Schmid on the Aristotelian proof

A fellow named Joseph Schmid has written a number of articles and blog posts critical of various ideas and arguments of mine, such as the Aristotelian proof defended in chapter 1 of my book Five Proofs of the Existence of God.  Until this week, I hadn’t read any of this material, though for some time now I’ve been getting an increasing number of requests that I comment on it.  Many of these have been anonymous and weirdly insistent or adulatory toward Schmid, which made me suspect sock puppetry rather than genuine widespread interest.  My attention in recent months has, in any event, been focused on the book on the soul that I am working on and which is way behind schedule (as well as on other existing writing commitments, most of which have deadlines).  I also have an article forthcoming in Religious Studies responding to Graham Oppy’s objections to the Aristotelian proof, and after working on that I was inclined to give the topic a rest for a while.  Hence my neglect of Schmid.  But the squeaky wheel gets the grease.  So, in hopes of appeasing the Schmid enthusiasts, this week I read his recent article “Stage One of the Aristotelian Proof: A Critical Appraisal.”  Let’s take a look at it.

Schmid develops three main lines of criticism.  The first is directed at my claim that for an ordinary substance to persist in being at any moment requires that something actualize it at that moment.  The second is directed at what I say about essentially ordered causal series.  The third is directed at my claim that the unactualized actualizer or first cause of the existence of things must be purely actual.  I’ll reply to these in turn.  I’m not going to repeat everything I say in Five Proofs or everything that Schmid says in his article.  So, what follows will presuppose that the reader is familiar with that material.

Concurrent actualization

Consider any ordinary substance that is a compound of actuality and potentiality, such as the water in a certain glass.  I claim in Five Proofs that “the existence of [such a substance] at any given moment itself presupposes the concurrent actualization of [its] potential for existence” (p. 35).  I note that there are several ways one could conceive of this actualization.  One could, in Aristotelian hylemorphist fashion, think in terms of prime matter’s potential to be water being actualized by the imposition of the appropriate substantial form.  Or one could, in Thomistic fashion, think in terms of the essence of the water having existence conjoined to it.  Or one could, in reductive naturalist fashion, think in terms of the particles that make up the water being made to constitute water, specifically, rather than some other substance.  For the purposes of this part of the Aristotelian proof, it doesn’t matter which of these models one goes with.

For expository purposes, let’s go with the last model, and think in terms of a collection of particles of type P which, considered just by themselves, could potentially constitute (a) water, or (b) separate quantities of hydrogen and oxygen, or (c) some other substance or aggregate of substances.  (I say “particles of type P” rather than making reference to atoms, quarks, bosons, or whatever, so as avoid getting sidetracked on questions about the particular physical and chemical facts, which are not relevant to the specific issue being addressed here.)

The basic idea is this.  Consider a collection of particles of type P which constitute water at time t.  Though they actually constitute water at t, there is nothing in the particles qua particles of type P that suffices to make them water rather than one of the other alternatives mentioned.  Again, qua particles of type P they have the potential to constitute water, or separate quantities of hydrogen and oxygen, or some other substance or aggregate of substances.  So, there must at t be something distinct from the collection which actualizes its potential to be water, specifically.

In response to this, Schmid writes:

Feser claims that it is the matter’s potential to exist as water that is presently ‘being actualized’. But ‘being actualized’ is arguably a notion of causal actualization.  Instead of claiming the matter’s potential is presently being actualized, then, a neutral description would say that the matter’s potential to exist as water is presently actual.

But not all actualities consist in or involve reductions of potency to act.  There are things that are (i) actual but (ii) whose actuality is not an actualized one – that is, not one consisting in the (concurrent) reduction of potency to act (by some causal actualizer).  For Feser, one example of this would be God… For those who do not already accept… that substances are concurrently reducing from potency to act in respect of their actual (substantial) existence – one example of an actuality that is an unactualized actuality may very well be the present existence of the water...

Now, notice that once we alter the phrasing to the neutral ‘the matter’s potential to exist as water is presently actual right now,’ we cannot straightforwardly infer the need for a concurrent, sustaining efficient cause of the water’s existence.

End quote.  Now, there are three problems with this.  First, Schmid is wrong to claim that my characterization of the situation is not neutral.  The implication is that no one who did not already agree with my argument would characterize what is going on as the potential of a collection of particles of type P to be water “being actualized” at t – because this would entail that there is a cause of this actualization, which is precisely what is at issue.

