#1
Hey all, just an interesting topic that's come up in one of my lectures, I thought I'd deliver it to the UG forum and see if anything interesting comes of it.

The question is, is all composition algorithmic, as in, is there always some form of system at play behind the music?

There is such a thing as algorithmic music that is music generated by certain sets of rules, this can incorporate people that write software that generates music or experimental work from people such as Iannis Xenakis or John Cage who invent (not necessarily computerised) systems which sometimes even resemble games that create music.
But at the same time couldn't one also say the rules of counterpoint for instance are algorithmic and then, any sort of composition one could think of?

A possible exception I thought of is some free jazz stuff where everyone just sort of goes off on one when someone says go, different keys, everything.
Seems fairly free of a defining system, but at this point can we still talk about composition?

Tell me what y'all think!
#2
I would say yes. And no. Even when we try to break free from the "rules", i think we subconciously stick to them a bit too. Dissonance rarely sounds good, so we will automatically stick to certain rules and principles to things sound nice to us. Weird chord progressions and weird chords sound weird, so we rarely use them.
#3
Ya im gonna say yes and no because most music does follow a certain set of rules. But then you have guys like Van Halen and Hendrix who know absolutly no musical theory. or you can listen to metallicas Kill Em All as they didnt get into theory until cliff taught it to them before RTL
Member of UG's Gain \/\/hores - pm gpderek09 to join
#4
Well, actually that's not true. Van Halen and Hendrix had large amounts of theory knowledge, but couldn't sight-read a lot. Hendrix was a theory genius, which is why his compositions were so amazing. dont mix up theory knowledge with the ability to sight-read. Both knew their guitars inside out and the same for the theory behind the guitar.

I agree Metallica were limited in theory, but still knew what they were doing. Burton was the dude though, and taught harmony to the band.
#5
Hendrix knew no musical theory?!!!? wtf!?!?! Dude, he knew everything, that is why he could go and play anywhere on the fretboard, any key, any song.
Quote by ibanez901
I brush my pubal forest into a combover.
#6
But then even without musical theory, people still know the difference between what sounds right and wrong, just through a natural ear for music. Like, you may not understand different chords and scales, but you'll still be able to pick when you're playing out of key.

This makes me think of Captain Beefheart's music, Trout Mask Replica inparticular. On first hearing it, it sounds like dissonant garbage. But then on further listening, you hear the intricate composition work involved in it.
#7
heh heh, SOME people can tell when they are in or out of key! Ever heard someone whistle who is tone deaf? It's painful. But mostly i agree. However, using your ear to build your musical theory knowledge is still giving you that musical theory knowledge.
#10
Quote by Applehead
I would say yes. And no. Even when we try to break free from the "rules", i think we subconciously stick to them a bit too. Dissonance rarely sounds good, so we will automatically stick to certain rules and principles to things sound nice to us. Weird chord progressions and weird chords sound weird, so we rarely use them.


Are you familiar with the 'Pitch Axis' method? I've been getting some extremely bizarre chord progessions through using that, and they sound beautiful. They're not dissonant though, as you're just modulating each new chord. You pick chords from the various modal families built on the same root and string them together, not dissonant, but you can up with some really weird stuff!

/off topic

And you might want to take 'Banned' out of your user title, I'd assume when the mods see someone who's supposedly banned posting new posts, it gets confusing
#11
^ No, not heard of it. Sounds interesting. Have you got an links or info? Would appreciate it.

As for the title.. If they are confused, that's their problem (until it's mine too)!
#12
Quote by Applehead
^ No, not heard of it. Sounds interesting. Have you got an links or info? Would appreciate it.

As for the title.. If they are confused, that's their problem (until it's mine too)!


Er... I'm sure if you look it up you'll find some stuff, but I'll give you a quick run through and example.

Okay, so you pick a root note to function as your axis, for this purpose I'll choose D (Because it's the one I'm working with at the moment).

Then, like I said - you pick chords from the various modal families, built on your root - being D - and string them together.

So, lets say we're doing a simple four measure progression, we take four modes and then list our available chord options (or the ones used most commonly for said mode).


