Two editions of Ravel's piano Trio.
Maurice Ravel. Klaviertrio = Piano Trio. Herausgegeben von = Edited by Peter Jost. Fingersatz der Klavierstimme von = Fingering of piano part by Pascal Roge. Munich: G. Henle, 2012. [Preface in Ger., Eng., Fre., p. iii-vii; analytical note by Maurice Ravel in Ger., Eng., Fre., p. viii-ix; score, p. 1-45 and.2 parts (violin and violoncello); Bemerkungen = Comments, p. 46-53; Ubersetzung der Vortrags- und Tempobezeichnungen = Translation of Expression and Tempo Marks, p. 54. ISMN 979-0-20180972-4; pub. no. HN 972. [euro]25.1
In the early 1990s, when Ravel's works briefly emerged from copyright, the first critical editions of his piano music (from Edition Peters.) opened with a brief discussion by their editor, Roger Nichols, of the delicate relationship between manuscripts and editions. Nichols's first two paragraphs end respectively: "The apparently laudable desire to go back to what the composer originally wrote needs therefore to be tempered with a certain amount of common sense." and "in the course of conversations with a number of composers of our own time, I am given overwhelmingly to understand that they would actually be angry if future editors ignored their carefully prepared printed scores and went hack automatically to their original autographs for a so-called true reading" (Roger Nichols, "Ravel's Piano Music--A New Edition," in Maurice Ravel, Gaspard de la nail [London: Edition Peters, 19911, 3).
This general philosophy has long informed critical editing, to the point that the term urtext ("original text") has tacitly come to encompass the composer's Fassung letzter Hand. The recent reemergence of Ravel's music from a further stint in copyright has now prompted editions of his Piano Trio from Barenreiter and Henle, both of which favor Ravel's manuscript fair copy of 1914 over the first edition of 1915. Completed in September 1914 in St-Jean-de-Luz, the manuscript. shows notable variants from the first edition (published 1w Durand in June 1915, several months after the work's concert premiere on 28 January), particularly in the "Pantoum" second movement where it presents some different instrumental textures and one less measure on its last page. As this was the manuscript submitted by Ravel to his publisher Durand, the first edition's variants evidently result from intermediate emendation.
Henle's and Barenreiter's editors explain their decision as having been prompted by one reported document and a pair of published ones: a set of instructions now lost. drawn up by Ravel as he was preparing to enlist in the armed services in early autumn 1914, reportedly containing details of exactly how to print his newly completed Trio in the event of his unavailability or demise; and the memoirs of Ravel's friend Alfredo Casella (i he pianist at the Trio's premiere), who recalled, decades later, that Ravel "entrusted me with the correction of the proofs, as he himself was at war" (as quoted in the Henle edition, p. v; original in Alfredo Casella: Gli anni di Parigi: Dai documenti, ed. Roberto Calabretto [Florence: Leo S. Olschki, 1997], 87). The Barenreiter edition quotes a slightly variant memoir: "Ravel was very grateful to me [for the Trio's first performance] and wished me to correct the proofs of the work" (p. ix; cited from Music in My Time: The Memoirs of Alfredo Casella, trans. and ed. Spencer Norton [Norman: University of Oklahoma Press, 1955], 122). On that basis, both editions conclude that the first edition's variants from the autograph manuscript cannot reliably be regarded as authentic.
The obvious implication is noted in the Barenreiter introduction: "Whether Casella ultimately adhered to Ravel's notes in correcting the proofs or made changes of his own can no longer be determined today" (pp. ix-x), while Peter Jost, in the Henle preface. observes, "It remains unclear whether, and/or to what extent, Casella observed Ravel's above-mentioned remarks, since they, like the proofs, have been lost" (p. v). In his comments, Jost concludes: "As mentioned in the Preface, Ravel was unable to supervise the publication of his Trio himself, but entrusted Alfredo. Casella with the task," adding that the first edition's changes "presumably derive from the experiences of the musicians in rehearsals and in the first performance" (p. 50)--which Casella reported as having, been marred by some "blunders" from the "mediocre" string players, Gabriel Willaume and Louis Feuillard (Music in My Time, 122).
