knowledge, justification, and assertion
I’m sure you’re all dying to for more about ought, but for now I’m going to let you continue to die. This post is about knowledge and assertion (and I’m betting Clayton has already had some of these thoughts). I want to argue that being committed to the following three positions is a sorry state to be in:
knowledge-assert: One may assert
if and only if one knows
justified-not-knowledge: One can be justified in believing
belief-assert: If one is epistemically warranted in believing
I take it that most defenders of the Knowledge Account of Assertion (KAA) want to be able at least to accept Knowledge-Assertion and Justified-Not-Knowledge. Williamson makes it clear that Justified-Not-Knowledge is compatible with E=K+his account of justification. Moreover, he seems to think that it would be a bad thing if he was forced to give up Justified-Not-Knowledge. I also think that Hawthorne and Stanley (e.g.) want to hold both Knowledge-Assertion and Justified-Not-Knowledge (although, I should say that I’m not as familiar with those two books as I am with Williamson).
Belief-Assert is also very intuitively plausible. Imagine what would make it false: A case in which it is perfectly within your epistemic rights to believe
Here’s why it’s a sorry state to hold all three positions: If you hold Justified-Not-Knowledge, it’s possible for someone to be epistemically warranted in believing
There are several fairly common moves one might employ to get rid of this. One might make some type of objective/subjective distinction about ‘warranted’ and say that one is objectively warranted iff one knows and one is subjectively warranted iff one is justified (or make the distinction between objective ‘may’ and subjective ‘may’). I’m skeptical that will work, but encourage people to try.
Barring arguing that there is an equivocation, the only other way to get rid of the problem is by giving up one of the three views. I take it that most who hold Knowledge-Assert will want to give up Justified-Not-Knowledge (‘Sutton’ it, as it were). But that means that justification=knowledge. That means that that Gettier guy was all wrong! Surprising result.
I say let’s give up Knowledge-Assert.
I say give up on Knowledge-Assert, too.
Errol,
Thanks, I have had similar thoughts before–but only because I read them from pages written by someone else. In his recent book, Without Justification, he argues from the inconsistency of the above that we ought to identify knowledge and justification. He thinks that the case for knowledge account of assertion is pretty strong, far stronger than the case for distinguishing K from J. Like you, I think he thinks that the assertion/belief link ought to be treated as non-negotiable.
But, you’re right that W is not keen on identifying K and J. It would be interesting to know what he thought the proper response to this is. I think there’s a forthcoming anthology in which he might address a critic who raises a worry in this ballpark (if I’m not mistaken Kvavnig defends the J account of assertion and insists that J and K are distinct). Unfortunately, my library fines have reached a point where I can’t get the book through interlibrary loan. My research is all apriori these days.
[...] Knowledge, justification and assertion. Three positions Errol feels we can’t simultaneously be committed to. Like him I think we ought throw out the idea we ought only assert what we know. [...]
Knowledege, Justification and Assertion : Mormon Metaphysics said this on August 15, 2008 at 8:37 am |
“Some also argue (e.g. Jon Kvanvig and Kent Bach) that the fight about the norm of assertion is really just a fight over the norm of belief. Obviously Belief-Assert falls out of that.”
I don’t find this obvious at all. On its most (only?) plausible interpretation the belief norm states a necessary condition on proper assertion, but not a sufficient condition. The picture is supposed to be that one ought to assert only what one knows, and any evidential norms on assertion are thereby derivative from such norms on belief. We don’t need to posit a separate knowledge norm for assertion, since all its work is already taken care of. If the belief norm specifies a sufficient condition for proper assertion, then one can properly assert something even if one is utterly ill-informed, unjustified, epistemically irresponsible, etc. I don’t think that’s what defenders of the belief norm had in mind. (Certainly I haven’t found anything in Bach that suggests he had a sufficient condition in mind, though I could be corrected on that.
In general, I think more needs to be said in defense of belief-assert. I’m far from a fan of knowledge-assert, but I don’t find the idea that warranted assertion is more demanding than warranted belief at all disturbing. I’d like to see more on why we should regard belief-assert a cost.
From my comment above:
“The picture is supposed to be that one ought to assert only what one knows, and any evidential norms on assertion are thereby derivative from such norms on belief.”
Obviously that isn’t the picture at all. I meant ‘only what one believes’.
