While living in Vienna, probably during his tenure as Director of the Opera, Gustav Mahler sketched a Scherzo in C Minor and a Presto in F Major. These drafts do not match any of the movements in the ten known symphonies. They may constitute parts of a single work unknown in the Mahler repertoire; at the least, they seem to have been composed in close chronological proximity. Although neither sketch was brought to full fruition, both stand as near totalities in form and substance. The manuscripts exist in a state comparable to that of the Tenth Symphony; and, as in the case of the Tenth, a performing version of these drafts is not only possible but desirable.
The Scherzo in C Minor is located in the Wiener Stadtbibliothek since being sold to the City of Vienna by Hans Moldenhauer in 1980. There is a brief description of the manuscript in the News About Mahler Research published by the Internationale Gustav Mahler Gesellschaft.1 The Presto in F Major is located in the Pierpont Morgan Library in the bequest of Mrs. Wolfgang Ros the wife of Mahler's nephew. J. Rigbie Turner first drew attention to this manuscript in an article in Nineteenth-Century Music.2 At some point, probably in the 1920s, both sketches were studied by Alban Berg, who wrote cover pages giving capsule descriptions. About the Scherzo he wrote:
7 pages of manuscript written on single sides, completely unknown, belonging to none of Mahler's familiar works; also not to the Tenth Symphony. It [the sketch] appears to behave like a scherzo type of movement in 6/8 meter with a possible trio (4) in 2/4 meter; earlier, if not from the earliest time, however, the many blue informational numerals appear to stem from a more recent time, as if Mahler had later thought of possibly utilizing the thematic material of these old sketches. . . .3
Berg described the Presto in somewhat similar terms to the way he had described the Scherzo, again emphasizing his belief that the draft came from an early stage in Mahler's compositional output and was also completely unknown in the Mahler repertoire. It is likely that the cover pages were written as the result of access afforded to Berg by Alma Mahler, who probably owned both sketches before they came into the hands of the Rosés and Hans Moldenhauer.
Mahler did not date either manuscript, and (surprisingly enough in view of the likelihood that Alma owned the sketches) they are not mentioned in Mein Leben4 or Gustav Mahler: Erinnerungen und Briefe.5 The only known references to symphonies outside the ten in the repertoire are in the memoirs of Natalie Bauer-Lechner6 and in Mahler's letters; these refer to at least three works Mahler drafted while a student in Vienna in the 1870s: (a) a symphony in A Minor mentioned by Bauer-Lechner; (b) another symphony mentioned by Bauer-Lechner rehearsed by the orchestra of the Vienna Conservatory;7 and (c) a "Nordic" Symphony or Suite discussed by Mahler in a letter of 1879.8 Berg may well have borne these references in mind when considering the possibility that the Scherzo and the Presto were written early in Mahler's career. However, the clues we have point to a provenance much later than Mahler's student days for both the Scherzo and the Presto.
First, the handwriting in both sketches bears a close resemblance to that of the manuscripts of the last decade in the century, notably the Fourth Symphony. Second, the style of the melody and harmony in both sketches suggests the period of the early to middle symphonies, i.e., the last decade of the nineteenth century and the opening years of the twentieth. However, the best clue is afforded by the type of paper on which both drafts were written. This is an oblong paper generally showing the logos of "J.E.&Co." associated with the Viennese lithographer Josef Eberle. This paper appears in sources beginning with the Fourth Symphony, the first work to date from the time after Mahler came to Vienna on being appointed Director of the Opera. Such paper was apparently not available to Mahler outside Vienna. Steven Hefling has pointed out that pages here and there do not show the Eberle colophons; but the fact that these exceptional pages are similar in form to those bearing the colophons—plus the fact that Mahler used them only sporadically—makes it impossible to attach significance to the discrepancy.9 With or without the Eberle colophons, three forms of paper most often appear, involving 18, 20 or 22 staves. The types are indiscriminately mixed, but in both the Scherzo and the Presto the 22-stave paper is predominant. The problem then becomes one of establishing whether the paper was being made while Mahler was a student in Vienna in the 1870s, or dated only from the 1890s when he returned to Vienna as Director of the Opera. Alexander Weinmann provides the answer to this question: "The press [Anton von Haykul], after passing through the hands of several interim owners, was transferred to R. von Waldheim in 1864; in 1892 he took Josef Eberle as a partner."10
According to Weinmann, Josef Eberle became a partner of Waldheim in 1892, although Waldheim had taken over the firm of Anton von Haykul as early as 1864. As Eberle is not mentioned in connection with any other firm in Vienna in the course of Weinmann's article, we may assume that the paper in question was not made until Eberle joined Waldheim and that Mahler did not use it until he returned to Vienna in 1897. Therefore, Berg's suggestion that the Scherzo and the Presto were written early in Mahler's career appears to be erroneous; Mahler could not have written either sketch while a student.
