Aquinas’s First Cause Argument, Part 2: Efficient Causes

In my first post on Aquinas’s First Cause Argument for the existence of God, I discussed several preliminary issues which it is important to understand/note before beginning to look at the argument itself. Here’s a recap of those preliminary issues:

  1. Aquinas’s First Cause Argument, or the Second Way, is just one particular version of many first cause/cosmological arguments, the most popular of which is the Kalam. It is important to distinguish the Kalam from Aquinas’s Second Way: the Kalam is based upon the premise that the entire universe began to exist, the Second Way is not.
  2. The notion of “causation” itself is extremely contested amongst both scientists and philosophers, especially due to modern thinkers such as Hume and Russell. The deeper metaphysics of causation can be complicated, and I’ll be getting into it a little bit, but, for reasons which I made clear in the last article, I will be taking the reality of causation as a given.
  3. This argument seeks to establish the existence of a “first cause.” This does not mean a cause which is first temporally in a series, but rather a cause which is first in causal power, meaning it is more fundamental than all other causes.
  4. The Second Way is, in many ways, extremely similar to the First Way (see my posts on the First Way here, here, and here), and there are different ways to interpret it. I will be discussing this issue further in this post and future ones.

With all of that having been laid out and hopefully made clear, let’s begin to take a look at the argument itself, using Aquinas’s own words as a starting point for examination/extrapolation.

Aquinas begins in his magnificent Summa Theologica:

“The Second Way is from the nature of the efficient cause” [1].

As I mentioned somewhat briefly in my Prime Mover Trilogy, Aquinas’s philosophical system has its roots in the metaphysics of Aristotle, and for Aristotle there are four types of causes: material causes, formal causes, efficient causes, and final causes. The Second Way deals specifically with efficient causes. An efficient cause, as philosopher Edward Feser explains, “is what brings a thing into being, or, more generally and technically, what actualizes a potentiality in a thing” [2]. Now there are several issues which arise immediately from this. The first is that, for anyone who read my Prime Mover articles, the terms “actualizes” and “potentiality” will be immediately familiar, for these terms are what that whole argument is based upon. So if the Second Way deals with efficient causes, and if an efficient cause is that which actualizes a potentiality, one might wonder, how is this argument any different from the First Way? This question will be addressed below.

The second possible issue has to do with the claim that an efficient cause is that which brings a thing into being. One might then ask, if an efficient cause is that which brings something into being, would the beginning of the universe itself not be an instance of a need for an efficient cause? The answer gets into an interesting and significant point. In a debate with William Lane Craig, philosopher Peter Millican, in response to the Kalam argument, makes the comment that, in our experience, nothing actually seems to “begin” to exist; rather, in our experience, there is only change. That is, one thing might change into a new thing, but this change is only the “rearrangement of existing materials.” He goes on to say that we never experience anything actually coming into existence from nothing, so we cannot infer that such “creation from nothing” requires a cause [3]. Now, what is somewhat ironic is that Millican does admit the reality of change, and even admits that such change is causally governed, both of which facts happen to comprise the first two premises of the Prime Mover argument, which argues from the existence of change to an unchangeable or “unmoved” changer/mover. Whether or not his comment actually works as an objection to the Kalam is irrelevant to this present discussion; but the comment itself raises an interesting question about causation. After all, in common sense understanding, we speak and think of things “beginning to exist” all the time: a newborn baby, a new star, a new table, etc. But Millican is correct that in each of these cases, no new particles or new matter or energy are actually coming into existence; they’re just being reordered into new arrangements. Similarly, in another interesting counter argument to the Kalam, Jeffery Jay Lowder proposes what he calls the “Anti-Creation Ex Nihilo Argument,” which goes as such:

  1. Everything that had a beginning comes from pre-existing material
  2. The universe had a beginning
  3. Therefore, the universe came from pre-existing material [4].

Now, Mr. Lowder does not himself think this argument works, because he does not himself think that the universe (meaning the sum total of all space, time, matter, energy) had a beginning. But he uses the argument as a counter to the Kalam, and it displays a similar objection to Millican’s comment above. So, if an efficient cause is that which brings something into being, and if nothing ever actually “comes into being” in the sense that any new material comes into existence, are there really any efficient causes?

