Bilingual Development in Childhood

De Houwer
In the first decade of life, children become bilingual in different

language learning environments. Many children start learning
two languages from birth (Bilingual First Language Acquisition).
In early childhood hitherto monolingual children start hearing
a second language through day care or preschool (Early
Second Language Acquisition). Yet other hitherto monolingual Child Development
children in middle childhood may acquire a second language
only after entering school (Second Language Acquisition).
This Element explains how these different language learning
settings dynamically affect bilingual children’s language
learning trajectories. All children eventually learn to speak the
societal language, but they often do not learn to fluently speak
Bilingual
Bilingual Development in Childhood

their non-societal language and may even stop speaking it.
Children’s and families’ harmonious bilingualism is threatened
if bilingual children do not develop high proficiency in both
languages. Educational institutions and parental conversational
practices play a pivotal role in supporting harmonious bilingual
development.
Development
About the Series Series Editor
in Childhood
Child development is a lively and engaging, Marc H. Bornstein
yet serious and purposeful subject of National Institute
Annick De Houwer
academic study that encompasses of Child Health
myriad of theories, methods, substantive and Human
areas, and applied concerns. Cambridge Development,
Elements in Child Development proposes Bethesda
to address all these key areas, with unique, Institute for Fiscal
comprehensive, and state-of-the-art Studies, London
treatments, introducing readers to the UNICEF, New York
primary currents of research and to original City
perspectives on, or contributions to,
principal issues and domains in the field.
Cover image: Shutterstock / Creativika Graphics ISSN

Downloaded from https://www.cambridge.org/core. IP address: 188.54.28.56, on 22 Apr 2021 at 08:12:24, subject to the 2632-9948Core
Cambridge (online)
terms
of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108866002 ISSN 2632-993X (print)
Downloaded from https://www.cambridge.org/core. IP address: 188.54.28.56, on 22 Apr 2021 at 08:12:24, subject to the Cambridge Core terms
of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108866002
Elements in Child Development
edited by
Marc H. Bornstein
National Institute of Child Health and Human Development, Bethesda
Institute for Fiscal Studies, London
UNICEF, New York City
BILINGUAL DEVELOPMENT
IN CHILDHOOD
Annick De Houwer
Harmonious Bilingualism Network
and University of Erfurt
University Printing House, Cambridge CB2 8BS, United Kingdom
One Liberty Plaza, 20th Floor, New York, NY 10006, USA
477 Williamstown Road, Port Melbourne, VIC 3207, Australia
314–321, 3rd Floor, Plot 3, Splendor Forum, Jasola District Centre,
New Delhi – 110025, India
79 Anson Road, #06–04/06, Singapore 079906
Cambridge University Press is part of the University of Cambridge.

It furthers the University’s mission by disseminating knowledge in the pursuit of
education, learning, and research at the highest international levels of excellence.
www.cambridge.org
Information on this title: www.cambridge.org/9781108791397
doi: 10.1017/9781108866002
© Annick De Houwer 2021
This publication is in copyright. Subject to statutory exception
and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without the written
permission of Cambridge University Press.
First published 2021
A catalogue record for this publication is available from the British Library.
ISBN 978-1-108-79139-7 Paperback
ISSN 2632-9948 (online)
ISSN 2632-993X (print)
Cambridge University Press has no responsibility for the persistence or accuracy of
URLs for external or third-party internet websites referred to in this publication
and does not guarantee that any content on such websites is, or will remain,
accurate or appropriate.
Elements in Child Development
DOI: 10.1017/9781108866002
First published online: April 2021
Annick De Houwer
Harmonious Bilingualism Network and University of Erfurt
Author for correspondence: Annick De Houwer, [email protected]
Abstract: In the first decade of life, children become bilingual in different

language learning environments. Many children start learning two
languages from birth (Bilingual First Language Acquisition). In early
childhood hitherto monolingual children start hearing a second
language through day care or preschool (Early Second Language
Acquisition). Yet other hitherto monolingual children in middle
childhood may acquire a second language only after entering school
(Second Language Acquisition). This Element explains how these
different language learning settings dynamically affect bilingual
children’s language learning trajectories. All children eventually learn to
speak the societal language, but they often do not learn to fluently
speak their non-societal language and may even stop speaking it.
Children’s and families’ harmonious bilingualism is threatened if
bilingual children do not develop high proficiency in both languages.
Educational institutions and parental conversational practices play
a pivotal role in supporting harmonious bilingual development.
Keywords: bilingual, children, learning, language, acquisition
© Annick De Houwer 2021

ISBNs: 9781108791397 (PB), 9781108866002 (OC)
ISSNs: 2632-9948 (online), 2632-993X (print)
Contents
1 Introduction 1
2 Becoming Bilingual in Infancy: Focus on Bilingual First

Language Acquisition 6
3 Bilingualism in Early Childhood: Bilingual First and Early

Second Language Acquisition 25
4 Bilingualism in Middle Childhood: Bilingual First

Language Acquisition, Early Second Language
Acquisition, and Second Language Acquisition 42
5 Socioeconomic Status and Bilingual Development in

Childhood 52
6 Summary and Conclusion 54
References 60
Bilingual Development in Childhood 1
1 Introduction
This Element concerns young children growing up with more than a single
spoken language. Three life stages form its organizing structure (Steinberg
et al., 2011, p. 5): (i) infancy, until about age 2 (infants); (ii) early childhood,
until about age 6 (toddlers and preschoolers); and (iii) middle childhood, to
about age 11 (schoolchildren).
The transitions between (i) and (ii) and between (ii) and (iii) coincide with
fundamental changes in children’s language use worldwide. When infants
begin to speak, they do so mostly in just single words. Shortly before
their second birthdays, toddlers combine words into short units that start
to resemble sentences. Around age 6, schoolchildren start to learn to read
and write.
Many newborns start hearing two languages from birth. This is a Bilingual
First Language Acquisition (BFLA) language learning setting (Meisel, 1989).
BFLA children have no experience with monolingualism. They are learning
Language A and Language Alpha (Wölck, 1987/1988). This terminology
expresses the lack of chronological difference between languages in terms of
first regular exposure.
Other children first hear just a single language. For many of these initially
monolingually reared children the transition between stages (i) and (ii)
coincides with fundamental changes in linguistic environment: Children
may start to regularly hear a second language through day care or preschool.
Children who grew up monolingually in a first language (L1) but start
regularly hearing a second language (L2) in late infancy or early childhood
are growing up in an Early Second Language Acquisition (ESLA) setting (De
Houwer, 1990).
For yet other initially monolingually reared children, it is the transition
between (ii) and (iii) that coincides with fundamental changes in linguistic
environment. Children who grew up monolingually throughout infancy and
early childhood may start attending school in a new second language (L2)
that differs from the one people were talking to them before (their L1). These
children in middle childhood are growing up in a Second Language
Acquisition (SLA) setting (De Houwer, 2019c). The latter differs from
ESLA because SLA schoolchildren are simultaneously learning not only to
understand and speak but also to read and write the new L2.
Bilingual children in very early infancy are typically BFLA children.
Bilingual children in late infancy and early childhood are either BFLA or
ESLA children. Bilingual children in middle childhood include BFLA, ESLA,
and SLA children (see Figure 1).
2 Child Development
Infancy Early childhood Middle childhood
BFLA----------------------BFLA------------------------------------------BFLA--------------------------
ESLA---- ESLA------------------------------------------ESLA-------------------------
SLA---------------------------
Figure 1 Three kinds of bilingual learning settings in three life stages

Legend: BFLA = Bilingual First Language Acquisition; ESLA = Early Second Language
Acquisition; SLA = Second Language Acquisition
Many people are bilingual because of a migration background. Bilingualism

also occurs in regions where two or more languages have been used side by side
for centuries, as in Papua New Guinea, much of South America, the Indian
subcontinent, and many regions in North America and Europe (Bhatia &
Ritchie, 2013). There is usually a social hierarchy between languages, with one
particular language and its users having higher status, social value, and power. The
language locally used in public life, government, and education tends to be the one
with the most prestige and is henceforth called the societal language (Soc-L). All
other languages are non-societal languages (Non-Soc-Ls). Non-Soc-Ls are not
used in education (except in foreign language classes); what is a Non-Soc-L in one
region may be a Soc-L in another. In some regions two languages may be used in
public life or in schools, or the local Soc-L differs from that in other parts of the
country, but even in those settings there is a language hierarchy. The specific
language environments and local language hierarchies individuals find themselves
in strongly influence their bilingualism (De Houwer & Ortega, 2019). Local
language hierarchies likewise affect bilingual children, regardless of whether the
bilingual environment is one of BFLA, ESLA, or SLA.
BFLA children may hear a Non-Soc-L and the Soc-L at home, or they may
hear two Non-Soc-Ls at home and learn the Soc-L in an (E)SLA setting at (pre)
school. Because we know hardly anything about BFLA children in the latter case,
the review in this Element only considers BFLA children with a Non-Soc-L and
the Soc-L at home. Families who speak a Non-Soc-L and the Soc-L with BFLA
children usually want their children to learn to understand and speak both
languages from the start (De Houwer, 2017c). There are emotional and cultural
bonds with both languages, and children are typically expected to speak both at
home. This is different for families with (E)SLA children, who only use a Non-
Soc-L at home. These families generally have a much stronger emotional and
cultural connection to the Non-Soc-L than to the Soc-L that children learn
through day care or (pre)school: Children are expected to speak the Non-Soc-L
at home, not the Soc-L.
The term “bilingual children” refers to typically developing, normally hear-

ing children under age 12 who need to learn to communicate in more than
a single language in daily life, leaving unspecified to what extent children are
able to communicate in two languages. The focus is on the untutored, so-called
“naturalistic” learning of several languages as a result of life circumstances that
are not easy to change. Children are born into a bilingual family (BFLA).
Children born into families speaking just a Non-Soc-L (ESLA and SLA) usually
have no other choice than to attend (pre)school in a Soc-L that differs from the
home language, and must learn to function in the Soc-L.
Aside from a Non-Soc-L and a Soc-L, children may learn a foreign language
(For-L), that is, a language that is not a local Soc-L. The For-L may be a Non-
Soc-L for some children in the classroom and a Soc-L in other regions and
countries. Although a For-L often carries high prestige and its learning is an
asset, learning a For-L well is not fundamental for children’s future in
a particular region: They do not need a For-L in daily life. In contrast, all
children, bilingual and monolingual, need to learn the local Soc-L well to
function in society. Furthermore, people’s emotional and cultural connections
with For-Ls are generally different from those they have with languages learned
at home and/or through residence in a new country. Muñoz and Spada (2019)
and Juan-Garau and Lyster (2019) offer overviews of For-L teaching and
learning.
Much of the heavy social and financial investment in the early teaching of
For-Ls (especially English) in Europe and elsewhere rests on the assumption
that children need to learn a language from an early age to learn it well (De
Houwer, 2015a). As summarized in Section 6, the joint findings from studies on
child bilingual development in naturalistic settings do not support this notion of
“the earlier, the better.”
The focus in this Element is on oral language use (Tang & Sze, 2019, discuss
how children acquire a spoken and a sign language or two sign languages;
Murphy, 2018, reviews how bilingual children learn to read and write).
Excluded from discussion are transnational adoptees who have replaced their
L1 with a single L2 (Genesee & Delcenserie, 2016) and children with develop-
mental challenges such as autism and Developmental Language Disorder
(Marinis et al., 2017; Patterson & Rodríguez, 2016).
It is possible for children to hear more than two languages. In referring to
bilingualism, this Element includes those more complex multilingual situ-
ations, although the research basis for early multilingual rather than narrowly
bilingual development is still thin (some excellent studies are Chevalier,
2015; Cruz-Ferreira, 2006; Montanari, 2010). Children may acquire different
varieties of the same language (e.g., Appalachian dialect and Standard
4 Child Development
American English) or varieties of different languages (e.g., Mexican Spanish

and Standard American English). Given the paucity of research on the first
case (Chevrot & Ghimenton, 2019; Durrant et al., 2014), this Element
exclusively refers to learning different languages, although learning very
different varieties of what is considered the same language is also a form of
bilingualism.
Child bilingualism is a widespread global phenomenon. Precise statistics are
not available. The fact that many societies in the Global South1 are highly
multilingual (Bhatia & Ritchie, 2013) means that children living there come
into contact with several languages as a normal part of everyday life. If children
do not encounter several languages before starting school, they often learn an
ex-colonial language at school that they do not hear at home but that functions
as a Soc-L (e.g., French in the south of Morocco; English in South Africa).
Children in the Global North2 are more likely to live in societies dominated by
a single language in which monolingual ideologies reign (Fuller, 2019). In
many of those societies, various types of statistics suggest that a fifth to over
a third of children do not solely hear a Soc-L at home: for instance, the 2016
Australian census3 found that more than one fifth of people spoke another
language than English at home; in Germany, well over a third of children
under age 10 had a migration background in 2016 (Autorengruppe
Bildungsberichterstattung, 2018) and thus may have heard another language
than German at home; the United-States-based Annie E. Casey Foundation
brought together data suggesting that 23% of US children speak a language
other than English at home;4 and survey data collected in the officially Dutch-
speaking region of Flanders, Belgium, in the 1990s (De Houwer, 2003) sug-
gested that in one out of eight families, children heard a language other than
Dutch at home.
In what circumstances do children become bilingual? We know little about
the proportions of BFLA, ESLA, and SLA children. However, BFLA may
occur about three times as often as ESLA and SLA combined (calculation
based on survey data from different continents in De Houwer, 2007 [N =
1,899; Belgium], and in Winsler et al., 2014a [N = 1,900; United States];
1
North–South Divide in the World. (n.d.). Wikipedia [website]. https://en.wikipedia.org/wiki/
North%E2%80%93South_divide_in_the_World
2
North–South Divide in the World. (n.d.). Wikipedia [website]. https://en.wikipedia.org/wiki/
North%E2%80%93South_divide_in_the_World.
3
2016 Census: Multicultural. (n.d.). Australian Bureau of Statistics [website]. abs.gov.au/ausstats/
[email protected]/lookup/media%20release3
4
Children Who Speak a Language Other than English at Home in the United States. (2020,
October). The Annie E. Casey Foundation: Kids Data Center [website]. https://datacenter.kids
count.org/
when families reported using two languages at home, children likely heard these
from birth; when families used only a Non-Soc-L at home, children likely grew
up with that language as an L1 and learned an L2 later). A rare study of 681
bilingual students in Germany probing children’s language learning histories
found that 51% of children had grown up in a BFLA setting, 24% in an ESLA
setting, and 25% in an SLA setting (calculations based on several tables in
Ahrenholz et al., 2013). A majority of bilingual children thus likely grow up
bilingually from birth rather than with a single L1 that is later complemented by
an L2.
Different language learning environments (BFLA, ESLA, and SLA) have
different effects on bilingual children’s language learning trajectories in the first
decade of life. It is the main aim of this Element to elucidate these language
learning trajectories. The online presentations HaBilNet Class 1: Trajectories
for Early Bilingualism5 and HaBilNet Class 2: BFLA Compared to ESLA6 on
the HaBilNet Vimeo channel offer a quick overview for BFLA and ESLA.
Lay people and researchers alike often want to know how bilingual children
compare to monolingual peers in the single Soc-L monolinguals are learning.
“Success” for bilinguals is frequently measured solely in terms of bilinguals’
performance in the Soc-L compared to monolinguals. If bilinguals perform
worse than monolinguals the blame is often laid with the fact that bilinguals
are acquiring another language. Even though bilingual–monolingual compari-
sons can elucidate theoretical questions, a unique focus on how bilinguals
resemble monolinguals in Soc-L performance rarely leads to a better under-
standing of child bilingualism. Bilingualism is not a sort of double monolin-
gualism (Grosjean, 1989). This Element mainly discusses bilingual
development on its own merit.
The review here is based on a wide array of data collection methods, ranging
from detailed case studies based on parental diaries to parental surveys yielding
information on thousands of children. Aside from parent reports, studies may
rely on experiments, standardized language tests, specially designed tasks, or
direct observations of children’s language use. Studies report on individual
children’s language use or, as has been much more often the case of late, on
levels of language use or language behavior that are averaged across groups of
children (i.e., group studies).
Terms in this Element that are ambivalent as to whether they refer to ethni-
city, citizenship, or language refer to language unless otherwise indicated. The
5
HaBilNet [Screen name]. (2020, May 6). HaBilNet Class 1: Trajectories for Early Bilingualism
[Video]. Vimeo. https://vimeo.com/415653440
6
HaBilNet [Screen name]. (2020, May 9). HaBilNet Class 2: BFLA Compared to ESLA [Video].
Vimeo. https://vimeo.com/416621250
6 Child Development
term “family” designates any private household made up of at least one child
under age 12 and one adult who is responsible for the child. The term “parents”
refers to the adult(s) who is/are part of such a family. This Element follows
Bornstein’s transactional and dynamic perspective on the family, according to
which “[c]hild and parent bring distinctive characteristics to, and each changes
as a result of, every interaction; both then enter the next round of interaction as
changed individuals” (Bornstein, 2019, p. 279). At the same time, macrosystem
patterns of beliefs and values influence the interpersonal experiences individ-
uals have (Bornstein, 2009). The influence of such macrosystem patterns is
particularly relevant to bilingual children and their families. Societal attitudes
toward early bilingualism and the languages involved affect stances and behav-
iors toward bilingual children and their families and may affect their socio-
emotional well-being (De Houwer, 2020a). Educational approaches in (pre-)
schools play a large role in this dynamic.
In addition to describing bilingual children’s oral language learning trajec-
tories, this Element examines the nature of children’s bilingual language
learning environments. Aside from aspects such as the quantity of child-
directed speech, these environments include parental conversational practices
and effects of the aforementioned language-related attitudes. A separate sec-
tion (Section 5) briefly examines the possible role of socioeconomic status in
bilingual development.
Children’s language learning environments determine the degree to which
they and their families experience harmonious bilingualism; that is, “a subject-
ively neutral or positive experience that members of a family in a bilingual
setting have with aspects of that setting” (De Houwer, 2020a, p. 63). This
Element concludes with a summary of the main points and a plea for more
research attention to harmonious bilingualism.
In brief, this Element focuses on the oral language development of three kinds
of bilingual children: (i) BFLA children acquiring a Non-Soc-L (Language A)
and a Soc-L (Language Alpha) from birth (followed from infancy to middle
childhood); (ii) ESLA children acquiring a Non-Soc-L as their L1 and a Soc-L as
their L2 (traced from early to middle childhood); and (iii) SLA children acquiring
a Non-Soc-L as their L1 and a Soc-L as their L2 in middle childhood.
2 Becoming Bilingual in Infancy: Focus on Bilingual First

Language Acquisition
Infants may hear two languages from birth within the family. Infants may also
be reared bilingually in spite of not living in a bilingual family. They may hear
one language at home but may start to be regularly addressed in an L2 well
before the second birthday, often through childcare outside the home, and are
thus growing up in an ESLA setting. There are virtually no reports on ESLA
infants’ language development (but see Pavlovitch, 1920; Vihman, 1999).
Hence the current section focuses exclusively on infants growing up with two
languages from birth.
The section starts by examining what it means to be born into a bilingual
family, and focuses on aspects of the language input to infants, that is, on
what they hear. We know about input through written records kept by parents
(often in the form of diaries), audio and/or video recordings, and parental
questionnaires. Infants must learn to make sense of what they hear. How
infants with bilingual input from birth do so is discussed in Section 2.2 on
early speech perception. Most research here relies on ingenious experiments.
Infants must learn to categorize sounds into units that are meaningful in each
of their two languages. Learning how to do this is part of their phonological
development, that is, the development of the sound system of a particular
language. Section 2.3 goes on to describe the first steps in BFLA infants’
word comprehension. Learning to understand words is part of infants’ lexical
development and has been studied through observation and sometimes
experiments. The size of infants’ comprehension vocabulary is assessed
through parental questionnaires. The best known are the MacArthur-Bates
Communicative Development Inventories (CDIs; Fenson et al., 1993) and
their different language versions. CDIs were first developed in American
English, but currently exist in about 50 languages and varieties (see the
official CDI website).7 CDIs are standardized report instruments asking
caregivers to tick off on a list which words or phrases children understand
and/or say. For children between 16 and 30 months, CDIs also ask about early
word combinations. CDIs are important both in research and in clinical
practice. For many languages CDI norms8 have been established that allow
clinicians to decide whether an individual child is developing as expected or
not. BFLA infants’ use of words and word combinations is the subject of
Section 2.4. That section also discusses BFLA infants’ phonological devel-
opment in production.
Both in early comprehension and in production BFLA infants may develop
each language at a different pace. This uneven development is the subject of
Section 2.5. Section 2.6 examines factors that may influence early bilingual
development in BFLA. A brief summary concludes Section 2.
7
MacArthur-BatesCDI. (n.d.) [website] https://mb-cdi.stanford.edu/
8
Vocabulary Norms. (n.d.). Word Bank [website]. http://wordbank.stanford.edu/analyses?
name=vocab_norms
8 Child Development
2.1 Born into a Bilingual Family

Once parents start talking to newborns, they may use both Language A and
Language Alpha, creating a bilingual language input environment from the very
start. BFLA newborns finding themselves in a bilingual family may hear each
parent speak both languages to them. A second possibility is that one parent
speaks both languages to newborns and the other parent(s) just a single lan-
guage. Alternatively, each parent may address the newborn in just a single
language (OPOL, the (in)famous “one person, one language” setting). The first
two patterns occur about equally frequently; the OPOL pattern occurs least
often (De Houwer, 2007; Yamamoto, 2001). Regardless of the language(s)
parents use to address them, babies may overhear parents address each other
in a third language (or more). The variation is large.
Language input patterns that infants experience from their parents are com-
plemented by those from other people both inside (e.g., siblings) and outside
(e.g., family friends) the home. Added to this variation comes variation in the
number of people children come into regular contact with.
In addition to the variation in the languages BFLA babies are hearing and
whom they are hearing them from, there is wide interindividual variation in how
parents verbally engage with infants. Parents may speak to babies a lot, or less
so. They may speak clearly, or less so. They may use a lot of typical infant-
directed speech (IDS) with exaggerated intonation patterns, short utterances,
and frequent repetitions, or less so. Listen to excerpts of American English IDS9
that contain stimuli used for, among others, Byers-Heinlein et al. (2021). Listen
to typical Portuguese IDS10 (and young BFLA infant vocalizations) in the
Portuguese–Swedish MCF Corpus11 collected by Madalena Cruz-Ferreira
(Cruz-Ferreira, 2006, is an in-depth study of the BFLA children featuring in
this corpus). Parents may be verbally quite responsive to babies (illustrated in
this video of a father interacting with his infant),12 or less so. Parents may
regularly read books and enact rhymes and songs with babies, or less so.
To what extent variability on parameters of parent–infant verbal engagement
has to do with parents’ bilingual status or with their status as being part of
a bilingual or monolingual family has hardly been explored. Parents of bilingual
and monolingual infants did not differ in the modalities of action, language, and
9
Audio Stimuli. (n.d.). Google Drive [website]. https://drive.google.com/drive/u/0/folders/
0B4NwkcR_udMNUmdOVzZVVkgtZFE
10
https://media.talkbank.org/childes/Biling/MCF/Karin/020403.mp3
11
MCF Bilingual Corpus. (n.d.). Talkbank [website]. https://childes.talkbank.org/access/Biling/
MCF.html
12
Acery [Screen name]. (2019, June 5) Dad Has Full Convo With Baby [Video]. YouTube. www
.youtube.com/watch?v=0IaNR8YGdow
gesture during parent–child interaction with their 14-month-olds (Gampe et al.,

2020). Mothers in bilingual and monolingual families addressed the same
amount of speech to their 13- and 20-month-olds (De Houwer, 2014).
However, within-group variability was large, with some monolingually reared
infants having very silent mothers, and some bilingually reared infants having
very talkative mothers, as well as the other way around (Orena et al., 2020,
likewise found large variability among bilingual families in the number of
words addressed to BFLA infants). Although it is often claimed that bilingual
infants hear less of each language than monolinguals of their single one, the
wide variability within each parent group as well as the lack of intergroup
differences render this assumption doubtful (De Houwer, 2018b). Assumptions
of bilingual–monolingual differences in frequency of input in a language are
further complicated by possibly large variation within bilingual families in
global use of a particular language, with parents speaking Language Alpha far
more frequently during trips to the “home“ country (Leopold, 1939–1949;
Slavkov, 2015), or with families hosting grandparents who speak Language
Alpha for weeks or months at a time (Leist-Villis, 2004). Such changes in
environment directly affect the amount of speech that bilingual infants hear in
each language.
Van de Weijer’s (1997) acoustic study is unique in presenting analyses of
intonation patterns in speech addressed to a BFLA infant and in comparing that
IDS to adult-to-adult speech (ADS) by the same adults. Data were collected in
the home for about 90% of the time that the Dutch–German infant was awake
between 6 and 9 months of age (720 hours in total). A total of 4,376 utterances
produced by mother, father, and a regular babysitter over a selection of 18 days
were the basis for analysis. Compared to ADS, in IDS adults spoke much more
slowly, used a much higher pitch, larger pitch variations, and more simple
intonation contours that made utterance boundaries quite clear (van de Weijer,
1998, furnishes more details), confirming findings for adults in monolingual
families. Van de Weijer’s analyses did not focus on the family’s bilingual nature.
Focusing mainly on maternal speech, De Houwer (2009, pp. 121–123) analyzed
information in van de Weijer (2000, 2002) about IDS to the infant and her sister,
who was 2 years older. On average the mother spoke about twice as often to her
toddler than to her infant, suggesting that IDS speaking rates of parents in
bilingual families change as a function of children’s ages, as was later confirmed
in longitudinal studies of maternal speech to BFLA 13- and 20-month-olds (De
Houwer, 2014; Song et al., 2012).
A study of language input within two days in the lives of 58 bilingual (at least
46 were likely BFLA) 10- to 12-month-olds in Paris found that many infants
heard both French and one or more of 16 other languages within the same half-
10 Child Development
hour, rather than just a single language (Carbajal & Peperkamp, 2020). About
half the families followed an OPOL approach. Regardless of parental input
patterns, infants encountered more people who spoke French to them than their
other language.
Parents in both bilingual and monolingual families may address infants in
a language they themselves learned later in life. They may speak that language
with a “foreign accent” (that is, an accent influenced by a language learned
earlier in life). The chance that parents in bilingual families speak a language
with a “foreign accent” is high. If they speak that language to infants, their use
of IDS may show traces of a “foreign accent” as well. Fish et al. (2017)
demonstrated how Spanish–English bilingual parents pronounced English
words addressed to infants with both Spanish- and English-like characteristics.
Fish et al. suggested that the use of specific bilingual characteristics of IDS may
have implications for bilingual infants’ early speech perception, the topic of the
next section.
2.2 Phonological Development: Bilingual Infant Speech Perception

