are displayed in 1. The view of Poplack (1980) and
other language researchers is that code-switching is not a form of unusual or deviant
language behavior; rather, it is part of the normal language repertoire of a bilingual
(Gumperz, 1971, 1976).
1. a. Me iban a lay off. They were going to lay me off’.

b.
Leo un magazine. I read a magazine.
There are likely to be many reasons that adult bilinguals engage in code-switching.
Heredia and Altarriba (2001) suggested that code-switching may occur when the speaker
lacks proficiency in one language. When speaking in L2, bilinguals may find that they do
not know or cannot remember at that moment the particular L2 word needed. An L1 word
can be substituted. Heredia and Altarriba (2001) pointed out that some code-switching may
be the result of word-finding difficulty. Monolingual and bilingual speakers alike some-
times experience difficulty coming up with a known word. These occurrences have been
called tip-of-the-tongue states, or TOT states (James & Burke, 2000). The bilingual speaker
may experience a TOT state in one language but be able to retrieve from memory the word
in the other language.
Analyses of sentences containing code-switching have shown that speakers appear sen-
sitive to syntactic structure when planning code-switched utterances. For example, speakers
do not produce multimorphemic words in which some morphemes are in one language and
the other morphemes are in another language. Poplack (1981) referred to this as the free
morpheme constraint. The view that code-switching adheres to structural principles known
to the speakers is supported by research showing that bilinguals show agreement about
which code-switched sentences are grammatically acceptable and which are grammatically
unacceptable (Aguirre, 1980; Gingras, 1974; Gumperz, 1976; Timm, 1975). For example, in
Spanish, adjectives are placed after the nouns that they modify (e.g., quiero un tomate verde,
which means I want a green tomato). Lederberg and Morales (1985) pointed out that the
grammaticality of code-switched Spanish–English sentences containing adjectives
depends on the rules of the language of the adjective. If the adjective appears in English, it
is only grammatical when it precedes the noun as in 2a versus 2b. If the adjective appears
in Spanish, it is only grammatical when it follows the noun, as in 2c versus 2d. (The asterisks
indicate that the sentences are ungrammatical.)
2. a. Quiero un green tomate.

*b.
Quiero un tomate green.
c.
I want a tomato verde.
*d.
I want a verde tomato.
Code-switching is also more likely to be used in some situations than in others

(Gumperz, 1976). Some of the factors that may be related to the frequency of code-
switching include the topic being discussed, the social status of the individuals taking
part in the conversation, the language proficiencies of those individuals, where the con-
versation is occurring (i.e., home, work, school, church), and the level of emotion
involved in the conversation. There appears to be a particularly strong relationship
between the emotionality of the topic being discussed and the likelihood that code-
switching will be used.
Research has shown that when talking about stressful or traumatic past experiences,
bilinguals tend to prefer to use their L2 rather than their L1. Since the 1980s, therapists
working with bilingual clients have reported that bilinguals often show a preference of
discussing embarrassing or upsetting topics in their L2 (Bond & Lai, 1986). Santiago-Rivera
and Altarriba (2002) proposed that bilinguals use the L2 to discuss emotionally charged
information because it serves to distance them from the upsetting content. Their explanation
for this phenomenon is that bilinguals “represent emotional words differently in their two
languages and typically associate these words with a broader range of emotions in their
first language” (Santiago-Rivera & Altarriba, 2002, p. 33). They also suggested that memo-
ries for emotional events contain information about which language was used at the time
of the event. When bilinguals recount memories, the language that was being used in the
episode may have a privileged role in remembering and talking about the event (e.g., Javier,
1996; Rubin, Schrauf, Gulgoz, & Naka, 2007). In a recent study, Santiago-Rivera, Altarriba,
Poll, Gonzalez-Miller, & Cargun (2009) interviewed nine bilingual therapists about the situ-
ations that led them to use L1 or L2 in sessions with clients and situations that led clients
to use L1 or L2. They found that when clients wanted to discuss an emotionally charged
topic (e.g., one involving anger), they usually used L1. At times in which they appeared to
want to distance themselves from the event, they would switch to L2. In contrast, therapists
reported switching from L1 to L2 as a strategy to redirect the conversation with the client.
One therapist was quoted as saying, “I may challenge some beliefs, and if I feel I am not
being successful in one language, I try to present it in another language” (Santiago-Rivera
et al., 2009, p. 439).
The relationship between the language mixing produced by bilingual children and the
code-switching behavior produced by adults is not well understood. Future research is
needed to determine whether children’s utterances containing language mixing appear to
show the same syntactic characteristics as adult utterances involving code-switching. Fur-
thermore, language development researchers would be interested in knowing whether the
amount of language mixing observed in bilingual children is related to the amount of adult
code-switching that children hear in their environments. As we learned in Chapter 1, theo-
ries of language development differ with regard to the role that language experience plays
in children’s acquisition, particularly in the acquisition of morphological and syntactic
structures. The study of children’s language mixing (or code-switching) appears to be well
suited for the testing of these theories.
There are situations in which an entire community may participate in language mixing,
and, over time, the language mixing leads to the formation of a new language. Text Box 7.1
describes how this can happen when a pidgin leads to the formation of a creole.
Text Box 7.1 Diversity of Human Languages: Pidgins and Creoles
When speakers of different languages and cultures work and live together on a daily basis, they
often devise ways of communicating, using a small number of words drawn from one or more
of the represented languages. Over time, the speakers come to understand one another. The
term lingua franca is used to refer to communication systems that develop to enable speakers
of different languages to communicate. In some cases, a lingua franca may become a pidgin,
when individuals who speak different languages work together over long periods. Around the
world, pidgins have developed in busy seaports, where there is an abundance of international
trade (Todd, 1990). For example, in the early 20th century, Hawaii was home to immigrants from
Japan, Korea, and the Philippines who worked on sugar cane plantations for English-speaking
landowners. Out of this unique situation developed Hawaiian Creole English (Bickerton, 1981,
1984). The communicative power of a pidgin is in its vocabulary rather than its grammar. Most
pidgins have a highly simplified grammar and most lack complex clause structures. However, an
interesting thing happens when children are exposed to a pidgin and acquire it as their L1; the
language that the children end up speaking is grammatically more complex. The word creole is
used to refer to a language that was once a pidgin but has subsequently become a native lan-
guage for the next generation. Bickerton (1983) pointed out that creoles are grammatically more
similar to one another, despite being located in distant parts of the world, than they are to the
languages from which the pidgin predecessor languages were derived. He identified 12 charac-
teristics that creoles around the world share, including an SVO word order, the verb-tense system,
and the formation of questions and negative sentences. Bickerton proposed the grammars of cre-
oles are similar because children who learn pidgins use aspects of innate universal grammar (UG)
as they learn pidgins. In his language bioprogram hypothesis, he claimed that using their innate
knowledge, children who acquire a pidgin as an L1 can add in grammar where it is missing. An
alternative view to explain why creoles are so syntactically similar is that some of the grammati-
cal consistencies observed in creoles may be due to similarities found in the languages that make
up the pidgins. Most creoles in the world have come from pidgins that have at least one language
from the Indo-European language family. Here are examples from pidgin and Hawaiian Creole
English from Bickerton (1991).
Pidgin Hawaiian Creole English
Now days, ah, house, ah, inside, washi . . . Those days bin get [there were] . . .
. . . clothes machines get, no? Before time . . . , . . . no more washing machine . . . ,
. . . ah, no more, see? And then pipe no more . . . , . . . no more pipe water like get [there . . .
. . . water pipe no more. . . . is] inside house nowadays, ah?

