Earliest works

Vaughan Williams’s first sustained engagement with orchestral music began in the second half of the 1890s and came to fruition initially with a number of performances in 1901 and 1902. As can be seen from Table 4.1, the first piece on which the composer embarked, the Fantasia for Pianoforte and Orchestra, was in fact never performed, though he worked on it on and off between 1896 and 1904. The Fantasia was one of a number of early orchestral works that were revised (in some cases more than once) after an initial phase of composition, and sometimes after a first performance; this complicates an already patchy documentary record of compositional and performance history, though we are at least helped by the fact that at this stage of his career the composer often recorded dates of composition and subsequent revision in his manuscripts, a practice unfortunately abandoned in later life. Some works, for example the Symphonic Rhapsody, were apparently withdrawn after a single performance, whereas others, such as the Bucolic Suite and the Heroic Elegy and Triumphal Epilogue, received several. In most cases the works were not withdrawn in a single and definitive act but simply allowed to lapse into obscurity. The biggest obstacle to a full assessment of the pre-1908 works, however, is the fact that for the most part we cannot hear them in performance and must rely on the scores alone; this situation is beginning to change, as was noted above, but it will be some time before a fuller picture can be formed. This would be unfortunate with any composer, but in Vaughan Williams’s case the real impact of his music is often especially difficult to imagine with any accuracy from the page alone.

Nevertheless, the scores alone can still tell us a great deal. The four works composed at the turn of the century show a thoughtful, and at times impressive, handling of late Romantic forms and orchestration, with attractive and at times compelling melodic and harmonic invention that, while not yet steeped in the fully developed modality that would characterize his later work, does contain flashes of the mature Vaughan Williams that would emerge over the next decade. The titles are abstract, or invoke generalized semantic associations rather than a specific programme; while all four works pre-date the composer’s intensive involvement with folksong, they do partake of the kinds of generic pastoral elements well established in nineteenth-century music. Vaughan Williams’s models for the Bucolic Suite and the Serenade seem to have been Brahms and Dvořák primarily (filtered in part through his RCM teachers, Stanford and Parry), along with Max Bruch, with whom Vaughan Williams studied in Berlin. In contrast, as Michael Vaillancourt has noted, the Piano Fantasia and the Heroic Elegy and Triumphal Epilogue suggest the more self-consciously ‘progressive’ central-European stream of Liszt, Strauss and Mahler, in terms particularly of development and thematic transformation within one-movement or cyclic formal structures (the Symphonic Rhapsody, whose manuscript the composer seems to have destroyed, may well have reflected similar influences).⁴ In terms of orchestration, the composer used a variety of different ensembles, with an often deft and poetic handling of his resources; this is especially true of the Bucolic Suite, which includes a range of percussion, and wind and string exchanges in the second movement that bring to mind Tchaikovsky. Though Vaughan Williams would later write that by the end of 1907 he felt that his music was ‘stodgy and lumpy’, and that it was this that led him to take lessons with Ravel,⁵ he was certainly capable of a light touch in these early works.

A more monumental tone is evident in the Heroic Elegy and Triumphal Epilogue, which uses the largest orchestral ensemble Vaughan Williams had yet mustered, with triple wind, a large complement of brass, and harps and organ ad libitum (the work was published by Faber Music in 2008 and has now been recorded).⁶ It is the most obviously impressive of this group of four works, and it certainly received the warmest critical reception.⁷ There are echoes of Wagner, Sibelius and Tchaikovsky, and several developmental passages call to mind Strauss and Elgar; the brass-heavy final pages in particular, building on earlier fanfare passages, suggest a new level of orchestral ambition. The two movements are separate but share thematic material; they were originally to have formed the second and third parts of a three-movement Symphonic Rhapsody, but it is unclear whether the first movement was even sketched (the title was revived, however, for the later work by that name).⁸ The relatively unusual choice of B minor as tonic key for the Elegy brings to mind Tchaikovsky’s Sixth Symphony, and the first movement of the Russian’s Fourth Symphony may have inspired subsequent tonal motion through a chain of minor thirds, a feature found in several other early works by Vaughan Williams, including the Bucolic Suite and Burley Heath.⁹ Yet the varied range of influences does not obscure an individual stamp; furthermore, the Triumphal Epilogue contains several striking adumbrations of the later and more consistently individual Vaughan Williams, in particular A Sea Symphony, on which he would begin work fairly soon afterwards. The developmental passages around rehearsal letter R look forward to the Scherzo of the symphony, while the D major melody that launches the Triumphal Epilogue, and returns in grandeur as its peroration (see Ex. 4.1), adumbrates the ‘Token of all brave captains’ theme of the symphony’s first movement. And this latter element has a broader significance: aspiring, gapped D major melodies of this kind would go on to become a staple thematic archetype for Vaughan Williams throughout his career.

Ex. 4.1. Heroic Elegy and Triumphal Epilogue, Epilogue, bars 256–63.

Landscapes with and without figures

As was noted above, the Heroic Elegy and Triumphal Epilogue received positive critical notices, and it was singled out by several critics familiar with other music by the composer as a landmark work confirming his potential.¹⁰ But in 1902 Vaughan Williams began to move in new directions, producing over the next five years a series of one-movement pieces that carry programmatic titles specifying particular geographical locations. These works are relatively short, but a number of them seem to have been conceived with the idea of larger multi-movement cycles (in 1903 Vaughan Williams also began work on A Sea Symphony, of which more below).¹¹ Though they all share an intensified interest in modal harmony, combined with more chromatic elements in sometimes experimental ways, the eight pieces divide into two groups according to whether or not they make clear reference to English folksong. This is quoted explicitly, in the form of complete tunes, in the three Norfolk Rhapsodies, and is more loosely evoked in the principal thematic material of In the Fen Country. These four works were the earliest substantial creative fruits of Vaughan Williams’s folksong conversion experience, as it were, of December 1903, when he first encountered this music in a personally compelling manner, and moved beyond scholarly appreciation to active engagement. This found expression not only in composition but in extensive activity as a collector.

The four other works, of which Boldre Wood is now lost and Burley Heath survives only incompletely, were all dubbed ‘Impression’ (the generic designation also of In the Fen Country) and relate to locations in and around the New Forest, the expanse of unenclosed heathland and forest between Southampton and Bournemouth. This was an area that Vaughan Williams knew well, from family holidays and other connections to the area, and he lectured in Bournemouth in 1902.¹² The manuscripts for the earliest pair of impressions, Burley Heath and The Solent, indicate that when he began work in 1902 the composer planned a set of four pieces, under the broader title of ‘In the New Forest’; it is not at all clear, however, whether the later Boldre Wood and Harnham Down were originally related to this scheme in any way, or were instead a fresh start at representing this particular geographical locale in music. The Solent and Harnham Down are prefaced by poetic inscriptions, from Philip Marston and Matthew Arnold respectively, but these shed little interpretative light.¹³ It is a great pity that so many unresolved questions surround these works, as musically they are of considerable interest. As Michael Vaillancourt has noted, they show a new structural compression, harmonic sophistication and subtle attention to orchestral sonority. Though the title ‘Impression’ hints at French influence, more striking in many respects is the anticipation in certain passages in The Solent of the kind of string-writing that the composer would a few years later develop with such distinction and originality in the Fantasia on a Theme by Thomas Tallis. Of course, landscape, and the natural world in general, had a long history of inspiring composers to textural and other kinds of musical innovation, and it is perhaps no surprise that a composer becoming increasingly restless in his search for a style both individual and somehow rooted in a wider national identity should turn to such subject matter.

There is an experimental quality at times in these New Forest works, and the composer’s broader restlessness during this period is reflected in the tally of works left incomplete, or discarded after a single performance, or (more commonly) subjected to multiple revisions. In the latter category is In the Fen Country, which has the distinction of being the composer’s earliest orchestral work to be allowed eventually to form part of his public oeuvre. It was composed initially in 1904, but then subjected to multiple revisions over the next few years, with a final retouching of the orchestration in 1935. The work shares with much of the composer’s music of this period a sometimes awkward combination of Wagnerian or Straussian chromaticism, fleeting echoes of Debussy, and fresher modal perspectives, and at times it relies too heavily on imitation or strained modulations to generate momentum from contemplative musical material. Nevertheless, it contains many compelling and beautiful passages and sustains its sense of musical direction overall. A glowing chordal passage heard at various points, in which high strings answer brass, strikingly anticipates A Pastoral Symphony of some fifteen years later. Although the opening cor anglais solo certainly suggests the kind of generic lonely shepherd heard as far back as Berlioz’s Symphonie Fantastique, the contour of the melody is also undoubtedly shaped by the English folksongs that Vaughan Williams had recently begun to collect, as Elsie Payne has shown.¹⁴

A sense of the wide, flat vistas of the Fens, where land, water and sky blur together, is conjured once again in the Norfolk Rhapsody No. 1, the only other orchestral work from the period before the end of 1907, when the composer began his studies with Ravel, to have survived Vaughan Williams’s severe self-criticism. Two further rhapsodies, the three together apparently intended eventually to form a Norfolk Symphony, with the second combining slow movement and scherzo, were eventually deemed unsatisfactory and not performed after 1912.¹⁵ The third rhapsody is now lost, but the second survives in a largely complete manuscript and has been recorded.¹⁶ All three rhapsodies drew on folksongs collected by the composer in East Anglia during 1905, and were designed in part to bring these tunes to a wider audience, though the composer also published vocal arrangements of a number of them. The rhapsodies are, in fact, the composer’s only orchestral works to present complete folk tunes, with the exception of Five Variants of ‘Dives and Lazarus’ and some folk dance arrangements; indeed, in his later career Vaughan Williams would rarely quote even fragments of specific songs, but rather evolved a melodic language shaped by generic features of English folksong or of certain families of tunes.

The original manuscript of the first rhapsody is unfortunately lost, and so we are unable to reconstruct with any certainty the work as it existed before its publication in 1925, though it seems likely that it was in essentially its final form by the time of a performance in Bournemouth in May 1914. The programme note for the 1906 premiere, however, does indicate that at this date the piece ended rousingly, rather than with the return to the quiet opening stasis that we hear today, and that it originally contained the stirring song ‘Ward the Pirate’, which in the third rhapsody formed the basis of a climactic final section, functioning as a cyclic recall of the opening number.¹⁷ The eventual withdrawal of the third rhapsody presumably played a part in the removal of ‘Ward the Pirate’ from the first. The songs that remain in the final version of the first rhapsody are memorable; this is especially true of the ‘The Captain’s Apprentice’, whose elegiac mood opens and closes the work, and to an extent pervades it, despite the boisterous energy of the second main melody, ‘On Board a 98’ (‘A Bold Young Sailor’ is also included). The desolate opening, in which lonely birdcalls keening across a coastal landscape are eventually joined by the more human voice of a solo viola, which weaves an improvisatory solo around the ‘The Captain’s Apprentice’, is truly haunting.¹⁸ The rhapsody does not entirely avoid the endemic pitfalls of works of this kind, especially in occasional awkward changes of gear between songs, but it is much more than a medley; Ian Bates has pointed to a number of subtleties of modal treatment in the work, and to various levels of symmetrical structure that suggest a more careful compositional process than is often associated with the term ‘rhapsody’.¹⁹ The second rhapsody also begins and ends quietly; slow outer sections, based on the tunes ‘Young Henry the Poacher’ and ‘All on Spurn Point’, frame a central scherzo that focuses on ‘The Saucy Bold Robber’. Though the third rhapsody is lost, reviews suggest that it was essentially a quick march with trio. Why Vaughan Williams withdrew all but the first rhapsody must remain a matter of speculation, though the second rhapsody does not seem to cohere as convincingly as the first (to this listener, at least).

Ravel and after

Vaughan Williams produced an impressive number of orchestral works in the period between 1902 and 1907, but he clearly remained deeply dissatisfied with his achievements (not only in the orchestral field, it should be noted), and it was at the end of 1907 that he travelled to Paris to begin his work with Maurice Ravel. As Byron Adams has discussed in Chapter 2 of this volume, despite a frustrating lack of concrete evidence as to the exact scope of his studies with Ravel, the time he spent with his younger French colleague was clearly transformative, at least judging by the compositional fruit of the next few years. Though this did not manifest itself immediately in the form of independent orchestral works, the substantial score (albeit for a 24-piece orchestra) that the composer wrote in 1909 for Aristophanes’ play The Wasps does show the impact of Ravel’s instruction in ‘how to orchestrate in points of colour rather than in lines’, as Vaughan Williams would later describe his teacher’s primary lesson.²⁰ The suite that Vaughan Williams later drew from this incidental music, scored for a rather larger ensemble than the original, went on to be a successful concert work, and at its publication in 1914 was in fact the first of his orchestral works to appear in print; the Overture proved particularly popular, and in 1925 it was one of the two works that the composer chose for his first venture into the recording studio.²¹

But 1910 would bring what were ultimately more important landmarks. The autumn of that year saw within just over a month the premieres of two crucial works: the Fantasia on a Theme by Thomas Tallis at the Three Choirs Festival in September, and A Sea Symphony at the Leeds Festival in October, both conducted by the composer. At the end of 1911 Vaughan Williams wrote in a letter to Ernest Farrar that the Fantasia ‘[is] the best thing I have done’,²² and most commentators have endorsed that judgement. Indeed, its stature remains impressive even in the light of the composer’s later accomplishments. The Fantasia is Vaughan Williams’s most widely performed and recorded major work, achieving in particular an international currency that has so far eluded most of his other music – though it was some time before the work took hold in this way and even in England it would not become a staple until the 1930s. It has been widely regarded as the first work in which Vaughan Williams fully realized the individual stylistic synthesis that would form the basis of his mature compositional output. He apparently discovered Tallis’s tune for Psalm 2 during work on The English Hymnal (1904–6), for which Vaughan Williams adapted it as ‘When Rising from the Bed of Death’. Nathaniel Lew has plausibly suggested that the idea of using the tune as the basis of a longer work may have originated in 1906, in connection with the Reigate pageant based on John Bunyan’s The Pilgrim’s Progress (the composer’s first engagement with the Bunyan work, which went on to become a lifelong fascination, of course).²³ This remarkable Phrygian-mode tune is certainly richly suggestive in terms of harmonic, rhythmic and other subtleties of construction, as a number of commentators have noted;²⁴ yet the process of revelation that Vaughan Williams drew from it nevertheless constitutes an extraordinary imaginative feat, for which the word ‘visionary’ is for once entirely free of hyperbole. Particularly striking is the manner in which a closed thematic model is daringly broken down to a condition of almost complete quietus, and then gradually reanimated, eventually culminating in a luminous climax. The climax itself, a passionate homorhythmic declamation, confirms the fact that harmonically the work is a profoundly original meditation on the power of unadorned triads: without a single appoggiatura, but instead juxtaposing triads riven by false relations that evoke sixteenth- and seventeenth-century English music, Vaughan Williams creates a climax as intense in its way as anything in Wagner or Tchaikovsky (see Ex. 4.2).

Ex. 4.2. Fantasia on a Theme by Thomas Tallis, bars Q.17–21.

The rhythmic unison of the climax is especially dramatic because of the array of multifaceted textures that has preceded it. Indeed, perhaps the most immediately striking (and most influential) feature of the work is its new and spacious approach to string sonority. The composer describes the ensemble as ‘double stringed orchestra with solo quartet’. Elgar’s Introduction and Allegro of 1904 had employed the concerto-grosso-like contrast of a quartet with a larger body of strings, but Vaughan Williams adds an intermediary second orchestra, essentially a double-quartet with double-bass added (the two orchestras are thereby of unequal size). The second orchestra is to be placed apart from the first ‘if possible’. These resources are exploited to the full, particularly to create spatially suggestive effects of light and shadow, with wide registral expanses evoking a vertical dimension that complements the recessed textural perspectives. This aspect of the work has understandably been linked by some writers to the experience of being within a Gothic cathedral such as that at Gloucester, where the work was premiered (though there is no direct evidence that the composer was thinking in these terms). Choral antiphony has also been suggested as a model, but the disparity in the size of the ensembles militates against this analogy, at least in terms of true double-choir music; the contrasts between different organ manuals offer a closer parallel. More precise architectural and spatial metaphors underpin the most recent re-examination of the work, in which Allan Atlas suggests that the structure of Gloucester Cathedral may have helped suggest a tension between two different proportional schemes in the Fantasia, between what Atlas identifies as a quarter-half-quarter (or 1:2:1) shape, and various manifestations of the so-called ‘Golden Section’ proportion.²⁵ Any spatial scheme has nonetheless to take into account (as does Atlas) the strong narrative thrust that develops during the work. Anthony Pople has conceived of this in part as an evolutionary historical account of musical style, invoking plainsong, folksong and organum before arriving at Tallis’s polyphony.²⁶ As Pople notes, this particular narrative weakens as the work progresses and Vaughan Williams’s own modern voice takes over; nevertheless, this is the composer’s most self-consciously historicizing work up to this point in his career. One earlier model that Pople essentially dismisses, however, is that of the one-movement Jacobean fantasy, as promoted during this period by the wealthy musical amateur Walter Willson Cobbett through his prize for chamber music: as Pople points out, Vaughan Williams violates a number of the requirements set out by Cobbett.²⁷ Both Atlas and Pople take into account important differences between the original version of the work and the published score of 1921; by the time the latter appeared the composer had subjected the Fantasia to at least two rounds of revision, cutting a total of thirty-three bars from the score, which included a second statement of Tallis’s theme at the end of the work where now there is only one.

Three symphonies

Though the Tallis Fantasia would ultimately eclipse it in fame, it was A Sea Symphony, for orchestra, chorus and soprano and baritone soloists, that established Vaughan Williams as a composer of large-scale works. It was first performed on 12 October 1910 – Vaughan Williams’s thirty-eighth birthday – at the Leeds Festival, conducted by the composer (later in the programme Rachmaninov played his Piano Concerto No. 2). Its long gestation began in 1903 and thus straddles several important junctures in Vaughan Williams’s early development. This is reflected in a disparate range of styles and influences, the latter including Brahms (Ein deutsches Requiem in particular), Parry, Stanford, Elgar, Wagner, Tchaikovsky and (to a lesser extent) Debussy and Ravel, along with folksong and hints of Tudor music. Though many aspects of its compositional chronology remain obscure, we do know that a short-score draft was complete by 1906; at this point the work contained an additional movement, ‘The Steersman’, placed between the Scherzo and the Finale.²⁸

Vaughan Williams had begun to set Walt Whitman (1819–92) in 1902 and would remain preoccupied with his work for the next decade, and to some degree for the rest of his life; Stanford and Charles Wood had already made notable Whitman settings before him, but it was Vaughan Williams in the years before the First World War who would engage most deeply with the American’s transcendent and universalist vision of the modern world.²⁹ In A Sea Symphony Whitman inspired a work of extraordinary originality and ambition. With a running time of around an hour and twenty minutes, it must surely have been the longest British symphony written to date. Its status as a true symphony has been disputed, and it is certainly a hybrid work in terms of genre, combining elements of symphony, oratorio and cantata. It is more fully choral than, say, Mahler’s symphonies with voices, in that the choir or soloists are heard virtually throughout; this necessarily dilutes its ability to pursue some more traditionally symphonic processes, particularly in developmental sections, and the form is sometimes episodic, especially in the massive finale. Yet the tonal and thematic strategies across the work ensure that it is much more than a loose succession of character pieces, and the outer movements in particular generate passages of gripping symphonic momentum. The overall tonal scheme of the four movements, essentially D–E–G–E♭, is unconventional, but hardly unprecedented against the background of Mahler’s ‘progressive tonality’.³⁰ And despite the stylistic disparities, there are moments of dazzling audacity. The exhortation of ‘Behold, the sea itself’ that launches the first movement, ‘A Song for All Seas, All Ships’, is one of these: its dramatic shift from B♭ minor to D major harmony at the word ‘sea’ conjures a visceral sense of space opening up before us, and surely constitutes one of the great opening gestures of musical history³¹ – a gesture intensified further at its second appearance by the unexpected addition of a C bass note under the D harmony (see Ex. 4.3). In the finale, ‘The Explorers’, the build-up to the prophetic moment of revelation, when the ‘Son of God shall come singing his songs’, is similarly electrifying; the rugged diatonic dissonances here are original and bracing, with the explosive urgency of the bass line pushing it ahead of the upper-voice harmonies, a kind of disjunction that might almost suggest the influence of early Stravinsky, were it not for the fact that it was composed before either The Firebird or Petrushka. The hushed and ambiguous ending, ostensibly in E♭ but with melodic emphasis on C and a low G in the bass undermining the stability of the tonality, is also striking.

Ex. 4.3. A Sea Symphony, ‘A Song for All Seas, All Ships’, bars 42–50.

The enormous last movement is certainly sui generis, and here Vaughan Williams’s reach exceeds his grasp (even if the listener may remain grateful that he reached that far); despite certain recurring thematic and tonal elements that provide landmarks in the soul’s cosmic journey of exploration, the structure remains relatively loose. The second

and third movements are more obviously in line with symphonic expectations: ‘On the Beach at Night, Alone’ is a slow ternary form, and the third movement, ‘The Waves’, is a scherzo. The third movement is the most impressionistic part of the symphony, replete with Debussy-like whole-tone passages among other elements. For the most part, however, the sea functions throughout the symphony as a spiritual metaphor rather than pictorial inspiration, reflecting the poet’s own perspective. The sea was a common theme in British music at this time, with obvious nationalistic resonance, but Whitman’s perspective is a universal one, reflecting his all-encompassing post-Christian sense of spiritual quest, his embrace of both the mystical and the mundane, and his visionary conception of global democracy. Such considerations, along with the poet’s highly original and fluid approach to metre and diction, clearly influenced Vaughan Williams’s choice of texts for the symphony; there is an almost palpable sense in a number of passages of an artist discovering the full extent of his powers for the first time, as Whitman’s exaltation becomes the composer’s own.

Though boldly innovative in a number of ways, A London Symphony is easier to place than its predecessor within the evolution of the symphonic genre since Beethoven, and it represents the culmination of Vaughan Williams’s development as an orchestral composer before World War I. According to Vaughan Williams the work began as sketches for a symphonic poem about London,³² but while it has important programmatic associations, these are subordinate to more traditional elements of musical coherence, including a unified tonal plan and cyclic returns of material in different movements. Nevertheless, one specific programmatic impetus was revealed in 1957, when, in correspondence about the symphony with Michael Kennedy, the composer remarked laconically: ‘For actual coda see end of Wells’s Tono-Bungay’.³³ This appears to refer to the final chapter of H. G. Wells’s novel, which traces a journey taken by the protagonist down the River Thames, from some way upstream to the open sea, a progress that is framed in both geographical and historical terms, and which Wells describes as the three movements of ‘a London symphony’. Tono-Bungay was first published in 1908; I have argued elsewhere that it was most likely the initial spur for the symphony, and surely influenced more than just its ending.³⁴ That said, it is not clear exactly when Vaughan Williams conceived of an orchestral work related to London, or made his initial sketches, though the symphony was apparently underway by mid-1911, and finished by the end of 1913.

The symphony made a strong impression at its premiere in March 1914, and confirmed for many Vaughan Williams’s position as the leading figure in the cohort of British composers who had emerged since 1900. Though a more traditional work than A Sea Symphony, it is scarcely less ambitious or original, and lasted almost an hour in the version performed in 1914. Its full impact, however, was initially stifled by the outbreak of war, and it would not be heard in London again until 1918. Between 1918 and 1920, when the work was first published, the composer made draconian cuts, amounting to roughly 25–30 per cent of the original in terms of playing time; additional, though less severe, surgery would be performed before a revised version of the score appeared in 1936.³⁵ Vaughan Williams apparently agreed with those critics of the original version who had, despite generally positive reactions, expressed concerns about its length and prolixity. Of the four movements the first remained virtually intact in the revisions, but the slow second movement, the scherzo (which lost an extended subsidiary section), the finale and the long Epilogue that flows out of the finale were all heavily cut.

But the bold ambition of the work is represented more by breadth and depth of musical complexity than by sheer length, and these dimensions remain impressive even in the revised versions. Vaughan Williams created in this work a Mahlerian, at times almost Ivesian, range of stylistic and social reference.³⁶ Though the composer’s own musical personality sounds strongly throughout, the clear imprint of Debussy, Wagner and even Stravinsky can he heard at various points. More significant in programmatic terms, however, are references to the urban street soundscape, including not only actual music – from modern ragtime inflections in the first movement and barrel-organ and harmonica in the scherzo, to the implicitly pre-modern lavender-seller’s cry of the second movement – but also ambient sound, as it were, in the form of hansom-cab jingles, the chimes of Big Ben, or what seem to be the metallic shrieks and rumbles of trams and trains. And although the detail is intimately particular at times, the perspective is nevertheless implicitly global, not just local or national. The London that Vaughan Williams portrays, as centre of the British empire, is the capital of the modern world, a dimension emphasized by its status as a major port, intimately connected with the boundless sea. Nature, humanity of all classes, the technological challenges of modernity, and relationships with a multi-layered past are brought together in a heady mix. But they are not simply juxtaposed: Vaughan Williams confronts apparently irreconcilable contrasts and conflicts, but also seeks to integrate and reconcile. In this respect the work stands directly in the tradition of the Beethovenian symphonic paradigm, and indeed the grinding semitone dissonances at the beginning of the allegro section of the first movement, which are then taken up again in the finale, seem to evoke directly the opening of the finale of Beethoven’s Ninth Symphony. A sense of strain, particularly in the last movement, is only fitting: if the symphony seems on the verge of collapsing under its own weight and self-contradiction at times, so did the city, nation and empire that it represents, riven as they were at this time by social and political turmoil (most notably widespread strikes and growing crisis in Ireland).

Programmatic interpretations of this kind must for the most part remain conjectural, since Vaughan Williams himself was ambivalent and inconsistent in shedding light on this aspect of the symphony (in common with most composers of programme music, it must be said). In 1925 he did tentatively acknowledge that the allegro of the first movement represented ‘the noise and hurry of London, with always underlying calm’, and that one might think of the slow movement in terms of Bloomsbury Square on a November afternoon, and of the Scherzo (subtitled ‘Nocturne’) as Westminster Embankment at night. He offered no such clues to the finale, however, or to the Epilogue into which the finale leads without a break. Nevertheless, the Epilogue, which is based on the slow introduction to the symphony, is clearly the ‘coda’ the composer mentions in reference to Tono-Bungay; the Wells connection confirms what is already strongly hinted at in the musical material itself, namely that both the beginning and the ending of the symphony evoke the Thames, and eventually the sea: the speed and turmoil of urban human modernity are thus framed by a mystical and implicitly timeless natural world. More broadly, Wells’s novel portrays a society in decay, and a vision of London (one common at the time) as a vast and uncontrollable – almost unrepresentable – cancer of modern society, a mindlessly churning cauldron of capitalism engulfing its environs and its people. Though Vaughan Williams’s view was surely not quite as bleak, there are many threatening shadows in the symphony, even moments suggesting nightmare, especially in the last movement and in a number of passages cut during the revision process. On the whole, the symphony darkens as it progresses, and the finale negates any notion of the kind of triumphal peroration beloved of many nineteenth-century symphonists, on the model of Beethoven’s Fifth and Ninth symphonies (though of course Brahms’s Fourth, Tchaikovsky’s Pathétique and Mahler’s Ninth all offered powerful precedents for such negation). The work is certainly much darker than Elgar’s portrait of London in his Cockaigne overture of 1901, which is for the most part by turns exuberant and tender.³⁷

Though the musical means by which Vaughan Williams expresses diversity and conflict are often original, the techniques by which he imposes coherence are for the most part well established in the symphonic tradition. Thus the interaction of tonal, modal and chromatic elements generates both dissonant conflicts, and the framework by which these are eventually absorbed into the overall G major tonality of the symphony.³⁸ E♭ emerges early on as a chromatic alteration of the sixth degree of G, in the form of a keening D–E♭ melodic figure in the slow introduction, and the interaction of scale-degrees 5 and 6, in a number of different tonal and modal contexts, looms large throughout the symphony. At the beginning of the allegro section of the first movement the D–E♭ semitone is verticalized, as a shrill chromatic progression of parallel triads in the upper parts clashes with a thunderous bass line, suggesting the threatening and inhuman dimension of modern city life (see Ex. 4.4).

Ex. 4.4. A London Symphony, first movement, bars 34–44, transition from slow introduction to Allegro.

The bifurcation of the texture here into aggressively independent layers constitutes one of the most conspicuously modernist elements of the score, with clear parallels in the music of Stravinsky and Ives, and a powerful representation of the colliding simultaneities of urban experience. Such heterogeneity takes on a more positive aspect with the populist high spirits of the second-group material, which eventually brings the first movement to a rousing conclusion – a rarity in this composer’s symphonies. But although this mood will return in parts of the deft Scherzo (especially the trio, which echoes Petrushka), the remainder of the symphony is mostly much darker. Twilit melancholy dominates the ternary second movement, and after the Scherzo itself burns out to end in ominous darkness, the finale embarks on an epic and ultimately tragic journey.³⁹ It is tempting to interpret this whole movement, not just the Epilogue into which it eventually leads, in terms of Wells’s Thames progress, which travels through London’s history as well as its geography; a solemn march gives way to more turbulent material, the hectic striving of which fails to achieve closure, and eventually collapses into a return of the opening of the first movement allegro, before giving way to the Epilogue. The niente ending echoes that of A Sea Symphony, and despite its attempt to absorb the symphony’s residual tonal conflicts, it offers more dissolution than resolution. As was suggested above, if the finale seems to strain too hard, and to fall short of the expectations and responsibilities generated by the preceding movements, such qualified failure is not inappropriate as a metaphor for the city and empire that inspired it – it may indeed have been inevitable, given the scale of the task that Vaughan Williams set himself in this work. The composer nevertheless retained an attachment to the symphony throughout his later life, and he conducted it more often than any of his other works.

Vaughan Williams made his first cuts to A London Symphony while on leave from active service in 1918 (he had enlisted in 1914). Though not obviously incapacitated, the composer was deeply marked by the war, not least through the loss of younger friends – most notably George Butterworth, to whom A London Symphony was dedicated upon publication. It is tempting to relate the removal of some of the work’s most apparently subjective passages to the impact of war, representing perhaps a search for some objective, even stoical distance from a pre-war world irrevocably lost. Whether or not this was the case, such objectivity and emotional restraint is a striking feature of the composer’s next symphony, made all the more telling there by isolated outbursts of anguish. A Pastoral Symphony was begun in 1916 and completed in 1921; it was the first major work that the composer completed after demobilization, and its unveiling in London in January 1922 was Vaughan Williams’s first major premiere of the post-war period.

