First Article of the First Question
1.1.1 On Actuality and Potentiality in General
We have actual being, and there actually are things around us. It is a significant and not a trivial thing to say that something is actual, a desk, perhaps, or a wall, (1) because we contrast actuality with something, and (2) because things that are actual are liable to act. Let us take these two points in order.
(1) Actuality has something with which it can be contrasted. If I say that a chair actually is, or is actual, or that it actually is blue, this can be contrasted with claims that the chair does not exist, or that the chair is only potentially blue, or the like. As St. Thomas says (PN 1), Note that sometimes a thing can be although it is not, whereas sometimes a thing simply is. That which can be is said to be potentially; that which already is, is said to be actually. Likewise, if we recognize that something red is potentially blue and later come back, we may see that it is now actually blue; we take these to be closely related. Moreover, whether or not something is potentially blue depends on what it actually is. We thus associate actuality with two things, albeit in different ways, namely, being and potentiality, and contrast it, albeit in different ways, with two things, namely, nonbeing and potentiality, and that potentiality is both in some way associated with actuality but in some way contrasted with it is important for understanding the world around us.
Using the principle of noncontradiction itself in a changing world requires some recognition of this; for if we say with Aristotle (Metaph. 4.13..1005b) that "it is impossible for the same thing to belong or not to belong at the same time to the same thing in the same respect", it is clear enough that we could equally say that it is impossible for the same thing to be actual and not actual at the same time to the same thing in the same respect. But applying this consistently to changing things requires us to recognize that a caterpillar is able to be potentially a butterfly while not actually being a butterfly; that is it is not merely a non-butterfly, but is, in some way, a butterfly despite not being an actual one. If you ask, "Where is the butterfly?" and I point to the caterpillar, saying, "It is there, still a caterpillar rather than a butterfly," I have told the truth, not contradicted myself by saying that the butterfly is not a butterfly, simply speaking. Likewise, and on the other side, if I ask, "Where is the caterpillar?" and you point to the butterfly it became, saying, "The caterpillar is a butterfly now, not a caterpillar" you have indicated the truth; it is the actual butterfly of which the caterpillar was the potential butterfly, not contradicted yourself by saying that the caterpillar is not a caterpillar, simply speaking. This is quite different from, say, trying to say that a caterpillar is an eel.
To some extent, then, we can say that anything that is not actual is not, or does not exist, but it clearly will not do the leave the matter at this. Some things can be but are not, while others actually are. Those that can be are said to be potential, or to be potentially, whereas those that are, are said to be actual, or to be actually. In other words, being potentially actual cannot be treated as if it were simple non-being. This is because what is potential must in some way be actual. An actual caterpillar is a potential butterfly, an actual acorn is a potential oak, and so forth. It is the actual that makes the potential identifiable at all; when we want to find the potential we look, as we must look, to what we know to be actual. As St. Edith Stein says (PA 104), Potentiality can only be understood from the viewpoint of actuality. Because of this the potential cannot be conflated with nonbeing; we can say of being that it is divided into the actual and the potential. Thus St. Bonaventure says (Itin. 5.3), anything that is understood is understood either as nonbeing, or as potential being, or as actual being. This is of considerable significance, for a common way of going wrong in discussing change, causation, and the natural world is to assume that all things understood are understood only as nonbeing or as fully actual being. This severs much of the order and relatability of one thing in the natural world to another, and often leaves things inexplicable.
If we needed confirmation for this, we could easily look to our explanatory practices, all of which refer to the actual and the potential in some way. To explain why a pig does not sprout wings and fly, for instance, we need nothing more than to identify those actual features of a pig that are inconsistent with its having the potential to sprout wings and fly, and to show that it does not actually have the features by which it would have the potential to sprout wings and fly. This is indeed a quite common pattern. If I wish to explain how a population of lizard-like creatures became a population of rodent-like creatures rather than a population of different lizard-like creatures over millions of years, I need to identify the actual features of the original creatures and the actual features of their likely environment that made it so the lizard-like population had the potential to be the rodent-like population, and I need to show either that features that would have led to being the other lizard-like population were not there, or that the environment did not have the features that would have led to one population becoming the other. Predictive practices, where we are not simply guessing, work the same way. A reasoned prediction that a chemical reaction will result in salt is well-founded to the extent that it is based on being able to identify what is actual, what is potential, and what is simply not actual. If I fail to recognize one of the actual chemicals with which we begin, or if I am mistaken about the ability of those chemicals to combine, or if I am assuming that a chemical is actually there that is not, any of these three can cause problems for my prediction.
