Major articles published since 1990
(except those collected in Matters of Metaphysics and Mind, Meaning, and Reality)
Nature's Joints: A Realistic Defence of Natural Properties Ratio 25 (2012), 387-404.
This paper attacks two contrary views. One denies that nature has joints, taking the properties we call natural to be merely artefacts of our theories. The other accepts real natural properties but takes their naturalness to come by degrees. I argue that both are wrong: natural properties are real, and their naturalness no more comes by degrees than does the naturalness of the things that have them.
Levi's Chances, Knowledge and Inquiry: Essays on the Pragmatism of Isaac Levi, ed. Erik J. Olsson, Cambridge: Cambridge University Press, 2006, 111-124.
Isaac Levi's principle of direct inference, from an agent's knowledge of a chance to that agent's corresponding credence, is central to his account of chance. He holds moreover that this principle shows the 'gratuitous, diversionary and obscurantist character' of frequency, propensity and other metaphysical theories of what chances are. In this contribution to Levi's Festschrift, I argue that, on the contrary, his direct inference principle commits him to just such a theory, the propensity theory.
Accepting the Universe, Think, 11 (2005), 55-64.
This article is a revised version of Too Many Universes in Mind, Meaning, and Reality.
Dispositions, transl. Jean-Maurice Monnoyer, La structure du monde, ed. Jean-Maurice Monnoyer, Paris: J. Vrin, 2004, 315-326.
This is a French translation of an earlier version of 'The Semantics and Ontology of Dispositions' in Mind, Meaning, and Reality. It was discussed at a colloquium on Australian metaphysics held in Grenoble on 9-13 December 1999.
For Facts as Causes and Effects, Causation and Conditionals, ed. John Collins, Ned Hall and L. A. Paul, Cambridge, Mass.: MIT Press, 2004, 309-323
In this article I argue that singular causes and effects are not particulars but facts, in the ontologically vacuous sense given by the principle that 'P' is true iff it's a fact that P. This does not make causation a relation between facts, since relations need real relata, which (e.g.) negative existential facts cannot be, although they can be causes and effects (as in 'She didn't die because she didn't drink the poison'). I argue that what makes such a causal 'E because C' true is not a relation but a truthmaker for conditionals like 'If C then E' and 'If ~C then ~E'. These truthmakers need not be relational -- since 'If C then E' entails neither 'C' nor 'E' -- and are usually particulars possessing properties, like an inertial mass, which is what makes it true that a net force F will cause a particular of mass M to accelerate at F/M.
Real Metaphysics: Replies, Real Metaphysics: Essays in Honour of D. H. Mellor, ed. Hallvard Lillehammer and Gonzalo Rodriguez-Pereyra, London: Routledge, 2003, 212-238.
In this article I reply to the other papers in this collection, written by David Armstrong, David Lewis, Peter Smith, Chris Daly, Tim Crane, Frank Jackson, Paul Noordhof, Peter Menzies, Isaac Levi, Alexander Bird, Arnold Koslow, Gonzalo Rodriguez-Pereyra and L Nathan Oaklander. The topics covered include truth and truthmaking, success semantics, subjective facts, physicalism, epiphenomenalism, causation, dispositions, laws of nature, change and time.
The Time of our Lives, Philosophy at the new Millennium, ed. Anthony O'Hear, Cambridge: Cambridge University Press, 2001, 45-59.
In this article I show how McTaggart's distinction between A- and B-series ways of locating events in time prompted and enabled the twentieth century's most important advances in the philosophy of time. It argues that, even if the B-series represents time as it really is, because having A-series beliefs when they are true is indispensable to the causation of timely action, the A-series represents 'the time of our lives'.
Realistic Metaphysics: an Interview with D. H. Mellor Theoria 67 (2001), 4-21.
This discussion with Anna-Sofia Maurin and Johannes Persson covers: how independent I think metaphysics is of physics, logic and semantics; how I use the Ramsey sentence of all laws of nature to say what properties there are; why I deny the existence of complex properties (and mereological sums); what I take truthmaking facts to be, and why these must include laws of nature unless the latter are necessary; why I reject the distinction between dispositional and other properties; why I think so-called determinables (e.g. colour) are properties of properties (e.g. red); and what I think of recent and likely future developments in metaphysics.
Transparent and Opaque Causation, Current Issues in Causation, ed. W. Spohn et al., Paderborn: Mentis, 2001, 9-20.
Although our actions seem to be caused by our intentions, most theories of causation cannot explain how they can be. This is because these theories treat causation as a relation between events. But reports of a relation must be 'transparent' -- i.e. it must not matter how we refer to a cause or effect -- since whether two events are causally related cannot depend on what they are called. Yet reports of intentions and other mental states are 'opaque': I can, for example, intend to reply to a letter from Fred without intending to reply to Fred, because I misread his signature. By showing how reports of causation can be opaque, this paper removes a major objection to the idea that our intentions and other mental states really can have causes and effects in their own right, and don't therefore have to be identified with physical states of our brains.
