The Philosophy Resource

Teaching resources for philosophy students and teachers

A more radical proposal

Drawing on work from Richard Foley, I have proposed in recent work that in light of Gettier cases we should think that the theory of knowledge is independent of the theory of justified belief. I call this the DIVORCE thesis.

DIVORCE Thesis: The theory of epistemic justification is independent of the theory of justified belief.

Having examined the broad contours of the field in the first three steps, we are now in a position to take a look at my proposal. So, If you fancy a challenge, read on!

My argument begins with reminding us of what people take to be the, original motivation for a justification condition in the analysis of knowledge. That knowledge is incompatible with luck and that the justification condition is there to try to take on board that thought. So, when Gettier came along and showed that justification was not doing the job it was meant to, that spelled real trouble.

But, against the orthodoxy, I think it is a mistake to think that justification is there to eliminate luck. It has a more important function. If it has the function I say it has, then DIVORCE is defensible, so I argue.

Recall the three options I mentioned with respect to how we should respond to Gettier-cases.

Option 1:

Add another condition to the three conditions in the tripartite analysis that deals with luck better.

Option 2:

Get rid of the justification condition and replace it with a better anti-luck condition.

Option 3:

Develop an understanding of justification which successfully does the necessary anti-luck work.


Taking option 1

If we take option 1, then we must ask what work justification is doing in our account of knowledge. If a fourth, tailor-made condition is there doing all the anti-luck work, then either the justification condition is redundant (doing no “work”), or else it is there to take care of another intuition than the intuition that knowledge is incompatible with luck.

Taking option 2

So, we might conclude that it is redundant, and so opt for option 2. But this is slightly counter-intuitive also. We have an intuition that something important has been left out of epistemology if justification is completely discarded.

Taking option 3

So we might then be motivated to take option 3. But the trouble here is that there is something in the very notion of justification that means that it must be possible to have justified, false beliefs. To think that we cannot have false, justified beliefs is to commit to infallibilism. But as we mentioned in Step 3, infallibilism leads us to scepticism. However, once if we allow false, justified beliefs, some forms of Gettier-case will always be possible in theory. And so justification will not do the necessary anti-luck work.

So we must think that justification does some other work than anti-luck work in the definition of. What work is this? What is it that we feel has been left out of epistemology when we leave justification out of it? I propose that it is the intuition that we ought to be responsible believers, that we feel that we have certain duties as believers – for instance to believe in accord with the available evidence and not believe the things that Neo-Nazis or members of the flat-earth society believe – and that we have done something wrong when we violate those duties. Put differently, we want to have justified beliefs, and we want to know how. I call this the Deontic Intuition.

Deontic Intuition: our duty or obligation not to have certain beliefs

The word ‘deontic’ means something like ‘duty-based’ or ‘duty-related’. For example, we have a duty or obligation not to believe that racism is a good thing. Or, to pick another example, a duty to believe, given the overwhelming evidence, that climate change is occurring. Crucially, we have this intuition independently of whether we think that justification is necessary for knowledge. In other words, we don’t necessarily come to think that we have a duty to believe that racism is a good thing through trying to define what knowledge is.

Deontic Intuition & the DIVORCE thesis

I argue that once we have identified the Deontic Intuition, we can make the DIVORCE thesis defensible. We can say that justification figures in epistemology in the service of the Deontic Intuition – we do not leave it out of epistemology altogether if we leave if out of an analysis of knowledge. If we leave it out of an analysis of knowledge we do not need to make it infallible, in order to deal with Gettier cases.

And we also have the means to diagnose why we might have the intuition that knowledge is a deontic notion (or involves a deontic component). Put differently, the intuition that in order to know we must have fulfilled certain intellectual duties. We know that p, for instance, only if we have fulfilled the duty to believe in accord with the available evidence.

Another Diagnosis

If we have the intuition that knowledge is deontic, we also have the Deontic Intuition – it would be incoherent to have the deontic intuition only about knowledge and not about justification. Because of the long history of the idea that knowledge is justified true belief, the fact that we have the Deontic Intuition makes us think that knowledge is deontic. That is, we mistakenly think knowledge is deontic because justification is. But if we have the Deontic Intuition we can coherently hold that justification is not necessary for knowledge, but nevertheless does other, important work in epistemology.

Go to…

Step One

Go to…

Step Two

Go to…

Step Three

Go to…

Step Four

Go to…

Step Five