Liszt: The Complete Music for Solo Piano, Vol.53 - Music for piano & orchestra II

Liszt: The Complete Music for Solo Piano, Vol.53 - Music for piano & orchestra II

  • 流派:流行
  • 语种:其他
  • 发行时间:1998-11-01
  • 类型:录音室专辑
  • 歌曲
  • 歌手
  • 时长

简介

Death as a Friend (1851) by Alfred Rethel (1816-1859) CDA67403/4 Recording details: July 1998 Budapest Studios of Hungarian Radio, Hungary Produced by Tryggvi Tryggvason Engineered by Tryggvi Tryggvason Release date: November 1998 DISCID: FD12A211 66123B19 2B042104 Total duration: 173 minutes 46 seconds 'Exemplary, and superbly recorded' (BBC Music Magazine) 'Hyperion's production, as always, is superb. For the Lisztian, this is indispensable' (Fanfare, USA) 'Howard's dedication is clear in all his playing here, with clear, crisp articulation vividly caught in finely balanced sound' (The Guardian) Allowing for arrangements of other composers’ works, Liszt’s surviving oeuvre for piano and orchestra comprises some fifteen pieces, and the earlier version of Totentanz is so different from the final conception that we must extend the total to sixteen, and to seventeen if Liszt’s contribution to the Ungarische Zigeunerweisen (Konzert im ungarischen Styl) is considered proven. There are other Liszt works for piano and orchestra which may well have existed, or which we know he performed, but which have not come down to us: two concertos that he wrote as a teenager have disappeared, and concertante versions of five of his major solo pieces are mentioned in correspondence or by publishers (Clochette Fantasy, Puritani Fantasy, Niobe Fantasy, El Contrabandista and Heroischer Marsch im ungarischen Styl) but still remain tantalizingly undiscovered. Of the music we have, Liszt himself played very little in public—just three works, in fact—and he did not see much of it through the press. It is easy to imagine the teenage Liszt, in his early conquest of the musical world, being asked to produce works of his own to play with orchestra, and we know that something was produced quite quickly, because there are many contemporary accounts of Liszt playing unspecified concertos of his own composition. But the earliest surviving concertante works were completed in the 1830s, by which time Liszt’s mature style had completely evolved. It is difficult to establish incontrovertibly a chronology of composition, but by the mid-1830s Liszt must have drafted the so-called Malédiction Concerto and the Lélio Fantasy, and probably an early version of the First Concerto. Only one of these works was performed by Liszt at this stage, and that was never prepared for publication: the Lélio Fantasy, which Liszt played under Berlioz’s direction in 1834, when it proved a far more successful piece than the extraordinary Berlioz work from which Liszt took his themes. In that same year, Liszt produced his largest concertante piece—the instrumental psalm De Profundis, which he never performed and which was not published in his lifetime, although he returned to it for material for other works in later years. By the end of the decade, Liszt had given several performances of his orchestral version of the famous Hexaméron variations, but this, too, remained unpublished in its orchestral form. At the end of the 1830s Liszt was working on three concertos. Two of them would wait twenty years or more for their final published form, and the runt of the litter would be abandoned in rather an unsatisfactory state of completion, to be published a century after Liszt’s death. A further work which would occupy a decade in the search for a final form was Totentanz, whose beginnings date from 1849 and whose first complete version contains material derived from, but never identical to, De Profundis. During the 1850s, when Liszt was concentrating on repairing earlier works for what he regarded as their definitive versions, he revised his Capriccio alla turca, adding the opening march and issuing the work as a Fantasy on themes from the Ruins of Athens, completing the work in 1852. At the same time, he created a version with orchestra of his Grand Solo de concert, but, because he later revised the solo version of the piece, he never published either the solo or the orchestral version of his first conception. However, he did issue his one potboiler: the Hungarian Fantasy, which harks back to the Tenth of the Magyar Dalok, and forward to the Fourteenth Hungarian Rhapsody. He also made his arrangements of the Weber Polonaise and the Schubert ‘Wanderer’ Fantasy at this time, and they achieved instant popularity. By the end of the decade, Concerto No 1 had undergone its final revisions after the premiere Liszt played in Weimar under Berlioz, and the final form of Totentanz was ready although Liszt never played it. Concerto No 2 was finally ready in 1861, but by this stage Liszt was only prepared to conduct the work and never to play the solo part. Nor did he play his revised solo part for the Weber Konzertstück which was published in the early 1870s. Incontrovertible evidence of the