For bilingually reared infants to gain entry into each of the two languages they
are hearing, they must be able to pay auditory attention to the way people
verbally engage with them. Although BFLA infants can learn from any kind of
language addressed to them, the special features of IDS capture infants’ atten-
tion to speech. BFLA infants (N = 333) from various countries preferred to
listen to IDS rather than to ADS (Byers-Heinlein et al., 2021). As Ronjat noted
in 1913, BFLA infants not only use auditory but also visual cues as present on
a speaker’s face to help them process and distinguish the dual language input
they receive (Weikum et al., 2007).
Newborns born into either bilingual or monolingual families are able to
globally distinguish among languages (Byers-Heinlein et al., 2010).
Languages differ widely in their use of particular sound patterns, that is, in
their phonology. As part of phonological development, infants gradually learn
to order the multitude of speech sounds they hear into perceptual categories that
are relevant to the language(s) they are learning. They simultaneously learn to
ignore perceptual differences that are of no consequence to meaning creation in
the language(s) they are learning. BFLA infants must keep an “open ear,” so to
speak, toward a larger variety of speech sounds than monolinguals. They must
develop sufficiently separate perceptual categories relevant to each input lan-
guage to be able to learn words in each language.
Most words are independent units whose meaning relies on a particular
combination of sounds. Languages differ widely in what kinds of sound
patterns can form words and which are relevant to shaping word meaning.
Most languages in the world use lexical tone, (i.e., pitch variations within the
word unit) to help distinguish among words. For example, “ma” in Mandarin
means <mother> if pronounced without pitch variation and <horse> if pro-
nounced with falling and rising intonation. In non-tone languages like English
and Dutch, words do not change meaning as a function of whether they are
said in a monotone, with rising intonation, or with falling intonation: “ma”
means <mother> in both Dutch and English regardless of the intonation
pattern. BFLA infants acquiring non-tone languages hence do not need to
pay attention to lexical tone (Liu & Kager, 2017). However, if one of the
languages they hear relies on lexical tone for meaning distinctions, infants
need to pay attention to word-level variations in pitch for that language (but
not for a non-tone language they may be acquiring simultaneously). In learn-
ing new words, 12- to 13-month-old Mandarin–English infants relied on
lexical tone in a Mandarin context but disregarded tone as a clue in an
English context (Singh et al., 2016).
Words consist of one or more phones. Phones are speech sounds produced
through one articulatory position. For example, “ma” consists of the phone
[m], produced by closing the mouth and vibrating the vocal cords, followed by
the phone [ɑ] or [a], produced by opening the mouth fairly widely and
vibrating the vocal cords (the difference between [ɑ] and [a] depends on the
position of tongue and lips). When the air flow from the vocal cords is
constricted, people produce a consonant (like [m]); they form vowels (like
[ɑ]) when the air flows freely (find out more about phones through this IPA
vowel chart with audio).13
Languages differ in the type, number, and order of phones they use to make
words. They also differ in the extent to which two phones contrast with each
other. If they contrast with each other, the phones are phonemes. If they do not,
they are allophones. Which phones contrast with which other ones depends on
the language. Consider [w] and [v], both produced by pouting the lips a bit and
vibrating the vocal cords. In [w] the air flow is fairly unimpeded. In [v] it is
somewhat obstructed by tongue and lips. In German [w] and [v] do not represent
a phonemic contrast but are allophones: When they are used in a word that is
otherwise the same, there is no meaning difference. Thus, [wIr] and [vIr] sound
the same in German: They mean “we” under either pronunciation. Infants
learning German need not learn to distinguish between [w] and [v]. However,
infants learning English do need to learn to distinguish between the two. That is
13
IPA Vowel Chart With Audio. (n.d.). Wikipedia [website]. https://en.wikipedia.org/wiki/
IPA_vowel_chart_with_audio
because using either creates a meaning difference: “a wet vet” is funny but “a
vet wet” just doesn’t make any sense (Germans will often fail to hear the
difference). In English, [w] and [v] build the phonemic contrast /w/-/v/.
Learning to distinguish phonemic contrasts in each language is an important
step toward deciphering the meanings of words. Studies of BFLA infants’
speech perception in the first year of life have primarily focused on the early
perception of consonant contrasts (Burns et al., 2007). Studies on the early
perception of vowel contrasts (Sundara & Scutellaro, 2011) are rarer.
Phonological perceptual learning continues well into the second year (Singh
et al., 2017). There has also been attention to BFLA infants’ perception of
lexical stress, another feature that is differentially relevant depending on the
language (Bijeljac-Babic et al., 2016).
Upon hearing utterances containing several words, infants need to learn
where a word begins and ends. Statistical learning and perceiving overall
pitch variations within an utterance likely play a large role in this process
(Höhle et al., 2020). BFLA infants are able to distinguish both monosyllabic
(Bosch et al., 2013) and bisyllabic (Polka et al., 2017) words in a speech stream
well before the first birthday, and in some cases by 6 months. The growing
literature on perceptual aspects of new word learning and familiar word recog-
nition in the second year of life shows that BFLA children differ in the extent to
which they use detailed phonetic information to process words, partly as
a function of the specific contrasts investigated (Havy et al., 2016), and partly
as a function of children’s specific sensitivities to acoustic information (Fennell
& Byers-Heinlein, 2014).
Even though the highly complex experimental research on BFLA infants’
developmental paths in early speech perception is still quite “fragmentary”
(Höhle et al., 2020), likely by the end of the first year BFLA infants will have
learned to appropriately categorize some language-specific aspects of the
phonological systems of the two languages they have been hearing. Whether
this means infants are approaching their languages as separate phonological
systems is, however, an open question (Höhle et al., 2020, and chapters 8 and 9
by Byers-Heinlein in Grosjean & Byers-Heinlein, 2018, offer more detailed
treatments of bilingual infant speech perception).
Early speech perception abilities develop concurrently with infants’ grow-
ing ability at interpreting contexts and people’s communicative intentions.
These combined abilities are the main basis for the emergence of early lexical
comprehension. The ability to understand words has been experimentally
documented for monolinguals as young as 6 months (Bergelson & Swingley,
2012). The next section discusses early language comprehension in BFLA
infants.
2.3 Bilingual Infant Language Comprehension

Depending on the language pair BFLA infants are hearing, there may be a lot of
overlap between words across both languages. Words that sound and mean the
same across two languages are cognates. Parents talk about quite concrete
things to infants that are relevant in both languages. Thus, infants also hear
translation equivalents (TEs). TEs are words from each language that differ in
form but mean more or less the same thing, such as Spanish “comida” and
German “Essen” (food, eating). In contrast to TEs, infants may hear some things
discussed only in a single language. For instance, if a father reads Turkish books
about zoo animals with his Turkish–German son but the German-speaking
mother does not mention animals, the child only has the chance to learn
names for zoo animals in Turkish. In addition to the degree of lexical variation
and overlap between languages A and Alpha the absolute number of words and
the number of different words that BFLA infants hear in each language are
likely different, with concomitant differences in relative exposure to words in
each language (De Houwer, 2009).
The complex nature of BFLA infants’ total lexical input may be seen as an
arduous language learning environment. One might expect BFLA infants to
take longer to start understanding new words and to build up their lexical
comprehension repertoire more slowly than children growing up in a less
variable, monolingual environment. One would therefore expect monolingual
infants to have a larger comprehension vocabulary size than BFLA peers. So
far, there is no evidence for slower bilingual development; in fact, there is
evidence suggesting that a bilingual environment from birth boosts word
comprehension.
The overall timing of first word comprehension by BFLA infants is no
different from that reported for monolinguals (Clark, 1993). BFLA infants
may understand words in both languages well before the first birthday. Three
Portuguese–Swedish children first responded to their names at ages 4, 5, and 7
months (Cruz-Ferreira, 2006, p. 146). English–German Hildegard understood
her name at 6 months, understood several English words by 8 months, and
understood directions in both languages by 8 months (Leopold, 1939–1949).
Cruz-Ferreira’s (2006, pp. 146–147) Portuguese–Swedish children responded
to simple commands at 7 months, and responded nonverbally but appropri-
ately to routine questions such as “where is the clock?” by 8 months, thus
showing some level of comprehension. Welsh–English 11-month-olds were
able to recognize both Welsh and English words (Vihman et al., 2007). By the
first birthday, BFLA infants routinely understand words in each of their
languages.
Both clinical and research interest in early vocabulary size is great because of
its strong relation with later language abilities (Bornstein & Haynes, 1998).
BFLA average comprehension vocabulary size in one particular language does
not differ from that of monolinguals. There were no differences among bilingual
and monolingual infants (average age: 11 months) in the average number of
French words understood (Carbajal & Peperkamp, 2020). Dutch–French first-
born 13-month-olds (N = 31) understood as many Dutch words as 30 demo-
graphically matched Dutch-learning monolingual peers (De Houwer et al.,
2014). Data collected for these same infants at 20 months likewise showed no
bilingual–monolingual differences for Dutch comprehension vocabulary size.
Rather, both ages showed large intragroup variability (De Houwer et al., 2014).
The fact that studies comparing early comprehension vocabulary size for
a single language found no bilingual–monolingual differences suggests that
BFLA infants experience no particular difficulty in learning to understand new
words.
We do find bilingual–monolingual differences when infants’ total vocabulary
sizes are compared. However, monolinguals perform far worse than bilinguals.
BFLA infants’ total vocabulary size includes both languages; monolinguals’
vocabulary covers just a single language. On average, 13-month-old bilinguals
in De Houwer et al. (2014) understood 60% more words than monolingual
peers, with the top-performing bilingual understanding 564 words compared to
just 358 understood by the best monolingual. Legacy et al. (2016) likewise
found larger total comprehension vocabulary in 50 French–English infants aged
16 to 18 months compared to monolingual French-speaking peers, with bilin-
guals understanding 39% more words, and the top-performing bilingual under-
standing 693 words compared to the best monolingual’s 387 words understood.
In neither study were the best monolinguals performing at ceiling.
One factor driving BFLA infants’ word learning may lie in TE comprehen-
sion. De Houwer et al. (2006) examined the understanding of 361 TE pairs
across Dutch and French in the same 13-month-olds as studied in De Houwer
et al. (2014). All infants understood TEs, words from two languages that meant
the same thing (this does not imply that infants realized that one word was the
translation of another). All infants also understood just single members of TE
pairs. There was large interindividual variability in the number of TE pairs
understood. Infants who understood more of the 361 meanings represented in
the TE pairs compared to others tended to understand them in both Dutch and
French rather than in just a single language. Learning to understand more words
in early bilingual development thus appears to be (partly) connected with
learning to understand the TE for an already known word. Legacy et al.
(2016) similarly reported a correlation between comprehension vocabulary
size and the proportion of TE pairs that their somewhat older sample
understood.
We do not know how BFLA infants who understand (or produce, see Section
2.4.4) both members of a TE pair interpret each of these members, nor whether
children understand that meanings overlap. Yet the term “conceptual vocabu-
lary” (Pearson et al., 1993) is widely used not only in the research literature but
also in clinical practice. It refers to vocabulary size counted in so-called
“concepts,” the number of lexicalized meanings that children know, abstracting
from actual word forms across two languages in a TE pair (for monolingual
children, “conceptual vocabulary” is identical with vocabulary size counted in
words). To what extent such abstraction is valid is up for debate. However,
studies comparing bilingual and monolingual infants’ “conceptual vocabulary”
comprehension sizes have found no intergroup differences (De Houwer et al.,
2014; Legacy et al., 2016).
We know hardly anything about bilinguals’ early semantic development, or
about what drives this development. Experimental studies such as Byers-
Heinlein’s (2017) with 9- to 10-month-olds, Henderson and Scott’s (2015)
with 13-month-olds, and Kalashnikova et al.’s (2018) with 18- and 24-month-
olds are starting to address some of the word-learning heuristics that bilingual
infants use to learn to understand words.
2.4 Bilingual Infant Language Production

2.4.1 Babbling
Soon after BFLA infants start to understand words (or concurrently), they start
to produce repeated syllables. Before the first birthday, these syllables consist of
a consonant followed by a vowel. The use of repeated syllables by babies is
known as babbling. When infants are a bit older they start to vary the syllables
within one breath. Babbling appears to have little, if any, communicative intent.
Ronjat (1913) was the first to closely describe babbling in a BFLA infant and
noted the importance of prosody (review in De Houwer, 2009, pp. 167–171).
Ronjat did not discover any specific bilingual aspects of babbling, but in 1913 he
did not have the technological tools for acoustic analyses that are available now.
Such analyses showed that Spanish–English 12-month-olds used different pros-
odic patterns in babbling depending on whether they interacted with a Spanish or
English interlocutor (Sundara et al., 2020). In doing so, infants reflected prosodic
characteristics of, respectively, Spanish and English. These findings confirm
earlier studies (Andruski et al., 2014; Maneva & Genesee, 2002).
Babbling is generally fairly short-lived, but there is wide variation among
children in how long they continue to babble. Some BFLA infants stop babbling
once they start saying their first “real” words, whereas others continue to babble.
In order to communicate with others, words are crucial.
2.4.2 First Words

Early word comprehension predates early word production in most children.
Toward the first birthday BFLA infants start to say syllables or combinations of
syllables that parents identify as attempts at words (i.e., forms with a meaning
attached to it). This BFLA milestone occurs around the same age as for mono-
linguals. Some BFLA (and monolingual) infants initially produce longish
uninterpretable utterances with sentence-like intonation patterns concurrently
with single word-like forms (listen to such uninterpretable utterances in this
conversation).14
It is rarely clear what language BFLA infants’ first words are aiming at. That
is because the structure of children’s first words is highly individual. However,
there are commonalities across children and languages (Vihman, 2016).
2.4.3 Phonological Development: Learning to Produce Sounds

As they add words to their production repertoire, BFLA infants have to learn to
use the appropriate sounds. Learning to use the right sounds entails selecting the
correct consonant and vowel phonemes (Section 2.2) for each syllable, putting
phonemes and syllables into an order that makes sense in a particular language,
controlling one’s voice to produce appropriate intonation patterns, and much
more. All this pertains to the production side of phonological development.
Phonological development in production is a long-drawn-out process that takes
many years.
In comparison to adult pronunciations, both bilingual and monolingual
infants delete or substitute sounds, shorten words, or repeat syllables. These
phonological processes make it difficult to decide what word infants are trying
to articulate. For BFLA children, phonological processes additionally make it
hard to decide what language infants are using, and hamper attempts to decide
whether children are developing language-specific phonological systems (the
question of whether BFLA infants develop two separate linguistic systems has
been of great interest to scholars).
As reviewed in De Houwer (2009), there is wide interindividual variation
among BFLA children in the development of phonology. As children grow
more mature and are better able to integrate what they know about how others
14
Coldquads [Screen name]. (2013, October 3). Baby’s Conversation with Grandmother [Video].
YouTube. www.youtube.com/watch?v=9gsjGAW18rk
speak and how they hear themselves speak, separate phonological systems may
start to slowly emerge (Vihman, 2016).
2.4.4 Lexical Repertoire and Vocabulary Size

Not only does word comprehension predate word production, but like mono-
linguals, BFLA infants understand more words than they say (De Houwer,
2009). On average, 31 Dutch–French 13-month-olds understood 14 times as
many words as they said (De Houwer et al., 2014). Whether the extent of this
comprehension-production vocabulary gap persists is not known, but it is
a common experience for people everywhere to understand more than they
say.
By 12 months most BFLA (and monolingual) infants have started to speak. If
there is no somewhat recognizable word production by this age, children’s
hearing should be checked. BFLA infants gradually start saying more different
words (Porsché, 1983; Ronjat, 1913). Some bilinguals experience a vocabulary
spurt: They suddenly start saying several new words every day. Others show
a generally much more gradual increase in the number of different words said
(Pearson & Fernández, 1994), but in all children an increase is expected. The
average total number of different words that Dutch–French infants had in their
production repertoire jumped from an average of 17 at 13 months to 254 at 20
months (De Houwer et al., 2014). Words come but also go again: Previously
produced words may no longer be used. Patterns of bilingual early lexical
development reflect those found in monolinguals, except that, according to
CDI norms for several languages,15 the total number of words that most 20-
month-old monolinguals say is much lower than the average of 254 found in De
Houwer et al. (2014).
There is extensive variation among infants in the speed with which they learn
to say new words (and hence in their vocabulary size). Case studies show one
BFLA child saying 100 words by 17 months, while another child did so only 3
months later (Vihman, 2016). The total number of words individual BFLA
infants in De Houwer et al. (2014) produced varied between 0 and 82 at 13
months, and between 14 and 1,234 (!) words at 20 months. Legacy et al. (2018)
reported similar ranges of variation. Cote and Bornstein (2014) reported nar-
rower ranges for 20-month-olds (Spanish–English: 6–363, Japanese–English:
2–429, Korean–English: 13–590), but there was wide variation within each
group. Notably, there is great variation among monolingual 20-month-olds as
well (English: 4–487, Spanish: 1–402, Korean: 3–253, Cote & Bornstein, 2014;
15
name=vocab_norms
Dutch: 19–531, De Houwer et al., 2014; CDI norms for monolinguals acquiring
various languages likewise show vast ranges of variation).16
Bilinguals may start saying words first in Language A, and only quite a bit later
in Language Alpha. Not all BFLA infants produce words in both languages
(13-month-olds, De Houwer et al., 2014; 15- to 19-month-olds; Legacy et al.,
2018). Both these longitudinal studies found that all infants were producing
words in both languages at the later age data were collected for (20 months
[N = 31], De Houwer et al., 2014; between 20 and 26 months [M = 24 months,
N = 38], Legacy et al., 2018). Bilinguals may also start saying words in both
languages at the same time. After age 1.5 most BFLA infants say words in each of
their languages (De Houwer, 2009). Children produce different numbers of words
in each language (De Houwer & Bornstein, 2016a) and the gap between lan-
guages may widen with age (Legacy et al., 2018).
BFLA infants start producing TEs at different ages. They also differ in how
many TEs they produce and in what proportion, relative to total production
vocabulary size (David & Li, 2008; Pearson et al., 1995).
It is often claimed that bilingual children generally have a smaller vocabulary
than monolinguals. So far there is no evidence to support this. Rather, CDI and
CDI-like studies comparing production vocabulary size for young bilinguals
and monolinguals have had mixed results. Because early lexical learning is very
much linked to the cumulative number of words heard (Head Zauche et al.,
2017), comparisons only make sense if the overall time for learning words has
been equal. Hence the following brief review only considers studies of children
with either bilingual or monolingual input from birth (because CDI studies may
report on children up to around 2;6 (years;months), toddler findings are
included here).
A first type of comparison considers children’s total production vocabulary
(always used as a basis for monolinguals). For bilinguals, this covers words in
both languages combined. Studies failing to find any bilingual–monolingual
differences are De Houwer et al. (2014; Dutch–French at 13 and 20 months),
Pearson et al. (1993; Spanish–English, 16- to 27-month-olds), Legacy et al.
(2018; French–English, 21–26 month-olds, M = 24 months), Hoff et al. (2012;
Spanish–English; 22, 25, and 30-month-olds), Marchman et al. (2010; Spanish–
English; M = 30 months), and Poulin-Dubois et al. (2013; French–English, 24-
month-olds). When the French–English children in Legacy et al. (2018) were
younger (15 to 19 months, M = 17 months), they produced more words than
monolinguals. The total production vocabulary size of Spanish–Catalan infants
16
name=vocab_norms
between 12 and 24 months far exceeded that of monolinguals in either language

(Águila et al., 2007). In contrast, Spanish–English 20-month-olds had smaller
total vocabulary sizes (Cote & Bornstein, 2014).
A second type of bilingual–monolingual lexicon size comparison considers
a single language for bilinguals (representing only part of their production abil-
ities) and compares this single language with monolinguals’ production vocabu-
lary in the same language. Studies finding no differences include De Houwer et al.
(2014; Dutch at 13 and 20 months), Legacy et al. (2018; French and English,
M = 17 months; the language with the most words, M = 24 months), and Pearson
et al. (1993; English between 16 and 27 months). In contrast to Pearson et al.
(1993), Cote and Bornstein (2014), Hoff et al. (2012), and Marchman et al. (2010)
reported smaller English vocabulary sizes for Spanish–English bilinguals (ages 20
months, 22, 25, and 30 months, and 30 months, respectively). Marchman et al.
found the same for Spanish. The 24-month-olds in Legacy et al. (2018) did worse
than monolinguals in the language bilinguals knew fewer words in. Poulin-Dubois
et al. (2013) found that 2-year-olds speaking just English or French produced more
English or French words, respectively, than bilinguals.
These mixed results for bilingual–monolingual production vocabulary size
comparisons are likely attributable to: (i) the ages at which infants were
compared to each other; (ii) the way parent report data were collected; (iii)
differences among parent report instruments; (iv) whether studies compared
raw scores or CDI norm percentiles; (v) the extent to which bilinguals and
monolinguals were of similar socioeconomic status (see Section 5 below);
and (vi) for single-language comparisons, the high variability among BFLA
infants in the number of words produced in each language and whether
studies made sure to consider only children’s strongest language (see
Section 2.5 below). Re (ii), for BFLA infants it is particularly important to
get as comprehensive information on children’s abilities as possible. Mothers
whose children speak Language A to them but not Language Alpha will not
be able to assess children’s lexical repertoire in Language A. Thus, both
researchers and clinicians should ask all regular caregivers to complete CDIs
in the language(s) they address to children and/or children speak to them (De
Houwer, 2019a).
Case studies (Cruz-Ferreira, 2006; Leopold, 1939–1949; Porsché, 1983)
have traced some aspects of semantic development in BFLA infants (De
Houwer, 2009), that is, about how children construct meanings and use words
to express them. Holowka et al. (2002) investigated the lexical domains of three
BFLA children’s early words. However, we know very little about semantic
aspects of BFLA infants’ word use (DeAnda et al., 2016b). Research such as
that by Jardak and Byers-Heinlein (2019) is sure to help change that.
2.4.5 Word Combinations

BFLA (and monolingual) infants’ first words are usually produced in isolation
(not as part of a longer utterance). These so-called “holophrases” often appear to
express big meanings and several pragmatic functions. For instance, for several
months German–English Nicolai used the form [das] to ask a question, confirm
the correctness of something, express that he agreed with something, express
excitement, and several other functions (Porsché, 1983). Eventually, BFLA
infants start combining two words into a single utterance, as in “mamã peix”
(mommy fish), said by a Portuguese–Swedish 20-month-old while her mother
was drawing a fish (Cruz-Ferreira, 2006, p. 155). The earliest age reported is 15
months, but there is wide variation in the ages at which BFLA children start to
combine words, as there is among monolinguals (De Houwer, 2009, p. 257;
Patterson, 1998; Vihman, 2016). At any rate, the second half of the second year
sees most BFLA (and monolingual) infants combining words. BFLA infants
combine two words from Language A, two words from Language Alpha, or
a word from each language. Utterances combining words from the same
language are unilingual utterances. Utterances combining words from two
languages are mixed utterances. Like first words, early two-word combinations
do not necessarily appear in both languages at the same time (De Houwer, 2009;
Lindquist & Gram Garmann, 2021).
The ability to combine two words into short utterances is a major milestone in
early language development for bilinguals and monolinguals alike. If by
the second birthday BFLA infants have not started combining words in at
least one language, or in a mixed utterance, parents would do well to have
children’s hearing checked. Many BFLA infants are already able to produce
short “real” sentences before the second birthday (De Houwer, 2009; see
Section 3.1.2).
Production vocabulary size and word combinations are linked. BFLA infants
say about a total of 50 different words before they start combining words,
regardless of language (Patterson, 1998). Marchman et al. (2004) and Conboy
and Thal (2006) found that the more English words bilingual infants (mostly
BFLA, between about 1;6 and 2;6) produced, the longer their English utterances
were. The same was true for Spanish. These relations between vocabulary size
and utterance length held within each language, rather than across languages
(see further Serratrice, 2019).
Word combinations increase the length of children’s utterances. Utterance
length is an important measure of early child language complexity and thus
level of development. English–Spanish 17- to 30-month-olds (M = 24 months;
N = 113; most of them BFLA children) were on average able to say equally long
utterances in each of their languages, but there was large variability within
each language (1 to 9 English words per utterance, M = 2.3; 1 to 7 for Spanish,
M = 2.6; Marchman et al., 2004).
2.4.6 Language Choice

By age 1.5, BFLA infants produce words that do not clearly belong to a single
language, either because they still sound too immature and it is impossible to
say which language they are aiming at, or because there is lexical overlap
between their two input languages. Children also produce words that clearly
belong to Language A, and words that clearly belong to Language Alpha. For
words that have a TE in the other language, children may say both members of
the TE pair, or just one.
At any one time, bilinguals have to select among the words they know: They
cannot pronounce two words at once. Bilinguals are not necessarily aware of
which language they select, but language choice is not random (De Houwer,
2019b). BFLA infants’ language choice is not random, either. From 1;6
onwards, BFLA infants often choose words and word combinations in the
same language that their interlocutor speaks with them (De Houwer, 2009, pp.
141–145, 238–241). Infants are able to do so with both familiar (Köppe, 1996;
Nicoladis, 1998; Sinka & Schelletter, 1998) and unfamiliar (Genesee et al.,
1996) interlocutors. This ability may be related to BFLA infants’ early ability to
take another person’s perspective (Liberman et al., 2017). However, BFLA
infants do not always adjust to their interlocutor’s language choice, and there
are BFLA infants who hardly ever do (Section 2.6). For BFLA infants who are
able to produce words in both Language A and Language Alpha, the “wrong”
lexical choice may occur because they do not know a particular TE in the other
language.
2.5 Uneven Bilingual Development in Infancy

Uneven development refers to cross-linguistic differences in the pace of
bilingual children’s language development and thus in their comparative
proficiency in each language. Already at the earliest stages in comprehension
BFLA infants do not necessarily develop each language at the same pace.
Between ages 13 and 15 months, German–English Hildegard lost her prior
ability to understand English and continued to understand only German
(Leopold, 1939–1949). This occurred after her family had traveled to
Germany from the United States when Hildegard was 11 months and she
stopped hearing any English. It took two weeks after the family’s return to the
United States before Hildegard started understanding a few English words
again (summary in De Houwer, 2009, p. 196). For some BFLA 12-month-

olds, individual Spanish comprehension scores differed widely from those in
English (Sundara et al., 2020, Appendix A). Only a small minority of 30
Dutch–French 13-month-olds understood about equal numbers of words in
each language (De Houwer & Bornstein, 2016a). Most understood more
words in one language than the other. Some infants in this sample understood
many words in a single language but did not understand their translation in the
other one (De Houwer et al., 2006).
Also in production BFLA infants may show cross-linguistic differences in
the pace of development (Section 2.4.4). The timing of first word production in
Language A may lag behind that in Language Alpha. Infants may say more
words in Language A than in Language Alpha, or about equal numbers of words
in each language. These balance patterns may be in flux over time. Two thirds of
30 Dutch–French BFLA infants showed a change in balance patterns for word
production from age 13 to 20 months (De Houwer & Bornstein, 2016a). Six
children showed a complete reversal and said more words in Language A at 13
months, but more words in Language Alpha 7 months later. Such drastic
changes may help explain some of the mixed findings for bilingual–monolin-
gual vocabulary size comparisons (Section 2.4.4).
Infants who understand more words in Language Alpha do not necessarily
contemporaneously produce more words in that same language, or the other
way around (De Houwer & Bornstein, 2016a). Both the changes that can occur
in balance patterns for production and the fact that infants may show different
balance patterns for comprehension and production suggest that speaking of
“dominance” in a particular language as a general characteristic may be
inappropriate. Uneven development also applies to word combinations
(Section 2.4.5): Infants may be combining words in just a single language and
not in the other.
2.6 Factors Affecting Bilingual Development in Infancy