Are There Cognitive Benefits When Children Are Bilingual?

Parents and caregivers of bilingual children are likely to be surprised to learn that there is
ample evidence that there are concrete, measurable cognitive benefits when a child learns
more than one language. The first of the studies suggesting the benefits of childhood bilin-
gualism was reported in 1962 by Peal and Lambert. They showed that 10-year-old French–
English bilingual children in Montreal performed better on verbal and nonverbal tasks than
10-year-old monolinguals from the same school. The authors concluded that the bilingual
children were cognitively more flexible than monolingual children. Ricciardelli (1992)
reported similar results in a study with a group of Italian–English bilingual children.
Bilingual children who were proficient in both languages outperformed monolingual chil-
dren on tests of creativity.
In a study with Hebrew–English bilingual children, Ben-Zeev (1977) found that they
performed better than monolingual English-speaking children and monolingual Hebrew-
speaking children on two cognitive tasks. The three groups of children were between the
ages of 5 and 8. One task required children to study a display of nine cylinders of varying
sizes arranged into three columns and three rows. They were then asked to describe the
pattern of the cylinders and transpose it. In a second task, children played a language game
with the interviewer. The interviewer told the child that an airplane was called a turtle. The
interviewer then asked the children questions, such as Can a turtle fly? and How does a
turtle fly? The author concluded that bilingual children are better than monolingual chil-
dren at “seeking out rules and for determining which are required by the circumstances”
(Ben-Zeev, 1977, pp. 1017–1018).
Hakuta and Diaz (1985) studied a large group of Spanish–English bilingual children in
Puerto Rico. All children were from low socioeconomic backgrounds and were more profi-
cient in Spanish than in English. Children who showed the highest degree of bilingualism
performed better than other children on tests of nonverbal intelligence and metalinguistic
awareness. The study was conducted longitudinally, enabling the researchers to test children
multiple times. They found that those children who became more proficient in English over
time also showed increased nonverbal intelligence and metalinguistic awareness.
Attentional Processing
Perhaps, the most convincing evidence of all has been obtained by Bialystok and colleagues
(Bialystok, 1999, 2005; Bialystok & Majumder, 1998; Bialystok & Martin, 2004). They com-
pared the attentional processing of 30 Chinese–English bilingual children and 30 monolin-
gual English-speaking children. Children were between 3 and 7 years old. In one task,
children were shown an object and a card on which the object’s name was written.
Children were then distracted by two toy bunnies. The bunnies caused the card to be
moved to a position under a different object. After the distraction and while the card was
near the wrong object, children were asked what the card had said. The interviewer then
pointed out the fact that the bunnies had disrupted things. The card was moved back to the
original object, and children were asked again to report what the card said. In a second task,
children were given a set of cards and asked to sort the cards into two piles. They were
instructed to sort according to a characteristic displayed on the cards (e.g., color or shape).
When they completed the task, they were asked to sort the cards again using a different
characteristic. Both tasks required children to exert attentional control during the tasks.
Carlson and Meltzoff (2008) found that bilingual children’s advantage in attentional
processing over monolingual children occurs in tasks in which there are conflicting atten-
tional demands but not in tasks in which participants must control impulses—as when a
response should be given after a short delay. Their study involved 6-year-old English–
Spanish bilinguals. The bilingual advantage over monolinguals in attentional processing
has been observed with infants as young as 7 months old (Kovács & Mehler, 2009) as well
as in adults (Bialystok, Craik, & Luk, 2008; Costa, Hernández, & Sebastián-Gallés, 2008).
The bilingual advantage over monolinguals that has been observed in healthy children
and adults has also been observed among older adults with dementia. Bialystok, Craik,
and Freedman (2007) compared the progression of symptoms in 93 bilinguals and
91 monolinguals. The results showed that bilinguals were diagnosed with dementia an
average of 4 years later than monolinguals, suggesting that the bilingualism served as a
protective factor.
Language Skills
Bilingual children also appear to be better judges of grammaticality than monolingual
children. Galambos and colleagues (Galambos & Goldin-Meadow, 1990; Galambos &
Hakuta, 1988) compared the ability to judge and to correct syntactically ungrammatical
sentences in a group of Puerto-Rican Spanish–English bilinguals and English-speaking
monolinguals. Both groups were from low-income families. Bialystok and colleagues
(Bialystok, 1986, 1988; Bialystok & Majumder, 1998) pointed out that grammaticality judg-
ments require the participant to simultaneously consider the structure and the meaning of
the sentence. When the two aspects of the sentence are in conflict, as occurs when one is
ungrammatical, the participants must resolve the conflict by directing attention to the
appropriate aspect of the sentence. Following this reasoning, they view the task of one
involving attentional processing. They compared the ability of bilingual and monolingual
children between the ages of 5 and 9 to judge sentences that were anomalous in terms of
either grammar or meaning. They found that bilingual and monolingual children per-
formed similarly judging grammatically correct sentences with sensible meaning (i.e., they
correctly labeled sentences such as Apples growed on trees as ungrammatical); however,
bilinguals were better than monolinguals at judging grammatically correct sentences that
did not have sensible meaning (i.e., accepting sentences such as Apples grow on noses as
grammatically correct). The results have been replicated with Italian–English bilinguals
(Ricciardelli, 1992) and Swedish–English bilinguals (Cromdal, 1999).
Studies have also found that children who have learned another language show gains in
their knowledge of English language structure and vocabulary (Curtain & Dahlberg, 2004;
Dumas, 1999). Other studies found that bilingual children performed better than monolin-
gual children in English as well as other subjects, such as social studies and math (Andrade,
Kretschmer, & Kretschmer, 1989; Armstrong & Rogers, 1997; Masciantonio, 1977;
Kretschmer & Kretschmer, 1989; Rafferty, 1986; Saunders, 1998). Armstrong and Rogers
(1997) found benefits in math performance for children after they had studied an L2 for
only 90 minutes per week for just one semester.
What Brain Changes Are Associated With Bilingualism?

