I’ve never been quite clear on Aquinas’ famous distinction between per se and per accidens ordered series, though I really like the clarity of Ed Feser’s explanation. Abridging greatly:
An instrumental cause is one that derives whatever causal power it has from something else. … [A]ll the causes in [a per se] series other than the first are instrumental [and thus] are said to be ordered per se or “essentially,” for their being causes at all depends essentially on the activity of that which uses them as instruments. By contrast, causes ordered per accidens or “accidentally” do not essentially depend for their efficacy on the activity of earlier causes in the series. To use Aquinas’s example, a father possesses the power to generate sons independently of the activity of his own father … .
The problem here is that it’s really hard to think of any examples of purely instrumental causes in this sense. Take Aquinas’s example of a per se series where the hand moves a stick which moves the stone. That may work in his physics, but not in ours. Every stick is basically a stiff spring—there are no rigid bodies. So, for ease of visualization, let’s imagine a hand that pushes one end of a spring, and the other end of the spring pushes the stone. When you push your end of the spring, the spring compresses a little. A compression wave travels down the spring and the tension in the spring equalizes. The spring is now “charged” with elastic potential energy. And it then pushes on both the hand and the stone by means of the elastic potential energy. There is an unavoidable delay between your pushing your end of the spring and the other end pushing the stone (unavoidable, because physical causation doesn’t exceed the speed of light).
Now, once the spring is compressed, its pushing on the stone is its own causal activity. We can see this as follows. Suppose God annihilated your hand. For a very short while, the other end of the spring wouldn’t notice. It would still be pushing against the stone, and the stone would still be moving. Then the spring would decompress in the direction where the hand used to be, and the stone’s movement would stop. But a very short while is still something—it’s enough to show that the spring is acting on its own. The point isn’t that the stone would gradually slow down. The point, rather, is that it takes a while for the stone’s movement to be at all affected, because otherwise we could have faster-than-light communication between the hand and the spring.
What goes for springs goes for sticks. And I don’t know any better examples. Take Feser’s example in his Five Ways book of a cup held up by a desk which is held up by a floor. Feser says the desk “has no power on its own to hold the cup there. The desk too would fall to the earth unless the floor held it aloft”. Yes, it would—but not instantly. If the floor were to disappear, the tension in the desk’s legs—which, again, are just stiff springs—would continue to press upward on the desktop, which would press upward on the cup, counteracting gravity. But then because the bottoms of the legs are unsupported, the tension in the legs would relax, the legs would imperceptibly lengthen, and the whole thing would start to fall. Still, for a short while the top of the desk would have been utterly unaffected by the disappearance of the floor. It would only start accelerating downward once the tension in the legs dissipatated. It takes a time of at least L/c, where L is the length of the legs and c is the speed of light, for that to happen. Again, the legs of the table are charged-up springs whose internal tension is holding up the desktop.
If this is right, then we don’t have any clear examples of the kind of purely instrumental causality that Feser—and, fairly likely, Aquinas—is talking about. Now, it may be that the deep metaphysics of causation is indeed such that indeed all creaturely causation is indeed of this instrumental sort, being the instrument of the first cause. But since Aquinas is using the idea of per se causal series to establish the existence of the first cause, we need an argument here that does not depend on the existence of the first cause.
18 comments:
Your argument seems to work because you can break down the stick into smaller elements that operate over time. If we were dealing with fundamental particles (or whatever is fundamental) it might not work because when particle one causes particle two to move (or however fundamental elements operate) it would be the same as particle two being caused to move. For particle one to be a cause it would have to bring about an effect, so if you remove particle one before it brings about an effect in particle two, it cannot be said to have started causing, but if it can be said to have started causing then the effect would have already been produced or will be produced regardless of particle ones continued existence, so I don't think you can break down fundamental causation in the same way you did with the hand stick rock example in your post.
You are pressing a physics worry against a metaphysical distinction. On Aquinas’s terms, a series “per se” is not a line of earlier-and-later pushes; it is a hierarchy of here-and-now explanatory dependence. What makes it “essentially ordered” is that some members act only by a power they do not have as principals but as instruments—i.e., their very efficacy in the relevant act is derivative on something higher in the same concurrent order. A series “per accidens,” by contrast, is a chain ordered by temporal succession, where downstream causes can continue to operate because they have their own native causal powers even after the prior members have ceased.
