I recently received an interesting comment to my 2009 piece Moore’s Paradox and the Belief in False Things. I’ve decided to reply to it as a letter-post.
Here is the comment that AnH posted:
Thank you for posting this article (I guess this comment is a little out of place, as it is five years after the original publication date). The essay is thoughtful, but I personally would disagree with a number of your interpretations of the key ideas presented in the text.
First of all, Moore's paradox has very little to do with epistemology. The question at hand is not one about what knowledge is, but about the absurdity arising from asserting two contradictory statements. Wittgenstein's remark, namely that there cannot be a verb that means "to believe falsely," is simply intended to illustrate the difficulty presented in Moore's paradox, and not to claim that we cannot believe false propositions. If there were some verb V that meant "to believe falsely," then that verb could be inserted into a sentence of the form "I V p," where p is some arbitrary proposition. However, to believe is to take something to be true (perhaps this is where you think epistemology enters, though epistemologists are mostly concerned with belief as a condition of knowledge, and not as a subject of study in and of itself), and thus to state that you take something to be true and that you believe that what you think is true is false is something of a contradiction.
So the issue at hand is not that you can't believe false propositions, it's quite obvious that you can (many people falsely believed that the earth was flat for many years). The issue is that you can't believe one thing and simultaneously believe that what you believe is false.
On to verificationism. You state that verificationists assert that all unverifiable statements are false. However, verificationism is not a theory about truth and falsehood of statements or propositions, but about what kind of statements have meaning. On the verificationist view, unverifiable statements are not false, they are meaningless.
Lastly and most importantly, your claims about knowledge seem to contradict everything common sense tells us about knowledge. It would not be the case that if Elvis were discovered alive tomorrow that our knowledge of his death would cease, at that moment, to be knowledge. If we discovered Elvis alive tomorrow we would not say that it used to be the case that we knew that Elvis was dead and now we know that he is alive, we would say that we thought we knew one thing, and it simply turned out we didn't. "I thought I knew..." is a common phrase in English.
It is generally an axiom of epistemology that there are things that we can know. Epistemology dismisses skepticism and instead asks, "assuming knowledge is possible, what are the necessary and sufficient conditions that must be met for an instance of knowledge to obtain?"
Very few contemporary epistemologists are not fallibilists. To say that it is necessarily the case that we can only have knowledge if what we believe is true is not to say that we can only have knowledge of necessary truths, or that knowledge is an infallible mental state. Thus I feel your comments about human pride and our "unwillingness to admit that we believe false things" are perhaps a little unwarranted. In fact I can assert that I believe falsehoods without invoking Moore's paradox (as long as I don't make any such claims about specific beliefs).
Last comment: your solution of Moore's paradox seems to rely on a definition of belief that does not involve truth. But this definition, for you, follows from your claim that we believe unverifiable statements, and unverifiable statements are false. Like I said above, however, unverifiable statements are not false, they are meaningless. Thus unverifiable statements, such as moral claims, cannot be proper objects of belief. An emotivist may argue therefore that moral claims are essentially just sentences that are used to illicit certain kinds of responses from others. Of course verificationism seems to be a very strong theory of meaning in the sense that it keeps certain sentences that should be said to have meaning from having meaning, but it cannot be used to dispute an understanding of Moore's paradox.
My reply is as follows:
Hi AnH,
Many thanks for your detailed and thoughtful rebuttal of this post! Not at all a problem that it's coming ‘five years late’ - all material at this blog is always up for grabs whenever you want to talk to me about it. As is usual for older posts, however, my understanding of these issues has improved greatly since the time that I wrote it, and I accept much of what you say in your rebuttal. I would like to raise a few points of interest...
Belief and Knowledge
"(perhaps this is where you think epistemology enters, though epistemologists are mostly concerned with belief as a condition of knowledge, and not as a subject of study in and of itself)"
Yes, this is where I am asserting epistemology comes into play – I’m not certain how much being concerned with ‘belief as a condition of knowledge’ excuses the need to understand belief as a phenomena. It seems to me the latter is a prerequisite for studying the former! However, I am primarily a moral philosopher and a metaphysician and not an epistemologist. As you may appreciate, my specific beliefs make epistemology hard to commit to!
"The issue is that you can't believe one thing and simultaneously believe that what you believe is false."
This is the claim I dispute. But the nub of my disagreement is that the suture between propositional logic and belief fails, and once it fails the ‘law of non-contradiction’ ceases to have force. In order to understand belief (whether as a condition of knowledge or otherwise) it is necessary to see that it is not propositional in nature, nor is it a single phenomena. For one interesting thread of this discussion, see Tamar Gendler’s productive concept of ‘alief’.
"On the verificationist view, unverifiable statements are not false, they are meaningless."
You are right, of course. What I need to make my intended point is to stress not falsehood but epistemic relevance. Both false and meaningless statements are claimed to lack epistemic relevance (in two different ways). For the way I approach epistemology (mostly from the outside!), I can treat both of these dismissals as closely related. Other people will naturally approach this matter differently.
Some Flat Earth Pedantry
"(many people falsely believed that the earth was flat for many years)"
I believe it would be fairer to say that many people falsely believe that many people believed that the earth was flat for many years. I say this because if you dig into this mythology you'll find it is far older than most people give credit for, and was not (to give the most famous misunderstanding) a factor in Columbus' voyage. If you consider the number of humans that actually ever believed in a flat earth, it is vastly less than the number of people who believed that ‘many people believed in a flat earth!’
The Truth of Moral Claims
"Thus unverifiable statements, such as moral claims, cannot be proper objects of belief."
Moral claims are not unverifiable. The claim that they are unverifiable assumes an absolute yardstick against which they could be judged. But in fact, the only thing that we have that constitutes such a yardstick is logic (and mathematics, which Stephen Yablo demonstrates contains truth only because it is logically constituted). Consequentialist attempts to suture mathematics and morality have been nothing short of a disaster for moral philosophy (which thankfully consequentialists are recognising and fixing with a new focus on decision procedures). For more on this, see either Allen Wood’s rebuttal to Derek Parfit in the latter’s On What Matters, or my own forthcoming Chaos Ethics.
Moral claims can be verified provided their context is understood. In this respect, they are in a very similar situation to knowledge claims, including scientific knowledge claims, whose verification is always dependent upon the relevant theoretical background, a context that cannot ever provide an independent foundation to such claims. There simply isn’t an absolute yardstick beyond logic, and we thus run awry when we believe, along with Plato, that we can get out of the cave and discover information that can still be safely imported back into the cave. The cave is its own world! Stare at the sun as long as you like, you’ll never learn anything about the cave.
In this regard, I find a useful point of reference is Peter Lamarque's exploration of Wittgenstein's concept of a practice. Lamarque makes the point that the value of the queen in chess is objective within the practice of chess. Extending this argument, I would say that moral claims can be objective within the practices that define them – for instance, ‘taking steroids is wrong’ is a true claim for the practice of professional sportsmanship (anyone who disagreed with this claim is not participating in the relevant practice). The apparent ambiguity of morality comes not from any aspect of unverifiability but from the diversity of competing moral practices.
But this takes us a long way from Moore’s paradox!
Many thanks for the interesting points of discussion,
Chris.