Immanuel Kant
As I mentioned in the previous lesson, Hume's problem caused a great deal of consternation. Kant himself tells us that his "recollection of David Hume was the very thing which many years ago first interrupted my dogmatic slumber." It was his principal objective to solve the problem of skepticism. Some of our philosophical speculations may seem very strange but this was a serious problem that needed to be addressed and solved if possible. Kant did solve, in large measure, this problem, and in doing so, ushered in a new era of philosophy that gave greater emphasis to the mind and its relationship to the world. Among other things, Kant was directly responsible for the rise of perceptual psychology as a scientific discipline.
Before we get to Kant's solution we need to introduce some new terminology. For Kant, there are two kinds of propositions: analytic and synthetic. All propositions contain two parts: the subject and predicate terms. In analytic propositions, the predicate term merely restates the subject term and adds no new information to the subject term. For example, "All triangles are three-sided objects." The predicate term here is "three-sided objects" but that doesn't tell us anything new about "triangles"; it only restates it differently. Synthetic propositions, on the other hand, contain predicate terms that do add new information to the subject term. For example, "That car is silver." The predicate term here is "silver" which tells us something new about the subject term "car," namely its color.
We gain knowledge of propositions in two ways. We can determine the truth of the analytic proposition "All triangles are three-sided objects" without actually looking at all triangles. We don't have to look at any triangles at all to judge the truth of it. We understand the concept and that tells us that the proposition is true. In other words, we gain knowledge of it before sense experience so we say it is known as a priori. On the other hand, most synthetic propositions, such as "That car is silver," require sense experience to tell whether or not they are true. So they are known as a posteriori.
So, analytic propositions are a priori and synthetic propositions are a posteriori. We can now restate Hume's problem in Kantian terms. According to Hume, we could not have certainty concerning matters of fact. In Kantian terms, we would say you cannot have synthetic a priori knowledge. But is this true? To tell, we first need to think of an example of a synthetic a priori proposition and then ask whether we know of it or not. Kant gives us two examples; one from the realm of mathematics and one from the realm of metaphysics. The mathematical example he gives is 7+5=12. There is some debate as to whether this is a synthetic proposition and since we are not concerned with mathematical insights we'll pass over it to consider the metaphysical example. This, after all, is where the significant contribution of Kant lies.
The example he offers us is "Every event must have a cause." This is a synthetic proposition since the predicate term "cause" adds new information to the subject term "event." So would you say that the statement is true or false? If you're like most people you'd say it’s true. In other words, we know this. Interestingly enough our knowledge of this has to be a priori since we have not experienced every event. So we do have synthetic a priori knowledge. The real question is not "Is synthetic a priori knowledge possible?" The real question is "How is synthetic a priori knowledge possible?"
How could Hume have missed the possibility of synthetic a priori knowledge? You remember that Hume began with the empiricist assumption that all knowledge comes from sense experience. Another way of putting this is to say that our knowledge must conform to objects of sense experience. So concerning causality Hume recognizes, quite correctly, that we have no sensory experience of causality. However, he concludes from this, quite incorrectly, that we do not know causality. The mistake does not lie in his belief that causality cannot be found in sense experience. His mistake lies in his belief that sense experience is all we have.
So Hume was half right. To illustrate this Kant points out in the preface to his monumental Critique of Pure Reason, "that all our knowledge begins with experience there can be no doubt." The real story, however, is told later, when he adds to this by saying "but although all our knowledge begins with experience, it does not follow that it arises from experience." Thus, we are on our way to the solution.
To see exactly how Kant gets there let's consider another problem and solution that impressed Kant very much. If you remember from astronomy the story of Copernicus this will help you understand Kant. The problem, at the time, was how to explain the movements of the planets around the Earth. This turned out to be very difficult to do and so Copernicus thought to try something revolutionary. What if I assume that the Sun is at the center instead of the Earth? Making this change made all the difference and allowed Copernicus to explain planetary motion much easier. This became known as the Copernican Revolution. Kant wanted to do the same in philosophy. He wanted to have his own Copernican Revolution. So, instead of assuming, as Hume did, that our knowledge must conform to objects of sense experience, Kant decided to see what might happen if we assumed that objects of sense experience had to conform to our knowledge.