There has been a great deal of interest among neuroimaging researchers regarding the
organization and functioning of the bilingual brain. Several studies have suggested that
early learned languages and late learned languages are handled somewhat differently by
the brain; later learned languages involve more frontal and more bilateral activation than
early learned languages (Dehaene et al., 1997; Hahne & Friederici, 2001; Hernandez,
Martinez, & Kohnert, 2000; Kim, Relkin, Lee, & Hirsch, 1997; Marian, Spivey, & Hirsch,
2003; Weber-Fox & Neville, 1999, 2001). Certainly, because this area of research is so new,
there are many more questions than there are answers at this point. The excitement
comes in knowing that in the coming decades, there are likely to be important discoveries
made that have the potential to change the way that we think about and also engage in
L2 learning.
So far, multiple research studies have shown that there are brain-related differences
related to proficiency. Some studies, but not all, have found brain-related differences
related to the age at which the bilingual began using the L2. A study by Wartenburger and
colleagues (2003) found that there were differences in bilingual brain organization related
to both language proficiency and the age at which one was first exposed to the L2. Chee,
Soon, Lee, and Pallier (2004) found brain differences related to language proficiency, which
were unrelated to the age of acquisition. Kim and colleagues (1997) compared brain activa-
tions during language processing for bilinguals who learned their L2 early in childhood or
later in life. They had participants use each of their languages while their brain activation
was recorded. They found that there was an area of activation in the left inferior frontal
gyrus that was observed during the use of each language. For early bilinguals, the areas of
activation during use of L1 and L2 overlapped. For late bilinguals, the areas of activation
during the use of L1 and L2 did not overlap. For both groups, there was a second region of
activation that did overlap during the use of L1 and L2; the region was located in the supe-
rior temporal gyrus. Some have questioned these results because language proficiency was
not assessed and the participant groups may have differed in ways other than when they
learned L2 (Abutalebi, Cappa, & Perani, 2005).
One of the most intriguing studies is that by Mechelli and colleagues (2004) who
compared the densities of gray and white matter in the brains of monolingual adults and
two groups of bilingual adults. One group of bilinguals had acquired an L2 before the
age of 5; the other group had acquired an L2 between the ages of 10 and 15. Both groups
of bilinguals reported using their L2 regularly. They found that in the left parietal cortex,
the density of the gray matter was greater for bilinguals than for monolinguals. The
density of gray matter in this area was greater for bilinguals who had learned the
L2 early in childhood versus later in childhood. Last, in correlational analyses, the density
of gray matter increased as the participants’ reported proficiency increased and
increased as the participants’ age of acquisition decreased. Past research has shown that
the particular area in the left inferior parietal cortex is involved in verbal fluency (Poline,
Vandenberghe, Holmes, Friston, & Frackowiak, 1996; Warburton et al., 1996). These
results suggest that the differences observed between bilingual and monolingual speak-
ers may one day be related to specific brain-related changes that occur when one begins
to learn an L2.
A number of recent studies have found

that specific brain regions are activated
when a bilingual switches languages
(Hernandez et al., 2000; Price et al., 1999;
Quaresima, Ferrari, van der Sluijs, Menssen,
& Colier, 2002; Rodriguez-Fornells, Rotte,
Heinze, Nösselt, & Münte, 2002; Rodgriguez-
Fornells et al., 2005). The regions include
the dorsolateral prefrontal cortex, inferior
frontal cortex, anterior cingulate, and
supramarginal gyrus. In these studies,
bilinguals carried out two language tasks
(e.g., a word-reading task and a translation
 Photo 7.3 High school students are studying in a typical L2 task). The language in which the words
class. Did you take an L2 in high school? were presented was either kept the same
or varied so that every other word appeared
in the other language. The results suggest that the differences in attentional processing that
have been observed between bilingual and monolingual children may also be related to
specific brain-related changes that occur when one begins to learn an L2.
Learning Second Languages
What Teaching Methods Are Used in Language Classes?

There are multiple ways for one to become bilingual. For many, learning an L2 involves
some classroom instruction. Over the last 150 years, numerous methods of L2 teaching
have been developed (Krause, 1916; Richards & Rodgers, 2001). Some methods have
become brand names, sold to those who would like to learn a language at home (e.g., Berlitz
and Rosetta Stone). Methods typically differ in the extent to which the L1 is used during
L2 instruction and the extent to which grammar is explicitly taught. The grammar-
translation method (also called the traditional method) utilizes the L1 and also explicitly
teaches students the grammatical rules of L2. At the other extreme, there is the direct
method, which involves students learning the L2 without the use of the L1. The grammar
of L2 is learned inductively, through exposure to L2 sentences.
In most L2 classrooms in the United States, instructors use the grammar-translation
method. Lessons are focused on having students memorize grammatical rules and vocabu-
lary. They are also asked to translate from L1 into L2 and vice versa. Exercises involve
students’ practicing writing or speaking sentences containing verbs in their different tenses
(e.g., present tense, past tense, future tense). In languages in which nouns must be marked
with suffixes indicating gender and/or number, exercises will focus on students using the
appropriate adjective–noun or subject noun–verb agreement.
There are numerous methods that place an emphasis on developing speaking and lis-
tening skills but involve varying degrees of grammar instruction. For example, the audio-
lingual method (Richards & Rodgers, 1987) emphasizes the use of spoken language but also
includes grammar exercises. It is based on the principles of behaviorism. Students are

instructed to avoid errors at all costs. The instructor, who is a native speaker, is viewed as
the model to emulate. The communicative approach emphasizes the idea that language is
primarily for communication and the expression of meaning. Students are encouraged to
use language with others. The long-term goal for students is to achieve communicative
competence. The immersion method involves having students use only the L2 to learn the
vocabulary, grammar, and pragmatics of the L2. They aim to acquire the language through
interactions with speakers of the language. Classes may be taught with complete immer-
sion (i.e., 100% of instruction is conducted in L2) or partial immersion (i.e., 50% of instruc-
tion is conducted in L2).
There are few empirical studies comparing the effectiveness of the various teaching
methods. The research that has been done has been conducted in Canada, where English–
French bilingualism has been a priority since the 1960s and the immersion method
became popular. Numerous studies comparing students’ performance in classrooms using
the direct method versus the immersion method have been conducted in Canada (Baker,
1993). In the 1960s, Canada implemented immersion language programs in schools to
address perceived inequities between French and English speakers in Quebec (Genesee,
1987). Programs were classified as early, middle, or late immersion. Early immersion
classes began when children were 5 or 6 years old. Middle immersion programs began
when children were 9 or 10 years old. Late immersion programs began when children were
11 or 12 years old. Studies showed that students in early immersion programs performed
less well on literacy skills, such as spelling and punctuation, in L2 compared with students
in classrooms in which the L2 is taught using the direct method. Differences diminished
over time. The evidence suggests that children can learn an L2 comparably well in either
type of classroom. Following the ruling, there have been a number of changes to the law,
most of them allowing schools flexibility in meeting the needs of their students.
Do All Countries Have Bilingual Education?

