Review of Understanding Interfaces
The layout and content of the book are as follows: Chapter 1 introduces the basic tenets of Minimalism and the role that interfaces play in its design. Domínguez briefly first addresses the now standard assumptions of a tripartite structure composed of syntax, and the two interfaces, LF (Logical Form) and PF (Phonetic Form). She points out that based on Chomsky (1995), these are considered to be the only levels necessary for the construction of linguistic expressions. However, in her view, the bare bones computational system presents a problem: it does not capture the syntax-pragmatics interface. Take, for example, the Extended Projection Principle (EPP), which requires the subject position to be filled in all languages. In Spanish this requirement may be satisfied even if the subject is not phonetically realized or does not move to the designated subject position. In addition, Spanish allows for free word order. Thus, while a range of possible structures are all well-formed, they are not, in fact, in free distribution, as they express different informational content and are appropriate in specific contexts only. This, Domínguez argues, is the key problem: how are these grammatical alternative linguistic expressions sorted out by the tripartite UG system? To solve the problem, Domínguez contemplates Reinhart's (2006) and Jackendoff's (2002) models. For Reinhart, the solution lies in enriching LF (Logical Form) with more fine-grained modules that handle inference and context in addition to concepts. For Jackendoff, the solution lies in loosening the tripartite model. Specifically, Jackendoff advances a more leveled view of the syntax-semantic-phonology interaction: for him, each of the grammar modules contributes equally. Domínguez, in turn, asks how these alternatives are sorted out in L2 acquisition.
Chapter 2 presents an examination of the syntactic, prosodic, and pragmatic constraints that are known to affect the position of subjects in a Spanish sentence. To summarize, Domínguez recounts that (i) subjects first merge in [Spec, VP] and optionally move to IP to satisfy EPP; (ii) a null referential pronoun, a null expletive, and either an overt or null adverbial can also satisfy EPP; (iii) subjects can optionally stay in their original position.
The properties of scrambling, fragment answers, and clitic left dislocation are discussed in great detail. Domínguez convincingly argues that the movement component of the syntactic operations may be optional, and is required to meet interpretative requirements of the interfaces. That is, while the position of phrases in sentences is conveyed by syntax, it is ultimately determined by interfaces.
Chapter 3 is devoted to presenting, analyzing, and assessing the current interface hypotheses in the light of L2 acquisition. Domínguez's main claim is that syntactic structure may be impaired in acquisition of L2, if L2 and L1 differ in language specific properties. Based on Spanish data presented in Chapter 2, she argues that, for example, variation in subject properties across languages affects acquisition of L2. Thus, an English speaker would potentially stumble in the acquisition of Spanish subject distribution. Null and postverbal subjects are prominent in Spanish, in contrast to English. These contrasts can be reduced to a difference in the inventory of abstract features that drive syntax.
Chapter 4 discusses the second language acquisition of two interface phenomena by English learners of Spanish: null/overt and post-verbal subjects, both constrained by focus and syntax. First, Domínguez provides an overview of the previous studies on null/overt subject and problems observed. Overgeneralization, for example, is a common problem in both overt and null subject acquisition. For the purposes of her study, the relevant observation gleaned from the literature is that pragmatic deficits are not as persistent as L1/L2 differences would lead one to believe. Next, Domínguez turns to word order studies, which have been much rarer. Here, again, the relevant observation is that inversion of subjects is slower and more gradual than acquisition of null subjects. For Domínguez, this is an inconsistency because the current view would predict there should be no divergence here: both null and inverse subjects are constrained by syntax and pragmatics. Finally, Domínguez describes her own empirical studies, (i) in comprehension; and (ii) in production. The focus of both studies is on the acquisition of subject realization and word order variation.
Data were obtained from three groups of L2 Spanish speakers and a group of native Spanish speakers. The data were collected using a range of tasks: story retelling, a paired discussion task, and an interview with an investigator. The sixty learner participants were native speakers of English learning Spanish. Ages ranged from complete beginners (13-14) to very advanced undergraduates in the final year of a Spanish degree. The results of both studies indicate that a possible syntactic deficit persists in the grammars of English speakers of Spanish; that is, the results of the studies cannot be due only to pragmatic impairment.
Chapter 5 covers the same issues as Chapter 4 except the native grammars of late Spanish-English bilinguals are the source of data. The focus is on non-pathological native syntactic attrition where an adult grammar is modified, restructured, or partially lost as a result of quantitative or qualitative changes in L2 input or extensive exposure to input from another language. The participants of the production and comprehension studies were Spanish-English bilinguals living in two different areas: Miami, Florida (US) and south-east England (UK). Thirty-one native speakers of Spanish (average age 61) living in an English-speaking country for an average of thirty-five years were interviewed. The results of the studies show that (i) word order variation and subject realization are, indeed, vulnerable areas in attrition-prone environments; (ii) the observed attrition cannot be due only to pragmatic impairment. Moreover, Domínguez notes that community specific differences in L1 input need to be taken into account in future studies of attrition.
Chapter 6 brings the book to an end with findings, conclusions, and implications.
Linguists interested in the acquisition-driven theory are the target audience for this book. From the point of view of an acquisitionist, Domínguez challenges a common key premise: that the syntax-pragmatics interface is the primary candidate for variation in L2 acquisition.
Having examined the learner and native speaker data, Domínguez posits that syntactic representations also can be impaired, and that they are subject to the same difficulties as properties at interfaces. Moreover, she argues that there is no clear theoretical support for differentiating between internal and external interfaces in any of the models of grammar reviewed.
The book’s main proposal that interface-based models need to accommodate the possibility that ''core'' syntactic representations can also be impaired. Essentially, Domínguez suggests that we should eliminate the distinction between ''core'' syntactic and interface phenomena. In line with Reinhart (2006) and Jackendoff (1997, 2002), she argues that all syntactic structures can be subject to impairment and vulnerability. Her arguments in support of this view rely on the acquisition and attrition data from Spanish.
In Spanish, subjects can take multiple positions within the same sentence. The two empirical studies presented in the book converge in showing that syntactic properties of subject realization and word order variation can be a source of divergence of L2 acquisition and native attrition. This does not support the main prediction of the currently assumed Interface Hypothesis, where divergence is viewed as an issue reflecting the syntax-pragmatics interface. Domínguez shows that advanced L2 speakers incorrectly overproduced structures with inversion in syntax-only contexts.
Domínguez argues for the view that parameterization -- feature selection (lexical parameterization) and language specific interface mappings -- are the key factors in explaining problems in L2 acquisition. She maintains that current minimalist approaches focus either on lexical parameterization or on interface mappings, while the two factors should be tackled simultaneously.
The author successfully challenges the current view of Interface Hypothesis from the minimalist perspective, grounding her arguments in solid empirical data. Her arguments remind us that a constant re-evaluation of theory should shape our assumptions and research questions. The intriguing possibility would be to see if her arguments would withstand a cross-linguistic test; e.g., if her findings would be similar for other Romance languages and beyond.
This book is aimed at an informed reader who is well versed in theoretical as well as empirical issues of second language acquisition. In order to appreciate the Spanish-English contrasts and similarities, the reader is expected to already have a solid knowledge of the specific empirical data, such as cross-linguistic variation in subject manifestations, the range and relevance of word order alternatives, and the like. Moreover, cutting edge knowledge of the current trends within generative grammar is required, too. The relevance of its contribution to the study of language acquisition and how it relates to theory will make this book useful to researchers who worry about supporting their theory with solid empirical data.
Chomsky, N. (1995). The minimalist program. Cambridge: MIT Press.
Jackendoff, R. (2002). Foundations of language: Brain, meaning, grammar, evolution. Oxford: Oxford University Press.
Jackendoff, R. (1997). The architecture of the language faculty. Cambridge: MIT Press.
Reinhart, T. (2006). Interface strategies. Reference-set computation. Cambridge: MIT Press.
| ABOUT THE REVIEWER:
Solveiga Armoskaite is currently a Visiting Assistant professor at University of Rochester. She is particularly interested in how syntactic categories emerge, and what makes them vary across languages, e.g. to what extent the notion of nounhood is universal. She also explores what syntactic features are, how they cluster, interact and drive syntactic derivation, e.g. how a feature like grammatical gender or definiteness manifests itself in unrelated languages. Solveiga is passionate about any kind of fieldwork data driving theoretical speculation.