|
ANTHONY GILBERT, unclassified composer
J. McL. Glover
Bells and birds, poetry, painting, puns and political crisis, stained glass and secret messages, the art and music of the orient - these are the kinds of things that set Gilbert’s whacky musical imagination in motion. It is forty years now since he had his first public performance, and over that time the music he has produced has remained resolutely modernistic, but also startlingly varied. To be sure, the earliest works - those written in London in the 1960s - were largely abstract in character, having no particular extra-musical associations. They observed certain principles of advanced serialism, leaned strongly towards the European avant-garde and in consequence tended to be performed at the international festivals. Their titles reflect their character: Regions, Sinfonia, Spell Respell, Treatment of Silence; there were, too, a Symphony and two Piano Sonatas. One might be forgiven for imagining they embody stern, fierce, unyielding music, and of course they do, to a greater or lesser extent. Spell Respell for electric basset clarinet and piano is an extreme example, taxing players and listeners alike with its rigorous structural logic, its engagement with maximal contrasts. You might say these works show the composer enjoying his technical skills to the utmost. But they also contain a breadth of expression which allows room for stillness, humour, colour, even sensuousness. The middle movement of the Sinfonia, for instance, has a dark, throbbing intensity which betrays the composer’s admiration for the music of Tippett, and the finale of the Symphony, a powerful elegy to Judy Garland, comes close at certain points to the blues. In among this series of works, framed by them as it were, are anomalies - striking departures from the tightly-controlled forms and structures of those around them. In 1967 two such appeared: Brighton Piece, for percussionists and ensemble and Nine or Ten Osannas, a chamber work in, in fact, 14 movements. Both contain hidden messages. The movement titles of Brighton Piece: ‘Introit’, ‘Gradual’ and two Chorales suggest liturgical connections (perhaps as interludes in a performance of Gilbert’s Missa Brevis); the music is wild, improvisatory and brilliantly anarchic. The titles of Nine or Ten Osannas are even more enigmatic: ‘Osanna for C’, ‘for the colours about some people’, ‘for one wild grey ghost’; what can these mean? The composer assures me that they do represent the emotional states, events, objects and of course people that the music celebrates, but that the listener does not need to know who or what they are. The music encompasses extremes of stillness, violence, colour, pace, brevity (one movement is about 2.5 seconds long), while the longest takes up about a third of the work’s time.
The move away from the supportive environment of London’s contemporary music scene in 1970 seemed to loosen Gilbert’s already fairly elastic ties to the European avant-garde. For the first time, too, he began to consider words in relation to music. Two commissions made it possible to explore the relationship between poetry and music: Love Poems, written for Jane Manning, show how metaphor in a text can be matched and elaborated musically, so that the music itself becomes a metaphor, even a vehicle for the poet’s half-hidden feelings. All the metaphors are of trees and birds, an early indication of much that was to follow in the succeeding decades. The one-act opera The Scene-Machine, with its verse libretto by George MacBeth, is likewise a metaphor for life, and for what can happen when one sacrifices principles for success. One London critic missed the point entirely, and was unimpressed by the deliberate surface banality of the story of protest-singer turned pop-star, and the inclusion of a rock-band. Words reappear in other works of this period, either as text as in the wonderfully dreamy Inscapes, embodying Gerard Manley Hopkins’ private meditations on natural phenomena, or increasingly as an inspiration for purely instrumental music, such as Crow-Cry, a chamber orchestra piece reflecting upon one of Ted Hughes’ sombre, troubled Crow poems. Alongside this growing involvement in poetry there came a more evident application of the techniques, though not yet the idioms, of Indian classical music. Modes, row-like melodic shapes, cyclic and cellular rhythms, games with time, all these things began to dominate Gilbert’s technical vocabulary. Put together with poetry they produced The Chakravaka-Bird, a wholly unique work, a 77-minute meditation for radio based upon the actual words of its protagonist, the 13th-century Carnatic poet and mystic Akka Mahadevi written during her life-long quest for union with the god Shiva. The music, in 3 closely-related cycles, alternates dialectic and inner contemplation, becoming increasingly dream-like as the poet approaches her moment of apotheosis, and reaching a level of spiritual intensity found in few other living composers.
In 1978 Australia called, and the possibility of living in a wholly different cultural environment presented Gilbert with the means of realising an abiding, deeply-felt need to divorce himself entirely from European compositional concerns and explore fully the things that music of the East had to offer. The results were interesting, to say the least. First came Towards Asávari for piano and orchestra, a plangently beautiful raga-inspired work paralleling The Chakravaka-Bird in its exploration of longing and fulfilment, but this time in the terrestrial domain, with suggestions of birdsong, the sound of wind, trees and water and above all physical movement; then Long White Moonlight, settings of Eastern poetry for Jane Manning’s and Barry Guy’s virtuoso soprano and double-bass combination. This cycle took a backward look at Gilbert’s expressionist idiom of 15 years earlier and blended it effortlessly with idiomatic and technical ideas from Korean, Indonesian, Thai and Indian classical music. Further than this he could not go without perhaps engaging in cultural theft. At all events, the influence of Eastern music, which had helped define Gilbert’s technical language for the whole of that 15 years, seemed to have peaked with Long White Moonlight and thereafter the works, superficially at least, begin gently to turn in other directions. The sounds of indigenous Australian tradition provided a new inflection to the existing language. Moonfaring is perhaps the first major work in which this can be heard - ritualistic, pulsing, dream-like music following its own powerfully mysterious laws. A whole series of shorter works in similar vein followed among which String Quartet no.2 predominates. The second half of that work remains one of my own most haunting musical experiences. But something else was needed, as Gilbert readily admits, before he could complete his explorations of this area. Aurally and visually the Australian experience had been intense, but where was the poetry to parallel all this? Gilbert devoted the first half of 1988 to seeking it out. The British-born Australian Sarah Day’s connected, with its celebration of the light and sound and rhythms of that world. A door opened and works flowed, perhaps not obviously ‘Australian’ but nevertheless inspired by the poet’s and his own perceptions. Dream Carousels for wind, a reflection upon Day’s poem Cycles, the orchestral song-cycle Certain Lights Reflecting, Tree of Singing Names, ...into the Gyre of a Madder Dance, again for wind, all these are full of the vibrancy and rhythm and music of the Day poems. His latest setting, Handles to the Invisible, has yet to be heard.
