Two Versions Of The Cosmological Argument
TWO VERSIONS OF THE
COSMOLOGICAL ARGUMENT
R. Douglas Geivett
I had that world-weary feeling you sometimes get when traveling. So, hunkered down in my window seat, it was mildly comforting to know that the next seat over was unoccupied. The young woman on the aisle appeared to favor a quiet, conversation-free flight—another positive. But in my gut I knew this arrangement wouldn't last. The plane was filling up, and the nice buffer I had hoped for would no doubt be taken by another passenger. Now I feared that it would be someone with a talkative streak. In fact, a shadow of conviction spread over me that whoever it was would definitely be an extrovert. The book I had with me would come in handy. I could look like I was reading and not to be interrupted, even if the words on the page were a blur to my tired eyes and mind. I could also break open my laptop in the event of an emergency.
I set my props in place for the forthcoming act and peered over the edge of the book. I wanted to monitor the entrance to the plane and see if I could identify my predestined seatmate without the benefit of indisputable evidence. I had this powerful intuition that I should expect some jaunty young fellow or perhaps a perky octogenarian. Some minutes later a guy bounced onboard, full of vigor and bonhomie. He was exuberance itself. That had to be the one. It was.
Without delay I buried myself in the lifesaving book. And for most of the journey, it worked. But the act required total concentration. Exuberance guy was watchful, ready to pounce on the slightest opportunity to engage. The trick was to turn the unread pages at a natural rate and carefully avoid eye contact. And the plan would have worked until touchdown if it hadn't been for the kindness of a flight attendant, who, in complete innocence, offered me water. As I reached to accept this small gift—I assure you, the act was as brief as it could be—exuberance guy interposed himself between us and looked me in the eye, saying, “Hey, what's that you're reading there?”
I answered with less than admirable compactness: “It's a book.”
Not to be put off, he pressed, “What's it about?”
“It's a book in philosophy.” This did not have the customary effect.
“Are you a philosopher?”
“Yes.”
“Really!” he said, with truly over-the-top enthusiasm. “I've always wanted to talk to a philosopher! I have so many questions.”
Resigned to my fate, I asked the inevitable question: “What would you like to talk about?”
Turns out he had a friend at work that enjoyed debate, mostly about religion and politics. The two would disagree on most things. Exuberance guy told me that one topic came up most frequently. He said his friend believed in God; so he wanted to know if I could give him some ammunition that would blow up the case for God's existence. I said I might be able to help, but I needed to ask a couple of questions.
“Fire away,” he said (or something to that effect).
“OK, you say your friend believes in God; so apparently you do not. What do you believe instead?”
“Oh, I believe in a higher power,” he answered matter-of-factly and without hesitation—not an utterly clear and distinct concept.
So I asked, “What do you mean by a higher power?”
For a moment he looked genuinely perplexed, like, “What kind of a fool philosopher doesn't know what a higher power is?” My impassive expression must have told him that I really did expect an answer. Here's what he said; I remember his exact words. “A higher power is … a power … that is … really high!” He spread his arms to gauge the magnitude of the power in question. I must confess, this struck me as a pretty superficial response, one for which I was, just possibly, overtrained.
I had one question left: “Why do you believe in a power that is really high?”
At that point things improved. He said, “Look around. All of this must have been caused by something.”
I glanced through the window and looked down at the majestic landscape, stretched out some forty thousand feet below us. I couldn't disagree. I looked back at him and said, “If you believe that, then I'm not sure why you don't believe in God.”
We were now fully engaged. The rest of our conversation was interesting, sometimes even a little animated. But the details are irrelevant for the purposes of this chapter. The important point is that this guy, who didn't believe in God, had just stated a low-tech version of the cosmological argument for the existence of God: the universe must have had a cause, and that cause must have been equipped with a load of power.
Kinds of Cosmological Arguments
All cosmological arguments reason from the cosmos to the existence of God, or cosmos-Maker. More precisely, cosmological arguments infer the existence of God from the need to explain the existence of the physical universe.
