Bruckner's Symphony No.9: Finally, a Finale? Recordings of the Completed Finale

Sidebar: Recordings of the Completed Finale

The missing passages composed by the various "completers" surveyed below—particularly the coda, for which the least evidence in Bruckner's hand has survived—cannot be considered to be what Bruckner himself would have finally composed or approved or wanted performed. They are thus, ultimately, speculative interpolations, and can never be more than that. However, each completion gives us a good idea—some so good as to be utterly convincing—of at least the scale of the work Bruckner originally envisioned. And in the best of them, every note is in Bruckner's own idiom, if not his hand.

In the brief description of each recording that follows, I discuss only the Finale. Unless otherwise noted, assume that the first three movements are one of the standard editions of the Ninth Symphony, and that any remarks about the conducting or playing of the Finale, or the sound of the recording, can be applied to the first three movements as well. With the exception of the Eliahu Inbal recording, all include the symphony's first three movements.

Most of these recordings are in print and easily available through or Others are available in the US only through, a fascinating and comprehensive discographical website devoted to recordings of Bruckner's orchestral music; I indicate which these are in the head matter.

I gratefully acknowledge the assistance of Benjamin-Gunnar Cohrs, William Carragan, and John Berky in the preparation, revising, and updating of this article.

The Fragments
Nikolaus Harnoncourt, Vienna Philharmonic
RCA/BMG 54332-2 (1 SACD/CD, 1 CD). TT: 58:54. Fragments: 17:58

The best place to begin an exploration of the Finale is Nikolaus Harnoncourt's fascinating lecture-demonstration of only the fragments left fully orchestrated by Bruckner—about 18 minutes' worth of music. In his narration before a live audience, Harnoncourt, the highest-profile conductor to even approach "the Finale problem," however gingerly—he seems not to want to even consider performing any of the completions—tells the story of Bruckner's last days, then conducts the first fragment, after which he explains how many bars are missing. He then conducts the next fragment, and so on. The entire sequence is recorded twice, once each with commentary in German and English. Also included is an alternate version of a fragment in which Bruckner's plangent dissonances were "corrected" by an early editor, to give an idea of how long it has taken for this music to be understood on its own terms. Harnoncourt ends with a heartfelt plea to anyone who might be hoarding any of the still-missing pages of the Finale to please send them along—anonymously if necessary, photocopies accepted. Since the release of this set, more fragments have indeed surfaced, but it remains a solid introduction to the task faced by the various completers, to exactly how much of the finished movement has come down to us in Bruckner's hand, and to how important this music is. Beautifully performed by the Vienna Philharmonic, in limpid high-resolution sound. (Peter Hirsch and the Berlin Radio Symphony Orchestra have also released, on Sony Classical SK 87316, a performance of the fragments. However, it was never released in the US, is currently out of print, and I have not been able to track down a copy.)

Alternate Finale composed by Peter Jan Marthé
Peter Jan Marthé, European Philharmonic Orchestra
Preiser PR 90728 (2 CDs). TT: 100:41. Finale: 30:14

Peter Jan Marthé is a musico-mystical Austrian wildman of a conductor who recorded three of his Reloaded versions of Bruckner symphonies (3, 5, 9) before his European Philharmonic Orchestra disbanded in 2009. For the Finale of the Ninth, instead of composing links no longer than needed to bridge Bruckner's surviving fragments and complete the orchestration, he came up with his own free—and freewheeling—composition. It is only loosely based on the surviving fragments, some of which he rewrites, or discards altogether. He also adds cymbals, triangle, and contrabassoon to the entire symphony. The result is based at least as much on Marthé's conviction that he was channeling Bruckner's spirit than on what Bruckner wrote. As he says in his liner note, Marthé believes that Bruckner composed not so much "symphonies" as "shamanic . . . archaic rituals of sounds . . . the completed Ninth [is] a four-piece act of initiation."

