To: Craig S.
From: Geoffrey Klempner
Subject: Does knowledge require existence of self-justifying beliefs?
Date: 6th March 2009 12:48

Dear Craig,

Thank you for your email of 25 February, with your essay for the University of London Epistemology module, in response to the question, ''In order to amount to knowledge, a belief must be justified. So, unless some beliefs are self-justifying, there is no knowledge.' Discuss.'

Regarding your comment on Quine being a 'philosopher's philosopher', he is in my view a very good writer (in the American tradition of crisp, pungent prose), or at least when he is at his best. The collection of articles, 'From a Logical Point of View' is a classic.

You've found a reading of this question which hadn't occurred to me, a reading according to which foundationalism would be an *alternative* to the view that 'some beliefs are self-justifying'. Your argument seems to be this: when a belief is formed in response to immediate experience -- say, a sensation of red -- the appropriate description would be in causal terms, rather than in terms of justification. That is true 'foundationalism' because we have found some thing (?) which operates in analogous way, e.g. to the foundations of a building. The foundations of a building are not part of the building, but rather the extra bit at the 'bottom' which you need to hold the building up.

On my reading, foundationalism just *is* the view that some beliefs are self-justifying, while the alternative view would be better described as a variety of externalism. On the alternative view, we form beliefs in response to the impact of experience, but these beliefs are not 'justified' because (in the normal case) there is no room for rational assessment. An observer, however, would conclude that so long as our cognitive apparatus is operating according to its design (by the theory of evolution) then it delivers up knowledge.

The classic example of foundationalism is Descartes' 'Meditations' where Descartes specifically identifies subjective experiences -- like the experience of 'red' -- as a kind of 'thought' which cannot be doubted. In other words, my experience of red *is* a self-justifying belief, 'I see red now'.

Incidentally, what I have described as the 'alternative' to the self-justifying view is fairly close to the account given by Quine in his essay 'Epistemology Naturalised' where Quine rejects traditional Cartesian foundationalism: in his often quoted statement, 'There is no first philosophy.' ('First philosophy' is the term used by Aristotle to refer to writings which are now known as the 'Metaphysics'. The full(er) title of Descartes 'Meditations' is 'Meditations on First Philosophy'.)

Naturalised epistemology would be one way to block the argument. It does so denying the demand for internalist 'reasons'.

One can be externalist about reasons (as you describe). Arguably, as you state, this merely pushes the problem back. The alternative externalist view would be to embrace naturalised epistemology. It is a matter of scientific inquiry and explanation how human beings are so 'good' at acquiring knowledge. The purpose of this inquiry is not to provide some kind of warrant or assurance that our knowledge is what it purports to be, for such warrant is neither necessary nor indeed possible.

Like you, I can't see any great attraction in infinitism. However, a variant on this theme would be 'indefinitism', where human knowledge gets stronger and more coherent, as we pursue the quest for reasons further and further back. This is the opposite of the picture of an infinite regress. It is also consistent with the line which you take at the beginning that justification can be fallible. There are different 'directions' in the pursuit of knowledge: we can learn about more things, and we can also learn more about how we know and why.

Regarding coherentism, almost as an afterthought you insert the remark that coherentism is improved with a 'whiff of foundationalism', which broaches a huge topic. Is pure coherentism coherent? The classic objection is that you can have two mutually inconsistent systems of beliefs, each of which is fully justified by the criterion of coherence. However, as soon as you inject 'experience' or 'contact with the real world' or whatever, you have re-introduced the very thing that coherentism was designed to avoid -- the appeal to foundations. It is merely truistic that coherence of beliefs increases our warrant for holding them.

-- Regarding all these distinctions, I have to admit that I feel somewhat the way you do, that the industry that has grown around epistemology is 'Byzantine', and, frankly, scholastic (in the worst sense). However, there are core questions about knowledge which are gripping and the thing to do is to explore the texts with the core questions constantly in view.

If you are looking for something 'gripping', I would suggest Barry Stroud's book 'The Significance of Philosophical Scepticism'.

All the best,

Geoffrey