On both cover pages Berg emphasized a lack of connection between these drafts and the Tenth Symphony. These statements are curious and, superficially, unhelpful; but they are reasonable if we allow the possibility that he wrote them at a time when considerable attention was focused on the Tenth because of the publication of a facsimile of that work by Paul Zsolnay of Vienna.11 A connection between the two sketches and the Tenth Symphony would have been understandable, given the fact that all the materials in question were written on oblong paper usually showing the Eberle colophons and that none of the materials were finished. But the melodic and harmonic style of the sketches is quite different from the style of the Tenth; and Berg's description, while limited in positive usefulness, does have the merit of warning against confusion of one unfinished work with another.
Both the Scherzo and the Presto appear to be short scores accompanied by sketch variants of certain passages. On four leaves in the Scherzo Mahler laid out a modified scherzo-trio form:
Scherzo in C Minor/C Major, 145 bars (6/8 meter)
Trio in A-flat Major, 71 bars (2/4 meter)
Bridge passage in E Minor, 39 bars (6/8 meter)
Scherzo in C Minor/C Major, reprise (implied by a verbal note)
The other leaves comprise sketches for passages of various lengths parallel with portions of the short score. There are elements in the sketches not found in the short score. This raises questions about the relative provenance of the sketches vis-à-vis the short score; usually sketches are considered as predecessors to a short score, but exceptions to that rule exist.12 However, in the case of the Scherzo, items found in the sketches rather than the short score are usually individual notes enhancing the harmonic texture; they would probably have been inserted into the short score eventually. They do not constitute sufficient grounds to assume that the sketches are later than the short score.
In most cases, the sketches correspond to the short score in tonal detail.13 This shows that Mahler had his tonal scheme well in mind.
Mahler indicated that he intended to make a da capo from the bridge passage back to the Scherzo material proper with the word "Reprise" at the end of the fourth page of the short score, rather than writing out the Scherzo again.14 The thematic character of the bridge passage constitutes supporting evidence of Mahler's plan: Mahler combined themes from the Scherzo and the Trio in this section, making simple rhythmic adjustments in the use of the trio motives which had to be altered from the 2/4 meter to the 6/8 meter. Unfortunately there is a major problem: Mahler indicated the reprise but did not demarcate a clear ending for the movement after the reprise.
The Presto appears to be a short score with variants of certain passages, set out on three leaves covered on single sides except for one bar on the reverse of the first leaf. The first two leaves were drafted with a key signature of one sharp, obviously G Major. However, Mahler wrote a verbal note: "Presto in F Major," indicating that he planned later to change the key signature to one flat and lower the material on the first two leaves by one step. (There is no logical reason to believe that he planned a tonal center of F Major only to write everything out in G Major instead.) On the second leaf there is a passage above which Mahler wrote "1 Ton höher ," but this note applies only to the passage in the second system; Mahler may have considered a momentary modification of the bars in question, or perhaps the note reflects his plans for the system following, which reproduces the material one step higher. The time signature on the first two leaves is alla breve.
The third leaf differs from the first two in both time signature and tonal center, and these discrepancies initially suggest that this leaf does not belong to this work at all. The time signature is 3/2 rather than alla breve; the tonal center is D Major, not G Major.15 However, close study reveals a common harmonic basis between this material and the material on the preceding leaves: much of the material on the third leaf is a reworking of the principal theme of the Presto, especially in its insistent closeness to the tonic/subdominant/dominant harmonies with only brief deviations. Indeed, Mahler wrote a verbal note, "Coda der Themen," above the fifth bar in the first system on the third leaf.
