In an elementary course on philosophy of language ( at the highschool level) , I try to explain to students the distinction betweeen semantics, syntax and pragmatics.

Referring myself to Carnap/Morris , I say :

  • Semantics deals with meaning ( sense and reference), with the relation linguistic signs have with the world or with thought ( considered objectively).

  • Syntax deals with the relations linguistic signs have with one another in a sentence.

  • Pragmatics deals with the relations holding between signs and the person that uses them.

However, nolens volens , when I want to explain the interest of each branch, I inevitably arrive at :

  • syntax gives the necessary conditions ( though not sufficient) for a sentence to mean anything

  • pragamtics shows that one can produce a meaning that is different from the litteral meaning of a sentence by using this sentence in a particular way in a particular context.

Hence the question : can it be suspected that, in fact, syntax and pragmatics belong to semantics? or are sub-field of semantics? Are not semantic considerations omnipresent in lingusitics?


In generative grammars, as far as possible the syntax is designed to be automatic and meaning-free. Jim McCawley used to distinguish syntax from the cluster of topics involving "meaning" that he called semantics/logic/pragmatics, viz (from the list of research topics on his home page):

  1. Syntax (of English, and when native speakers are available to be exploited, also Japanese, Chinese, Spanish, and Hindi), within the revisionist version of transformational grammar that I operate in (and to which I refuse to give a name - if you feel you need a name for it, go and make one up), which exploits what I regard as the fruitful ideas of transformational grammar (constituency, multiple syntactic strata, the cyclic principle) and chucks out what I regard as counterproductive ideas (the metaphor of a "base" structure, the idea of categories and structures as remaining constant throughout derivations, the fetish for keeping syntax and semantics separate). For detailed exemplification of this approach to syntax, see my The Syntactic Phenomena of English (University of Chicago Press, 1988; 2nd edition 1998).

  2. Semantics/logic/pragmatics (it's impossible to talk in any detail about any of these three fields without getting into the other two, so I don't even try to keep them separate). I teach courses on logic from a linguist's point of view, taking a broad view of the subject matter of logic (logic has suffered from 23 centuries of myopia, which I try to make up for) and giving full weight to linguistic considerations in revising (or replacing) existing systems of logic to maximize their contact with natural language syntax and linguistic semantics. (See my book, Everything that Linguists Have Always Wanted to Know About Logic (but were Ashamed to Ask) , University of Chicago Press, 2nd edition, 1993). I also from time to time teach courses in lexical semantics, tense and aspect, and speech acts (with Erving Goffman sharing top billing with J. L. Austin).

However automatic syntax may be, its functions can be discerned in many cases, and these have to do with processing, which is to say, communication -- and therefore "meaning" (an undefined term to avoid if possible in philosophy, as Wittgenstein pointed out).

For instance, English has dozens of syntactic phenomena like Extraposition and Raising that all operate on different words, with different constraints, and in different constructions, but all have the same effect, of leaving large and complex constituents at the end of the sentence, where they are easier to process in a right-branching language like English.

I suspect you might find McCawley's books good reading for a situation like yours. He's clear and complete and considers all the alternatives. And he deals nicely with philosophical matters, particularly in his logic book.

| improve this answer | |
  • 1
    I endorse your recommendation of McCawley, of course. That doesn't mean I agree with his disdain for the logicians' discovery that syntax and semantics should be divorced. – Greg Lee May 26 '19 at 21:54
  • I think he was just convinced that if there was any purpose to syntax, it was to facilitate semantics and pragmatics. Syntax was what fell out -- or could be made to fall out -- automatically – jlawler May 27 '19 at 1:55
  • 1
    Actually, semantics is what just falls out. Syntax is where it's at. – Greg Lee May 27 '19 at 2:18

To answer your first question, yes, meaning can be avoided when defining syntax (or doing it), and, presumably, also for pragmatics. Without dealing with meaning, following Carnap and Morris, we have a "logical syntax", which is a theory of implication. That is, we study what follows from what. There is plenty to do, there, and this is one popular way of introducing symbolic logic to students -- there is no reference to truth tables (which would make the account semantic).

| improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy