Anton Bruckner's 9th symphony

completion of the Finale

The first time I heard Bruckner's unfinished Finale, I felt not only all the greatness but also the exceptional and huge potential that arises from the score despite its truncated state.

For his ultimate symphony, Bruckner planned an ambitious and masterly Finale whose concluding apotheosis would very likely be a summation of his symphonic work. Although some interpreters consider the adagio (the third movement) to be a satisfactory conclusion to the symphony, there is not the slightest doubt that until his last days, Bruckner intended his Ninth Symphony to have not three but four movements.

Nikolaus Harnoncourt made a complete recording of the work in August 2002 with the Wiener Philharmoniker, presenting a « workshop » under the name of « Dokumentation des Fragments » published by the Musikwissenschaftlicher Verlag Wien.

This recording was for me the very first significant "catch" with the Finale. Harnoncourt's sharping and involved interpretation, the massive presence of the Vienna Philharmonic's qualities were indeed evidence of something. The second "catch" happened in 2007, with the encouragements of Nicolas Couton, a young French conductor, when we listened to the creation of the new last version of the Finale prepared by Benjamin Gunnar Cohrs and Nicola Samale, creation that took place in Aachen (Germany) and was played by the symphony orchestra of the city under the baton of Marcus Bosch.

The different realizations I had heard until then, including this Samale/Cohrs completion, always left me quite puzzled and deeply unsatisfied. Between 2006 and 2012, some minor improvements were evidently added to the earlier 1992/1996 Samale-Phillips-Cohrs-Mazzuca version. However, I have remained disappointed by the 'minimalistic' approach regarding the coda, both in terms of shape and content. As a composer, and based on my understanding of Bruckner's sonata form — particularly in this ultimate Finale — I have always believed that the coda required a revitalizing and conclusive build-up of tension, on a scale comparable to the Fifth Symphony. In my opinion, this level of magnitude was missing from all existing completions (including SPCM, William Carragan's versions 1 and 2, Nors S. Josephson, and the rather peculiar 'free-style' realization by P.J. Marthé). With regard to the SPCM completion, major problems — encompassing numerous serious aesthetic, technical, and intellectual issues — were obviously never addressed or resolved and may have even worsened in the subsequent 2021/2022 Phillips revision : paper 1 ; paper 2.

I did not find anything substantially new or convincing in the later completions by Gerd Schaller, Martin Bernhard, or Roberto Ferazza. Moreover, with regard to the entirety of the remaining available material from the composer's manuscript, these works also notably lack rigor and sufficiently deep analysis of the sources. The late William Carragan was one of the rare individuals who were simultaneously well-informed, transparent, and honest in his work on completing the Finale — particularly in distinguishing between what we objectively know and what remains unknown.

From May to September 2007, I worked on a first version or 'draft' which shortly after I modified, by changing number of orchestral details, by having a shorter recapitulation of the choral in the coda (28 bars instead of 36) and by inserting in the « coda of the coda » ; the « Song of praise to the Dear Lord » (or « Hallelujah ») theme taken from the Scherzo's Trio :

The score was completed in August 2008, with slight adjustments made in November/December of the same year following the premiere recording of my version with the MAV Symphony Orchestra Budapest in October. Later, some additional orchestral and contrapuntal details were improved, but the overall structures of the completed parts remained fundamentally unchanged.

Every realisation of Bruckner's Ninth Symphony Finale obviously cannot be compared to what Bruckner himself would have achieved, especially a Bruckner in full possession of his intellectual powers. Indeed, we know that his health declined in the last two years of his life and that the slow and laborious work on the Finale caused him difficulties.

Any ideal that aims for the most scrupulous respect, or at least claims to do so on strong musicological grounds, risks being too rigid and/or oversimplified, especially when completing a work of such high stature. As far as I was concerned, I wanted to avoid any obtusely rigorous quirkiness and strike a balance between scrupulous respect for the manuscript sources, necessary speculative composition, and carefully considered invention. The problem is particularly crucial regarding the missing coda, for which we objectively have very few elements. Its content and design remain completely unknown.