There is great variability among BFLA infants in how they learn to process,
understand, and speak two languages. An unsurprising reason lies in a child-
internal factor, namely, in children’s ages: Generally, the older they are, the
higher infants’ levels of performance on a particular language measure.
Perceptual abilities partly depend on age but differ among peers and underlie
some of the variability among BFLA infants: Finnish–Russian infants who were
comparatively better at discriminating a Finnish sound contrast at age 7 months
had higher levels of both Finnish and Russian word production in the second
year of life (Silvén et al., 2014). Garcia-Sierra et al. (2011) found similar results
for Spanish–English infants. BFLA infants also differ in the speed with which
they recognize words, and there likely is a bidirectional relation between the
speed of online word processing and receptive vocabulary growth (Legacy
et al., 2018). How well children are able to remember sounds (their phono-
logical memory skill) affects both languages similarly (Parra et al., 2011).
Variability in early speech discrimination abilities may also be related to
characteristics of the input. Orena and Polka (2019) focused on possible links
between parents’ use of mixed utterances (i.e., utterances with words from two
languages, see Section 2.4.5) and bilingual development: The more parental
mixed utterances 16 8-month-old and 20 10-month-old French–English infants
heard, the better the infants were at segmenting bisyllabic words in either
language, whereas infants whose parents reported using fewer mixed utterances
tended to be good at segmenting words mainly in the language they heard more
often.
Comprehension levels may also be affected by the input: BFLA 10- to 12-
month-olds who heard two languages more often in the same half-hour block of
time had lower French comprehension scores than BFLA peers whose language
input in each language was more separated (Carbajal & Peperkamp, 2020). No
such negative effects were found for the word production of older BFLA
toddlers who frequently heard their languages within the same half-hour
block, however (Place & Hoff, 2016). Instead, the proportion of the estimated
time that BFLA infants hear each language can often help explain why children
produce more words in one language than the other (Cote & Bornstein, 2014;
David & Li, 2008; Legacy et al., 2018; Marchman et al., 2004; Pearson et al.,
1997; Vila, 1984).
French–English infants’ production of TEs across the second half of
the second year correlated with changes in the proportion of exposure to each
input language (David & Li, 2008). In contrast, for somewhat older French–
English infants no such link was found (Poulin-Dubois et al., 2013). As De
Houwer (2009, pp. 197–198) noted, “adults will often tell children what the
translation equivalent of a word in Language A is in Language Alpha. They may
do this by saying the words from each language one after the other.” Parents
may also actively engage infants by teaching them routines across different
languages (watch how parents are encouraging their 16-month-old’s
comprehension of words for the same things in English and Spanish).17
Variations in such TE teaching strategies may help account for variation in
infants’ TE comprehension and production.
17
MisterO [Screen name]. (2011, October 23). Bilingual Language Acquisition [Video]. YouTube.
www.youtube.com/watch?v=68rP2-ecPDM
However, research comparing observed parental engagement with infants in

bilingual families and bilingual development is rare. Andruski et al. (2014)
found that some aspects of bilingual babbling correlated with individual fea-
tures of parental speech. Spanish–English bilinguals (N = 18; likely mostly
BFLA) with greater exposure to IDS (Section 2.1) in English one-on-one
parent–child interaction at ages 11 to 14 months had larger English vocabulary
sizes at 24 months (Ramírez-Esparza et al., 2017). The same held for Spanish.
The more bilinguals between 21 and 27 months were read aloud to in Spanish,
the more Spanish words they were able to say (Patterson, 2002). The same was
true for English. These effects were significant even when children’s overall
degree of exposure to each language was controlled for. Indeed, book reading
can be of particular importance to BFLA infants’ word learning (Cruz-Ferreira,
2006, pp. 182–184).
Finally, but importantly, parents socialize infants into (i) speaking only one
particular language with them or (ii) speaking both languages or any language
with them (Lanza, 1992). Parents do so through their discourse behavior in
conversation. Infants may speak a language parents are not using with them in
a particular conversation. In response, parents may use clarification requests
(“Minimal Grasp Strategy”) or requests for confirmation and/or correction as in
“did you mean X?”, where X is the translation of what the infant said
(“Expressed Guess Strategy”). These monolingual discourse strategies encour-
age infants to repeat their original utterance in the other language and help to
keep a conversation in a single language only. Parents may also simply repeat
children’s utterances in the other language. This “Adult Repetition” is neutral in
terms of creating the need for the use of a particular language. Bilingual
discourse strategies allow a conversation to take place in two languages. One
is the “Move On Strategy,” whereby parents just continue speaking the language
they were speaking and do not interrupt the conversation after infants used
another language. This conversational response signals to children that their
language choice was fine. The second bilingual discourse strategy consists of
parents themselves switching to the language infants used (Lanza, 1992).
Already in the second year of life BFLA infants are sensitive to their variable
linguistic environment (De Houwer, 2017b) and to parental discourse strategies
(De Houwer & Nakamura, accepted for publication 2021). Within
a conversation, infants whose parents use mainly monolingual discourse strat-
egies speak the same language that parents speak to them. Parents who use more
bilingual discourse strategies tend to have infants who do not. If infants are not
socialized into using a particular language, they may end up not speaking it.
Parental discourse strategies thus have a very important role to play (De
Houwer, 2009; see also Sections 3.6 and 4.6). In turn, infants may influence
parental language choice patterns, to the extent that parents at least occasionally
adapt to their infant’s apparent preference (De Houwer & Bornstein, 2016b;
Lindquist & Gram Garmann, 2021; Mishina, 1999) or give up speaking
a particular language to infants (Eilers et al., 2006), thus following universal
tendencies for interlocutors to accommodate to each other in conversation (De
Houwer, 2019b).
2.7 Summary
By their first birthdays, BFLA infants have managed to sufficiently categorize
the sounds they hear in each of their two input languages to be able to
understand many words in each language. Many BFLA one-year-olds have
also started to say a few words. Their vocabulary size in each language steadily
grows throughout the second year, and when they reach their second birthdays,
most BFLA children are able to combine words into sentence fragments. Many
BFLA 2-year-olds are able to choose the expected language in speaking to
parents.
BFLA infants’ two languages do not necessarily develop at the same pace.
The pace of development in each language relates to how often children hear
each language and to their communicative need to actually speak two lan-
guages. This need is negotiated through parental socialization practices.
3 Bilingualism in Early Childhood: Bilingual First and Early

Second Language Acquisition
As BFLA infants grow into toddlers and preschoolers, their world widens. This
mostly happens through children enrolling in group childcare or more formal
early education classes (henceforth: ECEC, Early Childhood Education and
Care). As Sections 3.1 and 3.5 review, BFLA children’s widening world has
important effects on their bilingual development.
Many monolingual toddlers and preschoolers likewise enroll in ECEC. For
a large proportion of monolingual children, such enrollment represents first
regular contact with a second language (L2) if the main language spoken in
ECEC is different from the L1 children hear at home. Children’s language
environment thus has changed to an Early Second Language Acquisition
(ESLA) setting (Section 1). Section 3.2 describes ESLA children’s early bilin-
gual development.
Sections 3.1 and 3.2 both start with a discussion of children’s developing
language comprehension. Many studies rely on tests like the Peabody Picture
Vocabulary Test (PPVT; Dunn & Dunn, 1997), currently available in several
validated language versions (henceforth, the term PPVT refers to any language
version of the original PPVT). Unlike the CDI (which is designed for younger
children; see top of Section 2), the PPVT measures comprehension vocabulary
size indirectly. Children have to pair a word they hear with one of four pictures
they see. Only one picture is correct. Children’s performance on the limited
number of words queried through the PPVT is considered a proxy for children’s
overall comprehension vocabulary size. Standardized means established on the
basis of test results from monolinguals are taken as the norm against which new
test results are compared. Results of PPVT tests of young bilingual children are
thus compared to monolingual-based norms (it is doubtful, however, that this is
good practice; see, e.g., Haman et al., 2015). The available research often
combines data from BFLA and ESLA children. Results are discussed in either
Section 3.1 or Section 3.2 depending on whether most children in a study were
likely BFLA or ESLA, respectively.
Discussion of children’s comprehension is followed by an outline of language
production in BFLA (Section 3.1.2) and ESLA (Section 3.2.2) toddlers and
preschoolers. Information on young bilingual children’s language production typ-
ically relies on observational studies employing parental diaries, labor-intensive
transcriptions of audio recordings (with or without added video), or a combination
of these. Many more such data are available for BFLA than for ESLA.
Section 3.3 compares the language development of BFLA and ESLA pre-
schoolers and Section 3.4 discusses their language choice. Before the summary in
Section 3.6, Section 3.5 explores both child-internal and child-external factors
that may help account for patterns of bilingual development in early childhood.
3.1 BFLA Preschoolers’ Language Development

3.1.1 Perception and Comprehension
Research on speech perception in BFLA children has mainly focused on
infancy. What little research there is shows that in early childhood BFLA
children continue to develop their speech perception abilities (Sundara et al.,
2006). In contrast, there have been many studies trying to understand processes
guiding BFLA preschoolers’ comprehension of new words (Brojde et al., 2012;
Yow & Markman, 2011). As with research on early BFLA speech perception,
this body of work heavily relies on comparisons with monolinguals. For
instance, monolingual preschoolers are strongly guided by the Mutual
Exclusivity Constraint (MEC; Markman & Wachtel, 1988): When children
already know the name for an object, they often reject a second name for it.
In bilinguals a strong application of the MEC would block learning new words
for the same object in another language. This does not happen: BFLA infants
understand translation equivalents from early on (De Houwer et al., 2006;
Section 2.3). It is therefore not surprising that experimental research has found
BFLA preschoolers’ use of the MEC to be weaker than monolinguals’, both for
learning new object names (Davidson et al., 1997) and for learning new object
property labels (Groba et al., 2019).
As for infants, there is limited research on BFLA toddlers’ and preschoolers’
comprehension vocabulary sizes. Smithson et al. (2014) found great interindi-
vidual variability among 77 French–English preschoolers’ (a combination of
BFLA and ESLA) comprehension scores on both the English and French PPVT.
These bilingual children scored higher than the standardized means on both.
Raw score comparisons with monolingual preschoolers acquiring either English
or French found no bilingual–monolingual differences. Other studies with
French–English preschoolers (mostly BFLA) equally failed to find any bilin-
gual–monolingual PPVT differences (Comeau & Genesee, 2001; Sundara et al.,
2006).
Although BFLA preschoolers continue to understand two languages, their
level of comprehension in the Non-Soc-L may be much lower than that in the
Soc-L (MacLeod et al., 2013), likely a direct effect of children hearing the Soc-
L more frequently in ECEC (see further Section 3.5). BFLA children may show
uneven development in comprehension already in infancy (Section 2.5), but in
early childhood uneven development may become more pronounced.
3.1.2 Production
BFLA toddlers and preschoolers continue to use phonological processes

(Section 2.4.3) that often make it hard to say which language they are speaking
(Navarro et al., 1998). Although such phonological processes decrease with
time, making it possible for unfamiliar strangers to understand most of what
any four-year-old (bilingual or monolingual) says, phonological development
continues over many years. Scholars have wondered to what extent BFLA
toddlers and preschoolers construct separate phonological systems, and to
what extent one language influences the other. Separate development is in
evidence for some aspects, but less so for others (Marecka et al., 2020). Much
depends on what is examined: phonemes (considered by themselves or in
combination with one another; see Section 2.2) or prosody (intonation patterns
on the lexical, phrase, clause, or sentence level). In addition, any findings are
restricted by the limited number of language pairs-in-acquisition that have
been studied.
As in infancy, BFLA toddlers and preschoolers produce translation equiva-
lents (TEs). By age 2;6 (or even earlier), BFLA toddlers often spontaneously
offer TEs, saying a word in Language A and immediately afterwards saying its
translation in Language Alpha. Many are able to provide a TE on request (De

Houwer, 2009). Such abilities show children’s developing metalinguistic
awareness, which appears to develop somewhat earlier in BFLA children than
in monolinguals (De Houwer, 2017b). Translation abilities depend on children
knowing the words for the same thing in each language. If BFLA toddlers
exhibit large differences in production vocabulary size in each language (as
found by Marchman et al., 2010; Hurtado et al., 2014; and Ribot & Hoff, 2014)
the chance that they know TEs diminishes. The fact that BFLA toddlers have
much larger production vocabularies in one language than the other also means
they develop language-specific vocabularies from early on. As in infancy, this
difference among languages is evidence of BFLA children’s uneven
development.
The few longitudinal studies of BFLA children’s vocabulary development
also show uneven development. A CDI study (top of Section 2) traced the
growth of production vocabulary in 23 Russian–Finnish children from when
they were infants (14 and 18 months) to when they were 24 and 36 months
(Silvén et al., 2014). In infancy, Russian and Finnish vocabulary sizes kept pace
with each other, but during early childhood Russian (the Non-Soc-L) vocabu-
lary size stopped growing much, while Finnish (the Soc-L) vocabulary size
grew quite quickly. Based on tests, Hoff and Ribot’s (2017) longitudinal study
found similar results for older children: From 30 to 60 months, BFLA pre-
schoolers’ Non-Soc-L scores (Spanish) were lower and increased far more
slowly than Soc-L scores (English). The fact that the Soc-L developed faster
was likely a direct effect of children attending ECEC in the Soc-L (see further
Section 3.5).
In early childhood BFLA children greatly expand their speaking abilities. In
this video you can hear 3.5-year-old Annabelle fluently speak German and
Russian.18 The short word combinations from late infancy have expanded into
veritable sentences with four or five words. Sentences may contain appropriate
inflection morphology (that is, parts of words attached to word roots that
express grammatical meanings, such as plural -s marking in Spanish; see De
Houwer, 2009, for deeper explanations) and different parts of speech, including
function words such as pronouns and articles. Inflection morphology, function
words, and word order constitute the main devices that languages of the world
use to connect lexical roots with each other into sentences. These devices are
known as morphosyntactic devices. Each language has its own range of mor-
phosyntactic devices and its own rules for using them. Morphosyntax refers to
18
Mama Nadia [Screen name]. (2020, August 21). Zweisprachige Kindererziehung: Klappt Das? ||
Tipps und Tricks für eine Bilinguale Erziehung [Video]. YouTube. www.youtube.com/watch?
v=VSVHJp_KrJ8
the totality of devices that structure the combination of words in an utterance so

that meaningful sentences are created.
The range of morphosyntactic devices that both bilingual and monolingual
toddlers and preschoolers use is limited in comparison with adult uses, but
continuously growing. By their third birthdays, many BFLA children produce
long clauses and complex sentences consisting of a main clause and
a subclause. By the time BFLA children are four years old, they can tell
short stories, are usually fairly easy to understand for strangers, and have
a large vocabulary (De Houwer, 2009). The global timeline here is similar to
that for monolingual peers. As for monolinguals there is wide interindividual
variability.
BFLA preschoolers also show cross-linguistic variability: For morphosyn-
tax, BFLA children develop each of their languages independently of the
other. In accordance with the Separate Development Hypothesis (De
Houwer, 1990, 2009), BFLA children’s unilingual utterances (i.e., with
words from just a single language) use the morphosyntactic devices and
rules of each of their languages separately: Utterances with words just from
Language Alpha show a word order and use of bound morphemes and function
words that fit Language Alpha (bound morphemes are grammatically mean-
ingful sound(s) or syllable(s) attached to word roots; an example is -ing in
English, which makes a gerund of verb roots such as sing). Utterances with
words just from Language A show a word order and use of bound morphemes
and function words that fit Language A. Morphosyntactic influence from one
language on the other language is rare in BFLA and, if it occurs, is not
systematic. Furthermore, BFLA children’s unilingual utterances in each lan-
guage are structurally very similar to those produced by monolingual peers
(for mixed utterances, see Section 3.3). However, the main hallmark of
children producing words and sentences is that children use them in conver-
sations with others. The continued relevance and importance of conversations
for children’s speaking abilities is explained in Section 3.4.
Some BFLA children regularly produce mixed utterances (Section 2.4.5),
whereas others produce none. On the whole, though, BFLA toddlers and
preschoolers tend to produce unilingual rather than mixed utterances (De
Houwer, 2009; Ribot & Hoff, 2014).
BFLA toddlers and preschoolers may not be equally proficient in each
language. The language they speak less well usually is the Non-Soc-L (De
Houwer, 2009). Many BFLA children even stop speaking the Non-Soc-L
around age 4 or 5. This likely effect of attending ECEC in the Soc-L (Section
3.5) shows stark uneven development, with BFLA preschoolers continuing
to understand two languages but speaking just one.
3.1.3 Summary
BFLA preschoolers continue to understand two languages. Many increase their
production ability in each language, although one language can develop faster
than the other, then slow down, and the other language can then “overtake” the
first. Unilingual utterances in Language A show no systematic morphosyntactic
influence from Language Alpha (nor the other way around). The morphosyn-
tactic characteristics of utterances produced by BFLA preschoolers resemble
those of monolinguals in the corresponding language. At the end of early
childhood, BFLA preschoolers have high speaking abilities, often in two
languages. By the time BFLA preschoolers enter school, they tend to speak
the Soc-L better and at levels that are comparable to those of monolinguals
speaking the same Soc-L as their single language.
3.2 ESLA Preschoolers’ Language Development

3.2.1 Comprehension
When hitherto monolingual toddlers and preschoolers start to have regular
contact with a new L2, they mostly encounter the L2 in group contexts, with
only little supportive one-to-one child-directed speech (Jahreiß et al., 2018). For
hitherto monolingual children, hearing structurally complex input in an
unknown language in an unknown context first comes as a shock: Suddenly
they can no longer understand what people are saying. This does not contribute
to children’s well-being (De Houwer, 2015b, 2020a): The first few months in an
institutional environment in which ESLA toddlers and preschoolers do not
understand anything can be an unpleasant, stressful, and even traumatic experi-
ence (Dahoun, 1995). Children may silently withdraw and disengage from
interaction at preschool for months (Itoh & Hatch, 1978) or even years
(Drury, 2007), resulting in an apparent rejection of the new language.
It will take some time before ESLA children start to pay attention to the new
language and understand enough of it to correctly interpret what is being said.
How long is difficult to say. Much depends on when children started to hear the
new language and on a variety of other factors (Sections 3.5 and 6.2). ESLA
children who regularly started hearing their L2 as of 2;8 showed a clear increase
in L2 comprehension after a year (Schulz, 2013, and Korecky-Kröll et al., 2016,
showed similar findings for somewhat older children).
Oral vocabulary learning is very much driven by how many words children
have cumulatively heard (Head Zauche et al., 2017). Given that ESLA toddlers
and preschoolers have heard their new L2 for a far shorter time compared to
monolingual peers who have heard that language as their only language from
birth, it is not surprising that bilingual–monolingual comparisons of L2-English

comprehension have found ESLA preschoolers to lag behind monolingual peers
(Bialystok et al., 2010).
3.2.2 Production
ESLA children who start regularly being addressed in an L2 soon after
their second birthdays will be able to say many words in their L1 and will
have started using word combinations and short sentences in their L1. The older
ESLA children are when they first start regularly hearing the L2, the more
advanced their L1 speaking skills will be.
After ESLA children have developed some level of L2 understanding, some
gradually start to speak the L2 a little, but only in short formulae and in one- to
two-word utterances (Itoh & Hatch, 1978). Other ESLA children soon start to
speak the L2 a lot and quite well. Yet others do not speak the L2 even after 2
years (or longer) in preschool (Thompson, 2000). This “silent period” is often
considered a “natural” and inevitable phenomenon of ESLA, but it is not. Young
children who go through a very long “silent period” in the L2 are not doing well
at a socio-emotional level (De Houwer, 2020a; Section 6.2).
Once ESLA preschoolers start to produce spontaneous utterances in the L2
that are not (partly) formulae or direct imitations, they may apply L1 morpho-
syntactic rules in their L2 utterances. An example in L2-English is “I something
eating,” said by an L1-Turkish 4-year-old (Haznedar, 1997, p. 247), in which the
child used Turkish word order. Such utterances showing clear morphosyntactic
transfer (that is, influence from one language on the other) are to be expected in
ESLA preschoolers (Li Wei, 2011; Pfaff, 1994; Zdorenko & Paradis, 2007). The
proportion of such sentences compared to the totality of early L2 production is
not known, but appears to be frequent and systematic within a limited learning
phase (Schwartz et al., 2015). Inappropriately missing function words in the L2
can also point to transfer from the L1 (Blom, 2010; Reich, 2009; Rothweiler,
2016). Conversely, the new L2 can influence morphosyntactic phenomena in
the L1 (Gagarina & Klassert, 2018).
In phonology, transfer from L1 to L2 is common in ESLA preschoolers
(Babatsouli & Ball, 2020). Many ESLA children eventually speak without the
“foreign accent” that is the result of such transfer, though.
Forms showing transfer or errors may persist as children grow older or may
be replaced by the “correct” forms as children gain expertise in their L2. There
is, unfortunately, little research on ESLA preschoolers’ language development
into the school years. We do know there is extremely large interindividual
variation in the speed with which ESLA toddlers and preschoolers develop
their new L2 (Hammer et al., 2012; Thompson, 2000; Winsler et al., 2014b).
This variation makes it difficult to assess whether children are following
a “normal” course of bilingual development. To make a proper assessment,
children should be evaluated in both their L1 and their L2 (De Houwer, 2018a).
A problem is that we know very little about the course of L1 development in
ESLA toddlers and preschoolers who have started to regularly hear an L2. Most
ESLA preschoolers likely continue to speak their L1. However, we know little
about how well older ESLA preschoolers are able to speak their L1 after they
have started to speak their L2. Reich’s (2009) unique longitudinal study of 36
ESLA preschoolers who started hearing L2-German around the third birthday
found that up to the fourth birthday most children gradually improved their
speaking abilities in both the L1 and the L2, with increasingly complex sen-
tences in both. At first, the L2 was less well developed than the L1. This
changed in the fifth year of life, when for several children the L2 became
significantly better developed than the L1.
As an example of an ESLA preschooler, listen to Petra. She started learning
Mandarin after moving to China with her parents from the United States at age
2. After 14 months she started to speak some Mandarin, but with reluctance, at
least in this video,19 in which she also speaks some English, one of her first
languages (Petra likely is both a BFLA and an ESLA child). One year hence,
after 2.5 years in China, Petra learned to speak Mandarin fluently20 and was able
to communicate well and easily in various circumstances.
Uneven development is strongly present in ESLA, particularly at the first
stages of L2 development: ESLA toddlers and preschoolers have already learned
to speak their L1 to some extent but are just starting to acquire their L2. At first,
their Non-Soc-L is far better developed than the Soc-L. Studies have explicitly
addressed uneven development for lexical comprehension (Ertanir et al., 2018)
and production (Budde-Spengler et al., 2018; Ertanir et al., 2018; Rinker et al.,
2017). Older ESLA preschoolers may show larger comprehension and production
vocabulary in the Soc-L than in the Non-Soc-L (Kan & Kohnert, 2005). This may
be indicative of L1 speaking skills starting to stagnate.
3.2.3 Summary
ESLA toddlers and preschoolers understand and speak their L1 (a Non-Soc-L).
Only very gradually do they learn to understand and speak a new L2 (the Soc-
19
Real Life Cinema. [Screen name]. (May 28, 2013). American Child Learning Chinese Natively
[Video]. YouTube. www.youtube.com/watch?v=026h0X0lD6I
20
Real Life Cinema. [Screen name]. (June 16, 2014). American Child Speaking Fluent Chinese
Mandarin [Video]. YouTube. www.youtube.com/watch?v=xbiqxkoLiO8
L), although some ESLA preschoolers may learn the L2 quite fast. By the time
ESLA preschoolers enter school, many speak the Soc-L quite well, but many
others do not. ESLA preschoolers’ proficiency in the L1 may continue to
develop but it may also stagnate.
3.3 BFLA and ESLA Preschoolers Compared

Sections 3.1 and 3.2 revealed some major differences between BFLA and
ESLA toddlers and preschoolers. The differences in their cumulative linguis-
tic experience first of all affect language comprehension. BFLA children have
learned to understand two different languages simultaneously from the start.
By the time they enter ECEC, they understand hundreds, if not thousands, of
words in two languages. In contrast, ESLA children have no experience with
people speaking different languages and can understand only a single lan-
guage once they enter ECEC. This parallels the monolingual experience.
ESLA children have yet to start learning to understand L2 words once L2
input starts. This has obvious implications for the comprehension of larger
structures. A rare experimental study comparing bilingual children’s sentence
comprehension revealed that ESLA preschoolers performed far worse in
German, their L2, than BFLA peers who had heard German from birth
(Roesch & Chondrogianni, 2016).
Differences in BFLA and ESLA toddlers and preschoolers’ overall lin-
guistic experience also affect language production. By the time they enroll in
ECEC, BFLA children can usually speak two languages. ESLA children start
off speaking only a single language and may only very gradually start to
speak another language (some children are faster). This difference in timing
is particularly relevant for the Soc-L (ESLA children’s L2, BFLA children’s
Language A). Even if BFLA children are performing better in the Non-Soc-L
than the Soc-L before they start in ECEC, they speak the Soc-L much better
than ESLA preschoolers do. BFLA toddlers’ and preschoolers’ attending
ECEC in the Soc-L will soon lead to increased speaking ability in the Soc-L.
In contrast, it may take ESLA children several months or longer to start
speaking the Soc-L. Once they do, they initially perform much worse in the
Soc-L than BFLA peers. In a rare direct (albeit numerically small) compari-
son, Dicataldo and Roch (2020) showed this to be the case for 28 bilingual
preschoolers aged between 4 and 6. BFLA children in this group had heard
the Soc-L, Italian, from birth and were doing far better on various Italian
verbal tasks than ESLA peers who were acquiring Italian as an L2.
Once ESLA preschoolers are able to produce Soc-L sentences their
morphosyntactic structures often differ from those that BFLA children use
in the Soc-L. BFLA children’s Soc-L sentences show no systematic influ-

ence from the Non-Soc-L (Section 3.1.2). In contrast, transfer from the Non-
Soc-L is common in ESLA preschoolers’ sentences in the Soc-L (Section
3.2.2). ESLA preschoolers may also use Soc-L sentence formation patterns
that do not show a direct Non-Soc-L influence but that differ from patterns
used by BFLA and monolingual children who have heard the Soc-L from
birth (Granfeldt et al., 2007). A rare detailed comparison of French morpho-
syntactic structures as used by BFLA and ESLA German–French preschool-
ers showed that in L2-French, ESLA children made errors that BFLA
children did not (Meisel, 2008). In the first few years of speaking the
Soc-L, many ESLA children have a “foreign accent,” which BFLA children
do not.
The Non-Soc-L may also exhibit major differences among BFLA and ESLA
children: many BFLA preschoolers no longer speak the Non-Soc-L. This has
rarely been reported for ESLA preschoolers.
There are also similarities between BFLA and ESLA children in early
childhood. Two-year-old BFLA (Ronjat, 1913) and ESLA (Idiazábal, 1984)
toddlers may comment on who speaks what language. Spontaneously offering
a translation equivalent for a word has been noted for both BFLA (Section
3.1.2) and ESLA (Idiazábal, 1984) toddlers. In both BFLA and ESLA pre-
schoolers, mixed utterances mostly concern the embedding of a single free
morpheme (mostly a noun) of language X in a statement in language
Y (Cantone, 2007; De Houwer, 2009; Paradis & Nocoladis, 2007; Reich,
2009). BFLA and ESLA preschoolers’ language choice patterns show both
similarities and differences (Section 3.4).
Both BFLA and ESLA children may refer to some things only in one
language (Rinker et al., 2017). Indeed, like this lexical usage, uneven develop-
ment characterizes most BFLA and ESLA preschoolers’ bilingual skills.
However, BFLA preschoolers usually know the Soc-L better, whereas ESLA
preschoolers usually know the Non-Soc-L better.
3.4 Language Choice in Early Childhood