extent of Liszt’s role in the preparation at Schloss Itter in 1885 of the Ungarische Zigeunerweisen remains elusive, and in any case it is not a work of the front rank, but the piece is not without interest and, on the grounds that Liszt may have had something to do with it, it is included in this recorded survey by way of a bonus. The version for two pianos of the Grosses Konzertsolo—the Concerto pathétique—became very popular, and two of Liszt’s pupils produced versions with orchestra, one of which Liszt took to his heart and then to his desk for comprehensive rewriting and extension, making it one of the last things he completed—although he then muddied the waters by keeping his part in the work anonymous in the published score. Concerto No 2 in A major has always been especially beloved of the most musicianly Liszt interpreters who are wary of the temptation of the tradition of performing the First Concerto for empty spectacle and in the shortest time possible (Alfred Brendel has written with typically incisive vigour and wit upon this subject and has noted that it has been whipped through in less than fourteen minutes by someone who ought to have known better) choosing instead the work which, for all its demands and grand gestures, remains fundamentally poetic. Of course, Liszt differentiates the characters of the two concertos clearly: the contrast in tonality is maximum, the openings could not be more unlike, the four-movement underlying structure of the First is replaced in the Second with a single movement which extrapolates the boundaries of the sonata principle in the same way that many of Liszt’s symphonic poems do, and the Second Concerto relies on fewer but more distinguished themes than the First. The first theme is striking for its juxtaposition of distant seventh and ninth chords which nevertheless fail to unsettle the principal key, clear from the opening chord. The piano enters with gentle arpeggios for the second statement, which is then extended into the transition to the second theme in a mellifluous passage with solo horn, oboe and cello. In a reversal of the usual constituents of a sonata exposition, the second theme is powerful and triadic, and is presented in D minor. A second transition presents a new theme which takes the music to B flat minor and a forceful tutti, bringing the exposition to an end. The development may be said to begin from the moment the soloist rejoins the orchestra, when the connection between the tutti theme and the end of the first theme is made manifest, and underlined by the little piano solo which calms things almost to the state of the beginning of the concerto. The tutti theme is then transformed in the orchestra into a gentle introduction extended by a short cadenza, and a reworking of the first theme in 4/4 rather than 3/4, in D flat major, with a solo cello. This is lyrically continued by the piano, eventually joined by oboe then flute in another transformation of the tutti theme, under which the violins play a phrase which at once derives from the first two chords of the concerto and yet outlines the melody (and sentiment) of a Liszt song Freudvoll und leidvoll. The material of the first transition informs the cadenza which leads to the recapitulation of the second subject, in a robust D flat major, the lower strings playing the tutti theme in counterpoint. The recapitulation continues with a further transformation of the first transition, development of the tutti theme and second subject combined, and a transformed version of the tutti itself. The second transition theme ensues, now in A minor, and the general increase in tempo is finally reigned in with the recapitulation in the original key of the first theme, now transformed into a march and punctuated with fragments of the ubiquitous tutti theme. (Various commentators have been rude about this passage, noting its martial vulgarity and generally failing to see that it is the permissible moment of triumph at the final point of recapitulation of the first theme and the first time we have seen the home key since the opening pages of the work. Even Searle refers to this passage as occurring in ‘the finale’, showing not much appreciation of the structure.) As if to silence potential critics, Liszt uses the second transitional theme again to reintroduce the first theme in the most magical form, running on with a version of the same lyrical extension we heard with the cello solo, and continuing in the same manner through the Freudvoll und leidvoll phrase to a short cadenza. This cadenza, like so much of the binding material of this work, is derived from the alternate falling semitone and tone from the first theme, and these intervals now immediately generate the material of the animated coda. The coda is so superficially appropriate a peroration that it requires a second look to see how well it draws the whole argument together, with every theme represented in one way or another, right to the closing bars. On 14 January 1835 Liszt wrote to the Abbé de