UDK 78.082.1Mahler Niall O'Loughlin
Loughborough University Univerza v Loughboroughu
The Movement Order of Mahler's Sixth Symphony: Musical Considerations
Sosledje stavkov v Mahlerjevi Šesti simfoniji: glasbeni premisleki
Prejeto: 1. avgusta 2008 Sprejeto: 5. septembra 2008
Ključne besede: Gustav Mahler, Šesta simfonija, sosledje stavkov, Erwin Ratz, Jerry Bruck
Izvleček
Mahler je svojo Šesto simfonijo napisal s Scher-zom kot drugim stavkom, pred Andantejem, a je pred prvo izvedbo zaobrnil njuno zapovrstje, ki naj bi veljalo za vse poslejšnje izvedbe in natiske. Nekateri vidni glasbeniki kljub temu vztrajajo pri tem soseldju. Na podlagi številnih povezav med Allegrom in Andantejem ter Scherzom in Finalom se Mahlerjevo zapovrstje kaže kot popolnoma prepričljivo.
Received: 1st August 2008 Accepted: 5th September 2008
Keywords: Gustav Mahler, Sixth Symphony, movement order, Erwin Ratz, Jerry Bruck
Abstract
Mahler composed his Sixth Symphony with the Scherzo preceding the Andante, but reversed them before the first performance for all performances and published editions. However, some prominent musicians still insist on this order. With numerous thematic and harmonic connections between Allegro and Andante, and Scherzo and Finale, Mahler's revised order is completely convincing.
Gustav Mahler composed the two middle movements of the Sixth Symphony at his country retreat at Maiernigg (Majrobnik) on the Wörthersee (Vrbskem jezeru) in the summer of 1903. It seems that he also sketched some of the first movement during the same period. The opening movement and the finale were composed in draft form the following summer. The autograph score was ready in May 1905 for the copyist, with the Scherzo placed before the Andante, and the score being published in this form some three months before the premiere in Essen in May 1906. However, in the final rehearsals in Essen, Mahler reversed the order of the middle movements for the first performance.
He immediately issued instructions for a correction slip to be inserted into the remaining unsold copies of the published scores and prepared a new edition to take account of the movement order and the revised orchestration. All the performances in Mahler's lifetime, whether conducted by him or not, adhered to this order. Mahler had discussed the symphony in 1907 with his friend the Dutch conductor, Willem Mengelberg, who performed the Sixth Symphony in 1916 with the Andante-Scherzo order. For reasons that may be connected with the existence of copies of the uncorrected first edition study score of the symphony, in 1919 Mengelberg wanted to check that Andante-Scherzo was really the correct order.1 He contacted the composer's widow who reportedly sent a famous telegram on 1 October 1919 saying 'Erst Scherzo dann Andante herzlichst Alma'.2 We do not know whether this represents the composer's last thoughts on the issue, which is conceivable but otherwise totally unsubstantiated, or whether this represented her own opinion as to how the music should be performed, which is also possible. It, of course, contradicts the statement in her book Gustav Mahler: memories and letters that the Scherzo was the third movement.3 On the strength of this telegram, Mengelberg marked his score 'Nach Mahlers Angabe II erst Scherzo dann III Andante'4 and performed it accordingly in 1920. Virtually all performances from then on,5 however, followed the revised published order of movements (Andante-Scherzo), in public performances and recordings until the late 1950s when Erwin Ratz cast some doubt on the authenticity of this order, persuading some conductors to use the movement order, but not the orchestration, of the first published score.6 Conductors such as Charles Adler, William Steinberg, Eduard van Beinum and Norman Del Mar adhered to the Andante-Scherzo order for the middle movements. Then Ratz put his ideas into concrete form with his edition of the work for the Internationale Gustav-Mahler-Gesellschaft Critical Edition of 1963, placing the Scherzo before the Andante.7
The Dissenters
While the factual basis for Ratz's decision has been conclusively repudiated by Jerry Brack and Reinhold Kubik,8 there have been persistent voices saying that musically the order Scherzo-Andante is the 'correct' one. Various points have been raised by numerous commentators in support of this position, ones which purport to represent the
Jerry Bruck: 'Undoing a »Tragic« Mistake: Determining the inner-movement order of Mahlefs Sixth Symphony', in Gilbert Kaplan (ed): The Correct Movement Order in Mahler's Sixth Symphony (New York: The Kaplan Foundation, 2004). I am very grateful to Mr Bruck for allowing me to read an early version of his paper in advance of publication.
Henry-Louis de La Grange: Gustav Mahler, Vienna: Triumph and Disillusion (1904-1907) (Oxford: Oxford University Press, 1999), p. 815.