(2) Thus we come to the second point, which is the link between acting and being actual. To act or be active is a sort of expression of actuality itself; it is when something actual is not merely actual in itself but in such a way as to be to or in or through or with something. This may be by way of changing something, as when a man, walking, is actual in and through the potential of his legs; or it may occur in some other way, as when a man remembers a dog he once had, and thus can be said to be actual in a way that is directed to the dog. In this we can see several things. First, nothing can act without being actual. Second, everything acts in the mode and manner in which it is actual.
Given what we have concluded about actuality and potentiality, we can use it to discuss three important features of the world as we find it: change, composition, and the ability to be. All three of these starting points lead to some actual cause that has features that people take to be divine.
1.1.2 On Actuality and Potentiality in Change
The difficulty of giving an adequate account of change is partly due to the difficulty of identifying something that is clearly prior to it; but it is also certain that change admits of analysis. It is for this reason that some have attempted to account for change entirely by the same and different, while others, following Aristotle, who saw the limitations of this approach, account for it in terms of the actual and the potential.
It is extremely clear that change has something to do with the same and the different; even discussing the question, "What is change?" with students will quickly raise further questions about sameness and difference in the context of change, and it is not surprising that Plato in the Timaeus describes the changing universe as circles of difference within the circle of the same. If I dye the shell of a white egg blue, first it is white, not blue, then it is blue, not white, and it is the same shell. The change involves a difference (white/not-blue, blue/not-white) and something the same. If nothing were ever different, it would be absurd to say it changes. On the other hand, though, not every difference is a change. If I walk into an ordinary room and cry out that the wall is changing into the ceiling because they are not the same, you will either assume you misheard, or that a misspoke somehow, or that I have gone mad. They are simply different; that your nose is not your toe is not a reason to say that one is changing into the other. Likewise, if I tried to claim that change was entirely constituted by different things at different times, this would cast too broad enough; a shoe yesterday and a sun tomorrow are different things at different times, but one does not change into the other. The shell, on the other hand, changes, because the difference it has is the difference of the same.
How to make sense of this paradox, that the changing thing must be both different and the same, is, as Hume rightly notes (Treatise 1.4.3), one of the fundamental questions of ancient philosophy, or, indeed, of any natural philosophy at all. It seems that there are only a few possible ways in which it might be handled.
(1) One might hold that changing things are intrinsically the same but only extrinsically different. On such a view that of which our minds aware actually has no differences at all, but our minds attribute differences to them. This position is sometimes associated with Parmenides. One obvious problem is that it seems that making sense of our awareness of anything requires that there be differences in changing things, e.g., in the mind's own changes. Nor can we make sense of our sensible experience by saying that all differences we sense are simply attributed by the mind.
One could perhaps have a different version of a position in this family by arguing that we have change is when the same thing exists at different times. But whatever the exact relation between change and time, it is clear that we can only identify different times by comparison, and in particular by something capable of functioning as a clock. However, for a thing to function as a clock requires that it be a regular change, i.e., a change for which we can give a general rule. We might use as a clock a mechanism of precise gears, or an oscillation of a physical medium, or the reflection of light between mirrors, or any number of other things, as long as they are changes for which we can give some rule by which we can apply them as measures of other changes. Thus the claim that we have change when the same thing exists at different times seems equivalent to the claim that change is difference as measured by change, which is a faulty and circular account.
(2) One might hold that changing things are intrinsically different but only extrinsically the same. An example would be Hume's own account, in which that of which our minds are aware are all distinct and separate, but our minds attribute a unity to them. One obvious problem with this is that the mind itself is a changing thing, and to make sense of this position for other things, it does not seem that mind's own unity can be merely attributed. Further, to describe changing things as constituted entirely out of discrete separate immutable bits is an entirely implausible account of our experience of change, which requires, for instance, that the change that is going on between light and visual system when we see something is somehow unified.
(3) One might hold that changing things are intrinsically different and intrinsically the same, in different respects. This is the path taken by Aristotelians and many others, but it is obvious that on this position simply saying that change is the difference of the same is not itself adequate. The next question that therefore needs to be asked is about the different respects by which something may be both intrinsically different and intrinsically the same. Actuality and potentiality provides the way to do this.