The answer is unequivocally yes. For one thing, on the Aristotelian-Thomistic system, one certainly would not accept that just because no new matter comes into existence, that therefore nothing new can come into being. After all, Aristotelian-Thomists hold that everything that exists has a “form,” or an essence/nature, that which makes something what it is. When a baby is born, no new matter may come into existence, but the pre-existing matter is rearranged in and instantiated with a new form. And this is just what we mean in normal, everyday, common sense language when we say that something begins to exist. So while pre-existing matter might be a necessary condition for something to “begin” to exist, it is certainly not a sufficient condition. A new form must also be instantiated in that matter.

But, besides this, we don’t have to argue that new objects actually come into existence to argue that efficient causes are real; for everything (material) that exists has certain “potencies” which can be actualized, and when a potency is actualized, a new aspect of an object’s being has come into existence. A brown table has the potency to be green or blue or purple. A seed has the potency to be a flower, etc. And as I argued in my Prime Mover articles, there must be something which actualizes those potencies, some cause which brings these new features “into being.” That is what an efficient cause is, fundamentally. Philosopher Edward Feser writes:

“An efficient cause, whether of a thing’s existence or of some change to it, always actualizes some potency or other. The principle of causality (which is concerned with efficient causality…) tells us that if a potency is actualized, that can only be because some already actual cause actualized it…The basic idea is that since a potency qua potency is merely potential rather than actual, it can’t do anything. In particular, it can’t actualize anything, including itself. Hence if it is actualized, something already actual has to be what actualizes it. Other formulations of the principle of causality are essentially just applications of this idea” [5].

But efficient causes also have different “manifestations,” if you will, and this is where the distinction between the First and Second Ways becomes clear. Feser continues:

“Hence the theses that whatever is changed is changed by another and whatever comes into existence has a cause are straightforward applications of the principle, since to change or to come into existence is just to go from potency to act” [6].

The thesis “whatever is changed is changed by another,” is, as some readers will recognize, one of the central premises of the Prime Mover argument. And the thesis “whatever comes into existence has a cause” is similar to the aforementioned premise of the Kalam, that “whatever begins to exist has a cause.” But efficient causes do not just stop with change and coming into existence. Philosopher W. Norris Clarke, S.J. in his book The One and the Many: A Contemporary Thomistic Metaphysics gives an outlined list of all the different types of efficient causes, the first two being those which we have already discussed:

  1. Every being that begins to exist needs an efficient cause
  2. Every being which undergoes real intrinsic change requires an efficient cause
  3. Every composed being requires an efficient cause
  4. Every finite being requires an efficient cause
  5. Every being which, in order to exercise its natural properties, must belong to a system, requires an efficient cause [7].

It would take much more space to go over each one of these in any amount of detail, so I’d like to go over just two more. While Clarke does not explicitly list this next type of efficient cause, Edward Feser does, and I believe it can fit under Clarke’s list:

“A contingent thing is such that its existence is distinct from its essence, where its essence is in potency relative to its existence, which actualizes is…To cause a contingent thing to exist is thus to actualize a potency. Hence the thesis that whatever is contingent has a cause is also an application of the principle that a potency can be actualized only by something actual” [8].

Now this gets into a bit of new terminology. Talk about “contingent” versus “necessary” beings/truths is common in contemporary philosophy, but Scholastics mean something a bit different. In contemporary philosophy, a contingent being is normally understood to be a being which exists in some, or at least one, possible world (another contemporary philosophical term, which I will have to explain elsewhere), but not in all possible worlds; while a necessary being is a being which does exist in all possible worlds. For Aquinas, however, something is contingent just if “it is possible either to exist or not exist, as is evident from the fact that [it is] generated and corrupted, coming into being and passing away” [9]. When something is generated, its essence is given existence (more on this later), which also requires an efficient cause.