A Pastoral Symphony may lay claim to have been the composer’s most misunderstood work, at least during his lifetime. While its reception cannot detain us here, insensitive critical reactions early on, including from some of the composer’s friends, did much to forge the persistent image, which has only in recent years finally weakened its hold, of Vaughan Williams as a purveyor of insular pastoral nostalgia and meandering rhapsody. The composer himself was perhaps partly to blame (though his reticence was understandable), in that he failed to make clear until much later that the landscape evoked in the work was primarily that of wartime northern France, not the idyllic English vistas assumed by many commentators.⁴⁰ Yet a degree of musical incomprehension was understandable, not least on account of the genre expectations engaged by the work’s title. Despite its imaginative orchestration – the deftly shifting and melding planes of sonority indicate a new and entirely individual absorption of the example of Debussy and Ravel – the symphony largely eschews traditional symphonic rhetoric. It is subdued in dynamic level for the most part, and employs almost unremittingly slow or slowish tempi; even the Scherzo has a relatively heavy tread for a movement of this kind.

In terms of deeper levels of construction, however, A Pastoral Symphony is in fact the composer’s most rigorously disciplined symphony up to this point in his career. Several recent commentators have demonstrated the tightly knit motivic and formal coherence that underlies the apparently rhapsodic surface of the music; out of melodic material steeped in English folksong yet involving no direct quotation, the composer weaves a subtly compelling and evolving symphonic narrative.⁴¹ Like its metropolitan predecessor, the symphony takes G as its fundamental point of departure (though in this case it will not return to end there); through subtle modal interplay, and the drawing-out of the implications of parallel triadic harmonization – a debt to the ‘Nuages’ movement of Debussy’s orchestral Nocturnes is evident at the opening of the symphony – the composer introduces tonal tensions that impel the music forward (see Ex. 4.5).

Ex. 4.5. A Pastoral Symphony, first movement, bars 1–12.

A sense of teleological trajectory, in both the melodic and tonal dimensions, is pursued across the course of the work, even if it goes underground at times, as it were. An important element in this is the inconclusive ending of the first movement, which leaves unresolved the tension between G and C♯ opened up early on in the movement, a tension bound up with the prominent role played by A major as a subsidiary tonal centre in the rotational sonata form.⁴² In the finale it is D major that emerges as the strongest point of tonal arrival, and a belated resolution of earlier tensions. This coincides with a melodic fulfilment, in that for the first time in the symphony a fully fledged, rounded melodic statement is allowed to reach convincing closure. And yet this sense of arrival and completion is ultimately frustrated, in that the wordless and unaccompanied soprano solo that began the movement returns unchanged at the end, leaving the symphony floating ambiguously on an unharmonized A. The soprano solo is one of two elements in the symphony that seem to reach most obviously out towards a more programmatic level of meaning. It is not difficult to hear in wordless female keening a lament for the war dead; yet there is a remote and ritualized quality to the expression that might suggest instead a natural world oblivious to human suffering. The cadenza for natural trumpet in the second movement has a much clearer significance: it was suggested to the composer during the war by hearing an army bugler practise, and its realistic but gentle evocation of wrong notes has an almost unbearable poignancy – no surprise, perhaps, that the symphony flares into one of its rare moments of protesting anguish immediately following the cadenza. Michael Kennedy is surely right to argue that this symphony constitutes Vaughan Williams’s ‘war requiem’.⁴³

In June 1922, a few months before his fiftieth birthday, Vaughan Williams made his first trip to the United States, having been invited to conduct the American premiere of A Pastoral Symphony at the Norfolk Music Festival in Connecticut, where Sibelius had premiered his tone poem The Oceanides in 1912. The invitation was probably made on the strength of A London Symphony, which was introduced to America in New York at the end of 1920 and was heard with some frequency there during the decade that followed. By the mid-1920s, Vaughan Williams had established himself as an orchestral composer of international standing. Through more than two decades of prolific, yet intensely self-critical, composition (and revision), he had succeeded in making the orchestra, that heavily freighted flagship of nineteenth-century musical culture, a flexible vehicle for his own profoundly original musical visions, including three extraordinarily contrasting symphonies. What no one could have predicted in 1922, however, is that this was just the beginning: that Vaughan Williams would go on to write six more symphonies that would help ensure the continued vitality and relevance of the genre for a tumultuous century and beyond.

English song and the marketplace: royalty ballad and art song c. 1900

In 1902 Vaughan Williams wrote to his cousin Ralph Wedgwood:

Vaughan Williams’s typically self-deprecating attitude towards this publishing deal belies what was in fact a strongly held belief in what he was doing and the music he was creating. Yet he is defiantly unashamed – as he was to remain throughout his career – of simplicity and popularity. The extract also says a good deal about contemporary attitudes to song publishing. For decades there had been considerable press coverage of what were seen as the iniquities of the ‘royalty ballad’ system, whereby publishers paid singers a royalty every time they performed a certain song.¹² At the turn of the century, the issue was still current. In 1901 The Times reported on an address by Frank Sawyer to the Incorporated Society of Musicians. Sawyer talked about song publishers and singers, explaining that

Often royalties were paid not only to singers but also to composers, who if they did not sell their song outright, received a royalty for each copy sold, as in Vaughan Williams’s deal with The Vocalist.¹⁴ The Vocalist had been launched in April 1902 and was aimed at ‘all those who consider themselves in the category of singers, whether they be elementary or advanced, amateur or professional’ as well as those ‘interested in musical art, and. . . “fond of singing”’.¹⁵ Readers were promised ‘four good new songs’ in each issue.¹⁶ The first issue included Vaughan Williams’s ‘Linden Lea’, a setting of words by William Barnes described as ‘a Dorset folksong’. This was Vaughan Williams’s first published work and he was to contribute several articles and songs to the early issues of the magazine.¹⁷

In the late nineteenth and early twentieth centuries, songs were a part of the programme in every type of public concert, as well as sung in every kind of home – either to family and friends around the ubiquitous piano or, in high-class circles, to invited audiences in lavish music-rooms. Publishers promoted their songs through Ballad Concerts, such as Boosey’s popular series at St James’s Hall, which had started in the 1860s. Study of surviving programmes shows that Ballad Concerts could in fact provide a wide range of vocal music, including songs by canonical composers such as Schubert or Schumann as well as an array of ballads.¹⁸ Nevertheless, the attitude in the musical press towards both the Ballad Concerts and the ‘royalty ballad’ remained consistently critical. Vaughan Williams joined in the criticism in his ‘Sermon to Vocalists’, published by The Vocalist in 1902. In this article he addresses amateur vocalists and implores them to use their intelligence when selecting songs, and to choose the sincere rather than the false. He describes the royalty ballad as possessing ‘neither melody nor rhythm’, and asks singers whether they really imagine that it is ‘so called because it is patronized by Royalty?’¹⁹

What other songs were being published and heard in London’s concert halls at the beginning of the twentieth century? Although by the turn of the century Vaughan Williams’s teachers Charles Villiers Stanford (1852–1924) and Hubert Parry (1848–1918) were involved primarily with high-profile operatic, orchestral and choral works, they continued to publish songs. Stanford’s An Irish Idyll in Six Miniatures, Op. 77, a collection of songs setting poetry by Moira O’Neill, appeared in 1901; and in 1902 Novello published the fifth set of Parry’s English Lyrics, settings of various British authors, including Shakespeare, Scott and Julian Sturgis.²⁰ Another leading figure was Arthur Somervell (1863–1937), who, like Vaughan Williams, had studied at Cambridge and the Royal College of Music with Parry and Stanford. On 7 March 1901, at St James’s Hall, he gave a concert of his own music that included the popular Cycle of Songs from Maud (Alfred, Lord Tennyson) and Love in Spring Time, which might aptly be termed an anthology, setting as it does poems by Tennyson, Dante Gabriel Rossetti and Charles Kingsley.²¹

Other composers active at this time were primarily songwriters. One of the best known today is Roger Quilter (1877–1953), remembered for his Tennyson setting ‘Now Sleeps the Crimson Petal’ (1904) and his many settings of sixteenth- and seventeenth-century English poets. In the summer of 1901 Quilter returned from studies in Frankfurt to accompany a performance of his Four Songs of the Sea (to his own words), at the Crystal Palace. This work was also published, as his Op. 1, in the same year. Quilter was a good friend of one of the most popular songwriters of the 1890s, Maude Valérie White (1855–1937). At the start of the century White had moved to Taormina in Sicily, largely for the sake of her health, but returned regularly to England to organize concerts of her songs. On 20 May 1903, for example, she gave a concert of her ‘tasteful and well-written songs’ at St James’s Hall.²² The selection of songs, typical of White’s diversity, included settings of German, French and English lyrics as well as arrangements of Sicilian tunes. The British poets represented included Burns, Swinburne, Herrick, Shelley and both the Brownings. White’s friend Liza Lehmann (1862–1918) was another important songwriter of the time. Her song-cycle for four voices and piano, In a Persian Garden (1896), which sets parts of Omar Khayyám’s Rubáiyát, in the then fashionable translation by Edward FitzGerald, remained popular far into the twentieth century. On 4 January 1902 it was performed at one of the ‘Popular Concerts’ at St James’s Hall, a series that, despite its name, showcased the ‘highbrow’ end of the song spectrum.²³ Lehmann’s songs, like those of White, were also heard at the lighter Ballad Concerts, and she herself clearly divided her compositions into ambitious or serious work, such as her Tennyson cycle In Memoriam (1899), and lighter, more popular and financially rewarding pieces.

The changing reception of Lehmann, White and their music highlights the difficulties of clearly defining the difference between the royalty or drawing-room ballad and the ‘art song’.²⁴ Both these songwriters heard their work performed at a wide range of public and private venues. In their early careers both were regarded as raising the level of the English song and, particularly in the case of White, choosing a better class of lyric to set.²⁵ In 1903 Edwin Evans published an article on Lehmann in his series ‘Modern British Music’ for The Musical Standard, which showcased composers ‘who actually are modern, breathe the modern spirit and write modern music’²⁶ (Vaughan Williams was featured in another article in the series). In these articles, Evans had decided ‘to avoid the vortex of the English song world’ but made an exception for Lehmann, feeling that there were several reasons, such as her introduction of the song-cycle, for writing about her ‘apart from the artistic excellence of her work’.²⁷ Nevertheless, later in their careers both White and Lehmann came to be regarded only as ballad composers, and after their deaths their contribution to British song was almost entirely overlooked and forgotten. The reasons behind this neglect are complex; the fact that they were both women, and therefore not expected to be capable of producing ‘great’ music, undoubtedly played a significant part. But just as important is the fact that they were primarily songwriters, concentrating on a genre so consistently regarded as insignificant. The two issues of gender and genre are of course not unrelated: the prominence and success of female songwriters at the turn of the century undoubtedly contributed to the later downgrading of British song as a genre, despite the indisputable power and beauty of songs by composers as diverse as Parry, Quilter, White and Vaughan Williams.

The early songs: from Herrick to Housman

It is worth highlighting that Vaughan Williams was nearly 30 years old when ‘Linden Lea’ appeared in The Vocalist and he first stepped into this world of song production. ‘Linden Lea’ was, of course, not the first song he had written: there are several songs surviving in manuscript or published later that pre-date what was to become his best-selling song.²⁸ But what do we know about his experience of song in his first three decades? The answer is surprisingly little. Both Ursula Vaughan Williams’s biography of the composer and Vaughan Williams’s own brief ‘Musical Autobiography’ are curiously reticent about his early exposure to song.²⁹ Unfortunately, very few letters to or from Vaughan Williams survive from before the late 1890s, and even then they tend to be unforthcoming about his musical life and experiences.³⁰

As a child at Leith Hill Place, Vaughan Williams learned the piano and harmony from his mother’s sister, Sophy, who taught him theory from The Child’s Introduction to Thorough Bass in conversations of a Fortnight, between a Mother and her Daughter of Ten Years Old, published in 1819.³¹ When Ursula Vaughan Williams mentions this book, she gives only the first six words of the title, omitting the interesting fact that it was clearly aimed at educating female children. Neither Ursula nor Vaughan Williams himself mention singing as part of the family music-making, other than hymns on Sundays.³² Vaughan Williams’s childhood compositions included operas for his toy theatre, and it seems unlikely that he would have engaged in this genre with no experience of vocal music other than hymns. It is also unlikely that a family clearly interested in music would not have owned and enjoyed songs. Ursula Vaughan Williams records that the family went to the Three Choirs Festivals,³³ and here they would certainly have heard songs as well as orchestral and choral works.

At the age of seven Vaughan Williams started learning the violin, and he continued to be involved in music at both his preparatory school and then at Charterhouse, a fairly progressive school as far as teaching music to its male pupils was concerned. In his account of his childhood and schooldays Vaughan Williams emphasizes his exposure to canonical keyboard and chamber music, which included playing piano-duet arrangements of music by Handel, Haydn, Mozart and Schubert with his brother and sister. In his preference for the violin over the piano, he was, of course, moving away from the dangerously female world of piano-playing, an association reinforced by his first textbook and the largely female household in which he grew up. He does recall, however, that he had also been exposed to a more popular, less highbrow kind of music. Remembering an early composition for piano trio, he writes that he ‘must have got the theme from one of the French or Belgian imitators of Franck whose salon music was popular in those days’.³⁴ We can only guess, but it seems likely that he would have heard this popular salon music in the homes of his music-loving relatives, alongside songs by composers such as Lehmann or White. Vaughan Williams was in his late seventies when he recorded these reminiscences, and it is only to be expected that he would have a selective memory of the musical experiences of his youth, ignoring the kind of music that had fallen so far from acceptance in the post-1945 musical world. But it is important not to overlook what must have been a significant aspect of Vaughan Williams’s early musical experience. It seems unlikely that the composer of his earliest surviving songs, such as the Herrick setting ‘To Daffodils’ (1895) or the Tennyson setting ‘Claribel’ (c. 1896), was not aware of White’s Herrick or Tennyson settings, not just for her choice of poets but also for her compelling, drawn-out melodies, finely judged sense of rhythmic propulsion, and subtly independent piano accompaniments that often avoid expected cadences or harmonies.

The songs through which Vaughan Williams first reached a wider audience included those published in The Vocalist between 1902 and 1905, his two Dante Gabriel Rossetti song-cycles, The House of Life (1904) and Willow-Wood (1909) and the Songs of Travel, in which he set poems from the collection of the same name by Robert Louis Stevenson.³⁵ The Vocalist songs were two Christina Rossetti settings (‘If I Were a Queen’ and ‘Boy Johnny’); two Tennyson settings (‘Tears, Idle Tears’ and ‘The Splendour Falls’); ‘A Cradle Song’ to a lyric by Coleridge; a setting of Robert Louis Stevenson’s poem ‘Whither Must I Wander?’;³⁶ three settings of lyrics by William Barnes; and two arrangements of German folksongs.

The Barnes settings and German arrangements were written before Vaughan Williams had started collecting folksongs himself, and they represent a long-standing British and German tradition of folk-like settings and arrangements of folksongs from a variety of different cultures and countries.³⁷ Vaughan Williams’s Tennyson settings reflect some of the approaches taken by other composers when choosing lyrics by this much-set poet. ‘Tears, Idle Tears’ (1903) is a sombre and substantial work that stands out among Vaughan Williams’s early songs for its length. The weighty piano opening, and indeed the overall atmosphere and low tessitura of the song, recall Liza Lehmann’s Tennyson cycle In Memoriam.³⁸ Vaughan Williams turned to Christina Rossetti for several early songs for solo voice – the mournful ‘When I Am Dead, My Dearest’ (1903), for example – as well as in works for multiple voices, including ‘Sound Sleep’ (SSA, 1903) and ‘Rest’ (SSATB unaccompanied, c. 1904–5).

Vaughan Williams’s ambitious song-cycles setting sonnets by Dante Gabriel Rossetti deserve to be better known. Willow-Wood was first performed in 1903 in a version for baritone and piano; an orchestral version was also made, and in 1909 the composer added a wordless female chorus, but following the first performance of the revised version he seems to have withdrawn the cycle. The House of Life appeared in 1904; one song from the cycle, ‘Silent Noon’, was performed and published before the rest, and went on to become a recital staple. The popularity of this evocative setting has not, unfortunately, led many singers to perform the cycle in its entirety. Michael Kennedy suggests that Vaughan Williams’s interest in Pre-Raphaelite poetry was inspired by the example of Claude Debussy, who in 1887–8 had set Dante Gabriel Rossetti’s poem ‘The Blessed Damozel’, as La Damoiselle Elue.³⁹ But there was a group of composers closer to home, and to Vaughan Williams’s age, who were also drawn to such poetry. These were the men who came to be known as ‘the Frankfurt group’, because they had all spent time studying in that city: Norman O’Neill (1875–1934), H. Balfour Gardiner (1877–1950), Cyril Scott (1879–1970), Percy Grainger (1882–1961) and Quilter. Grainger felt that what set the group apart from other British composers was ‘an excessive emotionality (& particularly a tragic or sentimental or wistful or pathetic emotionality)’ and claimed: ‘Perhaps it might be true to say we were all of us PRERAFAELITE [sic] composers’.⁴⁰ Yet while the Frankfurt group may have helped spur Vaughan Williams’s interest in Rossetti, he took a very different approach: in The House of Life he expressed not an intense or excessive degree of emotion, but a measured and contemplative depiction of emotion that was perhaps closer in spirit and a certain ideal of beauty to the work of the Pre-Raphaelite brotherhood, as well as to the approach of earlier British songwriters. Some songs of the cycle, for example ‘Heart’s Haven’, clearly echo the expressive expansiveness and harmonic colourings of White’s song-writing. It is possibly this association with Victorian song-writing that has discouraged later singers from performing the whole cycle.

An unashamed sense of beauty can also be found in Songs of Travel, a song-cycle with a complex performance and publication history. The first eight songs of what is now regarded as the complete cycle were performed at Bechstein Hall on 2 December 1904. But Vaughan Williams’s publishers in this case, Boosey, clearly did not feel that the complete cycle was commercially viable and insisted on publishing it instead as two collections, with songs 1, 3 and 8 as Book i (1905), and songs 2, 4, 5 and 6 as Book ii (1907). The seventh song, ‘Whither Must I Wander?’, had been published earlier, in The Vocalist of 1902. The ninth and final song, thematically related to others in the cycle, was discovered posthumously.⁴¹

Songs of Travel, both Stevenson’s poems and Vaughan Williams’s songs, are usually taken as representing a particularly masculine breath of fresh air in their depiction of love in relation to the outdoor life of a wanderer. This quality is clearly a relief to commentators trying to find an escape from what they see as the hothouse femininity of a work such as The House of Life.⁴² Kennedy, for example, writes of Stevenson’s ‘virile open-air verses’, while Banfield describes the work as ‘fertile in its Romantic wayfaring images’.⁴³ The first performance was sung by Walter Creighton and the cycle was taken up by Harry Plunket Greene, but it is perhaps surprising to note that other early performances were given by women. The cycle was, for example, sung ‘very charmingly’ by Norah Dawney at the opening concert of the 1905 Hovingham Music Festival,⁴⁴ and four years later the well-known Australian contralto Ada Crossley included ‘The Roadside Fire’ in a recital at Bechstein Hall.⁴⁵

Like most songwriters Vaughan Williams relied on particular performers to promote his songs. In these early years, the bass-baritone Jack Francis Harford (1867–1948) and baritone James Campbell McInnes (1874–1945) gave important performances of Vaughan Williams’s songs; neither was particularly well known, but both built reputations for promoting the work of young English composers. Harford gave the first performance of ‘Tears, Idle Tears’, which was dedicated to him, at St James’s Hall in 1903, and sang Vaughan Williams’s two arrangements of fifteenth-century French songs, ‘Jean Renaud’ and ‘L’amour de moy’, at the same hall in 1905.⁴⁶ In the early years of his career he was compared to Harry Plunket Greene, a singer who was also to become associated with Vaughan Williams’s songs.⁴⁷ McInnes made his London debut in 1899.⁴⁸ ‘Boy Johnny’ (1902) was dedicated to him (although the first London performance was given by Harford), and McInnes gave the first performance of Willow-Wood in 1903. Both men were studying at the Royal College of Music when Vaughan Williams returned there in 1895, and their association with the composer may have started at that time. Much better known was the tenor Gervase Elwes (1866–1921), the other singer who regularly performed Vaughan Williams’s early songs. Elwes is particularly associated with the song-cycle On Wenlock Edge, giving the first performance, with the Schwiller string quartet and pianist Frederick Kiddle, at Aeolian Hall on 15 November 1909. Elwes’s wife recalled that ‘Gervase was delighted at being asked . . . to give the first performance of the song-cycle, in which he discovered rare beauties’.⁴⁹

Written after orchestration lessons with Ravel in Paris and famously described by the composer himself as ‘a song cycle with several atmospheric effects’, On Wenlock Edge, a setting of six poems from A. E. Housman’s popular collection A Shropshire Lad, demonstrates a move away from the song world of Parry or White. The cycle has attracted attention from critics ever since its first performance. Edwin Evans and Ernest Newman had a notorious disagreement about it in the pages of The Musical Times in 1918. Evans felt that Vaughan Williams had realized ‘the inner qualities of the poems’ and summed up his achievement by saying:

Newman retaliated by saying of Vaughan Williams’s settings that ‘they do not mate happily with the prosody of the poems’, pointing to the importance of rhythm in this respect and claiming: ‘We shall not get our great English song-writer until we get some one who can take up the rhythm of a poem into his music without distorting or mutilating it.’⁵¹ He goes on to accuse Vaughan Williams of Wagnerian word-painting rather than conveying the sentiment at the heart of the poems, and in the song ‘Bredon Hill’ finds ‘another of Dr. Vaughan Williams’s disastrous attempts to imitate folk-song’. Later writers have been more inclined to agree with Evans, and in this work Vaughan Williams clearly found a distinctive and compelling voice.

It is curious that Vaughan Williams does not seem to have been associated with the upper-class music-making that played such an important role in the lives and careers of other British composers, such as Elgar, Grainger, Lehmann, Parry, Quilter or White.⁵² It is perhaps not surprising that this is passed over by Michael Kennedy and Ursula Vaughan Williams, who were writing in the 1960s when biographers tended to stress the professionalism of musicians and to distance them from the tainted world of the Victorian and Edwardian drawing-room and salon culture, by then perceived as frivolous and unimportant. But there is very little mention of Vaughan Williams’s participation in private musical worlds in his own letters or in the diaries, letters, or memoirs of others. He does not seem to have been involved in the private concerts or music-making of patrons and music-lovers such as Muriel Draper, Lady de Grey, Frank Schuster, or Edgar and Leonora Speyer, although both his class and his ability to play the viola would have made him very welcome at such gatherings, which in turn would have provided a significant venue for his music and a space in which to meet other composers and musicians.

Vaughan Williams did take part in musical gatherings at the Westminster home of his friend Lucy Broadwood (1858–1929), with whom he was, of course, also connected through their shared interest in folksong.⁵³ Another circle of private music-making with which he is known to have been involved is represented by parties held in Percy Grainger’s rooms, at which Grainger’s friends would try out part-songs,⁵⁴ and the musical ‘jamborees’ instigated by Gervase Elwes’s brother-in law Everard Feilding, and held at the Elweses’ Mayfair house in the winter of 1904.⁵⁵ Other participants at the ‘jamborees’ included Balfour Gardiner,⁵⁶ Grainger, W. Y. Hurlstone, Poldowski (Lady Dean Paul), Quilter and Scott, as well as Vaughan Williams’s cousin Diana Montgomery-Massingberd and her sister-in-law Margaret Massingberd.

The Massingberds provided Vaughan Williams with other significant musical connections and opportunities. Diana was a member of the Magpie Madrigal Society, an amateur choir, organized by Lionel Benson, whose members included many prominent music-loving members of the upper classes. At one point Parry was its president.⁵⁷ The society sang contemporary music as well as English madrigals from the sixteenth and seventeenth centuries, and in 1902, at St James’s Hall, gave the first performance of Vaughan Williams’s Christina Rossetti setting, ‘Rest’, for unaccompanied voices in five parts. His lively madrigal ‘Ring Out Your Bells’ was also dedicated to the society, who doubtless sang it at one of their private gatherings.⁵⁸ Margaret Massingberd, married to a cousin of Vaughan Williams, was one of three musical sisters immortalized in the painting The Home Quartet (Mrs. Vernon Lushington and Her Children) (1882–3) by the Pre-Raphaelite artist Arthur Hughes. She ran the East Lincolnshire Musical Festival at Spilsby, near the Massingberd family home, Gunby Hall. Both McInnes and Elwes took part in the 1901 festival,⁵⁹ and Vaughan Williams wrote his Christina Rossetti setting ‘Sound Sleep’, a trio for female voices, as a test piece for the 1903 festival.⁶⁰

Choral works and later songs

In general, much more is known about Vaughan Williams’s involvement with choirs and choral singing than is known about his exposure to solo-song culture. Particularly at the start of his career, he was constantly involved in organizing and conducting choirs, many of which would have tried out and performed his choral works. As a student at Cambridge in the 1890s he conducted a small choral society, and in his first job, as organist at St Barnabas Church in South London from 1895, he was responsible for training the choir, and also founded a small choral society.⁶¹ When the Leith Hill Festival was founded in 1905 by his sister Meggie and Lady Farrer, they turned to Vaughan Williams to conduct the Festival choirs, and during the First World War he formed a choir from the ranks of his fellow soldiers. From 1903 he had been a member of the Bach Choir, and in 1921 he became its conductor, having previously directed both the Palestrina Society (from 1912) and the Handel Society (from 1919).

Vaughan Williams’s close connection to a variety of choral groups is reflected in the assurance of his part-songs and other choral music. Even a comparatively early work such as ‘Rest’ (1902) demonstrates his awareness of effective a cappella writing. The sense of concord and comradeship represented by a choir, whether of well-to-do Londoners or Lincolnshire villagers, was to have a lasting impact on Vaughan Williams’s attitudes to the value of music-making. And this perhaps explains his avoidance of the ‘At Home’ aristocratic music culture. Like his close friend Gustav Holst, Vaughan Williams was a lifelong political radical, embracing socialist ideals from an early age. He may not have played the viola at high-society concerts, but he was happy to play at the Passmore Edwards Settlement, where Holst was musical director from 1904.⁶² The Settlement, established in Bloomsbury by the novelist and campaigner Mary Ward, promoted ‘equalisation in society’ by offering opportunities for local people to learn skills and enjoy a variety of entertainment.⁶³

Vaughan Williams’s drive to write music for the voice had been firmly established by the end of the first decade of the twentieth century, and he continued to explore a variety of approaches to vocal writing in the years that followed. The rethinking of traditional accompaniment for song found in On Wenlock Edge, in which a string quartet is added to the standard pairing of voice and piano, was continued in works such as the Four Hymns (1920) for tenor, piano and viola, and the strikingly effective Along the Field (1927),⁶⁴ eight Housman settings for voice and violin. In 1940 he would write: ‘Why should the voice always be accompanied by the pianoforte? There seem to me great possibilities in voices and instruments in combination.’⁶⁵

Four Hymns was dedicated to and first performed by tenor Steuart Wilson, who was to become the principal exponent of Vaughan Williams’s songs after the tragic early death of Gervase Elwes in 1921.⁶⁶ In 1925, for example, Wilson gave an all-Vaughan Williams recital at Aeolian Hall which included On Wenlock Edge, Four Hymns, Merciless Beauty (a 1922 setting of three Chaucer roundels for voice and string trio or piano), and several brand new works: Three Poems by Walt Whitman, Three Songs from Shakespeare and Four Poems by Fredegond Shove. The reviewer for The Musical Times was not impressed with the Whitman settings but found the Shakespeare songs ‘admirable’, and remarked that the Chaucer settings ‘are surely among the best of modern English songs’.⁶⁷ Much has been written about Vaughan Williams’s engagement with Whitman, realized most ambitiously in A Sea Symphony. Byron Adams has perceptively written of ‘Vaughan Williams’s search for a masculine but visionary poet [landing] him, metaphorically speaking, in the arms of Walt Whitman’.⁶⁸ But this 1925 collection of three songs is often overlooked by both critics and singers. ‘Nocturne’ is a sombre and poignant setting of the poem ‘Whispers of Heavenly Death’, underpinned by an ominous and relentless ground in the piano. A solemn equilibrium seems to be reached in the second song, ‘A Clear Midnight’, while the final song, ‘Joy, Shipmate, Joy’, is a reflection of a still sober and perhaps somewhat contemplative joy.

After several decades during which most of Vaughan Williams’s work with the voice was on a larger scale (in opera, for example), he returned to solo song in the 1950s, most notably with the Ten Blake Songs (1958) for voice and oboe – perhaps the composer’s most radical rethinking of song accompaniment – and the posthumously published Four Last Songs (1960), which set the poetry of his second wife, Ursula Vaughan Williams.

To claim that Vaughan Williams was not a ‘great songwriter’, as Day did in 1961, begs the question – what exactly is meant by the term ‘great songwriter’? The history of British song, and the place of a variety of song genres within different musical cultures in the decades around 1900, is extremely rich, but also complex, and frequently misunderstood or misremembered, bound up as it is with a historical distrust of a feminized genre associated with feminized musical space. But Vaughan Williams certainly has a central place in the world of British song. He was a persistent and important songwriter who followed in the footsteps of composers who had gone before him and then, through the slow but determined maturing of a powerfully individual musical language, brought together his love of voice, and what it represented for him in terms of the human spirit, with a relentless belief in simple beauty, creating works of a profound and enduring impact.

