Home Psychology Metaphysics and the philosophy of science : new essays
The Negative Content of ETMG
In respect of the negative content of ETMG concerning analytic metaphysics, the
main claims include the following:
Cian Dorr’s review of ETMG (2010) is one ofvery few sustained attempts to defend
analytic metaphysics against our critique.11 It is endorsed by Brian Weatherson and others and seems to have met with general approval. Dorr argues that our criticisms miss the mark because metaphysicians do not after all appeal to common sense or intuitions as a source of evidence or information about how things are. According to his account, we are quite wrong to suppose that common sense or intuition is taken as a reason for belief, rather “saying ‘Intuitively, P’ is no more than a device for committing oneself to P while signalling that one is not going to provide any further arguments for this claim” (Dorr 2010). Dorr contends that one of the important virtues of analytic metaphysics is the precision with which it investigates claims that have the form of conditionals, and their antecedents need only be assumed for the sake of argument. Hence, he argues against (1) above. According to Dorr, analytic metaphysicians do not generally take intuitions to have probative force. On the contrary, “very [often, ‘intuition’ talk is playing no such distinctive role,” but merely serves to introduce a premise and could be dispensed with in favor of a “pure and chilly” style involving simply their bald assertion.
This is puzzling because there are so many examples in the literature of metaphysicians using talk of what is commonsensical or intuitive as evidence in favor of a particular view. The first chapter of ETMG provided various quotations from prominent metaphysicians, and Dorr does not explain why they do not count as evidence for (1). Jonathan Tallant (2014) (who is concerned to defend analytic metaphysics against us) offers various citations from metaphysicians whose work is central to the literature, such as Ned Markosian, Ted Sider, and Dean Zimmerman, explicitly claiming that being intuitive and counterintuitive are epistemically significant features of theories or propositions. For example, Zimmerman (2008, 222) says that being commonsensical counts “very strongly in favour” of a proposition, and Markosian (2008) makes explicit the cost-benefit approach, and among the benefits he adduces, he explicitly includes satisfying intuitions. Indeed, Dorr appeals to intuition in various ways himself in a paper on relations (2004), saying, “I have not been shy of appeals to modal intuitions in this paper: in fact my method of argument relies essentially upon them” (183). Furthermore, at the end of the paper, Dorr offers an explicit statement of the cost-benefit analysis methodology of analytic metaphysics criticized in ETMG as follows: “Of course, these advantages must be weighed against the disadvantage of conflicting with our intuition that non-symmetric relations are possible” (191). In the light of all this, it is hard to believe that Dorr and his supporters are ignorant of the role intuitions play in analytic metaphysics; rather it seems that they are in denial about it.
(2) is not at all original to us. ETMG cited others making the point most eloquently, and it is not one that has been much contested. However, in another recent paper, Tallant (2013) produces evidence that “intuition” and cognate terms are used a lot in physics. Even if this is true of the sciences more generally, it does not show that scientists use intuition in anything like the way metaphysicians do. Often “intuitive” in mathematics and science is used to mean that something is readily comprehensible using a certain model or set of concepts. Sometimes it is used to describe insight and judgment. In such cases the intuition is schooled—often over very many years—making it a product and repository of existing knowledge. In any case, in science experiment and in mathematics, proof remain the certifiers of community acceptance and theory choice. Intuition is not a determinant of theory choice in anything like the way that it is allegedly deployed in analytic metaphysics, where it is the sole constraint other than logical consistency and theoretical virtues such as simplicity.
If Dorr was right that (1) is false, then (2) would be irrelevant and the argument for (5) in ETMG would be seriously undermined. However, of course (5) could still be true and it could be correct that current analytic metaphysics “contributes nothing to human knowledge” (vii). (3) and (4) provide further arguments for (5). ETMG offers various examples to substantiate (3) and (4), and in their reviews Kyle Stanford and Paul Humphreys (Stanford et al. 2010) take it to be pretty obvious that we are right about them. Dorr does not explicitly address (5), but he does seem to think that it does not follow from (3) and (4), though he concedes that “there is a fair amount of truth” (1) in our claim that metaphysicians often proceed with false presuppositions. He does not seem to be troubled by this, or to think that it is worth mentioning. This is because he does not regard those presuppositions as really playing any role, since according to him they serve only to provide concreteness and vividness and are dispensable (1). He points out that in the past, scientists made “great contributions to human knowledge despite making many false presuppositions”
(1). The difference, of course, is that as science has progressed, agreement has been reached about so much that we can be said to have a large stock of scientific knowledge. The Earth and the other planets orbit the Sun, germs cause disease, the heart pumps the blood to provide oxygen to cells for metabolism, all matter is composed of the elements of the periodic table, and so on. By contrast, as ETMG pointed out, metaphysicians continue debates about atomism and universals and cannot agree on any common core of metaphysical truths that have been established. The simplest way to refute our claim would be give an example of a contribution to human knowledge due to analytic metaphysics, but so far none has been forthcoming.
Of course, the claim that analytic metaphysics contributes nothing to human knowledge is ambiguous, since a contribution to human knowledge could take the form of propositions that can be added to what is known, or it could take the form of indirect contributions such as concepts, heuristics, and models. Indeed, metaphysics has arguably greatly contributed to empirical knowledge, albeit indirectly, for example, in terms of the mechanical philosophy, locality, physicalism, and atomism. Furthermore, some metaphysical systems have arguably been refuted indirectly— Aristotelianism, Cartesianism, vitalism, and, perhaps in the light of Bell’s theorem, local realism. This is the account of metaphysics that Peter Godfrey-Smith (2012) offers. According to him, metaphysics is about modeling reality, and so often involves simplifying and idealizing in order to represent the aspects of interest.
Of course, simplification and idealization are often essential to scientific representation, and that per se is not the problem with analytic metaphysics. Rather it is that the models that predominate are based on intuitive pictures of the world based on common sense and the manifest image, and to a lesser extent on domesticated science. This, especially given the current revival of Pre-Socratic, Aristotelian, and Mediaeval metaphysics, makes analytic metaphysics conservative and means that it does not encode much of our scientific knowledge of the world. Hence, it is unlikely to be of any use to individual sciences, which continually seem to require conceptual innovation, as Ladyman and Ross (2007) point out, and it is not helping with the unification of science. The critique of analytic metaphysics in ETMG concerns both methodology and content. The method is a priori and out of contact with science, and so accordingly the content tends to involve homely examples of substances and properties and corresponding thought experiments about change of parts and properties over time, or modally.
Much of analytic metaphysics contains no information derived from advanced science, does not engage with the scientific image and questions of interest to scientists, and embodies conceptual conservatism and seeks to domesticate science.
The analogues of metaphysical hypotheses in science—namely, highly theoretical high-level principles and symmetries—have evolved and are not discussed in the same terms as they were hundreds or thousands of years ago. ETMG correspondingly proposes reform of method and content. Of course, there are many metaphysicians and philosophers of science who exemplify naturalistic metaphysics broadly conceived.
|< Prev||CONTENTS||Next >|