Anton Bruckner's symphonies:

Aspects of performance practice


© Benjamin Gunnar Cohrs, Bremen (2008)

also available in


Intonation, Balance, Size and Seating of the Orchestra

At Bruckner's times, the intonation in the orchestra was quite different - not the muddy adoption of the >compromised< intonation of the modern piano, but largely a pure intonation (fifths and thirds), as it was available in Berlin, due to the influence of Joseph Joachim and his pupils, almost until ca 1940. Only the pure, >naturally harmonic< intonation would bring all the colours of Bruckner's harmony, so full of enharmonics and chromaticism. The fabric of contemporary woodwind instruments tests the case: Before 1900 - as, for instance, preserved in collections in Vienna - they had keyholes bored in a different intonation. For this reason, ourday's contemporary orchestras which still use those old instruments achieve better results in Bruckner (such as the Orchestra of the Age of Enlightenment or the Orchestre des Champs-Élysées), because even modern-trained string players automatically adopt their playing to what they hear from the woodwind and brass. On the opposite, modern woodwind and brass instruments have our typical >compromised< intonation; this, together with some peculiar bad habits of string intonation routine, like sharpening sharps and flatening flats, brings bad results for Bruckner - there should be an audible difference between, let's say, F sharp or G flat major. The conductor should at least be aware of such shortcomings. The knowledge on contemporary instruments may already answer important questions regarding the balance. The brass instruments were at about a third smaller, and at about half as loud as today. The large F-Trumpet, handmade Horns and Trombones of that time produce a much more pungent, full, and rich sound, even if not so horribly powerful like ourday's batteries. Also woodwind instruments were much more colourful those days. With modern, large brass instruments, the conductor should in any case always double woodwind instruments to improve the balance, and care for at least 8 Basses, 8 Celli, and 10 Violas. For Bruckner's four-part-writing of harmony, it is essential to make audible particularly the inner voices, Alto and Tenor. (The outer parts are audible anyway!) The brass should never outpower the woodwind. Not ourday's monstrous Horns and Trombones, of which one only can destroy an entire orchestral balance, but the woodwinds should dominate the wind sound even in the loudest fortissimo! Bruckner referred in his scores to the practice of the Wiener Philharmoniker, including a string orchestra of ca. 12-12-10-8-8. It was common already to double the woodwinds with such a number of strings. Traditionally, in Vienna the two Violin groups were placed on the left and right side of the rostrum, the Double Basses (four-stringed) in one row behind the orchestra. (Note that every concert orchestra of the 19th Century before 1912 disposed the Violins antiphonally!) Celli were sitting left behind the first, Violas right behind the second Violins. However, it might be helpful to reverse this and place the Violas left behind the first Violins, so that they can reflect their sound directly to the audience instead of to the rear wall. Furthermore, the second Violins feel much more comfortable with Celli behind them. In so doing, the Tenor part, so important in Bruckner's voice leading and design, will be better audible, which contributes significantly to the spaciality of sound. Hence it might be possible as well to place the Horns to the left behind the Violas, Trumpets, Trombones and Tuba to the right, behind Celli. This would make the sound of the brass less heavy. In this context it is interesting to learn that already in the old Gewandhaus in Leipzig, the Violas stood (!) behind the second Violins, which, however, stood to the left from the conductor, and the first stood to his right, with Celli sitting behind them. Only after the move into the new Gewandhaus in 1874 this practice was given up, and Violin and Viola players were allowed to sit down in concert.

— Recommended: Daniel J. Koury, Orchestral Performance Practices in the 19th Century: Size, Proportions, and Seating. UMI Research Press, Ann Arbor, Michigan, 1986, ISBN 0-8357-1649-X . Jutta Stüber: Die Intonation des Geigers, Köln 1989.