Next we have a “composed” or composite being. Feser explains:

“Aquinas’s further thesis that ‘every composite has a cause, for things in themselves different cannot unite unless something causes them to unite’ (Summa Theologiae I.3.7). But as Aquinas goes on to say, ‘in every composite there must be potentiality and actuality…for either one of the parts actuates another, or at least all the parts are potential to the whole.’ So for a composite to exist is just for the potency of its parts to comprise the whole to be actualized. Every composite has a cause, then, insofar as only what is already actual can actualize the potency in question” [10].

I realize this all has been very dense. All you really need to take away here is that an efficient cause is that which actualizes a potency, and potencies require actualization in a number of different instances, such as whenever something changes, whenever something begins to exist, or whenever something is a composite (comprised of parts), just to name a few. Why is this important? Because these different instances of efficient causation represent different arguments which can be constructed. The Prime Mover Argument begins with the first instance, that of change, and argues from change to an Unmoved Mover. The Third Way, as I will explain in a future series, begins with the reality of contingent beings (as Aquinas understood them) and argues to a necessary being. So which instance of efficient causation does the Second Way refer to? Well, that’s where some of the controversy of interpretation lies, as Aquinas doesn’t make this entirely explicitly clear; he just refers to efficient causation in general. Most defenders of the argument (that I’ve read) seem to interpret it as referring to things coming into existence, or else as referring to things coming into existence in conjunction with one of the other possible instances. We’ll look at this below; but first, a quick point. Although this claim may seem premature, I think we can actually say that there are prima facie reasons to believe that the Second Way works, simply because the First Way works (or, at least, I have argued that the First Way works). The First and Second Ways both have the same structure and employ the same methods; they just have different starting points. Where the First Way starts/argues from motion, which is just a particular  instance of efficient causation, the Second Way starts from efficient causation in general (or, perhaps, a separate specific instance of efficient causation, depending on the interpretation). So if the First Way works (as I think it does), then we might have reason to believe that all instances of efficient causation could be constructed into arguments for the existence of God. But, nevertheless, it is important to defend the Second Way by itself, which is what I am here trying to accomplish. So let’s take a look at the first premise which Aquinas offers:

“In the world of sense we find there is an order of efficient causes” [11].

In other words, just as we observe with our senses that things move/change, we also observe with our senses that there are series of causes and effects. For now, let’s just consider efficient causation as it refers to things coming into existence. So, for example, consider a marble sculpture such as Michelangelo’s masterpiece the Statue of David. It is obvious that this sculpture was “caused” in one of the revenant senses explained above (it began to exist, it is a composite being, a contingent being, etc). The most obvious efficient cause of such a sculpture would be the sculptor, such as Michelangelo, who brought it into being. Thus we have an efficient cause and its effect, which constitutes “an order of efficient causes.” Aquinas goes on:

“There is no case known (neither is it, indeed, possible) in which a thing is found to be the efficient cause of itself; for so it would be prior to itself, which is impossible” [12].

This should be fairly self-evident and uncontroversial, and really just reiterates the premise from the Prime Mover argument that something cannot actualize its own potencies. But it also refers back to philosopher Michael Augros’s distinction between causes “before all other causes in time” and causes “before all others in causal power” [13] which I mentioned in the first post. In his book Who Designed the Designer, he explains further. We often think about causes being before their effects, or, as he calls it, having “temporal priority.” So the artist must exist prior to the statue. So also, when a rock is thrown and smashes into a window, the throwing of the rock happens prior to or before the smashing of the window. But this is actually mistaken. In reality, causes and effects occur simultaneously. Or, at least, the immediate cause of an effect occurs simultaneously with that effect. Insofar as the smashing of the window would not have occurred if the rock had not been thrown, the throwing of the rock can be considered as a cause for the smashing of the window. But the direct or immediate cause of the window actually breaking, is the rock smashing into it. When I throw a rock, and the rock is in midair, it is having no effect on the window. It is only at the precise moment when the rock hits the window that the window shatters. Before the artist actually starts sculpting the statue, he is not actually causing anything; he only has the potential to do so. It is only when he starts working the marble that there begins the series of cause and effect [14]. So to reiterate: nothing can be the efficient cause of itself, for in order to do so it would have to be prior to itself. This applies to things beginning to exist, because obviously if something does not yet exist it cannot cause itself to exist. But it can also apply to other types of efficient causes, insofar as nothing can be “causally prior” to itself either. In addition, all efficient causes are simultaneous with their effects, which, as we shall see, is an important point.

