Naming and Necessity

Naming and Necessity is one of the classics of 20th century philosophy. If you haven't read this book, I suggest you drop your current reading and pick this up. Simply put, metaphysics isn't dead (although Kripke is broadly speaking in the tradition of analytic philosophy).

If you have read it, you might want to follow up with some of the work of Scott Soames.

Originally published in 1972, Saul Kripke's Naming and Necessity is a compilation of three lectures given by the author in January 1970 at Princeton University. Aside from transcripts of the lectures (with minor editing), the text includes a brief preface and postscript (or agenda as Kripke calls it) with some helpful points of clarification. Kripke is regarded by many as the pre-eminent philosopher of recent times- while Naming and Necessity is widely viewed as the most significant piece of post-Wittgenstein analytic philosophy.

In the book Kripke discussion of a range of issues and questions that has altered the trajectory of modern philosophy including:

* Accidental and essential properties,
* Theories of reference (direct reference v. descriptivist)
* Epistemic and metaphysical necessity (he poses the possibility of necessary a posteriori truth and contingent a priori truth)

Readers unfamiliar (or rusty) with Kripke may find the pertinent chapters in Scott Soames' excellent Philosophical Analysis in the Twentieth Century, Volume 2: The Age of Meaning helpful in preparing to engage Naming and Necessity. The small text `On Kripke' in the Wadsworth series is also useful and even more introductory.

Overall, this is an important work in analytic philosophy that would make a valuable addition to any collection. As with much modern philosophy in the analytic tradition familiarity with the genre and subject matter is a perquisite to fully understanding and appreciating the discussion (that said this book has a nice flow). My comments pertain to the 2005 reprint by Harvard.

Saul Kripke's "Naming and Necessity" is arguably the greatest piece of philosophy of the twentieth century. It has reshaped philosophy of language and metaphysics, and has significantly influenced philosophy of mind and epistemology, among other areas of philosophy.
"Naming and Necessity" is largely a transcription of three lectures that Kripke delivered when he was 29 years old - without lecture notes. The book reflects the informal style of these lectures; it is friendly and engaging, albeit sometimes unclear.
Among other things, Kripke refutes descriptivist theories about the meaning and reference of proper names and natural kind terms, and develops a new account in their place; defends the modal concepts of necessity and possibility, and distinguished between necessity and a prioricity; and argues that there are necessary truths knowable only a posteriori and contingent truths knowable a priori.
Of course, the book has some critics. However, in my opinion, Kripke does not commit some of the errors of which he has been accused in other reviews here.
I think that the book will be very difficult for those unfamiliar with analytic philosophy, especially philosophy of language. I recommend "Naming and Necessity" for upper-level undergraduate and graduate students in philosophy. I also recommend Soames's two-volume "Philosophical Analysis in the Twentieth Century", which culminates with an excellent exposition of "Naming and Necessity" but which is also difficult for beginners.

I really love this book and ought to give it five stars, but I can't make myself do so, because it's just all wrong. Even so, it's a great, great book--there simply is no more engaging book of analytic philosophy.

Kripke's error is basically one that is symptomatic of all modern analytic philosophy...overattachment to modal considerations, trying get them to do the work that only Bayesian analysis can do properly. Bertrand Russell said it best--"There is only one world, the 'real' world." Philosophers should have listened. The meaning of (even non-literal) utterances reduces to the transmission of information, and everything you ever wanted to know about transmission of information can be found in conditional probabilities. The space you're forced to deal with is the space of ways the world might *be* (for all I or you know), not ways the world "might have been."

A misreading of Wittgenstein (Kripke later based an entire book on a misreading of Wittgenstein) is representative: Kripke says that Wittgenstein was wrong to say that you can't say (just W's way of saying that it's non-informative to say) the standard meter stick in Paris is a meter long. According to Kripke, it's not only informative, but "contingent." (There are contingent a priori statements, according to Kripke, which everyone knew to be obvious nonsense before he came along and messed with their heads with a few carefully disguised modal equivocations.) It's *not* informative, obviously. If we accept the romantic story about the standard meter stick, and Kripke does, then all it means to be a meter long is to be the same length as the standard meter stick. There aren't any "other worlds" where the stick's a different length. Anyway even if there were other worlds, the stick wouldn't be there. It's here. (Well, it's in Paris.)

Modal considerations can paint some awfully pretty pictures, and I think Kripke got lost in the abstraction. Anyway you have to read this book one way or the other...hopefully you won't get lost, too. Unfortunately I'm not betting on it; Kripke is dangerously persuasive.

I would happily have hiven this book 4 stars if it hadn't been for the third lecture - it very nearly provoked a 1 star review. Why?

Let's start from the beginning: Kripke makes a very persuasive argument to the effect that we know named objects simply by their names, rather than via a definite description. Thus, the example that one can know of 'Feynman' without knowing anything about him other than that he is a famous physicist, a description so far from being definite that it could apply to any number of people. So, a name is a 'rigid designator', something we use to point out a particular item, but not itself simply shorthand for a description.

This is all good stuff, though it has the mild flaw that such rigid designators cannot achieve any meaning beyond one individual's mind unless they have some association to a description. For example, if you hear me talking about 'Cicero' without giving any description, then you have heard the rigid designator, but you have nothing to attach it to. You need a description, even if it may change over time, and it may be inadequate, to get some idea of what the rigid designator refers to.

Unfortunately, in the third lecture everything falls apart. First Kripke attempts to argue that 'heat' and 'molecular motion' are necessarily identical. This may be so if you're prepared to attach a description to the rigid designator 'heat', something along the lines of 'the sensation which I feel under certain kinds of sensory stimulation, which can be shown by experiment to correspond with the action of causing mercury to expand and contract'. If you don't have that description, you have two entirely separate concepts: the intuitive sense of 'temperature' derived from sensation, and 'calorific heat', a scientific concept based around mean kinetic energy in ensembles of molecules. The description is what links the two, and so without that description, Kripke's argument collapses. But the description is precisely the thing he refuses to accept. In fact the argument collapses anyway, because in no way is it necessary that calorific heat and the sensation of temperature should be identical.

Now things get worse. In this argument Kripke acknowledges quite happily that 'heat' = 'stimulation of certain nerve fibres'. But then he attempts to argue that while that equation is necessary, the equation of pain with stimulation of other nerve fibres isn't! He commits such open lunacy as the following: '...it would seem that God need only create beings with C-fibers capable of the appropriate type of physical stimulation . . . it would seem, though, that to make the C-fiber stimulation correspond to pain, or tbe felt as pain, God must do something in addition to the mere creation of the C-fiber stimulation; He must let the creatures feel the C-fiber stimulation as pain, and not as a tickle, or as warmth, or as nothing.' But this applies equally well to heat, so why was Kripke happy with that equation?

The problem appears to be a very elementary mistake in modal logic, which I would not have expected of Kripke. Viz, he assumes that any conceivable counter-factual counts as a possible world accessible from this one (e.g. a counter-factual in which I feel pain but my C-fibres are not stimulated). But this is only the case in the strong S5 modal logic. In more realistic modal logics, possible worlds are joined by a complex accessibility relation, such that necessity in a world depends on truth in all worlds accessible from that one, not in all possible worlds. Just because Kripke can imagine such a counter-factual, it does not make it a world accessible from this one.

So, in conclusion: buy it, read it, but don't treat it as holy writ. Personally, I'd stick to Quine.