The difference in tonal center between the first two leaves and the third leaf is not an important factor in Mahler's modus operandi in the long run;16 but the technique of recasting harmonic or melodic material in different meters is significant. The most obvious example among many is Das himmlische Leben.17
In form, the Presto appears to be a rondo varied by three episodes (the first and third being essentially the same) and rounded off by a coda. The layout is as follows:
Built into the rondo form of the Presto is a da capo, and Mahler indicated this in two ways: (a) by use of the word "Anfang" at the end of the second leaf; and (b) by a reprise of the opening bars of the movement in diminution. Comparison with other examples of the rondo form—notably the scherzando movement of the Third Symphony—suggests that Mahler would have diverged to the coda from the reprise of an episode rather than the theme itself. Unfortunately there is no clear demarcation of the diverging point—a problem somewhat similar to the one in the Scherzo, and equally awkward.
Lacking secondary documentation as proof that the two sketches belong together, I offer the following reasons as support for the theory that they do:
First, the fact that Mahler preserved both manuscripts (contrary to his usual habit of destroying material he considered useless or inferior) shows that he had plans for them. The plans are unlikely to have affected one draft alone considering Mahler's bias toward multimovement symphonies rather than one-movement overtures or tone poems.
Second, in both manuscripts there is a prominent use of large Arabic numerals in blue pencil in the margins, to which Berg referred in his cover pages. Although the significance of these signs is unclear, they may represent shorthand signals of Mahler's plans for continuity. As Berg suggested, the figures may well show that Mahler worked on the material some time after he sketched it. Some figures are common to both drafts. They support the theory about Mahler's preservation of the drafts cited in the previous paragraph.
Third, the style of the music in both drafts suggests the period about the turn of the century, i.e., the time of the Fourth and Fifth Symphonies.
Fourth, there is a tonal balance between the movements which must be familiar to anyone who has studied the Mahler works in the repertoire; the composer Alan Stout has compared these two drafts with the two Nachtmusiken of the Seventh Symphony. (Both cases involve movements in C Minor/C Major with contrasting sections in A-flat Major, and these movements balance off with movements in F Major.) Although the two sketches should not be considered as potential substitutes for the Nachtmusiken, the resemblance in tonal layout is too striking to be passed over.
Finally, the forms as well as the keys balance each other. A survey of Mahler's use of the scherzo-trio and rondo forms in general shows that the scherzo with trio is always an internal component of the four- or five-movement scheme, and—as a rule—the rondo form is also an internal component.19 It seems possible that Mahler visualized a four- or five-movement work in which the Scherzo and the Presto played internal roles. Had the project advanced further than it apparently did, the Scherzo and the Presto might have been flanked by slow movements in the style which began to grow in the middle symphonies and reached full-grown status in the late ones.
Having described the condition in which Mahler left the manuscripts of the Scherzo and the Presto, I would like to discuss methods useful for the task of making performing versions. In making my version, I have tried to be as conservative as possible. I have done nothing unjustifiable on the basis of comparison with Mahler's own methods in the repertoire works; and I have added as little as I believe to be safely possible.
There are five obvious tasks which must be undertaken in order to make credible performing versions. First, there is the sort of work faced by all editors of Mahler's music: correction of errors and omissions in notation, which applies to accidentals, rests and note values. This work is necessary for all types of sources from rough sketches to published orchestral scores.20
Second, in certain places it is necessary to fill in harmonic and contrapuntal texture. There are three ways in which this can be done:
(a) Transplantation of elements from the sketches to the short scores, facilitated by comparison of the parallel versions. As implied previously, this method is particularly appropriate in the case of the Scherzo.
(b) Simple harmonization using plain voice-leading. This course of action is facilitated by the apparent conservatism of Mahler's harmonic language; harmonization can often be deduced from his melodic outlines, rendering voice-leading possible with relative ease.
(c) Motivic counterpoint based on Mahler's melodic lines, transferred to subsidiary voices after use in primary voices. This is the most difficult of the alternatives for filling in texture, but it is the most rewarding if it can be accomplished. The resulting contrapuntal texture is quite characteristic of Mahler's style at the time of the Fourth and Fifth Symphonies.
An unusual way of extending Mahler's resources seemed indicated in the coda of the Presto. Mahler had written a figure on the dominant, in the bass line, at the beginning of the four bars preceding the coda; although he wrote it only in the first bar of the four, it recurred on the tonic at the beginning of the coda proper. This strongly suggests that Mahler would have extended the figure from the first bar through the entire four bars and moved without pause from the dominant to the tonic. From there it was only a step to the next move, i.e. extension of the tonic figure from the first bar of the coda itself. This figure serves a double function: rhythmically it is an ostinato, harmonically a pedal point (being based on a single note). My decision to extend the figure may appear to be a liberty not in keeping with Mahler's style. However, such devices can be found in other Mahler works, notably the figure in the timpani at the beginning of the penultimate movement of the First Symphony21 and the figure in the cellos and basses in the fourth movement of the Third Symphony.22 Although these examples involve two pitches rather than the single pitch in the figure in the Presto, the functions of ostinato and pedal point are similar. The fact that the figures in the First and Third Symphonies involve tempos much slower than the pace in the Presto suggests that Mahler was prepared to use such devices for substantial time blocks; the span of time involved in the use of the device in the Presto is only a fraction by comparison.
