There are good books and bad books; but among the books you find good there is the very proper subclass of those books which you find opening a wholly new view on a subject matter for you. These are the books whose reading is like coming to the top of a mountain from which you can see a landscape never seen before. (In some cases you later find out that the land so discovered is partly a fata morgana; but even to find out this is a valuable experience.) The books are listed in the order in which my encounters with them took place.
 
 

Pirmin Stekeler-Weithofer: Grundprobleme der Logik, de Gruyter, Berlin, 1986.

In the Introduction to my Doing Worlds with Words I noted that this book ”awoke me from my dogmatic slumber” – and this is how I really felt reading the book many many years ago. As I also wrote in the introduction, ”before encountering the book, I had seen philosophy and, in particular, logic as a way towards the firmest of truths, thereafter I realized that logic is a mere tool that can be used both to elucidate and to blur, and that philosophy cannot be made sense of in any other way than as the Wittgensteinian struggle with the bewitchment of our reason by our language.”
 
 

Willard Van Orman Quine: Word and Object, MIT Press, Cambridge (Mass.), 1960.

I do not think anybody who has read Quine need to be explained how captivating his philosophical narrative is; and Word and Object is, no doubt, his masterpeice. Quine’s unique style is seductive, his picture of language and its relation to the world is so self-contained and so elegant that it is hard not to fall in love with it. It is however necessary to keep in mind that seductiveness may sometimes be dangerous.
 
 

Richard Rorty: Philosophy and the Mirror of Nature, Princeton University Press, Princeton, 1980.

I am convinced that I am far from the only one engaged in analytic philosophy who sometimes feels that although this species of philosophy came into being to overcome empty speculations and sterile metaphysics, it provides cases of precisely what it was to eradicate. Rorty’s controversial book makes this feeling into a philosophy – the brisk, engaged kind of philosopy by which Rorty brought so much excitement into the camp analytical philosophers. Rorty is great in making you see things from an angle you have never considered before; and whether you agree with all the consequences he proposes or not, this is a true ‘adventure of philosophy’.
 
 

Willard Van Orman Quine: Set Theory and its Logic, Belnap Press, Cambridge (Mass.), 1963 (revised edition).

In Cantor’s mind, set theory originated as almost a natural science: he believed that he was studying what was going on on the number axis in essentially the same way in which natural scientists study what is going on within nature. Later this led to a ‘metaphysical’ approach to set theory, postulating a realm of sets as Platonistic objects, which came to be somewhat implausible when it turned out that many characteristic of such a realm (e.g. the validity of the axiom of choice) must be chosen, by us, more or less arbitrarily. Now Quine, following the idea of Russell, attempts to reconstruct the whole of set theory as a non-metaphysical, ‘instrumental’ enterprise, resulting from various developments of the introduction of ‘TÎ {x | P(x)}’ as a mere shorthand for ‘P(T)’. The resulting picture is liberating: we can get rid of the convoluted ontological commitments which we would have to bear if we stuck to its metaphysically conceived variant. Neeldess to say that the reconstruction is made with the Quinean superior elegance.
 
 

Daniel Dennett: Intentional Stance, MIT Press, Cambridge (Mass.), 1987.

What does it mean ‘to have mind’? Almost nobody is satisfied with the Cartesian notion of soul as simply an entity mysteriously interlinked with the body; however the Rylean way of saying that the soul or the mind is in fact nothing also do not seem to be too attractive. Dennett came with a different proposal: according to him something ‘has a mind’ if it displays the peculiar kind of complexity which makes us ascribe desires, beliefs, etc. to it. Every entity can be seen from the physical stance, i.e. simply as causally operating mechanism; but for some entities, which are physically to complicated, this stance is not really helpful for understanding and making predictions, so we adopt other stances, such as the intentional stance, from which we understand the entity in terms of ‘beliefs’ and ‘desires’ which it ‘has’. And the fact is that it is this stance which works for our fellow humans (but partly also for some other animals, for chess programs etc.) Does it mean that minds exist merely in the eye of the beholder? Well, if it is good for us to take something as having mind, them it is hard to find a reason to say that in fact it has none ...
 
 

Simon Blackburn: Spreading the Word, Clarendon Press, Oxford, 1984.

Every philosopher has a theory of the relationship between language and the world; so when he attepts to expose this problem, he almost inevitably slips into praising those theories which support his one and despising those which are in conflict with it. It is very hard not to take sides in advance – but this is, it seems to me, what Blackburn has managed to do in this book. He does have his own theories, but he is great in not letting them become preconception; and so his book comes to contain a genuinely unbiased analysis of the language-world relationship.
 
 

Stewart Shapiro: Foundations without Foundationalism: A Case for Second-Order Logic, Oxford University Press, Oxford, 1991.

I have been always fascinated by the mathematically intricate phenomenon of higher-order logics, which apparently are in one sense stronger than the good old first-order one, while in another sense being weaker. Fragmentary results on this score can be found in various logical papers and textbooks, but for a long time I was not able to find a detailed, systematic and illuminating exposition of this problem. Shapiro’s book contains precisely what I longed for: a clear and exhausting characterization of the mathematical properties of higher-order logics and their relationship to the first-order one.
 
 

Robert Brandom: Making it Explicit, Harvard University Press, Cambridge (Mass.), 1994.

I have always felt that there are two kinds of theories of language: those which see expressions as primarily labels stuck to some kind of entities, and those which see them (in Wittgenstein’s words) as items of a toolbox. I have laways thought it is the latter which we should adopt to get rid of many philosophical pseudoproblems. I have tried to articulate this distinction and its consequences in various ways, but then I came across Brandom’s papers in which he accounted for this distinction (between representationalism and inferentialism, in his terms) with an acuity I would never reach. I immediately became his fan; and when there appeared his more than eight hundred pages long opus magnum, I expected it to be a smash. It was: Brandom rethinked almost the whole of philosophy from the inferentialist veiwpoint, so dear to my hearth.
 
 

Alberto J. Coffa: The Semantic Tradition from Kant to Carnap, Cambridge University Press, Cambridge, 1991.

Before encountering this book, I had long been in pursuit of a survey of the history of the classical analytic philosophy. Coffa’s book (unfortunately published posthumously) not only reconstructs its development in its quality of an integral process, but provides for a ingenuously penetrating analyses of the thoughts of the protagonists of the movement (Frege, Russell, Carnap, Wittgenstein, Schlick, Neurath, Popper) up to the middle of the thirties. I do not know about any other book of comparable depth on this theme.