Du allein weisst was es bedeutet [You alone know what it means]
– annotation by Mahler on the manuscript of the symphony

When the Austrian composer Gustav Mahler died in May 1911 aged only 50, he left two completed but unperformed works: the symphonic song-cycle Das Lied von der Erde [The Song of the Earth] and the Ninth Symphony. Both works were given posthumous first performances by Mahler's close friend and protégé, the conductor Bruno Walter: Das Lied von der Erde in Munich in November 1911, and the Ninth Symphony in Vienna in June of the following year. In addition to these two compositions, Mahler left sketches and manuscripts for an incomplete Tenth Symphony.

Composition

The music of the Tenth Symphony as Mahler left it is the product of a single summer, 1910, a summer in which he would experience both the highest of highs and the lowest of lows.

Since May, Mahler had been heavily committed to rehearsals for the upcoming world première of his Eighth Symphony – the so-called Symphony of a Thousand – due to take place in September. With the vast resources involved, including, as well as a huge orchestra, eight solo voices, two mixed choruses and a children's chorus, rehearsals were many, and of necessity held in different cities in Germany and Austria. The logistics of these rehearsals, together with the many organisational arrangements to be made for the première, exhausted Mahler. He arrived in the south Tyrolean alpine village of Toblach¹ – to which he had come each summer since 1908 to compose in peace – on 4 July, just three days before his 50th birthday. By the end of August, most, if not all, of what is now left to us of this final symphony had been written down in various sketches and the manuscript.

Mahler's composing peace was to be short-lived, shattered at the end of July by the discovery of his wife Alma's infidelity and her affair with a young architect by the name of Walter Gropius². Mahler was devastated and we can understand the effect this had on the music he wrote; the two months of composing in 1910 are divided by the event. That he was able to achieve anything at all after this revelation is remarkable, but the music reveals the turmoil that occupied his mind that August. The manuscript bears physical witness to it.

Mahler's normal practice was to do as much original composition as possible during his summer holidays, and to work on the orchestration of the summer's work in whatever spare time he had available during the winter months of the opera and concert season. Mahler's commitment as conductor of the New York Philharmonic Society, recurring bouts of illness and making revisions to the Ninth Symphony precluded any significant further work being done on the Tenth Symphony before his death on 18 May, 1911.

What did Mahler leave us?

In 1924, Alma Mahler permitted a high quality, photographic facsimile of the manuscript to be published in Vienna. For the first time it was possible to study the composer's final, but incomplete work. The 165 pages³ of manuscript revealed a plan for a symphony in five movements:

1. Adagio – 275 bars⁴
2. Scherzo I – 522 bars
3. Purgatorio – 170 bars
4. Scherzo II – 578 bars
5. Finale – 400 bars

The first movement Adagio, and about the first 30 bars of the Purgatorio movement are written out in draft full score. The remainder exists only in what is known as Particell – a short score, similar to a piano reduction. What is important however is that the symphony's skeleton is complete. There are no missing chapters; the musical plot, every bar of it, is all there. Mahler himself referred to it as: a work fully prepared in the sketch. We cannot be certain however that had Mahler been able to finish the symphony, the movements would have remained in their present order. There is evidence that some re-ordering had already taken place, and it is known that Mahler made similar changes to some of his previous symphonies at a very late stage in their composition.

Genesis of a Performing Version

At Alma's request, an edition of the Adagio and Purgatorio movements was made by the composer Ernst Krenek⁵. Fellow composer and Mahler devotee Alban Berg was invited to proofread the edition. He made a number of suggestions, none of which were incorporated into the score, but some changes, probably by the conductor Franz Schalk and by Alma's former music teacher, Alexander von Zemlinsky, were introduced. This edition – given its first performance by the Vienna Philharmonic Orchestra conducted by Schalk on 14 October, 1924, – was played a number of times in concerts thereafter. Up until only shortly before her death in 1964, Alma Mahler forbade any performance or 'completion' of the work, other than the Krenek edition. During the 1940s, requests to the composers Arnold Schönberg, Dmitri Shostakovich and Benjamin Britten to complete the symphony were (quite rightly) turned down.

