Michael Dummett was born in London on 27 June 1925. He was educated at Sandroyd School and Winchester College. In 1944 he was converted to Roman Catholicism. From 1943 to 1947 he served in the armed forces. Afterwards he studied PPE at Christ Church, Oxford, graduating with first class honours in 1950. He was then elected to a fellowship at All Souls, Oxford and was assistant lecturer in philosophy at Birmingham in the year 1950–51. In 1951, he married Ann Chesney and over time they had five children.
For his final examinations in 1950 Dummett was required to read J.L. Austin’s translation of Frege’s Die Grundlagen der Arithmetik. Dummett was so impressed by this book that, in order to obtain a better understanding of Frege’s work, he began to study German and mathematics. This willingness to engage in depth with mathematics led him to adopt a philosophical method quite different from that associated with J.L. Austin and Gilbert Ryle , the two most celebrated Oxford philosophers at the time. Dummett’s philosophical writing is dense, employing technical vocabulary and making no concessions to the reader who is unwilling to follow a complicated train of thought. Indeed, when he submitted his first book for publication to Oxford University Press in 1958, Austin recommended that it be published, but on grounds of style, asked that it be rewritten. It was never rewritten to Dummett’s satisfaction.
In 1962 he became reader in the philosophy of mathematics at Oxford and by 1964 his first book on Frege was almost completed. However, he chose to give up philosophical research in order to devote his time to the fight against racism. In 1965 he was one of the founding members of the Oxford Committee for Racial Integration; from 1965 to 1966 he was a member of the Executive Committee of the Campaign Against Racial Discrimination; and from 1966 to 1968 he was a member of the Legal and Civil Affairs Panel of the National Circle for Commonwealth Immigrants. He then decided that because of the alienation of minority groups in Britain, a white person could not play a leading role in the struggle against racism. Although he did not completely abandon his political activities, he was able to resume his study of Frege and by 1973 Frege: Philosophy of Language, his first published book, was completed. Following its publication, he moved on from being reader in the philosophy of mathematics to take up a position as senior research fellow of All Souls in 1974.
Frege: Philosophy of Language consists of an evaluation of Frege’s general account of the workings of language. Dummett explains that Frege thinks that our understanding of the meaning of a statement is to be explained in terms of our grasp of what it is for that statement to be true or false. According to this position, which Dummett labels ‘realism’, a meaningful assertion is always determinately true or false, although we may never be in a position to discover which. In order to assess Frege’s position, Dummett compares it with what he considers to be the strongest alternative, although it is not an alternative that Frege himself ever contemplated. Dummett calls this alternative position ‘anti-realism’. The anti-realist explains understanding in terms of our ability to recognize whether an assertion has been verified or falsified. Thus the anti-realist has no reason to support the principle that every meaningful proposition is determinately true or false, that is the principle of bivalence, and rejecting this principle entails rejecting the laws of classical logic. The Dutch topologist L.E.J. Brouwer had made just such a proposal in the early years of the twentieth century, and developed an alternative to classical logic, which is known as intuitionistic logic. In 1977 Dummett published a study of intuitionistic logic, Elements of Intuitionism.
Dummett had in fact been publishing articles dealing with realism and anti-realism, amongst other topics, since the 1950s, and in 1978 his first anthology, Truth and Other Enigmas, was published. A recurring theme in this anthology is that there are a number of analogous but independent debates between realists and anti-realists. Each debate deals with a different subject-matter, such as the reality of the past, the reality of numbers or the reality of physical objects. The debates are analogous because, in each case, what is at stake is whether one should accept the principle of bivalence, or should instead follow the intuitionists in proposing a revision of classical logic. They are independent because one might be a realist with regard to some areas, and an anti-realist with regard to others. Dummett argues that such debates may be solved by providing a satisfactory theory of meaning, that is an account of what it is to understand a language. This formed the topic of the William James Lectures that he delivered at Harvard in 1976. These lectures were published in a much expanded form in 1991, entitled The Logical Basis of Metaphysics. This book is Dummett’s most detailed study of how a theory of meaning can be used to settle disputes about realism and anti-realism.