Aquinas’s next premise:

“Now in efficient causes it is not possible to go on to infinity, because in all efficient causes following in order, the first is the cause of the intermediate cause, and the intermediate cause is the cause of the ultimate cause, whether the intermediate cause be several, or only one” [15].

Now this premise, that there cannot be an infinite regress of efficient causes, is really the key to the whole argument. Admittedly, the equivalent premise in the Prime Mover Argument, that there cannot be an infinite regress of movers, seems more obvious; because in our experience, whatever moves something else is itself moved, thus requiring its own mover, which leads eventually to an unmoved mover of Pure Act. Every material thing that exists consists of both potency and act, and it is limited by that potency, so that it cannot move without being moved. But must this really be the case for series of efficient causes? After all, at the same time a sculptor is crafting a sculpture, there isn’t anything else which is also crafting the sculptor. The artist was caused by his parents, but once he exists, he certainly doesn’t need his parents to keep existing. If his parents have died, he doesn’t suddenly cease to exist; and this certainly does not prohibit him from exercising hist artistic abilities in making a sculpture. So why could there not be an infinite regress of such causes?

There are three options here, as Augros explains. One is that the series does have a first cause. But perhaps this first cause need not be anything like “God,” perhaps the artist himself just is the first cause of the effect of the statue being made. If “there is no first in such a series, there are two possibilities left: a circular series of causes, or an infinite regression of simultaneously acting causes” but “purely circular causation is obviously impossible…You can’t give your own existence to yourself or receive it from yourself” [16].  This is similar to Aquinas’s premise that nothing can be the efficient cause of itself. A statue cannot cause itself to be made, for the simple and obvious reason that it does not yet exist. Potencies cannot actualize themselves, for they do not actually exist. So since a circular series is ruled out, the only other possibility besides a first cause is an infinite regress. Here it might be useful to examine a different example other than a statue, such as a lamp hanging by a length of chain from a ceiling (this is the example Augros uses in his book). Although this is not an instance of something coming into existence, it is still an instance of efficient causation, for the lamp hanging is caused by the chain hanging it, and the chain hanging is caused in turn by the ceiling hanging it. Suppose that the length of chain has ten links. In that case, the actual hanging of the lamp is caused by the one link to which it is attached; and the cause of that link hanging is the link immediately preceding it, and so on until we reach the ceiling. If there were no ceiling, could the lamp hang? No, it would drop, for (normal) lamps do not have the power to float/suspend themselves in midair. Notice that this doesn’t change if we add more links to the chain. If there were twenty links in the chain, but no ceiling to which it was attached, the lamp still could not hang. If there were five hundred links, or a thousand, or even an infinite amount, the lamp still could not hang without a ceiling. If we think of each link as a cause in a series of efficient causes, it becomes clear why there cannot be an infinite regress. As I explained in my Prime Mover articles, Aquinas distinguishes between essential causal series and accidental causal series. An accidental causal series is one such as a father who begets a son who begets his own son. The existence of the son depends on one causal act of the father, but not on the continued existence or continued sustaining power of the father. An essential causal series, on the other hand, is one such as a woman playing a flute. The sound of the flute depends upon the continued playing of the flute by the woman. As soon as she stops playing, the music ceases. There cannot in principle be an infinite regress of such essentially ordered causes. You cannot have an infinitely long chain holding up a lamp if it is not attached to something at the other end. The ceiling is then the “first cause” of the hanging lamp. Philosopher Richard G. Howe writes:

“In a per accidens [accidental] infinite, the cause of an effect is only accidentally related to the effect being itself a cause whereas in an infinite per se [essential] the cause of the effect is what causes the effect itself to be a cause” [17].