Alma Mahler: Gustav Mahler: memories and letters (London: John Murray, 3rd edition, 1973), p. 70. The German edition was originally published in 1940. It is not clear if Alma is referring only to the order maintained in performances during her husband's lifetime or what was considered to be the definitive order in the published score of the second and third editions. de La Grange: op. cit., p. 815.
It appears that Webern's performances in 1932 and 1933 followed the order of the discarded first edition. See Henry-Louis de La Grange: Gustav Mahler, A New Life Cut Short (1907-1911) (Oxford: Oxford University Press, 2008), p. 1583. Two conductors in the late 1950s who followed Ratz's ideas were Hans Rosbaud and Dmitri Mitropoulos. Gustav Mahler: Symphonie Nr. 6 in vier Sätzen für großes Orchester (Lindau-Bodensee: Kahnt, 1963).
In Gilbert Kaplan (ed): The Correct Movement Order in Mahler's Sixth Symphony (New York: The Kaplan Foundation, 2004).
composer's musical thoughts. A small number of these have some plausibility, but most of them are highly subjective or misguided and are often argued from mistaken factual evidence, while others are quite obviously invented without any supporting evidence to provide corroboration. We can safely ignore the writers of superficial programme notes who express unsubstantiated opinions that have little or no basis in fact,9 but there are established Mahler experts who still maintain the discredited movement order and who should be taken very seriously. Let us take these in turn.
Before the publication of Jerry Bruck's researches, even if there were some dissenters, Ratz's ideas were widely accepted. Hans-Peter Julg, for example, in 1986 expressed the mistaken view (later disproved) that Mahler had returned to the original order for the Vienna performance of 1907, using this as part justification for the Scherzo-Andante order employed in his study.10 In 1992 a short article by the then current editor of the Internationale Gustav Mahler Gesellschaft, Karl Heinz Fussl, set out his view of the position, backing Ratz.11 His summary is missing some important points and talking about the middle movements, he wrote, 'It remains something of a mystery why their reversal took place several times.'12 There is no hard evidence for this statement, only speculation; from all that is known Mahler changed his mind only once. Fussl wrote that the first movement is parodied by the scherzo, which is a reasonable assumption, and that they should be played consecutively, which is not necessary; and that the key structure was such that the Andante is better positioned just before the finale. He also put forward the idea that the scherzo was a 'developing variation', in much the same way as the second movement of the Fifth Symphony connected to the first movement.13 There is some truth in this, but the first two movements of the Sixth Symphony are so completely different from those of the Fifth, especially in their proportions and their length, that the argument is difficult to sustain. It may well have been the case that Mahler was thinking on similar lines to the first two movements of the Fifth Symphony when he composed the Sixth, but what appears satisfactory on paper may not have worked well in practice. Three years later Robert Samuels in his book on the symphony carefully avoids the issue in the earlier part of his book by discussing the individual movements out of order, but later makes clear that he follows Ratz's order when he makes speculations about a possible programme.14 However, in this book he never really addresses the issue of the order of the middle movements.
The first major challenge to the new arguments that were produced by Jerry Bruck was that of the outstanding Mahler scholar Henry-Louis de La Grange in the new English version of his monumental Mahler biography.15 In an appendix devoted to the Sixth
Remarks like 'Mahler changed his mind many times.'; 'Mahler was constantly unsure of the right order.' Hans-Peter Julg: GustavMahlers Sechste Symphonie (Munich: Emil Katzbichler, 1986), pp. 110-13.
Karl Heinz Fussl: 'On the Order of the Middle Movements in Mahler's Sixth', News about Mahler Research 27 (Vienna: International Gustav Mahler Society, March 1992), pp. 3-7. Ibidem, p. 5. Ibidem, p. 6.
Robert Samuels: Mahler's Sixth Symphony (Cambridge: Cambridge University Press, 1995), p. 150.
Henry-Louis de La Grange: Gustav Mahler, Vienna: Triumph and Disillusion (1904-1907) (Oxford: Oxford University Press, 1999). De La Grange made an unfortunate mistake in a note to his main text in which he wrote the following: 'For the première, the Oct. performance in Berlin and the Nov. performance in Munich, the Andante followed the Scherzo. In Vienna in Jan. 1907 Mahler apparently reverted to the original order of the movements.' (p. 412 note 63) In all these performances, however, the Andante preceded the Scherzo, as is clear from other parts of his text. The author graciously accepted this mistake and apologised for it in a personal communication (2000).