Crafting a choral tradition: 1906 to 1914

Over the course of eight years, Vaughan Williams exploited much of what the British choral tradition he inherited had to offer. For most of the nineteenth century, festivals were the focus of music-making in areas outside of London; they were also the location of many choral premieres.² Repertoire at these festivals consisted of choral works, predominantly oratorios and cantatas in the mould of Handel’s Messiah and Mendelssohn’s Elijah, though in the last decades of the nineteenth century the importance of producing excellent instrumental works, such as symphonies, increased.³ But Vaughan Williams’s inheritance (and that of his contemporaries) was greater than the few festivals still known and celebrated today: the last quarter of the nineteenth century witnessed a revolution in choral music, and music festivals and choral societies were founded at an amazing rate. The music for festivals and choral societies was further disseminated to more parts of British society through the introduction of sight-singing methods, such as Tonic Sol-fa, which aided the middle and working classes in both reading and appreciating music.⁴ Such democratizing trends were general tenets of the so-called English Musical Renaissance, which promoted better training for British musicians – for professionals at institutions such as the Royal College of Music (hereafter RCM), and for amateurs at ordinary schools because of the inclusion of singing classes – and strengthened the infrastructure in which both professionals and amateurs could make music. This in turn led to more potential commissions for compositions. Vocal music publishing, both for the newly founded festivals and for smaller choral societies, also greatly increased during this time.⁵

Between 1906 and 1914, seven festivals performed Vaughan Williams’s compositions (see Table 6.1). These included smaller festivals like Hovingham; younger ones like that at Cardiff; the venerable major festivals of Leeds and the Three Choirs (which rotated among Worcester, Hereford and Gloucester); and the Music League Festival, meant to celebrate new foreign and domestic compositions.⁶

Table 6.1 Music by Vaughan Williams at major British choral festivals: 1906–14

P = premiere of composition

C = conducted by Vaughan Williams

(G) = Gloucester

(H) = Hereford

(W) = Worcester

How important was a premiere at a choral festival to the young Vaughan Williams? The money given for a premiere, while not excessively generous, was certainly not inconsequential. Vaughan Williams was paid 10 guineas for the premiere of Toward the Unknown Region at the Leeds Musical Festival in 1907; three years later, the Festival Management Committee considered paying him between 25 and 30 guineas for the premiere of A Sea Symphony.⁷ For a virtually unknown composer (as Vaughan Williams was at this point), the offer of a premiere was provisional, based on whether or not the Committee and the festival conductor liked the completed work; fortunately for Vaughan Williams he had an ally at Leeds, as the festival conductor in both 1907 and 1910 was his old teacher from the RCM, Charles Villiers Stanford.⁸ Once a festival accepted a work from a composer, though, it meant additional royalties (the Leeds Musical Festival bought 400 copies of the vocal score of A Sea Symphony from Breitkopf & Härtel), and a great deal of publicity (the festival placed advertisements in 27 provincial newspapers, 7 London papers and specialist journals like The Musical Times).⁹ Advance columns in most of the papers gave the entire festival programme.¹⁰

Vaughan Williams’s offerings to these festivals were a mixture of standard fare and innovative ideas. A Sea Symphony and Five Mystical Songs were such works, in that they blended genres. A Sea Symphony is in practice less a symphony than a secular cantata cast into the four-movement form of a symphony, with a structure that is almost narrative at points. Vaughan Williams refuses to promote a nationalistic ideal or even a triumphant one within this work: it ends with the listener cast out into the infinite, and without any sense of harmonic assurance.¹¹ Similarly, Five Mystical Songs – an orchestrated song-cycle on sacred texts with ad libitum choral accompaniment in four of the songs – was not the developed religious composition expected in a venue such as Worcester Cathedral. The genre itself was new to the musical festival. By contrast, Toward the Unknown Region and the cantata Willow-Wood were, in spite of their unusual texts (Walt Whitman and Dante Gabriel Rossetti respectively), exactly what festival audiences of the time expected for shorter choral works: non-narrative compositions that balanced orchestral textures with easy choral declamation. Each sets its text carefully, with the orchestral accompaniment emphasizing structural divisions. In Toward the Unknown Region, the care Vaughan Williams took with the text is particularly evident. The first four of Whitman’s five verses are separate syntactical units, with Vaughan Williams providing a sense of closure for each before moving to the next. At the end of the first verse, for instance, the composer returns to the three motives which opened the piece, though in reverse order (a horn fanfare, pizzicato lower strings, and a falling pentatonic gesture, now in a stripped-down orchestration), before presenting a new, arch-like chromatic motive to introduce the second verse. To end the piece with a greater sense of drama, Vaughan Williams avoids such an obvious structural break between the fourth and the fifth verses. After establishing continued motion into the fifth verse, Vaughan Williams sets this segment as an exercise in triumphant rhetoric, including several climaxes, a grand pause before a homorhythmic choral declaration, and a copious use of exclamatory brass and percussion.

Vaughan Williams’s choral experience before the First World War also included competition festivals. Whereas the charity and civic festivals described above brought music to middle-class individuals, the competition festivals in general were open to members of all classes, and frequently aimed paternalistically towards the working classes.¹² Such was the case with the Leith Hill Musical Competition, founded in 1905 in part by Vaughan Williams’s sister, Margaret Vaughan Williams (she was Leith Hill’s first Honorary Secretary). A feature of this festival was a massed concert of the choral forces at the conclusion of the competition. In 1905 Vaughan Williams became the conductor of this massed concert, a position he would hold (aside from his years of service during the First World War) until 1953.

Vaughan Williams worked assiduously and benevolently for Leith Hill. While he refused payment for his services to the festival, the strength of his commitment and the energy he gave to the institution were extraordinary.¹³ He was for many years part of the Musical Committee that decided not only the repertoire for the massed concert, but also the competition test pieces. His duties as conductor, especially in the early years of the festival, included rearranging music as needed to suit the amateur voices, occasionally improving translations of the texts (as he did in 1907 with the Novello scores to J. S. Bach’s Cantata No. 140, ‘Wachet auf’), and even reassigning trombone parts to the strings to save money by slimming the ensemble (he saved the Festival £5 10s. this way in 1909).¹⁴ Such work with Leith Hill gave Vaughan Williams both experience with amateur singers (which improved his own composition for such groups) and an outlet to explore what he came increasingly to call ‘good music’ from the choral tradition, such as works by Handel and Bach.

While the early part of Vaughan Williams’s career saw his reputation as a composer grow through the performance of his works at the major festivals, his immersion in the realm of the competition festival provides evidence of a desire to improve standards in music-making at a local, community level. Thus he exploited the traditional infrastructure of British music while at the same time attempting, through his pedagogical work, to preserve and enhance it.

Celebration and consolidation: 1920–38

If the period before the First World War was a time during which Vaughan Williams used the musical festival to introduce himself to British audiences, during the interwar years he consolidated his position vis-à-vis the festivals to become one of the most important working composers in Great Britain. When the festivals resumed after the war, they brought Vaughan Williams’s music into a new prominence (see Table 6.2). For instance, the only interwar year without a Vaughan Williams composition at the Three Choirs Festival was 1922. Some years featured several (three in 1931, 1933, 1934, 1936 and 1938, four in 1935), with Vaughan Williams frequently conducting his own compositions. And if the Three Choirs Festival still did not enlist Vaughan Williams to compose a major new choral work during the era, smaller commissions did come from his pen: Four Hymns for tenor and strings (originally written for Worcester in 1914, but not performed there until the first Three Choirs Festival after World War I, in October 1920, a few months after it had received its premiere in Cardiff) and Two Hymn-Tune Preludes for Orchestra (1936). In addition, the festivals at Leeds and Norwich regularly featured his music, and Norwich commissioned a major choral work, Five Tudor Portraits, for the 1936 festival.

Table 6.2 Music by Vaughan Williams at major British choral festivals: 1920–38 [abbreviations as in Table 6.1]

Aside from the festival performances, commissions for commemorations also signalled Vaughan Williams’s arrival as a composer. Sancta Civitas was premiered in 1926 at the so-called ‘Heather Festival’ in Oxford, which was a commemoration of the 300th anniversary of the establishment of a chair of music there; the concerts featured both new and old music, with a decidedly English bent for the former.¹⁵ Dona Nobis Pacem was premiered in 1936 as part of a series of concerts to celebrate the centenary of the Huddersfield Choral Society, while Serenade to Music was written in 1938 for the concert celebrating Henry Wood’s golden jubilee (fiftieth anniversary) as a conductor.

This consolidation of Vaughan Williams’s position at the forefront of the musical festival and within British celebratory music meant that there was a shift in his compositional output. His works became more deliberately cultivated for such performance settings, and began using somewhat more ‘traditional’ genres and texts. His festival compositions during this time included an oratorio (Sancta Civitas) and a secular cantata (Five Tudor Portraits); and he set major English literary figures, such as John Skelton in Five Tudor Portraits and William Shakespeare in Serenade to Music (which is largely reworked from Act v of The Merchant of Venice). Although his compositions increasingly acquired a more modernist bent, their musical language remained accessible to the choral singer.¹⁶ While his earlier, experimental works continued to be programmed at the musical festivals – A Sea Symphony, in particular, was a perennial favourite, and performed more often at interwar festivals than any of his other compositions – and even his non-choral symphonies found a space within them, it was these more traditionalist compositions of the interwar years that usually garnered the most critical attention in contemporary reviews.

Such a nod to tradition may also be seen in Vaughan Williams’s continued work with Leith Hill. After the First World War, the competition’s importance to the composer grew, especially because of the didactic ‘massing’ of the choirs. Increasingly, Vaughan Williams took advantage of this venue to present even more ‘good music’ to the people of Dorking and the surrounding area. One highlight was the festival’s 1931 anniversary concert, which featured a performance of Bach’s St Matthew Passion, with a force of 800 performers. The performance was planned as a special event, beyond the competitions themselves, and subsequent performances of this work (1938 and 1942–58, with Vaughan Williams conducting them until his death) showed the educational and taste-making (or taste-raising) component that Vaughan Williams believed should be a part of Leith Hill. In the 1930s, Vaughan Williams introduced many other large and important works to the massed choirs, including excerpts from Brahms’s Ein deutsches Requiem (1933), Elgar’s The Dream of Gerontius (1934), Dvořák’s Te Deum (1938) and his own Dona Nobis Pacem (1938).¹⁷

Choral music at this time also provided Vaughan Williams with an opportunity to respond to the horrors of the First World War. In this, he was not unique: many creative veterans directly or indirectly shared their battlefield experiences within a more artistic space in the decades that followed the war.¹⁸ Eric Saylor notes that the difference between Vaughan Williams and many of his contemporaries was that instead of dwelling entirely on the ravages of war, Vaughan Williams always left room for hope, ‘if not in this world or at this time, then in a time and place yet to come’.¹⁹ Within Sancta Civitas, this hope may be seen in the quick juxtaposition of the rhetoric for the Angel of the Lord descending on his white horse (the forces of righteousness; rehearsal figures 9–17) versus the armies of earthly kings (the forces of despair; figures 17–21). Vaughan Williams uses the same primary rhythmic motive in each segment: a grouping of dotted crotchet – quaver – crotchet. Furthermore, each segment is also introduced by the baritone soloist in a quasi-recitative style. But the first segment is rhetorically more insistent and triumphant, as a homorhythmic choir narrates the Angel’s appearance, skill and power; this section is further emphasized by fanfare-like flourishes in the brass. The armies of the worldly kings, in contrast, have no coherent organization. The chorus here only supports the baritone recitation, instead of driving the entire segment, as it does for the Angel. The chorus also does not unify itself into a homorhythmic structure, instead attempting imitative points, and even these Vaughan Williams destabilizes. In the first point of imitation the bass, tenor and alto parts enter at a rhythmic interval of five beats (beginning one bar after rehearsal figure 19). The soprano entrance (three bars later) is at an interval of four beats, and Vaughan Williams further undermines the metre by shifting from 5/4 to 3/4. The second point of imitation is even more disorganized: the five beats between the bass and tenor parts contract to three between the tenors and altos and the altos and sopranos, and the entry loses the energetic force of the 5/4 metre after only one bar, flattened instead quickly into 3/4.

The battle of good and evil in Sancta Civitas has a clear outcome of victory for the righteous. Not so for Dona Nobis Pacem, which is often seen as a presage of the Second World War. In the fifth movement of this work there is also an angel, but this is the Angel of Death, and the choir responds in a pleading fortissimo explosion in the second bar after rehearsal figure 30, before the soprano soloist cries out for peace over almost her entire range: from a high A♭ down to a middle C. There is hope here, too, but it is a hope that is distinctly qualified: the soprano solo that begins the work, chromatically asking God to grant peace using the verbal formula from the Mass Ordinary (‘Agnus Dei, qui tollis peccata mundi, dona nobis pacem’), becomes the mostly diatonic final plea at the end of the work (five bars after rehearsal figure 44). The ending also features a 27-bar structural decrescendo, removing first all of the instruments from the texture, then the chorus, finally leaving only the soprano soloist. While the choir articulates a solid C major chord for almost four bars, it is so low within the range of all of the voices that at a ppp dynamic, it is almost inaudible. Further, the ending is not reached from a dominant–tonic resolution, with that progression’s traditional connotations of finality. Instead, Vaughan Williams presents a third relationship, in a final harmonic progression of vi⁷–I in C major. Thus when the soprano articulates her final pitch of E, it does not bring a sense of completion or resolution: peace may be present, and it may be possible, but it is a tentative, fragile thing. While Sancta Civitas drew its vision of war’s horror and resolution mostly from biblical texts, Dona Nobis Pacem saw Vaughan Williams using an admixture of texts, including poems by Walt Whitman, a parliamentary speech by John Bright and texts from the Mass Ordinary, the Old Testament and the Gospel of St Luke. One contemporary critic thought this juxtaposition to be ‘peculiar’,²⁰ but the Whitman and Bright texts are used throughout the composition to focus on human tragedy, while the biblical ones engage with its reconciliation in peace.

Both these works became staples of the musical festivals in the years that followed. Five Tudor Portraits, with its premiere at the Norwich Musical Festival (1936), should have had a similar reception, but was not presented at the Three Choirs during the composer’s lifetime. It was popular enough, though, that in the 1938–9 season, British choral societies programmed it no fewer than nine times – the same number of performances as there were that year of Elgar’s Dream of Gerontius.²¹ In Five Tudor Portraits, Vaughan Williams presented a large choral work incorporating elements of his folksong style – something absent for the most part from both Sancta Civitas and Dona Nobis Pacem. The folksong style works better within Five Tudor Portraits, perhaps, because of its subject: what Jennifer Oates has called a ‘tonal genre-picture’ of John Skelton’s early sixteenth-century English poetry, in contrast to the biblical Armageddon of Sancta Civitas or the prayers of Dona Nobis Pacem.²² Indications for the soloist to emphasize a nasal quality (just after rehearsal figure 24), as well as spatial effects such as the placement of a few solo choral voices at the backs of their sections (figure 28), add some twists to the modal melodies and occasional rhetorical riffs, like the sea-shanty sound at figure 17. A genial roughness and modern tonal fluidity, with movements sometimes ending far from where they began, differentiates the Skelton settings from the ravishing Serenade to Music, which followed two years later. The Serenade is a tonally self-contained work, beginning and ending in D major, and prominently rearticulating motives and rhetorical devices (such as passages featuring the solo violin) from the beginning of the work to its end. A cinematic rhetoric inhabits this composition, whether it be fanfare-like figures for an evocation of Diana, the Huntress between rehearsal letters H and K, or moving to much more chromatic material when discussing those who do not appreciate music (letter M and following). The work has a clearly audible structure, and is sweet enough that it purportedly brought Sergei Rachmaninov to tears at the first performance.²³

Indeed, all of the compositions written for festivals or celebrations that Vaughan Williams composed during the interwar period are accessible, both to choral singers and to audiences. As he ascended to become the most important living composer in Britain, Vaughan Williams took the promotion of a choral heritage for British music extremely seriously. He did so with the understanding that he would not be considered a revolutionary, and that critics might even value his work less on account of that: ‘I think, sometimes, that I ought not to try to do the greatest thing on earth, which no fellow will understand, but to use my skill, such as I have, for doing useful work. E.g. things for Div:2 kind of people to sing & enjoy.’²⁴ The choral institutions were only too happy to accept such compositions from him, and responded through occasional commissions, by inviting him to conduct, and by drawing on his compositions to emphasize the important festal moments in interwar British music and culture.

Eminence and pre-eminence: 1939–58

The period from 1939 to 1958 saw Vaughan Williams become the acknowledged master of both festivals and choirs. Though the major festivals ceased for the duration of the war in 1939, not to return until 1946, Vaughan Williams’s music and choral activities took on a greater importance than ever before. By 1943, as the tide of the war was turning in favour of the Allies, the BBC commissioned him to compose a celebratory piece that would become Thanksgiving for Victory, a composition for soprano, full SATB chorus, children’s chorus, orchestra, organ and speaker. The work was recorded to be ready when Allied victory was assured. The execution of the premiere left a great deal to be desired. The composition was effectively hidden from public hearing, as the BBC did not announce or advertise the performance, and presented it at 9.30 a.m. on 13 May 1945. Yet it was Vaughan Williams, not another composer, whom the BBC had approached for this task.²⁵ As Edward Lockspeiser noted in his review of the work for Music & Letters, ‘Vaughan Williams was the right composer to choose [for this commission], not so much because of his acknowledged place among English composers, but because no one would have been so sure to avoid any suggestion of rhetorical pompousness.’²⁶ The competitions at Leith Hill also ceased during the war. Vaughan Williams, though, ensured that music – and good music – continued, to provide recreation for those soldiers, refugees, evacuees and townspeople in the area around Dorking. The massed choir concert continued each year, and in 1942 Vaughan Williams began conducting Bach’s St Matthew Passion annually.

In the years after the war, Leith Hill was a good friend to Vaughan Williams, and he to it. He continued his duties as director of the massed choirs there, finding them to be amenable to his pet projects, including a 1947 performance of Bach’s Mass in B minor in English. To commemorate the composer in 1949, Leith Hill organized a concert by the London Symphony Orchestra for a ‘birthday present’ performance of A London Symphony, the Sixth Symphony, Fantasia on ‘Greensleeves’ and Five Variants of ‘Dives and Lazarus’.²⁷ Vaughan Williams officially retired from conducting the festival in 1953; as a tribute, every concert in the 1954 festival included one of his compositions.

When the larger music festivals resumed after the war, Vaughan Williams’s music was again a central feature (see Table 6.3). Each year from 1946 until his death in 1958, the Three Choirs Festival included at least one major work by him, and in most years included several. Between 1946 and 1951, such festival programming focused on the performance of one of the symphonies (Nos. 3, 5 and 6); after 1951, the major works included both vocal and instrumental compositions. And with 1954 came the landmark of Vaughan Williams’s first and only major choral commission from the Three Choirs, for the Christmas cantata Hodie.

Table 6.3 Music by Vaughan Williams at major British choral festivals: 1946–58 [abbreviations as in Table 6.1]

Hodie was premiered on 8 September 1954 in Worcester Cathedral. The review in The Musical Times noted that it ‘was found to be easy on the ear and quickly acceptable’.²⁸ The work has had its detractors, including Donald Mitchell, who found it ‘grossly over-praised’, primitive and even simply ‘downright unacceptable’; but James Day found the work to be wholly approachable, even almost familiar because it is ‘childlike, not childish, for Hodie is a wide-eyed “once-upon-a-time” child’s view of the Nativity told with all the wit and understanding of what lies behind the story, even if he does not believe it literally’.²⁹ As a Christmas cantata, there is no need for the work to be dramatic. Instead, Vaughan Williams concentrates on the evocative, and this he does with an easy fluency. The texts chosen are a mélange of the sacred (passages from the Gospels of Luke, Matthew and John together with snippets of the Mass and the Christmas Vespers, as well as hymns by William Drummond, Martin Luther, John Milton and Ursula Vaughan Williams) and the profane (poetry of William Ballet, Thomas Hardy, George Herbert and Ursula Vaughan Williams). As Eric Saylor notes, ‘Hodie . . . follows the textual layout of Bach’s [St Matthew] Passion by alternating between scripture and meditative poetry’.³⁰ The assignment of much of the narration to a boys’ choir singing in unison underscores this alternation, and the Bach connection is reinforced by the inclusion of four-part chorales at two points in the work (even though the style of the chorales is unmistakably that of Vaughan Williams).

The chorales and the narration, however, pale in comparison to the stand-alone movements, such as ‘The March of the Three Kings’. Like most of the movements within Hodie, this is compelling and wholly evocative of its subject. The text (from a poem by Ursula Vaughan Williams) has two separate subjects: the three kings from ‘kingdoms secret and far’ seeking Christ, and the ‘star of morning’, the sign of Christ himself. Vaughan Williams alternates two basic styles of music to capture the shifting moods of the text. The three kings march into and out of the texture over an ostinato of three crotchets (D, C, A), which is metrically dissonant with the 4/4 metre of the passage. The scoring includes a good deal of percussion, which accompanies a brass melody and fanfare that is mostly in D Aeolian, with an overlying pentatonic figure. When the first strain of the star music is heard (four bars following rehearsal figure 2), Vaughan Williams removes the percussion, creating instead an organ-like texture out of a choir of woodwinds, horns and tuba. He also shifts from Aeolian modality to a freer, more chromatic scale that generates cross-relations in the harmony,³¹ and carefully structures the exchange of solo and choral writing to provide a natural climax within the movement. The initial narration of the three kings’ march is given to tenors and basses alone. When each of the three kings in turn explains his gift (rehearsal figure 6 to four bars after figure 10), there is a response by the full chorus. The gift solos are built up from the lowest voice, baritone (gold), through the tenor (frankincense) and finally to the soprano (myrrh). Only at the climax of the movement (figure 12) do the three soloists sing together, and they do so to an immediate response by the choir. The vocalists reach to the highest points of their ranges in this exultant moment (the solo soprano, for instance, hits a high A in the bar after figure 13). The entire movement closes with a return to the instrumental ostinato, presumably as the three kings march home.

By far the largest festival in this last period of Vaughan Williams’s life was the Festival of Britain in 1951, a post-war multi-disciplinary event meant to promote British culture and raise morale (in addition to the arts, design was featured prominently). The Serenade to Music was heard at the ceremonial opening concert at the Royal Festival Hall, on 3 May, alongside compositions by Handel, Arne, Parry, Purcell and Elgar, making it the only piece on the programme by a living composer. A few days later (6 May), the Schools’ Music Association of Great Britain premiered the choral work The Sons of Light, to a text by Ursula Vaughan Williams, with 1,100 children’s voices accompanied by the London Philharmonic Orchestra.

If the period between the wars was one of consolidation for Vaughan Williams, where he settled into his role as one of the most eminent British composers of his generation, in the years between the onset of the Second World War and his death he became the eminent composer of his generation. This, to be sure, led to his being castigated by some younger composers and critics – those with something to prove – and because of the narrative of ‘progressive revolution’ required by the modernist enterprise, these individuals were often suspicious of compositions and composers that forwarded discernable didactic aims. But it was precisely these didactic aims – to promote good music, to write music that was at once modern and comprehensible by the amateur performer and the casual listener, to compose as ‘befits an Englishman and a democrat’ – that make these compositions by Vaughan Williams so valuable to the festival and choral tradition, in their own time, and today.

Folksong arrangements

Vaughan Williams’s arrangements of English folk music take many forms. They are scored for small orchestra, chamber ensemble, military and brass band, solo piano, chorus (mostly mixed but also women’s and men’s) and solo voice (usually accompanied by piano); are intended for home, school, concert and church use; and are included as incidental, dancing, occasionally even diegetic, music for plays, masques, pageants, ballets and operas. So numerous are his settings, indeed, that a final reckoning is virtually impossible, especially as the more ambitious arrangements (like those found in the suite English Folk Songs (1923) and other borderline ‘free compositions’) typically quote tunes piecemeal or as a kind of contrapuntal quodlibet. Arrangements of individual folksongs (i.e. arrangements identified as such and not intended as part of a larger musical design) are somewhat easier to tabulate, though even here we need to differentiate between folksongs (i.e. tunes with mostly secular texts), folk carols (tunes with religious texts) and folk dances (tunes without texts). Collating these, we arrive at 47 separate publications containing 260 distinct arrangements (at least 12 more are unpublished). Settings of Scottish, Welsh and Irish as well as continental folksongs and carols also appear in these and other sources, adding a further 31 individual arrangements to this total.⁶

There were of course long-term compositional benefits to be gained from this work. Crafting so many different arrangements in so many different formats had the potential to bring folksong within reach of a wide segment of the population, and thus deepen public appreciation of native musical traditions. This, it was hoped, would help solve the perennial problem of generating an audience for the English composer who had long been passed over in favour of continental musicians. (A cornerstone of this strategy, pursued by Cecil Sharp and enthusiastically supported by Vaughan Williams, was to introduce folksongs to the populace at an early age by including folksongs in the school curriculum.)⁷ But social and humanitarian concerns were a possibly more important motivation, for folksong was held up as a cultural artifact common to the experience of all English men and women; its dissemination throughout society as a whole would thus serve as a reminder of a shared cultural heritage and help forge connections between social classes. It would also transfer to a new, literate class of performers long-standing oral traditions of making music ‘for yourself’.

It was an ambitious agenda, especially since not everyone shared the composer’s belief in folksong’s contemporary relevance. The ingrained prejudice against folksong as antiquated and extraneous to the modern world was strong, and countering it required careful presentation. One approach was to focus primarily on those songs that seemed relevant by virtue of their recent collection. Vaughan Williams occasionally arranged folksongs that he found in antiquarian sources like William Chappell’s Popular Music of the Olden Time (1853–9), but by far the great majority of his settings are of songs collected by Cecil Sharp, Lucy Broadwood, H. E. D. Hammond and others, himself included, active in the contemporaneous Folk Revival. A second, and more important, strategy was to select the most attractive folksongs for publication – those, as he put it, that possessed ‘beauty and vitality’⁸ – and clothe them in equally attractive arrangements. ‘Beauty and vitality’ are subjective values, and while Vaughan Williams dropped occasional hints about what made one folksong better than another,⁹ he refrained from identifying a specific method for arranging them. It was essential to focus on the melody – the arranger must be ‘in the grip of the tune’ so that the setting ‘flows naturally from it’, he wrote¹⁰ – but beyond that his own tendency was to pursue a wide range of compositional solutions, tailoring his arrangements to the abilities of those who would probably perform them. As a result, simple piano and choral arrangements, derived from parlour-music and popular choral traditions, stand alongside more elaborately ‘worked’ settings that draw on art-song and choral part-song repertories. The separation between styles is never absolute, however, as even the simplest arrangements show the traces of the skilled composer.

‘I Will Give My Love An Apple’, from Folk Songs for Schools (1912), for unison singing and piano accompaniment, offers a case in point (see Ex. 7.1). Vaughan Williams always reserved his ‘simplest’ manner for children, on whose musical tastes and preferences much of the hope for the future evidently depended, and the setting is clearly designed with their inexperience in mind. The tune in the vocal line is doubled throughout by the pianist’s right hand, while the steady three- (occasionally four-) part texture, with its unvarying crotchet rhythm, provides further encouragement. The Aeolian melody, with its occasional tricky skips, might seem to represent a challenge for children, but Vaughan Williams’s strict modal diatonicism and careful placement of each chord lend it an inevitable air. Carrying this off successfully is harder than it looks and is in fact rooted in part-writing techniques taken from the ‘learned’ tradition. The many chord voicings and inversions, the careful control of suspensions and other diatonic dissonances, the well-conceived bass that moves mostly by step and usually in contrary motion to the melody – these sustain a sense of forward motion that splendidly enhances the tune. Contributing to the overall effect are the unexpected move to the F (VI) chord in bar 15, the late (and once-only) employment of which lends extra weight to the end of the strophe, and the careful integration of the tune’s frequent quaver-quaver-crotchet rhythm throughout the accompaniment. Such concern for the overall architecture of the setting demonstrates the degree of ‘composerly’ skill that Vaughan Williams was prepared to bring to even his most modest folksong settings.

Ex. 7.1. ‘I Will Give My Love An Apple’.

The settings for adult vocal groups, by contrast, run the gamut from simple to more complex in tone and construction. To some extent, the level of difficulty or elaboration is a function of the specific choral subgenre drawn on in each case. Thus a dozen settings for unaccompanied men’s voices (TTBB), largely homophonic and with the melody principally in the 1st Tenors, invite comparison with the eighteenth-century glee and especially (given the folksong material on which they are based) the early nineteenth-century German Männerchöre repertory. ‘The Jolly Ploughboy’ (1908), ‘The Ploughman’ (1934) and the old English air ‘The Farmer’s Boy’ (1921) are exemplary. Other TTBB arrangements, however, combine this ‘direct’ style with the varied textures and more intricate part-writing of the contemporaneous part-song. Thus ‘Bushes and Briars’ (1908) and ‘Dives and Lazarus’ (1942, from Nine Carols for Male Voices) – to take two widely separated publications – pass the tune engagingly between different voices and use overlapping suspensions to create piquant harmonies that express the text. The striking polychords at the final cadence of the former, in particular – parallel ⁶₄ chords sounding against a dissonant pedal – convey the despair of the final verse while also briefly anticipating the bitonal effects of the composer’s music of the 1920s. Timbral contrast between texted soloist and wordless choir, first essayed in ‘The Winter Is Gone’ (1912) and incorporated into the Fantasia on Christmas Carols of the same year, becomes an especially favoured device, and contributes much to the appeal of celebrated arrangements like ‘The Turtle Dove’ (1919) and the Scottish ‘Loch Lomond’ (1921).

Elaboration is especially pronounced in the twenty-six settings whose scoring for unaccompanied mixed voices (SATB and its variants) draws even more firmly on the part-song tradition. Shifting textures and thoroughgoing counterpoint are the very lifeblood of arrangements like ‘A Farmer’s Son So Sweet’ (1926) and the SATB versions of ‘Bushes and Briars’ and ‘The Turtle Dove’ (both 1924), as well as of a number of non-English settings like the Manx ‘Mannin Veen’ (1913) and the Scottish ‘Alister McAlpine’s Lament’ (1912) and ‘Ca’ the Yowes’ (1922). The timbral innovations of this last, in particular – wordless female voices ‘distanced’ from the texted lower voices by means of their high register and parallel motion – resemble those found in Vaughan Williams’s purely original compositions from the period. Similar effects inform the Five English Folk Songs (1913), probably the composer’s most famous folksong settings, in any medium, and still frequently heard today. The vocal virtuosity demanded by these arrangements is unsurpassed in his output (Michael Kennedy surmises that they were composed for a choral competition).¹¹ Frequent vocal divisi results in as many as six different parts (in ‘The Spring Time of the Year’) while rapid-fire voice exchange (‘The Dark-Eyed Sailor’, ‘Just as the Tide Was Flowing’) creates a dizzying effect. In ‘The Lover’s Ghost’, folksong material combines with techniques drawn from the sixteenth-century madrigal, a distant forebear of the part-song. Textural contrasts and points of imitation punctuate the structure while, in verse 3, competing melodic fragments of the tune are presented in both ‘real’ and ‘tonal’ contexts (and in original and inverted forms) to create Elizabethan false relations.

