Why learn music theory?

Ask general music theory or songwriting questions, get feedback!

Moderators: MattKingUSA, khz

merlyn
Established Member
Posts: 1392
Joined: Thu Oct 11, 2018 4:13 pm
Has thanked: 168 times
Been thanked: 247 times

Re: Why learn music theory?

Post by merlyn »

It's worth drawing a distinction between music theory and harmony. They have a similar relationship as arithmetic does to maths. Arithmetic is a limited subject that can be be learned reasonably fully. Maths is open ended and a person could spend the rest of their life on it.

In the traditional approach to learning an instrument students sit grade exams. Before sitting a grade eight performance exam students must have passed grade five theory. There's not really that much to grade five theory. In the traditional approach it involves notation, key signatures, time signatures and rhythms. Basic theory includes twelve keys in major and minor versions and intervals.

Grade eight theory starts to introduce four part writing which is how students learn harmony and that can go on forever. Performers don't need a huge amount of theory. For a composer theory informs harmony which is material to work with.

There is also jazz theory which is the same material presented in a different way. It has been said elsewhere that theory involves being able to count to thirteen. :D A thirteenth chord contains all the notes from a scale. An improvising musician is like a composer who operates in real time and so harmony is more relevant to an improvising musician.

Every musician knows some theory. Learning music theory gives what they already know widely accepted labels.
GFX_Garage
Established Member
Posts: 13
Joined: Fri Jan 24, 2020 6:06 pm

Re: Why learn music theory?

Post by GFX_Garage »

Humans can reconcile roughly about 12 pieces of information. This explains a lot of the structure of music. Classical for instance, pushes the envelope with the tension; and even throws in more ascension prior to release. Of course humans are moved by such strategies (for better or worse); but the single note in context is something that baffles me. Of course it's influenced by the process of build and release; but I don't have an explanation of the appeal of pentatonic and diatonic patterns, for instance. Theory doesn't really address this.

There is some understanding of the appeal of symmetry vs asymmetry. Considering scale patterns as symmetry; too much of it might not seem realistic. This is something that is observed in our perceptions of beauty. A lot of symmetry and a single flaw tends to be more appealing. This explains the appeal of a beauty mark on a womans' face; for instance. Small amounts of asymmetry in runs can be more appealing than sticking to scale patterns; and can evoke emotional responses (for better or for worse), that make a piece more aesthetically interesting.

It might be safe to assume that scale patterns contain some form of symmetry; but as to what that particular symmetry is, I just don't know. It's thought that music was born of primitive communication and ritual; so linguistics may have some information to bare on the subject. The chosen frequencies may correlate with patterns in vocal expression in language. But it even goes deeper than that. In linguistics, speech is often thought of as a latent ability; and not so much a learned ability. The reasoning is more clear when considering that newborns can communicate their needs and emotional state. This is where it gets baffling. There is no theory for the morphology of this innate behavior. This is just one angle of attack; to demonstrate one of the brick walls that we run into, in trying to explain the appeal of resonant patterns.

I hope I'm explaining my issue clearly. It's quite a can of worms. :)
merlyn
Established Member
Posts: 1392
Joined: Thu Oct 11, 2018 4:13 pm
Has thanked: 168 times
Been thanked: 247 times

Re: Why learn music theory?

Post by merlyn »

GFX_Garage wrote: I hope I'm explaining my issue clearly. It's quite a can of worms. :)
I think the questions you're asking go outside music theory into anthropology and neuroscience :D

Have you heard of the book This Is Your Brain On Music?

https://en.wikipedia.org/wiki/This_Is_Y ... n_on_Music
User avatar
thumbknuckle
Established Member
Posts: 51
Joined: Sun Oct 19, 2014 6:16 pm
Location: Western Massachusetts
Contact:

Re: Why learn music theory?

Post by thumbknuckle »

merlyn wrote:
GFX_Garage wrote: I hope I'm explaining my issue clearly. It's quite a can of worms. :)
I think the questions you're asking go outside music theory into anthropology and neuroscience :D

Have you heard of the book This Is Your Brain On Music?

https://en.wikipedia.org/wiki/This_Is_Y ... n_on_Music
I would argue that anthropology and neuroscience (as well as linguistics and mathematics) are topics one should be studying as part of a comprehensive music theory pedagogy.