This procedure could immediately be justified by prominently labeling both editions "1914 manuscript version "--a matter of importance to performers, since the variants render the editions incompatible in rehearsal or performance with the established Durand score or parts. Since neither edition does so, responsibility must weigh on them to establish an adequate case against the first edition, given the serious risk (should their surmise be wrong) of losing the crucial final stratum of composer's corrections and revisions in a major repertoire work. The hazards are well illustrated by a Wiener Urtext publication of Debussy's first book of Preludes (Michael Stegemann, ed. [Vienna: Wiener Urtext Edition, 1985]), which prioritized Debussy's manuscript over the first edition (Durand, 1910), undoing some substantial variants in the latter and flagging others as doubtful--unaware of a set of proofs in the Durand archives with annotations in Debussy's hand authenticating the first edition readings. For the present case, relevant data can be sought in four areas; Ravel's relations with Durand and Casella (along with what can be imputed from Casella's statements); Ravel's whereabouts and activity over the period in question; how the sources relate; and the musical and technical character of the variants themselves.
Scrutiny is also prompted in view of what might be implied about the probity of Casella, and of Jacques Durand, the most reputable French publisher of his era and Ravel's exclusive publisher from 1906 onwards. Much about Durand's international profile and professional philosophy can be gleaned from his memoirs, including his stated disapproval of performers tampering in print with composers' texts (Quelques souvenirs d'un edileur de musique: 2e serie (1910-1924) [Park: A. Durand, 1925], 62). Indeed a host of legal and personal complications is likely to have exploded had he, of all publishers, allowed unsanctioned interference in a major new work by Ravel (of all composers). Yet Ravel's subsequent correspondence with Casella and Durand shows not a word of discord rippling his ever-cordial relations with them both (his letters to Durand continue to start "Mon cher ami"). This from the composer who, in June 1914, had a letter of protest published in major London newspapers after Diaghilev omitted the chorus from London performances of Daphnis et Chloe, and who reacted with fury in 1931 when Paul Wittgenstein meddled with the piano writing of his Piano Concerto for the Left Hand. Neither the Barenreiter nor Henle edition takes any prompting from this well-documented aspect of Ravel's character. The Henle editor, Peter Jost, alone briefly considers whether Ravel's now-lost publication instructions might have exceptionally authorized changes, and decides it unlikely (p. 50). (If Ravel did so, on what basis did both editors overrule it? None is stated.)
Regardless of that. the case in both Barenreiter and Henle stands or falls by the presumption of Ravel's unavailability to supervise the Trio's publication--the circumstance Ravel foresaw in August 1914. Barenreiter's case rests on two key statements, both unreferenced: "Ravel was not present [at the premiere on 28 January 1915], as by this time he was already doing war duty as a military truck driver"; and following the quotation (in footnote 9) of a letter from Ravel to Helene Kahn-Casella dated 21 September 1914, the editor, Juliette Appoold, remarks, "A short while later Ravel was approved for war duty as.a truck driver" (p.
Both statements are spectacularly wrong. After being rejected for military service at the end of August 1914, Ravel subsequently returned to Paris (by no later than November), remaining there for one of his longest ever uninterrupted Parisian sojourns, living at the family home in avenue Carnot for almost a year and a half until March 1916. He is documented as taking driving lessons in Paris in November 1914 (Roger Nichols, Ravel [New Haven, CT: Yale University Press, 2011], 179), composing the choral Trois chansons between December and February, and being a constant and refreshing presence .at. rehearsals for the Trio's premiere (Rene Chalupt and Marcelle Gerar, Ravel ad miroir de ses lettres [Paris: Robert Laffont. 19561, 121. confirmed by Casella's memoirs, as partly quoted in the Henle preface, p. v). There is no reason to believe he was not at the premiere (though it makes little difference to the present case); to have absented himself would have been quite out of character with his habitual courtesy in such matters. There could not have been a better time and place for him to supervise at least the Trio's early publication stages.
