Music ought to be one of the greatest human mysteries, the one kids say they're going to grow up and solve.
Why does it exist? Why do we love it?
You'd think we would have made some progress figuring out these questions, given the past two millennia of effort. In fact, if you read a "music theory" book, you might leave with the impression that we actually understand music. Music theory books are teeming with arcane mathematical nooks and crannies, from harmonics to timbre, and from diminished sevenths to the cycle of fifths.
But music theory books aren't on theory at all, not, at least, on "theory" as a scientist knows it.
Music theory is better described as accumulated musical lore, an inelegant pile of regularities for how music tends to work, and rules for what tends to sound "musical." Such repositories of musical knowledge are critical to our eventual understanding of what music is and why it exists -- but not because this knowledge is "theory."
Music theory is important because it amounts to data. Lots of data. Millennia of data! On rhythm, beat, scales, chords, harmony, melody, timbre, dynamics, style and more.
Phillip Ball's book, The Music Instinct, is a music theory book, in the sense that it covers that terrain. But The Music Instinct is not, by any means, a typical music theory book.
For one, it doesn't pose as a pseudo-mathematics book. Ball's book extends beyond the standard terrain, too. You won't find discussions on the origins and evolution of music, or on the psychology of music, in other music theory books.
You'll also find Ball's own erudite commentary and analysis sends the message that the current hypotheses and the interpretations of psychology experiments are highly debatable.
In looking for a big message in the book, one might take Ball's book as a metaphorical throwing-down-of-the-gauntlet to the scientific community. Look at all these fascinating regularities found in music! The nearly-ever-present beat. And the fact that it is pitch, not loudness, that tends to change from note to note. Surely we're poised to explain what it all means.
Ball's not trying to craft that explanation himself, though. He's quick to say it is dangerous to make generalizations about music. But aren't these music-theory regularities special? Or is it that nearly any random regularity can be learned and enjoyed? And if the latter, is music theory just arbitrary?
From the fact that some musical regularity is not universal (some cultures like one thing, other groups like another), people have a tendency to throw up their hands and say, "Anything goes." Drums tend to be discriminated against in classical music, for example. What matters, though, is not total universality, but strong empirical tendency, and by this weaker standard, there are a loads of musical regularities needing explanation. In the thousand-dimensional space of possible auditory patterns we could listen to in the car, nearly all Earthly music hovers tightly in one spot. That is to say, differences between genres are dwarfed by the similarities in most music relative to other types of noise.
There must be a good explanation for why we have music, and why it is structured as it is, an explanation grounded in biology, psychology and evolution.
But on that topic, scientists have almost nothing to say about why music is structured as it is, and much of what is said is ridiculously simple.
My own view is that the complex theoretical foundation of music is the sound patterns found among humans when we move. Music, I argue, sounds like people moving. Rhythm and beat are identified with the rhythmic gait sounds, melody with the Doppler shifts occurring when a person moves, and loudness modulations ("dynamics") with the variations in spatial proximity of a mover.
Why, for example, does music almost always have a beat? Because movers always have a regularly-repeating beat. And beats have many of the properties of footsteps, not least which is that they tend to come at about 1 to 2 per second.
And why does music modulate its melodic pitch quickly but its loudness relatively slowly, rather than the other way around? Because changes in the Doppler shift of a mover occur whenever a mover changes direction, and that can occur as quickly as a single step. Changes in proximity, however, occur much more slowly because it requires that the mover take many steps through space.
We evolved auditory mechanisms for sensing other people -- their direction, their distance, their tempo, their mood, their behavior -- and music has culturally evolved over time to sound like that, thereby harnessing our auditory human action recognition mechanisms. That's the topic of my upcoming book, Harnessed: How Language and Music Mimicked Nature and Transformed Ape to Man, which should be out next year.