CHAPTER 4

Language Development In

Simultaneous· Bilingual Children

"fi'
",$~
•

his chapter discusses the language development of simultaneous bilingual

T children-those who acqUIre two languages from birth or at least before the age
of3-from infancy to the school-age years. James, Bistra , and Gabriela, profiled in
Chapter 1, are all simultaneous bilingual children. The contexts in which James, Bistra,
and Gabriela are growing up are the most common , but they are not the only routes to
simultaneous bilingualism . Some parents expose their children to another language from
birth via a full-time caregiver who speaks another language. An example of this is an
English-speaking family living in Los Angeles who hires a live-in, Spanish-speaking
nanny when their child is an infant (see Chapter 1). If the child's exposure to Spanish is
fi'equent and sustained over the preschool and school-age years, she could be considered
a simultaneous bilingual, even though the child's parents do not speak Spanish. Children .
of parents who travel or live outside their home country for business or personal reasons
might also experience simultaneous bilingual acquisition through child care arrange­
ments. Finally, it is important to keep in mind that we have set the distinction between
simultaneous bilingualism and second language acquisition at 3 years of age in this book,
but this distinction is not a clear-cut one. Some of the information on school-age bilingual
children in this chapter might be applicable to children who begin to leam a second
language shortly after age 3. Conversely, some of the information in Chapter 6 on second
language learning could be applicable to children who begin to learn an additional lan­
guage at age 2){
Our primary goal in this chapter is to provide information about what is typical in
the language development of simultaneous bilingual children during the preschool and
60 Understanding Bilingual and Second Language DeI!elopmel1l Language Development il1 Simultoneo/·/s Bilingual Children 61

elementary school yea rs. Regardless of whether simultaneo us bilingual children are
destined to become bilingual adults, their parents, educators, speech-language patholo­ BOX 4.1
gists, and o ther health practitio ners sho uld be aware of the impact of simultaneous dual Raising Children Bilingually: Questions from Parents
language exposure on children's language development in the preschool and early
sc hool years. In particular, they need to set appropriate expectations for these children's Dear Dr. Genesee,
oral language abilities in order to gauge whe th er children's lan guage development
We hope you can shed some light on a question about infant bilingualism. My
appears to be typical or shows signs of delaylimpairment. We often use monolingual
Russian wife and (hope to raise our soon-to-be-adopted 70-week-old infant
children as a refere nce point in our descriptions , and we highlight th e similarities and
bilingually. (The infant comes from Georgia, a former Russian republic.) My
differences between typical bilingual children and typical monolingual children in
wife insists on the ii-nportance of almost exclusively speaking Russian. I was
order to permit better identification of aspects of bilingual del1elopment that truly
born and raised in C;alifornia and have only begun to learn Russian. I would
should be a cause for concern. like my wife mostly to speak English to 01,Jr infant when I am around. My ques­
The information in this ch ap ter is organized according to the following
tion: Willmy wife's speakingRussian almost exclusively increase the chances
questions: Df our child becoming bilingual? I suspect there are both cognitive and affec­
1. Do children exposed to two languages from birth learn bilingually at first' That is, do
tive factors .
they have one .single language system or two separate language systems'
. rhank you for any assistance,

2. .If bili ngual children have two separate language systems, do the languages interact in
A father

development' Does this interaction cause bilingual children to show some unique

patterns from monolingual childrell'

Dear Dr. Paradis,
3 Are bilingual children slower to learn language than monolingual children, and if so,
We live in Western Canada and our son is in a French day care, where there is
how much slower' Do bilingual children show similar rates of development in both
very strong encouragement to speak Ftench in the home in order to maintain a
of their languages'
French-speaking environme~t at the day care. Because of this, we committed to
4. Are multilingual children distin ct from bilingual children' Is learning three or four speaking French as much as possible, even though my French is at an intermedi­
languages early on too burdensome or confusing for children' ate le vel. However; now that he is 2, I have begun to question whether my
Although there may not be as many differences between bilingu al and monolingual French language abilities have been Constraining him, and more importantly,
language development as one might expe ct, the previous questions involve comparisons. feeling like I cannot fully express myself in my mother tongue to him-which I
Therefore, before we begin to answer this list of qu estions, we briefly discuss alternative feel will constrain our language communication and relationship development. I
ways of viewing differences between mon olingual children and bilingual children. His­ have now decided to speak English, though I recognize the challenges this poses
torically, bilingualism in yo ung children was thought to put them at an intellectual disad­ in being a parI of a French day care, and later school. Perhaps you could suggest
vantage, which co ncerned educational policy makers (see Chapter 3). Beca use of this some helpful resources that will coniinue to help me understand what is best for
historical attitude, differences between bilingual and monolingual language development my son in a mixed-language family. .
are often viewed negatively. Sincerely,
Bilingual children have often been considered typically developing only if they A mother
appear to be similar to monolingual children; they are considered to be not developing
typically or to have disabilities if they show any differences. This kind of attitude results Parents often ask us what the best way is to raise ch ildren bilingually. Although
when monolingual children's development is taken as the norm, even though, as discussed many different methods can work; the one. that peop le use the most is the one par­
in Chapter 1, childhood bilingualism is most likely as common worldwide as monolin­ ent-one language method. In this method, each parent speaks his or her native lan­
gualism . Given how widespread bilingu alism in childhood is, we cannot help but guage exclusively with the child (see Chapter 5 for more about this method). This
ac knowledge that the human mind is just as capable of bilingual development as mono­ en~ures that the child will, get sufficient and ri ch input in both languages. This is
li ngua l development, even if there are some differences (see Chapter 3). Any differences especia lly important whenorie language is a minority language and spoken very lit­
between the two should not be taken inunediately to imply that bilingualism has perni­ tle outside the home. It is also important for parents to fee l able to fully express
cious effects on language development as a whole. We believe th at a m ore appropriate themselves when speaking with their children.
attitude is that there is more than one path to acquiring language, and that one of these
paths is to acquire two languages at a time.
r­
62 Understanding Bilinglwl al1d Second Lal1gllage DevelopJllellt Langllage Development ill Sil1111 ilaneolls Bilingual Chilrlren 63

DO BILINGUAL CHILDREN

HAVE ONE LANGUAGE SYSTEM OR TWO?

Whe n an infant is experiencing input from two languages, the infant is not consciously
aware that he is learning two langu ages, and it is not until a bilingual child is mu ch
older-perhaps closer to 3 years of age-that he demonstrates explicit awa re ness ofbein g

CU

a dual language learner by talking directly about it . It is therefore not surprising that hien
researchers have asked whether children exposed to two languages acquire those lan­ JUS

guages bilingually from the outset. In other words, do they represent the language input
they hear around them in their minds as a single or duallangua.ge system at first? From f!})
the begi nning of the 20th century until the early 1990s, researchers tended to support the
idea that bilingual children do not acquire language bilingually at first . An influential
~
model oft~i s , the Unitary Language System Hypothesis (see Figure 4.1), was put Figure 4.2. 'Dualla.nguage syst~ms in early bil ingualism.

forth by Virginia Volterra and Traute Taeschner (1978). They proposed that bilingual chil­
dren begin the acquisition process with a single language system that combines the words unitary vocab ulary and grammar, then one would expe ct them to frequ ently mix words
and the grammatical rules fi'om their dual language input. At the next stage, the words dif­ and phrases from both languages together, regardless of language context or conversa­
ferentiate into two vocabularies/ lexicons, but the system of grammatical rules remains the tion partners. One might also expect them to use 'grammar rules from one language
same for both languages. In the final stage, around 3 years of age, th e system of grammat­ with words from another and even blend the rules from both languages together. Fur­
ical rules becomes differentiated, and the bilingual child can be said to have separate lin­ thermore, it· is possible that the process of differentiating two langua ge systems could be
guistic systems, as bilingual adulrs- do. An alternative view to Volterra and Taeschner's cognitively costly for a bilingual cllild, causing a slowdown in th @ir language develop­
mode.! is one put forth by Fred Genesee (1989), the Dual Language System Hypoth­ ment between the ages of 2-3 yea rs and uniqu e patterns of language use during [his
esis (see Figure 4.2) , which assumes that childre n exposed to two languages from birth reorganization. However, current' research on bilingual acquisition overwhelmingly
establish two separate linguistic systems from the outset of acquisition. According to this supports the Dual Language System Hypo thesis. We revIew this research herein with
resp ec t to several domains oflanguage.
view, children with simultaneous du al language exposure never go through a noticeable
period in which their linguistic representations are unified and later separa te from one
into two systems . Speech Perception
Both th e Unitary Language System and the Dual Langu age System Hypotheses
[n Chapter 3, an infan t's fundamental capacity to cope with du al language input in the
refer to the nature of the language representation in a child 's mind, so it is releva nt to
first year of life was disc ussed. This discussion included evidence that infants who will
ask whether this is a meaningful distinction for understanding the child's language
be bilingual can hear the difference between their ·two languages early on and go
behavior, especially for the practical concerns of parents, educators, and health care
through a developmental reorganization in their perception of phonetic segments, such
practitioners. I t is meaningful, because these two viewpoints make very different pre­
as consonants and vowels , at the same time as infants exposed to just one langu age. Not
dictions about how yo ung bilingual children will use their languages and what devel­
only does the timing of this developmental reorganization support the view that mfants
opmental stages they will go through. If yo ung bilingual children initially have a
are born with the capacity to acquire more than oue language early on, but also the
details of this process provide evidence in fa vor of the Dual Language System Hypo[h­
esis. Two teams of Canadian researchers examined the speech percep tion abilities of
infants exposed to French and English from birth. Burns, Yoshida, Hill, and Werker
(2007) tested these infants on their abilities to perceive subtle acoustic differences in
how [p) and [b) are pronounced in French and English. Burns and colleagues found
Juice, Chien,) that bilingual infants were able to perceive langu age-specifIC aco ustic properties of [p)
IlUck,' jus . and [b) when they were between 10 and 12 months of age-the same age when mono­
~ I lingual infan ts become langmge-specifi c listeners. French and English also have subtle

I differences in the acoustic properties of the phonetic segment [d), and Sundara, Polka,

W and Molnar (2008) found that infants exp osed to both French and English were able to
perceive [d) in the language-specific English way at age 10-12 months, but French
monolingual infants could not. The results of both studies suggest that infants exposed
Figure 4.1 . Unitary language system in earl y bilingualism. to two languages are establishing separa te perceptual systems for each language, rather
r

