Fingerprinting

[Euan Mackinnon]

Attribution in Art is important, for economic reasons if no other.

Several members here are certain that musical compositions, at least those by major composers, have a 'fingerprint'. (I quote a few examples below.) After all we must hear something in the music of a (major) composer that leads us to say, as Chris does "I have never heard this before, but I KNOW that's by Beethoven".

Forensic fingerprinting uses a combination of basic patterns and other features to uniquely identify a given human.

By analogy, is it possible to define the elements that make up a musical fingerprint? Philip and Bonn1827 (to name just two) make several suggestions - see quotes below.

Finally, is it then possible to map these elements into some form of mathematics and thereby take a piece of music, extract its 'fingerprint', and say with a high degree of probability that it is (or is not) by Beethoven, or Mozart, or Schubert, or etc?

I suggest the initial two questions are:

Can music be fingerprinted? And, if so,

What are the minimum component parts necessary to define a given fingerprint?

Euan


Here are just a few quotes from the thread Beethoven's uniqueness.

Chris (2nd March 2010)
I remember hearing pieces on the radio and thinking, "I have never heard this before, but I KNOW that's by Beethoven", and I was always right. It's unmistakable.

Peter (3rd March 2010)
All great composers have a unique fingerprint or else they are mere imitators.

As one might expect, Philip goes further and suggests what the elements of any fingerprint in (Beethoven's) music might be:

Philip (5th March 2010)
I think it is a little bit simplistic to reduce B's "fingerprints" to intervallic quanta; after all, there isn't a lot of choice. Beethoven, taking his lead from Haydn perhaps, favours the monothematic approach, where the motivic "germ" functions as both "melody" (if one can call such fragments 'melody') and "accompaniment". I would rather agree that his personal "stamp" resides in his idiosyncratic harmony, instrumental idiom and orchestration/instrumentation. To reduce Beethoven's melodic fingerprint to being pre-eminently focused on the interval of the 4th or 5th is, I feel, erroneous. It is, at its most basic level, simply a reflection of a harmonic (and classical) cliché : IV - V - I (or subdominant - dominant - tonic). (My emphasis)

And in a postscript:

Philip (5th March 2010)
And above and beyond the intervallic / harmonic, we must not forget B's rhythmic "punning", which is also a key fingerprint.

Bonn1827 adds the idea of 'movement across time':

Bonn1827 (2nd March 2010) Musical "fingerprints" are so obvious, yet so inexplicable in many ways. Yesterday I was listening to Elgar's "Introduction and Allegro" with my (non-musical) husband. I said that, on first acquaintance, once could say it was Dvorak that we were hearing but that it would shortly become obvious that it had "Elgar's fingerprints" all over it. I tried to explain what this meant and crudely suggested it was his soaring, noble melody line reminiscent of your basic "Pomp and Circumstance" or the symphonies. But that, somehow, didn't seem like an adequate explanation. LvB's fingerprints are unique and I tend to think it has something to do with the explosive passion and temperament of some of the work, but also his chord and harmonic progressions. You could play a "Beethoven chord", for example, and smile knowingly to yourself (like a mad relation?). (My emphases)

[Sorrano]

Great topic, Euan! Somewhere and long ago I read an article about fingerprints in music and there was quite a bit about Beethoven with specific examples. Just from memory, and this was years ago, I recall that there was a descending note pattern that Beethoven used frequently during certain dramatic situations in the music. Unfortunately, I don't remember anything else, but do wish to note that there is probably information out there and that studies have been done.

[Quijote]

Blimey, Euan. I see that I shall have to be very careful about what I post in future. I stand by what I said, by the way. Indeed, as you say, I could go even further, but I'll hold my tongue for the moment.
I would like to say that Ed C's "colour analyses" would offer no insights to this concept of "fingerprints" nor hold water in a comparative analysis either between two works by the same composer, or two works by different composers. This does not mean (read this carefully, everybody) that I am pouring scorn on what clearly represents a lot of work, I am merely questioning the principles underlying the exercise.
Euan, you don't say if you yourself can identify "fingerprints" by just listening.

[Sorrano]

There are certain composers who do have some very distinctive styles. For example, I find it very difficult not to recognize a Bruckner Symphony. Beethoven is another composer that I find readily recognizable. Why, I don't always know. Even between two contemporaries, such as Mozart and Haydn, I find some distinctive differences that I can often figure out which is which. (Of course, with Bruckner, there is always that giveaway 2+3.)

[Quijote]

There are certain composers who do have some very distinctive styles. For example, I find it very difficult not to recognize a Bruckner Symphony. Beethoven is another composer that I find readily recognizable. Why, I don't always know. Even between two contemporaries, such as Mozart and Haydn, I find some distinctive differences that I can often figure out which is which. (Of course, with Bruckner, there is always that giveaway 2+3.)

