Founding Editor Rob Barnett Editor in Chief
John Quinn Seen & Heard Editor Emeritus Bill Kenny MusicWeb Webmaster
David Barker Postmaster
Jonathan Woolf MusicWeb Founder Len Mullenger
Support us financially by purchasing this from
Ralph VAUGHAN WILLIAMS (1872-1958) A London Symphony (1920 version) [51:51] Sound Sleep (1903) [5:28] Orpheus with his Lute (ca 1901/03) [2:36]
Variations for Brass Band (1957) [12:27]
Elizabeth Watts (soprano); Mary Bevan (soprano); Kitty Whately (mezzo-soprano)
Royal College of Music Brass Band
BBC Symphony Orchestra/Martyn Brabbins
rec. Henry Wood Hall, London, 2016. DDD HYPERION CDA68190 [72:22]
In 2015 I reviewed what was then the second recording – but the first in modern times – of the 1920 version of A London Symphony. That was a Dutton Epoch recording by Martin Yates and the Royal Scottish National Orchestra. Now Martyn Brabbins has given us a second modern recording of the score.
It may be useful if I recount briefly the history of the symphony and its various revisions. Vaughan Williams completed the symphony in 1913 and it achieved its first performance the following year. Over the next few years VW made significant revisions to the score, mainly between 1918 and publication of the symphony in 1920. Further revisions followed in the 1930s and the score that we know today was finally published in 1936, superseding the 1920 version. In all these revisions the first movement remained unaltered.
In 2000, with the consent of Ursula Vaughan Williams, Richard Hickox recorded the original 1913 score for Chandos (review). That was a recording which came as quite a revelation to admirers of the work, especially because it gave us the chance to hear significant stretches of discarded music. When Martin Yates recorded the 1920 score he was following in the footsteps of Sir Eugene Goossens who recorded it in 1941 with the Cincinnati Symphony Orchestra. That was at a time when orchestral parts of the 1920 version were still in circulation in the USA so that version was used. The Goossens recording has appeared on CD (Biddulph WHL 016) but the sound is, inevitably, somewhat limited.
There are marked differences between the 1913 version of the score and the familiar 1936 edition. However, there are fewer differences between the 1920 and 1936 versions of the symphony. VW pruned the 1913 score radically in his first revision but subsequent surgery was much less invasive and included re-touching the scoring in places. Looking back at Lewis Foreman’s notes accompanying the Yates recording, I see that in the post-1920 revision VW excised a total of 12 bars from the slow movement and, in the finale, a further 36 bars, 25 of which were taken out of the Epilogue. These passages can be heard in the Brabbins and Yates recordings. Some indication of the extent to which VW cut his first thoughts can be gleaned by noting that the Hickox recording of the 1913 score played for 61:19. By contrast Brabbins takes 51:51 over the 1920 score and Yates takes 48:41.
A comparison of the Brabbins and Yates recordings has been stimulating and revealing. Brabbins opens the first movement softly and mysteriously. He’s broader in his approach than Yates, arriving at the Allegro risoluto at 3:38; Yates gets there at 2:40. I like Yates’ unfolding of the passage but when one hears them side by side I don’t think there’s much doubt that Brabbins is the more atmospheric.
In my review of his disc I expressed some reservation about Yates’ brisk pace for the Allegro risoluto. The energy he generated was bracing but I felt that the music wasn’t allowed sufficiently to breathe. I still feel the same way, especially now that I’ve heard Brabbins. If I say that Brabbins’ speed is more conventional I don’t mean that negatively; it’s more a case of what I’m used to. Even though he’s steadier than Yates – indeed, perhaps because he’s steadier – there’s still all the colour and bustle you need in this music. Brabbins is also very good in the quieter, more gentle passages, achieving poetry amid the surrounding urban bustle. The playing of the BBCSO is excellent. Round one to Brabbins, I think.
Both conductors are satisfyingly spacious in the wonderfully atmospheric Lento. Brabbins took longer over the first movement than Yates but here the boot is on the other foot though, in truth, the differences of pacing are not significant. Focusing just on the Brabbins account for a moment, the Lavender Seller episode (5:24) is ushered in by a lovely viola solo and the clarinettist who takes over the spotlight a few bars later is no less winning. In fact, this is a good point at which to say that Brabbins is very well served by a number of solo players in this movement and, indeed, that the BBCSO as a whole plays with great finesse. Brabbins brings this passage to an ardent climax. One of the key excisions made by Vaughan Williams is a passage six bars long in which the violins have very distinctive chromatic sul ponticello writing, accompanying solo phrases from the horn and cello. (11:02-11:40 on the Brabbins disc). Robert Matthew-Walker mentions in his notes that Bernard Herrmann tried in vain to persuade VW not to discard this passage. Overall, I’d say that Brabbins and Yates are pretty well matched in this movement.