LC

i;
Li
14
Symphony16 he includes a section entitled 'The Order of the Middle Movements'.17 Here he sets out the main facts and restates the arguments of Karl Heinz Füssl: 'The Scherzo belongs after - and with - the opening Allegro,' 'The Scherzo uses the same keys as the first movement', 'The key of the Andante ... is furthest removed from the end of the Allegro', while the C minor final introduction links the E flat of the Andante with the main key of finale, and 'A slow movement precedes the Finale in five other of Mahler's symphonies.' This is tantamount to telling the composer his business: one can only speculate about what Mahler might have said in reply. What is also a little disturbing is the appearance of some statements of doubtful authority, for example, '... given that Mahler himself changed his mind so many times, it is understandable that a conductor might nowadays wish to stand by the order in the second version, if he is deeply convinced that he can serve the work better by doing this.'18 Did Mahler really change his mind so many times? This suspect statement has been regularly repeated since by other writers. Further de La Grange suggests without any real evidence that one of Mahler's friends and disciples said to him that the similarity between the first and second movements was a weakness and should be changed.19 To state categorically that the irony and parody of the scherzo is lost when separated from the first movement is an opinion that is just as easily denied as stated (see my example of Liszt's Faust Symphony below).
A more detailed presentation of this position was given by David Matthews a year later.20 One possibility that he puts forward is the idea that Mahler changed the order of the middle movements for the first performance because of a 'fear of the Symphony's prophetic power, and an instinctive wish to diminish it.'21 This surely is mere speculation. In actual fact, Mahler had plenty of opportunity to return to his original order if he really thought this was needed to put across the work's prophetic power undiminished. The basis of this statement is completely undermined by the fact that Mahler stood by his decision in two more performances which he conducted and in the new editions which he instructed his publishers to undertake. It is my opinion that the revised order strengthens the power of the symphony rather than weakens it. This point will be addressed when dealing with the narrative issues involved. More significant is the investigation of key oppositions that are noted by Matthews: his point about the avoidance of A minor in the later parts of the first movement so that the impact of the A minor of the Scherzo is enhanced may well have been in Mahler's mind, at least to start with. It is as well to remind ourselves that Mahler knew this music so well that he must have been acutely aware of exactly what he was doing and why.
Three recent books have continued to assert that performers should use the discarded Scherzo-Andante order. A collection of Donald Mitchell's outstanding writings on Mahler includes an appendix by Gastón Fournier-Facio entitled 'The 'Correct' Order of the Middle Movements in Mahler's Sixth Symphony'.22 While he accepts Jerry
Ibidem, pp. 808-41. Ibidem, pp. 814-16. Ibidem, p. 815. Ibidem, p. 816.
David Matthews: 'The Sixth Symphony', in Donald Mitchell and Andrew Nicholson: The Mahler Companion (Oxford: Oxford University Press, 2000), pp. 366-75. Matthews: Ibidem, pp. 372-73.
In Donald Mitchell: Discovering Mahler: Writings on Mahler, 1955-2005 (Woodbridge: Boydell Press, 2007), pp. 633-47.
16


19


Brack's findings as historical facts, he insists that the musical evidence tells a different story, even if no convincing reason emerges from his text as to why Mahler changed his mind. He repeats Fussl's arguments about 'developing variation' and harmonic connections between the first movement and the Scherzo and between the Andante and Finale. His text cites the arguments of David Matthews who makes a strong case for the progress from E flat major to C minor to A minor (that is moving directly from the Andante to the finale). This of course may well have been in Mahler's mind when composing it, but he did not feel strongly enough on the issue not to reverse the order of the middle movements. Donald Mitchell has maintained a neutral position on the order, suggesting that there are two versions.23 For Mahler, of course, there was only one version, and only one version was ever publicly performed in his lifetime, unlike Das klagende Lied or the First Symphony, both of which were actually performed in different versions by the composer. The recent Cambridge Companion to Mahler includes a chapter by Stephen Hefling which outlines the history of the problem, but keeps to the Scherzo-Andante order.24 The most recent statement has come with a revisitation of the problem from Henry-Louis de La Grange in an appendix to the fourth volume of his biography of the composer.25 This distinguished of Mahler biographers takes a careful and low-key approach to the issue, presenting various documents that cast light on the situation, but provide no really convincing proof that Mahler did return to his original order - suspicions possibly, but no proof. He, like Donald Mitchell, does seem to settle for the rather unsatisfactory solution that there are two versions and performers should choose whichever they prefer.
