About three years ago I became a theist because I found the cosmological argument irresistibly compelling at a rational level. I did not become affiliated with any religion straight away, although I became a Christian not too long after. Today I no longer believe the cosmological argument is sound or can be known to be sound. Let me emphasise this point because it has two parts: (1) I do not find any version of the cosmological argument known to me to be rationally compelling and (2) I do not believe that such an argument can be formulated such that the conclusion (that God exists) is known with certainty.
Where that places my Christianity or theism more broadly is unclear to me. Some people contend that it should remain entirely unaffected because I have a personal relationship with God which transcends rationality. However, relationships seem to me to be based on elements of reason and other elements which go beyond reason. This can be illustrated with the usual analogy used for arguing faith as a virtue: if there seems to be evidence that my spouse is cheating on me, I give her the benefit of the doubt because of how much faith I have in her. But what is the basis of this faith? It is the experience of my spouse in the past and from there, the knowledge of her character. It is a faith built on evidence.
Belief in God, I would say, is much more like online dating without the benefit of cameras or telephones. Certainly, over the course of an online friendship, two people would get to know each other and some degree of trust could establish itself. In the back of their mind, however, is always the knowledge that in the past these people have turned out to be fake, always the knowledge that there exist master deceivers online in these forums. So faith is possible but it seems to lack certainty. Is this a problem? Perhaps not. Few things in life are certain. For me, nonetheless, the cosmological argument had given me a scrap of the closest thing to certainty I could have outside of the truths of mathematics and those most likely truths gleaned from the natural world. So for the argument to be unsound is tantamount to having the online dating website put a marker on the beloved's profile marking it as "Doubtfully Genuine." Any semblance of a relationship is viewed with suspicion and as perhaps the cunning ploy of a deceiver.
Let me enumerate the reasons that I do not think that the known cosmological arguments are sound. First of all, I have never been Aristotelian in my metaphysics (I have viewed with much more agreement the views of Locke and on occasion Hume and Kant, though I have no established metaphysics within which I operate) which means that I invariably find the arguments of St Thomas Aquinas to need reformulation, if only linguistic and cosmetic at points. Aristotle's metaphysics suffers from the same problem, in my opinion, which besets his natural philosophy: despite beginning with some observation, the addition of wildly speculative elements and, more importantly, ontologically realising his linguistic constructions renders a metaphysics which seems common sense but lacks critical grappling with real metaphysical problems. Aristotle's defenders continually accuse his critics (at least in my non-professional experience) of rejecting first principles which should be assumed as axiomatic, and yet this label given to common sense principles does not allow for their critical analysis. It is not the job of philosophy to make common sense reputable, but rather to seek for actual truth, and if common sense is misleading or false, then all the worse for common sense.
Secondly, to his first and second ways I say that I am entirely comfortable with the possibility of an "actual infinite," and so deny that there needs to be some first mover or first efficient cause. I can see why a naive view of infinities leads to paradoxes but in my studies of mathematics I have found no logical contradictions when infinity is rigorously defined; this is by no means to say that the conclusions of operating with infinities are common sense. However, un-intuitive should not be taken to mean false lest one makes of one's common sense an idol.