The extent to which L2s are used in educational settings is often determined by the law of
the land. In the United States, the Bilingual Education Act was established in 1968 (Stewner-
Manzanares, 1988). Originally introduced as a bill by Texas senator Ralph Yarborough, its
aim was to fund programs for students entering school without adequate English language
skills. The concern in Texas was schools servicing Spanish-speaking students. The Bilingual
Education Act of 1968 focused generally on the needs of schools educating bilingual
students. Financially, the act was a tremendous help to school districts, as it provided
$7.5 million for programs, teacher training, and curriculum development. Those programs
that could demonstrate success were funded for 5 years.
In 1974, the Supreme Court ruling in Lau v. Nichol mandated bilingual education. The
case was a class action suit that began because an attorney had heard about a client’s son
who was performing poorly in school because he could not speak English. The attorney filed
the suit against the San Francisco Unified School District, claiming that the Chinese students in
the district (approximately 1,800) were not being provided with equal education because of
the lack of bilingual education in the school. The case was lost at the lower court level but
was taken all the way to the U.S. Supreme Court, where the attorney prevailed.
Canada’s Official Languages Act became law in 1969 (Office of the Commissioner of
Official Language, n.d.). The law established English and French as official languages for
government and education and affirmed the equality of the two languages in all venues. It
also affirmed the right of the individual to receive services in either English or French.
Canada has supported immersion programs for students learning French. Students without
prior experience with French usually can begin immersion classes in kindergarten or first
grade and complete their secondary education completely in French. The availability of
the program depends on the province. Some regions provide French courses starting in the
fifth grade. Other regions offer French-immersion from kindergarten through ninth grade.
In regions in which French is the dominant language (e.g., Quebec), students may enter
similar English immersion programs. The indigenous languages of Canada have also been
part of Canada’s bilingual education efforts. Programs have been developed for Blackfoot,
Cree, Mohawk, Ojibwe, Mi’kmaq, Inuinnaqtun, Inuktitut, and the Pacific Coast Salish lan-
guages. The development of bilingual programs for the indigenous languages was particu-
larly important because of historical oppression of indigenous Canada. In the 20th century,
many were forced into residential schools in which they were treated poorly and forbidden
from using their native language. Similar residential schools were operated in the United
States for Native Americans (Child, 2000).
Around the world, children’s access to bilingual education varies widely. In Europe,
children typically have the opportunity to learn an L2 earlier than children in the United
States. For example, in Belgium, there are three official languages: (1) Dutch, (2) German,
and (3) French. By law, children have the right to be educated in one of these three lan-
guages. The language of instruction varies by geographic region. In the Flanders region, the
language of instruction is Dutch. In the Waloonia region, the language of instruction is
French. Throughout the country, English is frequently taught and may be required. In
Belgium, as in most of Europe, children begin their study of an L2 (i.e., a language different
from that in which general instruction is provided) in the elementary school grades
(Eurydice, 2005).
How Does First Language Influence Second Language Acquisition?

Regardless of the method used to learn an L2, one can expect to make errors. One may
find that characteristics of one’s L1 influence how one learns and/or uses one’s L2. The
term language transfer has been used to describe circumstances in which one carries
over a language rule or structure from L1 to L2. Positive language transfer occurs when
the rules in L1 are the same as in L2. Transfer helps the speaker to generate a correct
usage in L2. Negative language transfer occurs when the rules of L1 are different from
those in L2 and typically results in speakers making errors when using L2. The most
common form of negative language transfer is speaking with an accent that reflects the
phonological rules of the L1. For example, native speakers of Japanese who learn English
typically have great difficulty pronouncing the sounds /r/ and /l/, because in Japanese
these sounds are not distinct phonemes. Similarly, native speakers of English have great
difficulty pronouncing many vowels in French, which are distinct phonemes in French
but are not in English.
Language transfer may also be observed in vocabulary. When a bilingual’s two languages
share vocabulary words, one can benefit from the positive language transfer. Translation
equivalents that are similar in pronunciation and in written form are called cognates. Some
L2 words may appear similar to L1 words in sound and written form but differ in meaning,
such as the Spanish word embarazada, which means pregnant instead of embarrassed, and
the Spanish word balde, which means bucket. Such words are called false friends or false
cognates. Another example of how vocabulary can be involved in language transfer comes
from an interaction with a friend of mine whose L1 was not English. He once walked up to
me holding his arm. He had clearly injured it somehow. Before I could ask him, “What’s
happened to your arm?” he said, “I hurt my hand.” I was very confused. “Hand?” I asked.
“Don’t you mean your arm?” He paused and then realized his error. “In my language, the
word for arm and hand is the same word. I make that mistake a lot.” He once also said,
“Something’s wrong with my finger,” while he was holding his thumb. On that occasion, I
was able to infer that in his native language, the word for thumb is the same word as for
the other fingers.
The use of pronouns can often be influenced by language transfer. If one’s L1 uses the
same word to refer to he and she but one’s L2 has different words to refer to he and she, one
may find that one comes to rely on one of the L2 words for both pronouns. In Finnish, the
word hän is used to refer to males and females. When Finnish native speakers learn
English, they sometimes may use the pronoun he in English to refer to female antecedents,
as in *I met Mary; he is nice (the asterisk indicates that the sentence is ungrammatical). In
Chinese, the pronouns for he and she are pronounced the same but spelled differently. For
native speakers of Chinese acquiring English (or other languages in which there are differ-
ent words for he and she), pronoun errors in L2 tend to occur.
Language transfer may also be observed in the syntactic and morphological rules that
speakers apply. Native speakers of English frequently transfer the word order of English
when producing German clauses. An example of an English sentence containing two
clauses is provided in 3a. An example of a grammatically correct German sentence is
provided in 3b. In 3b, the adverb heute, which means today, is ordered before the verb
in German, because in German, the main verb usually appears in clause-final position.
The relative clause in 3c is the grammatically incorrect German form sometimes pro-
duced by English native speakers (the asterisk indicates that the sentence is ungram-
matical). The positioning of the adverb in the relative clause follows the verb rather than
preceding it.
3. a. I know when she arrives today. English

b.
Ich weiß, wann er heute ankommt. Correct German
*c.
Ich weiß, wann er ankommt heute. Incorrect German
L2 learners may also transfer morphological rules from L1 to L2. For example, in
English, one forms a comparative adjective by adding -er if the adjective is one syllable or
ends in -y, as in taller and happier. For longer words, the word more is placed in front of the
adjective, as in more important and more intelligent. Superlative adjectives in English are
formed similarly, by adding -est to an adjective or preceding the adjective by the word
most, as in tallest, happiest, most important, and most intelligent. Some languages, such as
Spanish, form comparative and superlative adjectives using one rule. The word mas is
placed before an adjective to form the comparative and the words le mas (or la mas) are
placed before the adjective to form the superlative. When Spanish native speakers learn
English, they often overuse the words more and most when forming comparative and
superlative adjectives in English. In recent research, Kennison and Bowers (2011) found that
the overuse of more and most in comparative and superlative adjectives in English by
Spanish native speakers was negatively correlated with how long participants had resided
in the United States, suggesting that exposure to the L2 in everyday life was related to fewer
errors in L2 resulting from language transfer.
The prior research on language transfer explains why language learners face greater
difficulty in mastering an L2 when the rules of the language differ a great deal from the rule
of their native language. Conversely, L2 acquisition may be easier when one’s L1 and L2
share many of the same grammatical distinctions and rules. For example, learning Italian
when one’s L1 is Spanish may be easier than learning Chinese when one’s L1 is Spanish.
As you learned in Chapter 1, languages belonging to the same language family are more
similar in grammatical rules than languages belonging to different language families.
While many students in the United States struggle to learn an L2 in high school and/or
in college, there are some individuals who appear to have a knack for languages. Those
who have mastered three, four, or five languages usually are regarded has being particularly
gifted in learning languages. The term polyglot is used to refer to individuals who speak
many languages. They are able to learn multiple languages rapidly and with apparent ease.
It is difficult to estimate the percentage of the population who possess above average
language-learning abilities. Text Box 7.2 describes a particularly prolific polyglot: the
German Emil Krebs.
Life With Two or More Languages

