# Can Free Will Escape Causation?

Speaker: Reed Winegar
Course: Kant’s Three Questions
Watch: https://ergo.org/videos/reed-winegar-can-free-will-escape-causation
Transcript: https://ergo.org/videos/reed-winegar-can-free-will-escape-causation/transcript/
YouTube: https://www.youtube.com/watch?v=LOhIdHvKU_Q

When quoting, name the speaker and link to the watch URL.

In this lecture, Reed Winegar walks through Kant's theory of knowledge, beginning with the fundamental distinctions between analytic and synthetic judgments and between a priori and empirical knowledge. He explains why Kant believed that synthetic a priori knowledge, such as mathematics and causation, is possible, drawing on the famous example of how we judge causal connections to be necessary, even though experience alone could never show them to be more than regular. Winegar then traces the limits of this framework: while knowledge applies reliably to objects of possible experience, it cannot reach beyond those bounds to things in themselves. This limitation, however, opens a surprising door. Through the Third Antinomy, Kant reveals a paradox in causal reasoning that leaves room for a special kind of uncaused freedom. Winegar concludes by exploring how this gap between appearances and things in themselves allows Kant to argue that humans might possess genuine moral autonomy, choosing between morality and desire in a way that escapes thoroughgoing causal determination.

## What Can We Actually Know?

One of Kant's most famous contributions to philosophy is his theory of knowledge. In particular, his claim that we're able to have knowledge for objects of possible experience, but that we're not able to have knowledge of alleged objects beyond the bounds of possible experience. The kinds of objects that Kant has foremost in mind that would be beyond the bounds of possible experience are things that he recognizes have, for many people, a strong existential kind of significance, things like freedom, the soul, and God.

What I'm going to do in this particular section is I'm going to start off by describing some of Kant's claims regarding the knowledge that we're able to have in relation to objects of possible experience. And then I'm going to turn to, in particular, the notion of causation and Kant's claim that we're able to have knowledge regarding cause and effect for objects of possible experience. And then the topic of causation is related for Kant to the notion of freedom and free will. And so then I'm going to turn to Kant's discussion of freedom, which is one of the things that he claims we're not able to have standard theoretical knowledge of.

And so using this topic of causation in order to look both at the type of knowledge that Kant says we can have for objects of possible experience, but then also what the limits are for that kind of knowledge.

## Analytic vs Synthetic Judgments Explained

Kant, as we've mentioned, has these distinctions between different types of judgments and then different types of knowledge. He talks on the one hand about analytic judgments. For an analytic judgment, what he says is that the predicate is contained in the subject, or the predicate is contained in the concept of the subject. A standard example for an analytic judgment, it's not his own example, but a standard example would be that a bachelor is unmarried. We don't think you need to do any fancy experiments or polls in order to figure out if this statement is true.

Rather, it's just part of the concept bachelor that it contains already the notion of being unmarried, where in order to be a bachelor, one has to be unmarried. And so that predicate, unmarried, Kant would say is already contained in the subject concept, and so this would be an analytic statement. A statement where the predicate is not contained in the subject, that gets called synthetic. So if I say, for example, this table is brown, it's not somehow definitionally true of this table that it's brown. It's not true of all tables as such that they're brown. But this particular table happens to be brown.

And so if we have this kind of statement, Kant would say that this predicate, brown, is not simply contained in the subject concept. So Kant refers to these kinds of statements where the predicate is contained in the subject concept as analytic and to these types of statements where the predicate is not contained in the subject concept as synthetic. These terms seem to be related to chemistry, where a kind of analysis is taking things apart, dissolving them, and then a synthesis is combining things together.

## A Priori vs Empirical Knowledge

Kant then also has this distinction that we've mentioned between two types of knowledge. A priori knowledge on the one hand, and then empirical knowledge, or his term is usually a posteriori, knowledge on the other hand. So we would take ourselves to know independently of experience, typically that two plus two equals four. So that would be a standard piece of knowledge that we take ourselves to have and that we take ourselves to know a priori independently of empirical observation or experience.

And then a posteriori or empirical knowledge, those would be bits of knowledge that we take ourselves to be justified in possessing on the basis of empirical observation. So if I say it is sunny outside today by looking outside the window and seeing the weather, then that's a piece of empirical knowledge, a posteriori.