BFLA toddlers and preschoolers usually select each language as a function of
their interlocutor, addressing persons who speak Language Alpha with them in
Language Alpha, and persons who speak Language A with them in Language
A. However, with adults who they can expect to understand both languages,
children may speak both languages, even if adults usually address them in just
a single language (De Houwer, 1990; Paradis & Nicoladis, 2007; Ribot et al.,
2018; Tare & Gelman, 2010). Ribot and Hoff (2014) showed 2.5-year-olds
using different patterns of language choice depending on which language

parents spoke to them. Children showed a preference for the Soc-L: When
parents addressed children in the Non-Soc-L, children were more likely to
respond in the Soc-L than the other way around (children were less likely to
respond in the Non-Soc-L after parents addressed them in the Soc-L). BFLA
toddlers and preschoolers may also speak only the Soc-L to parents, even if
parents always address them in the Non-Soc-L (De Houwer, 2009; Eilers et al.,
2006). Even if they occasionally speak the Non-Soc-L with people outside the
family, children may soon lose the ability to do so. Non-Soc-L loss in early
BFLA is not unusual, at least in production (De Houwer, 2017c). Thus, lan-
guage choice in interaction helps explain uneven development: If young chil-
dren do not regularly speak a language, they may lose the ability to use it.
Uneven development also affects language choice: You cannot speak the
“right” language with someone if you don’t speak their language. When ESLA
children first start hearing an unknown L2 in ECEC, they usually remain silent
rather than attempt to speak their L1. As soon as they can speak the L2 a bit,
they mostly speak the L2 in ECEC, and the L1 at home (Reich, 2009).
However, some ESLA children may start using the L2 at home, even though
their parents do not speak it to them. If parents do not use monolingual
discourse strategies (Sections 2.6 and 3.5), even ESLA preschoolers may
lose the ability to speak their L1 (Kaufman & Aronoff, 1991; Lindquist &
Gram Garmann, 2021).
When children are able to speak two languages they can switch between them
without hesitations and basically speak the language that is spoken to them (this
video with nearly 5-year-old Spanish–English BFLA Liam shows an example
of this; Liam is the same boy as in the video referenced in Section 2.6 earlier).21
Bilingual children are also able to repair their language choice upon request.
Such responsiveness to interlocutors is only possible if children are sensitive to
other people (Gampe et al., 2019).
This video22 illustrates some dynamics of language choice in child–adult
interaction when children speak two (or more) languages. About 30 seconds
into it, 4-year-old Emi shows an English-speaking adult (likely her mother)
a pink toy couch. She calls the toy a “divan,” which is the correct Russian word.
Her mother does not confirm that Emi has correctly described the toy, but briefly
interrupts the exchange by saying the English word “sofa?” in a questioning
tone. This use of this monolingual strategy, the Expressed Guess Strategy
21
MisterO. (2015, June 3). Bilingual Development Update (Long Overdue)[Video]. YouTube.
www.youtube.com/watch?v=T361-N1i3Dk
22
13ruskie [Screen name]. (2012, October 3). 4 Year Old Girl Speaks 3 Languages [Video].
YouTube. www.youtube.com/watch?v=jATItFipzBg
(Section 2.6), signals a confirmation query. In response, Emi quickly repairs her
earlier language choice by switching to English and says, nodding: “yeah yeah,
sofa,” after which she continues in English. A bit later the mother speaks
Russian with Emi, and Emi responds in Russian. At around minute 2:32, Emi
compares a toy table to a piano, saying in English: “It’s like a piano,” but
signaling nonverbally that she is not quite serious. Her mother chuckles and
repeats in a light-hearted tone “It’s like a piano?”, after which Emi’s face shows
satisfaction that her little joke was successful. After this little aside she con-
tinues in Russian again. This example shows that briefly switching to another
language can serve a pragmatic function. When in the Russian conversation
Emi refers to the pink couch as “divan” (minute 2:44), her mother this time
repeats the word in Russian with an admiring tone and makes confirmatory
noises. Emi, appropriately, does not repair anything after that but just continues
talking about the couch (in Russian). In the mostly Spanish conversation with
a male adult that follows, Emi speaks Spanish, but both adult and Emi occa-
sionally use a mixed utterance, inserting an English word into a Spanish utter-
ance, thus adjusting to each other.
3.5 Factors Affecting Bilingual Development in Early Childhood

Children learn language through engagement with others. In bilingual settings,
children’s engagement with others is always linked to patterns of language
choice as used by children’s interlocutors. As was the case in infancy, the
dynamics of language choice in parent–child interaction are crucial in helping
to ensure that toddlers and preschoolers speak both languages. BFLA children
who had been fluently speaking the Non-Soc-L to a parent may suddenly stop
speaking it. This often happens after children come home from their first day in
a Soc-L-speaking preschool (De Houwer, 2009, 2017c, 2020b). ESLA children
may also bring the Soc-L home (Section 3.4). Parents can turn this tide through
the sensitive use of monolingual discourse strategies (Section 2.6). Parental
discourse strategies and language choices encouraging children to use the Non-
Soc-L in conversations with them remain greatly important throughout child-
hood (De Houwer & Nakamura, accepted for publication 2021).
Children need many hours of using and thus practicing a language to become
expert speakers (Clark, 2003; Hurtado et al., 2014). Ribot et al. (2018) showed
that BFLA preschoolers who spoke the Soc-L more often spoke it better.
Conversely, less or no use of the Non-Soc-L by children who are not yet fluent
speakers may lead to language loss (at least in production). This downward
spiral may be exacerbated by Non-Soc-L-speaking parents switching to the
Soc-L with children, resulting in diminished input in the Non-Soc-L. As
outlined in Section 3.5.2, continued rich exposure to both languages is import-

ant for BFLA and ESLA alike.
Basic differences among BFLA and ESLA preschoolers in Soc-L skills relate
to the fact that BFLA children have heard the Soc-L from birth whereas ESLA
children have not: ESLA preschoolers simply have had less time to learn it.
Thus, it is not surprising that German–Turkish BFLA toddlers up to 30 months
of age produced many more German words than ESLA peers who only heard
L1-Turkish at home (Budde-Spengler et al., 2018). Length of exposure was
a major factor in older BFLA and ESLA preschoolers’ German word compre-
hension and production (Vaahtoranta et al., 2020). The story is of course more
complicated: BFLA and ESLA children are also at different developmental
stages when they first start to regularly hear the Soc-L. Additionally, when
BFLA infants first start to hear the Soc-L, they know very little about language,
but ESLA toddlers and preschoolers already know quite a bit.
The factors affecting the course of bilingual development are multiple,
complex, and dynamically interact with each other (Hammer et al., 2012).
They include factors particular to children as growing individuals (known as
child-internal factors) and factors relating to the environments in which children
are growing up (child-external factors). Child-internal and child-external fac-
tors are closely linked, and disentangling them is a challenge (Chondrogianni &
Marinis, 2011; Sun et al., 2020). Many studies have examined factors that may
help explain bilingual development. These studies usually do not focus on
individual developmental trajectories but aim to explain intragroup variability.
The remainder of this section is a highly selective review (for more details, see
Armon-Lotem & Meir, 2019, and De Houwer, 2018b).
3.5.1 Child-Internal Factors Affecting Bilingual Development in Early

Childhood
The most obvious child-internal factor is chronological age (see also Section
2.6). Age is a proxy for increases in cumulative language exposure, cognitive
maturity, memory skills, interactional skills, and much more. Unsurprisingly,
older children typically have higher developed language skills than younger
ones. Studies focusing on internal factors have mostly concentrated on ESLA
children.
Not unexpectedly, L2 vocabulary size and morphology are better developed
in older than younger ESLA preschoolers; children’s phonological short-term
memory and nonverbal intelligence are also strong predictors (Paradis, 2011;
phonological short-term memory refers to children’s ability to keep sounds
buffered in short-term memory). ESLA preschoolers’ phonological memory
abilities are related to both L1 and L2 vocabulary size (Ertanir et al., 2018).
Furthermore, ESLA differences in children’s early socio-emotional compe-
tence help predict their L2 learning (Ertanir et al., 2019; Winsler et al., 2014b),
as does personality: Shy ESLA preschoolers take longer to start speaking the
new L2 (Keller et al., 2013; Reich, 2009), whereas children who are eager to
learn the new L2 and ready to engage with peers develop it faster (Schwartz
et al., 2020).
A child-internal factor specific to bilinguals concerns their knowledge of the
other language. The greater ESLA children’s L1-Spanish vocabulary size was at
age 2, the better they scored on L2-English comprehension and production
measures 2.5 years later (Marchman et al., 2020). This longitudinal relation was
partly moderated by toddlers’ efficiency at processing Spanish words at age 2.
The supporting role of ESLA children’s L1 for later L2 development has also
been found for older children: The better developed L1-Spanish was at age 4,
the faster ESLA preschoolers acquired L2-English a year later (Winsler et al.,
2014b). Confirming similar results from earlier studies involving different
language pairs in different countries, ESLA preschoolers’ L1-Turkish compre-
hension vocabulary size predicted L2-Dutch comprehension vocabulary size
(Sierens et al., 2019). In contrast, neither in ESLA (Ertanir et al., 2018) nor in
BFLA (Marchman et al., 2010) did production vocabulary size in one language
concurrently predict production vocabulary size in the other. However,
Marchman et al. (2010) found strong concurrent within-language links between
BFLA toddlers’ ability to understand familiar words in a lexical processing
task, on the one hand, and production vocabulary size, on the other; children’s
processing speed in Language Alpha did not correlate with Language
A vocabulary size (see also Hurtado et al., 2014). BFLA 30-month-olds who
produced more Spanish than English words as rated through the CDI (top of
Section 2) had higher Spanish than English PPVT-based comprehension scores
6 months later (Hurtado et al., 2014). Furthermore, Hurtado et al. (2014) found
strong within-language links between these 3-year-olds’ real-time word pro-
cessing abilities and comprehension as measured by the PPVT. In addition,
there were significant correlations between children’s relative exposure to
Spanish vs. English and their relative speed of processing of Spanish vs.
English words. Thus, even for an internal characteristic such as processing
ability, external factors play a role. Palacios et al. (2015) showed how a child-
internal factor (level of language development) was dynamically influenced by
an external factor, namely, parental interaction: Spanish–English bilingual
children’s different levels of language development elicited different levels of
supportive language input from parents, which in turn affected children’s
linguistic skills.
3.5.2 Child-External Factors Affecting Bilingual Development in Early

Childhood
As indicated just before Section 3.5.1, differences between BFLA and ESLA
children’s development in the Soc-L can (partly) be attributed to differences in
cumulative amount of input. Amount of input plays a fundamental role in other
ways as well.
As for infants, the relative amount of input in each language helps explain
why BFLA toddlers and preschoolers have a better command of Language
A than Language Alpha: The language that is heard more often is the one that
is better developed. This has been shown for growth in comprehension vocabu-
lary (Scheele et al., 2010) and production vocabulary size (Marchman et al.,
2010; Nakamura & Quay, 2012; Place & Hoff, 2011) as well as grammatical
development (Blom, 2010; Hoff et al., 2012). ESLA preschoolers’ levels of
comprehension and production in each language likewise related to the concur-
rent relative amount of input in each language (Marchman et al., 2020).
However, even though Finnish–Swedish BFLA toddlers and preschoolers in
Silvén et al. (2014) heard the Non-Soc-L (Finnish) more often, the Non-Soc-L
did not develop as fast as the Soc-L (Swedish). Silvén et al. (2014) attributed the
Non-Soc-L’s slower development to children becoming more involved with
caregivers and peers outside the home who spoke the Soc-L as they grew older.
Additionally, school-aged bilinguals may foster their younger bilingual pre-
school-aged siblings’ learning and appreciation of the Soc-L through play and
literacy-based activities in the Soc-L (Kibler et al., 2014; Mirvahedi &
Cavallaro, 2020). You can see an example of how older siblings bring the
Soc-L into the home through play-school activities starting at about minute
3:40 in this French–Dutch video.23 Families that do not send their preschoolers
to ECEC stand a greater chance of having children who speak the Non-Soc-L
(Verdon et al., 2014).
In addition to children hearing unilingual utterances, they may also hear
mixed utterances (which combine words from two languages; see Section
2.4.5). Often, people have negative attitudes toward the use of mixed utterances
(De Houwer, 2019b). One might worry that hearing mixed utterances negatively
affects children’s bilingual development. So far, there is no evidence that they
do (Place & Hoff, 2016).
One language may be heard more often than another, but absolute amount of
input may be low in both languages, or high. Does this difference matter?
Marchman et al.’s (2017) study suggests that it does: Absolute measures
23
HaBilNet [Screen name]. (2017, September 26). Les enfants bilingues [Video]. YouTube. www
.youtube.com/watch?v=qJ9Uihrqx2A
tallying the actual number of Spanish and English words addressed to 3-year-
old bilinguals (likely a mix of BFLA and very early ESLA children) far better
predicted children’s language outcomes in each language than relative meas-
ures. As was the case for adults addressing BFLA infants (Section 2.1),
Marchman et al. (2017) found large variability in the amount of adult speech
to preschoolers.
Typically, the language spoken by staff in ECEC is a variety of the local Soc-
L. Staff generally do not understand, let alone speak, whatever Non-Soc-L
BFLA and ESLA children have learned to understand and speak. BFLA chil-
dren in ECEC will likely hear the Soc-L more often than they did before they
were attending ECEC. This increased absolute amount of input is likely partly
responsible for BFLA preschoolers’ relatively greater proficiency in the Soc-L
in the preschool years (compared to the Non-Soc-L). At the same time, the
hours spent in ECEC diminish opportunities to hear and speak the Non-Soc-L.
This decreased absolute amount of input is likely partly responsible for many
BFLA preschoolers’ slower development in the Non-Soc-L in the preschool
years (compared to the Soc-L).
Absolute amount of input is also important in ESLA: A longitudinal study of
preschoolers with L1-Turkish and L2-German found faster L2 comprehension
vocabulary development in children who spent more time at a German pre-
school, and thus had higher amounts of L2 input (Czinglar et al., 2017). This
effect was independent of the length of time children had spent at preschool,
although as in Blom and Paradis (2015) the overall duration of L2 contact was
also an explanatory factor, with children who started attending preschool earlier
having better L2 skills. Methods developed for assessing possible L2-German
delays in preschoolers take into account length of contact with the L2 (Schulz &
Tracy, 2011).
However, as in infancy (Section 2.1) there can be no assumption of stable
comparative levels of input in each language throughout bilingual children’s
early language learning years. The absolute amount of input in a particular
language is highly variable over time, and affects relative amount of input. This
variability makes it hard to correctly evaluate bilingual input environments.
In addition to addressing children in the Non-Soc-L, parents may start
speaking the Soc-L to them once children start attending preschool, thus
leading to decreased input in the Non-Soc-L. A longitudinal study reporting
on a mix of BFLA and ESLA preschoolers showed that the average percentage
of Non-Soc-L input to children decreased from 66% when children were 2.5 to
58% when children were 5 years old (Lauro et al., 2020). Parents usually start
speaking the Soc-L to children in response to children addressing parents in
the Soc-L (Kuo, 1974; Luo et al., 2020; Prevoo et al., 2011). The fact that
children no longer want to speak the Non-Soc-L may be the result of children
developing negative attitudes toward the Non-Soc-L (a child-internal factor),
which, in turn, may have been shaped by negative experiences in ECEC.
ESLA children who do not yet speak the Soc-L well may experience bullying
by peers (shown by Chang et al., 2007, in the United States and by von
Grünigen et al., 2012, in Switzerland), which helps deliver the message that
children need to focus on quickly improving their Soc-L skills, and forget the
Non-Soc-L. Because BFLA children’s Non-Soc-L is usually ignored at pre-
school, BFLA preschoolers may soon learn that the Soc-L is socially more
important (De Houwer, 2015b, 2020b) and turn against the Non-Soc-L.
Whereas one expects increased or decreased input in language X to affect
children’s use of language X, there is no expectation that input in language
X directly affects language Y. Yet Willard et al. (2021) found exactly this to be
the case: Parental literacy activities in a Non-Soc-L contributed to ESLA
children’s comprehension of the Soc-L. Children’s language experience thus
must be considered holistically.
Yet on a more detailed level, the kind of input (often called “input quality”) to
children may play a language-specific role. Bilingual parents may have unequal
levels of proficiency in each language and thus may address children at variable
proficiency levels (Hoff et al., 2020). These variable proficiency levels may
affect children’s language development: Preschoolers who heard the Soc-L
(Dutch) from parents who spoke it well, regardless of whether parents spoke
Dutch as an L1 or L2, did better in Dutch than children whose parents spoke
Dutch less proficiently (Unsworth et al., 2019).
Young BFLA children’s language use may show apparent and unexpected
influence from the other language. Studies comparing such apparent influ-
ence from the other language in young BFLA children’s language use with
highly proficient bilingual parents’ input showed that idiosyncrasies in
parental input could account for children’s language use (Bosch & Ramon-
Casas, 2011; Meisel et al., 2011; Paradis & Navarro, 2003; but see
Nakamura, 2015). Given that the actual speech that children hear is the
basis for their language intake (De Houwer, 2017a), studies trying to explain
unexpected patterns of bilingual development should include measures of
actual speech to children. Only then will correct interpretations of children’s
language use be possible.
As stated at the top of this section, parental discourse strategies in response to
children’s language choice play a crucial role in creating a continued need for
children to speak the Non-Soc-L. Absent this need, children may simply not
speak the Non-Soc-L, regardless of how much they hear it or how well parents
speak it.
3.6 Summary
BFLA and ESLA toddlers’ and preschoolers’ fundamentally different lan-
guage-related experiences lead to different trajectories for language develop-
ment in early childhood. Although there is great interindividual variability, the
following generalizations can be made. Whereas BFLA children understand
and speak two languages at the beginning of early childhood, ESLA children
understand and speak just a single language (their Non-Soc-L). In the early
childhood years, both BFLA and ESLA children make gains in their proficiency
in the Soc-L, but even by the end of early childhood, BFLA children understand
and speak the Soc-L at far higher levels than most ESLA children. Although
many BFLA and ESLA preschoolers continue to understand and speak the Non-
Soc-L at high levels, BFLA preschoolers are more at risk of losing the Non-Soc
-L than are ESLA preschoolers. Both BFLA and ESLA preschoolers adjust their
language choice to their interlocutors and are sensitive to the degree to which
interlocutors socialize children into responding in the Non-Soc-L. Such lan-
guage socialization efforts are crucial in supporting the Non-Soc-L. Further
support for the Non-Soc-L comes from the amount of language input in the
Non-Soc-L. Frequency of input also matters for the Soc-L.
4 Bilingualism in Middle Childhood: Bilingual First Language

Acquisition, Early Second Language Acquisition, and Second
Language Acquisition
The beginning of middle childhood coincides with many children starting to
attend primary school. Section 4.1 describes aspects of BFLA and ESLA
schoolchildren’s bilingual development. Section 4.2 zooms in on Second
Language Acquisition (SLA) children, namely, on hitherto monolingual school-
children who start hearing their new L2 at school, and who need to learn it from
scratch. Section 4.3 explains how, as at earlier ages, bilingual schoolchildren
show evidence of uneven development. Whenever bilingual children speak
more than a single language, they must select a specific language in speaking.
Language choice is the topic of Section 4.4. Before the brief summary in
Section 4.6, Section 4.5 explores factors that can help explain bilingual devel-
opment in middle childhood.
Unless otherwise specified, reference to schools in this section assumes that
schools use only or mainly a single Soc-L as a vehicle of instruction. In these
schools, children usually do not get much specific support to help them develop
the Soc-L but are mostly in a “sink-or-swim” situation (Kim et al., 2015; note
there is a range of educational approaches). This circumstance differs from
many preschools, which often have programs in place to help ESLA children
develop the new L2 through one-on-one supportive interaction with preschool

staff.
4.1 BFLA and ESLA Schoolchildren’s Language Development

Upon school entry, both bilingual and monolingual children still have a lot to
learn in the Soc-L used at school (Pearson, 2019). BFLA children who have
heard the Soc-L from birth will normally be well positioned to cumulatively
learn the kind of language expected at school and can normally already speak
the Soc-L well. In contrast, ESLA children who started learning the Soc-L in
early childhood may as yet lack sufficient proficiency to be able to optimally
function at school. Many ESLA children are likely to understand the Soc-L well
enough to understand most of what is going on in school, but their speaking
skills may as yet be underdeveloped.
Many BFLA schoolchildren are able to fluently speak two or more languages,
although not necessarily at similar levels. In this video you can listen to a girl
speaking Spanish, French, German, and English.24 Like monolinguals, BFLA
children have not, however, learned all the complexities of the sound systems of
each of their languages yet. BFLA schoolchildren may show occasional influ-
ence from the other language in the production of certain sounds (Mayr &
Montanari, 2015). Some studies have found signs of influence from the other
language in morphosyntax as well (Argyri & Sorace, 2007; Serratrice, 2007),
but there have been no reports of systematic transfer in BFLA schoolchildren. In
middle childhood, the strong differential effects of BFLA and ESLA children’s
language learning trajectories on Soc-L morphosyntactic forms that were
apparent in early childhood may start to level off (Gathercole & Thomas,
2009; Unsworth, 2013), largely because of older ESLA children’s gains in the
Soc-L. Research interest in the morphosyntactic features of ESLA schoolchil-
dren’s L1 is fairly new (Chondrogianni & Schwartz, 2020).
Uccelli and Páez (2007) traced production vocabulary and narrative develop-
ment in both the L1 (Spanish, the Non-Soc-L) and the L2 (English, the Soc-L) in
24 mostly ESLA children from the first to the second school year. Children
made strong gains in the L2 but, unsurprisingly, were not (yet) performing at
English vocabulary levels normed for children with input to English from birth.
Gains in the L1 were only seen in storytelling. A longitudinal study of likely
both BFLA and ESLA children spanning 5 primary school years (N = 139)
examined children’s speed and accuracy on a timed picture-naming task in both
24
Multilingual Education – Playing, Learning, Growing [Screen name]. (2016, October 30).
Bilingualism and More – Switching: Spanish – French – German & English no accent |
Switching Codes [Video]. YouTube. www.youtube.com/watch?v=cSy1490WSU0
Spanish and English (Oppenheim et al., 2020). In both languages, children

improved with time. Both at group and individual levels, children performed
better in the Non-Soc-L prior to around the eighth birthday, but afterwards the
balance reversed, and children performed better in the Soc-L (English). Dubiel
and Guilfoyle (2017) found similar results for L1-Polish children in Ireland.
Aarssen (1996) did so for L1-Turkish children in the Netherlands. A study
relying on processing rather than production measures demonstrated that not all
ESLA children perform better in the L2 than the L1 after age 8: L1-Italian/L2-
German ESLA 8.5-year-olds who had started hearing the L2 in preschool and
were attending an Italian-German school performed much better in their L1 than
their L2 on accuracy and speed in lexical judgment tasks (Persici et al., 2019).
Finally, in early childhood BFLA and ESLA children’s mixed utterances
mainly embedded a content word from language X into an utterance in
language Y. To be able to produce more complex mixed utterances, better
dual language skills may be necessary. Indeed, BFLA and ESLA Spanish–
English schoolchildren with longer experience with both their languages
produced more complex mixed utterances than just insertions of a single
word or phrase (Zentella, 1997).
4.2 SLA Schoolchildren’s Language Development

There are few detailed studies of the course of language development in
schoolchildren who started hearing an L2 in middle childhood. Very little is
known about the development of L2 comprehension in SLA schoolchildren.
Uchikoshi (2006) showed that 150 children acquiring English as an L2 (a mix of
ESLA and SLA children) made rapid gains in L2-English comprehension
throughout the first school year, achieving the level of 4.5-year-old English
monolinguals by the end of the first year of school. Whether that level is
sufficient for full participation in school at the beginning of second grade is
unclear. School test results from primary school children worldwide suggest
that it may take a long time for SLA children to understand the L2 sufficiently to
fully participate in school (UNESCO, 2016).
For production, and similarly to studies of ESLA toddlers and preschoolers,
the few extant studies tend to focus on the L2. Wagner-Gough (1975) as partly
reprinted in Hatch (1978) analyzed the very first L2-English utterances pro-
duced by Homer, who started speaking L2-English just a few weeks after
arriving in the United States and enrolling in an English-speaking school, just
before turning 6. He had an American 5-year-old friend right after his arrival.
Homer first imitated English words and phrases and replied to yes/no-questions
by repeating them without question intonation. In the second month, Homer
went beyond imitations and started forming sentences of his own making. The
Homer example shows that in SLA there need not be a prolonged “silent period”
as experienced by so many younger ESLA children.
Itani-Adams et al. (2017) described a longitudinal case study of L1-
English 7-year-old John learning L2-Japanese through immersion in
a Japanese-medium school in Australia. John started attending school at
6;3. By age 7 he was producing many Japanese words with verbal inflection
morphology. Soon after, phrasal morphology emerged and 4 months later
inter-phrasal morphology. Two years after enrolling at school, John was able
to use a variety of complex structures in L2-Japanese. Itani-Adams et al.
compared John’s L2 development with the development of Japanese as heard
from birth by a BFLA toddler and preschooler with excellent age-appropriate
speaking skills in both Japanese and English. After just 2 years of L2 input,
John used much more advanced morphosyntactic structures than the BFLA
preschooler with nearly 5 years of input in Japanese. Mobaraki et al. (2008)
reported on two L1-Farsi siblings who started hearing L2-English at school
after arriving in the United Kingdom at 7;3 and 8;3, respectively. Three
months after arrival these children were producing multiword utterances in
their L2, but they used some word orders that would have been fine in Farsi
but are not English-like (as in “We tennis play”, p. 220). These early struc-
tures are similar to structures showing morphosyntactic transfer in ESLA
preschoolers (Section 3.2.2). Over the 20 months of observation, the siblings’
English became gradually more English-like, and Farsi-like word orders in
English disappeared. Pienemann (1981) studied the development of auxiliar-
ies and pronouns in L2-German in two L1-Italian 8-year-olds who had just
moved to Germany.
As in ESLA, there is great interindividual variability in the rates at which
SLA schoolchildren learn to speak their L2. A longitudinal study of 89 SLA
children found that some could speak L2-English well after just 1 year at school
(MacSwan & Pray, 2005). Others needed 6.5 years to do so. Most (92%)
children took 5 years to reach a high level of L2-English proficiency. Oller
and Eilers’ (2002) cross-sectional study largely confirmed this picture. A large
longitudinal study of low-income children in the same bilingual area (Miami,
Florida) found that 90% of children were considered proficient in English by
third grade (Kim et al., 2014). This percentage likely covers data from BFLA,
ESLA, and SLA children. Focusing on academic achievement, Collier and
Thomas (2017) reported that it takes SLA schoolchildren in English-only
instruction in the United States between 7 and 10 years to achieve age-
appropriate levels across the English school curriculum and stay at that grade
level (or better) throughout the rest of their schooling (the authors based their
findings on their longitudinal studies involving millions (!) of children, from

kindergarten through grade 12).
At the more detailed level of morphosyntax, Haberzettl et al. (2013) found
differences in the pace with which SLA children learned to correctly order L2-
German verbs and to use correct verb forms. Like MacSwan and Pray (2005),
Haberzettl et al. (2013) found that the older SLA children were when they
started learning the L2, the less time they needed to develop good L2 skills
(recall that Itani-Adams et al. (2017) found much faster development of the
same morphosyntactic structures in an older SLA than a younger BFLA child).
The authors attributed this difference to greater cognitive maturity in older
learners and to greater experience with learning an L1 (for more discussion of
factors affecting bilingual development in middle childhood, see Section 4.5).
4.3 Uneven Bilingual Development in Middle Childhood