Those who use more than one language in daily life may be described as balanced bilin-
guals, which means that they are equally proficient in both of their languages. Most often,
bilinguals report having a dominant language or a language that they feel most proficient
in and use more often. The dominant language is not always L1. Some individuals may
reach a high level of proficiency in two languages and end up living in a setting in which
L2 is used predominantly. For example, someone raised in Canada becomes proficient in
both English and French. After college, the person gets a job in France and works there for
several decades, using mostly French in daily life. English is used only to speak to family
or friends, which occurs only several times a month.
What Factors Predict Proficiency?

Bilinguals certainly differ from one another in terms of proficiency. Some bilinguals
become proficient in speaking, reading, listening, and writing L2; however, others may not
master all aspects of L2 usage. In research with bilinguals, researchers mostly rely on
Text Box 7.2 Extraordinary Individuals:

Emil Krebs: An Extraordinary Polyglot
Emil Krebs (1867–1930) was born in Germany,

the son of a carpenter. By the end of his life, he
had become proficient in speaking and writing
60 languages (Matzat, 2000). He had studied
120 others. He is aptly described as a polyglot,
which is a term used to refer to someone who
knows many languages. In one written account,
Krebs was described as learning languages rapidly.
For example, when he studied Armenian, he mas-
tered it in about 9 weeks (von Hentig, 1962). He
spent 2 weeks learning the grammar. He studied
Old Armenian for 3 weeks and focused on the spo-
ken version of the language for the remaining 4
weeks. Between the ages of 13 and 17, he attended
gymnasium (roughly equivalent to a college pre-
paratory high school) and studied Latin, Hebrew,
French, and classical Greek. By the time he com-
pleted his gymnasium studies in 1887, he spoke
12 languages, including Modern Greek, English,
Italian, Spanish, Russian, Polish, Arabic, and  Photo 7.4 Emil Krebs lived between 1867
Turkish. In that year, he entered the Berlin School and 1930 and is known for being an amazing
of Law, where he continued studying languages. In polyglot. Which language would you start
learning today if you were assured that you
1890, he passed a translator’s exam for Chinese.
could master it?
In 1913, he went to China, where he worked as a
translator for the government until 1917, when dip-
lomatic relations between Germany and China dissolved. One can find Krebs’s personal library
of approximately 3,500 books and his writings in over 120 languages at the National Archives
in Washington, D.C. Among the other languages that he knew were Egyptian, Ainu, Albanian,
Armenian, Burmese, Georgian, Hebrew, Japanese, Javanese, Korean, Manchurian, Mongolian,
Nivkh, Persian, Sanskrit, Syrian, Tibetan, and Urdu. In his life, he achieved notoriety for his suc-
cess with languages. As a result, following his death, his brain became part of the collection of
the C. and O. Vogt Institute for Brain Research at the Heinrich Heine University in Düsseldorf.
In a recent research study, Krebs’s brain was compared with 11 others obtained from individu-
als without exceptional abilities (Amunts, Schleicher, & Zilles, 2004). The researchers found
that Krebs’s brain was architecturally different from the others in the areas of the left hemi-
sphere associated with language processing.
self-report measures of proficiency (Marian, Blumenfeld, & Kaushanskaya, 2007). Studies

have found that self-reported measures of proficiency are significantly correlated with
bilinguals’ performance on standardized language tests. For example, in a study of Spanish–
English bilinguals, Delgado, Guerrero, Goggin, and Ellis (1999) found that proficiency was
significantly correlated with participants’ scores on the Woodcock–Muñoz Language
Survey (Woodcock & Muñoz-Sandoval, 1993). Correlations were stronger for L1 than L2.
Proficiency in L2 speaking and understanding were not correlated with performance on
the language survey. Other studies have investigated the relationship between self-reported
proficiency and performance on language processing tasks. Bahrick, Hall, Goggin, Bahrick,
and Berger (1994) observed stronger correlations between proficiency and vocabulary
tasks than between proficiency and oral comprehension tasks.
Numerous studies have investigated factors that are related to L2 proficiency. One of the
strongest predictors is age of acquisition (Flege, MacKay, & Piske, 2002; Hyltenstam &
Abrahamsson, 2003; Johnson & Newport, 1989). As you learned in Chapter 1, since the
1960s, language researchers have recognized that childhood is the prime time for learning
language (Lenneberg, 1964, 1967). If language is learned after puberty, proficiency may not
be achieved. For bilinguals, those learning L2 early in life are likely to attain a higher level
of proficiency than those who begin learning L2 later in life (Johnson & Newport, 1989;
Kovelman, Baker, & Petitto, 2008).
Other studies have considered the length of time that L2 speakers have lived in the geo-
graphic region where their L2 is used predominantly (Birdsong, 2005; Espenshade & Fu,
1997; Flege, Yeni–Komshian, & Liu, 1999; Genesee, 1985; Stevens, 1999). For example,
Flege and colleagues (1999) investigated the relationship between how long bilinguals had
resided in the United States and their performance on grammatical tasks in English. They
found that performance was better for those who had been in the United States for the
longest period of time. For children who begin learning an L2 in school because the L2 is
the language of instruction, research suggests that most are able to become proficient in
the language in 5 years (Hakuta, Butler, & Witt, 2000). Conger (2009) conducted a longitu-
dinal study of schoolchildren in New York City for whom English was an L2. He found that
by the end of the first year of school, between a quarter and a third of students had become
proficient in English. After 3 years, half of the students had become proficient. The results
showed that children who entered school at younger ages reached proficiency faster than
children who had entered school at older ages. Proficiency was influenced by demographic
variables, such as sex, socioeconomic status (SES), and ethnicity.
Research by Hakuta, Bialystok, and Wiley (2003) suggests that aging may be a factor in
L2 proficiency. In their study, they used 1990 census data in the United States to investigate
the factors that were related to respondents’ self-reported L2 ability. The census data con-
tained responses from 2.3 million immigrants who spoke English as an L2 and Spanish or
Chinese as an L1. The results showed that several factors were related to respondents’
proficiency, including age of immigration, socioeconomic level, and amount of formal
education. They found that in both groups of speakers (i.e., L1 speakers of Spanish and L1
speakers of Chinese), as the years since their immigration to the United States increased,
proficiency in English declined. Stevens (2004) suggested that different results might be
observed using different sampling techniques and statistical procedures. Wiley, Bialystok,
and Hakuta (2005) reported the results of additional analyses, following the recommendations
of Stevens (2004). They observed results similar to those reported by Hakuta et al. (2003).
They concluded that the most plausible explanation for the steady decline in L2 profi-
ciency was aging, but other factors may also play a role (e.g., amount and quality of daily
experience with the language).
Over the past three decades, an impressive amount of research has investigated lan-
guage processing in bilingual individuals. In the past decade, researchers have begun to
report studies in which processing differences for bilinguals and monolinguals have been
observed. For example, in picture-naming tasks, research has shown that bilinguals are
generally slower than monolinguals, even when bilinguals named pictures using their L1
(Ivanova & Costa, 2008). Roberts, Garcia, Desrochers, and Hernandez (2002) found that
bilinguals made more errors in picture-naming than did monolinguals. Rosselli and col-
leagues (2000) found that on tasks of fluency in which one must produce as many words
as possible within a minute that belong to a specific semantic category (e.g., fruits or ani-
mals) or begin with a particular letter (i.e., s, a, or f), Spanish–English bilinguals did not
perform as well as monolinguals who were the same age. Rogers, Lister, Febo, Besing, and
Abrams (2006) observed that bilinguals’ ability to perceive words embedded in white noise
was poorer than monolinguals’. Bilinguals also experience TOT states more often than
monolinguals (Gollan & Acenas, 2004; Gollan & Silverberg, 2001).
A topic that has received relatively little attention is how bilinguals use both languages
in school and work. Sometimes, one may acquire information in one language but have to
use that information in an L2. There is preliminary research suggesting that the knowledge
that can be gained in one situation and applied in another is not dependent on the language
that is used to gain the information. This research is described in Text Box 7.3.
How Are Multiple Languages Stored in Memory?

