Over two episodes, we discussed Ludwig Wittgenstein's Philosophical Investigations,Part I, sections 1-133 and 191-360. Here's a version from the web. The full crew was present along with Philosophy Bro for episode 55, and that group minus Seth (who went to Portugal) was there for #56.
The Investigations was published posthumously in 1953; book one was originally ready for publication in 1946 (but Wittgenstein pulled it). The book reverses many of the positions laid out in his earlier work, The Tractatus Logico-Philosophicus, where he conceived of language as providing a picture of reality, where in basic cases, a word directly stands for a thing we experience, and a sentence expresses relations between those things. On that conception, to understand that particular sentence fully, you only need to see which things in the world the words hook to and what relation is being asserted.
Wittgenstein argues, instead, that the meaning of a word can best be understood by its role in the actions with which it is associated. So "red" is not a raw sense datum with a word attached to it, but a property of things we deal with every day, helping us to communicate about them. A philosopher might ask whether my red is different from your red, but so long as this supposed possible difference has no effect on our practices (i.e. we both call the same things red), then it's a confused question. Wittgenstein thinks that all philosophical problems are a result of taking words out of the context in which they are established and understood in daily life and being confused by that.
So (though this is not an example he uses), "good" has a definite role in certain activities (he calls them "language games"): I communicate my approval of something to you by calling it good, and you do likewise. If we ask what good is in itself, apart from anyone's approval of something, then we've been confused by language. If we start looking at difficult moral questions far outside the range of situations for which the term was developed, then there's no reason we should expect to agree on whether the word "good" applies there or not.
Similar to the "red" case, Wittgenstein asks about the difference between pain and pain behavior. He's not saying that we don't have these private experiences, but the only reason we know to call them "pain" is that they manifest themselves in pain behaviors like crying out. We don't understand that someone else is in pain by looking at our inner experience and trying to generalize this to other people; rather we learn that the word "pain" is associated with those signs, and in that way learn to describe ourselves as in pain, even when we aren't exhibiting those behaviors. Generalizing to all mental phenomena, Wittgenstein famously denies that we could have a private language wherein we name the inner contents of our consciousness. Language is inherently public, and characterized by the fact that we can get some expression wrong and be corrected on it. If we're naming something essentially private, then how can we ensure that we have the same thing in mind every time we use that word? Linguistic meaning, for Wittgenstein, is not about "having something in mind" at all, but again, about a role in some language game.
In addition to the private language argument and his behaviorist tendencies, we got into a few tricky questions:
For instance, if language is a bunch of "language games" all stacked on each other, how do these games interact? Do we all have to be playing the same game?
Wittgenstein explicitly says that most of these games don't have firm rules--definitely not rules that cover every application (which would be impossible). So what does it mean then to play the game appropriately if this doesn't mean referencing some rule set in your head?
Wittgenstein argues that Plato and others were wrong to ask for strict definitions for words representing natural kinds: there's no reason that there should be necessary and sufficient conditions expressible for being, say, a bird, much less for words like "justice" or "knowledge" or "truth." Typically, the alternative attributed to Wittgenstein is to say that there are paradigm (obvious) cases (i.e. whatever kind of bird you likely think of when that concept occurs to you) and borderline ones (penguins, emus, things you might not even be sure if it is a bird, or pictures of birds or robot birds, etc.). However, is there really a "border" at all between such concepts?
We also spent some time relating this back to the Gilbert Ryle section of our Philosophy of Mind episode. Ryle follows in Wittgenstein's footsteps by saying that "mind" and "body" just aren't words in the same category; they're not substances whose relation needs to be determined, but each concept is used in the context of a certain set of experiences and social practices.