That's why Aquinas starts chapter 29's methodological development by tying it to the introduction of perfection in chapter 28—'Now on that basis one can consider how likeness to God can or cannot be found among things' (28.269)—and, more precisely, to his warning about the easy misinterpretation of the word 'perfect' as applied to God. It provides the immediate occasion for considering the possibility of likenesses between God and other things despite the eliminative method's establishment of so many unlikenesses.
PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com)
© Copyright Oxford University Press, 2006. All Rights Reserved
In several places Aquinas develops a general account of kinds of
10 Besides SCG 1.29, see e.g. QDP 7.7c; QDV 23.7, ad 11; ST la.4.3c.
beginning with the basic observation that if X can literally and truly be said ii to be like Y in any way at all, then X has some form that Y also has.
11 See e.g. 29.273: '[W]hat is called like something is what possesses a quality or form of it'; ST la.4.3c: '[L]ikeness is associated with agreeing in or sharing a form.'
Fundamentally, then, likeness is conformity, sameness in respect of sharing at least one form.
Likeness shows up in many different contexts, of course. But our present concern is solely with likeness between God and ordinary things, and its context is the metaphysically primary relationship of anything else to God, the relationship of causal dependence—which is of course also the relationship on the basis of which our cognition of God is drawn inductively from our cognition of things. And so an understanding of the relationship of causal dependence is indispensable to natural theology not only theoretically but also practically, since it serves as the basis for the project of deriving truths about the ultimate explanatory principle from observations about the things the principle is meant to explain. For purposes of investigating the possibility of genuine likeness between God and ordinary things with a view to justifying the application of ordinary terms to God, we can, then, focus exclusively on Aquinas's analysis of the kinds of likeness obtaining between an effect and its cause. And although his agenda for the eight methodological chapters is oriented toward drawing practical conclusions regarding the use
of language about God,
12 The titles of the chapters give a pretty accurate impression of Aquinas's aims: The likeness of created things [to God]' (29); 'Names that can be predicated of God' (30); 'Divine perfection and the plurality of the divine names are not incompatible with divine simplicity' (31); 'Nothing is predicated univocally of God and of other things' (32); 'Not all names are said purely equivocally of God and of created things' (33); 'Things that are said of God and of created things are said analogically' (34); 'The several names said of God are not synonyms' (35); 'How our intellect frames a proposition about God' (36). (With the exception of ch. 34, which he left untitled, all these chapters were titled by Aquinas himself; see Ch. Four, n. 20.)
my present interest in this material is focused on the metaphysical underpinnings of those linguistic conclusions, the basis for the relational method.
Aquinas, of course, thinks of ordinary efficient causation not as a mere regular concatenation of events but, instead, as either the natural generation or the artificial production of effects. He understands this in terms of an agent's (or active subject's) initiating the sharing with a patient (or passive object) of some form the agent end p.145
Kretzmann, Norman , (deceased) formerly Susan Linn Sage Professor Emeritus of Philosophy, Cornell University, New York
The Metaphysics of Theism
Print ISBN 9780199246533, 2001 pp. [ 146]-[ 150]
possesses antecedently, often in some way quite different from the way in which the patient comes to possess the shared form. For present purposes I'm adopting Aquinas's understanding of efficient causation, and so, as a
reminder, I'll write in terms of'agent causation'.
13 My use of this term here is broader than, but includes, its standard contemporary use, especially in discussions of free will, where 'agent causation' is regularly contrasted with 'event causation'.
Some sort of likeness between an effect and its cause is an immediate consequence of this notion of agent causation, since agent causation shares a necessary condition with likeness: if A is the agent and P is the patient, then A antecedently somehow has some form, f, that P also somehow has, consequently. In agent causation the effect that is brought about by the actualizing of some potentiality in A or by A's exercise of some power is the
informing of P with f.
14 Herbert McCabe observes that the notion that effects are like their causes is one 'that the modern reader is likely to find most puzzling', but the puzzlement should be reduced if not eliminated by limiting the application of the notion to agent causation. As McCabe observes, Aquinas's 'typical causal proposition ... is not concerned with two events but with a thing, a form, and a subject into which the form is introduced by the thing. His general causal proposition would be something like "A brings it about that F is in B", where A is a thing, the efficient cause, F is a form and B is the "material cause", the subject upon which /A's causality is exerted. According to St Thomas what F is depends on the nature of A, so that if "/A" is a name expressing the nature of A, the meaning of "F" will be related to the meaning of "/A". What the effect will look like will depend not only on F but also on B' (McCabe 1964: 101).
Agent causation, then, entails a con-formity between cause and effect: 'Since every agent does something like itself in so far as it is an agent, but each thing acts in keeping with a form belonging to it, it is necessary that there be in the effect a [consequent] likeness of a form belonging to the
15 See also SCG 1.29.270, quoted on pp. 149-50 and 151 below.
Clearly, 'likeness' (similitude>) is a technical term in this context, closer to 'correspondence'than to 'resemblance' in the ordinary sense, even if in some cases the correspondence may be detailed enough to count as resemblance. The only immediately relevant con-formity between an agent cause and its effect is the presence in the effect of characteristics that could serve to identify, or at least to type, the agent—physical or metaphysical fingerprints providing the basis for an inductive argument to the agent's existence and some aspects of its nature.
Agent causation does not include the generation of accidental end p.146
effects: 'what is generated by something accidentally is not generated by it in so far as it is of such-and-such a sort, and so in what generates something there need not be a likeness of what is generated', just because in cases of accidental generation there is no antecedent likeness of the effect, which is at least often what we would call a chance effect. 'For example, the discovery of a treasure has no [antecedent] likeness in the person who finds the treasure accidentally while digging in order to plant something' (In Met. VII: L8.1443). On the other hand, the person's deliberately digging in order to plant something does have an antecedent likeness in his ideas and intentions, and is an instance of (artificial) agent causation. And if the treasure had been uncovered, instead, by a storm's uprooting a tree, (natural) agent causation would account only for features of the cause that could be inferred from the effect, such as the direction and force of the wind, forms belonging to A that constitute in it an antecedent likeness of the effect in P.
Was this article helpful?