64 Understanding Bilil1gHal and Second Language Development Language Development in Simultalleous Bilingual Children 65

than a unified perceptual system for both, toward the end of the first year of life. How­ labeled. When children start to use translation equivalents, this is considered evidence for
ever, there are further nuances and some conflicting findings in the research on bilin­ t\vo vocabularies, and hence for two language systems. Although the two girls studied by
gual speech perception in infants; see Werker~Byers-H einlein, and Fennel (2009) for a Volterra andTaeschner (1978) may have had few translation equivalents, they appear to be
comprehensive technical review. anomalous among the many other bilingual children studied. Researchers examining
Spanish-English, Portuguese-English, and French-English bilingual.~)1ildren have all
found translation equivalents from the earliest stages, even before children have vocabu­
Phonology laries of 50 words (Nicoladis & Genesee, 1996; Nicoladis & Secco, 2000; Pearson ,Fernandez,
Some researchers have presented evidence to support the Unitary Language System & Oller, 1995; Quay, 1995). In a study examining 27 Spanish-English bilingual children
Hypothesis with respect to the sound system of children's spoken language. For exam­ in the Miami area, Barbara Pearson and her colleagues found that on average, 30% of
ple, Celce-Murcia (1978) noted that the Frencfi-English bilingual child she studied bilingu al toddle rs' early vocabularies are translation equivalen ts (Pearson et ai., 1995).
had the same phonological substitution processes in both languages . Phonologica l Thus, there is compell ing empirical evidence to suggest that bilingual children establish
substitution processes occur when a child uses an easier-to-pronounce phonetic seg­ two vocabularies from the onset of acquisition.
ment instead of the correct segment. For example, many young children substitute a It is worthwhile to ask why bilingual children would have only 30% translation
vowel for an [I] at the ends of words so that the word ball would be pronounced equivalents in their vocabularies. Why don't bilingual children have a larger number of
"ba-o" or "ba-u ."This kind of evidence for a unitary language system is problematic, translation equivalents? The answer to this question becomes clear when we consider
be ca use these substitution processes are very common across children and languages: how words are learned. Children learn new words as they interact with the world around
Therefore, two monolingual French- and English-speaking children the same age as them while hearing speech. Children do not duplicate every experience in both lan­
Celce-Murcia's bilingual child co ul d easily have displayed identical phonological guages. A bilingual child may acquire ce rtain words duri ng an activity with her Spanish­
substitutions in thei~ speech. speaking grandn!.other and not learn the English equ.ivalents for some time, because she
Given that the early phonological system of children is so similar cross-linguistically, ·does not engage in that ac tivity with English-speaking playmates. If English is spoken by
how can we tell whether bilingual children have one or two phonologies? Johnson and the child mainly while at child care, then she will learn many English words related to
Lancaster (1998) examined the use of phonetic segments exclusive to either Norwegian that environment, and there may be many home-related vocabulary items lacking in her
or English phonology in the word productions of a Norwegian-English toddler younger English vocabulary. Even fluent bilingual adu lts have gaps in their vocabularies due to
than 2 years of age. Even at this early age, the production and distribution of these soun ds experiential factors. Therefore, bilingu al individuals- particularly bilingual children­
il~ the boy's words showed that he was building separate sound segment inventories for his should not be expected to have a translation equivalent for absolutely every word. Pearson
two langLlages. Johanne Paradis (2001) used an elicitation task to determine whether and her research group in Miamj looked at how the proportion of singlets, or words that
French-English bilingual children age 2)1 had separate phonological systems in produc­ do not have a translation equivalent, changes over' time in Spanish-English bilingual
tion. Children were asked to repeat four-syllable-long nonsense (made up) words in both child ren's vocabularies. The percentage of singlets in their dual language vocabularies was
languages. Children this age often omit· syllables when they repeat long words, but which 50% in fir st grade (age 6) and declined to 30% by fifth grade (age 11) . Bilingual students
syllables in the words are omitted differs between French and English, because the lan­ attending college still had an average of 10% singlets (pearson, 1998). In other wo rds,
guages differ in their rules for syllable structure.The bilingual children omitted syllables in these bilingual children acquired more and more translation equivalen ts as their language
French and English differently, ·showing similar patter ns to their monolingual peers. This development and experiences expanded, but they never reached a point at which they
result would not have occurred if they had had a unitary phonological system for both had 100% translation equivalents.
languages.
Morphosyntax
Vocabulary Morphosyntax refers to two elements of language tha t combine to form grammar:
On~ of Volterra and Taeschner's (1978) sources of evidence for a unified vocabulary in morphology and syntax. Morphology refers to freestanding function words such as
production was the initial absence of translation equivalents in bilingual children's articles (the fa), auxiliary verbs (she is. going, ~ she like milk?), prepositions (to, at, in, on,
productive vocabularies. A translation equivalent is a word that has the same meaning in etc.), or negative markers (she does not like milk), and to inflections that attach to words
two lan guages, such as zapatos in Spanish and shoes in English. The two girls studied by such as nouns and verbs in the form of suffixes or prefixes, as in -s in he walk-s, the
Volterra and Taeschner had few pairs of translation equivalent words in their combined cat-s, or M0111my- ~ sock, or -ed for the past tense, as in walk-ed, help-ed. Syntax mainly
vocabularies. These researchers suggested that bilingual child ren learn lexical labels for refers to the order of function and content words in a clause or sentence. Volterra and
concepts on a one-to-one basis and avoid learning words with the same meaning. For Taeschner (1978) claimed that the bilingual children they studied had a unified
example, if a child exposed to Spanish and English acquired the word zapatos, she would syntactic rule system for their two languages until they were 3 years old. Once the
not have shoes in her vocab ulary right away, because this concept had already been children appeared to have differentiated vocabularies (i.e., when translation equivalents
f~
lF ,

66 Undmtanding Bilingual and Second LaNguage Development tal/if/age DeiJelopmenl ill Sillluitaneolls Bilingual Childrm 67

emerged) , they sti ll seemed to use the same syntactic rules to make sentences in both amo!1g researchers that bilingual children have two language systems fi'om the onset of
languages, eyen when this resulted in erroneo us sentences. For example, one child their development. In other words, researchers believe that the human language faculty is
formulated most of her negative sentences in both Italian and German-by putting a pel{ectli capable of sorting out dual language input and establishing nvo linguistic sys­
negative marker (such as not in English) after the verb. This word order corresponds tems, and there does not appear to be a stage unique to bilingual development in which
roughly to German syntax but not to Italian syntax, in which the negative marker a child's language system undergoes differentiation. Additional evidenc~Jor separate lin­
comes before the verb. guistic systems, presented later in Chapter 5, comes from young bilingual children's ability
Since Volterra and Taeschner's study, a wealth of research has emerged that shows that to choose which language to use based on the context of the c.onversation and to whom
bilingual children can havc separate morphosyntactic systems-in some cases, even from .they are talking-an ability difficult to explain with the Unitary Language System
the beginning of their first word combinations. For example, similar to German and Hypothesis. Therefore, parents, early childhood educators, and speech-language patholo­
Italian, French and English differ ill terms of the placement of the negative marker. In gists can -expect to see bilingual children developing separate language systems and to go
French, the ncgative marker pas comes after the main verb, but in Engli5h the negative through the same kinds of stages that they have seen in monolingual children, on the
marker 110t comes in between the main verb and an auxiliary verb such as do, can, or is. whole. This point was also made in Chapter 3 with respect to early milestones in language
Compare thIS French sentence with its English translation in which tbe negative markers development. Does this mean that bilingu.ai children are "two monolingual children in
and main verbs are underlined: Ie hebe l1e bait pas Ie laitl the bab)' is not drinking the milk. If one" in their language development? Not entire ly, and in the following two sections on
French-English bilingual children had a unitary syntactic system, they might go through interactions between languages and on rate of development, we discuss the ways in which
a stage in which they adopt one of these rules for both languages, so they might say Ie hebe bilingual children can sometimes appear different fi'om monolingual children.
ne pas hoitle lait in French, or the bab)' is diinking no! the milk in English-neithfr of which
is a grammatically well-formed sentence. In language samples of 15 hench-English bilin­
DO THE LANGUAGE SYSTEMS OF BILINGUAL
gual children ages 2-4 years, Paradis, Nicoladi" and Genesee (2000) found only spo­
radic use of the wrong order fot' the placement of the negative marker in either language.
CHILDREN INTERACT DURING DEVELOPMENT?
There was no evidence from any child for a stage in which a unitary rule for negative The majority of researchers agree that bilingual children have dual language systems from
marker placement existed in their grammars. early on, even at the onset of language acquisition, but this does not necessarily mean that
This example of differentiated systems with respect to the syntax of negative markers bilingual children's two languages are hermetically sealed and utterly autonomOllS in
holds true for other aspects of morphosyntax and for other bilingual children. The results developmen t. As researchers have mewed beyond the Unitary versus Dual Language Sys­
of research projects that have followed groups of children acquirll1g the same language tem debate, they have been engaged in understanding the degrees of interconnectivity and
pair over time, such as the Montreal-McGill University project in Canada (e.g., Paradis & separation between bilingual children's two developing languages. Research to date shows
Genesee, 1996, 1997; Paradis et aI., 2000) and the DUFDE project in Hamburg, Germany that bilingual children show' some cross-linguistic influence between their two devel­
(e.g., Meisel, 1989, 1994), overwhelmingly support the claim that bilinguai children oping languages in all linguistic domains that have been studied. To introduce the concept
demonstrate the same language-specific patterns and stages in thei r morphosyntactic of cross-linguistic infl~lence, let's borrow from our previous discussion 'lbout negative
development as their monolingual peers, and thus demonstrate that they have differenti­ sentences in French and English. Suppose a French-English bilingLlal child produced a
ated morphosyntactic sys tems. sentence in English such as "the baby drin.k 110t the milk," in which the 1101 is placed after
Another source of evidence for the Dual Language System Hypothesis that straddles the main verb, as it would bein prench This sentence appears.to have some influence fi"Om
both lexical and morphosyntactic development comes from studies examining the associa- . French syntax, although it does not include French words as does a code-mIXed sentence
tions between vocabulary size, vocabulary composition (e.g., proportion of nouns,verbs, (see Chapter 5). If we see these kll1ds of sentences used often for a certain period of time
function words), and utterance-that is, sentence--length in bilingual children's speech in by a bilingual child, we assume that the child has a rule in his English grammar that allows
both languages. Barbara Conboy, Virginia Marchman, and colleagues used a standardized for these kinds of sentences, and this rule reflects the rule in his French grammar. It is pos­
parent report questionnaire to measure young bilingual todcllers' lexical and morphosyntac­ sible, then, that this English rule is the product of cross-language contact during develop­
tic development in each language (Conboy & Thal, 2006; Marchman, Sussman-Martinez, & ment. As mentioned previously, we have found very few cross-linguistic structures with
Dale, 2004). Marchman and colleagues also compared their parent report findi ngs with direct negation in our corpus of 15 French-English bilingual children. So these children not only
observations oflanguage abilities in a subset of the children. Both groups ofresearchers found lacked a unified system for negation, but did not seem to have much cross-linguistic influ­
that within each language, there were close connections among vocabulary size, vocabulary ence in this part of the grammar, either. However, cross-linguistic influence has been found
composition, and utterance length, but these connections were weak across languages.These in numerous other studies of simultaneous bilingual children. Our discussion of cross­
findings are consistent with the view that dual language learners are establishing separate lin­ linguistic influence will be limited to those aspects }\Ie believe would be most important
guistic systems, vocabulary and grammar combined, from an early age. for parents, educators, and health care professionals to be aware of, but the research on this
To summarize, looking across evidence from speech perception in infmcy, early topic is complex and growing. See Nicoladis and Genesee (2007), Paradis (2007), and Yip
phonological production, vocabulary building, and l110rphosyntax, there is a consensus and Matthews (2007) for comprehensive technical reviews.
 68 UndWlm1di11g Bilingual and Second Lang/lage Development Language Developmet// in SimllllaNeoHs Bilil/gllal Children 69

