Learning and Knowing

I used to think the following was a nice little analytic truth.

  • If immediately prior to t, S does not know that p, and at t she does know that p, then at t, S learns that p.

But now I’m convinced there are counterexamples to it. Here are four putative counterexamples, some of which might be convincing.

A few months ago, Alice learned that the President McKinley was assassinated. Soon after, she forgot this. Just now, she was reminded that President McKinley was assassinated. So she now knows that President McKinley was assassinated, and just before now she didn’t. But she didn’t just learn that President McKinley was assassinated, she was reminded of it.

Bob starts our story in Fake Barn Country. He is looking straight at a genuine barn on a distant hill, and forms the belief that there is a barn on that hill. Since he’s in fake barn country, he doesn’t know there is a barn on the hill. At t, while Bob is still looking at the one genuine barn, all the fake barns are instantly destroyed by a visiting spaceship, from a race which doesn’t put up with nonsense like fake barns. After the barns are destroyed, Bob’s belief that there is a barn on that hill is knowledge. So at t he comes to know, for the first time, that there is a barn on that hill. But he doesn’t learn that there is a barn on that hill at t; if he ever learned that, it was when he first laid eyes on the barn.

Carol is trapped in Gilbert Harman’s dead dictator story. She has read the one newspaper that correctly (and sensitively) reported that the dictator has died. She hasn’t seen the copious other reports that the dictator is alive, but the existence of those reports defeats her putative knowledge that the dictator is alive. At t, all the other news sources change their tune, and acknowledge the dictator has died. So at t, Carol comes to know for the first time that the dictator has died. But she doesn’t learn this at t; if she ever learns it, it is when she reads the one true newspaper.

Ted starts our story believing (truly, at least in the world of the story) that Bertrand Russell was the last analytic philosopher to win the Nobel Prize in literature. The next day, the 2011 Nobel Prize in literature is announced. A trustworthy and reliable friend of Ted’s tells him that Fred has won the Nobel Prize in literature. Ted believes this, and since Fred is an analytic philosopher, Ted reasonably infers that, as of 2011 at least, Bertrand Russell was not the last analytic philosopher to win the Nobel Prize in literature. This conclusion is true, but not because Fred won. In fact, Ed, who is also an analytic philosopher, won the 2011 Nobel Prize in literature. At t, Ted is told that it is Ed, not Fred, who won the prize. Since Ted knows that Ed is also an analytic philosopher, this doesn’t change his belief that Bertrand Russell was not the last analytic philosopher to win the Nobel Prize in literature. But it does change that belief from a mere justified true belief into knowledge. But arguably it is not at t that Ted learns that Bertrand Russell was not the last analytic philosopher to win the Nobel Prize in literature, since just like in the last two cases, Ted’s evidence for this conclusion does not improve.

8 Replies to “Learning and Knowing”

  1. Up front, I don’t find any of the putative counter-examples convincing. The first comes closest, but I want to say that what happens in that case is best described as re-learning, and I take that to be learning plus something. For the other three, I want to say that the learning/knowing subject does not learn the target claim until time t, the time at which the subject comes to know the target claim.

    I wonder how you are aiming to use “learn” in these contexts. Would you endorse the following principle:

    If S learns that p, then there is a time t such that S knows that p at t.

    Or more directly:

    If S learns that p at time t, then S knows that p at time t.

    If you endorse either principle, then I think you should at least deny that the last three counter-examples have any force against your original principle. Here is why. Suppose that in your example cases, the learning/knowing subject is destroyed before the big event in each story — before the alien attack, before the other news organizations change their tune, before Ed is announced as the prize winner — but after being exposed to the initial stimulus — after seeing the real barn, after hearing the true news story, after hearing that Fred will win the prize. In that case, the learning/knowing subject never knows the target claim, but for your examples to work, we have to say that the subject has learned the target claim. I can see how one might have learned that p without knowing that one has learned that p, but I don’t see how one could have learned that p without knowing that p.

  2. Well, I have doubts about the first case too, but I think the other three are better (although I like the fourth less). In the first case, it seems appropriate to me to say that Alice learned the fact again, if she had genuinely forgot it, so that she could not have ever recalled unless told again. Otherwise, she really knew all along.

    As I said, the other cases are more convincing to me. To reply to the previous comment, I see two accounts of learning that are compatible with the counterexamples. On one hand, one might think of learning as forming a true belief. Then one should reject the principles Jonathan proposes. But let us grant that those principles are attractive. Still, one does not need to identify learning with coming to know. It might be that learning is a specific way of coming to know, for example forming at time t a belief that constitutes knowledge at t. Therefore, one should not overlook the possibility that the subjects comes to know something without EVER learning it.

    Surely that a subject knows something does not entail that she learned it at some previous point. God could know things without learning them, and we could too, for example if we had innate knowledge. Of course, the cases are different in that the subjects came to know the relevant propositions, while in my previous examples there is no time at which the subject does not know. But again, if we do not identify coming to know and learning, we can save what I take (with Brian) to be our initial judgement on the cases, and also the principles Jonathan proposed.

  3. I agree in the first case that it is possible to re-learn things. I just don’t think this happens every time we forget something.