## Synthetic A Priori: The Revolutionary Category

Kant thinks that a lot of previous philosophers tended to conflate the distinction between analytic and synthetic judgments and a priori knowledge and empirical knowledge, and that previous philosophers tended to think that analytic judgments were judgments that we knew a priori, and those were the only kinds of judgments we could know a priori. And then synthetic judgments were known on the basis of experience, and it was the case for any synthetic judgment that it needed to be known just on the basis of experience. But Kant thinks that in conflating these two distinctions in this way, philosophers have overlooked various options that are out there regarding types of judgments and the ways in which our claims to knowledge regarding certain types of judgments might be justified. And so what Kant says is that we should go ahead and tease apart these distinctions. And so we can represent that in the following way with the various boxes.

Kant certainly thinks that there are analytic judgments, the truth of which we know a priori. A bachelor is unmarried would be an example. He certainly thinks that there are synthetic judgments that we know a posteriori. That it's sunny outside would be an example. He thinks that if a judgment is analytic, then you're able, in principle at least, to know it's true or false a priori. And so he's not interested in empirical justification for the truth of analytic statements because you don't need empirical justification. You can just know those things independently of experience. So this is uninteresting to him.

And then the box that he's extremely interested in is this one. The box that would include synthetic judgments that we took ourselves to be justified in knowing were true a priori in a way independently of experience. And what Kant claims is that there's three traditional claimed bodies of knowledge that when we look at them carefully, turns out would go in this box. So the examples that he gives are the following.

One is math. Kant's going to claim that arithmetic and geometry involve synthetic a priori knowledge. A second one is what Kant calls pure natural science. What he has in mind here are claims like, every alteration results from a cause. Every alteration is the effect of some cause that it follows according to a general rule, according to some kind of law of nature. So that kind of very general claim that things are causally governed belongs to what Kant calls pure natural science, and he's going to say that that type of knowledge would be synthetic a priori knowledge. More fine grain types of things that we know in science, like how many species of fish there are or something like that. That's not a priori. That's not pure natural science. But it's these very general claims, like the claim about things being governed by cause and effect belong to pure natural science.

And then finally, the traditional notion of metaphysics. What I have in mind here is what Kant terms transcendent metaphysics, metaphysics that goes beyond the bounds of possible experience. So philosophers who would try to, for example, give arguments for the existence of a first cause or who try to give ontological arguments for the existence of God. Those are philosophers dealing in what Kant thinks of as transcendent metaphysics, because first causes and God are not the sorts of things that are able to be given to us in experience according to Kant.

And what Kant is going to argue is he's going to say that there's a certain condition that makes synthetic a priori knowledge in some cases possible. And it's present for math, and it's present for pure natural science. And so we're able to have synthetic a priori knowledge in these two domains, but it's absent in the case of transcendent metaphysics. And we're not able to have knowledge regarding the alleged objects of transcendent metaphysics.

So, again, I'm going to say a little bit more about math to fill that in, but then the main focus is going to be on pure natural science and in particular this notion of causation. And we'll look at Kant's claim that for all things in space and time, that they're governed by cause and effect. And then for transcendent metaphysics, we'll look at his claims regarding freedom, which is also related to causation, but where he says since it's beyond the bounds of what can be given in possible experience, we're not able to have knowledge of it. So what's this thing that makes synthetic a priori knowledge possible for math and pure natural science but not for transcendent metaphysics?

## Intuition, Concepts, and Cognitive Faculties

For Kant, it's what he calls intuition. And in particular, a certain type of intuition, a priori intuition. Let me get this off the board and then I'll explain that. Kant maintains that there's various cognitive faculties that we possess. And two of them are what he calls understanding and sensibility. Understanding is a faculty of concepts, which Kant takes to be general representations like the concept chair. That's general in the sense that it's not about some specific chair. It's the concept under which all chairs fall.

Kant thinks that sensibility is a faculty that provides us with intuitions. And for Kant, intuitions are not general representations. Rather, they're singular representations. And if I'm looking at this chair here in front of me, I'm seeing this. I'm in a cognitive relation of some kind with this individual. That for Kant is an intuition.

