This question began with my curiosity about the close votes and decision on this recent question:

Help me figure out the mode of this Howard Shore piece

The rationale given for closing is as follows:

Basic analysis questions, such as "What key is this song in?", are off-topic. Questions should be substantial and refer to a well-defined work or subsection, including a concrete reference (sheet music, etc.).

In trying to understand the close decision, I've also looked on Meta at other previous questions (and their answers) recently on close votes, particularly these two:

How "basic" is basic for "basic analysis question" from the off-topic close description?

On closing vs. tolerating weak questions

Note that I am including this information to show that I have apparently done enough previous "research" to try to avoid this Meta question being closed while I'm asking it, though maybe this isn't enough.

I'm not being sarcastic in the previous sentence. I am genuinely concerned overall at the number of questions closed here, particularly for new users (as in the question I cited at the outset). I think, without sufficient help to the new user, it can be an incredibly unwelcoming and discouraging process. And it should only be done if a question violates a serious community guideline, if the user is unresponsive to suggestions for improvement (and not for the merely 30 minutes (!!) suggested in one of the answers in the meta question I linked), or if the question is obviously inappropriate for this site or the StackExchange format in general.

I know that most StackExchange communities follow general guidelines across the site for avoiding opinion-based questions, avoiding list-based questions, etc. I completely get and support those things. But I'm increasingly confused by the criteria on this site, where I believe we need to support users of all ability levels.

Let's look at that rationale cited above again:

Basic analysis questions, such as "What key is this song in?", are off-topic. Questions should be substantial and refer to a well-defined work or subsection, including a concrete reference (sheet music, etc.).

Now, it is true that this question is literally asking about the key (or mode) of a piece, though my sense is that this guideline was not declaring that such questions are always off-topic as much as giving an example of a case where they might be. I can understand if the question can be answered by simply looking at a key signature in a score, or listening to a recording in of the piece and simply matching up the audio sound with the most prominent initial or final chord or whatever. I would agree that such a question is generally not "substantial."

But let's examine the criteria cited. The question should:

  • not be a "basic analysis" question
  • be "substantial"
  • "refer to a well-defined work or subsection"
  • "includ[e] a concrete reference (sheet music, etc.)"

In this case, OP referred to a well-defined work and identified the section in the question (essentially the first page of the sheet music). The OP included sheet music as a "concrete reference." Given that I can guarantee I've known many students who have taken four semesters of college music theory and would have trouble trying to sort out what was going on harmonically in the first page of this piece to talk about the key/mode, I'm pretty sure this isn't a "basic analysis" question.

So, the only criterion to meet seems to be whether the question is "substantial." By looking at previous Meta questions on what constitutes "off-topic," it seems what makes a question "substantial" is often the questioner demonstrating prior research or attempts to solve the issue.

In this case, the OP demonstrated some basic knowledge of mode (though misspelling Mixolydian) and made a justifiable -- though, I think ultimately incorrect -- attempt to classify the mode in the second page of the score. The OP noted some tendencies of the composer with regard to mode usage (demonstrating at least an attempt at prior research). The OP further noted some details about the first page of the score (the focus of the question) and OP's attempts to understand what sort of scales it might fit into.

To me, this demonstrates that OP made a good faith effort attempting to answer a question that, as I said, would be challenging even with people who have much more significant theoretical knowledge. A more knowledgeable person may have framed the question slightly differently, realizing the first page was essentially in some D-based key/mode with harmonic excursions elsewhere, but again, that's a pretty "substantial" analytical question, not a "basic" one.

And yet this question was closed.

As far as I can tell, the reasons why this was closed might be either that this nebulous "substantial" qualifier was used to jettison the question and/or the specific example given in the stereotypical rationale (i.e., "such as 'What key is this song in?'") was used to dismiss the concern as too "basic" for this site.

If the latter was the problem, I think we may perhaps need to reword that stereotypical boilerplate rationale: while asking what key/mode a piece is in is generally an elementary question, it can sometimes not be obvious at all.

Comments also identified some concern about whether the question was about the "whole piece," as if that's a problem. The boilerplate only mentions a "well-defined work or subsection," which the question clearly is about (i.e., primarily the first 14 bars, also mentioning the next few bars, which by the way, does NOT constitute the entire "piece" from the movie, as the first two pages of music is not the entire track). But a "well-defined work" implies one can ask a question about an entire piece. (Again, ambiguity in language that makes it difficult for a new user to navigate these rather absurd and complex requirements just to not have a question closed.)

Let us note what is specifically declared on-topic here:


That list includes: "technical analysis of a specific, complete work, or well defined section thereof."

