David Hume
We have seen several alternatives to Cartesian metaphysics and so now we should turn our attention to an alternative to Cartesian epistemology. Remember that as a rationalist Descartes believes that at least some of our knowledge is innate. In particular, our knowledge of self and God are innate though there may be other aspects of knowledge that are innate as well.
So how much worse can it get? Well, as we've seen empiricism does have a few problems which led to the metaphysical position of idealism. For Berkeley, to be a consistent empiricist entailed the denial of material substance. As we'll see, Hume shares that belief as well; and more. Before we get to that here are a few preliminary thoughts of David Hume.
As an advocate of empiricism, of course, Hume believes that all knowledge comes from perceptions. There are two types of perceptions: impressions and ideas. The ultimate source of all of our ideas is impressions. “Impressions” is the term Hume gives to sense experience; which Locke referred to as sensations. Impressions give rise to ideas which, in Hume's phrasing, are "less lively copies" of impressions.
Hume recognizes that it might be difficult to believe that all ideas arise from impressions. After all, it seems that we can have some ideas without any corresponding impression. Take, for example, a golden mountain. It appears that here we have an idea that cannot possibly come from sense experience. But, consider, says Hume, that we do have a sensory experience of gold and a sensory experience of the mountain. All we've done, in the case of the golden mountain, is to combine in a new way, ideas that arose from prior impressions.
There are two ways the human mind can combine ideas. This Hume calls "relations of ideas" and "matters of fact." “Relations of ideas” is the term Hume uses to refer to ideas that are connected as a matter of necessity as if by definition. A good example of this would be the idea of a triangle and the idea of a three-sided object. These ideas go together in such a way as to be incapable of being thought of as different. Though we gain these ideas by sense experience we can judge their necessary connection before sense experience.
Matters of fact, as Hume points out in his Enquiry Concerning Human Understanding, "are not ascertained in the same manner." Since we connect these ideas based on sense experience "the contrary of every matter of fact is still possible." This may sound like a strange thing to say but sense experience cannot tell us what will be only what is. So while I can observe the sun rising today I cannot infer that the sun must always rise. The ideas of "sun" and "rising" are not necessarily connected. So "That the sun will not rise tomorrow is no less intelligible a proposition, and implies no more contradiction than the affirmation, this it will rise."
To gain a better understanding of why denying matters of fact implies no contradiction let's investigate what is the nature of the connection between these ideas. Matters of fact are based on causality. The causal relationship itself has three elements. First is the element of temporal succession. There is a certain order in which cause and effect occur; cause always comes before effect. Second is the element of spatial proximity. Cause and effect have to be in some contact with one another (remember Aristotle's problem of separation?). Now, both of these elements are observable but they are not sufficient to understand causality. Just because two events are in the right time sequence and in the proper spatial proximity we cannot infer that they are causally related. Therefore, there is a third element we need to address: necessary connexion (I'll use Hume's spelling since it's unique!). For two events to be causally connected we need the idea that the effect could not happen but for the cause. This is the critical element of causality but it gives us a problem.
While we can observe the first two elements, the necessary connexion is unobservable. This, of course, spells a problem for the empiricist. And Hume is nothing if not consistent in his empiricism. Since we cannot observe the necessary connexion we cannot know of it. And since we can't know of it, we can't know matters of fact. Thus, the contrary of every matter of fact is still possible.
Perhaps, an example will illustrate. Let's look at a couple of billiard balls; an appropriate example since Hume was a bit of a pool shark. Suppose ALL we know about a pair of billiard balls, labeled A and B is the following: Billiard ball A is moving towards billiard ball B and will make contact with it. Now, as a result of this contact what must happen? The problem is that we can think of many things that may happen but no one of them must happen. Billiard ball B moves? Why would that be necessary? Suppose it’s firmly affixed to the surface of the table. Or perhaps it’s ten times heavier than ball A. You might say, well, something has to happen. OK but what exactly. Saying something must happen is tantamount to no knowledge at all.
So empiricism can only tell us what is happening or what has happened. But there is no rational justification for supposing that the future will be like this. All we have to make this judgment is our previous experience and the habit or custom of expecting the future to be like this. That it always has worked before is no sure sign that it will continue to work. It's no wonder Hume is often considered a skeptic about knowledge. But, as I've been warning, it gets worse.
Clearly, from an empiricist perspective, we have to deny material substance. After all, we have no sense of the experience of it. So on this point, Hume agrees with Berkeley. But Hume doesn't want to stop there. In his book A Treatise of Human Nature, Hume asks “from what impression does our idea of the self arise?” Remember that all ideas must arise from impressions; the self is no different. But when I think about it, I can find no impression:
"For my part, when I enter most intimately into what I call myself, I always stumble on some particular perception or other, of heat or cold, light or shade, love or hatred, pain or pleasure. I never can catch myself at any time without a perception, and never can observe anything but the perception." So what are we to conclude from this? It seems we have only two options: There is no self. The self just is these perceptions. See, the same logic which required us to deny material substance leads us to deny mental substance as well. That's just not an option. However, it seems impossible to suppose that there is no self (of course, this is exactly what the Buddhist says but more on that much later!). So we are left with the only possible conclusion. The self just is what Hume, somewhat unromantically, calls "a bundle of perceptions."
This did distress Hume to some degree and he confesses as much in the Appendix of A Treatise of Human Nature. However, it was unclear to him how to free himself from this problem. And so this is where he left it. We now have, then, a new metaphysical position. We began with Cartesian dualism which Locke also adopted. However, due to the problems with the material substance we were led, by Bishop Berkeley, to idealism. Now, our position maintains that there is no substance at all; neither material nor mental. This view is called phenomenalism.
So let's sum up. By a consistent application of empiricism (an innocent enough theory when we first considered it), we have eliminated all substance. The self is now a bundle of perceptions and, one supposes that objects are also nothing but a collection of perceptions. We've come a long way towards the complete elimination of an independent objective reality (so a long way from Plato) and we're only in the 18th century! And, as is the case for most philosophers, Hume's greatest contribution to philosophy was a problem. OK, another problem. This one later became known as "Hume's Fork." About epistemology, we have two options. We can either have certainty in knowledge but it won't be about sense experience (relations of ideas) or we can have knowledge of sense experience but it won't be certain (matters of fact). This problem gave philosophers fits for years and caused Immanuel Kant to lose sleep. In this case, a good thing because he solved it. Well, sort of.
So how much worse can it get? Well, as we've seen empiricism does have a few problems which led to the metaphysical position of idealism. For Berkeley, to be a consistent empiricist entailed the denial of material substance. As we'll see, Hume shares that belief as well; and more. Before we get to that here are a few preliminary thoughts of David Hume.
As an advocate of empiricism, of course, Hume believes that all knowledge comes from perceptions. There are two types of perceptions: impressions and ideas. The ultimate source of all of our ideas is impressions. “Impressions” is the term Hume gives to sense experience; which Locke referred to as sensations. Impressions give rise to ideas which, in Hume's phrasing, are "less lively copies" of impressions.
Hume recognizes that it might be difficult to believe that all ideas arise from impressions. After all, it seems that we can have some ideas without any corresponding impression. Take, for example, a golden mountain. It appears that here we have an idea that cannot possibly come from sense experience. But, consider, says Hume, that we do have a sensory experience of gold and a sensory experience of the mountain. All we've done, in the case of the golden mountain, is to combine in a new way, ideas that arose from prior impressions.
There are two ways the human mind can combine ideas. This Hume calls "relations of ideas" and "matters of fact." “Relations of ideas” is the term Hume uses to refer to ideas that are connected as a matter of necessity as if by definition. A good example of this would be the idea of a triangle and the idea of a three-sided object. These ideas go together in such a way as to be incapable of being thought of as different. Though we gain these ideas by sense experience we can judge their necessary connection before sense experience.
Matters of fact, as Hume points out in his Enquiry Concerning Human Understanding, "are not ascertained in the same manner." Since we connect these ideas based on sense experience "the contrary of every matter of fact is still possible." This may sound like a strange thing to say but sense experience cannot tell us what will be only what is. So while I can observe the sun rising today I cannot infer that the sun must always rise. The ideas of "sun" and "rising" are not necessarily connected. So "That the sun will not rise tomorrow is no less intelligible a proposition, and implies no more contradiction than the affirmation, this it will rise."