Yates is considerably swifter than Brabbins in the Scherzo, taking a minute less overall. Brabbins is much steadier. Ideally, I’d like a pace somewhere between the two. While I admire the dexterity of the RSNO on the Yates recording the performance is just a little too hasty for comfort, I feel. Brabbins is to be preferred.
Brabbins is impassioned at the start of the finale. Yates is equally strong but almost immediately there’s a significant divergence between the two conductors. In my original review I observed that Yates’ tempo for the Maestoso alla Marcia (quasi lento) was a bit on the brisk side, not respecting sufficiently the quasi lento qualification. At the time I came round to Yates’ view but having now heard Brabbins’ performance I believe my initial instincts were right. Brabbins’ more measured pacing invests the processional with nobility and weight, qualities that I find missing in the Yates performance. As a result of his more measured speed Brabbins arrives at the Allegro section at 3:49 – Yates is there at 3:11. Brabbins is a fraction steadier than Yates in the Allegro and I think that works better; sometimes the phrases sound snatched in the Yates account and overall the music is better articulated under Brabbins. When VW returns to the Maestoso alla Marcia material Brabbins’ more measured speed means that in his hands the tri-fold climax has even more anguish and heft than is the case with Yates. At 8:38 (Brabbins) the harp gently intones the Westminster Chimes, prefacing the epilogue. My goodness, this epilogue is an inspired evocation and Brabbins shapes it with sensitivity, bringing out all the quiet poetry.
In every respect Brabbins leads a splendid performance of this marvellous symphony and he’s supported by the BBCSO on top form. The Yates performance has a lot to commend it but as my comments will have shown I have a strong preference for Brabbins at almost every turn. You may ask why you should hear the 1920 version of the score. I stand to be accused of heresy by fellow VW addicts but I think the 1913 version of the symphony, superlatively recorded by Hickox on Chandos, is for high days and holidays. True, it gives us the opportunity to hear a good deal of beautiful and highly original music that the composer later discarded. However, I think VW was absolutely right to make the cuts. I love this symphony but it does have something of a tendency to sprawl amiably. That tendency is all the more marked in the 1913 version and I’m in no doubt that VW tautened the structure significantly and beneficially in his revisions. Matters are not so clear-cut when it comes to the 1920 score which is significantly closer to the familiar 1936 definitive score than the original version is. That said, there are audible differences and just as I think all VW devotees should hear the 1913 version so the same applies to the 1920 score. And I’m in no doubt that the best way now to hear that 1920 score is in this new Martyn Brabbins performance.
When I received this disc, I thought that Brabbins’ couplings appeared to be something of a mixed bag, especially when one thinks that Yates’ coupling is the Concerto for Two Pianos. However, what Brabbins does here is to give some less familiar VW works welcome exposure by the side of a mainstream piece. The two vocal numbers are a delight. Sound Sleep is a Christina Rossetti setting. It was originally written for unaccompanied SSA voices to be sung at a 1903 festival in Lincolnshire. So successful was it that VW later re-worked it for three solo voices and a small orchestra. The music is utterly charming and quite rapturous in tone. Here we have a trio of delectable voices performing it and the three singers blend and complement each other expertly. It’s uncertain when Orpheus with his Lute was written but Robert Matthew-Walker suggests that the subsequent orchestration for eight woodwind and strings may imply that the arrangement was done for a theatrical purpose. Elizabeth Watts sings it delightfully and the delicate scoring works beautifully.
The Variations for Brass Band was composed as the test piece for the 1957 National Brass Band Championships. Robert Matthew-Walker is absolutely right to cite this piece, together with the much better-known last three symphonies, as evidence of VW’s delight in experimenting right to the end of his life with what were to him new sounds and new musical possibilities. Here he seems to relish the gleaming, silvery sound of the cornets, the mellowness of euphonium tone and the rich depth provided by a quartet of tubas. The work plays continuously and contains 11 concise, inventive variations. It seems to me to be a challenging, resourceful and rewarding piece and this recording should give it wider exposure outside the brass band fraternity. I admire both the crispness and, at times, the poetry of the playing of the Royal College of Music Brass Band. Incidentally, the choice of that ensemble for this assignment is as logical as it is pleasing since Martyn Brabbins is Visiting Professor of Conducting at the RCM.
The recorded sound throughout this CD is excellent, as you might expect given that the engineer is Simon Eaton and the producer, who has no little experience of producing VW recordings, is Andrew Keener. The competitor recording of the symphony from Dutton Epoch, also excellent, comes in SACD format but the Hyperion CD need not fear the comparison. Robert Matthew-Walker’s notes for Hyperion are most interesting.