Being can be divided into the actual and the potential, but some things may be midway, so to speak, between the actual and the potential by being in one way actual and in another related way potential. In changing things, what is merely potential is not yet changed; what is already actual has been changed; what is being changed is partly potential and partly actual. Thus change is neither the potentiality of what is potential nor the actuality of the actual. Instead it is the act of the potential precisely insofar as it is potential to some actuality. For instance, clay is able to be shaped, and this implies the potential for shape. But as something actually draws this potential into act, it is being shaped; by its nature the thing was potential both to complete act, in which the potential is fully realized, and to incomplete act, in which case it is being realized.
Thus change involves a twofold ordering to potentiality and a twofold ordering to actuality. A changing thing is thus actively potential, in such a way that its ordering to prior potential is called 'act' and its ordering to further potential is called 'potential being'. Water that is being heated is ordered as act to a prior potential state, and as potential to a further potential state. Or, to use the example of St. Edith Stein (PA 12), My present being contains the possibility for future actual being and supposes a possibility in my earlier being, and (FEB 38-39), My present being is simultaneously actual and potential being; and insofar as it is actual, it is the concrete realization of a possibility which antecedes my present actuality. Everything potential, by the same token, is potential to two acts, an act that completes the change and the act that is the change itself. When we put water on the stove to boil, it is potential both to being heated and to being boiling hot. Thus we see the rationale of Aristotle's definition of change as the act of the potential insofar as it is potential.
Changeable things, then, are something actual in such a way as to have a relative possibility, or potentiality, to be some other actuality; change is acting in precisely this way that they are relatively possible. However, as we have noted, nothing acts except insofar as it is actual because acting is the expression of actuality; but potential things, insofar as they are potential, are not actual. Therefore in change, the act of the potential must be the act of something actual as well; that is, change is the expression of the actuality of one thing in something potential. As Aquinas says (ST 1.75.1 ad 1), Since to be changed is to pass from potential to actual, the changer gives what it has to the changed, inasmuch as it causes it to be actual. Nothing, however, can be potential and actual in precisely the same respect, so in any particular change, the thing that is undergoing change is distinct from the thing that is changing it. Even in things that we think of as capable of changing themselves, the actual parts are what act in the change of the potential parts. Thus Maimonides says (Guide 2.1), "We constantly see things passing from a state of potentiality to that of actuality, but in every such case there is for that transition of a thing an agent separate from it." However, in the change itself there is also a kind of unity, since the act of what is changed is the act of what changes it; in Aquinas's words (In Phys. 3.4), Change insofar as it proceeds from the changer to the changeable is the changer's act, but insofar as it is in the changeable from the changer, it is the changeable's act. Without something like this, there is no tie between cause and effect.
We could look at it in another way, by recognizing that 'whatever is changed is changed by another' is equivalent to another claim, when it is properly understood, namely, 'Whatever begins to exist has a cause'. Just as it could be understood in terms complete and incomplete act, so also change can be understood in terms of complete and incomplete being, for, as Stein says (PA 12): As modes of being, actuality and potentiality are contained in the sheer fact of being and from it they are to be inferred. If something is changing, it is thus far not a complete being, but incomplete being. Thus a caterpillar, while it has complete being as a caterpillar, in changing to be a butterfly is incomplete butterfly. To say that something is beginning to be is to say it is incomplete being; thus to say that something is beginning to be is to identify a change. What is more, such a change requires a principle prior to what is beginning to be; as Stein says (PA 127), What begins to be can only be brought about by an activity that goes forth from something actual. Thus what begins to be is changed and, as it is changed by another, that other is what we call a cause of it.
