Piano World Home Page Forums Our Most Popular Forums Adult Beginners Forum Debussy plays Clair de Lune

I'm not sure I get your point.

One of the big issues with singers (I'm recalling from reading treatises years ago) around Schubert's day and after was that pianists were starting to follow the singers, rather than providing the tactus that they were used to with an orchestra that allowed them to sing around the beat.

The comment was, and I paraphrase, "What are you doing? I don't know where the beat is. Just play the bloody notes!"

As far as piano rolls vs. acoustic recordings, I don't have any evidence that there was sufficient difference in the recording technology to cause random massive fluctuations in tempo. You are implying that the publisher, either willfully, or with Debussy's permission, somehow took a direct punched-to-master roll and meticulously altered the relative spacing of the holes in order to create the huge fluctuation in tempo, rather than accepting the possibility that the unfettered solo performer might choose to play his work in the way recorded.

It would be possible to take the master roll, and while playing it back, have another second generation master made from it while someone altered the speed of the first master in order to create the tempo flucuations. Why would they do that? Either that is what the public expected, or they wanted to destroy Debussy's career?

Isn't the more likely solution found in Occam's Razor? Debussy actually plays that way.

I've heard lots of piano rolls recordings over the years, via BBC radio - everyone from Busoni to Mahler to Gershwin. Every one of them didn't sound anything like what their acoustic recordings did. Part of the explanation is here:


Hand played rolls are created by capturing in real time the hand-played performance of one or more pianists upon a piano connected to a recording machine. The production roll reproduced the real-time performance of the original recording when played back at a constant speed. It is industry convention for recordings of music intended to be used for dancing to be regularized into strict tempo despite the original performance having the slight tempo fluctuations of all human performances, as due to the recording and production process, any fluctuations would be magnified/exaggerated in the finished production copy and result in an uneven rhythm.

Reproducing rolls are the same as hand-played rolls but have additional control codes to operate the dynamic modifying systems specific to whichever brand of reproducing piano it is designed to be played back on. Reproducing pianos were beyond the reach of the average home in the original era of popularity of these instruments and were heavily marketed as reproducing the 'soul' of the performer - slogans such as 'The Master's Fingers On Your Piano' or 'Paderewski will play for you in your own house!' were common.

I also go with the theory that these rolls fluctuate a little bit, or at least can magnify fluctuations. The bit that Bennevis has bolded above suggests artificial rhythmic distortion in the finished product, designed to "bend" fluctuations back into place, but ending up magnifying some of them as an unfortunate side effect.

Reminds me of the modern day auto-tune a little bit.

bennevis: So you are saying that the Debussy 'Clair de Lune" was meant strictly for dancing? If so, the company didn't do a very good job of getting it 'regularized into strict tempo." Or, are you saying that they simply used the same production process for all piano rolls, whether or not they were meant for dancing. Could be.

I agree that piano rolls don't sound anything like acoustic recordings, but then I don't think DPs sound anything like APs either. You fake something, it is going to sound fake. But, I don't believe for a second that any piano roll artifact of production could possibly account for the Debussy recording I posted. Were are talking seconds here, not milliseconds. Come on, science has to enter here somewhere.

zaphod: Did you listen to the recording? The level of tempo variation is, to my mind, well beyond any artifact of production or reproduction. You say '...fluctuate a little bit...'. I agree. Even a magnification as an unfortunate side effect couldn't possibly follow the logical variations that Debussy made in note lengths, as well as tempi.

I'm surprised you don't understand my message: I thought it's pretty clear what I meant, and why I quoted the article. Please read the article again.

I'm saying the Debussy wasn't meant for dancing, therefore the piano roll wasn't regularized into strict tempo, hence the exaggerated tempo fluctuations - which is what the article says.

I suggest you listen to pianists from the early 20th century who made both piano rolls and 78s, and compare the rolls to their acoustic recordings.

I've read it again, and I don't understand your point.Yes, and I understand that to mean that we hear precisely the way Debussy played it.I have, and it is easy to hear through the minor differences, much in the same way we can tell an S&S D from a Acrosonic, even on a cheap car radio.

I am not saying I am correct. I often get something wrong until the correct answer is beaten into me or I have an epiphany. You're not beating me hard enough yet.

What I take from this quote: "...despite the original performance having the slight tempo fluctuations of all human performances, as due to the recording and production process, any fluctuations would be magnified/exaggerated in the finished production copy and result in an uneven rhythm."

Is that we are not hearing precisely the way Debussy played it because any recording that was not regulated into a strict rhythm for dancing would, conversely, have exaggerated fluctuations in tempo. The fluctuations in tempo, then, are not Debussy's but the technology's.

Regards,

BruceD, I interpreted it just as you did.

I must disagree to some extent. I have heard tape wow, 78rpm warbling, piano roll artifacts, seen hand turned movie camera footage, and the like, and none of them, by any stretch of the imagination, showed random, exaggerated time based activities that were not originated by the performer. They all, however, show, cyclical (even if variable due to the hand turning of the crank) variations in tempo, easily discernable and easily discountable.

OK, I've gone back and listened to this recording several times. Each time I hear what I think are some piano roll artifacts, but each time I get more and more involved in Debussy's peformance. When I first heard this, it was so strange and frightening, I almost had to turn it off. Now, all I hear is the incredible passion and musicality of this man. This is the real thing, artifacts or not.

I challenge you to sit back, ignore the noise, and actually listen to the man play. This is really something! Stop boxing yourselves in.

