How different is Wittgenstein's Language game from Contextuality?

Question

Contextuality

Language game

Per Wittgenstein:

1. Words are missing essence
2. Meaning is use

The way I understood it, a word's meaning is decided by the language game it is part of. This is either identical to or is too similar to contextuality where word meanings are determined by context e.g. the word "table" means a piece of furniture in "this table is made of wood" and a particular arrangement of data in "use the table to plot a graph"

If language games is identical to contextuality then Wittgenstein hasn't said anything new; we always knew words vary in meaning with context.

If language games is only similar to contextuality, there's got to be a subtle difference between the two. I however am unable to put a finger on what the "subtle difference" is.

Question: Is Wittgenstein's language merely a restatement of an already well-known fact, contextuality?

Answer 1

The principle of contextuality is the claim that a word only has a (determinate) meaning in the context of a sentence (in particular: a proposition). This claim can be traced back to Frege and others in the 19th century. (See for instance: Compositionality: its historic context, T.M.V. Janssen, 2012

Wilhelm Wundt, the neurophysiologist and empirical psychologist (quoted by Janssen) formulated it as:

The real thinking consists of thoughts, and separated from a thought in which it enters a concept has no existence. The same holds for an isolated word in ordinary language when used as a sign for a concept; it has reality only in the context of a sentence. (Wundt, Logik. Eine Untersuchung der Principien der Erkenntnis und der Methoden wissenschafliche Forschung, 1883)

A related, but different, and apparently contradictory claim is the principle of compositionality which is the claim that the meaning of a sentence is composed of the meanings of the individual words. Frege seems to have been dead-set against that because it can threaten "psychologism" (the reduction of mathematics to psychological operations or mere mental "concepts") which seems to make the objectivity and certainty of mathematics inexplicable.

Wittgenstein was strongly influenced by Frege, especially in his early work , but some influence persisted in his later work. The slogan "meaning is use" is a somewhat crude, popularized summary of Wittgenstein's core ideas in the Philosophical Investigations. In Philosophical Investigation 43 he wrote

For a large class of cases of the employment of the word ‘meaning’ —though not for all — this word can be explained in this way: the meaning of a word is its use in the language.

In the Philosophical Investigations this idea is closely connected with his "language games"; it's mainly developed in the context of a pretty far-reaching critique of essentialist (platonist, realist) and psychologist ideas about how words (especially names) and phrases acquire meaning. The critique also implies a thorough repudation of some of his own earlier ideas. For instance in PI 46 he writes:

What lies behind the idea that names really signify simples? (goes on to quote Plato, a quote that ends with) "... For the essence of speech is the composition of names." Both Russell's 'individuals' and my 'objects' (Tractatus Logico-Philosophicus) were such primary elements.

The term "language game" is primarily used in the PI to refer to certain artificial toy "languages" - languages that are much simpler than full-blown everyday language, and that thereby may make it easier to get a handle on the question how meaning is acquired or how meaning can change. Witgenstein uses these "languages games" as a conceptual tool, as thought experiments, to investigate those tricky questions.

The suggestion then is to use game playing as metaphor of how language is generally used (how meaning can be acquired/established/changed/explained/justified). In the form of a statement you could summarize this as: "Words (or phrases or sentences) acquire meaning by their embedding in social game-like activities."

The way I interpret this is that this is a broader, more general, and vaguer claim than contextualism. The 'ultimate' context for language, you might say, is "forms of life", in other words, certain social customs, behavioral patterns -- but that is an extension of the idea of contextuality. Wittgenstein's "language game" idea can be seen as compatible with both the narrow principle of contextuality and the principle of compositionality.

Wittgenstein's ideas about "meaning as use" and language as family of language games get a more precise, operational, empirical content in computational linguistics. The basic hypothesis (basic for the development of word-embeddings and LLMs) is the distributional hypothesis:

Linguistic items (of whatever kind) with similar (syntagmatic or paradigmatic) probability distributions, have a similar (syntactical) role or (semantic) meaning.

As we've seen, the resulting models have certain peculiar weaknesses, but they are also extremely powerful. The dialogue systems that are built on top of them essentially also parse and generate language in a game-like environment (for instance, there is very strict turn-taking; the human asks a question; the system generates a response). And the rules of the game are - as in Wittgenstein's toy language games - never completely fixed or determined (apart from deep rules that are not accessible to the surface system).

