When did Rembrandt get old? Such questions eventually lead us to the problem of vagueness. Williamson traces its history, questions conventional theories and defends the realist view that vagueness is a kind of ignorance.
Williamson's book is a landmark. It is admirably precise and synoptic. It is, in my view, a chore to read Williamson, but careful attention to each chapter of this book will be rewarding in the end.
The book is a bit of a dry slog but it's pretty convincing. I came in pretty confident epistemicism was wrong and exited thinking it's pretty likely that it's true. The main thing that moved me was Williamson's noting that epistemicism violates the law of non-contradiction--suppose you think it's vague whether something is a heap. Then it's both not a heap and not not a heap. But the proposition "it's not a heap" is both true and false. So one need not merely give up on one tenet of classical logic--excluded middle--they need to give up on the law of non-contradiction.
How gloriously stupid is analytic philosophy. Almost 400 pages and lengthy excursuses on fuzzy logic and supervaluationism for Timothy Williamson (the eminent Wykeham Professor of Logic himself) to argue that there is an exact number of grains of sand required to make a heap of sand, but, uh oh, we can never know what that number is. So much unnecessary precision and rigor brought to bear on something so silly and so obviously wrong that I can't help but be delighted by it. Can't believe I tricked them into paying me to have takes on this stuff. Very funny to imagine that this was what Williamson was thinking about when he was doing his "look, I've DONE these things!" rant at Paul Horwich.