Among researchers interested in bilingualism, there is still a debate regarding whether the
languages that a bilingual knows are stored in memory in a single, shared memory system
(Francis, 1999; Klein, Milner, Zatorre, Zhao, & Nikelski, 1999) or whether there are separate
memory systems, one for each language (Dehaene et al., 1997; Durgunoğlu & Roediger, 1987).
Numerous studies involving adult bilinguals have revealed how the knowledge about
words from multiple languages is organized in memory. Of particular interest is whether
words in the language being used are selectively accessed or whether all the words that
the bilingual knows are activated, regardless of the language being used. Numerous stud-
ies have shown that there are differences in how quickly bilinguals can retrieve L2 and
L1 words during language processing tasks. Some studies have observed differences
using the bilingual-translation task. Bilinguals view a series of words in one language and
are instructed to pronounce translation equivalents as rapidly as possible. Kroll and
Stewart (1994) have observed that participants can translate an L2 word into L1 faster
than they can translate an L1 word into L2. They proposed that memory links between
L2 and L1 words are stronger than links between L1 and L2 words because of the pro-
cesses involved in L2 learning. One more frequently associates an L2 word with an L1
word than vice versa.
Text Box 7.3 Research Discovery: Bilingual Problem Solving
In everyday life, bilinguals often acquire knowledge from one language, either via reading or
listening, and then apply the knowledge to a task in which another language is used (García,
2008). Very little is known about how knowledge transfer of this type occurs. Researchers
have begun to investigate knowledge transfer by bilinguals in studies involving problem
solving in which there is the opportunity for participants to apply a previously encountered
solution to a subsequent problem. The research utilizes the analogical transfer paradigm
(Gick & Holyoak, 1980, 1983; Holyoak & Koh, 1987; Holyoak & Thagard, 1989; Spellman &
Holyoak, 1992). When participants successfully apply a solution to a subsequent problem,
they are viewed as having drawn an analogy between the two problems and having trans-
ferred solution of the first problem to the subsequent problem. Thus far, there have been
only three bilingual studies using the analogical transfer paradigm (Bernardo, 1998; Francis,
1999; Fukumine & Kennison, 2011). The results of these studies indicate that bilinguals can
acquire knowledge from a source problem in one language and transfer it to solve a problem
in another language.
In a large study, Francis (1999) investigated how Spanish–English bilinguals solved prob-
lems that were written either in the same language or in different languages. She tested four
groups of participants. For two groups, the source problem and the target problem were in the
same language, either L1 or L2. For the other two groups, the source problem and the target
problem were in different languages. One group received the source problem in L1, and the tar-
get problem in L2. The other group received the source problem in L2 and the target problem in
L1. The results showed that analogical transfer occurred comparably often when the problems
were presented in the same language as when they were presented in different languages.
Across conditions, analogical transfer occurred most of the time (i.e., over 70%). The most
surprising result in the study was that participants’ language proficiencies were not strongly
related to problem-solving performance. Consequently, Francis concluded that the knowledge
that is transferred during problem solving across languages is stored in memory in a manner
that is language-free or language-neutral.
Francis’s (1999) results and those similar to them (Fukumine & Kennison, 2011) suggest
that in settings in which bilinguals are asked to acquire knowledge rapidly, there is no reason
to limit them to materials in one language. Educators and policymakers are likely to believe
that using materials prepared in different languages might create barriers to learning. The
research suggests the opposite. Once a bilingual acquires knowledge from a text or speech, it
appears to be stored in memory in a representation that is not linked to any language. When
the goal is acquiring knowledge, bilinguals’ performance is likely to be facilitated by allowing
them to access any useful materials, regardless of the language in which they are prepared.
Kroll and colleagues (Kroll & Stewart, 1994; Kroll, Van Hell, Tokowicz, & Green, 2010)
proposed the revised hierarchical model to describe the organization of bilinguals’ memory
for L1 and L2 words. Figure 7.1 displays the model, showing that conceptual representations
are connected to representations for L1 words and to representations for L2 words. The
representations for L1 words are connected to the representations for L2 words and vice
versa. The lines that connect the boxes indicate these connections. The darker the line in
the figure, the stronger the link is believed to be. Research by Bowers and Kennison (2011)
compared bilingual translation for words learned early in childhood and words learned after
the age of 8. They found translation from L2 to L1 was faster than L1 to L2 only for L1 words
that were learned early in childhood—perhaps because memory links are stronger for
words learned early in childhood than for words learned later in childhood.
Critics of the revised hierarchical model (Brysbaert & Duyck, 2010) believe that the Bilingual
Interactive Activation Plus (BIA+) model (Dijkstra & Van Heuven, 2002) emphasizes the inter-
connectedness of the bilinguals’ language knowledge, particularly when recognizing words.
Numerous studies have shown that when bilinguals recognize words, they retrieve and use
information about the words from both of their languages (Lam & Dijkstra, 2010). For example,
studies have observed evidence that Dutch–English bilinguals who view the letters work briefly
activate not only similar English words (e.g., word and cork): they also activate similar Dutch
words (e.g., werk, wolk, and worp). The studies require participants to hit a key on a keyboard
as quickly as possible when deciding whether a group of letters (e.g., work) is an actual word or
is a nonword (e.g., wark). Research with monolinguals has shown that the time that participants
take is related to the number of words that the person knows that look similar to the target
Figure 7.1 Kroll and Stewart’s revised hierarchical model proposes that bilingual
memory involves separate representations for L1 and L2 words. Both types
of representations can activate the meaning or conceptual representation.
The solid arrows indicate stronger memory links than the dotted arrows
L1 L2
Concepts
Source: Kroll and Stewart (1994).