By 1979, when he succeeded A.J. Ayer as Wykeham Professor of Logic, (relinquishing his position at All Souls), Dummett’s reputation as a contributor to contemporaiy philosophical discussion was firmly established, but doubts were expressed about his historiography. In Frege: Philosophy of Language Dummett’s attention was focused on considering how Frege might have responded to the work of his successors, rather than placing Frege in his intellectual context. He made no systematic attempt to engage with the existing secondary literature about Frege and, by a remarkable oversight, the first edition contained practically no textual references. In 1981 Dummett completed work on a second edition of Frege: Philosophy of Language, complete with footnotes, and a lengthy defence of his exegesis of Frege, entitled The Interpretation of Frege’s Philosophy. This book provoked further debate about the interpretation of Frege’s work, and in 1991 Dummett published a collection of papers entitled Frege and Other Philosophers. In these papers Dummett examines Frege’s work in the light of his contemporaries and predecessors, rather than his successors.
In 1987 he gave a series of lectures in the University of Bologna which were mainly devoted to a comparison of Frege and Edmund Husserl. These lectures were published as ‘Origins of Analytical Philosophy’ in the Italian journal Lingua e Stile in 1988. Joachim Schulte translated the lectures into German, and interviewed Dummett. The lectures and interview were published as Ursprünge der analytischen Philosophie in 1988, and an English version of this book, with a revised text, was published in 1993 as Origins of Analytical Philosophy. Dummett argues that both Frege and Husserl recognized that, in order to study the nature of thoughts, philosophers should not study private mental sensations. A ‘thought’, in this context, means the content that is grasped by several people who can be said to share the same thought. Obviously, it is essential to a thought, in this sense, that it is something that can be shared, that is public rather than private. The founding insight of analytical philosophy, according to Dummett, is that only by studying the structure of language can we understand the structure of thought. Although Frege does not explicitly subscribe to this doctrine, it is suggested by the way that he deals with philosophical questions, and so Dummett describes him as being the grandfather of analytical philosophy. He credits Wittgenstein as being the first philosopher to state explicitly the doctrine that was implicit in Frege’s best work. Origins of Analytical Philosophy is not simply a description of what analytical philosophy is and how it began. It is also an attempt to justify the primacy that analytical philosophers attach to philosophy of language.
In 1991 Dummett published Frege: Philosophy of Mathematics, the long-awaited sequel to Frege: Philosophy of Language. Although he concludes that Frege was ‘the greatest philosopher of mathematics yet to have written’ (Frege: Philosophy of Mathematics, p. 321), still Dummett sees it as his main task to pinpoint where Frege’s philosophy of mathematics went wrong. On one level the answer to this question is already well known to anyone who has studied even a little philosophy of mathematics. Frege’s aim was to prove that arithmetic was a branch of logic, and he thought that the first volume of his Die Grundgesetze der Arithmetik, published in 1893, provided the logical foundations of arithmetic. But, as is well known, in 1902 Bertrand Russell demonstrated that it was possible to derive a contradiction from Frege’s fifth axiom. Dummett’s concern is to explain the underlying philosophical error that induced Frege to believe his fifth axiom was acceptable. He argues that Frege’s error was to suppose that infinite domains, such as the sequence of natural numbers, exist independently of our ability to prove things about them. Dummett argues that rather than thinking of such domains as having a permanent existence as infinite totalities, we should instead think of them as constructions that can be indefinitely extended. Once again, acceptance of the anti-realist position requires that we abandon classical logic in favour of intuitionistic logic, although in this case the argument would only affect mathematics.
It was in 1992 that Dummett retired as Wykeham Professor of Logic, becoming Wykeham Professor Emeritus. His valedictory lecture was included in his 1993 anthology The Seas of Language. He has continued to publish on philosophy, and in other fields where he is an expert: the history of card games, voting systems and the question of immigration. Although he has held positions at Oxford for the whole of his professional career, he is widely travelled, having held visiting positions at many other universities. He was knighted in 1999.
One of the most striking features of twentieth-century British philosophy is the linguistic turn. To some, excessive interest in philosophy of language has led to a trivialization of philosophy, because traditional metaphysical problems have been ignored. To others, the value of the linguistic turn is precisely that it provided a way out of metaphysical debates that could never be conclusively settled, because the positions under discussion lacked any meaningful content. Dummett’s position is that philosophy of language is of central importance because it enables us to understand and solve long-standing metaphysical debates.
The title of his most extensive treatment of this topic, The Logical Basis of Metaphysics (1991), indicates not merely that one can use logical arguments within metaphysical debates, but that metaphysical debates, or at least the metaphysical debates between realists and anti-realists with which Dummett is concerned, are to be understood as debates about logic. When the realist and the anti-realist mathematician disagree about the validity of a proposed mathematical proof, because they do not agree about the correct set of logical laws for mathematics, it is clear that we do not have a pseudo-disagreement that lacks real content. So, if Dummett is right, in order to resolve a metaphysical debate we need to determine the correct means for justifying logical laws.
