How far apart are these descriptions? I was approaching the issue from the perspective of erotetic logic, and my intuition is that self-evidence is when a proposition is evident from its erotetic format ("S is P," is self-evident iff the question, "Is S P?" is evidence for, "S is P"), but being self-explanatory is subtly different. I.e. instead this is: "S is P," is self-explanatory iff our ability to ask the question, "Is S P?" is what explains how S is P.

This seems as if it would restrict self-evidence and self-explanation to pure erotetic relations, maybe. Like, only Cartesian ur-sentences (about the faculty of doubt) would be self-evident or self-explanatory.

The last disjunct of my question(!), then, is whether there are self-justifying imperatives and self-questioning questions. "Why do the right thing?" would seem to self-justify the imperative, "Do the right thing," in the sense that it's unreasonable to ask what the right thing to do is unless you care enough to act on that knowledge. As for the last category, this would be things like:

Is this a question? / Q: Is Q a question?

What is the first question on the Test?

(Although that last of the last things, here, is a riddle "too"...)

  • Evidence is about justification, not explanation. Euclid's axioms may be self-evident but not self-explanatory, ten commandments may be self-explanatory but not self-evident . I do not see how ability to ask questions or question formats are at all relevant, both are malleable and have little relation to the substance of what is asked.
    – Conifold
    Commented Nov 7, 2020 at 21:48
  • I was looking for a criterion of genuine "self-evidence," etc. Not "unquestionable" propositions, but ones whose epistemic function depends purely on erotetic concepts/abilities. These would seem to be the only candidates for things that don't "beg the question," arguably, whatever else may be said about other "axioms." (Or so it seems to me.) ---And self-explanation should at least be a cousin or other relative of self-evidence (in Wittgensteinian family-resemblance terms, you might say). Commented Nov 10, 2020 at 22:48
  • 1
    If you mean something that carries its evidence entirely within itself I do not think that works, everything must beg the question. Any question relies on external concepts used in it, and inferential relations they enter that depend on other "axioms". This is why cogito does not work the way Descartes intended. Explanation may facilitate finding evidence, but conceptually the two are very far apart, I think. A question can be as clear to us as morning dew, while we have no clue what the answer is, and evidence can leave no room for doubt, while we have no clue why things are so.
    – Conifold
    Commented Nov 10, 2020 at 23:38


You must log in to answer this question.