Audiences hate modern classical music because their brains cannot cope

[jochanaan]

...Now you say that Schoenberg jives well with Brahms, Mahler etc in terms of orchestration and rhythm.  Okey doke.  But the thing is that what we associate with romantics is not those things, it's the harmony.  When we hear chromatic music we think of Strauss, when that rule is broken it doesn't sound romantic anymore.  The harmony is the key defining feature of music.  Schoenberg's early works sound romantic because they are.  His later works sound very different because they are.  Let me give an example that illustrates how overwhelming important harmony is: Alot of people can listen to and enjoy Bax (he can actually receive radio play) even though he is rhythmically more complex/strange than a traditional romantic, and that is because he is harmonically closer to that chromatic sound than more avant garde composers.

Remember that Wagner and Liszt had both taken tonality to its outer limits with increasingly chromatic, indeterminate harmony.  Schoenberg's atonality and twelve-tone method were only the next inevitable step.

[Scarpia]

Aside from the usually piffle that fills these threads, the research makes an interesting point (at least in the more nuanced version that Brian posted).  An important element of "classical" music is predictability.  When we listen to music at any given music we have a feeling we know what comes next, dissonance gets resolved, one chord tends to progress to another, we feel one note naturally leads to another in a melody.  It is not a a mathematical certainty, but it allows the composer to chose between surprising us or giving us what we expect at each point in the work.  (And of course there are an infinite number or ways to surprise us.)  This can be described by probabilities.  Surprise happens when we hear a note that there is a low probability of hearing in a given musical context.   The researchers found that this characteristic, that one note is more likely than another to follow in a given musical situation, is not found in some forms of modern music.    This suggests that this characteristic of music, predictability and the perception of surprise, is absent from some forms of music.  If this statistical analysis is correct (and it may be a difficult to assess if their analysis is exhaustive enough) this would indicate something missing from some forms of modern music that no amount of listening or familiarity will remedy.

Now, the very speculative conclusion I would propose is that people who find serial music as interesting as tonal music are not, as they would have us believe, more musically astute, but are less musically astute.  The "predictability" of tonal music escapes them, so they do not notice that it is missing in atonal music.  It is like a color blind person who thinks a black and white photograph is just as vivid as a color photograph.   

[jochanaan]

jochanaan, the nineteenth century was a time of great instability, of ceaseless revolution and counter-revolution. It was the time when the power began to shift inexorably to "the people." We think of the twentieth century as volatile; the nineteenth was equally so. (Otherwise, loved your post, just by the way!!)

That's very true, of course; but the nobility and royalty had not yet been quite reduced to the powerless figureheads they are now, despite the revolutions in the air and streets.

...Now, the very speculative conclusion I would propose is that people who find serial music as interesting as tonal music are less musically astute.  The "predictability" of tonal music escapes them, so they do not notice that it is missing in atonal music.  It is like a color blind person who thinks a black and white photograph is just as vivid as a color photograph.   

Present company excepted, I suppose.

[Franco]

I find it a bit interesting that prior to 1700 and after 1850 what we think of the "tonal system" was much less entrenched, and really only the Classical Period composers (1730-1830) can be said to fully exploit the relationships of keys and tones that are the hallmark of "tonality".  This is not to say that J.S. Bach did not compose in the tonal system, just that his treatment of keys was different than Haydn, Mozart or Beethoven, among others, whose exploitation of the relationships of key centers led to the development of the sonata form, which I think of as the quintessential formal expression of tonality.

Yes there is a physics of the tone which produces the basic foundation of the tonal system - but all the notes are there in the array of harmonics, the upper regions are quite dissonant.

The composers after 1850 began a concerted effort at stretching the relationships that Haydn, Mozart and Beethoven brought to an apogee, which over time led to a complete abandonment of the principles of compositional organization using a hierarchal system of keys or key centers.

Arnold Schoenberg attempted to codify this abandonment with a new system of organization based on twelve equal tones.

I consider his system to be a major achievement in musical composition both as a summing up of what had come before and a stage setter for what came after.  But his was by no means the only technique composers chose in the wake of the demise of tonality.

I find listening to atonal music very rewarding - and always been somewhat surprised that it is the focus of such scorn.

[karlhenning]

. . .  An important element of "classical" music is predictability.

Sure.

And another equally important element in art of all media is (let's call it, in light of present discussion) unpredictability.

Nobody (or, hardly anyone) need music which, even when they hear it for the very first time, they can tell you everything that's going to happen, before it happens.

Undeniable interest in the actual, non-tendentious science notwithstanding, attempts to scientify the aesthetics fall on their white-smocked faces (to mix metaphors), IMO.

[karlhenning]

. . . I find listening to atonal music very rewarding - and always been somewhat surprised that it is the focus of such scorn.

Hear, hear.

[Scarpia]

Sure.

And another equally important element in art of all media is (let's call it, in light of present discussion) unpredictability.

Nobody (or, hardly anyone) need music which, even when they hear it for the very first time, they can tell you everything that's going to happen, before it happens.

Undeniable interest in the actual, non-tendentious science notwithstanding, attempts to scientify the aesthetics fall on their white-smocked faces (to mix metaphors), IMO.

The investigators fully appreciated the importance of unpredictability in music, which is why they employed a statistical approach.  If you reach a point in a score and there is a 90% chance that a certain note will follow (6 4 inversion in Mozart) there is a big distinction between hearing an expected note and an unexpected note.  If you reach a point where there are many subsequent notes that are equally likely, (an unstable harmony in a development section) then you are an unpredictable juncture.  In tonal music the full spectrum is employed.  In Mozart there are points where you feel you know exactly what's coming and it does, or when you feel you know exactly what is coming and something else happens instead, and points where you have no idea what is coming.  The composer is playing with you in that way.  If, in some modern music, predictability is absent, that aspect of music is simply missing.

