What is truth? Paul Horwich advocates the controversial theory of minimalism, that is that the nature of truth is entirely captured in the trivial fact that each proposition specifies its own condition for being true, and that truth is therefore an entirely mundane and unpuzzling concept.
What this book does well, it does very well. It is clear, accessible, and concise. Horwich's efforts to demonstrate the utility and appeal of truth Deflationism (He calls it 'Minimalism' but that name was already taken.) are often effective. When this book tries to be a cross between a sketch and a manifesto, it does very well. A special mention should be given to the insightful (if very quick) criticism of Putnam's Pragmatism. There are some issues, however.
First, his response to worries about propositions mainly miss the mark. He assumes that the rejection of propositions must be based on a dogmatic Physicalism, but this is not necessarily so. All sorts of epistemological worries, for example, could be raised. Horwich is either unaware or unimpressed by these objections.
Second, Horwich's insistence that a theory of truth must be given a priori required greater support. If a theory of truth is part of our study of semantics as a branch of empirical linguistics - something that Horwich wishes to argue against, not assume to be false - then there is good reason to think that a theory of truth is given a posteriori like any other linguistic theory. This is especially pertinent because it is unclear how Horwich's defence of propositions can be sustained without an appeal to methodological naturalism, and such an appeal depends on the empirical success of theories utilising propositions. Hence, his theory has only the appearance of being a priori. Horwich must allow for theories of truth to be a posteriori.
Third, Horwich thinks that what he says for truth can easily be moved over to reference. While criticism of inflationary theories of reference can be given (for example, by Field), Horwich does not prosecute that case. It is not obvious to me how Horwich can explain how the name 'Danny DeVito' comes to pick out the flesh-and-blood person that is Danny DeVito. It is less than obvious that "'Danny DeVito' refers to Danny DeVito" is all that must be said on the matter. To be more technical, I don't see how Horwich's view can't be catastrophically Skolemised Putnam-Style.
I have my criticisms, but all in all it is a very well executed book that is good at what it aims to do.