But that is not true.  Someone (a Humean, for example) could agree that the potential in question is being actualized at t, and still go on to claim that there is no cause of this actualization – that it just happens without anything making it happen.  To be sure, I don’t for a moment think that this would be a plausible claim (for the reasons I give in the book when criticizing Hume, defending PSR, and so on).  But that is beside the present point, which is that someone who does not already agree with the overall argument could nevertheless concede the claim that Schmid is criticizing.

A second problem is that Schmid’s proposed alternative way of characterizing the situation is incoherent.  For the claim that “the matter’s potential to exist as water is presently actual right now” (emphasis added) suggests that the collection of particles of type P is both potentially water and actually water in the same respect and at the same time.  But it is a well-known Aristotelian-Thomistic thesis – one which is famously given expression in Aquinas’s First Way, and which Schmid does not challenge – that nothing can be both potential and actual in the same respect and at the same time. 

My own characterization of the situation, unlike Schmid’s, does not imply otherwise.  Again, what I say is that at t, the collection of particles of type P considered just in respect of being particles of type P is only potentially water (and potentially other things too).  The collection is actually water only when considered as particles of type P together with the actualization of the potential in question (which, I also claim, requires a cause to make it happen – though, again, that is an additional thesis).  So, while I say that the particles are potential and actual at the same time, I do not say that they are potential and actual at the same time and in the same respect. 

Again, though, Schmid’s formulation incoherently suggests that they are potential and actual at the same time and in the same respect.  In particular, it suggests either (i) that considered just qua particles of type P, the particles are both potentially water and actually water, or (ii) considered qua particles of type P together with the actualization of the potential in question, the particles are both potentially water and actually water.  But neither of these makes sense.  If, as in (i), we consider the particles just qua particles of type P, then while they are potentially water, they are not actually water.  If instead, as in (ii), we consider the particles qua particles of type P together with the actualization of the potential in question, then they are actually water, but not potentially water.  (By the way, I am well aware that the last couple paragraphs might need to be re-read a couple of times in order to understand them!  Some of Schmid’s paragraphs are like that too.  That’s just in the nature of the subject matter we are debating and the subtle distinctions it involves, sorry.)

To avoid incoherent formulations like (i) and (ii), Schmid needs to put his point some other way.  The natural way to do it is to characterize the situation as one in which the collection considered just qua particles of type P is merely potentially water, but where the collection considered qua particles of type P together with the actualization of the relevant potential is actually water.  But that way of putting it would really amount to returning to my formulation after all, rather than offering an alternative formulation.

The third problem with Schmid’s criticism is in his glib suggestion that if God can be actual without being actualized, then – for all I have shown – the water too might be actual without being actualized.  For the view of Aristotelian-Thomistic philosophers (like me) is, of course, not that it is possible in principle for things in general to be actual without being actualized, but rather that it is possible only for something of a very specific type to be actual without being actualized – namely, for something that is purely actual and thus without any potentials standing in need of actualization.  Water is obviously not like that. 

So, whereas Schmid seems to be appealing to some point of common ground between us as the basis for his objection, in reality he is doing no such thing.  His objection presupposes that something other than what is purely actual might be actual without being actualized – a presupposition no Thomist would accept and for which he has given no justification.  So, the objection simply begs the question.

Schmid makes some further points in response to replies he imagines I might give to his objection.  Since, for the reasons I’ve just given, that objection fails, his further points are moot.  But I’ll briefly say something about some of them anyway.

He imagines, for example, that I might appeal to PSR as grounds for holding that the existence of the water at t requires some cause at t.  But in response he says that what happened prior to t plausibly explains the water’s existence at t.  Now, I am happy to concede that what happened prior to t is part of the explanation of the water’s existence at t.  But what is in question is whether what happened prior to t is by itself sufficient to explain the water’s existence at t.  Schmid says nothing to show that it would be sufficient.  Meanwhile, I argue in Five Proofs that it is not sufficient, and (as we have just seen above) Schmid’s attempt to undermine that argument fails.

Schmid also appeals in passing to the idea of “existential inertia” as a purported alternative explanation of the existence of the water at t.  But I have criticized atheist appeals to existential inertia at length (e.g. in this article and more briefly in Five Proofs at p. 233) and Schmid says nothing in reply to those criticisms.  (At least, he does not do so in the present article, which is the only one I’ve read.  But he ought to say something about them in the present article, since the article will beg the question otherwise.)

As I pointed out in one of my recent exchanges with Oppy, one problem with the kind of existential inertia scenario he and Schmid favor is that it is viciously circular.  Existential inertia would be an attribute of any substance that has it.  But attributes are ontologically dependent on substances.  So, Schmid’s proposal amounts to saying that the water’s existence at t depends on its attribute of existential inertia, and that its attribute of existential inertia depends on the existence of the water at t – a metaphysical merry-go-round.  (Only in the case of something in which there is no distinction between substance and attributes – that is to say, something strictly simple or non-composite – does this circularity problem not arise.  That is why God alone can have existential inertia.)