In all these views, however, there is an enormous assumption made by these writers: that they know better than the composer how his music should be performed. There are various suggestions that he was so emotionally upset on the day of the first performance that he could have made a 'mistake'. The fact that once the emotional tension of the first performance was over, he then changed the movement order for the new edition and for subsequent performances suggest that it was no accidental decision taken in a moment of stress. This must lead to a very important point that should be stressed: the question about choosing the musically 'correct' order of the middle movements is both impertinent and mistaken. Mahler decided the definitive order of the movements before the first public performance, used this in the second and third editions of the score and maintained this in all the performances in his lifetime, whether he conducted them or not. The first edition was superseded on the instructions of the composer and there is no credible evidence that he changed his mind again. The only legitimate musical question to ask is not whether one version is better or not, but what musically made Mahler change his mind.
Trying to understand the mind of a composer, especially one as complex as that of Gustav Mahler, poses a great many difficulties, an issue that the present study takes fully into account. For this reason one must progress with care. Even with this caveat, we may
23	Ibidem, pp. 387 and 644.
24	Stephen E. Hefling: 'Song and Symphony (II)' in Jeremy Barham (ed): Cambridge Companion to Mahler (Cambridge: Cambridge University Press, 2007), pp. 108-27, especially pp. 119-24.
25	Henry-Louis de La Grange: Gustav Mahler, A New Life Cut Short (1907-1911) (Oxford: Oxford University Press, 2008), Appendix 1I, pp. 1578-87.
begin to understand what made Mahler make the decisive move, before the premiere, to establish the Andante-Scherzo order for the middle movements of the Sixth Symphony for all public performances in his lifetime and for all the published material apart from that issued before the first public performance.
The most relevant areas of investigation would seem to be the following:
1.	Thematic connections between separate parts of movements and more importantly between different movements.
2.	Harmonic links within movements and between different movements.
3.	The narrative issues.
4.	The views of conductors who have gone through the same experience as Mahler himself.
Thematic Connections
What questions can be asked? What relationships are there between the themes and motifs of the four movements? What significance is there in them? Do connections between the themes have any bearing on the order of the movements?
There are a number of thematic connections that may have had some relevance to the movement order that Mahler eventually adopted. Norman Del Mar associated the first theme with the composer himself, having no doubts about its significance, 'And it required no great stretch of imagination to see the stormy but heroic material of the primary subject matter as a self-character-study'.26 The second main theme of the first movement was said by the composer's wife, Alma, to represent her in music.27 Its expansive arching melodies and yearning character and above all its emphatic major key make this one of the most positive elements in the movement. Interestingly the two themes have some melodic parts in common. The first of these themes seems to be parodied in the Scherzo. No doubt this was intentional and may have been influenced by the way the first two movements of the Fifth Symphony are connected. The question whether the Scherzo should follow the first directly or whether the slow movement should intervene was probably not in Mahler's mind at this stage.
At the rehearsals for the first performance, and possibly also at the earlier run-through in Vienna, there was probably some concern in Mahler's mind about an important thematic connection that could have persuaded him to adopt the Andante-Scherzo order. This relates to the chorale which separates the two main thematic areas of the first movement and which appears briefly in the E flat major 'pastoral' episode. Significantly it then returns as a 'memory' at the climax of the Andante. The connection between this calm episode in the first movement and the general peace of the Andante can be no coincidence. In performances which place the scherzo second, this subtle connection is completely lost.
Norman Del Mar: Gustav Mahler's Sixth Symphony a study (London: Eulenburg, 1980), p. 16.
Some doubt has been raised about this supposed association by Robert Samuels in his book Mahler's Sixth Symphony (Cambridge: Cambridge University Press, 1995), p. 157. The validity of this association does not affect the thematic argument which follows.

The similarity between the opening of the first movement and the opening of the Scherzo, however, would hardly be missed even by the least attentive listener. This awareness is not reduced at all by placing the Andante between the two movements. The suggestion that it does not make its true parodistic effect unless it follows the first movement is patently untrue. The assertion that the Scherzo simply 'develops' the music of the first movement, rather than parodies it, is an equally difficult position to maintain, because the appearance of the menacing xylophone and the brass and woodwind trills make this one of the most sinister movements in all of Mahler's music. One can point to a parallel with another work which Mahler may well have known, but does not seem ever to have conducted, Liszt's Faust Symphony. In this work the third movement representing the devil parodies the first movement which stands for Faust. There is no problem understanding the intended parody, despite the fact that the gentle 'Gretchen' slow movement is played second.