For instance, the ÖNB 3194/3r sketch reveals neat handwriting from the composer, unlike most of the last documents of Bruckner's hand during the final months of his life, which show signs of suffering and shakiness. This ÖNB 3194/3r sketch, likely earlier to the composition phase of the coda, is attributed — perhaps wrongly? — to the ending of the Finale because of its 'C' time signature (4/4), implying a slower, maybe 'conclusive' tempo compared to the usual 'alla breve' in 2.

However, it begins with an E-flat major chord (in the 6th position), which is not the main key — D minor — of the symphony. Starting the coda in the home key of the symphony has always been a persistent characteristic of Bruckner's Finales. Transposing these 24 bars a fourth interval lower (without any evidence to support it, cf. the SPCM completion) so that the first bass note is 'D' does not resolve the issue, as the first chord of the coda then becomes B-flat major in 6th position rather than D minor...

There exists another legitimate hypothesis that gives more weight to the obvious observation that ÖNB 3194/3r could very likely be an earlier sketch. Was it a first version of the transition leading to the presentation of the Chorale in the exposition? The 'C' (4/4) time signature might suggest that it was initially designed to unfold after the 'Trio section' of the B 'Gesangsperiode' group, seemingly implying an extensive accelerando.

The main reason for using the ÖNB 3194/3r sketch for the coda is that it is the only one that is substantial and thematically defined, notably with a new long motive suggesting a link to the main theme of the 8th Symphony. I utilized this motive to integrate a full quotation of this theme right before the final appearance of the Chorale.

Sketch ÖNB 3194/3r

The other remaining sketches identified for the coda are barely determined or not even characterized, consisting mostly of empty bars — some of them simply scratched — with very limited notations.

Sketch ÖNB 6085/45r

Sometimes, certain notes or some basso continuo markings only allow guesses of the chords (e.g., sketch ÖNB 6085/47r) :

Sketch ÖNB 6085/43r

Sketch ÖNB 6085/47r

Given the considerable uncertainty — and possibly permanently lost material — regarding the actual content of the coda, this naturally invites speculation and thus creative compositional invention.

At the end of the recapitulation, as I already explained, I have always felt that this Finale required a sort of second development phase, similar to the Finale of the 5th Symphony, where all the accumulated tensions need to be resolved.

The main purpose was to attain the best *inspiration* as possible ; since this music essentially has a transcendent range. This inspiration has to be found as possible from an aesthetic intimacy with the composer's compositional technique which more than obviously must guide and direct the work's every step. I do not claim however to have satisfied these conditions in an absolute way. Nonetheless I hope that anyone who listens to my completion will realise in it the devotion and the love for this incomparably unique and great symphonist that Bruckner was.

Finally, I would like to express my infinite gratitude to Nicolas Couton and Lionel Tacchini for their help and valuable friendship.

-------------------------------

Bruckner's manuscripts, drafts and bifolios available in the « Finale Faksimile Ausgabe » edited by Leopold Nowak for the Musikwissenschaflicher Verlag Wien (MVW) were used as the main source for this new completion of the Finale to Bruckner's Ninth Symphony.

As previously mentioned, other documents, thesis, and realizations have also been studied. While they provide a significant amount of information, they sometimes appear highly questionable both in terms of speculative assumptions and the quality of musical realization. Nonetheless, my own reflections have always drawn from critically analyzing these other sources.

The video below displays most of the significant material that has survived from Bruckner's hand, completely debunking the myth that there was almost nothing substantial or coherent left from the old composer. On the contrary, we can see how far Bruckner had progressed in composing his ultimate opus despite his severe illness.

As part of the 'Bruckner Casco Festival', which took place from September 13 to 15, 2024, at the Muziekgebouw in Amsterdam, the closing concert featured Anton Bruckner's 9th Symphony. Camerata RCO, composed entirely of members from the prestigious Concertgebouw Orchestra of Amsterdam, gave the world premiere of this symphony in its four-movement form, concluding with the Finale, which was completed by Sébastien Letocart in 2007/2008 and rearranged for 16 musicians in 2022/2023.

