9

(tl;dr version: "Primarily opinion-based" is designed to stop "What's the best X?"-type questions. It's not supposed to be used on every single question that involves opinions.)

Until last year, one of the reasons that questions could be closed was because they were "not constructive":

As it currently stands, this question is not a good fit for our Q&A format. We expect answers to be supported by facts, references, or expertise, but this question will likely solicit debate, arguments, polling, or extended discussion. If you feel that this question can be improved and possibly reopened, visit the help center for guidance.

In 2013, the question closing process was revamped across the SE network, in an effort to make it appear less hostile and to more effectively help users improve the quality of their questions. Several existing close reasons were retired, and new ones were introduced in their place. The new option closest to "not constructive" is "primarily opinion-based":

Many good questions generate some degree of opinion based on expert experience, but answers to this question will tend to be almost entirely based on opinions, rather than facts, references, or specific expertise.

While this change was entirely well intentioned—after all, nobody likes to be told they're not being constructive, and it hardly helps them improve—an unintended consequence of the change is that "primarily opinion-based" is being used at ELU to close questions it was never designed to close. To understand why, we need to look at why we had a "not constructive" close reason to begin with, as well as our own guidance about what constitutes a constructive question.

The archetypal "not constructive"/"primarily opinion-based" question is "What is the best [car, game, pet, movie, dictionary, study method, etc.]?" There is no right answer, and it can only lead to rep-pimping and mental masturbation, as everyone and their dog stops by to leave their two cents and shoot the breeze. It was widely apparent fairly early on in the history of SE that these questions have no place on the network, and it's easy to see why. Questions like this one are cute, and they can be very popular, but if they start to absorb all the energy on the site, real questions suffer. (Incidentally, this is a big part of the reason the "community wiki" option exists: it provides a mechanism to stop users from gaining thousands of undeserved rep points for lazily pasting their favorite cartoon into an answer.) Our FAQ provides an excellent list of the kinds of questions that are not constructive:

To prevent your question from being flagged and possibly removed, avoid asking subjective questions where …

  • every answer is equally valid: “What’s your favorite ______?”
  • your answer is provided along with the question, and you expect more answers: “I use ______ for ______, what do you use?”
  • there is no actual problem to be solved: “I’m curious if other people feel like I do.”
  • you are asking an open-ended, hypothetical question: “What if ______ happened?”
  • your question is just a rant in disguise: “______ sucks, am I right?”

Unfortunately, many people at this site think "primarily opinion-based" should be used to shut down discussion of any and all topics having to do with opinions about English. To select the most convenient example: this question, an intriguing and well-written query about the reasoning behind linguistic prescriptivism, currently sits with 4 close votes, and may in fact be placed on hold before I finish this meta question. All four voters in favor of closure have chosen the "primarily opinion-based" option. But why? The user isn't asking whether prescriptivism is better than descriptivision. He clearly doesn't care for prescriptivism, but he's not just saying "Prescriptivism sucks, am I right?" It's a legitimate question about the philosophical underpinnings of prescriptivism. It should be entirely possible to formulate a comprehensive and useful answer that invokes Robert Lowth and Strunk & White and explores the role of Greek and Latin antiquity in forming the opinions and assumptions of the learned class in Britain and America in the eighteenth and nineteenth centuries and so forth. Instead, people see the title of the question—"What is the point of prescriptive grammar?"—and the "primarily opinion-based" light bulb lights up in their brains, and the close votes pile up. And that's not right. Just because a question asks about opinions does not mean it is asking you for your opinion.

It should be obvious to everyone here that a great deal of English usage boils down to subjectivity and opinion. Register, formality, and etiquette are vital components of usage, yet talking about them unavoidably means talking about why one word or phrase is considered "better" than another in certain contexts in the greater society, or in or between various subsets of it. Linguists have spilled barrels of ink exploring exactly these questions in classrooms and in scholarly articles. To suggest that they should not be asked here is insanity.