There has been some discussion of the principle in the form 'whatever begins to exist has a cause'. Hume claims (Treatise 1.3.3) that this principle is neither intuitive (i.e., determinable by considering the ideas) nor demonstrative. (He does not actually reject the principle; he is only arguing that it can only arise from custom by experience.) He has very little argument directly for this, although he does give two weakly developed arguments and responds to some arguments otherwise. He takes it to be the case that all intuitive principles must involve four and only four relations: resemblance, contrariety, degrees in quality, and proportions in quantity and numbers. His reason for denying causation a place in this list is just a flat claim that "the power, by which one object produces another, is never merely discoverable merely from their idea" (T 1.3.1 SBN 69), so that we could not predict any specific phenomenon without the help of our memory and experience. As Shepherd will note (An Essay upon the Relation of Cause and Effect pp. 58-60), Hume often confuses two different questions, the general question about whether one object requires the existence of another (which is the question of the relation itself) and the particular question of how and why it does so, and this is a good example of him doing so. In reality, when we think of something genuinely beginning to be, that it is from something strikes the mind immediately, without requiring second examination. Moreover, Hume's manner of arguing confuses the directions; 'whatever begins to exist has a cause' starts with effect and concludes to there being some cause or other, but his only argument for why causation cannot be an intuitively discerned relation is the claim that if we start with a specific cause we cannot conclude to the specific effect without the help of memory and experience.
However, his primary argument on this point is based on the recognition that 'whatever begins to exist has a cause' is equivalent to 'nothing begins to exist without a productive principle'; that the latter is neither intuitive nor demonstrable he argues on the claim that everything distinct is separable, so as the idea of cause is distinct from the idea of effect, it is easy to imagine an object first nonexistent, then existent, without thinking of a cause. This argument as stated, of course, confuses idea and object, as Hume often does, and assumes falsely that we cannot have different ideas of the same objects, whereas these are quite common, as when one bidirectional road can be thought of in the leftward direction and thought of in the rightward direction. Thus not everything distinct is separable. It further confuses change and difference; that is, it has to assume that beginning to exist is the bare difference between nonexistence and existence in different times, whereas most cases that we label as 'beginning to exist' are cases in which one existing thing is turning into another thing at a given time. And, of course, it is absurd in itself to hold that a thing might not have a cause merely because you do not think of it, in the same way that it is absurd to hold that a number might not have a successor because you don't think of one. Each number is distinguishable; therefore each one on Hume's position is separable; therefore each one might be thought alone; therefore, you can think of the number 2 without thinking of the number 3; therefore there is no intuitive or demonstrable connection between the two. Or: two resembling things are distinct; therefore they are separable; therefore each one can be thought alone; therefore there is no intuitive or demonstrable connection between the two resembling things. Or: one thing is on the left and the other is on the right; left and right are distinct; therefore they are separable; therefore they can be thought alone; therefore there is no intuitive or demonstrable to deny that something could be to the left without anything being to the right of it. Or: what is before and what is after are distinct, therefore separable, therefore imaginable alone, therefore it is not absurd to say that something comes before yet has nothing after it. This is not a manner of argument that gives confidence in its ability to give a just assessment of any relation at all. The fundamental problem, of course, as is seen by the bare contrast of nonexistence and existence, is that Hume is attempting to account for changes entirely in terms of nonbeing and actual being; he has no room for potential being and therefore cannot relate things except by associations in the mind.
Moreover, as Shepherd recognizes, Hume is wrong to think that there is no analysis of the situation in which 'nothing begins to exist without a productive principle' yields a contradiction. For assume, she says (ERCE, p. 35), that we have something beginning to exist without a principle, "what is this starting forth, beginning, coming into existence, but an action, which is a quality of an object not yet in being, and so not possible to have its qualities determined, nevertheless exhibiting its qualities?" That is to say, "the notion of beginning an action (the being that begins not supposed yet in existence) involves a contradiction in terms" (ERCE, pp. 35-36). All beginning to exist involves something being new; something new is a difference; introduction of a difference just is what we call causation; the difference cannot be introduced by what presupposes the difference. This is a line of argument that we find independently in Bl. Antonio Rosmini, who argues (New Essay concerning the Origin of Ideas, 5.3.2, sect. 569) that to say that what does not exist, acts, is a contradiction. But to say, A happening is without a cause, is equivalent to saying, what does not exist acts. Our conception of an action is always of something acting; but if we are speaking of something beginning, that is an action. Thus, he continues, If, then, this action has no cause, it is conceived as isolated, without any being to which it belongs. It is therefore an action without being, or which is the same thing, that acts which does not exist. He likewise says (NECOI, sect. 352), in responding to Kant, that the steps are: coming into existence - change - new operation - prior existence - cause. What begins to be is changed, so involves a new action that must be the action of something acting, namely, the cause.