On 10 March 1915, three months before the Trio was published, Ravel was finally accepted for military service--and assigned to driver training and truck servicing no farther away than the army transport depot at rue Vaugirard, within walking distance of Durand's offices. During that period he was also editing Mendelssohn's piano works for a new Durand wartime edition (the first volumes appeared in 1915). involving frequent interaction and transfer of documents with Durand (see Arbie Orenstein, A Ravel Reader Correspondence, Articles, Interviews [New York: Columbia University Press, 1990], appendix C, pp. 512-16). His correspondence (Nichols, Ravel. 181-82: Orenstein, A Ravel Reader. 159) attests to continued social life until he eventually left Paris on active service a full year later, on 14 March 1916--driving his truck "Adelaide," a very different role from that anticipated in 1914. From near the Verdun front he maintained correspondence with Lucien Garban (his house editor chez Durand and a lifelong friend) over proofing and publishing details for his Trois chansons (Arbie Orenstein, "La correspondance de Maurice Ravel a Lucien Roger Garban I," Cahiers Maurice Ravel 7 [20001: 19-68). The idea of the same composer, a year earlier in Paris, ignoring proofs of his far more complex Piano Trio, while in constant contact with Durand over other matters, is untenable. (Nor indeed would Durand have allowed any such situation.)
Ravel's constant presence and easy availability until at least early March 1915 readily the loss of his earlier publishing instructions for the Trio--a document whose redundancy he had noted just two weeks after first mentioning it (in the letter to Helene Kahn-Casella of 21 September 1914. quoted in the Barenreiter introduction, p. ix). Possibly never delivered to Durand, the document would have been rendered permanently redundant by Ravel's presence at rehearsals, always, his favored forum for musical testing and revision through interaction with performers.
The Henle preface offers a more guarded (though unreferenced) chronology, stating that the Trio's publishing timetable proceeded more slowly than usual,
apparently because of the war. This is particularly important inasmuch as Ravel, clue to his slight build and light weight., was initially rejected as unsuited for military service and, upon his return to Paris in December 1914, would have been able to supervise the printing himself. When the Trio was finally sent to the printer's in spring 1915 [the German has "im Fruhjahr 1915 in Satz ging," the French "fut donne a graver au prin temps 19151, he was no longer in a position to correct the proofs.... (p. v)
Whichever way one reads this (sent either to engraving or to print, in either early 1915 or spring 1915), its final phrase is unsustainable, and in passing pinpoints the other pivotal assumption that both Henle and Barenreiter take as given: that the first edition's variants were introduced at proof. This in turn rests on a further assumption stated explicitly in both Henle (p. 50) and Barenreiter (p. x): that Ravel's manuscript (formerly in the Durand archives, now in the Harry Ransom Humanities Center at the University of Texas in Austin) served as engraving copy. In fact it shows no sign of having done so. Completely absent is the normal panoply of publisher annotations and instructions to the engraver, or engraver's pagination marks for the first edition's score or parts: the only additions are a couple of Durand house stamps, a few penciled in-house revisions (remedying the likes of the odd missing clef change), and lightly marked numbers throughout that suggest different pagination casts for copying parts and another score. (There are no signs of its having been performed from.) This would suggest that Durand, doubtless in consultation with Ravel, had performing copy made from it, and delayed engraving until the work had been tried out.
Where the work was really tested emerges from a source unmentioned by Barenreiter or Henle but cited above: Jacques Durand's Quelques souvenirs (pp. 79-80). First, Durand recalls Ravel "bringing" him the manuscript, presumably in November 1914 (regarding Ravel's apparent decision not to send it from St-Jean-de-Luz, see M. Delahaye, "Quatre letters inedites de Maurice Ravel a Jacques Durand," Cahiers Maurice Ravel 14 : 98). "Peuapres," Durand continues, he attended a house rehearsal of the Trio given by Casella, Georges Enesco, and the distinguished cellist Joseph Salmon., culminating in an unforgettably superb performance which the players then encored. There, obviously, was Ravel's ideal opportunity to test the Trio with the very best performers. It is inconceivable that Ravel was not central to this initiative involving Enesco, one of his oldest musical friends. There and then, too, was Ravel's obvious window for revision (with or without input from the concert premiere) on an engraving copy now lost, avoiding expensive proof changes. (Proof amendments also tend to leave visible traces or spacing irregularities on the printed page, something the Durand edition does not show, not .even on the last page of the Pantoum around the added measure.) Whichever proof stage or stages Casella may then have checked (a score as complex as this one would likely have run to four or five proofs), his role would have involved no more than checking for accuracy. His help may have been of particular relevance later on. if Ravel was more embroiled in rue Vaugirard ditties (and possibly also knee-deep in Mendelssohn proofs), by which stage no substantial amendments would be allowable. (This could also account for Casella's inaccurate "at war" mention, allowing for vagueness of memory, decades later, that also led him to misstate the month of the 'Trio's premiere.)