Sorrano, you are a naughty boy, as you well know I am forbidden to make reference to any 2+3 rhythms for reasons I cannot go into.

[Quijote]

There are certain composers who do have some very distinctive styles. For example, I find it very difficult not to recognize a Bruckner Symphony. Beethoven is another composer that I find readily recognizable. Why, I don't always know. Even between two contemporaries, such as Mozart and Haydn, I find some distinctive differences that I can often figure out which is which. (Of course, with Bruckner, there is always that giveaway 2+3.)

This is becasue you have two functioning ears and a brain in between, Sorrano.

[Sorrano]

This is becasue you have two functioning ears and a brain in between, Sorrano.

Yes, the ears, I was aware of those.

[Megan]

I can usually guess right If I am listening to say, Sibelius, Arvo Part, Mozart, Haydn, Beethoven as I did that just today, Sometimes I get it right with Wagner. Handel, and most definately Vivaldi , and Monteverdi. Bach and Biber , most of the time but not always.

[Sorrano]

For those who really wish to delve into some of the mathematics of music, here is a link I ran across. Math was not my best subject so much of this is a bit over my head. But there may be some tidbits that the more adept here might glean.

http://us.metamath.org/mpegif/mmmusic.html

[Quijote]

I can usually guess right If I am listening to say, Sibelius, Arvo Part, Mozart, Haydn, Beethoven as I did that just today, Sometimes I get it right with Wagner. Handel, and most definately Vivaldi , and Monteverdi. Bach and Biber , most of the time but not always.

You too have two functioning ears and a brain, Megan. I am jealous. Well, at least as far as the Sibelius goes. You like Arvo Pärt? Or rather, you listen to Arvo Pärt? Can we talk about him for a few postings?

[Euan Mackinnon]

Philip 0:22 today (My emphases)
I see that I shall have to be very careful about what I post in future. I stand by what I said, by the way. Indeed, as you say, I could go even further, but I'll hold my tongue for the moment.

I hope you won't 'hold your tongue' for too long especially with respect to the two questions I posed in the initial posting. I, for one, am very interested in what you have to say.
(The same applies to anyone else on this forum - Sorrano, for example - he/she started me off on this track.)

I would like to say that Ed C's "colour analyses" would offer no insights to this concept of "fingerprints" nor hold water in a comparative analysis either between two works by the same composer, or two works by different composers. This does not mean (read this carefully, everybody) that I am pouring scorn on what clearly represents a lot of work, I am merely questioning the principles underlying the exercise.

'[T]he exercise'? Do you mean Ed C's exercise or 'fingerprinting' along the lines I raised in this thread?

Euan, you don't say if you yourself can identify "fingerprints" by just listening.

No, I can't really claim to be able identify 'fingerprints' by "just listening]"

I'm afraid I have absolutely no musical training or ability. That said, I often feel that I can recognise the style of a composer - particularly Beethoven, Verdi, Brahms, Bruckner (perhaps), Schubert, Mozart (although I could easily mistake Haydn or J C Bach for Mozart), J S Bach, and so on. I would have absolutely no chance of identifying (or even knowing about) most of the composers you, Philip, often mention.

However, I think it is highly likely that there are ways to 'fingerprint' composers like Beethoven and, for me, mathematics offers the likeliest route. Hence the thread.

Euan

[Megan]

You too have two functioning ears and a brain, Megan. I am jealous. Well, at least as far as the Sibelius goes. You like Arvo Pärt? Or rather, you listen to Arvo Pärt? Can we talk about him for a few postings?

Ok, have started a new thread

[Megan]

I think most composers worth their salt can be identified after a few bars. But for me, a more interesting question would be, who can you not tell who it is, after really even quite a long intro.
I forgot to mention that I can identify if I am listening to Brahms.

I am afraid I am at a loss Euan, about having to go the mathematics route.


I think the early symphonies of Shostakovitch is difficult to recognize.

[Euan Mackinnon]

For those who really wish to delve into some of the mathematics of music, here is a link I ran across. Math was not my best subject so much of this is a bit over my head. But there may be some tidbits that the more adept here might glean.

http://us.metamath.org/mpegif/mmmusic.html

Sorrano

Thanks for the reference. I have bookmarked it and will get back to you when I have read it.

All the best

Euan

[Sorrano]

The key in this thread to recognizing a given composer is the why. What is it that is so recognizable about Sibelius, Beethoven, etc.? Do we recognize the composers in various genres and instrumentation equally?