The Great Mahler Debate

[Sergeant Rock]

but i don't think he cared for blondes..... he married a brunette

But all his girlfriends were blonde. He had to marry a woman his mother would approve of.

Sarge

[mahlertitan]

Gustav liked the blondes, i know that much.

[zamyrabyrd]

This is a totally different subject, but I have to comment on this since this is the stuff that really interests me.  

The more I think about it, it seems to me that every language is equally difficult to learn, if you plan to be able to speak AND write it. It seems for everything easy that there is for a language, there's also something hard.

Language and Music might be a good subject for the Diner.

What I was hoping to say in a few short paragraphs (impossible, though) that defining spectrum of events might be different cross-culturally in music as it is in language. In tonal music as in a Mahlerian Symphony events are usually harmonic, rather than polyrhythmic, for instance. The proportional shape of a motive usually remains the same even if it is fragmented, augmented or subject to dimunition. In fugues this is what makes a theme recognizable.

In other words, one usually doesn't have to deal with complexity on all levels, some fields are relatively stable or simpler. There seems to be a similarity to language in this respect cross-culturally. I still think 71db's idea, "spectrum of events" in music might be an interesting subject.

ZB

[greg]

Language and Music might be a good subject for the Diner.

What I was hoping to say in a few short paragraphs (impossible, though) that defining spectrum of events might be different cross-culturally in music as it is in language. In tonal music as in a Mahlerian Symphony events are usually harmonic, rather than polyrhythmic, for instance. The proportional shape of a motive usually remains the same even if it is fragmented, augmented or subject to dimunition. In fugues this is what makes a theme recognizable.

In other words, one usually doesn't have to deal with complexity on all levels, some fields are relatively stable or simpler. There seems to be a similarity to language in this respect cross-culturally. I still think 71db's idea, "spectrum of events" in music might be an interesting subject.

ZB

That's pretty interesting, I wonder if they teach stuff like that in music school. I wonder what language Babbitt could be compared to, having a spectrum of events that changes constantly, in all fields (harmony, rhythm, dynamics, tone color, etc.). Maybe the language of unicorns?

[71 dB]

I still think 71db's idea, "spectrum of events" in music might be an interesting subject.

ZB

My initial motivation to come up with this concept was realising that so called popular or commercial music doesn't have much long or short events and everything happens in medium time frame. The spectrum of events is narrow. More sophisticated music like classical has broader spectrum. There are lots of long events like development of thematic material and short events as well. It takes more effort from a listener to follow long and short events simultanuously. this "difficulty" makes music much more interesting at least for me.

If we have say 5 "event frequency" points in our spectrum we might have for a pop track:

Very long events: 1 %
Long events: 8 %
Medium length events: 70 %
Short events: 20 %
Very short events: 1 %

For a classical piece the spectrum would be much more balanced, perhaps:

Very long events: 15 %
Long events: 22 %
Medium length events: 25 %
Short events: 21 %
Very short events: 17 %

[MishaK]

My initial motivation to come up with this concept was realising that so called popular or commercial music doesn't have much long or short events and everything happens in medium time frame. The spectrum of events is narrow. More sophisticated music like classical has broader spectrum. There are lots of long events like development of thematic material and short events as well. It takes more effort from a listener to follow long and short events simultanuously. this "difficulty" makes music much more interesting at least for me.

If we have say 5 "event frequency" points in our spectrum we might have for a pop track:

Very long events: 1 %
Long events: 8 %
Medium length events: 70 %
Short events: 20 %
Very short events: 1 %

For a classical piece the spectrum would be much more balanced, perhaps:

Very long events: 15 %
Long events: 22 %
Medium length events: 25 %
Short events: 21 %
Very short events: 17 %

How do you define "event" and how do you define "long"? What is an event? One figure of an ostinato pattern in a Bruckner symphony may be one short event that is but one element of a very long event which is the exposition section, which may be made up of several less long events. Classical music is full of ideas within ideas within ideas. Its wealth derives exactly from the possibility of looking at it simultaneously through a microscope and a telescope.