These arguments begin with observation of the obvious and at least a moderate degree of curiosity about the obvious. The observation: There is a physical universe. What could be more obvious? We don't expect much disagreement about that. For many, though, the existence of the physical universe is so obvious it never strikes them as odd that there should be such a thing. It's simply taken for granted. And why not, as long as they wake up in the same world each day and the familiar laws of nature (like gravity) continue working in the usual way.
I can relate. Even for me, a philosopher by trade, the practical problems of life can muscle aside my inclination toward philosophical rumination. Maybe if I woke up in one of Robert Heinlein's parallel universes and wanted to hitchhike my way back to this one, the space-time coordinates of this world would matter more to me.1
During a quiet moment you might ask yourself this question: “Why is there something rather than nothing at all?” You don't have to be a trained professional to go there. You can do it at home on your own.
You'll soon realize that this question is an extension by many degrees of a question most of us do contemplate at least occasionally: “Why do I exist? I might not have existed, but here I am.” We know that we would not exist if the universe did not exist. So there's a connection of interest between my existence and the existence of the cosmos. We might well reflect, “Why does the cosmos exist? It might not have existed. But here it is.” Part of solving the mystery of our own existence is solving, if we can, the mystery of the existence of the cosmos, the physical environment where we conduct our lives and which determines the potentialities of our existence.
So what is it about the physical universe that needs explaining? Its existence, of course. But why? For two reasons: because of what sort of thing the universe is and because of the temporal longevity of the universe. These two reasons give us two broad categories of cosmological arguments. Within each category there are various formulations. There are far too many versions of the cosmological argument to do justice to more than two of them here. In my opinion, they are “best of breed.”
A Contingency Argument for the Existence of God
The first broad category of cosmological arguments zeroes in on a specific feature of the universe: its contingency. Contingency is a mode of existence: an entity is contingent if it is not a necessary being, if it does not exist necessarily. Every actually existing entity is either necessary or contingent. (All “entities” that could exist but do not would also be contingent if they existed since the fact that they are possible but not actual guarantees that they are not necessary.)
How are we to determine whether an entity is contingent or necessary? Our criterion is just this: if it is in the nature of a thing to exist, then its existence is necessary, and it is a necessary being; otherwise, it is a contingent entity, depending for its existence on something else. But how to tell whether it is in the nature of a thing to exist? Certainly, if a thing begins to exist or passes out of existence, it is a contingent thing. It is not in the nature of the thing for it to exist; it is not necessary. On the contrary, it depends on something else for its existence. If it were not for some appropriate cause, any entity that is contingent simply would not exist.
There are three major components in any contingency argument for the existence of God:
The argument I've selected to represent this type of cosmological argument was developed in the eighteenth century.
Samuel Clarke's Cosmological Argument
Samuel Clarke (1675-1729), an English philosopher and Anglican minister, produced an influential version of the cosmological argument that has enjoyed the attention of philosophers for three centuries. Everyone agrees that it is an argument from the contingency of the universe. But in the formulation I present, following Clarke's own language closely, I use the terms dependent and dependently, rather than contingent and contingently.2
Clarke turns next to a lengthy and sophisticated discussion of the nature of this self-existent being. There isn't space to do that here. But I do want to explain each statement in the argument, and the progression of one statement to another.
1. Something must have existed from all eternity. Samuel Clarke has an independent argument for this claim, but I leave it as homework for you to check that out. The statement seems obviously true. But it needs to be noted that what is initially most obvious is that the universe exists. From this it follows that something exists. The present claim is that if something exists, then something—whether or not it is the same thing—has existed from eternity; something has always existed.
Consider the implications if this is denied. The universe exists, but it has not always existed. And neither has anything else. So why is there a universe that has not always existed? This question leads directly to a different kind of argument for the existence of God, namely, a first-cause type of argument. In other words, if someone denies the first premise of Clarke's cosmological argument, we can pursue a different line of argument for the conclusion that God exists as the cause of the beginning of the universe. That sort of argument is developed later in this chapter.
