My belief is almost surely false
I'm perplexed. Arnold Zwicky at the Language Log just wrote,
Zwicky acknowledges in two consecutive sentences (a) that he believes an expression was in common use in the 1950s, and (b) that this expression "almost surely" was not in common use in the 1950s. He has just asserted "I believe P; but P is almost surely false". This is a form of expression which philosophers (following Wittgenstein and G.E. Moore) claim you just can't say--utter nonsense--and I've always been inclined to agree with them. The canonical example is, "I believe it is raining; but it is not raining". (More precisely, you can't assert such a thing. You can say it in non-asserting contexts, like "Suppose I believe it is raining, but it is not raining.") And yet here is Arnold Zwicky asserting it loud and clear--and worse, I seem to understand what he is talking about. Maybe. I can't tell.
The problem, you see, is that normally when you assert something, you are putting the force of your beliefs behind it. If I tell you (sincerely and without qualification) that it is not raining, you have the right to conclude that I believe it is not raining. But if I also tell you that I believe that it is raining, then it would seem that I believe two contradictory statements.
It seems like Zwicky is describing believing as a purely perceptual experience, without any connection to what he takes to be true. He's treating "believe" like "seem to see": you can say "I seem to see that it is raining, but it is almost surely not raining", in which you are describing on the one hand a subjective experience you are having involving images of falling water, and on the other hand your beliefs about what is objectively the case about the weather. But "I believe it is raining" doesn't work the same way, because you can't drive a wedge between your beliefs about the world and, well, your beliefs about the world. You see the problem?
Maybe Zwicky is treating himself as two distinct people. We relativize other people's beliefs all the time: there's nothing weird if I say, "Arnold believes it is raining, but it is almost surely not raining." But to adopt that same detachment toward yourself...the mind boggles.
Try this little experiment: ask a number of friends (linguists or not) how long they think the idiom the whole nine yards has been around; if they're over 30, ask them if they remember reading or hearing it when they were young. I myself believe that it was in common use when I was in high school and college (in the 50s and 60s).
My belief is almost surely false, since much tedious digging by lexicographic types has gotten attestation of the idiom back only to the early 70s. I still find this astonishing.
Zwicky acknowledges in two consecutive sentences (a) that he believes an expression was in common use in the 1950s, and (b) that this expression "almost surely" was not in common use in the 1950s. He has just asserted "I believe P; but P is almost surely false". This is a form of expression which philosophers (following Wittgenstein and G.E. Moore) claim you just can't say--utter nonsense--and I've always been inclined to agree with them. The canonical example is, "I believe it is raining; but it is not raining". (More precisely, you can't assert such a thing. You can say it in non-asserting contexts, like "Suppose I believe it is raining, but it is not raining.") And yet here is Arnold Zwicky asserting it loud and clear--and worse, I seem to understand what he is talking about. Maybe. I can't tell.
The problem, you see, is that normally when you assert something, you are putting the force of your beliefs behind it. If I tell you (sincerely and without qualification) that it is not raining, you have the right to conclude that I believe it is not raining. But if I also tell you that I believe that it is raining, then it would seem that I believe two contradictory statements.
It seems like Zwicky is describing believing as a purely perceptual experience, without any connection to what he takes to be true. He's treating "believe" like "seem to see": you can say "I seem to see that it is raining, but it is almost surely not raining", in which you are describing on the one hand a subjective experience you are having involving images of falling water, and on the other hand your beliefs about what is objectively the case about the weather. But "I believe it is raining" doesn't work the same way, because you can't drive a wedge between your beliefs about the world and, well, your beliefs about the world. You see the problem?
Maybe Zwicky is treating himself as two distinct people. We relativize other people's beliefs all the time: there's nothing weird if I say, "Arnold believes it is raining, but it is almost surely not raining." But to adopt that same detachment toward yourself...the mind boggles.
Posted by Jeff on Tuesday, February 21, 2006
7 Comments:
I only read those statements as moderately weird. First, the key point is the "almost" in "almost surely false." If you reworded it to say "I believe X, but there's a good chance I'm wrong," then that sounds a bit better. Secondly, I see the intent as being more an exaggeration of "I believe X, but there's no way I'm going to be able to find sufficient evidence to prove it." Though calling it "almost surely false" does seem to be a bit over the top for that.
On a side note, I know Arnold from shape note singing. It's always amusing to see people pop up in different contexts like this. :-)
Graham brings up a good example, but I'm pretty sure it's an example of something different than what Zwicky is doing. "I believe X, but there's a good chance I'm wrong" is a hedge. It's like saying, "It's raining--well, maybe it isn't raining." Hedging effectively cancels what you originally said. Or partly cancels it. Or at least weakens it. Or maybe it doesn't... At any rate, I don't think Zwicky is just hedging here, unable to make up his mind. It sounds to me like he's quite clear about what he wants to say.
"...it is difficult to deny one's own experience or what one's memory has constructed as an experience"--I agree that this is often difficult. But the thing is, as I read Zwicky, he has denied his experience. He at least has said that it is "almost surely false", which is pretty darn close to a flat-out denial. In other words, confessing that his belief is almost surely false amounts to abandoning that belief.
This leads me to my latest theory. Zwicky isn't quite treating himself as two people, but he is treating himself as one person at two different times--which is pretty similar in a lot of ways. He uses the present tense for extra force, but what he's really describing is a belief that he once held (and perhaps is still sympathetic toward), but has in fact given up in the light of the evidence. I just can't make sense of it otherwise.