An entire cycle by Zinman, Sym no 1 is just available. The new Baremboim 9th. Another 9th Sinopoli, I think never previously issued. A new Mahler 1 Jansons with the Concertgbouw, a new Chicago SO No 2 from Haitink, a reissue of Rattle's Das Lied from EMI. Apart from this we know Boulez is recording the 8th. And so it goes.

The Haitink/CSO is a No.3, not No.2. I guarantee it will be outstandingly excellent, as I was at the concert when this was recorded. The Barenboim 9 is also fantastic. See also this discussion.

[Israfel the Black]

My initial motivation to come up with this concept was realising that so called popular or commercial music doesn't have much long or short events and everything happens in medium time frame. The spectrum of events is narrow. More sophisticated music like classical has broader spectrum. There are lots of long events like development of thematic material and short events as well. It takes more effort from a listener to follow long and short events simultanuously. this "difficulty" makes music much more interesting at least for me.

If we have say 5 "event frequency" points in our spectrum we might have for a pop track:

Very long events: 1 %
Long events: 8 %
Medium length events: 70 %
Short events: 20 %
Very short events: 1 %

For a classical piece the spectrum would be much more balanced, perhaps:

Very long events: 15 %
Long events: 22 %
Medium length events: 25 %
Short events: 21 %
Very short events: 17 %

The difficulty in such a comparative analysis is the fact that a pop song is merely a 3 to 5 minute track of a single musical idea. This is not an inferior form of construction, as it roots from the Art Songs of the early Romantics. It would be more respective to compare let's say a symphony, an opera, or a complete full size orchestral work, which usually consists of about 45 minutes to an hour of music, to a full-length pop rock album. The spectrum of events, so to speak, would be slightly more reasonable to compare in this case. I view modern albums as the equivalent of musical suites or song cycles in terms of scale and creation. Therefore, to be fair, a song from say Schubert's Winterreise, when compared to a song from Radiohead's OK Computer, would likely contain a slightly less balanced spectrum of events to the latter.

[mahlertitan]

My initial motivation to come up with this concept was realising that so called popular or commercial music doesn't have much long or short events and everything happens in medium time frame. The spectrum of events is narrow. More sophisticated music like classical has broader spectrum. There are lots of long events like development of thematic material and short events as well. It takes more effort from a listener to follow long and short events simultanuously. this "difficulty" makes music much more interesting at least for me.

If we have say 5 "event frequency" points in our spectrum we might have for a pop track:

Very long events: 1 %
Long events: 8 %
Medium length events: 70 %
Short events: 20 %
Very short events: 1 %

For a classical piece the spectrum would be much more balanced, perhaps:

Very long events: 15 %
Long events: 22 %
Medium length events: 25 %
Short events: 21 %
Very short events: 17 %

did you get these data from the "Spectrum-meter"?

[karlhenning]

did you get these data from the "Spectrum-meter"?

The same measuring device wherewith he determined that he uses 10x as much of his brain to listen to Elgar as to Sibelius 

[Florestan]

I wonder what very long events might mean for a 3-minutes pop music piece.

[greg]

The same measuring device wherewith he determined that he uses 10x as much of his brain to listen to Elgar as to Sibelius 

makes me wonder how he even understands Elgar 

[karlhenning]

I wonder what very long events might mean for a 3-minutes pop music piece.

Letting the high-hat ring? 

[71 dB]

How do you define "event" and how do you define "long"? What is an event? One figure of an ostinato pattern in a Bruckner symphony may be one short event that is but one element of a very long event which is the exposition section, which may be made up of several less long events. Classical music is full of ideas within ideas within ideas. Its wealth derives exactly from the possibility of looking at it simultaneously through a microscope and a telescope.

Straighforward technical way to detect events is to use some kind of modified Cepstral analysis. Perhaps even as simple method as Spectral analysis to squared (=power signal) and low-pass filtered audio signal would do the trick.

Mentally we detect the events pretty much automatically. Every event should create a vibrational field in our brain.