Think of all that what you will, this needs to be heard to be believed. This 30-minute Finale sounds only intermittently like Bruckner and is, in a phrase, over the top. Marthé at times paints himself into musical corners from which he extracts himself only with the most extreme form of musical bootstrapping, with bombastic, even histrionic orchestral rhetoric that ends up sounding less like Bruckner than what those who loathe the composer's music are convinced it sounds like.

Still, Marthé should not be automatically dismissed. Despite some fanfares that sound more like Mahler, there are moments of genuine inspiration: a new brass chorale Bruckner might not have been ashamed to have written, a final quote of the Adagio's main theme, and then a greatly extended coda based on the climbing modulations of the Adagio of Symphony 7. No, they're not Bruckner, Marthé's composition doesn't ultimately hang together, and he doesn't know when to stop. But as a conductor he is a genuine Brucknerian who, to my ears, deeply understands what is most important about Bruckner's music—the combinations of richness and spareness, sonority and space, silence and sumptuousness, vastness and humility—and, in the first three movements, proves his ability to get an orchestra to embody it all in sound. His recording of the Reloaded Symphony 5 is probably one of the best ever made—Marthé makes the final double fugue of that symphony "speak" more clearly than anyone since Sergiu Celibidache.

Marthé's recordings of Bruckner, including the Ninth, are among the slowest, for reasons of his own taste and because all were recorded in the vast cathedral space of the Stiftsbasilika St. Florian, where Bruckner was organist for so many years and is now buried—Marthé conducts the hall as much as his band. The sense of space is immense as, in the pauses between Bruckner's great periods, the reverberation seems to endlessly die away.

Completions by William Carragan

1983 edition
Yoav Talmi, Oslo Philharmonic
Chandos CHAN 7051(2) (2 CDs). TT: 81:56. Finale: 21:58. Fragments: 15:53 (out of print on CD; available as a download from

This completion is elegantly idiomatic. A decided minimalist, William Carragan does as little as possible in constructing transitions between the fragments, and they are all the more convincing for it. Though occasionally awkward, these are genuinely creative in sounding not as one would expect Bruckner to sound, but as one might expect to be surprised by Bruckner. This is one definition of creativity or genius: that which is utterly surprising while perfectly conforming to the voice or idiom already established. Carragan has been careful to incorporate every note that Bruckner wrote, sacrificing nothing original. But Bruckner, of course, was not himself a minimalist: I find Carragan's coda, and his Finale overall, to at least feel too short—coda, movement, and symphony all seem to end too soon, skewing the scale of the entire work, which, at the close, sounds almost abruptly attenuated.

This set, like Harnoncourt's, includes separate recordings of the fully orchestrated fragments, though these are even more fragmentary—Harnoncourt had the benefit of another two decades' worth of further scholarship and rediscovered autograph pages. Yoav Talmi's sensitive leadership of the Oslo Philharmonic makes great sense of all four movements with clarity and careful phrasing, and the recording is warm, spacious, and luminous.

2006 edition
Akira Naito, Tokyo New City Orchestra
Delta Entertainment DCCA-0032 (CD). TT: 77:36. Finale: 22:51 (available from

Carragan's recent revision of his original completion is better in every way, though this recording is marred by too-forceful timpani strokes at the precise midpoint of the movement, in the great "sunrise" fanfare—a leap from mf to ff that obscures the bounding string figures that here so perfectly counterpoise the brass, and that fulfill such an important role throughout the movement. Still, the coda is much improved, with pealing, cascading brass; in fact, this is the most joyful of all attempts at a coda, though in this recording it still sounds as if wrapped up too soon. However, in a November 2009 performance in Scottsdale, Arizona, by Warren Cohen and MusicaNova, it worked astonishingly well. That event was the first performance of Carragan's new, 2009 edition of the Finale, which in summer 2010 will be performed and recorded at the Ebracher Festival, in Ebrach, Germany, by Gerd Schaller and the Philharmonie Festiva. Another curiosity of this disc is Carragan's revision of the second-movement Scherzo, in which he has reinstated Bruckner's original and far inferior Trio, which the composer discarded early on. As far as I know, this is the only recording of this curiosity.

