It's interesting to me that in the comments beneath Mr. Shiny and New 安宇's answer, no one seems particularly enthusiastic about the "no lists rule" and yet the consensus is that we must obey and enforce it.
Many worthy questions may seek answers that are, in effect, short lists. From one point of view, that is essentially what single-word requests are—invitations for answerers to list (and sometimes briefly comment on) the most appropriate words to match the questioner's criteria. Though I'm not a huge fan of single-word requests, they are obviously one of the most popular types of questions on this site, perhaps because coming up with one or two or three relevant suggestions is easy and fun.
I don't see any crucial difference between a question formulated as "What word means X?" (the standard single-word-request formulation) and "What word or words in Category X of words satisfies/satisfy neither criterion Y nor criterion Z?" (the formulation used initially by the OP of the question at issue here). I can easily imagine questions of either type that may have excessively many "correct" answers— such as "What word means 'fast'"? or "What nouns begin with neither the letter a nor the letter b?"—but in those instances, the more apt criticism of the question is that it is too broad or too general, not that it asks for a list.
I agree with Jay's comment above that a question that asks for one or more examples that satisfy a narrowly defined set of criteria is distinguishable from a request for an exhaustive list of such examples. It ought to be possible for users of this site to ask and answer such questions without first having to contort the questions so that they can't possibly be interpreted as list requests. What matters most is the value of the question and of the answers it draws, not whether its form can be interpreted as offending some arbitrary but longstanding taboo. Common sense is a better guide.
Followup Remarks (10/13/2014): At this moment, the question Other ways to say "I'm rooting for you?" has attracted 48,701 views over the three years since it was asked, and it is likely to garner several hundred more views in the next five days, since it carries a bounty of 100 points for most of the next week, offered because "This question has not received enough attention."
The body of the question (as opposed to the head already cited above) is "What are other ways one can say that have the same meaning as, 'I'm rooting for you?'" I see no evidence of any research effort on the poster's part, and both the head and the body of the question look very much like requests for a list of phrases similar to "I'm rooting for you." So if I were looking for an instance of a minimally researched list request to make an example of by voting to close it, "Other ways to say 'I'm rooting for you?'"—odd punctuation and all—would be a great candidate.
And yet... and yet... There is something deeply troubling about trying to disqualify a question that tens of thousands of site visitors have cared enough about to view. This is the fundamental contradiction that (in my view) haunts EL&U with regard to list questions and single-word requests: Questioners love to ask them, answerers love to respond to them, and site visitors love to read them. In fact, the only drawback of such invitations to play Name (or List] That Synonym is that they have so little to do with what I imagine are the core concerns of linguists, etymologists, and serious English language enthusiasts: thoughtful, informed investigations of grammatical structure, word history, and the like.
We are extremely inconsistent in our barring or not barring of questions that ask for lists, just as we are extremely inconsistent in our barring or not barring of questions whose askers have shown little or no effort to research a question before posting it. Our inconsistency doesn't mean that we should abandon any attempt to bar such questions; but it does mean that a person trying to figure out whether a question is good enough to post, based on what is already up on the site and hence (effectively) approved, will see many misleading examples of what we apparently consider "good enough." And if their question gets shot down because it asks for a list or doesn't show adequate research effort, they are likely to feel that the decision to close their question was based on arbitrary and unfair selective enforcement. Viewed objectively, it seems to me, they wouldn't be far wrong.
How can we improve? I favor two steps: rectification of names (identifying reasons for closure that reflect what we actually don't like about the flawed questions), and consistent enforcement (closing all questions that run afoul of those reasons). From other discussions here on Meta, I've gotten the impression that some descriptively accurate reasons for closing bad questions (such as "Too Elementary for Our Site; Try Asking at ELL" or "Already Answered in the Comments and Not Likely to Draw Any Further Response" or "Too Specific to Be of Interest to Anyone but the Poster") may be deemed too harsh in tone or in some other way inappropriate for the EL&U.
But if the alternative is to get rid of bad questions by selectively playing the "Too General" card or the "Insufficient Effort by the Poster" card, we're not doing the rejected posters or ourselves any favors. We don't help the posters understand what exactly was objectionable about their questions that was not objectionable in various equally general or equally unresearched questions that we've let stand. And we don't help ourselves by giving prospective future posters of bad questions a clear sense what our standards are—a shortcoming that encourages posters to hope that they may be one of the lucky ones whose questions don't get closed despite (for example) minimal research effort.