Now, after 16 years spiritually exploring the other side of the world, Gilbert’s imagination has again returned to Europe. At 60, in 1994, he began what was to be the major work so far of this latter period. A violin concerto bearing the title On beholding a Rainbow, it explores for virtually the first time in Gilbert’s output the implications of classical form. As in many of the works that precede it, there a hidden message, though this message, perhaps also for the first time since the comparative naivety of The Scene-Machine, appears to concern life as it is now - in this case social and political oppression. The work, even within the canon of Gilbert’s often highly expressive output, is of an unprecedented intensity, written in an idiom which brings together for the first time virtually the whole gamut of Gilbert’s rich linguistic vocabulary without in any sense losing focus. Modality relates illuminatingly to serialism, rhythmic modes relate to rhythmic cycles, harmonic movement to sustained pedals, formal control to a near-improvisatory freedom, dialectic to obsessive repetition, all nearly but not quite breaking the bounds of the work’s clearly-defined structural premisses. Social and political oppression, and also war, lurk between the bars of several succeeding works. Another Dream Carousel springs from Gilbert’s early, vicarious experiences of pre-war Vienna recounted to him by a young refugee; Unrise takes this further. Here there is a poetic connection: a posthumous fragment by Viennese refugee Avraham ben Yitzhak tells of a cock’s empty crowing for no dawn. Terror predominates. In String Quartet no.4 a complex of similar feelings, predominantly anger at political and religious hypocrisy, is half-hidden within music of a strangely ironic quality: highly expressive at times, but held in check with the musical equivalent of wry half-smiles. However, not all of Gilbert’s later music is dark and pessimistic. The Cathedrals of Chartres and Bayeux, or specifically the rose windows and labyrinth, provide him, in Rose luisante, Réflexions, rose nord, Sinfin and Worldwhorls with ideas that glow and dance; Os in its turn has much of Gilbert’s early crazy fantasy. Terror there is, but it’s the terror experienced in a morality play rather than at a 3 am banging on the street door, or a street kidnapping or suicide bombing...
| | |
GROOVIN’ not bloggin'
PUTTING IT OUT
It would be pointless to pretend that this is a genuine blog, as I understand the term. Composition, for me, is far too fluid, far too linear an occupation to be recorded in day-to-day or even week-to-week diary notes. Like a river it has blind creeks, floods bringing down loads of débris, dry seasons. More to the point, whereas a blog, or any log for that matter, records what has happened, my aim has been to record what I am striving to make happen. Anyway, even a month-by-month account forces one to record things which may prove irrelevant; all the same, that’s the approach I’ve opted for, as a compromise. It also means that strains on the memory are fairly limited.
SETTING IT UP
So this is an account of the composition of a substantial dance-concerto, a work intended as a gesture of gratitude and tribute to Stephen Plaistow, whose work as Producer and ultimately Head of Contemporary Music for Radio 3, and with other contemporary music organisations, did so much to raise the profile of real contemporary music (as opposed to pop) in British consciousness over more than three decades.
It first occurred to me to write the work while actually writing to Stephen in June 2005 to thank him for his beautiful production work on the BBC recording of Certain Lights Reflecting, which had been released by NMC earlier that year, so in that letter I asked him whether he would object to being its dedicatee. His reply momentarily scared me with its rather exaggerated indignation, until I realised he was simply and modestly protesting that composers should never need to show gratitude. Later, when I met him at a concert and he asked how his piece was progressing, I was further reassured.
My initial idea was to write a single-movement work of perhaps 10’ duration, full of dance and fire, to be called Ablaze. But as I worked on sketches in the succeeding months it became more and more evident that there was going to be rather more dance than fire, and, moreover, that possibly a whole library of earlier sketches, closely related to each other and dating from about 2001, had finally found their true home. The account of how readily these settled in, creating a scheme and character of their own, is in the next section, Getting to Work, written as the work emerged.
Obviously if one is going to undertake a major work the sensible thing is to get some organisation to commission it. That way at least one performance is guaranteed. So fairly early on in the process, when the work began to look like a possibility, I wrote to the Hallé in the person of Mark Elder, and to the BBC in the person of Roger Wright. I also wrote to Northern Ballet and Orchestra of Opera North suggesting a joint commission. I got a rejection from the former, who clearly misunderstood my perhaps unclear proposal, and a holding reply only from the latter. One positive and helpful reply came from the Ida Carroll Trust, who generously agreed to contribute a generous sum by way of commission fee, so the work is at least commissioned, though without performance. Phil Venables, Promotions Manager with University of York Music Press which, presumably, will take the work on, has also approached a range of performance outlets, equally unsuccessfully. On checking with Arts Council England, I find as a possible explanation that their policy for funding the creative arts seems virtually to exclude any possibility of straightforward commissioning of works for the concert-hall though mine could also be for the stage, as it turns out.
So I thought angry thoughts, and mentally more or less washed my hands of the English 'establishment'. It was going to get written anyway, and will not be economical to put on if it’s as difficult as it promises to be. We'll see.
GETTING TO WORK
Summer 2005. Semi-abstract musings.
