VAUGHAN WILLIAMS'S JOB.
Even those who are very familiar with Vaughan Williams's Job: A Masque for Dancing will find revelations in Julian Rushton's new edition, published by Oxford University Press as a study score. Most startling among these is the first publication (in appendix 1) of music that was cut from the work. As the new score is so affordably priced, this additional material is reason enough to acquire the edition. There are plenty of other reasons to recommend it: the note-text has been entirely reset, so it is a lot easier to read than the various photo-reduced scores derived from the 1934 full score, starting with the earliest-abnormally tiny-miniature score in 1935 (the review by Scott Goddard in Music af Letters 16, no. 4 [October 1935]: 355, noted that it was "for good eyes only"), and its successors in a more usable format (a study score first appeared in 1947). It is also the first to include measure numbers (which restart at the beginning of each scene), as well as replacing rehearsal letters with numbers: a sensible procedure since the old edition (and the old orchestral material) had so many rehearsal letters that it ran to fig. HHh. Here that becomes the much more practicable fig. 52. 1 hope Oxford University Press will make available a large format conducting score of this edition in due course (as they have for the new editions of symphonies nos. 4-9).
The publishing history of Job began in 1931 when Oxford University Press issued an arrangement for piano by Vally Lasker. By the time this appeared, Vaughan Williams had conducted both the premiere at the Norwich Festival on 23 October 1930, and a BBC broadcast on 13 February 1931 (when it was preceded by Hoist's Savitri, conducted by Adrian Boult). The stage version was given at the Cambridge Theatre in London on 5 July 1931, conducted by Constant Lambert, using his arrangement for the reduced orchestral forces that could be accommodated in a small theater pit (with his extensive experience of ballet, Lambert had been Vaughan Williams's choice for the task of making this reduction) and this was repeated at the New Theatre, Oxford, on 24 July 1931. By the time the piano score came out, some small cuts had been made to the work by Vaughan Williams, and Job was fixed in the form we know it, though as yet it carried no dedication. Several concert performances quickly followed, all of them conducted from the autograph full score. Vaughan Williams gave the first public performance in London at Queen's Hall on 3 December 1931. Adrian Boult conducted it for the first time in a BBC broadcast on 19 February 1933 and gave it again with BBC forces at the Festival of Music and Drama in Canterbury Cathedral on 8 June 1933. A few months later, he included it as the final item in the BBC Symphony Orchestra concert at Queen's Hall on 14 February 1934 (the first half comprised Weber's Oberon Overture, and Brahms's Piano Concerto no. 1 in D minor with Artur Schnabel). In his diary the previous day (13 February), Boult noted "Job rehearsal and dedication." This dedication was closely connected to the publication of the score: at Boult's request, the cost of engraving the full score for publication was subscribed by the members of the Bach Choir as a farewell present to Boult when he gave up the conductorship of the choir due to his BBC commitments. When Vaughan Williams heard about this, he dedicated the work to Boult. As Michael Kennedy described it in his biography (Adrian Boult [London: Hamish Hamilton, 1987], 176), at the rehearsal on 13 February 1934 the conductor "was amazed, on opening the manuscript full score to find that it was now dedicated to him." The printed dedication "To Adrian Boult" duly appeared on the full score, which was prepared from the autograph. Vaughan Williams subsequently gave the autograph manuscript to Boult who presented it to the British Library in 1968 (where it is Add MS 54326).
For the rest of his life, Boult was the most devoted advocate of Job: he made four studio recordings of it (in 1946, 1954, 1958, and 1970), and took it around the world, performing it in Salzburg with the Vienna Philharmonic (in 1935), with the Boston Symphony Orchestra (1946), Amsterdam Concertgebouw Orchestra (1947), Chicago Symphony (Ravinia Festival, 1949), and Vienna Symphony Orchestra (1965), among others. In Salzburg, Boult later recalled that "Bruno Walter came to me afterwards with tears in his eyes and said it was the most beautiful music he had ever heard" (quoted in Journal of the RVW Society 19 [October 2000], 7). He conducted numerous BBC broadcasts of Job (in 1933, 1934, 1940, 1943, 1947, 1948, 1949, 1950, 1951, 1962, 1966, 1968, and 1970), and on 12 October 1972, he gave it as the final work in the Vaughan Williams Centenary concert at the Royal Festival Hall (televised; published on DVD by ICA Classics 5037 )-a performance that Ursula Vaughan Williams described in an unpublished letter to Boult as "the Job of a lifetime." On 16 March 1973, Boult even conducted Constant Lambert's reduced orchestration with the Polyphonia Orchestra (probably the only time he performed this version). At the age of 88, on 17 August 1977 in the Royal Albert Hall, Boult conducted Job with the BBC Northern Symphony