That is why the hand–stick–stone illustration is only a heuristic, not the linchpin. The point is not rigidity or instantaneous propagation but whether, under the description of the effect in question, the intermediate member is acting as principal or as instrument. Your spring refinement simply shows that the case you describe ceases to be a pure instrument: once charged with elastic energy, the spring now contributes by a power it presently possesses. In Thomistic terms, we have slid from a strictly instrumental transmission to a principal secondary cause acting from a form now in it (tension), itself received from prior action. That does not undermine the distinction; it confirms it. We say: while the spring is merely transmitting the mover’s act, its causality is instrumental and essentially dependent; when it has been altered so as to store energy, it has a new, proper ground of action and so can act for a time per accidens. One and the same set-up can instantiate both orders under different descriptions and at different instants.
Likewise with the desk. The question is not whether the top begins to move at t + L/c or at t + 0, but whether the desk’s “holding up the cup” is an efficacy it has as principal or one it has only by the concurrent operation of deeper causes. On any contemporary account, the desk’s upward support is the resultant of lattice forces and a gravitational field configuration; remove the relevant higher-order conditions and, once the altered boundary conditions propagate, the support ceases. The brief persistence is just the inertia of an already actualized state (stored stresses), not independence from the higher causes specifying and sustaining the very kind of action at issue. A fan that coasts for three seconds after you pull the plug does not thereby generate electricity; its motion continues per accidens from an already-present form.
This is exactly the shape Aquinas intends. “Per se” ordering is hierarchical and synchronic: musician–bow–violin–sound; nervous impulse–muscle tone–finger pressure–key actuation; power supply–filament incandescence–light emission. In each, the lower member is a true cause, but precisely as an instrument, and only so long as the higher act is in place. Turn off the current and the bulb glows for microseconds; walk away and the violin emits no melody. None of this trades on rigid bodies or superluminal signals. It trades on whether the lower cause’s efficacy, as that cause of that effect now, is underived or derived.
Once that is clear, two further points follow. First, you do not need to assume a First Cause to draw the per se/per accidens distinction; you need only the modest thesis that some causes operate instrumentally—i.e., their efficacy, in actu, is dependent on another. That is a neutral, phenomenological feature of ordinary agency and of physical systems with standing boundary conditions. Second, Aquinas’s argument from a per se series to a first in that order does not require that any natural example be “purely” instrumental in every respect. It requires only that, where the order is instrumental as such, an infinite regress of nothing-but-instruments is impossible: remove all principals and there is nothing to confer the very power by which instruments instrumentally act. A guitar cannot play while there is literally no player anywhere in the order; an electrical network cannot deliver current if there is, at no level, anything that has power to supply it non-derivatively; a field cannot mediate if, at no level, anything grounds its existence and laws.
Accordingly, your spring and desk cases do not show that there are no essentially ordered series; they show that some popular illustrations are mixed and that physics adds propagation lags and stored states. Thomists are happy to concede that and then relocate the “per se” dependence where contemporary science actually puts it: in the concurrent hierarchy of conditions that specify and sustain an effect now—fields, boundary conditions, material structures, formal configurations—none of which, taken as mere instruments, explains its own present efficacy as that cause of that effect. The argument to a first that is not instrumental is built on that ontological dependence, not on pre-relativistic kinematics.
"We say: while the spring is merely transmitting the mover’s act, its causality is instrumental and essentially dependent": But there is no time during which the spring is "merely transmitting the mover's act". The standard analysis of the spring is it that it *always* acts via the stored elastic energy on the stone, which stored elastic energy is being constantly reloaded by the hand at the other end.
"On any contemporary account, the desk’s upward support is the resultant of lattice forces and a gravitational field configuration; remove the relevant higher-order conditions and, once the altered boundary conditions propagate, the support ceases. The brief persistence is just the inertia of an already actualized state (stored stresses), not independence from the higher causes specifying and sustaining the very kind of action at issue. A fan that coasts for three seconds after you pull the plug does not thereby generate electricity; its motion continues per accidens from an already-present form."
These two cases are different. The coasting for the full three seconds is indeed inertia (while the initial part of the coasting is due to the fact that there is still electrical energy in the wires). But when you remove the floor from under the table, if it were just a matter of inertia, the tabletop would have *instantly* started accelerating downward, with velocity continuously increasing from zero at the time of removal to non-zero at every subsequent time. But it doesn't instantly accelerate downward, because if it did that, we would have transmitted information about the removal of the floor faster than at the speed of light.