Other mixed-voice settings display a similar thematic integration. The ‘derived’ counter-melody of the 1924 ‘Bushes and Briars’ increasingly dominates the setting as it proceeds, culminating in the anguished clash of descending lines in the final verse. The rocking tertian progressions of the whimsical SATB ‘An Acre of Land’ (1934), themselves drawn from the skipping thirds of the melody, inform the tonal plan, which ambles no less whimsically between tonic and relative minor. Even the opening modal mixture (E♯ in the soprano vs. E♮ in the tenor) that ‘summarizes’ the tune in the celebrated SATB ‘Greensleeves’ (1945) turns out on closer inspection to encapsulate the harmonic conflict between tonal and modal forms of the dominant and ultimately between the setting’s principal keys of F♯ and A. Similar links between modal inflection and tonal argument are found in the Serenade to Music (1938) and the Sixth Symphony (1944–7) from the same period.

No obvious stylistic divide between subgenres emerges when it comes to the ninety-five settings for solo voice and piano accompaniment. The dominance of the late Victorian royalty ballad and parlour song, already a problem for those who would establish an art-song tradition in England, was keenly felt by those seeking to disseminate folksongs through arrangements for the domestic market. For a high-minded composer like Vaughan Williams, contemptuous of the commercial music industry¹² but cognizant of its popularizing potential, compromise was inevitable. Not that his arrangements merely duplicate the stereotyped arpeggiations and simplistic harmonies of the parlour ballad (a few of the twelve French and German folksong settings from 1902–3 come close, but are saved by the tasteful manner in which the formulae are applied). Rather, his tendency is to combine Victorian elements with the more sophisticated techniques of the solo art song. Sometimes the competing styles are kept separate – as in ‘The Lost Lady Found’ and ‘As I Walked Out’, Nos. 4 and 5 of Folk Songs from the Eastern Counties (1908), where uniformly simple triads and textures in one setting give way to melting diatonic dissonances and an elegantly shifting formal arrangement in the next. More typically, the two styles are combined in the same setting, as with ‘Tarry Trowsers’ (No. 2 from the same collection), in which popular ‘oompah’ rhythms mix with Bachian counterpoint, and ‘The Lincolnshire Farmer’ (No. 12), where broken-chord vamps are clothed in Debussian harmony. ‘The Saucy Bold Robber’ (No. 10), likewise, begins with simple crotchet rhythms in the accompaniment, only to pick up speed in later verses and conclude with chromatic runs worthy of Liszt.

Folk Songs from Sussex (1912), comprising fourteen arrangements of folksongs collected by W. P. Merrick, offers a stronger dose of art song. The continental styles that were mostly hinted at in the 1908 collection are more prominent here; in some cases, they form the core of the setting. The atmospheric effects and passionately descending parallel seventh chords of ‘Low Down in the Broom’ (No. 2) are deeply indebted to Debussy, for example, while Wagner stalks the wildly chromatic passages of the carol ‘Come All You Worthy Christians’ (No. 9). Fauré, too, permeates ‘The Seeds of Love’ (No. 11, with violin ad lib.), where lush harmonies, richly arpeggiated, reveal a side of the composer glimpsed virtually nowhere else. By no means are these and other settings from the collection pale imitations of their models, however: integrating continental styles with the strong diatonic lines of folksong, they embody the hard-won eclecticism of Vaughan Williams’s first maturity. ‘O Who Is That That Raps at My Window’ (No. 5) virtually summarizes this in its tremolo effects, static polychords and shifting pentatonic pitch collections – elements that contribute to a dramatic intensity not unlike that of On Wenlock Edge (see Ex. 7.2). The gapped head-motive, a Vaughan Williams fingerprint here drawn from the first phrase of the tune and presented in a variety of timbral and harmonic contexts, anchors the setting formally, while the motive’s rhapsodic extensions in bars 10 and 14–16 anticipate those of The Lark Ascending, begun two years later. The non-functional juxtaposition of triads, observable beneath the surface activity in bars 10–15, is another mark of the mature style; their return during the dramatic final verse, forte and spaced out over as many as four octaves, is in the manner of the Fantasia on a Theme by Thomas Tallis and other works that reserve purely triadic harmony for climactic moments.

Ex. 7.2. ‘O Who Is That That Raps at My Window’, bars 1–16.

Whatever pressure he felt to conform to Edwardian song culture, ‘O Who Is That That Raps at My Window’ shows that Vaughan Williams believed that there was room for innovation so long as it was properly balanced by an awareness of the needs and expectations of amateurs.¹³ Thus we find the same mixture of novelty and convention, elaboration and simplification, throughout all his later arrangements for voice and piano – in the Eight Traditional English Carols (1919) and Twelve Traditional Carols from Herefordshire (1920), notable for their frequent modal mixture, as well as in Folk Songs from Newfoundland (1934) and Nine English Folk Songs from the Southern Appalachian Mountains (c. 1938; published 1967). Here, as before, idiosyncratic and ‘uncomplicated’ settings stand side by side, while more than a few combine elements of both. Reduced textures – slighter chords, single-line accompaniments, etc. – increasingly characterize the last two publications, but this is less a concession to routine than a reflection of the relative austerity of Vaughan Williams’s interwar manner. Like the Shove and Whitman settings of the mid-1920s, the reduction of means results in concentration of material, as pithy head-motives and rhythmic figures, even the occasional counter-melody, distil the essence of the vocal line. An outwardly simple arrangement like ‘The Ploughman’, from Six English Folk-Songs (1935), packs all this and more into its sixteen bars, making it a small masterpiece of structural integration.

After 1945, Vaughan Williams turned almost exclusively to choral settings. The choice is partly explained by the fact that twenty-two of the thirty-seven arrangements from this period were written for Folk Songs of the Four Seasons (1949) and The First Nowell (1958), two large-scale choral works based around the idea of the folksong ‘anthology’. The retrospective aura surrounding these works – reinforced by the familiarity of most of the tunes selected – is not, however, matched by the arrangements themselves, which offer much to surprise. The unusual scoring (for women’s voices with orchestral or piano accompaniment) and striking modal innovations of the 1949 work, in particular, closely parallel the timbral and tonal experiments of the composer’s last decade. No two settings are quite alike – even the ‘unison’ settings vary in the forces used for accompanying descants – while the synthetic modal scales and contrapuntal knottiness of a cappella two- and three-part settings like ‘The Sheep Shearing’ and ‘The Unquiet Grave’ make them among the most interesting of his entire career.

Hymn tunes

Among Vaughan Williams’s best-known – and certainly most widely disseminated – folksong arrangements are the seventy or so that he adapted as hymn tunes and published in The English Hymnal (1906; revised edition, 1933) and its later offshoot Songs of Praise (1925; enlarged edition, 1931).¹⁴ Tunes like ‘Monks Gate’ (adapted from ‘Our Captain Calls’), ‘Kingsfold’ (‘Dives and Lazarus’) and ‘Forest Green’ (‘O Little Town of Bethlehem’, adapted from ‘The Ploughboy’s Dream’) have spread well beyond the Church of England to occupy a permanent place in the repertories of Catholic and Reformed congregations alike. But the composer’s involvement with hymnody went much deeper than mere folksong campaigning, for he served as music editor of these hymn books,¹⁵ an enormous undertaking that brought him into close contact with a wide range of musical traditions, many from outside his native shores. These volumes have been much praised for their editorial discernment and clarity – he exhumed old versions of tunes and greatly improved on contemporary practices of source identification – and for the quality of their musicianship. Arrangements by Vaughan Williams (and his adaptations of others’ arrangements) are among the most common versions of tunes like ‘Helmsley’ and ‘Winchester Old’, met with in the hymn books of many different faiths, while his redaction of ‘Lasst Uns Erfreuen’ (‘Ye Watchers and Ye Holy Ones’), used for the doxology in many English-speaking churches and homes to this day, is possibly one of the best-known melodies in Christendom. His eighteen original hymn tunes, likewise, are among the most popular written in the twentieth century; ‘Sine Nomine’ (‘For All the Saints’), ‘Salve Festa Dies’ (‘Hail Thee, Festival Day’) and ‘Down Ampney’ (‘Come Down, O Love Divine’), among others, have spread across the globe, lodging themselves within many hymnodic traditions, and show no signs of disappearing.¹⁶

Much has been written about the apparent incongruity of a self-proclaimed atheist – one who ‘drifted into a cheerful agnosticism’ later in life – devoting so much time and energy to religious music.¹⁷ Whatever the precise nature of his beliefs, there can be little doubt that writing for the Church of England provided for Vaughan Williams the opportunity to put his deepest artistic beliefs into action. Recognizing that for most people, church attendance represented their only regular opportunity for music-making, he set aside any theological scruples in order to focus on this crucial venue for amateur singing. Placing the needs of the congregation before those of the trained choir, he pitched tunes as low as possible, insisted on slower tempi and unison singing, and provided a wide variety of accompanimental patterns and arrangements, including various methods of combining congregation and choir.¹⁸ He also introduced much unfamiliar English material, notably folksongs but also early psalm tunes by Tallis, Gibbons and Henry Lawes. Exposure to these ‘bracing and stimulating’ melodies would, he believed, help create a demand for the music of contemporary English composers who were inspired by this same source material. It would also improve a debased musical taste by weaning congregations from the ‘languishing and sentimental’ hymn tunes of the Victorian church, and ultimately ease the grip of the commercial interests that had foisted these tunes, unwanted, upon them. Replacing what was essentially an alien and repugnant musical style with one that more truly ‘represent[ed] the people’, his reforms would improve the musical health of the nation.¹⁹

The mixture of paternalism and progressivism will be familiar from Chapter 1. On the one hand, he assumed the insufficiency of the people to make good musical choices for themselves, especially in the face of commercial pressures, and worked to ‘improve’ their taste. On the other, by insisting on unison singing and suggesting new methods of performance, he effectively elevated the congregation to a status of artistic equality with the choir. The unlikely combination of aesthetism and populism was one he shared with other early twentieth-century hymnodic reformers, notably the Poet Laureate, Robert Bridges, whose Yattendon Hymnal (1899) may be said to have initiated the trend, as well as the left-leaning Anglo-Catholics making up The English Hymnal committee, and the founders of the ‘Hymn Festival’ movement, Hugh Allen, Walford Davies and Martin Shaw among them.²⁰ But it was Vaughan Williams who did most to mainstream these ideas, with the result that he has been singled out by critics who question how a self-styled ‘populist’ could dismiss an entire repertory of Victorian ‘favourites’.²¹ Close examination of the sources shows, however, that, whatever he may have said about them, Vaughan Williams was surprisingly indulgent towards Victorian hymns, retaining 118 tunes by Victorian composers in the 1906 English Hymnal, where (by his own admission) he felt some pressure to do so, and more than 70 in Songs of Praise Enlarged, where he was under no such obligation.²² It is true that he generally viewed the strong diatonic outlines of folksong and the relative austerity of early English psalmody as a salutary antidote to nineteenth-century emotionality. But then this was the attitude of many Victorian editors, particularly those of High Church leanings, who maintained earlier traditions of adapting hymn tunes from folksong and who did much to infuse the English hymn-tune repertory with music taken from older non-English sources.²³ Indeed, it was their example that inspired Vaughan Williams to expand his hymn books’ offerings even beyond theirs, so that English tunes new and old now coexisted with an unprecedented number of Scottish, Welsh, Irish, continental European and even American tunes of all periods. As he later wrote of his English Hymnal experiences, ‘I determined to do the work thoroughly’ so that the book might be ‘a thesaurus of all the finest hymn tunes in the world’.²⁴

The eclecticism of Vaughan Williams’s editorial work is often overlooked, a casualty of the insularity that often characterizes the reception of his music, and yet it holds the key to one of the meanings of his nationalism. For what he really sought with hymn tunes was not the propagation of a specifically ‘English’ musical style but rather the consolidation of a repertory of tunes, native or otherwise, that reflected English musical practice. Basing hymn tunes on early English music and folksong was only one way to achieve this; drawing on ‘outside’ sources that had made their way into English tradition was another. In these terms, the Genevan psalter, brought to England by sixteenth-century Protestants returning home after the Marian Exile, counted as ‘national music’, as did those Lutheran chorales, harmonized plainchant melodies and continental folksongs and carols (notably those adapted from the 1582 Piae Cantiones) that had entered into English usage, largely at the hands of Victorian editors. He was interested not only in retaining well-known tunes, however, but also in introducing those that might become popular. Hence his striking out into French Diocesan and American revivalist repertories as well as his digging yet deeper into sources first tapped by the Victorians. (Here, his introduction of recently collected folksongs, not just those taken from antiquarian publications, was a real innovation.) If these efforts to expand the repertory conveniently coincided with his weeding out of Victoriana, it is important to stress again his deliberate retention of a large number of these tunes. Taken to the hearts of the people, Victorian hymnody had unquestionably become part of the nation’s cultural heritage and could not be ignored.

Thus was Vaughan Williams’s approach to hymnody closer to Victorian practice than his words might suggest. The similarities extend to his own original hymn tunes which, if they carefully avoid the ‘sickly harmonies of Spohr’ and the ‘operatic sensationalism of Gounod’, still partake of the ‘grand effects’ of nineteenth-century art music that Nicholas Temperley has identified as the overriding characteristic of the Victorian hymn tune.²⁵ This is the case with ‘Sine Nomine’ and ‘Salve Festa Dies’, with their ‘orchestral’ organ parts and ‘symphonic’ treatment of vocal forces, no less than with ‘Down Ampney’ and ‘Magda’, whose lyrical warmth and intricate part-writing clearly emulate tunes by John Bacchus Dykes and Samuel Sebastian Wesley. Even the composer’s concern for a tune’s ‘singability’ – its accessibility to congregations of limited musical experience – seems to have been anticipated by the Victorians. It is no coincidence that all of the tunes mentioned here are firmly tonal, nor that his folksong adaptations tend to blunt the idiosyncrasies of the source melody by harmonizing modal and gapped scales in a tonal context and by ironing out irregular rhythms and phrase lengths. Such alterations were an inevitable consequence of the need to match tunes with metrical texts; even so, it is plain that he took extra care to bring ‘unorthodox’ tunes into line with Victorian norms.²⁶

Not that innovation was lacking. The slower tempi, lower pitching and varied arrangement offerings were all radically new, as was the insistence on unison congregational singing. Indeed, one of the most striking features of The English Hymnal was the adaptation of the ‘unison song’ to the hymn-tune format. This very English choral subgenre, whereby harmony is relegated to the accompaniment while all voices (here, congregation and choir) crowd on to the melody line, had been adapted to church hymnody by Stanford and Parry but not to the extent that Vaughan Williams used it. Meanwhile, his revival of psalm-tune settings by Ravenscroft and others, in which the congregation was directed to sing the tune (in the tenor) against the choir’s four-part harmony, contributed to the emergence of the popular ‘descant’ style of Anglican hymnody after 1915.²⁷ Purely stylistic innovation, by contrast, was slower to emerge. Possibly the success of ‘Kingsfold’ and ‘Kings Lynn’, two adapted folksongs whose modal eccentricities were not ‘edited out’ of the 1906 English Hymnal, convinced him that congregations could sing such music; whatever the inspiration, Songs of Praise introduced a spate of original hymn tunes that emphasized the ‘characteristic’ features of folksong and early English music. Tunes like ‘Mantegna’, rooted in the Phrygian scale and with wildly unpredictable harmonies throughout, and ‘Oakley’, whose sudden turn to Elizabethan dance rhythms halfway through makes for a striking formal experiment, drink deep at the well of native tradition and unabashedly proclaim a ‘national’ style.

These tunes represent the extreme, however. By far the more common procedure in the later hymn books was to combine native elements with mainstream techniques, as in the celebrated ‘King’s Weston’ from 1925 (see Ex. 7.3). The rugged Dorian melody and repeated VII–i progression (occurring in bars 1–2, 6, 10 and 15–16) announce a boldly modal music, while a descending stream of parallel ⁶₃ chords in bar 13 – concluded, after an interruption, on the third beat of bar 14 – introduces a taste of fauxbourdon. Yet these ‘characteristic’ features are carefully blended with traditional methods of part-writing, dissonance treatment and long-range formal planning whereby an insistent melodic pattern of rising pitches and repeated rhythms mounts until its sudden reversal in bar 13. So dramatic a trajectory is enhanced by the infiltration of the modal E minor tonic by its relative G major, a competing ‘tonal’ key area that reinterprets the crucial D major chord (VII in E) as a dominant (note its strong reinforcement by the secondary dominant in bars 11–12). The ‘true’ key is ambiguous because the composer carefully refrains from directly emphasizing either tonic – here the parallel ⁶₃ chords at precisely the moment we anticipate resolution are a stroke of genius – and it is only with the arrival at the E minor chord in the last bar, at once surprising and expected, that harmonic tension is released.

Ex. 7.3. ‘King’s Weston’.

In its mixing of native materials and common-practice techniques ‘King’s Weston’ embodies a kind of synthetic ‘middle ground’ between innovation and convention, ideological commitment and broad-minded inclusiveness, that we find in Vaughan Williams’s engagement with hymnody as a whole. Even the details of the arrangement reflect this blended approach, for the massed voices and unconventional three-part accompaniment draw both on the ‘unison song’ format of the reformers and on the Victorian practice of giving ‘orchestral’ weight to the organ part. The forceful tone and ‘Sine Nomine’-like march tread, meanwhile, conform to the ‘muscular Christianity’ of Caroline Noel’s very Victorian text, with its militant references to ‘captains’, ‘empires’ and ‘victory’.

Church music

Vaughan Williams’s early atheism appears to have affected his attitude towards other forms of liturgical music for, apart from hymn tunes and a few student works, he wrote no music for the church until 1913. This was the isolated anthem ‘O Praise the Lord of Heaven’, composed for the fortieth anniversary of the London Church Choir Association. Music appropriate to Anglican worship does begin to appear with greater frequency after World War I; still, many of the works from this period were either commissioned or written with specific performers in mind. Altogether, only eight of the composer’s twenty-seven mature liturgical works (comprising small-scale anthems, motets, canticles and organ preludes, as well as a small clutch of simple prayer settings for children) were composed entirely ‘on spec’.

Like his hymn-tune arrangements, in other words, Vaughan Williams’s church music would seem chiefly to be an outgrowth of his philosophy of musical citizenship. A number of these works – the Te Deum in G (written for the enthronement of the Archbishop of Canterbury, 1928), the Festival Te Deum ‘founded on traditional themes’ (for the coronation of George VI, 1937), ‘The Souls of the Righteous’ (for the dedication of the ‘Battle of Britain’ Chapel, Westminster Abbey, 1947), O Taste and See and the The Old Hundredth Psalm Tune (for the coronation of Elizabeth II, 1953) – were specially composed for state occasions. Others, like the Magnificat and Nunc Dimittis (‘The Village Service’, 1925) and the simple Anglican-chant setting of Psalm 67 that he wrote for St Martin’s Church, Dorking in 1945, were explicitly geared to the limited resources of the parish choir. Taken together, these works display an extraordinary disparity in their intended audience and venue, and yet are united in a shared focus on filling a specific need. Moreover, many of them can be performed using ‘alternative’ forces, while some are even suitable for non-liturgical performance at outdoor festivals, public ceremonies and concerts.

The use of popular hymn tunes in many of these works confirms this broadly democratic approach. ‘St Anne’ (‘O God Our Help in Ages Past’) appears in both ‘Lord, Thou Hast Been Our Refuge’ (1921) and ‘O How Amiable’ (1934), while the entire text of the 1954 Te Deum and Benedictus is grafted, somewhat bizarrely, on to a series of well-known psalm tunes. This last choral work has a part for the congregation, as do the four so-called ‘hymn-anthems’, of which the The Old Hundredth Psalm Tune, mentioned above, is the best known. This genre, popularized by the interwar Hymn Festival but appropriate to worship services, combined congregation and choir by presenting the different verses of popular hymns in varied forms of arrangement (a descant was obligatory), sometimes with brief instrumental interludes linking the strophes.²⁸ A more striking experiment involving untrained voices is the complete set of Services (Morning, Communion, Evening) in D minor written for performance at Christ’s Hospital boarding school in 1939. No hymn tunes are employed, but an ingenious arrangement – a unison congregational part, seldom rising above d′′, combines with more complex music for SATB choir – allows (as the score states) for ‘a large body of voices [to] share in the musical settings of the service’. Here, the ‘unison song’ of the school assembly joins with the traditional forms of Anglican church music to create a work in which all can take part.

About a third of Vaughan Williams’s church music invites congregational participation; the remainder is for the trained choir alone and occupies a sliding scale from relatively easy to extremely difficult – further proof of his determination to provide music for every situation. Whatever the difficulty, liturgical and musical tradition is closely observed. Thus the Te Deum settings provide cuts for the authorized ‘shortened form’ of that canticle, while the D minor Communion Service follows the Prayer Book’s sequence of movements to the letter, with the Gloria placed last (Some Responses are also included). It is true that motets (only one of them in Latin) outnumber anthems, the more strictly Anglican genre, but this may reflect the development of a late nineteenth-century type of cathedral anthem, pioneered by Stanford, to which the name ‘motet’ was generally affixed.²⁹ Stanford’s stylistic innovations, indeed, are palpable in the prominent organ parts (often quite independent of the voices) of several works as well as in the tight construction of ‘O Clap Your Hands’ (1920), the Te Deum in G and others. The Irish composer’s application of cyclical elements to all the movements of a service, thus creating a unified musical experience spanning a whole day, is actually carried further in the D minor Services, as nearly all the material can be traced back to the first two themes (bars 1 and 36) of the opening Te Deum. Finally, the antiphonal effects of the cathedral tradition are prominent in many works, achieved through the juxtaposition of decani and cantoris choral groups (or their equivalents) and by the playing of soloists off against the full choir in the manner of verse anthems and services.

The selection and handling of texts, likewise, show an impressive familiarity with church history and tradition. Most are taken from The Book of Common Prayer and the Authorized (King James) Version of the Bible, though the composer pushes past Anglican orthodoxy by also setting the dissenting prose of Bunyan (‘Valiant for Truth’, 1940) and mysterious Catholic texts by Skelton (‘Prayer to the Father of Heaven’, 1948) and the Office of Tenebrae for Maundy Thursday (‘O Vos Omnes’, 1922). Hymns and metrical psalm texts are also prominent – in the ‘hymn-anthems’, of course, but also in the remarkable ‘Lord, Thou Hast Been Our Refuge’, which interleaves the Prayer-Book version of Psalm 90 with the first verse of Isaac Watts’s metrical paraphrase of the same. Critics who complain of the work’s stylistic incongruities – Watts’s poem is set to the sturdily diatonic ‘St Anne’, its ‘proper’ melody, while the Prayer-Book words are set to a rhythmically and modally flexible plainchant – have possibly overlooked this textual nuance.

So intimate a knowledge of the liturgy, texts and music of the Established Church speaks strongly of the composer’s cultural nationalism – his love for the words of the Authorized Version, his respect for patterns of worship that have prevailed in England for centuries, even his ‘inclusive’ embrace of the range of religious traditions (Puritan, Catholic, centrist) making up the Anglican compromise. (The works pointedly based on English folksong – the 1937 Te Deum, and the Benedictus and Agnus Dei that he contributed to J. H. Arnold’s Oxford Liturgical Settings (1938) – clearly also embody this ‘cultural’ view of the Church.) And yet, the intensity of expression in certain works suggests the possibility of an added personal motivation. It seems hardly coincidental that the stream of Vaughan Williams’s liturgical choral works effectively began with his return from the Great War, an experience that most commentators now agree triggered in him some kind of emotional if not metaphysical crisis. In particular, the immediate post-war works – ‘Lord, Thou Hast Been Our Refuge’, ‘O Vos Omnes’ and the Mass in G minor (1922) – communicate a sense of spiritual urgency that is conveyed by texts that emphasize human weakness and by settings that are predominantly sombre, occasionally anguished, but also flecked with unmistakable moments of radiance. The mysterious, even devotional mood of these works, one of which was directly inspired by Catholic ritual,³⁰ is decidedly private, not public, in expression, and may well mark a shift from the composer’s early atheism to a mature agnosticism or possibly to something approaching orthodox Christian belief. For the charged atmosphere of these works is not exclusive to the immediate post-war period. It resurfaces in ‘Valiant for Truth’ and in ‘Prayer to the Father of Heaven’, as well as in ‘The Voice out of the Whirlwind’ (1947) and ‘A Vision of Aeroplanes’ (1956), works whose harrowing Old Testament imagery asserts the base inadequacies of human beings.

We should be wary of interpreting these works too securely in conventional religious terms, however. They may acknowledge human fallibility and insignificance but this does not mean that Vaughan Williams necessarily addressed himself to a specifically Christian God. He was certainly drawn to the ethical teachings of Christ and found solace in the Christian focus on the life of the spirit. But he would never have asserted that Christianity had a monopoly on what he called the ‘ultimate realities’: in the tradition of philosophical idealism, no religious faith had exclusive access to the realm that lay ‘beyond sense and knowledge’.³¹ Indeed, the inscrutability of these regions prepared him for a stoical acceptance of the possibility that the spirit did not exist at all. Whatever his uncertainty, the aspiration remained, and he viewed it as the role of the artist to share intimations of spirit with others in a language that they could understand. Hence his adoption of the ready-made frameworks of Christian ritual, and hence his focus on the everyday needs of the human community around him. The anguished cry that concludes the Agnus Dei and thus the Mass is a plea for peace directed not merely to the Divine but to all of humanity in the wake of World War I. The triumphal entry of the hymn tune ‘St Anne’, sweeping away all doubts at the conclusion of ‘Lord, Thou Hast Been Our Refuge’, celebrates God’s role as protector even as it proclaims the power of a shared musical culture to comfort and sustain us.

Like the folksong settings but unlike the hymn tunes, the church music roughly follows the mainstream of Vaughan Williams’s stylistic development. Thus the antiphonal forces of ‘O Praise the Lord of Heaven’ – two full choirs and semi-chorus – employ the textural and dynamic contrasts of the Fantasia on a Theme by Thomas Tallis, written three years earlier, while the blatant root-position parallel triads of ‘O Clap Your Hands’ announce the new simplification of means of the 1920s. The stylistic possibilities of retrenchment, meanwhile, are explored in ‘O Vos Omnes’, where chromatic parallel chord streams (at ‘desolatam’, for example) anticipate the bitonality and localized atonality of the large-scale experimental works of the interwar years. Expressive devices peculiar to the post-1945 music are observable as well – the shadings of light and dark resulting from subtle shifts of mode and tonality (‘The Souls of the Righteous’, ‘Prayer to the Father of Heaven’) and the captivation with new timbres and scales (‘A Vision of Aeroplanes’). Even works written in his patented ‘public’ vein – remarkably consistent across his entire career – echo more distinctive compositions from the same period. The melismatic jubilations of the 1928 Te Deum resemble those heard in the Benedicite, while the Jubilate from the D minor Morning Service shares material with the finale of the Fifth Symphony.

But it is the personal synthesis of elements taken from a wide variety of historical styles and periods that most strongly links the church music with Vaughan Williams’s output as a whole. This can be observed anywhere but is perhaps best illustrated by the Mass, a work whose neo-Tudor associations have obscured awareness of a wider eclecticism. Techniques favoured by sixteenth-century English church musicians – false relations, fauxbourdon-like textures, contrasts between soloist(s) and the full choir – are indeed present, but they are combined with others – canon and points of imitation, sectional division of the text (articulated by textural contrasts), emphasis on the church modes – that were the lingua franca of the period, common to English and continental music alike. Even these Renaissance techniques are but a ‘starting-point’³² for what is clearly a highly personal essay, however. The false relations derive less from contrapuntal logic than from Vaughan Williams’s idiosyncratic modal harmony (itself adapted from Debussy), while the traditional ‘sectional’ organization is cross-cut by nineteenth-century methods of thematic recall, both within and across movements. The striking employment of two SATB choirs, ‘answering’ one another and often combining into eight parts, likewise, would seem to derive less from Renaissance models than from the imaginative reconstructions of the nineteenth-century Cecilian movement.³³ (Extreme dynamic markings – reaching pppp in the Credo – suggest this as well.) Folksong plays a role too, not merely in the frequent pentatonic gestures (as at ‘Gratias’ in the Gloria) but also in the way those gestures contribute to localized and even long-range motivic argument.

Though not lengthy, the Mass is undeniably a major work, a masterful utterance that the composer never sought to duplicate. The same cannot be said of the other works discussed in this chapter: small-scale ‘functional’ works chiefly intended for amateurs that for the most part trace well-worn patterns of genre, performance and style. What is remarkable is that, within this limited framework, Vaughan Williams produced such interesting music, much of it paralleling (and in some cases even anticipating) developments in his musical language. In the final analysis, though, innovation was secondary to the overarching social goals that he conceived for this music. With it, he sought to familiarize generations with a wealth of beautiful melody, raise awareness of a common cultural heritage, and above all inspire individuals to make music for themselves.

Operas

With the possible exception of Tchaikovsky, no composer’s operatic career was less emblematic of his success elsewhere than Vaughan Williams’s. An avid opera-goer who loved the works of Mozart, Verdi and Wagner, he was well-versed in the genre’s conventions, as revealed in his first letter to his librettist for Hugh the Drover, Harold Child: ‘I have no objection to the structure being more or less formal & conventional – as I like duets, trios, quartets or even quintets – I think all opera has to be conventional (or perhaps I should say not realistic)’.² But he then used the ‘not realistic’ qualifier to make a more radical assertion: ‘Slow – long tableaux – or long dramatic pauses are always good, as the music takes a long time to speak, much longer than words by themselves – in fact, one wants purely musical effects in an opera just as one wants purely poetical effects in a drama.’³ Indeed, Vaughan Williams repeatedly turned to dramatic scenarios that were inspired by, conceived as, or offered opportunities for static tableaux – not just in his operas, but in his masques and pageants as well. Such an attitude may explain the inconsistency of his libretti, implying that he was less interested in the literary quality of texts per se than in the potential he saw in individual scenes for eliciting effective emotional (and therefore musical) responses.⁴

His first attempt to put these principles into practice came in the form of his ‘Romantic ballad opera’ Hugh the Drover. He began composing the work in 1910, had largely completed the vocal score by the time he volunteered for military duty in 1914, and finished scoring it in 1920. Soon after, representatives from the short-lived British National Opera Company approached him about arranging a performance, which took place at His Majesty’s Theatre in London on 14 July 1924. Since then, Hugh has proved remarkably resilient, its tunefulness and local colour appealing to generations of amateur and student performers. Its emphasis on English music and imagery – ballad opera, English folk tunes, West Country villages, Maying festivities – reflected the composer’s advocacy for a distinctively English approach to opera, which he felt could be achieved by presenting clear alternatives to established continental traditions. This was his goal from the outset, as he wrote to Child – ‘I have an idea for an opera written to real English words, with a certain amount of real English music and also a real English subject [which] might just hit the right nail on the head’⁵ – but it deviated from most mainstream attitudes about opera in England, undoubtedly colouring critical opinions towards its reception.⁶

In letters to Child, Vaughan Williams expressed strong views about the characters’ portrayals – specifically, that the members of the English peasantry should be treated as sympathetic figures rather than comedy rustics – but in the attempt to avoid one set of stereotypes, he fell into another. The characters occupy generic roles: the mysterious outsider (Hugh), the star-crossed heroine (Mary), the dastardly villain (John the Butcher), the stern authority figure (the Constable), the sympathetic matron (Aunt Jane) and so on. This might have lent itself quite well to a light comic setting, but the characters’ earnestness – and the general lack of humour in the libretto – reduce them to stock character types rather than fully fledged personalities. None of the revisions Vaughan Williams made to the opera over the next three decades overcame this initial shortcoming, but despite the dramatic problems, there is much to enjoy in the music.⁷ As with most of his operas, the vocal lines are delivered in a flexible arioso style occasionally interrupted with more conventional numbers. Some of these are in fact arrangements of traditional English songs, which the characters usually perceive as diegetic music, but non-folkish arias occupy the usual expressive space reserved for such tunes in more conventional operas.⁸ Here the composer finds a vein of almost Puccinian passion that transcends the more narrowly parochial aspects of the story, foreshadowing similar effects in works to come.

