Think you need to talk about headedness. (p. 12) [via PaperHive@docloop]

[docloop[bot]]

Annotation imported from PaperHive.

Regarding this part:

In addition to such abstract rules

Laura Michaelis wrote:

The presence of head as the grand mediator between syntax and semantics makes it possible to have these abstruse rules. Because the head will always tell what can or must accompany it on whatever basis (semantics, lexical class, etc.). So heads are doing all the work in such theories. Whereas in CxG constructions are doing a lot of the work of licensing combinations. In fact I suggested (Ivan was upset) that CxG is a headless theory because no *syntactic* heads are needed in a theory where constructions determine what category a phrasal unit is and what its sisters are.

Link to original comment. About docloop.

[docloop[bot]]

Annotation imported from PaperHive.

Stefan Müller wrote:

Saying CxG is headless because something is not needed is like saying Stefan does not have money since he does not need it. And on top of this I think that Stefan needs money and CxG needs heads. This is the point of much of my work on CxG during the last 15 years (not the money part ...). I think linguistic knowledge involves a strong lexical component. This is also what I argue for in this book. I guess we have to start with the question of what a construction is. If constructions are form meaning pairs + conventionalized things (Goldberg, 2016), then lexical items are Constructions. If a lexical item has valence as in SBCG, HPSG and Minimalism, then you have heads in your theory. The X-Bar-Schema is neither abstruse nor wrong. You have the same rules (or call them Constructions) in your SBCG grammars. (1c) is the Head-Complement Phrase, (1b) is the Head-Adjunct Phrase and (1a) is the Specifier-Head-Phrase. You may call these rules or schemata Constructions but this does not make them any better. What is wrong about the x-bar schema is the claim that it is universal (Fanselow & Felix) and the assumption that this is all there is (a variation of this theme can be found in Minimalism). As I discussed in the grammar theory text book, X-bar does not even work for German. Fanselow, Gisbert & Sascha W. Felix. 1987. Sprachtheorie 2. Die Rektions- und Bindungstheorie (UTB für Wissenschaft: Uni-Taschenbücher 1442). Tübingen: A. Francke Verlag GmbH. Müller, Stefan. 2018. Grammatical theory: From Transformational Grammar to constraint-based approaches. 2nd edn. (Textbooks in Language Sciences 1). Berlin: Language Science Press. http://langsci- press.org/catalog/book/195. Sag, Ivan A. 2012. Sign-Based Construction Grammar: An informal synopsis. In Hans C. Boas & Ivan A. Sag (eds.), Sign-Based Construction Grammar (CSLI Lecture Notes 193), 69–202. Stanford, CA: CSLI Publications. Sag, Ivan A., Hans C. Boas & Paul Kay. 2012. Introducing Sign-Based Construction Grammar. In Hans C. Boas & Ivan A. Sag (eds.), Sign-Based Construction Grammar (CSLI Lecture Notes 193), 1–29. Stanford, CA: CSLI Publications.

Link to original comment. About docloop.

[docloop[bot]]

Annotation imported from PaperHive.

Laura Michaelis wrote:

CxG is headless because it does not have to identify syntactic and semantic headedness as syntacticocentric theories do. The former conception of head is 'category determinant'. The latter conception of head is 'a selector', and since we can have more than one daughter that is a selector then there is no unique semantic head in a phrase. Again, the CxG turn is about distinguishing between semantic and syntactic heads--in canonical patterns they coalesce but need not. See my 2004 Cognitive Linguistics paper. The failure to disentangle semantic and syntactic head properties results in stuff like DP theory. The whole rationale for DP theory as I see it is that determiners are selectors. Well, they're semantic heads in the selection sense but this doesn't entail they're syntactic heads. Nothing about my use of ABSTRUSE was intended to imply 'wrong' btw. My point was simply that in CxG we are returning to lots of category-specific rules, and such rules have both syntactic heads and semantic heads, but the two are not necessarily identified. In addition, of course, in CxG we allow for exceptions to what Zwicky calls 'strictly categorial determination'. For example, we allow a NP to have a V head, as in the hybrid gerund 'Fred's reading poetry'. That kind of datum suggests that our phrasal rules can't all have classic syntactic heads.

