¹ The set of entities which appear in the future theory only needs to be ‘unique-ish’ rather than unique in order to permit realism in some strong form, since even the most committed realists are prepared to admit that science can only specify what there is within a reasonable margin of error. The possibility of realism in the future would become untenable, however, were there to be a wide range of empirically equivalent theories which postulated radically different theoretical entities from each other.

² For some interesting scepticism about genetic theory, see the previous Supplement to Philosophy: Philosophy, Biology and Life (Cambridge: Cambridge University Press, 2005)Google Scholar, passim, especially Dupre, John's article ‘Are there Genes?’, 193–210.Google Scholar

³ I will not pause to consider the difficulties involved in clarifying the concept of ‘approximate truth’.

⁴ See, for example: Boyd, Richard N., ‘The Current State of Scientific Realism’, in Leplin, Jarrett (Ed.), Scientific Realism (Berkeley: University of California Press, 1984), 41–2Google Scholar; Fales, Evan, ‘How to be a Metaphysical Realist’, in French, Peter A., Uehling, Theodore E. and Wettstein, Howard K. (Eds.), Midwest Studies in Philosophy XII: Realism and Anti-Realism (Minneapolis: University of Minnesota Press, 1988), 253–4Google Scholar; Jennings, Richard, ‘Scientific Quasi-Realism’, Mind 98 (1989), 240Google Scholar; Matheson, Carl, ‘Is the Naturalist Really Naturally a Realist?’, Mind 98 (04 1989)Google Scholar; Papineau, David, Theory and Meaning (Oxford: Clarendon Press, 1979), 126.Google Scholar

⁵ See Kukla, Andre, Studies in Scientific Realism (Oxford: Oxford University Press, 1998), 10Google Scholar; Psillos, Stathis, Scientific Realism: How Science Tracks Truth (New York: Routledge, 1999), xix–xxi.Google Scholar

⁶ Devitt, Michael, ‘Scientific Realism’ in Jackson, Frank and Smith, Michael (Eds.), The Oxford Handbook of Contemporary Philosophy (Oxford: Oxford University Press, 2005), p. 769.Google Scholar

⁷ Ibid., 769.

⁸ I use the term ‘unobservable’ with some hesitation, since the distinction between what is observable and what is not is notoriously difficult to define. What I have to say in this paper will not rely on there being a principled distinction between observables and unobservables, nor on our being able to make it.

⁹ See almost any introductory text on the philosophy of science for a review of the standard arguments.

¹⁰ Contrary to what some scientific realists seem to think (including Devitt, , op. cit.)Google Scholar, this objection does not require that the empirically equivalent rival theory T₂ does not refer to any unobservables at all, merely that it refers to some different ones from those postulated by T₁. This makes it more plausible that there are such empirically equivalent theories, although I will not pursue this matter here.

¹¹ Thomson's model of the atom treated electrons as being embedded in a cloud or soup of positive charge, like plums embedded in plum pudding, hence the name; his student Rutherford directed the ‘gold foil experiment’ which led to this model being abandoned in favour of a model of the atom as having a nucleus orbited by electrons.

¹² This need not go as far as the provision of constitutive identity and individuation criteria for the members of a category of entities, although it may do so.

¹³ In fact, phlogiston was initially replaced by Lavoisier with the almost-as-mysterious caloric, an ‘invisible’ and ‘weightless’, ‘imponderable fluid’ which was only later absorbed into thermodynamics by Joule's identification of heat with the kinetic energy of molecules.

¹⁴ I have defended this strategy at greater length elsewhere. See ‘Disorder at the Border: Realism, Science and the Defense of Naturalism’, Philo 7, No. 2 (2004), 183–185.Google Scholar

¹⁵ I take it here that the prospect that a theory could be empirically confirmed or disconfirmed is a reasonable indication of its hypotheses being physically possible, since the prospect of empirical confirmation indicates that such hypotheses may well actually be true.

¹⁶ It would be a mistake to think that scientists in the late 19th Century did not think their theory (their physics, at least) to be almost complete, or at least making excellent progress on the right lines with only a few loose ends remaining to be tidied up. (See, for example, Albert Michelson's 1894 speech, quoted in Horgan, John, The End of Science (New York: Broadway Books, 1997), 19Google Scholar; or Lord Kelvin's 1900 address to the British Association for the Advancement of Science.) This optimism was shattered by Max Planck's proposed solution to the problem of black body radiation in 1900 which introduced quanta of energy and led to the formulation of quantum theory.

¹⁷ There is widespread disagreement about whether von Neumann's postulated collapse of the wave function should be realistically construed, and if so, whether and how it is brought about by measuring quantum particles (known as The Measurement Problem), or if the collapse also occurs where no measurement is being taken. (See Cartwright, Nancy, ‘How the Measurement Problem is an Artefact of the Mathematics’ in her How the Laws of Physics Lie (Oxford: Clarendon Press, 1983)CrossRefGoogle Scholar for a version of the latter view.) I will not go into the intricacies of this issue here, however, since there are difficulties enough for the notion of deterministic causality from the Schrödinger equation alone; should it turn out to be preferable to treat the indeterministic collapse of the wave function realistically, then the defence of determinism will be significantly more difficult than I suggest here.