Although Carragan's Finale is now the longest in sheer number of measures, it and the coda still feel too short to me—coda, movement, and symphony all seem wrapped up too soon. However, this may be as much a result of the tempo chosen as of the completion itself. Carragan, who has made a study of the steady slowing of the tempos taken in Bruckner's symphonies since the composer's time, insists on a brisk pace that I find too fast overall, at least under Akira Naito's baton—ultimately, a matter of taste.

Postscript: Gerd Schaller’s recording of Bruckner’s 9th with the Carragan Finale was Stereophile's "Recording of the Month" for November 2011.

Completion by Sébastien Letocart
Nicolas Couton, MÁV Symphony
Lirica CD-107 (2 CDs). TT: 82:04. Finale: 24:45 (available from

This intelligent and authentically Brucknerian completion by the Belgian-born composer Sébastien Letocart, still in his 30s (the completion and its exegesis comprise his PhD thesis), is intermittently somewhat awkward and extremely convincing. Overall it works very well, but its real strength is its coda. Letocart is perhaps the only completer to quote all of the themes from the symphonies 5, 7, and 8 that Bruckner may or may not have intended to incorporate into the coda. Here Letocart has written some thrilling modulations and superimpositions of themes—as many as four at once—that not only work superbly on their own terms, but also within the contexts of movement and symphony. And if Letocart's coda is still not quite of the scale Bruckner may have envisioned, it comes very close (but see Samale-Mazzuca-Phillips-Cohrs 2008, below). I would love to hear a revision of this, should Letocart write one.

The conductor is second-rate, the orchestra third-rate, the sound muddy. However, the liner notes include not only references to the folio and bar numbers of the score, but also to the minutes and seconds of this recording. You can watch your CD player's time display to hear exactly where Bruckner leaves off, Letocart picks up, and Bruckner then resumes. (The only other recording to do this is the Carragan/Naito.)

Completions by Nicola Samale, Giuseppe Mazzuca, John A. Phillips, and Benjamin-Gunnar Cohrs

1985 edition (Finale only, with Symphony 5)
Eliahu Inbal, Frankfurt Radio Symphony
Teldec 8.35785 (242 426-2) (2 CDs). Symphony 5, TT: 65:03. Finale: 20:44 (out of print)

Nicola Samale and Giuseppe Mazzuca were still feeling their way in this, the first edition of their long project of reconstruction. Some of the new linking passages are awkward and timid, not built fully or boldly enough to sound like much more than flimsy scaffolding connecting large blocks of ruins. The quotations from the earlier movements, and the very few from earlier symphonies, sound perfunctory. All of which makes this first, rudimentary stage of this series of completions, in combination with Eliahu Inbal's brisk pace, sound less like a completion than a condensation.

Still, Inbal is never less than deftly musical, with a cantabile feel very like that heard throughout his Mahler cycle but seldom heard in Bruckner, let alone any other completion of the Finale. The FRSO sounds lovely in the spacious acoustic of Frankfurt's Alte Oper. For some reason, Teldec paired the Finale with Inbal's recording of Bruckner's Symphony 5. Inbal's recording with the FRSO of the first three movements of Symphony 9, once available separately, is also out of print.

1992 edition
Kurt Eichhorn, Bruckner Orchester Linz
Camerata 30CM-275-6 (2 CDs). TT: 92:47. Finale: 30:11

Samale and Mazzuca were joined by composer-scholars John A. Phillips and Benjamin-Gunnar Cohrs to create this edition in which the coda has been completely reworked, and now includes themes from Bruckner's Symphony 7 and Te Deum. The result sounds like Bruckner almost all the way through, though parts of the new coda still sound like boilerplate musical rhetoric, however Brucknerian. (Those who have heard the coda of Bruckner's original version of the first movement of his Symphony 8 will know what I mean.) This is partly the fault of the conductor, the late Kurt Eichhorn (1908–1994), a Bruckner specialist of the craggy Knappertsbusch school: ie, deep feeling married to a great sense of a work's overall structure and arc, but considerable disregard for polish and tight ensemble work. Though Eichhorn is unfailingly musical, his very slow pace, although never too slow, at times simply loses momentum, most tellingly in some of the new transitional passages. But there are glorious moments here: the "sunrise" fanfare and its reprise get a more richly sumptuous treatment than in any other recording. The horn section alone seems to number dozens, all playing instruments of solid gold.

1996 edition
Johannes Wildner, New Philharmonia Orchestra of Westphalia
Naxos 8.555933-34 (2 CDs). TT: 82:47. Finale: 23:28

Now incorporating themes from Symphony 8 as well, this is the first edition of the Samale-Mazzuca-Phillips-Cohrs Finale that sounds entirely like Bruckner from beginning to end, with many of the new linking passages subtly revised to be brought more fully within the composer's idiom. Wildner's hand on the orchestral tiller is very sure, the orchestra is disciplined, the sound very natural. Wildner has great control of orchestral ensemble and dynamics. Don't let the brevity of this capsule review dissuade you: This is a great first choice of completed version, following a full digestion of Harnoncourt's presentation of the fragments alone. But see the 2008 edition, below.

2005 edition
Marcus Bosch, Sinfonieorchester Aachen
Coviello Classics COV 30711 (SACD/CD). TT: 69:54. Finale: 20:19

The Samale-Mazzuca-Phillips-Cohrs completion is further improved, especially in the coda, as well as by the inspired insertion of a horn descant over the first, piano reappearance of the chorale following the fugue and "sunrise" fanfare. The further development and "stacking" of themes from the first three movements just before the coda proper has also been refined, each theme now more clearly voiced. However, while Marcus Bosch makes his brisk pace work well in the Finale, I found his tempos in the first three movements simply too fast for the music to register in any way that could be actually felt. And while the high-resolution sound is rich and lush, the acoustic of the sizable sanctuary in which this concert performance was too distantly miked, in combination with tempos too fast for the space as well as the music, result in a muddy, swimmy sound in which entire choirs of trombones or Wagner tubas are buried, virtually unheard under mounting layers of reverberation. Bosch is now more than halfway through a complete Bruckner cycle, but on the evidence of this recording seems to have oddly little feel for or sympathy with the composer's music.

2008 edition
Friedemann Layer, Musikalische Akademie des Nationaltheater-Orchesters Mannheim e.V.
Deutschlandradio Kultur (2 CDs). TT: 83:24. Finale: 25:29 (available from

What lifts this most recent edition of the Samale-Mazzuca-Phillips-Cohrs completion into another league entirely from earlier editions and all other completions is the addition of a final restatement of the chorale, followed by an extended development in the brass of the leaping theme from the Finale's first few bars, and a considerably expanded final peroration, again in the full brass. And in the recapitulation just before the coda proper, Layer's combination of dynamic balance and slow tempo brings to the fore the quotation of a theme from Symphony 8—a chromatically descending four-note passage for solo oboe, repeated in various modulations—in a way that balances the mounting excitement of the prefiguring of the coda to come with the certainty of its fulfillment. Here, for the first time, Bruckner's aural edifice sounds fully constructed as he might have intended, all scaffolding removed, and on a scale in proportion with the rest of this astonishing work. Also for the first time, I can imagine someone familiar with all of Bruckner's other symphonies except the Ninth hearing this set, listening to all four movements, and never once thinking that Bruckner didn't actually write every note.

Conductor Friedemann Layer is fully up to the task, even if the Mannheim orchestra is a bit rough in spots. His articulation of the fugue is terrific—the stretto is crystal clear. Here, apparently, Layer gave the violins detailed bowing instructions that bring out references to the leaping theme from the movement's first bars; hitherto this has seemed little more than a supporting figure, but it now sounds as important as the fugue's main subject, and adds another dimension to a fugue already overwhelming in its complexity.

