The Black Eyed Peas on Assertion and Iterated Attitudes

So I’ve been spending time that I should be spending writing about disagreements about taste, fragmented belief, perception, and funny kinds of context dependence, thinking instead about a lyric from a Black Eyed Peas song (“Latin Girls”) that I was listening to while surfing the web writing about all of the very important topics that I ought to be writing about. Anyway, here’s the lyric:

“Girl, you know I know you know what I mean.”

And I started wondering (as one does):

(a) What’s the difference between the overall communicative effects of asserting,

(i) You know what I mean
(ii) You know I know you know what I mean

(b) Whatever the differences are, can you get an adding-to-common-knowledge view of assertion to predict them?

On (a): There do seem to be some, but I don’t feel like I’ve got my finger on what they are, exactly. They seem to be useful for either encouraging people not to engage in two different kinds of mock-ignorance, or else for giving people two different kinds of reasons for not engaging in the same kind of mock-ignorance. Or something like that.

On (b): Here’s the obvious reason to think not. On this sort of view, asserting either (i) or (ii) is going to add exactly the same stuff to common knowledge. You can’t make (i) common knowledge without making (ii) common knowledge, because as you iterate up the attitudes from (i), you’ll hit (ii), and then you keep going forever, just as you would if you’d started with (ii). And making (ii) common knowledge will also – given some pretty weak closure-under-dead-obvious-consequence conditions – make (i) common knowledge, on account of the factivity of knowledge. (Since you can’t know I know you know what I mean unless I know you know what I mean, and I can’t know you know what I mean unless you know what I mean. I’m sorry, that probably wasn’t the clearest way to say that, but it probably was the most fun way.)

A guess at what’s going on, due to Ben Blumson: The difference in effects is due to the fact that the upward iteration of attitudes probably does, in actual cases, stop after not too many steps.

Problem: Say just what the difference in effects of the two assertions is, and just which attitudes are bearing the load in explaining the effects in the two cases.

6 Replies to “The Black Eyed Peas on Assertion and Iterated Attitudes”

  1. It might help to notice that even a single iteration produces the problem. There seems to be a difference between asserting \‘You stole the cookie\’ and asserting \‘I know you stole the cookie\’. But, for the reasons Andy gives, there is no difference between making it common knowledge that you stole the cookie and making it common knowledge that I know you stole the cookie. So there appears to be a counterexample to the thesis that what is asserted is what is added to common knowledge.

    Nevertheless, I think the counterexample can be avoided, because in some contexts it will be possible to make it common knowledge that you stole the cookie by saying \‘I know you stole the cookie\’ but not by merely saying \‘you stole the cookie\’.

    Suppose, for example, that you believe I don\‘t know whether you stole the cookie. If I said merely \‘you stole the cookie\’ in this situation it might not become common knowledge that you stole the cookie, because if you believe that I don’t know whether you stole the cookie you may not be inclined to add my statement that you did to our common beliefs. So my assertion may be unsuccessful.

    On the other hand, if I say \‘I know you stole the cookie\’ then I address your belief that I don’t know directly and so I may be more likely to persuade you that I did steal the cookie. The difference between the two cases is then not that in the one case I asserted that you stole the cookie whereas in the other I asserted something else, but that in one case my assertion was successful but not in the other.

    The same point may be made about the difference between \‘You know what I mean\’ and \‘You know I know you know what I mean\’. There may be contexts in which it is possible to assert that you know what I mean by uttering \‘You know I know you know what I mean\’ even when it is impossible to assert that you know what I mean merely by uttering \‘You know what I mean\’.

    Another issue that might be relevant is that a few well-known examples suggest that what is asserted is not what is added to common knowledge, but what is added to what participants in a conversation act as if or pretend is common knowledge. This change is needed to accommodate examples like lying or obvious remarks about the weather.

    I\‘m not sure whether this subtlety would avoid the counterexample. But the point seems relevant because part of the difference in effect between the ordinary and iterated cases seems to be that the iterated cases debunk some kind of pretence.

  2. Here’s another suggestion for why this is different. Time-index the knowledge claims. Now “I know you stole the cookie” means that I knew at a time before the update to common-knowledge took place, whereas “you stole the cookie” doesn’t. The former indicates the status of some of my previous actions in a way that the latter doesn’t. And similarly for the song lyric that started all this.

  3. Well, I am thinking that the Monty Python skit won’t be nearly as funny with “You know I know you know what I mean. Wink, wink. Nudge, nudge.

  4. Yo Andy and Marc, what’s up!

    Hey Andy, what do you mean by “two kinds of mock-ignorance” eh?

    And if Williamson is right, “You know I know you know what I mean” is making a convention explicit, but in (i) it is not. But how does making a convention explicit explain why one has a reason not to pretend to be ignorant, that’s weird!

Leave a Reply