Thursday, October 30, 2008

Socratic ignorance

Okay, I’m going out on a limb here...

I’m not an Ancient Philosophy scholar, nor do I claim to have any special insight on Socrates/Plato.

But I can’t resist musing about Socrates on this fine, unseasonably warm Fall evening in late October, especially inspired by some thought-provoking conversations with my colleague Eric Snider (our local ancient philosophy aficionado).

Socratic Duplicity

Socrates is famous for claiming that he does not posses knowledge. Does he mean what he says?

Most say that Socrates is not being sincere. He is being duplicitous with good intentions. By feigning ignorance, he is trying to inspire his conversation partners to pursue after the truth with all that much ardor. In order to get them to pursue the truth, however, Socrates must knock out their false foundations of confidence.

I’ve heard (through the philosophy grapevine) that there are pretty unambiguous texts in Plato that suggest this interpretation is not right.

Merely True Belief

The other main interpretation is to deny that Socrates is playing ignorant. This would mean that Socrates is being sincere when he claims that he lacks knowledge. What he has, however, is still pretty good. He possesses true belief. This falls short of knowledge, but it’s better than nothing.

This would require that we view the pursuit of Socrates as a search for true beliefs but not knowledge. This doesn’t sound quite right either. One thing that I think is true of Socrates is that he equated virtue with knowledge. If Socrates must be taken to have given up on the project of knowledge, then he must also be said to have given up on the project of virtue and happiness. For a guy like Socrates, this is tantamount to saying that he has given up on the hope of becoming good (which for him was the task of philosophy). That sounds distinctly un-Socratic.

The Problem

On the one hand, Socrates’ claim of ignorance is sincere. He is not pretending when he claims that he lacks knowledge. On the other hand, Socrates claims that he pursues after knowledge. He is not satisfied with true belief. Socrates is not only dissatisfied with true belief, but he actually takes himself to possess knowledge.

Socrates the contextualist?

It’s possible that claims to knowledge might mean different things in different contexts for Socrates. Does this help illuminate what’s going on with the alleged Socratic duplicity? Maybe... but I’m just musing out loud (or whatever the equivalent is when applied to free-flow typing).

Think about the way we ordinarily use language, especially the word “know.” I drove my car to campus today. If someone asks me about the location of my car, there’s a perfectly natural context in which I would say, “I know my car is parked outside.” [In more precise jargon, the appropriate contrast-class of defeaters that needs to be eliminated is set by the appropriate features of the context.]

Suppose, however, that some annoying philosophy student fresh out of an introductory level epistemology class and drunk on his newfound insights, presses me by asking, “Do you really know that your car is parked outside?” (Or perhaps he just shouts “Cartesian Demon!”) This alters the context of inquiry. The conditions for confidence are now higher and more demanding. Maybe I might answer, “Well, ask Rita the Meter Maid; she’s outside standing next to where my car ought to be parked. She knows, and perhaps I don’t know after all.” [Philosophical aside: I think it’s fascinating that “knows” doesn’t appear to admit of degrees under analysis. For instance, strictly speaking, I don’t say “she knows better than I do” as a way of describing a degree of knowing, but rather to express that she’s in a better position to know simpliciter. Does this compete with contextualism? I don’t think it does.]

My saying this, however, does not thereby imply that my earlier claim to knowledge was inappropriate when I uttered, “I know my car is parked outside.” It is perverse to think that what I ought to have said in that context is, “I have a merely true belief that my car is parked outside.” Why? Arguably, the context of inquiry, while not wholly determinative of the conditions for justification, contributes something to those conditions. So, my original claim is appropriate in one context and less appropriate in another. One is less strict and the other stricter.

What this illustrates is that there is an acceptable practice of claiming knowledge in different contexts. [Philosophical aside: I know this is controversial, and I have colleagues who think this is just nuts.] The different contexts fix the rules for when a knowledge claim is legitimate and when it is less legitimate.

Something like this flip-flopping between a strict versus looser contexts of knowledge claims might be going on with Socrates.

I don’t think this is any brilliant insight. I’m sure there’s some Plato scholar who has mapped this territory already.

Did Plato Have a “theory” of Knowledge?

Here’s the standard view of Plato: A condition for a claim to count as knowledge is that the claim must be infallible and therefore certain. This sounds like such a strong condition—infallible certainty! Can it really be what Plato wishes as the distinction between, say, opinion and knowledge? A moment’s reflection indicates that it does. If Plato were to soften the infallibility criterion to something like a simple truth condition—viz., that one of the conditions on knowledge is that a claim must be true—he loses the distinction between opinion and knowledge, since there are such things as true opinions.

