Abstract
Both realist and anti-realist accounts of natural kinds possess prima facie virtues: realists can straightforwardly make sense of the apparent objectivity of the natural kinds, and anti-realists, their knowability. This paper formulates a properly anti-realist account designed to capture both merits. In particular, it recommends understanding natural kinds as ‘categorical bottlenecks,’ those categories that not only best serve us, with our idiosyncratic aims and cognitive capacities, but also those of a wide range of alternative agents. By endorsing an ultimately subjective categorical principle, this view sidesteps epistemological difficulties facing realist views. Yet, it nevertheless identifies natural kinds that are fairly, though not completely, stance-independent or objective.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
More modestly, the author of Classification Made Simple observes that without categorization “shopping would be very difficult and time consuming”(Hunter 2009, p. 3).
As will become clear, I use realism throughout this paper to indicate a commitment to the mind-independence of facts about which kinds are natural, and not to indicate an ontological commitment to universals, a further claim that some but not all advocates of realism (in my sense) will endorse.
By appealing to our universe’s causal architecture in this way, the anti-realist view I will develop exclusively concerns facts about which kinds are natural, and does not extend to the causal order itself (the mind-independence of which I am here agnostic). Naturally, keeping these two domains separate requires a conception of causal order such that facts about it do not themselves depend on facts about the natural kinds whose mind-independence is in question.
How does this approach apply to mind-dependent individuals (or processes), such as human mental states (which are constitutively dependent on us) or particles only created in supercolliders (which are causally dependent on us)? Though it might appear that my definition of realism would—rather unfortunately—make natural kinds of such mind-dependent individuals impossible, this is a mirage. What must be mind-independent for the natural kind realist are facts concerning which groups form, or do not form, natural kinds, not the existence of the members of those groups.
By ‘causal theory’ I mean to pick out any theory that understands natural kinds as groups enjoying some particular kind of lower-level causal architecture rather than to refer to the ‘causal theory of reference’ articulated by (among others) Saul Kripke. (Needless to say, a causal theory of natural kinds is perfectly compatible with the correctness of the causal theory of reference for particular natural kinds, such as gold, water, swan, etc.)
There are really two strands in Kitcher’s recent work on this topic, one maintaining that there are no natural kinds, and another suggesting that there are natural kinds, but that they are relative to our ends. I emphasize the second strand here because it presents a more interesting contrast to my own position.
In suggesting that, for the anti-realist, the natural kinds might have been different had we been different, I have made an assumption about how to interpret this counterfactual that is worth noting. I have assumed that for the anti-realist, the natural kinds are picked out, in a given world, by those groups that play the ‘natural kind role’ in that world, a role that anti-realists—whether promoting a content-dependent theory or one that is only status-dependent—will claim is partially determined by features of the categorizers in that world. Therefore, I am assuming that we are not availing ourselves of two rather squirrelly ways of interpreting the counterfactual: (1) holding our present aims and capacities fixed by packing them into the ‘natural kind role’; or (2) treating the term ‘natural kind’ as a rigid designator by tying its content to whatever physical property (if any) happens to play the natural kind role in the actual world. Having done either of these things, views I would consider to be anti-realist might maintain that, even had we been different, the natural kinds would have been the same. However, as David Lewis put it in a related discussion of laws of nature, this is “a cosmetic remedy only. It doesn’t make the problem [of mind-dependence] go away, but only makes it harder to state”(1994, p. 479). For more in-depth treatments of this issue, played out within discussion of normativity and metaphysics, see Street (2006: §7; 2008; ms.) and Sider (2012: §4.4) respectively.
Advocates of the species-as-individuals thesis (e.g., Hull 1978; Ghiselin 1997), who will deny that biological taxa are natural kinds and will instead consider them to be natural individuals, can re-read all of my claims about species classification as those about individuation. This is possible because, though this paper targets realism about natural kinds for reasons of space, identical issues face realism about natural individuals; the realist claims that there are mind-independent facts about what the natural individuals are, while the anti-realist maintains that any such facts are relative to us. (See Boyd 1999a, p. 73 for congenial comments on this matter.)
Some may wonder whether I have a right to presume realism about features on such thin dialectical ground. After all, one might worry that the kind of critique I develop in this section against natural kind realism—a form of genealogical critique—could be extended to features as well. Though I cannot explore the issue here, I believe the anti-realist is on much weaker ground when arguing against feature realism. Critiques of the kind I develop require mind-independent causal claims, causal claims that are often thought to themselves depend on facts about the features (or properties) that individuals possess. In consequence, arguably such arguments cannot be extended to the domain of features without being self-undermining.