Vaughan Williams’s next opera, and the only dramatic work he classified as such, was Sir John in Love (1924–8). Based on Shakespeare’s The Merry Wives of Windsor, Sir John in Love demonstrates that Otto Nicolai, Gustav Holst and even Verdi did not exhaust the tale’s possibilities. With a large and engaging cast, a well-crafted libretto (adapted by the composer), and a ravishingly tuneful score, Sir John stands as Vaughan Williams’s finest full-length opera, possessing a musical and dramatic range far beyond Hugh the Drover. Folk music is present once again, but much more subtly and effectively employed than in Hugh – no longer mere musical display, it can indicate characters’ emotional states, or help facilitate or foreshadow dramatic developments.⁹

The musical idiom Vaughan Williams employed in Sir John is somewhat more conservative than in contemporaneous works, but its emphasis on lyric vocal modality – encompassing a wide expressive range, from Falstaff’s cocksureness to Ford’s jealousy to Anne’s and Fenton’s tenderness – plays to one of Vaughan Williams’s greatest strengths as a composer, and his admiration for Shakespeare comes through in the exquisite text setting. That is not to say that productions of this work lack challenges. Whereas Verdi’s Falstaff has only ten principal parts (plus chorus), Sir John has twice that number; it also demands considerable acting skills, and the emphasis on small groups and ensembles limits the opportunities to showcase individual performers. Nonetheless, following its revival by English National Opera in 2006, Sir John in Love is ripe for reconsideration. Far from being second-rate Verdi, it is first-rate Vaughan Williams, and would be a welcome addition to the standard repertory.

If Sir John is Vaughan Williams’s best full-length opera, then his best one-act work – and one of his finest works in any genre – is Riders to the Sea (1925–32), a setting of the eponymous text by J. M. Synge. This introspective and fatalistic tale (‘a tragedy in undertones’, as Frank Howes put it)¹⁰ was well suited for Vaughan Williams: its unpretentious vernacular language supports a nearly actionless plot, thus allowing many opportunities for musical narrative. Naturalistic declamation dominates the vocal lines, underscored by octatonic harmonies that several writers have compared to those of Mussorgsky’s Boris Godunov or Debussy’s Pelléas et Mélisande. It is equally likely, however, that Vaughan Williams was following the example of his former tutor, Ravel, who was similarly obsessed with symmetrical scales and capturing the cadence and inflection of the spoken word.¹¹

Riders to the Sea was composed simultaneously with Sir John in Love in the late 1920s, a period when Vaughan Williams began expanding his compositional idiom into previously unexplored regions (not just octatonicism, but polytonality and polymodality).¹² In some ways, Riders could be Sir John’s tragic reflection. Synge’s tale, like Shakespeare’s, is populated by realistic characters with clear and plausible emotions; yet in Riders, as Howes observed, the characters ‘are real but oppressed and overwhelmed’ by the forces of the world, a profoundly modernist conceit.¹³ While ‘modernist’ is not often a term applied to Vaughan Williams’s works, Riders to the Sea expresses the same unflinching view of modernity as the Fourth Symphony, the vision of a man deeply engaged with the cultural and political challenges of his own era. Hugh Ottaway recognized this after hearing a radio broadcast of the work in 1952: ‘If, for a moment, the composer be substituted for Synge’s islanders and the sea taken as a symbol of the overwhelming forces in the modern world, then the work will assume its true proportions as a remarkably complete expression of Vaughan Williams’s character and outlook’.¹⁴

Walter Aaron Clark also observes that the characters of Synge’s play find no solace in religion, which he points to as another factor that may have attracted the agnostic composer to the work.¹⁵ Certainly there is no overt reference to organized religion, but Maurya’s stoic acceptance of her son Bartley’s death (‘They are all gone now, and there isn’t anything more the sea can do to me . . . it’s a great rest I’ll have now, and it’s time surely’) parallels the passage in The Pilgrim’s Progress when Christian, his burden having finally dropped from his back, says ‘He has given me rest by his sorrow, and life by his death’.¹⁶ Vaughan Williams emphasizes this theme of redemption through loss in several works, but despite its decidedly Christian overtones, the sacrifice that ultimately redeems the protagonist in Riders (as well as in The Pilgrim’s Progress, Job, and other works) is caused by indifferent Nature or an inscrutable God, not by another person of their own free will. (This may account for the opera’s harmonic ambiguity, appropriate for a work with such an unsettling dramatic resolution.) This dark echo of the Wagnerian tenet of redemption through love is a distinctly modernist phenomenon, and is perhaps understandable coming from a veteran whose memories of the dead and dying he encountered on the fields of France – and the apparent senselessness of their deaths – must have haunted him for years.

One other opera had its genesis during the late 1920s: The Poisoned Kiss; or The Empress and the Necromancer. Begun in 1927, the plot for The Poisoned Kiss was derived from two short stories: Nathaniel Hawthorne’s Rappaccini’s Daughter (1846) and Richard Garnett’s The Poison Maid (1888), itself based on Hawthorne’s tale. Vaughan Williams thought Garnett’s story ‘had the makings of a light opera’,¹⁷ and it shows considerable potential as a comic fairy-tale that could live up to its billing as a ‘Romantic Extravaganza’; by the end, all of the characters pair off in the best tradition of mannered comedies by Oscar Wilde or W. S. Gilbert. And with its discrete numbers separated by spoken dialogue, The Poisoned Kiss outdoes even Hugh the Drover in its resemblance to conventional ballad opera. Unfortunately, as with its predecessor, the libretto is highly problematic (which may explain why the opera was not recorded until 2003). Librettist Evelyn Sharp – journalist, author, suffragette and the sister of Vaughan Williams’s late colleague and fellow folksong advocate Cecil Sharp – simply was not up to the task, and the composer ended up doing much of the work himself.¹⁸ By the time he died, Vaughan Williams had substantially revised Sharp’s text three times, ultimately replacing the prose dialogue with rhymed couplets by his second wife, Ursula.¹⁹ As with any patient subjected to multiple operations, however, these continual revisions weakened the work as much as they strengthened it. Flawed from the start by not committing either to a frothy romantic comedy or to a satirical fairy-tale, Vaughan Williams tried to split the difference between the two, and failed at both.

But again, as with Hugh the Drover, the poor libretto belies the considerable beauty of the music. Indeed, this may be part of the problem Vaughan Williams faced when composing: the balance between lyrical romanticism and light comedy skews heavily towards the romantic side. Consider Tormentilla’s touching lullaby for her pet cobra in Act i, and the duet that follows with Amaryllus (‘Blue larkspur in a garden’), Amaryllus’s own ‘Dear love, behold for good or ill’ from Act ii, or the gorgeous duet between Tormentilla and the Empress from Act iii (‘Love breaks all rules’). Even minor characters, such as Tormentilla’s attending trio of Mediums, receive such delights as the tango-inflected ‘Behold our mystic exercises’ in Act iii. However, when it came to patter songs or deliberately humorous texts, Vaughan Williams simply could not adopt a Sullivanian manner. Musical humour was never the composer’s strongest suit; on those occasions that he succeeds, it is when he uses long-breathed, lyrical tunes as parodies (e.g., the lovely ‘O, who would be unhappy me’ at the end of Act i, as a takeoff on the standard lover’s complaint), a technique also employed in Jane Scroop’s Lament from the contemporaneous Five Tudor Portraits.

Masques, ballets and pageants

With one exception, Vaughan Williams’s masques and ballets are not particularly well known, perhaps because he eschewed most balletic conventions. Several of his dance works have no significant female lead; he also shunned the use of classical techniques (in Job, he went so far as to stipulate that there should be no dancing en pointe), had little interest in Diaghilev’s wildly popular productions for the Ballets Russes, and even avoided referring to his own works as ‘ballets’.²⁰ This parallels his disinclination to use the term ‘opera’, and perhaps for the same reason; in both cases, he wanted to create something distinct from continental models, and his idiosyncratic descriptors hint at the sources of his inspiration.

His first effort, Pan’s Anniversary (1905), was written with Gustav Holst for a performance of Ben Jonson’s eponymous masque at Stratford-upon-Avon. Comprising a series of hymns, fanfares and dances (the latter arranged by Holst from various folk tunes and Elizabethan keyboard works), it is more notable for being Vaughan Williams’s first attempt at writing for the stage than for any of the actual music.²¹ Most of the original numbers he wrote are both tuneful and diatonically tonal – perhaps surprisingly, given that the presence of shepherds, nymphs, and other quintessentially pastoral figures might imply a strong gravitation towards modality. The tunes are not especially distinctive (although the closing half of Hymn ii, ‘Pan Is Our All’, recalls the final two phrases of Vaughan Williams’s own ‘Sine Nomine’), and exploit standard dramatic choral effects, such as echo choruses in Hymn iii.

Nearly two decades passed before his next contribution to the field, Old King Cole (1923), the only work that Vaughan Williams designated a ballet rather than a masque. It re-enacts an embellished version of the nursery rhyme: King Cole of Colchester is visited by his daughter, Helena, who has brought him an elaborate hookah pipe as a gift. When it breaks, the King calls for a drink – brought forth in a bowl – and then presides over a contest among his fiddlers three. He awards the prize to the third fiddler (although his daughter preferred the romantic tune of the second), and all march off to dinner in a general dance. Unsurprisingly, given his interest in English folk customs, Vaughan Williams chose to showcase morris dances in this work, while the three fiddlers’ solos also featured traditional tunes (‘Go And List For A Sailor’, ‘A Bold Young Farmer’ and ‘The Jolly Thresherman’, respectively), though only the first and third Fiddlers’ tunes are danced – the first to a morris jig, and the third to a sword dance.

As was typical for Vaughan Williams’s dramatic works, Old King Cole’s debut was mounted by amateurs (on this occasion, musicians from the Cambridge University Musical Society, and dancers from the English Folk Dance Society), and presumably very nervous amateurs at that, since the composer apparently continued making corrections to the score following the dress rehearsal. That Vaughan Williams countenanced, even encouraged, such performances may seem odd, but he strongly supported amateur musicians; without them, he argued, there could be no artistic environment in which professionals could thrive.²² This may explain why, even as an elder statesman of English music, he deigned to compose such comparatively humble works as a folk tune medley for a masque mounted by the English Folk Dance Society (An EFDS Medley, 1937), or a short passage of Solemn Music for the Masque of Charterhouse (1950) as a gift to his old public school.

On Christmas Night (1926) was the first work that Vaughan Williams specifically referred to as a masque. Adolph Bolm, a former dancer with Diaghilev, proposed a scenario based on Dickens’s A Christmas Carol (the original name of the work); by 1935, the composer had revised it for a performance at Cecil Sharp House with Douglas Kennedy providing new choreography.²³ The work is a quodlibet of Christmas tunes, primarily traditional English carols and dances, awkwardly shoehorned into a freewheeling adaptation of Dickens’s story: only one spirit visits Scrooge; there are no visions of Christmases yet to come; and the lively closing feast has been replaced, rather oddly, with a tableau of the Nativity in the manner of ‘an Italian “Sancta Famiglia”’.²⁴ Vaughan Williams was aware of the thematic conflicts that this engendered, and considered jettisoning all references to A Christmas Carol altogether in the revisions of 1935. Yet he did not, and the resulting dramatic hodgepodge probably discouraged later performers’ interest. (Vaughan Williams did undertake a more straightforwardly religious Christmas piece when he composed and arranged the music for The First Nowell (1958), a nativity play that turned out to be his final work.)

Given that Vaughan Williams’s early contributions to dance literature were not especially noteworthy, the depth and sophistication of Job: A Masque for Dancing (1930) comes as something of a shock. Unlike most of Vaughan Williams’s other dance works, Job eschews folk dances in favour of more formal Tudor-era genres, such as pavanes, galliards and sarabands, which may go some way towards justifying its classification as a masque. Job was the brainchild of surgeon and scholar Geoffrey Keynes, who approached his sister-in-law, Gwen Raverat, about designing the sets and costumes for a ballet based on the scenes in William Blake’s Illustrations of the Book of Job. She in turn recommended that Keynes approach Vaughan Williams – her cousin – to provide the music. As Alison McFarland has explained, each of the participants brought their own interpretation of Blake with them, but ‘by the end of the project [Vaughan Williams] had become the dominant voice, and the finished drama bore little resemblance to what Keynes and Raverat initially had written’ – and for that matter, to what Blake himself had drawn.²⁵

Vaughan Williams’s facility in composing Job may have stemmed from his interest in dramatic tableaux. Keynes originally selected only eight of Blake’s twenty-one illustrations as the basis for the ballet, but Vaughan Williams refers to seventeen of them in his scenario and stage instructions. He also reordered several illustrations (sometimes over the strong objections of his collaborators), though he provided instructions in the published score on how to restage Scenes 4 and 5 if producers wished to retain the original order of Blake’s images.²⁶ In some cases, Vaughan Williams undermined or even contradicted Blake’s own pictures. He pointed to paintings by Rubens and Botticelli as useful models for the blocking,²⁷ and insisted that Job not be seen to play an instrument at the ballet’s end, despite Blake’s depiction of him playing a harp in celebration. Once again, Vaughan Williams sees Job’s acceptance of his loss as the primary vehicle for his redemption: following his ordeals, Blake’s Job is raised up and made prosperous by God once more, but Vaughan Williams’s Job is left, as the stage directions indicate, an old and humbled man content to gaze upon his cornfields and bless his daughters as the curtain falls. He has accepted the punishment that he believes to be his due, and in this understanding of his own shortcomings, achieves some measure of spiritual tranquillity and inner peace.²⁸

In spite of the tensions with his colleagues, it seems that Vaughan Williams tried to find musical analogues for the work’s visual components. Keynes and Raverat were impressed by Joseph Wicksteed’s analysis of Blake’s pictures, in which he argues that Blake represents ‘good and evil by the use of pervasive symbolism regarding spiritual and corporeal, that corresponds to right and left in the illustrations’.²⁹ Vaughan Williams read Wicksteed’s book at the urging of his collaborators, but told Raverat ‘I’m not going to worry about the left foot and the right foot’.³⁰ In terms of the choreography and set design, this is true; however, perhaps taking a cue from Stravinsky, he did employ contrasting musical systems to represent the opposing forces at work in the ballet. McFarland points out that the music associated with Satan (or his influence) emphasizes minor thirds and tritones, short and metrically ambiguous motives organized contrapuntally, and extensive dissonance. The music of God and heaven, by contrast, features lyrical modal melodies with restrained accompaniment, homophonic textures, and generally diatonic harmonies. Job, appropriately, lies between: his main themes are long-breathed and either modal or pentatonic (i.e. heavenly), but often receive contrapuntal settings and may return in fragments (i.e. satanic).³¹ Thus while professing his lack of interest in systems of symbolic interpretation, Vaughan Williams apparently gleaned enough from them to imagine a way of applying the visual imagery within a purely aural context.

Vaughan Williams’s last masque was The Bridal Day, which he originally wrote for the English Folk Dance and Song Society. It was his first collaboration with Ursula Wood – later Ursula Vaughan Williams – whose transformative effect on Vaughan Williams’s life is well documented.³² Their deeply passionate love affair helped the composer break through his professional crisis of the mid-1930s, and The Bridal Day was the first fruit of this relationship. Scored for strings, flute and piano, its first performance was scheduled for the autumn of 1939, but was cancelled because of the war; after fourteen years and some revision, it received a television broadcast by the BBC in June 1953, though both Wood and Vaughan Williams found the production unsatisfactory. More extensive revisions took place in 1957, when the composer recast the work as a cantata titled Epithalamion, the name of the poem by Edmund Spenser featured in both pieces.³³

The masque celebrates the marriage of a Bridegroom and his Bride; nymphs, gods and the Graces participate in the festivities alongside the couple’s family and friends. The guests take their leave as the couple, serenaded by musicians and blessed by the goddess Juno, enters the bridal chamber. Although choral singing, ritualistic gesture and mime play important parts in this work, many of the short scenes are organized as dances, and most are separated by monologues delivered by a Speaker; according to the stage instructions, whenever the Speaker delivers his verses, ‘everyone on the stage should be still, forming a tableau’.³⁴ The use of scenic tableaux and ceremonial gestures is applied in a more Dionysian context here than in some of his earlier compositions (appropriately, considering the presence of the god Bacchus in the masque). The solemnity of the event notwithstanding, this joyous and festive work uses dance to celebrate physical love as much as spiritual, and does so without either prurience or prissiness. The emphasis on the solo flute and viola signals this context, as these instruments were closely associated with two of Vaughan Williams’s most sensual works: the viola with Flos Campi (1925), inspired by erotic texts from the Song of Solomon, and the Magnificat (1932), in which he ‘thought of the flute as the disembodied, visiting spirit and the alto solo as the voice of a girl yielding to her lover for the first time’.³⁵

The least familiar of the composer’s dramatic works – and destined to remain so because of their occasional origins – are the three pageants: the London Pageant (1911), the Abinger Pageant (or The Pageant of Abinger, 1934) and England’s Pleasant Land (1938). Pageants are episodic historical plays (usually affiliated with a particular locale) intended for outdoor performance. Of the three in which Vaughan Williams was involved, the London Pageant was unquestionably the most audacious. It engaged some 15,000 participants on the grounds of the Crystal Palace, and was one of the high points of the festivities surrounding the coronation of King George V and the accompanying Festival of Empire.³⁶ Vaughan Williams was one of several composers recruited by the musical organizer, W. H. Bell, to write the music, and was assigned the fifth scene of Part ii, a setting of sixteenth-century May-Day Revels entitled ‘The London of Merrie England’. Roger Savage describes the scenario as ‘standard Victorian antiquarian picturesque stuff’, that is to say, a warmly nostalgic view of England’s supposed golden age ‘before Puritans, Satanic mills, and suburban sprawl’.³⁷ Vaughan Williams arranged a medley of English folk and traditional music for the scene’s twelve short movements.³⁸

The Pageant of Abinger and England’s Pleasant Land were smaller affairs, and involved collaboration with E. M. Forster (who wrote the scenarios) and Tom Harrison (who produced the performances). The three men all lived near each other in rural Surrey, and conceived the works as benefits for the local communities: the Abinger Pageant was designed to aid the Abinger Church Preservation Fund, while the proceeds of England’s Pleasant Land went to the Dorking and Leith Hill District Preservation Society.³⁹ Vaughan Williams’s own contributions consisted primarily of folksong and hymn-tune arrangements, although he did compose a few short pieces of new music; among these was the anthem ‘O How Amiable’ for the Abinger Pageant, which went on to enjoy an independent existence.⁴⁰ For England’s Pleasant Land he recruited several composers to contribute new works and arrangements along with his own,⁴¹ and also borrowed some pre-existing pieces, among them Holst’s ‘I Vow to Thee, My Country’ and Parry’s ‘Jerusalem’, along with several traditional tunes.⁴² Although much of the music Vaughan Williams composed for the pageant has been lost, what remains includes two passages (a procession for the Ghosts of the Past and a funeral march set to poetry by Horace) that have connections to the Fifth Symphony, work on which began in the same year (though exactly when is not clear). The Exit of Ghosts resembles the chorale-like tune heard in the Scherzo of the symphony, while much of the music accompanying the setting of Horace foreshadows the oscillating and palindromic figures associated with the opening of the first movement. Ironically, these seeds were planted at a time when Vaughan Williams feared that he was creatively ‘dried up’.⁴³

Incidental music

A simple inventory reveals that most of Vaughan Williams’s dramatic works can be broadly classified as ‘incidental music’ (whether for stage plays, radio dramas, or film); however, in terms of their aggregate length, such works constitute less than half of his dramatic output. With a handful of exceptions, this music is not particularly well known. Much of it has never been recorded (or, as with much of the film music, has not been widely available until recently), and performances or broadcasts are rare. Nonetheless, these works are significant not just on their compositional merits, but for what they reveal about Vaughan Williams’s interest in artistic collaboration and his own personal creative growth.

His earliest (and one of his finest) efforts in this field came in the form of incidental music for Aristophanes’ The Wasps (1909), which he wrote for the Cambridge Greek Play Series. The Wasps is a bawdy political satire that criticizes the Athenian legal system via the protagonist, Philocleon (or Procleon), and his antagonistic son, Bdelycleon (or Anticleon). It was performed in Greek (although the vocal score published by the Cambridge Greek Play Committee featured English translations of the songs and choruses), required a relatively large, all-male cast,⁴⁴ and featured a wide array of comic styles, from lowbrow puns to sophisticated social parody to a surrealistic dream sequence featuring enormous dancing crabs.⁴⁵

Although the overture to The Wasps is one of Vaughan Williams’s most popular instrumental works, the score is rarely heard in its entirety. This is a great shame, for as he had recently completed his tutelage with Ravel, Vaughan Williams displays in the work a level of confidence that he had hitherto lacked. To describe it merely as ‘incidental music’ does not do it justice: while there are entr’actes and other passages for orchestra alone, there are also nine choruses (including a twenty-minute Parabasis) and several passages of melodrama in which the orchestra plays a critical role. It is also quite witty – much more so than Hugh the Drover or The Poisoned Kiss. For example, No. 17a in the full autograph score (the dream sequence with the Sons of Carcinus) features quotations from Mendelssohn (the famous Frühlingslied, Op. 62, No. 6), Offenbach (the ‘Apache Dance’ from the ballet Le Papillon and the opera Le Roi Carotte) and Lehár (a much-slowed and ominous-sounding version of the waltz from The Merry Widow), though none of these are present in the published vocal score. Instead, there is a note at the end of No. 17: ‘Here follows the entry music of the “sons of Carcinus” which will vary according to season’.⁴⁶ This statement not only implies that Vaughan Williams fully expected The Wasps to be revived, but that future producers should not feel obligated to use the music he provided for this scene if they did not find it suitable or topical. This practice of suggesting possible performing modifications within the score became a hallmark of his later compositions, but the latitude he provided here is exceptional.

In 1911, Vaughan Williams continued his forays into Attic drama by turning to three plays by Euripides – The Bacchae, Iphegenia in Tauris and Electra – that had been translated by Gilbert Murray, then Regius Professor of Greek at Oxford. Dancer and choreographer Isadora Duncan had proposed using parts of these plays as the basis for a larger collaborative project involving other creative luminaries such as Vaughan Williams, Murray and the actor Gordon Craig; unfortunately, Duncan soon lost interest in the idea, and so productions were never staged. Nonetheless, Vaughan Williams set several choruses from the three plays to music (and as three independent works, rather than as a single Euripidean cantata), and may have conducted performances of them with the Palestrina Society.⁴⁷ They consist almost exclusively of extended choral monologues, but are interspersed with passages for soloists and by different configurations of semi-chorus. The declamation is more evocative of Greek theatre than is the music from The Wasps, in part because of the strong emphasis on unison choral textures. A footnote in ‘Dark of the Sea’ from Iphegenia makes clear the effect that Vaughan Williams sought: ‘In this and similar passages the note values are only approximate – the passage should be declaimed freely as in good reciting’.⁴⁸

Not long after completing the Euripides choruses, Vaughan Williams was invited to serve as music director for the Shakespeare Memorial Theatre Festival at Stratford-upon-Avon for the autumn and spring seasons of 1912 and 1913, respectively. Archibald Flower, Chairman of the Theatre Board, engaged him ‘to conduct the Theatre orchestra, to rearrange music for Richard the Second and Henry the Fifth, and eventually to reconsider music for all the Histories’, although this daunting final request did not come to fruition.⁴⁹ Despite Vaughan Williams’s initial enthusiasm for the task – probably because it gave him practical stage experience as he worked on Hugh the Drover – his efforts met with indifference from the festival’s long-time manager, Frank Benson, and lackadaisical musical attitudes from the performers. This must have frustrated him no end; for one performance of Richard III, he had to sing an offstage passage of plainchant himself because no singers were available.⁵⁰ Nonetheless, his efforts were singled out for praise in the Stratford-upon-Avon Herald, and he was presented with a ‘silver-mounted baton of ebony and ivory’ by a group of long-time festival attendees as a token of their esteem.⁵¹

The number of Stratford productions for which he completed music was impressive for such a short period: these included The Merry Wives of Windsor, Richard II, Henry IV Part 2, Richard III, Henry V and Twelfth Night; he may also have written music for Much Ado About Nothing and Romeo and Juliet.⁵² Benson’s company also mounted a performance of Shaw’s The Devil’s Disciple, for which Vaughan Williams also contributed three arrangements. All of these works emphasize simple, short musical numbers, mainly fanfares and entrance music carefully timed to suit the actors’ needs. While the scores themselves are not particularly distinctive or significant, the challenges of working to deadlines and the opportunity to observe production techniques were unquestionably of long-term benefit to the composer.

Some other incidental music from this period serves a very different kind of drama: the plays of Maurice Maeterlinck. Vaughan Williams composed scores for both The Death of Tintagiles and The Blue Bird in 1913; the former was performed privately the same year, but it is unclear if the music for The Blue Bird ever received a performance.⁵³ Unusually for one of Vaughan Williams’s early works, the music for Tintagiles was privately commissioned; Philip Sassoon asked Vaughan Williams to write the music on the strength of a recommendation from Cecil Sharp.⁵⁴ Although the dark, mysterious music is perfectly suited to Maeterlinck’s symbolist text, the first performance met with an unreceptive audience. The surviving music for The Blue Bird is considerably different, to judge from the piano score; much of it is dancelike in character, but echoes of Debussy, Tchaikovsky and Wagner’s ‘Loge’ motif evoke the magical events taking place on stage.

Later in his career Vaughan Williams had the chance to write music for a handful of radio plays, two of which allowed him to revisit earlier works. His first effort was for a version of The Pilgrim’s Progress (1942), which will be discussed later. He also returned to Shakespeare in 1944, composing an entirely new score of Richard II for a BBC radio performance.⁵⁵ Although in the end it was not in fact used for the production, its incorporation of reminiscence themes (for Richard, John of Gaunt, Bolingbroke and ‘Impending Tragedy’) and longer, more chromatically complex numbers marked a significant advance from the music he wrote for the same play at the Stratford Festival. His last radio work was on a new subject, Thomas Hardy’s The Mayor of Casterbridge (1950). This is very brief compared to the other two, consisting of only three movements (‘Casterbridge’, ‘Intermezzo’ and ‘Weyhill Fair’), though the last is divided into five subsections. The first two movements feature a broadly pastoral idiom and are rather longer than most of the movements of incidental music he wrote (nearly three minutes each). ‘Casterbridge’ also features an arrangement of the carol ‘On Christmas Night the Joy-Bells Ring’, published separately in 1953 as Prelude on an Old Carol Tune.

This practice – arranging and renaming individual numbers from incidental music scores – was even more conspicuous in the case of Vaughan Williams’s film music, his involvement with which began in 1940 when he was asked to write the score for the Oscar-winning 49th Parallel. Some arrangements were fairly conventional: he fashioned a concert suite out of the film score for The Flemish Farm (retitled Story of a Flemish Farm, 1945), a song-cycle from The Vision of William Blake (the Ten Blake Songs, 1957) and, in perhaps the most famous and complex example of such activity, drew liberally upon the music he wrote for Scott of the Antarctic in composing Sinfonia Antartica (1949–52). But he also allowed others to make such arrangements, most notably Muir Mathieson, musical director for several films in which Vaughan Williams participated, who crafted suites out of the scores for Coastal Command (1942) and The England of Elizabeth (1957).⁵⁶

As Daniel Goldmark rightly points out, however, it would be wrong to assume that Vaughan Williams simply viewed his film scores as ‘the starting-point for larger and greater (and more utilitarian) chamber and orchestral pieces’, even if that was the prevailing critical response to them during the composer’s own lifetime.⁵⁷ In fact, Vaughan Williams was powerfully attracted to the medium of film composition, claiming it ‘contained potentialities for the combination of all the arts such as Wagner never dreamt of’.⁵⁸ And while the creative possibilities drew him to the genre in the first place – it is perhaps worth noting that one of Vaughan Williams’s only extended descriptions of his own compositional processes came in his essay ‘Composing for the Films’ (1945) – he was also impressed by the discipline that it imposed, viewing it as a particularly apposite craft for young composers ‘apt to be dawdling in their ideas, or whose every bar is sacred and must not be cut or altered’.⁵⁹

Most of the films he worked on were either quasi-documentaries or propaganda efforts, often under Mathieson’s musical direction; these included 49th Parallel (1941), Coastal Command (1942), The Flemish Farm and The People’s Land (both 1943), Stricken Peninsula (1945), Dim Little Island (1949), The England of Elizabeth (1957) and The Vision of William Blake (1958). He also received more conventional studio commissions from Ernest Irving – the music director for Ealing Studios and eventual dedicatee of Sinfonia Antartica – including The Loves of Joanna Godden (1946), Scott of the Antarctic (1948) and Bitter Springs (1950).⁶⁰ As the titles suggest, many of these films celebrated English cultural, historical and political accomplishments, either overtly or implicitly. Given Vaughan Williams’s close association with such markers of Englishness in his own music, it is unsurprising that Mathieson thought he might be a suitable recruit for similarly evocative cinematic works.

His approach to writing for films was idiosyncratic. Rather than compose the soundtrack in post-production or after viewing a rough cut of the film (the most typical practice), he would start writing the music based on the scripts and cues provided to him before shooting actually started, then collaborate with the musical director after the film was complete to adjust it as necessary.⁶¹ He was less inclined to use musical stingers or short, evocative motives to punctuate ‘every action, word, gesture or incident’ than to ‘ignore the details and . . . intensify the spirit of the whole situation by a continuous stream of music’ – that is, to create a self-contained number that would evoke a general mood or atmosphere rather than reflect a specific image on screen, a practice well in keeping with his established approach in writing for the stage.⁶² Jeffrey Richards and Daniel Goldmark have both made admirable assessments of the style, function and larger cultural significance of Vaughan Williams’s film music – particularly those scores written during wartime – which reveal considerable dramatic sophistication residing in music intended for mass consumption.

The Pilgrim’s Progress

The reasons for Vaughan Williams’s fascination with John Bunyan’s allegorical novel remain obscure; after all, one would hardly expect that a wealthy, intellectual, agnostic artist of the twentieth century would find much common ground with a working-class, unschooled, evangelical Puritan of the seventeenth. Yet Vaughan Williams kept returning to Bunyan’s writing for nearly half a century, setting it in part and in whole, in small and large forms alike, despite its limited dramatic potential. While full of sharply defined settings and characters, the story’s allegorical nature does not encourage naturalistic behaviour. Vaughan Williams’s various truncations of the story only heightened this condition, as he eliminated many of the confrontational challenges that Christian/Pilgrim faced in favour of ritualistic and/or transcendently religious scenes, yet another reflection of his penchant for scenic tableaux. The result was that the pilgrimage often came off less as a quest for redemption and more as a watered-down paean to the values of the spiritual life, ‘attended by a few obstacles but nothing that a mildly proper spiritual exercise and ritual action could not overcome’.⁶³

Vaughan Williams’s first setting of The Pilgrim’s Progress (1906) came at the request of Evelyn Ouless and Joanna Hadley, who produced an adaptation of the story for a community performance in Reigate, Surrey. Of Vaughan Williams’s four versions of this work, the Reigate production is the most iconoclastic: it has a larger cast than any of the others (featuring major figures such as Faithful, Hopeful and Christiana); the opening is set not ‘in the similitude of a dream’, but in the City of Destruction itself; Christian’s and Faithful’s trial in Vanity Fair is much longer; and the tune ‘Monks Gate’ – one of several folk and traditional tunes the composer set in this production, including Tallis’s Third Mode Melody and the hymn tune ‘York’ – is sung to the words ‘Who would true valour see’, which received a different melody in later settings.

Nathaniel Lew’s analysis of the Reigate performance provides an invaluable description of how the story was adapted to suit the needs of a small, amateur company.⁶⁴ In particular, he points out the producers’ frequent recourse to tableaux, which constitute fully half of the play’s twelve scenes.⁶⁵ This is a less eccentric decision than it may seem at first glance, as tableaux vivants were familiar nineteenth-century dramatic staples, and would have effectively circumvented any major limitations of the amateur actors involved. More importantly, the tableaux allowed the music to proceed without significant dramatic distractions, emphasizing an aural rather than a visual focus. Lew argues that this formative experience powerfully shaped Vaughan Williams’s approach to later settings of Bunyan’s work, though not always advantageously.⁶⁶

Vaughan Williams’s next major setting came in 1922, when he completed a one-act ‘pastoral episode’ entitled The Shepherds of the Delectable Mountains, based on the concluding section of The Pilgrim’s Progress. It is a very compact work, with a twenty-minute duration and a cast of five (though it also calls for the ‘Voice of a Bird’ and a small chorus offstage). Although not a tableau in the strict sense, the dramatic climax (Pilgrim’s crossing of the River of Death) takes place offstage as the Shepherds watch, a technique recalling the epilogue of Glinka’s A Life for the Tsar. The action on stage is decidedly understated, as is much of the music; as with the Pastoral Symphony of the same year, the musical efficacy of The Shepherds comes from the quiet intensity of its expression rather than though large, demonstrative gestures, which would have ill fit the tranquillity of the hills overlooking the Celestial City.

While Vaughan Williams’s libretto closely follows Bunyan’s prose, the composer excised and reordered several scenes, and borrowed additional passages from Christiana’s arrival at the Delectable Mountains from the second part of the story.⁶⁷ This version was also the first one where Vaughan Williams used the name ‘Pilgrim’ for his protagonist, a choice he explained many years later to Rutland Boughton: ‘I, on purpose, did not call the Pilgrim “Christian” because I want the work to be universal and apply to any body who aims at the spiritual life whether he is Xtian, Jew, Buddhist, Shintoist or 5th [sic] day Adventist’.⁶⁸ Vaughan Williams also chose not to include Pilgrim’s companion, Hopeful, even though the two reached the Delectable Mountains together in Bunyan’s original. This seems a dramatically sensible choice for a work of this length, but may have shaped his more questionable decision not to include a companion in the ‘morality’ of 1951, the final act of which assimilated much of the music and text from The Shepherds.

Twenty years after the premiere of The Shepherds of the Delectable Mountains, Vaughan Williams composed incidental music for a radio play of The Pilgrim’s Progress. Adapted by Edward Sackville-West and starring John Gielgud, it was first broadcast in 1943. The score features thirty-eight discrete musical numbers, though some of them feature subentries while others simply reprise earlier passages from the play.⁶⁹ As a whole, the score marks a considerable advancement over the incidental music he wrote for Reigate; in fact, much like his radio score for Richard II, its use of reminiscence themes and confident scoring resembles the compositional practices associated with his film music.

That is not to suggest the radio play completely broke with the past. Like the Reigate production, the radio play prominently features the hymn tune ‘York’ in the prologue and epilogue; unlike Reigate, however, it greatly expanded the use of Tallis’s Third Mode Melody – better known as the theme from Fantasia on a Theme by Thomas Tallis – which was heard only at the end of the Reigate production.⁷⁰ The score of the radio version reproduces the arrangement from the Tallis Fantasia at its first appearance in the prologue, and the tune returns in scenes evoking the Celestial City or its representatives (as when Christian arrives at the House Beautiful, again after the death of Faithful, and several times throughout the scene at the Delectable Mountains). A third tune, replacing ‘Monks Gate’ from Reigate and accompanying Bunyan’s text ‘Who would true valour see’, is equally persistent. Foreshadowed by a trumpet fanfare as Christian arrives at the Wicket Gate, it is associated with Christian himself. It returns in a triumphant brass arrangement following the defeat of Apollyon, and again throughout the scene in the Delectable Mountains. All three tunes are heard in sequence at the end of this scene, a marvellous point of culmination.

In some ways, this may be Vaughan Williams’s most effective setting of Bunyan precisely because the visual element is removed. Deprived of the scenic tableaux he felt were so critical to staged performances, Vaughan Williams could allow the music to carry more dramatic weight without any concomitant loss of interest on the part of the audience, now primed to accept the story as a purely aural performance. With the dramatic expectations thus adjusted, the music can accompany audience members’ inner visions of Vanity Fair, the House Beautiful, the confrontation with Apollyon in the Valley of Humiliation, or the Delectable Mountains – visions undoubtedly more sumptuous and magical than the stage could allow.

With his four-act morality of 1951, Vaughan Williams came full circle in his personal pilgrimage with Bunyan’s work – quite literally so. As already noted, the dramatic and scenic organization of this opera closely resembled that of the Reigate production.⁷¹ This is somewhat unusual, as one might expect that the larger budget and scale of a full-length Covent Garden performance – not to mention the passage of several decades – would have inspired Vaughan Williams to new heights, leading him to introduce characters or scenes absent from the earlier version (say, ‘failed pilgrims’ like Talkative, Pliable, or Ignorance; tempters like Mr Worldly Wiseman; or obstacles such as the Slough of Despond, or Giant Despair and Doubting Castle). In fact, the opposite was the case: removing Faithful and Hopeful, for instance, eliminated opportunities for Pilgrim to commiserate or celebrate with others, and in so doing reveal his humanity to the audience. Instead, the ritualistic and static aspects of the drama were thrown into even sharper relief, leading several critics to suggest that performances of the work might be better suited to cathedrals than concert halls.

Despite these problems – heightened by the questionable direction and design of the first performance – the music is beautiful, if something of a stylistic jumble. Most of the music from The Shepherds of the Delectable Mountains was incorporated into the final act, and, as is well known, portions of the Romanza from the Fifth Symphony accompany the loss of Pilgrim’s burden from his back. Such ethereal sounds contrast with Apollyon’s doleful monotone, the oleaginous huckstering of Lord Lechery in Vanity Fair – a later addition set to a text by Ursula Vaughan Williams – or the cheerfully vacuous duet for Mr and Mrs By-Ends, though these more evocative passages are in the minority. Vaughan Williams retains a certain degree of dramatic consistency in all versions of the story, however, by once more returning to the theme of redemption through loss. True, Christian/Pilgrim achieves life everlasting in the Celestial City, but does so at the cost of his mortal existence; moreover, in order to demonstrate his worthiness to an allegedly all-loving and all-forgiving God, he is obliged to abandon his family, friends and way of life for a lonely and dangerous quest, and will be condemned to everlasting torment should he fail to surmount all of the obstacles placed before him. This, of course, reflects Bunyan’s own uncompromising Puritanism, but also the tropes of abandonment and divine caprice seen in several of Vaughan Williams’s other dramatic works.

Conclusion

Vaughan Williams’s dramatic music spans an enormous range, encompassing everything from deadline-driven hackwork to the inspired heights of sublimity only attainable by great artists. Taken as a whole, the quality of his output in the field is inconsistent, and this unevenness has almost certainly led to the underestimation of his abilities. But this is not surprising, considering the array of performers, settings and occasions for which he wrote: rushed schedules, amateur performances and works of brief topicality are not conducive to legacy-building. Still, many of the shortcomings in Vaughan Williams’s writing for the stage stem more from his musical reach exceeding his dramatic grasp than from an absence of inspiration or compositional facility. Nowhere is this more evident than in his various settings of The Pilgrim’s Progress. Although the music is often transcendently lovely, and well suited to the more meditative and reflective portions of the story, it seems to come at the expense of similarly effective music for passages driven more by action or character study. While it is perhaps unfair to criticize the composer for not adhering to dramatic goals that he did not set for himself (after all, he was quite open about his preference for letting the music carry the story rather than the text), his disinclination to follow established stage conventions more closely, for whatever reason – a desire to establish an independent ‘English’ musical identity within a particular genre, personal dislike for current musical trends, limitations of time or resources or collaborators – opens the door for such critiques. Several of his attempts to establish new (and, in many cases, specifically English) models and subjects for the stage may have confused or alienated critics and audiences alike, thus rendering them less viable for adoption by larger, more well-established performing organizations.

Yet for all his idiosyncrasies in writing for the stage, Vaughan Williams did create several works of great power and beauty. Given the degree to which he refused to compromise his artistic vision, and the long odds most English composers of his generation faced in having their stage works realized – not to mention the lack of encouragement from the English musical world in general – the fact that he met with any success at all is remarkable. But if the time has come to stop dismissing Vaughan Williams’s dramatic works out of hand, then it is also time to forgo special pleading for their value. That not all of his stage works were triumphs is no mark of shame. It is to posterity’s discredit, however, if received wisdom about the nature and quality of his efforts in the field is allowed to blind us to his very real and significant accomplishments within it.⁷²

The early chamber music: finding a voice

Vaughan Williams’s first acknowledged chamber work is the String Quartet No. 1 in G minor (1908–9; revised 1921). A number of earlier pieces were withdrawn by the composer, and remained in manuscript until his widow, Ursula, lifted the embargo forty years after his death, and allowed a number to be published by Faber Music. The earliest of these is effectively Vaughan Williams’s first completed attempt at an extended original composition, the String Quartet in C minor (1898). This shows some flair for string-writing, even though the textures never exceed the conventional. Particularly impressive is the fleet, scherzo-like music in the Trio, which derives much of its impetus from rhythmic disruption. If the liking for (and, in some ways, the manner of) development suggests a Beethovenian model, the minuet-like Intermezzo suggests Haydn in its asymmetrical phrasing. Meanwhile the modality for which the mature Vaughan Williams is known is incipient at most. The opening of the first movement is consistently Aeolian up to bar 28, the only harmonic deflection from the tonic C minor triad being the Neapolitan in bar 15. More varied colours, in both timbre and harmony, are found in the expansive Quintet in D major for clarinet, horn, violin, cello and piano of the same year.³

Breadth of utterance is another characteristic of the mature composer and is a notable feature, too, of the Piano Quintet in C minor (1903), which comprises three big, ambitious movements, each lasting around ten minutes. These movements are, generally speaking, impressively paced, suggesting an embryonic symphonist. A sense of trajectory across the span of the work is achieved by inserting pre-echoes of the finale’s main theme in the first two movements (from bars 186 and 277 in the first movement, and from bars 55, 99 and 171 in the second). The style is post-Brahmsian, nowhere more so than in the bold opening gesture. But the music also shows the influence of Parry in its investment in the expressive potential of diatonic dissonance (see, for example, bars 1–7 in the second movement)⁴ – though non-dissonant diatonic contexts such as from bar 139 in the first movement also make their mark. Indeed, it is these moments – mostly wistful, often slightly pained – that make the biggest impression, rather than the chromatic working-out. All three movements end quietly, with melodic descents and a sense of withdrawal – precursors of the ‘niente’ endings for which Vaughan Williams was later to become famous. There is a tendency towards over-writing, and his writing for piano (for which he was criticized throughout his career) has more than a few awkward moments, especially when he is attempting to forge his own manner of modal harmony and voice-leading (in, for example, the harmonized repeats of the final movement theme).

Three years later, in the Nocturne and Scherzo (1906), the problem of how to integrate separate and radically different resources (a vein of late Romantic chromaticism that is more reminiscent of Bridge, and two kinds of diatonicism – the appoggiatura-laden version inherited from Samuel Sebastian Wesley and Parry noted in the Piano Quintet (see n. 4), and the modality of the folksongs in which he had an increasing interest) seems to come to a head. As Kennedy observes of the contemporaneous In the Fen Country: ‘The harmonic idiom is a curious mixture of the chromaticism of Strauss and his imitators, the pure diatonic style in which Butterworth was to write, and occasional glimpses of the Vaughan Williams of a few years later’.⁵ While there is no hint of folksong in the Nocturne movement, various moments of melancholic yearning convey an English tone. See, for example, the events from bar 91 to the end – the long descent from the climax, the nostalgic return to the opening bars and the listless cadence on to an added-sixth chord. Folksong does play a part in the Scherzo movement (which is actually subtitled ‘founded on an English Folk-song’: Vaughan Williams had started collecting folksong three years earlier in 1903), principally in the Lento section towards the end, though elements of the song do also appear in the Scherzo section proper.

When the String Quartet No. 1 in G minor received its first performance in 1909, the purportedly specialist audience was, if the report in The Musical Times is accurate, if not hostile then certainly dismissive:

To the audience of today, cognizant, say, of Schoenberg’s works of 1909 (Erwartung Op. 17, Five Orchestral Pieces Op. 16, Das Buch der hängenden Gärten Op. 15 and Drei Klavierstücke Op. 11), the notion of Vaughan Williams’s music representing ‘the extreme development of modernism’ seems absurd. Indeed, despite the modal resources (the quartet begins with a theme in G Dorian, and the main material of all four movements is modal) and a classicizing tendency at certain points (perhaps the result of studying with Ravel in Paris the previous year), much of the work is conservatively Romantic. It is possibly these elements that the Grove writers are thinking of when they say that the ‘traces of former ways, usually involving a chromatic expressiveness’ constitute ‘a serious handicap’.⁷ Peter Evans observes that ‘A sonata plan can be read into the first movement, but the progress is desultory and quartet dialogue minimal; the surprisingly grazioso minuet and the jiggish rondo-finale are more successful’.⁸ He does not mention the third, slow movement. Entitled ‘Romance’, it is notable for its obliquity: no cadence is other than equivocal, while the opening theme suggests several possible ‘tonics’ (the proper modal term would be ‘final’), none of which is confirmed by supporting harmony. The coda is characterized by the unexpected ‘discovery’ of the C major triad, one of very few pure triads in the movement; even this is presented in its least stable $$ position, and has the seventh, B, pitted against it. Nevertheless, it is a moment of ‘revelation’ that is perhaps the most Vaughan Williams-ish aspect of the movement. Particularly uncharacteristic of the mature composer is the Romantic subjectivity: he adopts a personal, if not actually confessional, utterance, especially in the outer sections, rather than the ‘communal’ tone found in later movements titled ‘Romance’ or ‘Romanza’ (those, for example, in Symphony No. 5, the Piano and Tuba Concertos, String Quartet No. 2 and indeed in The Lark Ascending, which is subtitled ‘Romance’).

In both the Minuet and Trio and the Finale (Rondo capriccioso) a structural role is accorded to modal elements. Some of the elaborating chromaticism derives from the extension of modal qualities: for example, the parallel whole-tone scales in the violins in the Minuet at two bars after rehearsal letter C are readily heard as an extension of the whole-tone step at the top of E Mixolydian scale, E–D♮. This is a far more convincing use of the whole-tone scale than at the end of the first paragraph of the Romance, where it seems unmotivated. The whole-tone scale is also generated in this way in the Finale (see the three bars before rehearsal letter B and from five bars after rehearsal letter E to letter F). It is presumably these (relatively short) examples of whole-tonery that caused one of Vaughan Williams’s friends to say that he ‘must have been having tea with Debussy’ when he wrote the work.⁹

Writing the year before String Quartet No. 1 was revised and again saw the light of day, Edwin Evans noted that it

The Phantasy Quintet was actually composed in 1912, two years after the Fantasia on a Theme by Thomas Tallis, Vaughan Williams’s ‘breakthrough’ work (in technique if not in terms of initial public recognition). So it is not surprising that there should be a much clearer sense of identity. Some of the Fantasia’s technical means of achieving rapt contemplation are on show in the Prelude, while the lento Alla sarabanda foreshadows the luminous, diatonic hymnic tone of the final movement of A Pastoral Symphony and the slow movement of Symphony No. 5. The work was commissioned by W. W. Cobbett, presumably to exemplify his chamber-music ideal of a one-movement, multi-sectional composition, though this particular piece is in essence divided into four separate movements (each linked to the next by an ‘attacca’).¹¹ Vaughan Williams does, though, employ an integrative strategy he had essayed on a much larger scale in A London Symphony, on which he was concurrently working – the return of the opening melodic material in a ‘window’ near the end of the fourth movement (Burlesca). Material from the Prelude also returns in the development section of the Scherzo.

The success of the Prelude results in large measure from the simplicity of both materials and structure. The movement begins with a pentatonic melody rising from the bottom of the viola’s range. The pentatonic scale is devoid of semitones, with the consequence that, firstly, there is no inherent hierarchy: it is the metrical and rhythmic emphasis of the notes F and C that creates the sense of F as tonic. Secondly, there is no inherent sense of tension and release (there is no leading note demanding to be resolved to the tonic). It is primarily the latter that engenders the atmosphere of contemplation. After the second of the two types of material – triads moving in parallel – is brought into play, a second melody is introduced, employing the same pentatonic scale but this time descending from an f″ on violin I. The shape of the movement arises from the disposition and combination of this material. In the first section, the F centre is elaborated by shifts to the flattened mediant, A♭. The second section, meanwhile, starts with a shift to A♮: the harmony focuses on the A major triad (though this subsequently alternates with, and is supplanted by, the C that provides the route back to F), while both melodies are transposed to A: see Ex. 9.1. One could say that the melodies are counterpointed against each other, except that there is none of the propulsion normally associated with counterpoint: they are simply allowed to unfurl in the same modal ‘field’. The third section is the most exploratory harmonically, visiting as before first the flat, then the sharp side, but extending a little deeper into both. The movement is then rounded off with a literal repeat of the first melody. It is partly the recognition that regions of experience have been explored but ‘things remain the same’ that lends this short movement such a profound sense of melancholy. The Grove writers observe that in the Tallis Fantasia ‘hidden depths’ are revealed ‘contemplatively and obliquely rather than through direct dialectic’. The Prelude demonstrates this in miniature. It is a world immediately recognizable as Vaughan Williams’s own. It is not without dissonance (a ninth results from the voice-leading at four bars after rehearsal letter B, for instance, and the exact parallelisms after letter D result in a succession of false relations), but this enhances the contemplative state rather than supplying a destabilizing element.

Ex. 9.1. Phantasy Quintet: shift to A, 5 bars before rehearsal letter C.

It would be a mistake to view this as representing Vaughan Williams’s only authentic mode of utterance, however. As we have already seen, some of his finest apprentice music is to be found in scherzos, and the Scherzo in this Quintet strengthens the notion that he has a particular affinity for the genre. It is cast in sonata form, and it is in the development section that the material from the Prelude makes its reappearance: the initial harmonic succession recurs from five to seven bars after rehearsal letter E and the first melody from seven bars after letter F. Like the latter, the main theme of the third movement, Alla sarabanda, is pentatonic, and because of this a kinship with the Prelude may well be felt; but there is no actual thematic connection (at least, I cannot detect one, despite Vaughan Williams’s own statement that ‘There is one principal theme . . . which runs through every movement’¹²). The Burlesca, which starts on the cello (that instrument having been silent in the sarabanda) confirms the overall centre as D, though the final cadence in the subdued closing section is preceded by an alternative harmonization of the violin’s serene a‴ by F major, the tonic triad of the first movement.

The Phantasy is evidence that Vaughan Williams had purged himself of the stylistic traits of nineteenth-century musical Romanticism, even if his world-view remained in at least some sense Romantic.¹³ It could be argued that his voice was yet to emerge fully (the revision of A London Symphony in 1918 shows that he was tussling with how to respond to aspects of modernism), but the quality of lyricism was fully established. As I have suggested above, this is characterized by a sense of the collective: both the folksong resonances in the Phantasy, and derivations from English Renaissance music in the Tallis Fantasia, conjure up a repository of experience suspended outside time and evoke an authorial voice that is representative rather than personal and individualistic, even though the music is of course both of those things. This is consolidated in The Lark Ascending, which was initially completed in 1914, the year in which, in an article entitled ‘British Music’, Vaughan Williams discusses (however sketchily) the relationship between compositional individuality and communal utterance.¹⁴

Modality and melancholy: The Lark Ascending

Though composed just before the First World War, The Lark Ascending was not performed until 1920, by which time the composer had revised it. Subtitled ‘Romance for violin and orchestra’, it might be called a meditation on George Meredith’s sentimental (and, at best, third-rate) poem of the same name, extracts from which are reproduced at the beginning of the score. Vaughan Williams’s response is also often dismissed as sentimental, and too slight to be accorded serious scholarly attention.¹⁵ It seems likely to have been one of the works that provoked the epithet ‘cowpat music’.¹⁶ Yet it is a work of greater subtlety than has been generally acknowledged.

The term ‘Romance’ signals lyricism, of course, and, on this occasion at least, an essentially non-dialectical approach to form, as Table. 9.1, a précis after David Manning, shows.¹⁷ The Grove writers describe the work as ‘wholly idyllic, and therefore different in feeling from the post-war pastoral works’.¹⁸ It must be questioned, though, whether a ‘wholly idyllic’ scene is possible. As other writers have noted, only the outer sections are representational, the violin’s cadenzas enacting the lark’s song and ascending flight. I would suggest that a distinction between representation and commentary leads to an undermining of the idyllic – a distancing from the scene depicted that creates a powerful sense of loss.

Table 9.1 Formal summary of The Lark Ascending*

* adapted from David Manning, ‘Harmony, Tonality and Structure in Vaughan Williams’s Music’, II, 79

The opening of the work is reproduced as Ex. 9.2. The orchestral ‘introduction’, as Manning labels it, provides a frame that allows the representational world to come into being. Modality does some crucial work here. Firstly, for the listener used to common-practice tonality as the norm (arguably all of us in the industrialized West) it offers something ‘other’. Secondly, Vaughan Williams’s particular construction of modality here allows the underpinning major ninth chord to be dissonant without needing resolution, creating a symbol of time suspended. The pentatonicism of the cadenza similarly symbolizes effortlessness. However, when the orchestra becomes a mobile accompaniment in section A (see Table 9.1), and begins to introduce melodic material derived from the soloist, it takes on a commentating role that develops with the introduction of solos from four bars before rehearsal letter B. The passage for full orchestra at letter D is the first move away from the established modal field. Initially, this proceeds without the solo violin for the first time, and with a more forceful tone (a fuller sonority), but then the soloist joins in and actually effects the climax. In so doing, the soloist moves decisively away itself from representation into commentary;¹⁹ though the climax quickly subsides, it is not until the return of the opening material at four bars after rehearsal letter F that the initial point of view is recaptured, and it could be argued that even here the presentation is more of a reminiscence than ‘the thing itself’ (the reprise is, after all, truncated).

Ex. 9.2. The Lark Ascending, opening.

Section B begins with a new, folk-like melody in the flute, shifting the focus from the sky to ground-level and human activity: the cadenza heard earlier contained folk-like shapes, but these point towards the environment rather than connoting, as now, a community. The soloist’s music is essentially decorative in the first two of the three subsections, B1 and B2. Indeed, at the end of B2, from five bars after rehearsal letter Q, it becomes pure decoration once it has jettisoned the thematic oscillating minor third. This process of liquidation has the effect of emphasizing the soloist’s taking up of the flute melody in the return of B1 at letter R. In some ways this section can claim to be the linchpin of the work, in that the various reworkings create the clearest sense of nostalgia: the soloist’s assumption of the flute melody, the repeated cadences supporting the soloist’s languorous double-stopped descents, and the gently yearning suspensions in the soloist’s subsidiary line from six to ten bars after letter R (one of only two times the soloist has them²⁰) all support the effect. After this the second return of the cadenza material is imbued with an even greater sense of loss, the ambivalent modal final and niente ending intensifying the melancholy as an already distant vision vanishes.

The importance of endings: three concertos and a quartet

The Lark Ascending is obviously a virtuoso work, even though the virtuosity most memorably consists in being still and quiet rather than loud and flashy. But it is also the relationship between the soloist and the orchestra that marks this as essentially a concerto – the clear differentiation of roles if not the outright opposition that is commonly viewed as the essence of Romantic manifestations of the genre. The Concerto in D minor for violin and strings (1924–5), written for the Hungarian-born but London-based virtuoso Jelly d’Aranyi at around the time she gave the first performance of Ravel’s Tzigane (1924),²¹ is cast in a more traditional mould. It was originally entitled ‘Concerto Accademico’, though exactly why has been a source of debate. Hubert Foss sees a hearkening back to pre-Romantic approaches: ‘The implication is not derogatory, as “academic” (a word of different meaning in our modern world) is used in England: only that the form derives from the eighteenth-century concertos of Bach and his fellows’.²² A. E. F. Dickinson wonders whether ‘the ulterior motive is surely some jape, some desire to avoid the burden of a full-blown concerto, by a mock reference to the earlier type’.²³ The Grove writers, meanwhile, see the work as ‘Vaughan Williams’s nearest approach to a Bachian “neo-classicism”’, and Kennedy observes a ‘tightly-wrought synthesis of neo-classicism, folk-dance rhythms and triadic harmony’.²⁴ But it would be misleading to imply that the work represents a serious engagement with neoclassical principles. The structural organization derives from Vaughan Williams’s modal practice, and the neoclassical elements are largely surface phenomena, particularly in the second and third movements. In the first, the debt of the first-group material to baroque shapes and rhythms gives a certain degree of impetus, but has the effect of depriving the music of much sense of personality; and the more folky material, such as the new pentatonic theme that initiates the second half of the development at rehearsal letter L, has the ‘all-purpose’ air to which Vaughan Williams’s thematic invention is sometimes prone. But even if it is recognized that, like many composers who have a natural bent for symphonic composition, Vaughan Williams is generally more impressive for what he does with his material than for the quality of the material itself, the first movement is insufficiently inventive at the formal/structural level to make much impact.

One aspect worth noting, though, is the technique of planing. Ex. 9.3 shows two instances: (a) is the beginning of the pentatonic theme mentioned above, in which the soloist and lower strings, which inhabit the same mode, are pitted against a sustained E♭ in the bass; the more protracted (b), which begins at rehearsal letter R, sees an f″-based layer pitted against a lower G♭-based layer. Both dissonances are quickly resolved, but planing (or bitonality, as it is usually called in the Vaughan Williams literature) takes on greater structural significance in later works, most famously in Flos Campi, completed in the same year as the concerto.²⁵ It is employed in the second movement, too. Ex. 9.4 shows the coda, in which the upper strings elaborate a G triad (with the parallel triadic movement that was firmly established as a Vaughan Williams fingerprint by this stage) while the lower strings descend by step until there results a clash of E♭ against G that is sustained for three bars. As in the first-movement examples, the tension is relatively local, almost colouristic: the E♭ level simply drops away to leave G unchallenged.

Ex. 9.3. Violin Concerto, first movement, instances of planing: (a) rehearsal letter L; (b) rehearsal letter R.

Ex. 9.4. Violin Concerto, second movement, coda, instance of planing.

Tovey says of the ending of the Presto finale that it is the ‘most poetically fantastic and convincing end imaginable’.²⁶ It is certainly the most impressive movement, and again shows Vaughan Williams’s facility for the control of fast music. It is, though, the introduction of a broader melody from rehearsal letter U (a slowed-down version of the theme first stated at a bar after letter B) that sets up the ‘fantastic’ effect. The melody’s alternation between two versions, one in the acoustic mode²⁷ and one in the Dorian, focuses increasingly on the equivocation between major and minor mediants that is fully crystallized at the very end of the movement. This concentration on the play of modal alternatives is possible because of the emergence of a strong tonic: a long dominant pedal beginning at two bars before rehearsal letter U is followed by a stepwise bass descent to D via A♭, G, G♭, F and E.

It can be debated whether the ending of the Piano Concerto (1926–31) is equally effective. The work underwent various revisions in 1933 and 1934, and a version for two pianos was made in 1946, before the revised ending emerged in 1947, changing the closing key from G to B.²⁸ B is the key of a long piano solo that precedes the coda. As discussed further below, this solo passage prefigures the ending of Symphony No. 5 in its diatonic purity and ethereal ascent. But while the closing pages of the symphony complete a work-length trajectory, there is a patchwork quality about the Piano Concerto. This is epitomized by the first of the three movements,²⁹ Toccata, which for all its energy gives an overall impression of being rather static. This is partly due to transitions often being engineered over the last bar or even last couple of beats of a section. But it is mostly to do with the handling of the form. Vaughan Williams’s own rather terse programme note, with its use of the terms ‘development’ and ‘recapitulation’, implies a sonata-form background.³⁰ His listing of the thematic material suggests that the second group occurs at rehearsal figure 3, with a theme in B Aeolian. But this is not accorded the rhetorical force of a traditional second group, or indeed the length one would expect: whereas the work begins with twenty-four bars of C, the transition to the second subject group lasts just three bars; and the second subject group itself lasts only eleven bars (two short sentences of six and five bars in B Aeolian and F Aeolian respectively), before a further transition to the opening material in the tonic. Rather than setting up the traditional tonal dichotomy, Vaughan Williams’s purpose seems to be to set up a solid block of C major from which the rest of the music struggles to move away. The opening is sonically massive, the soloist playing a martellato semiquaver figuration over octave Cs supported by the brass, while the strings have a unison rising marcato line. It is the latter that embodies the straining against C, with a move from C Dorian to more dissonant Locrian inflections. In reference to the traditional expositional repeat, this music returns at rehearsal figure 5 with roles swapped. The hegemony of C is epitomized by the long C pedal in the development from figure 10, but also by the ease with which C returns after the ‘shortened recapitulation’ (as Vaughan Williams describes it) in A♭ for the final, Largamente section: the transition is effected in just three notes.

The Romanza is in ternary form. The opening of section A, which is itself ternary, is notable for its use of planing, the pentatonic tune being placed against various elaborated triads: see Ex. 9.5, which also shows the tune’s chromatic ‘tail’. In the central part of section A, E major emerges as the first clear key to support the initial melodic pivot, B. Such clarity, though, turns out to be fugitive: the harmonic underpinning of the next two phrases returns to ambiguity – a minor seventh built on C♯ major and the overlaying of E major and A minor triads. Section B, meanwhile, is largely diatonic and notable for chains of luminous suspensions: a putatively idyllic foil to the less stable outer sections.

Ex. 9.5. Piano Concerto, second movement, bars 7–13.

The final movement takes elements from the chromatic tail of Ex. 9.5 – in particular the pattern of a semitone followed by a minor third (e.g. C–B–G♯) amplified by the oboe and solo viola in the coda of the Romanza – as the basis for a chromatic fugue subject. This pattern, appearing in the bass, is the chief means of transition back to the tonic C in the first movement: see the bar before rehearsal figure 5 and the bar before figure 13 (this suggests a greater degree of unity across the work than the Grove writers allow). It is this fragmentary use of the hexatonic scale (so called because the replication of the interval pattern semitone–minor third across the octave results in six pitch classes), and the fugal writing itself, that most clearly marks the Concerto as ‘an interesting transition to the Fourth Symphony’.³¹ But it is the final section of the last of the three cadenzas, a reminiscence of section B of the Romanza, that is the most interesting music not only of the movement but also of the whole work. Cast in pure B major, it could be viewed, according to mood, as inspired or as a frankly lazy piece of composition. Ex. 9.6 shows the end of the section. As is the case throughout all of the section, there are three lines, each of which is filled out with parallel triads. The lines have to be arranged so that they are playable with only two hands, but the details of incidence are, I would suggest, otherwise inconsequential. Any dissonance is acceptable in this world so long as the tonic chord arrives at the end: the broad effect is what counts.

Ex. 9.6. Piano Concerto, third movement, end of final piano section.

The Oboe Concerto of 1944 (which, like the Violin Concerto, is for soloist and strings) can be heard as a satellite of Symphony No. 5, completed the year before. The most obvious connection with the larger work is the pastoralism: the first movement of the concerto is actually entitled Rondo Pastorale, and double-reed instruments have long been associated with rural contexts. And the concerto, too, has a cyclic element: a pentatonic theme spanning an octave marks the beginning and end of each movement.

The form of the first movement is outlined in Table 9.2. The movement opens ruminatively, with a gesture not unlike the opening of The Lark Ascending, and the main theme is supported by the relaxed rotation of three chords (i, VII and IV); even the ensuing short cadenza seems to be taking it easy. Momentum picks up in section B, though the varied return of section A at rehearsal letter B eases down before section C picks up the pace again. Section D is quasi-developmental, drawing on the dotted rhythm that characterizes section C; section A themes emerge from rehearsal letter E onwards, literally underneath section D material. It could be claimed that the events in Ex. 9.7, which open a ‘window’ into the return of section B, are the key moments of the movement, ensuring that it ends with disquiet rather than the cosy glow in which it begins: when the window opens, it is on to a very different kind of stasis. C is not a new centre (it underpins much of section D), but the mode, the Lydian, is new (and relatively rare in Vaughan Williams), producing slightly more tension than the Dorian and Aeolian patterns that have prevailed up to this point, especially when the line leaps to F♯ and thus highlights the tritonal and semitonal dissonances with the underlying C triad. The subsequent juxtaposition of C major and C minor furthers the tension, and the alternation of E and E♭ returns to colour the coda, which finally settles on a rather despairing A minor.

Table 9.2 Formal summary of Oboe Concerto, first movement

Notes:

^† G major is the same collection as the preceding A Dorian.

^‡ C Lydian is the same collection as the preceding A Dorian.

Ex. 9.7. Oboe Concerto, first movement, C-Lydian ‘window’ during return of material from section B, 4 bars after rehearsal letter F.

The journey of the final movement, the longest, is rather different. The main material is of scherzo character,³² but the form is essentially sonata form, with a development from six bars before rehearsal letter L and the recapitulation from thirteen bars after letter R after the interpolation of what is in effect a slow movement based on material from the second group. There is a greater sense of melodic expansiveness in this movement, culminating in a broad theme from nine bars after letter V. Kennedy refers to this as a ‘passionate and regretful episode’. Dickinson, in a highly dismissive commentary, sees ‘the penultimate resort to a new tune in G major, traditional in style if not in fact’ as ‘desperate’; furthermore, ‘the abandonment of the original E modal minor is quite unpersuasive’.³³ In fact, the ‘new tune’ is adumbrated at the end of the exposition, and E Dorian is little more than a starting point. It makes more sense to see the shape of this particular movement as being formed by balancing different characters of material and harmonic pacing rather than through tonal structure. The ‘new tune’ has some kinship with the hymnic music of the slow movement of Symphony No. 5, and as in that work the appoggiatura-laden diatonicism is indeed, despite the major mode, tinged with the sadness that Kennedy hears.

The last page of the contemporaneous String Quartet No. 2 in A minor (1942–4) also settles into pure diatonicism, though of the more serenely uplifting kind found in the closing pages of Symphony No. 5 (the correspondence is underlined by the use of the same key, D major, to which the music shifts from the F major of the first part of the movement). The final melodic utterance is given to the viola, which has a particular presence throughout because the work is dedicated to the violist Jean Stewart. The instrument also initiates all four movements. As Jeffrey Richards notes, ‘the main theme of the movement comes from the composer’s sketches for a score to a film about St Joan that never materialized’.³⁴ Regarding overall shape, it is with some justification that Evans writes of a ‘suite-like succession’.³⁵ Dickinson, too, says it is ‘Almost a short suite’, but he is surely wrong to say that it is ‘light-weight’.³⁶ The slow movement, which is the longest (it is the length of the others put together) and one of Vaughan Williams’s least romantic ‘Romances’, is serious in mood, its contrapuntal senza vibrato opening and the ecclesiastical chordal chanting of the contrasting idea and eventual climax suggesting an aspiration towards the tone of Beethoven’s late quartets. And both the first and third movements display a degree of angst not encountered in any of the other works in this chapter. As Richards observes,

The first movement (another bearing the title ‘Prelude’) is in fact the only one that has anything to do with the titular A minor, and this only unambiguously at the end, where it emerges more to articulate the moment of the music’s withdrawal than to celebrate a point of arrival. This moment represents the dissolution of the considerable amount of energy generated at the beginning from a dichotomy between E minor and F minor triads (an opposition also crucial to the Sixth Symphony). Embedded within their oscillation is the semitone–minor third pattern, and this plays an increasing role, leading to several statements of a complete hexatonic scale, A–A, in Violin I towards the end of the development section (from five bars after rehearsal letter G to one bar before letter H). Whilst this does not yet fully focus A as a tonal centre, as the passage includes all twelve pitch classes of the chromatic collection, it does prepare for the eventual outcome; the middle stage in the process is the A minor triad that underpins the G♯-based reprise of the second subject at the beginning of the reverse recapitulation (from nine bars after rehearsal letter H).

The beginning of the third movement is also ambiguous, but in a different way. It begins with a whole-tone motif on viola, G–F–E♭–D♭ (Richards’s ‘stabbing motif’), and during the opening section either G or D♭ can be regarded as the ‘tonic’. Initially, because the rest of the ensemble emphasizes it, it seems that G is the tonic and that the mode is G Locrian. When the violins and cello have the motif and the viola sustains the D♭, that note appears to be the tonic and the mode D♭ Lydian. Both of these pitch classes are later asserted as key centres: G Dorian at the climax at rehearsal letter E, then D♭ at eight bars after letter G before the shift to F minor for the close.

Envoi: Violin Sonata and Tuba Concerto

String Quartet No. 2 received its first performance on Vaughan Williams’s seventy-second birthday. On his eighty-second birthday he was present at the first performance of a work with the same titular key, the Sonata for Violin and Pianoforte in A minor (1954). The role of the titular centre is this time more traditional: the first movement, Fantasia, begins and ends in some form of A (Aeolian at the start; more chromaticized but closing on the A minor triad at the end), and the work concludes with an A major triad. The middle movement, another scherzo, begins and ends in a relatively traditional related area, the subdominant, D. While the quartet is compact and highly focused, and one of the most impressive works discussed in this chapter, the Violin Sonata seems (certainly in the outer movements) diffuse and often routine: Evans’s complaint about the late symphonies that ‘the habits of composition have outlasted the impulse’ is perhaps appropriate here.³⁸ He notes of the Sonata that the Fantasia ‘is in fact a sonata movement, propelled by those minor 3rd shifts prominent in the Fourth Symphony’,³⁹ but the rather episodic – if frequently energetic – approach suggests that ‘Fantasia’ is indeed a more appropriate title. The Scherzo, driven by unpredictable accentuation and harmonic deflection, compels more sustained concentration. While there are moments of comparative quiet – at rehearsal figure 10, for example, and in the coda from figure 14 – rhythmic or harmonic tension is maintained. The Variations of the final movement are based on the theme from the last movement of the 1903 Piano Quintet, delivered at the outset in octaves by the piano. This is representative of the utilitarian aspect that frequently mars Vaughan Williams’s writing for the piano, whose part here often seems like an orchestral reduction: there is little attempt (apart from a few moments in the Scherzo) to explore the instrument’s characteristic sonorities.

The Tuba Concerto (1954) is much shorter (under fourteen minutes in contrast to the twenty-seven or so of the Violin Sonata), presumably in deference to the physical demands of the instrument, whose capabilities Vaughan Williams took ‘great pains’ to explore.⁴⁰ The work benefits greatly from the reduced dimensions: there is a tighter structure, more sharply etched material, and it is much clearer what the composer is setting out to achieve. There is an obvious danger in writing a serious work for an instrument that, because of its size and perceived lack of agility, has often been characterized as comic, but Vaughan Williams takes the tuba into territory far away from caricature, especially in the second movement, Romanza. The music here perhaps risks sentimentality, particularly when the orchestra ‘corrects’ the tuba’s F♮ of the penultimate bar so that the movement ends on a tonic major triad that can seem a little too sweet, but otherwise the composer demonstrates once again the nostalgic, melancholic potential in pure diatonicism (see, for example, the tuba’s appoggiatura-laden descents from its second bar, and in particular the descent to low B, supported by a move from D major to B Aeolian, at the end of the first paragraph).

The outer movements may not be emotionally searching, but they sustain the listener’s interest in simple but effective ways: as Vaughan Williams’s own programme note announces, ‘The music is fairly simple and obvious and can probably be listened to without much previous explanation’.⁴¹ The first movement is an inventive sonata-form hybrid in which the development is concurrent with the recapitulation, such is the degree of reworking in the latter. The rumbustious finale – Rondo alla tedesca – relies on Vaughan Williams’s trademark alternative scale-degrees and chromatic sideslips to generate tension: see in particular the poco animato from five bars before rehearsal figure 9 that leads to the final cadenza. Set alongside the symphonies – including Symphony No. 9 (1956–8), which is relatively compact in comparison with all the others except No. 8 – the Tuba Concerto clearly has reduced ambitions. But it epitomizes Vaughan Williams’s (decidedly unmodernistic) desire, even to the end of his career, to meet the need for a broad range of musical experiences for performer and listener alike. And if it can be said that there is an aspect of ‘lateness’ about Symphony No. 9, the final works in both the chamber and concerto genres have nothing of this about them.

Contexts for the later symphonies

One of the most striking aspects of Vaughan Williams’s long career is his late flowering as a symphonist. When the composer turned sixty in 1932, he had three symphonic works to his name: A Sea Symphony, completed in 1909; A London Symphony, which dates from 1910–14; and A Pastoral Symphony, composed during and after the First World War. Between 1934 and 1957, Vaughan Williams completed six further essays in the genre, establishing himself as one of the great twentieth-century symphonists and contributing vitally both to the consolidation of British symphonism and to the reorientation of the genre after Mahler away from central Europe and towards the Slavic, Scandinavian, Francophone and Anglophone contexts.

Charting this progression is, however, neither historically nor analytically straightforward. The direction the symphonic cycle would eventually take is, for instance, not easily inferred from the first three symphonies. As a multi-movement cantata, A Sea Symphony is closer in conception to Mahler’s Symphony No. 8 or Rachmaninov’s The Bells than to any of Vaughan Williams’s subsequent essays in the genre; A London Symphony prefigures the programmatic aspirations of the Sinfonia Antartica, but not its technical means; and although many features of A Pastoral Symphony are carried forward in later works, as a whole it gives only a sketchy indication of the later symphonic idiom’s expressive and technical range.

Moreover, the division into earlier and later symphonic phases is itself problematic. I have assumed throughout that Symphony No. 4 should be regarded as the first work of a second phase, but this view has not received multilateral endorsement. Writing before the completion of Symphonies Nos. 8 and 9, Frank Howes divided Nos. 1–6 into three groups of two, thereby attaching No. 4 to its predecessor, notwithstanding their chronological distance: as he phrased it, ‘two are pre-war, two inter-war, and two in-war’.¹ Sinfonia Antartica by these terms stands alone as ‘a symphonic poem cast into a formal mould as a symphony’.² Michael Kennedy also distinguishes between a pre-war or ‘Edwardian’ manner in A SeaSymphony and A London Symphony and a more ascetic, Ravelian style in A Pastoral Symphony: however, he then groups A Pastoral Symphony with Nos. 4 and 5 as ‘inter-war’ works; and having the benefit of access to the complete cycle, he attaches No. 6 to Nos. 7–9 as ‘post-war’ symphonies, characterized by their shared debt to the composer’s wartime film scores, most overt in the relationship between the score for Scott of the Antarctic and the Sinfonia Antartica.³ Alain Frogley, by contrast, discriminates between the first two symphonies and a central cycle of four, which he further divides into two non-consecutive groups: Symphonies Nos. 4 and 6 emphasize ‘chromaticism, dissonance, and rhythmic turbulence’, while A Pastoral Symphony and Symphony No. 5 betray ‘a mode of expression shaped by the influence of English folksong, Tudor music and the works of Debussy and Ravel’.⁴ Unlike Kennedy, Frogley treats the last three symphonies separately, noting that ‘Although each of the final group of symphonies has its own particular character, all three works share a tendency to synthesize elements from the two polarized worlds of the symphonies from the “Pastoral” to the Sixth’.⁵ All these perspectives have merit; however, I will argue here for the subdivision advocated by James Day, Elliott S. Schwartz and Hugh Ottaway into early, middle and late trilogies respectively.⁶ Pace Frogley, I perceive a critical shift in technique and concept separating A Pastoral Symphony and Symphony No. 4; and pace Kennedy, I regard Symphony No. 6 as belonging to a line of development initiated with No. 4, apparent in a commonality of technical, formal and expressive means.

This partitioning is supported by contextual factors. Symphony No. 4, completed in 1934, not only occupies a key position in Vaughan Williams’s oeuvre, but is also his pivotal contribution to the interwar rejuvenation of British symphonism. The primary influence for this notion of symphonic modernism was Sibelius, whose music was advocated as a compositional model by a number of influential commentators, notably Cecil Gray and Constant Lambert;⁷ as J. P. E. Harper-Scott has recently observed, Sibelius was ‘the post-war influence of choice for British composers’.⁸ A substantial corpus of works flowed from this context, among them William Walton’s Symphony No. 1 (1934), the seven symphonies of Arnold Bax (1922–39) and the first five of Havergal Brian (1919–37). Vaughan Williams’s identification of Symphony No. 4 with the aspiration to compose a ‘modern’ symphony consequently implied above all a post-Sibelian symphony.

The larger context for Anglo-Sibelian symphonic modernism is the consolidation of what is often termed the ‘nationalist’ symphony, or the types of symphonism emerging outside of the Austro-German sphere in the late nineteenth and early twentieth centuries. If Sibelius and Nielsen furnished the dominant Scandinavian paradigms, coeval interwar strands of development can also be identified in Russia, where a version of symphonic modernism coalesced in Shostakovich’s symphonies Nos. 1–4 of 1926–36, and in the works of expatriate Russian composers, of which the most well-known examples are Prokofiev’s Symphonies Nos. 2–4 (1924–5, 1928 and 1929–30 respectively), Rachmaninov’s No. 3 (1935–6) and Stravinsky’s Symphony in C (1938–40).⁹ To this we must also add Francophone contributions in the wake of Saint-Saëns, Franck and D’Indy, notably Roussel’s Symphonies Nos. 2–4 (1919–34) and Honegger’s Symphony No. 1 (1930),¹⁰ as well as American symphonism in the aftermath of Charles Ives, primarily Copland’s three symphonies (1926–46) and Roy Harris’s Symphonies Nos. 1–6 (1933–44).¹¹ If Vaughan Williams’s Symphony No. 4 adopts an Anglo-Sibelian pose, it also responds to an international symphonic currency, the presence of which is markedly stronger after the First World War.

The following offers a broad analytical commentary on the six later symphonies, isolating salient structural characteristics and pursuing their extra-musical connotations. I have for the most part resisted the temptation to ground this commentary in a single theory or methodology, but instead draw out thematic, formal, tonal and post-tonal processes in each work, relating them where pertinent to commonly averred extra-musical meanings. The distinction between dynamic and static conceptions of form is a central concern, being vital to both the technical and extra-musical dimensions of Vaughan Williams’s evolving symphonic attitude.

Symphonies Nos. 4, 5 and 6

Symphonies Nos. 4–6 are most overtly Sibelian in their address on problems of formal integration.¹² A preoccupation with thematicism pervades Symphony No. 4 especially, both at an intra- and inter-movement level. Vaughan Williams himself identified two themes that, as he put it, ‘run through this Symphony’ (see Ex. 10.1), and also pointed out the derivation of the Finale’s main subject from the flute theme entering in bar 61 of the slow movement, and the reprise of the Symphony’s opening in the Finale’s coda.¹³

Ex. 10.1. Symphony No. 4, two main themes.

The composer’s analytical commentaries are notoriously cursory; here, as elsewhere, the situation is more complex than he suggests. In fact, the work is overlaid with a dense web of thematic links, clarified in Exx. 10.2a–d. As Ex. 10.2a shows, theme ‘A’ is a variant of its precursor (‘a’) from bar 1. ‘A’ recurs in three successively diminished forms in the Scherzo’s introduction, in the Finale’s development (bar 189), and as the first subject of the fugal epilogue. Variant ‘a’ reasserts itself in the coda’s recall of the symphony’s opening. The Scherzo introduction’s third variant of ‘A’ also supplies the accompaniment to a main theme derived from ‘B’, which as Ex. 10.2b shows, is adumbrated at the movement’s opening. Theme ‘B’ is also a critical ancillary idea in the slow movement, forming the substance of the introduction and recurring as a marker of each major climax. In the Finale, bar 177 in the development core supplies a shadowy reminiscence of the first movement’s closing-section theme (labelled ‘C’ in Ex. 10.2c). Lastly, the Finale’s main theme (Ex. 10.2d) retrieves the slow-movement flute theme.

Ex. 10.2(a). Symphony No. 4, examples of theme ‘A’ in first, third and fourth movements.

Ex. 10.2(b). Symphony No. 4, examples of theme ‘B’ in first, second, third and fourth movements.

Ex. 10.2(c). Symphony No. 4, recurrence of first movement ‘C’ theme in the fourth movement.

Ex. 10.2(d). Symphony No. 4, recurrence of second movement flute theme in the fourth movement.

The work’s most blatant cyclical device is, however, the Finale’s fugal epilogue, the complexities of which are hardly conveyed by Vaughan Williams’s bland observation that it combines ‘A’ with the movement’s other subjects. Table 10.1 summarizes its entry structure and the origins of the themes.

Table 10.1 Symphony No. 4, fourth movement, fugal epilogue

The table reveals that the epilogue more specifically consists of four fugues, each pitting ‘A’ against a different counter-subject. The first is a stretto fugue, in which ‘A’ is its own counter-subject, combined both as a whole-note subject and in diminution. The second pits ‘A’ against the Finale’s second main-theme-group idea, labelled ‘A2’, as well as a variant of ‘B’. In the third fugue, ‘A’ confronts the Finale’s second theme, labelled B1, combined from bar 385 with ‘A’ in stretto. The fourth fugue crowns the structure with the combination of ‘A’ in augmentation and the Finale’s main theme, labelled ‘A1’. Fugues three and four are linked by an extended episode, based on the Finale’s closing-section material, labelled C, given in dialogue with numerous variants of ‘A’, as rectus, inversus and (loosely) per arsis et thesis. As Table 10.1 explains, each fugue is organized into subject-answer pairings around a specific complex of pitch levels: fugue 1 is centred on F and C; fugue 2 on G♭, B♭ and E♭; fugue 3 primarily around A and D, with additional entries on F♯, C and E; fugue 4 returns to F and C, culminating on the tonic F.

Ex. 10.3. Symphony No. 5, fourth movement, 4 bars before rehearsal number 14.

Although neither Symphony No. 5 nor No. 6 matches No. 4’s dense cyclical logic, inter-movement relationships are nevertheless present. Symphony No. 5 has one clear cyclical event (Ex. 10.3): the first movement’s initial material interrupts the Passacaglia’s variation sequence four bars before rehearsal figure 14, pitting three of the first movement’s main-theme ideas against the Hauptmotiv of the finale’s ground bass. The latter occurs five times as an elaboration of the bass pedal C, above which the work’s initial horn call, labelled ‘a1’, appears seven times in complete or incomplete forms, counterpointed against the first movement’s principal subject from rehearsal figure 14.8 (labelled ‘a3’), and prefacing the response to the horn call (labelled ‘a2’) in figure 14.14.¹⁴ There are, additionally, three more generalized first-movement correspondences. Firstly, the passage also invokes the first-movement main theme’s accompanimental texture, evident in the consistent dotted figures. Secondly, the C pedal itself recalls the symphony’s opening; its embellishment with the Passacaglia Hauptmotiv is consequently also a cyclical synthesis, through which the harmonic milieu of the first-movement subject incorporates finale material. Thirdly, the first-movement material’s underlying progression from D major to D minor over a C pedal is also retrieved, a recollection having critical structural consequences.

Several issues converge in this passage. In one respect, it constitutes a ‘breakthrough’ in the Adornian sense, influentially defined by James Hepokoski as an ‘unforeseen in-breaking of a seemingly new (although normally motivically related) event’.¹⁵ The Preludio material descends on the finale from outside, interrupting the form’s flow (it interjects between the twenty-seventh and twenty-eighth variations) and imposing an unprepared harmonic event. The bass pedal is emblematic of this characteristic: because the form is driven by the repetition of a bass line, the pedal freezes its structural agent on a single pitch. The passage also elaborates a formal characteristic of the first movement: the Preludio’s analogous material (as far as rehearsal figure 2) establishes a recursive model as the basis of an additive process of varied recurrence acting against the implied sonata context. Because the reprise of this model in the Finale recovers all its elements, and (despite revisions) in the same twenty-bar span, it effectively continues the first movement’s formal process, as if Vaughan Williams had simply deferred its completion in order to accommodate the intervening movements. The ‘breakthrough’ consequently both interjects within the Finale, and reframes the material of the second, third and fourth movements as an interjection suspending the first movement’s formal action.

Symphony No. 6 forsakes No. 4’s dense thematicism and No. 5’s large-scale cyclical recall for more elusive cross-references. The epilogue’s opening retrieves three first-movement elements in a radically altered context (see Ex. 10.4). The implied clash between E minor and F minor in the first movement’s main theme is also skeletally reprised here, the bass settling on an E pedal while the first violins intone the rising trichord F–G –A♭. As Ex. 10.4 demonstrates, the violins’ continuation takes up and inverts the second motivic element of the first movement’s theme. These two ideas saturate the movement: instead of working synthetically with thematic combination (as in Symphony No. 4) or large-scale reprise (as in Symphony No. 5), Vaughan Williams trades in fragmentation: the Epilogue functions as an exhaustive, if understated, liquidation of first-movement material.

Ex. 10.4. Symphony No. 6, relationship between opening themes of first and fourth movements.

The E–F semitonal clash mobilizes a structural principle, which is basic to all three works and further differentiates them from the three earlier symphonies, namely the generative use of harmonic and modal duality. In Symphony No. 4, themes ‘A’ and ‘B’ are related by their common origin in the work’s opening gesture, a semitonal dyad projected both horizontally and vertically, which proves crucial to the entire structural narrative: the motive D♭–C in the violins and winds in bar 1, labelled ‘a’, is its own harmonic context, since D♭ also clashes with the bass C on the first beat of the bar (Ex. 10.5). Transposed, this ninth also forms the end-point of theme ‘B’, which peaks in bar 15 on a G♭ supported by F in the bass; the quartal ascent with which B begins is thus really a prefix to a harmonic variant of ‘a’. As if to clarify the relationship, Vaughan Williams reprises the initial material from bar 20, transposed so that the G♭–F dyad is reiterated.

Ex. 10.5. Symphony No. 4, first movement, semitonal clashes in themes A and B.

As the symphony progresses, ‘a’ is accorded greater harmonic substance (see Ex. 10.6). The first-theme return initiating the first movement’s development section transposes ‘a’ as E♭–D, the latter harmonized within a second-inversion D minor triad. At the start of the recapitulation, ‘a’ acquires a harmonic character with which it is associated for the rest of the piece: the G♭ participates in a G♭ minor triad, the third admittedly spelled enharmonically as A♮; and the dissonant ninth is now the frame for a clash between an F minor triad and a chromatic neighbouring chord. The slow movement’s twin introductory statements of ‘B’ harmonize its final pitch in a similar way, as Ex. 10.6 indicates: each statement peaks on a D♭ minor chord supported by a bass C. This in effect summarizes the movement’s gestural trajectory: its two pivotal climaxes (bars 56–7 and 90–1) are both signalled by the return of ‘B’ untransposed, bringing successive processes of melodic intensification to a point of culmination on the same critical sonority.

Ex. 10.6. Symphony No. 4, transformations of themes A and B.

In fact, the dissonant neighbouring chord is never allowed fully to resolve. The reprise of the symphony’s opening from bar 444 of the Finale retrieves the first movement’s grating G♭ minor–F minor clash. The work’s last harmonic gesture, in bars 451–5, categorically refutes the possibility of any smooth accommodation of G♭ within the tonic context. Theme ‘B’ builds in the brass towards a tutti, fortissimo reiteration of the F/G♭ sonority. Rather than a smooth descent of G♭ on to the tonic, Vaughan Williams isolates this dissonance with an abrupt caesura, and the final chord brutally enforces F at the expense of its modally defining third.

Arnold Whittall has noted a similar ‘destabilizing contrast’ in the opening paragraph of Symphony No. 5, albeit clothed in a very different expressive garb. The critical ambiguity here is between C as tonic and C as neighbour to D. Whittall pursues this as a basic structural device;¹⁶ elaborating on this reading, we can identify two successive ambiguities in the Preludio, both arising from the conflict between C and D as potential final tones. In bars 1–8, this is expressed as a tension between Lydian C and Mixolydian D: if we orientate our hearing around the bass C, then the music projects the former; if we orientate it around the horn call, then C behaves as the bass of a $$ chord, and the music projects the latter (Ex. 10.7).

Ex. 10.7. Symphony No. 5, first movement, opening, and alternative modal analyses.

The exchange of F♯ for F♮ in bar 9 introduces the second duality: we now have to choose between Ionian C and Dorian D. In both cases, the fundamental relationship is the major second C–D. At the end of the first movement, this tension is left unresolved. There is no convergence on a stable tonic; instead, the C–D dyad persists in the final bars without harmonic elaboration. The catalyst for resolution comes with the finale’s ‘breakthrough’ reprise of the symphony’s opening, in the wake of which the bass C is dispelled in favour of a grounding D pedal.

In Symphony No. 6, the question of modal conflict gains complexity, because the work’s initial material implies at least four possible modal origins, all of which the work exploits (see Ex. 10.8). At the start, the notation posits a conflict between semitonally distant minor modes (F and E) that is redolent of Symphony No. 4. Towards the end of bar 1, however, a contrary impression emerges of modal mixture over the same root: respelled enharmonically, the soprano A♭ could also be the raised third of E major. The material in bar 1 also implies two symmetrical pitch collections: altogether, it forms a tetrachord of the octatonic collection (4–3 in set-theoretical notation); the A♭–G–E tail suggests a hexatonic origin (3–3 being a subset of the hexatonic set 6–20), a context into which the underlying E minor triad also fits.

Ex. 10.8. Symphony No. 6, first movement, opening, and alternative modal analyses.

Vaughan Williams deploys these contrasting implications strategically throughout the symphony. The sense of modal mixture comes to the fore in the first movement’s coda. From bar 159, the reprise of the subordinate theme first heard from bar 82 exchanges E minor for E major, but the harmony moves freely between the two modes, and the climactic return of the opening in bars 187–190 respells A♭ as G♯, temporarily dispelling the incipient E minor/F minor duality. Later on, the hexatonic trichord gains harmonic significance. The climax in bars 117–29 of the slow movement alternates G♭ major and G minor triads over a B♭ pedal, anticipating in transposition the oscillating D♯ major–E minor progression with which the symphony ends. This arises from the pairing of two inversionally related forms of the trichord 3–3 around an ‘axial’ central pitch (Ex. 10.9). G♭ major and G minor hold the pitch B♭ in common; the pitch content of the two chords laid out in succession produces the five-note set 5–2.¹⁷

Ex. 10.9. Symphony No. 6, second and fourth movements, transformations of pairs of inversionally related 3–3 trichords.