Please note that nowhere in that guide is there anything mentioning that questions about identifying keys or modes are off-topic, even though OP's question was closed. What the guide to "on-topic" questions does say is that "identifying a ... chord progression" or "identifying notes/chords/meters/other elements in songs" can be off-topic. Given that those are potentially "technical analysis" question that conflict with the "on-topic" criterion I cited in the previous paragraph, I assume the "substantial" qualification is the one we're using to differentiate whether a particular "technical analysis" question is on-topic.

Also, the issue of the difficulty of identifying a single key in ambiguous situations was brought up explicitly in a question last year:

Can a song be in both a major and minor key at the same time?

More recently, another question that essentially boiled down to "what key is this in?" got closed and reopened, with Richard in comments wondering aloud about the "substantial" criterion:

How does one analyze a song that moves primarily between two non-diatonic chords?

One prominent user apparently also felt the need to reword the title of the question in editing it to get it out of the "what scale/key is this in?" genre of question, even though literally nothing in our on-topic guidelines says that such questions are off-topic. Literally no user will be aware of this apparent (arbitrary) criterion unless they've cast a close vote (which new users obviously can't do) or have had their own question closed.

To me, these sorts of actions indicate a potential obsession with criteria about what may be "off-topic" that is unhealthy. (Please note -- Aaron, if you're reading this, I'm not criticizing you for your edit. I appreciate your attempt to improve the question. I think you've made a good faith effort to follow community guidelines as you understand them, but the guidelines are themselves unclear and perhaps overbearing.)

We do have a lot of analysis questions -- that are generally NOT closed -- on this site that are effectively "What is this chord?" in the context of some piece. I'm not certain why "What is this key?" questions are targeted for closure except for the fact that that particular concern is used as an example in one of the boilerplate close rationales. At a minimum, I think we should consider changing the wording of that boilerplate close rationale to either eliminate that example, clarify it, or (my preference) give several examples of what might be "basic analysis" questions -- preferably also with some sort of qualifier or explanation that emphasizes only basic identification of a key/mode or whatever (not questions that may lead to more advanced theory concepts just to answer such a "basic" question).

In the discussion on Meta that apparently was used to create the boilerplate, originally two examples of "basic" questions were given, but for some reason that was trimmed down to only one, which has now apparently become a "gotcha!" reason for closing questions from new users.

Wording of the close reason for "bad" analysis questions

But in general, I'm also concerned about this nebulous notion of a "substantial" question, particularly when applied to new users on the site. I come here to help people answer their questions. I want new people to feel they can ask for help, and getting your first question marked as "closed" because it's off-topic when you actually take the time to do most of the things the site asks of you (providing sheet music example, referencing the specific piece/section, showing that you've tried to solve the problem yourself) can be a severely disheartening experience.

I've made an example mostly of one recent close decision, but it's a pattern I've seen elsewhere in quite a few closed questions, where citation of some technical (yet ambiguous) close criterion, implicit declaration that some question is not "substantial," or judging of a questioner based on lack of knowledge or insufficient English language skills results in summary dismissal of a question.

While many questions are apparently too "basic" or not "substantial," others are often judged "too broad." The latter can be a legitimate problem, but I also think it's possible for many supposedly "broad" questions to write an answer with a basic introduction to a topic in a few paragraphs that gives the questioner a place to begin to understand the topic. I myself have answered quite a few questions on this site that no one bats an eye at, but from my perspective are incredibly broad as they intersect with stuff that may have been covered in dozens of specialist books and articles in music theory. Yet, I know the question is likely an unknown topic outside specialist circles, so I answer in a few paragraphs and try to sum up the stuff from the field. (Please keep in mind that just because YOU know a lot about a topic and think a question is hopelessly "broad" doesn't mean that a summary intro to the concept can't be created in a few paragraphs in an answer, including perhaps some links to the user for more information when they're ready for it.)

"Broadness" is often another way of saying a question isn't "substantial," because a beginner may not know enough to ask a more specific question, so instead is asking a wide-ranging one that lacking in some details. (What's even more disconcerting is the number of "Sorry, but you're an idiot for asking that question -- if you knew anything, you'd realize how impossible it is to answer" or "If don't know anything about X, you're obviously not ready to think about the question topic" type of comments with that sort of tone that sometimes come at new users.)

