celestial harmonies
celestial harmonies
jean barraqué (1928-1973)
sonate pour piano (1952)
roger woodward
celestial harmonies 13325-2
recorded october 1972
abbey road studios, london
engineers: stuart eltham and robert gooch
producer: david mottley
hamburg steinway D
digital restoration: gideon boss
re-issue supervised by eckart rahn cover painting by pierre soulages (1964)



In 1961 the French musician and author André Hodeir, best known as a jazz critic, but also part of the Parisian circle of artists and intellectuals around the young Pierre Boulez, published a book called Since Debussy.  Its opinions (which largely coincided with those of Boulez) are generally scathing. Of Stravinsky, he says: “it would be almost indecent to mention him in the same breath with Debussy”, while Schönberg “who remained an exceptional creator as he was busy destroying, ceased to be truly creative as soon as he tried to construct”. Berg (though praised in part) “wasted the last years of his life in a sterile attempt to retrace his steps”. He refers to the “uncertainty and mediocrity” of Bartók’s last works, finds much to admire in Messiaen, but also finds that “even his best pages lack that assertive power which is the sign of the authentic masterpiece”.

So who survives this onslaught? Of the older composers, only the then relatively obscure Anton Webern, who is being hailed—almost sanctified—by the young avant-garde as the “Pioneer of a New Musical Order”. And of the young, just three composers, all of whom just happened to have studied with Messiaen in Paris: Pierre Boulez, Karlheinz Stockhausen, and Jean Barraqué.  Stockhausen, though not accorded a chapter of his own, is mentioned frequently with considerable respect. Boulez is hailed (somewhat equivocally!) as “one of the greatest precursory figures in Western art and thought, one of those men without whom things would not be what they are”, yet ultimately, as a young Moses whose work gets stuck at the border of the Promised Land: stuck not least through increasing caution, and an excessive preoccupation with polish and style (coming in the immediate wake of Pli selon pli, this is perhaps a rather prophetic comment).

All this would already have been enough to guarantee a degree of controversy, but in the final major chapter Hodeir goes several steps further. Its subject, Jean Barraqué, is virtually unknown outside the Boulez circle, and his acknowledged output consists of just four works (he will complete only three more before his death in 1973), of which only a couple of have been performed. But this doesn’t prevent Hodeir from hailing him as even more important than Debussy (which also implicitly raises him above Boulez). And the work on which he heaps the greatest praise is one that has not yet been performed in public (and will not be until 1967), a Piano Sonata (1950-2) that has recently been issued on LP, in a quasi-performance by Yvonne Loriod patched together in the studio from countless short ‘takes’, but whose score would not be available for a few more years.  He writes “one is amazed to think that this towering score [is] the work of a very young man. Certain works of Mozart and Schubert are astonishingly precocious; this one is terrifyingly so.” And he concludes: “It is unclassifiable, incomparable, and to some degree, still incommunicable music…this music lies outside the scope of our era, in any case; it can only belong to the future.”

Wild hyperbole? Maybe so, but Barraqué’s Sonata was, and remains, an utterly extraordinary work. By the time the recording emerged, it scarcely represented the latest style – it was (pace Hodeir) firmly lodged in the sonic world of early fifties serialism—yet even in Loriod’s somewhat dubious rendering, it had a unique, almost unsettling ‘presence’: obsessive, almost hypnotic. However seemingly esoteric, it gained a certain cult following, and this was reinforced by a first ‘authentic’ recording of the work, made in 1969 by Claude Helffer, at that time the most authoritative French exponent of new piano music. Good as this was, it was eclipsed a few years later by the present recording, which so impressed the composer that he promised Roger Woodward a Second Sonata. In reality, this was never going to happen—a year later, the already sick composer would be dead—but it is a striking testament to this remarkable performance, here in an extraordinary digital restoration of the original tapes.

Clearly, Barraqué struggled massively to create his own music, and rarely wrote about it. Yet he wrote with great fluency—and remarkable insight—about the music of other composers. Above all, he had two favourite topics: Debussy and Beethoven, both of whom are pertinent to the Sonata. Not that the work sounds like either: there are none of the Debussy reminiscences that pervade the work of his teacher Messiaen. Yet the whole sense of fluid tempo, of musical time as sort of Heraclitean flux, surely has its roots in Debussy, and as for Beethoven (and primarily, late Beethoven), the influence operates at many levels. As indeed it did for the young Boulez: the one work that obviously springs to mind as comparable to Barraqué’s Sonata is the Second Piano Sonata that Boulez had written just a few years earlier: a piece that Barraqué must surely have known, and regarded as a model to be contested and, if possible, surpassed.

We can be a bit more specific: Boulez’s four-movement Second Sonata is a deliberately anarchic, part-Oedipal response to Beethoven’s Hammerklavier (which he greatly admired), whereas for Barraqué the reference point is clearly the final (32nd) sonata, with just two movements. In practice, Barraqué’s Sonata has just one huge movement, but it falls into two parts, of which the first is predominantly fast, and the second predominantly slow (as in Beethoven’s last sonata). Though Barraqué never wrote a formal introduction to his Sonata—let alone an analysis—there is an interview from 1969, occasioned by Helffer’s recording, that sheds some significant light.  What appeals to him most in the sonata idea, he says, is the notion of duality—of how opposites are juxtaposed and mediated. Hence the two major parts, one fast and one slow. There are also many other ‘twofold’ aspects to the piece, such as loud and soft (there are relatively few ‘mezzo’ indications), and high and low.  But above all, the composer points to the contrasting of ‘free’ and ‘strict’ styles, which inevitably recalls Beethoven’s description of his Grosse Fuge as “tantôt libre, tantôt recherché”. The ‘free’ style, which one hears at the opening, typically consists of often extravagant, constantly varying figures, flung all across the keyboard, with fluid tempi and rhythms. The strict sections, in contrast, tend to consist of tiny, sharply etched little ‘cells’, often concentrated on the middle range of the keyboard, and almost claustrophobic in character, not least because each of 12 pitches always occurs at the same octave (a ‘fixed register’ technique Barraqué had picked up from Webern’s works); to that extent, one could think of the free sections as ‘open’, and the strict ones as ‘closed’ (almost caged-in).