[karlhenning]

If, in some modern music, predictability is absent, that aspect of music is simply missing.

Thank you for focusing on the shortcomings of the scientology here.  There is always context to predictability.

I think, too, that depending on how one quantifies predictability, the allegations of it being an essential aspect are open to question.

[Scarpia]

Thank you for focusing on the shortcomings of the scientology here.  There is always context to predictability.

I think, too, that depending on how one quantifies predictability, the allegations of it being an essential aspect are open to question.

If I understand correctly, the researchers did not impose any model of predictability, they simply looked for mathematically correlations among notes in the score.    When they ran a Bach Fugue through their program, correlations were detected, when they ran some modern music through, no significant correlations were detected.   The discretion of the researcher only comes into play when they assume that the obvious correlations they find are significant.

I did misspeak in saying that the analysis excludes predictability in general, it only address harmonic predictability.  Predictability can also be associated with rhythm, dynamics, orchestration, and other features not covered by the analysis.

[DavidW]

I think, too, that depending on how one quantifies predictability, the allegations of it being an essential aspect are open to question.

I don't think it's open to question, it's already well known that the majority of people want an element of predictability in entertainment and art.  Let me talk about movies, because I know 'em.  That's why movie trailers have spoilers: more people will actually go to see a movie that they know what's going to happen vs. being surprised.  Blockbusters will always succeed by simply vomiting out the Campbell's hero journey.

Music is a dialogue, if you don't understand what the composer is telling you because you don't understand his language, you're not really appreciating the music.  Those that say just let the music wash over you, well congratulations, you've just taken the Eric Anderson route of music appreciation!

Now I say there is good modern music and there is bad modern music.  Good modern music preserves that sense of dialogue, that is the ideal of classicism is still present.  Completely unpredictable music is garbage.  Without a sense of narrative you might as well waste time reading postmodernist drivel instead.

[karlhenning]

. . . Those that say just let the music wash over you, well congratulations, you've just taken the Eric Anderson route of music appreciation!

Yes, whether it's the Unpredictability Wash, or the Predictability Wash.

Completely unpredictable music is garbage.

Is completely unpredictable music even possible?

[DavidW]

Is completely unpredictable music even possible?

Well if you give a double blind test about predicting notes in a piece of music, and the participants perform statistically no better than just guessing, then I think one can conclude that the music is completely unpredictable.

[karlhenning]

Is that Best Practice? ; )

[some guy]

I think we may be limiting the scope of predictability and unpredictability, so that we are dooming ourselves to rotate, predictably, forever, without ever catching that elusive tail of ours.

Let me try to break us out of that cycle.

In any music, you can go up or down or stay the same. ("Up" and "down" are, of course, metaphors.) You can go louder or softer or stay the same. You can go from one sound to another sound or to no sound. You can go from one line to many lines (or one sound at a time to many sounds at a time). You can go from tight harmonies to loose ones or stay the same. All these motions/changes are as true for free laptop/turntable improv as they are for a Bach cantata.

In limiting the study to pitch (to synchronous pitch), I think the researchers have lost sight of all the things going on in any music. The "scientists" we've been talking about seem only interested in whether "people" can predict the next note or not. But that's only a very small part of the vast, complex web of things that can happen in music at any given time.

In any event, however random a "piece" of music is, our brains are not random. We humans can "make sense" of anything we perceive. (Can, note. Not that all humans are going to actually do it!)

[karlhenning]

I think we may be limiting the scope of predictability and unpredictability, so that we are dooming ourselves to rotate, predictably, forever, without ever catching that elusive tail of ours.

That is a predictable behavior in these threads.

Could be what some of us find enjoyable or essential about them . . . .

[some guy]

Indeed!!

[DavidW]

In limiting the study to pitch (to synchronous pitch), I think the researchers have lost sight of all the things going on in any music. The "scientists" we've been talking about seem only interested in whether "people" can predict the next note or not. But that's only a very small part of the vast, complex web of things that can happen in music at any given time.

I disagree.  In science one can only test one part at a time, not all parts at a time.  You seem too close to saying that it's too complex to be studied, and that's an anti-intellectual copout.  Life, the universe and all of its parts will always be more intricate than we can perceive in full, but that doesn't mean that knowledge and understanding can't be gained through simplistic analyses of it.  In fact all of science is done that way.  This is no different.

In any event, however random a "piece" of music is, our brains are not random. We humans can "make sense" of anything we perceive. (Can, note. Not that all humans are going to actually do it!)

I disagree, the study shows otherwise.

[Franco]

Well if you give a double blind test about predicting notes in a piece of music, and the participants perform statistically no better than just guessing, then I think one can conclude that the music is completely unpredictable.

If one group is "guessing" what is the other group doing to predict notes in a piece of music?

[Florestan]

Predictability... mathematical correlations... patterns... neuroscience...

All this leaves me as cold as dead. It's music we're talking about: a form of art, not of science. 

If it sounds good (i.e, if it touches my soul, if it captures my attention, if it makes me thoughtful, sad, melancholy, joyous, pensive or dancing) --- then it's good music for me and I couldn't care less whether it's modern or antiquated, tonal or serial, Boccherini or Bartok.

I'm a musical anarchist. 

[DavidW]

If one group is "guessing" what is the other group doing to predict notes in a piece of music?

Uh no.  The whole point is that each person would be given a sample and then asked to make a prediction.  There is no guessing group, you simply compare to the average score that would be attained from simply guessing.  If it's more than a standard deviation apart (higher) from the score from guessing, then one might conclude that they can actually correctly figure what is happening next.  You could apply this to pop music, classical era, romantic, baroque, modern etc etc If nothing interesting happens with pitch, test for other aspects.  Design similar tests.