You’ve probably heard the story about the old lady who tells a Famous Professor (whose exact identity varies) that the world is actually sitting on the back of giant turtle. 

“Then, what is the turtle sitting on?” asks the bemused professor.

“Another turtle,” the old woman replies. 

“Well and good. But what is that turtle sitting on?” asks the professor, smiling. 

“It’s no good, Professor,” the woman shoots back impatiently. “It’s turtles all the way down!”

A while back, I noted the rhetorical oddness when biologist Peter Corning told that story in order to explicitly identify his theory with the “turtles all the way down” view. But apparently this is becoming a trend. The Turtle Lady seems to be accumulating disciples — including quite a few of her former intellectual opponents, the Famous Professors. A new case of this can be found in the bestselling book Determined: A Science of Life Without Free Will, by Stanford neuroscientist Robert Sapolsky. I recently started reading the book because I was reviewing another book defending free will (which was also written by a materialist neuroscientist and had come out the same year) and I wanted to see what the other side was saying. And I have to say — Sapolsky is a delightful writer. He has a warm and humorous narrative voice, which he puts to good use retelling his version of the Turtle Lady story in the introductory pages of the book. After taking his time with the story, Sapolsky concludes:

Here is the point of this book: While it may seem ridiculous and nonsensical to explain something by resorting to an infinity of turtles all the way down, it actually is much more ridiculous and nonsensical to believe that somewhere down there, there’s a turtle floating in the air. The science of human behavior shows that turtles can’t float; instead, it is indeed turtles all the way down. [Emphasis in the original.]

Sapolsky argues that because everything must have a cause, there can be no free will. Instead, each instant is a snapshot of an arrangement of atoms and molecules that followed inexorably, according to the laws of physics, from the arrangement of atoms and molecules the previous instant — and so on, forever. The idea of something breaking this causal chain and just “doing as it pleases” is absurd, because this would be like a World Turtle floating in the air: an explanation without an explanation. An infinite regress, as strange is it may seem, is actually the more reasonable option.

It is to Dr. Sapolsky’s credit that he realizes that his worldview demands such infinite regress, and stakes his flag on this position from the outset. He is also correct that a realist view of free will requires something “floating in the air” so to speak — something that simply is what it is, and does what it does, without further explanation. 

Nevertheless, his choice of analogy is a bit unfortunate, because, embarrassingly enough, the Earth actually is “floating in the air” (or in empty space, rather — which is no better). To make matters worse, the explanation for why this is so — the law of gravity — also seems to be “floating in the air” without underlying explanation.¹ I’m not sure whether Sapolsky expects there to be an infinite regress of causes for physical laws — with the law of gravity dictated by another law that says, “There shall be a law of gravity,” which is itself dictated by another law, and so on ad infinitum — but that would seem to follow from Sapolsky’s principle that a turtle floating in the air is more absurd than an infinite regress of turtles. 

In fact, there are serious problems with the “infinite turtles” model of reality. The question really goes to the roots of metaphysics and what qualifies as an explanation, so it has bearing not just on the question of free will, but on many topics of debate — including intelligent design, and the naturalist paradigm itself. 

So let’s take a moment, and take the Turtle Lady and her disciples seriously. Let’s really answer the question: Which is more absurd — a floating turtle, or “turtles all the way down”? 

Critique #1: It’s Not What Science Shows Anyway

Before we get into the theoretical problems with “turtles all the way down,” there’s a more basic problem: the empirical scientific evidence doesn’t seem to point to that. 

And Sapolsky knows that. After basing his case on the relative reasonableness of a causal chain that extends back forever, and characterizing the alternative as laughable, Sapolsky admits (in a footnote) that he doesn’t really mean it:  “‘Forever’ might not really be the case,” he writes, “because, at some point in this regression, you get to the Big Bang and whatever came before that, about which I understand precisely zero.” 

This is not an inconsequential concession. Scientists up until the 20th century used to assume that the physical universe was eternal, and therefore without cause. That’s where Sapolsky’s worldview originates from. But the Big Bang validated the philosophers who had said all along that this was impossible (see below). Since time itself seems to have started in the Big Bang (based on physics that I don’t understand any better than Sapolsky does), it now seems scientifically impossible to invoke a causal chain of events stretching back in time forever.

