Aristotle on the Category of Relation (review)

It is often assumed that Aristotle cannot have an adequate understanding of relations, and in particular that "his substance-accident ontology and his reduction of propositions to the logical structure of subject-predicate form" (3) compel him to treat relations as monadic properties of the relata. But as Hood shows in this book, Aristotle's theory does indeed make room for dyadic (and generally polyadic) predicates—meaning not just that he employs such predicates in practice (that by itself would be no surprise), but that he does so in self-conscious connection with his theory of relatives. Moreover, Hood argues persuasively that Aristotle's characterization of relatives as holding "somehow" toward their correlatives suggests that he recognizes not only relational properties but relations as well. Along the way, she usefully distinguishes her interpretation from those of other commentators; especially helpful is Hood's diagnosis and critique of Morales's interpretation that relatives cease to be relatives when their correlatives are specified.

At first Hood seems to want to show that Aristotle's account of relations is not only adequate, but also consistent with his placing individual substances at the center of his ontology. Given the importance of this ontological orientation to Aristotle's overall philosophic project, it would indeed be valuable to be able to show that Aristotle can acknowledge dyadic relations without sacrificing his ontological commitments. But Hood's remark that Aristotle's relational theory transcends "the limitations of his other philosophical positions" (142) suggests instead that she doubts Aristotle's consistency, and perhaps dismisses his substance-based ontology as well-lost.

Similarly, while Aristotle's claim that the actualization of an agent affecting a patient is something located in the patient has sometimes been taken as symptomatic of Aristotle's confused assumption that predicates must have unitary subjects, Hood convincingly refutes this interpretation, showing that Aristotle explicitly describes the actualization as having both the agent and the patient as subjects (62). So far, so good. But one would then expect the moral to be that, when Aristotle speaks of the actualization's being located in the patient, he must mean something other than the actualization's having one subject rather than two. Hood concludes instead that Aristotle's "relational theory . . . may outstrip his causal theory" (63), as though the causal claim were embarrassingly indicative of a monadic approach to relatives after all. Left unconsidered is the possibility that Aristotle might have good reasons—unconnected with any confusion about relatives—for locating the actualization in the patient.

Hood's oddest omission is the lack of a clear explanation of what she takes items in the category of relative to be: relations? relata? relational properties? Hood glosses Aristotle's observation that "a man is not called 'someone's man' [but rather] 'someone's property'" (Cat. 8a22–24) as a distinction "between a specimen from the species and the species itself" (38); but I do not think that that can be right. For if Aristotle's reason for rejecting the expression 'someone's man' were that it would imply that someone owns the human species, then Aristotle's approval of the expression 'someone's property' would correspondingly imply that someone owns the species property. Surely what Aristotle actually means is that it is not qua man, but rather qua property, that a man belongs to someone; the distinction is not between specimen and species, but rather between a subject qua specimen of one species and the same subject qua specimen of a different species.

Now Hood does see that it is not Peggy's slavehood, but Peggy qua slave, that is supposed to be the relative. But it is unclear whether Hood thinks Peggy qua slave is just Peggy, or instead some complex (a "kooky object"?) constituted by Peggy together with her slavehood. She calls relatives "Complex Predicative Entities," for example, and writes that, while "Peggy's fundamental existence is as a primary substance, not as a relative," because Peggy is "the underlying...