De Houwer (2007) found that 1,356 6- to 9-year-olds in dual-parent families in
Flanders, Belgium, heard two languages from their parents at home. They can
thus be considered to have grown up in a BFLA setting. Nearly a third (30.3%)
spoke only the Soc-L (Dutch) and did not speak the Non-Soc-L that at least one of
their parents spoke (see further Section 4.5). It is rare for ESLA or SLA students
who only hear a Non-Soc-L at home not to speak it: only 3% of 422 6- to 9-year-
olds who only heard a Non-Soc-L at home did not speak it (De Houwer, 2007).
Summarizing surveys across the globe, De Houwer (2020b) found that one in four
bilingually reared children did not speak one of the languages heard at home.
Invariably, the language that children did not speak was a Non-Soc-L.
We know little about the comparative skills in each language of BFLA
schoolchildren who speak two languages. Persici et al. (2019) found no cross-
linguistic differences in Italian-German BFLA students’ performance on a lexical
judgment task. Many ESLA students perform better in the L1 than the L2 in the
first years of school, only to see this picture reversed around their eighth birthdays
(Section 4.1). SLA children are likely performing at a fairly high level in their L1
when they enter school at age 6 and start to learn a new L2 there. L1-Russian/L2-
German SLA students self-reported that they spoke the L1 much better than the
L2 (Ahrenholz et al., 2013). It remains to be investigated how SLA school-
children’s L1 continues to evolve as they gain more proficiency in the L2.
4.4 Language Choice in Middle Childhood

BFLA schoolchildren who speak two languages are able to easily switch between
them. As in the case of younger children, they often do so as a function of whom
they are talking to. In this video, a girl is fluently explaining a Lego plane she has
made (the “24,000 model“, as she calls it) in Spanish, English, and French.25 You
can hear adults ask her questions about the plane in each language in the same
setting. The girl responds in the language each adult speaks. This switching
behavior is typical. You can also hear switching in a triadic conversation:
About five minutes into this video you can hear how, in a game with both his
parents, 6-year-old Kevin speaks German with his mother but switches to Dutch
when he addresses his father.26
The fact that many BFLA schoolchildren in bilingual homes do not speak the
Non-Soc-L but only the Soc-L (Section 4.3) implies that they address Non-Soc
-L-speaking parents in the Soc-L. As is the case for younger children, BFLA
schoolchildren thus do not always use the same language that their parents
speak to them. Section 4.5 discusses factors that may lie at the root of these
language choice patterns.
Most (E)SLA children who hear only a Non-Soc-L from their parents at
home likely speak it to them as well (Gathercole, 2007; the findings from De
Houwer, 2007, imply this as well, see Section 4.3). However, most parents of
45 ESLA 5- to 6-year-olds with L1-Urdu reported that their children spoke
mainly L2-Norwegian, the Soc-L, with friends and family (Karlsen et al.,
2017). Parents and teachers of 139 BFLA and ESLA bilinguals reported that
both at home and at school, children increasingly spoke more English (the
Soc-L) than Spanish over the course of 5 school years (Oppenheim et al.,
2020). Kaufman (2001) found that a third of 30 ESLA and SLA children
between ages 6 and 13 spoke only L2-English to their Hebrew-speaking
parents. The relative use of the Soc-L vs. the Non-Soc-L in conversations
with peers was much higher in 8- to 10-year-old bilinguals than in younger
bilingual schoolchildren (Jia et al., 2014). Fuller’s (2012) ethnographic lon-
gitudinal study of L1-Spanish SLA students in the last three grades of a US
transitional bilingual (Spanish–English) program showed how children
started to use more and more L2-English rather than L1-Spanish with each
other as time passed (see also Pease-Alvarez and Winsler, 1994). Bilingual
students in Miami, Florida, almost exclusively spoke the Soc-L at school, even
if they did not speak it well (Eilers et al., 2002). Children may use mixed
utterances as well, but only in contexts where they feel comfortable using
them. Spanish–English bilingual children used hardly any mixed utterances at
school (Oller & Eilers, 2002).
25
Piba18 [Screen name]. (2012, September 28). Trilingual Polyglot Little Girl Speaks Three
Languages English French Spanish [Video]. YouTube. www.youtube.com/watch?
v=3KswIJwGknA
26
HaBilNet [Screen name]. (2017, September 26). Tweetalige kinderen [Video]. YouTube. www
.youtube.com/watch?v=OdzA-9Gcpug
4.5 Factors Affecting Bilingual Development in Middle Childhood

As for BFLA and ESLA preschoolers, the length of time that BFLA and
ESLA schoolchildren have heard each language may continue to affect their
skills in each. Persici et al. (2019) compared 6.5- and 8.5-year-olds with
bilingual input in German and Italian before age 3 (BFLA/very early ESLA)
and demographically matched ESLA peers who had heard Italian from birth
but started hearing L2-German at 3;6 or 4. Children’s language learning
histories affected language production and comprehension: Both groups
performed at equal levels in Italian, but BFLA/very early ESLA children
far outperformed ESLA children in L2-German. There were also differences
in accuracy and speed in lexical judgment tasks in both languages (Section
4.1): Whereas BFLA and very early ESLA students performed similarly in
both languages, ESLA students’ performance was far better in their L1 than
their L2.
Earlier exposure patterns also played a role in a longitudinal study of 228 (E)
SLA children’s Spanish and English skills at ages 6 and 8 (Collins &
Toppelberg, 2021). Children had heard L1-Spanish from birth. Age of first
regular exposure to L2-English started between ages 3 and 6. Children who
had earlier attended ECEC (Early Childhood Education and Care) in Spanish
had higher Spanish skills at ages 6 and 8 than those who had not. However, there
was no association between attending Spanish ECEC and later L2-English
skills, showing that maintaining the Non-Soc-L through ECEC is not a risk
factor for developing the Soc-L. In fact, as in early childhood, knowledge of the
L1 in (E)SLA can help develop L2 skills in middle childhood: The better ESLA
schoolchildren’s earlier narrative skills were in their L1, the better they were in
their L2 a year later (Uccelli & Páez, 2007). Relatedly, Collier and Thomas’
(2017) monumental study focusing on academic achievement in English
showed that SLA schoolchildren in programs that used both their languages
throughout the school years did much better than peers who were in English-
only programs.
The longer children have heard a language, the more frequently they will
have heard it. As in infancy and early childhood, frequency of input likely
continues to play a role for bilingual children in middle childhood (Unsworth
et al., 2014). Yet input frequency has hardly been featured in studies of bilingual
schoolchildren. Compared to when children were not attending (pre)school,
going to school in the Soc-L for about seven hours most workdays may increase
the time of exposure to the Soc-L and may decrease the time that children hear
the Non-Soc-L at home. How these changes affect the absolute amount of input
in each language is largely unknown. Oppenheim et al. (2020) found that BFLA
and ESLA schoolchildren’s changing patterns of performance over the years,

with them performing better in the Soc-L after they had performed better in the
Non-Soc-L, corresponded with third-grade classrooms using much more of the
Soc-L compared to the Non-Soc-L (earlier grades used both languages).
Children also spoke much more of the Soc-L at home, mirroring results in
Mancilla-Martinez and Lesaux (2011), who in addition reported much higher
relative use of the Soc-L by mothers in speech to 11-year-olds compared to
when their children were 4.5.
Language input at school is more diversified than that at home, is used in
a variety of circumstances, and comes from many different speakers. The
combined effect of these factors may help support the Soc-L. Gollan et al.
(2015) found that also for the Non-Soc-L the number of speakers is important:
The greater the number of people was who spoke the Non-Soc-L (Hebrew) with
them, the better 22 Hebrew–English BFLA children aged between about 6 and
10 were on a Hebrew naming task. However, some children did not speak the
Non-Soc-L.
Child-external factors relating to various aspects of the input as described so
far may have different effects depending on what language is being assessed
(Gutiérrez-Clellen & Kreiter, 2003). As outlined below, also child-internal
factors may differentially relate to one particular language and not the other.
SLA children’s motivation to learn a particular L2 may play a role in their
developing proficiency (Haberzettl et al., 2013). Schoolchildren who know they
will be moving to another country later in the school year likely want to invest
less in the L2 than children who know they will be staying in the new country
longer. Motivation may also have played a role in Homer’s (Wagner-Gough,
1975) eagerness to engage with English L2 right from the start (Section 4.2).
The fact that children typically learn to read and write mainly in the Soc-L likely
strengthens their positive attitudes toward it.
Once SLA schoolchildren are able to speak two languages, even at rudimen-
tary levels, they necessarily need to choose which one to use when. Children’s
language choices are not socially neutral. They may use each of their languages
to mark social boundaries and indicate affiliation with some but not with others.
Fuller (2012) documented how two SLA girls in the last years of a transitional
bilingual program used L2-English with each other to set themselves apart from
the other L1-Spanish children in their class. Attitudes and sensitivity to lan-
guage hierarchies also played a role. Other L1-Spanish children in the classes
that Fuller (2012) observed started to use more and more L2-English at school,
likely because they realized that English carried far higher prestige: It was used
exclusively in mainstream classrooms, and children aspired to transitioning to
those mainstream classrooms. Also younger (E)SLA children may be sensitive
to language hierarchies. Gibson et al. (2012) investigated the Spanish and

English abilities of 124 5- to 7-year-olds with input to L1-Spanish in the
home; 45% had started hearing L2-English at preschool (ESLA); 55% started
hearing L2-English at school (SLA). Unsurprisingly, ESLA children had better
L2 word comprehension and production abilities than SLA children. All chil-
dren, however, showed a much wider gap between comprehension and produc-
tion in L1-Spanish than in L2-English. Children were just not producing many
words in the L1. Tests were carried out at school. Gibson et al. speculated that in
the school environment, children were perhaps not keen on presenting them-
selves as Non-Soc-L speakers.
Positive attitudes toward the Soc-L may foster negative attitudes toward the
Non-Soc-L, leading to common observations that children forbid parents to
speak the Non-Soc-L when they are in earshot of classmates (review in De
Houwer, 2020a). Negative attitudes toward the Non-Soc-L may be shaped by
teachers’ stances toward languages not used in school. Teachers may ignore
children’s Non-Soc-Ls, thus erasing part of their identities. Worse, teachers
may explicitly express negative attitudes toward children’s Non-Soc-Ls, for
instance, by ridiculing or punishing children when they speak a Non-Soc-L at
school (anecdotally documented throughout many Western societies).
Through her in-depth ethnographic work relying on 300 hours of close
observation of 5- to 6-year-old bilinguals in a Danish school, Karrebaek
(2013) showed how children interiorized the monolingual, Danish-only ideol-
ogy communicated by the staff, and socialized each other into just speaking
Danish (the Soc-L) at school. If SLA children are bullied for not speaking the
Soc-L well (as was the case in von Grünigen et al., 2012), they may soon
realize that they need to focus on improving their Soc-L skills and not pay
much attention to the Non-Soc-L.
Not speaking the Non-Soc-L in middle childhood is at least partly related to
parental input patterns in the home. De Houwer (2007) surveyed families with
at least one 6- to 9-year-old and at least one parent who spoke the Non-Soc-L
at home. In 16% of 121 single-parent families, children did not speak the Non-
Soc-L. Most were reared by a parent who spoke both the Soc-L and the Non-
Soc-L. Non-Soc-L transmission in dual-parent families was even less success-
ful: In 24% of 1,778 dual-parent families no child spoke the Non-Soc-L. There
were large differences among families in proportions of children speaking the
Non-Soc-L. These differences were a function of parental input patterns.
ESLA children whose parents only spoke the Non-Soc-L usually also spoke
it (97%; Section 4.3). Other, likely BFLA, children heard both languages at
home, in bilingual families representing four parental input patterns: (i) Most
frequently, both parents spoke the Non-Soc-L as well as the Soc-L. In 79% of
these families, children spoke the Non-Soc-L. (ii) Next most frequent were
families where both parents spoke the Soc-L and one parent also spoke the
Non-Soc-L. These had the lowest chance (36%) of having children who spoke
the Non-Soc-L. (iii) Next came families where both parents spoke the Non-
Soc-L and one parent also spoke the Soc-L. These had the highest chance of
having children who spoke the Non-Soc-L (93%). (iv) Least frequent were the
so-called OPOL families (Section 2.1), where one parent spoke the Soc-L and
the other parent the Non-Soc-L. Only in 74% of these did children speak the
Non-Soc-L. In contrast to frequent claims to the contrary there was no parent
gender effect for Non-Soc-L transmission: Male and female parents who
spoke a Non-Soc-L had the same chance of having children who did so (see
also Gathercole, 2007). The different parental input patterns in De Houwer
(2007) may reflect differences in the absolute amount of Non-Soc-L input,
combined with the possibility that in bilingual families with a parent who did
not speak the Non-Soc-L at home, children identified with that parent rather
than the other parent.
Parental discourse strategies encouraging children to use the Non-Soc-L in
conversation remain important for bilingual schoolchildren (De Houwer &
Nakamura, accepted for publication 2021; Yu, 2014). Longitudinal analyses
of parent–child interaction in 68 US bilingual families showed that parents who
used monolingual discourse strategies and consistently spoke either Cantonese
or Mandarin (the Non-Soc-L) with their 6-year-olds had children who were still
proficient in the Non-Soc-L at 7;6 (Park et al., 2012). In contrast, 6-year-olds
whose parents used more bilingual discourse strategies with them, thus allowing
the use of English, the Soc-L, in a conversation, had very low proficiency in the
Non-Soc-L 1.5 years later, or had lost the ability to speak it. Parental use of
discourse strategies and of the Non-Soc-L were dynamically related to chil-
dren’s proficiency. Parents of 6-year-olds who were less proficient in the Non-
Soc-L used less support for the Non-Soc-L 1.5 years later, illustrating the many
transactional processes taking place in parent–child interaction over time (Park
et al., 2012).
Thus, as in early childhood, shifts in language choice patterns may lead to
shifts in children’s language proficiency. The combined effects of bilingual
schoolchildren’s often highly positive attitudes toward the Soc-L and their
more frequent and more diversified learning experiences in it may lead to
ever increasing proficiency in the Soc-L and to children’s lesser use of the Non-
Soc-L. Lesser use of the Non-Soc-L may lead to decreasing Non-Soc-L profi-
ciency, even in SLA children. Yet, parental engagement with children that
encourages continued Non-Soc-L use may be key in offsetting the overwhelm-
ing influence of the school language, leading to children who proudly speak
both the Non-Soc-L and the Soc-L, as shown in this video featuring several
bilingual children in the United States.27
4.6 Summary
We know very little about bilingual schoolchildren’s abilities in the Non-Soc-L,
and not enough about their abilities in the Soc-L. However, what evidence there
is suggests that bilingual schoolchildren greatly expand their abilities in the
Soc-L, regardless of whether they acquired the Soc-L from birth, as
a preschooler, or after age 6. By the end of middle childhood, most bilingual
children who first started regularly hearing the Soc-L before age 7 likely know
the Soc-L better than the Non-Soc-L they acquired from birth. Bilingual
schoolchildren generally appear to prefer speaking the Soc-L, and may have
stopped speaking the Non-Soc-L by the end of middle childhood (or earlier),
especially if they acquired the Soc-L from birth. As at earlier stages of devel-
opment, factors likely supporting children’s continued Non-Soc-L use are high
frequency of language input and a continued communicative need to speak the
Non-Soc-L as negotiated through parent–child interaction.
5 Socioeconomic Status and Bilingual Development

in Childhood
Research on monolingual children has found low(er) socioeconomic status
(SES) to be a risk factor for English language development in infancy
(Burchinal et al., 2008), early childhood (Levine et al., 2020), and middle
childhood (Linberg et al., 2019). Similarly, lower SES was correlated with
lower linguistic abilities in studies of Italian (Dicataldo & Roch, 2020) and
German (Linberg et al., 2019) monolinguals. In contrast, no relation between
SES and Spanish monolingual infants lexical development has been found
(DeAnda et al., 2016a; Jackson-Maldonado et al., 1993).
These contrasting findings for different languages complicate attempts at
investigating the role of SES for bilingual children’s development. A child
could be showing SES influence on a linguistic measure in one language but
not in the other. DeAnda et al. (2016a) investigated links between SES and 16-
month-old bilingual infants’ more frequently heard language. For 36 mostly
Spanish-hearing infants there was no relation between maternal SES and
vocabulary comprehension (see also Marchman et al., 2020). Yet for 25 mostly
English-hearing infants there was, with higher-SES mothers having better
performing infants. Although DeAnda et al. did not study the same children’s
27
Stephanie Meade. (2014, February 10). Many Languages, One America: The Voices of Our
Children [Video]. YouTube. www.youtube.com/watch?v=1FToY3BfHRU
performance in two languages, these results suggest that bilingual children’s

development may differentially relate to the same SES factor depending on
which language is assessed. This is exactly what Collins and Toppelberg (2021)
found for 228 Spanish–English schoolchildren: English proficiency was related
to SES, but Spanish proficiency was not, echoing findings for 90 Spanish–
English BFLA toddlers (Place & Hoff, 2016). Similarly, Dutch–Turkish 6-year-
olds’ Turkish vocabulary was not related to SES, whereas Dutch vocabulary
was (Prevoo et al., 2014). Furthermore, SES may correlate with measures
combining both languages but not with single-language measures. Ramírez-
Esparza et al. (2017) found this to be the case for Spanish–English 24-month-
old BFLA infants’ production vocabulary size. Also, SES differences may play
out differently in bilingual vs. monolingual children (Korecky-Kröll et al.,
2019).
If SES moderates child language development, it is through the ways in
which parents interact with their children (shown for monolingual parents in
Pace et al., 2017, and Scheele et al., 2010). In bilingual settings it may be
particularly important to take into account maternal as well as paternal SES
measures (Collins & Toppelberg, 2021; Prevoo et al., 2014; Scheele et al.,
2010). Fathers may speak a different language to children than mothers, or
may speak more of a particular language to children than mothers. However, the
extent to which SES functions as the only or main factor underlying how parents
verbally engage with children is not known. In bilingual families, where
languages and hence cultures are in contact, culturally determined patterns of
parent–child interaction may moderate or even clash with SES-related propen-
sities toward particular interaction patterns (DeAnda et al., 2016a; Gampe et al.,
2020; Ramírez-Esparza et al., 2017). SES and cultural interactions may help to
explain findings such Byers-Heinlein et al.’s (2020) showing that 6- to 9-month-
olds from higher-SES bilingual families preferred IDS more than peers from
lower-SES bilingual families. However, the evidence was not very strong, and
there was no SES difference for 12- to 15-month-olds (regardless of SES, all
bilingual infants preferred IDS over ADS; Section 2.2).
In bilingual settings involving immigrant parents it is unclear which SES
measures should be used (Gatt et al., 2020). Also, using just a single type of
measure may not adequately capture SES: Income measures may miss highly
educated parents who failed to monetize their social capital in the country to
which they emigrated; measures focusing only on education may miss parents
who are living in poverty in spite of being highly educated in their country of
origin. There is the added difficulty that bilingual children’s language learning
histories and associated exposure patterns need to be taken into account.
Dicataldo and Roch’s (2020) study of a heterogeneous sample of 4- to 6-year-
olds widely differing in both SES and degree of bilingual exposure demon-
strated that, for some language measures, differences in SES were more rele-
vant, regardless of bilingual experience, whereas other language measures were
more correlated with degree of bilingual exposure but independent of SES.
Byers-Heinlein et al. (2019) convincingly argued that, because of the ubi-
quity of child bilingualism and relations between bilingualism and many
aspects of development, developmental scientists should routinely document
children’s language backgrounds in any developmental research, including
precise ages of first regular exposure to each language for bilinguals. They
likewise proposed that scholars routinely report sample SES. These recom-
mendations are also relevant to ethnographic work undertaken in sociolinguis-
tically oriented research. Only with more extensive background documentation
in studies of child bilinguals will the connections between individual children’s
language learning trajectories and the contexts in which they develop their
bilingualism become clear.
6 Summary and Conclusion

6.1 Summary
There is a great deal of variability in children’s developing bilingualism, and
bilingual children form a very heterogeneous population in terms of the lan-
guage-related developmental trajectories they take. Yet, in spite of the great
number of different language pairs that children acquire, several generalizations
can be made as a function of the three basic environmental settings in which
children learn their languages: Bilingual First Language Acquisition (BFLA),
Early Second Language Acquisition (ESLA), or SLA (Second Language
Acquisition). These can be succinctly summarized:
1) Unlike in (E)SLA, BFLA children learn to understand two languages in

early infancy.
2) Unlike in (E)SLA, BFLA children say words in two languages when they
are 2 years old.
3) Unlike in (E)SLA, many BFLA children start to form short and then longer
and more complex sentences in Language A and Language Alpha at
virtually the same age. Typically, ESLA and SLA children first learn to
say sentences in their L1. Only much later do sentences in the L2 appear.
4) BFLA preschoolers speak each language without any systematic morpho-
syntactic influence from the other language. This is not the case for ESLA
preschoolers, nor for SLA schoolchildren, at least not in the first stages of
non-imitated sentence production.
5) The mixed utterances that child bilinguals in early childhood may use
mostly concern insertions of single words or phrases from language
X into an utterance otherwise in language Y.
6) BFLA children hardly have an “accent” in their two languages. (E)SLA
children frequently show influences from the L1 in the L2 at the sound
level, especially when they first speak the L2. In many (E)SLA children,
this “foreign accent” eventually disappears.
7) At school entry, the Soc-L is often the better developed language in BFLA
children. For ESLA children the Non-Soc-L is usually better developed.
SLA children only know their Non-Soc-L at the time of school entry.
8) Uneven development, where bilingual children of any age do not perform
equally well in both languages, is common.
9) There is vast variability among (E)SLA schoolchildren in the degrees to
which they develop proficiency in their Soc-L and in the rates of their Soc-
L development.
10) By age 11, many bilingual children have high levels of proficiency in their
Soc-L, even if they started learning it after age 6.
11) By age 11, many BFLA children have lost the ability to speak their Non-
Soc-L. Many other bilingual children may have low levels of proficiency in
their Non-Soc-L. BFLA and ESLA children appear at greatest risk.
12) Bilingual children can be highly proficient in two or more languages,
regardless of when they started learning them.
13) By age 11 and often long before, bilingual schoolchildren who are able to
form sentences in two languages can fluently switch from one language to
the other in a single conversation.
14) Bilingual children of any age address others mainly in a language they
know their interlocutors will understand.
Bilingual children’s different language learning environments (BFLA,

ESLA, and SLA) lead to a great heterogeneity of individual profiles for day-
to-day language use across the first decade of life. It is important to recognize
this heterogeneity both in clinical practice (De Houwer, 2018a) and in research.
The fundamental differences between BFLA and ESLA in infancy and early
childhood need to be taken into account much more than they are. Some studies
incorrectly refer to toddlers who started hearing their L2 at 1 or 1;6 as BFLA
children. Yet, early perception and word learning in BFLA infants’ first 12
months is influenced by the presence of two input languages, as is early word
production. This finding of different developmental paths warrants a clear
distinction with ESLA infants who started off with monolingual input and
added an L2 afterwards. By contrast, studies of preschoolers may refer to L1
and L2 even if children experienced no chronological difference between

languages. This failure to distinguish between BFLA and ESLA preschoolers
makes it impossible to correctly interpret any findings. If studies of bilingual
schoolchildren do not give information on children’s language learning trajec-
tories, many results will be equally uninterpretable.
Developmentalists may be surprised not to see child gender featured so far.
Child gender has hardly featured in studies of bilingual children focusing on
language. Studies that have considered the possible role of gender have not
found any relation with language outcomes (Groba et al., 2019; Kim et al.,
2014; Lauro et al., 2020).
In Western societies, BFLA is often held up as the ideal way to become “truly”
bilingual (many hundreds of students, parents, educators, and even researchers
have expressed this idea to me over the years). Most people do not realize that
many BFLA children do not, in fact, speak two languages after early childhood.
ESLA and SLA children actually fare better, because by the end of middle
childhood most are likely proficient in both the school language and the language
heard at home. The idealization of BFLA may underlie the notion that it is best to
start L2 learning as early in life as possible (hundreds of websites in various
countries subscribe to this notion). Whereas there are no inherent reasons why
children should not be given the chance to learn a new language from an early age,
the strong belief in “the earlier, the better” is leading many parents to take
up second jobs and loans in order to pay for (mostly English) foreign language
classes, and sets up high expectations for children’s rapid L2 learning. However, as
reviewed in this Element, learning to proficiently speak two languages in childhood
is not automatic: It takes a lot of time and practice. Supporting dual language
learning also requires a great deal of effort on the part of people engaging with
young children. The good news is that the older children are, the faster they are in
learning a new language naturalistically. It takes 10-year-old BFLA children and
monolinguals 10 years to learn to use a Soc-L at a 10-year-old level. SLA students
develop Soc-L skills at a much faster pace: If they start acquiring the Soc-L at age
6, SLA students do not need 10 years to reach a 10-year-old level. Many will reach
that level after 3 or 4 years. The main reasons for this later language learning
advantage may lie in SLA children’s greater cognitive maturity when they start
learning their new L2, and in their greater prior experience with their first language.
All bilingual children in Soc-L education eventually learn to speak the Soc-L.
The Non-Soc-L does not fare as well, and may be at risk of stagnating or being
lost. As explained in Section 6.2, this has repercussions for families’ harmoni-
ous bilingualism, that is, family members’ subjectively neutral or positive
experience with aspects of the bilingual setting they find themselves in (De
Houwer, 2020a).
6.2 Harmonious Bilingualism

