Hello!

Just a quick question. I was looking over Clair de Lune for the first time. First off, I had no idea it was in 9/8! I suppose the "Andante tres expressif" really has a way of disguising the time signature when the piece is performed. Still, when I looked over the music in the G-Clef (for the left-hand), I noticed that the first and ninth bars don't add up to 9 quavers!

Could somebody explain/correct this finding?

Many thanks,
Nick

CDL has a bunch of measures that combine triplets and duples. You play the duples (two notes) in the same time that you play the triplet with its three notes. At least, thats how I remember it when I gave CDL an aborted try last year.

edit: you should see a little 2 over the duples on the score, just as you see a little 3 over the triplets.

Yeah - best think of it in 3/4, then split the beat into 2 or 3 depending on how many quavers you count. Better yet, listen to a good recording and hundred times, and then don't count, just copy.

If you go listen to Hugh Sung's lessons on CDL in a thread by that name here on PW he has you play them to this kind of rhythm:

for a measure that goes tripet, triplet, duple you would count

"bum-ble-bee, bum-ble-bee, buzz-zing"


I drove my wife near insane with that cadence

JimF: There are no duples or triplets in the first and ninth bars. I can see duples and triplets in later bars though!

ten left thumbs: I would definitely prefer learning the piece in time and then making it more free, out-of-time myself. You're right, I could listen to somebody's recording of the piece; but I wouldn't really be getting a grasp of the 9/8 time for myself. I've found a good way to learn pieces in the past is to at first learn them strictly in time. As I become more comfortable, I make the piece more expressive, and this works well as I'm most confident.

Also, the issue isn't how to count 9/8 time, it's where to find the beat of the ninth quaver in the first or ninth bars in the left-hand.

I wonder if you're being led slightly astray by the notation. In the first measure, the first beat is a triplet rhythm consisting of a rest, followed by the left-hand quavers, then followed (in the triplet rhythm) by the right-hand quavers. But notice in the left hand that the opening notes are held for the remainder of the measure. Although it appears that this consists of a quaver tied to a dotted minim, the actual duration is a crotchet plus a dotted minim (thus filling out the count in the measure).

The notation is a bit ambiguous because the beaming of the opening triplet rhythm doesn't permit the left-hand notes to be shown as both quavers (in the beamed triplet rhythm) and crotchets (for the full time value). But the beaming is important to emphasize the triplet rhythm that is so fundamental to this piece, and the actual duration is obvious by the spacing of notes on the two staves in this measure.

By the same token, the ninth measure might have employed a few more rest symbols to fill out the two staves, but they are not really necessary since, by that point, you can see how the rhythm works and how it is distributed between the staves.

The practice of allowing some ambiguity in the strict notation where the performer can be expected to "fill in the blanks" is not uncommon.

7njap - I appreciate you want to understand how the rhythms work.



Can you work out where beats 1, 2 and 3 are? I'm not sure what you mean by the ninth quaver.

Let me try again with a picture. Here is the count for the first measure:



The notational issue is that the lower staff appears to consist of a quaver rest and a single quaver tied to a dotted minim. This appears to only count to 8 (1+1+6). But as the image suggests, the opening quaver in the LH is held for a crotchet plus a dotted minim. In order to explicitly show the first quaver as a crotchet, you would have to dispense with the beaming to the rest of the appregiated opening harmony. Instead, it is written like the other quaver triple rhythms, and you just assume the extra count because of the vertical alignment with the upper staff (and because it's the only thing that makes musical sense).

Packa - I understand what you are saying. The quaver is a crochet - I never noticed.

I'm interested in 7nap's question. I don't know what the 9th quaver is - the one between the eighth and the tenth? I'm confused.

Permit me to say - counting this as one two three four five six seven eight nine is going to make a player very very confused on several counts. There are not nine beats in the bar - there are three. The first thing is to find the three beats. Then, and only then, think about how they are subdivided.

Without being really clear in your mind about the three beats, this piece will confuse you dreadfully with the switches from triplets to duples. You will labour under the illusion that a quaver here is the same as a quaver there - it isn't.

I agree with you. I was not explaining how to feel or count the rhythm of the piece; I was trying to explain a small ambiguity in the written notation, which is what I thought the original question was.

9/8 here clearly means feeling 3 beats per measure where each beat is often expressed as a triple quaver or a quaver duplet.

One problem here is using the word "triplet". The strict definition ("3 in the space of 2" - see the Harvard Dictionary of Music) does not apply here. The quavers here are just quavers in 9/8, but they are grouped in threes to emphasize the triple meter of the piece. The duplets, however, that Debussy adds to the mix are really duplets ("2 in the space of 3"). This is why editions generally don't put the little "3" over the triple quavers, but do add the little "2" over the duplets. But whether we call them triplets or triple quavers or whatever, I think we're talking about the same thing.

Yes, packa I'm not that precise in how I use language!

Now I think we should let the OP respond before we say too much. Otherwise it can get terribly confusing.

Many thanks for this explanation, this answers my question, and many apologies if I wasn't clear enough with my question! I find it strange but also wonderful that composers would give their pieces such moments of ambiguity for the player, especially with an issue as strict as timing. This is the first time I've encountered this in a piece. Many thanks to packa and ten left thumbs!

This is an interesting example of how music notation functions as language. Music notation is not a precise formal system like a programming language. Rather, it functions more like a natural language: it doesn't explictly say everything; it often denotes and connotes as the same time; it can be ambiguous; sometimes it's downright confusing and illogical. But we're still able to see through the written symbols to reconstruct a coherent meaning.

I suspect that many folks who know exactly how the opening measure of CDL works musically have never become explicitly aware of the notational issue we have been discussing. Just as with their native tongues, fluent readers of music notation often immediately absorb the apparently intended meaning without dwelling consciously on the details of the written symbols.

Packa's hit it on the head!

I've been studying this piece quite closely lately as I'm helping someone learn it. What it seems to demonstrate it that the notation doesn't convey accurately how the music is to be played. The notation conveys what the composer/editor thinks you need to know in order to play it.

If they think something is obvious by the layout of the notes, then they don't clutter up the page with redundant waffle. In the second page they don't bother letting you know about the timing changes. They assume you will look at bar and say '9 quavers, divide beat into 3; 6 quavers, divide beat into 2.'

I suppose the principle has always been there, but this piece really brings it to the fore.