What an Efficient Cause Is

I. Preliminary Remarks

A. Suarez and the moderns

  • Unlike Hume's analysis of causality, and unlike most of the prominent contemporary analyses (e.g., those of Lewis, Mackie, Tooley, etc.), Suarez's definition is not meant to be reductive. That is, he does not mean to analyze causality in non-causal terms. Thus, he is not trying to replace causality with some philosophically more tame or less metaphysical concept like constant conjunction, regularity, counterfactual dependence, change in probability, etc. Indeed, it is precisely in order to avoid the notion of action, which stands at the heart of Suarez's analysis, that these modern theories have been developed. Hume's analysis was self-consciously aimed at avoiding the metaphysical implications of the notion of action and also of what he (perhaps mistakenly) saw to be the intimately related notion of causal or natural necessity.
  • In giving his analysis, Suarez presupposes that what we ordinarily take to be instances of causality in nature are just that. He argues for this presupposition in section 1 of Metaphysical Disputation 18, but for present purposes he is assuming that we have enough paradigmatic instances clearly in mind in order to construct an illuminating definition or, better, explication.


B. Creation ex nihilo

  • At a certain point in his discussion Suarez makes adjustments that are conceptually required in order for creation ex nihilo to count as an instance of efficient causality. Notice that these moves do not by themselves assume that creation ex nihilo is indeed metaphysically possible. Rather, they assume simply that if there is or can be such a thing as creation ex nihilo, then it should count as an instance of efficient causality. This seems rather modest and wholly acceptable.
  • This concern with creation explains, by the way, why Suarez talks about the communication of esse rather than about the communication of form. If every instance of efficient causality involved an agent acting on a patient, then we could simply speak of the communication of form to the patient that serves as matter of the change in question. However, creation is an action but not an action on an antecedently existing patient or matter. In the case where the effect of creation is a material substance, creation brings into being both the form and the matter. Hence, Suarez uses the more general term esse for the perfection communicated in efficient causality, rather than the more specific term form. However, whenever a created efficient cause acts (at least naturally), the terminus of its action is always esse qua form, be it a substantival form, or an accidental form, or (perhaps) a mode.



II. Reworking Aristotle's Definition

  • What we have in this section is a carefully crafted reworking of Aristotle's characterization of an efficient cause as that 'whence there is a first beginning of change or rest'. At each step Suarez makes an emendation and then raises a problem that leads to a further emendation.
  • Step One:

    • Proposal: An efficient cause is that whence there is a first beginning of change or rest.

    • Problem: This definition does not contain a proper genus.

    • Solution: Replace 'whence' with the term 'per se principle'. 'Principle' defines a general category under which all the various sorts of causal or explanatory notions fall. Literally, it means 'source' or 'beginning'. (Later on in section two Suarez will distinguish a principle quod from a principle quo. This amounts to a distinction between (a) the substance which (quod) is the agent in a given case of efficient causality and (b) the power or habit or faculty by which the substance acts. However, this distinction does not come into play at present.) The particle per se is meant to exclude anything (or any description) which is only accidentally related to the effect produced. For instance, if a doctor builds a house, he does so as a human being or as a builder, but not as a doctor. That is, his being a doctor is accidental or per accidens with respect to this particular instance of efficient causality. The term 'doctor' connotes a human being as having a certain set of powers, habits, and dispositions which define the medical art he possesses and is irrelevant to his house-building activity. So the principle quod here is best captured either by a natural kind term which leaves open just what the relevant principle quo is ('human being') or by a term which connotes the relevant powers ('builder').


  • Step Two:

    • Proposal: An efficient cause is a per se principle from which a change first exists or comes to exist.

    • Problem: The definiens here is common to causes other than efficient causes, in particular the matter or material cause, since the matter, like the agent, exists antecedently to the change or exercise of efficient causality. It is not enough to point out that the matter itself must first be made to exist by an exercise of efficient causality, since the same holds for all secondary (i.e., non-divine) efficient causes as well.

    • Solution: The efficient cause is at least conceptually prior to the matter, in that the matter receives because the agent acts, but not vice versa. So this is a legitimate sense in which the efficient cause is first and the matter is not. Further, the efficient cause is an extrinsic principle, unlike matter and form, which are intrinsic principles or causes. Later on, Suarez clarifies this further by pointing out that the matter and the form communicate their own esse to the composite, whose own esse derives from or includes that of the matter and the form. By contrast, the efficient cause communicates an esse that is numerically distinct from its own -- and this whether we are speaking of the principle quod or the principle quo.


  • Step Three:

    • Proposal: An efficient cause is a per se and extrinsic principle from which a change first exists or comes to exist.

    • Problem: The definiens seems to apply to the final cause or end rather than to the efficient cause, since the end is that for the sake of which an efficient cause acts and is thus prior to the latter. In short, an efficient cause 'aims at' a certain terminus. Note that the Aristotelian picture has a dynamism packed into it that later anti-Aristotelians found fit to reject. Some of them (Descartes) limited change to local motion, where finality is perhaps least evident; others (Hume) simply rebelled against metaphysics in general; still others (Malebranche, Berkeley) saw the connection between finality and efficiency and limited agency to God alone or to God and rational agents alone. What they all deny--or at any rate are agnostic about--is dynamism and real action in nature.

    • Solution: The efficient cause is first in execution and it alone has a real moving influence. The particle 'from which' is already sufficient to mark this difference, since the end is that for the sake of which a change exists, but not that from which a change exists. So no emendation is called for.


  • Step Four:

    • Proposal: An efficient cause is a per se and extrinsic principle from which a change first exists or comes to exist.

    • Problem: (1) This definition applies only to the First Cause, since no other cause is first, strictly speaking. (2) This definition does not apply to creation, since creation is not a change.

    • Solution: (1) By 'first' we mean first within a given genus or order of causes. So it can apply to any principal cause, and indeed to any instrumental cause, advising cause, or disposing cause--in short, to any cause that acts and thereby contributes to some effect. So each genuine efficient cause will be 'first' with respect to some effect or other within some order or genus of efficient causality. (There is some question here about instrumental causes, but this will be dealt with in the next section.) (2) This objection is well taken, and to accommodate it we should speak of action instead of change.

  • Step Five:

    • Proposal: An efficient cause is a per se and extrinsic principle from which an action first exists or comes to exist.

      Alternative 1: An efficient cause is a first per se and extrinsic principle from which an effect flows forth, or on which an effect depends, by means of an action.

      Alternative 2: An efficient cause is a first per se and extrinsic principle from which an effect receives its own distinct esse by means of an action.

    • Problem: What an action is is just as obscure as what an efficient cause is.

    • Solution: At the level of generality at which we are now operating, it is sufficient to understand by the term 'action' the effect's emanation from and dependence on its extrinsic cause. Thus, we need not make explicit mention of the effect in the formula, although we may ala the second alternative. Later, in Disputation 18, Section 10, we will say more about what an action is. (It will turn out that an action is a mode of the effect that has an essential relation to the agent as actually acting -- and it is this mode which is the agent's causality.)

  • Lingering questions:

    • How are we to 'picture' the causal relation? Is some entity transferred from the agent to the patient?

    • How can an action be a real modification of the patient or effect and not of the agent? Does this make sense?

    • How is this conception of causality related to notions such as constant conjunction, spatial continguity, regularity, counterfactual dependence, and variations in the probabilities of the effect-events?

    • Isn't it events, rather than substances, powers, and esse, that are the relata of the causal relation?