[B]Mode.[/B]                  [B] Chords.[/B]
Dorian.                 min7, min9, min11, min13.
Lydian.                 maj7, maj9, maj13, maj7#11, maj9#11, maj13#11.
Mixolydian.             7, 9, 13, 7sus, 9sus, 13sus.
Melodic Minor.          min/maj7, min/maj9, min/maj11.


(Feel free to correct me on any of those chords)

After you've done that, it's pretty simple. You experiment with all the different chords for each mode in our said key and see what you can come up with.

One I've recently come up with is a min9, maj7, 9sus, min7 progression in D - I've taken my first chord from Dorian, second chord from Lydian, third from Mixolydian and fourth from Melodic Minor - See how it works? - Sounds weird, but has an extremely interesting sound and feel.

You can extend it and mitch-match keys aswell... I just find it really helpful to come up with new progressions I'd never have thought of before. I think Satch uses this method religiously

Sorry, you guys can get back to your algorithms now, lol.
Last edited by Dave_Gray at May 30, 2006,
#13
Well, to a certain extent, yes, i think ALL music is algorithmic. Mostly due to the fact we are biologically and hence mentally built from a set of fractal patterns - at the end of the day, we are still designed to think on rails we dont know exist and hence will compose music along them.

Course, it is possible to try and break free, but it tends to be done in an equally constraining manner. To try and avoid white by playing black just means you sound like you're playing using the human concept of opposites and comparing it to the original, which is no more breaking free than going down the wrong way on the same road.

If you can follow all my metaphors, then i think i've made myself clear.

Although i do find it funny people tend to revert to purer mathematical forms of composition to break free of the "messy" mathematical forms of biologically intuitive composition. A movement from expression through the golden mean and it's spirals to expression through a discreetly divided system...
#14
So in theory if someone was different mentally, and had no concept at all of logical composition, so to speak, then they would follow their own little train of logic. That would be original composition.
#15
I would say most people on here have not studied composition so are working in the manner you speak of. So yeah..
#17
Well that's just BS. If you've ever listened to world music, chinese, indian etc... they use different chords, scales etc.. so they were NURTURED to use those scales, chords etc.. not biologically engineered. This aint Gattaca dude. We use the chords we use because they are familiar to us, we have heard them all before, so they sound good. if you'd been raised on dissonance, you'd love that **** as that is what you would know.
#18
Yes, i know, genius boy.

Perhaps the fact that almost every culture has a perfect 5th, 4th, and octave accepted as a consonant interval shows a little mathematical truism.

Those "scales and chords" you refer to are not music. They're part of music. Most of those rely on very biologically simple divisions of the octave, and most experimental on simple mathmatical divisions. Im talking about rhythm, articulation, and all the other aspects that make music music.

They're remarkably similar all over. And tbh, i'm not talking about what we "like", im talking about the way we think and hence compose. Stand up. Your hands (if they're by your sides) are at a very specific point relative to your total height. That same ratio is the basis for a huge portion of the structural construction of the natural world. Funnily enough, it's also the ratio of the pitch of a note to it's perfect 5th. Which you can find just about anywhere.

There is some stuff that is simply hardwired into the human mind - no matter where nurture takes you from there.
#19
Now now, don't get bitchy. This is an interesting topic which are both interested in. lets keep it civil.

What you are talking about is Phi, if i'm not mistaken.

But as far as perfect 5ths go, that is just a 3rd a long the length of the string. You are delving into science here, and frequency division. I was not aware Phi was involved in music, so thanks for that. I will look into it more.

I guess what you're saying is true, but i think nurture is as much a part of as anything else. You can't say a perfect 5th is hardwired into the brain. Sure it sounds nice because it is a division of the fiundamental frequency, which sounds nice. but whack a ring modulator in there and get some mad frequency come out the other side and it can still sound good. The question is, if you'd been raised on dissonance, if someone played you a major scale, would you like it? Yes, your probably would, but it would sound "weird" simply because you have been raised on dissonance.