Possibility, Chance and Necessity, Australasian Journal of Philosophy 78 (2000), 16-27.
A defence of the idea of chance as a real physical, contingent and quantitative kind of possibility. This entails the 'necessity condition' that, for any P, P's chance being 1 entails P, and P's chance being 0 entails not-P. This condition is defended by showing how all apparent counter-examples to it can be overcome.
Introduction (with Alex Oliver), Properties, ed. D. H. Mellor and Alex Oliver, Oxford: Oxford University Press, 1997, 1-33.
This is a systematic introduction to the metaphysics of properties and to the sixteen selected readings on the subject.
F. P. Ramsey, Philosophy 70 (1995), 243-62.
This is a slightly revised text of one of a series of public lectures on Cambridge philosophers. It is based on 'Better than the Stars', a 1978 radio broadcast about Frank Ramsey, who died in 1930 at the age of 26. It includes contributions by his younger brother (Michael Ramsey, the Archbishop of Canterbury), his wife Lettice, I. A. Richards, R. B. Braithwaite, A. J. Ayer and Richard Jeffrey, as well as biographical information and expositions and assessments of Ramsey's theories of truth, decision making, probability, the nature of scientific theories and of philosophy.
Probable Explanation, Theory, Evidence and Explanation, ed. P. Lipton, Aldershot: Dartmouth (1995), 123-33. First published 1976 in The Australasian Journal of Philosophy.
Hempel thought explanations should make explananda probable, because explanations are inferences and the conclusions of inferences should be made probable. Salmon and Jeffrey have argued rightly that explanations are not inferences, and wrongly that explananda need not be made probable. This paper shows why explanations should make their explananda probable, and why it matters.
The Need for Tense, The New Theory of Time, ed. L. N. Oaklander and Q. Smith, New Haven: Yale University Press (1994), 23-37. First published 1981 as chapter 5 of Real Time.
Unlike earlier holders of tenseless views of time, I argue that tensed -- i.e. temporally token-reflexive -- thought, and hence language, is indispensable. The reason is that it is the truth of beliefs which make the actions they combine with desires to cause succeed, i.e. achieve the objects of those desires; and the success of most if not all actions depends on when they are done. Thus to cause timely actions we need beliefs which are true at some times but not others: i.e. temporally token-reflexive and hence tensed beliefs.
'Thank Goodness That's Over', The New Theory of Time, ed. L. N. Oaklander and Q. Smith, New Haven: Yale University Press (1994), 293-304. First published 1981 in Ratio.
A tenseless view of time, according to which nothing in reality is past, present or future -- except in the tenseless sense of being earlier, later and simultaneous with other events -- has to account (a) for the objective truth and falsity of statements ascribing these temporal locations and (b) to the apparent temporal presence of experience. (a) is done by a token-reflexive account of the truth conditions of tensed statements; (b) by applying this in particular to judgements of the form 'this experience is present' made simultaneously with the experience.
The Unreality of Tense, The Philosophy of Time, ed. R. Le Poidevin and M. Macbeath, Oxford: Oxford University Press (1993), 47-59. Revised version of chapter 6 of Real Time.
I here restate, defend and develop McTaggart's proof that real tense, the thesis that events are really past, present or future, entails a contradiction. I rebut attempts to refute or evade McTaggart's argument and show how his conclusion follows from the tenseless token-reflexive truth conditions of tensed thoughts and statements.
An Interview with Hugh Mellor, Cogito 7 (1993), 3-10.
In this interview I recall how I changed from chemical engineering to philosophy and place my work in the science-based tradition of Cambridge philosophy. I explain how my view of chance grew from ideas of Ramsey and Popper and why I dislike the 'linguistic revolution' in philosophy, with its bad effects on metaphysics. I sketch my tenseless theory of time, my view of the relation between physics and metaphysics, my theories of causation and induction. and my objections to physicalism, and I defend my role in recent academic controversies in Cambridge.
Probability and the Evidence of our Senses, A. J. Ayer: Memorial Essays, ed. A. P. Griffiths, Cambridge: Cambridge University Press (1992), 117-28.
It is argued that Ayer's persistent and unsuccessful attempts to explain how our senses justify the beliefs they give us, by postulating sense data, stemmed from his failure to adopt the right conception of probability. This is the objective contingent single-case chance which causes give effects and which perceptual causes give the beliefs in them which our senses make them cause and which are thereby justified by being made probably true.
Causation and the Direction of Time, Erkenntnis 35 (1991), 191-203.
The article strengthens Reichenbach's probabilistic causal theory of temporal order by showing (a) how it follows from the causal mechanism by which temporal order is perceived, and (b) that closed causal loops, and hence backward causation and cyclical time, can -- contrary to Reichenbach's opinion -- be ruled out a priori.