Friday, June 24, 2011

Knowledge and too much evidence

I've been musing—nothing I care about rides on this—about this kind of case. Let's say that I have ten independent pieces of evidence that Socrates was killed by the Athenians. They give me knowledge, but it's more evidence than I need—any seven or eight of them would be sufficient. Unbeknownst to me, one of the ten pieces of evidence turns out not to work (maybe it's a complete fraud, maybe it's Gettiered, maybe it doesn't support the claim that Socrates was killed by the Athenians—there are all sorts of ways of filling this one, depending in part on what you think evidence is). The other nine pieces are fine.

Now, if you ask me how I know that Socrates was killed by the Athenians, I will give you all the ten pieces of evidence as my story. But it seems I'll be wrong: it is false that the ten pieces of evidence give me knowledge that Socrates was killed by the Athenians—at best, nine of them do that. Moreover, I am unable to give the correct story about how I know, because I don't know which of the ten is bogus. Do I know that Socrates was killed by the Athenians?

I think so. So it seems that it is not necessary for knowledge that I be able to give a story about how I know. In this case I know, but I don't know how I know. Nothing earthshaking here. But I thought it was an interesting case, and probably pretty common.

10 comments:

dogfreid said...
This comment has been removed by the author.
Jonathan D. Coffin said...
This comment has been removed by the author.
Jonathan Livengood said...

You write, "But it seems I'll be wrong: it is false that the ten pieces of evidence give me knowledge that Socrates was killed by the Athenians—at best, nine of them do that." I think there is an ambiguity that makes a pretty big difference to whether your assertion is reasonable.

Option 1: It is false that each piece of evidence by itself gives me knowledge.

Sure. That actually comes out by hypothesis, I think.

Option 2: It is false that the collection of all these pieces of evidence gives me knowledge.

Here I don't think I agree. If the nine genuine pieces of evidence are sufficient to secure knowledge and the tenth is not a defeater, then the whole collection secures knowledge for me.

Is there something I'm missing here? Maybe a concern about how much one has to know about the evidential relations among one's beliefs (or whatever)?

Eric S said...

Interesting...here's how I would see things.

I view knowledge as, in part, a set of truth claims. If I say, "I know X," I'm saying (or implying),

a. X is true
b. My belief that X is true rises to a level of certainty (it's rationally inescapable)
c. The evidence or support I have for X suffiently grounds that certainty
d. My belief that the evidences provide sufficient grounds rises to a level of certainty (it's rationally inescapable)

In reality, any of my claims could be true or false. I don't have absolute certainty that (abcd) are true (as from an objective perspective), rather I have virtual certainty (as much as is necessary to warrant a subjective claim of knowledge). The fact that I don't have God's perspective (omniscience) doesn't mean I don't have knowledge relative to human ability. I can't go beyond what's rationally inescapable.

X = Socrates was killed by the Athenians

Suppose I claim to know X (i.e. to be virtually certain that X is true), and I cite ten evidences to support my claim.

P = These ten evidences justify the claim, "I know X is true"

Now, if one of those evidences actually doesn't support my claim, then P is false. But I could just as well rephrase my claim.

P1 = Evidence A supports the claim "I know X is true"
P2 = Evidence B supports the claim "I know X is true"
...
P10 = Evidence J supports the claim "I know X is true"

Here, if P2 is false, I still have warrant for my claim. It isn't as if a faulty evidence equals no evidence at all.

Apart from this, however, who would know that P2 is false? It could be that no one (from a subjective perspective) could know that P2 is false, even if it was false. All (rational) people could accept P2 as self-evident. In that case, who would challenge my claim to know X? I can't both speak from my perspective and evaluate it from God's.

Belvy said...
This comment has been removed by the author.
dogfreid said...
This comment has been removed by the author.
Belvy said...

[Sorry, reposting with updated google profile.]

The question 'How do you know?' may be interpreted in two ways. One is the straightforward, every day sense of 'What makes you say that?' or 'What is your justification for saying that'. This is a call backing up one's claim in a typically non-rigorous or challenging way. Three, two, even one source may be sufficient to back up the claim involved in the original post. But the question, 'How do you know?', in its more technical sense, calls for a philosophical specification of actual, undefeated pieces of evidence. In the first sense, if nine pieces of evidence and one piece of nonevidence are presented as ten pieces of evidence, that tenth piece is superfluous due to the sufficiency of the other nine pieces. In the second sense, it renders your entire answer wrong, since your ten pieces do not all contribute to your possession of knowledge about the claim.

Belvy said...

[Again, reposting with updated google account.]

Jonathan, what I think you're missing is that if Pruss's questioner is asking for evidentiary support, then Pruss is, strictly speaking, mistaken if his response is to provide nine pieces of support and one piece that fails to actually support his claim. The only correct answer is to provide all, and only all, of his evidentiary support.

But as you and I have noted, this only comes into play if we take the question to be making this kind of request. Usually people who ask 'How do you know?' are not as interested in evidentiary exactitude as they are in evidentiary strength. Usually when nine pieces of evidence are advanced for a claim such as the one in the original post, nothing more is needed, even if the tenth piece is flatly false and obviously so.

Alexander R Pruss said...

I was thinking about the "How do you know that?" question as being a request for justification.

I guess a different kind of case to test this on is this?

A: I know that Smith is not to be trusted.
B: How?
A: Last year, I told him in confidence that we were considering hiring Jones, and Smith tweeted it in five minutes. Some time later, I had a look at Smith's dissertation, and noticed that he has a couple of pages that are taken word for word from an unpublished manuscript by Quine. Furthermore, Saturn has a solid core.

Suppose all of A's three claims are correct: Smith does tweet things he's told in confidence and he did commit plagiarism and Saturn has a solid core. Nonetheless, A's answer is incorrect. For that Saturn has a solid core is not a part of how she knows Smith isn't to be trusted.

Wes Salmon has a rather famous line, which I am not going to quote word for word as I don't remember it so exactly, that irrelevancies are harmless to arguments (adding true premises will never turn a sound argument into an unsound one) but fatal to explanations. The above case suggests that irrelevancies are bad for justifications, too.

Belvy said...

Nice points regarding irrelevancies. Another way to frame the issue is disjunctively. Logically speaking, one can always add a disjunct to a true statement and maintain the disjunction's truth, but metaphysically speaking, one cannot add a false disjunct to a true one and expect the disjunction to remain explanatorily accurate. An example of the former is adding 'water is dry' to any true disjunct, and an example of the latter is Dretske's disjunction problem.

But, again, it seems like a 'request for justification' can either be a call to provide the exact support one is relying on or the strength of support one is relying on. For what if you tell the person making the request that you have ten pieces of support but can only remember nine of them at the moment. Will only citing the nine make your answer incorrect? Not in cases where the request for justification is driving at the degree of strength of support.