Some early thoughts on musical chord progressions and human emotions

I’ve been mulling some thoughts on music and emotion, and thought I’d mention it here. They’re not smart thoughts yet. But it seems there might be something there. It concerns emotions and why music has the scale and chord structure it tends to have.

First, some of you may remember my book, HARNESSED. In that book I talk a lot about how music has culturally evolved to sound like a person moving evocatively around us. The melody has the properties of Doppler shifts, loudness modulations have the signature of proximity, tempo concerns the speed of the mover, and rhythm concerns the behavior / gait of a person moving.

But one thing that the book doesn’t get into is why we have 12 tone intervals, why 7 notes are in a scale, why so much music has a I, IV, IV chord structure to it. …and many many facts that are all circling around these issues.

One fascinating, little known thing about music is that, although there are seven chords within a key, it’s really approximately the case that there’s just three chord classes. So, for example, in the key of C…

• C, Em and Am are all in the same class. (They are I, III, VI.) That is to say, you can do these chords in any order and it doesn’t really much feel that the music is “going anywhere” or doing much. Yeah, there are subtle differences, but try it: the music is “stationary” in some sense.

• F and Dm are also in the same class. (They are IV and II.)

• And, G and Bdim are in the same class. (V and VII.)

So, in a sense, there’s really kinda sorta just three chords (ahem, three chord classes) that matter in much of music: C, F and G. (I, IV and V.)

This surely oversimplifies things, as there are differences that matter among the chords within a class. But, I think it helps to keep things really simple if we have a hope to maybe make progress on what these three chords mean to our brains.

Ok, now, given these three kinds of chord, which are also also called “tonic”, “subdominant” and “dominant” (i.e., I, IV and V), people have a very strong feeling summarized fairly well by the following (which I pulled from the web somewhere):

TONIC (I) = home

SUBDOMINANT (IV) = I am going somewhere / leaving home

DOMINANT (V) = tension! I wanna go home

I don’t take this “going” and “leaving home” stuff too literally. It could more generally just point to the idea that the tension is at baseline for a C chord (in the key of C, i.e,. a I chord), rises qualitatively to a higher level at the F chord, and once you get to G it’s for some reason the highest tension, after which it can only go lower, and “wants” to go back to low-tension baseline C.

Question is, Why would these three chords (within a given key) have such a universal psychological feeling?

And, for me, the question is, what human sorts of stimuli do these three chords sound like, and thereby harness?

Ok. My suspicion is this:

In conversation with someone, when I am at baseline, I have some fundamental frequency determining my key, and along with that are subtle baseline “chord” contributions from harmonics. That’s like a C chord. (A really messy one, with other notes outside the chord also being “played,” but just not too strongly.)

If I get a little bit more excited, or showing some intensity, other notes within the harmonics available for my throat size etc. become more salient.

The claim would be that more F chord notes become perceptible within the whole.

And if I get even more excited, trying to emphasize my point in conversation with you, and perhaps I’m angry or afraid or something, even more notes within the harmonics available for my throat become perceptible. And these tend to be notes within the G chord. That’s the climax, or highest level of intensity, after which I will usually go back to baseline again.

If you listen to people, and to yourself, engaged in conversation, we do perhaps seem to have a couple discrete levels of intensity, and then fall back to baseline again.

The idea would not be that our voice sounds like a beautiful C chord, then gets more intense and sounds like a pure F chord, and then still more intense to a pure G chord.

Rather, for an actual voice, the idea is that all these notes are always there even in the baseline, but that the trick to auditory recognition of these three modes is whether these F chord or G chord constituents are there in sufficient magnitude.

We would have evolved very sensitive recognition mechanisms for them, because it’s important to recognize the intensity level of vocalization of others.

Music would just end up being super extreme exaggerations of this, isolating the F and G chords all alone, rather than the more messy reality.

So, that’s the basic idea. A little more detail…

When humans vocalize, we have a baseline tone of voice, one that is due to my throat size and so on. When I speak, I get very particular harmonics due to my inherent size and shape. I will have some fundamental baseline key, and of course the harmonics at double that, four times that, and so on. Those are just the octaves above my baseline key.

But, there are -- for complex vibrating objects -- always other peaks in the frequency histogram due to other modes, and they tend to occur even between octaves, and are at key mathematical intervals, in particular corresponding to frequency intervals of notes in the scale.

So, for example, if my voice’s fundamental key is a C, there might be a little peak between it and the next octave C peak at a spot corresponding to the note of G. And actually one for each note in the scale. These are little tiny micro-peaks in between the octaves. (E.g., on a guitar string, if you play the harmonic at the seventh fret, that’s a harmonic note a fifth higher than whatever note the string is.)

Ok, so, my vibrating throat etc. has all these harmonics, and they’re mostly all “notes” within the key determined by my particular throat. Even if I raise my pitch, because my throat is still the same size and shape, those “notes” within the spectrogram are still there. (Of course, there are size changes as we speak, due to tongue modulations and so on. But let’s keep it simple.)

So, the claim for this general idea would be that, as we raise the intensity of vibrations of a physical throat system, that the F chord contributions tend to become more salient first, and then with even more intensity the G chord contributions tend to become salient.

There might not even be more after that. Or, as intensity rises, there’s no simple general regularity, as perhaps all the harmonics tend to rise at similar rates, and so little or nothing that we would have evolved to recognize beyond these two steps.

Many of you may have much better intuitions about harmonics than do I. Please chime in!