Lammenais, telling him that he was the dedicatee of a ‘little work’—an instrumental De Profundis, based on a plainsong which they both knew, and which he would send to the Abbé. Liszt himself was about to be ‘away for two months’, for which read much longer. Marie d’Agoult would be pregnant with their first daughter by the end of March 1835 (Blandine was born in December of that year) and events had conspired to remove Liszt (probably without much fuss on his part) some months previously from his voluntary retreat at Lammenais’s home at La Chesnaie. As Iwo and Pamela Zaìuski have written: ‘He would emerge after a spell of religious fervour to embrace worldly passions, and his need for solitude would give way to a need for lively-minded company. His impatience with humanity alternated with feelings of love for his fellow man.’ His particular need for Marie, however, has probably cost us an absolutely correct finished copy of Liszt’s largest concertante piece. The manuscript in Weimar which bears a dedication to Lammenais (we doubt that there was a fair copy made to send to Lammenais) is all but finished, but like many similar Liszt manuscripts, the very end awaits its final form, and may have done so until Liszt tried the work out in public, which he never did. He moved on, and never revived his interest in the work, even though the De Profundis plainsong features in his work as late as the early 1850s. We are indebted to Jay Rosenblatt for making the work performable, and for this recording one or two very minor alternative readings are employed and a short coda recalling the opening of the plainsong has been added. (The Rosenblatt score and parts are available for hire; the two-piano score published by Acs and the score and parts on hire with it are extremely defective, at one point lacking 52 bars of fully-scored music; a version by Michael Maxwell has been recorded, but it takes many liberties with the text, reorchestrates much of the piece, and adds a furious Mephistophelean conclusion.) It can be confidently stated that the orchestration is entirely Liszt’s own. Even more poignant than the use of the word ‘symphonique’, Liszt’s description of the work as being composed for ‘orchestre et piano principal’ tells us how significant the orchestral part is. The structure of the De Profundis is remarkable, both in itself and for what it presages in Liszt’s symphonic thinking: the piece is a vast sonata movement, containing a slow movement, itself based on the plainsong which does duty for the second subject, and a contrasting scherzo in the form of a polonaise, and ending with a coda based on the slow movement but transformed into a march. The principal tonality is D minor, but the slow movement is in A flat and the polonaise is in C sharp minor. The difficult question in terms of the serious nature of the work is how to account for the presence of the polonaise. There is certainly nothing in the text of the psalm to account for it. Had the piece come from Liszt’s later life it would be easy to see a tribute to the Princess zu Sayn-Wittgenstein. Here it is more puzzling, but the whole section suggests a capitulation to worldliness not unlike the temptations of Faust by Mephistopheles—Berlioz had introduced Liszt to Gérard de Nerval’s French translation of the first part of Goethe’s play in December, 1830—which are rejected by the sterner stuff of the main material, and which are prefaced by the extremely softened version of the melody advanced by the slow movement—possibly a premonition of Gretchen in the Faust Symphony, and certainly in the same key. It may not be too wide of the mark to see the Faust programme in the whole work—brooding, questioning, prayer, temptation and redemption are all in the piece—alongside the cry of the psalmist for God’s forgiveness: Out of the deep have I called unto thee, O Lord : Lord, hear my voice. O let thine ears consider well : the voice of my complaint. If thou, Lord, wilt be extreme to mark what is done amiss : O Lord, who may abide it? For there is mercy with thee : therefore shalt thou be feared. I look for the Lord ; my soul doth wait for him : in his word is my trust. My soul fleeth unto the Lord : before the morning watch, I say, before the morning watch. O Israel, trust in the Lord, for with the Lord there is mercy : and with him is plenteous redemption. And he shall redeem Israel : from all his sins. (Psalm 130 [Vulgate 129]) The work unfolds at an unhurried pace: the opening Andante presents two motifs—the first in octaves in the bass, the second in thirds in the piano. These alternate until the piano plays the first of several cadenzas, where the music is reminiscent in harmony and shape to the first of the Apparitions of the previous year (in Volume 26). The orchestra returns and the material reaches its first great climax, before subsiding to the opening phrase over a timpani pedal D in an attempt to re-establish the tonic key of D minor. When that tonality arrives, it is with a brisk new theme of repeated notes