Abstract
I question here the differentiation of kinds as substantive universals in Lowe and Ellis’ metaphysics, by taking up, for the argument’s sake, two extreme approaches on kind differentiation and kind change, a Heraclitan and a Spinozan approach. I show that, as things currently stand, Heraclitanism or Spinozism about kinds is consistent with the broad tenets of Lowe and Ellis’ metaphysics of kinds.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
Locke, An Essay Concerning Human Understanding, 3.6. 13.
See Klever (2012), esp. pp. 26–28; with respect to ‘essences’ and ‘essential properties’, Klever finds Spinozan echoes in Essay 2.3.6, 2.11.9, 2.23.3, 3.6.9, 3.6.26, 4.3.25, 4.3.30, 4.7.16–17.
In fact, that such problems exist should not be so surprising, given that, beyond British empiricism, the history of philosophy bears witness to an intricate thread of questioning of what we now call ‘differentiation’ of substantive universals, starting from Plato’s Parmenides and Sophist and going through the internecine debate between Thomists and Avicennian Franciscans in the thirteenth century over the thesis of unique substantial form in man; see Boureau (1999). I cannot go into this history here.
Lowe (2001, p. 179), italics added.
A piece of terminology for ease of exposition—as ‘particulars’ will be designated here kind members (and not tropes or modes)—where kinds could be countable or mass kinds. Members of countable kinds will be designated as ‘individuals’ and they will be the main focus of this paper. However, everything I shall have to say here about countable kinds is readily applicable to mass kinds as well and their member (which Lowe calls ‘quasi-individuals’; Lowe (2001, p. 58); and I shall in fact employ examples of mass kinds and quasi-individuals. In the same vein, the four-fold distinction I introduce in the following between two types of kind changes and two types of individual changes (which is strictly adequate only when applied to members of countable kinds) is readily convertible for members of mass kinds by substituting individuals with quasi-individuals and speaking of diachronic identity instead of numerical identity. However, since this four-fold distinction for individuals loads already the exposition I shall use it also with regard to members of mass kinds.
For instance, to use a stock example, changes of maturation for any animal offspring would amount to innumerable substantial kind changes, instead of the phase kind changes that common sense says they occur.
To appeal to another stock example, instead of the substantial kind change that common sense says it occurs, the transmutation of gold to lead would boil down to a phase kind change in which samples of gold, while changing their atomic number, would maintain their kind membership.
Lowe (2001, pp. 174–175). In spite of Lowe's admission that there are two types of substantial changes that particulars can undertake, Lowe does not put much emphasis upon the difference between individual and kind changes, and hence in certain places in his argumentation it is not very clear which of the two types of substantial changes is discussed, i.e. kind-related or individual-related (see for instance his 2001, pp. 178, 179 et passim, 1989, pp. 56–57). Accordingly, it is not very clear which of two possible sides of the Heraclitan and Spinozan strategy – directed at kinds or individuals—is criticised by Lowe and attempted to be refuted. Lowe appears at times to conclude that Heraclitanism about kinds can be rejected, using premises that seemingly could only refute Heraclitanism about individuals (see Lowe 2001, pp. 186, 187; Lowe 1989, pp. 56–57 et. passim.). I shall come back to this.
See Geach's famous example of the cat Tibbles apud Noonan (2006).
Locke, II, 31:6.
Locke, Essay, 3.6.39, italics added.
In Lowe’s sortal logic notation, for a particular a, a kind ϕ and a non-substantive universal F, ‘a/ϕ’ represents a relation of instantiation of a substantive universal, laws have the form ‘ϕF’ and the possession of the property F by the (kind member) a should be figured as ‘Fa/ϕ’ (see Lowe 1982, 1989, ch. 8; Lowe 2009, pp. 194–198). The non-substantive universal F ‘characterises’ the kind ϕ in such a way that the identity of ϕ depends on it being characterised by F (Cf. Lowe 2006, ch. 7 and 10, esp. pp. 155, 169, 170, 173; see also Heil 2006, p. 7).
Lowe (2001, pp. 54, 188, 198).
Lowe (2001, pp. 59).
Where this is consistent with the fact that individuals are always individuals ‘of a sort or kind’ such that ‘realism with regard to particulars…. implies realism with regard to sorts or kinds’ Lowe (1989, p. 4f), since the latter allows that individuals could swap natural kinds while remaining numerically the same, which is the possibility of ‘metamorphosis’. (Lowe 2009, p. 17), Lowe (2001, pp. 54–55).
Henceforth, for the ease of exposition, by saying that a particular ‘maintains or preserves its category’ I will also mean that and the specific diachronic conditions of the respective category are respected.
I use here my four-fold distinction between types of change in order to describe Lowe’s views, since I think his own terminology leads to a type of impasse that I shall try to describe in Sect. 3.
Lowe (2001, p. 186), italics added.
Lowe (2001, p. 176).
‘Where a change possessing features (1)–(4) occurs to a piece of stuff, it leaves us with numerically the same piece of stuff. This is because the diachronic identity-conditions for pieces of stuff require only that they retain the same material content forming a spatially connected whole, irrespective of any internal reorganization of that content or change in the shape, texture, or colour of the whole mass.’ Lowe (2001, p. 187).
Lowe (2001, p. 187). See also Lowe (2001, p. 179) where this sense of ‘survive’ associated with diachronic identity is clear Only in the light of metaphysics can we vindicate a judgement that a caterpillar survives its transformation into a butterfly whereas a pig does not survive its transformation into the flesh of the python which devours it’ (italics added).
‘It appears to be metaphysically possible for an individual living organism to start life as, say, a cat and yet to survive a process of transmutation which turns it into a dog. However, this wouldn't qualify as a phase change for that organism: it would qualify as a change of substantial kind, because contrary to the natural laws of development for cats’ Lowe (2001, p. 186).
Where, of course, the properties F1, F2, F3 and F4 instantiated by a would in fact be, in Lowe’s metaphysics, modes of the respective non-substantive universals.
Or, differently put the transition observed on the level of particulars from (F1&F2&F3)a/ϕ(?) to (F2&F3&F4)a/ϕ(?).
Or, more generally, the law L: Φ(1) (F1&F2&F3&F4).
Locke himself had asked whether monsters are distinct species (III, 6:17); see also Fales (1982). I am of course speculating here the fact that even if Lowe does have a conceptualisation of the accidental (pace Oderberg 2007, p. 133) this conceptualisation is not invoked in his construal of the phase/non-phase distinction for kinds. Recall that when it comes to the possibility that an animal might undergo changes contrary to its ‘natural laws of development’, Lowe cashes out this possibility only in terms of that animal’s undergoing a ‘change of substantial kind’; see n. 26 and Lowe (2001, p. 186).
See Lowe (2001, p. 178).
Of course, a Spinozan or a Heraclitan could also dismiss the intuitions for regarding, on a commonsensical basis, that certain transitions are ‘development changes’ governed by ‘development laws’. Again, the Spinozan and the Heraclitan could have intuitions of their own.
Recall that laws in Lowe could have exceptions on the level of particulars; among others, that is why substantive universals and kinds are invoked as truth-makers of laws and the latter are expressed in second-order quantified statements; these statements can justify and back up the so-called ceteris paribus clauses that are attached to the first-order quantified generalisations frequently appealed to in science; see Lowe (1989, pp. 153, 154) and Lowe (2006, pp. 13–17).
I have just taken up here the criticism addressed to Lowe that ‘normality’ is hard to ascertain on the level of particulars (see e.g. Mumford 2000) and have adapted it for the Heraclitan’s advantage.
See Lowe (2006, pp. 155, 169, 170, 173). Lowe avoids the of terminology of identity-dependence for substantive universals, preferring to say instead that (at least) some laws concerning natural kinds, when properly interpreted, emerge as metaphysically necessary (ibid. p. 155).
Cf. Lowe (2006, pp. 150–152, 164–165). Several other reasons are adduced against the metaphysical necessity of laws, as construed by the anti-Humean theorists. For instance, Lowe claims that universal physical constants could have been different (ibid. p. 151).
Ibid. p. 153.
Ibid. p. 154. Lowe claims that it also faces a metaphysical difficulty – to suppose that the identity of a non-substantive universal depends over its association with powers just seems ‘highly dubious.’ Ibid. pp. 164–165.
Ibid. p. 135.
See especially Lowe (2006, p. 170), where, on the same page, Lowe rejects the view that the identity of a non—substantive universal might depend on its association with a substantive universal (the example of electric charge) and intimates at the same time that kinds are ‘individualised’ by certain non-substantive universals they are characterised by (the example of electrons, positrons and unit charge).
Parenthetically, I should add, however, Heraclitanism or Spinozism about kinds does not appear as unacceptable as Heraclitanism or Spinozism about individuals. For instance, in its moderate version, Heraclitanism about kinds entails, as I said, that the number of kinds is much greater than the one commonly accepted, and, insofar as life sciences are concerned, there have been attempts in the philosophy of biology or medicine to show the plausibility of pluralism about the kinds in these sciences. The primary locus for this type of work is, of course, Dupré’s promiscuous realism in his (1993).
See Lowe (2009, p. 216). I thank one of the anonymous referees for drawing my attention to this aspect which I had previously neglected.
Think of an analogy with the Humean construal of laws, say in its Lewisian variant, as the axioms of a system that organises the facts of the entire history with the best balance between simplicity and strength (Lewis 1973, p. 73). Since actually there is no one who knew the facts of the entire history, the Humean theorist of laws would allow in actual scientific practice the employment of laws as they are currently formulated and as they converge with our knowledge so far. The Humean theorist might even allow the use of capacities-reasoning in the actual scientific thinking (see Papineau 1991) maintaining the caveat that ultimately, according to the Best System analysis, capacities are bestowed upon properties by the refined regularities constituting the axioms of the respective system (Lewis 1986, p. 223). Analogously, in my scenario, formulation of laws including kinds would be permissible, insofar as the co-instantiated properties delineating these kinds are useful for predicate projection, explanation, prediction, etc. maintaining the caveat that no one actually knows on what non-substantive universals depends the identity of what substantive universals. Perhaps the demon that ideally would be informed of the facts of the entire history in a Humean world (see Williamson 2009) would also know things about the identity of universals, and this is certainly encouraging. Note that this is a heuristic analogy, and my scenario with the metaphysically neutral employment of kinds would be consistent with an anti-Humean ontology.
Lowe (2001, pp. 182, 183).
Lowe (2001, p. 183).
Where this scenario is open to the possibility of Heraclitanism/Spinozism about kinds.
Lowe (2001, p. 188), underlines added.
Where this just a reformulation, or necessary consequence, of his distinction between substantial kind change and individual substantial change; Lowe (2001, p. 179).
Lowe (2001, p. 188), underlines added.
Lowe (2001, pp. 173–174).
Lowe (2001, p. 174).
Lowe (2001, p. 187).
which reads as: for all x, y and I, necessarily, if x and y belong to natural kinds K1 and K2 respectively, then in the circumstances Ci within a range of circumstance CI, they are intrinsically disposed to render the manifestations Ei, with the range of manifestations EI (Ellis 2001, p. 286). In line with his view of laws as metaphysically necessary, Ellis also adds a necessity operator within the scope of the universal quantifier, which I have not figured below since, as I said, I am interested in what happens with kinds in the actual world.
Lowe argues that structural properties such as shape are simply non-substantive universals (just like mass or charge), which characterize the kind, substantive universals. Lowe does not call the structural properties categorical, because he employs a different construal of the categorical-dispositional distinction; see Lowe (2006, pp. 17, 124–127).
In fact, it could not explain perfect similarities between all kind members because of the so called ‘problem of inference’, see Bird (2011).
See Ellis (2001: 68). Contra Bigelow and Pargetter (1990), Ellis argues that the universals standing for the atoms of carbon and hydrogen, cannot account for the shape of the methane molecule, because ‘universals cannot be spatially arranged’.
‘As soon as we speak about spatial arrangements of objects, we are already invoking the language of structural universals, since different groups of objects can presumably be arranged in exactly the same way’ (Ellis 2001, pp. 68–69).
Ellis (2001, pp. 70, 71).
‘…one might think of the property of being a hydrogen atom as a simple conjunctive universal (which is the simplest kind of structural universal). A thesis of this kind is certainly plausible for the fundamental constituents of matter, which presumably have no structure. But hydrogen atoms and the atoms and molecules of more complex substances all have structures, some of them quite complex. Yet their structures are no less essential to their identities as kinds’ (Ellis 2001, pp. 74, 75) One wonders though, if ‘universals are not spatially locatable’ (p. 75), how come substantive universals can fulfill for structures possessed by particulars the role that structural, non-substantive universals cannot. Ellis acknowledges in the final part of the respective section dedicated to substantive universals and structures that ‘tropes have the advantage of being spatially located, and relatable by primitive relations’. Campbell’s (1990) attempt to reduce structures to tropes is dismissed for the main reason that ‘it does nothing to explain the broad facts about the world alluded in the Basic Structural Hypothesis’ (Ellis 2001, p. 76), where the Basic Structural Hypothesis is the thesis that the world possesses the structure of natural kinds Ellis thinks it possesses (see Ellis 2001, p. 22).
Even if structural properties are mentioned at one point: ‘a trope of the structural property that all methane molecules have is the exemplification of the structure in any particular molecule.’ (Ellis 2001, p. 24, italics added). Compare with ‘substantive universals are always instantiated by objects or substances….and properties and relations are always instantiated by their tropes.’ (Ellis 2001, p. 73).
The same argumentation as for structures is advanced for all the other properties/powers possessed by kind members. I have chosen the example of structures because it is the only instance in which Ellis invokes explicitly in argumentation substantive universals. Otherwise, Ellis appeals only to rather general forms of his inference to the best explanation, of the sort: ‘[the perfect similarities between kind members] can best be explained by assuming that there are hierarchies of generic universals, the infimic species of which are all specific universals with identical instances.' Ellis (2001 pp. 67–68), italics added.
Bigelow et al. (1992).
Botterill (2005).
References
Bigelow, J., Ellis, B., & Lierse, C. (1992). The world as one of a kind: Natural necessity and laws of nature. British Journal for the Philosophy of Science, 43(3), 371–388.
Bird, A. (2005). Laws and essences. Ratio, 18(4), 437–461.
Bird, A. (2011). Are any kinds ontologically fundamental? In T. Tahko (Ed.), Contemporary Aristotelian metaphysics. Cambridge: Cambridge University Press.
Bird, A., & Tobin, E. (2008). Natural kinds. In E. Zalta (ed.), The stanford encyclopedia of philosophy, http://plato.stanford.edu/entries/natural-kinds/ (last accessed 28 Nov 2011).
Botterill, G. (2005). Review of Ellis: Scientific essentialism. Philosophical Books, 46(2), 118–122.
Boureau, A. (1999). Théologie, science et censure au XIIIe siècle: Le cas de Jean Peckham. : Belles Lettres.
Boyd, G. (1991). Realism, anti-foundationalism, and the enthusiasm for natural kinds. Philosophical Studies, 61, 127–148.
Bunge, W., & Klever, W. (1986). Disguised and overt Spinozism around 1700. Leiden: Brill.
Campbell, K. (1990). Abstract particulars. Oxford: Oxford University Press.
Crane, J. (2003). Locke’s theory of classification. British Journal for the History of Philosophy, 11(2), 249–259.
Dragulinescu, S. (2010). Phase and non-phase changes. Vox Philosophiae, 1, 143–166.
Dupré, J. (1993). The disorder of things. Cambridge: Harvard University Press.
Ellis, B. (2001). Scientific essentialism. Cambridge: Cambridge University Press.
Fales, E. (1982). Natural kinds and freaks of nature. Philosophy of Science, 49(1), 67–90.
Funkenstein, A. (1989). Theology and scientific imagination. Princeton: Princeton University Press.
Heil, J. (2006). Universals. Paper delivered at the Lowe conference, Buffalo University, http://ontology.buffalo.edu/06/Lowe/Heil.pdf (last accessed 14 Nov 2011).
Klever, W. (2012). Locke’s disguised Spinozism: Part 1. Conatus: Filosofía de Spinoza, 6(11), 61–82.
Lewis, D. (1973). Counterfactuals. Cambridge: Harvard University Press.
Lewis, D. (1986). Causal explanation. In Philosophical papers, volume II (pp. 214–240). Oxford: Oxford University Press.
Lowe, E. J. (1982). Laws, dispositions and logic. American Philosophical Quarterly, 19, 41–50.
Lowe, E. J. (1989). Kinds of being. New Jersey: Blackwell.
Lowe, E. J. (2001). The possibility of metaphysics. Oxford: Oxford University Press.
Lowe, E. J. (2006). The four category ontology. Oxford: Oxford University Press.
Lowe, E. J. (2009). More kinds of being. New Jersey: Blackwell.
Mumford, S. (2000). Normative and natural laws. Philosophy, 75, 265–282.
Mumford, S. (2002). Review of Scientific Essentialism by Brian Ellis. Metascience, 11(3), 324–328.
Mumford, S. (2005). Kinds, essences, powers. Ratio, 18, 420–436.
Noonan, H. (2006). Identity. In E. Zalta (Ed.), The Stanford encyclopaedia of philosophy, http://plato.stanford.edu/entries/identity/ (last accessed 14 Nov 2011).
Oderberg, D. (2007). Real essentialism. London: Routledge.
Papineau, D. (1991). Review: Correlations and causes. The British Journal for the Philosophy of Science, 42(3), 397–412.
Shoemaker, S. (1980). Causality and properties. In P. van Inwagen (Ed.), Time and cause (pp. 109–135). Dordrecht: Reidel.
Uzgalis, W. (1988). The anti-essentialist locke and natural kinds. The Philosophical Quarterly, 38(152), 330–339.
Wilkerson, T. (1995). Natural kinds. London: Ashgate.
Williamson, J. (2009). Probabilistic theories of causality. In H. Beebee, C. Hitchcock, & P. Menzies (Eds.), The Oxford handbook of causation (pp. 185–212). Oxford: Oxford University Press.
Acknowledgments
Work for the present paper was undertook during my doctoral studies at Lancaster University, for which I gratefully acknowledge financial support from the UK Arts and Humanities Research Council (fees doctoral bursary, awarded July 2007, Ref. No. 2007/135148/Lancaster University) and the British Society for the Philosophy of Science (doctoral scholarship, awarded July 2007). As always, I am very grateful to Rachel Cooper for advice and encouragement. I also thank the two anonymous referees of this journal for their suggestions and criticism.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Dragulinescu, S. A Challenge for Lowe and Ellis’ Differentiation of Kinds as Substantive Universals. Erkenn 78, 73–94 (2013). https://doi.org/10.1007/s10670-012-9401-8
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10670-012-9401-8