"Meaning" in these systems is in principle totally undetermined. That is, at any given moment, in actual use, relative to a particular system, relative to particular tasks, it may be determined. But a single word or phrase (or a sequence of letters) can not meaningfully be assigned a "meaning" outside of the system and the tasks that the system is trained on or expected to satisfy. So, like (or even more than) Wittgenstein's language games (and like William James' ideas of language) the implied philosophy of meaning is thoroughly pragmatic. "Context" -- again similar to considering a language game as "context" -- also has a much wider meaning than in the 19th century discussions (it's the context of a large corpus of utterances; both the horizontal/temporal/sequential context and the paradigmatic context; plus the context of different tasks, different modes of deployment, the goals of the community that encounters/uses/produces the system).

Answer 2

Mudskipper's answer is excellent, and I'll respond in a briefer fashion, to point you in a direction of work that might be of interest to you.

Wittgenstein's notion of the Sprachspiel cuts against his earlier philosophical work in Tractatus and his thinking defending what some call logical atomism. From WP:

Its principal exponent was the British philosopher Bertrand Russell. It is also widely held that the early works[a] of his Austrian-born pupil and colleague, Ludwig Wittgenstein, defend a version of logical atomism, though he went on to reject it in his later Philosophical Investigations.

For a more technical and deeper analysis than the WP article, see both the SEP's Russell's Logical Atomism and Wittgenstein's Logical Atomism.

The point of logical atomism, as best as I understand it, is to embrace the Principle of Compositionality and reduce propositions, the abstract notion of meaning bearers which form complete thoughts to analysis of relations expressed through syntax which bottomed out in terms. From there, if one can claim that the terms, most of which represent categories (SEP), can be grasped, then the combination of the terms and the logical structure which governed their relationship provided a complete meaning of the sentence. Propositional calculus in this way makes way for the more useful predicate calculus that Frege was instrumental in putting forth.

By the time of Philosophical Investigations, it became clear to LW that there was more involved in communication, and that in some regard not just the linguistic context, but the context of usage from the perspective of agent was involved in determining meaning. LW was therefore recognizing what modern linguistics call pragmatics. From the SEP:

The words ‘yes,’ ‘perhaps,’ and ‘no’ each has a perfectly identifiable meaning, known by every speaker of English (including not very competent ones). However, as those lines illustrate, it is possible for different speakers in different circumstances to mean different things using those words.

So, under this view of language, it's not enough to know the atomics of the proposition. Words themselves have meanings, not explicitly carried by the sign, but implicitly assigned by those who utter and interpret. Here, the order is inverted. The syntax doesn't function to compose the terms into units of meaning, rather, the word as meaning bearer now gets assigned by its role in the proposition. Or the wider passage of text, or as LW recognized: the meaning of words can actually be determined by the intention of the utterer.

Thus, what LW was drawing attention to was the intention of the speaker as providing knowledge about the language itself. The meaning of the words have to be inferred by the hearer by assessing the intent of the speaker. That's a far cry from earlier theories of language that build up meaning of proposition from terms and relationships. It fundamentally requires that one looks at the utterance and assesses what relevance the utterance has to the utterer.

In modern pragmatic analysis, informed by cognitive science, this thesis of LW's is taken to an extreme by the works of Dan Sperber and Deirdre Wilson who put forth relevance theory. From WP:

The theory takes its name from the principle that "every utterance conveys the information that it is relevant enough for it to be worth the addressee's effort to process it", that is, if I say something to you, you can safely assume that I believe that the conveyed information is worthwhile your effort to listen to and comprehend it; and also that it is "the most relevant one compatible with the communicator's abilities and preferences", that is, I tried to make the utterance as easy to understand as possible, given its information content and my communicative skills.

Thus, Wittgenstein with his notion of Sprachspiel set in motion the development of pragmatics and with it, changes in philosophy of language put developed further by Grice, Austin, Strawson, and others. This can be understood as the movement of understanding semantics to be conveyed by words from the externalism of meaning advocated by Frege, who was fervently an anti-psychologist to an internalized notion of meaning which is highly contextualized (no big surprise to a student of Japanese), a position supported by modern scientific realism.