"Accordingly, your spring and desk cases do not show that there are no essentially ordered series; they show that some popular illustrations are mixed and that physics adds propagation lags and stored states."
I am happy to agree that my discussion does not SHOW there are no essentially ordered series. But I have yet to see a clear formulation of an actual example of an essentially ordered series. I would love to see one.
Hi Prof.,
Harrison J. here. What do you think of these considerations: for Aquinas, causal powers (in creatures) are accidents of a substance. They depend for their being and actuality on the substance in which they inhere. It is properly and primarily the substance that acts/causes, but it does so through/by its causal powers. If this is the case, then would the causal power itself count as an instrumental cause (of the substance) in the relevant sense?
Harrison: That's really good. If so, then we have a short essentially ordered series: God->finite substance->causal power->effect.
But I am worried about this. The substance's motion consists in the actuation of the causal power--there is no other change in the substance. But that is also the same as the change in the causal power--it gets actuated. So we don't have a case where the substance's motion causes the causal power's motion. This makes the substance sound like an unmoved mover, and that can't be right.
How about this example, which builds off of other comments.
I believe, on the correct reading of Aquinas, substances sustain their accidents in existence. Consider the following:
(1) if x sustains y in existence at time t, then x (at least partially) causes the causal activity of y at t.
If substances sustain their accidents in existence, and (1) holds, then substances (at least partially) cause the causal activity of their accidents. A possible essentially ordered causal series would be: finite substance A -> accident of A -> finite substance B, where A is, say, the Higgs field and B is me.
"Take Feser’s example in his Five Ways book of a cup held up by a desk which is held up by a floor. Feser says the desk “has no power on its own to hold the cup there. The desk too would fall to the earth unless the floor held it aloft”. Yes, it would—but not instantly."
But I think you're supposed to abstract the temporal duration from the example here. The example *is* a case of an essentially ordered series because the desk has no power by itself to hold up the cup if not for the earth; or, if you'd rather, the desk has no power by itself to KEEP holding up the cup for long if the earth is removed. Whether or not this incapacity takes a little while to become apparent or not is irrelevant. By the time there is no earth, there is already a radical change in the causal series: you know the desk will stop being able to hold up the cup, even if this takes some time to happen.
The main thing about essentially ordered series of causes is not their simultaneity (though one could argue there is metaphysical simultaneity, too), it is the instrumentality of the causes in the series, I.e. the fact that they do not have, by nature, the full effect that they are causing. The intermediate causes are merely "passing ahead" what they have received.
We do have examples of essentially ordered series of causes in nature. For example, a series of mirrors reflecting light from a source: the mirrors are not truly the source of light, they are simply passing it ahead. A series of moons reflecting the light of the sun - no matter how many moons there are, the moons simply reflect light they receive, they do not generate it themselves. A train engine pulling a series of carts: each cart pulls the other, but their movement is essentially derived from the train engine and the whole thing would not be moving if not for the engine.
Aquinas holds that this is the case with existence (and any actuality). Things which do not have a certain actuality by nature must have received it from something that has it by nature. If there were no sun, the moons would not be giving any light.
This also sets up the structure of Participation. If you're not X by essence, you participate in it, and there must be a source for it.
Aristotle's culture, and the scholars of Aquinas' time after that one, may have believed their physics was empirically obviously true, or at least close enough to truth that it could be used convincingly to support metaphysical conclusions. We don't have the same starting viewpoint. One reconciliation is to say that we are not talking about actual "hands" and "sticks" in the example, just metaphysically appropriate ones. But this means we have moved the reasoning out of our world into a hypothetical one, where Aristotle's physics and biology are literally true, where maybe plants have genuine vitalist souls that might be weighed, and the sun moves around the earth. Is it legitimate, after that move into metaphysical equivalents of real things, to move our metaphysical conclusions back to the actual world?
One I believe valid approach is to see if we can build an equivalent metaphysics out of current physics by using arguments different from the classical examples, as Michael Staron proposes above with the Higgs field. But would Aristotle have developed the same metaphysics if he'd started with our biology and physics?
>”The example *is* a case of an essentially ordered series because the desk has no power by itself to hold up the cup if not for the earth; or, if you'd rather, the desk has no power by itself to KEEP holding up the cup for long if the earth is removed. Whether or not this incapacity takes a little while to become apparent or not is irrelevant. By the time there is no earth, there is already a radical change in the causal series: you know the desk will stop being able to hold up the cup, even if this takes some time to happen.”
