LINGUIST List 27.676

Thu Feb 04 2016

Review: Applied Ling; Lang Acq; Socioling: Paradis, Grüter (2014)

Editor for this issue: Sara Couture <saralinguistlist.org>


Date: 21-Sep-2015
From: Kate Riestenberg <katejriesgmail.com>
Subject: Input and Experience in Bilingual Development
E-mail this message to a friend

Discuss this message


Book announced at http://linguistlist.org/issues/25/25-4439.html

EDITOR: Theres Grüter
EDITOR: Johannes Paradis
TITLE: Input and Experience in Bilingual Development
SERIES TITLE: Trends in Language Acquisition Research 13
PUBLISHER: John Benjamins
YEAR: 2014

REVIEWER: Kate Riestenberg, Georgetown University

Reviews Editor: Helen Aristar-Dry

SUMMARY

Input and Experience in Bilingual Development brings together nine chapters from a variety of contributors in a volume edited by Theres Grüter and Johanne Paradis. The book takes up issues first presented in symposia organized by Paradis in 2008 and Grüter in 2011 at the congresses of the International Association for the Study of Child Language (IASCL). Some chapters report on a single experimental study, while others take the form of a review, summarizing a series of studies investigating a common research question. The overarching goal of the book, as stated in the introduction by Paradis and Grüter, is to show that while input is clearly a driving factor in language development, the relationship between the two is not necessarily straightforward or simple.

The research presented in the volume focuses on young children who have been regularly exposed to more than one language by about age three. A variety of global contexts and languages are represented, including more commonly studied language pairs such as Spanish-English (Grüter, Hurtado, Marchman & Fernald; Hoff, Place, Welsh, & Ribot), French-English (Thordardottir; Paradis, Tremblay, & Crago), English-Dutch (Unsworth), and Dutch-French (De Houwer), but also less commonly researched language pairs such as Russian-Hebrew and English-Hebrew (Armon-Lotem, Joffe, Abutbul-Oz, Altman, & Walters) and Wapichana-Spanish (Pearson & Amaral). While these chapters focus for the most part on contexts in which children experience simultaneous exposure to two languages, a chapter by Pierce and Genesee investigates parent-child interaction among Canadian children adopted from China, representing a special case of sequential bilingualism.

While the book does not commit to a certain theory of language development, most of the chapters in the volume approach the theme from a usage-based or interactional perspective (with the notable exception of the chapter by Unsworth, who argues for at least some innate architecture to explain her findings). The chapters by Grüter et al., De Houwer, and Pierce and Genesee are concerned with measuring and characterizing the quantity and quality of caretaker input to children exposed to more than one language. The chapters by Armon-Lotem et al. and Pearson & Amaral emphasize how factors in the larger sociocultural context of a speech community affect child bilingualism. Hoff et al., Thordardottir, Paradis et al., and Unsworth all address shortcomings related to how language outcomes have typically been measured in studies investigating the role of input in bilingual development.

Overall, the chapters demonstrate the complex task that lies ahead for researchers interested in the multifaceted roles of input and experience in language development. As Paradis and Grüter note in their introduction to the volume, “Bilingual development is both sensitive to and resilient against variation in input and experience” (p. 11). The chapters in this book demonstrate this notion quite well by presenting a variety of ways to investigate and analyze input, experience, and development.

“Defining and measuring input quantity”

Chapter 1. Language exposure and online processing efficiency in bilingual development

In this chapter, Grüter et al. discuss findings from a series of studies of Spanish-English bilingual children in California. They demonstrate clear relationships between amount of language exposure and children’s vocabulary size and processing efficiency, but they show that these relationships differ according to whether exposure and outcomes are measured in relative or absolute terms. While many studies of child bilingualism represent children’s input as relative proportions as reported by parents (e.g., 40% Spanish and 60% English), Grüter et al. use naturalistic recordings to establish an observed relative proportion as well as an absolute measure of frequency (“absolute word count”). They demonstrate that at least some parental reports do not provide an accurate picture of children’s actual language exposure. They also show that relative measures of input are good predictors of relative processing efficiency and vocabulary size, but poor predictors of absolute processing efficiency and vocabulary size.

Chapter 2. The absolute frequency of maternal input to bilingual and monolingual children.

De Houwer’s chapter also demonstrates the shortcomings of relying solely on relative measures of language exposure. Her central data are sets of short transcripts of mother-child interaction at ages 13 months and 20 months, divided into monolingual Dutch and bilingual Dutch-French groups. She employs a variety of absolute measures of frequency, speech rate, and density of parental input and finds no differences between a monolingual group and bilingual group. However, she finds a wide range of variation within both groups and shows that some of the children in bilingual Dutch-French environments heard more Dutch from their mothers than some of the children in Dutch-only environments.