'Anton Bruckner's ultimate Finale' - a documentary realized in Linz, Ansfelden, Saint Florian and Vienna in July 2013

American music critic David Hurwitz considers Bruckner's ultimate Finale to be extremely weak music compared to the three previous movements of the 9th Symphony, and views any attempt at completion by 'these Bruckner people comparable to Löwe and the Schalk brothers' (...) as superfluous and even disastrous... Hurwitz's judgments, especially regarding Bruckner, are often quite superficial, if not bordering on simplistic misinformation.

Notes to my Completion of the Finale to Anton Bruckner’s Ninth Symphony

I want to make it quite clear that my completion of the Finale of Bruckner’s Ninth Symphony is based strictly on Bruckner’s own material. I have orchestrated my realization as faithfully and discretely as possible. There are two main aspects to understanding the process of this completion:

Firstly, besides having to fill in some of the orchestration of the existing parts, there are six gaps in the development/recapitulation that have to be speculatively reconstructed by the recreation of coherent links. The gaps are located (the timings refer to this recording)

i. at the transition between exposition and development (pedal tone on E),

ii. in the middle of the first part of the development,

iii. at the end of the fugue (stretto),

iv. in the recapitulation at the transition to the third group,

v. in the middle of the third group (choral theme played by the oboe) and

vi. at the tense transition to the coda.

Secondly, my elaboration of the coda, however, shares neither the same task nor the same concern about the question “what would Bruckner have done” because it is quite simply impossible to know or to guess. We only have a few sketches of and some vague testimonies (Heller, Auer and Graf) about the Finale’s continuation, and we know nothing even about the precise number of bars, so these sources hardly give any idea of the global structure Bruckner had in mind. Nevertheless, I felt that this part of the Finale had to be as important as those in the Finales of the Fifth and Eighth Symphonies. My extrapolated coda in four parts (of 36, 28, 36 and 59 bars respectively), although only allowing itself to use the thematic contents and motives from the Finale itself, in my construction is partly inspired by the codas of the Fifth and Eighth Symphonies.

The coda begins (first part) with a long crescendo based on Bruckner’s 24-bar 'ÖNB 3194/3r' sketch which is built on a tritonic progression and which is linked thematically to the very beginning of the movement. I prolong this to 36 bars by adding another 12 bars of my own that culminate with the quotation of the main theme from the Eighth Symphony.

The second part begins with the last appearance of the chorale, i.e. the third group’s main theme, which is treated in the same manner as the Fifth Symphony with integrated quotations of the main themes of the opening movements of the Fifth and Seventh Symphonies.

The third part begins with a four-bar sketch (ÖNB 6085/45r) of Bruckner that recalls the coda (violins and brasses) of the Finale of the Fourth Symphony. It corresponds to the inversion of the descending brass line that appears just after the initial presentation of the main theme. The continuation is based on the saltus duriusculus (insistent repetitions of descending sixths and sevenths) of the Finale’s beginning. It is followed by the same rhythmic and thematic progression as in the exposition of the first thematic group but elaborated from one of Bruckner’s last sketches dated May 1896. It consists of a harmonic outline and a metrical structure of 16 bars. It is combined with the “heroic” motive (trumpets) that was heard for the first time at the end of the development (horns). The third part brutally ends on a climactic dissonant chord (parrhesia abruptio).

The fourth part, the “coda of the coda”, builds on a long and static D pedal, as Bruckner probably intended, which is a sort of mysterious and ethereal remembrance of the first movement’s coda, being a long crescendo based on what I identify as the “Halleluja theme” (2 Horns and 2 Wagner Tenor Tubas and then after 2 trumpets). As Richard Heller, Bruckner’s doctor, testified, a majestic “Alleluia” was to conclude the Ninth Symphony. Indeed Bruckner explained to Heller that the Finale had to end with “a song of praise dedicated to the dear Lord” based on a theme from the second movement. At the time Bruckner played passages of the finale on the piano to Heller however, was the order of the inner movements ‘scherzo-adagio’ or ‘adagio-scherzo’? We have neither a clear nor certain answer. In my opinion, this “Halleluja” theme seems to find its origin in the trio of the Scherzo (violins 1, bar 53 letter B / idem, bar 205 letter H – cellos-bassoon 1, bar 113 letter D) and not in the Adagio.