Thirdly, when it comes to arguments from contingency (as Aquinas' third way is), claims are made which are entirely unverified and implicitly deny, again, the possibility of an actual infinite. How does one know what sort of entity produces a contingent event? Clearly we know from experience that contingent events are produced by other contingent events. Why is it not possible for this to have gone on eternally in the past? More importantly, arguments from contingency are liable to the fallacy of composition which asserts that because the members of some whole have a property, then the whole must have it. This is sometimes true (if all the parts of a car are red, then the whole car is red) but it is not necessarily true (carbon atoms do not have the same properties as carbon allotropes like diamonds or graphite, or diamond and graphite would be identical). I see of no way of inferring that the universe is contingent based on its parts, particularly since:
Fourthly, even if a necessary agent was somewhere necessary in a sequence of contingent events, there is no reason to think that this agent is personal. It is quite conceivable that laws of nature are metaphysically necessary and so can produce contingent events. Why is the world the way it is? We explain everything in it by natural laws, so it is plausible that it can be explained as a whole by natural laws (though here I am asserting only possibility and plausibility). It is sometimes objected that natural laws only explain how to go from one state of affairs to another, not how the whole chain started, to which I reply with two points:
Fifthly, that the only natural laws we know are the ones that take one existing state of affairs to another by some means and mechanism, but this does not imply that these are the only natural laws that exist. This is particularly important because when cosmological arguments make claims about the beginning of all time-space and further claim that such an event would be impossible without a personal agent, they are implicitly are claiming knowledge about atemporal causation. What caused the Big Bang, they ask? But such a question requires a different type of causation to the one we are familiar with in day-to-day life because our daily notions of causality are that some event A which precedes B in time can be the cause of B. Yet there is no possible event of this type preceding the beginning of time, and so we must remain entirely agnostic as to what sort of causality is required. If the Big Bang was really the absolute beginning of all of time and space, then when it is said that it had a cause we are importing an entirely foreign notion to the event. When an argument like the Kalam says that everything that begins to exist has a cause for its existence, it makes a claim bigger than the one we have experience of: that everything that begins to exist within time has a cause preceding it in time for its existence. So to that premise I simply say: I cannot know.
Sixthly, the rather major point that I am an instrumentalist and so I do not believe that the scientific theory of the Big Bang can be thought to have metaphysical implications. If you want to know more about the point of view known as "scientific anti-realism", have a look at the article on the Stanford Encyclopedia of Philosophy. This means that in arguments like the Kalam where the universe is said to have a beginning, that premise must be proved by non-scientific means. If I was convinced of scientific realism, however, or even realism with respect to certain tenets, then I would furthermore believe two final points about quantum mechanics:
Seventhly, that the probabilistic nature of quantum mechanics (or in other words, its under-determinism) leads to profound implications for the Principle of Sufficient Reason (PSR) which underlies most if not all cosmological arguments. Where previously it was held to be certain that, if one measured something in a laboratory then it could in principle be explained uniquely by a preceding event, set of events or series of conditions, this need no longer be held. Every known result of quantum mechanics is consistent with the idea that an identical experiment can have different outcomes to which the only explanation is that the different outcomes were all possible. Because of Bell's theorem and its violation, I think a substantial case can be made that this result of quantum mechanics can be known to be metaphysically true.
Eightly, the interpretation of quantum mechanics which I favour as satisfactorily and elegantly explaining the results of physics is the many worlds interpretation. This actually is in dissonance with point seven and yet it still undermines some varieties of cosmological argument (albeit in a different way). This is because it removes the probabilistic nature of quantum mechanics so much that it makes every event uncontingent. In technical terminology, all of physical reality is in a superposition of quantum states which make up a wavefunction that never gets collapsed. The classical analogue would be like flipping a coin and both heads and tails came up in different worlds which branch off from each other, equally real, but we experience only one of the two. To those who believe that many-worlds is quantum mumbo jumbo, I refer this article from Sean Carroll. The consequence of the many worlds interpretation is that the sum of all the contingent worlds makes a necessary whole, much like the sum of the two outcomes of the coin toss makes a necessary whole. That falsifies the claim of the argument from contingency.
------------------
For those who are still reading I am going to add a final point in support of my claim that the soundness of the cosmological argument is unknowable with certainty by reason and experience that is available to us without extra revelation (which would be in contention anyway). It is, you might say, an argument from pessimistic induction: as our knowledge of science progresses and we learn more about the natural world, what we learn above all is how alien to our common sense it is. This makes evolutionary sense: the minds of human beings were not formed to contemplate quarks or comprehend how energy produces the curvature of space-time.
It is unsurprising that we intuit things that are false, but if we are honest with our findings we must acknowledge that the world is a weirder place than we could previously assume. The more we realise how unlike our daily life the real world is on scales that exceed it (the very small, the very fast, the very big, the very hot, the very cold, etc.) the more we must acknowledge how little we can justify on the basis of what appears obvious. Cosmological arguments are obvious arguments but the cosmos is not obvious. Therefore, to any claim of knowledge furnishes the premises of such arguments I respond "How do you know that?" Do not be surprised if I do not consider "It is obvious" to be an answer.
