Search Results

This chapter continues the argument that the ‘Q-based’ account should be extended to all other wh-fronting languages. The focus of this chapter is the constraints on pied-piping in many of the world’s languages, as well as the related phenomenon of so-called secondary wh-fronting. The chapter begins with a comparison of the pied-piping structures of Tlingit and English. It is shown that the latter are much more constrained than the former. This more restricted pied-piping found in English (and most other well-studied wh-fronting languages) is shown to follow from an assumption that ‘Q’ in these languages must undergo Agreement with the wh-word. The resulting theory is shown to make a number of other interesting predictions regarding the variation seen in pied-piping across languages. The remainder of the chapter concerns the methods that languages have for ‘obviating’ these constraints on pied-piping. The first, dubbed ‘secondary wh-fronting’, are cases where the wh-word undergoes movement internal to the pied-piped phrase. The second, dubbed ‘massive pied-piping’, are cases where the constraints on pied-piping appear weakened in some environments. Q-based analyses of both these phenomena are developed and defended.

The wh-constructions of Tlingit are introduced, as are the central empirical and theoretical puzzles concerning pied-piping structures. The proposed ‘Q-based analysis’ of Tlingit wh-constructions is sketched, as well as its potential consequences for the general theory of wh-fronting and pied-piping structures. These consequences include (i) elimination of the very concept of ‘pied-piping’ from the theory of grammar, and (ii) an account of ad-position stranding, whereby its ill-formedness is not a property of movement per se. Following this introductory discussion, the major results of the book are outlined.

This chapter begins the argument that the ‘Q-based’ account should be extended to all other wh-fronting languages. It first presents some general typological and learning-theoretic arguments. These include (i) the fact that the ‘Q-based’ account is transparently motivated for other wh-fronting languages (such as Edo), and (ii) the ability for the ‘Q-based’ account to provide a unified theory of the ill-formedness of P-stranding and left-branch extractions across languages. Following this, the consequences of the account for the theory of pied-piping structures are examined. In particular, it is shown that extending the ‘Q-based’ account to all other wh-fronting languages entails that the phenomenon dubbed ‘pied-piping’ need not exist at all. Finally, the chapter develops a ‘Q-based’ theory of multiple wh-questions in English and German. It is shown that the analysis predicts the complementary distribution of Superiority Effects and Intervention Effects in these languages. Finally, Intervention Effects in pied-piping structures are examined, and the ‘Q-based’ theory is shown to make an accurate (and surprising) prediction.

This chapter reconsiders the role of the Subset Principle in language acquisition and change, arguing that consideration of true formal optionality enables one to define grammars generating languages that are in inclusion relations. This in turn facilitates an explanation of diachronic changes where absence of sufficiently robust PLD led acquirers to ‘default’ to ‘smaller language’‐generating grammars.

This chapter asks whether wh in-situ can violate islands. The semantic challenge posed by pied piping of DPs/PPs, in English or in-situ languages, is discussed. One solution is to reconstruct. Another is to interpret the wh with scope over the island. Long-distance answers pairing a matrix wh with wh inside wh islands are considered. They can support the view that covert scope taking is island insensitive. Or the wh can be interpreted at the site of the island to yield a family of questions, with the list pairing the family of questions with the matrix wh. Adjunct islands raise similar questions. Either the wh inside them are impervious to syntactic constraints or there are mechanisms to interpret them locally. Single-pair vs. multiple-pair answers and their theoretical implications are discussed, as are trapped list answers. Co-ordinate structures and issues related to D-linking and scope are also discussed.

According to Noam Chomsky, Move is the combination of Agree, Merge, and Pied-pipe. It decomposes into three more primitive operations, whereas Merge takes two syntactic objects—α and β—and forms a new object, and is therefore the basic combinatorial operation of narrow syntax. Agree holds between α and β having interpretable and uninterpretable inflectional features, respectively. This chapter examines three possible ways of eliminating head movement from narrow syntax: structure preservation, chain uniformity, and the A-over-A Condition. It shows that the A-over-A Condition, but not structure preservation and chain uniformity, offers a viable way of forcing pied-piping of the maximal projection of the goal.

