Friday, August 9, 2013

Testimony and Credit

If one accepts testimony among the instruments of knowledge, then one is forced to reconsider one's idea of the role of credit within knowledge. In fact, in the virtue-epistemology understanding of knowledge, this can be "credited" to the knower. However, in the case of testimony, the credit seems to be little in the case of the listener and to rather regard the speaker ---who is not the "knower" of a testimonial belief.

Is there any possible way out?

This debate lies at the center of Jennifer Lackey's Knowledge and Credit (2009, available through Jstor at this link), and of a broader discussion possibly initiated by Lackey 2007. Lackey 2009 discusses the case of a tourist (called Morris) coming in Chicago and asking for directions. He happens to ask a Chiacago resident who gives him exact directions. Morris now knows where the Sears Towers are, although he deserves no credit for that!

—If one says that Morris does not know where the Sears Towers are, then one has to leave out of "knowledge" most cases of testimony.

—By contrast, one might try to say that Morris does deserve some credit (for instance, insofar as he asked a sober adult, instead of a child, an intoxicated person, etc.). However, Lackey explains, if this is the case, then credit should be granted also to "knowers" in Gettier-cases,
e.g., to Jack who has reasons to believe that his colleague John has 10 coins in his pocket and will get an advancement and as a consequence truly believes that someone in the room has 10 coins in his pocket and will get an advancement, although the one who has 10 coins and will get an advancement is Jack himself and not his colleague (you can find enough literature on Edmund Gettier, but since his article is only 3 pages long, the best way is just to read it, here),
although the Credit View of Knowledge should exactly aim at distinguishing knowledge (justified true belief) from beliefs which are only accidentally true (i.e., Gettier-cases).
Lackey shows how all attempts to reconfigure the role of Credit to meet the Chicago Visitor case are deemed to fail (or to include Gettier-like cases). Among these attempts are the idea (see E. Sosa 2007) that Credit can be shared (for instance, by speaker and hearer).

Unfortunately, Lackey does not spell out in her 2009 article the consequences of the dismissal of the Credit view of knowledge. Will this lead to a new paradigm (where true beliefs are knowledge, no matter whether they are justified or not)? Or will it just force epistemologists to find a new way to distinguish between knowledge and accidentally true beliefs?

The discussion has some special impact for scholars focusing chiefly on Indian philosophy and for readers of this blog in particular. One notices 1) the role of the listener within testimony, already highlighted on this blog (see below), 2) the conundrum implied in the case of testimony: Either one accepts only "safe" cases of testimony (notably the Veda, in the case of Prābhākara Mīmāṃsā), or the acceptance of testimony risks to lead to the acceptance of Gettier-like cases as well, 3) the similarity between virtue-epistemology and parataḥ prāmāṇya in placing the additional element of knowledge over true belief on the knower (and not on the piece of knowledge itself). Last, you might remember the articles by Sibajiban Bhattacharya arguing that in India (which meant, for him, Navya Nyāya) there is no need for a true belief to be also justified to count as knowledge. Gettier-like cases are also discussed and accepted as knowledge.

Does the Indian approach just deny the importance of the listener's competence? Does the burden only rely on the source of Linguistic Communication? And, more in general, do we need credit?

Further discussions on similar contacts between testimony and Gettier cases can be found in the Stanford Encyclopedia of Philosophy, here. On the role of the listener, see here and here.


Jayarava said...

I don't know the technical jargon but it strikes me that where you leave Morris he does not "know" where the Sears Towers are.

He has, at best, some decontextualised *information* about where the Sears Towers are likely to be (since he does not know from personal experience the layout of Chicago he cannot yet put the directions into a meaningful context). And presumably he has an evaluation of the trustworthiness of the informant - as a stranger he has only first impressions to go by. If he implicitly trusts the informant he may have a *belief* about where the Towers are. So we may only credit Morris with having untried information or an unsubstantiated belief. What does he know? He knows that someone has given him directions.

But he does not *know* until he tests the information. At best he believes, at worst he has some information of unknown value.

Information is not knowledge.

If we then asked Morris for directions to the Towers and ascertained the provenance of his information we would be dubious unless we could confirm it in some way. We would not credit him with knowledge unless we were incredibly naive. Personally I'd ask someone else.

An interesting variation would be to have Morris consult a published map. A published map is usually considered to be a highly reliable informant, and map reading is an acquired skill. Thus it might be said that Morris has some provisional knowledge if he is skilled at reading a map and has consulted a reliable map. One could attribute some knowledge to Morris in this example. But if I was Morris I would still not claim to know where the Towers are.

It seems to me that the discussion you are trying to have, you want us to believe that knowledge is an absolute - we either have it or not; and it is either 100% valid or 0%. But pragmatically this is untenable.

I don't see how "credit" in this context is even relevant. "Does Morris deserve credit" seems like a meaningless question. Even if we attempt to answer this question, why does it matter?

I can imagine trying to understand whether we can attribute knowledge to Morris (because I don't) but I can't imagine any real circumstances where I would be interested in crediting him or not. If Morris was *my* informant and someone asked me how I knew where the Towers were I'd have to say that I *do not know* where they are, but that Morris told me what someone told him about where they are. Am I therefore "crediting" Morris or attributing a statement to him? Credit is a term that seems to imply some value and I can't see what the measure of value is here.

elisa freschi said...

Dear Jayarava,

thanks for the comment, I was hoping for a different viewpoint on the whole debate. I do not know how much you are aware of it, but your position closely resembles the Buddhist Pramāṇavāda one, namely, that testimonially derived cognition does not amount to knowledge. As you say, only a cognition which leads to practical, testable, success (pravṛttisāmārthya) can be called "knowledge" according to the Pramāṇavādins. This position has some downsides, first and foremost that we end up with very little knowledge to base our lives on. From the moment we wake up, trusting the radio that it is (say) "5.15 am", to the moment we take the bus trusting the indications on the monitor and so on, our lives are permeated by information based on linguistic communications of various types.

One might well say that all these are not instances of knowledge and that, for instance, I do not really know what your surname is (after all, my only evidence about it is what you *told* me), nor that I was born on September (I only trust my parents' words about it), nor that my institute's director has a daughter and so on. However, one ends up with a very poor epistemology, which has little use in everyday life. It is, in short, too good to be useful.

Another interesting point you raise is that of degrees in knowledge. I tend to follow this approach of you, but I am not sure it would be accepted by Western and Indian epistemologists (after all, correspondentists insist that truth is correspondence with an outer world and a cognition X either corresponds to it or not, full stop).

As for credit, I agree with you that the point is not dealt with in detail in Lackey 2009. I tried to hint at the possibile significance of the notion of credit when I linked credit and justification and mentioned the fact that the idea of credit ---if true--- could be used to distinguish knowledge and accidentally true belief (the former could be credited to someone and the latter cannot).

Licenza Creative Commons
Quest' opera è distribuita con licenza Creative Commons Attribuzione - Non commerciale - Non opere derivate 2.5 Italia.