Chemical kinds

As soon as we decrease below 6 the temperature of the system whose mean composition is S, this system must form a heterogeneous mixture consisting of the mixed crystals of composition fi, and of the mixed aj crystals of composition fj, therefore, if we continue to cool our liquid, it will entirely freeze furnishing such a solid mixture this mixture is produced in the same way as the eidectic mixtures studied in Art. 207 like them, it has a definite mean composition like them, it is a heterogeneous mixture of two kinds of crystals only these cr3rstals, instead of being of definite chemical kinds, are mixed crystals each of the two kinds of mixed crystals enclosed in the eutectic mixture has, furthermore, a fixed composition. 

LaPorte, J. 1996. "Chemical Kind Terra Reference and The Discovery of Essence." Nous, 30 112-132. 

We have argued here the possibility of "making" at the chemical level serves to blur the natural/artefact divide. This has three consequences (1) chemical kinds are not the unproblematic, paradigmatic instances of natural kinds they have been taken to be in the philosophical literature (2) we are left without a clear and uncomplicated example of a "natural" kind (3) brands of realism, which are tightly tied to natural kinds, are no longer automatically the status quo position to take vis-a-vis chemistry. Of course, chemists may come to realism for reasons that have nothing to do with natural kinds. Our point here is that philosophers who have tended to use natural kinds as a route to a defense of realism may have to do some rethinking. 

Nalini Bhushan argues that chemical kinds are not natural kinds. They are not natural kinds because many chemical kinds are human-crafted—synthesized—kinds and do not occur naturally. However, they are also not natural kinds in the more important sense that chemistry does not offer a univocal way of carving up substances (synthesized or natural ) into kind categories. How a chemist classifies kinds, has to do with local chemical and functional needs, and is responsive to these needs. For some purposes, having a particular kind of reactivity will drive classification for other purposes, structural issues will drive classification and for yet other purposes, other classifications are appropriate. 

Cf. my distinction between the ontology of matter in general, to be dealt with in relation to developments in micro- and astrophysics and the ontology of particular kinds of matter, i.e., chemical kinds (van Brakel 1991). Cf. Kant s matter in general (21 307) and der Gemeinschaft verschiedener Substanzen (21 571). 

A goal of this essay is to contribute to this body of critical literature on chemical kinds, but from a different direction. It considers the impact of chemical synthesis on the natural/non-natural divide. The point is simply that the properties of synthesized compounds can satisfy criteria just as easily as naturally occurring compounds, and thus that the intuitive connection between essences (or necessities or rigidities) and natural kinds is not in fact warranted. For example, take naturally synthesized proteins versus artificially synthesized proteins that are in every other way identical to the former. Is one a natural kind, the other not What is the basis for categorization ... 

Chemical kinds are not the unproblematic, paradigmatic instances of natural kinds they have been taken to be in the philosophical literature. 

In contrast to the concept of natural kind sketched above that is suggested by Kripke and Putnam s writings, Quine s use of phrase natural kinds in his paper with the same title (Quine 1969) is to be understood as kinds that come naturally i.e., instinctively or innately to us. On this use of the phrase, chemical kinds would be a subset of natural kinds in being the most theoretically sophisticated of kinds to be found in the sciences paradoxically, therefore, the least natural in the sense that they come the least naturally to us. Also, Quine s use of kinds has no link to essentialism either. For him, a kind is the consequence of a similarity judgment that we are innately predisposed to make rather than a consequence of groupings that are... 

The argument in this essay against taking chemical kinds to be natural kinds has... 

I note in passing that although one might consider it an open question whether or not the chemical kind is to be identified at the molar (water) or molecular (H2O) level, that inhere no matter what, this is not so for Putnam who would take molar level properties to describe superficial characteristics that are contingent of the kind. 

In a recent paper, LaPorte argues that kind terms rigidly designate abstract kinds (LaPorte 2000). On this view, chemical kind terms refer, and refer rigidly, but not in virtue of picking out natural kinds. This is an interesting position, but I do not have the space to discuss it here. 

LaPorte, J. 1996. Chemical Kind Term Reference and the Discovery of Essence. Nous 30(1) 112-132. 

While I am sympathetic to the semantic externalist project, I think that the discussion of chemical kinds by philosophers of language has been rather badly oversimplified, hiding difficulties that arise when we try to coordinate scientific kinds with the natural kinds recognized by ordinary language users.2 In this paper, I will examine these difficulties by looking more closely at the chemist s notion of water. 

While the ordinary language user and even the baptizer need not have true beliefs about water in order to refer to it, someone in the community eventually will need to have these beliefs if the meaning of water is to be made exphcit. The obvious people to ask about the nature of water are chemists, and semantic externalists have often assumed that this is where we must turn. Putnam and Kripke are not very explicit about the details of the role chemists play in semantics, but this is where I believe the coordination principle is implicitly rehed upon. Chemists discover the natural kinds of the material world, which I will call chemical kinds. The coordination principle presumes that the very same kinds that chemists discover are the ones relevant to ordinary language. If this is the case, then when you describe a chemical kind in detail, you will have nailed down the semantics of the ordinary kind term associated with that chemical kind. Putnam and Kripke seem to believe that in the case of water, chemists describe a chemical kind whose members include all and only the molecules with molecular formula H2O. Appealing to this fact and to the coordination principle, they conclude that water is H2O. 

Closer examination of what water really is, I believe, shows that for chemists, water is not just the set of all molecules with molecular formula H20. There are multiple chemical kinds that might reasonably be coordinated with the ordinary language kind water. Because chemistry provides us with many different types of natural kinds, and because it does not provide us with rules favoring one set of kinds over another, deference to the findings of chemistry will not unambiguously allow us to discover the extensions of natural kind terms in ordinary language. A more nuanced version of the coordination principle, which has specific rules for picking out the appropriate chemical kind in particular circumstances, will be needed to carry forward the semantic externalist project. 

This description of chemistry s subject matter leads us to a criterion for the individuation of chemical kinds. In deciding whether two samples of a substance are of the same kind, a chemist examines their structure and reactivity at all three compositional levels. For example, the two alcohols methanol and ethanol are distinct chemical kinds... 

Chemical kinds are to be individuated with respect to structure and reactivity at the molar, molecular, and atomic levels. 

---