Content uploaded by Naomi Lapidus Shin
Author content
All content in this area was uploaded by Naomi Lapidus Shin on Feb 17, 2016
Content may be subject to copyright.
314 Heritage Language Journal, 12(3)
December, 2015
BOOK REVIEW
Bilingual Language Acquisition: Spanish and English in the First Six Years. (2014). Carmen
Silva-Corvalán. Cambridge University Press. 408pp.
Reviewed by Naomi L. Shin, University of New Mexico
In this outstanding book on bilingual language acquisition, Carmen Silva-Corvalán provides a
detailed and meticulous analysis of her two grandsons’ Spanish and English language production
during the first six years of their lives. The volume contributes a wealth of information that
furthers our understanding of bilingual/heritage language acquisition and offers profound
insights that are relevant to theories of language acquisition, language contact and bilingualism.
In addition, the research has important pedagogical implications: The younger child stopped
speaking Spanish at around 4 and ½ years old, but started using it again when he was told he was
going to attend a Spanish-immersion school. The availability of such a school in California
ensured that both children became highly fluent speakers of two languages. Thus, although the
book is primarily a linguistic analysis of bilingual language development, it is also a success
story that will inspire caregivers who wish to raise their children bilingually. In this review I
provide very brief summaries of each chapter, followed by an overview of some major
theoretical contributions of the volume.
1. CHAPTER SUMMARIES
Chapter 1 is a succinct, yet thorough overview of theoretical issues related to bilingual language
acquisition. Silva-Corvalán provides clear explanations of both nativist and constructionist
theories of language acquisition, and then reviews the history of bilingual first language
acquisition (BFLA) research. Next she delves into her conceptualization of crosslinguistic
interaction, and follows with a discussion of how complexity and markedness play a role in
language acquisition. Finally, she discusses how language dominance, proficiency, and amount
of linguistic input determine the outcomes of bilingual language acquisition. This chapter can
stand alone as an excellent introduction to bilingual language acquisition and is thus well suited
for courses on bilingualism and language acquisition.
In Chapter 2, which provides details regarding methodology, the reader witnesses the
impressive scope of the study. Silva-Corvalán recorded her grandsons speaking in both English
and Spanish between the ages of 1;3 to 5;11, and was able to compile large datasets (~20,000
Spanish utterances containing a verb and ~8,000 English utterances containing a verb). The
chapter also illustrates how much careful thought went into measuring factors such as language
proficiency. Silva-Corvalán calculated the mean length of utterance in words (MLUw) values for
Spanish and English, and also quantified amount of exposure to both Spanish and English. She
was thus able to demonstrate that both boys’ dominant language was English, and that the older
sibling, Nico, had a higher level of Spanish proficiency as compared to the younger sibling,
Brennan. These language dominance and proficiency profiles set the stage for explaining the
boys’ linguistic patterns. It is a recurring theme throughout the book that Nico more closely
mirrors monolingual Spanish-speaking children than Brennan does (e.g., pp. 327, 332).
315 Heritage Language Journal, 12(3)
December, 2015
Chapter 3 is a concise, yet deep, overview of the boys’ production data. The chapter provides
evidence that the boys reached developmental milestones at the same time as or earlier than
monolingual children typically do. This is worth underscoring, as laypeople sometimes fear that
learning more than one language results in an overall delay in language development. Nico
produced his first word before his first birthday, and Brennan did so by age 1;1.0 (p. 55). With
respect to English syntax, there were no delays (p. 62-63, 118, 232, 286). In fact, the boys
acquired some structures earlier than monolingual English-speaking children do (pp. 62-63, 312,
316-317, 349). For example, the boys produced the past progressive in English before age 3;0.
Also, the scarcity of their overregularization of the English past tense morpheme (e.g. sinked in)
lies in stark contrast to monolingual English-speaking children, who produce such
overregularizations more often and well into school age (p. 313). Silva-Corvalán suggests that
“the simultaneous acquisition of a rich inflection language, Spanish, made the siblings more
sensitive to tense morphology” in English (pp. 316-317) (cf. Bedore & Leonard, 2000). With
respect to Spanish morphosyntactic structures, the chapter covers early complex sentences,
conditional clauses, various adverbial clauses, clitic pronouns in Spanish, negation, and
interrogatives. The second half of the chapter is devoted to exploring crosslinguistic interaction.