The claim we have started with is not especially controversial. So we proceed to the next premise.
2. If something has existed from all eternity, then either there has always existed one unchangeable and independent Being, or there has been an infinite succession of changeable and dependent beings. Only two kinds of things can have existed from eternity. The first is an unchangeable and independent Being. The second is an infinite succession of changeable and dependent beings. These are not mutually exclusive. It's possible that both kinds of things have always existed.3
In his exposition Clarke expands on both kinds of things that can have existed from eternity. Clearly the universe is not an unchangeable, independent Being. But the universe could be an infinite succession of changeable and dependent beings. The picture here is of a string of changeable and dependent beings that are related to one another in succession (one after the other). Each individual being is changeable (for example, comes to be or ceases to be), and each is dependent on some other individual being in the string.
This string of changeable and dependent beings can have existed from eternity but only on one condition: it must contain infinitely many changeable entities standing in successive dependence relations. Clarke concedes that the universe may be like that. It's possible that the physical universe has always existed. But if it has, it exists, as a whole, as an infinite succession of changeable and dependent entities. These entities, he says, would be “produced one from another in an endless progression, without any original cause at all.”
So this argument does not require the assumption that the universe had a beginning.
3. Either there has always existed one unchangeable and independent Being, or there has been an infinite succession of changeable and dependent beings. This follows directly from premises 1 and 2. By a logical principle called modus ponens, if premises 1 and 2 are true, then statement 3 must also be true. Hence, no special argument is needed to establish this point.
4. An endless series of dependent beings either is caused by something external to itself, or it contains within itself the reason for its own existence. Here we come to a premise that makes use of an important principle, called the Principle of Sufficient Reason. There are various formulations of this principle. Clarke holds that for anything that exists, there must be a reason for its existence. Only two kinds of reasons for existence are possible. Something may contain the reason for its existence within itself. It may, in other words, be self-existent.4 If an entity is not self-existent, then the reason for its existence must lie outside itself.
We are now considering the universe as a possibly endless series of dependent beings. Since the universe exists, it is either self-existent, or the reason for its existence lies beyond the universe. Now we must consider whether the universe may be self-existent. If it may, then this argument will have entered a cul-de-sac where God cannot be found.
5. If an endless series of dependent beings is self-existent, then at least some one thing in the series must be self-existent. This statement tells us what it would mean for the universe, considered as a possibly endless series of dependent beings, to be self-existent. At least one entity within the series would have to be self-existent, or necessary. No series of dependent things can be self-existent if it has no self-existent parts, even if the series is endless—perhaps especially if the series is endless.
Suppose the series is finite and the universe has a beginning after all. Then perhaps the first entity in the series is self-existent and therefore supplies the reason for the rest of the series. But on our hypothesis, the universe is an endless series of things (all of them dependent, in virtue of progressing one from the other).
Some complain that premise 5 commits an egregious logical mistake called the fallacy of composition. If we're going to make a decent argument for the existence of God, it won't do to have a fallacy buried somewhere in the premises. The accusation here is that one cannot deduce that a whole set of entities has all of the same properties that each of its constituents has.
It's true that often a whole is “greater than the sum of its parts.” A sugar cube divided into exactly four parts having exactly the same proportions will weigh more as a whole than any of its four parts weigh individually. So the weights of physical objects vary between parts and wholes. But does this mean that none of the properties of the parts of our sugar cube will be properties of the sugar cube as a whole? Certainly not. For each part of the sugar cube will taste the same as the sugar cube as a whole. Each of the parts, as well as the whole, is sweet.
The question we must consider is whether the part-whole relation we have been considering in this argument is more like the part-whole difference in weight or more like the part-whole similarity in sweetness in our sugar cube example. The answer depends (here comes an unintentional pun) on what kind of property “being dependent” is. Clarke believed that if all of the parts of our universe are dependent, requiring a reason for their existence that lies outside themselves, then the universe as a whole series must be like that, too. I agree. If there is no internal reason any of the parts of the series should exist, then how can there be an internal reason they all exist as a whole?