To gain a better understanding of why denying matters of fact implies no contradiction let's investigate what is the nature of the connection between these ideas. Matters of fact are based on causality. The causal relationship itself has three elements. First is the element of temporal succession. There is a certain order in which cause and effect occur; cause always comes before effect. Second is the element of spatial proximity. Cause and effect have to be in some contact with one another (remember Aristotle's problem of separation?). Now, both of these elements are observable but they are not sufficient to understand causality. Just because two events are in the right time sequence and in the proper spatial proximity we cannot infer that they are causally related. Therefore, there is a third element we need to address: necessary connexion (I'll use Hume's spelling since it's unique!). For two events to be causally connected we need the idea that the effect could not happen but for the cause. This is the critical element of causality but it gives us a problem.
While we can observe the first two elements, the necessary connexion is unobservable. This, of course, spells a problem for the empiricist. And Hume is nothing if not consistent in his empiricism. Since we cannot observe the necessary connexion we cannot know of it. And since we can't know of it, we can't know matters of fact. Thus, the contrary of every matter of fact is still possible.
Perhaps, an example will illustrate. Let's look at a couple of billiard balls; an appropriate example since Hume was a bit of a pool shark. Suppose ALL we know about a pair of billiard balls, labeled A and B is the following: Billiard ball A is moving towards billiard ball B and will make contact with it. Now, as a result of this contact what must happen? The problem is that we can think of many things that may happen but no one of them must happen. Billiard ball B moves? Why would that be necessary? Suppose it’s firmly affixed to the surface of the table. Or perhaps it’s ten times heavier than ball A. You might say, well, something has to happen. OK but what exactly. Saying something must happen is tantamount to no knowledge at all.
So empiricism can only tell us what is happening or what has happened. But there is no rational justification for supposing that the future will be like this. All we have to make this judgment is our previous experience and the habit or custom of expecting the future to be like this. That it always has worked before is no sure sign that it will continue to work. It's no wonder Hume is often considered a skeptic about knowledge. But, as I've been warning, it gets worse.
Clearly, from an empiricist perspective, we have to deny material substance. After all, we have no sense of the experience of it. So on this point, Hume agrees with Berkeley. But Hume doesn't want to stop there. In his book A Treatise of Human Nature, Hume asks “from what impression does our idea of the self arise?” Remember that all ideas must arise from impressions; the self is no different. But when I think about it, I can find no impression:
"For my part, when I enter most intimately into what I call myself, I always stumble on some particular perception or other, of heat or cold, light or shade, love or hatred, pain or pleasure. I never can catch myself at any time without a perception, and never can observe anything but the perception." So what are we to conclude from this? It seems we have only two options: There is no self. The self just is these perceptions. See, the same logic which required us to deny material substance leads us to deny mental substance as well. That's just not an option. However, it seems impossible to suppose that there is no self (of course, this is exactly what the Buddhist says but more on that much later!). So we are left with the only possible conclusion. The self just is what Hume, somewhat unromantically, calls "a bundle of perceptions."
This did distress Hume to some degree and he confesses as much in the Appendix of A Treatise of Human Nature. However, it was unclear to him how to free himself from this problem. And so this is where he left it. We now have, then, a new metaphysical position. We began with Cartesian dualism which Locke also adopted. However, due to the problems with the material substance we were led, by Bishop Berkeley, to idealism. Now, our position maintains that there is no substance at all; neither material nor mental. This view is called phenomenalism.
So let's sum up. By a consistent application of empiricism (an innocent enough theory when we first considered it), we have eliminated all substance. The self is now a bundle of perceptions and, one supposes that objects are also nothing but a collection of perceptions. We've come a long way towards the complete elimination of an independent objective reality (so a long way from Plato) and we're only in the 18th century! And, as is the case for most philosophers, Hume's greatest contribution to philosophy was a problem. OK, another problem. This one later became known as "Hume's Fork." About epistemology, we have two options. We can either have certainty in knowledge but it won't be about sense experience (relations of ideas) or we can have knowledge of sense experience but it won't be certain (matters of fact). This problem gave philosophers fits for years and caused Immanuel Kant to lose sleep. In this case, a good thing because he solved it. Well, sort of.