As explained above, many of ourday's balance problems result in the changes of instrument manufacture during the 20th century. On >period instruments< such problems usually don't occur. Hence, one should generally consider the specific instruments for which Bruckner wrote: He was used to hearing spun gut strings, played with almost no vibrato, but sometimes with portamento, in more distinct intonation and in a >pure tone<: the habit to cover bad intonation with large vibrato only occurred after 1923, following the fatal suggestions by Carl Flesch in his Violin School. The Viennese Flutes had a distinct, wooden, but thin sound. Quite new, from ca 1875, were the Viennese Oboes (but in fact of German origin), with their pear-like mouthpiece, sounding quite similar to a Baroque Oboe d'amore; Clarinets and Bassoons of a particular Viennese fabric were common there until ca. 1900 - note in particular the Viennese Bassoon, sometimes even with a metallic sound cup, and a sharp, reedy sound. The famous Viennese Horns were handmade, more narrow bored, like all other brass instruments of that time, about one third smaller, and produced much less volume than today's brass. The smaller Bass-Tuba and larger Doublebass-Tuba were of a particular Viennese manufacture as well. Trumpets in D and C, but in particular the blazing, double-lengthed F-Trumpet and light, narrow bored Trombones were also common. Bruckner wrote for Trombones in the old-fashioned style, but the two upper parts were played on the same instrument, usually a Tenorbasstrombone in E flat. As Ken Shifrin pointed out in his dissertation, a real Alto Trombone is required only in Bruckner's early works before 1868; however, it is possible to play the f-minor Symphony and the g-minor Ouverture with an Alto Trombone.

— Recommended: Dieter Michael Backes, Die Instrumentation und ihre Entwicklung in Anton Bruckners Symphonien, Dissertation, Studio-Verlag Mainz, 1997; Musikinstrumente und Musizierpraxis zur Zeit Gustav Mahlers, Report of the Symposium of the International Gustav Mahler Society, ed. Reinhold Kubik; Böhlau Verlag Wien, 2007.

Music-Theoretical Background and Style

Only in 2001, the eminent scholar Wolfgang Grandjean published the first comprehensive study of Bruckners music-theoretical background, which is of inevitable importance for the style of performing his symphonies. He showed that Bruckner understood himself (as could have been expected already from his outstanding academic career) as an old-fashioned theorist, who gave living examples of his >scientific< thinking in his compositions. Of main concern to him were metrics, syntax, rhythm and form. Grandjean showed how Bruckner gradually developed a personal >system< of metrical theory in his works, which also largely explains his revisions and >regulations< of earlier works between 1876 and 1878 (N° I, II, III, IV). Most of Bruckners famous >metrical numbers<, which are an integral part of his autographs, were unfortunately not included in the score editions, because they were considered to be of no practical use (with the exception of my new critical edition of IX). However, they are by no means the fruit of >numeromania<, but in fact an acribic instance of controlling the weightening and structure of bar periods, in particular the disposition of heavy and light bars, main and intermediate harmony. The interested conductor may find most of them in the Critical Reports; I strongly recommend to transfer these metrical numbers into the personal conducting scores, because they are indispensable for analyzing the music, its phrasing, tempi, and weightening. By the way - a big misunderstanding of his instrumentation style occured by the assumption that it may be inspired by Organ registration and sound. But in fact, Bruckner did not write Organ symphonies for good reason: only the orchestra would allow him to bring out precisely this weightening, as indicated by his numerous accents. Different as in most romantic and late-romantic music, Bruckner's periods are dominated by the principle of the >emphasised beginning<, in opposition to Riemann's >principle of answering<. Bruckner's music swings in a pendulum of heavy and light bars, organised in superior, or >super-bar units< of at least two bars. Singular heavy or light bars, as in the Riemann system, usually don't occur in Bruckner's music.
The analysis of these macro-structures is very important for the musical interpretation, because the metrical syntax dominates those accents given by Bruckner; very often he does not give accents, even if a heavy bar has to be emphasised - see, for instance, IX/1., b. 167ff: No accents at all, but certainly the first and third bar need to be emphasised, because the music swings in two bars. Or IX/2., first twelve bars: The music swings in paired bars, starting with a rest, emphasising b. 3, 5, 7, 9, 11. Furthermore, the two-bar-pairs are grouped in four-bar super-bars. The metrical numbers are given as 1-2, 1-2, 1-2-3-4-5-6-7-8. This means: similarly strong emphasis of b. 3 and 5, then in the eight-bar period two possibilities: a) first note of b. 5 and 9 each strongest of the four-bar phrase, or b) first note of b. 7 and 11 each strongest, due to the fabric of the phrase, and in particular respecting the first tutti period (b. 43-50) with the resolution into the tonic on the seventh bar of the period (b. 49). Both solutions would make audible the idea of 2, 2, 4(=2x2) + 4(=2x2) bars. However, totally wrong, but most often heard is the following >interpretation< of b. 1-12: Accent on the first quaver of Oboe and Clarinets, but no weight at all on b. 3, 5, 7, and a crescendo to b. 8, accentuating the final note of the phrase (which, in Bruckner's own thinking, is in any case a >weak< bar, a >swing-back-bar< of a bar-pair). On the other hand, take the main theme of IX/1., b. 63ff: Bruckner gives ^-accents on b. 63, 65, 67/1&2, 69 and 70. However, the period structure swings in super-bars of two, organising a harmonical schedule of 4+2+2. Hence, b. 63, 67, and 69 need a heavier emphasis than 65. Bruckner himself underlined this by giving a > in the Timpani, b. 63 and 69.

