Representing Structure in Phonology and Syntax

Marc van Oostendorp, Henk van Riemsdijk, Marc van Oostendorp, Henk van Riemsdijk

• Publisher: De Gruyter Mouton
• Year: 2015
• Print ISBN: 9781501510663
• eBook ISBN: 9781501502255
• Edition: 1
• Topic: Languages & Linguistics
• Subtopic: Phonetics & Phonology
• Index: Languages & Linguistics

Formal grammars by definition need two parts: a theory of computation (or derivation), and a theory of representation. While recent attention in mainstream syntactic and phonological theory has been devoted to the former, the papers in this volume aim to show that the importance of representational details is not diminished by the insights of such theories.

Fußnoten

Doubly-Filled Comp, wh head-movement, and derivational economy

1	The term is kept although it dates back to a time in which S” was thought to dominate two complementizer positions. In the present article, we start out from the later assumption of a functional category C which projects a CP such that C can host a complementizer and SpecCP, a featurally matching specifier.
2	Speakers were instructed that 1 means “I could perfectly use this sentence in my own dialect”, whereas 6 means “I could never use this sentence in my own dialect”.
3	For the motivation of KP in German see Bayer, Bader and Meng (2001); for KP in general see Bittner and Hale (1996). As shown in detail by Seiler (2003), the dative is frequently found in southern German dialects to be “strengthened” by a preposition. One can see this preposition as the spell-out of K.
4	Interestingly, close parallels have been found in V2 and suspended V2 in Northern Norwegian dialects; see Vangsnes (2005); Westergaard and Vansgnes (2005), and the discussion in Bayer and Brandner (2008a).
5	Movement of the complementizer has been suggested for independent reasons in the T-to-C movement account of Pesetsky and Torrego (2001). Self-attachment of the verb to its own projection has been explicitly proposed by Platzack (1996), Koeneman (2000; 2002), Bury (2002), Fanselow (2002a), Suranyi (2003), Brandner (2004) and van Craenenbroek (2006). Donati (2006) assumes wh-head-movement for independent reasons. It has been implicitly assumed by many more. For a comprehensive overview and detailed theoretical discussion see Georgi and Müller (2010).
6	Although the present chapter does not focus on general issues of lexicon and morphology design, it should not be overlooked that there is a clear affinity to the program of Nano-Syntax (NS) as envisaged by Starke (2009) and publications quoted there. In NS, syntax projects from single features building morphemes and phrase structure alike. Thus, a lexical item – as defined by phonology – may associate with a syntactic phrase. The possibility of a combination of features which otherwise often distribute in phrase structure over comp (C) and wh (SpecCP) is expected from this perspective.
7	Thanks to Joe Emonds for his suggestions about this part.
8	The argument was placed in the larger question about movement as such. Movement was seen as an “imperfection” in the design of language. Head-movement was seen as movement that falls outside core syntax, essentially a PF-operation. In the aftermath of Chomsky’s argumentation, Müller (2004) went as far as suggesting reanalysis of a classical and so far undisputed case of head movement, namely Germanic V2, as phrasal movement in disguise (remnant VPmovement).
9	I am aware that the space which I can reserve here for the issue of derivational economy is far too small to cope with the problem. See Sternefeld (1997) for detailed discussion.
10	As I point out in Bayer (2006), this division cannot be accidental. It maps rather directly onto the semantic structure of embedded wh-questions for which the partition approach to questions (cf. Groenendijk and Stokhof 1982; Higginbotham 1993, 1997; Lahiri 2002) has argued independently. Wh-questions are like polar (or disjunctive) questions with the difference that they have a gap. John knows whether Bill smiled is true iff John knows that Bill smiled or that Bill did not smile. John knows who smiled is true iff John knows for each individual x (that may be a contextually relevant potential smiler) that x smiled or x did not smile.
11	Simplifying somewhat, I assume here that Pol is a subfeature of wh and can be set to + or –. Notice that in German as in various other languages there is lexical ambiguity between an interrogative and an indefinite reading. If there is no interrogative force in the left clausal periphery, wh-pronouns in situ receive an indefinite interpretation. In this case, βPol would be turned into –Pol and get deleted from the structure. Notice that the system of Georgi and Müller is more complex because it adds to their structure-building (subcategorization/merge) feature [● F ●] also a probe/agree feature [ F ] that may operate asynchronically in the derivation. For reasons of space I will not elaborate here on this aspect of head-reprojection.
12	Of course, it is not a trivial issue to determine at which point of a derivation a feature is deactivated. Nevertheless it should be clear for the core cases of scope taking C becomes irrelevant after it has been merged with TP; Pol becomes irrelevant after it has been merged with CP; Wh becomes irrelevant after it has been merged with PolP and has been subject to “scope freezing” (cf. Baker 1970 and following work).
13	In German, sentences with a V2-complement like (i) are known to hardly create parsing difficulties although there is a firm local ambiguity as shown in (ii). For lucid theoretical discussion see Gorrell (1994; 1995).
14	The clitic particle -n is absent in (26b) as it applies only to questions.
15	Pfalz (1918); Altmann (1984); Bayer (1984); Weiß (1998, 2005); Fuß (2005) among others.
16	The case is reminiscent of wanna-contraction. It has been argued that to cannot cliticize onto want across an intervening trace. This blocks *Who do you wanna die? If this reasoning holds water, cliticization across an empty complementizer should likewise be impossible, contrary to what (28b) shows to be actually the case.
17	Cf. Gutch (1992) for detailed discussion.
18	Alemannic uses n for epenthesis. As Ortmann (1998) shows on the basis of data from the Black Forest area, n is epenthesized for hiatus avoidance in cliticizations to the functional heads C and P. Consider the word-size wh-element wo, [vo:] (‘where’;) versus a comparable open class noun that is part of a genuine wh-phrase:
19	Relevant criticism had already been formulated by Marina Nespor (p.c.) at the time of publication of Bayer (1984). At that early stage of GB-theory, the dilemma could not be resolved.
20	A caveat must be added about PPs. It seems that for many speakers there is the possibility to inflect a simplex wh-item such as was even though it is part of a PP. Dialect speakers write on the internet in their dialect, e.g ‘Nothing forbidden and not what you already have thoughts about’ http://www.flf-book.de/Benutzer/Partybus.240.htm I tend to say that PP is a potential extension of the category in its complement. If the complement is X°, P+X° is also an X°. Evidence for this comes from the copying strategy in wh-scope extension that is possible in various German dialects. Significantly, no copying of genuine XPs i...

Table of contents

• Cover
• Title
• Copyright
• Contents
• Marc van Oostendorp and Henk van Riemsdijk Introduction
• Josef Bayer Doubly-Filled Comp, wh head-movement, and derivational economy
• Norbert Corver Interjections as structured root expressions
• Günther Grewendorf The internal structure of wh-elements and the diversity of wh-movement
• Jan Koster Relative clauses: Parallelism and partial reconstruction
• Alan Langus and Marina Nespor On the nature of word order regularities
• M. Rita Manzini On the substantive primitives of morphosyntax and their parametrization: Northern Italian subject clitics
• Jacques Mehler Language acquisition and the neuroscience of development
• Kuniya Nasukawa Recursion in the lexical structure of morphemes
• Marc van Oostendorp Final devoicing in French
• Markus A. Pöchtrager Binding in phonology
• Ludmila Veselovská and Joseph Emonds The cross-linguistic homes of mood and tense
• Edwin Williams Multidomination and the coherence of anaphors
• Fußnoten
• Language index
• Subject index