What most people call 'music theory' is really just nomenclature. Knowing how to spell chords and scales and stuff is necessary but not sufficient for a non trivial study of music theory.

To my mind music theory is about grappling with all sorts of different musics and trying to come to an understanding of their internal logics. It's not about giving names to what stuff IS but rather developing an insight in to how stuff WORKS.
Faster than a laser bullet.
Louder than an atom bomb.
merlyn
Established Member
Posts: 1392
Joined: Thu Oct 11, 2018 4:13 pm
Has thanked: 168 times
Been thanked: 247 times

Re: Why learn music theory?

Post by merlyn »

@thumbknuckle : I don't disagree. I use the term 'musical analysis' for the study of how music works.

https://en.wikipedia.org/wiki/Musical_analysis

Now we have theory, practice and analysis, plus history, composition, ear training ... and studying music involves all of these.

You may know that Newton didn't have a monopoly on theories of gravity. Descartes had his own theory of gravity involving 'corpuscles'. The difference between the two theories is that Descartes wanted to explain where Newton only attempted to describe. It was Newton's approach that proved more useful and modern scientific theories follow Newton's example by only attempting to describe. Theory describes, analysis explains.

In the context of this thread 'music theory' I thought referred to flat fifths, diminished sevenths and all that which is a necessary prerequisite for musical analysis.
GFX_Garage
Established Member
Posts: 13
Joined: Fri Jan 24, 2020 6:06 pm

Re: Why learn music theory?

Post by GFX_Garage »

What got me interested in this rabbit hole is the question, could strong AI make music that is virtually indistinguishable from human written music. There is of course enough theory to generate pretty nice pieces of music procedurally; but they aren't as interesting as human written music, and they don't seem to be expressive. I think about it sometimes; when I'm playing, and I'm even baffled as to how I can do it. :) So yeah, I'm far from answering the question.

I appreciate all of the input and the links. I'll check them out.
merlyn
Established Member
Posts: 1392
Joined: Thu Oct 11, 2018 4:13 pm
Has thanked: 168 times
Been thanked: 247 times

Re: Why learn music theory?

Post by merlyn »

GFX_Garage wrote:It might be safe to assume that scale patterns contain some form of symmetry
Symmetry applies to visual phenomena. What would aural symmetry sound like?

A tritone is considered symmetric because it is its own inversion. Inversion of an interval means 'put the low note up an octave or the high note down an octave'. So take G - C that's a perfect fourth. Now put the G above the C and we have a perfect fifth. But take C - Gb and it stays as a tritone when inverted. Kind of symmetric.

The term 'symmetrical scale' in conventional theory refers to diminished and whole tone scales and Messiaen's modes of limited transposition. Symmetry here means the scale has a pattern that repeats. So C = D = E = F# = G# = A# whole tone and there only are two whole tone scales.

The major scale has some symmetry. C D E F is the same step pattern as G A B C but this is less symmetry than a diminished or whole tone scale. So most music is not as symmetrical as it could be. :D
GFX_Garage
Established Member
Posts: 13
Joined: Fri Jan 24, 2020 6:06 pm

Re: Why learn music theory?

Post by GFX_Garage »

Symmetry is more about consistency in patterns.

The symmetry of scale patterns are somewhat fractal in that they repeat. For instance, the base pattern of the diatonic scale is the Do, Re, Me, Fa, So, La, Ti; and it just repeats. On a guitar neck the patterns of whole and simi-tones repeat at the 12th fret. It's more linear on a piano; due to there being no redundant positions but the same patterns repeat. The visual aspect of symmetry is still available though.

Inversion is an interesting pattern. It's kind of like a musical palindrome. They also repeat; relative to the key of focus, as do 5ths 7ths etc.etc..

The asymmetry happens when stepping out of the scale pattern. Playing out of scale is essentially what I'm referring to.
yama
Established Member
Posts: 43
Joined: Tue Dec 27, 2016 10:00 pm
Has thanked: 1 time
Been thanked: 1 time

Re: Why learn music theory?