Hume attempts to strengthen his brief argument by showing that common arguments, which he attributes to Hobbes, Clarke, and Locke, fail. The Hobbesian argument notes that points of time and space are exactly like; therefore for something to begin to exist at a particular point in time and space requires that there be something that fixes at which point in time and space it begins. The Clarkean argument is that if anything lacked a cause, it would have to produce itself, which is a contradiction. The Lockean argument is that if anything lacked a cause, it would be produced by nothing, which cannot be a cause. Hume gives the same answer to all of these: they beg the question by assuming that there must be a cause, whereas the supposition being considered is that something begins to exist without any cause at all. However, as is clear from what has already been said, and as Shepherd notes, Hume is not being fair to either the Clarkean or the Lockean argument. As Shepherd notes, they were not actually trying to give direct demonstrations of the principle itself. Rather, they "showed, that the mind of man was forced to look upon all things which begin to exist as dependent QUALITIES; and thus, that an object could neither depend upon itself for existence, nor yet upon nothing" (ERCE, p. 37). She also notes that Hume's own argument can be accused along similar grounds, since he simply assumes that he can imagine the effect without implying the cause.*
One of the things both Hume and Shepherd recognize is that 'whatever begins to exist has a cause' is the fundamental principle of experimental reasoning; one of Shepherd's criticisms is that Hume's account sharply and implausibly limits what can be inferred from experimental evidence. If we look at experiments, however, we can see that understanding this in terms of actuality and potentiality is capturing something essential to understanding experiments. In an experiment, one introduces something to processes and structures within some boundary so as to achieve a change that can be measured for the purposes of inquiry. This requires that what is within the boundary of the experiment have the potential to be in some way different; performing the experiment is activating that potential, which is the measurable change. There would be no point to an experiment involving a cloud chamber if it had no potential to be affected by anything; if one cannot infer from the droplet-trails that are formed in the alcohol vapor that there is some actual charged entity forming it, that the droplet-trails in being formed are formed by the expression of the actuality of something other than themselves, then it would not be possible to get very far. Indeed, any sensory observation is itself a change in our sensory organs that is caused by something, so it is not possible to do any observation at all without recognizing that when our sensory organs are changed, they must ultimately be changed by something other than themselves.
The practical cost to inquiry of denying that everything that is changed is changed by some other is thus very high. To deny it is to posit things that come to be simply of themselves or without any cause at all. But in all argument we must be consistent; an argument is not like a cab that can be stopped at any destination, but any argument we make carries us all the way to the end. If it is possible for things to come to be without a distinct cause in one case, then any other case of becoming could possibly be the same way.
On the other side, the benefits of accepting the principle seem great; as Whewell says (On the Motion of Points Constrained and Resisted, p. ix), "the fundamental principle of science" is nothing other than "No change can take place without a cause", which is equivalent to our principle that what is changed is changed by another. As he notes elsewhere, this provides a general structure for inquiry. It is still necessary to apply it to something, and indeed, to apply it to a change properly understood. We have to be careful to avoid confusing real change and merely apparent change, which may sometimes not always be easy to determine. Even when we know that there is a real change, it is not always clear what is undergoing it, since sometimes it may be the thing observed, sometimes our relation to it, sometimes our instruments, and sometimes ourselves. But these things are determined by repeated experience and testing and experiment. Once we have a well-defined kind of change, we can apply the principle. For instance, if we take bodies that change their direction or that accelerate, we can apply the principle "What is changed is changed by another" to get "[Bodies] that are changed [from uniform travel in a straight line] are changed by another." Further experiment can be done as to what kinds of 'other' are reasonable causes; eventually, by considering bodies causing other bodies to move, we might abstract a notion of 'force', and we might rule out other causes (for instance, we might rule out that time alone, or space alone, can cause such deviations). Then we could specify further, "[Bodies] that are changed [from universal travel in a straight line] are changed by [a force, i.e., the action of another body]." Further experimentation might adjust this up or down in various ways. For instance, we could discover that the principle could be applied more generally than just to bodies, or develop a more general conception of force such that action of a body is only one specific kind. We could also potentially discover that our principle, which covered so much, nonetheless does so only approximately -- perhaps there is an effect on travel by bare time and space that our previous instruments could not register, or perhaps there are rare situations not previously considered in which the deviations could be caused by something other than a force.** Whatever our experimental results may show, our inquiry will still be guided by the essential principle, whether we state it as "Whatever is changed is changed by another" or "Whatever begins to be must have a cause" or "No change can take place without a cause."