Importantly for Kant, there are a priori concepts and empirical concepts. And there are also a priori intuitions and empirical intuitions. The examples I just gave of the concept chair and then looking over here and seeing this, those are both empirical cases. The concept chair is an empirical concept. We see a lot of things around us. We abstract what's common between them in order to arrive at this notion of chair. Or you see this plant here. You see one of the trees outside. You see what's common between them. You leave out what's different and you arrive at an empirical concept of something like plant. Looking here, perception, having that intuition, that's empirical as well.

But Kant thinks that there's also a priori concepts. Indeed, he claims that in the *Critique of Pure Reason* that there's twelve particular a priori concepts that are able to provide us with synthetic a priori knowledge for objects of possible experience. And he calls these twelve concepts the categories. And one of these categories, we'll talk about in more detail in a bit, is the category of cause and effect.

And then for intuitions, Kant claims that we have some a priori intuitions. And in particular, he claims that we have a priori intuitions of space and a priori intuitions of time. For Kant, space and time are not taken to be features of things in themselves. Space and time also, we might say, are not just like containers that are out there that objects are in. Rather, for Kant, space and time are what he calls forms of intuition. So they, for Kant, are taken to be structures latent to our minds such that when we get sensation by being affected, then that sensation is processed by our minds in such a way that it is displayed to us spatially and temporally.

For Kant, space and time are forms of intuition. Any object of which we have experience will be given to us in one or both of these forms of intuition. And then Kant also says that we could have a priori intuition of space, of individual spatial regions or spatial objects and time. For example, in a priori intuition, I can construct a triangle. And that then is an a priori intuition of that spatial region, that triangular enclosed region. For Kant, what makes it the case that we're able to have synthetic a priori knowledge for things like math and things like pure natural science is the a priori status of things like the categories and of these intuitions of space and time.

## Why Math Is Synthetic A Priori

Let me talk about the math case first, and then we'll look at the pure natural science. If I draw these two points on the board and I ask you what's the straightest distance, or shortest distance rather, between these two points, the answer that you'll give me is that would be the straight line that you draw from one point to the next. Kant himself thinks that this claim that the shortest distance between two points is a straight line, he's just thinking about Euclidean space, he thinks that this is a synthetic judgment, but one that we take ourselves to know a priori.

Why does he think it's synthetic? He thinks it's synthetic because the notions of shortest and straight are too different from one another, where we can have one count as being included in the other. If I say that the shortest distance is a straight line, then Kant says, all right, so here's our general subject, shortest distance, and then here's what we're saying about it, that it's a straight line. Is it somehow contained in the concept of the shortest distance that it's going to be a straight line the way being unmarried is contained in the concept bachelor? And Kant says no.

Kant says that the notion of shortest is a kind of quantitative notion. It's a question about the measure of the distance. Whereas the notion of straight here is a qualitative notion. The type of line is a straight one rather than a curved one, but that's not something that's being given a kind of numerical value or being expressed as a kind of measurement. And so Kant thinks that this is a synthetic statement because this qualitative content is not contained in this more quantitative subject concept.

Now, there's been a lot of developments in mathematics since Kant's time. And I'm not going to pause here to evaluate the truth of Kant's suggestion for this particular example. But this is the kind of example that he provides to suggest that there's significant mathematical statements that are synthetic. But he thinks that we all take ourselves to know that this statement is true, certainly in Euclidean space. And so Kant says that this counts then as a synthetic statement that we take ourselves to know a priori. And so it's a piece of synthetic a priori knowledge that we take ourselves to have.

What is it that allows us to have this synthetic a priori knowledge? Well, it's not a matter of mere conceptual relations. If it was a matter of mere conceptual relations, then this would be an analytic statement, but we said it's not. And so Kant says it must be something else. It must be an intuitive element, and that intuitive element must be an intuitive element that's a priori for the statement to be one that we take ourselves to know a priori.

And in particular, Kant says it's the a priori intuition of space. We have this a priori intuition of space, in which we can construct different figures and as such are able to arrive at the knowledge that the shortest distance between two points is going to be a straight line. So Kant thinks that we're able to have this kind of mathematical knowledge a priori, independently of experience, even though it's synthetic, because he says space is something of which we have this a priori intuition.

