In my time here, I've noticed a lot of answers that say things like:

  • If it sounds good, do it.
  • You can do whatever you want.
  • Theory gives rules, but you can always break them.
  • Who cares what the rules say? Just do what you think sounds best.

In a way I understand where these answers are coming from. But when the OP asks "What is this rule?" it seems to me that "ignore the rule" is a non-answer. Or if someone asks "Is there a theoretical answer to why this chord follows this one?" I don't think "because it sounds good" is a worthwhile answer.

Should such answers be discouraged?

5 Answers 5


While I agree with Dr Mayhem, I have noticed this type of answer being used more and more as instead of a real meaty answer. There is a very real danger of using these answers instead of a sufficient explanation of ideas behind a piece of music.

For example I've see questions in the past with new users trying to understand a concept like modulation and an answer to that question just saying on the lines of

it sounds good so it works

which does not help anyone understand music composition, theory, or performance concepts any more then before the answer was posted.

There's a line somewhere that we may need to draw as a site with the first part being as Dr Mayhem describes of is the question just simply "Can I do/play X?" then the core issue is with the question. If the question seeks deeper understanding/reasoning, then the problem is with the answer.

I know I have down voted answers in the past that just give one of those phrases with no sufficient details to help the OP. I know it seems harsh, but if we want the site to be a resource for musical knowledge we really need to have and answers that actually can show people musical knowledge.

  • 1
    Agreed with both of you here.
    – user28
    Dec 19, 2017 at 16:21
  • 1
    It's not harsh; the answer isn't actually answering the question fully. Answers like that rarely get upvoted, which says something. And in reality, 'if it sounds good, it probably is'. However, without extra explanation, it's a pretty poor answer, which should be deleted, as pretty poor answers are. And, if the question is vague, then that should get adverse comments, or be deleted, as it's not really going to be much help to future readers.
    – Tim
    Dec 24, 2017 at 15:39

It's an interesting question.

The problem is that some questions can only have that answer - Music is a very subjective subject. Much more subjective than Stack Exchange was originally set up for, so there is a bit of a challenge in how the structure works here (similar to Parenting, Interpersonal Skills etc.)For example a question that asks if they can play something outside the rules will always have an answer along the lines of, "yes - of course you can ignore rules" so the problem is more around the question.

Would it be better to close those questions as off topic? Maybe, but I think we'd rather keep them if we can answer with something useful.


Music theory is always theory of some specific kind of music, and the problems described in this question result from people identifying or encountering rules, but without knowing the scope of the rules. If it sounds good, it follows some rules, even if you don't know what the rules are. If you broke the rules of some kind of music, then you just made different kind of music that has its own rules. Or you made something that was surprising and the surprise relies on there being rules to break.

When someone refers to "the rules", they still need to specify what kind of music they mean. Disco music has rules, rock has rules, punk has rules. There are no generic "rules of music" that are universally agreed upon and that apply to everything. Every style and genre has rules, otherwise you couldn't distinguish them from each other. The rule can be that punk music is listed under "punk" in a record store, and the rules can be completely subjective, debatable and controversial, but there are rules, there is logic. It's not random.


While I do answer this way at appropriate times, I try to explain the "meta rule" that really applies. Trivial example (that comes up lots of places, not just Stack Exchange) would be about parallel fifths or parallel octaves. There is a rule prohibiting these in some styles of counter point. In other styles of music, this rule does not apply but the reason for such rule is still applicable. In this particular case, the meta-rule is that "if one wants to have independence between voices, avoid parallels as these make the voices sound as if the voices merged." In popular music, parallels between the bass line and the melody may yield a thinner sound than one might like.

I think an answer that expands on the why make allow the "if it sounds good, do it" to convey some real information rather just being a cop out.


Semi-related: I find it interesting when music theory is used to analyze a song in the vein of "why does [insert random song] sound so good?". From my (still learning) perspective, music theory helps us describe, and prescribe methods how to create interesting music beforehand.

Therefore, it sometimes strike me as a somewhat overcomplicating afterthought when people struggle to cram a successful song into a theoretical frame in retrospect, especially when it concerns music that obviously weren't written from an orthodox theoretical standpoint (for example: Kurt Cobains music) to begin with. I've seen examples of this where the analysis has been full of ad hoc theory, kind of making the explanations a bit redundant.

It's a situation where i lean towards the it-simply-sounds-good-explanation, and i don't totally agree that it's a cop out since compositional music theory doesn't always add anything for understanding the music. Does this make sense?

  • 1
    Kurt Cobain's music very much fits into the narrative of analysis music theory gives you, it's just not as pretty as a Music Theory 101 class. His pieces have form, harmony, melody, instrumentation, use of dynamics, ect. all of which can be dissected to gain a deeper understanding of the pieces and find patterns not just in his music, but music you find similar. Just saying it sounds good doesn't dive into any of these aspects that can be explored.
    – Dom Mod
    Dec 26, 2017 at 20:03
  • 1
    Yes, that's exactly what i'm getting at. It's important to differentiate descriptive and prescriptive theory from analyzing music in hindsight. They offer different reliability based on how they're used. In my example i've seen analysis being imposed on Cobains music using prescriptive theory paired with a bunch of ad hoc assumptions. I agree to your elaboration that it can be meaningful to explore patterns for why it-just-sounds-good, but not all music theories are valid for that purpose afaict.
    – Erik
    Dec 26, 2017 at 21:21
  • I think we're pretty close in thought, but reaching different conclusions which is fine. There is one thing I'd point out though is theory is never in hindsight. It's always there regardless of whether you are looking for it or not. And if your using the wrong analysis tools of course it will seem ad hoc. For example, for Lithum if you are trying to use Roman Numeral analysis, you will just most likely confuse yourself since the the two halves of the progressions represents very different ideas which helps give the slightly off feel for the whole piece which is very much the intension.
    – Dom Mod
    Dec 26, 2017 at 23:55

You must log in to answer this question.

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