The Possibility of Language: Internal Tensions in Wittgenstein's Tractatus (review)

The Possibility of Language is a difficult, painstakingly detailed interpretation and evaluation of central doctrines of the Tractatus. It is not easy reading, but most readers who soldier through it will find their prospects of coming to grips with the Tractatus significantly improved.

Cerezo constrains her reading as tightly as she can by what she finds good textual and historical reasons to believe the early Wittgenstein was thinking about. She acknowledges and elaborates on what she takes to be serious internal tensions rather than forcing the Tractatus to make sense where she cannot find any. To avoid anachronism, she scrupulously refrains from trying to resolve difficulties by bringing the Tractatus into line with contemporary views. In keeping with her focus on Wittgenstein's text, she does not attempt to survey the extensive literature on the Tractatus. Instead, she limits herself to brief comments on selected bits of secondary literature to bring out and defend distinctive features of her interpretationsm, and to acknowledge the sources from which she draws.

She emphasizes the extent to which the Tractatus should be read as addressing philosophical concerns Wittgenstein inherited from Frege and Russell. As she points out, Wittgenstein's treatment of them is informed by methodological ideas from other sources, the most important of which is Kant. The Tractatus is Kantian in its concern to determine a limit to what can be thought or intelligibly spoken about, and to mark off questions which philosophers cannot fruitfully investigate. A further Kantian element is Wittgenstein's use of what amount to transcendental inferences to conclusions about ontological and logical preconditions of fact-stating discourse, thereby grounding what Frege and Russell thought were its fundamental characteristics. Among these are our ability to talk and think about things that do not exist; the meaningfulness of false assertions; and the possibility of using different signs to make claims with identical truth conditions. Cerezo's discussion shows how many of the problems Wittgenstein considers arise from Frege's and Russell's accounts of these features of ordinary and technical languages.

The Tractatus also addresses syntactical and semantic questions whose application to ordinary and technical languages (other than logical notations) is less clear. One such question is how the "logical constants" ('¬,' '&,' 'v,' '⊃,' etc.) contribute to the meanings of utterances in which they occur. It is relatively easy to see how this question, and Wittgenstein's answer to it, might apply to the logical notation of Whitehead's and Russell's Principia. But its relevance to 'or,' 'and,' 'if . . . then,' and other connectives in ordinary and scientific languages depends on some of the most difficult ideas Cerezo examines.

Cerezo's book is somewhat less historical in other respects. For example, although it tells us of Wittgenstein's concerns with Frege and Russell, it provides next to no historical narrative about how they arose in the ways we find them in the Tractatus. Furthermore, Cerezo neglects some philosophically important parts of the historical background, including connections (discussed by Susan Sterrett in her 2005 book, Wittgenstein Flies a Kite) between Wittgenstein's understanding of experimental models in aeronautical engineering and the development of the picture theory of language. At the same time, Cerezo is more historically oriented than other commentators (such as James Conant), who use Tractarian remarks about the nonsensicality of philosophical utterances (its own included) to bring the Tractatus into accord with their own understanding of Wittgenstein's later philosophy.

Tractarian communication is accomplished by using signs to assert or deny that the world is, or is not, as it is represented to be by various combinations of elementary propositions. Each elementary sentence is a concatenation of names which refer to ontologically basic simple objects. The fact that they are related as they are pictures a unique concatenation of the objects to which they refer, and asserts that those objects actually are so concatenated. An elementary proposition is true if the objects actually stand to one another as it...