Bottom line: if the universe is to be self-existent, it must contain within itself the reason for its own existence.
6. Every being within an endless series of dependent beings depends for its existence on some other being in the series. This is a tautology. It is true by definition. The phrase “an endless series of dependent entities” just means “a series where every entity within the series depends on some other entity in the series.” An opponent of this argument has the right to reject the definition, but then the universe will be a different sort of thing. Our argument draws out the implications of thinking about the existence of the universe the way we have been thinking about it.
7. An endless series of dependent beings cannot be self-existent. This follows directly from premises 5 and 6, again, by modus ponens: If P, then Q; P. Therefore, Q. Premise 6 says, in effect, that the series we are considering does not include any item that is self-existent. Given conditional statement 5, this means that the series itself cannot be self-existent.
8. An endless series of dependent beings must be caused by something external to itself. This results from following another logical rule, a rule governing disjunction. In premise 4 we have a disjunction. It states two possibilities connected by the word or. In statements like these, or can be interpreted in two ways. If it is construed as exclusive, then the statement tells us that exactly one of the two options is true; that is, either one is true, or the other is true, but not both. When the or is interpreted in the inclusive sense, the statement claims that at least one, and possibly each, of the options is true. In both types of disjunction, however, if one option is ruled out, the other option must be ruled in. That's what is going on here. Statement 4 tells us we have two options. Statement 7 excludes one of the options. Statement 8 just makes explicit the only option that remains. In this case an endless series of dependent beings must be caused by something external to itself. In other words, the series cannot be self-existent.
9. The external reason for the existence of an endless series of dependent beings must be either a dependent being or a self-existent being. Now we have another pair of choices. In this case the choices are mutually exclusive. It must be one or the other. If we can rule out one, then we'll know what the right one is. The next statement in the argument tells us which option must be ruled out.
(By the way, in Clarke's own exposition of the argument, he speaks of the need for an “external cause” of the endless series of dependent beings. This is because we seek an external reason for the existence of this series [i.e., the universe], and an external reason for the existence of such a thing will be causal in nature. For consistency in setting forth the argument in subsequent steps, I've used “external reason” in place of “external cause.”)
10. The external reason for the existence of an endless series of dependent beings cannot be a dependent being. If it is a dependent being that provides the reason for the existence of an endless series of dependent beings, then this dependent being, while appearing to play a special role, is itself little more than a part of the endless series. In causal terms this special dependent being will be just one more causally dependent being in the whole chain of dependent beings. It is not an external reason for the existence of the series if it is part of the series. (The passage from which this cosmological argument is derived is sketchy at this juncture, and I have developed the point in the way that makes the best sense in context.)
11. The external reason for the existence of an endless series of dependent beings must be a self-existent being. This step is the result of exclusive disjunction (statement 9) and the exclusion of one option at step 10. The only possibility is that the external reason we seek is a self-existent being, that is, a necessary being.
12. If an endless series of dependent beings exists, then a self-existent being exists. This statement simply makes explicit what follows directly from the assumption that an endless series of dependent beings exists and the conclusion, reached in step 11, that the reason for the existence of such a series must exist.
13. This self-existent being must have existed from eternity and be immutable and independent. A self-existent being will exist from eternity because the reason for its existence is internal, making it necessary. It cannot not exist. For the same reason it will be independent in its existence, not dependent on anything else for its existence. As for its immutability, readers should consult Clarke. He takes pains following this argument to explain what else may be said about this self-existent Being so that it is proper to speak of it as God.
A First-Cause Argument for the Existence of God
So far we have examined one of the more powerful cosmological arguments for the existence of God. We come now to another best of breed, the kalam cosmological argument. This argument is one version of a first-cause argument for the existence of God. Like the contingency argument, it has a simple three-step pattern:
The Kalam Cosmological Argument
Now for some clarification of each step:
1. The physical universe exists. This is our already familiar starting point.
2. The history of the universe is a series of events arrayed in temporal succession. This statement makes explicit how the events that make up the history of the universe stand in relation to one another. They stand in temporal relations where they occur in succession, one after the other.