In short, this recording, the most recent of any completion of the Finale, is also the most satisfying by far. It is the only completion with which, after listening to it, I have not found myself looking forward to the next edition's incremental improvements.

Bruckner's Ninth Symphony may never be finished—indeed, Benjamin-Gunnar Cohrs has just informed me that he and his colleagues are considering a new revision of the coda. But here, in the four movements presented in Layer's recording, it sounds at last complete.—Richard Lehnert

foscari's picture

Dear Richard Lehnert-First of all let me tell you that I have had no musical education. The fact that I love Bruckners music is that even though I am Jewish Atheist I have a lot of empathy for Bruckner and it one of the main reasons I cannot take even the latest finale of the 9th, seriously as the ending of the symphony.
I am just not sure how ill Bruckner was when trying to compose this finale but I expect that it was not just very serious physical illness but mental illness as well.
Thankfully, in my opinion he saw that he could not finish the work, even if his doctor Richard Heller said that he had played the coda to him, and finished his third movement with his "farewell to life" and this is my main point. I do not believe that the fourth move would have begun with the dissonance and fragmented music we hear after the way he finished the third movement. I do believe that the third movement would of finished in dissonance similar to ending of the first movement , left completely stark like the hellish dissonant utterances before the "farewell to life".
I just dont dont think through his health he was capable of completing , in my opinion, after the three greatest movements in the Symphonic repertory. You may say "what about the chorale"? Yes but the chorale in itself is knowhere enough . Alot of what was Bruckner reminds of music going back to remnants of the "Nulte". I happen to very much like the nulte, but this is a different world to this 20th century Bruckner.We are hearing late Mahler, Schostakovitch, Prokoviev and even early Schoenberg in this style. It is unique for the 19th century. Then we get this throwback.The Chorale is brilliantly arranged from the wide open chords we hear towards the end of the third, which has great similarity to the opening of the Tallis Fantasia by Vaughn Williams. I suspect Bruckner was able to compose this chorale during some kind of remission.
I am happy to hear these reconstructed finales as a very good curiosity. Perhaps Hans Zimmer was even inspired to compose his minimilistic ending to The Da Vinci Code after hearing this minimilistic theme after the reintroduction of the Chorale towards the end although I suspect that Carragan actualy "arranged" this.
However for me, I will take the three movements as Bruckners last "words".
You see Mr Lehnert as I have no musical education I can make up a conspiracy theory that we do hear something like Bruckners intentions for his coda to his beloved " Lord". Maybe , just maybe, Gustav Mahler paid his professeur a "late" visit and we hear this coda in The Resurrection !

Kind Regards From Barry Bernstein

sjlevine34's picture

I disagree with Mr. Bernstein.  Also being a Jew and an agnostic (leaning toward theistic), I have found the Finale of the 9th quite captivating and convincing.  While the movement's structure appears unusually tight for the composer, listening to the fragments as performed by Harnoncourt convinces one that it is what Bruckner intended.   And the music, I feel, is the composer at his finest.

dalethorn's picture

It seems that the data is mostly there, but the interpretation is in dispute. Not much different from Bible text then.

dieter's picture

Hi Guys
I'm not sure what being Jewish has to do with Bruckner: I'm a Goy, I love Mahler, so what.
I feel haunted by the reconstructed last movement of the 9th: it's awesome music, so Bruckner. I confess that Bruckner is up there with Bach and Beethoven in my Pantheon.

Terril's picture

I'm half jewish :) And a Bruckner lover for many decades. I found that it was quite difficult to fully "Grok" this finale for some time even though I can usually "grok" any music almost instantaneously. For example any oriental music some of which I learnt to play. My musical meducation is though quite limited. The version by Friedeman Layer of the finale I find absolutely inspiring and I consider it the greatest movement from a bruckner symphony. I far prefer it to the Berlin under Rattle, whether thats because of the altered reconstruction or Rattles conception I don't know.