Schmid also quotes a passage in Five Proofs where I speak of the existence of the coffee in a certain cup as being “actualized” by the existence of the water that makes it up, where the existence of the water in turn depends on the existence of the particles that make it up, etc.  Here, he suggests, I am dubiously characterizing what are in fact the constituents of a whole as if they were the efficient causes of the whole.

I can see why Schmid would say this, given an uncharitable reading of the passage in question, which perhaps I ought to have phrased more carefully.  But that he should have read it more charitably is, I think, clear from the fact that I there said that the potential existence of the coffee is actualized “in part” by the existence of the water.  Naturally, the constituents of a thing qua constituents are not efficient causes, but material causes.  But what I meant in that passage is that the existence of the coffee is explained by the presence of its constituents together with something that actualizes the potential of those constituents to be coffee, specifically, as opposed to some other kind of thing.

Essentially ordered causal series

A causal series ordered per se or essentially is one in which the members other than the first have their causal power only in a derivative or borrowed way.  A stock example would be a stick that can move a stone only insofar as someone is using the stick as an instrument to move it.  The stick is a secondary cause insofar as it can do its causal work only if there is a primary cause – a cause with built-in or underived causal power – working through it.  The notion of this kind of causal series plays a crucial role in the Aristotelian proof. 

Schmid claims to offer an alternative account of the notion of an essentially ordered causal series which would not have the implications the Aristotelian proof says it has.  He suggests that a necessary condition on essentially ordered series is that there is some natural tendency or causal power toward a certain outcome that the primary cause operating through the secondary causes is counteracting.  He has in mind cases like the one in which, because of the gravitational pull of the earth, a stick would fall to the ground and lie there inertly unless a person comes along to counteract this gravitational influence by picking up the stick and using it as an instrument to move the stone.

But this is simply wrong.  What Schmid is describing is at most a contingent feature of certain specific examples of essentially ordered causal series.  It is not a necessary condition of all essentially ordered series as such.  For B to be a merely secondary cause of C, all that is required is that B lack any intrinsic power to produce C.  There needn’t be (though of course in some cases there could be) a countervailing factor (whether some tendency within B or some causal power external to B) positively acting to prevent B from producing C.  There need merely be the absence in B of any positive tendency to produce C.  A primary cause A need merely impart to B the needed causal power.  A needn’t, either alternatively or in addition, counteract something that prevents B from exercising the needed power. 

Why would Schmid want to suggest otherwise?  (And suggest is all he does.  He does nothing to show that the counteracting of some opposite tendency is a necessary feature of any essentially ordered causal series.  The most he does is to propose that this is a plausible way of interpreting certain specific examples.)

The reason is that Schmid wants to suggest in addition that the existence of something like the water in our earlier example will need a cause standing at the head of an essentially ordered series only if there is some factor positively acting to knock the water out of existence – a factor which the primary cause in an essentially ordered series would have to counteract.  And the presence of such a destructive factor is, Schmid says, something I do not establish – so that (given his analysis of essentially ordered series) my conclusion wouldn’t follow.

Schmid’s own alternative account of what is going on with the water is this: The water which exists at time t – 1 will, in the absence of some factor positively trying to destroy it, simply carry on existing at t.  A primary cause standing at the head of an essentially ordered series would be needed only if there were some destructive factor that needed to be counteracted.  Since there isn’t such a destructive factor, there is no need to appeal to such a series.

The problem with this, though, is that it once again simply assumes Schmid’s “existential inertia” model of the continued existence of the water – something which, again, I have argued against in the book and elsewhere, and which Schmid does nothing to defend in the present article.  In particular, Schmid will have to assume that model in order to make sense of the suggestion that the water will continue to exist at t in the absence of any destructive factor working positively to knock it out of existence.  So, yet again he simply begs the question.

It is important to emphasize that there is indeed a burden on Schmid to defend his existential inertia model, in order for his objections in the current article to have any force.  He seems to think that it is enough for him that I have not proved that there is a destructive factor positively working to knock the water out of existence.  And indeed I have not proved that, but I have not tried to, because it is irrelevant.  What I have done is argue against the existential inertia model, and if my arguments are correct, then the sheer existence of the water at t will need a cause even in the absence of anything positively working to destroy it. 

Schmid also claims that I have failed to show that, in the case of the sheer existence of a thing at some time t, the essentially ordered causal series that accounts for it is to be understood according to my analysis of essentially ordered series rather than Schmid’s analysis (where, again, the latter involves the claim that such series necessarily involve the counteracting of some tendency opposite to that of the actual outcome).