A logical law can be justified by providing a semantic theory, a theory that explains how the truth-value of a complex expression is determined by its parts. An example of a logical law is the law of excluded middle, which tells us that for any proposition, P, ‘P or not-P’ is a true statement. Classical semantics, that is the type of semantics favoured by the realist, tells us that P has the value True or the value False; that it is determinate which of these values it has; that not-P has the opposite value to P; and that a complex proposition ‘x or y’ is true if and only if at least one of x and y is true. It is easy to see how these principles of classical semantics justify the law of excluded middle. What might seem puzzling is that this semantic theory is seen as constitutive of a realist metaphysics.
The connection between semantics and metaphysics becomes clearer when we consider a particular example, such as Goldbach’s conjecture: ‘Every even number is the sum of two primes.’ Currently, there is neither a proof nor a disproof of this conjecture, nor is there any guarantee that it will ever be possible to find one. If classical semantics is correct however, Goldbach’s conjecture must be determinately true or false. In explaining our deep-seated assumption that the conjecture must be true or false, we tend to invoke a certain metaphysical picture of the complete series of prime numbers existing in some platonic space, waiting for us to discover them, each of them either having or not having the property of being the sum of two primes. The anti-realist rejects in one stroke the semantic theory and the metaphysical picture that accompanies it. Dummett argues that the metaphysical picture has no content apart from the semantic theory: once we know what the correct semantic theory is, there is no further metaphysical question to be debated.
He then argues that a semantic theory is to be justified by providing the basis for a viable meaning-theory. When we are confronted with entirely new sentences in a familiar language, we are able to work out the meaning. Indeed, this does not usually require any conscious thought on our part. A meaning-theory explicitly states a set of rules that we implicitly follow when we interpret a sentence, or at least a set of rules that could serve this function. So, in order to evaluate a meaning-theory, we need to know exactly what it is to understand a sentence correctly. Dummett uses the term ‘theory of meaning’ for an account of what it is to understand a sentence correctly, so the task of a theory of meaning is to provide criteria by which we can evaluate a meaning-theory. Dummett thinks that questions about what it is to understand a language, while they may be exceedingly difficult, are capable of being definitively resolved. It is, after all, a matter of offering a satisfactory account of a human ability. It is on the basis of this long chain of connections that Dummett claims that what seems at first to be an intractable metaphysical puzzle can in fact be solved by proper attention to philosophy of language.
Dummett views Frege’s work on sense and reference as providing a model for the construction of a semantic theory. But when he comes to the crucial question of what it is to understand a sentence correctly, Dummett finds Frege’s suggestion unsatisfactory. Frege described our understanding of a thought as involving an unexplained human ability to grasp entities in a platonic third realm, beyond the world of spatio-temporal objects, or the world of the mental, that is the private domain of the individual mind. Dummett dismisses Frege’s theory as a myth, and turns instead to Wittgenstein’s suggestion that meaning is use.
As interpreted by Dummett, the slogan ‘meaning is use’ implies that to say that someone understands a language is not to speculate about an inner world to which nobody else has direct access, it is to ascribe to that person a skill that is manifested in their practice. The doctrine that meaning is use provides Dummett with a reason for being optimistic that his approach to metaphysical problems will lead to definitive solutions. It also provides him with a reason for doubting whether the solutions will be favourable to realism. When we respond correctly to evidence for or against a statement, by affirming or denying the statement, we demonstrate our ability to understand what has been said, and an account of what it is to understand a situation must include some reference to situations in which it is appropriate to recognize that the available evidence justifies or refutes a statement. What is distinctive about a realist theory is that it allows for occasions when a statement may be true, even though there is no possibility of justifying it, or false when there is no possibility of disconfirming it. We cannot however describe a situation in which someone recognizes that a statement is true but unjustified by the evidence, because for the person to act correctly in recognizing that the statement is true, there would have to be some evidence. The concept of a truth that completely transcends our ability to discover it seems to be superfluous to an account of how we respond to utterances of sentences.