According to our FAQ, constructive subjective questions:

  • inspire answers that explain “why” and “how”
  • tend to have long, not short, answers
  • have a constructive, fair, and impartial tone
  • invite sharing experiences over opinions
  • insist that opinion be backed up with facts and references
  • are more than just mindless social fun

Before voting to close a question as "primarily opinion-based," ask yourself whether it meets these criteria. If it does, leave it be. Try assuming that when the questioner asks something like "When should I say X instead of Y?", they're inquiring about cultural norms, not just bending your ear about what you, personally, think about it. If necessary, edit the question to clarify it. A word or two in the right place can save a question from the garbage and turn it into something that enlightens everyone.

Disclaimer: The aforegoing is all just, like, my opinion, man.

  • 7
    To me, that question reads exactly like a rant-peeve question. “Prescriptive grammar makes no sense, so what do people want with it?” seems to me to be its basic tenet. That’s why I closevoted. Even if it were worded more exploratively, I don’t think it would really be particularly on-topic on ELU. Prescriptivism and descriptivism are not English phenomena as such—they’re quite universal. I’d say it would be a good question for Linguistics (if less peevy). – Janus Bahs Jacquet Jun 4 '14 at 17:32
  • 2
    Yeah, I saw that thread over there. Too bad it got closed, imo. I thought it could be a good thread for a rainy day--if people could be civil and actually make constructive comments. I didn't think it would be any worse than all those is-there-a-word-or-phrase-to-replace-a-100,000-word-novel-and-having-the-same-meaning type questions. Anyhow . . . – F.E. Jun 4 '14 at 21:00
  • 6
    "Opinions about language are as interesting as opinions about arithmetic." -P. J. O'Rourke. – John Lawler Jun 4 '14 at 22:12
  • @JohnLawler - Is that good or bad? – RyeɃreḁd Jun 5 '14 at 13:56
  • Just pointing out that a great many people believe that grammar is a matter of personal opinion, and that the opinions vary greatly. – John Lawler Jun 5 '14 at 14:26
4

The purpose for closing questions is to remove questions that are not helpful for future visitors or would be extremely difficult to answer in the StackExchange format. "Opinion" questions can bump into the latter problem and the relevant question to ask oneself when deciding whether to close vote or not is:

Will this question provoke answers that are good, high quality answers?

With this frame of mind, here is how I would address your specific points:

In 2013, the question closing process was revamped across the SE network, in an effort to make it appear less hostile and to more effectively help users improve the quality of their questions. Several existing close reasons were retired, and new ones were introduced in their place. The new option closest to "not constructive" is "primarily opinion-based".

One thing to remember is that the purpose of closing questions remained the same. The reasons associated with closing a question changed in an effort to help users understand what questions to close; why a question was closed; and how to avoid posting questions that will be closed. The matter-of-fact tone behind "Not Constructive" was causing issues, yes, but it wasn't an invalid close reason. It just wasn't palatable.

While this change was entirely well intentioned—after all, nobody likes to be told they're not being constructive, and it hardly helps them improve—an unintended consequence of the change is that "primarily opinion-based" is being used at ELU to close questions it was never designed to close.

And that is a problem... why? If the community decides that a set of questions isn't worth having on the site, is that... wrong?

To understand why, we need to look at why we had a "not constructive" close reason to begin with, as well as our own guidance about what constitutes a constructive question.

The archetypal "not constructive"/"primarily opinion-based" question is "What is the best [car, game, pet, movie, dictionary, study method, etc.]?" [...]

I basically agree with this paragraph and only want to point out that the close reasons are the result of wanting to close bad questions. If a bad question exists and there isn't a valid close reason with which to close it, the close reason aren't doing their job.

Likewise, if a good question gets closed because it matches the letter of the law and not the spirit of the law, that is user error and the question should be reopened. StackExchange already has a contingency plan in place for this scenario. And it happens across all close reasons.

