Monday, June 13, 2016

Hume in se

My most popular post by a huge margin is on Hume and Popper, although I'm not sure if it gets so many hits because it is insightful or because philosophy professors keep sending their students to it as an example of how not to analyze philosophy.  I finally got around to reading Hume's famous work "An Enquiry Concerning Human Understanding" and I was curious whether it would cause me to rethink my other post.  In a word, no, because that post was mostly (I realize now) about Popper.  But I'm still glad I read it.

After slogging through Spinoza and Leibniz, I can't say what a pleasure it was to read a book in the Anglophone tradition, where the author's primary effort is to be understood.  That, and Hume's empiricist outlook, make this book a pleasure to read even where I disagree with it, and that does not include his generally sceptical approach.

Hume tries to show that we can only ever assume causality, never know it directly, which is unobjectionable to me.  We see one event follow another repeatedly, and assume that there is a causal link between them.  We can't "see" the cause, only infer it from observations.  However, I think Hume doesn't consider the nature of an observation sufficiently.  Science (not to mention practical life in general) is based on the assumption that nature follows laws consistently.  If it doesn't, we have no basis for science at all, not to mention that just staying alive would be quite a challenge.  So in that sense, we are reasonable to assume that the same event will be followed by the same consequences every time.

The trick is, what constitutes an event?  If we see one billiard ball striking another (as Hume likes to give as an example), what we're really seeing is one composite material striking another.  It may appear that the materials are the same from one instance to the next, and this may be a reasonable assumption, but someone could easily trick us by making some other material look like a billiard ball, or change a ball in some way, such as hollowing it out.  This is sometimes practiced in baseball, where batters may put cork inside their bats or pitchers may doctor balls.

But we know that these are not really the same things, and even if it appears the same to us, we are not surprised to find that they have been changed if their behaviour is different.  Similarly with bread (another of Hume's examples), which is even more heterogeneous, and in fact it would be reasonable to say that no two loaves of bread are quite the same.  What looks and smells like bread may have different proportions of various ingredients, including some that may not be nutritive and may even be harmful.

What if we could break matter down into its smallest units, i.e. atoms?  Then we would truly be surprised if the atoms behaved differently from one time to the next, because they are supposed to be identical one to another.  Molecules as well, when composed of the same atoms, are also supposed to be the same.  And this holds true even though atoms can actually be broken down further; a hydrogen atom with one proton and one electron is identical in behaviour to any other hydrogen atom, at least at a super-atomic level.  Subatomic particles are also identical with others of the same type, but at that level, natural laws, not to mention Newtonian mechanics, break down, and we cannot say that an electron will behave the same as any other electron it terms of its motion.

The trick to all this is that observations are, by their nature, empircal and uncertain.  We can never be sure that we have witnessed a truly "atomic" event (defined as one consisting entirely of homogeneous units that always behave consistently), but we can gain a degree of certainty (i.e., a strong probability) by repeated observations.  This is the thing that Hume considers so odd:  that we don't consider a single event to be determinative, even though it contains all the information that we get from multiple events that we do consider evidence, or even proof, of causality.  He is correct up to a point, but he overlooks the fact that it is the very repetition that allows us some certainty that we have been witnessing the same thing repeatedly and not a congeries of different things.

This doesn't disprove Hume's fundamental point, of course, that we cannot see causation and must infer it.  The thing that surprised me was that Hume seems to accept causality as fundamentally valid in spite of his sceptical analysis.  Later in the book, he takes causality as certain, even though we cannot know it a priori.  It is so certain to him that it is evidence that we do not have freedom of the will:  every material event has a material cause, therefore we cannot possibly have free will, which implies a non-material cause.  I feel that his argument on this point is much weaker than that on epistemology.  Surely, within a materialist framework, free will has no place.  On the other hand, Hume excuses the potentially negative consequences of his determinism by saying that it doesn't matter because we act as though we have free will.  We can't excuse people for their actions, nor blame God for determining everything.  He doesn't have much of an argument for this contradictory conclusion and really just seems to want to avoid the logical conclusions of the things he has just asserted with such confidence.

I should add that, apart from his slippery response to material determinism, Hume's analysis of human behaviour seems much more convincing in general than Spinoza's.  Hume tries to analyze actual behaviour, not construct an entirely consistent philosophical system from axia.  This allows him to make some observations that, while not provable, are nonetheless much more useful than anything Spinoza says.

Hume is the obvious forerunner to idealism in more than just his approach to causality:  he, like Kant and his successors, treats phenomena as fundamentally different from noumena.  This isn't really a contribution to philosophy, since Hume states it as the universal opinion, but obviously Hume's arguments put these ideas front and center.  What I don't understand about him and any of the idealists who build on him is why they  think phenomena bear no relation to the noumena from which they are derived.  True, our perception of "red" or "hot" may not correspond in an objective sense to what an object is (another Humian example), but it does correspond directly to something:  to reflected wavelengths of light, to energy in surface molecules, etc.  I may not be able to describe the sensation of heat to someone who has never felt it, but I could tell them something concrete and material about a hot object that would remain true for all people regardless of whether they actually felt heat.

Hume's most interesting observation, from my perspective, is that all of our reasoning is by analogy.  I have frequently thought this and have wanted to build it out into a more general theory of epistemology and psychology.  If we see one billiard ball strike another, and then we see one croquet ball strike another, we are not surprised that the result is similar because they are analogous objects.  Similarly if we see, for example, inflated round balls (soccer or basketballs, e.g.) strike each other, although the analogy isn't as complete in that case.  Scientists take this analogy with them when they consider the behaviour of atoms in a gas or a liquid, even though it may not be precisely the same.  And when subatomic particles seem to travel multiple paths at the same time, we have no analogy for it, and it just seems wrong.  We know it models their behaviour in a mathematical sense, but we can't really conceive it in conjunction with our general understanding of the world.

This becomes even more important, and perhaps less appreciated, in interpersonal relationships.  What happens when we face a new situation in a relation with another person?  I don't know about other people, but I tend to think of similar cases in literature that I have read or in movies that I have seen.  Sometimes it seems too close to fiction to be real, but the important part is that my mind is modeling the real events based on analogous fictional ones.  Even though I know that fiction is not real, I may not have any real-world examples to go on, or they may seem a weaker fit.  I believe Joseph Campbell used similar logic to argue for the importance of myths in human society:  even if we don't believe them, they still shape the way we think.  (I disagree with Campbell that it doesn't matter if we don't believe them, but I agree that they shape us in any case.)