It must be emphasized that the line of thinking that is summarized here is not presented by Dummett as a knock-down argument against realism. Dummett’s willingness to countenance a philosophical challenge to common sense and classical logic is a bold move, and it has, rightly, attracted a lot of attention. However, so much attention has been devoted to what Dummett says about the case for anti-realism that the impression is sometimes given that Dummett’s aim is to vindicate anti-realism. What he advocates is not that we adopt anti-realism as the correct theory, but that we carry out further research in philosophy of language, in the course of which he hopes we will uncover the principles that will enable us to solve disputes between realism and anti-realism. There is no presumption that either side will be correct in every case. In his valedictory lecture at Oxford, he described the situation in this way: ‘Justificationist [anti-realist] and truth-conditional [realist] meaning-theories do not stand opposed to one another as rivals. Neither is a worked-out theory: the justificationist principle is an unavoidable starting-point, the truth-conditional one no more than a hoped-for goal’ (Seas of Language, p. 474).
It is true that Dummett assigns a privileged role to intuitionistic logic: ‘If that [intuitionistic logic] is not the right logic, at least it may serve as a medium by means of which to discuss other logics’ (Logical Basis of Metaphysics, p. 300). Although the adoption of a semantic theory is intended to settle a dispute about what are the correct logical laws, the discussion of semantic theories cannot itself take place in a logical vacuum. Some logical laws will be invoked in the process of justifying a logical law by a semantic theory. So, it would be ideal if, in order to resolve disputes about controversial logical laws, logicians could agree upon a set of laws that are so far beyond reproach that they do not require a semantic justification. Dummett uses the technical apparatus of proof theory to establish which logical laws are in no need of self-justification, and on these grounds he assigns a special status to intuitionistic logic. It need not be the case that the logical laws which are agreed upon at this stage as being beyond reproach will be the same as the logical laws that are ultimately justified by whatever emerges as the preferred semantic theory, any more than we need suppose that a set of procedures followed when discussing a new constitution will themselves be incorporated in the final document. It is also true that in Frege: Philosophy of Mathematics Dummett comes very close to endorsing an anti-realist view of arithmetic, based on his argument about indefinitely extensible concepts. However, this is an argument that applies only to one particular debate about realism.
So far from being a dogmatic advocate of anti-realism, Dummett has argued that the thesis of global anti-realism, that is the thesis that the anti-realist position is always correct, may well be unsustainable (Truth and Other Enigmas, p. 367). Some of his most important papers are concerned neither with his general argument for anti-realism, nor with the history of philosophy, although a connection with issues of realism and anti-realism is often apparent. He has, for example, made important contributions to the philosophy of time and causation, the understanding of vagueness and the debate about the reality of possible worlds. Even a book such as The Interpretation of Frege’s Philosophy, which is primarily historical, contains lengthy and penetrating discussions of Peter Geach’s theory of relative identity, and Saul Krikpe’s view that proper names are ‘rigid designators’ picking out the same object in every possible world.
Dummett’s work has exercised a considerable influence. It is fair to describe the publication of Frege: Philosophy of Language in 1973 as marking a watershed in the study of Frege. It is true that there was already a growing revival of interest in Frege, but the book placed the work of Frege at the centre of attention for analytical philosophers, and set the agenda for the study of his writings for at least a couple of decades. More recently, attention has turned to Frege’s philosophy of arithmetic, and much of the most influential work on this topic has been carried out by philosophers who studied with Dummett: Crispin Wright , Bob Hale and the late George Boolos. Dummett’s work on anti-realism has inspired other philosophers, notably Crispin Wright and Neil Tennant, to work on issues of semantics from an anti-realist perspective. No less important is his influence on philosophers such as Christopher Peacocke who have tried to meet the challenge of his anti-realist arguments. However, to understand Dummett’s place in British philosophy, it is not enough simply to list those who have been directly influenced by him.
Dummett’s work exemplifies a certain trend within analytical philosophy, a trend that became dominant in Oxford in the 1970s. Since much of the inspiration for this trend came from the work of the American philosopher Donald Davidson, it is sometimes called the ‘Davidsonic boom’. Although it would not be correct to describe Dummett as a follower of Davidson, they share some ideas and themes, partly as a result of mutual influence. Important amongst these shared ideas are the primacy of philosophy of language, a focus on the link between meaning and truth, the desirability of adopting a systematic rather than a piecemeal approach to philosophy and a willingness to engage in metaphysical questions. The work of philosophers within this group is aimed primarily at other professional philosophers, since it cannot be understood without some prior study of formal logic. Dummett’s work combines a careful study of recent history of philosophy with a detailed proposal for a research programme, and thus he has a rationale for saying why certain philosophical methods have come to seem important at this point in history, and what results we can hope to achieve by using them. Whether or not he is correct, it is necessary for anyone who is concerned with the value of a significant movement in recent British philosophy to consider whether Dummett’s rationale is convincing. He can thus be described as one of the most important British philosophers of the latter part of the twentieth century.