One of the most salient types of cross-linguistic influence is the substitution of a bilingual. children , ages 6- 10 years, along with monolingual peers. In one study, they
word order rule from one language to another.Virginia Yip and Stephen Matthews have asked children to judge whether certain plural noun phrases were grammatical. The plu­
documented this in the speech of Cantonese- Engliili bilingual children (Yip & ral noun phrases had specific meaning, such as "the...dogs are in the park," and generic
Matthews, 2007). For example, relative clauses in Cantonese are placed before the noun meaning, such as "dogs are loyal animals"-the morphosyntactic structures for such noun
tlley modify, but in English they are placed after it. Yip and Matthews recorded several phrases differ between Italian and English (Serratrice, Sorace, Filiaci, &, J3aldo, 2009). The
instances of Cantonese relative clause order in Cantonese-English children's English for a bilingual children showed some cross-linguistic influence in their performance on the
period approximately between the ages of2~ and 4 years, when the children were more task, even in the oldest group, because they sometimes accepted noun phrases as gram­
proficient in Cantonese. For example, one child asked, ''Where's the Santa Claus give me matical in one language when those noun phrases actually had the morphosyntactic
the gun'" instead of the target English structure "Where's the gun that Santa ClatH give structure of the other language. It is unknown whether these bilingual children would
(gave) me'" (Yip & Matthews, 2007, p. 155). Susanne Dapke (1998,2000) followed a ever produce a noun phrase with the wrong morphosymactic structure in their everyday
group of German-English bilingual children in Australia over time, looking at word order speech, however. Research such as this suggests that interactions between dual language
in their sentences in both languages. She found that a certain proportion of the children's systems might be a permanent feature of being bilingual, and that some subtle differences
GennaH sentences had errors that mir~ored the word order of English during the pre­ between monolingual children and bilingual children in language representation extend
school years. Elena Nicoladis has found cross-linguistic infl uence in French-English across the lifespan.
children's use of compounds; namely, combinations Qf individual words that form other Two main explanations have been put forward to explain when cross-linguistic influ­
words, such as brosse a dents "toothbrush" or taille -crayon "pencil sharpener" (Nicoladis, encecan be expected in bilingual children's speech. The first explanation has to do with
2002,2003). Notice that the ordtlr of the words in the compounds is the opposite how the rules for a structure might overlap across languages. To take the example of null
. between French and English-brosse a dents tra.nslates as bwsh-teeth, word for word. subjects in Spanish, notice that although Spanish allows subjects to be null or overt, Eng­
Nicoladis found that French~nglish bilingu~ 1 children are much more likely than lish allows only overt subje~ts. The presence of overlap in overt s.ubjects between the two
monolingual children to reverse the word order in their compounds in both languages. languages and the optional use of null or overt subjects in Spanish might "invite" influ­
Foroodi-Nejad and Paradis (2009) found similar cross-linguistic ino.uence in the com­ ence fr9m English to Spanish, but the other way around is less likely, because English has
pound production of Persian-English bilingual 3-year-olds, so the phenomenon is not just one option. The second explanation that has been put forward is language domi/'/a/1ce, a
particular to the combination of French and English. concept discussed in detail in the next section. If children are more proficient in one lan­
Some forms of cross-linguistic influence do not result in outright errors in speech, guage than the other, cross-linguistic ino.uence might take place from the dominant to
but instead have less obvious effects. Paradis and Navarro (2003) looked at the use of sen­ the non dominant language, regardless of overlap and optionality. For example, there is
tential subjects over time in a Spanish-English bilingual girl ages 1)1-2~ years old, and zero overlap in the word order in relative clauses between English and Cantonese, and no
compared their observations to those of Spanish monolingual children the same age. In optionality either; thus, a logical explanation for cross-linguistic influence in this case is
Spanish, a speaker can omit the grammatical subject of a sentence, if the subject is under­ the children's dominance in Cantonese.
stood between the speaker and the hearer. 'Thus, Spanish conversations have numerous Whatever the reasons are for cross-linguistic structures, we would like to emphasize
sentences with null (omitted) subjects before the verbs. By contrast, English does not per­ that they are not rampant in the langu age of bilingual chi ldren. On the whole, simultane­
mit null subjects; thus, conversations 'in English would always have overt subjects, and if ous bilingual children use grammatical structures appropriate for their stage of develop­
the subject is known to both speaker and hearer, a pronoun subject is typically used. The ment in each language. But when parents or professionals dealing with bilingual children
Spanish-English bilingual girl in this study used more overt subjects in her Spanish come across structures that appea r to be ino.uenced by the child's other language, it is
sentences-subjects that were sometimes redundant in the conversation-than the important to bear in mind that this phenomenon is typical of bilingual development and
Spanish monolingual children her age. Paradis and Navarro suggested that the child's is not a sign of confusion or extraordinary difficulty coping with dual language input.
Spanish could have been influenced by the morphosyntactic rules of English, and that she
was therefore less likely to omit subjects in her Spanish. Use of an overt subject that is
ARE THERE DIFF ERENCES IN RATE OF DEVElOPMENT ,
redundant in Spanish is not a grammatical error per se, but it could make a child's speech
BETWEEN BILING UAL AND MONOLINGUAL CHILDREN?
sound pragmatically odd.
How long does cross-linguistic influence persist in development? Some researche rs, Many parents, educators, and health care practitioners have the impression that bilingual
including Dapke (2000) and Yip and Matthews (2007), have documented the rise and fall children go through the stages of language development more slowly than monolingual
of certain cross-linguistic structures in the speech of bilingual children over time in the children because they have twice as much language to acquire in the same amount of
preschool years. So, we know that some aspects of cross-linguistic ino.uence are tempo­ time.As discussed in Chapter 3, there is no evidence that bili ngual infants and toddlers are
rary. But research with school-age bilingual children conducted by Ludovica Serratrice, slower than children who are monolingual to pass through early critical milestones such
Antontlila Sorace, and their colleagues indicates that some very subtle aspects of cross­ as babbling, first words, and first word combinations. The concern we have in this section
I. linguistic influence could extend for a long time. These researchers studied .Italian-English is whethtlr this is also true for language development past the early months and for more
!I"'!""'""

Understanding Bilingual and Second Language Development Lal1guage Delle/opmel1l in Simultaneous Bilingual Children 71
70
fine-grained aspects of language. The research we discuss here shows that bilingual chil­
BOX 4.2 dren are within tbe normal range as defmed by monolingual children for some aspects of
language, but lag behind monolingual children far other aspects, in both the presehool
Determining tlie Dominant Language
and school-age years. In other words, bilingual children are not globally delayed in tbeir
Preschool children: How Ciln one tell what a young child's dominant language is? lexical and morphosyntactic development. The .differences in rates that are apparent from
The dominant language usually has a number of characteristi<;:s when compared research indicate that their rates of development are sensitive to factors such as amount of
with the nondominant language (se~ the following list). These characteristks have exposure to each language (see Box 4.2 and the following section on language domi­
been noted by various researchers and are based on spontaneous, natura.listic lan­ nance), the reiative complexity of the morphosyntactic structure being examined, and
guage samples from children approximately ]Ij,...3'/" years old. Researchers some­ possibly the minority/majority status of the language in the social context. Bilingual
times use a combin<)hon of direct measures of linguistic characteristics from the two children's sensitivity to these factors can sometimes cause them to lag behind their mono­
languages together with parental reports of which language iSI)"I(;>re proficient, or . lingual peers in at least one of their languages. Understanding where to expect differences
which language their child uses more often. We recomme[ld determining domi­ in rates oflanguage development between bilingual and monolingual children is particu­
.nance Using more than one measur~ alid, if poss'ible, across more than one ling~is- larly important for parents, educators, and health care practitioners who initiate referrals
.tic domain; that is, one vocabulary ilnd one morphosyntax measure rather than two and for specialists such as speech-language pathologists and psychologists, who conduct
morphosyntax measures. The dominant language would be the language for which language-based assessments. Before turning to research comparing bilingual and mono~
the child.has a higher score for most measures. Odd numbers ofmeasL!res, such as lingual rates of development and the sources of bilingual differences, we first discuss the
3 or 5, are good, bet;:ause a"tie" is not possible. Ira chii'd has very close or similar concept oflanguage dominance.
scores for each me~sure, then that child can be considered a bal~nc~d bilingual.'
Common characteristics of'a dominarit language: . Language Dominance
1. Longer mean length of utterance (MLU; average length calculated across all