So there was this mass of ‘raw material’ originally almost improvised, dating from 2000 and 2001: a sequence of 54 mirror-chords, originally used in Sinfin, the solo vibraphone piece reflecting on the south rose window of Chartres Cathedral, and a long, dancing permutation line applying a sort of Stedman’s principle to 12 (!) pitches in such a way as to create gradual, partial change of order, originally a response to the abstract, almost Islamic designs of the rose window of Bayeux cathedral. The permutations were in 92 stages, plus retrogrades and inversions: 368 stages in all, with dramatic differences between successive phases in their evolution differences between chant-like lines and plunging or rising gestures held together by a steady, consistent, motif-holding rate of change. The mirror-chords for the moment were simply that - a procession of sonorous harmonies varying between widely- and closely-spaced chords, from which I could possibly derive a harmonic structure an obvious, simplistic approach, as would have been simply to draw the linear material from the permutation sequence as it stood.
Besides being over-simplistic, there were a number of other reasons why this combination would not work, prima facie. Despite having been freely invented at around the same time in 2000-2001, there was no organic connection that I could perceive between the mirror-chord sequence and the prime form of the 12-pitch sequence or indeed any of its 367 derivatives. Both had a clear modality and, if transposed, could connect in that way, through their focal pitches. But the modal focus of the permutating line subtly shifted over groups of phrases, between C, E-flat, A-flat and G (I noted that inverting did not necessarily create an inversion of the focal pitch-relationship). This shifting would have put it out of kilter with the relative stability of the mirror-chord sequence. Transposition of a natural permutation process in any case undermines its essential clarity and integrity, so that could not be considered as a solution. As it stood, the set of chords themselves, though working quite well in sequence, one chord following the other, had few large-scale structural implications, and no rhythmic implications that I could detect. So for the moment the set was set aside.
Instead, long harmonic sequences were created, by careful superposition and by trial and error, from the permutation lines themselves, largely from the ‘chant’ segments thereof. Of these, one was found to work rather better than the others, having a strong linear force, interesting tension pattern and generous time-span. This was a superposition of 4 ‘chant’ lines from different segments of the prime, retrograde, inversion and retrograde-inversion of the 92-stage permutation line. Each of the superposed, often opposing lines consisted of 360 notes, and therefore in superposition created a sequence of 360 four-part chords, with lots of interesting near-recurrences, a structurally useful tension-pattern, meaningful colour-variety and quite dramatic expressivity, even when heard in equal values. This one, therefore, almost worked as a prime harmonic structure as it stood; by reordering certain segments I could make the motivic quality more structurally significant, and by reversing just the first 12 notes of the opening line in this new form I could create an initial growth-point, both harmonically and lineally. This I called my Prime Chord-Sequence. Now things could start to move.
September 2005. Semi-concrete experimentations.
Theoretically, the retrograde of this Prime Sequence, and the two inversions, could create three alternative forms. Whether they would or not would need to remain a supposition, whilst answers were found to 3 other basic questions relating to the Prime Sequence itself. Firstly, what were the rhythmic implications inherent in this long sequence of, for the moment, equal values? Secondly, did it matter that the rate of tension-change was, on the whole, gradual and subtle, supple even, and therefore far more suited to a fast than to a slow presentation? And thirdly, could the sequence survive the fact that its gestural implications, as it was built upon fairly level, chant-like lines, were not overly obvious? Fourthly, all this material was so restricted because it was drawn from less than a third of the whole 368-stage linear presentation. Was I being too restrictive?
Anyway, while I was mulling this over, I was in any case browsing through the original 92-stage permutation-line and its 3 derivatives, re-acquainting myself with those components of it which were strongly, even dramatically gestural. These were wild lines, plunging or rising through as much as 4 octaves if their component intervals were not doubled back on themselves for containment. And look, each of them consisted of 24 notes, the equivalent of 2 segments. They were crying out to be overlaid, as near-parallel lines. Overlaying 4 lines at a time would create problems of thickness and overall orchestral range; overlaying lines in pairs created too thin a sonority. Overlaying in threes, however, was crisp, harmonically incisive and, on the whole, instrumentally practicable. By trial and error I created four combinations in three parts each, two plunging and two rising, and each very different in character. For identification I called them: Gesture 1 (masculine); Gesture 2 (feminine); Gesture 3 (young feminine) and Gesture 4 (young masculine), for this is how I felt their characters. And each in its turn existed in four forms: prime, retrograde, retrograde inversion and inversion, making 16 gestures. Amazingly the character of each set was preserved in all its forms, and of course the inversion and retrograde of a plunge became rises, and vice-versa. So overall equality of plunge and rise gestures was possible (8 of each). Furthermore, 6 possible layerings of the three parts of each were possible some of which, it must be admitted, sounded less effective than the rest. There were some rhythmic implications in their tension-patterns, on the basis that the more interesting colours should be given the longer values, especially in the ‘feminine’ sequences, with a slight inclination towards consonance in the long values for ‘young masculine’, and towards dissonance in the same values for ‘masculine’. But this proved too random and disorganised without some kind of motivic unification, or, indeed, rhythmic modality. In the end I was able to rationalise the gestures into structures using just three values only: units of 1, 2 and 6 (semiquavers, as it turned out). I applied one rhythmic structure to each of the four categories of gesture, emphasising their already inherent characters. Now, in their upward or downward movements they sang out clearly, and especially, they danced. They were going, if I could see how, to provide the ideal foil to the 4-part chants.
So now, as well as a certain harmonic definition created by the superposition of, in one case 4 smooth, largely opposing parts making a gentle consonant/dissonant rhythm, and in the other case 3 near-parallel parts making a light-to-shade or shade-to-light progression, we now had the beginnings of a rhythmic definition. Since gestures, as opposed to extended lines, frequently depend for their characterisation on an element of strong rhythmic variety, in this case of 6 against 1 and 2, the remaining rhythmic structures should ideally have less contrast in their main workings, with perhaps only long values at climax points, cadences and maybe transitions.
The essential question now therefore was how to discover a rhythmic formation that would work on the broader scale of the 360-chord sequence and yet enable, or ideally create, cohesive interaction with the gestures. A tall order, but without it, the vision would remain blurred and unrealisable.