But in the end, the purpose is to close a bad question. In my opinion, it is better to close a bad question with a slightly-not-correct close reason than it is to leave the bad question open because there isn't a close reason that perfectly matches.

Unfortunately, many people at this site think "primarily opinion-based" should be used to shut down discussion of any and all topics having to do with opinions about English.

There really isn't any data backing this statement. At the very least, it wasn't presented in your post. It takes 5 people to close a question and only 5 to reopen one. If one particular group of 5 made a mistake than it should be rather easy to drum up support to get a question reopened.

But the idea that some massive group of users are closing good questions because they are misinterpreting a close reason seems unlikely. The methodology should be, "This is a bad question. Which close reason explains it best?" Not, "Here is a close reason. Which questions match this reason?"

So if there is a group of people out there "doing it backwards" then the lesson should be to teach them the appropriate approach. There certainly isn't anything unique about this close reason compared to all of the other close reasons you have posted about.

So what I consider more likely to be happening is that a bunch of people read the question and think, "This should be closed." Then they go through the awkwardly worded close reasons and chose the best option -- even if it doesn't quite technically fit.

To select the most convenient example: [...]. All four voters in favor of closure have chosen the "primarily opinion-based" option. But why?

Because it isn't asking a question that can be answered appropriately on this site. Theoretically, what would such an answer look even like? And who would be the one doing the answering? And how would the community vote on such a question?

Some of these concerns could have been avoided if the author had asked the question in a way that made it less of a loaded question. But they didn't and it was, therefore, closed.

The user isn't asking whether prescriptivism is better than descriptivision.

No, they just stated their opinion that it was better and more or less delivered an ultimatum to prescriptivists to respond to their argument. This kind of debate is not appropriate for ELU.

He clearly doesn't care for prescriptivism, but he's not just saying "Prescriptivism sucks, am I right?"

"Prescriptivism isn't logical, am I right?" is in the same family of problem questions.

It's a legitimate question about the philosophical underpinnings of prescriptivism.

Are you suggesting that philosophical underpinnings of linguistic theory is on topic at ELU? We aren't actually here to debate linguistic theory. We are here to answer question about English and its usage.

It should be entirely possible to formulate a comprehensive and useful answer that invokes Robert Lowth and Strunk & White and explores the role of Greek and Latin antiquity in forming the opinions and assumptions of the learned class in Britain and America in the eighteenth and nineteenth centuries and so forth.

And then we have to ask if such an answer would be appropriate here. Questions that are too broadly focused are another problem that can result in closing. Your suggestion of a hypothetical answer seems far too involved for ELU.

Instead, people see the title of the question—"What is the point of prescriptive grammar?"—and the "primarily opinion-based" light bulb lights up in their brains, and the close votes pile up. And that's not right. Just because a question asks about opinions does not mean it is asking you for your opinion.

You seem to jumping to this conclusion. There is nothing to suggest that people voted to close this question because of its title or without reading the entire post. Furthermore, your assumption that they closed it because of how the close reason is worded is just as unfounded (see above).

It should be obvious to everyone here that a great deal of English usage boils down to subjectivity and opinion. [...]

And those questions typically remain open and receive appropriate answers and discussion. The fact that such questions survive your supposed close-vote-gang implies that such a gang may not exist. Or, if they do exist, the more level headed contributors simply reopen the closed questions and all is well.

[...] Linguists have spilled barrels of ink exploring exactly these questions in classrooms and in scholarly articles. To suggest that they should not be asked here is insanity.

You appear to be assuming that all topics worth publishing are appropriate questions for this site. I don't understand why you would hold such an assumption.

Before voting to close a question as "primarily opinion-based," ask yourself whether it meets these criteria.

Before voting to close, ask yourself whether a question is good or bad for the site. The quoted notes on what makes a good subjective question are great examples of "good" questions and should absolutely be remembered and considered when choosing to vote a question closed.