— Recommended: Wolfgang Grandjean, Metrik und Form bei Bruckner, Hans Schneider, Tutzing, 2001,
ISBN 3-7952-1055-0


Bruckner's use of dynamic indications is much limited, similar as earlier in the works by Mozart or Haydn. In his symphonies he uses ppp, pp, p, mf, f, ff, fff, but not, for instance, mp, or ffff. The fff is limited to the real peaks of movements, to the most important moments. Local climaxes usually receive only a ff. Earlier scholars, for instance Manfred Wagner, found that the earlier symphonies often used terraced dynamics. Even if indeed sometimes terraced dynamics occur (in particular if connected with changes of register or colour), this is due to a misunderstanding and unfortunately also a wrong editorial practice: Since Bruckner added dynamics only in the late phase of nuancing, very often he put cresc. or dim. advice only in some lines of the score (usually under the string, brass, and woodwind system blocks). Even worse, often he marked (in particular before 1878) intermediate dynamics within a cresc. poco a poco, but not behind those marking further cresc. Quite often his crescendo passages, in particular in his early scores, may look like this: pp cresc. poco a poco -mf - f - ff, but in fact with the intention of pp cresc. poco a poco - mf cresc. sempre - f cresc. - ff ! Only rarely he wished a sudden ff or fff. This must be carefully observed by the conductor, and if necessary, supplemented. Even in his later scores, this happens sometime. Another important aspect of dynamics is that normally Bruckner did not write different dynamics for different instruments in a tutti, as for instance, to achieve a better balance, marking strings f, woodwinds ff, but brass only mf. He obviously wished to leave balancing to the descretion of the players and the conductor. Only occasionally he indicated passages with »Solo« or »hervortretend« if he wished a particular instrument to be more prominent. In his earlier scores we even find sometimes verbal instructions, as »stark hervortretend« etc. But later he reduced these kind of instructions very much. On the opposite, his early conductors and pupils edited his scores in a way that just this differenciated dynamics was added, in particular by Schalk and Löwe, who were also conductors and not only editors. Unfortunately, in doing so, they mindered the chances for flexibility which Bruckner himself had offered.

Such a limitation, as given by Bruckner himself, makes sense indeed, because his scores then don't look alltoo sprinkled with indications, and one should not forget that this is very clever indeed if you wish your pieces to be widely performed: Every orchestra of that time had its own traditions, size, seating and instruments, in a much larger variety as in our days. Some would double woodwinds throughout, others not; some would have louder brass, others softer; some a large, some a small string section; some would play in a small, some in a large concert hall. Instead of marking everything down to the tiniest detail, Bruckner decided to give some essentials and trusted in the understanding of the conductor. He could, however, not foresee that most of ourday's jetset-maestri would simply have no time to prepare things so properly and thoroughfully as his music would require .


Bruckner's articulation is largely determined by the metrical structure of his periods and phrases. Again like Mozart or Haydn, he needed amazingly little playing indications and without doubt assumed the conductor would have a certain basic knowledge on the music-theoretical parameters. We should not forget that at his times most of the conductors had a broad musical education and were very often composers themselves. We should also bear in mind that the music-making was most likely much clearer shaped, without those >endless melodies< of our days, today topped with much vibrato-sauce. Basically the sound was much more speech-like in phrasing, as only rediscovered by the historically informed practice in the 1970ies, for instance by Nikolaus Harnoncourt and Roger Norrington. This is clearly proven by the earliest recordings of orchestral music, such as the famous Beethoven V under Nikisch. (For Bruckner conductors of special interest: Beethoven, N° VI and VIII, Vienna Philharmonic, Franz Schalk, 1928, and Bruckner, N° VII, Berlin Philharmonic Orchestra, Jascha Horenstein, 1928.) Of importance for the weightening are basically two kinds of accents: the heavier ^ (= »Keil«), usually indicating a strong attack, then softer sustaining, and separating notes from eachother, often to be found on longer notes, and mostly at the beginning of a bar (unfortunately Bruckner did not use fz or rfz, and only rarely sf), and the lighter > (=»Druck«), emphasising important notes in phrases, and being of a more cantabile, bell-like character of sound. Sometimes, in particular in N° V, Bruckner combines this > with a ' (=»Strich«), to indicate they should be played better separated, but not with the weight of a ^. Furthermore it is incredibly important for the conductor to learn that weightening is not only based on notated accents! Following the findings of Grandjean, the common accents may be seen only as >internal<. On the other hand, it may well be that the beginning of a bar needs a special emphasis apart from notated (or even not given) accents, in particular where Bruckner's music follows the >arsis-thesis< -pendulum, as shown in the beginning.

