I have in mind two questions, largely the same in content and approach, but for which I offered opposing opinions on voting to close:
Is this a known Interpersonal “Anti-Skill”, if avoided could de-escalate conflict—or could it be original? I call it “barking” was a request for a word (or verification of whether or not a suitable word exists) which I voted to close.
I commented on this in the question itself and the associated post on meta, but my essential position was that word requests are off-topic here, and the question could only be definitively answered by someone who happened to have appropriate subject matter knowledge which is not really associated with this stack. I further thought that the homemade descriptions used in the question would make it difficult for the question to be useful in the future. There was a fair amount of IPS members that felt the question was on-topic.
Male-male dominance/status challenge is essentially the same type of question. But in this case I happened to have the requisite background to know the answer.
I think that we should consider codifying a stack-wide policy on these sorts of questions. My main issue is that, with the right background, these questions may be on-topic in the end but that determination can only be made post hoc. A collective "No one here knows" is not a definitive answer, and so a question like the former example might dangle forever, unresolved.
In the case of the latter question, that I could supply the answer gives it (potentially) more value to future users as it now contains the correct term and (in my opinion) sufficiently general descriptions in the question itself that it might be found by someone interested in the topic.
There's a problem with questions that are valid for the stack only in the case of a correct answer being provided-- we've little ability to know in advance if this will ever happen, and so a question being invalid due to the lack of an answer at some particular moment might be closed even though it could become valid despite the question not being able to stand on its own. We won't be able to know which questions belong to which category and so dealing with this on a case-by-case basis seems like it will be messy.
My personal feeling is that we should formally decline these sorts of questions. The alternative would seem to be that lots of hard-to-value questions (not necessarily valueless) will suffuse the site. A "consensus" approach, as we seem to have now, will lead to cases where questions I can't answer seem like they should be closed, while those I can will seem "good enough". We presumably don't want any boat-programming style questions, but the line between sufficient and insufficient relation to IPS seems unclear.