Instrumental Causality: A Principle Small in Size, But Great in Consequence

· Philosophy

In my last post, I said I would discuss this principle of instrumental causality, and show how it has very important consequences. There are two ways in which the principle is relevant to discussions of natural science: first, because it is a part of the order of natural causality itself, or in other words a part of the object of our science. Therefore that object cannot be understood without understanding the role that instrumental causality plays in it. My essay linked on the sidebar — “Animals, Inertia, and Projectile Motion”  — shows just how important this is, in fact, because the whole “culture of force” which originated in classical physics is, to a large extent, due to a failure to apprehend the role of instrumental causality in physical local motion. It is scarcely possible to underestimate the importance of this, even though it has been given almost no attention up until now.

But the principle of instrumental causality is relevant to the philosophy of science for a second, equally important reason: because it is involved in our own acts of knowing by which we learn about the physical world. And although it is present in our acts of knowing even from the very start, its role in science has not remained static. As the natural sciences have progressed in what I earlier called the order of concretion, the role of instrumental causality has necessarily become greater and greater, for reasons I will explain below. But this, again, is largely passed over in customary discussions of the philosophy of science. Yet it is of no small importance if one would have a true understanding of the nature of the physical sciences.

Let’s start with what instrumental causality is. An instrument is, by definition, what acts in subordination to a principle cause, towards an end given to it by a principal cause. In the order of final causality, it is thus the principal cause, and not the instrumental cause, that has the end primarily; the instrument has it by virtue of its subordination. But this is also true in the order of efficient causality: the instrument may be said to be doing something which it nevertheless does not do of itself; if it did then it would be a principal cause, and only apparently subordinate.

We can come at this another way, to make it clearer. Efficient causes give being to their effects. Insofar as an efficient cause is a cause and its effect an effect, there is nothing in the effect that is not first in the cause. But we understand an instrumental cause to be one through which a principal cause acts. Will it still be true, then, that there is nothing in the effect that is not in the cause — including the instrumental cause? It must be true, if the instrument is genuinely a cause. Yet instrumental causes do things, as instruments, that are not in their own inherent power. In an earlier post I gave the example of speech. In speech, we communicate acts of the mind, which are not in the inherent power of mere sounds to communicate. And yet it can be said that speech does have the power to communicate thought. It thus evidently has this power only in a transitory way, precisely by virtue of its instrumental subordination to a principal cause which in this case is reason. One can recognize that this must be true in all cases; an instrument is, by definition, subordinate to causing an effect which it really has the power to bring about, but only in a transitory way.

Readers who would like to study this further might consult especially St. Thomas’s Tertia Pars on the sacraments. It is through the distinctions just made that St. Thomas explains how the sacraments can confer grace, even though they are of themselves mere material beings.

How, then, does this bear on the acts of knowing whereby we know the physical world? If one trusts the grandfather of contemporary philosophies of science, Hume, it has no bearing whatsoever.

Hume’s first principle in his theory of human cognition is that “ideas are weak sense impressions.” That means, to put it another way, that the order of intellection is not different from the order of sensation, except in extrinsic and accidental ways. If it were my goal here to argue with the more radically positivist strands of contemporary philosophy of science, I should have to make a detour here to argue that Hume is deeply wrong about this; this is a matter, again, of no small importance. But I shall set that aside for another occasion, and take it as understood by my readers that the the grasp of universals cannot be the same thing as the act of either imagination or the external senses, since the latter never escape the confines of “here and now,” whereas the former quite assuredly does.

But I will note three things about this, the first two perhaps worth noting just in passing, whereas the last is essential to the present discussion. The first is that the failure to grasp the character of intellect as different from sensation has, historically, gone hand in hand with the failure to understand the very great difference between words and symbols. Certainly, at any rate, the new burgeoning of nominalism in Hobbes, Descartes, and subsequently in Russell and the other inaugurators of analytic philosophy was very distinctly traceable to the assumption that words do exactly what symbols do. Symbols do not aim at expressing the universal in the way that words do. Readers who want to look into this more carefully should read my “Heritage of Analytic Philosophy” on the sidebar.