But in fact my critique of existential inertia does double duty here.  If the water lacks existential inertia, then it simply will not exist even for a moment, including at t, without a sustaining cause at t.  No factor needs positively to act to try to knock it out of existence; the mere lack of existential inertia will suffice for its failing to exist at t if there is nothing causing it to exist then.  So, if something does cause the water to exist at t, then this won’t be a matter of its having to counteract some factor that is trying to knock the water out of existence (along the lines of Schmid’s model of essentially ordered series).  Rather, it will be a matter of the cause actualizing something (the water) that simply would not otherwise exist at t whether or not there is a factor that needs to be counteracted.  In other words, this will be a scenario that fits my model of essentially ordered series, not Schmid’s.

Now, suppose the water does have a sustaining cause C, and that this cause too lacks existential inertia.  Then C is ontologically in the same situation as the water.  It too will simply not exist at t – and thus will not be able to cause the water to exist at t – unless it too has a sustaining cause of its own.  And once again, there need be no destructive factor that this further sustaining cause is counteracting (as in Schmid’s model).  Now suppose that what causes C to exist at t is B, and that B too lacks existential inertia.  Then the same problem will arise yet again.  And once again we will have a case where (contrary to Schmid’s model of essentially ordered series) the need for a sustaining cause has nothing to do with there being some countervailing force that the sustaining cause is counteracting.

Indeed, in the case of the sustaining causes of things which lack existential inertia, we have perhaps the clearest possible example of an essentially ordered series fitting my description of how such series operate rather than Schmid’s description.  Hence to rebut his existential inertia model suffices to rebut what he says about essentially ordered causal series.

(By the way, Schmid’s objections so far seem to me similar to the ones Oppy raised in his own critique of the Aristotelian proof, and which I responded to in our two online debates and deal with more systematically in my forthcoming Religious Studies article.  So, the hoopla of Schmid’s fans notwithstanding, it doesn’t seem to me that there is much in Schmid’s first two objections that really adds much to the exchanges online and in print that I have already had with Oppy.)

One last matter before we get to Schmid’s third and final line of criticism.  Schmid proposes in passing that there is a tension between the Aristotelian account of causation as the actualization of potential, and the classical theist understanding of creation.  Prior to creation, nothing exists other than God.  So how, Schmid asks, can creation be a matter of actualizing potential?  For prior to creation there is nothing there, distinct from God, with potentiality waiting to be actualized.  And God himself, being purely actual, doesn’t have potentialities waiting to be actualized either.  So how can creation involve the actualization of potential?  Schmid says that this is not a problem for classical theism as such, but rather for reconciling classical theism with the Aristotelian proof.

But Schmid’s mistake here is his implicit assumption that causation as such requires some preexisting substrate that is altered in the act of causation.  And that is, of course, precisely what the doctrine of creatio ex nihilo denies.  Hence if Schmid’s alleged problem really were a problem, it would indeed be a problem for classical theism as such (since classical theism is committed to creatio ex nihilo) and not just for Aristotelian versions of it.

But in fact it is not a problem.  Certainly no Thomist would agree that it is, given the Thomist account of creation as the conjoining of existence to essence, where the latter, considered by itself, is merely potential until the former actualizes it.  Now, this is not a matter of altering some preexisting substrate, since prior to creation there is no substrate.  When we draw hydrogen and oxygen out of water, there is something already there – the water – in which the things we are drawing out preexist in a virtual way.  But creatio ex nihilo is a more radical kind of causation than that.  Actualizing the very essence of a thing by conjoining existence to it is analogous to actualizing matter’s potential to be hydrogen or oxygen, but it is not exactly the same sort of thing as that.  We need to extend our use of the relevant terms (“potentiality,” “causation,” etc.) beyond their application to the sorts of case in which the terms were originally applied (i.e. cases in which a preexisting substrate is altered).  There is nothing unique to Thomistic natural theology about this.  It is precisely the sort of thing we do in physics when, for example, we extend our use of the term “curvature” to apply to space itself (whereas in its original usage, it applies only to the objects that occupy space). 

Naturally, there are crucial assumptions here – concerning the Thomistic metaphysics of essence and existence, the Thomistic theory of the analogical use of terms, and so on – that require further elaboration and defense.  The point, though, is that Schmid is hardly raising some issue that no one ever thought of before.  On the contrary, there’s a mountain of stuff written on it that Schmid’s remarks simply ignore.  Hence those remarks hardly constitute a serious objection.