This chapter focuses on the Combinatory Categorial Grammar (CCG), a strongly lexicalized theory of grammar in which grammatical categories consist of a syntactic type defining valency, along with a logical form and a phonological form. It also discusses the Categorial Lexicon, the sole repository of language-specific information whose sounds and meanings are projected by a small universal set of type-driven combinatory syntactic rules onto the sounds and meanings of all and only the sentences of the language. CCG includes a number of restricted combinatory operations for combining categories that are strictly limited to various combinations of operations of type raising, composition, and substitution. Both reflexive/reciprocal binding and control are bounded under Condition A of the binding theory—that is, they relate elements within a single verbal domain. The chapter also examines relativization and relative pronouns, embedded subject extraction, pied-piping of wh-items such as which and who(m) in noun phrases, coordination of conjunctions, and the expressive power and computational complexity of CCG. Finally, it compares CCG with Categorial Type Logic and Lambek grammars.

Languages differ with respect to whether they allow for infinitival interrogatives and infinitival relative clauses. In order to explain this variation, this chapter postulates the “Wh-Infinitive-Generalization” that links the (non-)availability of infinitival interrogatives/relatives to morphological properties of the infinitival C-system. Based on synchronic and diachronic evidence, it is shown that wh-infinitives are impossible in languages in which the left periphery of the infinitive cannot be occupied by a phonetically realized prepositional complementizer. In contrast, languages with wh-infinitives do exhibit prepositional complementizers as a result of grammaticalization. In order to derive the “Wh-Infinitive-Generalization,” the author argues that infinitival C⁰ is “defective” in languages without wh-infinitives (/ infinitival relatives) where “defective” infinitival C⁰ is understood in analogy to defective T⁰def, i.e. C⁰def cannot bear the complete range of features specific for C⁰ (i.e. [focus]-, [wh]-, [topic]-, and [pred]-features). As a consequence, the specifier of C⁰def, like the specifier of T⁰def, may serve only as an intermediate but not as a final landing site of movement.

Headless relative clauses are investigated in the Zapotec variety spoken in San Pedro Mixtepec, Oaxaca, Mexico. Two related constructions are briefly introduced as well: wh- interrogative clauses and headed relative clauses. San Pedro Mixtepec Zapotec is shown to possess the three main kinds of free relative clauses that are attested crosslinguistically, each of which is distinguished by syntactic, morphological, and semantic properties: maximal, existential, and free-choice free relative clauses. The language also has two other kinds of headless relative clauses: light-headed relative clauses introduced by pronominals, and headless relative clauses introduced by relative pronouns.

The introduction outlines the scope of the volume and defines smuggling as involving two steps (Step A is a pied-piping, and Step B is extraction). It is pointed out that in terms of the order of the steps involved, smuggling is the reverse of remnant movement. The possible range of smuggling operations is discussed, including a case of smuggling in the domain of A′ movement from Italian. Lastly, the contributions to the volume are summarized.

/The properties of the English can’t seem construction call for a syntactic resolution of the syntax-semantics mismatch it exhibits. This chapter shows the can’t seem order must be derived from a [seem to [ . . . not can VP ] ] structure. Insights into the derivation come from verb clusters in Germanic OV languages, with complex verb formation and clustering verbs like can and seem playing a central role. Together with infinitival to, dative to, and downward entailing elements, these are instrumental in creating remnant constituents, triggering pied-piping and smuggling a remnant constituent up into the structure, until each element can reach its final landing site. Restrictions fall out from the particular sequence of merge which must hold for convergence, and from the role each element must play. The English derivation in turn sheds light on a potential syntactic resolution of a syntax-phonology mismatch with “displaced” zu in German verbal clusters.

Chapter 5 extends the fragment to handle wh‐question formation by introducing the type‐shifter FRONT. This type‐shifter captures the systematic syntactic relationship between in‐situ wh‐questions and fronted wh‐questions. Flexibility in the timing of the application of the type‐shifter accounts for pied piping. Because the type‐shifter adjusts the syntax of a wh‐question without affecting its semantics, the net effect is that the denotation of a fronted wh‐question will be identical to the denotation of the corresponding in‐situ wh‐question. Fronted wh‐phrases combine with question bodies that are clauses missing a DP argument somewhere inside of them. The position of the missing DP is occupied by a gap, where gaps are identity categories, both syntactically and semantically. The account of multiple wh predicts superiority effects through the same evaluation‐order mechanism as the explanation for weak crossover.

The chapter considers possible diachronic and acquisitional connections between the PP-ergative type and the DP-ergative type. As emphasized throughout this book, each type comes with a set of correlated properties, among which the inaccessibility to A-bar movement is just one of the manifestations of the PP-ergative type. If the P head in the PP ergative expression is no longer visible, a given PP-ergative language can embark on the path to develop properties of a DP-ergative language. These properties may not all develop at the same, and potential triggers of change may also vary, including the reanalysis of a PP as a DP or the relaxation of language-specific restrictions on P-stranding or pied-piping.