How much does music theory really shape your compositions?

[yukon3000]

Hey everyone,

I’ve been thinking about how much theory actually plays into the way we write music. Some producers lean heavily on scales, chord progressions, and all the “rules,” while others just go by ear and vibe until something clicks.

For me, I feel like theory can be super helpful when I’m stuck or trying to find new directions, but I also don’t want it to take away from the more instinctive, creative side of things.

How do you all approach it? Do you rely on theory as a foundation, or do you mostly follow your ear and let the music guide itself?

[lfm]

It started getting interesting as I got some songs done and felt I wanted notation on them. Maybe I would register with a publisher even.
- also worth hiring a real vocalist to them they needed notation as support to practice

So melodies in midi and you got accidentals on every note since default is C major key.
- so which key is this song, really?
- so I can put all accidentals in beginning of a notation stave

Then it went over to harmonizing chords and why some chords naturally were minor or major in a key.

Then doing more keyboard stuff, not only guitar, I started to learn about voicings and in which octave to put the 7 or 9 etc. Makes a huge difference and make it sound even more beautyful to spread different ways in two hands.

Then about modulations, inserting nice transitions from one chord to another.

You deepen understanding how it works. So very much worth the trouble.

[Opaque]

A lot actually, I always tell the ai to mind the rules of music theory while generating.

On a more serious note, it's similar to the technical/production stuff, try to understand and learn as much as possible, but then I go more on instinct while actually doing, and referring back to it just to build a foundation or to check where I am, especially if I get stuck.
But, I'm doing electronic stuff, at a certain point I'm working with sound over notes, this might be very different for singer-songwriters or composers of more traditional or orchestral work.

[jancivil]

Do you rely on theory as a foundation, or do you mostly follow your ear and let the music guide itself?

There is no dichotomy as far as I'm concerned. You can mostly, or entirely go by ear if "theory" is foundational. If we have such a grounding we can call a foundation for real, it's a certainty we'll have more to draw from ad libitum.

So, cases where my impetus for a track consisted of a principle of "music theory" were in adaptation of composer Erik Satie's works. I recognize in one of them that the theme is drawn from a scale that is the fourth mode of harmonic minor (and that the B section abandons that in the most direct way we can even talk about (the basic scale features a #4, B goes right to the IV chord.).
[F G Ab B C D Eb vs Bb]

Another instance was the recognition this material is not founded in a conventional scale. Today it's known as "The Satie scale". NB: there is no seventh degree here.
In trying to relate to one of the more basic scales we'll tend to see "Lydian Dorian no seventh", and


In that first example I improvised the melody for the A' section, in place of the "real" tune. Having the scale under my fingers now, I don't need to think about it. You'd have to call it playing by ear.
I was always a modal player. Sticking to 5, 6, 7 notes is natural.

In my own music I'm not burdened by thought in the moment of creation ("You can't think and play at the same time." - Sonny Rollins).
We might say the choice of an interval is something from "music theory" that I don't have to think about.
There is one case where I prepared or curated towards a mood, a sound world by creation of a small chord bible (ie., six or seven note chords that voice all members of a scale (scales I made up). I invested some time in thinking about it, the ramifications and so forth before I started. I'd have to analyze it to really know the extent I adhered to or violated it. So other than the Satie, a piece of music is made up on the spot and a fair bit of the former is improvised as well.

I was improvising melody before I knew much if anything of what things are called (half a century ago). I mean I knew what Dorian mode was by ear because I listened to Carlos Santana and Frank Zappa enough.
So when I improvise it's about intervals, not note names. I'm usually ignorant about the nuts and bolts of what I did after the fact. I knew what I knew when I made it, I doubt I'd learn anything going back to define things.


scare quotes around <music theory> because of a tendency of folk to think that means something it maybe doesn't.

[SoundOutWest]

Bear in mind that anybody with strong theory skills won't even realise they're making use of it - it's just ingrained at that point.

To me it's 50/50. On one hand music theory is the language of how to describe music, but it's not the music itself. Everything starts with improvisation for me and you don't need to understand what you're playing if it sounds good. That being said, like any language, it is hard to speak it without same baseline level of theoretical understanding.

I *definitely* lean on music theory when I have a very specific sound that I need to achieve, this is especially true of culturally-inspired music. I recently finished a string quartet which was themed around the British Isles and I restricted myself to the "rules" and patterns seen in that style of music. Those "rules" didn't leave a lot of wiggle room for creative innovation, but it worked, the piece was intensely British according to the auience.

[stoopicus]

Don't think of theory as a set of rules. Think of it as a shortcut to helping you understand what's going on and what will sound good.

If you think of theory as strict rules to follow, it's not going to be helpful. Instead think about it like "this is 2600 years of thinking about what sounds good, which I now don't have to figure out for myself when I get stuck."

[Stirner]

Bear in mind that anybody with strong theory skills won't even realise they're making use of it - it's just ingrained at that point.

To me it's 50/50. On one hand music theory is the language of how to describe music, but it's not the music itself. Everything starts with improvisation for me and you don't need to understand what you're playing if it sounds good. That being said, like any language, it is hard to speak it without same baseline level of theoretical understanding.

Very good explained!. Someone who hears a C3 on a piano as a functionally the same note as a C5 on a flute and calls both notes a "C", already uses musical theory and musical theory is a tool, a language for thinking about sounds with different pitches, textures and tembres.
It's not only for speaking with others, but also for composing in your mind. As a musician, you should understand what intervals and chords are and how they work and even be able to hear them in your head. This is the minimum musical theory that anyone should learn for getting efficient.

It maybe depends on your music, I think. I tried to work with a friend of mine, who makes some hip-hop. His approach is more about slicing samples and - in my opinion - randomly move notes around in a piano roll. It works maybe for simple melodies, but you can't create anything more substantial with such approach. I waited 20 minutes and in that time he could not get a single three-note-chord down, it was painful to see it.

[fatprice]

1000% for me it is theory for the right notes from there I like to experiment. I’ve begun lately breaking away from conventional key scales. I’ve been experimenting.

[Sonic Illusions]

What theory? I simply apply math regarded to the circle of fifths. Physics, essentially

[Starbright]

Overall 30% actively, intuitively probably 90%

[Dirtgrain]

Have we been inventing math or discovering it? Maybe this could similarly apply to music theory. It might be that we are discovering relationships of pitch (melody), harmony, rhythm and volume--relationships that just work on some fundamental level. Then again, most here have been raised while immersed the well-known music of the world--maybe that just taught us to value particular relationships of pitch, harmony, rhythm and volume.

There are some cultures with quite disparate systems of music that I cannot connect with.



Either way, given that we are immersed in it, it's likely that nearly all of the people playing by ear are still guided by music theory, even when they don't know the theory overtly. It's in us.

[FranklyFlawless]

How do you all approach it? Do you rely on theory as a foundation, or do you mostly follow your ear and let the music guide itself?

Both, I let them challenge each other and whatever prevails by the end of it gets primarily used in the final product.

[thecontrolcentre]

do you mostly follow your ear and let the music guide itself?

This ... (I don't know any music theory)

[TheMaestro]

Learn theory, as in the building blocks. That can make it easier for your workflow. Then make what sounds good.
Or just make it sound good. But theory can help save time with that.
And then you can break the rules.

[hotelpaintings]

I did the 'follow your ear' thing for several years - this can work okay so long as your ear is good. Mine is not, though. Once I started learning some basic music theory my productivity and "success" rate when up noticeably.