But this makes the notion of essentially ordered causes entirely trivial and uninteresting. Let’s stick with the cup example.
The cup *does* in fact have the power to hold itself up independently from the table, even if for the smallest measure of time, merely from the fact that it is able to do so once the table put it up and any counteracting forces were absent. A disposition to do something under certain conditions is precisely what a power is. Really, it’s no different in kind from a stone being thrown in a certain direction and breaking a window. The stone can break the window only if (1) I have actually thrown the stone at the right angle, and (2) the wind isn’t strong enough to change the stone’s direction. These are only simplified illustrations and obviously someone can modify them to take a mature mechanics/dynamics into account.
That doesn’t at all entail that the stone has no power of its own to break glass. Neither does it mean that the cup has no power of its own to stay up at that precise height off the ground. It’s just that in the case of the cup, the forces counteracting the cup’s act are more immediate than the forces counteracting the stone’s. Ultimately, this is a difference merely in degree, not in kind. There is an obvious sense in which both the cup and the stone depend on causal priors, but there doesn’t seem to be anything special about the cup’s dependence on the desk; it depends on the desk in the same way the stone depends on the thrower, or that a son depends on his father to beget him in order that he himself is able to beget a son. Both of these examples are occurring in a temporal continuum and hence both can be reduced to a description of an accidentally ordered sequence, not an essentially ordered one. And it seems that every purported example of an essentially ordered series can be similarly reduced as well, at least by my lights. Inertia saturates the universe.
Michael:
"if x sustains y in existence at time t, then x (at least partially) causes the causal activity of y at t."
I suppose so, but I don't think this makes for an essentially ordered series. After all, when a father causes a son, it seems right to say that the father _partially_ causes the son's activity. I think _partial_ causation is not enough for an essentially ordered series.
Consider, too, a case where the son does something in the first moment of his existence. That shouldn't affect whether the series is essentially or accidentally ordered. But in this case, the father does something rather like sustaining the son's existence--the father causes (at least partially) the son to exist at that very time.
I think that to make this essentially ordered one might combine this with Harrison's insight that the substance uses the causal powers of the accidents as instruments, and the accidents have no causal oomph on their own. Maybe, but I'm not sure.
William:
I like the suggestion that we take Aristotle's physics as a mere metaphysical possibility, and use that to leverage ourselves to the idea of an essentially ordered series. But I am not sure we have sufficiently strong reason to think that Aristotle's physics is indeed metaphysically possible. If we thought his physics was true, that would of course give us possibility, but otherwise, I am not sure.
"The cup *does* in fact have the power to hold itself up independently from the table, even if for the smallest measure of time, merely from the fact that it is able to do so once the table put it up and any counteracting forces were absent."
I think the fact that the cup can't do it for long is a relevant point, the way I put it. But if you want to drop the cup and table example, fine. Say that it is completely wrong and unhelpful, then. I don't see how it affects the deeper point I made or the examples of train carts and an engine; mirrors and moons reflecting light... I think these are real examples of what I talked about.
It seems quite clear to me that essentially ordered series are about intermediary causes not intrinsically having the powers they have to produce the full effect, being merely instrumental. The train cart does move another cart, but it is not the appropriate source of the movement in the whole story. It only moves the cart because it is itself moved. The mirror is not a true source of light, it merely reflects light from a true source.
For Aquinas, existence is like that with contingent things. Contingent things by nature can fail to be, don't have to be, their nature is not sufficient for their existence. If they cause something else, it is only because they themselves are also caused, and as such they are not really the appropriate source of existence in a series of causes. To say contingent thing X exists because of contingent thing Y is like saying a cart moved because another cart pulled it, or that there is light somewhere because the moon caused such light by reflection. It does not adequately explain the phenomenon.
The train cart under tension is a stretched out spring, charged up because the cart in front of it has pulled it. If you remove the cart in front of it, it will keep on pulling for a very short time, as long as the spring has tension. If this is a case of an essentially ordered series, likewise a case where you wind up a clock and the clock runs for a day is a case of an essentially ordered series--but it doesn't seem to be.