That’s the pipe dream of epistemology, and it has lured great minds into it for as long as recorded history. This is no less than the Cartesian project. Descartes had his own reasons for pursuing this, but for those of you who have familiarity with the continental rationalists, you will see the obvious connections to the ancient fascination with infallible certainty.

Satisfaction of this condition would enable one to say of a knowledge claim that one knows it, but also that one knows that one knows it. To claim that one knows that-K requires that one knows-that-one-knows-K (the so-called “K-K thesis”). Let’s call this gussied-up knowledge.

Loosey-Goosey Condition

Let’s recall that knowledge might function differently in different contexts from the perspective of ordinary linguistic practice. Is there a loosey-goosey context that might be appropriate for Plato, a context in which he would be amenable to having Socrates claim that he knows even if he does not know that he knows?

This is hard to say. It is worthwhile to ask if we today have a conception of knowledge that does not require infallible certainty. Indeed we do. We think we have made philosophical progress over a few millennia, and we by and large think it perverse to require infallible certainty for knowledge claims.

We take it to be consistent that we are both fallible and knowledgeable. Here’s how it works. [Philosophical aside: This is why we can also claim to be epistemic foundationalists about the structure of epistemic justification without having to be embarrassed by the crazier forms of foundationalism that tie themselves to further criteria about the nature of the foundations.]

Consider again my belief that my car is parked outside. Call that p. I have evidence for p. My evidence is my memorial belief that I parked my car outside, my memorial belief that I glanced at my car about five minutes ago, my belief based on testimonials from friends who told me that they saw my car parked outside, etc. Call that conjunction of evidence q.

I believe p on the basis of q.
p is true.
q counts as “good enough” evidence for p.
q, however, does not entail p.

Do I know that my car is parked outside? Sure I do. So long as all the conditions are satisfied, then I can be said to know that my car is parked outside. Do I know that I know that my car is parked outside. Clearly not. But the failure to satisfy the very strong condition for a second-order knowledge claim does not by itself show that my first-order knowledge claim is in jeopardy.

From here, epistemology gets very contentious; so I think it is good enough to leave it at that for our purposes.

Okay... the question is whether Plato has available something like this looser-goosier conception of knowledge.

I think the answer is “yes.” Call it Socratic elenchus.

Recall that the main obsession for Plato (and hence of Socrates) is knowledge in the domain of ethics, of “the Good.” On what basis does he and could he claim to know? In actual practice, it is usually through the dialectic of question and answer.

Could something like the model I used to analyze my knowledge that my car is parked outside be adapted to explain Plato’s knowledge by elenchus?

The essence of a fallibilist epistemology is that there is always a strictly logical gap between P and Q, where P is the object of an alleged knowledge claim and Q is the evidence.

When Socrates claims (dialectically) to know P, he must base it on Q.

Maybe Q is the evidence garnered from Socratic badgering, where no contradiction is uncovered after a lengthy, exhaustive inquiry and cross-examination.

So, in essence, it’s a negative condition: We did not uncover any contradictions.

If that’s true, then maybe Socrates is satisfied to claim that he knows (dialectically). To be sure, this is a looser-goosier claim than a claim about infallible certainty. It falls short of infallible certainty. After all, Socrates via the elenchus never claims to have perceived The Forms. Only the gods of Olympus get that honor.

Hence it is not the same as gussied-up knowledge. It’s just plain old “loosey-goosey knowledge.”

Socrates is saying that there is a kind of wisdom or knowledge that is appropriate only for the gods, because they have gussied-up minds. Human minds are loosey-goosey and so can contain only loosey-goosey content. He very clearly owns his human wisdom, but he denies that he has godly wisdom.

What Socrates is rewarded for as well as cursed with is his humble acceptance of the human condition. It is ultimately a religious, pietistic epistemology which disciplines human pretension. It is as if Socrates is saying, “I am not a god, and neither are any of you.”

At the end of it all, when Socrates denies that he has knowledge, he is denying that he has the mind of a god. He is denying any access to gussied-up knowledge.

When he claims that he possesses knowledge, he is claiming merely human wisdom which comes from the humble exercise of the elenchus which delivers loosey-goosey knowledge.

Socrates turns out to be a contextualist, which for him is an ethical epistemology.

Compared to gussied-up knowledge, loosey-goosey knowledge is a pauper. But what do you expect? We do not number among the children of the Titans.