Finally, the coda culminates on the same violent 'Phrygian' E flat dissonance as in the first movement and leads to a coagmentatio in D major of the four main themes of the work (fourfold piling-up at 24’04”), crowning the whole work exactly as in the Eighth Symphony. Note also that some small details have been rewritten or modified in the score of the whole movement after this first recording was made.

Concerning this coagmentatio, Max Graf wrote after consulting some hypothetically lost manuscripts in possession of Franz Schalk that there was "a 'Haupthema' (whether the first or the fourth movement's is unclear but most probably the first), a 'Fugenthema' (certainly the Finale's Fugue), a 'Choral' (also certainly the Finale's) and the 'Quintenthema' of the Te Deum and once these four themes are even combined all together (übereinandergestellt), there is a quadruple superposition (eine vierfache Thürmung) as we find at the end of the Eighth Symphony." Max Graf was a music critic and Max Auer, who also mentioned the same combination (probably repeating what he read from Graf), was a dilettante only. One can reasonably say that both these "amateurs" should not be considered as critical or musicological researchers. Furthermore, the idea of a coagmentatio of the four main themes of the symphony has now been discarded by the Australian musicologist, John Alan Phillips who considers Auer's and Graf's writings about this subject not very relevant. [1]

However, having no satisfaying alternative, the idea of a 'Halleluja' theme culminating with a 'coagmentatio' (superposition of the four main themes of the symphony) remains for me the most structurally convincing and the most musically effective option. Two other combinations that use the Te Deum motive, the chorale, the first movement's main theme and the fugue theme from the Finale as described by Graf/Auer have been easily realized but have been discarded because they are musically unsatisfaying :

Regarding the origin of the 'Hallelujah' theme, it is interesting to notice that in Psalm 150, a late work by Bruckner (1892), the chorus sings loudly the word 'Hallelujah' in the manner of a chorale (see below), with the upper melody having a very similar shape to that of the Choral in the Finale of the 9th Symphony. Actually, was the Choral theme this 'Hallelujah' mentioned in Doctor Heller's remembrance?

In his thesis [2], John Alan Phillips broaches the subject about the substitution problem for the Finale as planned by Bruckner (who was extremely troubled about the idea of dying) in the middle of the third thematic group (recapitulation) : a sudden transition for the insertion of the Te Deum … Obviously, this solution can not be taken seriously into account because of the question of the proportions and musical coherence: the duration of the Te Deum is approximately 22 minutes. What would such an “intrusion” mean after hearing almost ¾ of the Finale (about 15 to 17 minutes)?

I think we can understand it as a kind of capitulation by the composer through successive periods of strong doubt, after realizing that, due to his limitations, he could not elaborate the coda to a sufficient scope. He may have recognized that he did not have enough time to fully develop his ideas into a satisfactory musical form without a lot of extra work. However, he never gave up on the very idea of finishing his ultimate symphony.

The reasons for the Bruckner's difficulty in finishing his Ninth Symphony were probably a combination of physical and mental disorders. Nonetheless, this music still fascinates, although it is incomplete, as much by its grandeur, its power as by the wrenching enigma of its incompleteness. We acknowledge the same problem with another extraordinary symphony which, left unfinished at the composer’s death and although entirely sketched out, leaves this strange and enigmatic feeling of a process held forever in suspension. Of course, I mean the Tenth Symphony of Mahler.

_________________________

[1] John A. Phillips – Bruckner's Ninth Revisited Thesis, University of Adelaide, 2002; "Source (lost) of a Themenüberlagung" §3.1.10, E/138, II and "The Themenüberlagung" in §3.3.7

[2] John A. Phillips – ibid.; “The Te Deum as Ersatz” §1.1.11 and “The conjectural Te Deum transition” §3.3.8

Page updated

Google Sites

Report abuse