Sunday, September 25, 2011

Archers, abilities, beliefs, etc...

Consider the thesis that knowledge is just cognitive success from ability (KSA).

Pritchard and Whitcomb argue that KSA is mistaken and that the ability condition on knowledge does not give Greco and Sosa the anti-luck condition they need. Whitcomb offers an example that I think helps us see why KSA is mistaken:

Weights: Hoodlums at the shooting range put weights in most of the arrows’ tips. Champion archers shoot, and due to the weights they miss. Audrey gets to shoot, and by luck she gets the one quiver of unweighted arrows. Through skills that almost always bring target-hits, she makes those hits. Her shots are successful and, moreover, they are successful through virtue.

He says that this case is a case of practical success from ability and that it is analogous to Harman’s newspaper case. (Pritchard thinks that the case described earlier in which the targets are protected by force fields is analogous to the Ginet-Goldman fake barn case. ) Assuming that this is a case of practical success that manifests the agent’s abilities (which seems quite intuitive), we can state the Pritchard-Whitcomb argument against KSA as follows:

1. Weights and Fake Barns are analogous in all relevant respects.
2. Weights is a case of success from ability.
3. (Therefore) Fake Barns is a case of success from ability.
4. Fake Barns is not a case of knowledge.
5. (Therefore) Fake Barns is a case of success from ability without knowledge.

Because they defend KSA, Greco and Sosa have to reject (5) and reject one of the argument’s premises. Greco rejects (1). Sosa rejects (4). Let's focus on Greco's response.

Greco’s strategy for responding to the Pritchard-Whitcomb argument is to say that the practical and epistemic cases are not analogous. Abilities, cognitive or otherwise, should be understood as dispositions to reliably bring about some result under suitable circumstances. Cognitive abilities are understood as dispositions to bring about true beliefs in appropriate conditions and environments. Practical abilities can be understood as dispositions to bring about intended ends in appropriate conditions and environments. While Coop might have the ability to identify barns in, say, real barn country, Greco says that he does not have the ability to identify them in fake barn country. As such, judging correctly that the structure he sees is a barn is not success due to ability just as, say, Coop hitting a baseball blindfolded cannot be attributed to his ability to hit baseballs.
As for Audrey’s shot, he says this:

The ability to hit a target, like any ability, is defined relative to conditions that are appropriate for that sort of ability. In particular, we do not require that an archer is reliable (relative to an environment) in conditions involving arrow-weighting hoodlums. Accordingly, worlds where meddling hoodlums affect performance are not deemed relevant for determine whether [someone] has the ability in question, even if meddling hoodlums are in [her] actual environment and even … This is similar to Jeter’s ability to hit baseballs in Yankee Stadium – it does not matter whether there is some trickster in the stadium who could easily shut off the lights.

It is not clear to me whether our intuitions about the fake barns case is due to the fact that we think Coop does not have the cognitive ability to identify barns in the conditions or environment in which he forms his belief. We can describe the case that way, but it is not obvious to me that we have to in order to elicit the relevant intuition. Perhaps the best way to show that there could be cases of epistemic and practical success from ability that are analogous and that would threaten KSA is to focus on Weights and set the fake barns case aside.

Greco agrees that when Audrey hits the target she intends this is a case of success from ability. While he thinks that the Audrey’s shot is not analogous to Coop’s belief, I think he would think that if Audrey had beliefs about whether her arrows will hit her intended targets that these beliefs will be analogous to Coop’s beliefs. So, let’s think about Audrey’s beliefs in Weights. In drawing back the string, she thinks to herself that if she aims just like this while pulling the string back just like so that she will hit her mark if she releases the string. If Audrey truly is a skilled archer it might be that she would readjust her shot if she did not believe this.

I want to ask two questions about her means-end belief. Her means-end belief is correct in the case we described. Is it a case of cognitive success from ability? Is it a case of knowledge? To keep things clear, consider three propositions:
6. Audrey’s hitting the target is successful through her ability.
7. Audrey’s means-end belief is successful through her ability.
8. Audrey’s means-end belief constitutes knowledge.

My intuition, which I think Pritchard and Whitcomb share, is that while (6) and (7) are true, (8) is not. So, like the, I think we could use this case to show that KSA is mistaken without having to say that the case is analogous to the fake barns case or the newspaper case. I expect that Greco would agree that (8) is mistaken. Given his commitment to KSA, he has to reject (7). I think he should accept (6). Not only is it intuitive, his objection to the Pritchard-Whitcomb argument was that the cases are not analogous. Since he rejects (1), he should concede that there can be cases of practical success from ability where the agent’s actions are guided by belief’s that are not an example of cognitive success from ability. Perhaps the four of us can agree that her belief does not constitute knowledge because she easily could have drawn a weighted arrow, in which case her belief would have been mistaken and she would have failed to bring about her intended end in the way she intended. The problem with the position that I think Greco has to defend is that it seems that (6) would only be true if (7) were true.

If Audrey’s shot truly is a case of success from ability, there should be no principled objection to saying that she hit her intended target intentionally. Assume she did hit her target intentionally. Let’s think about how things have to be for her to hit her intended targets intentionally. If things had gone differently and she had grabbed a weighted arrow, she would have had a false means-end belief. If she had a false means-end belief, one of two things would have happened. She would have missed her mark or she would have hit her mark without hitting it in the way she intended to. Either way, she would have not hit her mark intentionally because she would not have had the necessary control over whether her intended ends would be realized.