3. Either the physical universe had a beginning, or it has always existed. This is an exclusive disjunction. It is a noncontroversial premise in the argument since it conforms to the logical law of excluded middle (tertium non datur), which has the form “either P or not-P,” where P is any proposition you like.
4. If the physical universe has always existed, then the history of the universe is a series of infinitely many events arrayed in temporal succession. Since the history of the universe is a series of events arrayed in temporal succession, the number of events in the series must be infinite in order for the universe to have no beginning.
5. The history of the universe cannot be a series of infinitely many events arrayed in temporal succession. This is the crucial and most controversial premise of the argument. While sets of infinitely many objects may not be impossible, it does seem that the events that make up the history of the universe could not be infinite in number. This is because they are arrayed in temporal succession, where one event follows another.
Select any event from the history of the universe; call it event E. If the universe did not have a beginning, then infinitely many events must have occurred prior to event E. But if infinitely many events must have occurred before event E could occur, and this would take an infinite amount of time, it seems that event E could never occur. But it has occurred. Since event E could be any event that actually has occurred, it seems that no event in the history of the universe could be preceded by infinitely many events.
6. The physical universe has not always existed. This follows from statements 4 and 5 by a pattern of logical implication called modus tollens: If P, then Q; not-Q. Therefore, not-P.
7. The physical universe had a beginning. This is equivalent to the previous statement, stated here in this way for convenient presentation of the premises that follow.
8. If the physical universe had a beginning, then the beginning of the universe had a cause. The rationale for this premise is that for anything to begin to exist is for an event to occur, and every event has a cause.
9. The beginning of the physical universe had a cause. This follows by modus ponens from statements 7 and 8.
10. If the beginning of the physical universe had a cause, that cause must be either an event or an agent. The events with which we are familiar are of two kinds. Many events are caused by other events. Many others are caused by agents.
If a baseball strikes a pane of glass at a certain velocity, it may break the glass. The breaking of the glass is an event. What caused the event? That depends on the circumstances. Suppose we wish to know what was the immediate cause of the breaking of the glass. If the baseball was launched by a pitching machine, then the movement of various parts within the machine cause the ball to be launched. That movement is an event. One event, the movement within the machine, has caused the ball to be launched, and another event has occurred, the breaking of the glass. This is an instance of event-event causation. But what caused the movement in the machine? Suppose some person pressed a button that activated the pitching mechanism. Now we have an agent, a personal being with powers and intentions, causing an event. This is an instance of agent-event causation (or simply, agent causation.)
Like any other event, the beginning of the universe must be caused either by an event or by an agent. (This, at least, is what we should think unless and until some other form of causation is discovered.)
11. The beginning of the physical universe was caused either by an event or by an agent. This follows by modus ponens from premises 9 and 10.
12. The cause of the beginning of the physical universe could not have been an event. Here we rule out one kind of cause for the beginning of the universe. The reason is simple. The beginning of the universe is, in the nature of the case, the first event in the history of the universe. If that event were caused by a prior event, it would not be the first. So the beginning of the universe could not have been caused by an event. We know now what must be the cause.
13. The cause of the beginning of the universe was an agent. Having excluded one of two available ways that events may be caused, we conclude that the cause of the beginning of the universe must have been an agent. This means that the origin of the universe was caused by a personal being, acting with powers and intentions suitable for such an act.
Is this agent God? As I've written elsewhere, the kalam argument allows us to infer that “there must be some personal and timeless being, powerful and intelligent enough to cause the universe to begin to exist. This fully transcendent Agent presumably acted for reasons when creating the universe. Even if that is all that can be said about the most likely cause of the beginning of the universe, it already appears to be a good candidate for the appellation God.”5
Summary
We have sampled two influential cosmological arguments for the existence of God. The first, developed by Samuel Clarke, is an argument from the contingency of the universe to the existence of God as a self-existent, necessary Being. The second, known as the kalam cosmological argument, reasons from the beginning of the universe to the existence of God as the First Cause of the Universe.
