As I often say, the mistakes of a great philosopher are worth the philosopher’s weight in gold. Kant said plenty of things to object to, but one thing Kant got right is that we can’t get free will and the laws of nature to play nicely together (at least not the current laws of nature — but that’s another story). Kant’s “solution” to this problem, in my view, is to impose yet another version of dualism. There’s another possibility — but first, let’s get clear on the problem.

From a “scientific” perspective, if you know the state of a system at a particular time and you know the laws that describe the behavior of all the elements of the system, then you can calculate any future state of that system. This is how we can calculate things like the positions of planets and the day and time of eclipses. We know the state of the solar system at some particular moment in the past (lots of moments, actually), we know the mathematical laws that describe the behavior of the sun, planets and their moons, etc., and all it takes is some math to calculate where and when something is going to happen.

To see the problem between this approach and free will, just think of yourself as a “system”: A bunch of chemicals and circuitry, inputs and outputs, and so on, all describable by means of mathematical laws. Yes, it’s incredibly complex and yes, we may not understand the system (yet?) or the laws (yet?), but at least hypothetically, we could know both the state of the system at some moment and the laws describing it. And so — as in the case of an eclipse — we could calculate what you’re going to choose for breakfast three months from this morning.

Now, contrast how you appear to be from that perspective with how you appear to yourself when you’re actually deciding what to have for breakfast. Better still, try this philosophical experiment: Wake up tomorrow and adopt the perspective of an observer: Watch yourself, and wait to see what you decide to have for breakfast.

If that sounds silly, it’s because it is — but the silliness is the philosophical payoff. You can’t think of yourself as a “determined” system in the way I described above and simultaneously as an agent who chooses. Something has to give, right? Well, not so fast. Suppose you have a medical condition and consult with your physician. Suppose your physician says, “Well, you’re an agent, so I recommend that you decide not to be ill.” My guess is that you’d be finding another doctor. But why?

A simplistic answer is because the doctor took the wrong perspective for the problem in that situation. I can tell a similarly silly mismatch story from the other side: Suppose you are in a particularly bad relationship and you ask your therapist what you should do. Suppose your therapist says, “Well, given your present state and the regularities of the laws of nature, I recommend you give it a month and wait to see what you do.” Same outcome: Look for a new therapist.

Kant acknowledges that we need both these perspectives, but for different reasons. We need the state/laws perspective to do science (like physics and medicine), and we need the agent/freedom perspective to do things like make choices and moral judgments. (If we’re all states and laws, then no one is actually choosing anything, in which case no one is accountable for what they did.) In this post, I’m going to be horribly unfair to Kant and justify it by noting the countless hours I’ve spent reading and teaching Kant. I’m just going to say that his solution was to (re)impose a kind of dualism — that I don’t think we need.

Let’s try a different approach, appealing to a postulate of continuity. Suppose reality is actually more complex than either of these perspectives, so that, no matter what perspective we take, there’s always more that’s “left out” of our thinking. (Adorno makes this point forcefully and beautifully in Negative Dialectics, for those keeping score at home.) If we take this path, then the accounts that result from these perspectives are “compatible,” but not for the reason you might think: They are compatible because they are always necessarily incomplete, abstractions away from the fullness of reality.

Except for one major problem: How can we talk about reality if what we say and think is always an abstraction, inadequate to reality? This, in a nutshell, is what I like to call “the reality problem” in metaphysics. It’s a chicken-and-egg thing: If we ask about the nature of reality, we have to have something in mind that the term points to. But if we do, then in some sense we already know, in which case it doesn’t make sense to ask. Socrates would be proud. (Or Plato, at least.) But for sure, if our inquiries always involve taking some perspective (which I like to call “taking a stance,” terminology I stole from my mentor, Doug Browning), and if taking up a stance always involves not capturing all of reality, then there’s no single stance from which we can capture reality with a term like “reality.” So, doesn’t this whole project collapse?

Not really. This is a subtle point, but what just happened in that discussion of perspectives and stances was that the idea of our concepts as a sort of “mirror” of what is independently real sneaks back again. And again, and again. It sneaks back because it’s a naive explanation for why our thoughts about the world are so good at helping us navigate in the world. Our thoughts guide us because they represent the world “as it is.”

Except that they don’t. We need another way to think about concepts that will allow us to address difficult philosophical problems — like the reality problem in metaphysics.

One approach to this problem is, of course, to say to hell with it and call metaphysics an unfortunate habit that we should break. I understand that response, but I’m unwilling to let exasperation at a hard problem dictate my philosophy. 😉

Suppose we approach this challenge a little differently. Suppose instead of assuming we know what the concept reality (or any other concept, for that matter) points to, we reframe the question: What must reality be like to allow for multiple perspectives that yield seemingly compelling ways of thinking? I nicknamed this “black-box metaphysics.” We don’t assume we know what’s in the box, which would be asserting a form of privileged access to reality, like that “mirror.” Rather, we ask how it’s possible — meaning, what are the conditions — for us to have the sorts of experiences we have and ask the sorts of questions we ask from the sorts of perspectives we take.

Kant’s approach, in my mind, appeals to a postulate of discontinuity: a “two realms” approach. The postulate of discontinuity has gotten a lot of airtime in philosophy: appearance vs reality, mind vs matter, noumena vs phenomena, etc. How about a little change of pace, metaphysically speaking?

Now you might be thinking, This is all fine and large, Matthew, but how does it help mortals live better lives? That’s an excellent question, and I have a response.

This may seem like an abstruse technical discussion for specialists, but it so happens that it applies to just about every aspect of our lives as rational animals. Want an example? How about last week’s post about the light switch. Was I determined in my behavior toward a light switch that I myself decided to install? Or was I determined to install it, too? Or am I deciding to follow my habits when I’m performing habits? Can a behavior slide from being chosen to being determined, if we train the animal according to our reasons?

Even insight into a relatively mundane mishap like putting your ice cream in the pantry instead of the freezer has the Problem of Free Will written all over it — if you know where to look.

Want to look more? Let’s talk.