In addition to their structural significance, these techniques are also programmatically suggestive, being deployed in the service of symphonic narratives, which, despite Vaughan Williams’s tight-lipped responses to such suggestions, resonate irresistibly with their cultural and political circumstances: for No. 4, the looming European crisis of the 1930s; for No. 5, written between 1938 and 1943, the war itself; and for No. 6, completed in 1947, its atomic aftermath.¹⁸

Symphony No. 5’s crucial interpretative associations have been, firstly, its apparent withdrawal from No. 4’s aggression, and secondly, its relationship with the contemporaneous opera on John Bunyan’s The Pilgrim’s Progress, completed in 1949. The former characteristic habitually licences interpretation as a plea for peace in the midst of war, the latter as a journey towards a musical analogy of Bunyan’s ‘Celestial City’, attained in the work’s valedictory coda.¹⁹ Attractive though these readings are, they conceal other, less pointedly metaphysical agendas. Symphony No. 5 for instance projects a suggestive music-historical narrative, evinced in the Preludio by the mediation of pre- and post-tonal idioms. The dualism of the opening is founded on pre-tonal modality, albeit projected as an interpretative simultaneity rather than a unitary resource. In contrast, the middle section exploits pentatonic heterophony: linear projections of the pentatonic collection are overlaid contrapuntally, such that the collection is also expressed as a vertical sonority, most forcefully in the climax in rehearsal figure 10.5–11. Between these extremes, Vaughan Williams introduces sequential and cadential progressions mingling archaism with post-tonal triadic harmony. We first encounter this in rehearsal figure 6.3–6.5, which establishes a 10–5 linear intervallic pattern followed by a modal v⁷–I cadence, which echoes the so-called ‘English’ cadence in its use of a lowered leading note. Shortly afterwards, a variant of this progression closes the section in rehearsal figure 6.10–6a.1; the cadence subsequently appears in G in rehearsal figure 13.4–13.5, and the whole progression recurs in its most expansive form as the substance of the climax in rehearsal figure 13.15–14.1.

The dialogue of these dimensions is rich in interpretative possibilities. The contrivance of a post-tonal medium from pre-tonal materials narrates the revivification of English music through recovery of its pre-tonal ancestry, a theme explicit in Vaughan Williams’s music at least from the Tallis Fantasia onwards. In the Preludio, this dialectic of tradition and innovation is marshalled as an agent of nostalgia: the modal cadences emerge as glimpses of a lost musical order, the subsequent dissipation of which is each time as pronounced as its mounting presentational force. The cadences invoke not simply a pre-war world, but more accurately a pre-modern one, antedating the tortuous conditions of musical modernism and political catastrophe.

Yet because it is caught between modal dualism and heterophony, the sequence–cadence model also cohabits with the arbiters of its own salvation, made most explicit at the end of the finale. The quiet resolution of the C/D duality on to a D pedal in rehearsal figure 15.2 ushers in a coda that projects a synthesis of the three dimensions introduced successively in the Preludio. The modal dualisms are dispelled, because the diatonic D modality is entirely consistent with the D pedal in the bass, but the harmony is modal-heterophonic rather than tonal, the various lines combining so as persistently to create diatonic pitch clusters, which only clarify to triadic harmony in the final D major chord. Vaughan Williams’s ‘Celestial City’ is a new musical order, in which there is no contradiction between modal and post-tonal harmonic means.

Although sharing Symphony No. 5’s mediation of pre- and post-tonality, Symphonies Nos. 4 and 6 by contrast suggest narratives of negation and resignation. Symphony No. 4 mobilizes the high-classical summative finale – apparent in the Beethovenian elision of Scherzo and Finale and the contrapuntal coda after Mozart’s ‘Jupiter’ Symphony – but in order to deny its potential for synthesis.²⁰ Vaughan Williams’s forcefully end-directed structure decisively rejects transcendental affirmation by dissociating thematic combination and resolution of the symphony’s propulsive structural dissonance. The synthetic efforts of the finale’s fugal epilogue come to nothing, because, at the end, the work’s generative modal duality remains starkly exposed; as Wilfrid Mellers writes, ‘Fugal unity fails; it cannot engulf, or gobble up, the contradictions’.²¹

Symphony No. 6 reformulates the same narrative of failure in valedictory terms. As with Symphony No. 4, so here also the narrative’s expressive core is the failure to attain the synthesis of modal duality through counterpoint. The Scherzo, for instance, narrates an unstable dialogue between a counterpoint that persistently congeals into melody and accompaniment, and a thematic homophony that persistently dissolves into counterpoint. The movement’s opening, analysed in Ex. 10.10, clearly reveals this. In bars 1–7, Vaughan Williams piles up successively diminished forms of the movement’s principal motive, but the entry structure and the diminution process are irregular: the ‘answer’ in bar 3 is initially literal, but the subject’s note values are diminished from bar 5; the entry in bar 5 is a strict diminution; and the entry in bar 6 shifts between diminution and double diminution. Bars 1–8, in brief, initiate a contrapuntal form, which they cannot sustain. The Shostakovichian music in bars 8–19 contrastingly comprises two themes (in bars 8–15 and 16–19) presented against syncopated chords, which fail to maintain their simple melody-and-accompaniment texture, instead tending increasingly towards counterpoint.²²

Ex. 10.10. Symphony No. 6, third movement, opening.

The Epilogue is not, as is sometimes suggested, a ‘subjectless’ fugue; rather, each entry cultivates a tension between stable and variable contrapuntal features, which, as we have already seen, are derived from the first movement. In one sense, the Scherzo’s contrapuntal impulse is fulfilled here, because the fugal entry structure is unchallenged as the Epilogue’s guiding formal principle. Yet we are still a long way from genuine counterpoint: fugal writing is made to serve a textural rather than a contrapuntal end. The oft-noted tonal ambivalence of the closing bars again signifies an abyss-like distance between the synthetic aspirations of contrapuntal forms and their ultimate impotence in the face of an irreparable rift in the music’s tonal fabric. Whereas in Symphony No. 4 this rift is formulated as violence, in Symphony No. 6 it informs a symphonic ‘farewell’, to which the oscillating closing triads furnish a gestural vanishing point. The parallel with Mahler’s Symphony No. 9, the Rondo Burleske and Adagio of which also contrast dissonant counterpoint and valedictory leave-taking, is striking.²³

Symphonies Nos. 7, 8 and 9

Although Symphony No. 6 and the Sinfonia Antartica are both coloured by Vaughan Williams’s wartime turn to film composition and hold technical characteristics in common, they are in many respects radically different, diverging particularly in their conceptions of large-scale symphonic process. Notwithstanding numerous inter-movement links, Sinfonia Antartica conveys a much weaker sense of dynamic trajectory, replacing linear narrative with a more episodic manner and a frequent concentration on static structures.²⁴ In truth, the last three symphonies are all in various ways imbued with these qualities, suggesting that the dialogue with modernism evident in Symphonies Nos. 4–6 has here ceded to a rather different perspective.

The genesis of the Sinfonia Antartica (1949–52) in the sketches for the music to the film Scott of the Antarctic (1948) moreover points to an overtly programmatic agenda, which locates the Sinfonia in a different symphonic subgenre. If Symphonies Nos. 4–6 engage with the ‘absolute’ symphonic model from Beethoven’s Fifth to Sibelius’s Seventh, then the Sinfonia is a programme symphony in the vein of Liszt’s Dante, Tchaikovsky’s Manfred and Rimsky-Korsakov’s Antar; but it embodies a uniquely twentieth-century take on this subgenre, supplanting the nineteenth-century affiliation of symphony and literary or poetic sources with a debt to filmic narrative.²⁵ The Sinfonia remains true to its genealogy in its looser attitude towards generic markers: it has five movements rather than four; and although the internal movements expand upon the traditional scheme – two dance movements (‘Scherzo’ and ‘Intermezzo’) frame a slow movement (‘Landscape’) – the outer movements pay no more than lip service to sonata and rondo forms. The programmatic lineage is also underscored by the provision of literary epigrams for each movement, from Shelley’s Prometheus Unbound, Psalm 140, Coleridge’s Hymn before Sunrise in the Vale of Chamouni, Donne’s The Sun Rising and Scott’s last journal respectively.

The composer’s programme note indicates four specific associations between film and Sinfonia.²⁶ Vaughan Williams explains that a ‘theme accompanied by deep bells’ in the first movement (rehearsal figure 12.9–12.16) ‘was supposed in the film to be “menacing”’. Two ideas in the Scherzo are then curtly related to more specific references: the first is associated with ‘whales’; the second ‘was used in the film to suggest penguins’ (the latter first appearing at rehearsal figure 5.4). Finally, the music ‘connected in the film with the death of Oates’ appears in the fourth movement at figure 7, following a recurrence of the Prelude’s ‘menacing’ theme.

Thanks to the work of Daniel M. Grimley and others, we now know that the links between film and Sinfonia run somewhat deeper.²⁷ Most obviously, Vaughan Williams fails to mention that the first movement’s opening theme is used, in an abbreviated form, to accompany the film’s opening credits; this material recurs both as the film’s postlude and in the last section of the Sinfonia’s Epilogue. Other relationships include the nature music accompanying the shots of the Antarctic wilderness in the film’s Scene 1, which became the second section of the Prelude and the closing music of the Epilogue; the accompaniment to the ascent of the Beardmore Glacier in Scene 6, which is used twice in the ‘Landscape’ movement from rehearsal figures 3.13 and 8.4 respectively; a brief fragment of the Intermezzo’s second section (figures 2.7–6.1) used in the conversation between Scott and his wife Kathleen in Scene 2; and the theme forming the basis of figures 5.6–5.9 in the Epilogue, which first appears in the film’s Scene 7 to depict the arduous process Scott describes as ‘manhauling’, the transporting of sleds and equipment on foot. Grimley’s study of the film score’s sketches reveals that Vaughan Williams gave the germinal forms of many of these ideas extra-musical tags. The music opening both film and Sinfonia is, for example, labelled ‘Heroism’, whilst the ‘nature music’ that follows it (from figure 5.6) represented the ‘terror and fascination of the S. [sic] Pole’, and the music with which the Intermezzo opens was conceived as a valse triste depicting Wilson’s wife Oriana.

The Sinfonia’s movement scheme is balanced by two means: the work as a whole is framed by the ‘heroism’ music, perhaps the most overt debt to the structure of the film; and a number of key motives are recycled at critical junctures, affording if not large-scale integration, then at least a degree of narrative continuity. Ex. 10.11 shows three variants of a pivotal idea drawn from the ‘heroism’ theme, marked ‘a’: its presentation in the Prelude’s main theme; the variant heard after the ‘death of Oates’ music in the Intermezzo; and its first appearance in the Epilogue.

Ex. 10.11. Sinfonia Antartica, transformations of ‘a’ theme.

Taken together, these occurrences chart a process of dissolution. In the Intermezzo, Vaughan Williams detaches the theme’s tail and varies its rhythmic design, metrical character (3/4 becomes 4/4) and pitch content. The crisis of metrical identity becomes more acute in the Epilogue: despite the 4/4 context, the theme initially retains its 3/4 groupings, but this soon becomes displaced against both the notated and implied metres, eventually forcing a temporary shift to 3/2. The gradual weakening of the theme’s stability reflects its programmatic contexts: heroism is expressed straightforwardly at the start, in the shadow of death in the Intermezzo and of tragic failure in the Epilogue, under which conditions the theme’s identity starts to buckle.

A pervasive technique in the Sinfonia is the composer’s habit of opposing post-tonal triadic harmony with a sort of collection-based heterophony. Typically, the former is associated with the human themes of heroism, adversity and death, while the latter underpins much of the work’s nature music. The ‘heroism’ theme, quoted in Ex. 10.12, gives the clearest example of the first type.

Ex. 10.12. Sinfonia Antartica, first movement, opening; chord transformations; bitonal projections.

In one sense, the harmony here can be understood as employing three kinds of chord transformation: e♭–G and b♭–D are fragments of major-third cycles, and as such have hexatonic origins; G–b♭ is a minor-third transformation, having an octatonic basis; and G–a♭–G is a semitonal shift, which David Lewin has termed a ‘SLIDE’ transformation.²⁸ At the same time, the passage also projects a kind of bitonality, alternating between E♭ minor and G major. This has large-scale ramifications: the Sinfonia’s framing tonality is G; but E♭ and its relations often intrude as a counter-pole. Tellingly, and in a manner redolent of the end of Symphony No. 5’s Preludio, the work ends not by confirming G, but by juxtaposing a G bass pedal and a melody that comes to rest on an E♭.

In the Prelude, the ‘heroism’ music is followed from rehearsal figure 5.6 by the first ‘nature’ music, which clearly demonstrates the turn to a collection-based mentality (see Ex. 10.13). The basis of the passage is the alternation of a $$ chord with its tritone transposition, redolent of the ‘Chez Pétrouchka’ scene of Stravinsky’s Petruschka. The underlying collection here, however, is not the octatonic set 8–28, but as Ex. 10.13 shows, set 8–25, expanded to set 9–8 at rehearsal figure 6.1, where the strings introduce an E♭. Unlike the opening, where melodic material elaborates a triadic scheme, here harmony results from the vertical projection of a pitch collection.

Ex. 10.13. Sinfonia Antartica, first movement, ‘nature’ music.

These two harmonic attitudes play a fundamental role in supporting the work’s programme, embodying not just the opposition of the human and the natural, but also that of symphonism and a repertoire of ecological ‘noises’, which, as Grimley explains, ‘are essentially unsymphonic, in that they are almost entirely static and resist any attempt at development or change’.²⁹ For Grimley, this juxtaposition functions symbolically, aligning the futile heroism of Scott’s expedition and the tragic heroism of the Great War, an affiliation that was prominent in the public consciousness. He thereby reads the Sinfonia as ‘a monument to the war dead’.³⁰

Each movement formulates the polarization of humanity and nature in a different way; and each perspective is clarified by Vaughan Williams’s prefatory poetic and literary epigrams. Thus the Prelude’s pitting of human endeavour against impassive nature is reflected in Shelley’s ‘woes which hope thinks infinite’ in the face of ‘power which seems omnipotent’, whilst Shelley’s conclusion that such struggle ‘is alone life, joy, empire and victory’ is embodied in the prematurely triumphal coda. The Scherzo and the ‘Landscape’ movement are both broadly naturalistic, reflecting scenes 3 and 6 of the film, which respectively portray life at the coastal base camp and the ascent of the Beardmore Glacier. In the second movement, the quotation from Psalm 104 succinctly appraises both the Scherzo material, which marshals a barcarolle topic to underscore the nautical context (‘There go the ships’), and the whales and penguins of the Trio (‘And there is that Leviathan’). Similarly, the transfer of the film’s glacier music into the ‘Landscape’ movement suggestively parallels the ‘Motionless torrents’ and ‘Silent cataracts’ of Coleridge’s poem.

The Intermezzo’s juxtaposition of love and death attends instead to a dichotomy within the human pole. It comprises a loose ternary form in which the A and A¹ sections employ material designed to represent Wilson’s wife Oriana and Scott’s wife Kathleen, contrasting a B section built from the ‘menacing’ theme and the music for Oates’s death. The sectional contrast here stresses a physical distance between the protagonists’ wives in England and Oates’s Antarctic demise, which, however, Donne’s poem transcends (‘Love, all alike, no season knows, nor clime’). Finally, the Epilogue retrieves the heroic theme, here couched as a tragic march. The ultimate collapse back into the Prelude’s main theme and vocalise nature music confirms a tragic circularity, which circumscribes human endeavour: Scott’s failure does nothing to quell human folly and has no effect on impassive nature. In symphonic terms, there is ultimately no teleological fulfilment, simply the reassertion of the germinal dichotomy.

If the retreat from ‘dynamic’ form in the Sinfonia Antartica results from its programmatic affiliations, in Symphony No. 8 (1953–6) it arises from an abstract preoccupation with subverting the teleological processes of the post-classical symphony. The outer movements, entitled ‘Fantasia’ and ‘Toccata’ respectively, both declare a baroque genealogy, which sidesteps classical and post-classical attitudes. And although the Scherzo and Cavatina seem more generically conventional, Vaughan Williams isolates them within the overall design as showcases for the winds and strings respectively (‘per stromenti a fiato’ and ‘per stromenti ad arco’). The symphony as a whole consequently projects a textural strategy, in which the tutti is established in the Fantasia, dismantled in the inner movements, and then re-assembled in the Toccata. The resulting instrumental differentiation resembles a stretching-out of the baroque concerto grosso’s ripieno/concertante distinction, to the point where it conditions an entire symphonic movement cycle. The baroque atmosphere is reinforced by various material allusions, most blatant of which is the quotation of the first strain of the chorale ‘O Haupt voll Blut und Wunden’ from Bach’s St Matthew Passion, which comprises the Cavatina’s main theme.³¹ Cyclical thematic connections are present (the flourish at the start of the Scherzo is, for example, a variant of the Fantasia’s initial motive), but they act as passing allusions rather than supplying a consistent, overarching process.

The design of the individual movements reinforces the withdrawal from teleology. The Fantasia’s ‘Variazioni senza Tema’ engineer a twofold structural displacement. The professed absence of a theme deprives the variations of their normal frame of reference, thereby removing the customary sense of the accumulation of variants around a fixed melodic-harmonic framework. Instead, Vaughan Williams exploits the developmental potential of three germinal motives stated in the first variation, quoted in Ex. 10.14 as ‘a’, ‘b’ and ‘c’, which are developed freely in the six subsequent variations, before being revisited in the brief coda. This produces a comparable displacement at the level of the movement cycle: the variations replace the generically normal sonata form and in so doing abandon the teleological impulse that sonata form customarily installs as a symphonic starting point.

Ex. 10.14. Symphony No. 8, first movement, motives ‘a’, ‘b’ and ‘c’.

The sense of goal-directedness is never recovered. The Scherzo is both self-contained and yet curiously provisional. It falls into six sections: sections 1–3 (rehearsal figures 0.1–2.12, 2.13–3.8 and 4.1–6.7) present four themes in succession, two in sections 1 and 2 and two in section 3, all couched in a simple melody-and-accompaniment context. Section 4 (figures 6.7–10.10), in contrast, is an increasingly dense fugato on theme 4, curtailed at its point of maximum density by section 5 (figure 10.11), which in its metrical, topical and textural character resembles a trio. Section 6 (figure 13.8) functions both as a coda and a heavily truncated reprise, retrieving first the opening and then the fugato. The result is an unbalanced implied ternary design, reflecting Vaughan Williams’s predilection for truncated reprises: B and A¹ together occupy the last 55 bars of a 181-bar structure. This scheme is weakened further by the tonal plot, which is not discretely correlated with the regions of the ternary form: sections 1, 5 and 6 are all centred on a Phrygian C minor; sections 2–4 are orientated around D minor. Altogether, the movement is understood better as projecting two formal ideas, which are mutually negating: an additive, sectional design, which is brought to a halt by the trio and has no time to reassert itself; and a ternary scheme, which is initially swamped by the sectional form and reasserts itself too late to balance the movement’s proportions. The sense of imbalance is compounded by the coda’s last-minute contrapuntal gymnastics, through which prime and augmented forms of theme 4 pile up in stretto and combine with theme 1, and by the elusive ending, which undermines any sense that closure is synonymous with the attainment of a structural goal.

The symphony comes closest to retrieving a sonata dynamic in the Cavatina, in which two contrasted sections (rehearsal figures 1–3.3 and 3.4–5.6) are reprised (from figure 8.14), following a rhapsodic intervening development. Yet this hardly compensates for the form’s absence in the outer movements; rather, any sense of an emergent sonata process is constrained by the material’s expressive intimacy. The Toccata embraces a greater sense of continuity, but again this is not grounded in structural conflict. It has the character of a loose rondo, in which contrasted episodes alternate with freely varied returns to the triumphant main theme first proposed in bar 9. The movement exploits two generic implications: it transfers the properties of the ostinato-based keyboard piece into an orchestral context; and it references the genre’s older usage as an operatic intrada, most famously in Monteverdi’s Orfeo, an expressive sense that the Finale paradoxically projects.

Symphony No. 9 occupies a middle ground between the extremes of programmatic depiction and abstraction evident in its two predecessors. Its form is if anything more conventional than that of No. 8, comprising a four-movement scheme, with an (admittedly idiosyncratic) sonata-type first movement, Andante, Scherzo and Finale. At the same time, as Alain Frogley has identified, its genesis is bound up with programmatic ambitions more obviously redolent of the Sinfonia.³²

The types of stasis evident in Nos. 7 and 8 play a central role in Symphony No. 9. The first movement’s engagement with sonata form is calculated to undermine key features of its structural dynamic, the operative technique being ‘[t]ension between the static tendencies of juxtaposition and the dynamism of expansion and development’, as Frogley puts it.³³ In the exposition, this is evident in the relationship between first and second themes. Vaughan Williams makes no clear distinction between a closed first theme and a modulating transition, instead framing the whole passage within a Phrygian E minor. The second group, following in a mixed mediant minor/major from rehearsal figure 4, is by contrast formally and tonally open-ended. Sectional divisions ensue at figures 8 and 10, but neither demarcates an obvious closing section, either tonally or gesturally. The music at figure 11 is at least grounded on a G pedal, and so as Frogley suggests potentially rounds off the group, and with it the exposition; the melodic momentum of figure 11.8, however, pushes into figure 12, which is harmonically mobile.³⁴ In the recapitulation, the relationship between stasis and mobility is reversed: the first group is now mobile and the second is framed by the tonic. A putative first-theme reprise enters at rehearsal figure 15, where its diminution appears in the tonic in the strings. This does not settle, but rather moves rapidly through a succession of key centres, attaining D minor by figure 16 and reaching a climax on a modally mixed C tonality by bar 17.5. The stability of this recapitulation is further challenged, retrospectively, by the coda, beginning at figure 23, which recovers the first theme’s initial form, key and orchestral texture, thereby fostering the impression that the coda is in fact a first-theme return in a reversed recapitulation. Despite localized chromaticism, the second theme’s reprise from figure 19 is altogether more secure, being circumscribed by the tonic and projecting the character of an enclosed formal episode.

The tendency to juxtapose dynamic strategies and episodic designs comes to the fore in the Finale. The movement begins with a large expository phase consisting of three main sections (rehearsal figures 0.1–2.5, 2.6–5.6 and 6.1–7.5), which is immediately reprised in varied form (figures 8.1–14.3). Rehearsal figures 16–31 are then cast as a single developmental span, culminating in the C major climax at figure 29.1 (the score contains no figure 15, thanks to a cut the composer made before publication). This music, however, is based almost exclusively on a new theme, presented in quasi-fugal fashion from figure 16.1. The result is that the functions of presentation and development are separated out, such that the latter does not flow from the former: excepting the return from figure 27 of elements of the post-cadential idea first appearing at figure 13, the development undertakes a self-contained process, which does not draw upon the first half of the movement. Rehearsal figures 32–35 then furnish a broad coda, which returns to the Prelude’s main theme and pulls the music back towards E major. The form seems purposely to isolate episodic and developmental ways of organizing material, such that each is denied the support of the other: the developmental potential of the initial material is not realized; and the development is also of necessity presentational, evolving out of its own initial thematic statement.

At the same time, Symphony No. 9 revisits thematic strategies more readily associated with Symphonies Nos. 4 and 6. The first movement’s ostensibly rhapsodic second theme is linked to the first theme by the motive in its second bar, identified in Ex. 10.15 as ‘b’.³⁵ This idea makes its first appearance in the saxophones in rehearsal figures 1.3–1.6, thereafter infusing the first group. Subsequently, Vaughan Williams subjects ‘b’ to a variety of transformations, the most dramatically effective of which is that employed at the climax of the first-theme reprise.

Ex. 10.15. Symphony No. 9, first movement, relationship between first and second themes.

The first theme’s Phrygian inflection furthermore supplies a cyclical device that recalls Symphony No. 6. Ex. 10.16 displays the theme and a number of critical relationships within and between movements, isolating the germinal Phrygian second as ‘a’. In a manner redolent of Symphony No. 4, ‘a’ is given harmonic substance from rehearsal figure 1.1 by the saxophones’ e–f–e triadic succession. As Frogley observes, the Scherzo’s initial gesture relocates ‘a’ around an F tonic, and its main theme develops out of this in its compound-melodic conflation of two chromatic neighbour-note patterns.³⁶ Frogley has also noted the debt the Finale’s first theme owes to ‘a’;³⁷ and the symphony’s opening is recalled directly in the coda, both in the recovery of a variant of ‘a’ from bar 32.3 and in the saxophones’ neighbour-note figure in bar 34.3.

Ex. 10.16. Symphony No. 9, appearances of ‘a’ in first, third and fourth movements.

In contrast with the Sinfonia Antartica’s mediation of film, symphony and historical narrative, Symphony No. 9 adopts a more straightforward literary affiliation. The association with Thomas Hardy’s Tess of the D’Urbervilles has been understood at least since Michael Kennedy drew attention to it in 1964, but has been given its most substantial formulation by Frogley.³⁸ His investigation of the sources reveals numerous connections between the material and Hardy’s novel, ranging from the generalized description of the first movement as a ‘Wessex Prelude’ (the domain of Hardy’s novels in general), through various links with associated locations (Stonehenge, Salisbury, Salisbury Plain) and characters (Tess herself) to the narration of parts of the novel.

The most sustained engagement with Tess takes place in the second movement, which Frogley identifies as tracing the action of chapters 58 and 59, from the flight of Tess and Angel Clare to Stonehenge after the murder of Alec D’Urberville, to Tess’s capture and execution. Ex. 10.17 shows the themes to which Vaughan Williams gave specific programmatic tags. This material is arranged into a ternary design with coda, such that the ‘Stonehenge’ and ‘barbaric march’ themes form the A and A¹ sections and the ‘Tess’ theme sustains B and the coda. For Frogley, this scheme shifts between specific narrative and character portrait: section A depicts Tess’s and Angel’s night-time arrival at Stonehenge and her arrest at dawn; A¹ describes Tess’s execution at Winchester (Wintoncester in the novel), the eight deep-bell strokes between rehearsal figures 14 and 19 signifying the town clocks chiming the hour of execution; the B section rather constitutes a ‘character study’ of Tess herself. The coda then has a transcendental function mirroring the end of the novel, Tess finally being released from the ‘sport’ of the gods.

Ex. 10.17. Symphony No. 9, second movement, opening and ‘Tess’ theme.

This by no means exhausts the symphony’s extra-musical associations. Frogley identifies numerous tangible allusions in other movements, including links between the Finale’s development theme, Elgar’s The Apostles and plainsong, the opening of the Prelude and the first chorus of the St Matthew Passion, and the main theme of the Scherzo with The Sorcerer’s Apprentice, as well as numerous inter-opus connections within Vaughan Williams’s oeuvre.³⁹ Drawing all of this together, he proffers a programme for the symphony that returns again to the problem of wartime sacrifice and the futility of heroism:

At a broader level of formal process, the agent of Vaughan Williams’s struggle with such existential questions is, in Symphony No. 9 and in the last three symphonies generally, surely the conflict between dynamic and static modes of symphonic discourse. In the Beethovenian tradition, the dynamic model of form is intimately related to the possibility of transcendence, or ‘utopian semiosis’ in Michael Spitzer’s apt phrase.⁴¹ But in Vaughan Williams’s late symphonies, this impulse is repeatedly halted by episodic structures: the nature music and the ultimate circularity of the ‘heroic’ music in the Sinfonia Antartica; the evasion of post-classical symphonic forms in Symphony No. 8; and the dissociation of presentation and development in the Finale of Symphony No. 9. Whereas Beethovenian dynamism still drives Symphonies Nos. 4–6, even as a negative presence, in the final three symphonies a more pessimistic, pointedly postmodern vision emerges, which suggests that the utopian aspirations of modernity are enslaved by a futile, goalless circularity.

Conclusions

In 1955, Donald Mitchell published two articles spearheading a critical reaction against Vaughan Williams, which have fundamentally coloured the composer’s subsequent reception.⁴² Where earlier critics marshalled superlatives, Mitchell sought to expose Vaughan Williams’s art as flawed and transient. Vaughan Williams’s music has of course endured this critique, and recent scholarship has done much to balance the Oedipal hostility it engendered. Nevertheless, its lingering consequence has been an ex post facto alignment with a kind of establishment conservatism, contrasting the modernism of a generation of post-war British composers who looked towards central Europe rather than England and Scandinavia for their model of musical modernity (what Dai Griffiths has acerbically called ‘grammar schoolboy music’).⁴³

The myopia of the post-war critique is nowhere more forcefully exposed than in the later symphonies. Even cursory engagement reveals a post-tonal vocabulary, which is no less progressive than that employed by Bartók, Stravinsky, Prokofiev or Shostakovich. Neither is this one isolated facet of Vaughan Williams’s compositional personality: indeed, a remarkable characteristic of this music is its derivation of post-tonal chromaticism and luminous modality as possibilities within an integrated mode of expression. Symphonies Nos. 4 and 5 demonstrate this very clearly: the semitonal bitonality of the former and pre-tonal modality of the latter both stem from the same underlying concept. There are, moreover, practices in these works that run considerably ahead of their time. The collection-based textures of the Sinfonia Antartica’s nature music, for instance, prefigure heterophonic techniques exploited by the second generation of the post-war avant-garde, most obviously in the spatial works of Ligeti’s first maturity (Lontano, for example).

Yet perhaps the most compelling argument for the relevance of these works is the ways in which they not only absorb, but also critique twentieth-century modernism. Symphonies Nos. 4–6 engage directly with the ideal of the teleological symphony, responding with narratives of brutal negation, valediction and leave-taking respectively. In the last three symphonies, the tendency to replace goal-directed processes with static, episodic or circular structures suggests more than an Adornian pessimism; it locates these works in a territory more readily associated with postmodern thought. This is given narrative immediacy in the Sinfonia Antartica, where progressive human ideals are not only seen to fail, but are more bleakly revealed as circling helplessly in the face of an impassive, purposeless wilderness. Consequently, although Vaughan Williams’s later symphonies tell us much about the development of the British symphony, about the European symphony in the twilight of the Austro-German tradition, and about how modes of musical expression change under the impact of the two world wars, they resonate at least as strongly with the present, expressing in nuce sentiments that came to dominate Western culture at the end of the twentieth century.