Piano World Home Page Forums Our Most Popular Forums Piano Teachers Forum Notation challenges: Feuilles mortes (Dead Leaves)

Debussy Preludes Book II

I talked about some interesting notation choices in this piece, the second prelude of Book II.

The above link is to the whole book.

Keystring has agreed to add images here of specific parts of the prelude. I always have trouble uploading pictures to this site.

Here is the place in question, four major chords with unusual spellings.

Check out, starting with: un peu en dehors, at the ppp mark.

What page is this on the score?

From page 8 to page 9. I merged two pages together, for the section...

The last time this came up I was too shy to say anything in public, but now that it's come around again here's what I see (with some help from LoPresti for learning to tolerate having radically different chords in the top vs. bottom staff):

On the top staff, respelling, Debussy has four major chords in root position with the root doubled: A, Eb, F#, C. (Or maybe A, D#, F#, C, since everything else has sharps, not flats, including the A# half-diminished seventh chord in third inversion in the left hand -- a.k.a. a rootless (F#)9/G# chord.)

There's a pattern to the chords: the roots in the top staff travel up a tritone, up a minor third, down a tritone, and (leading into the next measure) down a minor third.

My new theory for why Debussy chooses the spellings he does is to minimize the numbers of accidentals and notes being accidentalized and then de-accidentalized, or re-accidentalized, in the same measure.

My other theory is that the accidentals have to do with some sort of non-standard scale(s) that is implied by the chords, either together or individually.

OK, go ahead. Explain how I have it all wrong.

Mostly in the face of this music I feel that the little bit of theory I know is far too poor to describe what Debussy is doing.

And this is only based on looking at it; when I got home last night I forgot to try this out at the piano to listen to it. (Shame, shame.) I'll try to remember tonight.

I glanced through what you wrote, and I would never say you "have it all wrong". I didn't post this as a gotcha. I didn't want to play the know-it-all. I simply thought it would be good for everyone to have concrete examples of tricky notation to talk about. It would take people out of the world for the theoretical and put them into the world of the practical.

This particular example actually illustrates how the practical and the theoretical come together.

I think you are very close to right. In fact, there is no reason why what you are describing is not right. Here is what I see:

1) He has in mind some tonality. It's not always easy to see, but it could be C# minor. The last chord in the whole piece is C# major, so that's at least a place to start.
2) He continues to repeat a G# in the bass. That at least suggests some kind of dominant feeling. In more conventional music, you might have: A#7(5)/G# to G# to C#m. But he just sort of leaves that half-diminished chord floating, and eventually simply jumps to F#, the chord.
3) The middle staff is a melody in octatonic. E F#- G A--A# B#-- C# D#--E. Not all of it is there, only: E F#- G A--A#, although C# is in the following chord, major.
4) The roots of his major chords (A, D#, F#, B#) all follow from the octatonic scale. A, D#, F#, B#. The remaining notes in his major chords follow that, and they also follow the spelling a combination of the C# minor scales, if you consider natural and melodic minor.
5) The roots of the major chords form a diminished chord, which itself is also contained in the octatonic scale. This also shows why jazz players will often talk about such scales "belonging" to a particular chord. Obviously it will go great with a vii°7 chord, for example.
6) When you get all done, the only note that is not part of the octatonic scale is G#, which is both in the bass and doubled as the bottom note of the LH chord.

That's what I see.

That, again, is what I see (and hear).

This is where it gets really interesting. Is it a non-standard scale? My own ear tells me that I have heard Liszt use this scale, so he may have been one of the Romantics to exploit it. I have played almost none of his music, so maybe other people who have will weigh in here.

I think you were very close to having the whole thing. If it makes you feel any better, I did not see the octatonic scale at first. I SHOULD have seen it, but Debussy, like all great composers, has a way of disguising what he is doing, and it's not easy to see of hear all of this.