There is one passage in the trio of the Scherzo in which Mahler appears to have written subsidiary material rather than primary, and this problem is also not unheard-of in sketches and short scores of other Mahler works.23 In this case, a radical course of action seemed necessary. I wrote a melody into another voice which is not Mahler's own; however, it is not of my composition but is borrowed from an opera by a well-known contemporary of Mahler.24
The third task involved solution of the problems affecting the endings of both movements. In the Scherzo Mahler provided for a reprise; the question is how far forward to carry the reprise without going so far as to reach the trio again. (Since Mahler used both the classical and romantic forms of scherzo-trio in his other works, there are precedents for both forms to be used in this case; but a second trio in the case of this movement would probably throw the movement off balance without solving the problem of finding a place to stop.) There are several places in the opening scherzo section at which the line might be drawn suitably; they stand out from Mahler's score by virtue of double bars signifying the ends of sections at measures 19, 57 and 80. My solution was to take the reprise through the first eighty bars of the scherzo material—slightly over half the section—and, following the double bar after measure 80, add a five-measure codetta essentially based on the succeeding measures 81 and following. I then filled out the texture with an inversion of Mahler's melody in the bass line and static chords in other voices.
In the Presto there was a choice between two courses of action:
(a) Find a cut-off point after the reprise of the Presto and construct a codetta, similar to the solution of the problem in the Scherzo; or
(b) Diverge from the Presto into the transitional bars on the last leaf which precede the coda proper, in which case the task is to find a good divergence point.
The first alternative was untenable because it involved complete elimination of the third leaf, shortening the Presto by twenty-eight bars and weakening it thematically. Therefore, in choosing the second alternative, I made the divergence from measure 66. This is the unique bar mentioned earlier, which occupies the reverse side of the first leaf, quite alone; early in the movement it leads from the first episode into the truncated second statement of the main theme. It was obviously written on the reverse of the first leaf because of lack of space on the proper side, and must have been meant as a transition. Therefore I used it again as a transition to the bars preceding the coda; the inclusion of the coda gives the movement the most dramatic material in the entire draft, providing a good ending.
The fourth task involved alteration of the tonal center in the Presto from G Major to F Major, in accordance with Mahler's verbal note on the first leaf. For the first two leaves, this was a mechanical task; it merely involved transposition of every note down one step, alteration of the key signature from one sharp to one flat, and adjustment of accidentals. The third leaf, however, presented a different problem since it was based in D Major rather than G Major. Since Mahler experimented with progressive tonality, there are movements in his works which end in different keys from the ones in which they begin.25 However, in the early and middle works such examples were largely confined to outer movements. Internal movements, which the Scherzo and the Presto probably represent, generally went no further afield than the parallel major or minor.26 Therefore, given these precedents, I transposed all material on the third leaf from D Major to F Major in order to end the movement in the key in which it had begun.
The last task, orchestration, was in many ways the most difficult. That Mahler's concept was orchestral is clear from verbal notations in the short scores;27 although verbal indications are not numerous, I observed them carefully. However, since the bulk of the orchestration was mine, some observers may consider Mahler's lack of progress in this area to be sufficient grounds for disagreement with the entire idea of a performing version. But I would like to quote a statement written by Deryck Cooke while he worked on the Tenth Symphony, which seems appropriate in this case also:
I need no telling that to attempt to orchestrate Mahler would be a ludicrous impertinence. But 'orchestrate' is not quite the right word here: Mahler conceived his music orchestrally, and his short scores are blueprints for instrumentation; if studied and auralized persistently enough, they score themselves—in essentials.28
Cooke's observation is by no means mystical but quite pragmatic. Mahler wrote so characteristically for his instruments that it is possible to "hear" them when reading his short scores. For two reasons, a conservative selection of orchestral resources for the Scherzo and the Presto seemed highly desirable:
(a) The suggested provenance of both movements about the time of the Fourth and Fifth Symphonies recalls the unusual conservatism Mahler employed with wonderful effect in the Fourth, inviting comparison.