## Causation as A Priori Knowledge

The other major domain in which Kant says we can have synthetic a priori knowledge, like I mentioned before, is the domain of pure natural science. And the chief example here to focus on is the notion of causation or cause and effect. First, it's important to recognize that Kant thinks that the notion of cause and effect is a priori. He doesn't think that it's something that we get from experience. Following philosophers like Hume, he's happy to note that there's various kinds of regularities that we observe and experience.

Like each time I eat a sandwich, it satisfies my hunger. So we have this constant regularity. But the observation of those regularities is not sufficient to qualify as observation of cause and effect connections. Because it's part of the notion of one thing being the cause of an effect that there's a necessary connection between the two. Given one, the other one had to result. Must result of necessity, must come to be.

And so we can see that on a bunch of occasions. Whenever I ate a sandwich, it satisfied my hunger. But we can't somehow see the necessity that eating a sandwich is going to satisfy hunger. You can stare at a sandwich as long as you want, and you're not able to somehow see that sandwich will necessarily satisfy hunger. And so Kant, along with some people like Hume, says that if we have a notion of cause and effect, then it's going to be one that we possess a priori.

More specifically, what Kant suggests is that we can look to the different rules of logic, and in logic, we can note this logical relationship between ground and consequence, and then we can use that notion of ground and consequence to arrive at this a priori concept of cause and effect. So we have this concept a priori. Well, just because we have the concept doesn't mean the concept is one that is legitimate for us to use. It could be the case that our mind comes up with various concepts, but then there's nothing in the world that those concepts actually match or properly track and that maybe we have these concepts, maybe we can't even help but use them, but nevertheless count as being caught up in a kind of ongoing intellectual daydream or intellectual hallucination of a certain kind. Because although we have these concepts and are thinking in terms of them, it's just not the case that we can rightfully claim that they correspond to or track anything in the world. And so what Kant wants to do is he wants to overcome this type of worry.

## The Boat Argument for Causation

And one place where he gives an argument that attempts to overcome this type of worry is in a section of the *Critique of Pure Reason* known as the second analogy. Kant's basic argument in the second analogy, and I'll present it in a little more detail in a second, but the basic argument is that in order for us to be able to make objective claims, like intersubjectively valid claims, about what happened earlier in time and what happened subsequently in time, we have to appeal to the notion of cause and effect. We obviously take ourselves to have knowledge of what happened earlier in time, what happened subsequently in time. We take it to be the case that there is some, even in cases where we happen to not know what was first and what was second, because maybe we weren't there to observe it. We nevertheless take it to be the case that there is some objective fact of the matter about what counts as having happened first and what counts as having happened second.

And so Kant's argument in the second analogy is going to be that causation, the notion of cause and effect, plays a key role in our ability to make these types of claims about it being objectively the matter that one thing was first and another thing was second. And, as such, we're licensed to appeal to the notion of cause and effect, because it's a condition for this knowledge we all take ourselves to have about one thing that was first and another thing that was second.

So the example that Kant likes to give is the example of a boat floating down a river. And so we're going to say that the boat was initially upstream, and then the boat after that was downstream. Well, what is it that allows us to make this claim that the boat was initially upstream and then downstream? So imagine that you were at the regatta yesterday. You saw the boat upstream. You saw it downstream. So you have memories of these two different events.

But now what you have in your memory are just these two representations, the boat in one location and the boat in another location. And what is it that allows you to give a particular order to these two representations to say that one of them is representing the earlier event and the other one is representing the later event? The two representations, Kant maintains, don't seem to come with something like timestamps, where they are somehow labeled first and second in your mind. And so it doesn't seem like simply by looking at these representations in your mind's eye and comparing them that you're able to designate which of them counts as the first event and which of them counts as the subsequent event.

And so Kant says that in order for it to count as objectively true or intersubjectively valid that one of these representations represents the earlier event and the other one represents this later event, we need to say that there's something that makes one of these events necessarily first and the other one necessarily second. And what that will be will be the notion of causal laws. That given the causal laws, given the laws of nature, it had to be the case that the boat was first upstream before it was downstream.

So what is it that allows us to say that it should count as an objective fact of the matter that one of these representations is representing the earlier event and the other one is representing the later subsequent event? And Kant's claim is that what licenses this is an appeal to the notion of cause and effect. That given the laws of nature, it had to be the case that one of these events was earlier and the other event was later, because given the laws of nature, that's the only order in which these two events could have occurred. Somehow flipping the order of the events would count as a violation of the laws of nature.

