As, God help me, a professional philosopher, I try to argue for a clear thesis with compelling, unchallengeable arguments. This post feels too tentative to me, but I don't want to put any more work into it. So, here it is.
One of the most puzzling questions in philosophy is "Why is there something rather than nothing at all?" Theists say that this question must have an answer and the only way to give an answer is to appeal to the existence of God. That God exists, on this view, explains why the contingent, dependent world exists, and God requires no further explanation since God's nature as a perfect being explains why God exists.
Refining the Cosmological Argument
The Judeo-Christian tradition does not provide an explanation for why there is something rather than nothing since even in the first Genesis creation story (the more systematic one) God does not create something from nothing. Apparently borrowing from the Babylonian stories, Genesis has God creating the universe from a watery chaos. Obviously, then, the Judeo-Christian mythology does not explain why there is a watery chaos and, hence, does not explain why there is something rather than nothing.
The Judeo-Christian text does not provide the purported explanation, but perhaps the more abstract theist position can do so. Philosophical theists tend not to attach their beliefs to the highly flawed text of any particular religion, so they conceive God as (borrowing from Alvin Plantinga) a maximally perfect being (one that has all perfections--goodness, wisdom--essentially) who created and designed the universe. How should this supposed need for an explanation prove the existence of this abstractly-conceived God?
Taking the need for an explanation for the universe as an argument for the existence of God is to give the cosmological argument. St. Thomas Aquinas is the most famous proponent of this argument, but, unfortunately, every one of his four ways--four versions of the cosmological argument--to prove the existence of God are obviously unsound. (I read the last of his "Five Ways" to be a version of the design argument.) I cannot go into these flawed arguments in detail here, but one of the proofs, the argument from efficient causation, appears to many to have some plausibility.
Efficient causality is essentially what we ordinarily think of as causality. Efficient causes trigger or bring about an event. Here the argument is that every event has a cause, so there must be a causal sequence of events that has occurred over some period of time in which causes precede their effects. If the cause does not occur, Aquinas argues, the effects would not occur. Since there are, evidently, effects, there must have been a first cause to begin it all. This first cause must itself be uncaused, and so is not simply another event, and is, thus, supposed to be God.
There are two problems with this argument. First, there is no reason to think the causal series has to be finite. Aquinas seems to think that some event, call it "A", must be the first cause because if A did not occur, nothing could happen at all. But if one denies that there is a first cause, one is not denying that A occurs, one is only denying that A is a first (uncaused) cause. Second, if there is a first cause, there is no reason to think that being is God. Something's being a first cause of the universe implies nothing about it having the other characteristics theists attribute to God. It need not be personal, rational, a moral agent, powerful, or wise (let alone the omni-version of each of these attributes). It could simply be a trigger, like the big bang, that did not require a cause for its occurrence.
Various attempts have been made to salvage this argument. One claim (from G.E.M. Anscombe) is that everything with a beginning has a cause, and the universe has a beginning so the universe has a cause. The only thing that could be the cause of the universe is God, so God exists. The first problem with this argument is that it assumes, with insufficient justification, that the universe is a thing that can have a cause rather than a collection of things which need not, as a mere collection, have a cause. I'll get to this point in a moment.
The second problem with this argument is that it would show that God also has a cause since there is no space-time without matter (according to general relativity), and thus God must also have a beginning. This fact would rather undermine the point of the argument since it would imply that God also must have a cause.
The other attempt to salvage the argument, besides relying on beginningless causes, is to claim that the universe as a whole must have a cause or explanation because every part does. This argument is clearly fallacious. If the universe is only a collection of beings, then there is no reason to think the universe as a whole must have a cause. Just because every event has a cause does not mean there is a cause for all events. There might simply be an infinite series of such events with no need for a cause or explanation for that sum of events. (Bertrand Russell, somewhat famously, pointed this out.)
Clearly, reasoning about causes won't prove the existence of God. Thus, theists have turned to more abstract concepts in order to prove God's existence. One such concept is that of a necessary being. A necessary being is a being that exists in all possible worlds, as we philosophers like to say. It is a being such that, if it exists at all, it exists necessarily. There are, potentially, necessary beings besides God. If there are abstract objects, such as numbers or universals, then they exist necessarily. God, as an all-powerful being, is an inviting prospect for such a necessary being. The rest of the observable, physical universe, it is alleged, consists of only contingent beings, things that might or might not exist. For these contingent things to exist at all, the argument goes, they must depend on a necessary being.
