Before we get into epistemology, which is the topic for the next two days, I want to say something about the topic of meaningfulness. Last time someone very sensibly asked me to clarify what it means to say that a sentence is meaningful. We discussed the notion that it might the same as understandability; the idea was that, if a sentence is meaningful then we can understand it, and if we can understand it, then itís meaningful. But I raised the problem that of course there are some sentences which many of us canít understand, but nonetheless theyíre perfectly meaningful. Still, you might say: if a sentence can be understood by someone, then thatís enough to make the sentence meaningful.
Now, Iím still not satisfied by this account of meaningfulness, if for no other reason than that it doesnít say why sentences are meaningful. Look at it this way: if the worldís greatest genius can understand some incredibly complex, jargon-filled sentence, thereís something about the sentence that makes it meaningful, right? Itís not just the fact that the genius can understand it; there has to be some reason why the genius can understand it. Well, there has to be a meaning there for the genius to understand! And that just raises the question: When does a sentence have a meaning and when doesnít it?
Iím not going to go into this in much depth, but I just wanted to say that it clearly has something to do with the fact that the words in the sentence have a meaning. Roughly put: If a sentence is meaningful, then each word in the sentence has some meaning. That principle seems to me at least generally true, if not always true. Take the sentence, "Barry is a bachelor." Itís meaningful; and each word in that sentence, "Barry," "is," "a," and "bachelor," has a meaning in the English language. That doesnít mean that it has a reference, right? Because as I said, different kinds of words have different kinds of meaning. But after we have worked out our theory of meaning, then we can say when a word of a certain sort has a meaning, and what sort of thing its meaning might be. So even the words "is" and "a" have some sort of meaning that we would have determined.
By contrast, take the converse of the principle stated above: If each word in a sentence has a meaning, then the whole sentence is meaningful. Do you think thatís right? Well, consider the following string of word: "Had computer sky flatly in cat over." Each word in that "sentence," if you want to call it a sentence, is meaningful. But it certainly is not meaningful. So maybe you could amend the principle, like this: If each word in a grammatical sentence has a meaning, then the whole sentence is meaningful. But even this principle is incorrect. Hereís a counterexample: "The evil kitten used green songs to grate cheese." Thatís perfectly grammatical, but I donít think we want to dignify that sentence as meaningful.
So the question then is: what, in addition to having meaningful words in a grammatical order, does a sentence have to have, in order to be meaningful? Iím not going to try to improve the principle; but it seems pretty obvious to me that you could improve it in various ways.
I would be irresponsible if I did not mention, before I leave this topic, a very important theory of meaningfulness, called verificationism. Hereís a definition:
A sentence is meaningful iff it is at least in principle verifiable, i.e., there are some conditions under which it could be shown to be true or false.
If "Barry is a bachelor" is meaningful, on this theory, then there are some conditions that could be fulfilled, under which you would have shown that Barry is indeed a bachelor. But if there are no such conditions then the sentence, "Barry is a bachelor," would be meaningless; of course there are such conditions, so that sentence is meaningful. On the other hand, "The evil kitten used green songs to grate cheese" is meaningless, because there arenít any conditions under which this sentence could be shown to be true. How could you show that sentence to be true? When would I know that I was looking at an evil kitten? When would I know that the kitten was using a song? What is a green song? How could a kitten use a song to grate cheese? Well, the point is that you couldnít show this sentence to be true; there isnít any way to do it because we donít even know what observations we could make, that would confirm it in the slightest.
Very briefly I want to raise one problem for verificationism. Itís the same problem that I raised for the view that meaningfulness is understandability. Namely, this view doesnít seem to explain why a sentence is meaningful. It might be perfectly true that, if a sentence is meaningful, then it can be verified (or falsified). But it seems like the reason that it can be verified, or falsified, is the fact that itís meaningful. I mean, again, this view faces this problem: what is it about the sentence that makes it verifiable, or falsifiable? That is what will make the sentence meaningful.
And in my opinion anyway, itís probably going to turn out to be a function of a grammatical combination of meaningful words, plus some other conditions (like, the right words being joined together). I suspect that that is what would make a sentence understandable, and verifiable or falsifiable, and meaningful. But thatís all Iím going to say about meaningfulness. On to epistemology.