It seems that Bruckner refined his practice of notating such >nuances< only in the years between 1876 and 1878, when he was much devoted to developing a systematic of metrics, initiated by the composition of N° V. Strangely, after completing a first version of it, he decided to revise all valid earlier symphonies before giving N° V her final shape in 1878. For this reason, those works composed before 1876 (namely symphonies in f minor from 1863, N° I in c minor, 1865-8, in d minor, 1869, N° II in c minor, 1872/3, N° III in d minor, 1873, and N° IV in E flat major, 1874) bear much less articulations and almost no bowing indications. On the other hand, N° V sometimes exaggerates the necessary, perhaps because Bruckner tried to express his ideas as clear as possible, even if sometimes with apparently inconsistent, overdone indications, including also some particular, at first sight perhaps ood looking features, for instance in V/1., main theme, b. 55ff: The > is given here, because Bruckner wished it to be played not weak (as it should be would be there no accent). But it should certainly not be louder as the weight-note at the beginning of the bar, as sometimes heard.

On the other hand, the articulation of the main theme of the Finale suggests that Bruckner wrote accents over all notes which should be emphasised, see V/4., b. 31ff: This would mean the first note of b. 34 should not be emphasised, but instead the usually weak four in b. 32, as indicated by >. Generally Bruckners phrasing and articulation is in much respect not >romantic<, but >baroque<: The first note under a slur should be strong, the final note of a slurred passage always weak, as much as final notes of phrases, if not indicated different. Very often Bruckner indicates slurs which include repeated notes. Unfortunately, he gives an explaining tenuto line only rarely (in particular in N° IX); conductors should carefully check such passages; I recommend if necessary to insert tenuti by hand into the parts to make sure players would give a clear articulation and not tied notes.

Sometimes Bruckner also writes slurs above two notes which could be misunderstood as ties, in particular in syncopated situations, for instance, VI/2., b. 9ff, Horn, where we see only from the Viola part that these notes should be perhaps played portato, and not tied. (This is the more difficult to decide because Bruckner sometimes may indeed mean tied notes even when he avoided to write a dotted note.) Hence, such passages must be read very carefully. Sometimes Bruckner indicated a verbal »legato« and did not give precise slurs. This is particularly often the case in his earlier scores written on oblong paper, with little space only between the staves for long slurs. The Bruckner-Gesamtausgabe (= BGA) rarely added editorial slurs in brackets here; the conductor should better carefully add slurs in such cases. Spiccato playing for strings is occasionally indicated with small wedges, and sometimes, for writing convenience, also with a ' in the manuscripts. In the strings, he also often indicated upbows and downbows, sometimes together with accents. Since Bruckner was a Violin player himself, these indications should be strictly observed by the players, even if they seem to be uncomfortable. Very rarely he used the >Bogen-Vibrato< (dots under a slur, for instance in VII/2.), but of course staccato and tenuto.