Aidan,
I’m slightly confused about what you are objecting to. I take you are objecting to the following thought: Belief-Assert directly falls out of the picture that holds that belief is the constitutive norm of assertion, but knowledge/truth/justification is the constitutive norm of belief. But you say, if this is right, then one can comply with the norm of assertion iff one believes what one is asserting–even if one has no epistemic warrant for holding that belief. Granted, that person would be violating the norm of belief, but not assertion. Part of your objection is that this isn’t what the defenders of the ‘push-back’ view have in mind. Maybe you’re right about Kvanvig, but I think Bach is pretty clear about this in ‘Applying Pragmatics to Epistemology.’ He writes, ‘it seems to me that the only relevant rule on assertion is belief, since an assertion essentially is the expression of a belief…the knowledge rule of assertion has no independent significance. It is the combination of the knowledge norm on belief and the belief rule on assertion’ (pp. 10-11).
I don’t find the idea that warranted assertion is more demanding than warranted belief at all disturbing.
Here’s a reason why it might be disturbing. You might think that if you are warranted in believing
, then you are warranted in using 
in deliberation. And you might think that there is an analogous role for assertion. Warranted assertion directs inquiry in the proper direction (I guess if you think warrant isn’t factive, then it doesn’t necessarily lead inquiry in the proper direction). It, to use Stalnaker’s terminology, eliminates the possibilities one is warranted in eliminating. So, if this is right–and to me these seem like principles that it is hard to deny–then it is hard to see why it would be permissible for you to eliminate possibilities in your head but not permissible for you to eliminate possibilities out loud.
Well, you may be right about Bach. If so, the implicit argument you seem to be offering him seems pretty bad; assertions are essentially the expressions of beliefs, so an assertion is proper iff one believes what one asserts. Maybe I’m just missing something important, but I don’t see how the premise supports the conclusion.
In any case, there’s no need to take the belief account to be committed to the sufficiency claim. All the explanatory work the knowledge account does is done by the claim that one asserts properly only if one knows, and so if belief is necessary for proper assertion, and knowledge is necessary for proper belief, that’s enough to make the knowledge rule superfluous. I’m still not seeing any good reason to add on the claim that belief suffices for proper assertion, and it looks like that’s the claim that would deliver belief-assert.
One cluster of reasons to think the out-loud cases are different concerns the relationship between assertion and the transmission of knowledge through testimony. Defenders of the knowledge account have discussed this; Williamson has a short discussion in the section on the point of having a speech act that is constitutively governed by the knowledge rule, and Reynolds goes to town on it, seeming to make no distinction between asserting and testifying. So here’s a vague thought; the standards for assertion are more demanding than the standards for belief because of assertion’s role in the transmission of knowledge through testimony. Vague or not, presumably something like this thought often drives rejection of belief-assert, and something in this region strikes me as plausible.
Lottery cases seem to me to be prima facie counterexamples to Belief-Assert. I know that my friend’s holding a ticket in a million-to-one lottery, and that the drawing was held earlier this evening, though neither of us has heard the result. It seems that I’m highly warranted in believing that he’s lost the lottery. However, there’s something (apparently epistemically) wrong with my asserting to him, flat-out, “Your ticket lost the lottery.”
Aidan,
I’m still not seeing any good reason to add on the claim that belief suffices for proper assertion, and it looks like that’s the claim that would deliver belief-assert.
I might be getting lost in the dialectic again, but I take this claim to mean that the ‘push-back’ strategy gets Belief-Assert only if belief is sufficient for (epistemically) warranted assertion. This seems false generally. Belief-Assert is stronger than Belief-Assert*:
belief-assert*: Necessarily, A is epistemically warranted in asserting
iff A believes 
.
Obviously there are cases where one makes a warranted assertion if Belief-Assert* is true but doesn’t make a warranted assertion if Belief-Assert is true. Just plug what your theory is of epistemically warranted belief, and find a case where (1) A believes
and (2) A fails to meet the standards for epistemic warrant. The existence of a case like this shows that Belief-Assert doesn’t imply that belief is sufficient for epistemically warranted assertion.
As to your vague suggestion. I think it is a weird position to hold that warranted testimony is equal to warranted assertion and also to hold that the transferral of knowledge through reasoning isn’t equal to warranted reasoning. Think of it this way: If you had the intuition that the first view is true, why wouldn’t you have the intuition that the second view is true? I really don’t get it.
Geoff,
Thanks for commenting! It’s always nice to get new commentators.
I don’t trust my intuitions about lotteries any more; but insofar as I have the intuition you are warranted to think that the ticket will lose, I have the intuition that you can assert that it will lose; and insofar as I have the intuition that you cannot warrantedly assert that it will lose, I have the intuition that you cannot warrentedly believe that it will lose. I simply cannot pry those intuitions apart.
But if you have an argument for why I should, then I am eager to hear it.