In a monolingual setting parents quite naturally expect children to speak the
language they speak to them and to get better and better at speaking that
language. Parents in a bilingual family will often have the same expectation.
It comes as a shocking surprise to find that children do not speak the Non-Soc-L
(De Houwer, 2017c). Parents may feel regret, remorse, and guilt. They also may
feel anger at their children and may feel that children reject them by rejecting
their language, thus detracting from parental well-being, and, thus, harmonious
bilingualism (reviews in De Houwer, 2017c, 2020a). Children who speak only
the Soc-L may feel embarrassed that they can no longer communicate with
monolingual Non-Soc-L-speaking grandparents. As young adults, they may
feel that their parents did not do enough to support their development in the
Non-Soc-L and may deeply regret not speaking the Non-Soc-L (Nakamura,
2020). Language choices in the family strongly influence family dynamics and
interpersonal relationships among family members (Müller et al., 2020).
Bilingual children need not only develop skills in the Non-Soc-L. They need
to develop two languages. If they do so, their well-being is enhanced (De
Houwer, 2020a). Bilinguals in early and middle childhood with better skills in
each language than peers had better social-emotional skills (Sun et al., 2021).
However, on the way toward developing their languages children encounter
several hurdles that negatively impact their well-being, and thus harmonious
bilingualism. The clearest sign that children may feel unhappy in relation to
their linguistic environment comes from monolingual preschoolers who find
themselves in a group setting where a language is used that they do not
understand. As described in Section 3.2.1, many ESLA-to-be children withdraw
from the communicative goings-on in that group setting, may show signs of
depression, and are silent. At home they may also appear depressed, but they are
still speaking their L1 (Drury, 2007). Manigand (1999) described how he
observed several silent preschoolers in L2-preschools and tried to interpret the
nature of their silence, taking into account information about children’s home
and cultural backgrounds. With much patience he then tried to engage these
children in one-on-one interaction, at first nonverbally; however, without much
success, until he said a few words in the children’s Non-Soc-L, leading to smiles
and diminished reluctance to interact with him (see also Chang et al., 2007).
Eventually, through Manigand translating words between the Non-Soc-L and
the Soc-L and with the help of pictures and games, children started to say some
words in the Soc-L. Itoh and Hatch (1978) recounted a similar trajectory. SLA
schoolchildren may experience lower levels of self-control and interpersonal
skills and higher levels of internalizing problems compared to bilingual children
who speak the Soc-L at school entry (Han, 2010; Han & Huang, 2010). To avoid
such ill effects of a relatively late start in the Soc-L, Han (2010) proposed that
schools should implement high-quality L2 programs (Kim et al., 2015), rather
than the “sink-or-swim” approach characteristic of so many schools.
Children and families need positive support. Childcare centers and pre-
schools should ensure that ESLA preschoolers do not experience a possibly
traumatic long “silent period.” Pedagogical approaches that recognize and value
all languages that children bring to preschool are essential (Chilla & Niebuhr-
Siebert, 2017; Chumak-Horbatsch, 2012). With reference to Spanish–English
US families, Hoff and Ribot (2017, p. 245) wrote: “Just as pediatricians
encourage English-speaking parents to read books and use rich turn-taking
conversations with their monolingual children, they should offer the same
powerful advice to parents speaking Spanish with their bilingual children.”
Initiatives like the Háblame Bebé28 app developed by Baralt et al. (2020) and
the research-based Harmonious Bilingualism Network29 may help families to
guide their children on a constructive and positive bilingual path.
6.3 Parting Words

Research on bilingual children has grown exponentially in the last two and
a half decades (Bayram et al., 2018). New research on bilingual children is
published in widely different publication channels, reflecting the diversity of the
disciplines and subdisciplines in which bilingual children’s language develop-
ment and its supporting factors are investigated: developmental psycholinguis-
tics, sociolinguistics, cognitive and developmental psychology, education,
applied linguistics, sociology, neuropsychology, social work, psychopathology,
clinical linguistics, and more. These fields differ widely in their methods and
orientations. Given this diversity and the extraordinary wealth of information
available, this Element has necessarily had to be selective. Furthermore, state-
ments indicating lack of knowledge about a particular phenomenon may be in
error, either because I was unaware of the relevant research or because a new
publication has appeared since this Element was written.
However, in spite of the wealth of studies available, bilingual children’s
individual trajectories beyond the first 3.5 years of life have hardly been docu-
mented. This means there is little information on the developmental course of
bilingual children’s language comprehension and production beyond the middle
of early childhood. Studies tend to focus on language outcomes for groups of
children, often averaging findings. This obscures individual developmental paths.
28
Háblame Bebé (2018) [website]. https://hablamebebe.org/
29
HaBilNet (n.d.) [website] www.habilnet.org/
Even though many studies have explored factors influencing bilingual out-
comes at one or more points in development, “almost nothing is known about
the long-term impact of home language management on children’s . . . bilingual
development” (Schwartz, 2020). There is an urgent need for developmental
studies to trace the dynamics of family language choices and practices over the
years and to determine how families manage the challenges associated with
living in environments that do not necessarily support children’s bilingualism.
The surge in research interest in bilingual children has undoubtedly been
helped by the general realization that bilingual children account for a very large
and ever increasing proportion of the children in preschools and schools across
the globe (Section 1). This increase has gone hand in hand with highly increased
mobility worldwide. Although there have been severe restrictions on inter-
national mobility in relation to the SARS-CoV-2 virus, child bilingualism will
continue to be part and parcel of societies across the globe. It is my hope that
scholars everywhere will continue to give child bilingualism the attention it
inherently deserves, without the still all-too-frequent focus on comparisons with
monolinguals, and notions of “accelerated” or “delayed” development in bilin-
guals. A deep-seated monolingual bias in many Western societies is one reason
that bilingual children may fail to experience harmonious bilingual develop-
ment. The research community needs to support harmonious bilingualism by
investigating it with attention to the many factors that threaten or foster it. Only
increased knowledge of these factors will help us to truly understand bilingual
development.
References
Aarssen, J. (1996). Relating events in two languages: Acquisition of cohesive
devices by Turkish-Dutch bilingual children at school age. Tilburg: Tilburg
University Press.
Águila, E., Ramon-Casas, M., Pons, F., & Bosch, L. (2007). La medida del léxico
productivo inicial: Aplicación de un cuestionario para la población bilingüe. In
E. Diez-Itza (Ed.), Estudios de desarrollo del lenguaje y educación (Vol. 32).
Oviedo, Spain: ICE Monografias de Aula Abierta, 163–71.
Ahrenholz, B., Hövelbrinks, B., Maak, D., & Zippel, W. (2013).
Mehrsprachigkeit an Thüringer Schulen (MaTS) – Ergebnisse einer
Fragebogenerhebung zu Mehrsprachigkeit an Erfurter Schulen. In İ. Dirim
& I. Oomen-Welke (Eds.), Mehrsprachigkeit in der Klasse: Wahrnehmen –
Aufgreifen – Fördern. Stuttgart: Fillibach bei Klett, 43–58.
Andruski, J. E., Casielles, E., & Nathan, G. (2014). Is bilingual babbling
language-specific? Some evidence from a case study of Spanish-English
dual acquisition. Bilingualism: Language and Cognition, 17, 660–72. doi:
http://doi.org/10.1017/S1366728913000655
Argyri, E., & Sorace, A. (2007). Crosslinguistic influence and language domin-
ance in older bilingual children. Bilingualism: Language and Cognition, 10,
79–99. doi:http://doi.org/10.1017/S1366728906002835
Armon-Lotem, S., & Meir, N. (2019). The nature of exposure and input in early
bilingualism. In A. De Houwer & L. Ortega (Eds.), The Cambridge handbook
of bilingualism. Cambridge: Cambridge University Press, 193–212. doi:
http://doi.org/10.1017/9781316831922.011
Autorengruppe Bildungsberichterstattung. (2018). Bildung in Deutschland
2018. Ein indikatorengestützter Bericht mit einer Analyse zu Wirkungen
und Erträgen von Bildung. Bielefeld: wby Media. doi:http://doi.org/10
.3278/6001820fw
Babatsouli, E., & Ball, M. J. (Eds.) (2020). An anthology of bilingual child
phonology. Bristol: Multilingual Matters. doi:http://doi.org/10.21832
/9781788928427
Baralt, M., Darcy-Mahoney, A., & Brito, N. H. (2020). Háblame Bebé: A phone
application intervention to support Hispanic children’s early language envir-
onments and bilingualism. Child Language Teaching & Therapy, 36 (1),
33–57. doi:http://doi.org/10.1177/0265659020903779
Bayram, F., Miller, D., Rothman, J., & Serratrice, L. (2018). Studies in bilin-
gualism: 25 years in the making. In D. Miller, F. Bayram, J. Rothman, &
References 61
L. Serratrice (Eds.), Bilingual cognition and language: The state of the

science across its subfields. Amsterdam: John Benjamins, 1–12. doi:http://
doi.org/10.1075/sibil.54.01bay
Bergelson, E., & Swingley, D. (2012). At 6–9 months, human infants know the
meanings of many common nouns. PNAS, 109(9), 3253–3258. doi:http://doi
.org/10.1073/pnas.1113380109
Bhatia, T. K., & Ritchie, W. C. (2013). The handbook of bilingualism and
multilingualism (2nd ed.). Oxford: Blackwell. doi:http://doi.org/10.1002
/9781118332382
Bialystok, E., Luk, G., Peets, K., & Yang, S. (2010). Receptive vocabulary
differences in monolingual and bilingual children. Bilingualism: Language
and Cognition, 13, 525–31. doi:http://doi.org/10.1017/S1366728909990423
Bijeljac-Babic, R., Höhle, B., & Nazzi, T. (2016). Early prosodic acquisition in
bilingual infants: The case of the perceptual trochaic bias. Frontiers in
Psychology, 7, 210. doi:http://doi.org/10.3389/fpsyg.2016.00210
Blom, E. (2010). Effects of input on the early grammatical development of
bilingual children. International Journal of Bilingualism, 14(4), 422–446.
doi:http://doi.org/10.1177/1367006910370917
Blom, E., & Paradis, J. (2015). Sources of individual differences in the acquisi-
tion of tense inflection by English second language learners with and without
specific language impairment. Applied Psycholinguistics, 36, 953–976. doi:
http://doi.org/10.1017/S014271641300057X
Bornstein, M. H. (2009). Toward a model of culture↔parent↔child transac-
tions. In A. Sameroff (Ed.), The transactional model of development: How
children and contexts shape each other. Washington, DC: American
Psychological Association, 139–161. doi:http://doi.org/10.1037/11877-008
Bornstein, M. H. (2019). A developmentalist’s viewpoint: “It’s about time!”
Ecological systems, transaction, and specificity as key developmental prin-
ciples in children’s changing worlds. In R. D. Parke & G. H. Elder, Jr. (Eds.),
Children in changing worlds. Sociocultural and temporal perspectives.
Cambridge: Cambridge University Press, 277–286. doi:http://doi.org/10
.1017/9781108264846.010
Bornstein, M. H. & Haynes, O. M. (1998). Vocabulary competence in early
childhood: Measurement, latent construct, and predictive validity. Child
Development, 69(3), 654–671. doi:http://doi.org/10.1111/j.1467-8624
.1998.tb06235.x
Bosch, L., Figueras, F., Teixidó, M., & Ramon-Casas, M. (2013). Rapid gains in
segmenting fluent speech when words match the rhythmic unit: Evidence
from infants acquiring syllable-timed languages. Frontiers in Psychology, 4,
106. doi:http://doi.org/10.3389/fpsyg.2013.00106
62 References
Bosch, L., & Ramon-Casas, M. (2011). Variability in vowel production by

bilingual speakers: Can input properties hinder the early stabilization of
contrastive categories? Journal of Phonetics, 39, 514–526. doi:http://doi
.org/10.1016/j.wocn.2011.02.001
Brojde, C. L., Ahmed, S., & Colunga, E. (2012). Bilingual and monolingual
children attend to different cues when learning new words. Frontiers in
Psychology, 3, 155. doi:http://doi.org/10.3389/fpsyg.2012.00155
Budde-Spengler, N., Sachse, S., & Rinker, T. (2018). The relationship between
input factors and early lexical knowledge in Turkish–German children.
International Journal of Bilingual Education and Bilingualism. doi:http://doi
.org/10.1080/13670050.2018.1543647
Burchinal, M., Vernon-Feagans, L., & Cox, M. (2008). Cumulative social risk,
parenting, and infant development in rural low-income communities.
Parenting: Science and Practice, 8, 41–69. doi:http://doi.org/10.1080
/15295190701830672
Burns, T., Yoshida, K., Hill, K., & Werker, J. T. (2007). The development
of phonetic representation in bilingual and monolingual infants.
Applied Psycholinguistics, 28, 455–474. doi:http://doi.org/10.1017
/S0142716407070257
Byers-Heinlein, K. (2017). Bilingualism affects 9-month-old infants’ expect-
ations about how words refer to kinds. Developmental Science, 20, e12486.
doi:http://doi.org/10.1111/desc.12486
Byers-Heinlein, K., Tsui, A., Bergmann, C., Black, A. K., Brown, A., Carbajal, M.
J., … Wermelinger, S. (2021). A multi-lab study of bilingual infants: Exploring
the preference for infant-directed speech. Advances in Methods and Practices in
Psychological Science, 4(1), 1–30 [Registered report]. doi:https://doi.org/
10.1177/2515245920974622
Byers-Heinlein, K., Burns, T. C., & Werker, J. F. (2010). The roots of bilingual-
ism in newborns. Psychological Science, 21(3), 343−348. doi:http://doi.org
/10.1177/0956797609360758
Byers-Heinlein, K., Esposito, A. G., Winsler, A., Marian, V., Castro‖, D. C., &
Luk, G. (2019). The case for measuring and reporting bilingualism in devel-
opmental research. Collabra: Psychology, 5(1), 37. doi:http://doi.org/10
.1525/collabra.233
Cantone, K. F. (2007). Code-switching in bilingual children. Dordrecht, the
Netherlands: Springer. doi:http://doi.org/10.1007/978-1-4020-5784-7
Carbajal, M. J., & Peperkamp, S. (2020). Dual language input and the impact of
language separation on early lexical development. Infancy, 25(1), 22–45. doi:
http://doi.org/10.1111/infa.12315
References 63
Chang, F., Crawford, G., Early, D., Bryant, D., Howes, C., Burchinal, M.,
Barbarin, O., Clifford, R., & Pianta, R. (2007). Spanish-speaking children’s
social and language development in pre-kindergarten classrooms. Early
Education and Development, 18(2), 243–69. doi:http://doi.org/10.1080
/10409280701282959
Chevalier, S. (2015). Trilingual language acquisition: Contextual factors influ-
encing active trilingualism in early childhood. Amsterdam: John Benjamins.
doi:http://doi.org/10.1075/tilar.16
Chevrot, J.-P., & Ghimenton, A. (2019). Bilingualism and bidialectalism. In
A. De Houwer & L. Ortega (Eds.), The Cambridge handbook of bilingualism.
Cambridge: Cambridge University Press, 510–23. doi:http://doi.org/10.1017
/9781316831922.026
Chilla, S., & Niebuhr-Siebert, S. (2017). Mehrsprachigkeit in der KiTa.
Grundlagen – Konzepte – Bildung. Stuttgart: Kohlhammer.
Chondrogianni, V., & Marinis, T. (2011). Differential effects of internal and
external factors on the development of vocabulary, morphology and complex
syntax in successive bilingual children. Linguistic Approaches to
Bilingualism, 1, 223–248. doi:http://doi.org/10.1075/lab.1.3.05cho
Chondrogianni V, & Schwartz, R. G. (2020). Case and word order in Greek
heritage children. Journal of Child Language, 47, 766–795. doi:http://doi.org
/10.1017/S0305000919000849
Chumak-Horbatsch, R. (2012). Linguistically Appropriate Practice: A guide for
working with young immigrant children. Toronto: University of Toronto
Press.
Clark, E. V. (1993). The lexicon in acquisition. Cambridge: Cambridge
University Press. doi:http://doi.org/10.1017/CBO9780511554377
Clark, E. V. (2003). Critical periods, time, and practice. University of
Pennsylvania Working Papers in Linguistics, 9(2), 39–48.
Collier, V., & Thomas, W. P. (2017). Validating the power of bilingual
schooling: Thirty-two years of large-scale, longitudinal research.
Annual Review of Applied Linguistics, 37, 203–17. doi:http://doi.org/10
.1017/S0267190517000034
Collins, B., & Toppelberg, C. (2021). The role of socioeconomic and sociocul-
tural predictors of Spanish and English proficiencies of young Latino chil-
dren of immigrants. Journal of Child Language, 48(1), 129–156. doi:http://
doi.org/10.1017/S0305000920000203
Comeau, L., & Genesee, F. (2001). Bilingual children’s repair strategies during
dyadic communication. In J. Cenoz & F. Genesee (Eds.), Trends in bilingual
acquisition. Amsterdam: John Benjamins, 231–256. doi:http://doi.org/10
.1075/tilar.1.11com
64 References
Conboy, B. T., & Thal, D. (2006). Ties between the lexicon and grammar: Cross-
sectional and longitudinal studies of bilingual toddlers. Child Development, 77,
712–735. doi:http://doi.org/10.1111/j.1467-8624.2006.00899.x
Cote, L. R., & Bornstein, M. H. (2014). Productive vocabulary among three
groups of bilingual American children: Comparison and prediction. First
Language, 34, 467–485. doi:http://doi.org/10.1177/0142723714560178
Cruz-Ferreira, M. (2006). Three is a crowd? Acquiring Portuguese in
a trilingual environment. Clevedon: Multilingual Matters. doi:http://doi.org
/10.21832/9781853598395
Czinglar, C., Rüdiger, J. O., Korecky-Kröll, K., Uzunkaya-Sharma, K., &
Dressler, W. U. (2017). Inputfaktoren im DaZ-Erwerb von sukzessiv bilin-
gualen Kindern mit L1 Türkisch. In I. Fuchs, S. Jeuk, & W. Knapp (Eds.),
Mehrsprachigkeit: Spracherwerb, Unterrichtsprozesse, Seiteneinstieg:
Beiträge aus dem 11. Workshop “Kinder und Jugendliche mit
Migrationshintergrund.” Stuttgart: Fillibach bei Klett, 15–34.
Dahoun, Z. K. S. (1995). Les couleurs du silence. Le mutisme des enfants de
migrants. Paris: Calmann-Lévy.
David, A., & Li, W. (2008). Individual differences in the lexical development of
French–English bilingual children. International Journal of Bilingualism
and Bilingual Education, 11(5), 598–618. doi:http://doi.org/10.1080
/13670050802149200
Davidson, D., Jergovic, D., Imami, Z., & Theodos, V. (1997). Monolingual and
bilingual children’s use of the mutual exclusivity constraint. Journal of Child
Language, 24, 3–24. doi:http://doi.org/10.1017/S0305000996002917
DeAnda, S., Arias-Trejo, N., Poulin-Dubois, D., Zesiger, P., & Friend, M.
(2016a). Minimal second language exposure, SES, and early word comprehen-
sion: New evidence from a direct assessment. Bilingualism: Language and
Cognition, 19(1), 162–180. doi:http://doi.org/10.1017/S1366728914000820
DeAnda, S., Poulin-Dubois, D., Zesiger, P., & Friend, M. (2016b). Lexical
processing and organization in bilingual first language acquisition: Guiding
future research. Psychological Bulletin, 142, 655–667. doi:http://doi.org/10
.1037/bul0000042
De Houwer, A. (1990). The acquisition of two languages from birth: A case
study. Cambridge: Cambridge University Press. doi:https://doi.org/10.1017
/CBO9780511519789
De Houwer, A. (2003). Home languages spoken in officially monolingual
Flanders: A survey. Plurilingua, 24, 79–96.
De Houwer, A. (2007). Parental language input patterns and children’s bilingual
use. Applied Psycholinguistics, 28(3), 411–424. doi:http://doi.org/10.1017
/S0142716407070221
References 65
De Houwer, A. (2009). Bilingual First Language Acquisition. Bristol:

Multilingual Matters. doi:http://doi.org/10.21832/9781847691507
De Houwer, A. (2014). The absolute frequency of maternal input to bilingual
and monolingual children: A first comparison. In T. Grüter & J. Paradis
(Eds.), Input and experience in bilingual development. Amsterdam: John
Benjamins, 37–58. doi:http://doi.org/10.1075/tilar.13.03deh
De Houwer, A. (2015a). Early foreign language teaching: Some critical remarks
and some recommendations. Babylonia, 01(14), 14–21.
De Houwer, A. (2015b). Integration und Interkulturalität in Kindertagesstätten
und in Kindergärten: Die Rolle der Nichtumgebungssprache für das
Wohlbefinden von Kleinkindern. In E. Reichert-Garschhammer, C. Kieferle,
M. Wertfein, & F. Becker-Stoll (Eds.), Inklusion und Partizipation. Vielfalt als
Chance und Anspruch. Göttingen: Vandenhoeck & Ruprecht, 113–125.
De Houwer, A. (2017a). Bilingual language input environments, intake, matur-
ity and practice. Bilingualism: Language and Cognition, 20, 19–20. doi:
De Houwer, A. (2017b). Early multilingualism and language awareness. In
J. Cenoz, D. Gorter & S. May (Eds.), Language awareness and multilingual-
ism. Encyclopedia of language and education (3rd ed.). Cham: Springer,
83–97. doi:http://doi.org/10.1007/978-3-319-02240-6_6
De Houwer, A. (2017c). Minority language parenting in Europe and children’s
well-being. In N. Cabrera & B. Leyendecker (Eds.), Handbook on positive
development of minority children and youth. Cham: Springer, 231–246. doi:
http://doi.org/10.1007/978-3-319-43645-6_14
De Houwer, A. (2018a). Input, context and early child bilingualism:
Implications for clinical practice. In A. Bar-On & Dorit Ravid (Eds.),
Handbook of communication disorders: Theoretical, empirical, and applied
linguistic perspectives. Berlin: Walter de Gruyter, 599–616. doi:http://doi.org
/10.1515/9781614514909-030
De Houwer, A. (2018b). The role of language input environments for language
outcomes and language acquisition in young bilingual children. In D. Miller,
F. Bayram, J. Rothman, & L. Serratrice (Eds.), Bilingual cognition and
language: The state of the science across its subfields. Amsterdam: John
Benjamins, 127–153. doi:http://doi.org/10.1075/sibil.54.07hou
De Houwer, A. (2019a). Equitable evaluation of bilingual children’s’s language
knowledge using the CDI: It really matters who you ask. Journal of
Monolingual and Bilingual Speech, 1(1), 32–54. doi:http://doi.org/10.1558
/jmbs.11184
De Houwer, A. (2019b). Language choice in bilingual interaction. In A. De
Houwer & L. Ortega (Eds.), The Cambridge Handbook of Bilingualism.
66 References

.1017/9781316831922.018
De Houwer, A. (2019c). Uninstructed language acquisition in multiple lan-
guage learners. In J. Darquennes, J. Salmons, & W. Vandenbussche (Eds.),
Language contact: An international handbook. Berlin: Mouton de Gruyter,
183–196. doi:http://doi.org/10.1515/9783110435351-016
De Houwer, A. (2020a). Harmonious Bilingualism: Well-being for families in
bilingual settings. In S. Eisenchlas & A. Schalley (Eds.), Handbook of home
language maintenance and development. Social and affective factors. Berlin:
Mouton de Gruyter, 63–83. doi:http://doi.org/10.1515/9781501510175-004
De Houwer, A. (2020b). Why do so many children who hear two languages
speak just a single language? Zeitschrift für Interkulturellen
Fremdsprachenunterricht, 25(1), 7–26.
De Houwer, A., & Bornstein, M. H. (2016a). Balance patterns in early bilingual
acquisition: A longitudinal study of word comprehension and production. In
C. Silva-Corvalán & J. Treffers-Daller (Eds.), Language dominance in bilin-
guals. Issues of measurement and operationalization. Cambridge: Cambridge
University Press, 134–155. doi:http://doi.org/10.1017/CBO9781107375345.007
De Houwer, A., & Bornstein, M. H. (2016b). Bilingual mothers’ language
choice in child-directed speech: Continuity and change. Journal of
Multilingual and Multicultural Development, 37(7), 680–693. doi:http://doi
.org/10.1080/01434632.2015.1127929
De Houwer, A., Bornstein, M. H., & De Coster, S. (2006). Early understanding
of two words for the same thing: A CDI study of lexical comprehension in
infant bilinguals. International Journal of Bilingualism, 10, 331–347. doi:
http://doi.org/10.1177/13670069060100030401
De Houwer, A., Bornstein, M. H., & Putnick, D. L. (2014). A bilingual-
monolingual comparison of young children’s vocabulary size: Evidence
from comprehension and production. Applied Psycholinguistics, 35,
De Houwer, A., & Nakamura, J. (accepted for publication 2021).
Developmental perspectives on parents’ use of discourse strategies with
bilingual children. Routledge.
De Houwer, A., & Ortega, L. (2019). Introduction: Learning, using, and
unlearning more than one language. In A. De Houwer & L. Ortega (Eds.),
The Cambridge handbook of bilingualism. Cambridge: Cambridge
University Press, 1–12. doi:http://doi.org/10.1017/9781316831922.001
Dicataldo, R., & Roch, M. (2020). Are the effects of variation in quantity of
daily bilingual exposure and socioeconomic status on language and cognitive
abilities independent in preschool children? International Journal of
References 67
Environmental Research and Public Health, 17, 4570. doi:http://doi.org/10