But the cup *can* do it, and it *would* do it, under certain conditions. The reason why the cup tends to drop down to the ground is because an external force (the Earth’s gravity) is acting upon it. If the force of gravity were not present, it would just stick there in the same spot. There’s no intrinsic, necessary tendency of the cup to move to a certain location. Neither is there such a tendency in any other object. At least, it hasn’t been demonstrated in any object as far as we know. As I’m sure you are aware, we’ve generally moved past Aristotle’s physics.
So in point, the cup does indeed have the power to hold itself up. It’s just that this power is immediately blocked by gravity. Which means there isn’t an instance of an essentially ordered causal series occurring here. I think this is the case for your other examples (the train pulling the carts, moons reflecting light, etc) as well. Modern physics can describe these scenarios perfectly fine without the use of the notion of essentially ordered causes. Indeed, there doesn’t seem to be much room for them, given that it’s well-established that actions are never truly instantaneous (information cannot travel faster than light). In all of these examples, it’s not as if there is a causal source by which the others are instruments of it; rather it is that an earlier causal source transfers its own ‘sourcehood’ (which in these cases are to be understood as energy) to later members, and once those later members are themselves a source, the earlier source is no longer necessary. To be sure, there is still a sense in which we can speak of a sustaining cause of the cup, the train cart, the moon, etc. but it’s not in the sense Aquinas would understand it. Namely, insofar as there is a net causal force that tends an object towards not-X, a sustaining cause is that which is required to keep an object in X.
“For Aquinas, existence is like that with contingent things. Contingent things by nature can fail to be, don't have to be, their nature is not sufficient for their existence.”
Yes, but even Aquinas would admit that there is a difference between contingent in the sense of a per accidens series and contingent in the sense of a per se series, even in the case of existence. Certainly, I could have failed to exist. The fact that I exist however, can be traced back to my parents. So my existence is caused by my parents. Plausibly, I am also dependent on my internal components (both physical as in organs, and metaphysical as in body and soul) in order to exist at every moment. However, it doesn’t seem obvious that I require an *external* sustaining cause of my existence, for every possible moment that I do exist. So it seems my existence too is ultimately situated in a per accidens series, not a per se one. And such a series Aquinas would agree, could in principle be without a first member. Hence, existential inertia comes into play.
I don't think the sustaining is the important thing here. As I said, the temporal aspect should be abstracred away from the examples. The instrumentality is not in the sense that the intermediate cause must always, at every moment, be instantaneously sustained by the original cause - it is in the sense that the intermediate cause is only ever passing arround what it has borrowed, but is not in itself the true origin of the power. I thought I had made that clear.
In that sense, nothing you said seems to me to affect my examples. Yes, energy is being passed around and you can say the moon no longer needs the sun when it already has the energy, etc., but the point is that a series of moons is not sufficient in itself to account for the presence of light being passed around. Forget time, simultaneity, etc., this is a different topic.
Why not?
My whole point is that you should stop thinking about time and duration. The whole point about essentially-ordered series is not that the intermediate causes have to be continuously sustained through time. That is an interesting topic for metaphysics, but it is not the point. The point is that the intermediate causes are instrumental in the sense that they do not - of their own nature - have the sufficient power to account for the effect. A series of mirrors might reflect light into a room, but it can only do so if a true source of light has given it light in the first place. The fact that the mirrors, once the light was given, can just go on to reflect that light by themselves without any simultaneous sustenance is a different issue that doesn't change that fact. You needed a light source to get the effect going; mirrors were not enough.
Likewise, the carts pull each other and move each other, but this in itself is insufficient to truly account for the movement of the train. We need the train engine.
If we are to explain X, we need to have something which is X by nature, or which (to put it a bit differently) has the power to produce X by nature. That which is per aliud requires that which is per se.
Essentially-ordered series of causes are fundamentally about CAUSAL POWER, not about time.
Although I do think that in the end we need something like divine conservation, but this is because *existence* is a special case (nothing can act, or keep acting, or keep anything it has received, or be inertial, if it doesn't already exist at whatever time in which it has such and such properties or does such and such things). The essence of essentially-ordered series (heh) is not about conservation through time, but about causal power - the fact that the intermediate causes are by themselves insufficient to fully explain the effect, since they are not *by nature* appropriate sources of the effect. A series of moons does not appropriately account for the light; they needed a sun, or some other light source, to bring about the light
When things produce effects "by nature", they still depend on a bunch of other conditions. There does not seem to be a qualitative difference between charging up a human with food so that the human produces gametes for reproduction and charging up a spring by compressing it.
Post a Comment