"God" vs. Chickens That Make Themselves

Why a Circular Explanation is Worse than No Explanation

I recently argued that an infinite regress of causes implies a kind of circularity. The significance of the argument hinges on whether a circular explanation is worse than having no explanation at all. (This gist of the argument is this. An infinite chain can be divided into even-numbered members and odd-numbered members. Since the chain is infinite, the even-numbered members are explained entirely in terms of the odd-numbered members, and vice versa! This leads to a kind of circularity embedded in the infinity.)

But someone suggested to me in response that a circular explanation may not be any worse than no explanation. It has already been proposed, after all, that perhaps an infinite chain has no explanation as a whole. While there is an explanation of each member, there is no explanation of the entire chain. So if the infinite chain turns out to involve a circular explanation, perhaps that is not a worse scenario.

Here I’d like to suggest a reason to think that a circular explanation is actually much worse than having no explanation.

A couple disclaimers:

1. First, I acknowledge that a circular explanation may be no better than having no explanation. For example, suppose there is a chicken, and it simply appears from nowhere without any explanation. That is counterintuitive. It goes against our experience. But suppose someone says, “No, it has an explanation. That chicken produced itself.” That does not seem to help. The puzzlement remains. Why is there a chicken that produced itself? If we then say, “The chicken produced the producing of itself,” we have not removed the puzzle. We have only pushed the circle one step back.

   This suggests that the call for an explanation is, in many contexts, a call for an external explanation. If something calls for an explanation, then it is hard to see how the explanation could be the very thing that needed explaining in the first place.

2. Second, I also acknowledge that, when it comes to unexplained things, many candidate examples would be puzzling if they had no explanation. A chicken is one example. Chickens are the kind of things that have explanations. So if we compare a circular explanation of a chicken with an unexplained chicken, perhaps these cases are, in some ways, equally puzzling. They are puzzling precisely because chickens are the kind of things that seem to call for an explanation.

However, I think there is a difference between something that has no explanation at all and something that is explained in a circular way. A circular explanation violates a condition of explanation: explanations are anti-symmetric. On an ordinary, intuitive understanding of the term, it is in the nature of explanation to explain something else. If I want to know why something exists, and the answer is the very thing itself, then the answer seems to contradict a certain ordinary notion of explanation. (I leave to the side whether there might be other, more peculiar notions of explanation that admit of circles.)

There is another possibility: there could be a kind of thing that could not have an explanation. That kind of thing, even if only hypothetical, would be different from something that is explained in terms of itself. A self-explanation contradicts the asymmetry of explanation. But the absence of a further explanation need not do that.

For example, suppose there were a field of energy that exists of necessity by its nature. It cannot fail to exist. It never comes to be or ceases to be. And suppose its nature implies that anything with that nature cannot be preceded by any prior explanation. In that case, we may have a reason to think that this thing could not have a prior explanation. That is not puzzling in the same way as saying that it has a circular explanation or in saying that a chicken appeared from nowhere without an explanation.

Upshot

So I think it comes down to the kind of thing in question. I have argued in my work that the deepest possible explanation of all of contingent reality would be in terms of something that cannot have a further explanation, and that it would exist without the possibility of not existing. That is very different from something that must have a further explanation and then receives that explanation from itself.

We can distinguish three different puzzling situations: an unexplained chicken, a self-explained chicken, and an infinite regress of chickens. The infinite regress of chickens, according to the argument, may imply a kind of circular explanation. And a circular explanation may be no better than an unexplained chicken.

But this leaves open another possibility: something could be unexplained because its nature could not be explained in terms of anything else. That sort of stopping point, it seems to me, is not nearly as puzzling as a circular explanation. For while it raises questions about what kind of thing could be unexplained, it allows us to preserve an ordinary concept of explanation without violating the asymmetry condition.

p.s., I’m creating a course, The Truth-Seeker’s Guide to Power, for people who want to grow in clarity when life’s deepest questions feel confusing. This course will be provide you a systematic guide to your powers in a series of steps in a personalized way. Pre-order window is open until July 4th.