Book announced at https://linguistlist.org/issues/34.3163

AUTHOR: Ian Roberts
TITLE: Beginning Syntax
SUBTITLE: An Introduction to Syntactic Analysis
SERIES TITLE: Cambridge Textbooks in Linguistics
PUBLISHER: Cambridge University Press
YEAR: 2023

REVIEWER: Fauzia Mughal

SUMMARY

Ian Roberts’s “Beginning Syntax: An Introduction to Syntactic Analysis” is the first installment of a three-part series. This is an introductory textbook on generative syntax that is meant to guide undergraduate students through this framework without assuming any prior knowledge of the topic. It addresses core aspects of the theory that form the bedrock of any syntax course. With respect to its layout, the book is a bit over 200 pages in length. There is an Introduction, eight chapters, and a Conclusion. Every chapter begins with an introductory section that provides a brief roadmap, and the introductions to Chapters 2–8 further include recaps of the core concepts covered in the immediately preceding chapter. Each chapter ends with hands-on exercises, broad discussion questions, and an annotated list of references for further reading. The book ends with a glossary and an index. The much-anticipated subsequent textbooks (Volumes II and III), entitled “Continuing Syntax” and “Comparing Syntax”, will further lead students through syntactic theory at the intermediate and advanced levels, respectively.
The Introduction lays the foundation for the text. It does so by defining what it means to be a language as a natural object of study and what it means for syntax (and linguistics more generally) to be a cognitive science. There are three factors that constitute our capacity for language: 1) our innate genetic endowment for language, 2) the stimuli we are exposed to in early life, and 3) our general cognitive abilities. Universal Grammar (UG) is a theory that represents the first of these factors. The focus of syntax is to develop a formal theory of one aspect of UG, namely, how relatively simple units (words) are systematically combined to form relatively complex ones (phrases and sentences), by analyzing language from an internal, individualized perspective (i.e., I-language).
Chapter 1 discusses the fundamental properties of natural language syntax that form one’s tacit knowledge of a given language using a single English word: fish. Here, readers are introduced to conventional formalizations of their innate judgements about their native language (i.e., I-language), including structural ambiguity, phrasal categories, recursivity, competence vs. performance, and silent material, among others, that will be further explicated in the following chapters. The juxtaposition of English and Italian data throughout illustrates the (near-)universality of syntax.
Chapter 2 covers the two building blocks of syntactic theory: categories and constituents. Standard morphological, syntactic, phonological, and semantic criteria are first used to identify phrasal categories in English and to distinguish lexical ones from functional ones, with the caveat that these are not foolproof, particularly in isolation. Next, the ways in which categories combine to form constituents and constituent structures are explored. For ease of exposition, tree diagrams are used to illustrate the core hierarchical relations of (immediate) dominance and (immediate) constituency. Although the notion of category and constituent, as well as the X'-template, are assumed to be a part of UG, it is made clear that languages may differ with respect to the number and types of categories they possess and in their constituent structures.
Chapter 3 introduces the formal mechanisms involved in computing structural descriptions of sentences, viz., Phrase-Structure (PS)-rules and their ability to apply ad infinitum to their own output, capturing a crucial property of natural languages: a finite set of rules can be used to produce an infinite set of structures. PS-rules encode constituent information regarding linear precedence, hierarchical relations (viz., dominance/constituency), and category labels. Such rules and the tree diagrams (or labeled bracketings) that they produce are substantiated by a complete array of constituency tests, which are shown to be generally successful with respect to English data; the question as to whether such diagnostics provide clear results in other languages remains open.
Readers are met with X'-theory in Chapter 4, which constrains the possible PS-rules used to derive lexical and functional categories in English to those that are well-formed. With the proposed X'-schema comes the universal principle of endocentricity, requiring phrases to always contain a head. The basic structural relations of the X'-template are established here (viz., specifier, complement, and adjunct), with the pivotal differentiation among these being that only heads select for their complements. Constituency tests are used to independently justify the postulation of intermediate phrasal levels (e.g., N', P', and A'). Relatedly, the apparently exocentric categories S(entence) and S' that were assumed prior to X'-theory are analyzed as T(ense) P(hrase) and C(omplementizer) P(hrase), respectively.
Chapter 5 introduces three types of movement rules that augment the X'-theory established so far and that are to be discussed in greater depth in Volume II of this series. The first rule is head-movement, which is empirically illustrated by subject–auxiliary inversion in English questions that involve T-to-C movement. The second rule that is introduced is NP-movement—a transformation that is exemplified by English passivization, which incorporates the pivotal notions of argument structure and thematic roles (e.g., Agent and Patient, etc.). Here, it is described how the complement of a verb raises to become the subject of the sentence and the original subject is demoted to an optional by-phrase, though the constituents’ respective thematic roles remain intact. The third movement rule is wh-movement, which is illustrated yet again by English data, wherein wh-phrases raise to the specifier of CP to form content questions. Wh-movement uniquely produces constructions with an unbounded dependency, is triggered by features in the narrow syntax (viz., [Q] and [wh] on C), and is blocked by constraints in certain contexts (viz., syntactic islands). All of these movement rules leave behind an unpronounced copy (or trace) in the base-generated position, but they crucially differ from PS-rules in that they manipulate (rather than generate) structure.
Chapter 6 further enriches the proposed X'-theory by incorporating interpretative rules, with a specific focus on binding relations. This completes the text’s introduction to the core mechanisms underlying syntactic computation. Binding theory differs from PS-rules in that it does not generate structure, and it is different from movement rules in that it does not manipulate the structures that PS-rules produce. Rather, binding theory licenses the semantic interpretation of relationships between constituents that are generated by PS-rules. Specifically, Principles A, B, and C of binding theory, which capture the distribution and interpretation of anaphors, pronominals, and R(eferential)-expressions, respectively, are illustrated via English data. This discussion further introduces the fundamental notions of c(onstituent)-command and co-reference, variable binding, and the distinction between A-positions and A'-positions (wherein the former is an XP position associated with a grammatical function and the latter is not).
Chapter 7 extends the theory of syntax established in the previous chapters to facts beyond English, in keeping with the central aim of forming a description of a universal (rather than language-specific) model of syntax. It is emphasized that while universal principles and operations governing sentence structure exist (e.g., the X'-template, the principles of binding theory, and syntactic movement as a strictly upward process), such principles and operations are subject to parametric variation. With reference to cross-linguistic facts from the World Atlas of Language Structures (WALS) (https://wals.info/) and language-specific data, three parameters of two types are introduced for illustrative purposes: The first type relates to the presence/absence of movement rules such as 1) the V-to-T Parameter (which is set to positive in verb-raising languages like French and German and is set to negative for languages without verb raising like English and possibly Japanese); and 2) the wh-movement Parameter (which is set to positive for languages with overt wh-movement like English, French, and German and is set to negative for languages without overt wh-movement like Chinese and Japanese). The second type of parameter, referred to as 3) the Head Parameter, captures the various attested word orders of natural languages by establishing whether the categories of a given language are head-initial (as in, for example, English and French); head-final (as in, for example, Japanese); or a mixture of the two (as in, for example, German and Chinese). The question as to whether an NP-movement Parameter analogously exists is tentatively answered in the affirmative (based on the absence of passives in several natural languages).
Chapter 8 pieces together the various aspects of the syntactic theory outlined in the previous chapters and ties them into the general discussion in the Introduction. It provides a brief description of the assumed ‘inverted-Y’ architecture of the grammar. The lexicon houses idiosyncratic information and constitutes a numeration of lexical and functional items associated with c(onstituent)-selection and s(emantic)-selection features, a phonological representation, and a specification of said items’ semantics. The syntactic representations that PS-rules generate constitute D(eep)-structure; lexical and functional items (in line with the Projection Principle) are also introduced at this step of the derivation, as are arguments bearing specific thematic roles in particular structural positions (as far as these are structurally determined). S(urface)-structure comprises syntactic representations to which the various movement operations have applied, and it branches out into two independent interface levels: 1) P(honological) F(orm) takes the S-structure representations as input and outputs their phonological representations after hierarchically organized PF operations have applied (e.g., ‘downward’ T-to-V movement in English (i.e., m(orphological)-merger) precedes copy-deletion and ellipsis); such processes have no effect on the structures’ interpretation. 2) L(ogical) F(orm) takes S-structure representations and outputs their semantic interpretations. Operations that take place at the S-structure–LF interface (e.g., covert wh-movement in wh-in situ languages like Chinese and Quantifier Raising to derive quantifier-scope ambiguities) have no effect on the surface output of syntactic structures.
The Conclusion provides a brief summary of the main concepts covered in the various chapters and raises pertinent questions related to some of the key assumptions that the proposed syntax model evokes, several of which are to be addressed in Volume II of this series. These include the following: 1) To what extent can syntactic domains be expanded (akin to TP expanding to include categories like M(ood), Agr(eement), and Asp(ect)), and are there cross-linguistic differences in this regard? 2) Why does the [wh] feature on C trigger wh-movement, and what causes other movement operations (viz., head-movement and NP-movement) to occur? 3) What types of syntactic mechanisms can be parameterized (cf. head-movement and wh-/NP-movement, which concern the presence/absence of movement, and the Head Parameter, which does not); how much can be parameterized?; and how are parameter settings acquired? 4) Would it be more economical to abandon the distinction between D- and S-structure and alternatively assume that representations simply branch out to the interface levels (i.e., PF and LF) post-syntactically?

EVALUATION

This textbook is an excellent introduction to contemporary syntax for undergraduate students that have no prior knowledge of the matter; it thus succeeds in its goal. Roberts has the exceptional ability to discuss the highly abstract and complex notions that constitute syntactic theory with engaging and extremely succinct prose, which is by no means an easy feat (a sentiment that I recall feeling while reading his “Diachronic Syntax” as an undergraduate student myself). The data are clearly analyzable for a novice, and the use of real-life analogies for illustrative purposes are clever and effective.
The overall organization of the textbook is extremely coherent. The conceptual progression from one chapter to the next is seamless throughout, and the use of boldface font to highlight key terminology is a useful visual aide in discerning core takeaways—a property that students new to the subject matter will surely appreciate. In a similar vein, I believe that the text may have benefitted from incorporating movement arrows in its tree diagrams to better visualize and reiterate the direction in which constituents move, particularly in discriminating upward head movement from downward T-to-V movement (i.e., m-merger). The exercises and discussion questions at the end of each chapter are carefully constructed and will indeed serve as useful tools for instructors and teaching assistants alike in reinforcing key concepts in the classroom and/or tutorials, as intended (p. xiv).
This text’s discursive simplicity and overall brevity, I believe, have a two-pronged effect. While these characteristics certainly set this textbook apart from its competitors (e.g., Tallerman 2020; Carnie 2021) by etching the fundamentals of syntax in an arguably more effortless fashion, from a utilitarian perspective, its lack of in-depth discussion may make it perhaps difficult to extend its use beyond the intended audience (as in, for example, introductory graduate courses) in a way that some of its predecessors can. However, the novel conception of breaking down the different levels of syntactic theorizing into three smaller volumes may compensate for this, affording instructors the ability to use different installments of the series for courses of differing levels of academic rigor.
The discussion is largely based on English data, and a proper treatment of cross-linguistic facts is unfortunately limited to only one chapter (viz., Chapter 7), the impetus of which is “expository convenience” (p. 153). Nevertheless, I believe that a better interleaving of more cross-linguistic data throughout would not have been detrimental to the pedagogical straightforwardness of the text and would have only enriched it. From the student’s perspective, the use of English data in learning the complex and abstract concepts that underlie syntactic theory is only advantageous in so far as the student has a native competency in the language, which is not always the case, particularly in a classroom with students of diverse linguistic backgrounds.
Overall, Roberts has skillfully compiled a clear, precise, and easily digestible introduction to syntactic analysis, which is an immensely difficult task. This textbook is thus nothing short of a triumph and serves as an attractive pedagogical resource for both students and instructors.

REFERENCES

Carnie, A. Syntax : A Generative Introduction (4th edition). Oxford: Blackwell.

Roberts, I. 2007. Diachronic Syntax. Oxford University Press.

Tallerman, M. 2020. Understanding Syntax (5th edition). London: Routledge.

ABOUT THE REVIEWER

Fauzia Mughal is a Ph.D. candidate in Linguistics at the University of Ottawa, Canada. Her specialization is in theoretical syntax, and she is primarily interested in clausal ellipsis, with an empirical focus on Hindi-Urdu.