Not long ago on the always interesting and uber-nerdy Language Log, Mark Liberman directed readers’ attention to a recent paper supposedly studying what gives music its emotional resonance (Daniel L. Bowling et al., “Major and minor music compared to excited and subdued speech“, Journal of the Acoustical Society of America, 127(1): 491–503, January 2010). They explain:
The hypothesis examined here is that major and minor tone collections elicit different affective reactions because their spectra are similar to the spectra of voiced speech uttered in different emotional states…. the spectra of major intervals are more similar to spectra found in excited speech, whereas the spectra of particular minor intervals are more similar to the spectra of subdued speech.
Unfortunately the paper is available to subscribers only, so I can’t get at the whole thing. Apparently there’s some rather complicated acoustical analysis involved, but I don’t know the methodology and assumptions that went into it, so I can’t comment on that. If the conclusion implied in the abstract is true–that musical intervals can be correlated to the inflections of speech patterns–then it’s very interesting indeed. But I have a few reservations.
Of course, it’s no great surprise for anyone who knows a bit about music to learn that the cadences and rhythms of music may sometimes reflect the cadences and rhythms of spoken language. That’s how you write a song, after all, and it’s why we don’t think it’s entirely unnatural when characters in operas or Broadway musicals suddenly start singing their lines. When I’m setting a text to music, I often begin by doing an expressive reading of it, and find halfway through that a good tune has practically written itself.
But the researchers claim to have gone one better, and found a correlation between speech patterns and the structure of major/minor tonality. As quoted on Language Log:
If that’s the assumption that drove the research, then I’m immediately and implacably skeptical of everything else. My problem begins in the little four-word scientific cliche: “Other things being equal.”
When you’re doing a scientific experiment, of course, you want to make other things as equal as possible so you can eliminate all the variables. That’s fine for science. But I can’t see how on earth you could possibly get it to work with music. “Other things” in music are simply never equal in any way you can quantify.
“Intensity”? In what way? There’s all the difference in the world between a tune that’s being played intensely on a banjo or the same tune played intensely on a solo french horn. A chord played on a piano or the same chord played on an organ. A song sung by a boy soprano and the same song sung by a heroic Wagnerian tenor. A fortissimo passage by a full orchestra and the same volume from a rock band. And if you wanted to, you could notate them all precisely the same way in score. Forget other things being equal; the same things aren’t even equal.
“Tempo and rhythm”? I’ve read that someone once asked Johannes Brahms why he didn’t put metronome markings on his scores, because it seemed inconvenient to the performers. The composer retorted, “Do you think I always want to hear my music at the same speed?” Any good performer knows that a metronome marking is a suggestion–helpful for determining the composer’s intent, to be sure, but impossible to replicate with mechanical precision unless you’re actually playing with the aid of a machine. (Metronomes, computers, and click tracks come to mind–mechanical aids the lot of them.) A good performer will vary the tempo ever so slightly depending on the acoustics of the room and the emotional temperature of the crowd. Conversely, some entire genres of music (think Chopin) demand to be played rubato, slowing or speeding the tempo at will within a phrase to increase the emotional effect. The same piece never sounds the same twice.
That’s just where things are supposedly the same. Changing it up makes it even more complicated. The same chord in open voicing or closed voicing (if you’ll pardon a bit of theory shop-talk) has a completely different effect before it even gets off the page. Placing a melody in a different octave, or a different key, or a different chord progression, has a subtle but instantly noticeable effect. You can arrange the same song a dozen different ways. “Other things being equal”? Not very likely.
The only way to construct musical samples that are “equal” in all respects but one is to create lifeless clinical notes, things that can only exist in laboratories, not in concert halls. And that leads to a disproportionately fascinating point brought up by Liberman: In an unpublished study, subjects with musical aptitude were played computer-generated musical samples in which the only difference was between a major chord and a minor chord (raised or lowered third, for you theory folks). The subjects were by no means tone deaf or musically illiterate, but they couldn’t hear the difference.
That’s completely flabbergasting to me, but then I have perfect pitch and have spent a good part of my life determining the difference between notes as a profession, more or less, so I’m probably not the ideally objective clinical subject. If you’re inclined, you can test yourself with the audio samples on the link; I’d be very interested to know your results.
But, whatever the cause of that, it serves nicely to make the point: Even in an arbitrary, artificial, clinical instance where other things are made as equal as they can be, they still aren’t equal, unless they’re so equal it spoils the whole experiment.
That’s why the rest of the rest of the premise doesn’t work: It’s simply wrong to think of all music in major keys as “excited, happy, bright, or martial”, and all music in minor keys as “subdued, sad, dark, or wistful.” Not true at all.
I’m sure the examples that spring most readily to our minds do fit those patterns– “Happy Birthday” and the “Bridal Chorus” are in major tonalities, while the “Funeral March” is minor, to pick some painfully obvious ones. But escape the realm of the obvious into real-life music, and I’m sure you can find plenty of examples that work well outside the perceived box:
- That happy, bright, and cheery Christmas carol “God Rest Ye Merry, Gentlemen“? Minor key.
- Does any piece of music known to man more purely express “wistfulness” than Paul McCartney’s “Yesterday“? Major key.
- Lest you think this is only about associations with lyrics, have a listen to Chopin’s subdued, melancholy, and wistful waltz in A flat major (op. 69, no. 1).
- And in case all these melancholy major-key songs are getting you down, how about Hans Zimmer’s excited, happy, bright, and martial theme music from “Pirates of the Caribbean”? In a minor key, savvy.
- If you were wondering, my latest composition “Mick McGuire” is jaunty, cheeky, comical, and minor.
- [I bet you've got an example or two of your own. Add it in the comments!]
An exception may prove a rule, but a whole bunch of exceptions disprove it. Don’t misunderstand; although I’m pointing out the flaw of the simplistic thinking that 
says major is always happy and minor is always sad, I’m certainly not attempting to fall off the other side of Martin Luther’s horse. My point is simply that, although harmony and tonality clearly have some degree of built-in “affective reactions,” they aren’t nearly as distinct as the researchers seem to think. It’s not much good trying to discover the causes of a phenomenon that doesn’t exist.
I haven’t even gotten into another obvious fact: “Major” and “Minor” are fairly arbitrary and recent developments in western harmony–a way to adapt melodic modes to triadic harmony–and not even all western music follows them exactly. Lots of modern jazz, blues, and rock, for instance, can’t be placed strictly into “major or minor” categories at all, as they can be based on modes rather than scales. It’s even more pronounced if you listen to music from other cultures: Jewish Klezmer music, for instance, usually sounds “minor” to western ears, yet is easily capable of expressing sheer undiluted joy better than most western major-key songs can dream about. (Think “Hava Nagila” or “Hinei Ma Tov” if you’re lost here.) And a good bit of 20th-century art music did away with tonality altogether, to the general bewilderment of the listening public. Again, if you want to discover something universal about music, you’d have to think a lot bigger.
Now in spite of my gleeful nitpicking and faultfinding here–hey, I work as a copyeditor; it’s kind of what I do–there is a positive point to all of this: Let’s let music be music. When music tries too hard to be “equal,” it fails. A song that sounds just like another song is a boring song. But if you listen to enough good music, you’ll hear things that are completely and gloriously different. The differences are what makes them interesting, exciting, and worth listening to. A few obvious parallels to life might be drawn here.
Related posts:
Abigail (19)
OneSurvivor (11)
Meg Moseley (6)
Mr Pond (4)
Dodi (3)
Hillary (3)
Welcome to my world!
Most Commented