What was at least becoming clear was that if the rhythmic values in the extended 4-part chants (I should call them 'chorales') were to be constrained to simple units: values in the proportion of, say, 1, 2 and 3 as opposed to the 1, 2 and 6 of the plunge and rise gestures, this would create the effect of a near-stable pulse even a dancing pulse. Regular in that it would contain no irrational values, or durations outside its given canon except for structural emphasis, and yet alive in that, with everything being a simple multiple of the shortest value, the rhythm from phrase to phrase, from bar to bar, would be unpredictable in its simplicity. But I needed to find a logical rather than a purely random or intuitive way of creating these value-sequences, perhaps not too unrelated to what had revealed itself for the plunge and rise gestures.
If that could happen, the interaction between the strongly-characterised plunge and rise gestures on the one hand and the nearly-regular pulsing stream of the 360-chord sequences could create some sort of concertante effect an effect of conflict, of intrusion, even perhaps of disruption. If interaction was the right word, that is. For, for interaction, one needed a point of contact.
Then I made a very interesting discovery. Perhaps because they were formed from superpositions of related material, the 3-note harmonies of the gestures could also be found embedded in the 360 x 4-part prime-chord sequences. Indeed the initial harmonies of the gestures occurred with sufficient frequency, and regularity, as to suggest link or trigger-points for the gestures themselves. Those harmonies, where they occurred in the 360-chord sequences, would assume a colouristic, and even a structural significance by virtue of this connection, even where not used as trigger- or link-points, so long as the links were clearly established anyway. Their importance could be highlighted by accentuation, prolongation, or even by diminution. Now, a means of creating significant, non-mechanical, aurally-prompted rhythmic definition for the 360-chord sequences began to present itself.
All the above, little though it now seems, was the product of weeks of trial-and-error work with the long harmonic sequences. Although the overall structure of the work was still far from clear, its nature, that of a dance-concerto, began to assert itself in place of the original notion of a sort of fiery blaze. Rhythmically and texturally it wanted to be very simple but very fast, very possibly maintaining the same tempo from beginning to end. And the 360-chord sequence, as it stood, would be focal. Scoring, however, was far from clear.
All the same, the whole thing began to seem real enough for me to ask Phil Venables of UYMP to try and raise a commission for it something I’d imagined would be fairly straightforward, since in the past I’d never encountered serious problems. What happened or rather, didn’t over the succeeding weeks & months is outlined in the introductory paragraph.
October 2005. Messing around with building-blocks.
Much material was beginning to accumulate, in the form of sketches on paper, some of which had been transferred to computer. Nothing coherent, but it needed identifying and categorising. There were some very raw materials: the mirror-chord sequence and the long permutation-line. Derivatives of the latter fell into two main initial classes: the permutation combinations (5 x 360 4-part chord sequences) and the plunge & rise gestures (16 x 24 3-part chord-sequences plus all possible parts-inversions).
Two of the 4-part chord sequences stood out as immediately viable: the ‘Prime’ and its respaced mirror-inversion. 50 link-points for all 16 plunge or rise gestures were identified throughout these sequences far too many to be used as actual triggers, but an appropriate number for use as rhythmic/harmonic definers. The ‘Prime’ had a bright, energetic character, suitable for woodwind plus perhaps tuned percussion. Its respaced mirror-inversion was altogether darker, suggesting a predominance of longer values, and a string scoring. So at crotchet 96 the Prime was notated in semiquavers with occasional quavers and dotted quavers for link harmonies; its mirror-inversion seemed to need predominantly quavers, with occasional semis and dotted quavers for link harmonies, depending on their colour or tension-quality. All these sketches were at this point barred in simple 3/4.
Musically, it was fairly obvious that the mirror-inversion version should come first (dark progressing towards light), so for much of October I worked on turning this into music: 82 bars of string harmonies darkly dancing, inter-woven by plunge- and occasional rise-gestures springing from link harmonies, and all preceded by a slow emergence of the first harmonic sequence from a whole-tone trichord (C, E, F#). E, rather than C, began to reveal its significance as a focal pitch in this permutation-sequence.
However, imagining that further progress before receiving a commission might seriously hinder my chances of geting one, I now set that music aside and did a lot more structural thinking. Clearly an overall harmonic framework was necessary if the dances were to have any real solidity and structural strength. The luminosity of the mirror-chord sequence was irresistible, but as it stood it did not ‘belong’. There were no features in common with the permutation chord sequences that I could detect, and it lacked broad focus, to boot. As it stood it was in 4 parts. In an attempt to create focus I decided to try balancing each separate chord around a single central focal pitch C above middle C (shifting the whole sequence an octave lower would have lost it its luminosity). This was not entirely an arbitrary choice, since C was a focal pitch in much of the linear permutation work too. Anyway, it meant that quite a few of the mirror-chords had to be slightly shifted up or down so that they balanced around this C and not another pitch. Then, around this focal C, the chords were reordered so that they created a top line which followed the first 54 pitches of the permutation line. Then obviously, the bass line would follow the inversion thereof). This had much more clarity and ‘directedness’ than the original had (it was intended to work as a circling sequence in its original form), and produced a natural climax-point at the 29th or 30th chord - almost at the ‘golden section’. Adding this fifth, focal pitch deprived the original 4-part chords of some of their luminosity, but added strength and solidity.
November 2005. Mixing the cement, notionally.
Soon I made another useful discovery, which turned out later to be highly significant in terms of the overall structure of the work. These 54 mirror-chords now connected, in their new sequence, focusing on C, with 54 link-points in the mirror-inversion, 4-part chord sequence. A link-point, in this case, meant 3 pitches in common between a 4-part chord and a 5-part chord. They created an interesting harmonic rhythm over the 82 bars: a long (just over 10 bars) gap between the first 2 mirror-chords, several shorter ones, a long, several shorter, etc., concluding with a fairly rapid sequence of short gaps. These 82 bars lasted just over 2.5 minutes at crotchet 96. 2.5 x 8 = 20 minutes, the envisaged duration of the whole. Just over 10 bars x 8 = ca. 82 bars. Why not try applying this harmonic rhythm, enlarged by a factor of 8, to the overall structure of the work? Would this be too prescriptive, too likely to freeze the music into a mould before it was even written?