Orchestra. For those of us lucky enough to be there, or listening on the radio, it was an unforgettable experience (even finer, in my view, than the 1972 performance), and it turned out to be the last of his 251 appearances at the Proms (a recording of this magnificent performance is due to be released in autumn 2019 by CRQ editions). Boult conducted the first night of a new stage production at the Royal Opera House on 20 May 1948, and returned for further performances there in 1955 and 1970. The 1948 production was described at the time as the first staging to use Vaughan Williams's full orchestration, a claim often repeated since, but a letter from Constant Lambert to The Times (26 May 1948, p. 5) put the record straight: "My reduction for an orchestra of fewer than 30 players was a temporary expedient.... The moment the Sadler's Wells orchestra became enlarged to symphonic stature, my reduction was quite naturally dropped, having fulfilled its purpose." A review in The Times on 16 October 1935 (p. 12) confirms this: "For the first time [15 October 1935] Vaughan Williams's own full score was used instead of Mr. Constant Lambert's clever reduction." Though other conductors performed Job-notably Lambert, who did it on many occasions in the theater ("wonderfully" according to Vaughan Williams)-Boult's name is inextricably bound up in its performance history. After Boult's retirement in 1977 (he died in 1983), it found eloquent advocacy from the likes of Vernon Handley, David Lloyd-Jones, Richard Hickox, Andrew Davis, and Barry Wordsworth, all of whom recorded it. Others to conduct it included Walter Susskind, Bryden Thompson, John Pritchard, and Sakari Oramo. Andrew Manze's recorded cycle of the symphonies (for Onyx) has been warmly received, and in December 2019 he is performing Job with the London Philharmonic Orchestra. But concert performances are probably rarer than they were, and Job has never quite established itself in the ballet repertory either: it's very effective in the theater, but hasn't been staged with any regularity (though as well as occasional revivals at the Royal Ballet, it was staged by San Francisco Ballet in 1992, choreographed by David Bintley). Shortly before his death Diaghilev had rejected the proposed scenario by Geoffrey Keynes and Gwendolen Raverat as "too English and too old-fashioned," but Lydia Lopokova-who had been a dancer in Diaghilev's Ballets Russes-wrote to Keynes (her brother-in-law) on 7 August 1931 after seeing the first production: "My chief pleasure was that it differed from the Russian Ballet tradition-the most important merit of Job." Incidentally, it was one of the first ballets to be broadcast on BBC television: a slightly shortened version was given by the Vic-Wells company on 11 November 1936 (with Robert Helpmann as Satan). From a purely musical point of view, this luminous score has usually been held in the highest regard by commentators on Vaughan Williams. Frank Howes, in The Music of Ralph Vaughan Williams ([London: Oxford University Press, 1954], 299) described it unhesitatingly as "one of the greatest creations of its composer"-a view that I, for one, would not quarrel with-and Michael Kennedy believed it was "equally splendid" as both a stage work and as a concert piece, "where it has the stature and cohesion of a symphony" (The Works of Ralph Vaughan Williams [London: Oxford University Press, 1964], 221). He adds (p. 225) that "musically, it is a perfect reconciliation of the various elements in his style," though he also points out that "Job is regarded as a wholly English work. The Continental musicians who saw the ballet performed at Oxford during the I.S.C.M. festival in late July of 1931 cavilled at the music's lack of novelty." In spite of Bruno Walter's enthusiasm four years later, Job has only rarely been taken up by non-British conductors: the American Thor Johnson conducted it with the Boston Symphony in 1949, the Czech born Walter Susskind and Vilem Tausky conducted BBC broadcasts, and more recently it was given at the 2014 Proms by the Finnish Sakari Oramo.
Since the autograph manuscript of Job served as the engraver's copy for the 1934 full score, that edition had few serious problems and several minor errors were silently corrected in later reprints of the study score. Now Rushton has sorted out some more ambiguities and errors, and he has also tidied up dynamic markings and other details. His sources include not only the autograph full score and the 1934 first edition, but also the hand-copied rental score of Constant Lambert's version for theater orchestra and Vally Lasker's piano reduction (both of which predate the published full score by three years). Significant changes to the note-text of the 1934 edition are relatively few (they are listed in Rushton's textual notes). One that occurs early on is the placing of the violin I entry, 12 mm. after fig. C (rehearsal no. 3 in the new score). In the old score (p. 8, last m.) the violins enter half way through the third beat (of a 4/2 m.). This has always been a puzzle since in the 1931 piano score-and in Lambert's theater orchestration-the violin entry is firmly on the third beat. That is how it now appears in Rushton's score (p. 9, last m.).
