One word, one context, one speaker: a narrower game rather than a better definition.
Language game
A rule-governed practice in which words get their meaning from what is done with them.
- Theme
- Logic, language and interpretation
- Tradition
- Analytic
- Term
- Sprachspiel
- Term fixed
- 1953
In plain terms
Ask what a word means and the natural move is to look for the thing it stands for. Wittgenstein's proposal is to look instead at what people do with it.
On that view a word is more like a tool than a label. Water does different work as a request, a warning, a chemical claim and an answer to a question, and none of those is the real meaning with the others as variations. There is a practice, the word has a job in it, and that is all there is.
The word game is chosen deliberately. Board games, ball games, card games and children's games share no single feature - some are competitive, some have winners, some involve skill - and yet the concept works perfectly. Language is like that too, which is why looking for the essence of meaning was never going to succeed.
As a philosopher would put it
A language game is a linguistic practice together with the activity it is woven into. Wittgenstein's builders, exchanging four words on a site, are the opening example: the words have meaning because of the roles they play in a shared task, not because they name objects.
The argument is aimed at the picture theory his own earlier work proposed, and behind it at Augustine's account of learning language by ostension - which he quotes at the start of the Investigations precisely because it treats every word as a name.
Two consequences follow. Meaning as use makes the private-language argument possible: a word whose application only one person could check has no distinction between correct and seeming correct, so it has no use and no meaning. And forms of life mark the limit of justification - explanations of a practice run out, and what is left is that this is what we do, which is where Wittgenstein says the spade turns.
Where you meet it
Standard aviation phraseology reserves the word takeoff for one purpose only. Controllers say cleared for takeoff when granting the clearance and never at any other time; every other reference uses departure. Pilots do not use the word at all except to read back a clearance.
The restriction was introduced after the 1977 collision at Tenerife, in which two aircraft were destroyed on the runway and 583 people died. Among the contributing factors identified was a transmission from one cockpit stating that they were at takeoff - a phrase the tower and the other crew did not read as an announcement that the aircraft was rolling.
What the reform recognised is that the meaning of the phrase was not fixed by what it described. It was fixed by its role in a practice, and the practice had no agreed role for that use. The remedy was not a better definition but a narrower game: one context, one speaker, one meaning, enforced by procedure and drilled until automatic.
What is actually disputed
Whether meaning as use explains enough.
The attraction is that it dissolves problems rather than solving them - the search for what a word really means is shown to rest on a bad picture. The objection is that it is not a theory and does not try to be, so it cannot say why some uses are correct beyond the fact that the community treats them so, and that looks like it makes agreement constitutive of correctness.
Kripke's reading sharpens this into a sceptical paradox about rule-following: nothing in a person's past use determines what counts as going on in the same way, so meaning appears to require a community whose agreement fixes it. Whether Wittgenstein held that, and whether the conclusion is acceptable, has produced one of the largest literatures in the subject and no settled reading of a book that was written to be unsettling.
Commonly confused with
- game in the ordinary sense Language games are not amusements. A cross-examination and a marriage vow are both language games.
- playing with words The term names how words work, not a way of being frivolous with them.
Defined against
- Reference The naming model is exactly what the notion is built against, and the two cannot both be the fundamental account
Philosophers
- Ludwig Wittgenstein Where the notion is developed
- Augustine of Hippo The account of language learning it is written against
- Willard Van Orman Quine Meaning as behavioural and indeterminate
- Jacques Derrida A different attack on the same picture of fixed meaning
- John Dewey Language as practice, reached independently and earlier
Works
- Philosophical Investigations The central text
- Tractatus Logico-Philosophicus The picture theory being rejected, by its own author
- Confessions The passage quoted at the opening of the Investigations
- Experience and Nature Language as a mode of shared activity
- How to Do Things with Words Utterances as moves rather than descriptions
Schools
- Ordinary Language Philosophy Where the approach becomes a method
- Classical Pragmatism Where language had already been treated as practice
- Vienna Circle The programme it abandons
- Post-Structuralism Where fixed meaning is attacked from another direction
Related concepts
Meaning 1892 · Speech act 1955 · Interpretation 1819 · Sign and symbol 1868 · Discourse 1966
Wittgenstein introduces the term in the Blue Book and develops it in the Investigations; game is chosen precisely because games have no single defining feature, which is the point of the analogy.