Consider the following sentence:

(1) I don't know that John kissed Mary.

When I assert this sentence, am I contradicting myself? The reason is as follows: following Stalnaker's view on the factivity of know, by asserting (1), I presuppose that the embedded sentence "John kissed Mary" is true. In other words, in asserting (1), I am in fact undertaking a prior commitment to the fact that John kissed Mary, namely:

(2) (It's a fact that) John kissed Mary.

But if that is the case, how the assertion of (1) is possible? If (1) is possible, then by conjoining (2) and (1), a Moorean sentence arises:

(3) %John kissed Mary, but I don't know that.

However, (3) is trivially odd. This rules out the possibility of asserting (1). But clearly (1) can be asserted and in fact we can find many sentences of the form "I don't know that ..." in ordinary discourse.

Does anyone have idea about the problem presented by the above examples? Thanks in advance!

  • 2
    Sentence 1 doesn't presuppose that "John kissed Mary" is true.
    – curiousdannii
    Commented Dec 24, 2020 at 2:15
  • 2
    You could analyse it as a different that, equivalent to whether, or a different don't know, equivalent to not sure, or you can say that the negation makes the whole thing mean it is not the case that I know that..., which would cancel your presupposition. I would have thought all of those possibilities are worth exploring, although I'm sure this will already have been done.
    – rchivers
    Commented Dec 24, 2020 at 3:29
  • @curiousdanii Indeed, one meaning of "that" here is "whether" (not commonly in my idiolect, but I'm aware of it). But it also can be the same "that" as in "I know that John kissed Mary," under which reading I think the OP's analysis is good. No doubt that's why when we utter such sentences, we mean the "whether" sense. Commented Dec 24, 2020 at 5:12
  • 1
    @user6726 Thanks a lot! I have read the second and third links that you recommended. I found them interesting but less formally rigid. I am happy to see the first link on the connection between the (non)factivity of know and focus effects. In fact, this connection has been detailedly discussed in Barker (1974). I think that's very useful. More can be said about this, such as Simons (2001) and Abusch (2002, 2009). But still, the analyses they give appeal to some ad hoc princiniples that are perhaps linguistically valid but mathematically obscure. I don't know. That's why I posed this question.
    – Fred
    Commented Dec 28, 2020 at 11:19

1 Answer 1


Could we be looking at this from a different or incorrect view, but I do not think you are contradicting yourself.

I read your comment about "I don't know that John kissed Mary." to mean...

You as a person, could have been asked "Why did John kiss Mary" you could reply " I don't know that John kissed Mary." Implying that you don't "know" that it happened only that it was a rumor and you are distancing yourself from the fact or stating you don't have proof that he kissed her.

"I don't know that John kissed Mary." could imply you don't know who kissed who, Maybe she kissed him?

"I don't know that John kissed Mary." could be thought of in the line of a detective reviewing evidence of a crime of passion, "Why did John kiss Mary" from evidence that his lips had lipstick and she is known for wearing lipstick of the same color. The detecive could be questioning what was previously thought as a known fact and now he is looking for supportive evidence "that John kissed Mary" because John could be a cross dresser which would give reason for why he had lipstick on. Or it could support a theory that Judy kissed John which is why he had the lipstick and he never did kiss Mary.

I'm reaching for different ideas but hopefully this helps stir the creative juices?

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.