(Compare: Suppose I remarked, in an article critical of materialism, that it is difficult to see how consciousness could be explained in materialist terms, and left it at that.  Would that be an interesting objection?  Of course not.  A materialist could justifiably respond: “Well, maybe so and maybe not, but surely you realize that there’s been an enormous amount written on how such an explanation might go!  Do you have anything to say in response to it?”  That’s how Schmid’s remarks on creation are bound to sound to a Thomist, or indeed to any defender of creatio ex nihilo.)

The purely actual actualizer

Schmid’s final main objection is to claim that, even if it is granted for the sake of argument that the sheer existence of the water at t requires a sustaining cause, it doesn’t follow that this cause would be purely actual rather than a compound of actuality and potentiality.  In particular, he claims that all I am entitled to conclude is that there is a first actualizer at t whose own existence is not in fact being actualized by something else.  But that is consistent with the supposition that the existence of this first actualizer could in principle be actualized by something else.  And if it could be, then it would have potentiality, even if it is potentiality that is not being actualized at t.

But this simply makes no sense.  Naturally, if the first actualizer is operating at t, then it must actually exist at t, and not merely potentially exist at t.  But in that case, then (if it is not purely actual) how can it have some potential to exist that is not being actualized at t?  For if such a potential were there but not being actualized at t, then the first actualizer would not exist at t, and thus not be causing (or doing anything else) at t.  Yet if such a potential is being actualized at t, then we are not really talking about the first actualizer after all, since in that case there would be something distinct from it that is actualizing its potential to exist (and that other thing would be the true first actualizer). 

Or is Schmid saying that the first actualizer’s potential to exist at t is actualized, but that there is no cause that is doing the actualizing?  That can’t be right, because in this third objection, Schmid is, at least for the sake of argument, conceding the principle that the actualization of potential requires a cause.  (Or, if instead he rejects this principle, that would really just take us back to his first objection to the Aristotelian proof, rather than constituting a third line of criticism.)

So, the objection seems to me to be a muddle.  The subsidiary points Schmid makes in the course of developing it aren’t much better.  For example, he says that, even if the first actualizer were purely actual with respect to its existence, it might still have potentialities in other respects (for example, with respect to changes it might undergo).  But the problems with this suggestion should be obvious from other things I say in Five Proofs.  For one thing, if the first actualizer has potentialities even of the sort Schmid suggests, then it will be composite rather than simple.  But, Thomists argue, anything composite requires a cause, in which case this actualizer will not after all be purely actual even with respect to its existence.  For another thing, the Scholastic principle agere sequitur esse (“action follows being”), which I defend in the book, entails that the manner in which a thing acts reflects the manner in which it exists.  Hence, if something acts only by way of actualizing potentialities, then it would exist only by way of actualizing them; or, if instead it exists without the actualization of potentialities, then so too it acts without actualizing them.  No doubt Schmid would disagree with all this, but the point is that his objections simply presuppose that it is wrong, and do nothing to show that it is.

Schmid also oddly claims that my own position unjustifiably “presupposes the impossibility of changeable necessary beings” (emphasis added).  But in fact my position presupposes no such thing.  Rather, it claims to demonstrate the impossibility of changeable necessary beings.  (It seems to me that there may be a kind of unintentional rhetorical sleight of hand in Schmid’s remark.  If someone claims to show that X is impossible – whether X is a round square, two plus two equaling five, a changeable necessary being, or whatever – a critic could always say: “Well, your argument is correct only if X is indeed impossible.”  Which is true, but trivial.  It hardly entails that the argument presupposes that X is impossible!)

Now as Schmid acknowledges, the charge that a first actualizer need not be a purely actual actualizer is in fact one that I anticipate and respond to in Five Proofs (at pp. 66-68).  He quotes a remark I make there to the effect that “the first actualizer in the series is ‘first’, then, in the sense that it can actualize the existence of other things without its own existence having to be actualized… in order for it to exist” (p. 66, emphasis added).  Schmid responds that his scenario is not one in which a first actualizer has some potentiality that has to be actualized in order for it to exist.  Rather, he is simply claiming that it is one in which a first actualizer needn’t in any sense be purely actual.  For example, it might have at t a potential with respect to its existence that is not in fact actualized at t. 

But this simply misses the point I made above.  If a first actualizer has at t a potential with respect to its existence, then it simply will not exist at t unless that potential is actualized.  Hence its potential would indeed have to be actualized in order for it to exist.  Again, Schmid’s scenario simply makes no sense.

I appreciate Schmid’s interest in the argument and his attempt to engage with it seriously.  However, on close inspection the attempt seems to me to be riddled with confusions, begged questions, and missed points.