The difficulty in such a comparative analysis is the fact that a pop song is merely a 3 to 5 minute track of a single musical idea. This is not an inferior form of construction, as it roots from the Art Songs of the early Romantics. It would be more respective to compare let's say a symphony, an opera, or a complete full size orchestral work, which usually consists of about 45 minutes to an hour of music, to a full-length pop rock album. The spectrum of events, so to speak, would be slightly more reasonable to compare in this case. I view modern albums as the equivalent of musical suites or song cycles in terms of scale and creation. Therefore, to be fair, a song from say Schubert's Winterreise, when compared to a song from Radiohead's OK Computer, would likely contain a slightly less balanced spectrum of events to the latter.

Yes, pop music is very different from art music and that shows in the "spectrum of events". Comparing them is relevant as it illustrates the huge differencies. It tells us that one reason people into pop music have difficulties with classical music is because they are used to narrow spectrum of events. We can always think about what are the optimum spectrums of events for different music genres?

did you get these data from the "Spectrum-meter"?

I get them from my head. I made them up just to illustrate the concept of my ideas.

[mahlertitan]

I get them from my head. I made them up just to illustrate the concept of my ideas.

and that's exactly what the problem is, i hope you realize that by now.

[71 dB]

I wonder what very long events might mean for a 3-minutes pop music piece.

It would be practical to use logarithmic time scale. For example:

Very long events: 16 - 64 s
Long events: 4 - 16 s
Medium length events: 1 - 4 s
Short events: 250 - 1000 ms
Very short events: 63 - 250 ms

For events longer than 64 s or shorter than 63 ms we would need event frequency points "extremely long events" and "extreme short events".

and that's exactly what the problem is, i hope you realize that by now.

And Mahler's music didn't come form Mahler's head? 

[karlhenning]

It would be practical to use logarithmic time scale. For example:

No, actually, this strikes me as some of the most impractical discussion of music I've ever encountered.

[Israfel the Black]

Yes, pop music is very different from art music and that shows in the "spectrum of events". Comparing them is relevant as it illustrates the huge differencies. It tells us that one reason people into pop music have difficulties with classical music is because they are used to narrow spectrum of events. We can always think about what are the optimum spectrums of events for different music genres?

I believe you completely misunderstood my post. I was making the point that you were comparing a 2-3 minute pop song to a classical composition of undefined length and genre. As per example, if compared to an art song of say Schubert or Brahms, a modern song (depending on the genre/artist) probably contains a more complex spectrum of events than a repeated piano melody and vocal harmony. It's a bit rash to compare six centuries in vast different genres of orchestral music to a single musical form that is mostly exercised in the song-style. I'm pointing out the flaw in the initial comparison, not acknowledging simple differences in the quantitative so-called spectrum of events. If you are using a modern song as comparison to classical music, then compare a song to a song. It's quite simple reasoning.

[71 dB]

No, actually, this strikes me as some of the most impractical discussion of music I've ever encountered.

This strikes me as some of the stupidest comments I've ever encountered. You are overwhelmed by my intellectual contribution to this thread and this is what you do, write pointless comments.

Note lenghts in music are defined in logarithmic way (1/1 note, 1/2 note, 1/4 note,....)
Logarithmic time scale IS most practical in music.

[MishaK]

Mentally we detect the events pretty much automatically. Every event should create a vibrational field in our brain.

If our brains detected and processed all musical "automatically" none of this discussion, or indeed this forum, would be needed at all. While some people definitely have more of an intuitive grasp than otehrs, you still need to have a conceptual idea of the things you are listening to, the "grammar" of the music if you will. Otherwise it will just completely bypass you (which is apparently what is happening with much of Mahler for you at the moment).

As per example, if compared to an art song of say Schubert or Brahms, a modern song (depending on the genre/artist) probably contains a more complex spectrum of events than a repeated piano melody and vocal harmony.

I highly doubt that, given the harmonic poverty of current popular music, which rarely ever goes beyond the most banal 1-5 cadence.

[karlhenning]

This strikes me as some of the stupidest comments I've ever encountered. You are overwhelmed by my intellectual contribution to this thread

Good God! Yes! That's it, exactly!  I am overwhelmed by your intellectual contribution.

I bow to your brilliance.