While you will not intentionally Φ by Ψ-ing when you intend to Φ by Ψ-ing and believe falsely that you will Φ by Ψ-ing, you do not regain the necessary control simply because you happen to have a true means-end belief. Suppose Audrey grabbed a weighted arrow, fired, and the arrow cut just the path towards the target that she intended. Suppose it did this because the arrows were weighted with iron and a powerful magnet was placed just behind the spot that Audrey intended to hit. She hit what she intended to hit, her means-end belief might well have been true, but she knew nothing about the magnet and she did not intentionally hit her mark.
What does it take to put Audrey back in control if not a true means-end belief? Suppose Audrey needs to get into the safe, believes for no reason whatever that the combination to the safe is 12-34-56, and she unlocks the safe acting on her baseless belief. We would not say that she intentionally punched in the correct combination. Too much luck was involved. We might say that if she has a correct means-end belief that is rationally held we can deal with this case, but there are more cases in the offing that cause trouble for this proposal.

Suppose Audrey had excellent evidence that the combination was 12-34-56. Coop told her that this was the combination on Monday. Unfortunately, Coop was mistaken on Monday when he said this. On Monday, the combination was 12-34-55. The note that he fished from Ben’s coat pocket with the combination was smudged. Fearing that Coop discovered the combination to the safe Ben changed the combination moments before Audrey entered his office to 12-34-56. She punched in the correct code believing reasonably and correctly that the code was correct, but it still seems she did not intentionally enter in the correct code. The connection between what she intended to do and what it would take for her to bring about what she intended seems too accidental for us to say that she had the right kind of control to have brought about her intended end intentionally.

What do we need to add so that the agent who acts on a means-end belief brings about her intended end intentionally? Knowledge would do the trick. If Audrey knew that she would hit the target by firing in just the way she did, she would hit her target intentionally. If she knew that by typing in 12-34-56, the safe would open, we could say she intentionally entered in the right combination. If nothing short of knowledge will do the trick, however, Greco’s defense of KSA is bound to fail. If nothing short of knowledge would do, we would have to say that (6) is true in Weights only because (8) is. Remember that Greco rejects (7) and (8). Thus, he needs to explain how Audrey hitting the target could be a case of success from ability if the correctness of the means-end beliefs that guided her actions were not themselves a manifestation of her cognitive abilities. I think it cannot be done. When we think that the correctness of the agent’s means-end belief is not an instance of cognitive success through ability, we will be disinclined to say that the actions guided by these beliefs will be cases of success from ability. Thus, it seems (6) and (7) stand or fall together.

4 comments:

Aaron Boyden said...

Hmmm. I'm sympathetic to Greco's response, but I'm having trouble formulating my thoughts about your criticism. I guess I'm not sure Audrey has an ability to judge where her arrow would hit under circumstances like that. She judges successfully because she has an ability to judge where her arrow would hit under normal circumstances, and by luck an inappropriate application of that ability to her present abnormal circumstances happens to work (that is, 8 is false because 7 is false). But I don't feel inclined to make similarly fine distinctions when it comes to the ability to hit the target (so 6 is still true). I suppose I must then think 7 and 6 can't involve the same ability. I don't know if that shows that I'm inconsistent (nor do I know what I should revise if I am), but I have some inclination to think that perhaps knowledge just requires that more fine-grained distinctions among abilities be made, distinctions not appropriate to some other contexts where abilities are relevant.

Clayton said...

Hi Aaron,

So, one idea that I've been toying with that might be one you're sympathetic to is that the cognitive abilities that belief must result from in order to guide skilful and intentional action are distinct from the cognitive abilities that you must exercise in acquiring knowledge. We can thus remain agnostic as to whether some version of KSA is right, because the main point is that there's some perfectly good notion of success from ability that (i) falls short of K in environmental luck cases but (ii) must be met if the actions the belief guides are intentional.

For what it's worth, Sosa I think would say that (6)-(8) are correct and so would, I think, be sympathetic to my idea that you can't simply deny (6) if you want to defend (7). My main objection to Sosa's response is that he thinks (8) is true and I don't, but my intuitions about (8) are not terribly firm.

Anonymous said...

Hi Clayton,

I was interested to learn that Sosa thinks there is knowledge in the fake barn cases. Would you mind telling me where he says this?

Clayton said...

Hi Anon,

This is from a footnote on pp. 96 in Sosa's _A Virtue Epistemology: Apt Belief and Reflective Knowledge_ Volume 1:

"Our account does help to bring out, however, how not all Gettier cases are created equal. In some cases, such as Gettier’s two actual examples, and such as Lehrer’s Nogot/Havit case, the subject does not attain so much as animal knowledge: apt belief, belief that gets it right in a way sufficiently attributable to the exercise of a competence in its proper conditions. However, in other similar cases what the subject lacks is rather reflective knowledge. Our kaleidoscope perceiver, in Lecture 2, is a case in point. The Ginet/Goldman barns example arguably belongs with the kaleidoscope case."