Vienna Philharmonic Orchestra (Harnoncourt) – ‘Bruckner: Symphony No.9’ An SACD review by Mark Jordan

Is it possible to cheat death? As far as Bruckner’s ‘Ninth’ Symphony goes, we’re getting close. Anton Bruckner died in 1896, apparently leaving his ‘Symphony No. 9 in D minor’ unfinished. Or did he? The assumption that the finale was too sketchy for performance has led to a long-standing tradition of performing the first three movements as a self-contained entity, even though that clearly violates Bruckner’s original plans. As the end was coming, Bruckner said that if he didn’t live to finish it, perhaps his choral ‘Te Deum’ could be used in its place. Unfortunately that piece dates from much earlier in his career, and the stylistic problems make it an ineffective conclusion. Not only will audiophiles be interested in hearing this, RCA Red Seal’s first SACD; all Brucknerians will want to hear it for the revelations of Nikolaus Harnoncourt and the Vienna Philharmonic, who have teamed up with several Bruckner scholars to remind us that the question of the finale still hangs in the air, demanding some sort of answer. But it also begs an uncomfortable question. Which is worse: To sift through a composer’s unfinished sketches trying to extract a meaningful and coherent conclusion to a complex symphony, or to willfully misrepresent the work by leaving out a crucial part of its artistic conception?

This is, in fact, the dilemma that several famous unfinished works have left for posterity – Schubert’s ‘Unfinished Symphony’, Mahler’s ‘Tenth’, Berg’s ‘Lulu’. Completions have been made of all these, with varying degrees of success. But many have gone so far as to say that Bruckner’s ‘Ninth’ feels sufficiently complete in three movements, and that no finale is needed. I’ll be tilting at some sacred cows here, but that attitude strikes me as an astonishingly arrogant thing to say in the face of music of genius! It is the contention of musicologist Benjamin-Gunnar Cohrs that the mere exclusion of the sketchy finale to this work distorts Bruckner’s musical conception to the point that it no longer stands as representative of Bruckner. While not settling the issue, this “workshop” presentation of the finale fragments, along with a complete performance of the three finished movements, allows listeners to decide for themselves. The workshop is a presentation of the fragments (edited by John A. Phillips) by the full orchestra in a concert setting with spoken commentary by Harnoncourt (in both German and English versions). The regular movements also come to us as the first recording of the new critical edition of the score, edited by Benjamin-Gunnar Cohrs and published by the Complete Bruckner Edition.

First of all, let us examine Harnoncourt’s handling of the standard three movements of the symphony. As early as the opening moments of the third-movement ‘Adagio’, it is evident that Harnoncourt is taking into account the fact that if the finale stood in full finished form, it would alter the balance of gravity in the work. The usual performance of the three-movement version features a weighty first movement and an even weightier ‘Adagio’, with the feisty ‘Scherzo’ acting as keystone between the two epics. But should the slow movement be so epic? In living memory we’ve seen performances of the ‘Adagio’ running longer than the colossal first movement, such as the one by Carlo Maria Giulini (also with the VPO) on Deutsche Grammophon, which lasts over half an hour. But Harnoncourt avoids giving the movement extra weight, letting it run its questioning course in less then twenty-four minutes. It clearly is an approach that demonstrates that when played straight, the work does in fact feel unfinished. After the stern, larger-than-life perils of the first movement and the demonic terrors of the second, the ‘Adagio’ is no resolution at all. It merely attempts to come to terms with these tensions, leading to a huge dissonant outburst that rivals the one in the first movement of Mahler’s unfinished ‘Tenth’. Harnoncourt pulls no punches, nor does he attempt to gloss over this raw, vicious chord. Most conductors try to handle it so that it seems a release, but I’m convinced that Harnoncourt is revealing the composer’s vision accurately. Rather than resolving or romanticizing that death-knell, Bruckner is defeated by it – no wonder he saw the ‘Adagio’ as being his own death-song. Instead of conquering it or coming to terms with it, the movement is defeated by it, and it retreats into prayers that float up to the heavens. Thus we are left sitting on the edge of our chairs, waiting for the great epic battle to come in the finale… And still we wait…

