blogpost: The Second Syntax

 Posted by on 11/24/2018  Add comments
Nov 242018
 

DISCLAIMER: I don’t think the following thoughts are particularly “revolutionary” – i.e., I imagine other people have thought (and maybe said) the same things before. But for what it’s worth, I’m a syntactician who has been in linguistics for years and I’ve only recently come to realize these things; so perhaps there are others like me out there who might still benefit from reading this.

A while ago, on facebook, I was talking to Heidi Harley about my unease with proposals that resort to statements like “XP undergoes syntactic movement because it is of the wrong semantic type to compose in situ.” I mentioned I felt that there was something intuitively wrong with this idea, because when a movement requirement is not obeyed, the feeling I get is not that of coercion, i.e., trying to interpret something in the wrong semantic context. It is a crisply different feeling, of ill-formedness and, sometimes, outright word-saladness.

The original examples of coercion I came up with in the discussion involved things like the “universal grinder” (a bit of banana) and the “universal packager” (three waters). Seth Cable then corrected me that these examples had nothing to do with semantic type; banana and water are both predicates of individuals and, thus, equivalent as far as semantic types go. Some further back-and-forth followed and, eventually, Seth explained that “type-clash results in ungrammaticality, not a feeling of coercion.”

I think we can come up with examples that aren’t susceptible to the pitfall of the banana/water cases. For example, trying to interpret an individual as a verb, as in, say, He tried to Chomsky me to death – which, again, does not result in the same kind of sharp unacceptability that syntactic violations give rise to, and instead gives rise to something like a feeling of coercion. This, even though proper names and verbs do not share the same semantic type.

Even if such examples didn’t exist, though, I think there would still be something very wrong with the idea that type-clash results in ungrammaticality. Here’s why. If type theory is a purely formal mechanism for determining which constituents can and cannot stand in a sisterhood relation – in this case, at LF – it should worry us (a lot!) that we, um, already have a mechanism for that. It’s called “syntax” and we already have a theory of it.

This is worse than a mere duplication problem (which would already be bad enough, on its own). If you examine type theory qua theory of syntax, it amounts to a version of Categorial Grammar. And here’s the thing: Categorial Grammar is a really bad theory of syntax! (The reasons are manifold, and I really, really don’t feel like turning this post into a “what’s wrong with Categorial Grammar” post. That sounds tedious and boring. Suffice to say that it cannot handle but the very simplest case- and agreement-patterns.) Don’t you think that if we lived in a world where something like type theory underpinned constraints on syntactic sisterhood, the opposite would be the case? Wouldn’t we expect that Categorial Grammar would be a good (if not the best) theory of syntax?

Essentially, then, any theory that subscribes to “type-clash results in ungrammaticality” (and its offspring, “XP undergoes syntactic movement because it is of the wrong semantic type to compose in situ”) is a theory where there are two syntaxes running in parallel: one, the actual thing people like me would call “syntax”; the other, a syntax based on what is manifestly the wrong theory of natural language syntax (Categorial Grammar). This shouldn’t be anyone’s idea of a good overall theory of grammar.

Subscribe
Notify of
guest

26 Comments
Inline Feedbacks
View all comments
Jason Merchant
1 year ago

Type mismatch leads to noncomposability, which should lead to ill-formedness, as on theories of semantic well-formedness like Ladusaw, W. (1983) Logical form and conditions on grammaticality. (Plug: see http://home.uchicago.edu/~giannaki/pubs/cls.giannakidou.pdf for discussion.) It should never lead, as you correctly note, to movement per se. Movement of phrases should be syntactically motivated, not to “repair” type-mismatches. We sometimes speak this way out of a certain familiarity, but strictly speaking in the cases at hand, movement is merely… Read more »

Seth
Seth
1 year ago

What Jason said. I apologize if I said “ungrammaticality” back then. What I intended (if I did say that) is what Jason writes. Type mis-matches mean the semantics simply cannot interpret the structure, which leads to an intuition of ‘ill-formedness’. This is covered pretty well in the Heim & Kratzer (1998) textbook, as well as (I believe) the architectural nuances of “type-driven movement”. That is, the need for interpretation isn’t ever being access by the… Read more »

Seth
Seth
1 year ago

Admittedly, though, it’s very difficult to think up cases where (i) there is a predicted semantic type mismatch in some surface-form, (ii) there is not *also* a syntactic subcategorization mismatch (or violation of the functional hierarchy), and (iii) the result is intuitively ill-formed, and not ‘rescued’ somehow by null morphology (i.e., zero-derivation or coercion). Perhaps this is the source of your skepticism that type-mismatches lead to ill-formedness? But, note that this is an empirical challenge,… Read more »