Errol (not Error) and Geoff,
I think I’m with Errol wrt to the significance of lottery cases to belief/assert. There’s a nice paper by Matt Weiner arguing that there’s a story to tell about why you oughtn’t assert lottery propositions and if I recall from a conversation he and I had (I have a terrible memory) he doesn’t think that lottery beliefs violate the norms of belief. I’ve argued that it’s unlikely that that you can show that lottery assertions violate the epistemic norms of assertion unless the beliefs violate the norms of belief, but I think the issues are interesting. I’m curious–do you think there’s a good principled explanation as to why we oughtn’t assert lottery propositions?
Hi Clayton and Errol,
I’m working on a paper now where I argue that there’s an epistemic norm governing assertion with the consequence that your assertion of p is improper if you would have asserted p if p had been false (there are some bells and whistles but that’s the basic idea). So that’s my account of what’s wrong with (most) assertions of lottery propositions. The proposed norm does good work above and beyond explaining the manifest impropriety of lottery assertions. It also serves as the basis for an explanation for the (mistaken!) intuition that we don’t know lottery propositions. And, I argue, we’d expect a norm like this one if testimony were a basic source of knowledge, which of course it is. Someday this paper will be finished.
But my preferred explanation aside, lottery assertions are obviously improper; the only question is whether the impropriety is due to their being epistemically unwarranted or something else. The terms we naturally employ to criticize such assertions (“You don’t know that!” “How can you say that for sure?” “Hey, there’s a chance I might have won!”) strongly suggest the former. (I’ve read Matt W.’s Gricean explanation, and it’s heroic but I’m not convinced.) It seems the default position wrt lottery assertions has to be that they’re epistemically unwarranted.
But the case with lottery beliefs seems to me to be very different. If there are instances where P(P|E) = 0.9999999, I believe P on the basis of E, have no defeaters, etc., but am nonetheless not warranted in believing P, well that would be surprising, wouldn’t it? Again, it seems to me that the default position with lottery beliefs is that they’re warranted (though I do admit that this is not so crystal-clear to me as the impropriety of lottery assertions).
But of course this is only my sense of how things stand before you start giving arguments. I could be talked into giving up one of these default positions if I came across a compelling enough argument for belief-assert. (Can you point me to the argument you mentioned, Clayton?) But Errol didn’t give such an argument; rather, you posed a challenge to imagine “a case in which it is perfectly within your epistemic rights to believe p but not within your epistemic rights to assert p” — when I imagine this case, it seems to me that I’m imagining just such a case!
Hey Geoff,
I think Sutton thinks he’s given an argument for belief-assert parallels (if you (epistemically) oughtn’t assert p, you (epistemically) oughtn’t believe p) in his _Without Justification_, but I can’t recall offhand how his goes. Myself, I’m somewhat put off by what strikes me as the odd suggestion that there would be two levels of epistemic scrutinizing, the first of which determines whether one may believe p and the second of which determines whether one may assert that one’s belief is correct. (I could imagine additional constraints on assertion where epistemic standards are raised for the purposes of the conversation, but that seems consistent with what I’ve just said.) Those who think there is this additional scrutiny could help me by asking what questions are left open when one settles the question as to whether to believe p in asking whether one may assert it (or, what it takes to close the questions). Like I said, I have nothing compelling here to offer.
The thing about the lottery cases is that is sort of see this:
If there are instances where P(P|E) = 0.9999999, I believe P on the basis of E, have no defeaters, etc., but am nonetheless not warranted in believing P, well that would be surprising, wouldn’t it?
But, I also feel the force of this:
If there are instances where P(P|E) = 0.9999999, I believe P on the basis of E, have no defeaters, etc., but am nonetheless not warranted in asserting P, well that would be surprising, wouldn’t it?
Like you, I’m worried about Weiner’s story. I’ve argued that if the lottery belief wants for nothing epistemically, the Gricean story he tells won’t work because asserting lottery propositions would serve a legitimate conversational purpose. If, however, lottery beliefs are epistemically improper, the Gricean story seems otiose because it seems that we oughtn’t assert epistemically impermissible beliefs. (I think that’s something you’d agree to, it’s just the other direction that’s tricky.) In general, I worry that if there’s nothing wrong with believing p, there’s nothing wrong with asserting p for the consideration of others as well. If they believe on my say so, more power to ‘em. If they don’t, well, their loss.
[...] justification, and action In my previous post, I discussed the relationship between the knowledge account of assertion, a non-factive account of [...]
knowledge, justification, and action « The Excluded Middle said this on August 25, 2008 at 12:35 pm |