In other words, in an essentially ordered causal series, a cause actual transfers causal power to its effect, as it were; whereas in an accidentally ordered causal series, this is not the case. Feser, in his book Scholastic Metaphysics, extrapolates on several ways in which essential series differ from accidental series. The first, and most relevant to our discussion, is that in accidental series, “second causes” do not derive their causal power from a first cause. Once a son exists, he does not depend upon the father to be able to cause his own son. He has his own causal power. In an essential series, however, second causes do derive and depend on first causes for their own causal power. So a stick does not have the inherent power to move a stone; it is only when it is held and moved by a hand that it can move a stone. The hand has causal power which it imparts to the stick. One is a “principal cause” and the other is an “instrumental cause” which derives its own ability to cause from the principal cause [18]. An accidental causal series, then, could in principle go back infinitely, with no first cause. Essential causal series, however, cannot have infinite regresses, and must have a first cause.

But where exactly does this get us? After all, at first glance, an artist sculpting a statue doesn’t really seem to be a “series” of causes and effects; at first glance, it just seems to be one cause and one effect, so that saying there is a “first cause” is somewhat trivial. In the examples with essentially ordered series, such as a lamp hanging from a ceiling, the series of causes and effects does seem to terminate in a first cause, but this first cause is just the ceiling which is holding up the chains which are holding up the lamp. Furthermore, there are many different such causal series, such as the hand moving the stick moving the stone, which seem all to arrive at completely different, unrelated first causes. Is there anywhere further to go?

One option is to go from efficient causation referring to things coming into existence, to efficient causation referring to a different type. Here’s one possible solution, as explained by Augros:

“When something newly brought into being continues to be with some robustness, when it insists on its new unity even after the obvious causes of its coming to be have ceased to act, this is because of some other kind of cause at work…Sometimes a new combination of things persists due to causes of coherence acting from within…Other times, the new combination sticks together because an entirely external cause exercises some preservative action on it…Either way, a combination endures due to the ongoing action of something that exists before the combination itself. Every combination presupposes a combiner. Every combination that comes into existence depends on a combiner to produce it, and every combination that exists depends on a combiner to sustain it” [19].

This could actually be an instance of both efficient causes for composite beings, and efficient causes for beings “which, in order to exercise [their] natural properties, must belong to a system” (see Clarke’s list above) [20]. It is obvious that essentially ordered causal series of motion cannot have an infinite regress, because anything that moves something else is itself moving (in our experience). But, as I pointed out, an artist sculpting a statue is not himself being sculpted at that same moment. Or is he? We said that lamps do not have the power to float/sustain themselves in midair, so they must hang from something; and it does no good to posit an infinitely long chain from which they hang, because neither do chains have an inherent power to float/sustain themselves, so they too must be attached to something such as a ceiling. But is a ceiling a true first cause of this series? After all, the ceiling is held up by walls, and the walls are built upon a foundation, and a foundation rests in the ground, and the ground is just layers upon layers stacked up from the earth’s crust, and the earth is suspended in space by numerous factors such as gravity and the fundamental forces; all of which are acting here and now in conjunction to keep the lamp hanging. If one of these factors were not present, the lamp would not hang. But neither is it entirely correct to reduce all of these factors to just the most fundamental forces. The fundamental forces and particles are certainly necessary causes to keep the lamp hanging, but they aren’t sufficient. If there were no foundation and no walls and no ceiling, even with the fundamental forces, the lamp could not hang. All of the causes must work in conjunction, here and now. And so finally we come to understand that at least certain types of efficient causes (at least those causes of composite beings or beings which belong to a system) do indeed exist in essentially ordered causal series, and thus that there cannot be an infinite regress, and thus that we must arrive at a true first cause, which itself depends upon no other cause, and which itself imparts all causal power to other secondary causes. Aquinas concludes:

“Now to take away the cause is to take away the effect. Therefore, if there be no first cause among efficient causes, there will be no ultimate, nor any intermediate cause. But if in efficient causes it is possible to go on to infinity, there will be no first efficient cause, neither will there be an ultimate effect, nor any intermediate efficient causes; all of which is plainly false. Therefore, it is necessary to admit a first efficient cause, to which everyone gives the name God” [21].

