Hector BERLIOZ (1803-1869)
Harold in Italy (arr. Franz Liszt, 1882) [44.23]
Franz LISZT (1811-1886)
Romance oubliée (1880) [5.07]
Kurt ROGER (1895-1966)
Viola Sonata, Op.37 (1948) [22.23]
Philip Dukes (viola), Piers Lane (piano)
rec. Wyastone Concert Hall, Monmouth, 3-5 May 2012
NAXOS 8.573011 [71.53]
Franz Liszt was the great populariser of ‘modern music’ during the nineteenth century. During his long life he poured out a series of paraphrases, transcriptions and variations on music by fellow-composers. His sound instincts led him to espouse the music of the greatest of his romantic colleagues. These works fall basically into two groups. The first of these were the barnstorming virtuoso pieces intended for his own performance during his touring recitals. The second were more straightforward transcriptions designed to make the music available to good amateur performers to encounter and get to know the music in the privacy of their own homes. His transcription of the Berlioz Symphonie fantastique - in which incidentally he confirms that Berlioz imagined the bells in the final movement in the lowest register, and not the tinny tubular bells we so often hear today - helped to establish the fame of the composer in Germany. He followed this up with the version of Harold in Italy which we are given here. However although the transcription was written in 1838, it was later revised and was not published until 1881 by which time Berlioz was long dead.
It has to be said that Berlioz, the master of the orchestra, is not best served by the performance of his music on the piano, no matter how excellent the transformation of the orchestral textures in terms of the keyboard. However the Symphonie fantastique is given a new dimension in Liszt’s version, revealing facets of the score that otherwise might be overlooked. Harold in Italy is less successful. This is largely because of the nature of the work itself. It was originally intended as a virtuoso vehicle for Niccolo Paganini, who wanted to show off his recently acquired Stradivarius viola. What Berlioz produced was not the barnstorming concerto that Paganini anticipated, but a symphony with viola obbligato. The soloist remains silent for long stretches throughout the first movement and the viola plays almost nothing at all during the finale. Liszt faithfully adheres to Berlioz’s intentions, allowing the viola to deliver only what Berlioz wrote for the instrument. This is with one exception, to which I will come in due course. In a version for viola and piano this really exposes the disproportionate nature of the solo contribution. One can only imagine what the impression on stage would be during the finale as the viola stands silent while the piano ploughs through the extravagances of the brigands’ orgy. I understand that when Dukes and Lane performed the arrangement at the Purcell room a couple of years ago, Dukes left the stage during the finale and played his final dying bars from offstage, a dramatic effect that would work well although it is not clear whether that procedure was followed in this recording.
Even without the possibility of such visual distractions, however, the transcription otherwise works surprisingly well in this performance. This is largely due to the excellently contrived balance between viola and piano. Other recorded performances of this arrangement that I have heard tend to spotlight the viola, with the result that the balance of the work is disturbed. Here, quite correctly, the emphasis is placed on the piano, and the reverberant acoustic helps successfully to conjure up the richness of Berlioz’s writing. Indeed, even too much so in the opening bars, where the ominous bass mutterings sound considerably more present than Berlioz’s pianissimo marking in the orchestral score would imply. Philip Dukes is well integrated into the sound picture, and his ethereal arpeggios in the Pilgrims’ March have just the right sense of musing distance. Piers Lane is rather brisk in this movement, but one recognises that without the sustaining sounds of the orchestra the music could easily appear to grind to a halt, and the fault - if indeed it is one - errs on the right side. The mountaineer’s serenade is lively, and in the repeat of the opening material Dukes is given some additional material, appropriating the orchestral viola line to provide a sort of drone bass. I don’t recall this from other performances I have heard. Presumably it derives from one or another of Liszt’s revisions of the arrangement. It works so well that one wonders why Liszt did not go further and compose some additional material for the viola in the lengthy finale. This however generally comes over most successfully. Liszt does however rather miss the sense of sheer excitement in the reiterated violin figure in the central section, substituting some more conventional piano figurations which lack the requisite driven mania (track 4, 5.45). Nonetheless Dukes and Lane make out the best possible case for this version of the score.
More interesting however is the Liszt Romance oubliée, a transcription made in his later years of his 1844 song Oh pourquoi donc. Actually it is far more than a simple transcription, substituting the viola for the voice. It is a free improvisatory contemplation on the melodic material of the song, and the viola arpeggios in the final bars echo the similar use of the sound in the Pilgrims’ March in Harold - surely not a coincidence. The freely rhapsodic viola line has a sense of poised beauty which anticipates in some measure the similar use of solo string instrument and piano in later works such as Vaughan Williams’ Lark ascending. The work is quite familiar in its versions for violin and cello, but the original viola version is much rarer and it is a delight to encounter it here. There is also a piano Romance based on the same song, which bears but tangential resemblances to the version here.
Even better is the Kurt Roger Sonata, not at all the kind of work that one might have anticipated from a pupil of Schoenberg writing in 1948. Indeed the opening movement, with its swingeing chordal writing for the piano, could well have been written fifty years earlier. The booklet note informs us that the work is sometimes known as the Irish Sonata, and indeed from this sonata-form movement one could well imagine that one is hearing an undiscovered work by Stanford. There are indeed some Irish folk elements apparent, and in the later movements - with their strict counterpoint - one can also detect echoes of Moeran and Alan Bush. This is not simple imitation; there is an original voice at work here. The Gould Piano Trio have already given us a Naxos disc of Roger’s chamber music, and although only one item on that disc - the delightful Variations on an Irish air - is as immediately attractive as this Viola Sonata, one suspects that the composer’s output could well bear further investigation (review).
So buy this CD especially for the Roger Sonata, and also for the original version of Liszt’s Romance. You’ll also get a superb performance of the somewhat problematic Berlioz transcription as well.
Paul Corfield Godfrey
A superb performance of the somewhat problematic Berlioz transcription alongside the fascinating Roger Sonata and the original version of Liszt’s Romance.
Support us financially by purchasing this disc from