Though I believe it is substantial, I cannot precisely state the level of improbability (‘luck’) that the realist really faces. Its extent will depend on the size of the set of conceptually possible partitions of the universe into natural kinds and on how our credences are spread across them.
See Van Fraassen (1989, pp. 51–53) for a related discussion of realism about natural properties, with equally critical comments on the ability of realists to settle on the mind-independent set.
The principle coordination challenge should be most worrisome for naturalistically minded realists, those who cannot reasonably deny the influence of the scientific categories on their accounts of natural kinds. What then to say to the rare metaphysician who maintains that the natural kinds have nothing to do with the categories of science? She may question my suggestion that judgments of naturalness are invariably influenced by those categories. Though I cannot engage with this opponent here, I believe a parallel genealogical critique can be constructed against the a priori principles such a non-naturalistic philosopher would need to use in constructing their purportedly category-independent theories of natural kinds.
Philip Kitcher articulates a view relevantly similar to this in his (2007).
This is a peculiar aim in, among other ways, the fact that it is fully constituted by a desire to categorize things in one particular way, rather than concerning (as is far more customary) some other matter that was served by a particular categorization.
Note that I am overlooking many interesting properties of Magnus’s and Boyd’s theories, and focusing on just the feature that presents the most useful contrast with my own account.
Admittedly, it remains doubtful that any ordering of what is more or less partially objective could itself be objective, as that would require an objective individuation of points-of-view.
Robust clustering is most apparent when we constrain our focus to the kinds of features actually in scientific circulation and exclude highly disjunctive creations. As justifying such a constraint is notoriously difficult, I cannot pursue it here, but instead, as previously mentioned, take for granted here a broadly realist approach to ‘features’ in this paper.
The account just described allows one to define, not just the natural kinds, but two sorts of non-natural kinds. The pathological kinds correspond to categories that well serve neither actual inquirers nor most neighboring agents. The neutral kinds correspond to categories that well serve either actual inquirers or some group of neighboring agents, but not both.
Given its role in the openly anti-realist Categorical Bottleneck account, I am able to understand both the structure of the epistemic agent space and the closeness threshold that determines the set of neighboring agents as fixed by us, rather than being properly objective.
For these subjects to be truly distinct requires that they be analyzed in a way such that they do not themselves depend on the identity of the natural kinds. I am presuming an account of causation that does this. In this respect, my view does not differ from competing views; most accounts of natural kinds in the high-level sciences presume the same with respect to causation in particular. For instance, Boyd’s homeostatic property cluster theory makes facts about natural kinds depend on facts about causation—since causal claims go into defining which properties are in ‘homeostasis’—but not the reverse.
For instance, the advocate of the homeostatic property cluster theory may suggest, in light of the fact that natural kinds are groups in which clusters of properties are maintained by homeostatic means, that some particular group in which properties are so maintained is a natural kind.
Of course, our normative judgments are themselves grounded in some of our actual dispositions. In this way, even on the normative approach the correctness of the correct principle depends on our actual characteristics. However, it does so in a very different way than it does for the descriptivist, and thereby recommends a different procedure for uncovering the correct principle.
See Magnus (2012) for a helpful discussion of this sort of situation, though approached in a different way than I do here.
References
Boyd, R. (1990), Realism, conventionality, and ‘realism about’. Festschrift for Hilary Putnam, pp. 171–195.
Boyd, R. (1991). Realism, anti-foundationalism and the enthusiasm for natural kinds. Philosophical Studies, 61(1), 127–148.
Boyd, R. (1999a). Kinds, complexity and multiple realization. Philosophical Studies, 95(1), 67–98.
Boyd, R. (1999b). Homeostasis, species, and higher taxa. In R. Wilson (Ed.), Species: New interdisciplinary essays. Cambridge, MA: The MIT Press.
Boyd, R. (2000). Kinds as the “workmanship of men”: Realism, constructivism, and natural kinds. Rationalität, Realismus, Revision: Vorträge des, 3, 52–89.
Boyd, R. (2010), Realism, natural kinds, and philosophical methods.
Chakravartty, A. (2007). A metaphysics for scientific realism: Knowing the unobservable. Cambridge: Cambridge University Press.