Still, there are limitations with the concept of a necessary being. My shoe might be a necessary being if it exists in all possible worlds, but it may not be the creator of the universe in this one. Hence, something's being necessary is not sufficient for it being God. We need, further, some connection of this necessary being and a dependent universe. Typically, this connection is made by appeal to the concept of explanation which, so the theory goes, means saying why something had to be the case given whatever set of conditions were in place. To explain, one might say, why the moon orbits the earth, we must cite conditions such that, given that they are in place, no other alternative is possible. If the cited conditions do not entail that the event occur (or the state be the way it is), they cannot be considered a complete explanation.
This is a fairly intuitive concept of explanation. Suppose we learn that Bob did not get tenure at his job as assistant professor. To explain why he did not get tenure, we have to be able to say what those conditions were that led to him not getting tenure. It wouldn't be much of an explanation if all we could say was, "Sometimes people get tenure and sometimes they don't." Intuitively, you need to say why conditions were such that they had to work out as they did.
Theists have tended to think that this means that, ultimately, there must be a necessary being to explain why everything exists at all. After all, they say, if there are only contingent beings, then when you attempt to explain something, you've only just deferred the explanation to some further state that is also contingent. Atheists can, however, ask why there should be this uber-explanation. We know people give ordinary explanations all the time, for why Bob was denied tenure, for example, but we don't have any explanations that go back to first causes. We give explanations, but rarely of this uber-explanatory kind. (And some of our explanations appear to be even less complete in that they are only statistical, but more on that later.)
A better move is to demand that there be an explanation for why there is anything at all or why there is something rather than nothing. Explanations for the existence of individual things need never amount to an explanation for the universe as a whole, just as causes for individual events need not invoke a cause for the whole of the universe. But in order to explain why there is something rather than nothing, what is required is a being that explains why something's existing is necessary and which requires nothing else to explain why the explanans (the thing doing the explaining) exists.
William Rowe calls such a being "self-existent". A self-existent being is one whose existence is explained by the nature of the thing itself; its concept explains why the thing exists. Traditionally, God's existence is supposed to be explainable without reference to anything else because God is the greatest possible being and contingent beings (that might not exist) are less great than necessarily existent ones. (Please note this is not the ontological argument which tries to show that God exists through the concept alone. Here we only explain God's existence through God's concept, and the fact that we can do this for God and--presumably for nothing else--combined with the fact that something exists rather than nothing, is supposed to show that God exists.) Assuming the principle of sufficient reason (that every event that occurs and every fact have explanations), God's existence is explained by the nature of God, and God's creation of the universe explains everything else, so, on theism, everything has an explanation, whereas, on atheism, it is not clear that everything does.
Difficulties with God as ExplanatoryBefore I consider atheist responses, I want to point out that this idea that a self-existent being might create and maintain the universe in existence is less coherent than is sometimes granted. Recognizing the failure of cause-based cosmological arguments, theists have fallen back on explanation-based cosmological arguments that make no assumptions about time or efficient causes. In other words, because the argument that God must be the first efficient cause is unsound, they have banked on God explaining the existence of the universe as an underlying ground of being or fundamental essence of existence. Theists, and I'm thinking of Richard Taylor in his book
Metaphysics primarily here, consider the argument not to prove that there is a God who is the first in a series of efficient (triggering) causes, since any such argument is flawed, but as the "first" self-existent being that explains why the universe exists at all even if such a universe has existed forever with no temporal beginning. Instead, God is the most fundamental existent upon which everything in the universe currently depends for its existence. If God ceased to exist (a metaphysical impossibility on this view) or ceased to pay attention to the universe, then it would cease to exist.
In fact, there are theories that this has already happened and that's how we ended up with Newt Gingrich as Speaker of the House.
Sorry, needed some levity.
This idea of God as the fundamental ground of all existence is what leads to these odd formulations of the concept of God as the essence of being
qua being and that sort of thing. However, it is far from clear (1) what any of this "being
qua being" talk means, (2) that being
qua being is necessary for things to exist and (3) how God could be said to create the universe if God does not do so as an efficient or triggering cause but as a "fundamental ground of existence".
