Skip to content
Coordinates of ThoughtHome of Philosophy
Join
Question · Explain Newcomer-friendly

Norman the clairvoyant is reliably right, yet most people say he doesn’t know. Why?

Ethan Mercer Fellow
· 5 replies
Knowledge & beliefAnalyticAlvin GoldmanErnest SosaLaurence BonJour#clairvoyance#externalism#internalism#reliabilism
Follow

We did the Norman case in my epistemology tutorial this week and I came out more confused than I went in.

The setup, as I understood it: Norman has a clairvoyant faculty that is completely reliable. He has no idea he has it, no evidence for it, and no reason to think clairvoyance is even possible. One day he finds himself believing the President is in New York. The belief is true, and it came from the reliable faculty.

The tutor asked for a show of hands on whether Norman knows. Almost everyone said no. I said yes, mostly out of contrarianism, and then could not defend it.

Here is what bugs me. If what matters is getting things right reliably, Norman is doing great. A weather forecasting model that is right 95% of the time is something we trust, even if nobody can say exactly why a given forecast came out the way it did. So why is the "no" intuition supposed to be so strong?

I think the answer has something to do with the difference between internalism and externalism, but the readings explained the terms in a way that made them sound like the same thing with different emphasis. Could someone give me the clean version: what each side says, why Norman is supposed to be a problem for one of them, and whether there is a way of splitting the difference?

LikeAnswer
Follow5 answers

Join the conversation: members reply here, and joining is free. Reading is always free.

Join free to reply Sign in

5 answers

Accepted answer

  1. Ana Clara Ferreira

    ✓ Accepted answer Fellow

    Here is the clean version.

    Internalists say that what justifies your belief must be something you can access by reflection: your evidence, your reasons, how things seem to you. If you could not, even in principle, point to why you believe it, the belief is not justified, however it was caused.

    Externalists say justification can depend on facts outside your view. The best-known version is Alvin Goldman's process reliabilism (from his 1979 paper "What Is Justified Belief?"): a belief is justified if it comes from a reliable belief-forming process, whether or not you know that.

    Norman is Laurence BonJour's counterexample to reliabilism. From Norman's point of view the belief arrives from nowhere, and holding on to it anyway looks epistemically irresponsible. Irresponsible beliefs are not justified, however reliable their source.

    The externalist reply: you cannot explain how your eyes work either, yet you know there is a cup on the table. If Norman fails, so does ordinary perception. The internalist answers that perception at least comes with a way things look, which you can cite. Norman has nothing like that.

    To split the difference, Ernest Sosa distinguishes "animal" knowledge, reliable and apt, from "reflective" knowledge, where you also have a perspective on your own reliability. Norman might have the first and lack the second.

    Helpful · 2
    1. Ethan Mercer

      Fellow

      Ana Clara, thank you, this is much clearer than the reading. So Sosa's split basically lets both sides be right? Norman knows in the animal sense and doesn't know in the reflective sense, and the show of hands was just people reaching for different senses of "know"?

      1. Ana Clara Ferreira

        Fellow

        Not quite both right. Sosa's view is externalist at the base: animal knowledge needs no reflective access, so on his view Norman may well know something. The internalist will say that animal knowledge is not the kind they were talking about. So the disagreement survives, it just moves to the question of which kind of knowing matters for which purposes. Alistair's point is a good example of a case where the reflective kind is the one we owe each other.

        Helpful · 2

Other answers

  1. Lars Eriksen

    Fellow

    A version of the problem from the lab. When a new detector is first switched on, we do not trust it just because it happens to be reliable. We calibrate it against known sources, and only then do its readings count as measurements. If someone told me a box on the bench had been right about particle masses for years but nobody knew why, I would want it opened before I published anything based on it.

    So in the lab we behave like internalists about instruments, even if in daily life we are all externalists about our own eyes.

    Helpful · 3
  2. Alistair Gow

    Fellow

    From general practice: experienced GPs develop a "something is wrong here" feeling about a patient that they often cannot articulate on the spot, and the good ones I worked with learned to take it seriously. But they never treated it as the end of the matter. It was a reason to examine again, order a test, or ask a colleague. The feeling earned its place by being checked.

    That suggests a middle path: a reliable hunch is a starting point for knowledge, not knowledge itself, until you can stand behind it.

    Helpful · 1