In terms of comparison, then, this performance really isn’t comparable to any previous version. It is instead a thesis about the meaning of Bruckner’s final work of art. It can be accepted or rejected, but comparison is not especially useful. Suffice it to say that Harnoncourt responds to both the lyrical and the fiery sides of Bruckner’s artistic personality. Though I have never heard it discussed before, I suspect that Harnoncourt’s early years as an orchestral player in Vienna have come heavily into play ever since he widened his range from the period instrument movement and started conducting “modern” orchestras. When he played in the Vienna Symphony in the 1950’s, one of the conductors he worked with many times was the great Jascha Horenstein, and I think traces of Horenstein’s influence can be discerned in both Harnoncourt’s taste for intensity of sound, and in his sharply etched phrasing. Nowhere has this relationship seemed closer than in Harnoncourt’s performances of Bruckner. Consult Horenstein’s live 1970 ‘Ninth’ recently issued by the BBC for comparison to this release.

Only three other recordings have been made that include elements of the unfinished finale. The first one was by Eliahu Inbal and the Frankfurt Radio Symphony on Teldec in 1984. Sort of, anyway. The problem is that Inbal recorded a standard three-movement performance which was released by itself, and then he recorded the reconstruction of the sketches for the finale by Samale and Mazzuca and appended it to his recording of Bruckner’s ‘Fifth’ Thus, clearly, there is no conceptual re-thinking of the work and the performance of the sketches is quite tentative – not that the fault for that can be laid exclusively at the conductor’s feet. In their initial attempt, Samale and Mazzuca did not effectively weld together all the scattered parts of the finale and the stop-and-start coda just seems to fall apart, failing to make the transition that is needed for the movement to satisfyingly answer the despair and questioning of the rest of the work.

Around the same time, Yoav Talmi and the Oslo Philharmonic recorded a full rendition of the work with the finale as finished by William Carragan. It is often a more flowing affair than the Samale-Mazzuca, but it errs in the placement of some sketches, and in other places, seems to disregard them completely in blind attempts to fill in lost transitions with original material. Both versions make the same mistake of underplaying the “trumpet hymn” passage just before the coda, which Harnoncourt convincingly argues in his lecture was most likely Bruckner’s short-hand for a passage that should be fully-scored and triumphant, giving a plateau from which to launch the coda. The Talmi recording was made by Chandos and is currently available in the U.S. from the Musical Heritage Society, but I don’t particularly recommend it as it also feels tentative, with both orchestra and conductor remaining unsure about how to shape the finale, and again, it is just tacked on to a standard rendition of the first three movements.

More recently, Bruckner scholars John A. Phillips and Benjamin-Gunnar Cohrs added their insights and expertise to the Samale/Mazzuca project, including some newly discovered sketches. By fleshing out the existing sketches and reconstructing the implied transitions, they have provided us with a twenty-three minute finale that contains very little that is not directly derived from Bruckner’s sketches, filled in the gaps and improved the flow from the earlier attempts at finishing the movement. In the recording on Naxos, Johannes Wildner (a former Vienna Philharmonic violinist) leads the New Philharmonic Orchestra of Westphalia through a performance of the movement that comes very, very close to convincing me. But a question remains: Exactly how much of it is genuine Bruckner?

Judging by Harnoncourt’s performance of the sketches, what Bruckner left us is 80-85% finished. Harnoncourt’s sketches amount to about eighteen minutes of music, not including the fragments of the coda that have turned up, which Harnoncourt declined to perform until their shape and purpose becomes clearer. Indeed, the linchpin of the finale is the coda, as Bruckner seems to have been structuring this movement along the lines of the finale of his ‘Eighth’, where victory is not assured until the final minutes of the work. The structure of the movement leading up to that point seems fairly clear: A nervous, jumpy introduction brings back the unsettled mood of the earlier movements. After a typically thrilling Brucknerian build-up, the first theme of the movement proves to be a striding, brusque utterance, quite an effective foil to the grim main theme of the first movement. After the striding theme settles down, it leads into a warm, lyrical second theme that recalls the longing of the ‘Adagio’. Then comes a hymn-like theme in octaves, which gets derailed by the daring dissonances that typify the whole work. There follows a development of each theme in turn. The development of the first theme turns into a wild fugue that seems to be headed to crisis – when we are confronted with a missing gap. Harnoncourt points out that all the instrumentated pages are numbered, which means that in the places where there are now gaps, pages once existed. Harnoncourt pleads in his lecture for listeners to go in search of these pages. They may have been taken by friends or autograph hunters as mementos and may well be tucked away in document collections somewhere in the world right now. When the music resumes, the fugue is at its stern peak, wrestling the striding theme into some sort of control. The development of the second theme explores more lyrical territory, tying that into everything else going on in the movement and introducing a noble new theme in the horns, which seems destined to be an important player in the eventual triumph of the movement. Then here again, we are hit with another crucial gap of a page. The following third development seems sketchier, not completely scored, and if Harnoncourt is right, it should build to a full, roaring chorus of sound, which is then halted by a series of fanfares leading to… What? The coda, of course. But it is here that the sketches break down into the fragments that Harnoncourt declines to perform. This is the point of transformation, which Bruckner was at when he died.

