1 The ‘etc’ here covers something of a difficulty, in that there are many more complex objects composed of molecules. One possible continuation continues with cells, multicellular organism, and social groups. See Paul Oppenheim and Hilary Putnam, “The Unity of Science as a Working Hypothesis,’ in Minnesota Studies in the Philosophy of Science Vol. 2, H. Feigl, et al., eds. (Minneapolis: University of Minnesota Press 1958), 3-36.

2 This brief account is oversimplified in several respects. First, I say nothing about the nature of laws except to treat them as constituents of theories. For purposes of this discussion, I am treating a theory as encompassing all the laws applicable to a particular level. Second, the derivations to which I refer will clearly require premises identifying objects at higher levels with structures of objects at lower levels, what are often referred to as bridge principles.’ I say nothing about these here, though it will be clear from the discussion in the text that often these will not be available. Third, I am assuming that sub-atomic particles are the smallest parts of matter. But things might turn out to have no smallest parts and be infinitely and arbitrarily divisible. If this should prove to be the case, reduction-ism would require that a theory of some level of (small) objects be deducible merely from a theory of stuff.

3 This argument against reductionism, with special emphasis on biological contexts, is developed in more detail in my book The Disorder of Things: Metaphysical Foundations of the Disunity of Science (Cambridge, MA: Harvard University Press 1993), chs. 4-6.

4 I treat this issue at length in The Disorder of Things, chs. 1 and 2.

5 I am assuming in this discussion that we are considering only human hemoglobin. Of course, the term is normally also applied to functionally homologous substances in a wide variety of organisms. This extension would vastly increase the chemical diversity I have been describing.

6 Waters, C.K. ‘Why the Anti-Reductionist Consensus Won’t Survive: The Case of Classical Mendelian Genetics,’ in PSA 1990, Fine, A. et al., eds. (East Lansing, MI: Philosophy of Science Association 1990), 125-39Google Scholar

7 The Extended Phenotype (Oxford: Oxford University Press 1982), 21

8 Notably in The Selfish Gene (Oxford: Oxford University Press 1976) and in The Extended Phenotype.

9 A difficulty with this pluralistic conception of the levels at which entities can be causally efficacious comes from the assumption that there is some level (typically the microscopic) at which a totalizing causal account can, in principle, be given. These issues are beyond the scope of this paper, but I address them in The Disorder of Things, part 3.

10 Such a pluralistic account has been developed in detail by Sober, ElliottThe Nature of Selection (Cambridge, MA: MIT Press 1985)Google Scholar. Unfortunately, I wholly disagree with the account of causality on which Sober bases this account. Indeed, I suspect that Sober’s account is not even ultimately compatible with genuine pluralism of causally efficacious structural levels.

11 Kitcher, Philip ‘Genes’ British Journal for the Philosophy of Science 33 (1982) 337-59CrossRefGoogle Scholar