I've noticed that we're getting a lot of questions like this one. I'm picking this question not in specific, but because it's a good example of a general question I had.

Sometimes, questions like these are very easy to prove, but extremely difficult to disprove. This primarily comes up when dealing with questions of prior example - i.e. are there any previous examples of X happening? All you need to do is find one example for the answer to be "yes," but for the answer to be "no," you might have to review someone's entire body of work, or maybe as broad as an entire century of literary thought.

I'm not saying that the answer to the above example is no, but imagine, for a second, that it is. That would mean Bob Dylan didn't draw from Timrod in any other earlier works, and demonstrating that would require combing his other works in an authoritative and comprehensive way.

How would an answer go about helpfully and meaningfully showing a negative result? What counts as authoritative and succinct, particularly if it hasn't necessarily been studied or reviewed by another source before?

  • Here's a random example of how an "unanswerable" question can be answered. We can't prove a negative conclusively, but if people who know the subject spend a while searching for an answer and can't find one, there might be enough evidence to justify a "this doesn't exist"-type answer. – Rand al'Thor Jan 21 '17 at 12:51

Unanswerable questions are not bad.

In fact, if you can prove a question is unanswerable, that's gonna make for an awesome answer. And if you can't prove it, well, then we're just guessing it's unanswerable.

So if a question's answer is "no", and you can prove it, great. If the question can't be answered and you can show why, that's a great answer too. If you think it's "no" or unanswerable but can't prove it, leave it alone and let someone else try.

(The given example is about Bob Dylan, who has academics all in a tizzy every time he sneezes, so it's very likely that particular question will be supportable with exhaustive PhD theses.)

  • "So if a question's answer is "no," and you can prove it, great." - trouble is, it's often impossible to prove a negative. – Rand al'Thor Jan 20 '17 at 11:49
  • 3
    @Randal'Thor So, don't. Just because a question is literally unanswerable doesn't make it a bad fit for the site... and if we can't prove it, then we shouldn't close it on spec. – BESW Jan 20 '17 at 12:00
  • 1
    I'd argue that a "I can't prove it can't be answered, but I've looked here, here, and here, and found no answer, so it's possible that this is unanswerable." It's helpful to later researchers and it gives the best answer currently possible. – Sean Duggan Jan 24 '17 at 21:33

You must log in to answer this question.