Other thematic links, especially the major-minor 'Fate' motif of the first movement which is played at the end of each appearance of the main Scherzo, are equally memorable and are not in any way rendered less significant by the intervening Andante. Further, the fact that the same motif appears at numerous points in the finale then suggests that its significance in the parodistic Scherzo can be thought of as a preparation for the complex operations and ultimately tragic outcome of the finale. The appearance in the finale of episodes in which the cowbells are used recalls the Andante, but now parodied and played in the minor key. While Mahler may have originally decided that this thematic (and instrumental) reference to the Andante would be satisfactory with the Andante placed third, the idea of a more 'distant' memory would be enhanced by placing the slow movement second.
Harmonic Links
This is a very contentious issue. The move from the first movement to the second is a good point to start. There would be no problem harmonically for the listener when the first movement ends in A major and the second movement (the Scherzo) starts in A minor.28 Because of the point of possible thematic connection noted above, Mahler then had to consider the issue of whether the A major of the end of the first movement could be followed directly by the E flat major of the Andante. Adherents of Mahler's original order are horrified by the idea of making such a fundamental harmonic shift, especially as the reverse move would then have to be made after the end of the movement to return to the A minor of the Scherzo. This did not seem to worry Mahler. A few seconds' pause between the movements would be enough to separate them for the audience. But we need only go back a few minutes into the E flat major Pastoral
Peter Andraschke: 'Struktur und Gehalt im ersten Satz von Gustav Mahlers Sechster Symphonie', in Hermann Danuser (ed): Gustav Mahler, (Darmstadt: Wissenschaftlicher Buchgesellschaft, 1992), pp. 234-37, suggested that the A major of the ending of the first movement is then naturally followed by the A minor of the Scherzo, mirroring the notes of the »fate« motif. This argument is ingenious but difficult to support. The major-minor change in the motif is effected by sustained chords with a lowered third, but the first movement ends with only a quaver and the Scherzo starts with low As from timpani and string basses without a sustained A minor chord.
episode in the first movement to see how the composer extricates himself from this precise harmonic situation. From his E flat major tonality in the episode he moves through a very fluid tonality until at bars 250-251, he wrenches the music into B major ('sehr energisch'), a truly magical moment that might have surprised Beethoven. The path back to A major/minor for the recapitulation was then not difficult to achieve smoothly.
Let us turn to Adorno's support for Ratz's position: 'his [Mahler's] last arrangement of the movements, with the E-flat major Andante before the Finale, should be respected, if only for the modulation scheme; E-flat major is the relative of C minor, with which the Finale begins, only to decide, after long preparation, on A minor as its principal key.'29 Leaving aside the opening erroneous statement - it was not Mahler's last arrangement of movements - this insidious remark about key relationships involving major keys and relative minors has more to do with Beethoven than Mahler. By the late 19th and early 20th centuries, the predominance of tonic and dominant relationships no longer held such importance. Mahler found much greater interest in mediant relationships and major-minor relationships involving the same tonic.30
On the same point Andraschke wrote: 'the E flat of the Andante relates better to the C minor at the beginning of the Finale'. This arises from the Classical idea of relating major and minor keys with the same number of sharps and flats - the so-called relative minor/major. In Mahler's musical language A minor would connect just as well with the finale's opening keys. Again one can look nearly in vain for harmonic relations such as this in Mahler's music.31 Mahler's harmonic structures normally use mediant and subme-diant relationships in much of his work, including this symphony. Moving from the E flat major of the Andante to the C minor which opens the finale is one such move which Mahler must have considered satisfactory. In the revised order, however, the placing of the Finale immediately after the ending of the Scherzo also has a strong continuity which must have satisfied the composer even more. The Scherzo exhausts itself in fragments and whimpers, including a number of disguised appearances of the fate motif, ending in A minor with the double bassoon, double basses and timpani playing only the notes A and C. While Mahler did not indicate 'attacca' there is no problem linking this passage with the C minor of the first nine bars of the finale and of the A minor section which follows. Adorno's point about the long delay of A minor at the beginning of the final is not helpful, because the harmonic focus is in a state of flux, which is no surprise after the disruptions of the Scherzo.32 Again Mahler would have been able to recognise the harmonic implications of that very readily.
Theodor W. Adorno: Gustav Mahler (Chicago: Chicago UP, 1992), p. 85.
One can read the detailed evidence of Christopher Orlo Lewis in his Tonal Coherence in Mahler's Ninth .Symphony (Ann Arbor: UMI, 1984) to understand how this functions.
The Second Symphony starts in C minor and ends in E flat major, but the intervening movements do not support the idea of any traditional relationship between the two keys.
