One of the multiple-choice options that close voters are invited to select as justification for voting to close a question posted on English Language & Usage emphasizes the absence of evident research in the body of the question:
Please include the research you've done, or consider if your question suits our English Language Learners site better. Questions that can be answered using commonly-available references are off-topic.
Whatever the merits of this close reason, it doesn't on its face foreclose the possibility that someone other than the original poster will try to satisfy the identified shortcoming by adding research to the question. In doing so, of course, the person making the addition may be criticized for intentionally or inadvertently altering the nature and direction of the original question—and thus frustrating the OP's attempt to get a suitable answer. As a commenter put it yesterday,
The value of mindreading askers and adding research they never performed is a question for another day.
But let's make it a question for today: Should English Language & Usage forbid third parties from adding research to questions that are subject to closure for lack of research, on the ground that such additions necessarily involve guesswork about and approximations of what the OP wants to know, and on the ground that such editing may shanghai the OP's original question and refocus it in a direction unintended by the OP? Are there other good reasons to adopt a policy of leaving posters to sink or swim on their own when trying to ask questions that EL&U close voters find acceptable?
I admit that I have a strong bias in favor of intervening to save and improve potentially interesting questions. In part that's because, as a copy editor, I make my living reading the minds of authors who have expressed themselves ambiguously or incompletely, and trying to express their thoughts more clearly and seamlessly—often to the extent of adding or citing research that they have omitted. My experience (or professional hauteur, as the case may be) leaves me unapologetic about the mindreading aspect of the job; after all, I'm not the only reader who will have to struggle to make sense of the ambiguity or gap if the text remains as it is. And if I get it wrong, the author can set things right easily enough by clarifying what his or her intentions were.
But my bias in favor of intervention also comes in part from a sense that what matters long term at EL&U is recognition and refinement of questions worth answering—not careful adherence to the limits of a poster's original conception of the dimensions of the question posed. If a poster asks a question that has very interesting implications, but the poster exhibits no awareness of those implications, I think it is counterproductive to insist that we adopt what we take (using our mindreading powers again) to be the poster's narrow conception of the question—especially when doing so will lead us to close the question as unfit for the site.
My views on adding research to potentially interesting questions are similar: Why should we think that our primary duty in dealing with a question that lacks research is to uphold the poster's autonomy and integrity in having chosen not to include research? If the long-term goal of accumulating a full roster of useful answers to interesting questions is central to EL&U's reason for existing, I think we should take that goal seriously and do what we can to promote it. And I think we serve that goal better by making the improvement of potentially good questions our primary focus than we do by ensuring that we do not stray from strict fidelity to the question asker's original (and quite possibly myopic) intent.