Eric said...

A couple of things briefly. Dan's was a long post, and deserves a lot more than I will give it here. But this is what I will give it on a relatively warm Nov 1 evening.

I can't readily think of places in Plato's early (Socratic) dialogues (or even in the middle or late ones) where Socrates starts a sentence "I know that...." However, there are plenty of instances where he simply asserts beliefs in such a way that we might say "Socrates knows X." Some of them are rather mundane, like "I could bring my wife and children to the trial weeping to try to get a favorable verdict for myself" or "my wife's name is Xanthippe" or "I fought as a hoplite" or "I have lived my whole life holding Athens in high regard." Some of them are a bit more substantial, like "the unexamined life is not worth living for a human" or "one should care more for virtue, knowledge, and the best possible state of one's soul than for wealth, honor, and reputation." However, he consistently denies knowing what virtue is (and since he thinks that "to know what virtue is is to be virtuous," he must know he is not virtuous).

Now put these two together: one should care more for virtue than other mundane things, Socrates recognizes he does not know what virtue is and so is not virtuous, so he spends his life testing himself and others about virtue. Part of Socrate' hope in that pursuit is to at least remove mistaken beliefs he has about virtue, and at most come eventually to understand what virtue is and so become virtuous.

I guess that is all for now. But I have a lot more I can and I'd like to say about it.
Eric Snider

Keith DeRose said...

Hi. Found you with a Google search.

On "Socrates the contextualist?":
Gregory Vlastos had an important paper on just that topic:

Vlastos, Gregory 1985. "Socrates' Disavowal of Knowledge," The Philosophical Quarterly 35: 1-31.

I recently wrote a book on epistemic contextualism (not out til about May), and, in giving a brief history of the idea, wrote this short paragraph and footnote about Vlastos's paper (to give you some idea what Vlastos's paper is about):

Interestingly, a nice early statement of contextualism, not usually mentioned when early examples of contextualism are discussed, occurs in a study of ancient philosophy: Gregory Vlastos's (1985) uses contextualism to reconcile the disavowals of knowledge by the Socrates of Plato's early dialogues with the claims to knowledge made by the same Socrates.*

*The actual reconciling work is done by a distinction Vlastos draws between what he calls "knowledgeC" (certain knowledge) and "knowledgeE" ("elenctic knowledge"), which does not involve the "fantastically strong standards of knowledgeC" (1985: 18), but does require that beliefs be "elenctically justifiable," as well as true. Thus, through most of the paper, Vlastos seems to be utilizing what looks like some form of the "two senses of 'knows'" account, like the account of Malcolm's that I mentioned at the start of this section. (And, interestingly, Vlastos was a colleague of Malcolm’s for several years at the Cornell University philosophy department.) However, Vlastos begins the section of his paper in which he draws this distinction with an extremely interesting paragraph that begins with the sentence, "Let us reflect on our own use of the terms 'know' and 'knowledge'" (1985: 11), and then proceeds to give a very nice contextualist (though Vlastos doesn't use that word for it) account of the ordinary usage of those terms that seems to express a form of contextualism that's much like the current versions of the view, on which many different standards can govern uses of "know(s)", rather than "two senses of 'knows'" accounts. Vlastos seems to be thinking of the particular distinction he utilizes throughout much of his paper as a distinction between two of many possible different uses of "knows." Thus, though Vlastos seems to be using a "two senses" theory through most of his paper, his actual position would seem to be a form of contextualism like those that are currently popular. (Thanks to Gary Ostertag for reminding me of Vlastos's paper in connection with the history of contextualism, and for an interesting e-mail exchange about the paper.)

Dan said...

Keith, thanks for posting your comment. Strange and cool how Google ends up connecting people and their ideas!

You know, shortly after I posted that, a friend of mine back in California told me I should take a look at Vlastos’ article. It’s quite amazing that he’d be writing in that vein. I wonder if he had any encounters with Unger’s earlier stuff in Philosophical Relativity. Actually, I don’t know if the dates would work out.

I look forward to your book. I got interested in contextualism during a course I taught on Pyrrhonian skepticism. We used a recent collection of essays edited by Walter Sinnot-Armstrong. I recall one of the essays (I think it was one written by Sinnot-Armstrong) where the concept of a contrast-class of eliminable propositions that are inconsistent with the knowledge-candidate proposition is deployed as a way of articulating contextualism. This was then used as a way of interpreting what Pyrrhonian skeptics were actually trying to do.

In any case, thanks for the tip.

Back to grading!

– Dan