This point had already been made by Hans Redlich in 1963 in his article, 'Mahler's Enigmatic »Sixth«' in Otto Erich Deutsch Festschrift (Kassel: Bärenreiter, 1963), pp. 250-56. Redlich changed his position completely in his edition of the superseded first score (London: Eulenburg, 1968) though, significantly, he did not include the amendment to the movement order that Mahler instructed his publisher to include in the scores that were still remaining in the unsold stock. On p. xxv of this score Redlich makes a very surprising unsubstantiated statement: 'His intention to revert to the original sequence of movements ... was never incorporated in print because no further edition of the symphony was issued in his lifetime.'
29
J0
Ji

Christopher Hailey charts a very convincing large-scale tonal scheme that makes a great deal of sense of the order that Mahler always used for his performances. He also closely connects the idyllic E flat major episode in the development of the first movement with the key of the Andante, and catalogues numerous motivic connections between the two movements.33 The implication is that the two belong together just as strongly as many commentators suggest that the first movement and Scherzo belong. Hailey interestingly relates and interprets the motivic connections between Andante and Scherzo in two ways34 depending on which order is chosen, but he gravitates strongly towards placing the Andante first.
The motivic and tonal relationships between Scherzo and Finale are even more important.35 After discussing the first movement and Andante, Hailey emphasises this point: 'Still more compelling reasons, however, argue in favor of having the Scherzo immediately precede the Finale.'36 He points to the use of the rising octave in both the Scherzo and Finale and elaborates a number of harmonic connections that relate the Scherzo and Finale that are too complex to summarize. While Hailey felt that Mahler's indecision was well founded, he went on to write: 'I would argue on esthetic grounds for an AndanteScherzo ordering'.37 Of course, so did Mahler, and that is what really counts.
Musical Narrative
Since the publication of Theodor Adorno's study of Mahler's symphonies, one commonly encounters the idea of a musical narrative, not so much that the symphonies 'tell a story', but rather that there is a certain progression to the music, movement by movement. This can be seen in the early symphonies, but from the Fifth Symphony onwards it becomes a very strong feature.
The narrative idea can be seen very clearly in the Fifth Symphony.38 We can take part one (movements 1 and 2) as an interconnected series of recurring minor-key sections, much of march-like character, which reach their climax at the D major chorale near the end of the second movement. It is the rondo-like elements that build up the tension and expectation. The music collapses into A minor in some kind of 'failure'. Part two (the scherzo) reintroduces D major now in an exuberant dance-related context. Part three (Adagietto and Rondo-Finale), almost exclusively in major keys, presents another interlinked series of three 'rondo' materials: the middle section of the Adagietto and its related grazioso of the finale; the finale's main rondo theme itself; and the five fugal episodes. As in part one the music reaches a climax in the appearance of the same D
Christopher Hailey: 'Structure and Tonal Plan in Mahler's Sixth Symphony', in Hermann Danuser (ed): Gustav Mahler (Darmstadt: Wissenschaftliche Buchgesellschaft, 1992), pp. 265-66. Some of these connections are also given by Robert Samuels in Mahler's Sixth Symphony (Cambridge: Cambridge University Press, 1995), p. 26. Hailey: op. cit., p. 266. Ibidem, pp. 267-68. Ibidem, p. 267.
Ibidem, p. 268. Hailey's remark: 'there is no reason that the ordering should not be left to the discretion of the conductor' (also p. 268) is one point too many and one which flies in the face of Mahler's final decision.
I presented this material in considerably greater detail in a previous article: Niall O'Loughlin: 'Interconnecting Musicologies: Decoding Mahler's Sixth Symphony', Muzikoloski zbornik xxxix/1-2 (2003), pp. 31-49.



major brass chorale. In contrast to the collapse in part one, the music remains firmly anchored in D major right up to its triumphant conclusion. The narrative relates to the significance of major and minor keys, and the opposition of the march (the first movement) to the dance (the Scherzo).
In the Sixth Symphony the process is reversed, but only with the Andante-Scherzo order of the middle movements. In this form the first two movements then each contain triumphant climaxes, the third and fourth movements each include the collapses. The conflict between the A minor march material of the first movement (perhaps the music representing the composer) and the 'Alma' theme results in a triumph for the latter (and for the major key). The Andante follows with major key tranquillity and references to the first movement before coming to rest on a sustained E flat major.
The scherzo and finale now turn the triumphs on their head. The scherzo parodies the first movement, tonally, rhythmically and thematically, in a way that the listener would have no difficulty in recognising, even with an intervening slow movement. Its three-in-a-bar music sardonically mocks the march material, while the keys of the two trio sections (F and D major) mirror in parodistic fashion the predominant keys of the positive 'Alma' music. The main scherzo section is progressively contracted at each appearance with a devastating collapse at the third time. As noted above, this disintegration ends by emphasising the two notes A and C, significantly the key-notes of the two minor-key tonalities that dominate the early part of the finale which results in the biggest collapse in all of Mahler's music.