The second thing I want to note in passing bears not only on the question of nominalism versus essentialism, but on the validity of the whole conception of science which began with Aristotle’s Organon. The thing to note is this: typically, discussions of nominalism versus essentialism concentrate on the object of speech, rather than on the subject, which is ourselves. But this has an inadvertent and misleading consequence, which is that the question about what our words properly signify seems to most people to necessarily and directly resolve into the nature of the (mostly physical) objects themselves that we speak about. The “essentialist” view, therefore, comes to depend excessively on the postulate that the physical world around us actually consists of “essences” knowable “scientifically” (in the Aristotelian sense). This turns out, however, to be a difficult position to defend by anyone who really pays attention to the emerging discoveries about the character of the world around us. Historically, of course, advocates of “essentialism” have looked especially to Euclidean geometry as their example for defending the view that the world around us consists of essences knowable scientifically. The difficulties with that view don’t appear as readily as they would if, for example, one looked to astronomy or entomology. But they are there, all the same. That, however, is matter for another discussion. What needs to be seen here is that the question of nominalism versus essentialism should take its resolution more from the character of the human mind than from the character of Euclidean geometry, or rocks and minerals, or whatever. Once it is understood that the intellect is immaterial, the basis is laid for the claims made in Aristotle’s Organon; but it is a mistake the consequences of which haven’t yet fully been revealed to think that Aristotelian “scientific knowledge” is what it is immediately because of the character of the world around us.

The third thing to note — the thing which is really most relevant here — is that Hume’s effective denial of the immateriality of the intellect, and his consequent assertion that the order of intellection was the same as the order of sensation, went hand in hand with a failure to recognize that sensation (which here I take to include imagination) is precisely instrumental with respect to intellection. Because he granted nothing to the role of instrumental causality in human knowing, he took it as a first principle that the instrument of human knowing, namely sensation, must have fully within it the power necessary to bring about universal knowing; and he therefore proceeded to use that principle to try to better understand universal knowledge. He thus saw sensation and imagination not as instruments but as principal causes. That of course, is why he devoted an entire work to the futile endeavor to see how the universal could be derived out of acts of sensation and imagination as principal causes.

Well, I haven’t gotten very far with this, but I’m pretty much out of time already. Next time, I’ll start to say more about how this has an extremely important bearing on the progress of science. In the meantime, those who want to study this further could begin by reading De Koninck’s Introduction to the Study of the Soul. In that wonderful essay, De Koninck intimates — though only in a somewhat seminal form — the things I hope to elaborate on further. Unfortunately, life is too busy right now to promise that my next post will be very soon. I’ll see how it goes. But for those who want to continue to think about this, the thing to think about is this: if it is true, as it manifestly is, that human reason is capable of using the sensible as its instrument of knowing, should we not pay attention to that fact when we come to think about what we are doing when we do experiments, make models, and, in general, come to devise instruments of knowing in a more conscious way than is typical of prescientific knowing? Should we assume, gratuitously and with no conscious examination at all, that the limitations of our instruments are ipso facto limitations also upon our reason? Popper’s postulate that scientific propositions are never verifiable, but only falsifiable, is an example of what I am talking about. The view that our knowledge never transcends model-making is another example.


Comments RSS
  1. Blaise B

    It seems that our first notion of instrumental causality arise with reference to agent causality. Do you think we should think of instrumental causality only in the line of agent causality or can the notion be applied (albeit analogously) to other types of causality?

    • sdcojai

      Yes, certainly there is instrumental causality in the other kinds of causes as well; in fact, since the causes are not simply species of a genus but coordinated with one another (so that in physical beings all four causes are normally involved), instrumentality in one cause actually implies instrumentality in others. So, for example, a means may be seen as an instrumental agent, but also as an instrumental end; that is, a means is desirable as a means.

      In fact I think it is this very coordination among causes which makes instrumental causality intelligible. This is illustrated by contrast in theories which recognize no coordination of causes. A striking example, I think (but there could be many others) is Newton’s account of how light is involved in sensation. There is no recognition of any coordination of agent causality with final causality in Newton’s account, and the result is that he cannot think of light as an instrumental cause of sensation; it becomes no more and no less than the very object of sensation. … At least that’s true until the moment when the question of causality arises even in the internal causal order of sensation within our bodies, at which point even light ceases to be able to viewed as the object. Solipsism thus arrives….

      There is instrumental causality in matter and form as well. Agents use some forms as means to others, and some materials as means to others; that is right away enough to see, I think, that there is instrumental causality there too. And one could show that this instrumentality doesn’t just happen to accompany that of end and agent, but is essentially involved with the latter.

      As I began to suggest in my posts (before I ran out of available time; maybe after the school year is over I’ll come back to it), this understanding of instrumental causality is important not only for the account we give of physical reality, but also for the account we give of how science understands physical reality. The neglect of it is, as the example above from Newton might suggest, widespread, even among many of our friends. For example, we often hear allusions to Eddington’s fascination with the fact that the physicists’ experiments reduce to pointer readings. But if one is aware of the importance and meaning of instrumental causes in our theoretical methods themselves, the proposition that theories reduce to pointer readings can at once be recognized as doubtful.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: