To: Louis G.
From: Geoffrey Klempner
Subject: Perception and the nature and limits of knowledge
Date: 12 July 13:00

Dear Louis,

Thank you for your email of 2 July, with your fourth essay for Possible World Machine, in response to the question, 'What is perception? Explain the role of perception in an account of the nature and limits of knowledge.'

This is a good piece of work which you have thought a lot about. You have found some good examples to illustrate your argument.

What is perception? There are two ways to focus on this, from the example of our organs of sensory perception as you do - seeing, hearing, smelling etc. - or more abstractly, in terms of what it means to say that someone 'perceives' that such and such is the case, or perceives an object as opposed to knowing discursively that such and such is the case or knowing of the existence of an object.

While knowledge implies some process of thinking or inference, in perception the fact is somehow 'given'. There is no process of reasoning involved, we just 'take in' the object, or the fact in question.

It could be argued that there has to be perception, otherwise we could never get started in acquiring knowledge. There has to be something to reason from - e.g. what we can discover just by looking.

Consider the role of perception in understanding what someone is saying. Normally, we perceive what someone else is saying. It is only in exceptional circumstances that we have to consciously think about the words and what they are, or might be intended to mean.

What is the difference between sensation and perception? The red that I see when I push my eyeball, or the pain I feel when I pinch my hand are sensations. Arguably, in perception there is always something 'given' - in the matter of sensation - which we 'take as' an object or fact of some kind.

To explain perception in terms of the physical processes involved, as you do in your first paragraph, might give the false impression that the biology is all that is involved in explaining perception. Yet organisms which have the capacity to sense their surroundings do not necessarily 'perceive'. They respond to stimulus but are not aware of the world as a world, or of spatio-temporal objects as objects.

The big problem for the empiricists, especially Hume, was how the 'given' - which they termed 'ideas' - could possibly generate knowledge of spatio-temporal particulars. In a notable passage in the Treatise of Human Nature ('Of Scepticism With Regard to the Senses') Hume asks how we could possibly form the idea of objects which exist continuously, and distinct from our perception of them, given that our experiences are non-continuous and do not exist apart from us.

While Hume resorted to a theory of 'fictions' - spatio-temporal objects as a kind of mental make-believe - Kant's contribution was not just, as you say, to reconcile empiricism with rationalism but also to demonstrate (specifically in the 'Refutation of Idealism' from the 2nd edition of the 'Critique of Pure Reason') that perception is, as a matter of logic, necessarily perception of objects in space. Without the a priori notion of space, there can be no experience, because there is no logical basis for a unitary self which has the experiences. Without a unitary world, there is merely a succession of momentary 'selves' each submerged in the sensation of the moment. The identity of the subject thus presupposes perception of an external world.

Perception raises fundamental questions for metaphysics as well as for epistemology. The clash between idealism and realism is a question primarily for metaphysics. Kant's metaphysical view turns out to be a form of idealism - the world of objects in space is merely the 'form' that our experience must necessarily take, while there remains a 'noumenal' world beyond all possible knowledge which ultimately accounts for the 'phenomenal' world. By contrast, Kant correctly describes his epistemology as 'empirical realism'.

One of the most interesting questions for epistemology is whether the considerations which you describe - the selective nature of perception, the difference between the ways that different persons perceive - raises questions about the objectivity of knowledge. Is there, despite all the differences, something that we can call the edifice of human knowledge, or are all 'knowledge' claims ultimately relative? How would you argue against the relativist? Surely, there has to be something indisputably 'given' in order to provide a foundation for knowledge. If there is nothing 'given', or if different subjects are 'given' things differently, then it looks like the edifice of knowledge has nothing firm to rest on.

Issues of 'foundationalism', 'non-foundationalism', 'coherentism', 'relativism' in epistemology all trace back to the view one takes of perception.

All the best,

Geoffrey