As a community, I think we should reflect long and hard before closing questions where someone has clearly made some effort to think about a question and craft it according to most of the rules of this site. (Frankly, I think we should be more patient with new users in general, asking for clarification or more details if they don't hit all of our criteria for "substantial" questions, rather than reflexively hitting the "close" button.) As I said, I can understand that there are certain kinds of questions that StackExchange in general acknowledges as unanswerable (opinion-based, list questions, etc.), and I can understand that certain types of requests may be difficult to handle here and declared off-topic or the purview of another site (e.g., the stuff that now gets referred to the other Music Fans).

Closing a theory or analysis question immediately from a new user, from my perspective, should be a last resort unless there's an obvious duplicate (not just a "duplicate" that's vaguely on the same topic and may not answer the specific details brought up in the question) or a clear violation of general StackExchange principles that are designed to prevent vague, opinion-based, or unanswerable questions. If the user doesn't respond to requests to fix the question within several days, and it's still a bit vague, broad, off-topic, etc., then I understand why mods may choose to close. But I think our current close criteria are themselves quite vague and broad -- not to mention internally inconsistent and non-transparent, as I've pointed out above -- making it rather difficult for new users (particularly without a technical background) to navigate.

I'll conclude by citing an answer from a quite old Meta thread (bolding in original):

Music SE suffers from trigger happy closers

A strong SE community needs expert users, both experts in the subject matter (music, here) and experts in SE itself. A strong SE community also needs helpful, friendly users.

When new users post poor but salvageable questions, the best approach is to improve the question and the poster. Be friendly, be helpful, and encourage the users to salvage their own questions through better research effort and editing. If you succeed, you’ll have a better question and a user who is better equipped to contribute to the community if they stick around.

Personal note: If I continue to see these efforts to shut down the voices of those -- particularly new users -- who are just trying to learn about music theory (even if they may not have a lot of knowledge yet), I may reconsider whether it's worth my time to continue to contribute here.

  • 2
    I thought I'd wait a few days. Since not much has been said yet, other than a few upvotes, let me just clarify that I am sincerely appreciative to the work moderators on this site do, as well as those who volunteer their time to cast votes, etc. I just think we can be more welcoming and more transparent in these policies. At a minimum, if you feel the need to cast a close vote or are a mod closing a question for a new user, I hope you might also consider leaving a "Welcome to SE!" comment with a brief suggestion on how to do better. I've seen that on other SE communities much more than here.
    – Athanasius
    Aug 4, 2021 at 17:48
  • 2
    I also mentioned this in a comment to an answer below, and I know this makes more work for mods who are volunteering their time -- but I do believe "on hold" status could be more appropriate for many new users until they have time to perhaps edit and try to improve their first question here. Sure, some questions are obviously very bad, completely off-topic, or violate serious community guidelines and should be closed or deleted immediately. But if someone made a good-faith effort to ask a question about music theory, I think our first response should be to try to help them.
    – Athanasius
    Aug 4, 2021 at 17:50
  • 2
    "On hold" status was removed at the end of 2019. See New Post Notices are live network-wide.
    – Aaron
    Aug 6, 2021 at 5:39
  • @Aaron Ah. That's why I can't remember any recent on-hold notices.
    – user45266
    Aug 7, 2021 at 8:32
  • @Aaron: Thanks. I didn't realize this had disappeared. I do think that's unfortunate, as I felt in the past that the "on hold" notice was a little more forgiving to new users who are not used to the SE system. Oh well...
    – Athanasius
    Aug 7, 2021 at 18:45

2 Answers 2


While I'm a newcomer to the SE (and probably don't yet have "pet peeves"), I'll offer my two cents:

  • Yes, I was puzzled by that closure, since it did seem to be a substantial question showing research, documentation, and a useful question. Perhaps it caught our eye since the example given in the close-rationale is "What key is this in," but I imagine that that example is envisioning the sort of easily-answered, no-effort-given question like "What key is Sonata in C in," or "Hey, why are there are no sharps or flats where the key signature is supposed to go?" Perhaps well-intentioned questions about modal mixture, especially limiting their scope to specific passages, should be entertained.
  • At the same time, there are perhaps times when a quick closure is the best thing. Every day we humanely relocate "Hey, what's this song?" questions to Music Fans, and shut down "Who would win in a fight, Mozart or Bach?" In these cases, dallying about closure would just encourage misguided answers. Closing isn't a punishment*; it's an act of maintenance.
  • *Counter-argument: Though, it can feel like punishment, especially to the uninitiated. I'm sure this is ground that has already been covered in every SE and wiki, but it's new to me, and my take on it is: Online communities are still communities. If they hope to grow and survive, they must take sociology into account. If a newcomer sticks their head in the door and you hit them with the "shushing stick," they're probably not going to stick around. There are plenty of cases where we shouldn't be hesitant to close a question, but perhaps can do it humanely, including comments steering the OP to how they might improve it. After all, closure is reversible.
  • One of the problems inherent in any such community (I certainly felt it in Yahoo Answers years ago) is that the insiders, the Answerers, become worn down by providing the same smart answers to the same dumb questions day after day, but for every Questioner it's brand new. You pop in, ask "So soundtracks are Classical music, right?" and are surprised when someone blows up or is terse, as if they just lost heart after the 200th time they explained their position in detail. I'm not sure there's a good solution to this problem, except to repeat the calming mantra "It's new to them, it's new to them," and link to places that it's been answered well before.
  • 3
    Nice post. Regarding the last point, we shouldn't hesitate to flag as "unfriendly/unkind" any post of the type described. While such posts can't be prevented (aside from the calming mantra and self discipline), they can be swiftly addressed. (BTW, Bach would totally kick Mozart's ... aspirations. Was Mozart ever arrested after a bar fight? No. Bach? You betcha! Do not mess with the big B.)
    – Aaron
    Aug 3, 2021 at 17:05
  • 1
    @Aaron How about Bach vs Turlough O'Carolan, who ended a drinking contest by tying his buddy into his bed and sticking him with the tab? Aug 3, 2021 at 17:31
  • Touché. (You can't spell "Turlough" without "Tough".)
    – Aaron
    Aug 3, 2021 at 17:55
  • 1
    Thanks for your thoughts. I appreciate the comments. My one thought in reaction to your second and third points and how to resolve that is that mods do in fact have another option: they can put a question "on-hold." I'm pretty certain that by default a question "on-hold" reverts to "closed" status after five days unless a mod approves it. This strikes me as a better way to deal with new users than summary closing. Put a comment saying: "Welcome to SE! This question doesn't fit our guidelines because of X. Can you clarify?" Sometimes that happens, but a lot of time Q's are just closed.
    – Athanasius
    Aug 4, 2021 at 17:40
  • 3
    I will say that in some other SE communities I've participated in, the exchange I mentioned is much more common for new users. Even a "Welcome to SE! Unfortunately, your question doesn't fit into the guidelines for questions on this site, which are [HERE], so we have to close your question for now. If you rewrite the question to conform to the guidelines, it may be reopened. We encourage you to ask other questions following the guidelines..." or something like that which could even be cut-and-pasted would be great. Lots of new users have no idea how to navigate the restrictions.
    – Athanasius
    Aug 4, 2021 at 17:44
  • 1
    Hi Athanasius - closed and on-hold are exactly the same. When questions are closed guidance is given automatically from the system (okay, it's basic, but it points folks at the guidance) and often comments are left by whoever closes it, whether it is a few community members or a mod, if they feel that more guidance is needed. Any closed question that is edited to be suitable can be reopened.
    – Doktor Mayhem Mod
    Aug 11, 2021 at 7:54

I agree with you on almost all fronts, and I'm so glad you've provided such a compelling start to a (in my opinion) long-overdue conversation.

I think you're absolutely right that we often close things too quickly and that these quick closures can be disheartening for new users. This is a problem on just about every Stack Exchange site I visit, and I like to think it's something that everyone, site-wide, is trying to improve. (Note that I'm not saying this to excuse us; it's absolutely something we can do better.)

(As an aside, I wonder if there are data we could access for how long it takes, on average, for the community to close a question. There have been multiple instances where I've seen a question last for three or four days until five users vote to close it. I think that's a reasonable timeframe, but I'd be curious to hear what you think.)

I've been much better this year about leaving notes to users suggesting changes to questions/answers, and then coming back a day or two later to decide on an action. But it's also a bit easier for a mod; we have a "dashboard" that helps us keep track of these things, so that I (or another mod) can come back hours/days later, see what (if any) changes have been made, and act accordingly.

One thing I've considered in terms of deeming a question "substantial" is to emphasize the "concrete reference (sheet music, etc.)" listed in the help center. If my memory is correct, these criteria came about due to an influx of questions asking for keys/modes of random songs on the Internet. There were no scores, these were just aural tonic identification exercises. Of course we can't require sheet music, and for various reasons: plenty of music is never notated, the questioner may not have the skills or means to notate, etc. But for me this is at least some sort of initial barometer for evaluating a question.

Lastly, I view the “What can we ask?” and “What’s off-topic?” pages as working in tandem. I don’t personally think the lack of information on either page means we can or can't ask about something, since I view them as two halves of one whole. But perhaps we could merge the information on these two pages better just to be as clear as possible?

You must log in to answer this question.

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