First, I'm not sure what this "being
qua being" talk means. Certainly it is true that things that exist "have" being in some sense, at least in the sense that they exist. But it is hardly clear that there must be some essence that they must have in order to exist. I understand what it means to say that a whole depends for its existence upon its parts, and we can thereby see that in order for the universe to exist, there must be some units or parts which constitute it. But what is it to say that these parts (fundamental particles, if you like) require this being
qua being in order to exist? There must, so the principle of sufficient reason goes, be an explanation for why they exist, but this might be given in terms of causes. If PSR is true, there must also be an explanation for the fact that there are fundamental particles at all, but I'm not sure how saying that they depend on any fundamental essence or being
qua being constitutes such an explanation.
Second, it is not clear that the universe requires God, understood as this being
qua being, in order to exist. I'm open to the idea that some theory of everything in physics might explain why they exist, but that explanation would do well to be more than that they are grounded in
being itself. In short, the intuition bus stops for me at the idea that there must be some further explanation for the fundamental particles. Maybe there is an explanation for why there are fundamental particles at all. And there must be such an explanation if PSR is true. I can see why people might think for any event that it needs a cause. How else could it have gotten here if it were not caused? one might ask. But I just don't feel that same pull for the question, "Why are there fundamental particles at all?" I'm saying here that the intuitive support for PSR may not be so strong when we think about fundamental particles as it is when we think of events needing causes.
Third, it is not clear how God can be the fundamental essence of existence and explain the universe's existence. God would not, on this view, have brought the universe into existence as a first, temporal cause. But if God did not do this, then it's not clear what it means for God's existence (choices, etc.) to explain why the universe exists at all. If God didn't cause the universe or bring it into existence, I'm not sure how God's existence can explain anything at all. The explanation must be that God's existence somehow underpins the existence of anything at all, that God is the fundamental property of all existence that makes it such that it exists. I believe at this point all talk of God has either lost all touch with reality or been reduced to tautology.
Atheist ExplanationsAtheists are mostly naturalists in that they believe everything can be explained in natural terms using the tools of contemporary science. There are well-known exceptions to the principle of sufficient reason from quantum physics, but, for the most part, it is taken as given that events have explanations. So, naturalistically-minded people should require an explanation for why there is something rather than nothing unless they can show that there is an exception for the case of the universe or that there is something ill-formed about the question "Why is there something rather than nothing?"
It is tempting to say that the question, "Why is there something rather than nothing?" is a non-sensical question. It involves an implicit contrast between a state in which something exists and a state in which nothing exists, and it is incoherent to say that there is a state in which nothing exists for to say that something exists is to say that there is something rather than nothing. Thus, we cannot even conceive or state the alternative necessary for the question to make sense. The cosmological argument, then, must fail to gain a foothold since there is no contrast case for which an explanation is necessary. (I think this is Sartre's view, but I get this only from my largely uninformed reading of
Nausea.) Since one cannot even assert the possibility of nothing existing, one cannot require an explanation for why it doesn't.
Alas, the problem is not so easily dissolved. To say that nothing exists is not to reify nothing and treat it like an ordinary object. To say that nothing exists is to say that it is not the case that anything exists; it is not the case that there is anything at all. There is no incoherence in such a thought any more than there is to say that Socrates does not exist. If it were not the case that anything at all existed, I would not be here to ask the question, but there is nothing nonsensical about considering whether everything might simply not exist. The question does not appear to be defective.
Others have claimed that the universe might have popped into existence as scientists working in the field of quantum electrodynamics have discovered particles sometimes do. This, however, does not explain why there is something rather than nothing because it takes for granted that there is a quantum field, and whatever that is, it is not nothing. Hence, this explanation does not appeal to a real contrast in which nothing is a possibility but a contrast in which a quantum field exists but no particles pop into existence in it. Similarly, one cannot explain why there is something rather than nothing by appeal to quantum tunneling in which a particle travels back in time in order to bring itself into existence since such things only occur in quantum fields, and, to explain why there is anything at all, requires that we not assume that there is anything as part of our explanation (unless that thing can be established to be self-existent).
My students always want to say that the necessarily existent thing is matter or mass/energy. We know, they say, that mass/energy can neither be created nor destroyed. That means it must exist and must always have existed in order for it to be here now. No one much likes the idea of positing some physical being (mass/energy) as necessarily existent because it always appears sensical to ask, "Why should that exist?" in a way that does not if we accept the theist conception of God. Of course, we might be wrong about what's possible and impossible, and it might be metaphysically impossible for mass/energy not to exist (rather than just a law of physics that it cannot go out of existence once it exists). But it sure seems to be a physical law only that mass/energy cannot cease to exist. And, more important, it is unclear why a law that says mass/energy that exists cannot cease to exist should be, in any way, relevant to why that mass/energy exists in the first place.
Rowe, forthrightly, thinks that PSR is not adequately supportable. Thus, there is no strong reason to think PSR is true. The general principle, that everything that exists or any fact that is true, has an explanation is false. Questions such as, "Why is there something rather than nothing?" need not, therefore, have an answer at all.
The worry about this response is that it threatens to be the same sort of science stopper that theism is. If we cannot be certain that everything has an explanation, then for any particular scientific experiment we do, there may be no explanation for why it turns out as it does. Hence, we would prefer an argument that there are explanations in the natural, spatio-temporal world, but that demanding an explanation for the whole ball of wax is an impossible demand.
"Why there is something rather than nothing?" need have no answer whereas empirically testable why-questions must have answers (at least in the non-quantum realm). But how can one argue for this?
Victor Stenger, apparently, argues that we can explain why something exists by contrast with nothing because nothing is unstable. I think what he means by this is that a quantum vacuum is unstable; things pop into existence spontaneously in the quantum field, so nothing would not exist for long. As noted, this argument is flawed since a quantum field is not nothing. Nothing is not a vacuum or a probability field. Nothing exists when there is not anything of any kind, when there is no location or place in which that quantum field might exist. So, Stenger cannot contrast something and nothing but only contrast some particles in a quantum field and no particles in a quantum field.
Is this real, absolute nothing stable or not? It's tempting to say that nothing is stable since, as philosophers used to say, "nothing can come from nothing". On the other hand, how can nothing have any properties at all? So, I'm not sure whether it's stable or unstable, but I don't see a good case that nothing (rather than a quantum vacuum) is unstable or would very likely (or inevitably) give rise to something.
Robert Nozick starts with the assumption implicit in my formulation of the question that explanation is contrastive. To explain why X occurred, we must explain why other possibilities did not occur. It makes no sense to explain, in isolation, why the US soccer team lost, we explain it implicitly by contrasting it to their winning. The contrast is explicit in asking why there is something rather than nothing. If it is logically impossible for there to be nothing, for example, then no explanation for why there is something would be necessary. Nozick does not attempt to prove that nothing existing is logically impossible, however.
Instead, Nozick argues that explanations are given in contrast to a likely, normal, or natural condition. Thus, we need not explain why something exists if nothing existing is statistically unlikely, abnormal or unnatural. Nozick then claims that, since there are an infinite number of ways for something to exist, and only one way for nothing to exist, it is infinitely more likely that there is something rather than nothing. Hence, if nothing existed, that would require explanation, but one need not propose an explanation for why something exists since it is the more likely situation.
I do not see how one could ascertain the likelihood that nothing exists, and Nozick's argument does nothing to establish that nothing existing is unlikely. Moreover, this assumption, that only unlikely, unnatural or abnormal situations can be or need to be explained is simply not true; Nozick is confusing a pragmatic convention with an epistemic necessity.
First, is nothingness statistically more likely or less likely than something? I have no idea how one could answer this question meaningfully. Since there is something and not nothing, perhaps the answer is that the existence of something is a certainty, the existence of nothing is then a statistical impossibility, and we do not need an explanation of the existence of something. That would miss the point since we have to make our probability claims in deciding what needs an explanation absent our knowledge that the thing actually exists. Otherwise nothing that existed would require an explanation. I have no idea how one would argue that something is likely. No one has any idea what the probability is that our universe have the constants, laws of nature or basic features that it does. How could anyone even begin to estimate the probability that there be anything at all (without even considering what such an estimate of probability could even mean)? Without a scientific theory of why something exists at all (which is the question at issue), I don't see how any probability estimate is possible.
Second, the claim that there are infinitely more ways for something to exist than nothing to exist is completely irrelevant to the probability of nothing existing. There are an infinite number of ways in which I might cease to exist from one moment to the next, and only one way to exist (in conformity to the natural law), but I know it's more likely that I will continue to exist as an integrated biological object it is that all those other possibilities (that I will turn into a chicken, that my brain will explode, that I will become a slice of cheese, a bowl of petunias or a whale, etc.) combined will occur.
Making an argument that something is more natural or normal is even less likely to be fruitful than arguing for statistical unlikelihood. I have enough trouble figuring out what it means for something to be natural or normal in other (natural or normal?) contexts, I have no idea how to establish such a claim in this context. I don't think anyone else does either.
Pragmatically, we often look for an explanation for the unusual, abnormal or unnatural because it is unexpected, and we tend to want explanations for the unexpected more than we want explanations for the expected. But that need not be the case. If John, a human being, is born with flippers instead of hands, we might explain this by pointing out that his mother took thalidomide when she was pregnant and that thalidomide causes birth defects of this sort. But if John is a dolphin, and his being born with flippers is normal, natural and statistically expected, we can still explain why he's born with flippers, and it is perfectly reasonable to ask such a question. The statistically likely can still be explained, and we can still reasonably demand such an explanation. In short, pragmatically we often look for explanations for the unexpected, but it is not part of the logic of explanation that we are required to do this.
This leaves us only with the remaining possibility of rejecting PSR and living with the consequences. Is there any legitimate reason to reject PSR or would this have to be an ad hoc atheism-saving move? As it turns out there are two good reasons to reject PSR. First, it is empirically refuted by quantum mechanics. Some facts about the quantum world have no explanation. Why an electron drops from one orbital to another
now rather than later has no explanation. Second, we are quickly returned to the infinite regress the theist hoped to avoid if we ask for an explanation for why PSR is true.
Suppose we accept, as does the theist, that PSR is true. There must, therefore, be an explanation for why it is true. It cannot be true because God acts only for reasons and thus nothing would occur without an explanation. Rather amusingly Leibniz gave this as his reason for belief in PSR, yet he later turned around and used PSR to prove the existence of God. Genius!
What other explanation could there be for PSR? It isn't self-evidently true, so we cannot expect it, like God, to be explained in terms of its own concept or nature. So, it must be explained in terms of something else whose explanation, we can, of course, also demand. This means, I think, that we must reject the idea that everything has an explanation.
The theist, in particular, must be unhappy to reject PSR since it underpins the cosmological argument, but to accept PSR renders the whole cosmological argument moot. The point of the argument is to show that we need not be caught in an infinite regress of explanations but can find a first, self-explaining explainer. Unless we reject PSR, however, we are stuck with exactly the regress that the appeal to God is supposed to prevent. And, of course, rejecting PSR means we no longer have the cosmological argument to support belief in God.
The naturalistically-inclined atheist must be unhappy with rejecting PSR in that it potentially blocks scientific inquiry. It's not clear how to limit the damage here. We could accept the "Every event has an explanation for why it occurs" part of PSR (call these type 1 claims) since that does not entail that PSR itself has an explanation. So, this principle is, at least, not caught in our regress problem. The "Every fact has an explanation" part of the principle (call these type 2 claims) remains problematic since PSR is supposed to be a fact. Yet, it is not easily possible to imagine that all scientific demands for an explanation can be understood as type 1 claims. It appears reasonable to demand to know why a falling object has the acceleration that it has, and this is not to be explained (apparently) just in terms of events occurring. Rejecting PSR, while independently necessary, has problematic consequences for science that render it problematic even when considered only as a rejection of this part of the principle.
ConclusionMy conclusion has to be tentative. Theism is not adequately supported by the cosmological argument since it relies on a principle (and part of a principle) that is false. We must, tentatively at least, reject the idea that there is an explanation for why there is something rather than nothing. Even if there were such an explanation, appeal to God would be ontologically excessive and unnecessary. More limited posits would, presumably, be preferable to the extravagant claims of the theists. We must, apparently, reject the idea that there are explanations for all facts, and, indeed, the fact that there is something rather than nothing might be (and certainly appears to be) the kind of fact we cannot explain. We must reject the principle of sufficient reason that supports the idea that there is such an explanation.
The difficulty is in limiting the epistemic carnage wrought by rejecting this principle, even if we must only reject one part of it. To say that some facts lack an explanation is problematic since any apparently inexplicable empirical result might turn out to be such a fact. It would be nice if there were some way to reject PSR only for foundational or fundamental claims, but I doubt that there is a clear line one could draw or a way to justify drawing it. Apparently, explanations, as Wittgenstein said, have to come to an end somewhere. There is, at any rate, no reason to think that end has to be God.
Note: I had to revise this a second time after some revisions simply disappeared. Thus, the fact that this post is incomplete or imperfect is entirely the fault of the technical glitch that destroyed all my perfect ruminations. None is the fault of the author.