Editorsā Note
Between 1947 and 1952, Elizabeth Anscombe was Mary Somerville Research Fellow at the Oxford University college of that name. She took this position after a four year spell at Cambridge during which she met and studied under Wittgenstein, while continuing her Oxford DPhil with Fredrich Waismann. While being a Mary Somerville Research Fellow, Anscombe filed an annual report on her work. This report was submitted in May of the second year of her fellowship.
The progress that I make in philosophy is partly by advancing on a centre from many sides. It has another feature which I shall explain after describing the ring of problems on which most of my work has been done this year.
- Phenomenalism. I have been working on this subject with Platoās Theaetetus as a starting point, and principally on two questions; a) is there any special connexion between phenomenalism and the doctrine that man is the measure of all things, as Plato claims? I have tried to show that there is, in the sense that the doctrine is characteristically absent from āintellectualistsā philosophies such as Platoās, Aristotleās, Aquinasā, Descartesā; and I have done a certain amount towards shewing why it is absent from such philosophies; and I am interested in making out whether the idea of man as the measure of all things is implied in Oxford philosophical work of the present day which is neither intellectualist nor phenomenalist. b) the refutation of phenomenalism by consideration that every statement and every concept has a logic; including the names of sensible qualities and the epistemologically primitive propositions announced as data by phenomenalism (This work, together with much on the problem of false belief, will go towards the preparation of lectures on the Theaetetus next term).
- This problem immediately links up with that of mental events and contents ā e.g. by way of the ācriterion in oneselfā of which Plato makes the phenomenalist speak; or by the concept of a mental act, such as meaning āredā, as a symbol of a quite special kind. The peculiarity would consist in this; that a thing cannot be characterised as a symbol without being characterised logically, but the thing that is a symbol ā e.g. a mark on paper or a set of flashes ā can usually be described as something quite apart from and antecedently to its being a symbol. But the mental seems to be characterised essentially by what it is of, whether it is a sensation or an image or a thought. This I use to explain the Greek idea that ālike knows likeā, Platoās thought of the affinity between the word and its form; the Aristotelian āintellectual quodammodo omniaā, St Thomasās theory of āesse naturaleā, āesse sensibileā, āesse intelligibileā. In modern times the idea might be expressed by the statement that a thought cannot be regarded as standing in an external relation to what it is a thought of; I discuss whether the idea of an internal relation as implied here is not a cheating attempt to have it both ways, to speak of a relation but to avoid the consequences of doing so. Here also comes in Aristotleās idea that the mind is not anything until it has something in it. It would be a strange idea of a vessel whose existence consisted in there coming to be something in it; in short it is clear that in spite of certain expressions he is not thinking of the mind as a vessel at all. I have reached only negative conclusions on the cognate question.
- The soul as a substance. All that has become clear to me is that one cannot explain the mental by using such expressions as āeventā, āprocessā, āactā, āsubstanceā and so on, and adding the rider that they are immaterial. I have gone over the present day attack on the philosophical notion of introspection without so far being able to arrive at a clear view of the whole problem ā whose ramifications are extremely complex ā or to convince myself of a single point: except that I am satisfied that the notion of a private ostensive definition explains nothing. But the arguments that lead me to this push me to unacceptable conclusions about such activities as talking to oneself or calculating in oneās head, about which I have repeatedly written without being able to remove the perplexities what they involve me in ā Taking a completely different starting point, I am much interested in Aristotleās theory of the soul as the form of the body; principally because his approach is neither introspective nor behaviourist; and I have tried to get at what he means by calling the body the organ of the soul as the eye is of sight; but so far have not arrived at any conclusion except that the connexion between an organ and that of which it is the organ is a conceptual one.
- āThe subjectā, āIā. This problem is closely entwined with the others. I write on it with the aim of shewing that the conception of the metaphysical subject of consciousness is both barren and fallacious; but I am never able to get completely free of it. I have written principally on the thought āOnly an āIā can be an observerā, which involves a consideration of solipsism (āI am the only observerā) and of Descartesā āCogitoā. Further I discuss Descartesā assertion that the mind is of all things most intelligible to itself; I have not been able to make up my mind whether the point that he is making is true and trivial (I am satisfied that if it is true it is true only in some trivial sense) or whether it is a falsehood which would be in conflict with e.g. the Augustinian description of the soul as āan abyssā which is ādark to itselfā. I have tried to go into the meaning of such phrases as this, considering e.g. whether they are explicated by ancient and medieval conceptions of the āirrational partā, or by modern pscyho-analytical investigations, whether they have anything to do with empirical or philosophical psychology, with theory of knowledge, and with metaphysics, but though all of these have a certain indirect connexion with this idea of āan abyssā they none of them appear to me to touch it very directly.
These are the interconnected problems with which I am most centrally occupied and on which I have done most writing in the past year. There are others, whose connexion with these is not close, and which are useful for me to investigate principally because I have not got a clear and decisive line on any philosophical problem: if I had on one, I should be in possession of the method that would carry me through all; and the most important thing for me is to discover the kind of considerations and arguments which shall solve my problems. In particular, I am engaged in preparing a paper on āThe reality of the pastā which I have promised as a contribution to a book of essays in āanalytical philosophyā which is to be published from Cornell University next year. I am using this relatively small topic to work out my doubts about what is called analytical philosophy, and expects the conclusions about the methods employed in that manner of philosophizing, to be of direct use for my work on my main subject.