Almost all studies of simultaneous bilingual children have found that even though they
utterarices in a stretch ofdiscourse)

have been acquiring two languages from birth, or shortly after, the two languages do not
2. More frequent appearance of "advanced"morphosyntactic structures develop in perfect synchrony. The language in which bilingual children appear to have
3. Larger number of different word types-or verb types in particulai~used in a greater proficiency is commonly referred to as their dominant language and the other
stretch of discourse of a fixed number of utterances .. as the nondominant language. (Other commonly used terms are strol7ger language and
4. Fewer pauses or hesitations in a stretch of discourse of a fixed number afutterances weaker language.) Most researchers of simultaneous bilingual children consider domi­
nance to be a measure of relative proficiency between the two 'Ianguages that the child is
5. Greater volubility: more utterances in a stretCh of discourse of afixed time length
learning; dominance is not usually construed as meaning that a simultaneous bilingual
6. Fewer code-switched utterances used in a stretch of discourse conducted child is incompetent in one language or has only passive knowledge of one language.
(mainly) inthat ianguage Dominance is closely linked to the amount of input the child receives in each lan­
See Genesee et al. (1995), Paradis and Nic.oladis (2007), or Foroodi-Nejad and guage, which is seldom equal. For example, if we inquired about the preschool linguistic
Paradis (2009) for techoical examples of ho~ dO.minance can be determi~ed. See also· exposure of a child such as Gabriela, we might find that she hears mainly Spanish from
Yip aod Matthews (2007) for in-depth discu5sioos of the use.of MLU cross-linguis\ically. her mother and both Spanish and English from her father on weeknights and weekends
and· of differen\approaches to calc'ulating dominance.. . and hears English during the weekdays at a child care center. Gabriela is.receiving much
School-age children: Calcu lati ng measures such. a,sutterance length from spoQta­ more input in English, and it would be expected for her to be English-dominant at this
neaus language samples may be less revealing for school:age children, because stage in her life. Barbara Pearson and her colleagues found that Spanish-English bilingual
their language knowledge and ability have groyvnso much.ln coniplexity that such children who received less than 25% of their input in Spanish often did not become flu­
methods would nat easily captur~differences in language profLcieilcy. Old~r chil- ' ent Spanish speakers (pearson, Fernandez, Lewedag, & Oller, 1997). Therefore, very low
dren's dominant language could be ascert~inedby ca~paring':their scores on lan­ input for one language could render it a passive language rather than a nondominant lan­
guage tests in both languages, if the tests are comparable in 'termsofwhat they guage.This would mean that the child would not spontaneously speak that language, but
measure and how they measure it. It is important not to rely.on just one, measure, could understand, more or less, what is addressed to him in that language. Over time, even
such. as vocabulary size, but examine at least more than one lingui~tic domain, passive ability in a language can disappear entirely, unless input in that language increases.
Comprehensive information on the chi Id's exposure to each language at home, at Expectatiom of balanced bilingual development in preschool children may be unre­
school, and outside of school could also assist in determining an older child's qom­ alistic, and it is more likely that bilingual children will be dominant in one language. One
. inant language. . consequence of dominance is that a bilingual child may appear to be less advanced in the
development of the nondominant language than an average monolingual child of the
~

72 Undmtandlng BllIl7gnal and Second Langl~age Development Language Development i/1 Simullamou5 Bilingual Children 73

same age (see Box 4.2 for details on determining a child's dominant language). It is also 14-month-old bilingual children could successfully learn new \-vords for novel objects in
possible th~t a bilingual child's proficiency in the nondominant language could be so lim­ an experimental task (reported in Werker et aI., 2009).
ited that the child would display competence in that language that is below the normal However, the words used in this task were phonologically very- different from each
range of pelformance based on monolingual children. For this reason, it could be inap­ other. Language learners must eventually be able to learn word-meaning pairs when the
propriate to clinically assess a child in their nondominant language, and in particular, if words differ in only one sound segment, for example, bet-pel in English, or boire ("to
only the non dominant language is being assessed. (For more on this point, see our discus­ drink")-poire ("pear") in French. Such words are called minimal pairs, because a small
sion of mean length of utterance later in this chapter and see Chapter 9.) phonological difference-that is, one sound-signals a complete difference in meaning;
Dominance has another effect on young bilingual children: their language choices. all languages have rninimal pairs. Learning form-meaning mappings for minimal pairs
The McGill University research project included a series of studies examining the sensi­ requires coordination of the emerging phonological system with cognitive word learning
tivity of bilingual children to the language preferences or abilities of their conversational skills. Fennell, Byers-Heinlein, & Werker (2007) tested bilingual and monolingual tod­
partners (Genesee, Boivin, & Nicoladis, 1996; Genesee, Nicoladis, & Paradis, 1995; dlers' ability to learn new words for novel objects when the words were minimal pairs,
Nicoladis & Genesee, 1996). The researchers found that even bilingual children as young such as blh and dlh. They found that monolingual children succeeded in mapping form
as 2 years old can tailor-their language choice to their adult interlocutor. However, bilin­ and meaning at 1] months of age, but that bilingual children succeeded in doing this
gual children are often constrained by their dominance, so when they are speaking to an closer to 20 months of age. Thus, bilingual children appear to lag behind monolingual
adult in their nondominant language, they often interjed·individual words and whole children in this ability, although the gap is not huge.
utteral1Ces from their dominant language in the conversation. Furthermore, Paradis and Finally, there is evidence that language dominance can playa role even at the very
Nicoladis (2007) found that 4-year-old French-English bilingual children in a French early stages of children's lexical development. Conboy and Mills (2006) measured the brain
minority /English mJjority context showed an interaction between dominance and lan­ activity of 19- to 22-month-old bilingual toddlers while they listened to known and
guage status in their conversations with unfamiliar interlocutors. Specifically, English­ unknown words.The researchers gathered information on what words would be-known to
dominant children felt free to use English with either a French- or English-speaking the children before the experiment, and also on their exposure patterns. to each language
interlocutor, but French-dominant children were reluctant to use French with an English­ to infer language dominance.The timing and distribution of brain activity to known words
speaking interlocutor and thus were restricted in the' topics they could engage in when was different, depending on whether the word was in the children's dominant or nondom­
conversing in English. Thus, these 4c year-old children showed awareness of the fact that inant language. In sum, at the beginning of vocabulary learning, bilingual children show
. most French speakers in their community are bilingual, but English speakers are not. similar rates of development to monolingual children, generally speaking, but they may lag
More information abollt bilingual children's language choice is presented in Chapter 5.' behind with regard to more challenging aspects of vocabulary acquisition, and the amount
Language dominance can also be seen in school~age bilingual children, as differential of exposure they have in each language can affect how they process words.
amounts of exposure to each language continues to have an impact on their language abili­ Researchers, educators, and speech-language pathologists often measure the size of a
ties past the early years. The sources of differential exposure to each language in older chil­ child's vocabulary using a standardized test in order to assess whether the child's language
dren goes beyond parental language use at home and can also include community context, is developing "normally" or typically-in other words, whether it falls within the range of
lailguage of instruction in school, language use with 'fi'iends, media choices, and more. normal performance/ability as defined by a large cohort of children the same age; that is,
within the age-expected norms. (More details on normal curve distributions and these
kinds of tests are given in Chapters 9 and 10.) Differences between bilingual and mono­
Vocabulary
lingual children have been consistently documented in research using standardized
One of the earliest vocabulary-building skills IS the ability to recognize familiar word vocabulary tests, including toddlers up to school-age children. These differences some­
forms in the speech stream. For example, an infant might"be able to recognize the word times show bilingual children pelforming below the normal range for their age in one of
"cup" as a familiar·group of sounds when spokeri in isolation, but the infant also needs to their languages, yet in other cases, performing lower than a monolingual comparison
be able to recognize this word when she hears it in a stretch of discourse, because words group, but still more or less within age-expected abilities. Furthermore, how bilingual
do not usually appear in isolation in natural conversation. Such word recognition abilities children compare to their monolingual peers in vocabulary development depends on
emerge toward the end of the fmt year oflife.Vihman and her colleagues examined word input exposure factors, possibly including social context.
form recognition in English monolingual and Welsh-English bilingual infants ages 9-12 Young Spanish-English bilingual preschoolers in the United States usually score
months using both behavioral and brain activity measures (Vihman, Thierry, Lum, Keren­ lower on standardized vocabulary tests in each language than monolingual children
Portnoy, & Martin, 2007). They found that the timing of the bilingual infants' abilities to (Conboy &Thal, 2006; Pearson, Fernandez, & Oller, 1993). In addition, vocabulary size in
recognize familiar word forms in both languages was the same as that for English mono­ each language varies directly in proportion to relative amount of exposure to each lan­
lingual children: around 11 months. After being able to recognize word forms, the next guage (Marchman et aI., 2004)-the more exposure to a language, the larger the vocabu­
step is for infants to map meanings onto those forms in order to build vocabularies. lary, other things being equal. A large-scale study of Spanish-English bilingual children in
Byers-Heinlein and her colleagues found that, as with monolingual children of the same age, both English-only and Spanish-English instructional programs in Miami schools revealed
~

74 U/ldcrstanding Bilil1<ellal and Second Language Del'elopl11cl11 Langllage Developmenl ill SimultanCOI/S Bilil1gl/al Children 75