We observe the footprints of English in the boys’ Spanish syntax, e.g., preposition stranding and
genitive possessive -’s in Spanish utterances, and in their lexicon, e.g., lifa for ‘leaf’. The boys’
code-switches adhered to the grammatical patterns of each language (e.g., Poplack, 1980), but
did not appear to fulfill communicative functions (cf. Zentella, 1997). The last part of Chapter 3
presents analyses of picture book narrations, which indicate that the boys were able to produce
age-appropriate narratives in both English and Spanish. As Silva-Corvalán notes, “This is indeed
an extraordinary achievement, given the extremely reduced input and limited possibilities of use
of Spanish that characterize the siblings’ process of Spanish language acquisition.” (p. 117).
In Chapters 4 – 7 Silva-Corvalán focuses on particular grammatical structures. Chapter 4
provides evidence that the boys – especially Brennan – produced high rates of subject expression
in Spanish. Furthermore, Brennan diverges from monolinguals with respect to the impact of
discourse-pragmatic constraints. Silva-Corvalán proposes that both linguistic complexity and
crosslinguistic influence explain this finding. First, Spanish subject pronoun use patterns are
highly complex (Otheguy & Zentella, 2012), which may explain, in part, why these patterns take
a long time to acquire (Shin, 2015). Second, Silva-Corvalán suggests that the increasing
entrenchment of the English pattern of subject expression reinforces a tendency to express the
subject in Spanish (p. 169). Thus, linguistic complexity may render Spanish subject expression
‘susceptible’ to crosslinguistic influence, while English itself may also play a role by promoting
more abundant subject expression (Otheguy & Zentella, 2012; Shin & Montes-Alcalá, 2014, but
also see Michnowicz, 2015).1
Chapter 5 delves into subject placement. Researchers interested in this topic will be particularly
pleased by the detailed description of coding methods, especially the list of verbs categorized as
unaccusative (pp. 176-177). Silva-Corvalán finds that the boys’ rates of pre-verbal subjects in
Spanish were higher than those of monolingual Spanish-speaking children. At the same time, the
analysis of subject placement with particular verb lexemes and with animate versus inanimate
316 Heritage Language Journal, 12(3)
December, 2015
subjects suggests that, for the most part, they acquired the “semantic and discourse-pragmatic
principles that regulate subject placement (p. 215).”
Chapter 6 provides highly detailed analyses of the boys’ copula selection, and concludes that
their use of ‘be’ in English and ser and estar in Spanish is typical of patterns found among
monolingual children. This chapter also provides interesting evidence for piecemeal learning.
For example, the boys produced forms of the copula be in lexically specific units: [it’s N] and
[that’s N], and only later abstracted to a more general construction (p. 233). The imperfect forms
of Spanish copula estar initially occurred with only first and third person singular subjects (p.
242),2 and 78 of 90 adjective types were used only with ser or only with estar (not both).
The main focus of Chapter 7 is tense, mood, and aspect (TMA) morphology. The conclusion is
that the boys’ path of development of tense in English was comparable to that of monolingual
English-speaking children (p. 310), but their Spanish TMA system was simpler than that of their
monolingual Spanish-speaking counterparts. The boys did not produce the most infrequent and
complex of the Spanish tenses, such as the conditional and the present perfect indicative (p. 332).
Furthermore, the boys’ simplified Spanish verb system is similar to that of bilingual adults in the
U.S. Silva-Corvalán thus proposes that the verb system of bilingual adults is the result of
“incomplete acquisition” (p. 346) (but see Otheguy, 2015, for a critique). Chapter 7 includes
other findings that will also be of interest to language acquisition scholars. For example, Silva-
Corvalán finds examples of root infinitives (RIs) that lend support to the idea that RIs convey a
modal interpretation (Hoekstra & Hyams, 1998), for example yo cortar eso (‘I to-cut that’)
interpreted as meaning yo quiero cortar eso (‘I want to cut that’) (p. 277). Moreover, the boys
produced more person agreement errors than tense marking errors, which is predicted by Bybee’s
(1991) semantic relevance scale for verbal morphology.
Silva-Corvalán concludes the book in Chapter 8. Here she concisely summarizes some of her
main findings regarding questions posed in the Introduction. For example, she convincingly
demonstrates that amount of exposure to Spanish, which correlates with proficiency, had a clear
effect on the boys’ Spanish language development. In particular, compared to Brennan, Nico was
exposed to more Spanish and also consistently exhibited greater command of certain linguistic
structures in Spanish, including TMA verb morphology (pp. 352-352). On page 356, Silva-
Corvalán introduces her parallel structure hypothesis, which I describe in detail below.
