To: Alfred M.
From: Geoffrey Klempner
Subject: Difference between perception and introspection
Date: 6 February 2007 10:07

Dear Al,

Thank you for your email of 26 January, with your University of London essay in response to the question, 'The only difference between introspection and ordinary perception is that introspection reveals aspects of the subject's psychology while perception reveals features of the external environment.'

You have written a very interesting essay which explores the relation between our capacity for perception and powers of reasoning on the one hand, and our emotional or affective character which determines the features in our environment which appear salient to us at a given moment in time.

Obviously, you have done a lot of reading and thinking about the psychology and physiology of cognition, as well as about the nature of knowledge and perception.

However, it is fair to say that you have missed the point of this question.

Let me try to characterize the problem, in a way which I hope will be gripping to you.

Consider first the nature of introspection. We are not now talking about how one deliberates and forms a plan before action, or how one ratiocinates to a conclusion from given data. The basic act of introspection is to become aware of a mental 'object'. I feel a tickle in my left knee. I remember how hot it was in Turkey last year. I experience a pang of regret that I didn't put my name down for that conference.

The tickle, the memory and the pang have meanings which point beyond themselves. But they also appear simply as events in consciousness. As such, my awareness - of the tickle, or the memory of x surfacing in consciousness, or the pang as it hits me - is unmediated. One is tempted to say that there is a subjective aspect which I cannot be wrong about, irrespective of whatever further facts these mental events might point to. I cannot doubt that I feel a tickle at the moment when I feel it. A moment later, I might decide that it isn't a tickle but a tiny sharp stabbing pain, but then at that very moment the subjective character of my feeling has changed.

In short, the subjective mental events which we introspect have an aspect which seems in some sense self-validating. I cannot doubt that I feel x at the moment when I feel it, because, in some sense, there is no distance between the 'I' and its 'object'.

Now contrast the case of perception. As you note, we can sometimes be deceived. Take your red cat. I seem to see a red cat in the middle of the road outside but when I get closer I realize that it is a lady's brown fur hat. Or, as I sip my tenth bottle of beer, my friend Herbie the pink elephant joins me at the bar.

There is no doubt that something is presented to introspection, but there is room for doubt as to what is really out there in the external world.

Some philosophers have used this argument - known as the 'argument from illusion' - to put the case that whereas in introspection we are 'given' objects directly, perception of objects in the external world is always indirect, an inference from what is given to what is out there. In one of its forms, this view is known as the 'sense datum theory'.

This is a very challenging philosophical problem for the theory of perception. If the sense datum theory is correct, the problem of scepticism looms large. We never really perceive objects in the external world, we merely infer their existence from what is subjectively given.

Reacting against the sense datum theory, some philosophers categorically refuse to accept that in cases of erroneous perception, such as my example of the cat or the elephant, any 'object' is immediately given in consciousness. The whole truth about the subject is that he or she 'seems to perceive a red cat' or 'seems to perceive a pink elephant' and nothing further can be done to analyse that mental state. While other philosophers try to tread the thin line of arguing that although there is no such thing as sense data as conceived by the sense datum theory, nevertheless there is a sense in which there is always a content available to us which allows the possibility of referring to the 'seeming red cat' or 'seeming pink elephant' as actual mental entities.

Now, from what you say in your essay, one would naturally conclude that you do not subscribe to the sense datum theory. You say all the things that a proponent of some form of 'direct realism' about perception would say. What the exam question asks for, however, is an argument for this position. And that you do not give.

The following entry in the Stanford Encyclopedia of Philosophy will be helpful:

http://plato.stanford.edu/entries/perception-problem/

Or you can find references on this topic by looking up 'sense datum theory' in Google.

All the best,

Geoffrey