Was it truly not finished? According to Bruckner biographer Werner Wolff, Bruckner’s doctor commented that shortly before his death, Bruckner played the end of the finale for him at the piano, his trembling hands turning mighty and strong as he moved into the grand piling-up of themes in the closing moments, closing with a song of praise in D major. Composer and former Bruckner student Hugo Wolf stopped by to visit the master near the end and saw him staring off into space, smiling as he conducted with one feeble hand the music he could hear in his imagination. These testimonies seem to indicate that the piece was finished and that Bruckner was merely going through the task of writing down what he had already successfully conceived in his head. Is it possible that all the missing sketches still exist, but were lost or suppressed by friends who felt that Bruckner tried to push things too far with this feisty, eccentric finale? Who knows? But since there remains a chance, however slim, that we may yet be able to assemble a finished version of this great unfinished masterpiece, I will echo the pleas of Harnoncourt and Cohrs for everyone to spread the word to collectors to be on the lookout for these priceless pages. The collectors can keep the pages; we just want to hear the music on them! Benjamin Gunnar-Cohrs can be contacted at Postfach 10 75 07, D-28075 Bremen, Deutschland.

With such matters at stake, the value of this release goes far beyond any regular considerations of performance and/or sound. But that is not to understate the power of Harnoncourt’s performance. He truly has the measure of these mountains of sound and has the wizardry to bring them to life. He lets the Vienna Philharmonic strings sing in the places where longing is needed, but Harnoncourt doesn’t even come close to ladling on the schmaltz that most conductors heap on the lyrical parts of this work and his handling of the devastating crags of the first movement rival or surpass any other account. In the ‘Scherzo’, Harnoncourt lights into the pounding fury without any apparent hesitation or reservation, yet still he keeps the intricate writing layered and shaped. The fast, feverish trio is especially effective in these hands, as he gets the winds to play with a wickedly seductive playfulness that seems much truer to the fiery spirit of Bruckner than the usual poker-faced manner. What’s more, Harnoncourt pulls off the unlikely trick of making the second part of the trio theme seem to sink down into listlessness without actually slowing down his tempo, a rare feat indeed. Of a dozen recordings I reviewed for comparison, the other conductors either kept a tempo, making no attempt to capture the sinking feeling, or else they slowed down to a crawl and let the energy sag. The ‘Adagio’ is clearly going to be a different experience with Harnoncourt from the very opening phrase in the violins. Instead of grabbing your soul by the throat as Giulini does, Harnoncourt keeps the dynamics subdued so that the gesture is more introspective, wearier than passionate – more late Bruckner than late Mahler. The trumpets lead the following blaze of sound a couple of minutes into the movement, seeming to represent the elusive triumph toward which the symphony keeps fighting its way. And the fury Harnoncourt unleashes in the dissonant chord before the retreat into the tender closing pages of the ‘Adagio’ is unforgettable, a moment of crisis to be forever blazed into this listener’s memory and caught with great visceral impact by the recording. But Harnoncourt is not necessarily perfect, the overall shape of the first and third movements does not seem as balanced as in the performance by Gьnter Wand. Harnoncourt gets more caught up in the details than Wand does, so he never quite achieves the magisterial flow of Wand’s Bruckner, and, having finally heard Harnoncourt in concert with the Vienna Philharmonic last winter, I can attest to the fact that the at times aggressive sound of this recording is not an anomaly created by the recording team – at least not completely. Harnoncourt tends to play the extremes of soft and loud, slow and fast, glacial and fiery. When the fire is in the music, Harnoncourt in concert can unleash orchestral attacks that make you jump right out of your skin, and this recording is no exception.

