WAB: "Fragments" | The following contribution is an excerpt from Cora Diamond: "Peter Winch on the Tractatus and the unity of Wittgenstein’s philosophy", in: Alois Pichler and Simo Säätelä (eds.): Wittgenstein: The philosopher and his works, Working Papers from the Wittgenstein Archives at the University of Bergen no. 17, Wittgenstein Archives at the University of Bergen (WAB) 2005, pp. 133-163. Publication on WAB's website with kind permission from the author and the editors (2005.3.8).

Cora Diamond: Peter Winch on the Tractatus and the unity of Wittgenstein’s philosophy

Winch, Malcolm and the unity of Wittgenstein’s philosophy

One of Peter Winch’s most noteworthy contributions to philosophy lies in his writings on Wittgenstein. In the hope of making clearer what he achieved, I shall look at the evolution of his ideas about the unity of Wittgenstein’s thought.

He first expressed these ideas in the Introduction to Studies in the Philosophy of Wittgenstein (1969). He wanted, he said, “to combat the widespread view”, a view which he took to be “disastrously mistaken”, “that we are dealing with two different philosophers: ‘the earlier Wittgenstein’ and ‘the later Wittgenstein’”, and so he subtitled his essay “the Unity of Wittgenstein’s Philosophy” (p. 1).1 Winch believed that the idea of ‘two Wittgensteins’ reflected and grew from misunderstandings of both the Tractatus and the later work. He thought that the causality worked the other way round as well: i.e., that the two-Wittgenstein view led to misreadings of all of Wittgenstein’s work. So he was trying to break the cycle of misunderstandings by challenging both the two-Wittgenstein view and readings of Wittgenstein’s individual works, especially readings of the Tractatus. The essay is successful mainly as a programmatic essay; much that he said then in working out the program was clarified and changed later on.

What indeed happened later was that Winch was immensely stimulated by his discussions with Norman Malcolm, especially during the years Malcolm was Visiting Professor at King’s College. Winch had great respect for Malcolm, but was also very critical of Malcolm’s understanding of Wittgenstein. He once wrote that he thought Malcolm shied away from the radical nature of Wittgenstein’s thinking in the Tractatus and in the later writings, in parallel ways.2 Malcolm was a particularly forthright and steadfast defender of the two-Wittgenstein view; and I think we can find very clearly in Malcolm’s writings the complex dynamic I described: the two-Wittgenstein view drawing on certain misconceptions of early and later Wittgenstein, while those misconceptions themselves are encouraged by the idea of Wittgenstein as two philosophers. Winch was aware of that dynamic before he and Malcolm became colleagues, but the contact with Malcolm greatly sharpened his sense of how it worked, and helped him to revise his ideas about what was wrong with the usual readings of the Tractatus. As will come out in the rest of this essay, Winch’s understanding of Wittgenstein shows also the effect of discussions with another colleague, Rush Rhees.

In 1969, when Winch published that first essay on how Wittgenstein’s philosophy hangs together, the orthodox view was not only that there were ‘the early Wittgenstein’ and ‘the later Wittgenstein’, but also that the latter had dismantled the philosophical theories of the former, and was utterly distant from the former in method, aims, and concerns. That view of Wittgenstein was taken by almost every commentator, but there were two sorts of exception. First there was Rush Rhees, who had in 1966 laid the groundwork for an understanding of Wittgenstein as one philosopher by arguing for the continuity of Wittgenstein’s concern with logic, and specifically for the idea of Philosophical Investigations as a book on philosophy of logic.3 Rhees had also rejected the idea of Wittgenstein as having, in his later work, demolished his earlier system and replaced it by a new one. A representative of a very different kind of exception to the orthodox reading is Erik Stenius, who had argued in 1960 against the existence of deep differences between the picture theory and Wittgenstein’s later views.4 But Stenius’s defence of a one-Wittgenstein view rested on misconceptions about both early and later Wittgenstein, and on failure to grasp the character of the differences between them. He attacked the orthodox view on what was in fact a strong point, namely its insistence on the philosophical importance of Wittgenstein’s later critique of the Tractatus. (Stenius nevertheless deserves recognition for noting that many commentators were simply reading into the Tractatus any view that Wittgenstein criticised later.)

Back then to Winch in 1969: Prior to Winch’s essay, there had been no sustained attack on the established two-Wittgenstein view that had taken seriously the strength of such a reading, namely its recognition of very significant changes in Wittgenstein’s approach, and of deep-going criticisms in the later work of Wittgenstein’s earlier views.

Winch located as a primary continuity in Wittgenstein’s philosophy his concern with the nature of logic. If Wittgenstein is, in his later philosophy, still centrally concerned with the nature of logic, why (we might ask) does he spend so much of his time dealing with so many apparently quite different problems? Winch takes those discussions to belong to Wittgenstein’s new conception of how logic itself has to be treated. So the idea is not that Wittgenstein is turning from an interest in the nature of logic to an interest in quite different sorts of philosophical issue, but rather that the attention to these various topics itself reflects a new idea of how one should approach the philosophy of logic.