Thank you Gary! Octatonic -- nice to know there's a method to the madness here. So, going to that big F# chord, there doesn't seem to be any preparation, just, bang, he's there. Do the preceding two chords count as preparation in octatonic, or is this just part of what Debussy does: noodle along one way and then pick a chord, any chord, and settle on it?

Debussy was thoroughly grounded in what we call today "Common Practice". When he began to break away from conservative rules, he had to face critics. Nothing new there.

Today we here jolts like this and think nothing of it:

C, Gb, C, parallel major chords. This creates an effect. If you think about it, that is a tritone jolt. When you use root position triads to do this, you are literally using parallelism.

Similar: Play Am, Fm, Ebm. Those are the only three chords John Williams uses in the Darth Vader March, but of course he leaves himself free to keep moving the theme to different keys.

I think the reason Ed keeps arguing for rudiments is that when you jump into the world of "no rules", your choices are infinite. Just using hit and miss methods, you are unlikely to GET out of the box without running into pure chaos.

If we listen to totally untrained and inexperienced players try to "make up music", they end up with a combination of standard, cliche ideas, and odd things that just don't seem to have any reasoning behind them.

There is a reason why composers in the Barogue and Classical period could write so many symphonies, but Beethoven could only write nine. Brahms only four. The fact that Mahler wrote so many is astounding.

If you sit down to write music with no rules, no starting place, it is quite frightening. I don't know anyone who composes without some kind of organization. For instance, in order to write something that sounds totally outside of any tonal base takes PLANNING. It is a bit like planned chaos.

So now I've played through those measures, and as I thought they might, they shimmer. I wonder if shimmering is how I experience diminished harmony? I basically have these qualities of music that I hear (in general I mean, not specifically in this piece)): normal, sad, wierd, shimmering, and fauve.

Have also hunted up Claudio Arrau playing it, and I can actually hear that the harmony is unusual! Yay! This is remarkable for me and my recalcitrant ears to be able to notice. Then there are some parts that sound less unusual, so when I get the whole score printed out (I only printed out the one page, silly me), I can look and see if they correspond to Debussy actually doing less unusual things with the harmony.

This is fun! I really like diminished chords.

Here is yet another interesting point, and for this one I have only an idea of why the spelling here was chosen. This involves measure 1, 2 and 3 only, and the question is about the kind of chords Debussy may have had in mind...

Well, here the spelling seems to be chosen like this: if he has a note in the key signature, he uses it. If he has a white note not in the key-signature, he naturals it. If he has a black note not in the key signature, he uses its sharp name, which is a sensible choice given that it's a sharp key signature.

One way to see his opening chord is as F#7 in first inversion over a random G natural.

So what about that G natural? Inspecting the physical notes played, one discovers that, despite appearances, the bottom three notes of this chord are a basic Gdim triad (can I call it a triad if it's not written as a snowman? I'm going to, because its sound is the sound of a triad. So now it's like he's experimenting with a chord that starts out one way on the bottom (Gdim) and ends up another way on top (F#7). I don't know if there's a name for that.

In the second measure he moves everything down a whole step, and has Fdim that morphs into an E7 on top, and then in the third measure he moves back up a wholestep to where he started. But wait! He layers on even more, in the form of an outlined C# harmony in the low bass of the second measure, which rises a wholestep with everything else in the third measure, to D#.

So that gives us, in the second measure, the notes (not in this order): C# D, E F, G#, B. Which are most of the notes of our friend the octatonic scale (just missing G natural and A#). (And then by parallelism a different octatonic scale in the third measure).

Not that my ears can name any of this when I listen to the piece. It just sounds strange (in a good way) and shimmery and wonderful. But perhaps writing with these unifying ideas gives structure to the piece, so that unusual as it is, it hangs together with the unusualness being related instead of random.

I'm stumped. I posted this, not to show how smart I am, but to get input as to what is going on. My interest is that I think Debussy here is about 1/2 century ahead of himself. It reminds me of something I would expect to hear from the 2950s and later. It's "smokey".