first announced by the violas, and the orchestra develops the material sequentially until F major is reached, and with it the plainsong De Profundis appears for the first time. It must be said that this theme has not been located by the present writer outside the works of Liszt, and although Liszt reminds Lammenais in the letter quoted above of their mutual interest in the theme which he describes both as a plainsong and as a faburden, it is not to be found in any of the traditional Catholic repositories of plainchant, nor does it resemble any of the chants commonly applied to this particular psalm. Perhaps the melody is of later origin, and is merely something which Liszt and Lammenais knew as the commonly performed music for the psalm at some time and place when they were together (and it may then be of an origin similar to that of the Psalm which Liszt transcribes in the Album d’un voyageur, or of the Miserere—another psalm, of course—of the Harmonies poétiques et religieuses). In any case, the melody as such contains but five notes—the first is heard twenty-one times in succession, the second and the fourth ten times each, and the third, like the fifth, once only. The harmony may originate in some psalter, too, but Liszt is happy to vary the mode—the orchestra gives the whole verse in F major, then the piano plays alone in F sharp minor, at which point the Latin text of the first two verses is overlaid: De profundis clamavi ad te, Domine : Domine exaudi vocem meam. Fiant aures tuae intendentes : in vocem deprecationis meae. After a restatement of the psalm by piano and winds a long development section ensues in which the psalm rhythm is often applied to the accompaniment of the opening material, culminating in a close with four mighty E flats from the depths of the piano. The slow movement begins with a piano cadenza for which some of the material of the very first cadenza provides the springboard, but which is interspersed with phrases of the psalm, now in A flat major. Finally a new melody of disarming beauty emerges, clearly derived from the psalm, and at the end of the cadenza this is taken up by divided strings and piano. After a full close comes the scherzo in the style of a polonaise—except that it keeps sidestepping the traditional 3/4 for passages in 2/4. The form of this section is: polonaise; episode in 2/4; trio melody in A major; polonaise first variation; 2/4 episode first variation; brief quotation of psalm theme; trio first variation in B flat minor/major; truncated polonaise second variation; 2/4 episode second variation; trio second variation in D flat major; development of polonaise with psalm rhythm and main material leading to the recapitulation proper. This section is greatly telescoped, and we arrive quickly at the reprise of the psalm theme from the full orchestra, thence into a modulatory episode (with the soloist) which emerges in D major, with the concluding march. As we have seen, the march takes the slow movement melody as its basis, but after its mightiest statement it subsides over a timpani roll to the concluding phrases which combine fragments of the psalm with the march rhythm to end the work with a quieter confidence. We encounter the ‘symphonique’ epithet again in the Franz Schubert: Grosse Fantasie opus 15—symphonisch bearbeitet für Piano und Orchester—Liszt’s beloved ‘Wanderer’ Fantasy in his transcription which was for many years extremely popular, and which, frankly, lies much easier under the hands than Schubert’s original. Apart from the tiny cadenza which forms the transition to the E flat section of the first movement, Liszt adheres scrupulously to Schubert’s work, very rarely allowing himself very much in the way of decoration, let alone succumbing to the easy chance of adding counter-themes. (This version is not to be confused with Liszt’s later reworking of the piece for solo piano, S565a—Volume 49—but it does equate with Liszt’s published reduction for two pianos, S653.) In the first section, Liszt lets the orchestra speak first, and the piano enters only with the second, quiet statement of the principal theme. Liszt takes various melodic fragments into the orchestra—often as woodwind solos—and gives some of Schubert’s repeated chords over to the orchestra altogether. The slow movement is left to the piano to begin, the orchestra entering where Schubert’s left-hand tremolo is now taken up by both hands on the piano whilst the thematic material is passed to a dialogue between wind and strings. In the variation with the filigree writing in the right hand Liszt restores Schubert’s theme in the winds and in the left hand, where Schubert’s original merely hints at it, and he allows the whole orchestra to join in Schubert’s mighty climax. Exchanging phrases between piano and orchestra is Liszt’s starting point for the arrangement of the scherzo and the fugal exposition of the finale is left to the piano, with the orchestra joining gradually thereafter in a nicely calculated crescendo. Throughout the piece, Schubert’s