My trouble arises from this basic thesis: you can't believe something and at the same time think that it's false. Only slightly more tenuous: you can't believe something and at the same time think that it's almost surely false--or even likely to be false. Because I take it to be unassailable that if you believe something, then you think that it is in fact true. Isn't that right?
Here's a related question. Is it really possible to believe something, and also think that your belief is unreasonable? I'm less sure about this, but there's definitely something funny about a case like that.
What about the case of knowing that one's belief can't possibly be accurate, but believing it anyway because it is basically a best guess? This is a case of believing something while at the same time knowing that it is (to some extent) false (since I know that my best guess can't reasonably be %100 correct.) There's probably some word for this that isn't insanity seeing as how it appears with great regularity, for instance, in one's relationships with other people. I just don't know that word.
But then if that is possible it should also be possible for one's idea of a 'best guess' to get a bit muddled; I for one am not a consistent or philosophically industrious enough person to make the knowledge that I base my beliefs on and the knowledge I suspect to be the most accurate match up completely. That blurs the boundaries of what it is possible to know and not believe and vice versa.
Interesting. I'm going to try to unpack that a bit.
What does it mean for Bert to know that a claim P can't possibly be accurate? In particular, you're talking about knowing P to be false "to some extent", or not "100% correct". So if you mean that in the most straightforward way, then Bert would know that P is partly wrong. In that case, I maintain that Bert doesn't (entirely) believe P. Maybe he sort of believes P, or he believes sort-of-P; but he doesn't actually believe P itself--because he thinks that P doesn't capture the whole situation, or P isn't really a precise enough way of thinking about things, or P misses the point. Maybe he doesn't exactly disbelieve P, but there are plenty of options besides belief and disbelief.
But maybe what you're getting at is that Bert isn't 100% certain about the claim. He knows that he makes lots of mistakes, and that he can't reasonably expect himself to have gotten things down pat this time. He doesn't have any particular reason to think that P is false, but at the same time he acknowledges its fallibility. I wholeheartedly agree that it's possible (laudable, even) to hold beliefs and still maintain that kind of self-critical attitude (which I suspect is also what Graham and Dad were driving at). But there's a world of difference between "I could be wrong" and "I am wrong". The second statement really involves changing your beliefs--giving up the wrong belief. You don't believe what you know is wrong.
Here's another way of approaching the problem:
"I believe it's raining, but I could possibly be wrong."
"I believe it's raining, but there's a good chance I'm wrong."
"I believe it's raining, but it's extremely likely that I'm wrong."
"I believe it's raining, but I'm almost surely wrong."
"I believe it's raining, but I'm surely wrong."
The last sentence is patent nonsense (or a self-canceling hedge). On the other hand, the first sentence is patent sense. This means that somewhere along this continuum (if it really is a continuum?) we cross the line from sense to nonsense. I'm pretty confident it's before the "almost surely" version--but I'm picking up Graham's intution a little more as I look at this longer.
This comment has been removed by a blog administrator.
I haven't seen anyone play with the meaning of 'beleive' yet. So, what if beleiving something just means that you live as if it were true, whether it is or not? Or perhaps you beleive something, and even if it's proved wrong, you retain it for a while until it's convenient to change your beleif. Granted, you would not expect someone to categorize their own beleifs as such.
Using the example, since you beleive it's raining, you wear a rain coat. When you get outside you realize it is not raining. It may be simply too much bother for the moment to take off the coat you just put on, so you leave it on for a while until you get a chance to take it off. In that case you could easily beleive something that you knew was not true.
Though that is probably where the example broke down, there are probably many people who unintentionally do this, or don't necesarily realize they do this. I don't actually know who Zwicky is, so I don't know whether this is a plausible behavior for him. Just a thought.
(I took the liberty of deleting the duplicate comment. Hope you don't mind. Hi Elizabeth!)
"believing something just means that you live as if it were true"
That's a pretty popular way of analyzing belief (C.S. Peirce is one philosopher who has defended something like it). It's appealing, because there's definitely something a bit weird about somebody who acts inconsistently with what they say they believe.
But let's think about that example a little bit. Bert thinks that it's raining, for one reason or another. So he puts on his raincoat. Then he goes outside and sees that no drops of water are falling on him. But he doesn't take off the raincoat. Do you think it's a normal use of the word "believe" to say "Bert believes it is raining" in this case? I wouldn't say that. I'd say something like, "Bert doesn't believe anymore that it's raining (but it's too much trouble to take off his raincoat)."
Of course, this is kind of a silly example, because there's nothing really inconsistent about wearing a raincoat on a sunny day. There are lots of reasons you might do that, without believing that it's raining (In fact, I did that just this morning, because I wanted to have my coat with me this evening). It would be harder in a case where there is a real inconsistency--a situation where there's no way you would do such-and-such if you really believed this-and-that--but it's actually pretty hard for me to think of a case like that. Maybe one of you can?
But if it's right (and that's a big if) this kind of action-theory of belief is a good way to try to make sense of Zwicky's comment. He would be saying something like, "I have a tendency to behave in a particular way, even though I know that behavior is inconsistent with the facts." We can all sympathize with that kind of irrationality--and what Zwicky is trying to describe here is fundamentally a kind of irrationality, an illusion.
But a here's another way of putting my basic problem with this sort of answer: statements of belief aren't just descriptions--they're commitments. When I say "I believe it is raining", I'm not just describing something about myself--my perceptions, or my actions, or anything like that. Besides what I may be saying about those things, I'm also committing myself to the fact that it's raining. Examining my behavior, or my perceptions, or my brain won't tell you anything about the rain.
Post a Comment
<< Home