To: Alistair L.
From: Geoffrey Klempner
Subject: Knowing what one believes as inner perception
Date: 18th February 2010 13:13

Dear Alistair,

Thank you for your email of 9 February, with your essay for the University of London BA Epistemology module, in response to the question, 'Is our ability to know what we believe based upon a form of inner perception?'

In your email you raised a question regarding your previous essay, 'Did Descartes succeed in proving the existence of God?' How are you supposed to answer it?

What the question asks for, in effect, is *how good* is Descartes' argument, or arguments. Given that there are two arguments, the argument from the idea of perfection and the ontological argument, there are three ways to answer the question:

1. Say you think the argument from the idea of perfection is stronger and give your critical evaluation.

2. Say you think Descartes' version of the ontological argument is stronger and give your critical evaluation.

3. Critically evaluate both arguments.

In 1. and 2. you can get away with making a brief statement as to why you think the argument in question is stronger. The examiner will accept this.

Whichever path you choose, the primary question is, to repeat, 'how good the argument is'. If there appear to be missing premisses, then you can supply them. Imagine you had the chance to dialogue with Descartes: if you put to him certain objections, how would he reply? The principle of charity operates here, in that we are looking for the best argument that can plausibly be attributed to a philosopher.

You are right that modern versions of the arguments (e.g. Plantinga's or Norman Malcolm's versions of the ontological argument) are not really relevant to answering this question, if they involve notions which it is implausible to imagine that Descartes might have considered.

However, it is equally irrelevant to consider Descartes' influences, such as his training in the scholastic philosophy of the Jesuits, except insofar as this sheds light on a contested interpretation of what Descartes actually says.

In short, be prepared to consider the possibility that the argument is valid (even if you 'believe' or 'know' in advance that it isn't). Give it a good run for its money. In a similar way, if you get into an argument with a bar-room marxist who says, 'You only believe that because you're a member of the capitalist class', your reply should be, 'Don't tell me why I believe what I believe, respond to my argument!'

Your essay.

After considering objections to the idea of the Cartesian Theatre, you give an account of two opposed positions -- Armstrong who holds that awareness of what we believe is a form of inner perception, and Shoemaker, who offers a telling objection, in the concept of 'self-blindness'.

I think that the key question here is what is so special about the concept of belief, as opposed to knowing what you desire or want or feel about something. It is accepted that we are in a privileged position with regard to our own mental states, while at the same time (as you illustrate with your example of the romantic comedy, where the two protagonists are unaware that they love one another), there are plausible cases where we fail to know what we desire or want or feel about something.

Wittgenstein remarks somewhere, in response to the idea that love is a kind of 'feeling', 'Love is put to the test, pain is not.' You can think you love someone, but when 'put to the test' you realize that you don't. Or you can think you don't love them, but then discover that you do. Love is not just a subjective sensation but is about what you are prepared to do, and you can't always be sure of this through introspection.

On the other hand, if someone asks me what I believe, or, rather, whether I believe that P, the question for me is simply, 'What is my view regarding P?' You're not asking me whether in the past I believed that P but whether I believe that P now. I may or may not have considered the question regarding P in the past but that is all water under the bridge. Of course, it is perfectly possible for me to answer, 'I just don't know, I'll have to think about it more.'

It could be objected that there are cases where I 'know' or 'believe' something in a way that requires a rehearsal of behavioural moves, or the calling up of perceptual cues. 'Is the central library second on the left and first on the right?' I think I know the way to the library but I can't picture the route in my mind: I have to 'follow my nose' and then I will be able to tell you. However, in this case, I really don't at this moment have a belief either way about whether the library is 'second on the left and first on the right'.

Well, how about, 'Do you believe the ice is safe to walk on?' This looks superficially like the love example, in that I might say, 'Oh yes its definitely safe', and then, as I put my foot down I unexpectedly hesitate. Would you say, 'I didn't really believe that the ice is safe to walk on although I thought I did'? I wouldn't. I would say, 'I believed that the ice was safe to walk on but I've changed my mind.' Changing your mind is something you can do without being presented with new evidence. In this example, the stakes are raised, and so you think again.

You raise the question whether, 'the person as a whole can form beliefs through emotional and lower level unconscious mental processes that are not necessarily immediately available.' It would have been nice to have seen some examples. If it's like the route to the library case, then I don't really have the belief in question, even implicitly. The route to the library is something I believe or know (one has to allow the possibility that I am mistaken) but I don't believe or know it in that way or in those terms.

Can you see the problem here? We are looking for an example where it is the case that you already believe that P, but you discover this by introspection rather than by forming a judgement here and now about the question whether P.

Language is crucial, not just as Shoemaker says so that we can communicate our beliefs to others, or engage in rational discussion of our beliefs, but because there is a whole class of beliefs which cannot be understood in behavioural terms -- beliefs involving the past or future, or generality, or negation. In effect, that is the argument why language is needed for thought (see Jonathan Bennett 'Rationality' in the RKP Studies in Philosophical Psychology series).

The crucial point to make is that the process that we describe as 'introspection' is essentially and not merely accidentally verbal. It is the ability to say what is 'on one's mind'. This applies to all mental states not just beliefs.

In this regard you could look at what Daniel Dennett says about mental states and language in his book 'Consciousness Explained'. Dennett offers a powerful account of the nature of self-knowledge which recognizes that we have two different levels of 'awareness', depending on whether or not our 'language centre' is engaged. (As in the familiar example where you are driving along, thinking about philosophy, and realize that you stopped correctly at three red lights but have no recollection of doing so.) Dennett proposes a 'multiple drafts' model of the self, which is more radical than merely removing the ghost from the Cartesian Theatre of Consciousness. There is no single, self-aware 'I' as such, only autobiographies in progress, constructed to meet the needs of the moment.

All the best,

Geoffrey