And i dont agree scales aren't music. In fact, i look at scales as the composite parts of a song. If you've ever played musical conundrums then you'll agree. but no, in their most raw form they are just notes. we make them music with our mad skills
#20
Quote by Applehead
What you are talking about is Phi, if i'm not mistaken.

But as far as perfect 5ths go, that is just a 3rd a long the length of the string. You are delving into science here, and frequency division. I was not aware Phi was involved in music, so thanks for that. I will look into it more.

I guess what you're saying is true, but i think nurture is as much a part of as anything else. You can't say a perfect 5th is hardwired into the brain.


No, but i can say Phi is hardwired into the brain, and many other similar relationships - it seems logical to assume that that hardwiring will spill over into the way we compose and appreciate music. I think it's Beethoven's 5th that has Phi written all over it - Subjects introduced and moduation all according to phi.

If the most obvious pattern in the human body is that obviously all over the music that is perceived in most cultures as "well proportioned" then it's almost a given that the more subtle patterns will show up too - even if they're harder to spot.

I'm delving into science because my argument is a scientific one.

Im not trying to say that we are all bound by maths to compose exactly the same way - but rather that we are all bound by maths to think in a roughly similar way. And hence our composition...

I'm not denying nurture - im merely pointing out that no matter what you do with a human's mind, it remains human and biased heavily towards music it can understand and take in. Imagine we our brains natural worked in binary - what would our music sound like? Would we ever actually be able to comprehend what we see as music in the real world? And vice versa!
#21
Quote by Dave_Gray
Er... I'm sure if you look it up you'll find some stuff, but I'll give you a quick run through and example.

Okay, so you pick a root note to function as your axis, for this purpose I'll choose D (Because it's the one I'm working with at the moment).

Then, like I said - you pick chords from the various modal families, built on your root - being D - and string them together.

So, lets say we're doing a simple four measure progression, we take four modes and then list our available chord options (or the ones used most commonly for said mode).

(Feel free to correct me on any of those chords)

After you've done that, it's pretty simple. You experiment with all the different chords for each mode in our said key and see what you can come up with.

One I've recently come up with is a min9, maj7, 9sus, min7 progression in D - I've taken my first chord from Dorian, second chord from Lydian, third from Mixolydian and fourth from Melodic Minor - See how it works? - Sounds weird, but has an extremely interesting sound and feel.

You can extend it and mitch-match keys aswell... I just find it really helpful to come up with new progressions I'd never have thought of before. I think Satch uses this method religiously

Sorry, you guys can get back to your algorithms now, lol.


Can you direct me to a page with some more info on that? It looks very interesting.
#23
I find new substitutions for chords all the time by looking at older compositions and musicians.
UG Blues Mafia
Founding Members: Forklifterer, Steve Cropper, Crzyrckgtrst28, and TNfootballfan62


Member of the Anti-Whammy-Bar-Solo Club
#24
I find that Ryan Adams describes the songwriting method best:


"If you just sit around and dick around on a guitar for longer than 30 minutes and you might accidentally run into some chords that sound like a song, and as soon as I get on to something like that, I'm like... 'Aww, ****!' And then I have take that wherever it's going, and I feel really satisfied, like I did something today. Then I go out and get drunk, it's great."
#25
Quote by Freepower
Er, my whole post was to make the point that we are biologically engineered to work in a certain way, no matter the nurture influence.

That's very much untrue. The ear is trainable, distinctly, and very much so from birth. Western tonality sounds horrible in some areas, and exotic harmonic relationships sound horrible to us. Hell, even in western composition, complete-serialism takes a lot of training to actually be able to hear and appreciate properly... even when you can hear it, and appreciate it a lot of it still sounds like noise.

As for the algorithmic question... I'm not sure I can absolutely say yes to that. But I'm not sure I can provide any examples that are absolutely to the contrary... possibly sound clustering (sound mass, sound wall) effects, but even then you start to see structure in some areas. If you really want an answer, look into the sound mass composers of middle 20th century, and their responses to complete serialism; and late minimalist composers (Terry Riley, for instance).
Quote by les_kris
Corwinoid is God
I'm not even God-like... I've officially usurped the Almighty's throne.
Click here to worship me.

Member #3 of the Corwinoid Fan Club