Basically when you have notes clustered together, it is very harsh, but when they are used in open-voicing, almost anything seems to work.

The core of what I HEAR is this:

D#7(#9) and C#7(#9). The opening chord is where I consider it "rootless", because it reappears WITH a root in M3.

The problem is that b9 is there too.

Second problem: it is spelled completely incorrectly IF that is what Debussy had in mind.

So I can stick this chord and its voicing into my "toolbox", but I just have to pull the sound into my body and get the shape of it into my hands. I have no idea what to call it.

My big question: was he using a chord that we expect in jazz, in the 20th century, before anyone else thought of using it, notating it or naming it?

I don't know...

I looked up E. Robert Schmitz's take on this in the hope that it would clarify. But I'm afraid it (perhaps deliberately) obfuscates. I think one of Debussy's aims in the opening of this piece is to avoid tonality-related pulls and tensions. A big tool here and throughout is parallelism. The there and back again of the first three chords is subverted by the bass line so that there is no feel of tonic-something-tonic and also, in such a short space, by the F#9 being two beginnings of two quite different phraselets. The chromaticism at the end of the second phraselet is arbitrary in a tonal sense.

Another way of trying to understand this music is as counterpoint where the lines can be of a variety of thickness, or of complexity, or colour. It would also be interesting to compare Debussy's use of minor (flattened) 9th harmony here with Beethoven's use of the same chords which prevades his works and helps to give that special Beethoven sound.

Sorry if this is garbled.

John

Here is a snippet from the E. Robert Schmitz paragraph which I promise does not suffer from being taken out of context.

"In the three opening measures one finds that the use of D natural and A# consistently, obviates C# minor and F# minor or major as possible tonalities, despite the prominence of these tones as centers of attention. Are they centers of an, as yet, unnamed series? The closest tonality one can come to in the series of tones present is B minor (allowing for the exchange between G natural and G# as harmonic and melodic forms of the scale), yet nowhere is this tonal center in evidence -- a series of unresolved dominant-family chords?"

So far my impression is that other people don't quite know what is going on either, but they say so with a WHOLE bunch of words.

Thanks for weighing in. It's too late to change "2950" to "1950". Curse PWs editing system. I know why there is a time limit on changing a post, to keep people from playing games, but I wrote that before I went to sleep. I just got up and it is too late to change it.

My feeling: when you get to bar 3 and the bass note enters, with the 5th, it's pretty strong. I think we could take that same measure, isolated, use is as a V of V, slip to G#7, then resolve to C#m something. And it notates much better in flats. But that doesn't work immediately afterwards unless you flip to sharps.

Probably the problem is trying to somehow get from Common Practice thinking to TOTALLY out the box thinking. But the beginning still reminds me of the kind of thing someone like Bill Evens might have done - not the chord progression, but the chord itself.

Thanks for weighing in. I think maybe I posted this in the wrong forum.

I knew what you meant. I am accustomed to your typos by now.

I know. I am always in a hurry to get my thoughts out. If I do not take time to proof, forcing myself to read what I've written out loud, the result is horrendous. Thank God I don't play PIANO that way!!!

You could always repost it on non-classical and see what they have to say.

I have a question: does every chord have to have a name? Suppose there was a chord, say C D F# G# Bb E. (Voice it however you like: I just made this up.). Would that have to have a name, or would it just be a sound without a name other than just listing the notes?

Is there some boundary or dividing line in music that you hear, first in music that mostly just uses up to seventh chords mostly (including half and whole diminished seventh) like say Bach, and then music that uses these extended jazz chords like Debussy here, and then music that uses nameless chords, like my example chord? This is assimg that there are composers that use nameless chords.

Short answer (I wish I had time for the long answer ) - no. A name can be a useful label, but it can be misleading the further away from common practice harmony you go. Far better to describe characteristics, and describe how it's used. I personally don't think there's much point in describing GBDF as a dominant 7th if it occurs out of a tonal context, except as a shorthand version of "GBDF". But even so, it isn't shorter, is it.