A crucial question in the performing practice of Bruckner's symphonies is the choice of the tempo, because his tempo indications refer to his music-theoretical thinking, rarely understood by conductors. Bruckner's technique to give an entire symphony a motivic integrity, including such features as the return of themes from earlier movements later in the symphony, requires a refined analysis of tempo relationships, in particular because such motivic connections are basically made audible to the listener by a distinct rhythm. A good example of this technique is N° VIII, in which the rhythm of the main theme of the first movement returns in the closing period of the exposition of the Finale, and in the same tempo. Hence, much less as in works by other composers, the choice of the >correct< tempo is limited here; Bruckner allowed much less freedom or decisions out of personal taste. Unfortunately the first print editions, which included numerous alterations of playing indications by their editors, established a performing tradition often against Bruckner's own indications. Since these scores were at least performed until the BGA was established in 1932 (with IX and V), the particular, personal interpretations of early Bruckner conductors and editors such as Franz Schalk and Ferdinand Löwe set the guideline of what is known as the >Bruckner interpretation< even to this day. Eminent Bruckner conductors, such as Jochum, Karajan and Klemperer, never questioned this tradition, despite the publication of the often contradicting Critical Scores. One can only suggest to the conductor not to trust in any >playing tradition< at all, but study the scores in detail and come to own conclusions, based on profound analytical insight.
Generally, Bruckner's tempo indication is related to the real meter of the harmonic progression, the >pulse< of harmony. This explains why Bruckner gives »sehr langsam« to V/2. in alla breve and quaver triplets, but also to VII/2., which is 4/4, due to the smaller note values, but the pulse related to the harmony proceeding in minims as well. In his Scherzo movements, the indication »schnell« refers to the crotchets, which move rather quick, hence, even if taken in whole bars the music seems to be rather slow. Also, generally spoken, due to the motivic connections between movements, there is usually a particular proportion between the tempi of different movements. Very often the tempo for the first and last movement has to be choosen identical. The slow movement may offer two options: The minims of the first movement correspond to the quavers of the Andante movement, or the the quavers of the first movement may correspond to triplet crotchets of the Adagio movement. (Main correspondences are 1:2 or 2:3.) Likewise to the Scherzo: Very often the triplet quavers of the Adagio correspond with the 3/4 crotchets of the Scherzo. Bruckner was aware of such correspondences and used them intentionally, following the ancient Tactus-principle. This is proven, for instance, by a letter to Wilhelm Tappert in Berlin, dated 6. 12. 1876, preparing a performance of N° IV, in which Bruckner gives some specifications about the tempo change from Andante to Adagio in the second movement, explicitely writing: »Die Achtel wie früher die Viertel« (»quavers as earlier crotchets«; Facsimile: Bruckner, Briefe I, BGA, Vol. XXIV/1, p. 165). Hence, the conductor must closely study the connections and correspondences between the movements, to choose the correct tempo. Sometimes Bruckner's movements have more than one tempo, sometimes two, sometimes even three. In this case there are likewise strict metrical correspondences to be observed. Bruckner may even set such a clever feature as in VI/1. or VIII/4., in which the initial tempo of the main theme is, however, not the main tempo of the entire movement. Bruckner did not give metronome markings, with one significant exception - VIII/4. Already in the sketches (!) he marked the first tempo »Feierlich, nicht schnell« (>Solemn, but not fast<) as minims = 69, and the second »langsamer« as minims = 60. This latter tempo, for the second and third theme, seems to be the basic pulse of the entire symphony, because (as mentioned earlier) in the closing group of the exposition the rhythm from the main theme of the first movement returns. The first movement has no metronome marking, but the similarities between the second and the third theme of it and the finale as well as this rhythmic connection make clear that minim = 60 is also the tempo of the first movement - in particular since Bruckner referred to its end with the picture of a >death-clock<: »Somebody expires on his death-bed, but the clock at the wall continues to beat: tic-tac, tic-tac« (his own words), hence, one beat per second, or minim = 60. Bruckner marked this first movement »Allegro moderato« in alla breve; so we can assume that his later, moderate allegro first movements should often be taken close to minims = ca. 58 to 63, making only a slight difference between movements in 2/2 or 4/4. Also within a movement, the Tempi don't allow for much change. In particular at angles between sections the pulse must be strictly kept, if not designed otherwise, in particular when the closing period is developed from rhythmical material of the main theme, or when it is rhythmically prepared at the end of the song period, as, for instance, in V/1., which must have only one tempo, »Allegro«, opposed to the »Adagio« of the introduction. The development of the movement proves that the crotchets of the Adagio have to correspond with the minims of the Allegro. Bruckner usually counts on a firm basic speed. Only occasionally he allows for a local ritardando (using the word normally if followed by the preceeding tempo) or a ritenuto (mostly used if followed by a new tempo). In his earlier symphonies, in particular the initial versions of N° II and III, he also gave »rubato« sometimes, in later revisions replaced by »rit.« or slower indications. Hence he understood the word >rubato< as a local deviation from the basic speed, due to the wish of the conductor. It furthermore seems as if Bruckner in his earlier work phase (until ca. 1876) generally felt a faster basic pulse of ca. minims = 69-72 for his first movements, and only from N° V onwards we find his typical moderate allegro of ca. 58-63. The quicker pulse only re-appeared in the first movement of N° VI and the Finale of N° VIII (but only as a kind of >framing tempo<).
Conductors may feel uncomfortable with the idea that so much of the tempo choice is determinated by the compositional fabric - however, this is also the case in most of the Vienna classical composers! Perhaps Bruckner is only a bit more strict than, for instance, Brahms, due to the more profound theoretical aspects of his composing. On the other hand, the conductor must be aware of that an arbitrary choice of a tempo, due to personal taste, may significantly change the clearness of the formal disposition, and then achieve perhaps the opposite effect: Bruckner's huge works are best comprehensive if the entire fabric can unfold and all affects and contrasts are presented as close to the text as possible.

