Monday, April 25, 2011

Reference to One

I recently came across an objection to Aquinas's basic account of analogy that is interesting, in the sense that answering it clarifies some things about the account that people often find confusing. The objection is essentially this. Analogy is supposed to be something intermediate between univocity and equivocity: if a term is predicated analogically, it is predicated neither univocally nor equivocally. But this means that its meaning in the two instances considered has to be partly the same and partly the different. But pretty much every sort of equivocation is a case in which the terms involved are in some way partly the same. 'Bank' in the sense of a financial institution and 'bank' in the sense of a part of a river would be equivocal -- they are merely homonyms. But you still can find things that a financial institution and a part of a river have in common. For instance, they are both places where something might be located. So it seems that the doctrine of analogy has a problem.

As you might have guessed, I don't think so. When Aquinas says that the definition of the term is partly the same and partly different, he isn't talking about any old way in which they could be the same. What he means specifically is that either (1) one definition includes a reference to the other or (2) both definitions include a reference to the same thing. One reason why Aquinas uses the 'healthy' example so much (it is not the only example he uses) is that it nicely exhibits both of these. 'Healthy' can be applied to diets, to bodies, and to urine samples. Obviously it doesn't mean exactly the same thing in each case. But the healthiness of diets and the healthiness of bodies are obviously not completely different, either; and the reason is that you can't explain the healthiness of a diet without reference to the healthiness of bodies. That is, what makes 'healthy' in 'This diet is healthy' and 'This body is healthy' analogical is that the former use of 'healthy' logically depends on the latter use. Likewise, 'healthy diet' and 'healthy urine sample' aren't using 'healthy' in a univocal way. But they both are partly the same in the sense that an adequate account of both requires reference to healthiness in bodies -- healthy diets are healthy in the sense that they contribute to bodies being healthy, and healthy urine samples are healthy in the sense that they are signs of bodies being healthy. In general this reference to one is by cause, or by effect, or due to mental understanding (e.g., when 'God' is used of what someone thinks of as God and of God); I can't recall Aquinas giving another way in which the reference to one could exist, although conceivably there is some somewhere.

The point that is relevant to the objection, in any case, is that the partial sameness of analogy is in fact a relatively precise thing; it does not mean similarity but actual sameness, and it doesn't even many any sort of sameness, but sameness due to proportion between terms -- reference to one thing.