It may seem quite a leap for Aquinas to jump from “a first cause” to that first cause being God (much as it was in the First Way). Indeed we still have yet to show whether or not there could be multiple first causes for multiple series, or what sort of properties this first cause must have. In addition, there are a good number of difficult objections which have been offered against the Second Way. And, furthermore, we haven’t yet shown that there must be a first cause for things beginning to exist, just things which here and now are exercising causal power as a result of being a composite being or being a part of a system. I will be addressing all of these issues in the next post in the series. Until then, please leave your comments and questions below, and I’ll try to respond!




[1]. Aquinas, Thomas. Summa Theologica. 1265-1274. Trans. Fathers of the English Dominican Province. Coyote Canyon Press, 2010. ebook.

[2]. Feser, Ed. The Last Superstition: A Refutation of the New Atheism. South Bend: St. Augustine’s Press, 2008. ebook.

[3]. The debate itself, as well as commentary, can be found here: Knight, Wintery. “Video: William Lane Craig and Peter Millican Debate ‘Does God Exist?'” Wintery Knight., 29 Mar. 2012. Web. 3 Jul. 2016. <>.

[4]. Lowder, Jeffery Jay. “25 Lines of Evidence Against Theism.” The Secular Outpost. Patheos, 26 Jun. 2016. Web. 5 Jul. 2016.

[5]. Feser, Edward. Scholastic Metaphysics: A Contemporary Introduction. Piscataway: Editiones Scholasticae/Transaction Books, 2014. Print, 105.

[6]. ibid.

[7]. Clarke, W. Norris. The One and the Many: A Contemporary Thomistic Metaphysics. Notre Dame: University of Notre Dame Press, 2001, 181-185.

[8]. Feser, Edward. Scholastic Metaphysics: A Contemporary Introduction. Piscataway: Editiones Scholasticae/Transaction Books, 2014. Print, 105-106.

[9]. Feser, Edward. Aquinas. London: Oneworld Publications, 2009. Print, 90.

[10]. Feser, Edward. Scholastic Metaphysics: A Contemporary Introduction. Piscataway: Editiones Scholasticae/Transaction Books, 2014. Print, 106.

[11]. Aquinas, Thomas. Summa Theologica. 1265-1274. Trans. Fathers of the English Dominican Province. Coyote Canyon Press, 2010. ebook.

[12]. ibid.

[13]. Augros, Michael. Who Designed the Designer? A Rediscovered Path to God’s Existence. San Francisco: Ignatius Press, 2015. Print.

[14]. See page 147 of Feser, Edward. Scholastic Metaphysics: A Contemporary Introduction. Piscataway: Editiones Scholasticae/Transaction Books, 2014. Print.

[15]. Aquinas, Thomas. Summa Theologica. 1265-1274. Trans. Fathers of the English Dominican Province. Coyote Canyon Press, 2010. ebook.

[16]. Augros, Michael. Who Designed the Designer? A Rediscovered Path to God’s Existence. San Francisco: Ignatius Press, 2015. Print.

[17]. Howe, Richard G. “Two Notions of the Infinite in Thomas Aquinas’ Summa Theologica I, Questions 2 and 46.” Christian Apologetics Journal Vol. 10, No. 1 (2012): 15. Web. 3 Jul. 2016.

[18] See page 149 of Feser, Edward. Scholastic Metaphysics: A Contemporary Introduction. Piscataway: Editiones Scholasticae/Transaction Books, 2014. Print.

[19]. Augros, Michael. Who Designed the Designer? A Rediscovered Path to God’s Existence. San Francisco: Ignatius Press, 2015. Print.

[20]. Clarke, W. Norris. The One and the Many: A Contemporary Thomistic Metaphysics. Notre Dame: University of Notre Dame Press, 2001, 181-185.

[21]. Aquinas, Thomas. Summa Theologica. 1265-1274. Trans. Fathers of the English Dominican Province. Coyote Canyon Press, 2010. ebook.


4 thoughts on “Aquinas’s First Cause Argument, Part 2: Efficient Causes

Leave a Reply

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

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

Google+ photo

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

Twitter picture

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

Facebook photo

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


Connecting to %s