Note
- G. E. M. Anscombe, āMary Somerville Research Fellow Reportā (1948), Somerville College Archive, University of Oxford. Ā© M C Gormally.
PART 1 From Parmenides to Wittgenstein
Elizabeth Anscombeās philosophy stands at the intersection of two ostensibly incongruous traditions: (i) a strain of analytic philosophy that originates in Frege and is developed in the work of Anscombeās teacher, Ludwig Wittgenstein, and (ii) the Aristotelian-Thomist tradition1. To chart these two influences in their entirety would be to recapitulate far too much of Anscombeās thought. But there is a surprising occupant of this Anscombean middle ground that is especially illustrative of her attempt at reconciliationāAristotleās metaphysics of substance. This topic is located at the nexus of these movements in two ways. First, Wittgensteinās influence is manifest in Anscombeās historical engagement with Aristotleās theoryāin the idiosyncratic interpretation she offers of his metaphysical texts. Second, Anscombe accepts and employs several core elements of Aristotleās theory of substance throughout her contemporary writings, but these positions are, once again, inflected by her Wittgensteinian commitments. In this essay, I will discuss several aspects of Anscombeās Aristotelian metaphysics in both its historical and contemporary registers. In doing so, I aim to illuminate how unique the variety of Aristotelianism she adopts is.
Few would describe Anscombeās attitude towards her philosophical predecessors as hagiographic and Aristotle receives the same kind of sharp and terse dismissals she infamously levels against other historical figures. For example, she claims that Aristotle āmisconceived the importance of the categorical syllogismā and that the theory of scientific explanation he builds upon this misconception in āhis worst bookā, the Posterior Analytics, is fundamentally flawed.2 More generally, she deems Aristotleās formal logic āto be of no more than scholarly and historical interestā (7).
But Anscombeās dismissiveness doesnāt extend to all of Aristotleās views. In particular, she notes the importance of Aristotleās metaphysics. She says,
his theory of substance, predication and existence [ā¦] seems to be the most fundamental and the most central topic in his philosophy; so much so that, apart from Aristotleās account of syllogisms and his ethical, aesthetic and political writings, most of his philosophical work can hardly be understood at all without it.
(5)
This acknowledged centrality doesnāt inoculate Aristotleās metaphysics from criticism. For example, Anscombe confesses that āI do not understand Aristotleās āformā, and I do not yet know whether he got clear about it himselfā and her response to Aristotleās claim that matter and form are the same, the former in potentiality and the latter in actuality, is simply that āthis is still Greek to meā.3
But there is much in Aristotleās theory of substance that Anscombe finds compelling.4 I will focus on her understanding of three of the theoryās central elements: Aristotleās conception of substance as he presents it in the Categories (§1), Aristotleās account of matter and its role as a principle of individuation (§2), and the concept of essence (§3).
1 Categories and substance
Consider the totality of things that we can sensibly say about a subject, for example, a particular human being: āShe is a human beingā (or āShe is an animalā), āShe is six feet tallā, āShe is tannedā, āShe is twice the height of her deskā, etc. According to Anscombe, there are āsignificant logical differences betweenā the āfairly simple kinds of things [ā¦] that might be said about a subjectā and it is these logical differences that ground Aristotleās list of categories: substance, quality, quantity, relation, etc. (15).
What is it for predications, or what they signify, to be significantly logically different from (or significantly logically similar to) each other? Elsewhere, Anscombe calls the sense of logic at issue here grammar. To understand somethingās grammar is to grasp its āstructureā, āgeneral formā, or ālogical shapeā.5 Somethingās logical/grammatical form is manifest in the intelligible statements in which it can figure, the reasonable questions we can ask about it, and the appropriate answers someone can give in response to such questions. As an example, Anscombe notes that ā[a] quality of a body like āwhiteā or āhotā differs from a quantity in that, for example, you can ask whether it is white or hot all over, whereas such a question hardly makes sense for predications of quantityā (15). There are presumably dozens of similar differences and, if they collectively meet some threshold of significance, they warrant our making a categorial distinction. To possess the grammatical understanding that would constitute knowledge of what a category is, one has to be fluent in the linguistic practices that surround the relevant variety of predication. This grammatical understanding is the common possession of those who can competently participate in the practice of making such sensible statements and asking and answering such questions intelligibly. Indeed, this grammatical understanding constitutes the practice itself and gives the statements, questions, and answers their unique senses.
Is Aristotleās list of categories, or any such list, complete? Why these ten categories? Why not additional categories, e.g., weight or shape? Anscombeās appeal to a threshold of significance for the logical difference is (appropriately) imprecise and she is consequently pessimistic that we can answer these questions clearly and determinately. If this is correct, then āit looks a hopeless task to construct a complete list of categoriesā and Anscombe concedes that Aristotleās doctrine is, at best, āa relatively crude sketchā (15, 16).6 Nevertheless, āthe idea of a category-difference [ā¦] is certainly a useful oneā and Anscombe appeals to the logical differences we find between Aristotleās preferred categories to elucidate central features of his theory of substance (15). We can see this if we follow Anscombe and contrast her interpretation of Aristotleās account of substance with a broadly Lockean view that she opposes.
The Lockean account of substance goes hand in hand with the rejection of nominal essences for individuals and comprises two principal commitments.
(L1) One can (re)identify a proper nameās referent without identifying it as being some kind or other.
(L2) Simple, non-relational predicates signify qualities.
Aristotle and Locke agree that particular substances (what Aristotle calls first substances) are subjects of p...