MODERN philosophy of language is saturated with skeptical doubts. Wittgenstein questions the efficacy of ostensive definition. Searle argues that nothing in the mechanics or algorithms of computation can endow the manipulated symbols with meaning or the computer with understanding. But Searle’s argument does not use the fact that computers are implemented in silicon. Shouldn’t his skepticism transfer to animals, for instance? Nagel argues that no amount of neurophysiology can tell us what it is like to be a bat. Can biochemistry endow the neurological processes of a bat – or a whale, or a chimpanzee – with meaning? But what about other humans? Quine invites us to put ourselves in the position of a field linguist in an alien culture. A rabbit runs out of the bush and a native shouts gavagai. It is consistent with the observed facts that gavagai means rabbit to the native, but any number of other possibilities are consistent with the observed facts. Gavagai might mean running rabbit or good to eat or even temporal slice of a rabbit. (After reading Searle, should we add the possibility of no meaning at all?) Quine concludes that without some preexisting shared system of language we can never know what the native means by gavagai. Quine is willing to follow his argument to its logical conclusion. In principle, the same problem is faced by people in the same culture – by any people who communicate. And he points out that his skepticism about translation is just one facet of a more general skepticism about induction, grounded on the underdetermination of theory by evidence.

Where does the skeptical philosophy of language lead? An influential group of literary critical theorists goes much further than Quine does. They give up not only on intensional meaning but also on truth and denotation, reducing language to the bare existence of text. But if there is no meaning, there is no distinction between symbol and non-symbol, text and non-text. These theorists should quietly go out of business. (And, as of this edition, they have gone out of business – or at least out of fashion.) After looking into the abyss, it is tempting simply to dismiss this skepticism as unproductive, as requiring too much of knowledge and as neglectful of non-demonstrative inference. But these skeptical musings do raise important scientific questions for naturalized epistemology.