Have very many linguists studied the structure of sentence fragments used in informal conversation in a given language?

In the course of informal conversation, we English speakers use lots of sentence fragments to answer questions (e.g. (It's) "In the barn," (It's) "The roses." (You can do/have/etc.) "Whatever you want." We also use fragments for other purposes, as in this fragment: (Look out,) "The wolf!" (is back). I imagine that the same is true in most languages.

However, some sentence fragments are clearly not permitted. For example, we English speakers can't say "Take the." (peas) for example.

Since linguists professional linguists can document the types of structures that are and aren't permitted in a language, I wonder which syntacticians have studied sentence fragments in informal conversation, how much work has been done in this area.

  • I believe a lot of work has been done in this area, because syntacticians quickly realized that fragments could pose problems. Can't think of any study in particular at the moment; but have you consulted the Cambridge Grammar of the English Language yet? I know we have studied fragments in my field, the classics.
    – Cerberus
    Apr 3, 2012 at 4:19

2 Answers 2


Of course! The usual term is a (sentence) fragment, some started using the term "nonsententials". What language(s) are you interested in? What syntactic theory?

Something to start with, based on English, purely descriptive. Have a look at section 14.3 The constructional principles of spoken grammar in the monumental "Longman Grammar of Spoken and Written English", esp. 14.3.3. Syntactic non-clausal units and 14.3.4 Ellipsis in clausal units. Or "From utterance to discourse" in the "Cambridge Grammar of English" by Carter and McCarthy.

From a cross-linguistic point of view, see The Syntax of Nonsententials: Multidisciplinary perspectives, eds. Progovac, Ljiljana, Kate Paesani, Eugenia Casielles, and Ellen Barton. 2006. Amsterdam: John Benjamins.


Randy Thrasher has. Details here

  • This isn't paper, duplication of citations is ok :) Besides, I find this a more interesting question than the one pointed to. Only one working on fragments? Tsk. Should think they would be useful for discovering islands and constituents and similar.
    – kaleissin
    Apr 3, 2012 at 22:16
  • It seems to operate almost entirely on a phonological level; if the speaker can predict what words would occur, they can be omitted, until a word appears that is "resistant", i.e, unpredictable and therefore undeletable in conversation. Plus, this was 1974, at the height of the Linguistics Wars, and this was Generative Semantics. As some of the posts linked to indicate, this is still terra incognita for Generative theories.
    – jlawler
    Apr 3, 2012 at 22:20
  • .. might there be enough anecdotes for a "Linguistics Wars the second" soon? Not that we can talk about it, "these are not the droids you are looking for".
    – kaleissin
    Apr 3, 2012 at 22:56
  • @jlawler Do you mind rielaborating here? This answer looks a bit poor.
    – Alenanno
    Jul 25, 2012 at 13:39

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.