“Experiential factors beyond input quantity”

Chapter 3. Language input and language learning: An interactional perspective

Pierce and Genesee report on a set of studies investigating joint attention episodes between Canadian parents and their internationally adopted (IA) children. They show that parents of IA children tend to use interactional strategies similar to those employed by parents of children with cognitive delays related to language development. They also demonstrate that while birth fathers spent less time engaged in joint attention with the child than birth mothers, there was no such difference between fathers and mothers of IA children.

Chapter 4. Language exposure, ethnolinguistic identity, and attitudes in the acquisition of Hebrew as a second language among bilingual preschool children from Russian- and English-speaking backgrounds

Armon-Lotem et al. examine the effects of distal factors (ethnolinguistic identity, attitudes towards speakers, parents’ education level, family size, and birth order) on language development among L1 Russian-L2 Hebrew and L1 English-L2 Hebrew preschoolers in Israel. They show that identity and attitude factors are less variable for the English group, who they argue have a religious motivation for immigrating, than the Russian group, who they argue have an economic motivation for immigrating. As such, identity and attitude factors only showed a significant effect on Hebrew proficiency for the Russian group. The authors also examine the effects of more proximal factors such as chronological age, age of onset, and length of exposure, and find these to have a greater effect on Hebrew proficiency than the distal factors measured.

Chapter 5. Interactions between input factors in bilingual language acquisition

Pearson and Amaral report on the language vitality of Wapichana, an Amazonian language spoken in parts of Brazil and Guyana, using UNESCO’s (2011) guidelines. They first argue that the UNESCO guidelines emphasize quantity of input, and consider “how much” Wapichana input is necessary to maintain bilingualism among Wapichana children. They then address several questions regarding input that are less well represented in the UNESCO guidelines, namely, from whom children receive language input, in what social domains, and at what age of onset. They go on to explore the complexities involved in assessing Wapichana vitality, comparing and contrasting the situations in Brazil and Guyana.

“Comparing bilingual and monolingual rates of development across linguistic domains”

Chapter 6. Properties of dual language input that shape bilingual development

Hoff et al. summarize a set of studies comparing Spanish-English bilinguals with English-speaking monolinguals matched for socioeconomic status (SES). They find that the difference in the amount of input that bilinguals and monolinguals receive in English is a good predictor of their English development, and that variation in the amount and type of input also explains within-group differences among the bilinguals. These findings differ from previous studies of child bilingualism that have reported no difference between the language abilities of monolinguals and bilinguals because the latter fall within a given “normal range” established for monolinguals.

Chapter 7. The typical development of simultaneous bilinguals

Thordardottir reports on findings from a set of studies comparing preschool children in three groups: monolingual English, monolingual French, and bilingual English-French. She demonstrates clear language-specific patterns in accuracy, errors, and acquisition order. Her results show a strong relationship between amount of input and measures of vocabulary development, grammatical productivity, and sentence imitation, but no relationship, in most cases, between amount of input and performance on a nonword repetition task. She further shows that nonword repetition accurately identified children with language impairment regardless of bilingualism.

Chapter 8. French-English bilingual children’s sensitivity to child-level and language-level input factors in morphosyntactic acquisition

Paradis et al. review a set of studies investigating the acquisition of French direct object clitics among French-English bilinguals and French-speaking monolinguals. They show that 6-year-old bilinguals with balanced exposure performed similarly to monolinguals on a direct object elicitation task, but that bilinguals who spoke mostly English at home performed worse. However, among 11-year-olds, bilingual groups performed similarly to French monolinguals regardless of amount of exposure.

Chapter 9. Comparing the role of input in bilingual acquisition across domains

In the final chapter of the volume, Unsworth examines the effect of exposure on Dutch-English bilinguals’ development of two distinct Dutch linguistic features: a morphosyntactic/lexical feature (gender marking on definite determiners) and a compositional semantic feature (scrambling). She shows that there is a strong relationship between amount of regular Dutch exposure and grammaticality judgments of gender marking among bilinguals from ages 5 to 17, but that there is no such relationship for truth value judgments involving scrambling. Instead, all speakers perform at ceiling on the scrambling task by about age seven.

EVALUATION

Overall, the chapters of Input and Experience in Bilingual Development cohere and complement each other quite nicely. Although the chapters do not all follow the same template, they each contribute different insights relevant to the theme. The major strength of the volume is its critical look at the measures traditionally used to study input and experience, on the one hand, and bilingual development on the other hand.

With regards to INPUT, the chapters by Grüter et al. and De Houwer show the importance of examining absolute measures of input in each of the child’s languages in addition to the more commonly employed relative measures, which view the input bilingual children receive as two proportions of a whole. In terms of EXPERIENCE, Pierce and Genesee break new ground by drawing attention to the potentially different discourse behaviors exhibited by caregivers in multilingual versus monolingual contexts.

