Classical Music / A Faustian compact with form: Before tonight's performance, Bayan Northcott reappraises the unique synthesis of Berlioz's La damnation de Faust

In 1856, working up his courage to tackle the last of his grand projects, Les Troyens, Hector Berlioz wrote to Liszt's mistress, the Princess Sayn-Wittgenstein: 'I am for music which you yourself call free. Yes, free and wild and sovereign; I want it to conquer everything, to assimilate everything.' But he felt compelled to add: 'What is going to be tremendously difficult is to find the musical form, for without form music simply does not exist, or is merely the humiliated slave of words.'

And there one has the nub of what subsequent generations of sceptics have held to be the Berlioz problem. Debussy - who actually admired at least some of the works - was merely giving way to a tradition of exasperation when he characterised Berlioz as 'an exception, a monster. He is not a musician at all. He creates an illusion of music by means borrowed from literature and painting. Besides,' he added, 'there is, as far as I can see, little that is French in him.' The last cut is the unkindest - and not only because Berlioz spent so much of his time and substance in trying to interest the Parisian public. What, after all, could be more specifically French than the Revolutionary ceremonial of the Grande symphonie funebre et triomphale, or the exquisitely refined Gautier settings of Les nuits d'ete?

Granted, his attempt to assemble a musical manner from such influences as Gluck, Beethoven, Spontini and Weber, and his pursuit of concepts inspired by writers as varied as Virgil, Shakespeare, Goethe and Byron, suggests from the start an almost unprecedented, pan-European aspiration 'to conquer everything, to assimilate everything'. And doubtless this in turn helps to explain why he was often more warmly received in Germany, England, even Russia, than at home. But what Debussy was really raising was the old question of whether, by gifts or training, Berlioz ever quite commanded the musical technique to sustain, integrate and resolve the massively mixed projects he undertook.

His musical origins were certainly the most exceptional of any great composer up to his time. There was no piano in his remote provincial childhood, no chance of hearing an orchestra; only a few theory lessons plus flute and guitar practice to feed his creative aspirations. Then, suddenly, Paris at 18 with its concerts and operas, and in an astonishingly few years the young man had not only acquired a mastery of orchestral effect but advanced to the very forefront of the entire Romantic movement. Yet some would question whether his basic compositional technique ever quite caught up. His gift for projecting long melodies can sometimes seem short-circuited by oddities and non sequiturs in his harmonic fillings-out. And while many of his most exciting passages consist of two or more streams of music separately stated and then superimposed - a device familiar in operatic ensemble but never before carried so far in orchestral composition - not all would agree that the effect is as integrally musical as it is dramatic.

And above all, there is the matter of form. So many of Berlioz's larger works are strange hybrids. The five staggeringly innovative movements of the Symphonie fantastique are certainly coherent within themselves as character pieces, but without the programme and the idee fixe link, we might well wonder how they belong together. Again, it is difficult to think of any other composer who could preface a symphonic allegro not just with one, but two slow introductions, as Berlioz does in Harold en Italie. As for the 'dramatic symphony' Romeo et Juliet, with its often inspired but disparate sequence of full-scale movements, songs, scenas and mimetic outbursts, it takes an exceptional performance to transcend a sense of falling between all the genres the work invokes.

Opera might have seemed the answer, but in the event the motley provenance and the volatile continuity of the Berlioz style prove too much even for the theatre, and actual stagings of Benvenuto Cellini never quite seem to add up to the sum of its wonderfully inventive musico-dramatic parts. Throughout his twenties and thirties, indeed, Berlioz could be seen as searching for the ideal form to realise his idiosyncratic gifts. In his early forties, at a time when the circumstances of his life were least conducive to sustained work, he found it, in effect, by casting off the impedimenta of both symphonic and operatic composition.

As so often in his creative life, the stimulus lay in his youth. In his early-twenties, excited by Gerard de Nerval's translation of Goethe's dramatic poem, he had composed and published Huit scenes de Faust - and then withdrawn them when the great man failed to respond. But pulling them out in his early-forties, at a time when his failing marriage and shortage of cash drove him on an almost endless round of European tours, he realised they could be elaborated piecemeal into an entire 'concert opera' or, as he finally subtitled it, 'Legende dramatique'. Nor were the eight scenes the only items chucked into the structure: surprised by the howling success of an arrangement he had made of the Rakoczy March, Berlioz calmly transferred his opening tableau to the plains of Hungary so that he could play it again. Other sections of La damnation de Faust were scribbled under streetlights and in hotel rooms half across the Continent. Such an assembly job ought to have been a disaster. In fact, it was salvation.

For a start, the huge resulting four-part collage allowed Berlioz to place in balance so many of the fundamental attitudes and conflicts of Romanticism: the solitary versus the collective, the sacred versus the profane, human love versus the love of nature. Again, the ad hoc continuity enabled him to lock the most diverse moods and genres of which he was master into tight sequences - march, song, scena, chorus and ballet, cross-cut and superimposed in a sweep of expression from the most affectingly tender to the most terrifyingly grotesque - and scored with all his inimitable orchestral imagination from the minute delicacy of the Danse des sylphes to the surging power of Faust's Invocation a la nature.

But above all, the form liberated Berlioz's quite special command of musical space and time. Few composers have possessed a comparable skill for evoking in the mind's ear so varied a sense of aural perspective from the closely intimate to the vastly panoramic, and such shifts play a fundamental role in articulating the structure of La damnation.

As for timing, no 19th-century master, not even Mendelssohn, could move as fast as Berlioz: the total, split-second transformation at Mephistopheles's first entrance is only the most obvious example. In the final tableau, and at a pace to leave Wagner's Valkyries standing, Berlioz sets Faust and Mephistopheles galloping across country to the hellish abyss, scattering the devout in their way and gathering an ever more gruesome entourage of spectres behind them. In performance, and still more in the memory, this devil's ride seems to achieve epic proportions. Yet it actually last little over three minutes.

Berlioz being Berlioz, his newest conception was bound to provoke resistance and the Parisian premiere in 1846 duly proved the biggest flop of his career. A pity, for a Faustian approach to Les Troyens might have rescued that work's many beauties and excitements from their dispersal amidst tracts of grand opera recitative. Nor has the structure of La damnation inspired many follow-ups, unless one counts Schoenberg's far less volatile Gurrelieder - or until, if ever, Robin Holloway's concert opera in progress on Peer Gynt reaches performance. Perhaps the most fully realised large-scale form in Berlioz's output, La damnation also remains his most tantalising in its potential.

'La damnation de Faust': 7.30pm tonight, Royal Festival Hall, London (071-928 8800); broadcast 3.10pm tomorrow on Radio 3