In some trepidation I made a chart of a whole such work in theoretical 3/4 bars (the ‘bars’ hereafter referred to, at a notional crotchet = 96), spaced out the occurrences of the initial mirror-chords over 8 times the number of beats in their original spacings but preserving their exact time-proportions, and studied it. I discovered that several distances between chord occurrences measured 82 bars, suggesting 3 alternative sets of placings for further 82-bar dance-blocks. The most interesting, i.e. structurally helpful set of placings seemed to be the regular one, with placings occurring at approximately 164-bar intervals. These, leaving approximate 82-bar spaces between them, suggested overall 4 main dance-blocks and a coda. Even more interestingly, the climax-point of the 54 link-chords, so expanded, corresponded exactly with the notional climax-point of the transplanted 82-bar sequence in its third placing, i.e. at notional bar 372, at the golden-section point between mirror-chords 29 and 30, and at the overall golden-section point. (These bar-numberings are for convenience in assessing time-spans only. The music might be barred overall in 3/4, and then again it might not, if performance practicalities dictate otherwise.)
However, there was something about the newly-fashioned mirror-chord sequence that didn’t satisfy, didn’t meet the perceived needs for adequate structural support. It had a poise and elegance of line, perceptible only on a smaller (say 82-bar) scale. For structural support over a notional 20 minutes the pedal tone needed to be lower or (in my dreams) higher. Simple transposition of the whole set of mirror harmonies to achieve this end would not work; their character as chord-colours would be completely lost. Higher, it lacked what I called harmonic ‘authority’; lower, it lacked necessary clarity. Two alternative derivatives presented themselves: the whole sequence of 54 chords with each chord transposed so that its bass note was E, the bass note of the first harmony in the sequence; and the same process in the opposite direction, so that the top note of each chord was A flat (the top note of chord 1, in the original sequence, clearly). These I called ‘bass-ped.’ and ‘top-ped.’ sequences, respectively. By the sound of things, the bass-ped. sequence had the necessary rich solidity to carry through over ca. 20 minutes; the top-ped sequence, as it stood, had a certain light intensity which befitted a much smaller-scale presentation.
December 2005. A momentary mortar-moratorium.
But again I had to interrupt work, to concentrate on some easy piano pieces while I awaited news of a commission for this Dance-concerto as it now began to define itself more clearly in my imagination. Some 45 years before, I’d copied for Schotts some of the parts for Priaulx Rainier’s Dance-concerto, Phala-phala, a work which I’d found, then and later, fresh and strikingly original (I can’t resist recording my Music-history tutor’s comment when I told him I was doing this job: “Oh, they’re such a lot of Rowdies, those people”, meaning, I guess, South-African composers). I realised that Priaulx had died in 1986, nearly 20 years before. So, as with Igórochki and Stravinsky, there was a coincidental 20-year connection. Rainure = groove. Had we a title here somewhere? Maybe two?
January 2006. Surveying architectural detail.
Still no signs of the commission, but I was approached by Nick Baragwanath at the RNCM, who advised me that I could in these circumstances apply to College for a Research grant, which would at least cover my costs, in part if not in total, for producing the work if no commission was forthcoming, on condition that I also applied for funds elsewhere. I gave this serious thought. But by now, despite lack of performance hopes, the need to explore the implications of this structural plan, of the new mirror-chord sequences and the potential variants of the 4-part 360-chord sequence became irresistible. I also needed to work out a practical way of notating this fast dance-music, with what promised to be continuously irregular successions of semiquaver, quaver and dotted-quaver values. On conductor Clark Rundell’s advice, I rebarred all the block-sketches of the mirror-inversion sequence (i.e. the opening dance), doubling the values and the tempo, so that the bar-structure was as irregular as the rhythmic groupings themselves. This avoided ties across barlines, which proliferated in the 3/4 version, and looked nice and clear. But were the bars too short to be beaten clearly at crotchet 192? We should see.
I called this Rondo-Dance 1; its 3 variants, when created, to appear about 5 minutes, 10 minutes and 15 minutes into the scheme, would therefore be Rondo-Dances 2, 3 and 4. The prime version, still in semiquavers in 3/4 at crotchet 96 because there were far fewer ties, would be simply called Prime-Dance. In energy terms, the Prime rated as high, the Rondo-Dances, as far I could see/hear, largely medium but hopefully pushing to very high at their climax-points. Some low-energy music was therefore needed for the episodes. Low-energy, but not slow. The underlying pulse needed to be constant throughout: a MM 384 effect, in fact: the ‘groove’. One way of creating quick-pulse, low-energy music is by freezing a line: a top-pedal or a bass-pedal, for instance using a mirror-chord derivative. The bass-pedal sequence gave broader structural strength; the top-pedal sequence, especially on muted horns or brass, had an interestingly bluesy, ‘still’ feel which would make it ideal but for one thing: it had little linear interest. However, on checking it for link harmonies to trigger plunge or rise gestures, these turned out to be ample, and the gestures themselves provided welcome spasms of light in an otherwise moderately shady texture.
The logical point for this first mirror-chord sequence to appear would be straight after the first Rondo-dance luminous stillness after dark energy, before the bright, brittle, brutal energy of the Primal-dance. The first Mirror-dance (which is what I called it) and the Prime-dance together would last just under 2.5 minutes, and could represent the first Episode. The rhythm of the top-pedal chords in the Mirror-dance could simply be that of their prime form in Rondo-dance 1, with their time-values halved to give the necessary 50% compression, and quantised for simplicity. The top-pedal will be the A flat above middle C.