If the idea is thus completely untenable, from any historical standpoint. of the first edition having bypassed Ravel's control, it remains relevant to consider the nature of its musical variants. The Heide commentary defines these mostly in two categories, either performing markings that leave pitch and structure unaffected, or "simplifications" that include "transferral of passages for the string instruments to the piano" (p. 50; a definition that might suggest 'influence from historical assumptions). In fact the variants seem entirely characteristic of revisions Ravel often made to works before and sometimes after publication: added articulation, phrasing and dynamics, useful cautionary accidentals at modaly complex moments, clarifying and tidying textures, and generally remedying omissions of accidentals and related details. All of it suggests the fruit of rehearsal, judiciously weighed and focused with a master composer's grasp of his own music. (Even apart from the experience with Enesco and Salmon. Ravel was easily experienced .enough not to be thrown by shortcomings of the string players at the concert premiere.)
Neither Henle nor Barenreiter discusses the technical feasibility of the original manuscript string figurations in the Pantoum. As a former professional violinist, I would aver that not even Jascha Heifetz could have managed those of measures 25-30 at Ravel's indication of * = 192 (long acknowledged as a good if demanding tempo). with their sudden triple-stops and repealcd notes, within a bow, and arpeggios ricocheting across almost three octaves, all within very fast triplet-eighth-note figurations. In practice they can only result in congestion, slowing down, or technical approximation; the first edition, far from compromising them, renders them (just) playable, thus demanding the utmost rather than the impossible. This it mostly does by the thoroughly Ravelian procedure of removing pitch doublings within or across instruments, as at figure 4 and between figures 16 and 22. (At figure 4 the first edition's retouches match what Ravel worked into the manuscript at figures 7 and 18, thus removing a textural disparity that Barenreiter and Henle reintroduce.) All this accords with Ravel's well-known bon mot that he had spent a year composing Le tombeau de Couperin, then a further three years removing the unnecessary notes; it can also be seen in post-publication retouches he made to his own printed score of Gaspard de la nuit (from his personal library, now deposited at the Bibliotheque nationale de France).
Regarding "transferral" of material, the nearest that can be seen is the rhythmic transferral of some triplet subdivisions across existing textures: a few right-hand piano lines in the Pantoum are thus rhythmically arpeggiated where the manuscript had presented them chordally (notably between figures 2 and 3), while a small handful of string triplet groups are simplified to quarter-notes. While providing some cover for the exposed string patterns, this mainly increases textural integration, and again makes a better match with some parallel passages, notably at figure 15. Cello retouches around figure 11 remove a few fast reiterations of harmonics unlikely to "speak" on cello strings, while a redirected cello line before figure 13 clarifies voice direction; a removal of some piano tie-overs (shortly after figure 15) makes no audible difference but averts a hand collision. just before figure 15, a manuscript measure in 3/2 for the strings is remetered in 3/4, ensuring more reliable ensemble with the piano at what all performers know is one of the scariest ensemble moments in the whole work. The largest revision is the first edition's added single-beat measure of piano bass anticipation at figure 22, allowing an extra split second for the subito pp to register (as well as For cello and violin respectively to change from pizzicato to arco and leap up three octaves), by dint of a characteristically Ravelian moment of metrical punctuation (almost exactly the same thing occurs in his song Sainte at mm. 8 and 15, or m. 28 of Une barque sur l'ocean). Barenreiter and Henle reverse every one of these Changes Henle shows a few as footnotes or ossias). Both editions also restore a matching pair of manuscript piano ties in the first movement, mm. 86-87 (the Henle edition tacitly, the Barenreiter commentary misstating their location), having perhaps overlooked that their removal in the first edition sets an important imitation point in clearer relief (answering the theme just enunciated by cello then piano right hand).
