Fallibilism is not self-contradictory

From the archives: Posted on 9th December, 1994

I had written that one of the things that seems to be necessary for the growth of knowledge is the assumption that one may be wrong in anything one says, to which someone responded:

“This statement is self-contradictory. It’s like the old no-rule-without-exceptions paradox.”

“Self-contradictory” is a term from logic. Statements of the form “this statement may be false” are not self-contradictory. To be self-contradictory, it would have to be of the form “this statement is false”. If one asserts “A” then one asserts that “either A or B is true”, then one is still allowing for the possibility of “A” being true. My statement that one should assume that one may be wrong in anything one says is not self-contradictory, because it allows for the possibility that one may not be wrong – that one may be right.

In logic, there are statements which are analytically or logically true, which can be proved through reasoning, without reference to the facts; and then there are statements which are logically false, which can be proved wrong just by pure reasoning, also without reference to the facts. These logically false statements are self-contradictory by definition. 

The contrary of a logically false statement is a logically true statement, so we can be sure that if we ever find a logically false statement, then the negation of it is logically true. So if you are saying that the statement “one may be wrong in anything one says” is self-contradictory, you are also saying that its negation is true.

The contrary of “one may be wrong in anything one says” is “one cannot be wrong in anything one says” So you are saying that the statement “one cannot be wrong…” is logically true.

It is not a logical paradox. The paradox comes from confusing “maybe” with “is”. If I say that I may be wrong, and that in particular, the statement that I may be wrong may be wrong, then I am admitting the possibility that I may be infallible, but if I am admitting that possibility, then that statement, that I may be wrong, may be false. If that statement is false, then it would be true, because I would have said something which is false, and so by definition I’d be fallible. So if it is true it is false, and if it is false it is true. The confusion is simply a result of confusing the word “maybe” with “is”. “This statement may be false” is not the same as saying “This statement is false.” If one takes this as a logical self-referential statement, then it is empty, but if one takes it as an epistemological statement, it is not at all empty. When one says “I may be mistaken” there is more to that than merely that “This statement may be false”.

“Does believing it to be absolutely true that one may be wrong in anything one says, make you blind to “any better ideas that do come up”?”

You are changing the emphasis of my statement. I was saying that I believe it to be true, but I am bearing in mind that I may be wrong about this. Being wrong about this would amount to the possibility that one is infallible. In other words, I am saying: “I believe I am fallible, but I could be wrong. It is possible that I am infallible.” “May be” is not the same as “is”.

“Or on the other hand, are you actually saying that it is possible for a person to say something without the possibility of being wrong? That is what is implied when you say that you could be wrong in saying that anything a person says can be wrong.”

So you are saying that my statement is either wrong, or it is asserting its own infallibility. It isn’t. It is asserting its own fallibility.

“This is an error that appears very often in everyday life, especially in religion and politics. (“I can’t stand intolerant people!”)”

That is a bit of a paradox, because it sounds as though one must be saying that one can’t stand oneself, but in logical terms, it is still not self-contradictory.

There is a philosophical point here, beyond this stuff about logic, about whether, when we say we are fallible, we are saying that the whole philosophy of reason might be false. You asked “Does believing it to be absolutely true that one may be wrong in anything one says, make you blind to “any better ideas that do come up”?” This is an example of over abstraction. What would it look like if I were blind to the other possibility?

Suppose I am arguing with my child about whether or not she should go to the ball. I say: “I don’t think you should go to the ball, but I may be wrong and I think that ultimately, this decision is for you to make.” And the child says: “You don’t think I should go to the ball, but you think you may be wrong; but you could be wrong about that. Have you considered the possibility that you may be infallible, and that therefore you would definitely be right about the ball, and I’d be wrong?” And I’d say: “Well, if I am infallible, then I am right that it is wrong for you to go to the ball. How does that seem to you? You can’t agree that I am infallible without also agreeing that I am right that you should not go to the ball. So do you agree about the ball?” If the child does not agree that she should not go to the ball, then she can’t agree that I am infallible either. On the other hand, if she does agree that she should not go to the ball, then she does not need to know whether I am fallible or not, because there is no disagreement!

This worry that one might lose something by favouring rationality is a mistake. Take the view that we are not infallible. It could be that everything we say is true. I might still be fallible even though everything I say is true. It might be an accident that everything I say is true. That does not imply infallibility. It might be coincidence. There is nothing in the statement “I am fallible” that prevents me from being right some of the time or even all the time, or on any particular occasion of taking into account the fact that I may be right.

Sarah Fitz-Claridge, 1994, ‘Fallibilism is not self-contradictory’, https://www.takingchildrenseriously.com/fallibilism-is-not-self-contradictory/