Welcome to the ultimate free lunch: existence without an ultimate cause.
According to my made-up and inarticulated measures, “why anything exists at all” is the central question. It is neither just the central question of metaphysics nor just the central question of philosophy — it is the central question independently of context or universe of discourse. It’s the supreme and ultimate question. This is just me being fancy and spouting non-sense, though.
It first struck me, perhaps, when I was 14 years of age, that the Universe, and anything that may lie “beyond” it, could have not existed. The whole of reality seemed to me contingent; not only all human experience, but all of being. (I also got a kick from imagining absolute non-existence, which is very weird and amuses the mind, but alas, it is not incoherent.)
And this fact, or this apparent fact, of contingency made me think there must be a reason for the existence of anything at all. Perhaps it was because of an innate human intellectual tendency or intuition, or perhaps because I roamed Enlightenment-driven areas of the internet, the principle of sufficient reason always seemed very compelling to me. So compelling that even necessary entities seem to me to need reasons for their existence. But the idea is that they themselves constitute the very reason of their own existence. That is how (many) philosophers have put the matter for a long time, and it seems to me correct. My reasoning will, then, abide to the principle of sufficient reason as well as to the idea that necessary entities, if they existed, would need to be their own reason for existing.
(Note: I am not sure who doubts the principle of sufficient reason. Perhaps, members of a distant primitive culture untainted by Western presuppositions do doubt it. (Side-note: it’s funny how the term ‘distant culture’ had a sort of veil of mystery surrounding it. Now we have satellite images and anthropological accounts of all these tribes; they don’t seem so distant anymore.) Surely many a a philosopher, from any culture, doubt this principle, but I don’t know who. Nor do I know why or how people doubt this. More on the strength of this principle later.)
I really don’t want to reject the principle of sufficient reason. It states everything has a reason, an explanation, something that caused it or made it so it was or could be the case. It states there are no brute facts, no facts that have absolutely no explanation. Colin McGinn suggests in an interview that explanations must end somewhere, and that this proves there are brute facts. Further on I will argue against this, that explanations don’t need to end anywhere and that throwing out the principle of sufficient reason is too high a price to pay. His argument is very interesting, though: you can perfectly explain why there are trees, you say something like “because of these other things”. But when you take the whole of reality, there is nothing else you can use to say: “anything exists at all because of these other things”. The question “why is there something rather than nothing?” has the same form of “why are there trees?”, but it cannot be answered in the same way. More on him later.
Some people may argue the Copenhagen interpretation of quantum mechanics proves some events happen without sufficient causes, and thus that the principle of sufficient reason is false. In it, there is no sufficient reason for a particle having done this or that. However, even then there is still the fact that there is a physical law which attributes probabilities to different outcomes and, more importantly, makes the outcomes possible in the first place. The particle wouldn’t go that way if it wasn’t physically possible, or if there wasn’t energy to do so. Even in quantum indeterminacy, then, there is still something prior to that particle’s movement which is involved in the production of that movement, even if not in a way to produce a sufficient reason. No thing happens without any reason at all. (Perhaps we should call our principle, then, the principle of prior reason.)
Now, if the whole of existence could have not been the case, then there must be a reason for it being the case (or, in a fancy manner: for it being). The whole of reality couldn’t have been its own cause of existence, because something must exist before being able to cause anything. On top of that, if it was something that caused its own existence, then how do we make sense of its contingency, of its could-have-not-been-ness? If something causes itself, then it is a necessary being. Since I was thinking reality is contingent, then there must be something outside (or “beyond”) it which was or is the cause of its existence. However, ex hypothesi, there is no thing beyond it, for we defined our universe of discourse so as to encompass all there is! This is what McGinn had in mind.
On pain of rejecting the principle of prior reason, then, we must suppose there to be some existent, something, which needs to cause outside it, but rather has its cause inside. In other words, there must be a necessary event or entity. So it seems we have arrived at the conclusion that something must be its own cause. However, this does not make sense: something has to exist before causing anything, and thus supposing something to cause itself begs its very existence. [Objection #2 attacks this point of the reasoning, and I answer it interestingly.] Thus, it makes no sense, at least in my conceptual framework, for something to cause its own existence. Only existents can cause anything.
This brings me back to McGinn. He seems to fall back on the principle of sufficient reason when he says brute facts imply some entities are self-subsisting, but self-subsistence does not make much sense. (But he said this in agreement to what the interviewer said, perhaps he doesn’t really think it but only assented to avoid being nit-picky in a short interview.) I say he falls back on the principle because self-subsisting is clearly attributing a reason for something’s existing: itself. Furthermore, as I have argued, self-subsistence is circular. The best you could say is that something keeps existing because its own existence makes it so it cannot stop existing. But you cannot say somethings exists in the first place because of itself. Necessary entities do not make sense and could not exist. There is nothing that is causa sui. (See my final note, at the very end of this essay, regarding this conclusion.)