word. When the target word is similar to

Table 7.2 Examples of Translation Equivalents many other known words of the same
(Noncognates), Cognates, and False length, participants respond more quickly
Cognates for German and English than when the target word is not similar to
other known words of the same length
German Word English Translation
(Andrews, 1989; Grainger, O’Regan, Jacobs,
& Segui, 1989; Grainger & Segui, 1990). In
Noncognates Zorn Anger studies with bilinguals, participants gener-
Brust Chest ally respond more quickly when a target
Hund Dog word is similar to a word in the other
Saft Juice language than when it is not.
The jury is still out regarding which of
Leute People
the two theories of bilingual memory best
Frage Question describes bilingual language processing.
Cognates Tiger Tiger Proponents of both theories have found
Steak Steak that there are some types of words that are
Garten Garden particularly useful in testing the models.
The types of words vary in terms of how
Freund Friend
similar they are to L1 and L2 words in terms
Gitarre Guitar of sound and memory. Most L2–L1 transla-
Papier Paper tion equivalents are completely dissimilar
False Cognates Gang Hallway in terms of pronunciation and written
form, as in the English word apple and the
Mantal Coat
Spanish equivalent manzana. In contrast,
Tag Day some L2 and L1 translation equivalents are
Kind Child highly similar, as in the English word map
Bad Bath and the Spanish equivalent mapa. Such
Teller Plate cases are referred to as cognates. Bilinguals—
particularly less proficient bilinguals—have
Source: Friel and Kennison (2001). been found to translate cognates more
quickly than noncognates (de Groot, 1992).
Research by Lotto and de Groot (1998) found that in an L2 learning task, participants learned
cognates faster than noncognates. A second type of word that has been particularly useful to
researchers has been called false cognates or false friends. Beauvillain and Grainger (1987)
showed that when bilinguals process a false cognate, they retrieve multiple meanings for the
word from L1 and L2. For example, when French–English bilinguals view coin, they activate
both the translation equivalent in English corner and the English word coin, which is related
to money. In order to better understand how words from different languages can be similar
in sound and meaning, refer to Table 7.2, which displays samples of translation equivalents
(noncognates), cognates, and false cognates for German and English.
Can One Forget One’s First Language?

Among middle-aged and older bilinguals, there are some who may feel that they are not
entirely proficient in any language. They may have acquired English and worked in an
English-speaking environment for many years. As a result, they may have grown unaccus-
tomed to speaking their native language if it is not used daily in their home. Without the
opportunity to speak the native language regularly, one may become rusty at speaking the
native language. One example of this was reported by Isurin (2000); a native speaker of
Russian was adopted by English-speaking parents at the age of 9. After that point, the child
did not speak Russian. After 1 year, the child’s vocabulary knowledge decreased by 20%. It
is common for children who are adopted before the age of 9 and do not use their native
language in their new homes to have little or no memory of their L1 (e.g., Pallier et al., 2003).
Several studies have found that bilinguals can experience some loss of their L1 over time (de
Bot, 1999; Levy, McVeigh, Marful, & Anderson, 2007; Linck, Kroll, & Sunderman, 2009; Seliger
& Vago, 1991). Linck and colleagues (2009) investigated the language performance of Spanish–
English bilinguals in Spain. One group of bilinguals was learning Spanish in an immersion
experience. The other group of bilinguals was learning Spanish in a classroom setting.
Participants’ comprehension and production abilities were measured in L2. Those participants
learning L2 through immersion outperformed those learning in a classroom. The results also
showed that L1 performance was worse for participants learning through immersion than for
those learning in a classroom. The authors concluded that the learning of L2 inhibited L1.
A recent study by Levy and colleagues (2007) reached similar conclusions in a study with
Spanish–English bilinguals. They utilized a procedure known as retrieval-induced forget-
ting (RIF) (Anderson, Bjork, & Bjork, 1994). In the typical RIF experiment, participants were
instructed to study category-exemplar pairs (e.g., fruits-apple, fruits-pear, drinks-whiskey).
After the pairs were learned, they were then instructed to practice remembering half of the
items (i.e., every other pair of items in the previous list). Last, they were instructed to recall
all of the words in the word pairs. Before performing the RIF task, they were asked to name
pictures either in Spanish (L2) or in English (L1). The number of picture-naming trials was
varied. The results showed that performance in the memory task in English was worse
when participants had previously named pictures in Spanish. Participants who struggled
the most with Spanish pronunciation showed the most interference in English on the
memory task. Their third experiment showed that the sounds of L2 words, rather than the
meaning of the L2 words, interfered with L1 words.
Summary and Theoretical Implications

Children who are reared in homes in which multiple languages are spoken develop lan-
guage along the same time frame as children who are reared in homes in which only one
language is spoken. Research shows that the perceptual abilities of bilingual and monolin-
gual infants may differ early in childhood; however, it is unclear whether these differences
impact later language development. Some studies have shown that bilingual children may
have smaller vocabularies than monolingual children. Related research shows that other
factors influence vocabulary size, such as sex, SES, and whether the L2 is a culturally
dominant or minority language. An increasing number of studies have shown that bilingual
children outperform monolingual children in tasks requiring attentional processing.
There are many methods that have been developed to teach L2s. Few studies have com-
pared how well learners fare using the different methods. When languages share vocabulary
and grammatical rules, learning is easier than when languages do not share vocabulary and
grammatical rules. Researchers debate about the extent to which the words in bilinguals’
two languages are stored together or separate in memory. In language processing studies
involving adult bilinguals, they tend to be slower and more error prone than monolinguals.
Language studies have shown that the L1 is vulnerable to interference from the L2.
The four theoretical approaches to language development each have potential to account
for some aspects of L2 acquisition and bilingualism. The behaviorist approach predicts that
learning an L2 would be facilitated through the mechanisms of classical and operant condi-
tioning. The social-interactionist approach predicts that learning an L2 would be facilitated
when the learning occurs in a supportive social environment, which is characteristic of a total
immersion language learning environment. The statistical learning approach is likely to have
compelling explanations for language transfer effects. Last, because of the generative
approach’s claim that aspects of language learning are innate, the approach is compatible
with the evidence that there is a critical period for language acquisition. With the current
state of knowledge on the topic of L2 acquisition, we are not able to judge any view superior
to any other. As research continues to be conducted, it may be possible to determine whether
any of these approaches can completely account for how L2s are learned and processed.
KEY TERMS
age of acquisition false cognates negative language transfer

balanced bilinguals first language (L1) polyglot
bilingual education free morpheme constraint positive language transfer
bilinguals fusion hypothesis second language (L2)
code-switching grammar-translation method sequential bilingualism
cognates immersion programs simultaneous bilingualism
direct method language transfer translation equivalents
dominant language monolingual trilingual
REVIEW QUESTIONS
1. What is the difference between simultaneous and sequential bilingualism?