that vocabular y size differences between 1110nolinguar children and bilingual children literacy skills in Spanish at home and participates in other linguistically rich activities in
persisted until at least Grade 5 (Oller.& Eilers, 2002). Exposure time to each language in Spanish on a regular basis. Furthermore, because Spanish is not a majority language in the
school made a difference. Bilingual children in Spanish-English instructional programs New·York City area (although it is a widely spoken minority language), ensuring long-term
had bigger vocabularies in Spanish than their peers in English-only programs. The bilin­ proficiency in this language will take some effort on the part of both schools and parents.
gual children in English-only programs had bigger vocabularies in English than their Why do bilingual children tend to have smaller vocabularies than ro.,onolingual chil­
bilingual peers in the Spanish-English program early on, but interestingly, this effect dren? Because bilingual children have the same cognitive abilities and limitations with
diminished between second grade and fifth grade (see Chapter 8 for more on the out­ respect to memory capaciry as monolingual children, one would expect that their vocab­
comes of bilingual education programs). ulary in each of their languages would be smaller than that of monolingual children, espe­
A Canadia n study of vocabulary size in French-English bilingual children attendi ng cially in the early years. Viewed this way, it is not surprising that bilingual children may
French-language schools in an English-majority city, Edmonton, showed similarities and not have accumulated the same absolute number of word- concept pairings in twO lan­
differences with this researc h from Miami (paradis, 2009). The main similarity was that guages in the same amount of time as monolingual children do for one language. Pearson
the size of bilingual children's vocabularies in each language was sensitive to the amount and her colleagues assessed . the vocabulary of bilingual Spanish-English toddlers in
of exposure to that language at home and in school. The main difference was that the Miami using parent report che·cklists of the words the children spoke and understood in
majority of bilingual children had vocabulary sizes equivalent to monolingual age peers both languages. They found that when the vocabularies of both languages we re com­
in French in first grade and sixth grade. In these French medium schools, English lan­ bined, and translation equivalents were counted once only, their total "conceptual"
guage arts instruction is introduced in third grade. This seemed to make a difference in vocabulary was similar in size to that of mo nolingual norms (Pearson, 1998; Pearson,
the children's English vocabulary scores, which were slightly lower than the monolingual Fernandez, & Oller, 1993). This shows that bilingual children are not slower than mono­
English average in first grade, but comparable to the monolingual English average in lingual children to assign word labels for concepts, the basic process that underlies vocabulalY
sixth grade. In sum, the magnitude ofbilingllal-monolingllal differences in-vocabulary size building, and that they are able to learn the same number of word-::.concept pairings as
was smaller in the Canadian study than in the Americail study, even though an identical monolingual children. .
measure of vocabulai·y size was used. One reaso n for this difference could be that because To summarize, bilingual children use the same mechanisms to acquire words as
schooling was nearly exclusively in the nunority language, French, this helped the Cana­ monolingual children, and thus are successful word learners. When bilingual children are
dian bilingual children to "balance out" the impact of English exposure in the larger compared to monolingual children , the size of their vocabularies in each language can
community. Whatever the reason is, the importance of the contrast between these studies seem smaller, but this depends on amount of exposure to each language, as determined by
is that parents and professionals need not assume that bilingual children will always lag home, school, and perhaps even the broader social context.
behind in their vocabulary development throughout the school years, but instead can
expect that bilingual children could lag behind in vocab ulary development, at least in Morphosyntax
some contexts.
Wha t do these findings suggest for our case study children such as James and In Chapter 3, we discussed researcb showing that exposure to two languages at the same
Gabriela? Remember, James is a French-English bilingual attending school in French in time does not delay when simultaneous bilingual children begin to use word combma­
a bilingual city, Montreal, and Gabriela is a Spanish-English bilingual attending a bilingual tions or simple sentences to express themselves. A common measure of children's early
school in the NewYork City area. First, the findings suggest that children in a social context morphosyntactic development is mean length of utterance (MLU) , which refers to
such as James's are more likely to acquire vocabulary similarly to monolingual children for the average length of a child's utterances, calculated across numerous utterances in a sam­
both languages earlier on than children in a social context similar to that of Gabriela. Sec­ ple of that child's spontaneous speech. In young children, "sentences" can consist of single
ond, they suggest that for the language not taught at school or taught less at school, dis­ words or phrases, which is why utterance is the preferred term rather than sentences.
crepancies between a bilingual child's vocabulary and monolingual-based norms would Miller and Chapman (1981) published a set of norms for MLU development in English
be expected, because literacy and academic language in school build vocabulary. In other based on the language of over a hundred monolingual children. MLU was calculated
words, the more literacy and academic language skills are developed in a langu age, the using morphemes per utterance, so a word such as cals would count as lVI'O morphemes:
more likely it is a child's vocabulary in that language will also be enhanced. Therefore, cal-s. Do the MLUs ofbilingllal children increase with age at the same rate according to
because James is attending a French-language school with conti nuous but limited English these monolingual norms! A query from a speech-language pathologist in the United
instruction, his parents might be advised to encourage him to read in English at home. States on this question (see Box 4.3) prompted our analysis here. Detailed information on
Because Montreal has a large English-speaking commu nity, as does Canada as a whole, it the MLUs of the three children studied in Paradis and Genesee (1996) are given in Box
is reasonable to expect that James will acquire English successfully. Gabriela's situation is 4.4, along with the normal range ofMLUs (one standard deviation above and below the
somewhat different. Because she is attending a transitional bilingual school with little or mean) from Miller and Chapman's sample of children.The bilingual children we exam­
no Spanish instruction after third grade (see Chapter 8 for more details on this kind of ined were gro\Ning up in the same context as the case study child,James. Some cautions.
.......I"'\rrr"lrnl h.:. ...
l", "rpnt<" TYl;nht "'I 1",f"'I hp ')(lU;l:pr! r() n1~kp <:nn=: t.hat she continues to de.ve1op
are in order, as these are nor ms for English and we are using them to judge French as
 76 Ullderstaltding Bilingual and Second Language Development Language Development ill Simultaneous Bilingual Children 77

BOX 4.3 . BOX 4..4

Mean length of Utterance as an Indicator ofTypical Development Mean length of Utterance in Bilingual Children and Monolingual Children

The MLUs from the bilingual children are from Para.dis and. Genese~ (1996, p.ll).

Dear Dr. Paradis,

They are calcu lated asMLU in morphemes and are co~ pared to the norms for

Is MLU (mean length of utterance) a good· indicator of a langvage impairment ranges of MLUs by age from Miller and .Chapman (1981, p.158). The two child~en

before the age of 4 for children 'developing two languages or hearing one lan­ with typical development, Gene and Oliver, have MLUs that fall close to the lower

guage spoken by parents and another in their community/TV/non-home ehviron­ bound or the middle of the ,normal range at all ages, The child with language delay, .
ment? In other words, should we expect an MLU of, saYt 2.0 for a 2-year-old who­ William, has MLU values that fall below the 1 standard deviation boundary (elr
is processing two languages-or 3.0 fora 3-year-old? I am having a difficult monolingual children .
time finding research on this exact topic, but thousands of Spanish-speaking
preschoolers in the United Slates are being identified as language impaired based Monolingual English
on their MLU before the age of 4, which concerns me greatly, even when their
-1 tQ +1 standard
English French deviation range
receptive language is measured as well within normalliinits.
. . Child Age MLU MLU for age'
Thank you for any help or direction you can give me in this endeavor.
'Gene' 1;11 2.04 1,92 1.47-2.37
A speech-language pathologist ' ..
(typically developing) . 2;7 . .2.17 2.12 1,97-3.11
The information in Box 4:4 illustrates the difference betyveen the MLUs of a bilingual 3;1 .2.44 .' . .. V6· 2.47....:...3.85
child with language delay and two bilingual children with typical development, as Oliver 1;11 1.55 . 2.32 1.47-2.37

referenced to monolingual children. For schoolcage Spanish~English bilingual chil­ (typically developing) 2;6 2,18 2.59 1.97~3.11

dren in the United States; normative information on MLLJs from a storytelling task 2;10 2.31 2.40 2.22-3.48 .

t, can be found on the SALT web site at http//www. languageanalysislab.com. William' 2;2 <1.29 1.26 1.72-2:74

(language delay) 2;10 1.54 US 2.22-3.48

3i 3 2.19 . 1.60 2:71~ .2 3

well; in addition, {here are some minor differences between our calculations and those
used by Miller and Chapman. Nevertheless, two of the children, Gene and Oliver, had
MLUs within or close to {he normal range in both their languages from the ages of 2 to morphosyntax during the preschool years, research shows some differences between
3 years, although their MLUs tended to be at the lower bound of the ran ge for monolin­ bilingual and monolingual children can occur in rates of development. Paradis, Crago, and
gual children. The MLUs of the bilingual children were not exactly equivalent in both Genesee (2005 / 2006) found that French-English bilingual 3-year-olds were as accurate
languages, which is most likely the result of unequal exposure to both langu ages, that is, as French monolingual children in their use of French object pronouns-a notoriously
language dominance. Similar findings on ea rly MLU development in three Spanish­ difficult aspect of French morphosyntax to acquire, unlike in English. In contrast, Perez­
English bilingual children were reported in Padilla and Liebman (1975). Leroux and her colleagues found that bilingual 3-year-olds lagged behind monolingual
We found that the pattern for Gene and Oliver held for other children in the Montreal­ chi ldren in their use of object pronouns in French (Perez-Leroux, Pirvulescu, &
McGill University study, except for William. Toward the end of the study, around the age Roberge, 2009). The difference in these findings could be due to social context. The
of 3 yea rs, William was identified by a speech-language patllologist as having expressive French-English bilingual children Paradis and colleagues studied were residing in the
language delay, and he received speech therapy services before entering school. This was bilingual city of Montreal, whereas the children studied by Perez-Leroux and colleagues
unknown to us or to his parents at the outset of the study. Notice that his MLUs are con­ lived in the English-majority city of Toronto. Nicoladis and colleagues found that
sistently lower than age-based monolingual expectations in both languages, Thus,Wiliiam French-English bilingual children's accuracy in using verbs in the past tense, such as ran
illustrates the development of MLU in a bilingual child who has language delay. The fact (rtm) or pick-ed, in both languages was slightly lower than their monolingual peers a{ 4-6
{hat his MLU values are distinct from those of Gene and Olivier shows how typically years of age (Nicoladis, Palmer, & Marentette, 2007). However, these researchers did not
developing bilingual children are not delayed in their growth in MLU compared to look at the children's performance in their dominant language in particular. Ina follow­
monolingual children, although they tend not to be at the upper end of the monolingual up study, Paradis, Nicoladis, and colleagues found that bilingual children could be as accu­
range. (More research on bilingual children affected by language delay and impairment is rate as monolingual children at this age in their me of {he past tense in their dominant
presented in Chapter 9.) Moving on to the acquisition of more specific aspects of language (Paradis, Nicoladis, Crago, & Genesee, in press). Comparisons between
 ·
r
l1li"""'""