(b) Even in big works, internal movements are often modest in orchestral scope, by contrast with outer movements of the same works quite heavily scored. The two Nachtmusiken in the Seventh Symphony and the Adagietto of the Fifth Symphony are obvious examples.
Alma Mahler wrote an account of an experience involving the Fifth Symphony which serves as a lesson about how far the composer may go:
I had heard each theme in my head . . . but now I could not hear them at all. Mahler had overscored the percussion instruments and side drum so madly and persistently that little beyond the rhythm was recognizable. I hurried home sobbing aloud. . . . At last I said between my sobs: 'You've written it for percussion and nothing else.' He laughed, and then produced the score. He crossed out the side drum in red chalk and half the percussion instruments too. He had felt the same thing himself, but my passionate protest turned the scale.29
Taking into account the possibility that Alma was being dramatic at the expense of strict accuracy, the final impact of her comments—so far as the editor is concerned—is that the composer may overscore but change his mind later with impunity; unfortunately, the editor has no such freedom.
However, a point made by Deryck Cooke during his work on the Tenth Symphony shows the other side of the picture by pointing out what the editor may do in such a task as this:
It is not at all impossible to produce a performing version . . . which shall allow Mahler's . . . conception to speak out clearly. Admittedly, to make the sketch as it stands performable, it is necessary to provide a certain amount of conjectural addition to the texture, and a large amount of conjectural orchestration; but these elements, in comparison with Mahler's own 'fully-prepared' thematic, harmonic, tonal and formal argument, sink to the level of the subsidiary. . . . It can be argued . . . that the result would be artistically unacceptable, since it would not be a definitive, perfected, fully-achieved work of art. The answer to this is that we are here dealing with a unique case.30
In the case of the Scherzo in C Minor and the Presto in F Major, as in that of the Tenth Symphony, we are indeed dealing with a unique case, in which time and the opportunity to hear the works in performance can be the only way to determine if this music can stand with the works in the repertoire as a representative of Mahler's creative process.
1News About Mahler Research, No. 8 (February 1981), p. 3.
2J. Rigbie Turner, "Nineteenth-Century Autograph Musical Manuscripts in the Pierpont Morgan Library: A Check List (II)," Nineteenth Century Music IV, No. 2 (Fall 1980), 164. "[Unidentified Sketches]. 3 leaves. 26 × 34.5 cm. The bequest of Mrs. Wolfgang Rosé."
3"7 einseitig beschriebene Manuscriptseiten/völlig unbekannt/zu keinen der bekannten Werke Mahler/gehörig; auch nicht zur X. Symphonie/Es scheint sich um einen Scherzoartigen/Satz zu handeln 6/8 mit/einem eventuell(en) Trio (4) 2/4[.] Jedenfalls aus früherer, wenn nicht/frühester Zeit!/Kommen die vielen blauen Orientierungsziffern/aber aus der letzten Zeit, so kann höchstens/angenommen werden, dass Mahler das/Themenmaterial dieser alten Skizzen eventuell/später noch zu verwenden gedachte. . . ." I am indebted to Professor Hans Tischler for his aid in deciphering this material, which is written in the old style of Germanic handwriting. In my translation I have edited for the sake of clarity.
4Alma Mahler-Werfel, Mein Leben (Frankfurt am Main: S. Fischer Verlag, 1960); English version, And the Bridge is Love (New York: Harcourt, Brace & Co., 1958).
5Alma Mahler, Gustav Mahler: Erinnerungen und Briefe (Amsterdam: Albert de Lange, 1940); English version, Gustav Mahler: Memories and Letters, transl. by Basil Creighton (New York: Viking Press, 1946). Subsequent enlarged editions of these two books also lack references to the Scherzo and the Presto.
6Natalie Bauer-Lechner, Erinnerungen an Gustav Mahler (Leipzig: E.P. Tal, 1923).
7Ibid., p. 1 (reference to the symphony rehearsed by the orchestra of the Conservatory), p. 39 (reference to the Symphony in A Minor).
8To Anton Krisper, c. December 14, 1879. First published in Hans Holländer, "Unbekannte Jugendbriefe Gustav Mahlers," Die Musik XX, No. 11 (1928), 807-13.