So, I'm quite new on the forum, but based on the threads I've read so far, I gather you guys just love a good old debate. It's all good fun though, and very interesting.

I never knew there was such a thing as HIP (that's a real field of knowledge, I take it), but it's great to have discovered it. And yes, I actually like this recording, it doesn't put me off at all; but then, I've only just started learning to play the piano, so maybe it's one of those things where you're more open to weirdness when you're not part of the crew. Thanks again for posting, and debating the historical accuracy of the interpretation!

I think there are two distinct lines of conversation/debate in this thread:

- One about how much Debussy deviated from his own music that he wrote years before this recording was made.
- And one about to what extent the tempo fluctuations in the recording stem from the particular recording equipment used, or reflect the way Debussy actually played.

I'm mostly following the first line of discussion, as I'm Jon Snow when it comes to recording techniques either contemporary or historical. And I think it goes to show how much more liberal the interpretation of music was in those days. I mean, we've all heard the stories about how Horowitz actually *changed* bits of Rachmaninov's works - with Rachmaninov's blessing, at that. Imagine being able to walk over to the likes of Rach and say "hey mate, I've changed around the ending of this thing, hope you don't mind," and him going: "Sure, work away, lad..."

This reminds me of discussions we've had about modern interpretations of Romantic-era music, and how some of it was considered "wrong" by some because it didn't follow the instructions from the composer to the letter. You could of course argue, that with the historical distance we have to the time when this music was written, it behooves us to be respectful and stick to what's there, because we no longer have the ability to take liberties according to what was the custom of a time period we've never experienced.

I don't think we'll ever all agree on these things, but I love listening to people's opinions about them and inevitably learning from them.

Not sure I get the Job Snow reference, but I’m definitely Team Targaryen / Stark.

People build off others’ knowledge and creativity all the time, in music and art, theater and film, math and science etc. Otherwise how would progress occur. Someone used Edison’s electricity model to further home lighting, which evolved into heating and air, etc. Someone used Watson and Cricks DNA model to further gene therapy, and they in turn would have been nowhere without Mendel. The wheel became the cart which became the wagon which became the automobile.

So why not use, interpret and further music of the great composers? I’m not of the level to improve upon it, or even close, but Horowitz certainly can add his touches to a Rach piece (or even dare I say Chopin?) and create something new.

Just means I "know nothing" in this respect

Completely agree with you on the subject of interpretation, btw.

Gotcha, Ygritte!

After having listened a few more times, I see nothing wrong with the way Debussy plays. HE wrote it, and can play it anyway he likes. Maybe he was just experimenting, or that was how the music moved him that day?

Those changes were nothing like the ones Debussy makes in the recording. They were mostly cuts or reshuffling in what Horowitz felt was an overly long piece. Even when Horowitz wrote significantly different arrangements of works like his versions of the Liszt Hungarian Rhapsodies No. 2 and Rakoczy March he did not play with an unsteady tempo.

I am still hoping that someone can post an historical recording by someone other than Debussy where the tempo fluctuations from the written score are as extreme as in the Clair de lune recording. If none exist, then I would conclude that Debussy is one off case.

pianoloverus - Time to accept that Debussy played Clair de Lune the way you hear it being played. If it weren't the case, he would NEVER have let the rolls be published. Here is a comment regarding the recordings AND a letter from Debussy.

"In 1913, Debussy recorded several pieces onto piano rolls using the Welte-Mignon reproducing piano system. This system was capable of recording the pedaling and dynamics of the performer as the performer played the instrument. There were no corrections or editing after the recording. This was the only technology at the time with this capability. Debussy was very happy with the reproduction quality of the Welte system. He went so far as to write to Edwin Welte the following note:

Paris, 1 November 1913

Dear Sir,

It is impossible to attain a greater perfection of reproduction than that of the Welte apparatus.
I am happy to assure you in these lines of my astonishment and admiration of what I heard.

I am, Dear Sir,
Yours Faithfully,
Claude Debussy

This is the only recording Debussy made of his solo piano works. "

You can image his astonishment at hearing, not a scratchy, low fidelity acoustic recording of his playing, but a real acoustic piano playing away in all its high fidelity and intimacy, as if he were sitting at the piano playing it himself.

I've basically accepted that even before any discussion about possible tempo changes due to recording techniques of the day.

What I'm really interested is if there are other recordings by pianists of the same era that exhibit such extreme tempo fluctuation.

I posted one by de Pachmann(the Mozart Rondo) that had pretty extreme fluctuations but perhaps not quite as extreme as the Debussy. But other recordings by de Pachmann, while probably extreme in fluctuations by modern standards, are nowhere near the Debussy CdL. And the Paderewski recordings I have heard are nowhere near as extreme in tempo fluctuation as the Debussy CdL although they are extreme in terms of chord arpeggiation.

I briefly listened to as much as I could stomach(the first minute or so of each piece) of the much longer posted Debussy recording of a large number of his pieces and felt that about half of them were as extreme as his CdL recording but the other half were less extreme.

Thank you for your thoughts. If we accept that this style, what we would call extreme playing, was normal and even anticipated at that time, it is a bit of a window into the expressive emotions of the day.

Like you, I don't play pieces that way in public today, but, at home, when only I am present to explore the emotional impact this music has on me, I am sure that I engage in the same very extreme fluctuations in tempo. I don't think that, in doing so, I am playing musically, just luxuriating in the sound of our great piano. I bet you do the same.