To: Max B.
From: Geoffrey Klempner
Subject: Justified true belief without knowledge
Date: 5th May 2009 10:40

Dear Max,

Thank you for your email of 28 April, with your essay for the University of London Epistemology module, in response to the question, 'Can you have a justified true belief that you are sitting an exam without knowing that you are sitting an exam? What consequences does your answer have for the analysis of knowledge?'

A situation in which you believe falsely that you are sitting an exam might be one where, e.g. you go into the wrong room by mistake where students were taking a practice exam (by coincidence, in the same subject). Even though you are doing exactly the things that a student who is genuinely sitting an exam would be doing, you are not 'sitting an exam' because you have, in fact, missed the exam that you were scheduled to sit. Further elaboration of this example can be used to describe a hypothetical situation where you are, in fact, in the right room but only by sheer luck. So although you are, in fact sitting an exam at this moment, you don't 'know' that you are.

I must admit that I first thought that you had interpreted the statement, 'You are sitting an exam' to mean, 'You will be sitting an exam [at such and such a date].' As you must realize, knowledge of propositions about the future is in itself problematic. As it stands, the statement can be read two ways: 'Your exam has been scheduled for such and such a date' (which does not describe a fact about the future but rather informs you of a decision that has been made concerning when you are to sit the exam), or, 'It will come to pass that you sit the exam on such and such a date.' I don't think that the latter can be known. Students fail to turn up for examinations for all sorts of reasons.

However, I then realized that the structure of your example was rather this: You are sitting in the correct exam room, writing your script. However, the information which you previously received which accounts for your belief that you are sitting an exam in fact came to you by accident. If the email had reached only its intended recipients, you would not have received the information. A serious error was made by the administrators, but fortunately this was cancelled out by a second error. The information shouldn't have reached you, but it did.

Isn't this a case of knowledge? I actually think it is. At any rate, it is a matter on which intuitions differ. You might want to say that any false assumption in the reasoning leading to the belief in question invalidates the knowledge claim. But this seems overly strict. Surely, knowledge can leak out by chance, even though the recipient doesn't realize how lucky they were to get it.

Apart from worries about your example, I had some difficulty following what you want to say about truth and time. There is an important difference between knowledge and belief in relation to the future, which we recognize when we exhibit caution in making knowledge claims which relate to events that have not yet occurred. That is not to say that one cannot have knowledge of the future. If I jump of a building, I know I will hit the ground a few seconds later. We know (and do not merely believe) what we will do through knowledge of our intentions. As I lift my hand to scratch my nose, I surely know that my nose will be scratched.

Interpreting belief in a quasi-Kantian way, however, any belief can be interpreted as implying falsifiable beliefs about the future. This is the situation that the sceptic standardly exploits. I believe that I am sitting an exam -- I received the email, followed the directions, found the table with my number -- then just as I start to write everybody turns round towards me and starts laughing: it was all an elaborate hoax.

One way to read this leads to an anti-realist conception of truth, where there are no given 'facts', because every statement implies non-falsifiability in all future time. Yet this is consistent with a notion of truth which accounts for the validity of logical inferences. If I am sitting the Epistemology paper then, necessarily, I am sitting the exam.

You mention solipsism at one point: this is apt, because it can be argued that Kant's view of truth and experience implies transcendental solipsism. The world is the world of 'my possible experience'. As experience extends indefinitely, I may repeatedly revise my picture of the world, but no statement that I make can be the final word.

If you answer a similar question in an exam, be aware that examiners don't have endless time to mark scripts. You need to explain your idea clearly and simply, and signal to the examiner early on that you are proposing an original solution to a problem. There may indeed be something original (and important) in what you say. But it will pass the examiner by, just as it has (largely) passed me by, because you have not expressed yourself with sufficient clarity.

All the best,

Geoffrey