Now, the first moment of truth. Would the bass-pedal mirror-chords, intended for broad structural definition, link with their harmonies at the hoped-for points in the broad rhythm-structure marked up in the structural plan? At the outset, we needed harmonic links for bass-pedal chords 2-3, close together. The link for chord 1 was a given, since it was the same as chord 1 in the middle-pedal sequence, in bar 1, though in the latter identity it was an octave lower than the rest - a useful pointer. Then links were needed for chord 4 after four bars in the notional 3/4 notation, for chords 5 and 6 at regular, wide intervals corresponding to bars 102-103 and 119 in the same notional notation, all in the first Mirror-dance. Amazingly, links for bass-ped. chords 2 to 6 existed in the mirror-dance, pretty well exactly where needed. The Prime-dance corespondingly fell neatly between bass-ped. 6 and 7 in their correct spacing, so needed no internal structural link-harmonies. So far, so good. But how would these occurrences be characterised? How scored? In Rondo-dance 1 the mid-pedal chords might be played on, say, harp and vibraphone, to ring out and ring on. It seemed logical to maintain that sort of sonority for the bass-ped. harmonies so long as their presence could be felt. Certainly if the pedal E was above middle C as it should have been if the C was in its original position in the mid-ped. sequence there would not be either the sonority or the staying power. But, as noted, the first appearance of the combined mid-ped. and bass-ped. chord was an octave lower than that a happy coincidence,
for with all the bass-ped chords shifted to this range they would achieve their purpose. The E pedal, now below middle C, would still need to be strengthened, perhaps by marimba and by tuned gong, whose chief function would be to draw attention to this note, and this moment, gamelan-fashion. The fact that the vibraphone featured strongly in the Prime-dance didn’t affect this approach, since there were no bass-ped. links in that dance. In fact, the vibraphone part therein emerged quite naturally out of the preceding mirror-chord, and could lapse gently into the next one. For the moment, these chords are to be struck once on each occurrence and allowed to ring. By the end of the work it’s possible that a more interesting mode of articulation might reveal itself.
February 2006. A little more concrete, but it needs more cement.
Suddenly now there were nearly 5 minutes of music, admittedly with rough edges, approximate scoring and a still rather uncertain opening. A lively but simple and logical overall structure had revealed itself and source-material for all subsequent sections seemed to be there, ready and waiting to be realised. All too easy, with a big risk of being mechanical and predictable. The question of continuity immediately presented itself. These were all separate sections, with no logical reason for one to succeed another beyond the simple structural rhythm already noted, rondo-like in form. There needed at least to be (a) a simple rite of passage from one dance to the next, ideally different each time but with related underlying logic; (b) a perceptible logic in the overall succession and (c) a sense of overall organic movement, with some growth to the golden-section climax, but not such that it overwhelmed the essential dance-like spirit of the work, or its unrelenting pulse-based tempo - its ‘grooving’.
An element of (a) could be provided by the fact that a bass-ped. harmony occurrence signalled the starting-point of each new phase. A transition harmony, or gesture, relating the end of one dance to the beginning of the next, would be based on the relevant bass-ped. and there seems to be enough difference between the character of successive bass-peds to provide variety. It is, after all, their function to create a sense of direction with change over a constant element. If gesture be used, it could be a timely ‘plunge’ or ‘rise’, triggered by a relevant link-harmony, or it could be a line deriving from the harmony of one dance or the other.
The question (b), sensible, logical succession, needs time and actuality of invention to resolve. In other words, I need to know much more about the musical nature of each dance before seeing/hearing how they could fit into/balance each other. Some ‘dreaming’ is necessary at an early stage in the realisation of, particularly, the episodes. The Rondo-dances should be related, but not too similar, and the whole conceived in such a way as to satisfy requirement (c). Dream hard!
February-March 2006. Building-work.
So it’s time to get down to solid invention. ‘Dreaming’ continues, but the result of that is that I can wait no longer for someone to commission the work. It demands to be set down, at least in rough. I am making computer files of each of the components as they crystallise: plunge & rise gestures, permutation-harmony chains (which for convenience I call combiperms), mirror-harmony chains, and now possible harmonies for Rondo-dance 2, needing refinement and stronger linearising. These are from pages 25-27 of the Combiperms 5 file, in which various orderings, inversions, retrogradings and even transpositions (despite my rule never to transpose permutations) have been experimented with. They need to have a tonal quality focusing on, almost certainly, C. Their closest connection is via the retrograde of the harmonic sequence for Rondo-dance 1, itself an inversion of the prime sequence.
But chords in Rondo-dance 2 we do not need at least not initially. The connecting link with the essentially homophonic Prime-dance must be linear. It can behave like a plunge gesture or a rising one, cutting through the chunky homophony of the Prime-dance from a link pitch, and only gathering its own homophony line by line gradually throughout the length of its progress, without too much sense of growth. This is a transparent Rondo-dance. Initially, its line can be drawn from notes in all 4 parts of the raw harmonic sequence so long as a chord-sequence-related pitch sequence is strictly maintained. Only later will the remaining 3 parts gradually accumulate around the line.
In the overall scheme, this Rondo-dance needs to be succeeded by the ‘still centre’ the longest episode, and the 5th section out of the notional 8.5 main sections of the work (i.e. the coda will only be half a section in length). Theoretically there are two choices of ways into this episode: either bang it straight onto the end of Rondo-dance 2, creating a dramatic shock, or make a long, gentle transition. There’s time in the structure to do the latter, and that would also answer the need for a ‘rite of passage’ perceived early in February as option (a). But what will the still centre itself consist of?? It needs to be brought into existence before I can know how it should be approached.
April 2006. Recycling some builder's waste.
Early in the conceptual process, whilst still seeking some sort of clear, strong harmonic structure for the whole, I created a succession of 5-part ‘background harmonies’ by overlaying the ‘flattest’ passages in the permutation lines to create a condition of near-stasis, with just gentle harmonic flexings. They were too bland overall to serve their originally-conceived purpose of a large-scale structural stability and strength, but perhaps they would work here. But they don’t dance, and they don’t sing. They need vitalising in some way. Predominantly nearly consonant, there are some pure consonances that can be prolonged, and a few acute dissonances that should be shortened to avoid too much unsettlement of the section’s overall stability. And to sing, they need a line, which can hopefully be drawn from the notes of the chords - one or two notes from each 5-part chord, in their correct associations.
