Dexter asked:

What constitutes knowledge?

and Romina asked:

How can we be sure of what we know?

============

You might think that whatever else knowledge might involve, it has at least got to entail that the
person who is said to knowthat P, is surethat P. How can you be said to know something if you're
not sure?

In that case, Romina's question would have to be understood as a rather careless way of expressing
the problem of scepticism. What Romina should have said is, How can we be sure thatwe know what
we think we know?

For the moment, I do not want to talk about scepticism. So let's stick with the question whether you
can know something even though you're not sure.

First off, it could be argued that there are cases where a person does know something, even though
they are not sure. The standard example is the 'nervous schoolboy'. The nervous schoolboy is asked,
'What is the capital of France?' The schoolboy knowsthat the answer is 'Paris'. After all, he's been on
holiday to France and even gone up the Eiffel Tower! Yet when put on the spot by an irascible
teacher, all his confidence vanishes, and he cannot bring himself to give an answer.

There is more than one way to analyse this example. Threatened with instant execution, there might
be all sorts of questions that you would have answered perfectly confidently, which you now feel not
quite sure about. How sure issure? Would you risk your life to assert that Paris is the capital of
France? Isn't there a tiny possibility that Ministers at the latest EU Summit agreed that Lyons should
be the capital in return for a subsidy for French beef?

An alternative explanation for failure to be sure of what you "know" is that there are certain states of
fear where one's brain is simply paralysed and will not let out the knowledge which is in there. Not just
the word 'Paris', but the city of Paris with all its buildings and inhabitants, has temporarily vanished
from the fearful schoolboy's mind.

This takes us into the general question of just what does constitute knowledge. The standard account
used to be that knowledge is 'justified true belief.' Someone cannot be said to know that P, if it is not
true that P. (What we would say instead is that 'They thought they knew.') Or we use scare quotes: I
"knew" that this question was not going to take me more than fifteen minutes, but I was wrong!
However, not every belief which is true, counts as knowledge. You've got to be able to justify your
claim by giving suitable reasons.

As the philosopher Paul Gettier showed in a paper which rocked academic philosophy in the 60's,
that's still not enough. You can have excellent reasons for believing that P, and your belief that P can
be true, but it can still turn out that it was only by a lucky fluke that your belief turned out to be true.
For example, I "know" that my next door neighbour Derek is at home because I can see him mowing
the lawn. In fact he ismowing the lawn. But what I didn't know is that his long lost twin brother Brian
has come to stay for a fortnight, and it could just as easily have been Brian, not Derek, whom I spied
through the window.

With the Gettier-type examples, the floodgates are opened. Take anything you could reasonably be
said to know. Like the fact that Derek is mowing the lawn. I ask myself, 'Do I know that Derek has not
got a twin brother?' If I can't say, 'Yes' then I don't know that Derek is mowing the lawn, even though I
can see him clearly. With a bit of ingenuity, you can do the same trick on just about any factual
proposition that you take yourself to "know".

Geoffrey Klempner