.3390/ijerph17124570
Drury, R. (2007). Young bilingual learners at home and school: Researching
multilingual voices. Stoke on Trent: Trentham Books.
Dubiel, B., & Guilfoyle, E. (2017). Language strength in bilingual children: The
Child HALA test. Heritage Language Journal, 14(1), 1–29.
Dunn, L. M., & Dunn, L. M. (1997). Peabody Picture Vocabulary Test (3rd ed.).
Minneapolis, MN: Pearson Assessments.
Durrant, S., Delle Luche, C., Cattani, A., & Floccia, C. (2014). Monodialectal
and multidialectal infants’ representation of familiar words. Journal of Child
Eilers, R. E., Oller, D. K., & Cobo-Lewis, A. B. (2002). Bilingualism and
cultural assimilation in Miami Hispanic children. In D. K. Oller &
R. E. Eilers (Eds.), Language and literacy in bilingual children. Clevedon:
Multilingual Matters, 43–63.
Eilers, R. E., Pearson, B. Z., & Cobo-Lewis, A. B. (2006). Social factors in
bilingual development: The Miami experience. In P. McCardle & E. Hoff
(Eds.), Childhood bilingualism: Research on infancy through school age.
Clevedon: Multilingual Matters, 68–90.
Ertanir, B., Kratzmann, J., Frank, M., Jahreiss, S., & Sachse, S. (2018). Dual
language competencies of Turkish–German children growing up in Germany:
Factors supportive of functioning dual language development. Frontiers in
Psychology, 9(2261). doi:http://doi.org/10.3389/fpsyg.2018.02261
Ertanir, B., Kratzmann, J., & Sachse, S. (2019). Sozio-emotionale Kompetenzen
mehrsprachiger Kindergartenkinder und deren Wechselwirkungen mit den
Sprachleistungen im Deutschen. Zeitschrift für Entwicklungspsychologie und
Pädagogische Psychologie, 51(1), 31–44. doi:http://doi.org/10.1026/0049-
8637/a000207
Fennell, C. T., & Byers-Heinlein, K. (2014). You sound like Mommy: Bilingual
and monolingual infants learn words best from speakers typical of their
language environments. International Journal of Behavioral Development,
38, 309–316. doi:http://doi.org/10.1026/0049-8637/a00020710.1177
/0165025414530631
Fenson, L., Dale, P., Reznick, J. S., Thal, D., Bates, E., Hartung, J., Pethick, S. J., &
Reilly, J. (1993). MacArthur Communicative Development Inventories: User’s
guide and technical manual. San Diego, CA: Singular Publishing Group.
Fish, M. S., García-Sierra, A., Ramírez-Esparza, N., & Kuhl, P. K. (2017).
Infant-directed speech in English and Spanish: Assessments of monolin-
gual and bilingual caregiver VOT. Journal of Phonetics, 63, 19–34. doi:
https://doi.org/10.1016/j.wocn.2017.04.003
68 References
Fuller, J. M. (2012). Bilingual pre-teens: Competing ideologies and multiple

identities in the U.S. and Germany. New York, NY: Routledge. doi:http://doi
.org/10.4324/9780203110645
Fuller, J. M. (2019). Ideologies of language, bilingualism, and monolingualism.
In A. De Houwer & L. Ortega (Eds.), The Cambridge handbook of bilingual-
ism. Cambridge: Cambridge University Press, 119–134. doi:http://doi.org/10
.1017/9781316831922.007
Gagarina, N., & Klassert, A. (2018). Input dominance and development of home
language in Russian-German bilinguals. Frontiers in Communication, 3(40).
doi:http://doi.org/10.3389/fcomm.2018.00040
Gampe, A., Hartmann, L., & Daum, M. M. (2020). Dynamic interaction
patterns of monolingual and bilingual infants with their parents.
Journal of Child Language, 47(1), 45–63. doi:http://doi.org/10.1017
/S0305000919000631
Gampe, A., Wermelinger, S., & Daum, M. M. (2019). Bilingual children adapt
to the needs of their communication partners, monolinguals do not. Child
Development, 90(1), 98–107. doi:http://doi.org/10.1111/cdev.13190
Garcia-Sierra, A., Rivera-Gaxiola, M., Percaccio, C. R., Conboy, B. T.,
Romo, H., Klarman, L., & Kuhl, P. K. (2011). Bilingual language learning:
An ERP study relating early brain responses to speech, language input, and
later word production. Journal of Phonetics, 39(4), 546–557. doi:http://doi
.org/10.1016/j.wocn.2011.07.002
Gathercole, V. C. M. (Ed.) (2007). Language transmission in bilingual families
in Wales. Cardiff: Welsh Language Board.
Gathercole, V. C. M., & Thomas, E. M. (2009). Bilingual first-language devel-
opment: Dominant language takeover, threatened minority language take-up.
Bilingualism: Language and Cognition, 12, 213–237. doi:http://doi.org/10
.1017/S1366728909004015
Gatt, D., Baldacchino, R., & Dodd, B. (2020). Which measure of socioeco-
nomic status best predicts bilingual lexical abilities and how? A focus on
four-year-olds exposed to two majority languages. Journal of Child
Language, 47(4), 737–765. doi:http://doi.org/10.1017/S0305000919000886
Genesee, F., Boivin, I., & Nicoladis, E. (1996). Talking with strangers: A study of
bilingual children’s communicative competence. Applied Psycholinguistics,
17, 427–442. doi:http://doi.org/10.1017/S0142716400008183
Genesee, F., & Delcenserie, A. (Eds.) (2016). Starting over – The language
development in internationally-adopted children. Amsterdam: John
Benjamins. doi:http://doi.org/10.1075/tilar.18
Gibson, T. A., Oller, D. K., Jarmulowicz, L., & Ethington, C. A. (2012). The
receptive-expressive gap in the vocabulary of young second-language
References 69
learners: Robustness and possible mechanisms. Bilingualism: Language and

Cognition, 15, 102–116. doi:http://doi.org/10.1017/S1366728910000490
Gollan, T. H., Starr, J., & Ferreira, V. S. (2015). More than use it or lose it: The
number-of-speakers effect on heritage language proficiency. Psychonomic
Bulletin and Review, 22(1), 147–55. doi:http://doi.org/10.3758/s13423-014-
0649-7
Granfeldt, J., Schlyter, J., & Kihlstedt, M. (2007). French as cL2, 2L1 and L1 in
pre-school children. In J. Granfeldt (Ed.), Studies in Romance bilingual
acquisition – age of onset and development of French and Spanish. Lund:
Lund University, 7–42.
Groba, A., De Houwer, A., Obrig, H., & Rossi, S. (2019). Bilingual and
monolingual first language acquisition experience differentially shapes chil-
dren’s property term learning: Evidence from behavioral and neurophysio-
logical measures. Brain Sciences, 9(2), 40. doi:http://doi.org/10.3390
/brainsci9020040
Grosjean, F. (1989). Neurolinguists beware! The bilingual is not two mono-
linguals in one person. Brain and Language, 36, 3–15. doi:http://doi.org/
10.1016/0093-934X(89)90048-5
Grosjean, F., & Byers-Heinlein, K. (2018). The listening bilingual: Speech
perception, comprehension, and bilingualism. Hoboken, NJ: John Wiley &
Sons. doi:http://doi.org/10.1002/9781118835722
Gutiérrez-Clellen, V. F., & Kreiter, J. (2003). Understanding child bilingual
acquisition using parent and teacher reports. Applied Psycholinguistics, 24,
Haberzettl, S., Dimroth, C., Wulff, N., & Czinglar, C. (2013). Erwerb des
Deutschen als Zweitsprache im Grundschulalter. In A. Berndt (Ed.),
Fremdsprachen in der Perspektive lebenslangen Lernens. Frankfurt am
Main: Peter Lang, 143–161.
Haman, E., Łuniewska, M., & Pomiechowska, B. (2015). Designing Cross-
Linguistic Lexical Tasks (CLTs) for bilingual preschool children. In S. Armon-
Lotem, J. de Jong, & M. Meir (Eds.), Methods for assessing multilingual children:
Disentangling bilingualism from language impairment. Bristol: Multilingual
Matters, 194–238. doi:http://doi.org/10.21832/9781783093137-010
Hammer, C. S., Komaroff, E., Rodriguez, B. L., Lopez, L. M., Scarpino, S. E.,
& Goldstein, B. (2012). Predicting Spanish-English bilingual children’s
language abilities. Journal of Speech, Language, and Hearing Research, 55
(5), 1251–1264. doi:http://doi.org/10.1044/1092-4388(2012/11-0016)
Han, W.-J. (2010). Bilingualism and socioemotional well-being. Children
and Youth Services Review, 32, 720–31. doi:http://doi.org/10.1016/
j.childyouth.2010.01.009
70 References
Han, W.-J., & Huang, C.-C. (2010). The forgotten treasure: Bilingualism and
Asian children’s emotional and behavioral health. American Journal of Public
Health, 100(5), 831–838. doi:http://doi.org/10.2105/AJPH.2009.174219
Hatch, E. M. (Ed.) (1978). Second language acquisition: A book of readings.
Rowley, MA: Newbury House Publishers.
Havy, M., Bouchon, C., & Nazzi, T. (2016). Phonetic processing when learning
words: The case of bilingual infants. International Journal of Behavioral
Development, 40, 41–52. doi:http://doi.org/10.1177/0165025415570646
Haznedar, B. (1997). L2 acquisition by a Turkish-speaking child: Evidence for
L1 influence. In E. Hughes, M. Hughes, & A. Greenhill (Eds.), Proceedings
of the 21st annual Boston University conference on language development.
Somerville, MA: Cascadilla Press, 245–56.
Head Zauche, L., Darcy Mahoney, A., Thul, T., Zauche, M., Weldon, A., &
Stapel-Wax, J. (2017). The power of language nutrition for children’s brain
development, health, and future academic achievement. Journal of Pediatric
Health Care 31: 493–503. doi:http://doi.org/10.1016/j.pedhc.2017.01.007
Henderson, A. M. E., & Scott, J. C. (2015). She called that thing a mido, but
should you call it a mido too? Linguistic experience influences infants’
expectations of conventionality. Frontiers in Psychology, 6, 332. doi:http://
doi.org/10.3389/fpsyg.2015.00332
Hoff, E., Core, C., Place, S., Rumiche, R., Señor, M., & Parra, M. (2012). Dual
language exposure and early bilingual development. Journal of Child
Hoff, E., Core, C., & Shanks, K. F. (2020). The quality of child-directed speech
depends on the speaker’s language proficiency. Journal of Child Language,
47, 132–145. doi:http://doi.org/10.1017/S030500091900028X
Hoff, E., & Ribot, K. M. (2017). Language growth in English monolingual and
Spanish-English bilingual children from 2.5 to 5 years. Journal of Pediatrics,
190, 241–245.e1. doi:http://doi.org/10.1016/j.jpeds.2017.06.071
Höhle, B., Bijeljac-Babic, R., & Nazzi, T. (2020). Variability and stability in
early language acquisition: Comparing monolingual and bilingual infants”
speech perception and word recognition. Bilingualism: Language and
Cognition, 23, 56–71. doi:http://doi.org/10.1017/S1366728919000348
Holowka, S., Brosseau-Lapré, F., & Petitto, L. A. (2002). Semantic and concep-
tual knowledge underlying bilingual babies’ first signs and words. Language
Learning, 52, 205–62. doi:http://doi.org/10.1111/0023-8333.00184
Hurtado, N., Grüter, T., Marchman, V. A., & Fernald, A. (2014). Relative
language exposure, processing efficiency and vocabulary in Spanish–
English bilingual toddlers. Bilingualism: Language and Cognition, 17(01),
189–202. doi:http://doi.org/10.1017/S136672891300014X
References 71
Idiazábal, I. (1984). Conciencia bilingüe del niño bilingüe. In M. Siguan (Ed.),

Adquisición precoz de una segunda lengua. Barcelona: Publicacions
i edicions de la universitat de Barcelona, 55–63.
Itani-Adams, Y., Iwasaki, J., & Kawaguchi, S. (2017). Similarities and differ-
ences between simultaneous and successive bilingual children: Acquisition of
Japanese morphology. International Journal of Applied Linguistics & English
Literature, 6(7), 268–276. doi:http://doi.org/10.7575/aiac.ijalel.v.6n.7p.268
Itoh, H., & Hatch, E. (1978). Second Language Acquisition: A case study. In
E. Hatch (Ed.), Second Language Acquisition: A book of readings. Rowley,
MA: Newbury House, 77–88.
Jackson-Maldonado, D., Thal, D., Marchman, V., Bates, E., & Gutiérrez-
Clellen, V. (1993). Early lexical development in Spanish-speaking infants
and toddlers. Journal of Child Language, 20, 523–549. doi:http://doi.org/10
.1017/s0305000900008461
Jahreiß, S., Ertanir, B., Sachse, S., & Kratzmann, J. (2018). Sprachliche
Interaktionen in Kindertageseinrichtungen mit hohem Anteil an mehrspra-
chigen Kindern. Forschung Sprache, 2, 32–41.
Jardak, A., & Byers-Heinlein, B. (2019). Labels or concepts? The development
of semantic networks in bilingual two-year-olds. Child Development, 90(2),
e212–e229. doi:http://doi.org/10.1111/cdev.13050
Jia, G., Chen, J., Kim, H. Y., Chan, P.-S., & Jeung, C. (2014). Bilingual lexical
skills of school-age children with Chinese and Korean heritage languages in
the United States. International Journal of Behavioral Development, 38,
350–358. doi:http://doi.org/10.1177/0165025414533224
Juan-Garau, M., & Lyster, R. (2019). Becoming bilingual through additive
immersive programs. In A. De Houwer & L. Ortega (Eds.), The Cambridge
handbook of bilingualism. Cambridge: Cambridge University Press,
213–232. doi:http://doi.org/10.1017/9781316831922.012
Kalashnikova, M., Escudero, P., & Kidd, E. (2018). The development of
fast-mapping and novel word retention strategies in monolingual and bilin-
gual infants. Developmental Science, 21(6), e12674. doi:http://doi.org/10
.1111/desc.12674
Kan, P. F., & Kohnert, K. (2005). Preschoolers learning Hmong and English:
Lexical-semantic skills in L1 and L2. Journal of Speech Language and
Hearing Research, 48(2), 372–383. doi:http://doi.org/10.1044/1092-4388
(2005/026)
Karlsen, J., Halaas Lyster, S.-A., & Lervåg, A. (2017). Vocabulary development
in Norwegian L1 and L2 learners in the kindergarten–school transition.
Journal of Child Language, 44(2), 402–426. doi:http://doi.org/10.1017
/S0305000916000106
72 References
Karrebaek, M. S. (2013). “Don’t speak like that to her!”: Linguistic

minority children’s socialization into an ideology of monolingualism.
Journal of Sociolinguistics, 17(3), 355–375. doi:http://doi.org/10.1111
/josl.12035
Kaufman, D. (2001). Tales of L1 attrition – Evidence from pre-puberty children.
In T. Ammerlaan, M. Hulsen, H. Strating & K. Yağmur (Eds.),
Sociolinguistic and psycholinguistic perspectives on maintenance and loss
of minority languages. Münster: Waxmann, 185–202.
Kaufman, D., & Aronoff, M. (1991). Morphological disintegration and recon-
struction in first language attrition. In H. Seliger & R. Vago (Eds.), First
language attrition. Cambridge: Cambridge University Press, 175–88. doi:
http://doi.org/10.1017/CBO9780511620720.012
Keller, K., Troesch, L., & Grob, A. (2013). Shyness as a risk factor for local
language skills of immigrant preschoolers. Journal of Applied
Developmental Psychology, 34, 328–335. doi:10.1016/j.appdev.2013.07.
001
Kibler, A. K., Palacios, N., & Simpson Baird, A. (2014). The influence of older
siblings on language use among second generation Latino preschoolers.
TESOL Quarterly, 48, 164–175. doi:http://doi.org/10.1002/tesq.151
Kim, Y. K., Curby, T. W., & Winsler, A. (2014). Child, family, and school
characteristics related to English proficiency development among
low-income, dual language learners. Developmental Psychology, 50(12),
2600–2613. doi:http://doi.org/10.1037/a0038050
Kim, Y. K., Hutchison, L. A., & Winsler, A. (2015). Bilingual education in the
United States: An historical overview and examination of two-way
immersion. Educational Review, 67(2), 236–252. doi:http://doi.org/10.1080
/00131911.2013.865593
Köppe, R. (1996). Language differentiation in bilingual children: The develop-
ment of grammatical and pragmatic competence. Linguistics, 34, 927–954.
doi:http://doi.org/10.1515/ling.1996.34.5.927
Korecky-Kröll, K., Czinglar, C., Uzunkaya-Sharma, K., Sommer-Lolei, S., &
Dressler, W. U. (2016). Das INPUT-Projekt: Wortschatz-und Grammatikerwerb
von ein-und zweisprachigen Wiener Kindergartenkindern. logoTHEMA, 13(1),
16–22.
Korecky-Kröll, K., Dobek, N., Blaschitz, V., Sommer-Lolei, S., Boniecki, M.,
Uzunkaya-Sharma, K., & Dressler, W. U. (2019). Vocabulary as a central link
between phonological working memory and narrative competence: Evidence
from monolingual and bilingual 4-year olds from different socioeconomic
backgrounds. Language and Speech, 62, 546–569. doi:http://doi.org/10.1177
/0023830918796691
References 73
Kuo, E. C.-Y. (1974). The family and bilingual socialization: A sociolinguistic

study of a sample of Chinese children in the United States. The Journal of
Social Psychology, 92, 181–191.
Lanza, E. (1992). Can bilingual two-year-olds code-switch? Journal of Child
Language, 19(3), 633–658. doi:http://doi.org/10.1017/s0305000900011600
Lauro, J., Core, C., & Hoff, E. (2020). Explaining individual differences in
trajectories of simultaneous bilingual development: Contributions of child
and environmental factors. Child Development, 91(6), 2063–2082. doi:http://
doi.org/10.1111/cdev.13409
Legacy, J., Zesiger, P., Friend, M., & Poulin-Dubois, D. (2016). Vocabulary
size, translation equivalents, and efficiency in word recognition in very
young bilinguals. Journal of Child Language, 43(4), 760–783. doi:http://doi
.org/10.1017/S0305000915000252
Legacy, J., Zesiger, P., Friend, M., & Poulin-Dubois, D. (2018). Vocabulary size
and speed of word recognition in very young French–English bilinguals:
A longitudinal study. Bilingualism: Language and Cognition, 21(1),
Leist-Villis, A. (2004). Zweisprachigkeit im Kontext sozialer Netzwerke.
Unterstützende Rahmenbedingungen zweisprachiger Entwicklung und
Erziehung am Beispiel griechisch-deutsch. Münster: Waxmann.
Leopold, W. (1939–1949). Speech development of a bilingual child: A linguist’s
record (4 volumes). Evanston, IL: Northwestern University Press.
Levine, D., Pace, A., Luo, R., Hirsh-Pasek, K., Golinkoff, R. M., de Villiers, J.,
Iglesias, A., & Wilson, M. S. (2020). Evaluating socioeconomic gaps in
preschoolers’ vocabulary, syntax, and language process skills with the
Quick Interactive Language Screener (QUILS). Early Childhood Research
Quarterly, 50, 114–128. doi:http://doi.org/10.1016/j.ecresq.2018.11.006
Liberman, Z., Woodward, A. L., Keysar, B., & Kinzler, K. D. (2017). Exposure
to multiple languages enhances communication skills in infancy.
Developmental Science, 20, e12420. doi:http://doi.org/10.1111/desc.12420
Linberg, T., Schneider, T., Waldfogel, J., & Wang, Y. (2019). Socioeconomic
status gaps in child cognitive development in Germany and the United States.
Social Science Research, 79, 1–31. doi:http://doi.org/10.1016/j
.ssresearch.2018.11.002
Lindquist, H., & Gram Garmann, N. (2021). Toddlers and their translingual
practicing homes. International Journal of Multilingualism, 18(1), 59–72.
doi:http://doi.org/10.1080/14790718.2019.1604712
Liu, L., & Kager, R. (2017). Perception of tones by bilingual infants learning
non-tone languages. Bilingualism: Language and Cognition, 20, 561–575.
doi:http://doi.org/10.1017/S1366728916000183
74 References
Li Wei. (2011). The early acquisition of English as a second language: The case
of young Chinese learners of English in Britain. In A. De Houwer &
A. Wilton (Eds.), English in Europe today: Educational and sociocultural
perspectives. Amsterdam: John Benjamins, 105–122. doi:http://doi.org/10
.1075/aals.8.07wei
Luo, R., Escobar, K., & Tamis-LeMonda, C. S. (2020). Heterogeneity in the
trajectories of US Latine mothers” dual-language input from infancy to
preschool. First Language, 40(3), 275–99. doi:http://doi.org/10.1177
/0142723720915401
MacLeod,A. A. N., Fabiano-Smith, L., Boegner-Pagé, S., & Fontolliet, S.
(2013). Simultaneous bilingual language acquisition: The role of par-
ental input on receptive vocabulary development. Child Language
Teaching and Therapy, 29, 131–142. doi:http://doi.org/10.1177
/0265659012466862
MacSwan, J., & Pray, L. (2005). Learning English bilingually: Age of onset of
exposure and rate of acquisition among English language learners in
a bilingual education program. Bilingual Research Journal, 29(3),
653–678. doi:http://doi.org/10.1080/15235882.2005.10162857
Mancilla-Martinez, J., & Lesaux, N. (2011). The gap between Spanish
speakers” word reading and word knowledge: A longitudinal study. Child
Development, 82(5),1544–1560. doi:http://doi.org/10.1111/j.1467-8624
.2011.01633.x
Maneva, B., & Genesee, F. (2002). Bilingual babbling: Evidence for language
differentiation in dual language acquisition. In B. Skarbela, S. Fish, & A H.-
J Do (Eds.), Boston University Conference on Language Development 26
Proceedings. Sommerville, MA: Cascadilla Press, 383–392.
Manigand, A. (1999). Le silence des enfants turcs à l’école. Psychologie &
Éducation, 37, 57–73.
Marchman, V. A., Bermúdez, V. N., Bang, J. Y., & Fernald, A. (2020). Off to
a good start: Early Spanish-language processing efficiency supports Spanish-
and English-language outcomes at 4½ years in sequential bilinguals.
Developmental Science, 23(6) e12973. doi:http://doi.org/10.1111/desc.12973
Marchman, V. A., Fernald, A., & Hurtado, N. (2010). How vocabulary size in
two languages relates to efficiency in spoken word recognition by young
Spanish–English bilinguals. Journal of Child Language, 37(04), 817–840.
doi:http://doi.org/10.1017/S0305000909990055
Marchman, V., Martínez, L., Hurtado, N., Grüter, T., & Fernald, A. (2017).
Caregiver talk to young Spanish-English bilinguals: Comparing direct obser-
vation and parent-report measures of dual-language exposure. Developmental
Science, 20(1), e12425. doi:http://doi.org/10.1111/desc.12425
References 75
Marchman, V. A., Martínez-Sussmann, C., & Dale, P. S. (2004). The

language-specific nature of grammatical development: Evidence from bilin-
gual language learners. Developmental Science, 7, 212–224. doi:http://doi
.org/10.1111/j.1467-7687.2004.00340.x
Marecka, M., Wrembel, M., Otwinowska, A., Szewczyk, J., Banasik-
Jemielniak, N., & Wodniecka, Z. (2020). Bilingual children’s phonology
shows evidence of transfer, but not deceleration in their L1. Studies in
Second Language Acquisition, 42(1), 89–114. doi:http://doi.org/10.1017
/S0272263119000408
Markman, E. M., & Wachtel, G. F. (1988). Children’s use of mutual exclusivity
to constrain the meanings of words. Cognitive Psychology, 20, 121–157. doi:
http://doi.org/10.1016/0010-0285(88)90017-5
Marinis, T., Armon-Lotem,S., & Pontikas, G. (Eds.) (2017). Language impair-
ment in bilingual children: State of the art 2017. Special Issue. Linguistic
Approaches to Bilingualism, 7 (3–4). doi:http://doi.org/10.1075/lab.7.3-4
Mayr, R., & Montanari, S. (2015). Cross-linguistic interaction in trilingual
phonological development: The role of the input in the acquisition of the
voicing contrast. Journal of Child Language, 42(5), 1006–1035. doi:http://
doi.org/10.1017/S0305000914000592
Meisel, J. M. (1989). Early differentiation of languages in bilingual children. In
K. Hyltenstam & L. K. Obler (Eds.), Bilingualism across the lifespan: Aspects
of acquisition, maturity and loss. Cambridge: Cambridge University Press,
13–40. doi:http://doi.org/10.1017/CBO9780511611780.003
Meisel, J. M. (2008). Âge du début de l’acquisition successive du bilinguisme.
Effets sur le développement grammatical. In M. Kail, M. Fayol, &
M. Hickman (Eds.), Apprentissage des langues premières et secondes.
Paris: Editions du CNRS, 245–272.
Meisel, J. M., Elsig, M., & Bonnesen, M. (2011). Delayed grammatical acqui-
sition in first language development: Subject-verb inversion and subject
clitics in French interrogatives. Linguistic Approaches to Bilingualism,
1(4), 347–390. doi:http://doi.org/10.1075/lab.1.4.01mei
Mirvahedi, S. H., & Cavallaro, F. (2020). Siblings’ play and language shift to
English in a Malay-English bilingual family. World Englishes, 39(1),
183–197. doi:http://doi.org/10.1111/weng.12417
Mishina, S. (1999). The role of parental input and discourse strategies in the
early language mixing of a bilingual child. Multilingua, 18(4), 343–367. doi:
http://doi.org/10.1515/mult.1999.18.4.317
Mobaraki, M., Vainikka, A., & Young-Scholten, M. (2008). The status of
subjects in early child L2 English. In B. Haznedar & E. Gavruseva (Eds.),
Current trends in child second language acquisition: A generative
76 References
perspective. Amsterdam: John Benjamins, 209–236. doi:http://doi.org/10

.1075/lald.46.11mob
Montanari, S. (2010). Translation equivalents and the emergence of multiple
lexicons in early trilingual development. First Language, 30, 102–125. doi:
Müller, L.-M., Howard, K., Wilson, E., Gibson, J., & Katsos, N. (2020).
Bilingualism in the family and child well-being: A scoping review.
International Journal of Bilingualism, 24(5–6), 1049–1070. doi:http://doi
.org/10.1177/1367006920920939
Muñoz, C., & Spada, N. (2019). Foreign language learning from early child-
hood to young adulthood. In A. De Houwer & L. Ortega (Eds.), The
Cambridge handbook of bilingualism. Cambridge: Cambridge University
Press, 233–249. doi:http://doi.org/10.1017/9781316831922.013
Murphy, V. A. (2018). Literacy development in linguistically diverse pupils.
In D. Miller, F. Bayram, J. Rothman, & L. Serratrice (Eds.), Bilingual
cognition and language: The state of the science across its subfields.
Amsterdam: John Benjamins, 155–82. doi:http://doi.org/10.1075/sibil
.54.08mur
Nakamura, J. (2015). Nonnative maternal input : Language use and errors in
a Thai mother’s interactions in Japanese with her child. Japan Journal of
Multilingualism and Multiculturalism, 21, 10–26.
Nakamura, J. (2020). Language regrets: Mixed-ethnic children’s lost opportun-
ity for minority language acquisition in Japan. Multilingua, 39(2), 213–237.
doi:http://doi.org/10.1515/multi-2019-0040
Nakamura, J., & Quay, S. (2012). The impact of caregivers’ interrogative styles
in English and Japanese on early bilingual development. International
Journal of Bilingual Education and Bilingualism, 15(4), 417–434. doi:
http://doi.org/10.1080/13670050.2012.665827
Navarro, A. M., Pearson, B. Z., Cobo-Lewis, A., & Oller, K. D. (1998).
Identifying the language spoken by 26-month-old monolingual- and bilin-
gual-learning babies in a no-context situation. In A. Greenhill, M. Hughes,
H. Littlefield, & H. Walsh (Eds.), Proceedings of the 22nd Annual Boston
University Conference on Language Development, Vol. 2. Somerville, MA:
Cascadilla Press, 557–68.
Nicoladis, E. (1998). First clues to the existence of two input languages:
Pragmatic and lexical differentiation in a bilingual child. Bilingualism:
Language and Cognition, 1(2), 105–116. doi:http://doi.org/10.1017
/S1366728998000236
Oller, D. K., & Eilers, R. E. (Eds.) (2002). Language and literacy in bilingual
children. Clevedon: Multilingual Matters.
References 77
Oppenheim, G. M., Griffin, Z. M., Peña, E. D., & Bedore, L. M. (2020).