So in the example, Kant's thinking that, to say that we're upstream and then downstream is to say that the water is flowing in this direction from here to here. And, given the laws of physics, assuming there's no wind blowing the sails and no rudders and the boat's just floating, then it has to be the case that the boat was upstream before it was downstream because the boat is not able to simply float upstream against the current. That would be violating the laws at play here. Rather, it has to be the case that the boat was first upstream and then after that downstream because of the laws of nature and the different cause and effect relationships that those laws of nature govern.

So this is one of Kant's arguments in favor of the notion that we should take ourselves to count as having synthetic a priori knowledge of cause and effect. Given that we take ourselves to have knowledge that there is a kind of objective temporal order of events, where one event counts as first and the other counts as second, and given that the only way that there could be that objective temporal order would be on the basis of there being cause and effect relations such that one of these events as the cause had to have been the first one and the other event as the effect had to be the second one, that that then legitimates or licenses us in saying that our a priori concept of cause and effect shouldn't count as something that we happen to have but doesn't have any correspondence to the empirical world.

It shouldn't count as something that we have to think in terms of, but where that's just kind of like an intellectual daydream. Rather, it's a concept that is shown to have purchase on our experience, that anything given to us in space and time is going to be subject to cause and effect relations, because anything given to us in space and time needs to have some objective place on the timeline for when it happened, and it's the cause and effect relations that determine that timeline such that we can say it's objectively true that this event was at this place on the timeline. Kant thinks that this argument, this type of argument, gives us reason to claim that we have synthetic a priori knowledge of objects given to us in experience.

## Where Knowledge Hits Its Limits

But we all know that Kant, even though he says we're able to have some synthetic a priori knowledge, also wants to prohibit a lot of knowledge that earlier philosophers and subsequent philosophers took themselves to have. In particular, he wants to claim that we cannot have knowledge of transcendent metaphysics. And one of the chief notions that he has in mind here is the notion of free will.

We can note that there's an intimate connection between the notion of freedom and causation, because if I'm acting freely, we take me, I take myself to be acting in a particular way where I'm the cause of some kind of effect. And there's, of course, a question of is there something that itself is causing me to act that way rather than some other way? And so causation and freedom are intimately related notions.

In the latter part of the *Critique of Pure Reason* and the transcendental dialectic section, Kant presents an antinomy that he calls the third antinomy. So a kind of contradiction that reason gets itself tangled up into, and in particular, in this case, a contradiction revolving around the notion of causation.

## The Third Antinomy: A Causal Paradox

What Kant says is that our faculty of reason demands the unconditioned in such a way that it gives rise to the idea of the notion of the world as a whole. And in relation to causation, the notion of the world as a whole should be understood in terms of a kind of total causal series leading up to the present moment in time. And Kant notes that that total causal series seems like it could have two forms.

One, it could be a finite series. So this caused this, this causes this, this caused this, but this itself then doesn't have a cause. Rather, this causally speaking is unconditioned. It's not conditioned by some prior cause. In respect to causation, it doesn't have a prior, a further condition. So it counts as an unconditioned cause. Or the series, it seems, could take the form of a causal series in which there is no individual member that is itself unconditioned, but rather that you would have an infinite series of causes and effects.

Kant maintains that there are problems with both of these options that get us tangled up into a kind of paradoxical situation of claiming both that this has to be true because this is false, and this has to be true because this is false. And in order to resolve that paradoxical conflict, Kant appeals to his theory of transcendental idealism, to his basic claim that space and time are forms of intuition.

## Appearances vs Things in Themselves

Objects of experience are given in space and time, but space and time are not features of things in themselves. We have this distinction then between appearances, objects as they appear to us in space and time, and then things in themselves, which underlie in some way those appearances. And Kant says that we can get out of our paradoxical conflict by saying that actually both these scenarios might very well be true, but one of them might be true at the level of objects of experience, and the other one might be true at the level of things in themselves.

So here we have the friendly world of experience. In the friendly world of experience, Kant says that it will be the case that any alteration in time and experience, it's going to be in time, any alteration in time, will be the effect of a cause that it follows in accordance with a necessary rule. So there are going to be laws of nature that govern everything in time, and every alteration in time is going to be the effect of a cause, such that there's nothing in the temporal realm that is somehow independent of having any alteration here that's somehow independent of having a cause as its condition.

But we also have this much more mysterious realm of things in themselves, which in some sense Kant takes to ground the appearances. There's a lot of debate over how exactly to interpret the notion of transcendental idealism, the distinction between appearances and things in themselves. And so I'm not going to get further into those weeds. But a traditional interpretation would hold that there's the appearances, and then the things in themselves that are the grounds in some sense of those appearances.

And Kant says that we're not able to have substantive knowledge of things in themselves. Some people want to interpret Kant as claiming we can have no knowledge of things in themselves at all. That sometimes seems to run into some problems because there's various places where it seems like Kant makes some limited knowledge claims about things in themselves, like maybe the limited claim that we know they exist, or maybe the limited claim that there's no contradictions at the level of things in themselves. He oftentimes seems to maintain that the principle of non-contradiction holds for things in themselves.

But at the very least, it seems like he doesn't want us to, he wants to ban any kind of really substantive or significant knowledge of things in themselves. So we don't really know what they're like in any detail. Maybe we have a minimal knowledge that they exist. They're not contradictory. That doesn't really tell us anything substantive about what they're like in any full way. Maybe we can have some negative knowledge, like we know they're not in space, we know they're not in time. But again, that doesn't tell us what they're like in a positive sense. It just tells us what they're not like.

So in any case, whether one thinks that Kant is literally banning all knowledge claims about things in themselves or if he's just banning a set of significant positive claims about things in themselves, either way there's strong prohibitions on substantive knowledge about things in themselves. And so we can't know whether or not it's the case at the level of things in themselves that things count as in some way causally conditioned by others, such that everything in itself would count as causally conditioned by others. But we also can't rule it out. The fact that we lack knowledge here entails that it might very well be the case for all we know that there is, at the level of things in themselves, a cause that counts as an uncaused cause, a cause that counts as an unconditioned cause.

Indeed, the item that allowed us to claim that everything in space and time is subject to prior causes was specifically this need to order everything in an objective timeline. But at the level of things in themselves, we're saying there is no time. There is not that alteration. There is no objective timeline of first and second. And so the argument that led to us claiming that every alteration in time is going to be the result of a cause doesn't apply at the level of things in themselves because it depends on time, and time's absent here at the level of things in themselves. So it might, again, we can't claim to know that the following is the case, but it might be the case that at the level of things in themselves, there is something that's causally unconditioned, maybe even something that's causally unconditioned but itself a cause of other things.

## Could Humans Have Uncaused Freedom?

Kant initially presents this notion in a very cosmological context, thinking about the notion in traditional metaphysics of a first cause. Could there be some kind of first cause that itself is not the result of some prior cause? And Kant is saying at the level of things in themselves, for all we know, maybe yeah. And so if we wanted to put a traditional label on that first cause, then it could, for example, for all we know, be the case that God exists. God wasn't caused to exist or brought into existence, but God nevertheless creates the world, counts as the cause of all the finite things, and is this unconditioned cause.

But Kant doesn't want to limit this picture simply to a creator as the first cause of the world. He rather wants to extend it to human actions and to the human will. And he says something like, "Look, if it's the case that there could be at least one unconditioned cause, maybe there could be others." Maybe it could be the case that we as human beings who exist here at the level of appearances, but also here at the level of things in themselves, where this is just the appearance of me as a thing in itself, maybe it could be the case that at the level of things in themselves, we count ourselves, our actions, as uncaused causes, that we can have a kind of causality that it's not the case that some prior cause brings it into effect. Rather it counts as an uncaused cause of actions. So the picture that Kant develops is the following.

## Noumenal and Phenomenal Character

So here I am as an appearance, and then here I am as a thing in itself. At the level of a thing in itself, it can be the case that I have what Kant is going to call a noumenal character. Noumenal here, don't get too hung up on it. But oftentimes when Kant is talking about us thinking about things in themselves, he describes the things in themselves in terms of noumenon, like objects of the intellect. So the objects of thought in this case. So we can think of ourselves as having some kind of character as things in themselves.

And then that character is the ground of what at the level of experience, of appearances, gets called the phenomenal character. And the term phenomenal here is just referring to the notion of appearance, of sensibility, that these are things that are part of the world of appearances, part of the world of experience. Kant is a little ambiguous or vague regarding precisely what he has in mind by the phenomenal character. But one maybe natural way to interpret it would be as a kind of basic psychological profile that one has. So the kind of empirical psychology, personality that one has as a being in the world of space and time.

And as a result of my phenomenal character, whenever I'm in different situations, I will necessarily act in particular sorts of ways. I'll be governed by my phenomenal character. So maybe due to my phenomenal character, I have really intense liking for ice cream. So whenever the ice cream is presented to me, I'm going to eat it. At the level of appearances then, my actions can count as being determined in accordance with these kinds of laws and rules, governing my behavior.

And indeed, Kant in some dramatic passages says that at the level of appearances, we don't count as having freedom at all. Sometimes he puts it provocatively in that sort of way. But my noumenal character could be something that is uncaused. Nothing made me, forced me to have this noumenal character. And my phenomenal character can be seen as the expression of my noumenal character.

So all of my actions in space and time that are resulting in my interactions with the world on the basis of cause and effect and are thoroughly governed by cause and effect relations. Indeed, sometimes Kant says, if we had enough knowledge, we could predict your behavior with the same exactness as we could predict the motions of the stars. All of that might nevertheless, in an extended sense, be seen as a product of freedom because it can all be seen as an expression of this noumenal character, which is something that very well might have the status of an uncaused cause or a causality that's not being brought into effect by some other cause. Whereas I could count as having free will at this level, and at this level, everything's causally determined. But at this level, I could still count as having free will in an extended sense because all of this stuff is grounded on this, which is free.

## Timeless Choice: Morality or Desire?

Now, there's two ways that Kant oftentimes talks about the relationship between these two levels. And the exact relationship between these two ways of talking is not obvious, and might very well be something that ultimately is a source of difficulty for Kant. But I'm going to go ahead and present both of them and then leave it as a question of how, if at all, they can really be brought adequately fully together.

Sometimes what Kant seems to present us with is the notion that as a timeless thing in itself, I make some kind of basic choice about the kind of being that I am. And that basic choice then comes to structure my entire life, including structuring my personality in certain ways, and then structuring the ways that I behave in my everyday interactions in the world. So for instance, in part one of his book, *Religion within the Boundaries of Mere Reason*, Kant presents the following picture. He says, "Look, on the one hand, we are creatures who possess sensibility and who are able to act on the basis of sensible desires and inclinations and passions. On the other hand, we're also, in a sense, moral beings who are subject to the moral law, and can either act in accordance with the moral law, or can act out of step with the moral law."

And Kant suggests that the human being makes a timeless choice about whether to prioritize sensibility, like sensible desires as motives, over the moral law, or whether to prioritize the moral law over sensible desires. He thinks that if somebody is prioritizing sensible desires over the moral law, then what they're doing is they're incorporating some deviations from the moral law into their most fundamental maxim, into the maxim that's the result of this basic choice. Whereas if they prioritize the moral law over sensibility, then no matter what temptations or threats we're presented with to our sensible desires or inclinations, we would always nevertheless aim to do what's right, not to incorporate deviations from the moral law into our fundamental maxim.

And Kant presents this as a timeless choice because he wants it to be a choice for which we count as responsible, and he takes it to be the case that for us to count as responsible for the choice, the choice can't be something that's causally determined. It can't be something that we were causally forced into or compelled into. Rather, it needs to be a free choice. And so for Kant, since everything in time is the result of these causal interactions, if that choice were taken to be a choice in time, it couldn't count as a choice for which we bore this kind of ultimate responsibility. And so Kant says, it should be a timeless choice, so it can be a free choice in the proper sort of sense where we can count as having responsibility for that choice. So this is one way that Kant sometimes talks about the notion of the freedom that we might have outside of time.

However, this way of talking gives rise to certain puzzles. So for instance, it seems like Kant's presenting a situation where we haven't made a choice, but then we made a choice one way or the other. And then he'll go on to talk about how we can have a revolution of the heart where we change our choice. And all of this talk of there being a kind of initial state and then a choice and then maybe a new choice of some kind is introducing notions of alteration, which Kant has specifically said are not part of the realm of things in themselves.