Not only does Sapolsky admit that his infinite regress might not actually be infinite, he admits that that “the majority of physicists accept the indeterministic picture of quantum mechanics” — so, looking at the problem from the other side, it’s really not clear what scientific basis he has for rejecting non-deterministic or “floating” events. 

It’s beginning to seem like just a stubborn preference. 

And we get the same thing yet again when we follow the casual chain in a different direction — going explanation by explanation, rather than temporal cause by temporal cause — and ask, “What is our conscious experience of free will? Can it be reduced to any physical explanation?”

“I don’t understand what consciousness is, can’t define it,” Sapolsky candidly admits. “I don’t understand philosophers’ writing about it. Or neuroscientists’, for that matter, unless it’s ‘consciousness’ in the boring neurological sense, like not experiencing consciousness because you’re in a coma.”

But free will is first encountered as part of our subjective conscious experience — we feel that we do things willfully, not randomly or according to pure necessity, and that’s why scientists are forced to write books explaining why free will is an illusion — so if Sapolsky won’t even try to explain the phenomenon of consciousness, he has essentially given up explaining free will.

Critique #2: You Cannot Cross Infinity

Let’s put aside the empirical validity of the infinite turtles theory, and consider its theoretical merits. In the classic materialist paradigm, anything that cannot be reduced to an infinite regress of physical causes, stretching back forever, is dismissed as a “supernatural” explanation. Something just floating there, without a natural cause, as if by magic — that’s unscientific, and therefore false. 

In one important way, this thesis is actually more absurd than the original Turtle Lady’s thesis. That’s because the Turtle Lady was merely invoking an infinite regress in space, but causal determinists are invoking an infinite regress into the temporal past. This is more difficult, because in time (but not space) you must have crossed the whole infinite progression in order to be at the “top” (i.e., the present). That is, with time it is assumed that you start in the past, and move future-ward. 

The problem is that you cannot traverse an infinity one finite step at a time. But to get to the present moment from an infinite past, you would have to have done so.

Why can’t you transverse infinity?  Because you never actually reach infinity by proceeding incrementally by finite steps: there never comes a moment when you say, “a gazillion and one… a gazillion and two… infinity!” Infinity is not any finite number plus one; any finite number plus one is just another finite number. 

And yet, if there is an unbroken chain of moments stretching back infinitely far into the past, then (a) there must be a moment infinitely far back (if there’s not, it’s not an actual infinity), and (b) we must have gotten to the present from that moment, by moving forward one moment at a time.

Since (b) is impossible, the model is false. 

This problem has been discussed by various philosophers, and there are some attempts to get around it. One is to say that we didn’t actually get to the present moment by inching our way through the past; in other words, time as we know it is an illusion. 

Sapolsky seems to hold this view of time, suggesting that the inexorability of cause and effect “means that past and future are identical, that there is no direction of time, that events one second in the future are already the past of two seconds in the future. Which makes me feel queasy, reminding me that I’ve already died somewhere in the future.”

This view of time is surprisingly popular, but it has been pointed out that it fails to explain why “you” (your conscious experience, that is) seem to be here in the present moment, rather than other times.

I suppose there are two ways to answer that. The first is to say that although the progressiveness of time is an illusion and everything past, present, and future exists simultaneously, we (that is, our conscious experiences) are moving along the static timeline from past to future. But then you would have to ask… moving in what? In time, clearly! So the temporality of time has not been removed, only hidden. You now have two timelines, instead of one. 

The second way to interpret this would be to say that we are not moving along the timeline at all. This, in turn, results in two possible interpretations: that our conscious experience actually exists only in this moment (the real “present”), and we merely think that we came from the “past” and will later be in the “future”; or else that our conscious experience is simultaneously present in multiple moments of our lives in the timeline (presumable all of them) — and every moment is being simultaneously experienced as the “present” by a copy of our conscious experience. Either way, the sense of movement is an illusion. 