— Recommended: Bruckner Interpretation, report of the Linz Bruckner Symposium 1982, ed. Linz 1983; Harry Halbreich, >Verlangt Bruckner ein einheitliches Tempo?<, in Bruckner-Jahrbuch 1981, p. 191-204, Linz 1983; Hans Swarowsky, Wahrung der Gestalt, Manfred Huss, ed.; Universal Edition 1979; ISBN 3-7024-0138-5

Conducting Bruckner: Style of Playing

Regarding the string playing, the style of that time required a vibrato which was expressedly treated as an ornament to colour and highlight phrases of particular expression. A non-vibrato was the start to develop a tone, and the end of a phrase in any case. However, note Bruckner's own indications in particular passages, where he gives »gezogen« or »breit gezogen«, as opposed to »ausdrucksvoll« or »innig«. Note that he also gives tenuto on repeated crotchets or minims in strings, with perhaps the same meaning. I can't help to feel that with »gezogen« he wanted to indicate a non vibrato playing by all means (that is, to use the bow only and move it in a controlled manner), as perhaps a reaction towards an already increasing general vibrato at that time, or to indicate that notes should not be played with a >messa di voce< ( < >). »Ausdrucksvoll« would then allow for more vibrato. Note, for instance, IX/1., b. 115: Violas and Doublebasses have »gezogen«, Celli »ausdrucksvoll«; both Violins play ornamental counterpoints. Such writing in multiple counterpoint in four parts requires much differentiation. The Celli come through only here if they play their line indeed expressive, and with vibrato; otherwise it could not be indentified by the listener as the leading line; Violas and Basses shall provide a >neutral< bass; the Violins would certainly play not expressive, that is, no permanent vibrato, and second Violins developing the long note at the beginning from a non vibrato. One should also not overlook that Bruckner, as an organist, was used to the effect of the >Tremulant< and >Vox coelestis< registers. Sometimes he used a real tremolo in similar orchestral textures, but sometimes such passages seem almost to cry for a dense vibrato, note, for instance, N° V, first movement, b. 247-58. One should perhaps better not dogmatically insist in a non vibrato throughout, because this robs the music an important method for more differentiation in the realisation of sounds. On the opposite, a discrete vibrato seems to be required sometimes, quite as much as in particular places (even if very rarely) a dense, intense vibrato, but certainly never the fat vibrato-sauce which is common in our days. One should also be aware of the typical string portamento of that time, as evident even from recordings up to the late 1920ies. For instance, the Trio of N° VII seems to cry for much portamento. Regarding the general style of playing Bruckner, I had some very interesting talks with Daniel Harding in Stockholm, where he performed Bruckner's Ninth. He spoke about the typical >Bruckner mode< into which every orchestra falls, however >historically informed< they might play other composers: This can be described as »playing Bruckner always sostenuto, always monumentalising each tiny little detail to an absurd extent, with no sense for declamation, speech-like phrasing and drama.« He also talked about that with Ken Ward, editor of The Bruckner Journal (Vol. 12, N° 1, March 2008, p. 41ff), of which I quote some lines in the following - observations which seem to be a fitting conclusion of this essay.

