@ACB,
ACB;111995 wrote:Can he know this even if he does not know the meaning of the terms "modus ponens" and "inference rule"?
If he knows that the premises P, and if P then Q, entail the conclusion Q, then I don't think it would be incorrect to say that he knows that modus ponens is a valid inference rule.
I think a lot of the potential for worry about this kind of thing lies in the difference between the linguistic and the logical. Say for example that it is true that S knows that Mount Everest is the highest mountain in the world; it also seems reasonable to say that S knows that Qomolangma (the Tibetan name for Everest) is the highest mountain in the world. From a logical point of view, both statements of knowledge seem to say the same thing, that S has a justified true belief that a certain relation holds between one particular mountain, and all of the other mountains. The 'particular mountain' being the one that is referred to by the two different names. We might say that names are the grammatical subjects of expressions, but particular objects are the logical subjects of expressions. On the other hand, S may have no idea that Qomolangma is the same thing as Everest, so if we were to ask S "Is Qomolangma the highest mountain in the world?" he might reply "No, Everest is."
My inclination is to say that S does know that Qomolangma is the highest mountain in the world, reasoning as follows:
Premises
S knows Everest is the highest mountain in the world.
Everest = 'Qomolangma'
Conclusion
S knows Qomolangma is the highest mountain in the world.
The worry arises because S does not know that Everest = 'Qomolangma'. This seems like a circumstance where it might be appropriate to say S does not know that he knows that Qomolangma is the highest mountain in the world, while in the original case it would make sense to say that one did not know that one knew that modus ponens is a valid inference.
This seems to raise an issue with iterated knowledge. If we use the above reasoning, which seems to be sound, to say that S does not know that he knows that Qomolangma is the highest mountain in the world; it still seems possible for S to know that he knows that Everest is the highest mountain in the world. This would certainly seem to be the case in the following exchange:
Q.
Is Qomolangma the highest mountain in the world?
A.
No, Everest is.
Q.
Do you know that Everest is the highest mountain in the world?
A.
Yes, I do know this.
However, in this case, there also seems to be a sense in which we may also say that S knows that he knows Qomolangma is the highest mountain in the world. The logical sense, in which we mean that S has a justified true belief that his belief about a relation that obtains between a particular mountain (the one that is picked out by the names 'Everest' and 'Qomolangma') and all the other mountains is true and justified.
Apparently the rules governing how we may use the word 'know' are very very subtle.