Ligeti: Ten Pieces For Wind Quintet 'Analysis'.

Following up on my previous Berio 'analysis' here is something from Ligeti (Ten Pieces For Wind Quintet 1968), this time looking at rhythm, voice leading and gesture as opposed to harmony as before.

Firstly I think this is a great piece, my 'favourite' wind quintet, as with most Ligeti I have heard (nearly all of it, I think) the music is paced just right, the overall duration of the work and its various movements are all digestible (to my ears) and even though it's a fairly 'complex' affair (at times) on first listen you walk away with an idea of the music still ringing in your ears, so to speak.

Also I think the scoring is quite practical, the ideas, however complicated, are expressed with clarity (I am not suggesting these qualities are positive musical 'universals', just my own preferences/predjudices, take them or leave them).

The element I want to look at today (listen up at the back please) is the way the music accelerates and decelerates in a precise way by means of various tuplets/groups as opposed to the more commonplace accelerando or rallentando/ritardando technique.

This acceleration and deceleration is some extent an illusion, it might be better to describe the effect as one of increasing and decreasing density. Why?, because the tempo or pulse is constant, it's the number of attacks per beat that changes. A good analogy, perhaps, are gears.

Think of a bicycle (go on, in your favourite colour), imagine that each revolution of the pedals is one beat, this stays the same, what changes is the gear on the rear wheel. For each turn of the pedals the number of revolutions of the rear wheel differs, increasing or decreasing in number (3 per turn/beat, then 4 then 5, then 7 or going in the other direction to slow down or reduce the density of notes).

Starting with the eighth bar/measure you can see/hear the flute and clarinet moving into some tuplet patterns (after some fairly slow moving polyphony), first a quintuplet and then a sextuplet (for the clarinet) then both instruments play a septuplet in unison before playing a series of straight demisemiquavers/32nd notes. This results in a 'winding up' effect, or a change of gear.

Then the clarinet holds a steady trill pattern while the remaining instruments (except the horn) play a (rather spiffing) melody over the top (in bars/measures 10-13 ish). The flute joins the trill fest with the clarinet and finally, they both decelerate or 'wind down' or 'change gear' using the same technique, first straight hemidemisemiquavers/64th notes then into a diminishing pattern, septuplet, sextuplet, quintuplet and a straight four in hemidemisemiquavers/64th notes, then down into various groups of semiquavers/16ths, then quavers/8ths then a semibreve/whole note with a fermata to end.

This technique is broadly speaking 'micropolyphonic' (one of Ligeti's inventions more or less).

Why is this worth discussing you might ask?, good question..

A more standard accelerando or rall/rit sound wouldn't sound the same at all, it would be much smoother, and arguably, considerably less interesting. The effect Ligeti achieves is something like the sound of a slightly clunky mechanism winding or gearing up and down and he achieves this via fairly simple instructions (even I can read/intepret it so it can't be that complex now can it).

It's almost an illusory effect, the tempo or speed (in one sense) remaining the same while the density of attacks changes. Whether in some ontological sense this results in a speeding up or down is something of a semantic issue (an interesting result for a piece of music perhaps).

This seems a common feature of his music in this period (ive not looked at his later music properly yet, i.e. from late 70s onwards so I can't comment on it).

In terms of the voice leading and harmony, briefly, as you can see the motion between the clarinet and flute is contrary in the decelarating/winding down section (bars/measures 14-15) and mostly contrary in the previous accelerating gesture though the relationship between the two voices is more complicated or less ordered (from a cursory examination anyway). This more- or-less-the-same-but-slightly-different effect is another 'micropolyphonic' gesture/trademark.

The harmony (referring to the flute and clarinet) is generally speaking chromatic and voices move by step, it's clear in the decelarating section that both instruments are playing a repeated semitone gesture/trill in contrary motion on the same two pitches (Bb/10 and A/9) , with the clarinet moving up to B/11 at the end and the flute finishes on A/9 (making this final gesture symmetrical, a tone or major 2nd leaving an axis tone in the middle, a Bb/10, this axis is implied not sounded, perhaps a 'pensanto' if you like, a note 'imagined, not played' - Perle - or perhaps, it's just a major 2nd).



For more info (from a competent theorist) on the voice leading and pitch relationhips see 'Stepwise Continuity As A Structural Determinant in Gyorgy Ligeti's Ten Pieces For Wind Quintet' by Charles D. Morrison in Perspectives of New Music (which I haven't read for ages and probably should have done before posting this, I might read it over the next day or two in case it contradicts anything I have written here).

So there you have it, apologies for the brevity of this 'analysis' and if I have made any mistakes (i'll check it again over the next few days). I only analyse pieces like this for my own compositional purposes, not because I intend to be any sort of theorist or analyst so this is something of a cursory, idiosyncratic reading of the piece, better than nothing though eh?.

The score is available here and is published by Schott, the music fragment in the video is from this CD by the Albert Schweitzer Quintet (also featuring a rather good quintet by Kurtág amongst other Hungarian quintet music including one of Ligeti's and Kurtág's teachers, Sándor Veress).

(UPDATED: 08/02/2009 to include the 'gears' analogy)

(UPDATED 15/03/09, sorted out pitches in final harmonic 'analysis' bit)

The score fragments below are marked Lento (crotchet/quarter note = 40). Also, something that caught me out, it's a transposing/conductors score not a usual concert pitch study score (*grumbles.)

pgs 13 & 14 (click for larger images).