dog cat horse ox run jump climb fish green blue orange apple pear swiftly Jane BillThose probably aren't the actual words, but that doesn't matter. He then told us, "Organize these."
"According to what," someone asked. It might have been me, but I don't remember and it's not important.
"Not my problem," he answered with a shrug. "Just organize them however you want." I put them in alphabetical order: apple Bill blue cat climb dog fish green horse Jane jump orange ox pear run swiftly. Done. Our instructor asked us how we organized them - one person had organized them into parts of speech, with "fish" going under both the Noun and Verb headings, "orange" being both a Noun and an Adjective, and "Jane" and "Bill" under the Proper Names subheading of Nouns. Another student had divided them into Domesticated Animals and Other. We had hit all the obvious ones, and then the instructor put some others on the overhead and had us try to guess what they were.
"You might like to organize them like this:"
Bill apple orange
I'm not sure why vowel count would be important to you, but yeah, I guess that's a legitimate way to organize words. "If you were an OSU football player," he said, "You might organize them like this:"
ox cat Bill apple orange swiftly
dog blue climb
run Jane green
It took us a moment to realize that they were organized according to letter count, and about a week to realize that this guy really had it out for the OSU football team. Another one had us completely stumped:
See if you can figure that one out. I won't spoil it for you.
This was our first lesson in historiography: the same information can be organized and categorized differently, depending on what your agenda is at the moment. A primary source document, for example, could be filed away under any number of headings, depending on what it's being used for. You might file a diary in one place if you were interested in local history, another if you were documenting evolution of a local dialect.
Here's another example that strikes closer to the point I'm trying to make: take the numbers one, two, three, and four. You can divide them in half according to whether they're even or odd. You could divide them into another half depending on whether they are more or less than 2.5. You could divide them into yet another half depending on whether they're prime or not, whether they're factors of six or not, or whether they are within 0.75 of 2.5 or not (same output, different ways of getting there). With all these different ways of halving the same four numbers, a question naturally arises: How many halves are there?
As many as you go to the trouble of making.
That's one consequence of the fact that categories are artifacts - you can come up with as many as you like, arbitrarily. What's behind it is the fact that reality doesn't come "cut up" for us, it just is what it is and that's that. We come along and see regularities, patterns, and we give them names; justly so, for a great many of them are useful, but not rightly so, for language can't "stick" to the world the way we want it to. To really unravel this problem, let's consider whether propositions about the future have truth value.
Consider the proposition, "The Sun will rise tomorrow." Is it true or false? I don't mean "pick one," I mean to ask if the statement itself has truth value. If it doesn't... well... OK, I didn't think that one through. I'm not sure how it wouldn't, though; I mean, you've got solid referents and you know what you mean, it's just that you're attempting to refer to an event that hasn't happened yet, and that's kind of what we're asking about... but at first blush, it sure seems like it ought to be either true or false that the Sun will rise tomorrow. I mean, if it does then it's true, if it doesn't then it's false. Right? But that brings up all kinds of weirdness.
Like, does the future event reach back in time and make it true all along? Or does it have to happen and then it "becomes" true but it wasn't before? Or did it always have to have been true, and it just so happens that someone said it and was right, but we've somehow proved determinism?
None of those things, actually. Because we've made a mistake, and I don't mean that propositions about the future don't have truth value, I mean our mistake is in how we're conceiving of truth value. We're assuming that "truth" is a "thing" that propositions "have" somehow, like it's not just some utterance that we talk about in some weird way. That is to say, that is what's going on: propositions are just utterances, and "truth value" is just a funny way of talking about them. When we considered the matter of propositions about the future having truth value, that exposed the thread we want to pull on, because we assumed that reality has an obligation to language. Like if we talked about things in the right way, that would mean something. That's the illegal move: we can't do that. I mean, we can try, but it doesn't work.
Try and think about it: what exactly is truth? At some level, we want it to be a property of language, right? Trot out all the philosophy you want on correspondence or coherence, it's about language, and language is made-up. So anything we want to say, and anything else we want to say about what we've said, can't possibly impose an obligation on reality - yet we think we can establish that things are this way or that way.
It sure seems to work a lot, and we've got a whole load of convenient language built up around the idea. It's useful, no doubt, and it's useful because we get stuff that works out of it. But that doesn't mean that it works in any kind of interesting way - the language, the descriptions, the math, all that doesn't do the work. Whatever it's supposed to represent, now - that thing, whatever it is, is what does the work. The language is a fancy tag we hang on it so we can keep track of it. It doesn't do anything else.
Even the chemical elements, which would seem to be the closest we'd need to get to natural kind terms, don't cut the mustard. Hydrogen is hydrogen, we're taught in high school chemistry, except when it's deuterium, which doesn't usually make a difference unless it makes heavy water, which actually slows down the reactions in your body enough to kill you. Oops. "Oh, but it still acts like hydrogen, the extra neutron just affects chemical reactions." But that's just it: so does the proton. What we call "hydrogen" is just what "one proton" chemically acts like, and even that proton is made up of quarks held together by gluons. Even then, a proton needs one red quark, one blue, and one green, two of which must spin up and the third of which must spin down. That's more like a recipe, really: get these things together and bam, they'll do this. Gather up the components for a proton and mash 'em together, spin an electron around it, and hooray, you've got a hydrogen! Get two of those together, and you've got a noble gas! Guess what comes next? That's right, an alkali metal! Keep it up, and you'll figure out periodicity and things like that. But periodic properties are patterns, not things, and when you get right down to it, even the "things" we think of as things are also patterns.
To tie things back in some, we were talking about truth before (it's a word we made up), and how picking at that makes it a little more obvious that language doesn't "stick" to the world, all of which was in order to show how things can be organized pretty much however you want. Because they're not actually organized in any way at all.
Except that some people disagree with this. Some people think that there are these things and they behave in these ways and if you're doing something not like one of the ways then you're wrong. But how could you be? You're just doing what you do, acting according to your nature, whether you're a proton or a pedophile. OK, bad example, and protecting children is like way more important than letting pedophiles be. Point is: they're acting like there's a normative component to reality. The only restriction that reality seems to impose is that it has to be possible, and then it can be done (or may be inevitable? I don't know). There's this bizarre notion I encounter quite a lot, that physical laws are somehow "enforced" and moral laws require a "lawgiver," and sure they're distinct notions but they seem to run together an awful lot. Like somehow morality boils down to "doing what you're told." Well, then of course you'd need a moral lawgiver, if that's how morality worked. But it's not. It turns out that "good" is also one of those words we made up, and there are different ways of working with the idea that can be evaluated according to various sets of criteria; but then we're doing meta-meta-ethics which is probably dangerous. If there were moral laws, then they would be inviolable, just like physical laws - which are not pronounced by anyone, and in fact only as inviolable as we find them to be (if they do get violated, then we weren't dealing with genuine physical laws in the first place).
It's only by conceit of our perspective that we're able to come up with these bizarre notions, growing up as we do in a sea of patterns upon patterns that self-perpetuate based on their ability to be regular enough to be worth naming. But even then, there also needs to be an ability to be a little irregular as well, or you get stuck. What these patterns are of, by contrast, is much subtler and finer than we had ever suspected: what we think of as "solid matter" isn't actually solid at all, and our idea of what "matter" actually is breaks down so quickly under scrutiny that it's a wonder we ever thought it was coherent in the first place. Probing for the ultimate nature of reality has shown, in stark contrast to Plato's Forms, that reality is in fact formless.
Ironically, since this makes it a lot harder to keep track of, it's a good thing we have all these fancy tags after all!