Fine, in theory, but the reality nearly caused despair because the initial results were completely characterless. Then in late February, after 10 days of concentrated trial and error, suddenly, after some useful ideas about how to articulate the harmonies themselves, there had emerged a shmoozy waltz-tune for oboes, with a counter-melody for horns. I had set them aside, but now, in April, it’s time to give them proper consideration. To give equal prominence to both, I’ve decided that each segment of the oboe tune was to be given twice, the first time without the horns’ counter-melody, the second with the counter-melody predominating. The main focal pitch is F sharp. Since C was the quasi-tonic of Rondo-dance 2, there’s a clear indication that a link-passage is needed. C and F sharp are two of the four pitch-classes of the main focal harmony of the work the pitches of the opening chord, in fact; E is the third, and A flat/G sharp the fourth. The opening few permutations of the source permutation-line focus on E we could try linking with these. A flat is an important link-pitch too, and appears almost as a quasi-dominant throughout much of the permutation sequence. And as a matter of fact, the first 22 permutations of this running permutation-line in its prime form fit exactly with the ‘background’ harmonies in their defined rhythmic structure, this structure working actually as a background, on, say, strings, whilst the permutation-line is played in quavers, quietly, on, say, horns, detached and possibly muted. Elegant!, some would say...
But this now forces me to confront an important structural issue once again. Fitting exactly is what the bass-pedal harmonic structure is supposed to do with each new phase, each new dance. This is an imperative even before it’s determined how the bass-ped. harmonies are to be articulated, orchestrally.
Miraculously, the slowly-expanding durations of the predetermined bass-ped. structure supporting Rondo-dance 2, i.e. chords 7 to 13 with chord 14 as the link-harmony to the linking-passage to the still-dance (yet to be written), find links at the right time-points in the harmonies of the Rondo-dance a link meaning harmonies with 3 pitches in common. The long linking-passage also finds its connections with the succeeding 4 bass-ped. harmonies (chords 16 and 17 are pretty much identical in their pitch-content as well as being almost simultaneous); chord 18 is the bridge-chord into the main still-dance, which for the moment I’ll call ‘Chillchamber’. Amazingly, bass-ped. chords 19 to 23 are all there, if not quite at the right spots, in the chillchamber harmonies too. This is one of those chains of coincidences that over the years I’ve almost come to expect in the composition process once a lot of careful thinking has taken place. The unconscious mind seems to be able to do ‘calculations’ of a far, far more elaborate nature than one could ever do consciously in such a relatively effortless manner.
Chillchamber (alias Cool-chant) continued.
That’s all the link-passage needed to be, though cutting itself off at the moment the smoochy waltz started rather defeated its initial purpose all the same. But first, let me write the chill-chamber dance. It needs to run smoothly, have a regular crotchet 192 pulse, rather than quaver 384, not overly-goal-oriented but at least with a sense of arrival towards its end. The oboe and horn lines could be slightly refashioned to fulfil this structural purpose; a measure of structural unpredictability to counterbalance the pulsed regularity could also be achieved by having the repeats at irregular points. Fine. And this way the bass-ped. harmonies land time-wise pretty well spot-on their target links. But: it’s all a bit airy-fairy, a bit vapid. And does music of this rather dreamy, near-tonal waltzy nature really belong? The honest answer had to be: No.
But there also remained the unresolved problem of the cut-off running quavers of the linking passage. Now the linking passage, at 44 bars, was a little under one third of the length of Chillchamber. Its little permutation dance in quavers used 22 permutations out of a total of 92 in the prime line almost the same ratio for both cases, therefore. In fact, if I continued the perm-line to its completion as a little spine through the Chillchamber dance, it would run out just immediately after the melody-line reached its arrival-point, its gentle climax, and consequently provide a link-line, comparable to the linking-passage before Chillchamber, and now leading into the gentle coda to the Chillchamber dance. Perfect, because not only would it solve the interruption problem at the onset of Chillchamber, but would add a little sub-plot, remove the blandness, enrich the harmony, keep the music moving and even add a touch of what I perceived as humour. If lightly scored using a variety of sonorities including tuned percussion, muted trumpets and flutes, say, it would add colour. Overall, it might just solve the problem of ‘belonging’. The only other way, indeed, that I could conceive of solving that was to have each successive dance now in a further pastiche mode, now so fashionable. In death, indeed. Oh gosh, that way I could earn enough money to buy a 4x4, and help us all towards Armageddon.
May-June 2006. Driving round the back.
But now that the drive of the first two Rondo-dances and the Prime-dance has been suspended in favour of this ‘cool’ music, how is it to be recovered in time to create a convincing and fairly powerful climax in Rondo-dance 3? My preferred solution is to eliminate the ‘coolness’ first by turning it completely to ice, than allowing a single chant-line, drawn as before from all 4 parts of the presented harmonies (in this case simply the Prime-dance harmonies retrograded and parts-inverted) for the first phase of the Rondo-dance, then to build up, in the manner of a round, quasi-canonically, relentlessly, inexorably, unstoppably, phase-by-phase from bottom E or middle E, each time followed by D sharp and F sharp. The notional tonic, final or sonant of the Cool-chant thus prevails through these four entry phases, is emphasised by the bass-pedal harmonies’ pedal E, especially just before and after the climax, and is confirmed at the climax itself by the top E pedal on strings and woodwind, a happy coincidence producing wonderful vertical symmetry at the very point, the overall and local golden-sections, when all the other forces coincide. As far as its powerful E-focus is concerned, Rondo-dance 3 is less a rebirth after the Chillchamber/Coolchant’s relaxed lyricism, than a reawakening of its sustaining force as a dominant area to the gentle, ambiguous A-minor final of the 2nd Rondo-dance.
