Desktop version

Home arrow History arrow A Minimalist View on the Syntax–Semantics Relationship: Turning the Mind into a Snowflake

Not ugly, but unique

Eliminating pair-merge

A route which may be followed in this situation would thus retain relevant properties of (14) without letting in free availability of replacement operations in narrow syntax. A line which might seem to be in accordance with the explanation of Chomsky (2015b: 12) would capitalize on morphosyntactic properties of the resulting structure:

Conventional treatments of head-raising assume that the raised element is adjoined to its host—pair-merged to it, in our terms. But that is not correct. For example, when V raises to T, a collection of inflectional features, the result is not V affixed to T, but T affixed to V More generally, the conventional theory of head-raising seems to have the story backwards: the host should be affixed to the raised element. (Chomsky 2015b: 12)

Exploiting affixal nature of syntactic objects, while having a noble pedigree within generative theory, should be accompanied by a distinction between strictly syntactic and post-syntactic operations, though, ‘the result’ in the quote above being a result of post-syntactic operations applied to a syntactic object; the strategy suggested in Chomsky (2015b) would thus be not to explain properties of head movement by morphological effects thereof, but rather to use the latter as an insight into properties of the former. Accepting the basic result, i. e. adjunction of v to R as a result of R-to-v head movement does not answer the question about syntactic properties of the objects in question which might be relevant for special characteristics of head movement. A pertinent property seems to be the relationship holding between R and v (and C and T), which might also account at least partly for unclarity of the status of other cases of head raising noted in

Chomsky (2015b): it is the relationship allowing (possibly not requiring, as already observed in section 1.3.6; see also below, section 2.4.3) feature inheritance. Suppose that it is this relationship that keeps LI’s in question susceptible to modification: remaining active in this sense makes them capable not only of transmitting and accepting features, but also of undergoing operations affecting their status in the syntactic structure. A modification of (13)-(14) takes care of this restriction, based on independently needed properties of X and Y:

  • (16) Operations in head adjunction
  • 1. Take an SO = {X, [1]}, where{WP} may = {ZP}, may Ф {ZP} or may be null, and X and Y are able to participate in feature transmission from X to Y;
  • 2. take two LI’s X and Y;
  • 3. self-merge X;
  • 4. merge the output of 3 and Y.


The adjunction structure in (17) makes X—together with its features—invisible for labeling purposes, as required. In spite of appearances, its creation does not involve substitution of a complex syntactic object for an occurrence of X—due to special properties of objects entering into a relationship of possible feature transmission. Furthermore, {{X}, Y} is considered acceptable by the labeling algorithm for the same reason for which adjunction structures generally are able to pass the labeling test, lack of label being in this case the expected behaviour (thus incorporating the ideas of Hornstein (2009); Pietroski and Hornstein (2009) without sacrificing the effects of lack of label for structures created by simple applications of merge predicted by the theory of Chomsky (2013c, 2015b))—the property of objects created by head movement, viz. that ‘raisingof R to v*yields an amalgam with v* adjoined to R, and the affix is invisible to the labeling algorithm. Note that although R cannot label, the amalgam [R-v*] can (Chomsky 2015b: 12), is a property of adjunction structures which are special in that both the adjunct and the host are LI’s (whence ‘amalgam’ in the quote from Chomsky (2015b: 12)), so that the structure as a whole may be treated by the labeling algorithm as one which will enable interpretation of the structure in which they participate in the C-I component.

  • [1] WP}, {Y, ZP
< Prev   CONTENTS   Source   Next >

Related topics