A London Symphony


A London Symphony is the second symphony composed by Ralph Vaughan Williams. The work is sometimes referred to as Symphony No. 2, though the composer did not designate that name for the work. First performed in 1914, the original score of this four-movement symphony was lost and subsequently reconstructed. Vaughan Williams continued revisions of the work into its final definitive form, which was published in 1936.

Instrumentation

The work is scored for:
Vaughan Williams said that while the title may suggest a programmatic piece, it was intended to be heard as absolute music. In a programme note in 1920, he suggested that Symphony by a Londoner might be a better title. However, he allowed the conductor Albert Coates to provide elaborate descriptions for the 1920 performance.
The symphony is in four movements.
;1. Lento – Allegro risoluto
The symphony opens quietly, and after a few nocturnal bars, the Westminster chimes are heard, played on the harp.
\relative c' After a silent pause, the allegro risoluto section, much of it triple forte, is vigorous and brisk, and the ensuing second subject, dominated by the wind and brass, is no less so. \relative c \relative c After a contrasting gentle interlude scored for string sextet and harp, the vigorous themes return and bring the movement to a lively close, with full orchestra playing fortissimo.
;2. Lento
The movement opens with muted strings playing ppp. Vaughan Williams said that the slow movement is intended to evoke "Bloomsbury Square on a November afternoon".
\relative c' Quiet themes led in turn by cor anglais, flute, trumpet and viola give way to a grave, impassioned forte section, after which the movement gradually subsides to its original quiet dynamic.
;3. Scherzo
In the composer's words, "If the listener will imagine himself standing on Westminster Embankment at night, surrounded by the distant sounds of The Strand, with its great hotels on one side and the "New Cut" on the other, with its crowded streets and flaring lights, it may serve as a mood in which to listen to this movement." In the definitive score, the movement revolves around two scherzo themes, the first marked fugato and the second straightforward and lively.
\relative c' The piece closes with muted strings playing pppp.
;4. Finale – Andante con moto – Maestoso alla marcia – Allegro – Lento – Epilogue
\relative c'
The finale opens on a grave march theme, punctuated with a lighter allegro section, with full orchestra initially forte and appassionato. After the reappearance of the march, the main allegro theme of the first movement returns. Following this, the Westminster Chimes strike again, this time the harp plays the first three-quarters of the hour chimes, and there is a quiet Epilogue, inspired by the last chapter of H.G. Wells's novel Tono-Bungay:
The last great movement in the London Symphony in which the true scheme of the old order is altogether dwarfed and swallowed up... Light after light goes down. England and the Kingdom, Britain and the Empire, the old prides and the old devotions, glide abeam, astern, sink down upon the horizon, pass – pass. The river passes – London passes, England passes.

History and versions

The symphony was composed from 1912 to 1913. It is dedicated to Vaughan Williams's friend and fellow composer George Butterworth who was subsequently killed by a sniper on the Somme during World War I. It was Butterworth who had first encouraged Vaughan Williams to write a purely orchestral symphony. Vaughan Williams recorded that:
"We were talking together one day when he said in his gruff, abrupt manner: 'You know, you ought to write a symphony'. I answered... that I'd never written a symphony and never intended to... I suppose Butterworth's words stung me and, anyhow, I looked out some sketches I had made for... a symphonic poem about London and decided to throw it into symphonic form... From that moment, the idea of a symphony dominated my mind. I showed the sketches to George bit by bit as they were finished, and it was then that I realised that he possessed in common with very few composers a wonderful power of criticism of other men's work and insight into their ideas and motives. I can never feel too grateful to him for all he did for me over this work and his help did not stop short at criticism."

The work was first performed on 27 March 1914 at Queen's Hall, conducted by Geoffrey Toye. The performance was a success. Shortly afterwards, the composer sent the score to the conductor Fritz Busch in Germany, and the original score disappeared in the upheaval of the outbreak of World War I. The second performance was given in Harrogate on 12 August 1914 by the Harrogate Municipal Orchestra under Julian Clifford. There was a short score, which had been prepared by Bevis Ellis, Francis Toye and George Butterworth, so it is possible that version was used instead. The composer, aided by Geoffrey Toye, Butterworth and the critic E. J. Dent, reconstructed the score from the orchestral parts, and the reconstruction was performed on 11 February 1915 by the Bournemouth Municipal Orchestra under Dan Godfrey.
The symphony went through several revisions before reaching its final form. Vaughan Williams revised it for a performance in March 1918, and again in 1919–1920. This second revision became the first published version, and was recorded for the gramophone in 1925 by the London Symphony Orchestra conducted by Sir Dan Godfrey. It was also recorded in 1941 by the Cincinnati Symphony Orchestra conducted by Sir Eugene Goossens. It had already received its American premiere on 20 December 1920 when the New York Symphony Orchestra played it under the baton of Albert Coates. It is most unlikely that OUP sanctioned the use of the 1920 version in Goossens' recording, since Vaughan Williams had withdrawn it and all scores of the 1933 revision carry the statement:
However, the Cincinnati/Goossens recording was made at a period during World War II when communications between Britain and the USA were difficult, and a "rogue" set was used. In fact, the 1920 version was already in the public domain in the US, having been published before 1923, so it may simply have been cheaper to record that version. It was still possible to buy scores and parts of the 1920 version in the USA until early in the 21st century, although this is no longer the case.
While he was working on his fourth symphony in 1933, Vaughan Williams made time to revise A London Symphony yet again. He regarded this version, which was published in 1936, as the definitive one, and it is this version that entered the repertoire, being played in concert and on record by many conductors.
In 2001, a new commercial recording appeared on Chandos of the original 1914 score, following assent from the composer's widow, Ursula Vaughan Williams, for a recording only, without live performances. The new recording of the original 1914 score attracted attention from various music critics, including some commentary that the composer had cut many bars of interesting music. Richard Tiedman commented:
Andrew Clements has separately remarked:
The main differences between the first and last versions may be summarised as follows:
Below is a summary of the changes made between the original and the two published versions. It shows the number of bars in each movement and the total for the whole symphony:
VersionMvt IMvt IIMvt IIIMvt IVEpilogueTotal
19144082023862271091322
1920407162398173851225
1933407150398162601177

The final version is more than twenty minutes shorter than the original, as some indicative timings show:
1914 version:
1920 revision:
1933/36 revision:
The reception accorded to the Chandos recording of the 1914 score persuaded Ursula Vaughan Williams to allow a live performance of the original version. In November 2003, Richard Hickox conducted the original 1914 score with the London Symphony Orchestra at the Barbican, in the first live performance of this version since 1918. The Proms presented an additional live performance of the 1914 version on 19 July 2005, with Hickox conducting the BBC National Orchestra of Wales.
In his liner note commentary to the Chandos recording of the original version, Michael Kennedy placed the status of the original score as subordinate to the final 1936 published version:

Recordings