This enables the G sharp + E relationship of the transition into the Cool-chant to reassert itself: G sharp now as the A-flat pedal of the top-pedal harmonies for Mirror-dance 2, and E as the pedal of the bass-pedal structural harmonies.
July 2006. Additional mirrors, also naming of the parts.
The aim was to make the second Mirror-dance a distorted reflection of the first. The rhythms needed to go forwards, because they belonged to that forward-moving set which underlay the whole. The harmonies,also in fact, refused to work retrograded, so other things needed to be done to create the distorted mirror-effect. One was to vary the scoring considerably, another was to allow the impact of the bass-pedal harmonies to disrupt things going on above and around them, such as forcing repeats of the final mirror-dance phrases and stretching or compressing other essential values somewhat, and a third and most dramatic, in the true concerto spirit, was to allow far more plunge & rise gestures to be triggered, and force space for themselves. After 3 or 4 tries it arrived at the wished-for rather wobbly condition; the repeated concluding phrases with their wild up-and-down motion in the lower two parts helped this enormously. A nice discovery was that if the final plunge, to be used as a linking-passage to the final Rondo-dance, was reduced to 2 parts, it rather elegantly defined a IV-V imperfect cadence in C major, or a modal cadence in G. This, in any case, linked beautifully with the opening harmony of Prime-dance 4.
By this stage in a work I was usually very clear about what it should be called, overall and in its component parts. In this case it was not that I had no thoughts, but that there were too many possibilities. The overall title wavered between Grooving the Chants, Chants to Groove, Chants to Dance or just Groovin’. The separate dances, however, named themselves on a long car-journey, as Entrance, First Rowdy-dance (remembering my Music-history Tutor’s comment), First Mirror-dance, Prime-dance, Second Mirror-dance, Into the chillchamber (the linking passage), Cool-chants, Third Rowdy-dance, Second Mirror-dance, Fourth Rowdy-dance and Outrance (pronounced the French way). Seems to be no way of escaping my bit of Huguenot ancestry, or Priaulx's.
August 2006. Some off-road driving
The fourth Rowdy-dance needed to seem to take the feeling of gentle disintegration a step further, yet maintain very close contact with key features of what had gone before. It was time to bring the Prime-dance back in a new guise. With the first 12 pitches in the 2 lower parts retrograded, and the whole thing transposed down half a tone, it joined perfectly onto the tailpiece of the 2nd Mirror-dance. Rhythms needed another connection, however, so instead of using those of either the Prime-dance or any of the previous 3 Rondo-dances, I created a structure of exactly the right length out of the plunge & rise rhythmic sets, arranged in an order that supported the inherent harmonic tension-pattern of the chord-sequence. And as for scoring, that needed to partake of the concertante energy created in those earlier dances by the interaction between the long, sustained 4-part dancing-lines and the subversive plunge and rise gestures, but this time internally. In other words, the process of gradual loosening-up in Mirror-dance 2 now applied itself to the scoring itself, so that it now became antiphonal rather than a continuum. And the little germinal 2-1-3 cell was enabled by these means to separate itself out, become emphatic and even disruptive. ‘Groove, perchance’, it said, dotted irregularly all over, like hecklers. It also at last provided me with the means of gesturally-articulating the bass-pedal harmonies on harp, vibraphone, marimba and gong. On harp, particularly, so I traced back and allowed the cell to attach itself to bass-ped. appearances where it could make a useful contribution to the rhythmic life of the dance in question. Later, more frequently than earlier in the piece.
Maybe now, in the 4th Rowdy-dance or even before, we’d arrived at the brilliant, even blazing texture I’d envisaged for the work at the very outset, 14 months before. The bass-pedal harmonies found their links, or if they didn’t I shifted them a little one way or another without disrupting their essental broad rhythmic structure. But that Mirror-dance tailpiece haunted me. It wanted to come back and join in the dance here, too. Not unless you can find a function and link for yourself, I said. So it linked itself to the closing bass-ped. harmony of Rondo-dance 4, and did exactly what it had done for Mirror-dance 2, only this time it linked to the coda. But having caught the disintegration disease, it broke off halfway through, with the arrival of essential permutation-combinations that sounded very much like an echo of the Prime-dance. Ah, I see, it wanted to play its own little game. Sure enough, it entered into a kind of plungey dialogue with the steady semiquavers, constantly interrupting itself and passing from one pair of wind instruments to another. While all this was going on, the expected acceleration in bass-pedal chords was taking place, which far from disrupting, was forcing everything to hold itself together despite the little concerto-game. And so we entered the Outrance, in course of which the exasperated bass-pedal chords finally managed to take control, in a sort of rôle-reversal from that perceived for them in February 2006 but not before the little tailpiece had had its final wag, its full statement, which in the end I felt should be on clarinets.
September-December 2006. Colour-schemes.
Because, you see, at that stage I was by no means certain about the scoring, or indeed the barring. I printed out a score as it stood and showed it to kindly Clark Rundell, who, master of rhythm that he is, confirmed my suspicions about the short bars in crotchet=192, and advised me to rebar all the dances except the Cool-chants and the lead-in thereto, in half-values and half-tempo, and double the lengths of all bars. But before I did that, I wanted to be absolutely sure of the scoring, which finally ended up rather more economical that I’d originally conceived. On 25th December I finally let it all be.
So that just left the main title. Back and forth over all the possible choices I went, even while I was extracting the parts. Then we had a storm, and while clearing up the débris on the front lawn it suddenly came to me. The heckling rhythm. ‘Groove, perchance’. Or better still, to reflect the fact that all the lines came from that chant-like succession of 368 linear permutations, why not spell it Groove, Perchants. Perchant, in French, is a colloquial word for very difficult, virtuosic, on the edge....
which it certainly is.
| |