I feel there are a couple problems with this theory. First, it fails to explain why we remember the past but not the future. It seems like under this model we should either know only the present moment, or else every moment that we inhabit. The second problem is that it has not explained where the illusion of temporal movement comes from; why do we feel that time is moving? Indeed, how could one suffer from such a fantastic delusion if there were no such thing as “movement” or temporal change in the universe in the first place? The third problem is simply that this model wildly contradicts our most basic experience of reality; it seems intuitively more absurd than either an infinite chain of turtles or a turtle floating in the air. 

Nevertheless, I’ll concede that such a state of things might at least be logically possible. So we’ll move on to the next problem with the infinite turtle model of reality. 

Critique #3: Infinite Regresses Don’t Actually Explain

The classic problem with infinite regress explanations is that you never get to a final explanation, so each explanation in the chain seems like, in the words of one philosopher, “a promissory note that is never paid.” 

A few contemporary philosophers disagree with this, arguing that the validity of the “local” explanations do not depend on there being a “global” explanation, and that because the explanatory power of each explanation does not depend on whether or not that explanation has its own explanation (*gasp*), each explanation is not a mere “promissory note” but a perfectly valid explanation in its own right.  

But to my mind, this falls apart pretty quickly. Consider the turtle tower again. The reason the explanation of “turtles all the way down” is ridiculous is that there is no explanation for why the whole turtle tower is not falling together. There seems to be no reason why the infinite regress of turtles wouldn’t be rushing downwards as a collective unit. And if the tower were collectively falling, then of course each turtle would also be individually falling. The failure of the whole is implicit in the failure of the parts, and vice versa. 

Critique #4: You Can’t Escape the Floating Turtle 

But suppose we concede this one too. The disciples of the Turtle Lady would not be out of the woods yet — there is still another problem. 

Suppose we say, “Okay, let’s accept that the universe is a branching network of explanations and causes with no end anywhere — receding into infinity in every direction. What explains the whole system?”

To answer that, you have two options: to propose a cause, or not. If you propose some cause for the system, this cause will by definition be part of the system — a cause on the causal tree. Yet, according to the no-floating-turtles rule, that cause did not cause itself; therefore, at least one part of the system has been excluded and left unexplained by it. Thus, the cause you proposed is not actually a valid explanation for the whole. So you are left only with the other option, which is to not propose a cause: to say that the whole system simply is, for no reason — it’s just “brute reality” (as materialists are fond of saying of the physical universe). But that is itself a floating turtle — the mother of all floating turtles, in fact.

That means that the infinite regress has failed to help you avoid the type of explanation you were trying to avoid — an explanation that is merely “floating in the air” without a preceding explanation or cause. Yet the initial justification for invoking an infinite regress was that this kind of explanation is not allowed! So you have lost your prime reason for invoking an infinite regress to begin with.  

In Conclusion 

Sapolsky notes that the statement “It’s turtles all the way down!” is a good punchline because one imagines the esteemed professor struggling to refute it. It is hard to refute, and I suspect that is why a version of the argument has become popular among esteemed professors. Making yourself irrefutable is one of the perks of possessing a towering intellect (I imagine). But that has its dangers — intelligence can be used to avoid the truth as easily as it can be used to find it. Which is probably why some of the most intelligent people seem to hold some of the silliest opinions. 

Come to think of it, expecting the most intelligent people to speak the most good sense is a bit like expecting the most powerful monarchs to be the most beneficent. Beneficence may be the right and proper use of their power, but it is not the way power is most often used. Intelligence is power too, and the more power you have, the greater the temptation to use it badly — especially when the alternative is particularly unappealing. 

In this case, the alternative is admitting that there must be an end to explanations, and therefore an end to your own wisdom. And you would have to begin seriously asking, what lies at the End? 

The danger of asking that question is that you might answer it — with simple knowledge, not causes, definitions, or explanations — and you might not like with what you find.  

Notes

1. If not the Newtonian law itself, then the quantum physical explanation a few steps deeper — once physicists can agree on what exactly that is.