On the basis of all that we have said about actuality and potentiality in change, we can conclude that there is a first source of change that changes other things without anything changing it; this is what is traditionally known as a first unmoved mover. We may summarize the argument as follows.
(1) Something is changed. It is clear that some things in the world are changed, and we must accept such claim if we are to have any account of the world that is consistent with the senses.
(2) What is changed is changed by something other than itself. What is changed is so precisely insofar as it is potential and is not actual; for being changed just is to be something actively potential. But the 'actively' is important, too; what is changed also cannot be merely potential, because being changed is itself an act. What changes something, however, must to that extent already be actual, for the act of the potential must be an act of something actual. As Bonaventure says (Myst Trin 1.1), if potential being is, actual being is, for nothing potential is reducible to act save by actual being, nor would there be potential if it were not reducible to act. But it is not possible for the same thing to be at once actual and potential in the same respect; thus one and the same thing cannot be a changer and be changed in the very same respect, which is to say, it cannot change its very self. While we can say loosely that something changes itself in the sense that an actual part of it activates a potential part of it, there are no strict self-changers, in which a potential part, precisely insofar as it is potential, acts on itself so as to make itself actual.
(3) Some things change other things precisely because they themselves are changed by something. If, however, what is changing something is doing so because it itself is changed, the act by which it is changed will itself be due to something else, in such a way that the changed is changed so as to change another by a changer that is changed so as to change another by yet another changer, and so on.
(4) It is not possible for a series of such changed changers to go back infinitely. This cannot, however, go back infinitely in such a way that there is no principal changer, because if we do, we get contradictions in which there are things that are both changed and unchanged. This is worth considering a bit more closely.Suppose we have something being changed. We can call it Z. Since it is changed, it has to have a changer, or source of change, which we can call Y:
Y → Z
Now if Y is causing Z to change because Y is itself undergoing change, then it has to have a cause that changes it in such a way that it changes Z:
X → (Y → Z)
We can extend this back, and there is particular length we can rule out for a causal chain:
R → (S → (T → (U → (V → (W → (X → (Y → Z)))))))
If we are considering this, we can identify as our changed effect either Z, or Y and Z together, or X and Y and Z together, or W and X and Y and Z together, and so on back. And for each of these, it is changed by another, and we can identify this cause of change; for instance, W is the cause of that change which consists of X changing Y to change Z. But then we can take together all things in the series that are changed; this whole series, in fact, is something changed. What is changed is changed by another; so there has to be something that is responsible for the whole change that is the infinite series. But the infinite series is a regress without a first changer, and therefore we have something, the infinite series, that is both changed and not changed.
We can divide the matter a little differently. Any means by which a changer changes the changed must both change and be changed, for it is changed by the principal changer, and it moves the further thing that is changed. For this reason every changed changer in the proper sense, i.e., in the sense that it is changed in such a way that it changes something else, is what might be called an instrument, or, as it is also called, a moved mover; it has something in common both with its changer and with what it changes, so that its changer changes the thing it changes through it. Everything that is changed by another is a sort of instrument of the first changer, as Aquinas says (CT 3). So if we consider the previous series, we can, instead of focusing on the effect, focus on the cause. Thus Y is instrumental to X in causing Z:
X → [Y] → Z
If X is instrumental to W, then X and Y together form an instrument for W:
W → [X → Y] → Z
Thus we can continue it further, as far back as necessary:
R → [S → T → U → V → W → X → Y] → Z
In an infinite regress, there is no first, and if there were no first, every changer would be an instrument. But a series of instruments changing other instruments so that they are actually instrumental is itself an instrument; that is, we can treat any series of instrumental causes that are instrumental to other instrumental causes as an instrumental cause. If we claim that the series of moved movers is infinite, then, claim that there is a changed changer (the whole instrumental series) that has no cause prior to it, an instrument that is not instrumental for anything. But everyone can see that it is impossible to claim that something is an instrument and not an instrument, or something that is changed and not changed, which is a contradiction.