Winch puts the point this way: the change here “turns upside down [Wittgenstein’s] view in the Tractatus that, once the central logical problems had been settled, the dissipation of other philosophical difficulties would in principle have been [achieved] at one blow, so that all that would remain to be done would be a sort of mopping-up operation” (1969, p. 2). Winch sees a radical change in Wittgenstein’s understanding of the role of generality in philosophy, of the kind of generality that he had taken to characterise philosophy. There is a totally new idea of the significance that attention to particular cases can have, attention to the problems that can surface in them. Wittgenstein’s later thought thus involves rejecting the point he had made at TLP 3.3421: that the only significance of particular cases in philosophy lies in what they can disclose of what is totally general, as for example the possibility of a certain kind of notation for identity might help us to grasp what all adequate notations have in common, through which they can express what they do. Winch’s point then is that this vital transformation in Wittgenstein’s conception of philosophical method can be seen in the right light only so far as we recognise its tie to the questions about the nature of logic which had been central to him all along. Winch mentions (p. 2n) that P.F. Strawson’s 1967 bibliography of works on philosophical logic includes only the Tractatus, not Philosophical Investigations – as if the latter were not concerned with philosophical logic. Things have changed somewhat since 1969: Michael Dummett, Saul Kripke, and others have given currency to the idea that Wittgenstein’s later work has important implications for issues in philosophical logic. But these philosophers fit, or attempt to fit, Wittgenstein’s ideas into a conception of philosophy which takes for granted the possibility of an entirely general examination of fundamental logical issues, like whether the meaning of words is fixed enough for what we say to have determinate consequences. Within that conception of philosophy there is no room for the idea that Winch was inviting us to take as central in Wittgenstein’s post-Tractatus thought.

Here a comparison with Malcolm suggests itself. In one of Winch’s last pieces of philosophical writing, he discusses again the relation between Wittgenstein’s ideas about logic and his later philosophical methods.5 Winch believed that Malcolm did recognise the importance in Wittgenstein’s later work of attention to particular cases, of not trying to extract from them a theory of what is essential. But, Winch argues, Malcolm’s own failure to see how questions about logic are involved in Wittgenstein’s later treatment of topics like belief and knowledge suggests that Malcolm didn’t fully see how Wittgenstein was addressing the sources of philosophical puzzlement. Winch was uncertain how deep his criticisms of Malcolm went, how far Malcolm was unaware of what Winch took to be at issue. I do not want to try to decide the question about Malcolm, but rather to make clear Winch’s continuing emphasis on the need to see Wittgenstein’s later ideas, including ideas about his own philosophical methods, as tied to his rethinking of questions about logic. The comparison between Malcolm on the one hand and Kripke and Dummett on the other goes like this: Kripke and Dummett are, as it were, hungry for logical implications of what Wittgenstein wrote, but are deeply disinclined to take his methods seriously, and are therefore unable to see how Wittgenstein genuinely does treat problems about logic, while Malcolm is committed to certain characteristic Wittgensteinian methods, including the eschewing of explanatory theory in philosophy, but fails to see the significance of those methods in relation to questions about logic, hence cannot see how the methods are relevant to someone caught up in puzzlement about logic. My suggestion now is that Winch’s insight in the 1969 essay is a first expression of a main theme in his work on Wittgenstein, that one cannot grasp what is radical in Wittgenstein’s philosophy without seeing how his continuing interest in logic is involved in the two later shifts: the shift in subjects being discussed,6 and the shift in his methods. Thus it is part of this suggestion that Winch’s critical relation to Malcolm is not as distant as it may seem from his critical relation to Dummett and Kripke; for each side misses half of what Winch took to be essential.




1Winch, P. “Introduction: the Unity of Wittgenstein’s Philosophy”. In Studies in the Philosophy of Wittgenstein, ed. Peter Winch. London: Routledge, 1969, pp. 1–19.
2Winch, personal letter, 1987. See also some related comments on Malcolm in Winch, P. “Critical Notice of Malcolm, Wittgensteinian Themes”. Philosophical Investigations 20 (1997): pp. 51–64; p. 57.
3Rhees, R. “The Philosophy of Wittgenstein”. Ratio 8 (1966): pp. 180–93. Reprinted in Discussions of Wittgenstein. London: Routledge, 1970, pp. 37–54.
4Stenius, E. Wittgenstein’s Tractatus. A Critical Exposition. Oxford: Blackwell, 1960.
5Winch, P. “Discussion of Malcolm’s Essay”. In Malcolm, N. Wittgenstein: a Religious Point of View, ed. Peter Winch. Ithaca, NY: Cornell University Press, 1994, pp. 95–135.
6It should be noted that one of Winch’s aims in the writings of the last few years of his life concerned a significant non-shift of topic: Winch argued that Wittgenstein’s interest in logical questions plays a similar role in his early discussion of ‘A believes that p’ and in his very late discussions of belief in connection with Moore’s paradox. (See especially Winch, P. “The Expression of Belief”. Presidential Address. Proceedings and Addresses of the American Philosophical Association 70 (1996): pp. 7–23.) He thought that Malcolm’s failure to see the logical significance of Moore’s paradox, as seen from Wittgenstein’s point of view, weakened Malcolm’s discussion of Wittgenstein on belief. And he connected this with Malcolm’s misreading, as he saw it, of the Tractatus discussion of solipsism. (See Winch, 1997.)