Remember, his claim in the third antinomy was that the reason why there can be an unconditioned cause at the level of things in themselves, at least maybe we can't rule it out, is because things in themselves are not in time, are not subject to alteration. And so that argument from the second analogy doesn't apply to them. But now it looks like maybe he's introducing some kind of alteration at the level of things in themselves in a way that it's not clear that he's justified in doing. Maybe there's a way around that, but the solution to it at least is not transparently obvious.

The other way that Kant sometimes talks about the relationship between the thing in itself and the phenomenal human being is less in terms of some kind of choice, but rather in terms of autonomy, in terms of a notion of self-legislation, and in particular, the self-legislation of the moral law.

## Moral Autonomy and Self-Legislation

What Kant will do is, he'll claim that we, as noumenal beings, legislate the moral law, that we, as noumenal beings, provide the moral law to ourselves, place the moral law upon ourselves. And Kant's picture then of the human being is a picture of the human being as morally autonomous, morally self-ruling. The human being has a sensible side of inclinations and desires, but also has a rational side that includes practical reason. And practical reason legislates the moral law, and in doing so, binds us, including our sensible side, to certain moral obligations and responsibilities.

For Kant, the moral law is not something imposed on us by society, or imposed on us by God, but rather a law that we as rational beings impose upon ourselves. And so sometimes when Kant seems to be describing the relationship between these two levels, what he seems to have mostly in mind is not some type of free choice, but rather this free legislation of the moral law that we give to ourselves, and the notion that the moral law can have a kind of efficacy on our actions, such that we count as most fully free when we're not simply being pushed around by whatever sensible inclinations or desires we happen to have or happen to have been born with, but rather where we're being determined by a law that we've given to ourselves, namely the moral law, because that moral law is having some kind of efficacy on our actions.

The way he spells out the way in which the moral law has an effect on our actions is in talking about the feeling of respect for the moral law. And he claims that the moral law infringes on our self-love, that it strikes down our self-conceit, and that in doing so, there's a kind of determination of our will from the moral law. And so another way that one might see the relationship between the noumenal character and phenomenal character would be less in terms of I chose to be a certain way, and then that gave me a certain personality here, but rather in terms of I legislate this moral law that's able to determine my sensible will. And that shows up at the phenomenal level in terms of moral feeling and the feeling of respect, which is then part of my everyday moral psychology.

So we have these two ways in which Kant talks about the notion of the free will, and the relationship between the noumenal character and the phenomenal character. And how exactly these two are supposed to line up, at the end of the day is not entirely clear. The most that Kant gives us to work with is the following. The most he gives us to work with is to say that we have a will, and there's two different aspects of that will. One aspect of that will is the legislative aspect that legislates the moral law and binds us to the moral law. The other aspect of that will is some kind of aspect of choice which chooses whether to subordinate the moral law or not.

But both these aspects of the will would belong to the will at the level of things in themselves. And how exactly this element of choice really could be present at the level of things in themselves, given that it might seem caught up with notions of alteration, is something that unfortunately Kant doesn't address nearly as head-on as we might like him to.

## Wrapping Up: Knowledge and Freedom

In any case, Kant certainly provides us here with what is a really compelling and influential picture of the human being as a free being, in particular in his conception of the human being as an autonomous being, as a being that gives itself a particular law and that counts as self-determining because it gives itself this law, that then is able to serve as a determiner of its different actions, and where for Kant that law is in particular the moral law that we all legislate to ourselves as free, rational beings.

So, where does that leave us? Kant on the one hand has claimed that we're able to have significant synthetic a priori knowledge in certain areas related to space and time, so in particular math, and then what he's called pure natural science, and in pure natural science, these claims about knowledge of causation. But he's also limited the extent to which causation is able to provide us with knowledge. He's claimed that we're not able to appeal to the notion of cause to have knowledge of things in themselves.

But that need not be seen as a glass half empty. It can also be seen as a kind of glass half full, because in not having that knowledge of things in themselves, it leaves open room for the possibility that we might ourselves have some kind of unconditioned causality, that we might ourselves have this kind of special freedom, that we might ourselves be these kinds of morally autonomous beings, that we might somehow be, in some way exempt from something like a full, thoroughgoing causal nexus of determination, and be able to retain an element of human freedom.