more extreme demands are modified: the octaves at the end of the first movement, the terrible leaps to the right-hand arpeggios at the end of the scherzo, and the final page of arpeggios in both hands which so often brings disaster in concert are entirely avoided, these last being replaced by more solid chords from the piano and fanfares from the horns and trumpets. The Concerto pathétique in the version recorded here represents Liszt’s last thoughts on a work which had occupied him on and off for over thirty-five years. As we have seen, the first version was the Grand Solo de concert for piano, S175a, and then with orchestra, S365 (see Vol 53a). Then came the extended Grosses Konzertsolo for piano, S176, (Vol 3) and the version for two pianos: Concerto pathétique, S258, was made in about 1856, and enjoyed much popularity amongst Liszt’s students, even though it was not printed until ten years later. A second edition of that version appeared in 1884, with a section of twenty-seven bars towards the end replaced with forty bars newly composed by Hans von Bülow—with Liszt’s apparent blessing. Possibly as early as 1872, Liszt’s student Eduard Reuss had made a version of the 1866 edition for piano and orchestra, and it is this version which came to Liszt’s notice in 1885. (Another of Liszt’s students—Richard Burmeister—had already made a concertante version of the work, but Liszt does not seem to have seen it. More recently, a version by Gábor Darvas has been published in Hungary and recorded, but this has nothing to do with Liszt either.) There was a lively correspondence between Liszt and Reuss as Liszt gradually took the piece under his own control; having first indicated to Reuss how pleased he was, Liszt then made copious alterations and revisions as late as January 1886, before he approved the work for publication. (The incomplete manuscript in the Library of Congress does not represent his final thoughts, and we do not have the engraver’s copy to see what changes were made at the proof stage. In any case, the publication took place after Liszt’s death, as did that of the two-piano reduction by August Göllerich.) Apart from altering the instrumentation, and adding a new introduction and coda, Liszt’s main aim was to lift the piano out of the orchestral texture by adding various new solo linking passages—a strange but effective plan, since they are almost all in the sparse style of his late works and in some contrast to the opulent textures of his middle years which inform the rest of the work. In his typically kind way, Liszt arranged for the piece to be published with full credit to Eduard Reuss, and with no mention at all of his own considerable collaboration and rewriting—a circumstance which has led to the work being unjustly omitted from most Liszt catalogues. The opening of the Concerto pathétique is a newly composed passage of ten bars in octaves for the orchestra, and the orchestra keeps the ensuing statement of the first theme to itself. The piano enters for a brief cadenza leading to the lyrical contrasting melody which still forms part of the first subject group, and plays throughout the agitated transition to the second subject which is preceded by the first of Liszt’s interpolated solos. This Grandioso theme is taken by brass and drums—the drum rhythm is significant, because it was newly invented by Liszt, who wrote to Reuss (on 10 January 1886) that it would produce a better effect than his original plan of quick reiterated fifths copied from the two-piano piece, and that it would also serve twice as a rhythm for new transition material, and again at the end of the piece. So there is a new added passage in this rhythm for solo piano before the music again follows the two-piano score with the next statement of the second theme with horn and harp (this is the only concertante work of Liszt to include a harp). Liszt modifies the end of this passage and inserts further new material in the link to the Andante sostenuto. Further new material is found before the second, decorated statement of this theme, and the following phrases characterized by initial falling fifths have each been extended with trills and other-worldly harmonies for three bars at a time. Another extension leads into the Allegro agitato assai, which is a further development of material from the first group, leading directly into a development of the second subject. This breaks off suddenly with another added solo derived from the opening bars, and the recapitulation begins, as in all the versions of this piece, with the reprise of the agitated transition material. Another interpolation precedes the Stretta—a developed reprise of the first theme—which, in this version, is taken by both piano and orchestra, and yet another new passage connects to the funeral march. Here we have muffled drums and solo trumpet in an astonishingly avant-garde piece of orchestration, and Liszt adds to the gloom by adding eight extra bars to the middle of it. The above-mentioned timpani rhythm returns in another new passage—and it is probably no accident that this rhythm is identical to that of the moment in the Stabat Mater speciosa in the oratorio Christus where it is associated with the text ‘Quando corpus morietur, Fac, ut animae donetur Tui Nati visio’ (‘When the body die, grant that the spirit receive a vision of thy Son’); Liszt does not seem to have used this rhythm elsewhere—and the Andante sostenuto is recalled with its previous phrase extensions and a new connecting passage. The succeeding reprise of the second subject was in E major in all the other versions of the piece. Astonishingly, and to great effect, Liszt drops the tonality by a semitone, and then engineers a marvellous chord change to restore E major for the triumphant coda, which is introduced by more new material and extended by some twenty-five bars of defiant but not virtuosic octaves from piano and orchestra. The young Liszt spent much time playing Weber’s two concertos, and only learned the more famous Konzertstück in his twenties, after which he never returned to the concertos again. But it was not until many years after he had retired from concert playing that Liszt produced his instructive edition of the piece, with its many intelligent suggestions for a text more apposite to the power of the newer pianos. Liszt’s revised text of the piece as a piano solo has already appeared in this series (Vol 49), and his suggestions for a revised version of the solo part in Weber’s original are recorded here. Weber’s orchestration remains unaltered, but Liszt offers many interesting solutions to passages which present problems of balance in the original, or which demand things like octave glissandos which work less well on the heavier instruments of the later nineteenth century (or most concert grand pianos since). The introduction remains virtually unaltered, but various tasteful and helpful embellishments are suggested in the principal Allegro. The most obvious difference lies in the march, where Weber has the piano silent except for a glissando interruption (the châtelaine’s beloved emerges from the distant crusaders’ march, according to the fanciful programme which Weber allowed to be attached to the score), but Liszt allows the piano to play distantly in a high register above the orchestra. In the final section Liszt again offers many practical suggestions for a more effective text without ever doing any musical violence to Weber’s work. The private owner of the manuscript of the first version of Totentanz declined permission to view to the present writer, and so we are obliged to take Busoni’s work in preparing the piece for publication in 1919 on trust. It is, therefore, unclear whether this version was completed in 1849, as Busoni suggests, or whether the revisions apparent in the manuscript from which he worked date from 1853 when we know Liszt carried out some intermediate revision on the piece. There are many differences between this Totentanz and the version commonly performed (see Vol 53a), most of which are immediately obvious to the ear: the introductory gong strokes require no signposts, but the rest of the introduction is, broadly speaking, the same as the final version, except for a number of rhythmic differences, slighter scoring, and an uprushing scale in chromatic chords which later became blind octaves. Variations 1–3 correspond, with minor differences, to Variations I–III of the final version. Variation IV of the later version is not found here, and the Variazione fugata corresponds to the later Variation V, although the scoring is much thinner and there is a cadential passage accompanied by chromatic bass rumblings which was later replaced with a trombone statement of the opening phrase, in tempo, over semiquaver chords from the piano. Then there follows no cadenza, nor the orchestral statement of the second theme with its horns. In the place of what Liszt later called Variation VI—itself a theme and six variations, we have an Alternativo, which corresponds to the first of the later miniature variations, and Variations 4, 5, 6 and 7 (this last placed separately on track 17, so that it may be compared to the earlier version which Busoni offers as an appendix to his edition, here appended as track 21), which correspond to the second, third, fourth and sixth of the later miniature variations—the fifth of the later variations is new. The ensuing cadenza in the earlier version is quite different, and leads to a brass chorale in B flat major—‘De Profundis’—a version of exactly the same theme that generated much of the Psaume instrumental. The chorale is repeated by flutes and clarinets before the piano leads the music into B minor (the same shift that it accomplished in the Psaume when the music moved from F major to F sharp minor), and then developed with much fantasy, and through a variety of distantly related harmonies until we stand again on the threshold of the D minor of the main body of the work. The coda of the later version is clearly derived from the