In both parts of the Sonata, the two styles alternate, and they do so in quite distinct, segregated sections which constitute the main formal articulation of the main parts. In the first part there are six such sections (free – strict – free – strict – free – strict), in the second part just three (free – strict –free). In Woodward’s recording, these occur at the following points:


First part                                                                               19’38”

1st free section:      0’00”   1st strict section:              2’05”

2nd free section:    3’08”   2nd strict section:           4’38”

3rd free section:        8’10”   3rd strict section:             9’42”


Second Part                                                                         28’18”

1st free section:      0’00”   strict section:                   16’14”

2nd free section:                                                                22’35”


So the work opens with a whirlwind of different figures, initially played at the fastest (articulate) tempo at which the pianist can manage. Many of these figures will recur frequently, but always transformed in some way—as with Beethoven’s last sonata, the ‘exposition’ is already a fully-blown development section.  Particularly notable is a three-note motive, of which either the first two or the last two notes are repeated; this will be a sort of audible Ariadne’s thread through this seemingly ‘athematic’ work. The frantic pace is not maintained absolutely throughout—there are ebbs and flows—but the general tone is rapid and extravagant. About half-way through this first free section, there are just a couple of trills. They seem almost incongruous amidst this resolutely anti-ornamental music; but they are discreetly planted seeds whose significance will not be fully revealed for another 30 minutes.

Towards the end of this free section we hear, for the first time, a sequence of chords, in a kind of wave form between high and low, and then comes the first strict section, beginning in the middle register, but subsequently including some higher notes. It’s short, but a lot happens: the music twitches frantically, using just short motives (rarely more than 3 notes), but in a fragmented two-part counterpoint. As in most of the strict sections, two notes (the D and F at the top of the treble clef) seem to recur particularly obsessively, but now the repetitions arise from the collision of single notes in different contrapuntal voices. The next (second) free section again begins wildly and explosively, but gradually slows down: from here on, there are all sorts of discrete preparations for the much slower music that will come in the Second Part. And the following strict section—twice as long as any of the preceding ones—now starts out at moderate pace (Modéré), though in general it exhibits the same ‘behaviours’ as its predecessor. The final free section starts out almost rhapsodically, with long, twisting melodic lines, but becomes increasingly abrupt, and ends at high speed.

As for its final ‘strict’ successor, this has once again blown out in length: it’s long as all the preceding sections put together, and it has a lot of work to do. Ultimately, it’s the last chance to develop the musical material at a fast, cohesive pace. And there’s a certain dialectic at work here: even if development involves ripping material apart, it also offers a possibility of momentary reconstruction, of glimpsing a new edifice that, such as it is, will be mercilessly demolished in the Second Part. It follows that it can’t really restrict itself to the typical hallmarks of the preceding strict sections; it has, increasingly, to also admit elements of the free sections as well. It’s not practical to give a blow-by-blow account of what happens here, but one can at least point to one very striking moment (12’06” onwards) where nine little figures, half of them already in the slow tempi that will be typical of the Second Part, are separated by pauses of increasing length: it’s a sort of premonition of what is to come.

The end of this long section gradually slows down, but even so, that scarcely prepares one for the dramatic change in the Second Part. Here, the first (free) section is on a completely different temporal scale to its predecessors: it is almost as long as the entire First Part. It seems to explore and chart a huge, bleak wasteland, occasionally punctuated by little nocturnal oases. The materials are essentially those of the First Part, from both free and strict parts, but slowed down and fragmented to the point of being virtually unrecognizable (the little repeated note figures are the only exception here). There are even passages that quote directly from the opening of the work (page 3 of the score proves a particularly fruitful source of pillage here), but they present the material in retrograde, so that only a slight increase in tempo and continuity gives a clue to their origins.

The (strict) middle part is much shorter, and despite the slower tempi and a greater concentration on the extreme top and bottom of the keyboard, it’s not so hard to hear the relationship to earlier ‘strict’ sections, and especially to the very first one. There, there were just a couple of trills that seemed almost incidental. The same happens here, but this time with drastic consequences:  a bit later (at 19’30”) there’s a sudden eruption of trills and tremolos, all fortissimo or louder, and they come back again at the end of the section. Now it becomes clear that these are not just ornamental trills: they’re a Hammerklavier invocation. In Beethoven’s masterpiece, these trills sometimes drive the music forward, but at other times they seem to impose a furious momentary stasis, and that’s the role they play here in Barraqué’s Sonata.

The final section too is relatively short, and for the first time, the music is dominated by chord sequences, often in waves going, for example, from one note up to six notes, and then back again. It’s as if the polyphony, the linearity of nearly all the preceding music had swallowed up into blocks of sound, though towards the end, this swallowing is contested by brief attempts to grasp back to the opening of the Sonata. As for the last moments, it’s hard to beat Hodeir’s classic, albeit melodramatic description: “Whole slabs of sound crumble and vanish beneath the all-engulfing ocean of silence, until only the twelve notes of the series remain, and even these are plucked off, one by one.”


Richard Toop


p.o. box 30122 tucson, arizona 85751
+1 520 326 4400 fax +1 520 326 3333