    But I got very confused by Jonathan’s principles. This seems true:

    If S learns that p, then there is a time t such that S knows that p at t.

    But the ‘more direct’ version seems clearly false.

    If S learns that p at time t, then S knows that p at time t.

    Compare the following two principles.

    If S kills X, then there is a time t such that X dies at t.
    If S kills X at t, then X dies at t.

    As Fodor pointed out years ago, slow poisonings are a counterexample to the second but not the first. The same is true, I think, with learning and coming to know. Learning may require eventual knowledge, but it doesn’t require knowledge at the time of the learning. This implies a weak kind of temporal externalism; whether S learns p at t depends on what happens after t. But I don’t see why that should be surprising. Whether S kills X at t also depends on what happens after t.

  4. Brian,

    Thanks for the follow-up. I agree that my more direct version is (logically) stronger and so it may be more susceptible to counter-examples. I like the Fodor story, but I think I want to draw a different moral than you do. I want to say that some actions are temporally spread out. Learning might be one of them. If we are considering a spread out case of learning, then it will be false that S learns that p at time t, and my principle will be vacuously satisfied. In a spread out learning case, I want to say that S learns that p over the time interval <t1, t2>. Then I’ll recast my principle as follows:

    If S learns that p over the time interval <t1, t2>, then S knows that p at time t2.

    Moreover, for times in the interval, I don’t want to say that S learns or has learned that p. Rather, I want to say that S is in the process of learning that p. Only upon completion of the process do I want to say that S learns or has learned that p. The same story seems to work for killing, but maybe I’m missing something?


    None of the principles precludes the possibility that the subject knows that p without learning it. Brian’s principle precludes coming to know that p without learning it, but mine don’t even do that.

  5. Jonathan, you wrote “but for [Brian’s] examples to work, we have to say that the subject has learned the target claim [in your modification of the cases].”
    But that’s wrong. The examples work against Brian’s principle even if the subject never learns the target claim, although the subject comes to know them (in the original cases, if the subject does not die). Do you have a reason for ruling that possibility out?

  6. Daniele,

    I think you are probably right about what I said. What I was picking up on was Brian’s claim that if S learns that p (in his stories), then S learns that p right at the moment of exposure to the cause of S‘s belief that p. That still seems wrong to me, and it seems to be what’s driving the counter-examples. That is, the examples have the form: S learns that p at time t1 < t2, but S doesn’t really know that p until time t2.

    I don’t think I can rule out the possibility you suggest, but I also think we could do a better job of generating examples that exploit your point. Here’s one — provided I’ve understood what you’re suggesting:

    Jack goes to see his local neuroscientist, Jill. Jack really wants to be a top defense attorney. However, he doesn’t have time to study the law. Jill the neuroscientist is eager to help. She puts Jack under anesthetic and manipulates Jack’s brain so that he has proficient knowledge of the law.

    When Jack wakes up, he finds that he has true, justified beliefs about the law. It looks like he knows things about the law that he didn’t know before the surgery. Still, I don’t want to say that Jack learned these things about the law.

    Does this sort of case fit your suggestion?

    All of that said, I’m not sure that those are counter-examples to the principles that I proposed in my comments. Though I hadn’t really noticed it until now, I turned around the ordering on the conditional in the principles. I want to go from antecedents about learning to consequents about knowledge, not the other way around. And from that perspective, it will be perfectly possible to have knowledge without learning. I would be surprised, however, to discover that one can have (completed or successful) learning without knowledge.

    How does that strike you?

  7. Jonathan,

    I quite like your story as an example of a case in which there is coming to know without learning. I still think that Brian’s last three original counterexamples are also cases like that. But I am more confused than I was before (or maybe I am more aware of my confusion) about what account of learning could support these judgements.
    In any case, I agree these are all counterexamples to conditionals going from coming to know to learning, but not the other way round.
    I am not sure what Brian thinks; although he clearly meant to leave the possibility open that the subjects in his stories never learn the target claim, comment #3 suggests that he thinks there are episodes of learning involved – the belief-forming episodes would constitute learning in virtue of the fact that the beliefs later turn into knowledge. On that view, Brian’s original cases would be counterexamples to all but the weakest of the learning>coming to know conditionals considered.
    I do not think there is anything problematic in principle about future events determining whether a property applies to a previous event. My buying a lottery ticket is buying-a-winning-ticket only if the ticket is going to win. But we are not forced to say that my buying-a-winning-ticket is a process that lasts until the extraction. My buying-a-winning-ticket is completed by the time I bought the ticket, although it depends for its being such by future events (I just now thought of a more fun example: conceiving a child. I’ll leave the development to the reader). So, while I do not agree with the view I was tentatively attributing Brian about his original cases, I see no general reason to discard it.

  8. I did mean to suggest what Jonathan says. That is, what happens at t2 might make it the case that what happens at t1 is a learning experience.

    The alternatives just all seem bad. Perhaps we can have knowledge without learning, but these don’t seem like such a case. So the learning must have happened sometime. And, especially in the Harman case, the agent is only engaged in what look like learning activities at t1. So the learning was at t1.

Leave a Reply