¹⁸ See, for example, Honderich, Ted, Mind and Brain. A Theory of Determinism. Volume 1 (Oxford: Clarendon Press, 1988)Google Scholar, sections 5.6–5.7; and How Free Are You? (Oxford: Oxford University Press, 1993), Chapter 6.Google Scholar

¹⁹ See Schrödinger, E., ‘Die gegenwärtige Situation in der Quantenmechanik’, Naturwissenschaften 23 (1935)Google Scholar. Translated into English by Trimmer, J. D.: ‘The Present Situation in Quantum Mechanics: A Translation of Schrödinger's ‘Cat Paradox’ Paper’, Proceedings of the American Philosophical Society 124 (1980), 323–338.Google Scholar

²⁰ These difficulties were initially formulated by Bell, J. S. (‘On the Einstein Podolsky Rosen Paradox’, Physics 1 (No. 3), 195Google Scholar. Reprinted in his Speakable and Unspeakable in Quantum Mechanics (Cambridge: Cambridge University Press, 1987))Google Scholar and gained some important empirical confirmation in Aspect, Alain, Dalibard, Jean, Roger, Gérard, ‘Experimental Test of Bell's Inequalities Using Time-Varying Analyzers’, Physical Review Letters 49 (No. 25, 1982), 1804–1807.Google Scholar

²¹ Bohm, David, Wholeness and the Implicate Order, (London: Routledge and Kegan Paul, 1980), 10.Google Scholar

²² Perhaps the most plausible interpretation of quantum mechanics which retains determinism is David Bohm's pilot wave theory. See Bohm, ibid, and Bohm, D. and Hiley, B. J., The Undivided Universe: An Ontological Interpretation of Quantum Theory (London: Routledge & Kegan Paul, 1993).Google Scholar

²³ For the purposes of this discussion, I will treat these as different names for members of the same category of entities, since they all share the functional role of providing the ontological grounding for the qualitative division of the mind-independent world and often share the criterion of being individuated in virtue of their causal roles. Whatever ontological differences there are between them are not relevant to the arguments I will make, so I will use ‘property’ as a neutral term to stand for the members of whichever of these categories the reader prefers.

²⁴ See Webb, J. K., Murphy, M. T., Flambaum, V. V., Dzuba, V. A., Barrow, J. D., Churchill, C. W., Prochaska, J. X. and Wolfe, A. M., ‘Further Evidence for Cosmological Evolution of the Fine Structure Constant’, Physical Review Letters 87 (2001), 091301.CrossRefGoogle ScholarPubMed

²⁵ See Sandvik, H. B., Barrow, J. D. and Magueijo, J., ‘A Simple Cosmology with a Varying Fine-structure Constant’, Physical Review Letters 88 (2002), 031302Google Scholar; also, Magueijo, J., Faster than the Speed of Light: The Story of a Scientific Speculation (Cambridge, Mass.: Perseus Publishing, 2003)Google Scholar for a more irreverent account of his research into VSL (Varying Speed of Light) theories.

²⁶ More precisely, alpha = e²/(ћ. c), where ћ = h (Planck constant)/2π. ћ might also be affected, but I am simplifying the matter a little here. Since alpha is dimensionless it is independent of any units.

²⁷ The nature of a particular property might not be entirely determined by its causal or structural role, but my point will affect all those philosophical views which regard what a property does as an essential feature of it.

²⁸ There also appears to have been confusion about those presocratics who did seem to understand change. For example, John Tzetzes, writing much later about Heraclitus, says the following: ‘Old Heraclitus of Ephesus was called clever because of the obscurity of his remarks: Cold things grow hot, the hot cools, the wet dries, the parched moistens.’ (Notes on the Iliad, 126H, in Barnes, Jonathan (ed.), Early Greek Philosophy (London: Penguin Books, 1987), 115.)Google Scholar Now, although there are different metaphysical accounts of change, we would not consider the idea that an entity can persist through change by changing its properties as being obscure.

²⁹ See my ‘Deepening the Controversy over Metaphysical Realism’, Philosophy 77 (2002), 519–541Google Scholar for further discussion about accepting metaphysical assumptions on the basis of inference to best explanation.

³⁰ Except, perhaps, as elements of a few extremely ‘despicable’ ones.

³¹ The possibility that the characterization of properties might not be correct might brings with it the danger that many philosophical positions outside purely metaphysical enquiry might stand in need of revision since they presuppose properties: properties (under this new conception) may alter the philosophical landscape by behaving in a different way, or may require replacement by some other category since they may not be up to their former task. It is equally possible, however, that some traditional philosophical problems may be resolved, were properties to become mutable (say).

³² I consider some issues to do with science and the supposed unity of the world's ontological structure according to the realist in ‘Disorder at the Border’ (op. cit.) including Margaret Morrison's example of how the unification of electromagnetism with the weak nuclear force to produce the electro-weak theory stands against the commonly held assumption that successful reductions support the case for realism about the unity of the world's causal structure. See her ‘Unified Theories and Disparate Things’, Proceedings of the Biennial Meeting of the Philosophy of Science Association. 2 (1994), 365–373.Google Scholar

³³ This option seems to be acceptable to Mellor, D. H., The Facts of Causation (Cambridge: Cambridge University Press, 1995), 124.CrossRefGoogle Scholar

³⁴ This account of the ontological interdependence between particular events and objects can be found in the later work of Donald Davidson; see, for example, ‘The Individuation of Events’ in his Essays on Actions and Events (Oxford: Oxford University Press, 1985), 174–5.Google Scholar

³⁵ Once again, if one accepts Cartwright's arguments, one might conclude that the apparent plurality of causal relations means that there is no causation, that there are many types, or that we must adopt a minimal conception of causation which can embrace them all.

³⁶ See my ‘Deepening the Controversy over Metaphysical Realism’ (op. cit.) for further discussion of this issue.

³⁷ For one attempt to defend this distinction against the arguments of Donald Davidson and Richard Rorty, see Baghramian, Maria, ‘Why Conceptual Schemes?’, Proceedings of the Aristotelian Society (1998), 287–306.CrossRefGoogle Scholar