Despite the ominous title “Take It or Levitt” (possibly chosen for lack of a better pun :-), Alan’s post effectively highlights a number of issues that I concede I have not treated clearly enough:

* How do community norms relate to “universally valid” knowledge?

* What does it mean for knowledge to be “reliable?”

* How do we know if we are being honest with ourselves?

* What is the proper relationship of epistemology to ethics?

[Read more] for my attempt to articulate a better answer to those questions.

Let me start with a definition of knowledge:

I. Knowledge is contextually accurate, paradigmatically-justified belief.

This is somewhat different than Plato’s definition of knowledge as “Justified True Belief.” In his view, I can’t validly say I have knowledge unless I know it is True. But that which requires Reasoning, which I also need to know is True, and so on to an ad infinitum.

I take this infinite regress problem seriously. If that is the only viable definition of knowledge available, I’d have to agree with the skeptics that genuine knowledge is unachievable. However, I choose an alternate path: I reject the notion of ‘absolute knowledge’. Instead, I affirm what may be called ‘contextual knowledge.’ In fact, I would argue this is much closer to what people mean when they say, “I know X is true.” Apart from analytic statements (true by definition like, 2+2 = 4), all human knowledge is necessary incomplete. There is no real data for which we have perfect confidence to infinite accuracy; we are always constrained by our context.

However, that doesn’t mean knowledge is arbitrary, or disconnected from reality. “Knowledge” in this definition still requires both some a priori justification, as well as a posteori validation. However, my formulation explicitly notes that the former depends on a paradigm, and the latter on a specific context of reality. I honestly can’t think of any other meaningful definition of knowledge: can you?

This of course still begs the second-order question, how do we know if our context and paradigm are themselves valid? Certainly we believe they are true, but (to use your phrase) how do we know we are not “self-deceived” —

Applying my original definition recursively, we need a paradigm that can test paradigms — and ideally test itself, to avoid the infinite recursion problem. Thus, in addition to the general definition of knowledge I asserted above, I am also asserting a *specific* paradigm within that definition. I’m afraid I blurred that distinction earlier, for which I apologize.

What is my paradigm? Well, it is basically that:

II. Knowledge approaches truth via honest, collaborative inquiry

That is, it assumes the existence of Character, Community, and Reality, and names a particular attribute associated with each — what we might call epistemic virtues: honesty, collaborativeness, inquisitiveness. The more those are present, the greater confidence I have in my knowledge. In particular, I assert I can use my paradigm to demonstrate its superiority to alternate paradigms (e.g., arguing purely based on Reason, Authority, or Experience).

To be sure, there are a few caveats:

* if you reject this paradigm, I have no way of proving it is true, so yes the argument is circular (I prefer “consistent” 🙂

* strictly speaking, this only defines epistemic virtue, which is technically different than “moral” or “social” virtue, though it may constrain them. I apologize for confusing the two.

On the other hand, I believe this does address the issue of “groupthink” — at least as much as it can be addressed. However, I take the opposite tack from Alan, who asks:

How can we account for these considerations in our epistemology? My inclination is to leave the idea of community outside the proper description of the epistemology, or at least to de-emphasize its role. But if I am understanding Ernie, it seems pretty important to him, so I would like to hear his take on this.
In my view, it is impossible to remove any possible community-centric bias — I’d love to hear how Alan thinks he can get around this. My version of wisdom is rather to recognize that every community defines knowledge relative to their own norms — including ours — which can create systematic deviations from truth. The obvious antidote is to attempt to integrate knowledge from different communities, to find a larger paradigm encompassing a broader range of experiences with reality. This merged community may of course have its own biases, but at least it should be closer to truth.

Which, as it happens, is exactly what I believe Alan and I are trying to do. I’m curious to see whether he agrees with me.

Update: Oops, forget one more point of clarification.

III. More accurate knowledge enables more accurate predictions

Hopefully that phrasing avoids the ambiguity of earlier attempts. See what happens when you hurry me? 🙂