One can easily see how Ravel, risk taker par excellence, might have perpetrated a few moments of technical overreach while composing in St-Jean-de-Luz away from his home piano and guinea-pig colleagues--until his return to Paris allowed test-driving to his usual exacting standards. Barenreiter eventually yields to the playability issue in the closing two measures of the Pantoum, by exceptionally according the first edition priority over the manuscript's "almost unplayable" (for which read "unplayable") piano figurations--thereby effectively negating its whole editorial premise.
Although Barenreiter and Henle present their editions as being based on the manuscript (and in the process. correct several old Durand misprints), both incorporate an unpredictable amount of detail from the first edition, including some slurs, dynamics, ties, and accidentals but not others (not even the piano's canceling naturals to D at the end of the first movement, mm. 13 and 15, or the sensible augmentation dots to the piano's half-notes in the fourth movement, mm. 7-8). No rationale of selection is stated, though both select a remarkably similar body of readings for respective inclusion and exclusion. Both remove the first edition's 8va bassa for piano left hand in the first movement, mm. 68-71 (Henle mentions it in a footnote), turning the first edition's steady bass descent over three pages (a standard Ravel footprint, as in "Scarbo" or the opening tableau of Daphnis) into leaps of a seventh upwards then a ninth downwards that also result in hand collisions. (The missing portion of indication coincides with manuscript system breaks, a perennial place for such blips. Both editions then immediately contradict what they have .just clone by tacitly adopting the first edition's correction of the manuscript's foreshortened 8va bassa marking in the following measure, 72.) Both editions also restore the manuscript's conflict of violin B against piano A in the fourth movement, m. 9 (surely an obvious slip, amended in the first edition to A for both), though Henle at least footnotes the variant.
Henle and Barenreiter thus both present, as a critically definitive text, an unexplained blend of two very disparate sources that accords only partly with their defined editorial approach. Additional to that is the matter of further retouches in later reprints from Ravel's lifetime (including a few important accidentals). listed in Henle and Barenreiter but generally accorded little more favor than the 1915 first edition. There is no reason whatever to regard these as other than authorial, probably implemented with the help of Lucien Garban.
The main differences between Barenreiter and Henle concern the latter's presentation of more variants on the page. Henle also indulges its habit of adding editorial piano fingering (by Pascal Roge), an element that might seem akin here to issuing Tour de France contestants with bicycle-riding instructions. (Why just the piano part? The string pails are equally hard.) More ironically, Henle shows some first-edition fingering removed (as also in Barenreiter) only to be overwritten by identical editorial fingering (5-1 in the first movement, m. 60; and 5-5 later at in. 75). Henle encloses various other editorial additions (though not its fingering) in parentheses, leaving readers to distinguish them from Ravel's own numerous parenthesized indications.