Seth
Seth
1 year ago

Oh wait! I thought of one (i.e., a case of semantic type mis-match not saved by coercion, *and* where there’s not an independent *syntactic* account of the ill-formendess). Consider sentences like this: 1) *Dave each came to the party 2) *Dave is all quickly running to the school. Each of these is ill-formed; none is saved by coercion. Moreover, if we assume that “each” and “all” are semantically distributive operators, these sentences would incur a… Read more »

Seth
Seth
Reply to  Omer
1 year ago

Regarding your comment (3): “Type mismatch” is a general label for cases where two meanings cannot be composed together by existing semantic rules. This will hold in cases where one daughter’s denotation is an <e,t> function restricted to non-atomic entities while the other daughter is a type e expressions that is atomic. So, you’re right – the daughter is type e – but it won’t be in the domain of the function, and so the… Read more »

Ethan Poole
1 year ago

Are there not cases where syntactic and semantic combinatorics clash? What I have in mind is the Williams Puzzle (discussed in Partee 1986—a paper which at some level is about this “two syntax” question, though not in that terminology), where DPs can occur in predicative positions, but not all DPs (can) denote the right kind of semantic object (namely an e->t function). I feel that a purely syntactic approach to this puzzle (essentially a CG-like… Read more »

Ethan Poole
Reply to  Omer
1 year ago

I disagree that semantic-type theory is necessarily a CG, because semantic-type theory does not inherently specify *how* things are combined. It is just an inventory of possible semantic expressions, or at least a convenient way of talking about them. This is not a grammar. Nothing about semantic types says that “Mary” (DP; type e) and “women” (NP; type e->t) cannot combine; this is the role of some kind of syntax that says DPs do not… Read more »

Ethan Poole
Reply to  Omer
1 year ago

What you describe is just type theory, but type theory is not CG and CG is not type theory. CG involves a set of typed expressions *and* a set of rules to combine the expressions. Semantic-type theory does not involve a set of combinatoric rules and thus is not “literally an instance of CG”. As for the Williams Puzzle, sure, the “something else” might not be semantics. That remains to be seen. But I should… Read more »

Ethan Poole
Reply to  Omer
1 year ago

Someone with more formal chops than me might need to weigh in here, but, as I understand it, Functional Application and Predicate Modification in H&K do not define a (formal) language and are not a system of combinatorics. That is, they do not define a class of well-formed expressions or even involve an alphabet. They are direct interpretation rules, without an intermediate language like formal logic. Of course, one could define them differently or go… Read more »

Ethan Poole
Reply to  Omer
1 year ago

Sure, that’s fine. The spirit of these systems is function-argument structure, which characterizes a lot of theories (including, imo, Minimalist Grammars). I am only responding to your claim that semantic-type theory is CG, CG is a bad theory of syntax (or not expressive enough), and therefore semantic-type theory should be worrisome. I am pointing out that they are similar, but not the same and that you can tease apart semantic types from any kind of… Read more »

Itai Bassi
Itai Bassi
Reply to  Omer
1 year ago

Hi! I agree with Ethan that type theory is not a version of CG, at least not the way I understand type theory or CG. Type theory is not committed to how the syntax works – it is a(n ingredient in a) theory of semantic interpretation, and as such is compatible with many theories of the syntax that give the input to that interpretation (including, as it happens, CG but also of course “standard” approaches… Read more »

Itai Bassi
Itai Bassi
Reply to  Omer
1 year ago

Ok, I suspect I now see where the disagreements (and agreements) stem from, namely from how strong interpretation one gives to the term “type theory”/”STT”. I quote: “(STT)… is, abstractly, a theory of categories in which whether a pair of categories can stand in a sisterhood relation is determined by the addicity encoded in the individual categories.” Well, I don’t agree with this characterization. I take “STT” to be something narrower: it merely specifies the… Read more »

Itai Bassi
Itai Bassi
Reply to  Omer
1 year ago

“Intuitively, most of the things that STT is employed to do strike me as part of a natural class with the latter, not the former” – I agree with the intuition wholeheartedly, but who said that the semantic component is homogenous w.r.t the feeling of unacceptability/ungrammaticality it gives rise to? In other words, it’s possible that type clashes result in something more severe than “drink humanity” even though their unnaturalness is a result of something… Read more »

Itai Bassi
Itai Bassi
Reply to  Omer
1 year ago

No, not type theory. The “some” sentence gives rise to a logical contradiction.

26
0
Would love your thoughts, please comment.x
()
x