Allegro which follows here, and the cadenza resembles the last cadenza in the later version, but beyond that there is no more music in this version which was carried over into the final one. Instead, we are offered a further variation on the second theme, albeit at a much faster tempo. The Dies irae returns and leads to a last cadenza before the final Allegro con fuoco combines the Dies irae and the De Profundis themes to close the work in shattering volumes of sound. The Fantasie über ungarische Volksmelodien—or ‘Hungarian Fantasy’, as it is commonly known—bears an ineluctable kinship with the Rapsodie hongroise No XIV, but it was conceived before that work, and diverges from it in many important particulars, the first and foremost amongst which is the tonality: the Fantasy is in E minor, the Rhapsody in F minor. Various reworkings in the Rhapsody proceeded at a different time and pace for those of the Fantasy so, although the Fantasy was the second of the two works to be published, it was the first to achieve its final form. The background material for the piece was the tenth of the Magyar Dalok and the Magyar Rapszódia No 21, S242/10 and 21 (see Vol 28). Liszt took the Magyar Rapszódia and arranged it for piano and orchestra, adding an extra bar of timpani roll to get things under way, and amplifying the cadenzas in such a way as to recall Magyar Dal No 10. A new cadenza leads to the main theme of the piece, since identified as the Hungarian folksong Mohacs Field. This theme is presented three times—from the piano, from the trumpet and then other wind instruments with an agile piano accompaniment, and from the full orchestra. The Molto adagio which follows derives from the Magyar Dal No 10, as does the succeeding material. Music in common with the fourteenth Rhapsody but deriving directly in this instance from Magyar Rapszódia No 21 is the theme marked Allegretto alla Zingarese, which leads to a reprise in D flat major of Mohacs Field. After a cadenza based on this theme, but punctuated by fragments of the Magyar Dal on clarinet and flute, a new theme is presented which will carry the work almost to the end—labelled Koltoi Csárdás in the Magyar Rapszódia. At the Prestissimo coda all the themes are brought together, and in this recording, the curious tradition of slowing the last statements of Mohacs Field to something less than a quarter of Liszt’s required continuation in tempo prestissimo is deliberately avoided. The history of the Ungarische Zigeunerweisen (Konzert im ungarischen Styl) is clouded with uncertainties. What is indubitably true is that, in 1892, at Sophie Menter’s request, Tchaikovsky prepared a score for piano and orchestra from material which she provided. Equally true is that the publication of that score was not seen through the press by Tchaikovsky, who died the following year, and that the published score and parts require a good deal of common-sense correction. What Tchaikovsky worked from has not been preserved, but it seems to have been some kind of short score. The question is: did Sophie Menter (a very famous pianist in her day, and a composer of salon trifles) compose the work, did Liszt write it, or did Menter take something to Liszt which he then got into shape for her in the period of exactly two days in which he is known to have worked at Menter’s castle—Schloss Itter—in 1885? Göllerich mentions the piece in his diary and suggests that Liszt would have had trouble completing it (failing eyesight and poor health being likely primary reasons; not wishing to write a virtuoso piece in a style which he had long abandoned no doubt being another). Liszt’s letter to Menter dated 3 August 1885 tells her that the ‘Sophie Menter Concerto’ is begun and that he would complete it at Schloss Itter. At this remove we cannot establish whether the work referred to as a Concerto in the Hungarian Style equates with the present piece, but on balance it seems very likely. It has to be said at the outset that the musical substance of the piece is not particularly Lisztian, but if Menter really collected the themes (which are unknown in Liszt’s works, although they are similar in style to melodies found in some of the Hungarian Rhapsodies or the Ungarischer Romanzero) and Liszt just helped over two days to arrange the short score, then his possible collaboration may be conceded. What appears exceedingly unlikely, although it is a theory that has been advanced anent this work, is that Liszt instructed Menter to take the piece to her friend Tchaikovsky for orchestration without mentioning Liszt’s name, since Tchaikovsky did not admire Liszt, especially since the publication of the transcription of the Polonaise from Eugene Onegin, and that Liszt’s composership of the work be hidden behind Menter’s name. This is just preposterous: Tchaikovsky orchestrated Liszt’s song Der König in Thule without demur, and he also orchestrated Liszt’s version of Mozart’s Ave verum corpus for his Mozartiana suite when he could just as easily have used the original. And