So if nothing can be the cause of its own existence, and the whole of reality needs a cause for it, then what explains the existence of anything at all? I propose an infinitary solution. If we have an infinite chain of causation going back forever, then every link in the chain — every entity, every event — is explained or caused by the preceding link. This is the case because, in an infinite chain, at least the one I am envisioning — which has no dead-ends going backwards — every link has a link preceding it. There is no first link, no first event. (Note: the real line, between 0 and 1, has an infinite amount of entities, even though it has a beginning. However, in the real line, no member has a definite ‘previous member’. Succession is not well-defined in the real line.)
Since all there is,exists, or happens are links in the chain, it follows that everything that is, exists, or happens, has an explanation.
Keep in mind that this sounds very weird and that it should be rejected if possible, but also keep in mind that all the alternatives seem even more absurd. See my response to objection #1 for a defense of the principle of sufficient/prior reason, and see my response to objection #2 for a stronger refutation of a self-causing entity.
EDIT: Backing me up, some 22 hours after writing this post I stumbled upon an interview with Quentin Smith in which he gave the very same answer as I did: every state of the Universe was caused by some previous state. I found it weird, however, that he did not emphasize his argument implies there has been an infinite amount of states of the Universe (or of reality), with no beginning. It is only through infinity, no-beginning, that this model works.
Objection #1: Something still lacks an explanation. Regarding the infinitary solution, we may feel like the mathematician Georg Cantor did when he made a profound new discovery about infinity. “I see it,” Cantor exclaimed, “but I don’t believe it.” (Adapted from Jim Holt’s ending of chapter 2 of his Why book.)
Even if all the links of the chain are explained, someone may object the existence of the chain itself has not been explained. I can see the appeal of this, there is something weird about anything existing at all with no ultimate explanation. However, the following seems to be true: in this model, the chain “itself” is entirely composed of individual links, and each individual link has a perfectly good explanation (the immediately previous link). So the whole chain has no part that has not been explained. Does someone want to say that a whole still needs an explanation even if all of its parts — and even the arrangement of the parts, their connections and relative positions! — are entirely explained?
Upon the force of this argument, I must accept that the infinitary model is, say, “closed under explanation”, even though on first blush there strongly appears to be something still in need of explaining. And upon the force of the previous arguments, I am forced to accept the infinitary model. At least, supposing we don’t throw out the supremely intuitive principle of prior reason (which seems an illegal move, if anything is!), which is the same kind of idea than ex nihilo nihil fit is (no thing can come from nothing, no thing can happen without anything making it possible in the first place). There can’t be no criteria for possibility.
Objection #2: Temporality, causality, and self-causation. Someone may object I am thinking temporally when I say a cause must precede its effect, which automatically rejects the possibility of (i) something causing its own existence, and (ii) that very same something coming into existence, at the very same time. However, I am not talking of a temporal beforeness, or a temporal preceding, but a logical one. If A causes B, and B is not overdetermined (i.e. also caused by things other than A), then if A does not happen or does not exist, then B does not happen or does not exist.
Thus, if some entity’s very existence causes its own existence, then the following counter-factual is true: if that entity did not exist, then it would not have existed. Since there is no argument establishing this counterfactual cannot be the case (in fact, we are criticizing the very argument that purported to show this!), then it is possible for such an entity to not have existed, and then we fall back to the need of providing an explanation: why did the supposedly self-causing entity exist in the first place, so that it could go on to cause its own existence?
Possible mistake: Conflating explanation for causation. Perhaps some principle of prior reason is true, but it states that everything has an explanation for its being (being neutral on whether this explanation is intelligible to human beings or intelligible to a cognitively ideal entity), without this implying everything has been caused by something. If identifying explanation with causation is a mistake, then my argument falls apart: necessary entities wouldn’t be explained as self-causing entities (which doesn’t make sense), but as something else which is not amenable to my circularity objections.
(Recall the principle of prior reason, in the causation-interpretation, was what lead us to say: something has either a cause outside it or inside it. This is what eventually led us to reject the possibility of necessary entities.)
EDIT (01/01/17): I want to read more on this: “The assumption that explanations must always involve “things” has been called by one prominent contemporary philosopher, Nicholas Rescher, “a prejudice as deep-rooted as any in Western philosophy.” Obviously, to explain a given fact—such as the fact that there is a world at all—one has to cite other facts. But it doesn’t follow that the existence of a given thing can be explained only by invoking other things. Maybe a reason for the world’s existence should be sought elsewhere, in the realm of such “un-things” as mathematical entities, objective values, logical laws, or Heisenberg’s uncertainty principle.” (Chapter one of Jim Holt’s book.)