But once a question has been weighed and determined bad, it should be closed. Backing up and deciding against the closure because the admins have worded the close reasons a particular way is a grave mistake.

  • 4
    "The methodology should be, 'This is a bad question. Which close reason explains it best?'" - I couldn't possibly disagree more. Treating close reasons as ex post facto rationalizations for ill-defined gut feelings about certain questions simply being "bad" turns the whole process into a farce. – phenry Jun 11 '14 at 18:13
  • 5
    @phenry: The opposite is worse and leads to exactly the behavior you think you are seeing. "Here is a hammer that closes questions. Time to find some nails!" I am not suggesting people willy-nilly close questions they don't like. There are still very helpful guidelines to determine whether a question is good or bad. The point is that the close reasons were created from those guidelines. The guidelines came first and the guidelines should be the authority. – MrHen Jun 11 '14 at 18:17
  • My entire purpose for writing this post and similar ones has been to examine what our guidelines actually say (and why), in comparison to how they're being applied. If that is to be the ground on which this discussion is to take place, I'm delighted. As far as I'm concerned, we're there already. – phenry Jun 11 '14 at 19:57
  • 1
    "It takes 5 people to close a question and only 5 to reopen one." Not strictly true, I have seen questions closed by one or two people. Often a user votes to close a question, a mod agrees and puts the question on hold. And a question is easier to close than to reopen, very few get a second leash of life, and if they do, it's too late. The expiry date on a question is shockingly short. – Mari-Lou A Jun 12 '14 at 6:08
  • @Mari-LouA: Correct, mods have the ability to immediately open or close and they just recently changed the rules to let people with Gold badges for a tag to immediately close as well. A question being easier to close than reopen is mostly likely evidence that most of our closed questions should remain closed. – MrHen Jun 12 '14 at 13:18
  • @MrHen "A question being easier to close than reopen is mostly likely evidence that most of our closed questions should remain closed." I don't understand this. Would you explain why you think so? – ivanhoescott Feb 6 '15 at 14:54
  • 1
    @ivanhoescott: It is easier to get people to agree to the question, "Should we close this?" versus "Should we reopen this?" A question being easier to close than reopen is most likely evidence that [those with permissions think] most of our closed questions should remain closed. If a question should not have been closed then those people would not have closed it. – MrHen Feb 6 '15 at 21:11
  • @ivanhoescott: I'm trying to be concise for the sake of comments but I don't have lots of time for an extended chat conversation right now, sorry. Short version: If more questions should be reopened we have plenty of people who can reopen them. They have chosen/voted not to. So it stands to reason they should stay closed (in the opinion of those who have the power to reopen). – MrHen Feb 10 '15 at 15:25
  • PS) We can always take this to English Language & Usage Chat if you want to talk in more depth. – MrHen Dec 5 '15 at 22:49
  • @ivanhoescott Try this: chat.stackexchange.com/rooms/32582/mrhen-ivanhoescott – MrHen Dec 6 '15 at 16:15
3

As I understand it, the crucial point in phenry's excellent post is that questions that ask "What is the rationale for X?" are not asking "What is your opinion of X?" Rather, they are asking for factual information—assuming that a rationale for X actually exists.

If I ask "What is the point of the subjunctive mood?" and make clear in my description that the question isn't merely one of rhetorical annoyance but a request for a clarification of what purpose the subjunctive mood serves or is supposed to serve, I am not asking a POB question. My question may still be inappropriate for EL&U, for some other reason, but rejecting it as "primarily opinion based" seems to me to misunderstand the nature of my question and to misapply the extremely valuable POB justification for closing "What do you think of X?" questions.

I would hate to jeopardize the tone of civility that prevails at this site by inviting pointlessly divisive discussions of controversial subjects. But on the other hand, it seems regrettable to limit the site's scope and usefulness because, at some level, we don't trust ourselves to behave well and to stick to facts when facts are asked for.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .