Mahler's Unfinished Symphony

Now that several versions of Mahler's Tenth have been recorded, listeners have a variety of ways to appreciate its greatness

SOON after Gustav Mahler died, on May 18, 1911, at the age of fifty, it became known that he had been working on a new symphony at the time of his death. It was clear that he had been unable to complete the work, but there were conflicting reports as to the state in which he had left it, and therefore as to whether it might someday be performed.

In a 1912 lecture Arnold Schoenberg, who was inclined toward numerology, alluded to the fact that neither Beethoven nor Bruckner had been able to get beyond a ninth symphony: "It seems that the Ninth is a limit. He who wants to go beyond it must pass away. It seems as if something might be imparted to us in the Tenth, for which we are not ready." Less mystically but just as pessimistically, Richard Specht, in his 1913 critical biography of Mahler, wrote, "It may be regarded as completely impossible that out of his mute symbols someone else -- no matter how intimately acquainted with Mahler's spirit and essence -- should fashion the full score." Yet where Schoenberg had spoken merely of "sketches," Specht referred to "an eleventh symphonic creation [he counted Das Lied von der Erde in with Mahler's nine numbered symphonies] drafted and completely written down in the short score" -- that is, on several staves, with suggestions for orchestration.

In the years following the First World War, Mahler's reputation steadily rose, especially among younger composers like Schoenberg and his pupils Alban Berg and Anton Webern. In 1920, to celebrate what would have been Mahler's sixtieth birthday, Willem Mengelberg and the Concertgebouw Orchestra, which Mahler had conducted on occasion, staged a Mahler festival in Amsterdam. A few years later Mahler's widow, Alma, decided that the world was ready to learn at least a little about her husband's last symphony.

Mahler had all but completed the scoring of the first movement, and so in 1924 Alma asked her son-in-law, the young composer Ernst Krenek, to touch up the first movement and to score the very brief third movement, which Mahler had titled "Purgatorio." These two movements were performed that October in Vienna under the direction of Franz Schalk, who may also have had a hand in the scoring. That same year Alma published a facsimile edition of Mahler's sketches for the Tenth Symphony. In her preface she wrote, "If at first I held it as my cherished right to preserve the treasure of the Tenth Symphony in secret, so I now know it to be my duty to disclose to the world the last thoughts of the Master."

The performances of the two nonconsecutive movements naturally gave little idea of what the symphony as a whole might be like. Moreover, the facsimile, although beautifully produced, was woefully incomplete, and most people who examined it thought it confirmed Specht's view that the work was unreconstructible.

Specht himself, however, in the 1925 edition of his biography, recanted that view. Moreover, he castigated what he termed the puritanism and false piety of those who said that performing the Tenth would dishonor Mahler's memory. Now that he had thoroughly analyzed the sketches, Specht wrote, he had concluded that although the three movements not performed by Schalk "obviously require filling out, in the counter-themes and inner voices, by a musician of high standing devoted to Mahler and intimate with his style," such a man "would surely find the right path to the goal." Specht suggested Schoenberg.

During the 1940s Alma Mahler and Jack Diether, of the New York Mahler Society, did appeal to Schoenberg, and also to Shostakovich. Both men, occupied with their own creative work, declined to take on the task.

IT was at about this time, in the mid-1940s, when I was in high school, that I became interested in Mahler. He was then still something of a cult taste in this country, and only a few of his major works had been recorded. One of the few friends who shared my love for his music was an older fellow named Clinton Carpenter, who worked at the Chicago music store where I bought my scores. In time Carpenter, a dedicated amateur, revealed to me that he was going to complete and orchestrate Mahler's Tenth Symphony. I had little idea what this might involve, but I still recall the neat timetable, taped to the glass door of his bookcase, that indicated exactly how long he planned to work on each movement.

I thought no more about the Tenth until around 1970, when someone gave me a copy of Pierre Boulez's superb recording of the first movement. A few years after that, driving home one night from dinner at a friend's house, I heard on my car radio what was obviously a Mahler symphony -- and a very fine one -- but not one of the canonical nine. It turned out to be the first recording of the entire Tenth Symphony, by Eugene Ormandy and the Philadelphia Orchestra, from an early draft of the "performing version" prepared by the English musicologist Deryck Cooke.

Clint Carpenter, I soon learned, had not been the only one interested in completing the Tenth. He had begun work in 1946, and was followed a few years later by an English composer named Joe Wheeler and a German, Hans Wollschläger. Cooke did not start work until 1959, but his first draft was almost finished a year later. That year, in a BBC broadcast in honor of Mahler's centenary, Berthold Goldschmidt, a composer and conductor who had been helping Cooke, led the Philharmonia Orchestra in sections of Cooke's score; Cooke gave a brief talk on the work. Soon afterward Alma Mahler, who had earlier approved Cooke's efforts, decided that she did not want the Tenth Symphony completed and performed after all. She issued a ban on all future performances. It seemed that Cooke's work, along with that of Carpenter and the others, had been in vain.

But in 1963, the year before her death, Alma listened to a tape of the BBC broadcast, was much moved by it, and changed her mind again. "I have now decided once and for all," she wrote Cooke, "to give you full permission to go ahead with performances in any part of the world." On August 13, 1964, the BBC aired the first complete performance of the first draft of Cooke's score, with Goldschmidt conducting the London Symphony Orchestra. A year later Ormandy recorded it.

Wollschläger eventually abandoned his score, but Carpenter and Wheeler completed theirs, in the mid-1960s. They had little success getting them played. Cooke had got there first -- and with Alma's blessing. Moreover, many people were firmly opposed to any performances of the entire Tenth Symphony, as distinguished from the first movement, on the ground that Mahler himself had not completed the work; to present it as his was to do his memory an injustice and so to commit an immoral act. It was perfectly understandable that an older man such as the great conductor Bruno Walter, who had assisted Mahler at Hamburg and Vienna and had conducted the first performances of Das Lied von der Erde and the Ninth Symphony, should take such a position. But coming from younger conductors, who had had no personal connection with Mahler, this seemed more of Specht's "false piety."

Another objection, made by conductors of all ages, was that the Cooke score at some points simply did not sound like Mahler. In the preface to his score Cooke had explained that it was "in no sense intended as a 'completion' or 'reconstruction' of the work." Mahler, one of the most fastidious and painstaking composers in the history of Western music, would surely have made many revisions and additions that only he could have imagined. "The idea that someone else can now reconstruct this process," Cooke wrote, "is pure illusion." All that Cooke sought to do was to "represent the stage the work had reached when Mahler died, in a practical performing version."

Thus it makes perfect sense that the absence of the contrapuntal writing that Mahler alone could have provided -- those "counter-themes and inner voices" mentioned by Specht -- might leave the Cooke score sounding too simple and bare to be genuine Mahler. The excellent Mahler conductor Michael Tilson Thomas once told me that he found Cooke's score "very empty-sounding." "There are so many places," he said, "where I just think, 'This is an outline, this is just an outline,' because none of the complex inner workings of the countersubjects and whatnot are present."

Lacking the ears and the training of a conductor, I hear no emptiness in the Cooke score. Anyhow, Mahler's textures in his earlier, completed works are often quite bare. More important, the sheer power and beauty of the music, as captured in Cooke's performing version, easily overwhelm whatever force these objections may possess. It is no wonder that over the past twenty years an increasing number of conductors have found Cooke's score very much worth performing and recording. As Cooke remarks in his preface, "The thematic line throughout, and something like 90% of the counterpoint and harmony, are pure Mahler, and vintage Mahler at that." Then again, if a conductor finds something lacking in Cooke's score, he is always free to add in performance what seems to be missing -- as several conductors have in fact done.

THE Tenth Symphony is the most perfectly and satisfyingly formed of all Mahler's major works. At either end stands a vast and complex movement twenty to twenty-four minutes in length, depending on the conductor. Next in from each end is a Scherzo, twelve to fourteen minutes in length. At the center sits the marvelous, enigmatic little "Purgatorio," which lasts barely four minutes. Resting complacently at the center of Mahler's long and often turbulent work, it reminds one of the secret in Robert Frost's two-line poem, "The Secret Sits":

We dance round in a ring and suppose,
But the Secret sits in the middle and knows.

Though the music of the longer movements is sometimes anguished, the Tenth Symphony is by no means as dark a work as either Das Lied von der Erde or the Ninth. Pliant, consoling dance rhythms keep breaking in upon the high drama of the first movement, reasserting the presence of the everyday world. The movement ends calmly and quietly. The two Scherzos are energetic, insistent, even at times menacing, but each one has two gracious, seductive trio sections. The magnificent finale begins very dramatically, with a slow introduction punctuated by repeated strokes on a muffled drum -- Mahler's memory of a heroic fireman's funeral procession that he and Alma watched from their New York hotel window in February of 1908. But the music gradually turns warm and serene. A vigorous Allegro follows, in which the battles of the other movements are fought once again: motifs from the "Purgatorio" recur in phrases that recall the first Scherzo, and the introduction to the first movement, there heard softly in the violas, is blared forth by the horns. But the earlier serenity gradually returns, and a three-note motif from the "Purgatorio" is transformed into what sounds like a remembered snatch of a touching folk song or lullaby. This gradually fragments and then -- after one brief, ecstatic outburst -- dies away, giving the ending a wry and gentle benedictory air.

It was this ending, so different from the deep sadness with which both Das Lied von der Erde and the Ninth Symphony end, that first attracted critical attention to the Tenth. In a 1962 essay, "The Facts Concerning Mahler's Tenth Symphony," Cooke spoke of the work's structure as presenting "an emotional and psychological statement as intelligible as those of Mahler's other symphonies"; he took "the 'message' of the symphony" to be "the final resolution of Mahler's lifelong spiritual conflict." In 1974, in his excellent little book on Mahler in the Master Musicians series, Michael Kennedy went further. He spoke of Cooke as having "completely altered the emotional interpretation of Mahler's last year" and as having demonstrated "that it is wrong to regard Mahler as having died in a mood of valediction, defeated or resigned to the inevitable." In contrast to Das Lied von der Erde and the Ninth Symphony, the Tenth "transcends thoughts of death and ends with a gloriously affirmatory and positive assertion of man's spiritual victory."

This seems to me to go too far: we have left the realm of music criticism and entered that of soap opera. It is perfectly appropriate to think of the ending of a symphony as exerting a special force in shaping our interpretation of the whole work, and hence to regard symphonies, especially perhaps Romantic symphonies, as having a sort of rough narrative form. But it seems wrong to string together a composer's symphonies -- even the symphonies of so autobiographical a composer as Mahler -- to form a larger narrative that is then assumed to be a blow-by-blow account of his emotional life.

It is surely right to say that the serene and consolatory ending of the Tenth is different from the endings of Mahler's two other late symphonic works. But to say that the Tenth thereby invalidates the popular image of Mahler as an egocentric, death-obsessed neurotic is to accord that image more attention than it deserves. For it has been transmitted to us not through Mahler's works but rather through Alma's memoirs. As even Kennedy argues, this image never did stand up well against the evidence of Mahler's letters, of other people's memories of him, or of his busy and highly successful conducting career. Its invalidation does not require the discovery (or invention) of anything so high-toned as an assertion of man's spiritual victory.

Despite their endings, Das Lied von der Erde and the Ninth no more support that image than the Tenth invalidates it. Like the Tenth, and like Mahler's other mature symphonies, they are far too varied in emotional content to admit of so simple an interpretation. In 1907, on a visit to Finland, Mahler had a discussion with Sibelius on the nature of the symphony, in the course of which he told Sibelius, "The symphony must be like the world. It must embrace everything." Nowadays these famous words are cited to show that Mahler wanted to open the symphony to all the highly personal, disturbing, lurid emotions supposedly eschewed by classicists such as Brahms and Sibelius. Mahler's words have been assumed to justify the popular view of him as an essentially confessional artist -- the representative composer of our troubled century, the supreme purveyor of musical angst. Every apparently cheerful Scherzo must be heard as eerily grotesque, with Hitler's jackbooted troops and the Holocaust lurking just around the next bend. The most beautiful slow movements are revealed as laden with pathos and bleak resignation. The gayest and most festive finales, closely inspected, show us that Mahler was merely whistling in the dark. It has been easy to forget that only one Mahler symphony, the Sixth, actually ends in the minor.

To interpret Mahler's analogy between a symphony and a world in this way is to miss its point. Of course there is a good deal of anguish and angst in his music; but there is a great deal else, too. He developed his analogy in conversations with his friends, particularly the violinist Natalie Bauer-Lechner, at far greater length than he evidently did with Sibelius. From his various remarks it is clear that for him a symphony needed to be all-inclusive -- to bind together through strictly musical means (primarily the development and linking of motifs) the most emotionally disparate materials imaginable, creating a self-contained world different from the world of any other symphony. Mahler did not, I think, intend each (or any) symphony to make a particular statement or convey a particular message in the sense that Cooke implied. The analogy stipulates that each work, taken on its own terms, will be far too emotionally wide-ranging to yield any such single statement or message. Much less did Mahler intend his successive symphonies to form a spiritual autobiography, even though they contain many incidental effects -- the muffled drum in the Tenth, for example -- of autobiographical origin. The respective worlds of the various symphonies are far too different from one another for the works to form a continuous narrative.

The predictable result of this misinterpretation of Mahler's intentions has been the view, common among music critics, that any performance of a Mahler symphony that fails to find bleakness, desolation, neurosis, and even hysteria in virtually every bar is to be dismissed as cold, pedantic, uninvolved. This goes dead against the all-inclusiveness that Mahler so consciously sought and, in his best works, achieved. The finest of our present-day Mahler conductors -- Boulez, Claudio Abbado, Bernard Haitink -- do not impose a one-size-fits-all interpretation on his symphonies. Rather, they take each movement as they find it, bringing out the extreme contrasts among and within movements to give the works their emotional breadth and depth. It is to be regretted that so far none of these three has recorded the entire Tenth Symphony.

YET there are excellent recordings available. The two best are Kurt Sanderling's 1979 performance with the Berlin Symphony Orchestra (Ars Vivendi 2100225) and Simon Rattle's 1980 performance with the Bournemouth Symphony Orchestra (EMI CDC 54006). James Levine's recording with the Philadelphia Orchestra (RCA RCD2-4553), made in 1978 (the first movement) and 1980 (the other movements), is also very fine, but has one drawback: Levine, apparently convinced that the Tenth Symphony ends tragically, takes the finale almost intolerably slowly.

The two earlier recordings, Ormandy's of 1965 and Wyn Morris's of 1972 with the New Philharmonia Orchestra, which was the first recording to use the final draft of Cooke's score, are not in print and not worth looking for. Two later performances, by Riccardo Chailly and the Berlin Radio Symphony Orchestra (London 421 182-2) and by Elihu Inbal and the Frankfurt Radio Symphony Orchestra (Denon CO-75129), from 1986 and 1992 respectively, are commendable but not the equal of Sanderling's, Rattle's, or Levine's.

All these recordings are of Cooke's score. In the past couple of years we have finally been given the opportunity to hear the Carpenter version, which has been recorded by Harold Farberman and the Philharmonia Hungarica (Golden String International GSCD 024A). Unlike Cooke, Carpenter decided that some free composition of inner voices was required if the Tenth, as it stood in Mahler's short score, was to be made to sound like a genuine Mahler symphony. To anyone who has listened for twenty years to Cooke's score, Carpenter's at first sounds very cluttered. Yet the more one listens, the more it sounds like a credible alternative. If we are really to judge Carpenter's work fairly, however, we need a better performance than Farberman's. All the movements except the "Purgatorio'' sound as though they are being taken too fast, and this turns out to be the case: the performance instructions that circulate with rental copies of Carpenter's score give timings for the five movements that add up to fourteen minutes more than the total time of Farberman's performance.

Equally interesting is the performing version by Remo Mazzetti Jr., completed in 1986, which has been given an excellent performance by Leonard Slatkin and the St. Louis Symphony Orchestra (RCA 09026-68190-2). Mazzetti, who first got interested in the Tenth in the mid-1970s, helped Carpenter to prepare his score for its New York premiere, in 1983. He soon came to feel that whereas Cooke had not added enough to Mahler's short score, Carpenter had added too much. So Mazzetti prepared a version that adds more than Cooke but less than Carpenter. On first hearing, Mazzetti's scoring seems overelaborate, though very elegant. But after so many years of Cooke any other score is bound to sound a little wrong at first.

Joe Wheeler's version has never been commercially recorded, though private tape recordings are known to exist. The score is, however, available for inspection, and a cursory reading reveals that Wheeler has added even less than Cooke, producing a score that looks very bare and simple.

Plainly, what is needed is for all four versions to be given adequate recordings, so that they may become familiar not only to ordinary listeners but also to conductors who might be interested in performing the Tenth. Slatkin has helped this process along by including a disc on which he makes several comparisons among the versions, with musical illustrations provided by his orchestra. Once all the versions have become well known, conductors can either choose one or prepare their own mix-and-match versions. This last may seem a wild suggestion, but the late Georg Solti, in his recently published Memoirs, wrote that he was planning to do just that. After remarking that he had not performed Cooke's score because "I think it lacks the contrapuntal element in Mahler's writing," Solti continued, "Three further versions of the Tenth Symphony exist or are in preparation, and in the summer of 1999 I would like to work on a solution to the symphony, putting together the different reconstructions and adding points of my own."

One's sadness that Solti did not live to carry out this plan is tempered by a sense that if such a suggestion could come from so firmly established a conductor, then Mahler's Tenth is at last ready to assume its rightful place in the permanent repertory. When it does -- when, that is, it ceases to be treated as an affront to Mahler's memory, or as an incomplete and thus necessarily flawed work, or as welcome proof that at the end of his life Mahler came to have faith in man's ultimate spiritual victory -- then perhaps it will be recognized for what it is: the greatest work its composer has left us.

William H. Youngren is a professor of English at Boston College. He has just completed his doctoral dissertation in musicology at Brandeis University, on the songs of C.P.E. Bach.

Illustration by Max Oppenheimer

The Atlantic Monthly; December 1998; Mahler's Unfinished Symphony; Volume 282, No. 6; pages 98 - 105.