Is Music Really Universal?

[emcee]

again could be wrong, isn't there some kind of Gregorian Monk thing that is the one note sung by them in a monastery high in the Andes that has no rhythm and lasts 40 minutes?

And the band on tattoine had earth like music

[aciddose]

It can be broken down to the level of entropy.

A single tone has very low entropy consisting of amplitude, frequency and duration. It's certainly multi-dimensional although only a single vertex.

White noise has the highest entropy nearing 100% although often significantly less so for common noise sources.

Music sits somewhere in between and further toward the low end of the scale.

With this objective definition in hand I believe it is possible to define the lowest entropy which could be considered melody or rhythm. This may very well be a single tone.

It should also be possible to define an upper limit beyond which it becomes unlikely for any human to objectively identify these elements as a part of the signal. (Multiple subjects giving the same definitions for the same signals.)

This may be a Gaussian distribution or a skewed Gaussian, it would be very interesting if it were otherwise.

[emcee]

It's really too early in the day over here to be pondering all that
My take - music as in broadly "organised noise" taking out different cultural differences (jungle drums vs Gregorian Chant vs Kanye vs Psy etc etc) is universal as every culture has it. Without resorting to Google, even ancient Egypt had music, displayed through hieroglyphic's - unfortunately they didn't have Pro-tools back there to record it but if we could hear a "tune" from back then we'd still identify it as music.. even if it were simply rhythmic clapping.

[JumpingJackFlash]

Without resorting to Google, even ancient Egypt had music, displayed through hieroglyphic's - unfortunately they didn't have Pro-tools back there to record it but if we could hear a "tune" from back then we'd still identify it as music.. even if it were simply rhythmic clapping.

Indeed, written music exists going as far back as 1200 BCE; it is still identifiable - in fact, it is perhaps remarkable how recognisable it is. But there's no way we can possibly "hear" it the same way they did back then.

[stonestreet]

I like this explanation from acidose.

These sorts of questions (thread title) have been going on a long time. I read an article titled "Disorder/Order: Discovery or Invention?"

I tend to favor Invention. I admit though that this is just one verse of the multiverse.

Have fun.

It can be broken down to the level of entropy.

A single tone has very low entropy consisting of amplitude, frequency and duration. It's certainly multi-dimensional although only a single vertex.

White noise has the highest entropy nearing 100% although often significantly less so for common noise sources.

Music sits somewhere in between and further toward the low end of the scale.

With this objective definition in hand I believe it is possible to define the lowest entropy which could be considered melody or rhythm. This may very well be a single tone.

It should also be possible to define an upper limit beyond which it becomes unlikely for any human to objectively identify these elements as a part of the signal. (Multiple subjects giving the same definitions for the same signals.)

This may be a Gaussian distribution or a skewed Gaussian, it would be very interesting if it were otherwise.

[woggle]

It can be broken down to the level of entropy.

A single tone has very low entropy consisting of amplitude, frequency and duration. It's certainly multi-dimensional although only a single vertex.

White noise has the highest entropy nearing 100% although often significantly less so for common noise sources.

Music sits somewhere in between and further toward the low end of the scale.

With this objective definition in hand I believe it is possible to define the lowest entropy which could be considered melody or rhythm. This may very well be a single tone.

It should also be possible to define an upper limit beyond which it becomes unlikely for any human to objectively identify these elements as a part of the signal. (Multiple subjects giving the same definitions for the same signals.)

This may be a Gaussian distribution or a skewed Gaussian, it would be very interesting if it were otherwise.

many years ago I had one of my masters students run some experiments on categorising music based on a measure similar to continuous entropy then modelling that measure and discriminating on the coefficients of the models. Showed some promise