9The discrepancy was mentioned during Mr. Hefling's response to a paper I presented at the 1983 meeting of the American Musicological Society in Louisville, Kentucky. Since, however, no colophon at all appears on these occasional pages, it is impossible to trace them to any specific lithographer in Vienna or any other location.
10"Die Druckerei [Anton von Haykul] ging über mehrere Zwischenbesitzer 1864 an R. von Waldheim über, der 1892 Josef Eberle als Gesellschafter aufnahm." Alexander Weinmann, "Wiener Musikverleger und Musikalienhändler von Mozarts Zeit bis gegen 1860: Ein firmgeschichtlicher und typographischer Behelf," Österreichische Akademie der Wissenschaften Philosophischhistorische Klasse Sitzungsberichte, 230. Band, 4. Abhandlung (Veröffentlichungen der Komission für Musikforschung, Heft 2, hrsg. von Erich Schenck, 1956), p. 21, section 9.
11Gustav Mahler, Zehnte Symphonie (Wien: Paul Zsolnay Verlag, 1924).
12There is a notable example at the end of the Adagio of the Tenth Symphony, in which Mahler refined his material in several sketches, made a short score—and then revised it with a truly remarkable sketch almost as good as the subsequent orchestral score. See Susan M. Filler, "Editorial Problems in Symphonies of Gustav Mahler: A Study of the Sources of the Third and Tenth Symphonies" (Ph.D. dissertation, Northwestern University, 1977), pp. 549-52, for a discussion of these materials.
13This in itself is an indication that the materials under discussion are not the earliest for the two movements; Mahler may well have made other sketches in various keys or used his sketchbooks.
14This may be compared with a similar expedient Mahler employed in other short scores, notably the Purgatorio movement in the Tenth Symphony.
15The key signature of two sharps is not found at the beginning of the leaf, but in the third system; however, the effects of accidentals in the first two systems result in a strong sense of D Major.
16It is found in the Tenth Symphony in the short score and sketches for the ending of the Adagio. See Filler, "Editorial Problems," pp. 533-52, for detailed discussion.
17Compare Das himmlische Leben, measures 95-104, with Third Symphony, second movement, measures 79-89. Another example of this technique can be found by comparison of Third Symphony, first movement, measures 14-23, with fourth movement, measures 1-17; the citation in the first movement is in common time, but the citation in the fourth movement switches back and forth between 2/2 and 3/2 time.
18This episode has some harmonic links with Episode A, which integrates the organization of the movement yet more closely.
19The Fifth Symphony proved to be an exception to this rule, since the rondo was the last movement; but since that movement was at least partially dependent thematically on the Adagietto preceding it, Mahler was obviously ready to suspend the usual rules due to the demands of cyclic procedure in the case of that symphony.
20For discussion of this problem, see Susan M. Filler, "The Case for a Performing Version of Mahler's Tenth Symphony," Journal of Musicological Research III, No. 3-4 (1981), 288-91; see also Filler, "Editorial Problems," pp. 554-56, 583-85, 594-96, and Parks Grant, "Mahler Research and Editing in Vienna," Chord and Discord III, No. 1 (1969), 101-15.
21Measures 1-18, reinforced by the string basses in measures 19-37.
22Measures 18-41, 47-56 and 94-99.
23Deryck Cooke pointed out such an example in the short score of the fourth movement of the Tenth Symphony, measures 301-08, in Gustav Mahler, A Performing Version of the Draft for the Tenth Symphony (New York: Associated Music Publishers, Inc.; London: Faber Music Ltd., 1976), pp. xix-xx, 175.
24Since Mahler is often criticized for lifting materials from other composers' works, in this case I acted at least spiritually in his tradition.
25The most obvious example is Das himmlische Leben, which ends in E Major although the main key of that movement is G Major.
26For example, the third movement of the Third Symphony begins in C Minor but ends in C Major.
27For example, in the Scherzo Mahler indicates second violins at measure 158 and first violins at measure 189.
28Deryck Cooke, "Mahler's Unfinished Symphony," Essays on Music: An Anthology from "The Listener," ed. by Felix Aprahamian (London: Cassell & Co., Ltd., 1967), p. 149.
29Alma Mahler, Gustav Mahler: Memories and Letters, p. 67.
30Deryck Cooke, "The Facts Concerning Mahler's Tenth Symphony," Vindications: Essays on Romantic Music (Cambridge: Cambridge University Press, 1982), pp. 93-94.