2. THEORETICAL CONTRIBUTIONS
Crosslinguistic Interaction
For Silva-Corvalán, there are two types of crosslinguistic interaction: transfer and language
influence (pp. 10-12). She further distinguishes between two types of transfer. The first involves
the incorporation of a form from one language into the other, such as lipo ‘lip’, estoquié ‘stuck’,
and, pueres ‘you pour’ (pp. 80-81). The second type of transfer involves the incorporation of
meaning of a form from one language into another. Examples include No puedo esperar X ‘I
can’t wait’ taking on the meaning of ‘keen anticipation’ (p. 12), tomar la siesta ‘take the nap,’
estar over ‘to be over’, and tener un bath ‘have a bath’ (pp. 79-84).3 Silva-Corvalán also finds
transfer in the boys’ Spanish morphosyntax. For example, the boys produced preposition
317 Heritage Language Journal, 12(3)
December, 2015
stranding, e.g. ¿Qué es eso para? ‘What is this for?’ (pp. 68-72), and genitive possessive (’s),
e.g. ¿Dónde está Bibi’s manzana? ‘Where is Bibi’s apple?’ Silva-Corvalán defines her second
category of crosslinguistic interaction – language influence – as follows: a “superficially parallel
structure in the more frequently used language (A) motivates a higher frequency of use of the
corresponding structure in the other language (B)” (p. 12). She finds evidence of ‘influence’ in
the boys’ patterns of Spanish subject pronoun expression, preverbal subjects, and [adjective +
noun] word order (pp. 76-77, 153, 194, 206-215).
As mentioned above, in her final chapter Silva-Corvalán presents the parallel structure
hypothesis. According to this hypothesis, crosslinguistic influence will only take place at the
morphosyntactic level if two requirements are met. First, the two languages must have a structure
in common, but crucially there is variation in usage in the non-dominant language. Consider
subject placement. Spanish and English have a parallel structure: subject-verb word order. But in
Spanish verb-subject order is also frequent, whereas in English it is not. The same reasoning
applies to subject pronoun expression. The parallel structure is the expressed subject pronoun.
But in Spanish subject omission is also frequent, whereas in English omission is rare. The second
requirement of the parallel structure hypothesis is that use of the structure in question is
constrained by discourse-pragmatic factors in the non-dominant language, but not in the
dominant language. For example, information structure constrains variable subject placement in
Spanish, but less so in English, as word order is much more fixed in English. Similarly, changing
reference across two consecutive subjects significantly predicts subject pronoun expression in
Spanish, but should have a weaker impact in English, as subject expression is more variable in
Spanish than in English.4 Silva-Corvalán proposes that if these two requirements are met, there
will be influence from the dominant on the weaker language, hence the increase of subject
pronoun expression and pre-verbal subject placement in Spanish, and a concomitant weakening
of the discourse-pragmatic constraints on usage in Spanish. She then argues that the lack of an
effect from English on Spanish copula selection is because the use of one copula rather than the
other involves a lexical change. As such, she adds to the parallel structure hypothesis a
modification that it must not involve a lexical change. (p. 357). Thus the book culminates with a
hypothesis that broadly captures the results of Silva-Corvalán’s own study and, as is typical of all
of her groundbreaking work, she leaves us with exciting ideas for shaping future research
questions.
Towards a Usage-based View of Language Acquisition
Silva-Corvalán provides ample evidence that Nico and Brennan acquired morphosyntax in a
piecemeal fashion, suggesting a lack of generalized syntactic rules at the beginning stages of
development. Lexically specific verb frames, such as the collocation I want it (p. 90), emerged
first and acted as a precursor to a generalized subject-verb-object word order in English (pp. 215-
216, also see Ambridge and Lieven, 2011). In fact, Nico produced 20 verbs types in English by
age 1;8.27, but 19 were used with one form and one verbal function only (p. 283). For example,
during this initial period, the verb break only occurred as past tense broke, and dance only
occurred as imperative dance. Similarly, most of the boys’ Spanish verb types were used in one
form with only one verbal function (pp. 290-295, 300, 344). Haber only occurred as present
tense hay (‘there is/are’). These data suggest that what looks like tense-marking is not
318 Heritage Language Journal, 12(3)
December, 2015
generalized over verb types (Tomasello 1992, p. 23). Further evidence for piecemeal learning
comes from the chapter on ser and estar. The boys’ acquisition of these copulas began with
“lexically specific constructions, memorized chunks and routines” (p. 252).