Darwin, C. (1859). On the origin of species, or the preservation of favoured races in the struggle for life. London: J. Murray.
Dupré, J. (1995). The disorder of things: Metaphysical foundations of the disunity of science. Cambridge, MA: Harvard Univ Pr.
Elder, C. (2004). Real natures and familiar objects. Cambridge, MA: MIT Press.
Ellis, B. D. (2001). Scientific essentialism. Cambridge, MA: Cambridge Univ Pr.
Fodor, J. A. (1974). Special sciences (or: The disunity of science as a working hypothesis). Synthese, 28(2), 97–115.
Ghiselin, M. T. (1997). Metaphysics and the origin of species. Albany, NY: Suny Press.
Goodman, N. (1978). Ways of worldmaking. Indianapolis: Hackett Publishing Company.
Hacking, I. (2007). Natural kinds: Rosy dawn, scholastic twilight. Royal Institute of Philosophy Supplement, 61(1), 203–239.
Hendry, R. F. (2010). The elements and conceptual change. In The Semantics and Metaphysics of Natural Kinds, pp. 137–158.
Hull, D. L. (1978). A matter of individuality. Philosophy of Science, 45, 335–360.
Hunter, E. J. (2009). Classification made simple: An introduction to knowledge organisation and information retrieval. Farnham: Ashgate Publishing.
Khalidi, M. A. (1998). Natural kinds and crosscutting categories. Journal of Philosophy, 95(1), 33–50.
Kitcher, P. (1984). Species. Philosophy of Science, 51, 308–333.
Kitcher, P. (2001). Science, truth, and democracy. New York: Oxford University Press on Demand.
Kitcher, P. (2007). Does ‘Race’ have a future? Philosophy & Public Affairs, 35(4), 293–317.
Kitcher, P. (2008). Carnap and the caterpillar. Philosophical Topics, 36(1), 111.
Kornblith, H. (1995). Inductive inference and is natural ground: An essay in naturalistic philosophy. Cambridge, MA: The MIT Press.
Lewis, D. (1994). Humean supervenience debugged. Mind, 103(412), 473–490.
MacLeod, M., & Reydon, T. A. C. (2012). Natural kinds in philosophy and in the life sciences: Scholastic twilight or new dawn? Biological Theory, 1–11.
Magnus, P. D. (2012). Scientific inquiry and natural kinds. New York: Palgrave Macmillan.
Mill, J. S. (1882). A system of logic, ratiocinative and inductive: Being a connected view of the principles of evidence and the methods of scientific investigation. Green: Longmans.
Psillos, S. (1999), Scientific realism: How science tracks truth. Psychology Press.
Quine, W. V. (1969). Natural kinds (pp. 5–23). Netherlands: Springer.
Reydon, T. A. (2009). Do the life sciences need natural kinds? Croatian Journal of Philosophy, 26, 167–190.
Scerri, E. R. (2011). The periodic table: A very short introduction. Oxford: OUP.
Sider, T. (2012). Writing the book of the world. Oxford: Oxford Univ Pr.
Skarsaune, K. O. (2011). Darwin and moral realism: survival of the iffiest. Philosophical Studies, 152(2), 229–243.
Stanford, P. K. (1995). For pluralism and against realism about species. Philosophy of Science, 65, 70–91.
Street, S. (2006). A Darwinian dilemma for realist theories of value. Philosophical Studies, 127(1), 109–166.
Street, S. (2008). Reply to Copp: Naturalism, normativity, and the varieties of realism worth worrying about. Philosophical Issues, 18(1), 207–228.
Street, S. (ms.). Normativity and water: The analogy and its limits.
Van Fraassen, B. C. (1989). Laws and symmetry.
Acknowledgments
For helpful discussions of this paper, thanks to Richard Boyd, Marc Ereshefsky, Andrew Franklin-Hall, Sharon Street, Michael Strevens, participants in the Corridor Reading Group (Errol Lord, Barry Maguire, John Morrison and Kristin Primus), and an audience at Cornell University. Many thanks are also due to Elizabeth Radcliffe for organizing this issue and for her patience with my contribution to it.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Franklin-Hall, L.R. Natural kinds as categorical bottlenecks. Philos Stud 172, 925–948 (2015). https://doi.org/10.1007/s11098-014-0326-8
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11098-014-0326-8