Longitudinal evidence for simultaneous bilingual language development
with shifting language dominance, and how to explain it. Language
Learning, 70, 20–44. doi:http://doi.org/10.1111/lang.12398
Orena, A., Byers-Heinlein, K., & Polka, L. (2020). What do bilingual infants
actually hear? Evaluating measures of language input to bilingual-learning 10-
month-olds. Developmental Science 23(2), e12901. doi:http://doi.org/10.31234
/osf.io/2qnhw
Orena, A. J., & Polka, L. (2019). Monolingual and bilingual infants” word
segmentation abilities in an inter-mixed dual-language task. Infancy, 24(5),
718–737. doi:http://doi.org/10.1111/infa.12296
Pace, A., Luo, R., Hirsh-Pasek, K., & Golinkoff, R.M. (2017). Identifying
pathways between socio-economic status and language development.
Annual Review of Linguistics, 3, 285–308. doi:http://doi.org/10.1146
/annurev-linguistics-011516-034226
Palacios, N., Kibler, A. K., Simpson Baird, A., Parr, A., & Bergey, R.
(2015). An examination of language practices during mother-child play
activities among Mexican immigrant families. International
Multilingual Research Journal, 9(3), 197–219. doi:http://doi.org/10
.1080/19313152.2015.1048543
Paradis, J. (2011). Individual differences in child English second language
acquisition: Comparing child-internal and child-external factors. Linguistic
Approaches to Bilingualism, 1(3), 213–237. doi:http://doi.org/10.1075/lab.1
.3.01par
Paradis, J., & Navarro, S. (2003). Subject realization and crosslinguistic inter-
ference in the bilingual acquisition of Spanish and English: What is the role
of the input? Journal of Child Language, 30, 371–393. doi:http://doi.org/10
.1017/S0305000903005609
Paradis, J., & Nicoladis, E. (2007). The influence of dominance and sociolin-
guistic context on bilingual preschoolers” language choice. The International
Journal of Bilingual Education and Bilingualism, 10(3), 277–297. doi:http://
doi.org/10.2167/beb444.0
Park, H., Tsai, K. M., Liu, L. L., & Lau, A. S. (2012). Transactional associations
between supportive family climate and young children’s heritage language
proficiency in immigrant families. International Journal of Behavioral
Development, 36(3), 226–236. doi:http://doi.org/10.1177/0165025412439842
Parra, M., Hoff, E., & Core, C. (2011). Relations among language exposure,
phonological memory, and language development in Spanish-English bilin-
gually developing two year olds. Journal of Experimental Child Psychology,
108, 113–125. doi:http://doi.org/10.1016/j.jecp.2010.07.011
78 References
Patterson, J. L. (1998). Expressive vocabulary development and word combin-

ations of Spanish–English bilingual toddlers. American Journal of Speech–
Language Pathology, 7, 46–56. doi:http://doi.org/10.1044/1058-0360
.0704.46
Patterson, J. L. (2002). Relationships of expressive vocabulary to fre-
quency of reading and television experience among bilingual toddlers.
Applied Psycholinguistics, 23(4), 493–508. doi:http://doi.org/10.1017/
S0142716402004010
Patterson, J. L, & Rodríguez, B. L. (Eds.) (2016). Multilingual perspectives on
child language disorders. Bristol: Multilingual Matters. doi:http://doi.org/10
.21832/9781783094738
Pavlovitch, M. (1920). Le langage enfantin. Acquisition du serbe et du français.
Paris: Champion.
Pearson, B. Z., & Fernández, S. C. (1994). Patterns of interaction in the lexical
growth in two languages of bilingual infants and toddlers. Language
Learning, 44, 617–653.
Pearson, B. Z., Fernández, S., Lewedeg, V., & Oller, D. K. (1997). The relation of
input factors to lexical learning by bilingual infants. Applied Psycholinguistics,
Pearson, B. Z., Fernández, S., & Oller, D. K. (1993). Lexical development in
simultaneous bilingual infants: comparison to monolinguals. Language
Learning, 43, 93–120.
Pearson, B. Z., Fernández, S., & Oller, D. K. (1995). Cross-language synonyms
in the lexicons of bilingual infants: One language or two? Journal of Child
Language, 22, 345–368. doi:http://doi.org/10.1017/S030500090000982X
Pearson, P. D. (2019). Time, complexity, and the enduring importance of words:
Key themes in language learning in the middle years. In V. Grøver, P. Ucelli,
M. L. Rowe, & E. Lieven (Eds.), Learning through language. Towards an
educationally informed theory of language learning. Cambridge: Cambridge
University Press, 159–169. doi:http://doi.org/10.1017/9781316718537.015
Pease-Alvarez, L., & Winsler, A. (1994). Cuando el maestro no habla Español:
Children’s bilingual languages practices in the classroom. TESOL Quarterly,
28, 507–535.
Persici, V., Vihman, M. M., Burroa, R., & Majoranoa, M. (2019). Lexical access
and competition in bilingual children: The role of proficiency and the lexical
similarity of the two languages. Journal of Experimental Child Psychology,
179, 103–125. doi:http://doi.org/10.1016/j.jecp.2018.10.002
Pfaff, C. (1994). Early bilingual development of Turkish children in Berlin. In
G. Extra & L. Verhoeven (Eds.), The cross-linguistic study of bilingualism.
Amsterdam: Royal Netherlands Academy of Arts and Sciences, 75–97.
References 79
Pienemann, M. (1981). Der Zweitspracherwerb ausländischer Arbeiterkinder.

Bonn: Bouvier.
Place, S., & Hoff, E. (2011). Properties of dual language exposure that influence
two-year-olds’ bilingual proficiency. Child Development, 82 (6), 1834–1849.
doi:http://doi.org/10.1111/j.1467-8624.2011.01660.x
Place, S., & Hoff, E. (2016). Effects and noneffects of input in bilingual
environments on dual language skills in 2½-year-olds. Bilingualism:
Language and Cognition, 19, 1023–1041. doi:http://doi.org/10.1017
/s1366728915000322
Polka, L., Orena, A. J., Sundara, M., & Worrall, J. (2017). Segmenting words
from fluent speech during infancy – challenges and opportunities in
a bilingual context. Developmental Science, 20, e12419. doi:http://doi.org/10
.1111/desc.12419
Porsché, D. (1983). Die Zweisprachigkeit während des primären
Spracherwerbs. Tübingen: Gunter Narr.
Prevoo, M. J. L., Malda, M., Mesman, J., Emmen, R.A.G., Yeniad, N.,
van IJzendoorn, M. H., & Linting, M. (2014). Predicting ethnic minor-
ity children’s vocabulary from socioeconomic status, maternal lan-
guage and home reading input: Different pathways for host and
ethnic language. Journal of Child Language, 41(5), 963–984. doi:
Prevoo, M., Mesman, J., van IJzendoorn,M., & Pieper, S. (2011). Bilingual
toddlers reap the language they sow: Ethnic minority toddlers’ childcare
attendance increases maternal host language use. Journal of Multilingual
and Multicultural Development, 32, 561–576. doi:http://doi.org/10.1080
/01434632.2011.609279
Poulin-Dubois, D., Bialystok, E., Blaye, A., Polonia, A., & Yott, J. (2013).
Lexical access and vocabulary development in very young bilinguals.
International Journal of Bilingualism, 17(1), 57–70. doi:http://doi.org/10
.1177/1367006911431198
Ramírez-Esparza, N., García-Sierra, A., & Kuhl, P. K. (2017). The impact of
early social interactions on later language development in Spanish–English
bilingual infants. Child Development, 88(4), 1216–1234. doi:http://doi.org
/10.1111/cdev.12648
Reich, H. H. (2009). Zweisprachige Kinder. Sprachenaneignung und spra-
chliche Fortschritte im Kindergartenalter. Münster: Waxmann.
Ribot, K., & Hoff, E. (2014). “¿Cómo estas?” “I’m good.” Conversational code-
switching is related to profiles of expressive and receptive proficiency in
Spanish-English bilingual toddlers. International Journal of Behavioral
Development, 38(3), 333–341. doi:http://doi.org/10.1177/0165025414533225
80 References
Ribot, K. M., Hoff, E., & Burridge, A. (2018). Language use contributes to
expressive language growth: evidence from bilingual children. Child
Development, 89(3), 929–940. doi:http://doi.org/10.1111/cdev.12770
Rinker, T., Budde-Spengler, N., & Sachse, S. (2017). The relationship between
first language (L1) and second language (L2) lexical development in young
Turkish-German children. International Journal of Bilingualism and
Bilingual Education, 20, 218–233. doi:http://doi.org/10.1080/13670050
.2016.1179260
Roesch, A., & Condrogianni, V. (2016). “Which mouse kissed the frog?”
Effects of age of onset, length of exposure, and knowledge of case marking
on the comprehension of wh-questions in German-speaking simultaneous
and early sequential bilingual children. Journal of Child Language, 43(3),
Ronjat, J. (1913). Le développement du langage observé chez un enfant bilin-
gue. Paris: Champion.
Rothweiler, M. (2016). Zum Erwerb der deutschen Grammatik bei früh
sequentiell zweisprachigen Kindern mit Türkisch als Erstsprache –
Ergebnisse aus einem Forschungsprojekt. Diskurs Kindheits- und
Jugendforschung, 1, 9-25.
Scheele, A., Leseman, P., & Mayo, A. (2010). The home language environment
of monolingual and bilingual children and their language proficiency.
Applied Psycholinguistics, 31, 117–140. doi:http://doi.org/10.1017
/S0142716409990191
Schulz, P. (2013). Wer versteht wann was? Sprachverstehen im frühen
Zweitspracherwerb des Deutschen am Beispiel der w-Fragen. In
A. Deppermann (Ed.), Das Deutsch der Migranten. Berlin: de Gruyter,
313–337. doi:http://doi.org/10.1515/9783110307894.313
Schulz, P., & Tracy, R. (2011). LiSe-DaZ. Linguistische Sprachstandserhebung –
Deutsch als Zweitsprache. (In Verbindung mit der Baden-Württemberg
Stiftung). Göttingen: Hogrefe.
Schwartz, M. (2020). Strategies and practices of home language maintenance.
In S. Eisenchlas & A. Schalley (Eds.), Handbook of home language mainten-
ance and development. Social and affective factors. Berlin: Mouton de
Gruyter, 194–217. doi:http://doi.org/10.1515/9781501510175-010
Schwartz, M., Deeb, I., & Hijazy, S. (2020). “How do you say it in Arabic, in
Hebrew, in English?” Towards a better understanding of children’s agentic
behaviour in novel language learning. Journal of Multilingual and
Multicultural Development. doi:http://doi.org/10.1080/01434632.2020.1711768
Schwartz, M., Minkov, M., Dieser, E., Protassova, E., Moin, V., & Polinsky, M.
(2015). Acquisition of Russian gender agreement by monolingual and
References 81
bilingual children. International Journal of Bilingualism, 19, 726–752. doi:

Serratrice, L. (2007). Cross-linguistic influence in the interpretation of ana-
phoric and cataphoric pronouns in English–Italian bilingual children.
Bilingualism: Language and Cognition, 10, 225–238. doi:http://doi.org/10
.1017/S1366728907003045
Serratrice, L. (2019). Becoming bilingual in early childhood. In A. De Houwer
& L. Ortega (Eds.), The Cambridge handbook of bilingualism. Cambridge:
Cambridge University Press, 15–35. doi:http://doi.org/10.1017
/9781316831922.002
Sierens, S., Slembrouck, S., Van Gorp, K., & Ağırdağ, O. (2019). Linguistic
interdependence of receptive vocabulary skills in emergent bilingual pre-
school children: Exploring a factor-dependent approach. Applied
Psycholinguistics, 50(5), 1269–1297. doi:http://doi.org/10.1017
/S0142716419000250
Silvén, M., Voeten, M., Kouvo, A., & Lundén, M. (2014). Speech perception
and vocabulary growth: A longitudinal study of Finnish-Russian bilinguals
and Finnish monolinguals from infancy to three years. International Journal
of Behavioral Development, 38(4), 323–332. doi:http://doi.org/10.1177
/0165025414533748
Singh, L., Fu, C.S.L., Tay, Z., & Golinkoff, R. M. (2017). Novel word learning in
bilingual and monolingual infants: Evidence for a bilingual advantage. Child
Development, 89(3), e183–e198. doi:http://doi.org/10.1111/cdev.12747
Singh, L., Poh, F. L. S., & Fu, C. S. L. (2016). Limits on monolingualism?
A comparison of monolingual and bilingual infants’ abilities to integrate
lexical tone in novel word learning. Frontiers in Psychology, 7, 667. doi:
http://doi.org/10.3389/fpsyg.2016.00667
Sinka, I., & Schelletter, C. (1998). Morphosyntactic development in bilingual
children. International Journal of Bilingualism, 2(3), 301–326. doi:http://doi
.org/10.1177/136700699800200303
Slavkov, N. (2015). Language attrition and reactivation in the context of bilin-
gual first language acquisition. International Journal of Bilingual Education
and Bilingualism, 18, 715–734. doi:http://doi.org/10.1080/13670050
.2014.941785
Smithson, L., Paradis, J., & Nicoladis, E. (2014). Bilingualism and receptive
vocabulary achievement: Could sociocultural context make a difference?
Bilingualism: Language and Cognition, 17(4), 810–821. doi:http://doi.org
/10.1017/S1366728913000813
Song, L., Tamis-LeMonda, C. S., Yoshikawa, H., Kahana-Kalman, R., & Wu, I.
(2012). Language experiences and vocabulary development in Dominican
82 References
and Mexican infants across the first 2 years. Developmental Psychology, 48,
Steinberg, L., Bornstein, M. H., Vandell, D. L., & Rook, K. (2011). Life-span
development: Infancy through adulthood. Boston, MA: Wadsworth Cengage
Learning.
Sun, H., Ng, S. C., O’Brien, B. A., & Fritzsche, T. (2020). Child, family, and
school factors in bilingual preschoolers’ vocabulary development in heritage
languages. Journal of Child Language, 47(4), 817–843. doi:http://doi.org/10
.1017/S0305000919000904
Sun, H., Yussof, N., Mohamed, M., Rahim, A., Cheung, W. L., Cheong, S. A., &
Bull, R. (2021). Bilingual language experience and social-emotional well-
being: A cross-sectional study of Singapore preschoolers. International
Journal of Bilingual Education and Bilingualism, 24(3), 324–339. doi:
http://doi.org/10.1080/13670050.2018.1461802
Sundara, M., Polka, L., & Genesee, F. (2006). Language-experience facilitates
discrimination of /d-δ/ in monolingual and bilingual acquisition of English.
Cognition, 100(2), 369–388. doi:http://doi.org/10.1016/j.cognition.2005.04.007
Sundara, M., & Scutellaro, A. (2011). Rhythmic distance between languages
affects the development of speech perception in bilingual infants. Journal of
Phonetics, 39(4), 505–513. doi:http://doi.org/10.1016/j.wocn.2010.08.006
Sundara, M., Ward, N., Conboy, B., & Kuhl, P. K. (2020). Exposure to a second
language in infancy alters speech production. Bilingualism: Language and
Cognition, 23(5), 978–991. doi:http://doi.org/10.1017/S1366728919000853
Tang, G., & Sze, F. (2019). Bilingualism and sign language research. In A. De
Houwer & L. Ortega (Eds.), The Cambridge handbook of bilingualism.
.1017/9781316831922.025
Tare, M., & Gelman, S. A. (2010). Can you say it another way? Cognitive
factors in bilingual children’s pragmatic language skills. Journal of
Cognition and Development, 11, 137–158.
Thompson, L. (2000). Young bilingual children in nursery school. Clevedon:
Multilingual Matters.
Uccelli, P., & Páez, M. (2007). Narrative and vocabulary development of
bilingual children from kindergarten to first grade: Developmental changes
and associations among English and Spanish skills. Language, Speech, and
Hearing Services in Schools, 38, 225–236. doi:http://doi.org/10.1044/0161-
1461(2007/024)
Uchikoshi, Y. (2006). English vocabulary development in bilingual kindergar-
teners: What are the best predictors? Bilingualism: Language and Cognition,
References 83
UNESCO. (2016). If you don’t understand, how can you learn? Global
Education Monitoring Report. Policy Paper 24. bit.ly/MLD2016
Unsworth, S. (2013). Assessing age of onset effects in (early) child L2
acquisition. Language Acquisition, 20, 74–92. doi:http://doi.org/10.1080
/10489223.2013.766739
Unsworth, S., Argyri, F., Cornips, L., Hulk, A., Sorace, A., & Tsimpli, I. (2014).
The role of age of onset and input in early child bilingualism in Greek and
Dutch. Applied Psycholinguistics, 35, 765–805. doi:http://doi.org/10.1017
/S0142716412000574
Unsworth, S., Brouwer, S., de Bree, E. & Verhagen, J. (2019). Predicting
bilingual preschoolers’ patterns of language development: Degree of non-
native input matters. Applied Psycholinguistics, 40(5), 1189–1219. doi:http://
doi.org/10.1017/S0142716419000225
Vaahtoranta, E., Suggate, S., Lenhart, J., & Lenhard, W. (2020). Language
exposure and phonological short-term memory as predictors of majority
language vocabulary and phonological awareness in dual language learning.
Bilingualism: Language and Cognition, 24(2), 319–332. doi:http://doi.org/
10.1017/S1366728920000541
van de Weijer, J. (1997). Language input to a prelingual infant. In A. Sorace,
C. Heycock & R. Shillcock (Eds.), Proceedings of the GALA ‘97
Conference on Language Acquisition. Edinburgh: University of
Edinburgh, 290–293.
van de Weijer, J. (1998). Language input for word discovery. Max Planck Series
in Psycholinguistics 9. Wageningen: Ponsen & Looijen.
van de Weijer, J. (2000). Language input and word discovery. In M. Beers,
B. van den Bogaerde, G. Bol, J. de Jong, & C. Rooijmans (Eds.), From sound
to sentence: Studies on first language acquisition. Groningen: Centre for
Language and Cognition, 155–162.
van de Weijer, J. (2002). How much does an infant hear in a day? In J. Costa &
M. João Freitas (Eds.), Proceedings of the GALA2001 Conference on
Language Acquisition. Lisbon, Portugal: Associação Portuguesa de
Linguistíca, 279–282.
Verdon, S., McLeod, S., & Winsler, A. (2014). Language maintenance
and loss in a population study of young Australian children. Early
Childhood Research Quarterly, 29(2), 168–181. doi:http://doi.org/10
.1016/j.ecresq.2013.12.003
Vihman, M. M. (1999). The transition to grammar in a bilingual child:
Positional patterns, model learning, and relational words. International
Journal of Bilingualism, 3(2–3), 267–301. doi:http://doi.org/10.1177
/13670069990030020801
84 References
Vihman, M. (2016). Prosodic structures and templates in bilingual phonological

development. Bilingualism: Language and Cognition, 19, 69–88. doi:http://
doi.org/10.1017/S1366728914000790
Vihman, M. M., Thierry, G., Lum, J., Keren-Portnoy, T., & Martin, P. (2007).
Onset of word form recognition in English, Welsh, and English-Welsh bilin-
gual infants. Applied Psycholinguistics, 28, 475–493. doi:http://doi.org/10
.1017/S0142716407070269
Vila, I. (1984). Yo siempre hablo catalan y castellano: datos de una investiga-
cion en curso sobre la adquisicion del lenguaje en niños bilingües familiares.
In M. Siguan (Ed.), Adquisición precoz de una segunda lengua. Barcelona:
Publicacions i edicions de la Universitat de Barcelona, 31–51.
von Grünigen, R., Kochenderfer-Ladd, B., Perren, S., & Alsaker, F. (2012).
Links between local language competence and peer relations among Swiss
and immigrant children: The mediating role of social behaviour. Journal of
School Psychology, 50(2), 195–213. doi:http://doi.org/10.1016/j
.jsp.2011.09.005
Wagner-Gough, J. (1975). Comparative studies in second language learning.
CAL-ERIC/CCL Series on Language and Linguistics, 26. Washington, DC:
Center for Applied Linguistics.
Weikum, W. M., Vouloumanos, A., Navarra, J., Soto-Faraco, S.,
Sebastián-Gallés, N., & Werker, J. F. (2007). Visual language discrimination
in infancy. Science, 316(5828), 1159. doi:http://doi.org/10.1126/science
.1137686
Willard, J. A., Kohl, K., Bihler, L.-M., Agache, A., & Leyendecker, B. (2020).
Family literacy activities and their interplay with family and preschool
language environments: Links to gains in dual language learners’ German
vocabulary. Early Education and Development, 32(2), 189–208. doi:http://
doi.org/10.1080/10409289.2020.1744404
Winsler, A., Burchinal, M., Tien, H.-C., Peisner-Feinberg, E., Espinosa, L.,
Castro, D., LaForett, D., Kim, Y., & De Feyter, J. (2014a). Early develop-
mental skills of diverse dual language learners: The roles of home language
use, cultural heritage, maternal immigration, and socio-demographic
variables. Early Childhood Research Quarterly, 29, 750–764. doi:http://dx
.doi.org/10.1016/j.ecresq.2014.02.008
Winsler, A., Kim, Y., & Richard, E. (2014b). Socio-emotional skills, behavior
problems, and Spanish competence predict the acquisition of English among
English language learners in poverty. Developmental Psychology, 50,
Wölck, W. (1987/1988). Types of natural bilingual behavior: A review and
revision. The Bilingual Review/La Revista bilingüe, 14, 3–16.
References 85
Yamamoto, M. (2001). Language use in interlingual families: A Japanese–

English sociolinguistic study. Clevedon: Multilingual Matters.
Yow, W. Q., & Markman, E. M. (2011). Young bilingual children’s heightened
sensitivity to referential cues. Journal of Cognitive Development, 12, 12–31.
doi:http://doi.org/10.1080/15248372.2011.539524
Yu, S. (2014). How much can migrant parents influence their children’s lan-
guage choice. Taiwan Journal of Linguistics, 12(1), 81–105.
Zdorenko, T., & Paradis, J. (2007). The role of the first language in child second
language acquisition of articles. In A. Belikova, L. Meroni, & M. Umeda
(Eds.), Galana 2: Proceedings of the Conference on Generative Approaches
to Language Acquisition North America 2. Somerville, MA: Cascadilla
Proceedings Project, 483-490.
Zentella, A. C. (1997). Growing up bilingual: Puerto Rican children in
New York. Oxford: Wiley-Blackwell.
Child Development
Marc H. Bornstein
National Institute of Child Health and Human Development, Bethesda
Institute for Fiscal Studies, London
UNICEF, New York City
Marc H. Bornstein is an Affiliate of the Eunice Kennedy Shriver National Institute of Child
Health and Human Development, an International Research Fellow at the Institute for Fiscal
Studies (London), and UNICEF Senior Advisor for Research for ECD Parenting Programmes.
Bornstein is President Emeritus of the Society for Research in Child Development,
Editor Emeritus of Child Development, and founding Editor of Parenting: Science and
Practice.
About the Series

Child development is a lively and engaging, yet serious and purposeful subject of
academic study that encompasses a myriad of theories, methods, substantive areas,
and applied concerns. Cambridge Elements in Child Development proposes to address
all these key areas, with unique, comprehensive, and state-of-the-art treatments,
introducing readers to the primary currents of research and to original perspectives
on, or contributions to, principal issues and domains in the field
Child Development
Elements in the Series

Child Development in Evolutionary Perspective
David F. Bjorklund
Gender in Childhood
Christia Spears Brown, Sharla D. Biefeld and Michelle J. Tam
The Child’s Environment
Robert H. Bradley
The Nature of Intelligence and Its Development in Childhood
Robert J. Sternberg
Child Welfare, Protection, and Justice
Murli Desai
Annick De Houwer
A full series listing is available at: www.cambridge.org/EICD

Bilingual Development in Childhood

Uploaded by

Copyright:

Available Formats

Bilingual Development in Childhood

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Bilingual Development in Childhood

Uploaded by

Copyright:

Available Formats

De Houwer

In the first decade of life, children become bilingual in different

Bilingual Development in Childhood

Cover image: Shutterstock / Creativika Graphics ISSN

Cambridge University Press is part of the University of Cambridge.

Elements in Child Development

Abstract: In the ﬁrst decade of life, children become bilingual in diﬀerent

Keywords: bilingual, children, learning, language, acquisition

© Annick De Houwer 2021

2 Becoming Bilingual in Infancy: Focus on Bilingual First

3 Bilingualism in Early Childhood: Bilingual First and Early

4 Bilingualism in Middle Childhood: Bilingual First

5 Socioeconomic Status and Bilingual Development in

6 Summary and Conclusion 54

Infancy Early childhood Middle childhood

Figure 1 Three kinds of bilingual learning settings in three life stages

Many people are bilingual because of a migration background. Bilingualism

The term “bilingual children” refers to typically developing, normally hear-

American English) or varieties of different languages (e.g., Mexican Spanish

2 Becoming Bilingual in Infancy: Focus on Bilingual First

2.1 Born into a Bilingual Family

gesture during parent–child interaction with their 14-month-olds (Gampe et al.,

2.2 Phonological Development: Bilingual Infant Speech Perception

2.3 Bilingual Infant Language Comprehension

2.4 Bilingual Infant Language Production

2.4.2 First Words

2.4.3 Phonological Development: Learning to Produce Sounds

2.4.4 Lexical Repertoire and Vocabulary Size

between 12 and 24 months far exceeded that of monolinguals in either language

2.4.5 Word Combinations

2.4.6 Language Choice

2.5 Uneven Bilingual Development in Infancy

again (summary in De Houwer, 2009, p. 196). For some BFLA 12-month-

2.6 Factors Aﬀecting Bilingual Development in Infancy

However, research comparing observed parental engagement with infants in

3 Bilingualism in Early Childhood: Bilingual First and Early

3.1 BFLA Preschoolers’ Language Development

BFLA toddlers and preschoolers continue to use phonological processes

translation in Language Alpha. Many are able to provide a TE on request (De

the totality of devices that structure the combination of words in an utterance so

3.2 ESLA Preschoolers’ Language Development

birth, it is not surprising that bilingual–monolingual comparisons of L2-English

3.3 BFLA and ESLA Preschoolers Compared

in the Soc-L. BFLA children’s Soc-L sentences show no systematic inﬂu-

3.4 Language Choice in Early Childhood

using different patterns of language choice depending on which language

3.5 Factors Aﬀecting Bilingual Development in Early Childhood

outlined in Section 3.5.2, continued rich exposure to both languages is import-

3.5.1 Child-Internal Factors Aﬀecting Bilingual Development in Early

3.5.2 Child-External Factors Aﬀecting Bilingual Development in Early

4 Bilingualism in Middle Childhood: Bilingual First Language

develop the new L2 through one-on-one supportive interaction with preschool

4.1 BFLA and ESLA Schoolchildren’s Language Development

Spanish and English (Oppenheim et al., 2020). In both languages, children

4.2 SLA Schoolchildren’s Language Development

ﬁndings on their longitudinal studies involving millions (!) of children, from

4.3 Uneven Bilingual Development in Middle Childhood

4.4 Language Choice in Middle Childhood

4.5 Factors Aﬀecting Bilingual Development in Middle Childhood

and ESLA schoolchildren’s changing patterns of performance over the years,