Head Movement in Linguistic Theory

Abstract

In this article, I address the issue of head movement in current linguistic theory. I propose a new view of the nature of heads and head movement that reveals that head movement is totally compliant with the standardly suggested properties of grammar. To do so, I suggest that head movement is not a single syntactic operation, but a combination of two operations: a syntactic one (movement) and a morphological one (m-merger). I then provide independent motivation for m-merger, arguing that it can be attested in environments where no head movement took place.

The Rich Agreement Hypothesis Rehabilitated

“…This becomes an issue when some constituent (e.g., an adverb or negation) intervenes between Arg 0 and v 0 . 22 For those cases, the literature provides two means of ensuring adjacency: (a) by moving the verb in the syntactic component, as has been traditionally proposed (see also Lechner 2004, Matushansky 2006, Roberts 2010 for recent proposals along these lines), or (b) by moving the verb postsyntactically (see, e.g., Chomsky 2000, Boeckx and Stjepanović 2001, Zwart 2001, Platzack 2012). Since our proposal does not hinge on the correctness of either approach and is fully compatible with both, we are not forced to make a principled choice.…”

Argument structure and morphologically underived nouns in Spanish and English

“…One of them has to do with the technical implementation of phrasal spell out. There are different options in the literature to give an account of situations where one exponent spells out a series of heads: head-movement (Matushansky 2006), fusion as a post-syntactic operation (Halle & Marantz 1993;Embick & Noyer 2001), having lexical items correspond to subtrees (Starke 2009;McCawley 1968;Neeleman & Szendröi 2007;Weerman & Evers-Vermeul 2002;Caha 2009) or spanning (Ramchand 2008, Dékány 2012. The first approach cannot be adopted in our analysis, given the assumption that roots lack a grammatical category -which would make the fact that they have to move to check categorial features in the syntax surprising-but deciding between the three remaining approaches clearly requires more research.…”

‘Verb floating’ and VP-ellipsis

“…Given general economy conditions, we may assume that, if a verb does not need to move, it cannot move. Thus, all raising verbs will raise from their base positions if they are to bear the T, +0 and +en affixes, but they will not move if they bear the +ing affix; rather, the verbal head and +ing will combine by m-merger in Morphology (Marantz 1988, Matushansky 2006 The fact that auxiliaries bearing the -ing affix cannot license ellipsis (in a headhead licensing account) was observed by Lobeck (1995) and Johnson (2001). This was dealt with by the stipulation that non-finite auxiliaries (but not infinitival to 17 )…”

“…10 Feature-checking must be done under strict adjacency, which obtains between a verbal head and an affixal head either in the case of a head-head configuration (head-head adjunction after overt movement of the verb to the affix, or a Spec-head relation if we follow Matushansky 2006 in assuming head movement is movement to a specifier), or when there are no intervening non-empty 11 projections between them. In Morphology the unaffixed inflection morphemes are lowered onto the main verb successively, and the one on the outside (i.e.…”

“…Thus the ordering of operations is Copy-Delete-Merge, and Delete always targets the element affected by previous operation. Now consider ellipsis: first X is Copied, then (re)Merged into the landing site for movement, YP; then, Delete applies to the element targeted by the previous operation, which in this case is the landing site for movement, YP, but to the exclusion of the newly created specifier (since this was not part of the element targeted by Merge; note that this requires head movement to specifiers, as in Matushansky 2006). This would derive deletion of the whole complement of the target landing site, i.e.…”