There have been suggestions that Mahler reversed the order of the middle movements to soften the impact of the work, but in performances that I have heard the opposite is the case. With the Scherzo first the Andante is left in limbo, but with the Scherzo third the Andante can fulfil the positive yearnings of the coda of the first movement. The first two movements (Allegro and Andante) represent the optimistic outcome, while the third and fourth movements (Scherzo and Finale) give us the frightening end with its sinister preparation. If this is what the composer wanted, it makes a great deal of sense. The original order (Scherzo-Andante) does pose problems for some performers and it is to those that one can turn to see how Mahler might have been thinking when he settled for the new movement order.
Conductors' Views
Why Mahler himself made the decision to reverse the order of the middle movements of his Sixth Symphony before the first performance is a matter of speculation. The writings of Klaus Pringsheim give some clue to his tortured state of mind during the rehearsals, and there have even been some suggestions that he was influenced by other people in this decision. Henry-Louis de La Grange is convinced that Mahler was persuaded, by implication against his better judgement, by those around him to change the order at the time of the first performance.39 The points mentioned above give pos-
39 Henry-Louis de La Grange: Gustav Mahler, Vienna Triumph and Disillusion (1904-1907), (Oxford: Oxford University Press, 1999), pp. 815-16.
sible reasons for Mahler to have changed this order, but they are by no means definitive. The best that one can say is that he imagined that the original order was satisfactory, until he performed the work. Then he realised that by reversing the order of the two middle movements, he could create the balance that he wanted. There is a great deal of evidence that the Andante-Scherzo order is completely satisfactory from the point of view of thematic links, harmonic connections and the somewhat disputed area of musical narrativity. The composer would never have insisted on the new order for his performances and for the second and third editions of the printed material if he had any real doubts.
Norman Del Mar is a conductor who performed Mahler's Sixth Symphony on at least two occasions,40 using the Andante-Scherzo order at both and, unlike Mahler, we know why he used this order. In his book on the Sixth Symphony Del Mar went out of his way to state his view that the 'correct' order of movements was that found in the revised Kahnt edition of 1906. His justification for this view still carries weight today: 'For Mahler's reaction, even during rehearsals, had been to realise that the Scherzo was too similar in style and dynamism to follow directly upon the enormously strenuous twenty-two-minute opening movement. Equally, for the Andante to precede the long slow introduction that opens the monumental Finale was not really satisfactory, whereas by reversing the order the necessary contrast and relief on both counts was solved at a single stroke.'41
Following the publication of Ratz's score in the Critical Edition in 1963, very few conductors continued to use the Andante-Scherzo order. Norman Del Mar, as mentioned above, and John Barbirolli were two. It is quite well known that until Jerry Bruck's research into the facts about Mahler's performances were well publicised, the only other conductor that consistently maintained the order was Simon Rattle. He had been convinced by the thematic resemblances noted earlier not to separate the first movement and Andante.
Since Bruck's revelations, many conductors have been convinced that this order is the one that should be followed. These include Glen Cortese, Leonard Slatkin, James Judd, Mariss Jansons, Charles Mackerras, Claudio Abbado and Ivan Fischer.42 The authority of both Jansons and Abbado must carry a lot of weight in correcting the situation. Jansons, for example, has made recordings with both the London Symphony and the Royal Concertgebouw Orchestras, in addition to numerous public performances, all with the Andante-Scherzo order. It is interesting to note that in 1979 Abbado had made a very well received recording with the Chicago Symphony Orchestra using the Scherzo-Andante order, whereas he changed his mind in his recent award-winning recording with the Berlin Philharmonic Orchestra and his 2006 performance at the influential Lucerne Festival in Switzerland43 which both use the Andante-Scherzo order.
With the BBC Symphony Orchestra at the Henry Wood Promenade Concerts in the Royal Albert Hall in London on 26 August 1963 and at the Edinburgh Festival in the Usher Hall on 2 September 1963. Norman Del Mar: op. cit., p. 90.
Of leading conductors, the only ones to my knowledge who now maintain the Scherzo-Andante order are Bernard Haitink, Pierre Boulez and Christoph Eschenbach.
With the Chicago Symphony Orchestra, Deutsche Grammophon 423 928-2 (1980); with the Berlin Philharmonic Orchestra, Deutsche Grammophon 477 5684 (2005); and with the Lucerne Festival Orchestra, Euroarts DVD 2055649 (2007).
