The beginnings of the modern onslaught on causality are clearly discernible in the Nominalists of the late Middle Ages. Continued by the Cartesians, the attack culminated in complete repudiation of causality by the British Empiricists of the 18th century. Thereafter, modern philosophy was to take it for granted that causality is but an illusion, or at best a subjective category. Among the circumstances which made for this radical denouement was the stand of some Cartesians that God alone can be a cause in the true sense of the word, in which case secondary causality (the name for the causality of creatures) had of course to be watered down considerably. And so it was - reduced to a mere occasion for God to act. Causality becomes occasionalism, with a capital, to be sure.
By far the boldest stroke, however, came from such as Hume, in consequence of the phenomenalist interpretation of experience. Here, for the moderns, cause was systematically reduced to a mere relation of succession. To illustrate: I roll a billiard ball against another. The second ball begins to move. What is the explanation? The first ball, you naturally say, caused it (the second) to move. But the phenomenalist Hume will not have it so. All you saw (he argues, rightly enough) was two motions, one after the other. And he allows that under the same or similar conditions the same thing will happen; one motion will succeed another, or one event another. Which is precisely the reason why, in his view, we come to regard the relation between the two motions as a relation of causal dependence and eventually, by some psychological habit, translate this (assumed) dependence into the absolute principle that "whatever is moved, is moved by another." So doing, however, we have gone beyond the presentations of sense; and this the phenomenalist view of things does not permit.
Kant, for his part, had every intention of safeguarding the necessary and universal character of the causal relation. But having posited causality as no more than an a priori (i.e. subjective) category, a mere law of the mind, he could not but deny that it had any objective validity. The truth is that Kant, even as his predecessors, fell victim to the phenomenalist view of sense knowledge; like them he denies in principle that the intellect can derive the intelligible from the sensible. Against these dissenting voices, which we have hardly more than called attention to, the reality or objectivity of causality must be uncompromisingly maintained; and maintained both in the realm of experience in the strict sense, and in the realm of metaphysics as one of its first principles.
1. The Experience of Causality
The causal relation is, in the first instance, come upon as a fact of experience. One thing, to all appearances, moves another, effects a change in another; we see this constantly, accept it implicitly. I touch the flame of a candle; instantly I feel my hand burning, and instinctively lay the burning to the flame. Instances of this kind are multiplied a thousand times daily. Granted, the cause is sometimes assigned mistakenly, for sensible experience can be complex and difficult to analyze. Yet there are evidences of causal dependence, especially within my conscious self, so direct and unimpeachable that I would find it extremely difficult, if not impossible, to disbelieve them. Suppose you want to raise your arm; and now you have raised it. Is there anything more certain than that it moved from one position to another and that you were the one who moved it, caused its motion? The whole march of practical life and, it is well to add, of scientific endeavor rests on the supposition that things in this world, the world of experience, act upon each other.
We do more, however, than experience the causal sequence, the relation of real dependence of one thing upon another. We generalize it, so that under the same set of circumstances we confidently expect the same sequence, the same dependence. Metaphysics, however, goes still further, to the affirmation of causality as an absolute principle. With the recognition and enunciation of this principle causality appears in full view, and not merely under the aspect of generalized experience; for what has then been seen is that causality, under certain conditions, is a law, an absolute requirement, of being.
2. The Principle of Causality
This principle refers, as a rule, to efficient causality, which is the causality that usually comes under attack, as in the earlier-mentioned criticism of it. Of the principle so understood, two cardinal proofs can be given, one more particular in scope, the other more inclusive and at the same time more penetrating.
a) Whatever is moved, is moved by another. This, as noted before, is how the principle of causality is usually formulated in Aristotelian thought: whatever is moved, is moved by another. While the principle can be vindicated on physical as well as metaphysical grounds, we shall speak only of its metaphysical basis, namely, the resolution of motion into act and potency; 10 for it is from the consideration of act and potency that the principle receives its highest support.
Assume, then, that motion is a fact - not just an illusion; furthermore, that it is a transition from potency to act, which for all practical purposes means any kind of becoming. We then argue: Motion is a transition from potency to act. But a being in potency does not become actual by itself but only by some being already in act"- No being, moreover, can be in act and potency at the same time and in the same respect. Consequently, the going from potency to act can only be effected through the agency of another, through a being in act. Thus the principle is established: whatever is moved, is moved by another - quidquid movetur, ab alio movetur.12
b) Being which is not of itself, is necessarily of another. 13 To argue this proposition we do not set out from change or motion but from being which is not of itself, from being, this means,whose existence does not necessarily follow from its nature or essence. Such being is contingent (as against the necessary); nothing in its nature says it must exist, it can exist and it can just as well not exist. All beings encompassed by experience (by the senses) are contingent.
Now, what is to be inferred from contingent being? Considered in itself such being, as we have said, can exist and just as well not exist; its essence entails neither one nor the other. Consequently, its existence is in some manner an addition to its essence. Whence it follows that contingent being is a union or composition of diverse elements: essence and existence. But things which in themselves differ from each other cannot, as different, constitute a unity. If, then, they are found as one, as a unity, it can only be through some extrinsic cause having united them.14 And such is contingent being, a unity or composition of diverse elements, hence a being that needs must have a cause.
c) Justification from the principle of sufficient reason. The foregoing conclusion on the necessity of a cause for contingent being may also be arrived at by considering the principle of causality as an application of the principle of sufficient reason. We would then proceed as follows. Every being that does not have its sufficient reason from itself must have it from another. The contingent is such a being; the sufficient reason of its existence is not in itself, in its essence, hence must be in another. Which comes to saying that the contingent is caused; causality, once again, is demanded.
d) Ultimate ground of the principle of causality. It is perfectly true, as more than one author has observed, that the principle of causality is not strictly analytical; the predicate "being of another" is not expressly contained in the subject "being which is not of itself." More concretely, I can have the notion of contingent being, say of this book before me, without adverting or tracing it back to its cause. But granted this, it must still be maintained that, derivatively, the principle of causality is self-evident (the meaning of "analytical"), for causality follows on the nature of the contingent somewhat in the manner of a necessary property. No sooner, that is, do I grasp "being which is not of itself" and "being which is of another" than I recognize that the two notions imply each other, that "being which is not of itself, is of another." St. Thomas says as much when he writes:
Though relation to its cause is not part of the definition of a thing caused, still it follows as a result of what belongs to its nature. For, from the fact that a thing is being by participation, it follows that it is caused. Hence such a being cannot be without being caused, just as man cannot be without having the faculty of laughing.15
What St. Thomas has just said rests, ultimately, on the impossibility of more than one "being which is not of another." Clearly, "being which is not of another" can only be "of itself." If, then, the contingent was not caused, it would be a "being that is of itself" and thus there would be more than one "being that is of itself." Yet "being that is of itself," whose nature in consequence is to be, cannot but be an only being. For the infinite has no similar; God is one. Accordingly, the contingent cannot but be caused - under pain of contradiction.16