What happens is that Hume's problem disappears. See Hume proceeded on the basis that sense experience was where the knowledge came from and so since causality is not in sense experience we do not know of it. Kant agreed that causality wasn't in sense experience. Causality is one of the conditions for having sense experience. Think of it like this. If you were wearing a pair of rose-colored glasses everything in your world would appear rose-colored. Now imagine you were born with those glasses and can never take them off. In some sense, the glasses are not part of your sense experience though they do determine what your sense experience will be like. For Kant, causality is like that pair of glasses. It’s a filter, so to speak, through which we view sense experience. There are more of these filters as well; things like space, time, and substance. Kant called them the categories of the mind and they constituted the essence of his "critical philosophy."
This is truly radical thinking. While, other philosophers and scientists were trying to understand the nature of space, time, causality, and substance in objective terms Kant was saying that they were part of the mind's ability to process experience. So, while Newton had said, in Principia Mathematica, that space and time were absolute frames of reference, Kant was saying that they were subjective conditions governing our perception of reality. Is reality composed of three space dimensions and a one-time dimension? Who knows! For Kant, what we do know is that we can only perceive reality in four dimensions. Who’s to say that reality isn't composed on 10 or 11 dimensions? Ironic that this is exactly what some contemporary physicists are saying who defend string theory.
Before we get to Kant's solution we need to introduce some new terminology. For Kant, there are two kinds of propositions: analytic and synthetic. All propositions contain two parts: the subject and predicate terms. In analytic propositions, the predicate term merely restates the subject term and adds no new information to the subject term. For example, "All triangles are three-sided objects." The predicate term here is "three-sided objects" but that doesn't tell us anything new about "triangles"; it only restates it differently. Synthetic propositions, on the other hand, contain predicate terms that do add new information to the subject term. For example, "That car is silver." The predicate term here is "silver" which tells us something new about the subject term "car," namely its color.
We gain knowledge of propositions in two ways. We can determine the truth of the analytic proposition "All triangles are three-sided objects" without actually looking at all triangles. We don't have to look at any triangles at all to judge the truth of it. We understand the concept and that tells us that the proposition is true. In other words, we gain knowledge of it before sense experience so we say it is known as a priori. On the other hand, most synthetic propositions, such as "That car is silver," require sense experience to tell whether or not they are true. So they are known as a posteriori.
So, analytic propositions are a priori and synthetic propositions are a posteriori. We can now restate Hume's problem in Kantian terms. According to Hume, we could not have certainty concerning matters of fact. In Kantian terms, we would say you cannot have synthetic a priori knowledge. But is this true? To tell, we first need to think of an example of a synthetic a priori proposition and then ask whether we know of it or not. Kant gives us two examples; one from the realm of mathematics and one from the realm of metaphysics. The mathematical example he gives is 7+5=12. There is some debate as to whether this is a synthetic proposition and since we are not concerned with mathematical insights we'll pass over it to consider the metaphysical example. This, after all, is where the significant contribution of Kant lies.
The example he offers us is "Every event must have a cause." This is a synthetic proposition since the predicate term "cause" adds new information to the subject term "event." So would you say that the statement is true or false? If you're like most people you'd say it’s true. In other words, we know this. Interestingly enough our knowledge of this has to be a priori since we have not experienced every event. So we do have synthetic a priori knowledge. The real question is not "Is synthetic a priori knowledge possible?" The real question is "How is synthetic a priori knowledge possible?"
How could Hume have missed the possibility of synthetic a priori knowledge? You remember that Hume began with the empiricist assumption that all knowledge comes from sense experience. Another way of putting this is to say that our knowledge must conform to objects of sense experience. So concerning causality Hume recognizes, quite correctly, that we have no sensory experience of causality. However, he concludes from this, quite incorrectly, that we do not know causality. The mistake does not lie in his belief that causality cannot be found in sense experience. His mistake lies in his belief that sense experience is all we have.
So Hume was half right. To illustrate this Kant points out in the preface to his monumental Critique of Pure Reason, "that all our knowledge begins with experience there can be no doubt." The real story, however, is told later, when he adds to this by saying "but although all our knowledge begins with experience, it does not follow that it arises from experience." Thus, we are on our way to the solution.
To see exactly how Kant gets there let's consider another problem and solution that impressed Kant very much. If you remember from astronomy the story of Copernicus this will help you understand Kant. The problem, at the time, was how to explain the movements of the planets around the Earth. This turned out to be very difficult to do and so Copernicus thought to try something revolutionary. What if I assume that the Sun is at the center instead of the Earth? Making this change made all the difference and allowed Copernicus to explain planetary motion much easier. This became known as the Copernican Revolution. Kant wanted to do the same in philosophy. He wanted to have his own Copernican Revolution. So, instead of assuming, as Hume did, that our knowledge must conform to objects of sense experience, Kant decided to see what might happen if we assumed that objects of sense experience had to conform to our knowledge.
What happens is that Hume's problem disappears. See Hume proceeded on the basis that sense experience was where the knowledge came from and so since causality is not in sense experience we do not know of it. Kant agreed that causality wasn't in sense experience. Causality is one of the conditions for having sense experience. Think of it like this. If you were wearing a pair of rose-colored glasses everything in your world would appear rose-colored. Now imagine you were born with those glasses and can never take them off. In some sense, the glasses are not part of your sense experience though they do determine what your sense experience will be like. For Kant, causality is like that pair of glasses. It’s a filter, so to speak, through which we view sense experience. There are more of these filters as well; things like space, time, and substance. Kant called them the categories of the mind and they constituted the essence of his "critical philosophy."
This is truly radical thinking. While, other philosophers and scientists were trying to understand the nature of space, time, causality, and substance in objective terms Kant was saying that they were part of the mind's ability to process experience. So, while Newton had said, in Principia Mathematica, that space and time were absolute frames of reference, Kant was saying that they were subjective conditions governing our perception of reality. Is reality composed of three space dimensions and a one-time dimension? Who knows! For Kant, what we do know is that we can only perceive reality in four dimensions. Who’s to say that reality isn't composed on 10 or 11 dimensions? Ironic that this is exactly what some contemporary physicists are saying who defend string theory.
The Kanizsa square is a good example of what Kant was trying to say. When you look at the picture you see a white square, right? There’s only one thing. There is no white square there! Your mind filled in the square. That’s an example of how the mind organizes sense experience.
What Kant was trying to do was find a middle ground between rationalism and empiricism. Both entail serious problems as we've seen. Rationalism forces us to conclude that some (if not all) ideas are innate while empiricism leads us to skepticism about such fundamental aspects of our knowledge as causality. Both are unacceptable though they do have useful elements. For Kant, the optimal solution is to take the best aspects of each epistemology and combine them. This is what he called critical philosophy. So, Kant denies innate ideas and says knowledge comes from sense experience. But, he does believe that the mind contains innate structures which govern how sense experience is processed (the categories). Of course, no good theory is without problems and Kant's is no different.
By way of considering these problems let's address a few other points about Kant's critical philosophy. One of the implications of the categories of the mind is that we do not perceive objects independently of them. All our sense experience is filtered through these categories so there is a difference between the objects of our sense experience and the objects independent of our sense experience of them. Kant terms these objects of sense experience "phenomena" and the objects as they are independent of our sense experience "noumena." There may be nothing too startling here, to begin with, but some of the implications of this may be problematic. That's not to say they aren't true of course but they may cause us problems.
One of Kant's concerns was to show the limits of human reason. There are some things that we cannot know of. For example, almost by definition, we cannot have a sense of the experience of objects as they are when they are not being perceived. These "things-in-themselves" can never be observed and so we can never know of them. He refers to objects like this, in somewhat Platonic terms, as transcendental. Unfortunately, there are other transcendental elements.
If objects that exist beyond our sense experience are beyond our knowledge this would also seem to include God as well. Certainly, God is transcendental so we are forever precluded from knowing him through pure reason alone. Fortunately, Kant postulates that there are other types of reason through which we can gain knowledge of God. It turns out that God can be known through practical reason which we need as a basis for our understanding of morality. We will be discussing Kant's approach to morality in a later chapter.