We could put the same point in a slightly different way. If X changes Y in such a way that Y changes Z, then it is clear that Y is an intermediate cause between X and Z. If X is changing Y because it is also being changed by Z, then it is clear that X is an intermediate cause between W and Y, and also that X and Y together are serving as an intermediate cause between W and Z. It makes no difference whether there is only one intermediate cause or several; all intermediate causes can be taken together insofar as they share the nature of intermediacy. Nor does it matter whether there are finite or infinite intermediate causes; so long as they share the nature of being intermediate they are taken together as intermediating between first and last. However, if there is an infinite regress, there is no first, and therefore we have an intermediate cause (i.e., all the causes prior to the last effect) that is not intermediate, which is a contradiction.
We can also look at the matter by considering the series as a whole. In an infinite regress there is no first. But a series of changers, in which something is changed by a cause precisely insofar as the cause itself is changed by another cause, and so on, can only get its character as such-and-such a series rather than some other from something other than itself, which will be a source of change constituting the series itself by its action. Indeed, even when not tracing back to a very beginning, such as when we are just looking at a particular order or category of motion, we still trace it back to something that is relatively first. Thus the ball is moved by the stick's being moved by the arm's being moved, and we always understand the series in terms of something that is taken to be relatively first. If the relatively first is the choice of the person who has the arm, that is one kind of change; if the relatively first is another person's hand pushing the arm, that is another kind of change; if the relatively first is a physical object hitting the arm, that is yet another kind of change; and this is true, despite the fact that ball, stick, and arm are shared by all three very different changes. There is no principled reason why this would not still be true regardless of how far you push it back. As an actual change, it has to have a particular character; it gets its particular character from something other than itself. Thus if there is no first source, there can be no definite character to the change by which all the changers act, and there will be no changers at all.
Likewise, we can see that circularity of sources of change is not possible because that would simply be a particular case of infinite regress, and the same arguments would apply; besides which, a strict circle of changers would involve every changer changing other things in such a way as to cause that very act of changing, which is a contradiction.
(5) Therefore there is a first unmoved mover. Therefore it is necessary to come to some primary source of change that is itself not changed by anything. As Aquinas says (ST 1.75.1 ad 1), Since everything changed must be changed by something else, a process which cannot be prolonged endlessly, we must allow that not every changer is changed.
This is all very much what one would expect. First, it is antecedently credible. There can be changed things that don't change other things (moved nonmovers) and changed things that change other things (moved movers), so unless there is some sign of a contradiction in it, it would be reasonable to suspect there to be something unchanged that changes other things (unmoved mover); but there appears to be no contradiction in the notion of such a thing. The denial of it would directly imply that the causes of every change infinitely regress, and thus that every change has infinite causes; this quite clearly goes beyond any experience we have, which could never on its own underwrite the claim that any change has infinite causes, much less that all do. The denial of an unmoved mover would therefore have to be based on some immediate proposition that can provide a self-evident ground for the claim that nothing can change another without itself being changed so as to do so. This implies that nothing can act without being acted upon, that every action is a being acted upon, and therefore appears to confuse actuality with potentiality. Thus we already have some reason to doubt that there could be an infinite regress in sources of change to begin with.
Reaching a first unmoved mover, however, one has reached something that has a description that is generally regarded as divine. Such a first mover must be purely actual in its causing, for it acts without being changed so as to do so; but what is purely actual cannot be changed by anything at all, because that would require that it be potential to something. Some attempt to wiggle out of this by suggesting that perhaps the first mover would be purely actual only in that respect in which it is causing. However, this is not to the point, for the first mover as such would then be the actual part, regardless of what arbitrary hypotheses one might make up about what potential parts might somehow be united to it. However, the hypothesis is also not possible; in unions of actuality and potentiality, the actuality is from another, and therefore to posit such a union is to posit something that is not first. We will see this is even more true when consider as well composition and capability for being. Thus the first mover, as such, is simply immutable. It is, moreover, unsurprising to find mutable things traced back in some way to something immutable in some way, because the typical order of explanation is from the more variable to the less variable, from the varying to the invariant.
