Must every questioned 'notable claim' include a referenced citation with a quote?

Question

FAQ: Welcome to New Users links to How should we enforce notability? in which the accepted answer includes the following:

It is not required to have an explicit cite for a claim in questions, though it is very helpful if the notability of the claim is questioned.

and

If you suspect that a claim is not notable, you should perform a minimum amount of research before acting. Performing a quick google search using search terms taken from the question is a good idea.

Is it true that an explicit citation is not required in questions? Even if it wasn't required in the past, should it be required in the future?

People have difficulty answering the question when there's no citation.

A typical example is Can the UK fire its nuclear weapons without US agreement?

The problems, the reasons why I think that all claims should include a reference by default, are:

• When a user posts an unreferenced claim, then the claim as it's phrased is often very brief, vague or imprecise ... which means that people need to use comments to question what the question really means and exactly what it's asking.

• If you do a quick Google search then you may find many slightly different variations on the claim, and it's not possible to know which one to answer

• The person who is asking the question should be able to do the Google search themselves, if the claim is notable. Their doing that minimum of research improves the question; but if/when someone else does the research and edits the question, that is a perhaps unwarranted change to the meaning or intent of the question.

In contrast, a referenced claim is better:

• The original reference is longer (with more text), so that people can read it to understand the context, the exact words being used (not a semi-accurate paraphrase which is merely the understanding/memory/summary of the person asking the question here) and therefore the implied or exact meaning of the claim.
• If or when the meaning of the original reference is unclear, nevertheless the reference itself is then ipso facto on topic (the fact that it's referenced makes it notable almost by definition, except when it's from a joke or fictional source which nobody was meant to believe). The lack of clarity or ambiguity then becomes on topic, and deserves being answered in an answer instead of being questioned in a comment.

Alternatively, is it appropriate to simply flag a moderator if you think that an unreferenced question should have a reference: instead of posting a comment to the user, which the user may not reply to? Moderators have the ability to post a standard/canned/boilerplate comment, and/or to put a question on Hold, which may be better than (or supplement) a user comment.

I don't understand why the "rule" is ambiguous: why or when is it "should" (usually) provide a reference in questions, instead of "must" (always) provide a reference?

See also FAQ: Must all questions be notable?

Answer 1

As agreed in that answer, it is essential that a claim be notable. It is important, but not essential, that the claim be represented with an example.

For all the reasons you give, and more, it is important to have a notable reference:

• Makes the question more specific, which in some cases it's necessary
• Helps identify strawmen, where the OP has misunderstood the original claim.
• Demonstrates notability in borderline cases

Therefore, it is better to have a notable reference. Where a question without a notable reference suffers from any of the above problems, and only where there is sincere doubt about the notability of the question (for example, after performing a google search and not finding examples) I think it is appropriate to vote to put it on hold until a reference can be found. (Slight tweak to your suggestion: The number of non-mods who have the privilege to vote to put a question On Hold outnumbers the mods. I would encourage them to directly vote to close, rather than flag.)

However, there are oral memes as well as written ones, and I am hesitant to exclude those as valid questions.

Here are some examples (and, yes, some of them suffer badly from a lack of agreed definitions, but there is no doubt the claims are notable):

• Does an "apple a day keep the doctor away"?
• Will a mother bird abandon her young if touched by a human?
• Are vegetables good for me?
• Will sticking a knife into a toaster electrocute me?
• What is the average number of legs per human?

The hurdles for asking a valid question on Skeptics.SE are high, and we almost certainly lose potential new users because of it. I don't propose that we should let through bad questions just to make it simpler, but I am wary about making the hurdles even higher.

In fact, we should struggle to make asking valid, notable and answerable questions as easy as possible, by collaborating with people asking questions rather than simply enforcing a rule.

If no-one but the OP has heard of a claim or if there is doubt about the meaning of the claim, let's push back (or find a suitable reference ourselves). But, where the meaning is clear enough and the claim is widely heard, there seems little reason to refuse the question.