Lal1gll~ge Development in Simultaneous· Bilillglinl Childrell 79

78 UJ1d erstal1ding Bilinglwl alld Sao/l d Language Development \

Spanish-English bilingual children and monolingual children in the United States have their grasp of subtle aspects of grammatical expression and knowledge. It is possible that
yielded parallel findings for similar aspects of morphosyntax (Gutierrez-Clellen, Simon­ the absence of schooling in one of a bilingual person's two languages could affect long­
Cereijido, & Wagner, 2008). In sum, it appears that for some aspects of morphosyntax, term outcomes. It is also possible that insufficient early exposure to Spanish, due to the
simultaneous bilingual children can keep pace with monolingual children-at least in societal predominance of English , causes what Montrul refers to as "incomplete acquisi­
their dominant language and/or in the majority language of their social context. tion." However, it is important to point out that the bilingual adults s"tUdied by Montrul
The studies of vocabulary discussed thus far that took place in Miami and Edmonton
were not necessarily incompetent in Spanish ; many were very proEcient in this Iangllage,
also included mtasures of children's morphosyntactlC development in both languages
particularly in a conversational context. Turning to our case study children, Montrul's
(Gathercole, 2007 ; Oller & Eilers, 2002; Paradis , 2009, in press). The results for mor­
Endings are more applicable to Gabriela than to James. Children such as James in any part
phosy ntacti~ meas ures largely paralleled those for vocabulary. First, bilingual children's
of Canada can have both their languages supported through schooling. Children such as
morphosyntactic abilities in each language were influ enced by how much expoSLlfe they
Gabriela lnight experience limited, transi'tional, or no Spanish language instruction in
had in that language in both studies. Second, in the Miam.i study, bilingual children lagged
school, reinforcing our earlier point about how important it is to encourage parents of .
behind monolingual children in both languages, but nearly closed the gap by fifth grade
children in a situation similar to that of Gabriela to enrich their children's language expe­
(Gathercole, 2007; Oller & Eilers, 2002). In contrast, in the Edmonton study, bilingual
riences in Spanish, including literacy, as much as possible.What is important here is to rec­
children, as a group, pelformed similar to monolingual children in French.The only bilin­
ognize that the linguistic limitations of the adults studied by Montrul do no t reflect
gual children who did not were those who spoke mainly English at home, and differences
lilnitations in these individuals' ability to learn both languages, but rather reflect the com­
with monolingual children were evident only in Erst grade and not in sixth grade (Paradis,
plexity of the learning environments in which they acquire them.
2009). For English morphosyntax, bilingual children lagge d behind monolingual chil~

dren in first grade, but bilingual children who came from homes where English was the
Bilingual Bootstrapping and Rate of Development
predominant language, or where English was spoken equally with French, perform.ed

What about cross-linguistic influences that could beneEt language development by facil­
comparably to monolingual children (Paradis, 2009, 2010).

As with vocabulary, the main distinction between the Miami and Edmonton studies
itating acquisition? To the best of our knowledge, the tenn bilingual bootstrapping
was that the magnitude of bilingual-mon olingual differences was smaller for the
was Erst use d by Gawlitzek-Maiwald andTracy (1996) to describe what they observed in
French-English bilingual children in Edmonton . The amount of the minority language
a German-English girl's use of code-mixing. The girl seemed to mix words from German
taught at school could be at the root of this difference between the studies, as mentioned
into English in order to stretch her expressive abilities in English; English was a nondom­
for vocabulary findings, but another reason could be the morphosyntactic structures exam­
inant language for this child . Since this study, bilii1gual bootstrappi11g has been used with a
ined. Because different aspects of morphosyntax were examined in each study, it is possible
slightly different meaning. It generally refers to the idea that a bilingual child's develop­
that the aspects of French and English that were studied were"easier" to learn than the
ment in one language can be advanced by the other, dominant language, andlor that the
aspects of English and Spanish studied in the children in Miami, giving the appearance of
two languages can be mutually advanced by virtue of sharing some linguistic-conceptual
smaller differences in the Edmonton study.The complexity of a morphosyntactic structure
knowledge. This means that development might proceed more rapidly than one might
can influence how much input and practice is necessary to learn it, which in turn could
expect if a bilingual child were acquiring each language in absolute isolation. For exam­
mean that bilingual children wo uld take a lot longer to acquire complex structures than
ple, a bilingual child might not have to "discover" all the linguistic concepts and develop
monolingual children. The impact of structure complexity was shown in a large-scale
all'the details oflinguistic representations twice. In this way, bilingual bootstrapping could
study of Welsh-English bilingual children in Welsh-medium schools (Gathercole,2007).ln
explain why bilingu al children, in spite of some lags relative to monolingual children, are
Welsh, there are numerous phonological changes in noun phrases when nouns have femi­
not substantially delayed in their overall language development. Put differently, some
nine gender, but fewer when nouns have masc uline gender.Virginia Gathercole and col­
bilingual children might receive half the input that monolingual children do, but not be
leagues found that bilingual children's accuracy with noun phrases in Welsh varied directly
twice as far behind monolingual children; on the contrary, they can display competence
with how much Welsh was spoken at home, but this variation was much more pronounced
that falls wi thin monolingual norms in some cases.
Is there direct evidence for bilingual bootstrapping? The studies discussed earlier on
for the more complex feminine forms (Gathercole, 2007).

Studies of simultaneous bilingual learners do not usually extend past the late elemen­
cross-linguistic influence show that the two langllages of a bilingual do not develop in iso­
tary school yea rs. This leaves us with the question of whether bilingual children ever
lation. However, generally speaking, this research has focused on unique patterns that arise
become indistinguishable E'om their monolingual peers in their morphosyntax, regardless
as a consequence of interaction between a bilingual's two developing languages, and not on
of dominance, instru ctional program, and complexi ty of structLlfes. In her 2008 book,
whether rates of development can be positively affected by connections between the lan­
Silvina Montrul reviews several studies she has conducted with adult university students
guages. Fabiano-Smith and Goldstein (2010) exalnined accuracy in pronouncing phonetic
who grew up speaking both English and Spanish, but attended school only in English.
segments in both Spanish and English in bilingual preschool children, as compared to '
Many of these adults would have been simultaneous bilingual children. She consistently
monolingual children.They put forward the proposal iliat phonetic segments iliat are si mi~
lar. or Sh:Hf.rl . hf':twfPn thp
r~"n.-1 'ho' ,h",p ;n,1;";,1,,ol , ,1;(("rp,1 <nmpwhot frnm Sn~nish monolimrual individuals in
h :u,"" bnCTll 'lrr pc rntrrht- h o ,::..,("; ,.,. .- f", .,. . 'h;1; ........... ... 1 .... l.:l,.L__ .. - ,. ___

L_
r

80 Understanding BilJl1gzlal and Second Language Developl1lent LangzlQge Development in Simullaneous Bilingual Children 81
than nonshared sounds because production or practice with shared sounds would be more learning a minority language along with English-for instance, Hebrew or Russian-in
frequent than with nonshared so_unds-children would use the shared sounds no matter the United States. These studies showed that rapid decline and loss ofsome morphosyn­
which language they are speaking. They argue that such facilitation might explain why tactic features of the minority language, especially upon entry to an English preschool
bilingual children can be more or less within the normal range for monolingual children in program, is common. Anderson (2001) reports erosion of some morphosyntactic and
their rate of phonological acquisition, despite less exposure to one or both languages. It is lexical abilities over time in two Puerto Rican children whose famifies moved to the
possible that what Fabiano-Smith and Goldstein found for phonology could extend to United States when they were preschoolers. Even though Spanish is less of a minority
morphosyntax as well. Concerning vocabulary, sharing at the conceptual and semantic lev­ language than Hebrew or Russian, it is important to highlight two points about the chil­
els for words is likely to take place in the developing bilingual lexicon (Gathercole, 2007). dren in Anderson's study: 1) they were residing in a small city in the midwestern United
Thus, bilingual children might need to learn separate phonological labels for words in each . States where there is less concentration of Spanish speakers and no Spanish instruction
language, but they might not necessarily need to learn separate conceptual and semantic in school, and 2) the parents were fluent English speakers and thus may have been using
features for all of them. In addition, facilitative interdependence ben'leen bilingual children's both languages comfortably at home. In contrast, our own research and that of our col­
languages has been noted by several researchers and in several reviews in domains related to leagues in Wales and in Miami are based on children such as James and Gabriela, who are
academic language abilities and literacy in particular (e.g., Oller & Eilers, 2002; Riches & growing up in both a bilingual family ancra bilingual community where they may have
Genesee, 2006). These studies will be discussed in some detail in Chapter 8. educational opportunities that support both languages. These children are expected to
It is also possible that bilingual bootstrapping reflects the sharing of cognitive processes develop their bilingualism past their early childhood years, although with some differ­
and/ or knowledge that underpins language learning, rather than the sharing of linguistic ences between them, as discussed earlier. In sum, it will be a challenge for Bistra's parents
constructs themselves. As discussed in Chapter 3, research by Ellen Bialystok has shown cog­ to ensure that she maintains fluency in Bulgarian past the preschool years and, as with
nitive advantages that can result from early bilingualism, particularly in what are called execu­ the other two children, it is recommended that acquiring literacy in Bulgarian could
tiveji.llzctiol15.These cognitive processes are important not just for other cegnitive functions or - help her to maintain the language.
academic skills, but also for learning language itself. Ii could be that bilingual development
enhances cognitive skills that, in turn, umher enhance bilingual development. However, not
enough research has been conducted on bilingual bootstrapping to know for certain which
EARLY MULTIUNGUAL DEVElOPMENT
aspects of shared linguistic or enhanced cognitive abilities might contribute to it. The information presented thus far has been based on children exposed to two languages
in their early years, but there are some children who learn three, or even four, languages
Development of the Minority Language: Dominance and Language Loss early·on; that is, multilingual children. In Chapter 3, we put forward evidence suggesting
that human infants have the capacity to learn more than one language, but what about
James, Bistra, and Gabriela are all simultaneous bilingual children, although a key differ­ three languages l There is very little research on children who are exposed to three lan­
ence between them is the minority/majority status of the two languages they are learn­ guages early on, and all are case studies. Typically, the children are exposed to two minor­
ing. Gabriela and James are learning two languages that are relatively widely used in the ity languages at home and to a third majority language as a result of attending day care or
cities where they live. Bistra, however, is learning Bulgarian and English in the United preschool programs at an early age. What these case studies tell us is that trilingual chil­
States in a city where there are very few Bulgarian speakers other than her mother. We dren are not much different fi-om the bilingual children we have just discussed in the early
have already discussed James and Gabriela in the context of the research, but what about preschool years, but developing comparable proficiency in all three languages as they
Bistra l grow older can be a challenge.
The importance of the minority/majority language distinction is how it can influ­ Simona Montanari has examined many aspects of the language development of a
ence children's early language dominance and their ultimate success at learning both Tagalog-Spanish-English trilingual girl growing up in California (Montanari, 2008,
languages, because it influences children's opportunities to experience rich and frequent 2009). The girl began to produce word combinations in all three of her languages just
input in both languages. Many simultaneous bilingual children such as Bistra may be after her second birthday, her emerging vocabularies included translation equivalents in
proficient in both languages when they are 3 or 4 years old but shift to become much all three languages, and the word order in her sentences reflected language specifiC pat­
more proficient in the majority language-typically the language of schooling-when terns from the outset. In addition, she tailored her language choice according to the inter­
they are older, and they may even end up not being bilingual adolescents and adults. locutor by using more of the interlocutor's preferred language when speaking with him
Much of the early research on bilingual preschool children was based on children such or her. Thus, trilingual exposure appeared to not cause major setbacks in basic milestones,
as Bistra, who are the children of so-called linguistically mixed marriages, in which the and differentiation of three languages was evident. Similar to that of bilingual children,
parents have different native languages and each parent speaks his or her native language this child's vocabulary size and MLU in each language reflected the amount of input she
to the child. In these cases, however, one language is not frequently used in the commu­ was receiving in each. Montanari reported that the girl's MLU in words at age 2 years, 4
nity outside the home. This kind of situation is called family bilingualism. Montrul months was 1.31 for Tagalog, 1.6 for English, and 1.27 for Spanish (Montanari, 2008,
(2008. Chaoter 4) reviews case studies of earlv bilimrual children such as Bistra who are p. 8).We cannot compare these to the table in Box 4.4. hFrOl1<F thmF MI I I, orp hocpri An
~