Post by yama »

What got me interested in this rabbit hole is the question, could strong AI make music that is virtually indistinguishable from human written music.
Does it matter?

This could mean that an AI is as good as writing the shittiest music a human could make. I think the music itself could be more interesting than this question.

For example it's really easy to tell when a machine is playing something that is beyond the limitations of the human body for playing in real time. But can you make an AI compose something that is humanly possible, but still completely distinguishable as machine-made? I think that's an interesting question too.
User avatar
thumbknuckle
Established Member
Posts: 51
Joined: Sun Oct 19, 2014 6:16 pm
Location: Western Massachusetts
Contact:

Re: Why learn music theory?

Post by thumbknuckle »

What got me interested in this rabbit hole is the question, could strong AI make music that is virtually indistinguishable from human written music.
Isn't that just an incredibly cumbersome form of algorithmic composition? Seems to me that whoever wrote the ai is really responsible for the music.
Faster than a laser bullet.
Louder than an atom bomb.
GFX_Garage
Established Member
Posts: 13
Joined: Fri Jan 24, 2020 6:06 pm

Re: Why learn music theory?

Post by GFX_Garage »

I admit Turing Test type comparisons aren't that interesting on the surface; but it could help us to better understand where we get the ability. Posing questions in different ways help us to attack problems from as many directions as possible.

There's a difference between the narrowly focused AI that exists today and AGI (strong AI). Strong AI is similar to humans in that our innate abilities are similar to its' source code; and our learned abilities are similar to what it might achieve with machine learning. One thing that might be more interesting for this community is the fact that the most advanced AGI project is open source. :) Its' also being applied to a blockchain based network; so that there will be community learning among strong and narrowly focused AI projects. With strong AI, the human steps away and allows the AI to evolve and develop; in much the same way that we have. i don't think the coder will be responsible for anything beyond innate ability; as nature is responsible for our innate ability.

My conjecture is that AI would have greater memory resources to work with; so they might employ more tonal frequencies and ascension and release might be more drawn out and complex. What is more interesting though (I think), is are simulated emotional responses likely to produce similar forms of expression? Maybe the expressive qualities of AGI music won't be humanly empathic. This might be a hurdle for AGI music; as we humans kind of have the patent on it.

This all begs the question what exactly is music? Also, what is good music? Theory can become very philosophical; due to the ambiguity surrounding music.
jeroen
Established Member
Posts: 9
Joined: Sat Nov 23, 2019 6:43 pm

Re: Why learn music theory?

Post by jeroen »

thumbknuckle wrote:
What got me interested in this rabbit hole is the question, could strong AI make music that is virtually indistinguishable from human written music.
Isn't that just an incredibly cumbersome form of algorithmic composition? Seems to me that whoever wrote the ai is really responsible for the music.
Not so much the programmer of the AI software, but the one who will be the trainer of the AI. Same reason why parents are responsible for their children (until they grow up). So, what would happen if we let the AI search for it's own training?

I don't think music theory is required to make music on any instrument including the computer. But it can be a great tool to further explore music, and will lead to new ideas for composition, chords, melodies, etc. I also like music theory as an observational tool, a bit the same as analysis of language. You don't become a better speaker, but it helps to better understand how music/language got transformed into it's current state. Music is playing with expectations. In that way, it's good to know the rules (expectations) in order to play with them. I think this is why experimental underground music influences the pop music some years later. Experimental music generally about either breaking rules or inventing somewhat new sets of rules. In both cases music theory will not apply, because there is not enough history of experimental music to analyse. While you don't need music theory to create experimental music, it can be fun to analyse how it becomes the new normal.
Mk2
Established Member
Posts: 193
Joined: Sat Sep 10, 2016 8:28 pm
Has thanked: 3 times
Been thanked: 2 times
Contact:

Re: Why learn music theory?

Post by Mk2 »

i learnt solfege & piano a bit,

=> i know some chords, gamme.. but if i had not learnt those, i'll be certainly 100% FREE OF mind to explore new stuffs !
Post Reply