Why do we need another book about unbounded dependencies? A review article on Chaves & Putnam’s Unbounded Dependency Constructions

Unbounded dependencies (UDs), in wh-interrogatives, relative clauses and other constructions, have been a major focus of syntactic research for more than half a century. The most widely assumed approach analyzes them in terms of movement and views island phenomena as largely a matter of syntax. Both these positions are problematic. Moreover, they stem from assumptions that have been at the heart of syntactic theorizing for many decades. Chaves and Putnam present evidence that both the dominant approach to UDs and the general approach to syntax from which it derives are flawed. They argue for a non-movement approach to UDs and a largely non-syntactic approach to island phenomena, and for an approach to syntax which has the relation between linguistic knowledge and language use and the complexity of acceptability judgements as central concerns. Their book is an important one that could have a major impact both on research on UDs and on syntactic research more generally.

Extraction types

This chapter provides a descriptive introduction to unbounded dependency constructions, and to the range of filler-gap dependency patterns they allow. These are two very different topics. The former concerns the repertoire of syntactic constructions in which extraction takes place, and their grammatical idiosyncrasies, whereas the latter concerns the types of interaction between fillers and gaps. We survey of the various kinds of interrogative, declarative, and subordinate UDCs that exist in English. The resulting picture is one of astonishing richness and complexity. There are three major families of UDCs which sub-divide into smaller families, each with their peculiar syntactic, pragmatic and phonological similarities, as well as their idiosyncrasies. Such idiosyncratic meaning and structure must be stipulated somewhere in the grammar, regardless of which theory one adopts. The chapter next focuses on the nature of the linkage between fillers and gaps, and shows that these can interweave and create complex dependencies, beyond what is recognized in the literature.

Extraction constraints

This chapter offers a detailed survey of the constraints that restrict filler-gap dependencies (island constraints), and argues that there are several different kinds of island constraints, due to different combinations of independently motivated factors. Most importantly, it argues that most islands are not cross-constructionally active. That is, most island phenomena are restricted to certain kinds of unbounded dependency constructions (e.g. interrogatives, or relative clauses). In particular, several island types are primarily caused by drawing the hearer’s attention to a fronted referent that is not at-issue, and is of little consequence to what the utterance convey. Such an account emerges naturally from the observation that not all propositions express equally likely states of affairs and that different constructions come with different biases with respect to how information structure is packaged, and consequently, to which referents it is pragmatically licit to single out. The chapter concludes with a discussion of resumption and supposed island effects in other types of construction.

The acquisition of UDCs

This chapter focuses on behavioral evidence concerning acquisition of unbounded dependency constructions and island phenomena. It provides a general overview of their acquisitional trajectory in normal developing L1, focusing on the emergence of filler-gap dependencies from one-word constructions to multi-clausal constructions, as well as island effects. The chapter argues that the gradual and frequency-based developmental evidence is consistent with an exemplar-based approach that contains rich morphosyntactic, semantic, and pragmatic information along the lines of the empiricist perspective. In particular, the extant evidence supports a usage-based view in which the growth of a grammar proceeds from simple units to more complex ones, exploiting the frequency of simple as well as complex exemplars. The chapter concludes by describing a exemplar- and chunk-based account of grammar development, composed of rich information that is probabilistic in nature, and shaped by experience.

Movement-based approaches

This chapter discusses how the Minimalist Program (MP) strives to model unbounded dependency constructions and island constraints, and discusses the empirical, theoretical and cognitive status of syntactic displacement (movement), as formalized in terms of Internal Merge. At the present time, modelling filler-gap dependencies via movement faces significant theoretical and empirical issues. There is no parsimonious account of successive cyclic movement in the MP because of the Triggering Problem, nor of convergent and cumulative filler-gap dependencies. Other problems concern island phenomena, which have been argued to follow from core architectural economy constraints, but which make incorrect predictions not only about islands, but also about unbounded dependency constructions more generally. Finally, the MP has also been difficult to reconcile with extant psycholinguistic evidence about language processing. All recent attempts to make the MP consistent with incremental sentence processing adopt phrase-structural information, and abandon movement altogether.

Non-movement-based approaches

This chapter compares movement-based conceptions of grammar and of unbounded dependency constructions with their construction- and non-movement-based antithesis. In particular, the focus of this chapter is on how unification and construction-based grammar provides not only a better handle on the phenomena than the MP from a linguistic perspective, but also from a psycholinguistic point of view. The flexibility of non-movement-based accounts allows a much wider and much more complex array of unbounded dependency patterns because it rejects the basic idea that extracted phrases start out as being embedded in sentence structure, and instead views the propagation of all information in sentence structure as a local and distributed (featural) process. The grammatical theory discussed in this chapter is also more consistent with extant models of human language processing than the MP, and demonstrably allows for efficient incremental and probabilistic language models of both comprehension and production.

Introduction

This chapter introduces the phenomena that will be discussed in the remainder of the book, discusses how these are modelled in a variety of frameworks, and how human beings process such filler-gap dependencies. The chapter concludes with five overarching questions, each of which is the topic of subsequent chapters: (1) What is the possible range of filler-gap dependency types? In particular, what patterns arise when there are multiple gaps?; (2) Is there a common constraint at work in most or all island phenomena?; (3) What are the advantages or disadvantages of movement-based versus non-movement-based approaches? (4) How can a theory of grammar account for the fact that some (but not other) island violations have gradient acceptability, are prone to frequency effects, and are sensitive to contextual information? (5) How can unbounded dependency constructions be learned by speakers? Does the evidence favor nativist approaches or domain-general experience-based approaches?

On a family of Welsh constructions

Research on unbounded dependency constructions (UDCs) has focused mainly on the properties that are shared by all UDCs, but a satisfactory theory of syntax also needs to capture the properties that distinguish specific UDCs and the properties that are shared by some but not all of them. Three Welsh unbounded dependency constructions – wh-interrogatives, free relatives, and cleft sentences – are of interest here because they show a challenging array of similarities and the differences. However, given a slightly expanded hierarchy of phrase types, HPSG can capture both the similarities and the differences in this area.

Phrase Structure

The phrase structure of English has been a central concern for most approaches to syntax, including various forms of Transformational Grammar, Head-driven Phrase Structure Grammar, Lexical Functional Grammar, and the earlier Generalized Phrase Structure Grammar framework. They have developed detailed analyses of verb phrases, nominal phases, clauses of various kinds, including unbounded dependency clauses and elliptical clauses, and adjective phrases and prepositional phrases, and coordinate structures. There are similarities and differences between the various approaches in all these areas. They differ in whether or not they are confined to binary branching, whether or not they assume that all phrases are headed, and in the extent to which they assume heads which are phonologically empty. More generally they vary in how complex they take phrase structures to be and in how much variety they see in the local trees that they consist of.