Another hallmark of the usage-based model of acquisition is the importance placed on the role of
input. Generative models invoke the poverty of the stimulus argument, which essentially claims
that linguistic input alone is insufficient for successful language acquisition (e.g., Chomsky,
1965). In contrast, usage-based models argue that children’s linguistic development closely
aligns with the input they receive. This argument is supported by studies showing frequency
effects, e.g., more frequent structures are acquired first (Ambridge, Kidd, Rowland, &
Theakston, 2015). Silva-Corvalán invokes frequency effects to explain the boys’ success with
gender agreement in Spanish (pp. 78-79)5 and lack of success with infrequent structures, such as
the preterit forms of ser and estar (p. 240) and infrequent verb tenses (pp. 311, 327-329). Also,
the general finding that Nico acquired some structures that Brennan did not is also evidence of
the impact of frequency in the input, as Nico was exposed to more Spanish input than Brennan
was. The crucial role of input is also apparent from the growing evidence that children match
patterns found in their caregivers’ production data – even when these patterns are probabilistic,
(e.g. Hudson Kam, 2015; Miller, 2013; Roberts, 1997; Smith, Durham, & Richards, 2013). Silva-
Corvalán’s study adds to this body of literature: the boys produced copulas be (231, 232), ser,
and estar (p. 252) in particular contexts at rates that closely mirror those of adults.
Complexity of Linguistic Structure
A particularly interesting result of Silva-Corvalán’s study illustrates how structural complexity
impacts language acquisition. Even though English was the children’s dominant language,
negation and wh-question formation were acquired earlier in Spanish than in English, most likely
because these structures are more complex in English than in Spanish (pp. 60-63, 350, 353).
Silva-Corvalán also argues that the children had more difficulty learning subject pronoun
expression than copula selection or subject placement because subject pronoun expression is
constrained by more factors, and is thus more complex (p. 354) (See also Shin, 2015).
SUMMARY
Silva-Corvalán has produced a labor of love that is chock full of precious data, analyses, and
clearly articulated hypotheses to be tested in future research. Those of us who investigate child
language, bilingualism, language contact, and linguistic complexity will read this volume many
times over, and will use it as a starting point for our own projects. The book is also useful for
graduate seminars on bilingualism, bilingual language acquisition, Spanish in the U.S., and
numerous other topics. Silva-Corvalán has yet again produced seminal work that will serve as a
benchmark for future research for many years to come.
319 Heritage Language Journal, 12(3)
December, 2015
REFERENCES
Ambridge, B., & Lieven, E. (2011). Child language acquisition: Contrasting theoretical
approaches. Cambridge, UK: Cambridge University Press.
Ambridge, B., Kidd, E., Rowland, C. F., & Theakson, A. L. (2015). The ubiquity of frequency
effects in first language acquisition. Journal of Child Language, 42, 239-273.
Anderson, R. (1999). Loss of gender agreement in L1 attrition. Preliminary results. Bilingual
Research Journal, 23(4), 389-408.
Anderson, R. (2012). First language loss in Spanish-speaking children. In B. Goldstein (Ed.)
Bilingual language development and disorders in Spanish-English speakers (2nd ed.) (pp.
193-212). Baltimore, MD: Brookes Publishing.
Bedore, L., & Leonard, L. (2000). The effects of inflectional variation on fast mapping of verbs
in English and Spanish. Journal of Speech, Language, and Hearing Research, 43, 21-33.
Bybee, J. (1991). Natural morphology: The organization of paradigms and language acquisition.
In C. Ferguson & T. Huebner (Eds.), Cross-currents in second language acquisition and
linguistic theory, 67-91. Amsterdam, Netherlands: John Benjamins.
Chomsky, N. (1965). Aspects of the theory of syntax. Cambridge, MA: MIT Press.
Hoekstra, T., & Hyams, N. (1998). Aspects of root infinitives. Lingua, 106, 81-112.
Hudson Kam, C. (2015). The impact of conditioning variables on the acquisition of variation in
adult and child learners. Language, 91(4), 906-937.
Michnowicz, J. (2015). Subject pronoun expression in Yucatan Spanish. In A. Carvalho, R.
Orozco, & N. L. Shin (Eds.), Subject pronoun expression in Spanish: A cross-dialectal
perspective (pp. 101-120). Washington, DC: Georgetown University Press.