4C
41
42
One of these conductors, Ivan Fischer, has given an explanation of his decision to settle finally for Mahler's final order. His decision and the reason for it are very revealing:
'There is the famous discussion about the order of the middle movements. Putting the scientific arguments aside I have been fascinated by the question what Mahler's doubts felt like when he suddenly abandoned his beautifully constructed original symphonic plan. To relive this experience we took the sixth symphony on a long European tour and changed the order of the middle movements every single concert. In the Scherzo-Andante performances the transitions from one movement to the next felt wonderful, the whole architecture made sense but I felt a clear unease about the size and weight of the Scherzo after the first movement. In the Andante-Scherzo concerts there was a fantastic balance and variety. I became convinced that Mahler's abrupt decision was a stroke of genius.'44
In effect, this is the same point that Norman Del Mar made and it may well have been what Mahler was thinking, too. Needless to say, Fischer's recording uses the AndanteScherzo order. The American conductor James Conlon also recognises the correctness of the Andante-Scherzo order and now performs it that way. Like Del Mar, Fischer and probably Mahler himself, Conlon feels that the Scherzo is too weighty to follow the first movement directly: 'On a purely subjective level, I found the contrast between the first movement and the beginning of the Andante much more satisfying than once again hearing the repeated thumping of the timpani immediately after the first movement. Its reappearance at the beginning of the third movement is far more effective.'45 He did express some regret about the loss of the Scherzo-Andante order, especially in what he calls the self-revelation of the Andante, which he thinks appears prematurely if played as the second movement.46 In the end, however, he wrote: 'None of this proves that Mahler was wrong to want to change the order of the inner movements of the Sixth. By definition he cannot be wrong: the composer's wish is our command.'47
Ivan Fischer on p. 6 of the booklet insert for his recording with the Budapest Festival Orchestra on Channel Classics CCS SA 22905 (2005).
'The Curious Problem of Mahler's Sixth', Gramophone (November 2007), pp. 50-51.
It is interesting to note that in recent compact disc recordings conducted by Abbado, Fischer and Valery Gergiev the Andante (played second) lasts less than 14 minutes with correspondingly less emotional stress being placed on this movement, rather than the normal 15 to 17 minutes with the Andante third. The duration given by Mahler in the correction slip of 1906 is 14 minutes, which suggests that the Andante as third movement has been made to carry more emotional weight than was intended by Mahler.
James Conlon: op. cit. p. 51.
44
Povzetek
Gustav Mahler je svojo Šesto simfonijo napisal s Scherzom kot drugim stavkom, medtem ko je Andante postavil na tretje mesto, in v tej obliki je delo tudi izšlo pred njegovo prvo izvedbo. Na zadnji vaji pred premiero pa je skladatelj zaobrnil zapovrstje the stavkov, nakar je svojemu založniku izdal navodilo glede novega sosledja, ki je bilo sprejeto v vseh nadaljnjih natiskih. Vse izvedbe so se za Mahlerjevega življenja držale tega načrta, vendar pa je po njegovi smrti dirigent Willem Mengelberg v dvomih povprašal Mahlerjevo vdovo o pravilnem zapovrstju stavkov. Alma Mahler je dejala, da bi Andante moral slediti Scherzu, česar se je dirigent tudi držal. Vendar pa je večina dirigentov sledila revidiranemu zapovrstju, dokler ni Erwin Ratz pred izdajo zbranih Mahlerjevih del prepričal nekatere dirigente, da so omenjena stavka izvajali v zapovrstju Scherzo - Andante, nakar sta Jerry
Bruck in Reinhold Kubik diskreditirala Ratzovo početje. Ne glede na to je precej uglednih muziko-logov smatralo, da original predstavlja »glasbeno konkretno« različico. Razni argumenti naj bi namreč podpirali to in tako odločitev, predvsem pa dejstvo, da bi si tematsko sorodna prvi stavek in Scherzo morala slediti in da bi se zavoljo tonalnih povezav med Andantejem in Finalom le-ta morala izvajati eden za drugim. Vse to so utegnile biti tudi Mahlerjeve misli, vse dokler ni v živo izvedel svojega dela. Na podlagi »dokaznega gradiva«, zbranega v štirih poglavjih - tematske povezave, harmonske zveze, poved skladbe in pogledi dirigentov - članek zastopa mnanje, da Mahlerjevih sprememb mišljenja ni pripisati nekakšni napaki, ki so jo implicirali nekateri zagovorniki vrnitve k prvotnemu zapovrstju. Predloženi argumenti kažejo, da je Mahlerjev revidiran načrt glasbeno smiseln, čemur pritrjujejo izvedbe skoraj vseh vodilnih dirigentov.