Link to original comment. About docloop.

[docloop[bot]]

Annotation imported from PaperHive.

Stefan Müller wrote:

But this is a mixed category, in't it? The head of a gerund is both a N and a V at the same time and hence the whole thing can function as an NP. (If I remember the details of Rob Malouf's book right) And DP/NP: We need selection in both ways. The quantifier in Det must be able to incorporate the noun's content but the element in the Det-slot ma also fill a role: (1) Peter's destruction of the car This does not decide the DP/NP issue. The original motivation was parallelism of structures and hence less difficult language acquisition (Abney, 1987). But views on language acquisition and the Poverty of the Stimulus changed. Most other argument for DPs are gone since then as well (Salzmann, 2019). --- - Abney, Steven P. 1987. The English noun phrase in its sentential aspect. Cambridge, MA: MIT dissertation. http://www.vinartus.net/spa/87a.pdf (25 February, 2018). - Müller, Stefan. 2018. Headless in Berlin: Headless structures in Head-Driven Phrase Structure Grammar. In Ulrike Freywald & Horst Simon (eds.), Headedness and/or grammatical anarchy? (Empirically Oriented Theoretical Morphology and Syntax). To appear. Berlin: Language Science Press. https://hpsg.hu- berlin.de/~stefan/Pub/headless.html (23 January, 2019). - Salzmann, Martin. 2018. Revisiting the NP vs. DP debate. In Ulrike Freywald & Horst Simon (eds.), Head- edness and/or grammatical anarchy? (Empirically Oriented Theoretical Morphology and Syntax). To appear. Berlin: Language Science Press.

Link to original comment. About docloop.

[docloop[bot]]

Annotation imported from PaperHive.

Remi van Trijp wrote:

I fully agree with Laura's claim here and in her 2004 paper that in a construction grammar, it is the construction that does a lot of the work that the "head" is doing in projection-based/lexical-licensing theories. In my own implementations, that allows me to e.g. implement a Determiner-construction that combines a determiner with an adjective, which can then behave as an NP even in absence of a "head" noun as in "the good, the bad, and the ugly" :). A question I do share with Stefan is what the status is in SBCG of heads... when I read Ivan Sag's informal synopsis, there is a lot of discussion on Head-Complement constructions, etc. Is that Ivan's view, or is it something that is assumed in SBCG?

Link to original comment. About docloop.

[docloop[bot]]

Annotation imported from PaperHive.

Stefan Müller wrote:

Well, the intro to SBCG by Sag, Boas & Kay argues for a lexical view. They basically discuss my arguments at length. Head-Complement Schemata are usually very abstract in HPSG (including SBCG). Ivan developed a series of subtypes of the Filler-Head Construction/Schema, but this does not affect specifier head and head complement schemata. Laura and Paul Kay have theories of idioms that are elaborations of Ivan's proposals and they are lexical. Ironically, I always thought they are completely wrong and they should go for phrasal analyses but there are some more recent papers that convincingly argue for lexical approaches. --- - Müller, Stefan. 2006. Phrasal or lexical Constructions? Language 82(4). 850–883. https://doi.org/10.1353/lan.2006.0213. - Müller, Stefan. 2010. Persian complex predicates and the limits of inheritance-based analyses. Journal of Linguistics 46(3). 601–655. https://doi.org/10.1017/S0022226709990284. - Sag, Ivan A., Hans C. Boas & Paul Kay. 2012. Introducing Sign-Based Construction Grammar. In Hans C. Boas & Ivan A. Sag (eds.), Sign-Based Construction Grammar (CSLI Lecture Notes 193), 1–29. Stanford, CA: CSLI Publications.

Link to original comment. About docloop.