The generally concise Henle preface fills one-and-a-half pages (in each language), as against Barenreiter's five-and-a-half, a disparity explained by the latter's generous supply of information including "Ravel added many dynamic marks to his Trio" (p. xi), "In several passages he calls for the use of mutes" (p. xi), "In Ravel's piano music the melody part is sometimes given to the left hand" (p. xii), and even this: "Ravel uses several virtuoso piano techniques as expressive devices. Among them are the very first measures of the opening movement, where the melody consists of parallel .chords in dotted rhythm" (p. xii). While the Barenreiter editing shows its best skills at spot-the-difference (an aspect carried out fairly thoroughly, if poorly tabulated), Henle presents its commentary more cogently (apart from the documentary mis-readings noted above), and embodies a larger number of sensible corrections (despite the ones it also removes). The Barenreiter commentary leaves several matters of musical literacy in doubt, repeating oddly pidgin-like terms such as "bar begin," "grace note with slash through stem," or "redundant accidental in round brackets" (for which read "cautionary," most of which the edition discards despite their practical use, while diligently reinstating other redundant accidentals, as in the third movement, m. 45). The term "cautionary" as such makes its first Barenreiter appearance in the note to the second movement, m. 50, erroneously applied to an active inflecting accidental. Many of the commentary notes betray loose phraseology or errors of pitch, octave definition, or instrument labeling (occasionally referring to pitches not present in the measures concerned), or leave German or French prepositions or articles in the midst of English sentences. Editorial interventions left unmentioned include some implausible piano rephrasing in the first movement, mm. 86-87 (also in Henle), and chord restemming in the third movement, in. 47, while numerous lines of commentary are taken up with notes about the vertical placing of metronome indications (and which, unless read very carefully, can convey the misapprehension that the irst edition omitted them).
Meanwhile some problems long familiar to performers remain. Probably the most notorious one concerns which clef is meant for the first movement's final piano right-hand dyad. Barenreiter leaves it without comment in default treble clef (not quite making sense); Henle editorially indicates bass clef without comment, having reportedly declined some source documentation offered by Pascal Roge (kindly conveyed in an e-mail message of April 2013 from him to myself). Henle generally does a better job than Barenreiter of lining up grace-note figurations that involve all three players (notably in the third movement, m. 57, where Barenreiter loses the manuscript's careful alignment), but the issue also needs. more editorial attention, notably in the first movement, where Barenreiter also undoes some judicious first edition alignments. A list of French musical terms and translations at the end of the Henle commentary again suggests the Tour de France analogy, but also needs better translation (for example "ziemlich Iebhaft/rather lively" instead of "ziemlich schnell/rather fast" for assez vif and "follow" not "continue" 16r suivez.).
At present one of the most usable texts For Ravel's Trio, besides Durand, is an annotated Masters Music edition by pianist Richard Dowling (Boca Raton, FL: Masters Music, 2007). Even if readers may disagree (I do) with some of his extrapolations and aspects or editorial presentation, he explains his procedures, shows the vital musical in situ, provides more relevant documentation than Barenreiter or Henle (though he too misses Durand's Quelques souvenirs, and runs somewhat with the Casella proofing hare), and prints the Pantoum in both its 1914 and 1915 versions, editorially correcting perceived misprints. His inclusion of Ravel's descriptive analysis of the Trio, unlike Henle's, references its first. publication (Arbie Orenstein, "Le trio de Maurice Ravel: Analyse par l'auteur," Cahiers Maurice Ravel 9 : 59-69); Barenreiter appears unaware of the document's survival. Informed by performing experience (if somewhat overloaded by it in places), Dowling's edition makes the point that a work like this needs input: from experienced performers as a vital part of editorial preparation--as Ravel doubtless understood when enlisting Casella's help.
The rear cover blurb on the Barenreiter edition reads, "Barenreiter Urtext is a seal of quality assigned only to scholarly--critical editions. It guarantees that the musical text represents the current state of research, prepared in accordance with clearly defined editorial guidelines." Its edition of Ravel's Trio comes nowhere near meeting such criteria. More surprising is to see the Henle edition, less ineptly drawn up, following so uncritically into the same traps. For such enterprises from two major publishers to be .based from start to end on misreading, ignorance, and disregard of easily verified data, and so shot through with musical unawareness, is more than astonishing, and calls for some urgent addressing to avoid bringing scholarly editing into disrepute will performers. Given all that is irreparably wrong in their present state, and the impossibility of just relabeling them "1914 manuscript version," 1 can see no proper option other than for these two editions to be withdrawn and re-edited.
Royal Academy of Music, Landon
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||'Trio pour Piano, Violon et Violoncelle' and 'Klaviertrio - Piano Trio'|
|Article Type:||Book review|
|Date:||Dec 1, 2013|
|Previous Article:||Arias for a tenor from Mozart's Vienna.|
|Next Article:||The vocal score for Messiaen's saint francois d'Assise.|