Interestingly, fans that have followed Harnoncourt on Teldec over the years will have no problem continuing with RCA Red Seal. The soundscape of the CD layer of this recording is very comparable to the Teldec discs, and for a very good reason: When Time-Warner absorbed the Teldec label and discontinued it as a separate entity, the Teldec engineers formed their own production company called Teldex Studios, and it is they who have been hired by RCA to make Harnoncourt’s recordings. Thus, the recording style is the same, albeit with a shift in allegiance from DVD-Audio format to SACD. Whatever the case, fans of previous Harnoncourt outings will recognize the somewhat distant, aggressively compact sound on display here. The focused “punch” of orchestral sound is very true to the tone Harnoncourt cultivates with the VPO, although I would hasten to add that the somewhat distanced perspective robs a little of the orchestral glow that this group makes live in concert. In the more lyrical, less heavily scored passages, the strings sing through with astonishing beauty, but more vivid in immediate memory is the blaze of trumpets capping the loud parts – thrilling, yes, but bordering on hard-pressed.

As noted above, the CD layer of this hybrid disc inhabits very much the same sound-world as Harnoncourt’s other Bruckner recordings. Moving to the stereo mix on the SACD layer of this DSD recording, however, it’s as if there is suddenly electricity in the air, a palpable presence. Here the perspective doesn’t bother me quite as much, because the high-resolution recording picks up more details of the scoring, as well as the crucial sound-color of the instruments (crucial because this is not a flashy work – most of Bruckner’s instrumentation is on the dark side in the ‘Ninth’). But moving into the 5.1 multichannel mix makes even the stereo SACD mix seem limited. Suddenly, you are sitting about halfway back in Vienna’s Musikvereinsaal, perhaps even a little farther, listening to a great orchestra playing under a distinctive conductor. You can feel the presence of the audience there, but they remain mostly quiet. For my taste, I would rather be closer to the stage, getting a wider orchestral picture. Here, with the distant miking, we get the bulk of the orchestral picture thrown to the center channel, tending to make the orchestral sound rather monolithic and forbidding. It is arguable that this is a valid perspective for this imposing work, but I feel the distance leaves the listener outside the envelope, confronting a wall of sound. Even so, the clarifications of the conductor and the technology make it clearer than most recordings of the piece. Throughout much of the first movement, I wondered if there was not enough back-wall reflection coming from the rear channels, but the second movement made it clear that there actually was. It’s just hard to isolate it in the first movement because of the way Harnoncourt has the players attenuate their phrase-endings in so many places, fading off into the room sound instead of cutting off abruptly. The ‘Scherzo’ could hardly be more of a contrast: The notes are stabbing staccatos, crisply carrying through the hall’s warmth, revealing a very natural bounce from the back. The trio of the ‘Scherzo’ becomes some of the most beguiling music Bruckner ever wrote, fleet and elusive, shimmering and vanishing, and the recording captures it magically, especially the breathy flute and parchy timpani. The ‘Adagio’ brings a return of the more lyrical playing style, blending in with the surround channels. In sum, a good recording, though live recording in the Musikvereinsaal remains tricky. One hopes the continued application of surround-sound to RCA’s Harnoncourt recordings will convince the Teldex engineers to move in and spread the sound-image a little wider. It would help draw the listener into the music.

For some reason, the workshop on the finale is only presented in regular stereo 44.1kHz 16-bit on a separate disc, but what is important is the intellectual point being made here: The work is incomplete as it stands in three movements, and distorting it does not make it finished. The most recent reconstructions make it clear that we are tantalizingly close to resurrecting the “lost” crown of Bruckner’s symphonic career, and I urge all Brucknerians to hear this disc, as it may well help the process along. Let us hope it will lead to the discovery of the missing sketches uniting the finale without the need for any conjectural contributions – which would also lead, no doubt, to a recording of the complete finale by Harnoncourt. At last, then – over a hundred years after the composer’s death – we would finally have Bruckner’s masterpiece.