As far as DEVELOPMENT is concerned, Hoff et al. show that while studies of child bilingualism have often reported no difference between the language abilities of monolinguals and bilinguals because the latter fall within a given “normal range” established for monolinguals, the variation in the amount and type of input that bilinguals receive is still a useful predictor of language outcomes. The chapters by Paradis et al. and Unsworth demonstrate the flaws of measuring language outcomes monolithically; they show differential effects of input on different linguistic features. In addition, Hoff et al. and Thordardottir both highlight the potentially differential effects of exposure on expressive versus receptive skills.

While certainly demonstrating a variety of methodological advances, the volume also inadvertently reveals several limitations still faced by this field of study. The first is the obvious difficulty involved in collecting representative amounts of actual child input data over time. Grüter et al.’s findings rely on just one 8 to 12-hour naturalistic recording, and De Houwer’s claims rest on a set of four 10-15 minute recordings. While certainly very interesting data, it may be prudent to keep in mind that these data are not necessarily representative of all of a child’s language exposure.

A second limitation is revealed through the emphasis in several chapters on so-called “distal” factors posited to affect bilingual development. For instance, Armon-Lotem et al. examine parents’ education level, family size, and birth order. Clearly these factors are likely to be associated in some way to bilingual language development, but because the authors treat the distal factors in the same way as proximal factors in the analysis--that is, by entering them in a regression model with L2 proficiency as the dependent variable--the acknowledged distance between the independent factor and the language outcome is washed out. A similar problem is seen in Pearson & Amaral’s chapter, which takes a more qualitative approach. They consider the relationship between UNESCO’s factors for determining language vitality, such as overall number of speakers and availability of language materials. The authors provide a thorough assessment of Wapichana vitality, but they fall short of showing a direct link between the UNESCO factors and children’s input. A more convincing approach is that taken by Hoff et al., who look at the effect of parents’ native languages and at the existence of an older sibling, first with input measures as a dependent variable and then with language outcome measures as a dependent variable. This allows the reader to see that intermediate, more proximal input-based factors link these distal factors to the language outcomes.

A final methodological issue revealed throughout the volume has to do with the measurement of frequency-based factors. Both De Houwer and Paradis et al. include measures of type and token frequency in their examination of children’s input, but they do so in slightly different ways. De Houwer appears to count up the number of different word tokens, or lexemes, as well as the different word types, or lemmas, in the mothers’ speech. In contrast, Paradis et al. conceptualize token frequency as a count of the number of times a lexeme (or morpheme, or construction) appears in the child’s input, and they conceptualize token frequency as the number of lemmas belonging to a given morphological class. A recent article by Ambridge, Kidd, Rowland, and Theakston (2015) addresses differences in how type and token frequency are conceptualized in language development research. They conclude that traditional notions of type and token frequency are not robust enough to account for the differential effects of frequency on a range of language outcomes. If input-driven accounts of language development are to retain their popularity, it will be necessary to elaborate and refine the methods used to measure input frequency.

In sum, this volume reveals both the strengths and limitations of current research on input and experience in bilingual development. The reader is left with the distinct impression that the interesting work presented in this volume has set the tone for a variety of new and exciting research in this area in the coming decades. This work is likely to take the form of continuing to address various methodological limitations. This may include, for instance, taking advantage of advances in storage and computing capabilities to examine input and development in larger data sets. This would allow for the emergence of more precise links between the large number of potentially measurable input-based factors (proximal and distal, child-level and language-level, internal and external, and so on) and the equally large number of possible linguistic outcomes.

It is perhaps because this body of work finds itself at such a juncture that the authors chose to remain relatively agnostic theory-wise. This book is most useful to scholars interested in current research practices and less useful to those interested in the theoretical implications of research on child bilingualism.

REFERENCES

Ambridge, B., Kidd, E., Rowland, C. F., & Theakston, A. L. (2015). The ubiquity of frequency effects in first language acquisition. Journal of Child Language, 42, 239–273. doi:10.1017/S030500091400049X

Grüter, T., & Paradis, J. (Eds.). (2014). Input and experience in bilingual development. Philadelphia: John Benjamins.

UNESCO Ad Hoc Expert Group on Endangered Languages. (2011). Language vitality and endangerment methodological guideline: Review of application and feedback since 2003. Paris: UNESCO. Retrieved from www.unesco.org.


ABOUT THE REVIEWER

Kate Riestenberg is a Ph.D. candidate at Georgetown University. Her research focuses on second language acquisition and bilingualism, phonology, and Zapotec languages. Her dissertation investigates the acquisition of lexical tone by child second language learners of Zapotec in a language revitalization program (NSF BCS-1451687). She recently completed a description of the morphophonology of Guadalupe Guevea Zapotec and a project investigating the use of task-based methods in teaching Zapotec to second language learners. She has co-authored papers on measuring awareness in implicit language learning and on cross-linguistic influence in third language acquisition. Upon completion of her dissertation, Kate hopes to continue investigating how input and experience differ for child (first language) versus later (second language) language learners.

Page Updated: 04-Feb-2016