There is a further transcendental element that we need to consider. This is the one that gave Hume so many problems. While the categories of mind can be understood as the elements of our sense experience, who is it that's having this experience. In other words, what is the nature of the self? Well, it’s not an object of sense experience since the self is what has sense experience. We can infer its existence though through the fact that our sense experience is unified. Something must be doing the unifying and this something is the self; which Kant terms the transcendental unity of perception.
So where does this leave us? We have partially solved Hume's problem of skepticism. We can have certainty about causality since it is a condition for our having sense experience. We can still maintain that knowledge comes from sense experience without buying into the denial of such fundamental knowledge as we have of substance, space, time, and causality. However, we do have to recognize that applying these categories to the transcendental realm of noumena will never be possible. While we gain a better understanding of the workings of the mind through the innate structures it contains, we don't have to postulate that some ideas are innate and fully formed at birth in the mind. So while we have defined very definite limits to our knowledge we have advanced far beyond the limits of rationalism and empiricism. Have we advanced as far as we can? To answer that we'll need to address what happened in philosophy after Kant.
What Kant was trying to do was find a middle ground between rationalism and empiricism. Both entail serious problems as we've seen. Rationalism forces us to conclude that some (if not all) ideas are innate while empiricism leads us to skepticism about such fundamental aspects of our knowledge as causality. Both are unacceptable though they do have useful elements. For Kant, the optimal solution is to take the best aspects of each epistemology and combine them. This is what he called critical philosophy. So, Kant denies innate ideas and says knowledge comes from sense experience. But, he does believe that the mind contains innate structures which govern how sense experience is processed (the categories). Of course, no good theory is without problems and Kant's is no different.
By way of considering these problems let's address a few other points about Kant's critical philosophy. One of the implications of the categories of the mind is that we do not perceive objects independently of them. All our sense experience is filtered through these categories so there is a difference between the objects of our sense experience and the objects independent of our sense experience of them. Kant terms these objects of sense experience "phenomena" and the objects as they are independent of our sense experience "noumena." There may be nothing too startling here, to begin with, but some of the implications of this may be problematic. That's not to say they aren't true of course but they may cause us problems.
One of Kant's concerns was to show the limits of human reason. There are some things that we cannot know of. For example, almost by definition, we cannot have a sense of the experience of objects as they are when they are not being perceived. These "things-in-themselves" can never be observed and so we can never know of them. He refers to objects like this, in somewhat Platonic terms, as transcendental. Unfortunately, there are other transcendental elements.
If objects that exist beyond our sense experience are beyond our knowledge this would also seem to include God as well. Certainly, God is transcendental so we are forever precluded from knowing him through pure reason alone. Fortunately, Kant postulates that there are other types of reason through which we can gain knowledge of God. It turns out that God can be known through practical reason which we need as a basis for our understanding of morality. We will be discussing Kant's approach to morality in a later chapter.
There is a further transcendental element that we need to consider. This is the one that gave Hume so many problems. While the categories of mind can be understood as the elements of our sense experience, who is it that's having this experience. In other words, what is the nature of the self? Well, it’s not an object of sense experience since the self is what has sense experience. We can infer its existence though through the fact that our sense experience is unified. Something must be doing the unifying and this something is the self; which Kant terms the transcendental unity of perception.
So where does this leave us? We have partially solved Hume's problem of skepticism. We can have certainty about causality since it is a condition for our having sense experience. We can still maintain that knowledge comes from sense experience without buying into the denial of such fundamental knowledge as we have of substance, space, time, and causality. However, we do have to recognize that applying these categories to the transcendental realm of noumena will never be possible. While we gain a better understanding of the workings of the mind through the innate structures it contains, we don't have to postulate that some ideas are innate and fully formed at birth in the mind. So while we have defined very definite limits to our knowledge we have advanced far beyond the limits of rationalism and empiricism. Have we advanced as far as we can? To answer that we'll need to address what happened in philosophy after Kant.