Conclusion
There are compelling reasons to conclude that there is a Necessary Being, a self-subsistent Creator and Sustainer of a universe that hangs onto existence by the skin of its teeth.6 We have good reason to believe that the universe, the theater of human existence, had a beginning and that there must be a First Cause to whom it and we owe our existence. Is this the God of Abraham, Isaac, Jacob, and Jesus? It could be. An argument with such a fantastic conclusion would take dozens of steps. But regardless of where further argument leads, a Necessary Being, a First Cause of everything else we know to exist, is a plausible candidate for the God of Christian theism.7 If there is but one Necessary Being and it happens that the God of biblical faith does exist, then they must be one and the same. If there is a First Cause of the physical universe, among Christians He goes by the name Yahweh, or Jehovah; He is the unique and sovereign God who has expressed Himself more fully in Scripture and in Jesus Christ. That possibility is worth more than a moment's casual reflection.
If the Being upon which our cosmological arguments converge is not the Greatest Conceivable Being, it is surely the greatest being that actually does exist. As such, it is the nemesis of naturalism. Our universe is haunted by a presence that cannot be conveniently contained by the dainty theories of scientific and philosophical naturalists. But that's not all. This greatest of all existing beings is, for all intents and purposes, a god; indeed, it is the only god there is, worthy of capitalization. We are put on notice that there is a God with whom we have to do. We ignore this God at our own risk, for this God is our God (see Acts 17:16—31). And we pay our God no compliments with comforting abstractions about some amorphous and remote “spiritual energy force” or the “Religiously Real” common denominator of all world religions. And we should know better than to refer to this God with effete monikers like “a Higher Power” (with or without capital letters).
1 See, for example, Robert Heinlein, Job: A Comedy of Justice (New York: Del Ray, 1985) and his The Number of the Beast (New York: Ballantine, 1980) or Isaac Asimov, The Gods Themselves (West Sussex, England: Gollanz, 2000).
2 Clarke's brief exposition of this argument appears in A Discourse Concerning Natural Religion (1705). It is conveniently excerpted under the title “Samuel Clarke: The Argument from Contingency,” in Louis P. Pojman, ed., Philosophy of Religion: An Anthology, 4th ed. (Belmont, Calif.: Wadsworth, 2003).
3 Clarke uses the capital letter B for an unchangeable and independent Being because of the superior kind of existence it has in comparison with changeable and dependent beings.
4 Clarke's choice of terms here is deliberate. We are not to confuse a self-existent being with a self-caused being. A self-existent being, while it does contain within itself a reason for its own existence, cannot be caused to exist by itself. For Clarke, a self-existent being is a necessary being. Such a being cannot not exist. Hence, it will not have been caused to exist at all. But we can understand why it exists if we understand that it is a necessary being that cannot, given its nature, not exist. This is what it means for such a being to contain the reason for its existence within itself.
5 R. Douglas Geivett, “The Kalam Cosmological Argument,” in To Everyone an Answer: A Case for the Christian Worldview, ed. Francis J. Beckwith, William Lane Craig, and J. P. Moreland (Downers Grove, Ill.: InterVarsity, 2004), 74. For more detailed exposition and defense of the kalam argument, please see the rest of that chapter. Kalam is an Arabic term meaning “word/ speech,” but it came to mean “natural theology”; it is the classic argument used by Muslims and some Christians in the Middle Ages to defend the existence of God based on the finitude of the past.
6 See Dallas Willard, “The Three-Stage Argument for the Existence of God,” in Contemporary Perspectives on Religious Epistemology, ed. R. Douglas Geivett and Brendan Sweetman (New York: Oxford University Press, 1993), 215.
7 For details on how a cumulative case might be made for Christian theism, see my chapter “David Hume and a Cumulative Case Argument,” in In Defense of Natural Theology: A Post-Humean Assessment, ed. James F. Sennett and Douglas Groothuis (Downers Grove,Ill.: Inter-Varsity, 2005), 297-329.