Why does it seem wrong to you? Maybe you can expand a bit to help get the conversation going.

If "to know that one knows" means "to be conscious that one knows" (which is how I would interpret it), then the principle would apparently deny the possibility of unconscious knowledge. Which, as the OP says, seems wrong.

If there is some other reading of the meaning of "knowing that one knows", I may be persuaded otherwise.

The question amounts to something like:

Is our knoweldge to us, in all ways, transparent? Why would we say we hold a belief that is a justified true belief, if we haven't reflected on it? With unconscious "beliefs", would we be better off with calling them casual dispositions instead of justified beliefs?

The KK principle says that, for any proposition p, if one knows that p, then one knows that one knows it. — Marty

Let's apply the justified true belief (JTB) standard, which is a bunch of bullshit, but easy to apply here. In order to know that I know p, the following must be true:

• It is true that I know p - that's given
• I believe I know p. By JTB, if I know p, I believe p. Is it possible to believe p without believing I know p?
• I am justified in believing I know p. By JTB, if I know p, I am justified in believing p. Is it possible to be justified in believing p without being justified in believing I know p?

I was wrong. This isn't helpful at all.

I was right, JTB is just a bunch of bullshit.

Is our knoweldge to us, in all ways, transparent? Why would we say we hold a belief that is a justified true belief, if we haven't reflected on it? With unconscious "beliefs", would we be better off with calling them casual dispositions instead of justified beliefs? — Marty

It's mostly a matter of how we decide to the use the word knowledge. A knowledge externalist would deny that one needs access to any justification. It's a short putt from there to accepting that knowledge can exist where there is no self-consciousness at all.

Knowledge internalism comes closer to the way we usually speak of knowledge. It has its own challenges.

The KK principle says that, for any proposition p, if one knows that p, then one knows that one knows it.


I think this principle is implicit in a lot of philosopher's epistemological foundations, but seems wrong. — Marty

Does knowing that one knows something imply absolute certainty, like really, really knowing it? How do you know Paris is the capital of France, do you really, really know it? The ancient skeptics held that to know something you need a criterion, but then how do you know the criterion is right? You need a criterion for the criterion. In other words, how do you know, how do you know that you know, know that you know that you know, and so on?

This approach seems hopeless, and as Spinoza said in response to this problem: in order to know that you know; you must first know. So, as someone else said the JTB model is "a bunch of bullshit". That's the way to disappear up your own arsehole in true anal-ytic fashion. :wink:

Does knowing that one knows something imply absolute certainty, like really, really knowing it? — Janus

Not certainty, just recognition of a relationship between the self and the proposition. It's a wild goose chase no matter what your ontological commitments are. Endless philosophy fun.

That it's my proposition, and I can cry if I want to? Fun? Yes... :lol:

You know it! :D

"S knows that P" -> "S knows that S knows that P."
"S knows that S knows that P" -> "S knows that S knows that S knows that P"

This seems like such a basic point that I'm sure proponents of the KK principle have thought of it, but what's the reply? How does this not imply that, in order to know something, I have to know that I know that I know that I know... ad infinitum? And what would that mean? If it's supposed to be intelligible that I can know that I know something, then the whole regress should be intelligible, right?

I suspect that this misses the point somehow, that someone who holds the KK principle would mean something totally different by it. Is the second K a different kind of knowledge from the first K?

It definitely depends on how we define knowledge.

Do you know how to tie your shoes? How do you prove it? You tie your shoes.

Is knowledge something that we can possess without being aware of it? You weren't thinking about how to tie shoes just before I mentioned it, but now that I mentioned it, you become aware that you know how to tie shoes. You know you know because you've done it and you can do it now. Just untie your shoes and try it. Do you know how to do it? Did you know how to do it before I mentioned it - before you just now became presently aware of it?

It seems to me that knowledge can be something that we are not presently aware of at any given moment. We access it when we need it. We aren't aware of all we know at any given moment. Therefore knowledge must be something like storage, or memory that contains information that we consciously access when needed.

Whether or not you believe in the KK principle depends on your notion of knowledge as others have said, which can be tricky since the idea of what constitutes knowledge is tricky. People say that knowledge involves a justified true belief, but what does it mean to believe something and what does it mean for something to be justified?

From what I have found, there are many different variations of belief. One can say one believes in something if one is disposed to affirm it, or one can also define it as a conscious affirmation in thought for instance. Depending on how flexible one is with belief, then the KK principle may either be obviously true (as in the former) or false (like in the latter). As for justification, the Gettier problems show that what properly counts as justification is also unclear. You may have reasons for believing in something that turns out to be true that is completely irrelevant to the truth of that something. Though that may not be an issue in this case though since one is already directly acquainted with the fact that they know something to base their meta-knowledge on, which just leaves the issue of belief.

"S knows that P" -> "S knows that S knows that P."
"S knows that S knows that P" -> "S knows that S knows that S knows that P"

This seems like such a basic point that I'm sure proponents of the KK principle have thought of it, but what's the reply? How does this not imply that, in order to know something, I have to know that I know that I know that I know... ad infinitum? And what would that mean? If it's supposed to be intelligible that I can know that I know something, then the whole regress should be intelligible, right? — Pneumenon

My feeling is that the problem is either solved through semantic collapse, i.e. "S knows that P" means exactly the same as "S knows that S knows that P" (similarly to how, say, "It is true that P" means exactly the same (?) as "It is true that it is true that P") - or else it cannot be solved and the KK principle is false. Which one it is will turn on the concept of knowledge used.

Okay, but that makes the KK principle vacuous.