Some alterations are more problematic. I was baffled by a change at the end of scene II in the new score: the percussion parts in the last five measures (timpani, bass drum and cymbal, all fff) have completely vanished, despite being in all earlier printings (compare p. 44 of the new score and p. 39 of the old one). Nor is this among the reductions to the percussion suggested by Gustav Holst (which Rushton helpfully lists in his appendix II). In fact, I understand that the omission of the percussion here was the result of a glitch during production; it should be a simple matter to restore it in a reprint. There's a questionable correction near the start of scene VI: on p. 78, mm. 1 and 2, the violin I part has "pizz." added. The commentary states that although this is not in the autograph or the 1934 full score (p. 69, mm. 1 and 2), the "pizz." marking is in Lambert's theater orchestration (which it is), and "in Boult's score" (as reported to Rushton by one of Boult's pupils). There are two surviving scores in Boult's library, a full score and a miniature score, one unmarked, the other annotated-both are in the possession of the conductor's family (my warmest thanks are due to Anthony Powers for his generous assistance with these scores). Neither of them has the "pizz." marking in the strings at this point. There is also a list of more than forty corrections in Sir Adrian's hand and that makes no mention of it either. It's a shame that Boult's errata list wasn't known about when the new edition was being prepared: some of the errors he noted were corrected in later reprints of the old edition; those corrections and several more have made it into the new edition, but a few have not (page references are to the new edition): on p. 55, rehearsal no. 25, the mf marking in the violas is an oddity that has appeared in every edition of the orchestral score (all the strings, including the violas, are marked pp up to this point, and all the others remain pp); Boult, almost certainly on Vaughan Williams's advice, says "delete mf." which makes musical sense-and Boult's removal of the mf marking is confirmed by Lambert's score where the dynamics remain unchanged. On p. 59, two measures before rehearsal no. 27, the third note of the trumpet I part should, I think, be a written C[natural] (not C[flat])-sounding B[flat]) matching the Bt in the first violins doubling the same line (this rogue trumpet note is also wrong in Lambert's score). The frequent changes of key signature in the "Pavane of the Sons of the Morning" can lead to problems with missing or incorrect accidentals and these have now been corrected, apart from one: on p. 96, last measure, the first note in the first cello part should be an F[natural], as in the harp part (not an F[sharp] as suggested by the key signature). A more substantial change that must have come from Vaughan Williams (given Boult's famous reluctance to tamper with scores) and for which Boult's errata is the only source, concerns the climax of scene VI: in the timpani on p. 88, m. 1, Boult suggests that the last three notes should double the cello line, playing E, A, B rather than the (uncharacteristic) A, A, A which is how they appear in all editions of the score.
There's a conspicuous wrong note on the previous page of the new edition: p. 87, mm. 4 and 5, the timpani roll should be on an A (doubling the organ pedal, tuba and basses), not a C as it appears here (it is correct in all earlier sources). One troublesome note comes thirteen measures from the end of the work (old score, p. 108, m. 2; new score, p. 120, m. 2): in the new edition a [flat] sign has been added to the last note (an E) in the cello I part. This is neither in the autograph nor in the 1934 edition-and the 1934 score does include a t on the first E in the following bar, seeming to suggest that the E preceding is a [natural]. In Boult's marked score, the cello note in question has a red pencil ring around it, indicating a problem to resolve, though it is not mentioned in his errata. Something certainly needs correcting here to reconcile the cellos with the harp, which has an E[flat]. Changing the harp note to an E[natural] is one option that could work musically, but persuasive evidence for Rushton's decision to opt for E[flat] comes from Constant Lambert's score which also adds a [flat] for this note in the cellos.
From a visual point of view, one feature of the original edition of Job-and of Oxford's house style at the time- was printing empty measures without rests, but as (literally) empty measures. This was the practice in Oxford scores until the late 1950s, but in this new edition, rests are shown as we would expect them to be. Other details of musical orthography have been modernized: the timpani tuning is not indicated on the prefatory staves for each scene, the presentation of the harp and percussion parts has been tidied up, and the general appearance of the score is both cleaner and clearer. The matter of the stage directions deserves some mention. In the 1934 score, and in this new edition, these are often shorter than the directions given in the piano score. I wonder if room might have been found in an appendix to give the fuller versions? With their references to works of art other than Blake's illustrations-including named paintings by Rubens and Botticelli-they are of interest in revealing some of Vaughan Williams's thinking about the visual aspects of his conception. Still, any readers wanting to compare the different versions of the stage directions can do so in Alison Sanders McFarland's article "A Deconstruction of William Blake's Vision: Vaughan Williams and Job" (International Journal of Musicology 3 : 339-71). Lasker's piano score is of general interest as the first appearance of Job in print, and a manuscript copy of it (marked "Copy No. II") in Oxford University Press's Archives includes several more suggested cuts that Vaughan Williams was unwilling to sanction (see Simon Wright: "Ballet in black and white: the 'piano reduction,' " blog.oup xom/2015/02/ballet-black-white-piano --reduction [accessed 1 September 2019]). In both the old and new full scores, these passages are indicated in footnotes "if required by the stage" and are never cut in concert performances.
To sum up: in spite of a few problems (notably the missing percussion at the end of scene II) which can easily be put right in a reprint, this new Job is an essential addition to any Vaughan Williams library. The appendix of music deleted by the composer adds significantly to our knowledge of the work, and Rushton's other introductory material provides valuable insights into the composition process. On a practical level, the note-text is not only more accurate but also considerably more legible than its predecessors. I hope Rushton's new edition will stimulate fresh interest in this glorious work, whether in the concert hall or in the theater.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Job: A Masque for Dancing|
|Article Type:||Book review|
|Date:||Dec 1, 2019|
|Previous Article:||GOUNOD'S MASS IN HONOR OF ST. CECILIA.|
|Next Article:||MUSIC RECEIVED.|