»I think Bruckner [.] works very, very well [.] the less you interfere with it. But then, to find that balance where you still get the right level of sensitivity and subtlety is extremely difficult. [.] The hard thing for a conductor, clearly, is to do with shape. You're dealing with enormous structures, and there are things where you can lose your way very, very early and very, very dramatically, and the whole thing can collapse around you. But as with all great composers, the real challenge is that it's never just one thing: perhaps we have this very clichèd idea of Bruckner - [.] the >Cathedrals of Sound<, and the idea of the music being >composed at the Organ<: Of course the large scale and the impressive and the enormous side of Bruckner is important, but just one aspect of the many sides. I don't like, generally, when music becomes soup, and for me a lot of the long-term tension in Bruckner comes from [.] gesture, and it's built up from gesture, and that's both harmonic and rhythmic - so you can have long stretches where the harmony is moving very, very slowly, and the thing that makes that relate, that holds it together or builds the tension or slowly releases the tension, is to do with repetition, to do with the rhythmical tension that's inside. If you don't concentrate on making those things audible, then in the worst case you can end up with [.] a 45 second D minor chord. It's the inner life inside that is very important. Young musicians are always told to >make the music sing<, which is of course important - it should sing. But there is also a lot of speech in music, and I think a clear pronunciation is important in Bruckner. [.]«

»I hope I'm not misremembering this because I remember reading it when I was a teenager, but somebody asked Barenboim what he's learnt from Furtwängler, and he said that Furtwängler had taught him that you build up very big lines from very, very small elements. And I think sometimes in our quest for the Holy Grail, which is the long line, we forget some fundamentals about how you really build up a proper long line. I remember Sir Simon Rattle when he first went to conduct the Vienna Philharmonic (there's an orchestra that's famous for it's long line!) he said what's fascinating is to see how they build that up from having an incredible control over small articulation. The thing is, composers don't write things generally by accident, and if you've got a detail in a score that's hard to hear, that's not an excuse for not hearing it! [.]«

»There's something Bach-like in the kind of perfection of the music on the page with Bruckner. Of course I can think of exceptions, and an exception is the very beginning of the third movement of the Ninth Symphony where he's deliberately having the musicians strain at the limits of what's possible and he wants you to feel that tension. But I would say in quite a lot of Bruckner you really want to try to realise the music as it is on the page, and being aware of what's technically difficult often doesn't help. [.] Sometimes he writes things for the Violins which are ferociously difficult, but they have to sound just as easy as if you suddenly play this note down here and then this very high note on an Organ. Whereas other composer are using the technical difficulties, I think in Bruckner a lot of the time you have to hide them. And the easiest way to hide them is to give people time to do them, and of course then the thing for the conductor, you want to find a way of keeping the shape, of keeping the skeleton of the thing in a rigorous way - because you can have wonderful moments along the way, but if you focus properly on >here you're going<, the feeling when you get there is much greater. And that's a very hard thing to do.«

»For musicians, Bruckner is one of those composers who occupies such a special place that we have this instinct built-in that we have to suddenly play his music in a way that we wouldn't play anything else. And orchestras who know how how to phrase in the most wonderful way suddenly get Bruckner and play every single note completely equally! There are moments in Bruckner where clearly the way he's writing accents or the way he bows things, he wants something to be the cliché >like it is on the Organ<, so everything is equal. But there is other stuff, there's Austrian folk music, there's choral music, and inflection and phrasing, and that what Ben Cohrs was talking about with Bruckner - the pendulum swing, which is very, very important, and the weighting and the articulating of large phrases through having a clear hierarchy of what's important, even the smallest sense of phrasing. There has been a tradition of making Bruckner monumental down to the smallest detail, and that has something to do with the question of security, because it's a much, much less frightening and exposing way to play for the orchestral musician. There is nothing that requires greater control over your instrument than phrasing, because then you're talking about the absolute subtleties of light and shade. And you know we're not talking about musical traditions that go back more than 50 years, which is a very short space of time, so I don't think it's something that we have to treat as a holy. I think it had a lot to do with fear and to do with finding a way of playing which is beyond reproach: You play exactly the full length of the note that's written on the page, sostenuto, no change in infection, and nobody can say you've done something wrong! It sounds like I am exaggerating, but in a way I am not. The wonderful thing in music is to try and understand what it means, not what it looks like, and for me listening to somebody like Harnoncourt conducting Bruckner [.] is such a revelation for me in terms of, you know, >wow, it's awesome music!< It's not just huge and impressive, but for the orchestral musicians it is much more difficult to play that way than it is to play in the monumental way.«