Firstly here’s a quote from the philosopher Gilbert Ryle in which he talks about Wittgenstein’s general distaste for generalisations:
“…he now [i.e., the late Wittgenstein] avoids any general statement of the nature of philosophy, not because this would be to say the unsayable, but because it would be to say a scholastic and therefore an obscuring thing. In philosophy, generalizations are unclarifications. The nature of philosophy is to be taught by producing concrete specimens of it.” (1951)
So what about Wittgenstein himself? He wrote:
“Our craving for generality has [as one] source … our preoccupation with the method of science. I mean the method of reducing the explanation of natural phenomena to the smallest possible number of primitive natural laws; and, in mathematics, of unifying the treatment of different topics by using a generalization. Philosophers constantly see the method of science before their eyes, and are irresistibly tempted to ask and answer in the way science does. This tendency is the real source of metaphysics, and leads the philosopher into complete darkness. I want to say here that it can never be our job to reduce anything to anything, or to explain anything. Philosophy really is ‘purely descriptive’.” (1933–1935)
When Wittgenstein used the phrase “craving for generality” to express his position towards what he saw to be Western philosophy’s prime vice, he was himself generalising about Western philosophy.
To state that
All traditional Western philosophy generalises about things which can’t be captured by such generalisations.
is to generalise (even if we drop the quantifier “all”). Similarly, the statement that
All traditional philosophy is “concerned with the general rather than the particular”.
is itself concerned with the general (i.e., all Western philosophy) rather than the particular.
If one classes oneself as a Wittgensteinian particularist (even if one doesn’t actually use that term), that statement is a generalisation about those who can be called generalists and also about what can be called generalism. (For that matter, it’s also to implicitly generalise about particularists and particularism.) It’s of no consequence at all if particular Wittgensteinians don’t use these terms (or indeed any ists or isms). They may not use such words; though they’re still tacitly committed to the concepts which these words express.
“All generalizations are unclarifications.”
is a generalisation about (all) generalisations.
In order for Wittgenstein’s position to have a point (or in order for him to demonstrate his own point) there must be (or there needs to be) at least one generalisation that’s not an “unclarification”. Though if there’s at least one such generalisation that isn’t also an unclarification, perhaps it could be (as Wittgensteinian might have claimed) the very generalisation that
“All generalisations are unclarifications.”
If the initial generalisation isn’t taken to be the sole exception to what it claims, then it can be rewritten in this way:
All generalisations — including this one — are unclarifications.
This self-referential statement therefore allows itself to be either an exception to itself (i.e., if it didn’t include the central clause) or to be an actual example of what it claims. Though if it’s not an exception to its own claim, then it too must be an unclarification. If it is an exception to what it claims, then there it may be the one and only one exception. Thus
All generalisations — including the one you’re now reading — are unclarifications.
can be taken as a kind of proof (or just an example) of what it claims (i.e., if it’s actually taken as an unclarification). If the statement about all generalisations implies (or entails) that it too is a generalisation (if about generalisations), then it must also suffer from the same kind of unclarity that’s taken to be a property of all other generalisations.
However, Wittgenstein’s statement about all generalisations is a second-order (or metalinguistic) generalisation. (Compare with Karl Popper’s Falsification Principle and the object-metalanguage distinction of Alfred Tarski.) That is, it’s a generalisation about generalisations: not a generalisation about events or things which aren’t themselves taken to be linguistic generalisations. If all first-order generalisations (i.e., those in the “object language”) are unclarifications, then a second-order generalisation about first-order generalisations can be seen either being free from the property unclarity (i.e., due to its second-order status); or as being itself a victim of such a property (i.e., perhaps it’s not an actual or genuine metalinguistic statement).
Note: Lycan’s Paradox (at least according to Michael Clark) also deals with generalisations. More specifically, it deals with the statement, “Most generalisations are false”. It states that “a minority [of generalisations] must be true”. However, since Wittgenstein’s own meta-generalisation is about generalisations (one which claims they’re all false or unclarifications), it’s even less likely to be true than the first-order generalisations it has as its target.