if he referred to Liszt in his diary as ‘the old Jesuit’ then at the least it is a lesser insult than that volume reserves for many of Tchaikovsky’s other colleagues (such as Brahms, for example: ‘a giftless bastard’!). In any event, this piece is missing from any other recorded accounts of Tchaikovky’s concertante works, and ought to be heard for that reason alone. We cannot say whether Tchaikovsky played any part in the actual composition, but towards the coda there is a harmonic sequence very familiar from Tchaikovsky’s concertos. Let us hope that our bets are hedged correctly and that Liszt had something to do with the piano figuration, especially in the earlier part of the piece. The structure of the piece is straightforward enough, and is clearly inspired by works such as the Hungarian Fantasy. The work begins with a theme from the orchestra that will not return, and a piano cadenza, full of imitation of the cimbalom, leading to the Andante—a soulful theme expounded in arpeggiated chords. There follows an Allegro variation and another cadenza, leading to a new theme marked Allegretto, given first by the piano and then joined boisterously by the orchestra. The Andante theme is recalled in the ensuing cadenza and a new theme is presented in the slow Andante (really an Adagio) which follows. One more reminiscence of the Andante theme leads to a variation on the Allegretto, with the piano playing in constant demisemiquaver octaves. Another short cadenza introduces a new theme in the horns, but it is shortlived and we find ourselves very soon in the coda, which is generated from a faster version of the Andante theme. All in all, the piece is effective without laying claim to any special importance, and considering the probable assistance of Liszt and the certain participation of Tchaikovsky, it was thought wise not to pass it by. Leslie Howard © 1998 Works on This Recording 1. Concerto for Piano no 2 in A major, S 125 by Franz Liszt ■ Performer: Leslie Howard (Piano) ■ Conductor: Karl Anton Rickenbacher ■ Orchestra/Ensemble: Budapest Symphony Orchestra ■ Period: Romantic ■ Written: 1839/1861; Weimar, Germany ■ Date of Recording: 1997-98 ■ Venue: Hungarian Radio, Budapest, Hungary ■ Length: 21 Minutes 4 Secs. 2. De profundis for Piano and Orchestra, S 691 (unfinished) by Franz Liszt ■ Performer: Leslie Howard (Piano) ■ Conductor: Karl Anton Rickenbacher ■ Orchestra/Ensemble: Budapest Symphony Orchestra ■ Period: Romantic ■ Written: 1834-1835; France ■ Date of Recording: 1997-98 ■ Venue: Hungarian Radio, Budapest, Hungary ■ Length: 36 Minutes 13 Secs. ■ Notes: The score for this recording was prepared by Jay Rosenblatt and Leslie Howard from the manuscript sources. 3. Wanderer Fantasy (Schubert) for Piano and Orchestra, S 366 by Franz Liszt ■ Performer: Leslie Howard (Piano) ■ Conductor: Karl Anton Rickenbacher ■ Orchestra/Ensemble: Budapest Symphony Orchestra ■ Period: Romantic ■ Written: by 1852; Weimar, Germany ■ Date of Recording: 1997-98 ■ Venue: Hungarian Radio, Budapest, Hungary ■ Length: 21 Minutes 55 Secs. 4. Totentanz, S 126 by Franz Liszt ■ Performer: Leslie Howard (Piano) ■ Conductor: Karl Anton Rickenbacher ■ Orchestra/Ensemble: Budapest Symphony Orchestra ■ Period: Romantic ■ Written: 1849/1859; Weimar, Germany ■ Date of Recording: 1997-98 ■ Venue: Hungarian Radio, Budapest, Hungary ■ Length: 17 Minutes 23 Secs. ■ Notes: Ver: 1849 (edited by Ferruccio Busoni) 5. Hungarian Fantasy for Piano and Orchestra, S 123 by Franz Liszt ■ Performer: Leslie Howard (Piano) ■ Conductor: Karl Anton Rickenbacher ■ Orchestra/Ensemble: Budapest Symphony Orchestra ■ Period: Romantic ■ Written: ?1852; Weimar, Germany ■ Date of Recording: 1997-98 ■ Venue: Hungarian Radio, Budapest, Hungary ■ Length: 16 Minutes 18 Secs. 6. Concerto for Piano in the Hungarian Style, S 714 "Ungarische Zigeunerweisen" by Franz Liszt ■ Performer: Leslie Howard (Piano) ■ Conductor: Karl Anton Rickenbacher ■ Orchestra/Ensemble: Budapest Symphony Orchestra ■ Period: Romantic ■ Written: 1885 ■ Date of Recording: 1997-98 ■ Venue: Hungarian Radio, Budapest, Hungary ■ Length: 17 Minutes 37 Secs. ■ Notes: This work may have been composed in collaboration with Sophie Menter, or assembled from melodic material provided by her; the nature of Liszt's involvement with it is unclear. This version for piano and orchestra was prepared by Tchaikovsky in 1892 at Menter's request. 7. Concerto pathétique, S 365 by Franz Liszt ■ Performer: Leslie Howard (Piano) ■ Conductor: Karl Anton Rickenbacher ■ Orchestra/Ensemble: Budapest Symphony Orchestra ■ Period: Romantic ■ Written: by 1857; Germany ■ Date of Recording: 1997-98 ■ Venue: Hungarian Radio, Budapest, Hungary ■ Length: 25 Minutes 22 Secs. 8. Konzertstück in F minor (Weber) by Franz Liszt ■ Performer: Leslie Howard (Piano) ■ Conductor: Karl Anton Rickenbacher ■ Orchestra/Ensemble: Budapest Symphony Orchestra ■ Period: Romantic ■ Written: by 1837 ■ Date of Recording: 1997-98 ■ Venue: Hungarian Radio, Budapest, Hungary ■ Length: 17 Minutes 28 Secs.

[更多]