I was wondering how has the field of linguistics was changed (altered? untouched?) by Ludwig Wittgenstein's theories in Tractatus and the Philosophical Investigations.

All Wittgenstein's work deals with language as a center. I know he is not specifically speaking of a particular language (I am not sure if it actually needs to be a spoken language as well). But in general, did any of his work have any consequences in the way linguists approach their work?

  • 3
    Long story short -- early W. had some effect on certain formal semanticists, late W. had more effect on cognitive and computational linguists. Cognitive linguists would feel that the language would have to be spoken, btw; spoken language is universal and evolved, but literacy is modern technology, not language, and it's the property of only a minority of speakers of a minority of languages.
    – jlawler
    Jun 17, 2016 at 0:43
  • 2
    @jlawler However, functional linguists like M.A.K. Halliday and Vachek consider writing to be an alternative means of access to language, with its own structures and properties distinct from speech. A similar point was made by Pullum in Linguistics of Punctuation, which I think was influential among programmers/natural language processing people. Sep 3, 2016 at 2:03
  • (continued) But no one disputes that speech is the natural ability and writing is a later, specialized development; and unlike speech in that it must be learned consciously, and so on. It's just that, for some linguists, writing is "an application of the principles of language", like e.g. dancing is an application of the principles of locomotion and rhythm; and, as such, it deserves to be studied with the toolset of linguistics. Sep 3, 2016 at 2:05
  • 1
    @leoboiko The Linguistics of Punctuation seems to be written by Geoff Nunberg, rather than Geoff Pullum. Feb 1, 2017 at 5:35

1 Answer 1


I would highlight Wittgenstein's idea of family resemblances, which served as a basis for a very productive field in semantics (specificaly prototype theory). This basically postulates that words and their meanings do not work in binary categories but are organised in a field-like manner, where at the centre of the semantic field you have the prototypes (the most typical exemplars of something) while towards the edges, you have the less clear exemplars.

E.g.: semantic field of furniture - table is a prototypical example of furniture, a lamp is a solid one too, a TV is not a superb one but still passable, computer not so much, alarm clock is a bad example of furniture, etc.

This goes very far and basically gets rid of almost any defining features that would trivialise the category to a binary choice - e.g. "adoptive mother" is still a kind of a "mother", much more so than for example a "vacuum cleaner" even though the term specifically denies the biological relationship in the family.

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.