Ul1derSI~nding Bili17gllal @d Second Language Development L~l1gnage Devel0pJ1lent ill SillJI.I110l!fOtl5 Bilillgual Children 83
82
Somali-Swahili bilingualism simultaneously from birth, with English introduced after
BOX 4.5 migrating to Canada around the age of 3 years. In these situations, families do not neces­
sarily drop the "old" majority language inunediately, and so a child such as Faisal might be
Multilingualism in Young Children
an early trilingual child, although his trilingualism is likely to be transitional because Eng­
lish is likely to replace Swahili over time, The issue of minority langua'ge status is impor­
Oear Professor Genesee, tant for all trilingual children. In Montanari's case study, as soon as the child entered an
I live in California. My daughter is 2. Her mother is Brazilian and speaks only English language preschool, her English advanced considerably (Montanari, 2008). It is
Portuguese with her. Our nanny is Mexican and speaks only Spanish. I speak possible that the language that has the least support, Tagalog, might recede as the child
English with her. We have been considering a French immersion school for her; enters school. One of the biggest challenges facing families who want to raise their chil­
as I am aware of some of the benefits of a bilingual education. My fear, however, dren as multilingual speakers is to arrange for them to have sufficient and rich exposu re
is that asking her to learn a fourth language will tax her too much, and over­ to all the languages over time. Without such rich and sufficient exposure, it is not likely
~~~. '
that children will maintain and continue to develop all of their languages.

Sincerely, '
A father KEY POINTS AND IMPLICATIONS
We often receive eniails from parents and professionals that concern rather complex Key Point 1
cases of multilirigualism in young children. Most of the time, these queries are about
whether children can learn three ,or four languages early in life. Our answers to
Children exposed to two languages from birth have two separate but interconnected lin­
these queries always include-one central point: children's capacity for full bi- or guistic systems from the outset. Developmental stages and patterns are the same overall for
. multilingualism is dependent on their ability to receive rich and frequent exposure
monolingual and bilingual children. For the most part, the same kinds of " errors" occur in
. to both/all of their languages over time. The more languages, the more of a chal­
the language of bilingual and monolingual children while they are en route to mastering
the, adult language system.When bilingual children occasionally produce unique errors in
" lenge this is.
their language, it is most likely the result of cross-linguistic influence, which is a natural
process in bilingual development.

morphemes, not words. But Paradis and Genesee (1997) calculated MLU in words for Implications
two French-English bilingual children at the same age: Yann had 1.49 for English and Parents, educators, and health care practitioners should not assume that bilingual children
1.58 for French, and Mathieu had 1.22 for English and 1.60 for French (Paradis & Genesee, will display unique stages or patterns in their language development because dual lan­
1997, p. 103).Thus, for MLUs, this trilingual girl does not appear to be lagging behind her guage input might confuse them. Bilingual children can be expected to appear as "two
bilingual peers. Other case studies of trilingual children have also noted language-specific monolingual children in one" much-but not ~ll-of the time, and can differentiate
word order patterns in early sentences, the children's ability to differentiate use of their between their two languages. Parents, educators, and health care practitioners should not
!,\:!
languages with different interlocutors, and how their development is sensitive to the be concerned ifhilingual children sometimes produce sentences in one language that fol­
amount of input they receive in each language (Hoffman, 1985; Maneva, 2004). low the grammatical rules of their other language; this kind of cross-linguistic influence is
Parents, educators, and speech-language pathologists often ask us if learning three or typical, is mainly temporary, and is not a sign of confusion or impairment. It is simply one
four languages would be "overload" for young children, or whether it is even realistic to of the few ways in which bilingual children are not always "two monolingual children in
expect a child to be proficient in three or four languages learned at once.The e-mail mes­ one" (see also Chapter 5 on code-mixing).
sage in Box 4.5 represents a typical query from a parent about the feasibility of early mul­
tilingualism. The results of the case studies point in a positive direction for trilingual
Key Point 2
children in the early stages; they seem to indicate that children can sort out and separate
input from three languages and that critical developmental milestones emerge over time. There is no scientific evidence that children's language learning ability is limited to one
However, it should be noted that the case studies reported in scientific reports are based language. On the contr:lry, research indicates that infants have the capacity to acquire two
on children from families where trilingualism is a conscious choice that is managed by languages without signifIcant costs to the development of early milestones. Although
the parents. Children can also be exposed to three languages in the preschool years for bilingual children do not always keep pace with monolingual children in every aspect of
other reasons. Recall one of the children profiled in Chapter 1, Faisal, whose family spent their subsequent lexical and grammatical development, they often do so in their domi­
time in a refugee camp in Kenya before COIning to Canada. Although we have profiled Faisa! nant language (see Key Point 3). The rate oflanguage development for bilingual children
" , 'f'rnnn bn~II'~f' If',rnf'r ~ chilo from a similar backeround could have experienced .;, <:pn,itl"p tn tl,pi," pvnn<:llr.:> t~Y'np. r.... ~ ... rh 1" ........... " .......... ~ .. 1~~~~ .... ~ ~_L~ .... l _ .• J ; .... 1-_

 - "~ '. ,...

84 Understanding Bilingual and Sewnd Language Development Language DeveioplJlenl in SilllullaneoJ,/s BilinglAal Children
85
community. Any apparent differences in the rate of development between bilingual chil­ majority language is the language of the clinician, but the ' child's more proficient lan­
dren and monolingual children are not a sign that dual language learning is too burden­ guage is a minority language, because the. clinician lacks competence in the appropriate
some for young children. Rather, such differences reflect the complex linguistic language of assessment. Strategies for dealing with this mismatch are presented in Chap­
environment of bilingual children and its impact on their developmental trajectories. ter 9. Testing bilingual children in their nondominant language could result in substantial
underestimation of the child's linguistic abilities. Clinicians should be 'Very aware of the
Implications pitfalls of testing bilingual children in their nondolll.inant language and be appropriately
cautio us in coming to conclusions from such testing.
Parents and health care professio nals should not assume that any clinically signifICant
delays or difficulties a bilingu al child is experiencing in language development are the
result of-dual language exposure. Dual language 'exposure is not a risk factor for language REFERENCES
development per se; we discuss this in greater detail in Chapter 9. However, bilingual
children might lag behind their monolingual peers for some aspects of development, and Anderson, R. (2001). Lexical morphology and verb use in child first language loss: A preliminary case study
investigation. Il1lemaliol1al JOllrnal of Biling"oliSlIl, 5, 377-402.
it is important to distinguish between variation in rates of d~velopment that might be due Burns.T.C.,Yoshida, K.A., Hill, K. ,& Werker,]. (2007). The developnlent of phonetic representation in bilin­
to having less exposure to a language and variation in rates of development that are truly gual and monolingual infants. Applied Psycholillguislics, 28,455-474.
clinically significant and signal an inherent language learning disability (more about this Celce-Murcia, M. (1978) .The simultaneous acquisition of English and French in a two-year-old child. In E.
Hatch (Ed.), Second language acqllisilioll:A book of readings (pp. 38-53). Rowley, MA: Newbury House.
issue in Chapter 9). One of the most C0111l11on results of variation in exposure is language Conboy, B., & Mills, D. (2006). Two languages, one developing brain: Event-related potentials to words in
dominance (see Key Point 3). What appear to be "deI'ays" in the development of one of a bilingual toddlers. Developmental Science, 9, F1 -F12.
child's languages could be due to less exposure to that language compared to his other Conboy, B., & Thai, D. (2006). Ties between the lexicon and grammar: Cross-sectional and longitudinal stud­
ies of bilingual toddlers. Child Deve/opllletJI, 77,712-735 .
language. In order to ensure full bilingualism~or even trillngualism-it is vital that chil­
Dapke, S. (1998). Competing language structures: The acquisition of verb p'lacement by bilingu;l German­
dren be given consistent, continuous, and rich exposure to their 'languages on a regular English children. JOIITlwl ~f Child u,nglloge, 25(3),555-584.
basis. Finally, it is worth keeping in mind that "the whole is greater than the sum of its Dapke, S. (2000). The interplay between Ianguage-speci6c development and cross-linguistic inOuence. In S.
parts" insofar as the combined communicative and linguistic competence of bilingual Dapke (Ed.), Crosslinguistic slTlWllres in sim"/raneo,,s bilingualislll (pp. 79-103). Amsterdam:John Benjamins.
Fabiano-Smith, L., & Goldstein, B. (2010). Phon ological acquisition in bilingual Spanish-English speaking
children is concerned, and viewed fi'om this perspective, any differences they display rela­ children.)ournal of Speech, u,nguage and Hearing Research, 53, 160-178.
tive to monolingual children may be unimportant when proficiency in the two languages Fennell, C., Byers-Heinlein, K, &Werker,j. (2007). Using speech sounds to guide word learning:The case of
combined is considered. bilingual infants. Child Developmel1t, 78, 1510- 1525.
Foroodi Nejad, E, & Paradis,]. (2009). Crosslinguistic transfer in the acquisition of compound words in
Persian-English bilinguals. Biling'lOlism: Language and Cognitioll, 12, 411--427.
Key Point 3 Gathercole,YM. (2007). Miami and North Wales, so far and yet so near: A constructivist account of morpho­
syntactic development in bilingual children. Inlemal ional Journal of Bilingual Education alld Biling"olistl1, 10,
Dominance or unbalanced develop ment of the two languages is expected and typical in 224-247.
bilingual acquisition, particularly among preschoolers. Bilingual children may appear Gawlitzek-Maiwald, I., & Tracy, R. (1996). Bilingual bootstrapping. Linglli5lics, 34, 901-926.
Genesee, E (1989). Early bilingu al development: One language or two' JOlnnal af Child Language, 6, 161-179.
more advanced in one of their two languages. Dominance is closely related to which lan-: Genesee, E, Boivin, I., & Nicoladis, E. (1996) .Talking with strangers: A study of bilingual children's commu­
guage the child hears and speaks more often in the home and in preschool/school set­ nicative competence. Applied PsyciJolingllistics, 17, 427-44 2.
tings. The majority/mi nority status of a language can contribute to the frequency of Genesee, E, Nicoladis, E, & Paradis,]. (1995). Language differentiation in early bilingual development. JOllmal
cfChild u,nguage, 22,611-631.
input in that language and, in turn, to language dominance.
Gutierrez-CleHen, V., Simon-Cereijido, G., & Wagner, C (2008). Bilingual children with language impair­
ment: A comparison with monolinguals and second language learners. Applied Psycholil1gl.listics, 29, 3-20.
Implications HofIinann, C (1985). Language acquisition in two trilingual children.Journal of Multilingllal and lvIulliwltu ral
Develapmenl, 6, 479-495.
Health care professionals should try to determine which of a bilingual child's two lan­ Johnson , c., & Lancaster, P. (1998).The development of more than one phonology:A case study of a Norwegian­
English bilingual child. Inlernalional Journal of Bilingualism, 2(3), 265-300.
guages is the dominant language before assessing the child's language development. This
Maneva, B. (2004). "Maman, je suis polyglotte" [Mommy, I'm a polyglot]: A case study of multiLnguallan­
can be achieved by asking the parents about the child's language environment in order to guage acquisition from 0 to 5 years. The InlemalianalJolmlO1 af Nhr/tilinglralism, 1, 109-122.
ascertain which language the child is exposed to and uses more often. The language of Marchman,v, Martinez-Sussman, C, & Dale, P. (2004).The language-specific nature of grammatical develop­
ment: Evidence from bilingual language learners. Developmental Srience, 7, 212-2~4.
greatest exposure is typically the dominant language. This can be determined by asking
Meisel,]. (1989). Early differentiation of languages in bilingual children. In K. Hyltenstam & L. Obler (Eds.),
parents a series of questions about language use around the child. More precise measures Bilingualism across Ihe lifespan: Aspeas of acqllisitiOI1, malllrity alld loss (pp. 13-40). Cambridge, England:
oflanguage dominance for preschool children would include examining vocabulary size, Cambridge University Press.
sentence length and complexity, and volubility in each language. When assessing a bilin­ Meisel,). (1994). Bilillgllal first language acquisition: Frel1ch and German grommatical development. Amsterdam:John
Benjamins.
gual child, the dominant language' is the one to. examine for the upper limits of that Miller,]., & Chapman, R. (1981). The relation between age and mean length of utte rance in morphemes.
rhilrt ,'<:: ilnO"lli.'ti r npo'UpjnnnlPnt Thi c;: r::lTl hp. nrnhlpl11::1tir Fnr r::l<:p<: ""hprp thp <::()rip t::li nr lOl-4rnal of Snurh and F-fpm-il1(J Rp((>lTrrfl ?d 1 t:..tl_1 (..1
86 U,lderslfll1dillg Bilil1gual and Second Language Developl1lent Language Development in Simultancolls Bilingual Children 87

Montanari , S. (2008). Pragmatic differentiation in early trilingual development.jollll/al of Child LOllglwge, 36, Serra trice, L, Sorace,A., Filiaci, F., & Baldo, M. (2009). Bilingual children', sensitivity 10 speCificity and gener­
495-527. iciry: Evidence from metalinguistic awareness. Biling"alislll: Lallgllage and Cogl1ilioll, 12, 239-257.
Montanari, S. (2009) . Multi-word combination, and rhe emergence of.differentiated ordering pattern, in Sundara, M., Polka, L, & Molnar, M. (200S). Development of coronal stop perception: Bilingual infants keep
early trilingual development. Bilingualisl11: Lallgllage olld Cogn ilioll, 12, 50}-S I9. pace with theinnonolingual peers. Cognition, 108,232-242.
Monrrul, S. (2008). Incomplete acquisilioll in hiiingliolislII: Re·examillil1g the agefaa or. Am,terdam:John Benjamins. Vihman, M., Thierry, G., Lum ,]., Keren-Portnoy, T., & Martin, P. (2007) . Onset of word form recognition in
Nicobdi" E. (2002).What's the difference between "toilet paper" and "paper toilet'" French-English bilin­ Engli'h, Welsh, and Engli'h-Welsh bilingual infants. Applied Psycholingliistlcs, 28, 47.5;;;493.
gual children', crosslinguistic transfer in compound nouns.jollmal of Child Langllage, 29, 84}-S63. Volterra, v., & Taeschner, T. (1978). The acquisition and developm ent of language by bilingual children.
Nicoladi" E. (2003). Cross-linguistic transfer in deverbal compounds of preschool bilingual children. Bilin· Journal of Child Lallg"age, 5, 311-326.
gualism: Langllage alld Cognilioll, 6, 17-31. Werker, j., Byers-Heinlein, K., & Fennel, C. (2009). Bilin gual beginnings to learning word,. Philosophical·
Nicoladis , E., & Genesee, I' (2007). Bilingual first language acquisition . In E. Hoff & M. Shatz (Eds.), Halld· D"OIIsoctions of lire Royal Sociely B, 364,3649-3663.
book of Langllage Development (pp. 324- 342). Oxford, England: Blackwell. Yip,v., & Matthews, S. (2007). Tire bilingual child: Early developmelll aud language COl/focI. Cambridge, England:
Nicoladis, E., & Genesee, F. (1996). A longitudinal Study of pragmatic differentiation in young bilingual chil­ Cambridge Univer,ity Press.
dren. Larrg"age Leaming, 46(3), 439-464.
Nicoladis, E., & Secco, G. (2000). Productive vocabulary and language choice. First Langllage, 20(5 S), 3-28.
Nicoladis, E., Palmer, A., & Marentette, P. (2007). The role of type and token frequency in using past tense
morpheme, correctly. Developmental Sciellce, 10, 237-254.
Oller, KD., & Eilet~ , R. (Eds.). (2002). Laligllage ond lileracy ill bilingual children. Clevendon, England: Multilingual
Matter,.
Padilla, A.M. , & Liebman, E. (1975). Language acquisition in the bilingual child. Bilingl/al Review, 2(1-2),
34-55.
Paradis,]. (2010). Bilingual children's acquisition of English verb morphology: Effects of language expo,ure,
structure complexity, and task type. Language Leaming, 60, 65 1-680.
Paradi, ,]. (2001). Do bilingual two-year-olds have separate phonological systems' linernatiol1aljolU/wl of Bilin·
gl1alism,5,1 9-38.
Paradis,]. (2007). Early bilingual ana multilingual acquisition . In P. Auer & Li Wei (Eds.), Handbooks ofApplied
Linguistics: Vol 5. Multilingualism (pp. 15-44). Berlin: Mou ton l de Gruyter. .
Paradis,]. (2009). Oral language development in French and Englilh and Ihe role <1{ horne input factors. Edmonton,
Alberta, Canada: Report for me Conseil Scolarie Centre-Nord [North-Central School Board]
Paradi,,]., & Navarro, S. (2003). Subject realization and crosslinguistic interference in the bilingual acquisition
of Spanish and English.)olm,"1 of Child Language, 30, 371-393 .
Paradis,]., & Nicoladis, E. (2007). The influence of dominanc e and sociolinguistic context on bilingual
preschoolers' language choice. The Intetnational Journal of Bilingualism and Bilingual Education, 10, 1-21.
Paradis,]., Crago, M., & Genesee, I' (200512006). Domain-'peciflc versus domain-general theories of the
deficit in SLI: Object pronoun acqui,ition by French-Engli,h bilingual children. Langlinge Acquisition, 13,
33-62.
Paradi,,]. , Nicobdis, E., Crago, M., & Genesee, F. (in press). Bilingual children 's acquisition of the past tense:
A usage- based approach.JolIlI/al·of Child Language.
Paradis,]., & Genesee, F. (1996). Syntactic acqui,ition in bilingual children: Autonomous or interdependent'
Studies ill Second Language Acq,.,isition, 18, 1-15.
Paradis,]', & Genesee, F. (1997). On continuity and the emergence of functi onal categories in bilingual first
langua ge acquisition . Language Acql/isilion, 6(2),91-124.
Paradis, J., Nicoladi" E., & Genesee, I' (2000) . Early emergence of ,tructural constraints on code-mixing: Evi­
dence from French-English bilingual children. Bilingualism: Lallgllage and Cognition, 3(3), 245- 261.
Pearson, B. (1998). Assessi ng lexical development in bilingual babies and roddlers. Inlernational jOlm101 of Bilin·
gualism , 2(3),347-372.
Pearson, B., Fernandez, S., Lewedeg,V., & Oller, O.K. (1997). The relation of input factors to lexical learning
by bilingual infants. Applied Psycholil1gllislics, 18, 41-5S.
Pearson, B., Fernandez, S.c., & Oller, D.K. (1993). Lexical development in bihngual infants and toddlers:
Comparison to monolingual norms. Language Leaming, 43, 9}-120.
Pearson, 8., Fernandez, S., & Oller, o.K. (1995). Cross-language ,ynonyms in the lexicons of bilingual infants:
One language or rwo' JourI1al of Child Langl/agr, 22, 345-36S.
Perez-Leroux,A.T., Pirvulescu, M., & Roberge,V. (2009). Bilingua lism as a window into the language faculty:
The acqui,ition of objects in French-speaking children in bilingual and monolingual contexts. Bi/ingl/al·
ism: Lang"age and Cognilion, 12, 97-112.
Quay, S. (1995). The bilingual lexicon: Implications for '[lIdies of language choice. j 011mai of Child Langliage,
22, 369- 387.
Riches, c., & Genesee , I' (2006). Literacy: InstruCti onal issues. In I' Genesee, K. Lind holm-Leary,W Saunders,
& D. Christian (Eds.), Educatillg English langllage learners: A synthesis of research evidence (pp. 109-175).
Cambridge, En·gland: Cambridge University Press.