Miller, K. (2013). Acquisition of variable rules: /s/-lenition in the speech of Chilean Spanish-
speaking children and their caregivers. Language Variation and Change, 25, 311-240.
Montrul, S., & Potowski, P. (2007). Command of gender agreement in school-age Spanish-
English bilingual children. International Journal of Bilingualism, 11(3), 301–28.
Otheguy, R. (2011). Functional adaptation and conceptual convergence in the analysis of
language contact in the Spanish of bilingual communities in New York. In M. Díaz-Campos
(Ed.), The handbook of Hispanic Sociolinguistics (pp. 504-529). Malden, MA: Blackwell
Publishers.
Otheguy, R. (2015). The linguistic competence of second-generation bilinguals: A critique of
‘incomplete acquisition.’ In C. Tortora, M. den Dikken, I. Montoya, & T. O'Neill (Eds),
Romance Linguistics 2013. Selected papers from the 43rd Linguistic Symposium on Romance
Languages (LSRL) (pp. 301-318). John Benjamins Publishing Company.
Otheguy, R., & Zentella, A. C. (2012). Spanish in New York: Language contact, dialect leveling,
and structural continuity. New York, NY: Oxford University Press.
Poplack, S. (1980). Sometimes I’ll start a sentence in Spanish y termino en español: Toward a
typology of code-switching. Linguistics, 18 (7/8), 581-618.
Roberts, J. 1997 Acquisition of variable rules: a study of (-t, d) deletion in preschool children.
Journal of Child Language, 24, 351- 372.
Shin, N. L. (2015). Acquiring constraints on morphosyntactic variation: Children’s Spanish
subject pronoun expression. Journal of Child Language. doi:
http://dx.doi.org/10.1017/S0305000915000380
320 Heritage Language Journal, 12(3)
December, 2015
Shin, N. L. & Montes-Alcalá, C. (2014). El uso contextual del pronombre sujeto como factor
predictivo de la influencia del inglés en el español de Nueva York. Sociolinguistic Studies, 8,
85-110.
Smith, J., Durham, M., & Richards, H. (2013). The social and linguistic in the acquisition of
sociolinguistic norms: caregivers, children, and variation. Linguistics, 51, 285-324.
Tomasello, M. (1992). First verbs: A case study of early grammatical development. New York,
NY: Cambridge University Press.
Torres Cacoullos, R., & Travis, C. (2015). Foundations for the study of subject pronoun
expression in Spanish in contact with English: Assessing interlinguistic (dis)similarity via
intralinguistic variability. In A. Carvalho, R. Orozco, & N. L. Shin (Eds.), Subject pronoun
expression in Spanish: A cross-dialectal perspective (pp. 81-100). Washington, DC:
Georgetown University Press.
Zentella, A. C. (1997). Growing up bilingual. Malden, MA: Blackwell Publishers.
321 Heritage Language Journal, 12(3)
December, 2015
NOTES
1. A minor typo is worth pointing out for researchers who explore subject expression: the title of
Table 4.3 on page 133 should have the word ‘pronouns’ removed, i.e., Overall percentage of
use of overt subject pronouns by Manuela and three monolingual children. The percentages
reported in this table reflect expression of all types of subjects, including lexical and
pronominal subjects.
2. At the same time, Silva-Corvalán notes that contexts for the use of plural forms were limited
in the data, which underscores the need to complement corpus studies with experimental work.
3. Although researchers agree that these are usages that emerge in situations of language contact,
Otheguy (2011) argues that these so-called phrasal calques are more aptly characterized as
instances of conceptual convergence rather than linguistic convergence. Otheguy’s position
calls into question whether phrases such as tomar la siesta are the result of a copy mechanism
that directly transfers phraseology from English into Spanish.
4. In fact, switching reference impacts English subject expression (Torres Cacoullos & Travis,
2015), but subject expression is much more variable in Spanish than in English.
5. Previous research on bilingual children has found evidence of gender mismatches between
nouns and their modifiers (Anderson, 1999, 2012; Montrul & Potowski, 2007). Silva-
Corvalán, however, finds that “gender assignment and gender agreement are acquired in an
amazingly errorless fashion, except with clitic pronouns.” (pp. 77-78). She attributes their
success with gender to the “regularity, transparency and productivity” of the Spanish gender
system (p. 78). More research is needed to better understand why some researchers have found
gender mismatches in bilingual children’s Spanish but others have not.