Margaret MacDonald and Gilbert Ryle: A Lost Philosophical Friendship

Michael Kremer (University of Chicago)

In the last years of his life, Gilbert Ryle (1900–1976) tried to destroy any evidence of his correspondence and unpublished work, so that there would be no Ryle Nachlass. However, upon his retirement in 1968, he donated a collection of his books to the library at Linacre College, Oxford, supplemented at his death. These books hide unintended clues to his life: postcards and letters that he presumably had used as bookmarks—for example, a postcard from Edmund Husserl and a letter from A.J. Ayer. They also provide evidence of an otherwise unremarked, personal and intimate friendship between Ryle and another philosopher, Margaret MacDonald (1903–1956)—librarian at St. Hilda’s College, Oxford before the Second World War, and reader in philosophy at Bedford College, University of London, afterwards. In this talk, I present the evidence for this friendship, beginning with a single 8-page letter from MacDonald to Ryle, written by one intimate friend to another, which I show is part of a now-lost longer correspondence.

In the light of the hypothesis that Ryle and MacDonald were close friends and philosophical interlocutors, I explore the relationship between their respective philosophies. I argue that MacDonald, who had studied with Wittgenstein at Cambridge prior to coming to Oxford in 1937, was an important source of Wittgensteinian ideas in Ryle’s philosophy. In particular, Ryle very likely derived his famous distinction between knowledge-how and knowledge-that from her (she in turn may have derived it from Wittgenstein’s Blue Book). I show this through an analysis of her contribution to a 1937 Aristotelian Society Symposium on “Induction and Evidence,” in which Ryle also participated. There, MacDonald discusses “knowing rules” in the sense of “understanding and being able to apply them,” and as distinct from knowing them to be true. This discussion preceded by three years the first appearance of the idea of knowing-how in Ryle’s work, in his 1940 “Conscience and Moral Convictions.”

Time permitting I may discuss MacDonald’s and Ryle’s evolving conceptions of philosophical methodology. In her papers “The Philosopher’s Use of Analogy” (1938), “The Language of Political Theory” (1941), “Natural Rights” (1947), and “Ethics and the Ceremonial Use of Language” (1950), she moves from a therapeutic conception of philosophy as uncovering misleading analogies in language, to a more positive conception of philosophical analysis as bringing to bear multiple analogies, none of them without flaws, to illuminate problematic concepts. I contrast this methodology with Ryle’s examination of conceptual categories and reliance on the idea of a category mistake in “Categories” (1938), “Philosophical Arguments” (1945), and The Concept of Mind (1949). I discuss MacDonald’s insightful 1951 review of The Concept of Mind in which she applies her own methodology in criticizing Ryle’s use of analogies. I suggest that the shift in the weight he places on the notion of categories in his 1953 Tarner Lectures, Dilemmas, may be due in part to MacDonald’s influence.


Part of the UConn History of Analytic Philosophy Workshop.