After the Second World War, it occurred to a small number of musicologists that perhaps there was sufficient detail in the facsimile manuscript and sketches to enable a 'Performing Version' to be realised. The first to be ready was by an American, Clinton Carpenter, who completed a version between 1946 and 1949. In Britain, amateur trumpet player Joseph H Wheeler was working on a version from 1953.

In 1959-60 however, there was a significant step forward. The English musicologist, Deryck Cooke, together with composer Berthold Goldschmidt, produced a version that was part-performed by the Philharmonia Orchestra, conducted by Goldschmidt. This partial performance had been tape-recorded, and was broadcast on BBC radio on 19 December, 1960, as part of a workshop lecture. Initially, Alma Mahler was angry about this, and demanded that no further performances take place, a veto strongly supported by Bruno Walter. However in 1963, after Walter's death, Alma was persuaded to listen to a tape of the broadcast. This was the turning point; Alma withdrew her objection to Cooke's version. In a letter to him, she wrote:

I was moved to tears, I had not realised there was so much Mahler in it.

Not only did she withdraw her objection, but she made some previously unpublished material available to Cooke. The veto hurdle cleared, a full public performance was now possible, and this took place at a BBC Promenade concert at the Royal Albert Hall in London, on 13 August, 1964⁶. This version will be referred to here as Cooke-I. Given the new material and the added confidence of Alma's acceptance of the first version, Cooke, assisted by Colin and David Matthews, went on to produce a second, more complete version (Cooke-II) which was published in 1976, shortly after Cooke's death⁷.

We must be absolutely clear that Deryck Cooke's Performing Version was in no way an attempt to complete Mahler's unfinished symphony – not a single bar of free composition was added. It represents, to use Cooke's own words:

…the stage the work had reached when Mahler died, in a practical performing version.

With great skill, Cooke's team took the available material, executed the orchestration where it is directly indicated in the manuscript, and provided some missing, but inferable, supporting harmonies and some counterpoint – in general adding only the absolute minimum to implement Mahler's clear intentions. Although the facsimile is of good quality, it is not a publisher's printed score, nor is it even a fair copy manuscript in the composer's hand. It is very much a work-in-progress document with scrawled corrections, additions and deletions. A good deal of what is written is unclear and is certainly open to alternative interpretations. In attempting to read the document, familiarity with Mahler's musical handwriting is essential.

We must also be clear that what we hear – less so perhaps with the Adagio – is very far from being the work Mahler would have left had he been able to complete it. Fairly complete though the structure may be, about half of the first Scherzo and most of what follows it, apart from the first 30 bars of the Purgatorio, is entirely bereft of detail. So why attempt a performing version? To quote Deryck Cooke again:

Mahler's music, even in its unperfected and unelaborated state, has such significance, strength and beauty, that it dwarfs into insignificance any uncertainties.

If nothing were done, other than to have the facsimile and other sketches in the public domain, only those with the ability to read the score and hear in their mind what is written on the page would have any concept of where Mahler was heading with his new symphony.

Musical Form and Outline

The overall structure of the work shows a high degree of symmetry: the two slow, large scale outer movements, the second and fourth movement Scherzos and the little central Purgatorio. Even in its fragmentary form, the Tenth Symphony shows a clear line of continuity with Das Lied von der Erde and with the Ninth Symphony.

Orchestration

Although Mahler did not leave a ready-made list of the instruments of the orchestra required for a performance, there is sufficient information in the draft scores and the sketches to infer with a fair degree of accuracy what that list should be. In addition to the usual complement of violins, violas, cellos and double-basses, the probable orchestra comprises:

Woodwinds:

• Four flutes, with one player doubling on the piccolo that is specified once
• Four oboes
• Four clarinets, including the one in E flat that is specified
• Four bassoons, with two of the players doubling on contra-bassoons

Brass:

• Four horns
• Four trumpets
• Four trombones
• Tuba, specified once

Percussion:

• Timpani
• Tam-tam
• Bass drum
• Cymbals

To these can confidently be added a bass clarinet, a cor anglais (doubled by one of the oboists) – these two having been used in all the other symphonies – and at least one harp (only the Sixth Symphony does not call for one).

First Movement

The long first movement is the most complete of the five. It is written out in draft score and is capable of being performed standalone – since 1924 it often has been. Unusual in being an Adagio first movement, it seems to continue in the mood in which the Ninth Symphony ended. It is based on two thematic subjects: the first – a cold, bare theme given to the violas alone, acting as a recurrent motto – opens the movement. This is followed by the main theme, unmistakably Mahlerian, for combined strings and trombones, then the motto theme in expanded form.

For most of the movement, these two themes form an uneasy alliance, the main theme remaining anchored firmly in the symphony's home key, F sharp. Then suddenly, about three-quarters of the way through the movement, the alliance fails, and a fearful, organ-like A flat minor brass chorale breaks out. Worse is to come, no sooner has this subsided when it is followed by a terrifying nine-note dissonant chord building across the whole orchestra, with the piercing top note of the chord being held over on a solo trumpet. For the moment, the crisis is spent and the movement ends in the contemplative sound world that pervades both the end of the Ninth Symphony and the concluding Der Abschied movement of Das Lied von der Erde.

Second Movement

The second movement Scherzo, the first of two, shows significant signs of having been written in a hurry. Apart from at the very beginning, the movement is very short on detail, with only the barest indications of orchestration. Can we infer from this that it was the last to be composed? If we can, then this movement really is the last music that Mahler wrote. However there is nothing to substantiate the inference, so unless any new documents come to light, it must remain no more than a possibility.

Taking the previous symphonies as a guide, we might expect a Mahler Scherzo movement to be in the familiar A–B–A format; often a Ländler⁸ with a middle Trio. Mahler has both Ländler and Trio elements here, but with a couple of twists. The movement has two Trio sections; the first, in the home key of F sharp, rather than being a clearly contrasting section to the Scherzo, is instead a variation of it, whilst the second takes the slow main theme of the first movement and presents it up-tempo as a Ländler in E flat. The Scherzo itself, together with the two Trio themes, are varied and repeated throughout the movement. In the Coda, the two themes combine, and the movement rushes to a conclusion redolent of the Scherzo of the Fifth Symphony.

Third Movement

As mentioned previously, it was in the middle of working on this symphony in the summer of 1910, that Mahler discovered his wife's infidelity. It is likely that this little B flat minor Allegretto movement was written immediately after that traumatic event. It is a short movement, shot throughout with the recurrence of a little three-note figure. The title page of the manuscript bears the inscription Purgatorio oder Inferno  – the word Inferno is then crossed out, although not at the time it was written. It is thought that the title might be a reference to a set of poems by Mahler's friend Siegfried Lipiner entitled Il Purgatorio, and perhaps to one in particular about the feeling of shame in betrayal. The lower part of the title page has been cut away, probably by his widow Alma, in which case it would have been done in order to hide from the world whatever Mahler had written there. She may also have been responsible for the deletion of the word Inferno. Mahler wrote a number of little annotations over the music in this and the subsequent movements. One such in the Purgatorio is: Erbarmen!! [Have Mercy!!], a direct reference to Richard Wagner's final music drama Parsifal⁹.

At the start of the recapitulation section, Mahler wrote simply Da capo [(repeat) from the beginning], but as Cooke writes:

The Da capo…would no doubt have been realised by Mahler with a varied and possibly extended repeat, rather than with an exact one…

Fourth Movement

Der Teufel tanzt es mit mir [The Devil dances it with me] is Mahler's inscription on the title page of this Scherzo movement, a swift-paced waltz in E minor. Like the first Scherzo, this one has two Trios, the second of which has a joyful air about it. It also includes a direct quotation from the opening movement of the composer's symphonic song cycle, Das Lied von der Erde.

The energy of the opening and the joy of the second Trio however cannot be sustained, and the dance collapses leaving only near-inaudible instrumental fragments, accompanied by bare percussion sounds. The movement ends with a loud, single stroke of a muffled bass drum. This relates to an incident that Mahler experienced in New York in February 1908. He was stood at the window of his hotel suite overlooking Central Park, when the funeral cortège of one of the city's firemen passed by on the street far below him. The only sound rising up to him was that of a muffled drum. It made a great impression, and Alma relates that tears streamed down his face as he watched. It is to this incident that Mahler is referring in his annotation quoted at the top of this Entry.

Fifth Movement

The muffled thud at the conclusion of the previous movement is carried over into the D minor opening of the final movement, revealing a world of abject nihilism. This is Mahler's soul, and anguish at his wife's infidelity, totally exposed. The drum stroke is all-dominant; any attempts by instruments of the orchestra to start an idea – among them the little three-note figure from the Purgatorio movement – are ruthlessly crushed each time by a resounding thud. Eventually, a solo flute establishes an extended melodic line of incredible beauty, its contrast with the desolation of the opening making it even more breathtaking. The upper strings, now in the glorious key of B major, carry the baton on, taking the music to a sumptuous climax, at the height of which the bass drum stroke again intervenes, and the desperate mood of the opening returns, the muffled thuds beating down all opposition.

Quite unexpectedly, the tempo increases for the Andante moderato section, making use of material from the third movement. Its initial comparative lightness provides a welcome contrast to the extreme gloom of the preceding section. Later, the flute melody from earlier is re-introduced, with an important role for a trumpet in its highest register. The Andante tempo returns but soon the mood darkens. The high trumpet is left by the orchestra holding a solo note, a prequel to two returns of the nine-note dissonant chord heard in the first movement, with the solo trumpet note sustained over the top. The note still held, the horns steal in with the viola theme from the opening of the symphony. All the anger now subsides, the conflict is resolved and Mahler gives us his final Adagio, formed around the embryonic flute melody which is taken up finally by the whole string section in the symphony's home key, F sharp major. The movement, the symphony and Mahler's journey ends quietly in peace and love.

Other Realisations

Although Deryck Cooke's (second) version is by far the best known, other realisations have been attempted. Mention has already been made of those by Clinton Carpenter and Joe Wheeler (who not without coincidence played in the same amateur orchestra as Cooke), which were developed pretty well concurrently with Cooke's work, but certainly without detailed knowledge of its content. Wheeler's fourth and final version was given a partial performance in 1966; that by Carpenter, although finished in 1949 and revised in 1966, was not performed until 1983¹⁰. A third version by Cooke (Cooke-III) with only small differences in detail to Cooke-II, was published posthumously in 1989. The 1999 live CD recording by Simon Rattle and the Berlin Philharmonic Orchestra uses this version, but also includes some additional changes by Rattle himself.

A second wave of workers has had the benefit of the knowledge of their predecessors. Remo Mazzetti, an American musicologist, began a Performing Edition in 1983 that was premièred in 1989. However, a performance of Wheeler's version in 1997 prompted Mazzetti to revise it yet again in 1999. There are also a Performing Version by the Russian conductor and viola player, Rudolf Barshai (2001), and a 'Reconstruction' by the Italian team of Nicola Samale and Giuseppe Mazzuca (also 2001).

To Hear or not to Hear?

… that is indeed the question. In touching work that was left uncompleted at the creator's death, do we perform an act of reverence or one of sacrilege? In his 1971 book Mahler: Eine musicalische Physiognomik (A Musical Physiognomy), the German philosopher and musicologist Theodor Adorno was unequivocal:

…precisely someone who senses the extraordinary scope of the conception of the Tenth ought to do without adaptations and performances. The case is similar with sketches of unfinished pictures by masters: anyone who understands them and can visualise how they might have been completed would prefer to file them away and contemplate them privately, rather than hang them all on the wall.

Contrast that with the view of the American writer and Mahler enthusiast Jack Diether:

It is much more important that what Mahler wrote should be heard, than that which he did not write should not be heard.

In the case of Mahler's Tenth Symphony, at least we know precisely what Mahler himself wrote – in so far as we are able to interpret his handwriting – whereas taking the example of another work left uncompleted at the composer's death –  Mozart's Requiem K.626  – from the Sanctus onwards we cannot be certain as to what is by Mozart, and what was added later by Süssmayr.

Those detractors who insist that the Tenth Symphony should have remained for visual study only should ask themselves this: What would Carl Maria von Weber have thought of Mahler's completion of Weber's unfinished opera, Die drei Pintos? For that work, in 1887, at the request of the composer's grandson, Mahler wrote a considerable amount of new music, albeit based on existing Weberian tunes.

Mahler's Tenth Symphony must be heard for what it is, a tantalising glimpse into an unknown and unknowable future.