2. In terms of perceptual abilities, how do bilingual and monolingual infants differ?
3. In terms of vocabulary size, how do bilingual and monolingual children differ?
4. What factors are related to vocabulary size in bilingual children?
5. What is code-switching? What evidence is there that code-switching involves grammatical rules?
6. What are the cognitive benefits for children who know more than one language?
7. What evidence is there that attentional processing differs for bilingual and monolingual children?
8. What differences have been observed in brain-imaging studies between bilingual and mono-
lingual brains?
9. What area of the brain has been identified as involved when a bilingual switches from one
language to another?
10. What is the evidence that one’s L2 interferes with one’s L1?
11. What is language transfer? What is the difference between positive language transfer and
negative language transfer?
12. How does the performance of adult bilinguals and monolinguals on language processing
tasks differ?
13. What are the most common ways in which people learn L2s and L3s? Has research shown
that one or more methods are more effective than others?
14. What has research shown about the effectiveness of immersion programs as compared with
other programs?
15. How did bilingual education come about in the United States in the 1960s and 1970s?
16. What is a polyglot? How did the brain of one famous polyglot, Emil Krebs, differ from the
brains of non-polyglots?
17. What is a cognate? What has research shown about how bilinguals process this type of word?
18. What is a false cognate? What has research shown about how bilinguals process this type
of word?
19. What evidence is there that bilinguals activate words from both languages when they are
recognizing words?
20. What are the two competing models of bilingual memory? How do they differ?
RECOMMENDED READING
Altarriba, J., & Heredia, R. R. (2008). Introduction to bilingualism. New York: Psychology Press.
Bialystok, E., & Hakuta, K. (1995). In other words: The science and psychology of second language acquisition.
New York: Basic Books.
Gardner-Chloros, P. (2007). Codeswitching. Cambridge, UK: Cambridge University Press.
Grosjean, F. (2010). Bilingual: Life and reality. Cambridge, MA: Harvard University Press.
Hakuta, K. (1997). The mirror of language: The debate on bilingualism. New York: Basic Books.
Kroll, J., & de Groot, A. (2009). Handbook of bilingualism. Oxford, UK: Oxford University Press.
Myers-Scotten, C. (2005). Multiple voices: An introduction to bilingualism. New York: John Wiley.
RECOMMENDED FILMS
ACTION
Jeffcoat, J. (Director). (2006). Outsourced [Motion picture]. United States: ShadowCatcher.
Nava, G. (Director). (2008). El norte [Motion picture]. United States: Criterion Collection.
Pakula, A. (Director). (1982). Sophie’s choice [Motion picture]. United States: Universal Pictures.
Schrieber, L. (Director). (2006). Everything’s illuminated [Motion picture]. United States: Warner Home Video.
Wang, W. (Director). (1982). Chan is missing [Motion picture]. United States: Koch Lorber Films.
SUGGESTIONS FOR CLASS PROJECTS
1. Investigate public opinion about bilingual education. Work alone or in groups to develop a
brief questionnaire. Questions may relate to the public’s understanding of the federal law that
mandates bilingual education, the ways in which local school districts implement bilingual educa-
tion, and attitudes about the advantages and disadvantages of bilingual education. Determine
whether respondents’ opinions are changed after they learn about the recent research demonstrat-
ing that children who acquire more than one language experience cognitive benefits.
2. Survey your peers in other classes about their views on whether an infant should be raised
to be bilingual from birth or whether parents of an infant who speak different languages should
pick a single language of the household to use with the infant. The questionnaire can be designed
to assess others’ opinions about which approach they would use if placed in that situation and also
to assess why they believe that option to be the best. The reasons that respondents give may reveal
beliefs about bilingualism that may or may not be supported by empirical research. Discuss either
in a class presentation or in a paper what commonly held views are or are not supported by
empirical research.
3. Survey up to five peers about their experiences trying to learn an L2. Be sure to ask respon-
dents their age when they began trying to learn the L2; whether they learned in a formal classroom
setting or in a naturalistic, immersion setting; and the level of proficiency that they achieved in the
L2. Share your findings with the entire class. The class can then analyze the group results, specu-
lating about the role of age of acquisition and method of acquisition in success in learning an L2.
For additional ancillary resources, please visit the companion website at www.sagepub.com/
kennison.
o Video Links o Web Resources o Flashcards

o Audio Links o Internet Activities o Web Quizzes
View publication stats

Introduction To Language Development: July 2013

Uploaded by

Copyright:

Available Formats

Introduction To Language Development: July 2013

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Introduction To Language Development: July 2013

Uploaded by

Copyright:

Available Formats

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

Introduction to Language Development

Book · July 2013

Parents and Children View project

Tolerance View project

The user has requested enhancement of the downloaded file.

LIFE WITH MORE THAN

BILINGUAL CHILDREN LEARNING SECOND LANGUAGES

2000). In the United States, it is expected

Country Population Official Languages

Country Population Official Languages

Source: Adapted from The World Factbook (2009).

How Do Bilingual Children Differ From Monolingual Children?

Studies of children’s receptive vocabu-

1. a. Me iban a lay off. They were going to lay me off’.

2. a. Quiero un green tomate.

Code-switching is also more likely to be used in some situations than in others

Text Box 7.1 Diversity of Human Languages: Pidgins and Creoles

Pidgin Hawaiian Creole English

. . . water pipe no more. . . . is] inside house nowadays, ah?

Are There Cognitive Benefits When Children Are Bilingual?

What Brain Changes Are Associated With Bilingualism?

A number of recent studies have found

Learning Second Languages

What Teaching Methods Are Used in Language Classes?

includes grammar exercises. It is based on the principles of behaviorism. Students are

Do All Countries Have Bilingual Education?

How Does First Language Influence Second Language Acquisition?

3. a. I know when she arrives today. English

Life With Two or More Languages

What Factors Predict Proficiency?

Text Box 7.2 Extraordinary Individuals:

Emil Krebs (1867–1930) was born in Germany,

self-report measures of proficiency (Marian, Blumenfeld, & Kaushanskaya, 2007). Studies

How Are Multiple Languages Stored in Memory?

Text Box 7.3 Research Discovery: Bilingual Problem Solving

Source: Kroll and Stewart (1994).

word. When the target word is similar to

Can One Forget One’s First Language?

Summary and Theoretical Implications

age of acquisition false cognates negative language transfer

1. What is the difference between simultaneous and sequential bilingualism?

SUGGESTIONS FOR CLASS PROJECTS

o Video Links o Web Resources o Flashcards

View publication stats

You might also like