A cause that is simply immutable is clearly what people regard as divine, and likewise it is common to characterize God as immutable. As St. Edith Stein says (PA 70), all mutable being, all becoming, points to an upholding outside of itself, to something immutable, to absolutely actual being, and Augustine says (De nat. bon. 1), God alone is immutable, and whatever beings He has made, being from nothing, are mutable. St. Thomas Aquinas echoes this, saying (ST 1.9.2), God alone is altogether immutable; whereas every creature is in some way mutable. Lady Mary Shepherd likewise says (Essays on the Perception of an External Universe, p. 397), "Changes, effects, require their proper causes, but not the mighty Being, which is no change, no effect, who is self-existent.” Likewise, the First Council of Nicaea anathematizes those who say the Son of God can be changed or altered and the third letter of Cyril to Nestorius, accepted by the Council of Ephesus, says the Word of God is unalterable and absolutely immutable and remains always the same, and his letter to John, accepted by the same Council, says that God is immutable by nature. The same is explicitly stated by the Fourth Council of Constantinople (869), the Fourth Lateran Council, the Council of Florence, the First Vatican Council, and many others. To be sure, other things can be more or less immutable, immutable in this way or that, but for a cause to be simply immutable is reasonably and widely seen to be a divine prerogative, since such a cause can act on others but cannot be changed by anything else.
Some have argued that it is inappropriate to regard pure immutability as a divine prerogative because to say that something is immutable is to imply that it is inert or unresponsive. But we can see the foolishness of this objection, which is based only on a false association of stability and inactivity in the imagination, if we look at a related characterization of God. Sacred Scripture characterizes God as The Rock (Dt 32:4) and God, my rock (Ps 42:9) and the Rock of my salvation (Ps 89:26). Hannah prays, There is no Rock like our God (1 Sam 2:2). David sings (2 Sam 22:2-3; Ps 18:2), The Lord is my rock, my fortress and my deliverer, my God, my rock, in whom I take refuge. Again, he prays (Ps 31:2-3), Be a rock of refuge for me, a strong fortress to save me. You are indeed my rock and my fortress. And again (Ps 62:2): He alone is my rock and my salvation, my fortress; I shall never be shaken. In another Psalm, we find written (Ps 71:3), Be to me a rock of refuge, a strong fortress, to save me, for you are my rock and my fortress. Speaking of the repentance of Israel, another Psalm says (Ps 78:35), They remembered that God was their rock. In yet another, we find (Ps 92:15), He is my rock, and there is no unrighteousness in Him. In yet another (Ps 94:22), But the Lord has become my stronghold, and my God the rock of my refuge. And another (Ps 95:1), O come, let us sing to the Lord; let us make a joyful noise to the rock of our salvation. The prophet Isaiah likewise predicts judgment for Damascus (Is 17:10), For you have forgotten the God of your salvation, and have not remembered the Rock of your refuge. Thus we find 'Rock' used as an epithet of God in the Torah, in the Writings, and in the Prophets, associated especially with refuge and salvation and strength. 'Rock' is used of God metaphorically, of course, but as has been noted previously, metaphorical statements may be true. Rocks are associated with strength, stability, and endurance. One might suggest that 'rock' is not an appropriate name of God because it suggests inertness and inactivity, but this is not even really true of ordinary rocks, whose surfaces resist change actively. Everything actual acts, and the most active is not necessarily the most obviously active. But in any case God is The Rock because He is strong, stable, and endures, and does so actively; indeed, all of these passages are filled with comments about God's actions. He is a sure refuge because no other can overcome Him in any way. Likewise, God is said to be immutability not from a lack of activity but from a fullness of actuality, which His activity expresses. Thus St. Gregory Palamas says (Capita, ch. 145), God acts without being acted upon and without undergoing change. The strength of God is that He cannot be moved by another, so that He may be a safe refuge, a sure recourse for salvation. As Malachi says (3:4), I the Lord do not change, so you, descendants of Jacob, are not destroyed, and James says (Jas 1:17) that every perfect gift comes from God, in whom there is no variation or shadow due to change, or, as some have it, no variation due to a shadow of turning. The responsiveness objection, then, is based on a misunderstanding of everything.
* This is also noted by Elizabeth Anscombe, "'Whatever Has a Beginning Must Have a Cause': Hume's Argument Exposed", Analysis, Vol 34 No 5 (April 1974), 145-151. In addition, Anscombe has a defense of the Hobbesian argument.
** For further discussion of a similar sort, see William Whewell, "Of the Nature of the Truths of the Laws of Motion", The Philosophy of Inductive Sciences, Volume II, pp. 573ff.