by Hartmut Haenchen

In the age of media, it is generally no longer about the representation of the work itself but about the self-representation of conductors and directors. Since the majority of media outlets, faced with a multitude of events, pay attention only to extremes — it must be extremely young or old, extremely fast or slow, extremely soft or loud, extremely beautiful or ugly — an overwhelming number of performers conform to this demand of the times in order to attract the attention of these media.

Proving this thesis does not seem easy at first. Yet it can be clearly demonstrated through the example of tempo selection. A study of the performance traditions of Wagner's works reveals that during roughly the first 70 years of their existence, a certain direction of tempo could be identified, passed on from one generation to the next, exhibiting no extreme deviations. This transmission of the tempi traceable back to Wagner himself took place without recordings (and without the influence of the recording industry's marketing, which promotes its products regardless of their actual quality). Starting around the middle of the 20th century — coinciding with the enormously growing influence of the media — a tendency began to emerge that seeks extremes demonstrably far removed from Wagner's original intentions; for, on the evidence of all historical documents, we can assume that the performance durations (that is, the tempo tendencies) of the premieres certainly stood closer to Wagner's intentions in essential respects than the later extremes did, and were if anything too slow rather than too fast. The following chart makes clear what is meant by this.

CD-Aufnahmen: kursiv

Naturally, tempo depends on many factors that cannot be detailed here. No conductor is in a position to achieve exactly the same tempo evening after evening — certainly not in an art form like opera. The deviations involved, however, amount to a maximum of 14 minutes (and that is already extreme) over the entire Ring for the same conductor in a single run, according to the Bayreuth records. On the other hand, the Ring is well suited to illustrate a general tendency, since it is music that has had to endure (in its hitherto extremes) a difference in performance durations of 2 hours and 19 minutes (!). One must, however, be aware that there are several thousand different tempi in the entire cycle, which are in turn linked by Wagner's proportional indications and from which a total performance duration results. But since all very fast tempi cannot (significantly) be played any faster, due to the technical limits of playing, tempo differences can take place only in the medium and slower tempo range. It is well known that the premiere conductor Hans Richter, who was master of nearly every instrument, could in every case prove to the players that everything Wagner had written was indeed playable — even if it took, as for the Bayreuth revival of the Ring, 46 orchestra rehearsals.

If we assume that today certain technical playing problems are easier to handle than in Wagner's time, then the difference from Wagner's tempo ideas in today's customary slower performances is to be considered even greater.

As always, only original sources can of course be the basis for judging the "right" tempo.

In our production in Amsterdam we are in a position to take all the records of the musical assistants of 1876 ( Porges, Levi, Mottl and Kniese ) into account in the interpretation for the first time. On the basis of Wagner's original performance ideas, the overall tendency in performing his works should therefore be somewhat faster than at the premiere. A few examples can make clear what Wagner was after.

Wagner, who as the author also directed the staging and of course had overall musical authority, needed conductors who were able and willing to enter into his conceptions without reservation and to realise them.

Wagner's rehearsal remarks of 1876 must be understood in this light.

For Walküre and Siegfried — to single out two universally valid examples — 715 remarks are recorded in total. The astonishing thing is that 208 of them are instructions calling for tempi faster than those realised by the premiere conductor Hans Richter in rehearsal. That is fully 30% of all surviving remarks. Against this stand only 135 instructions relating to slowings, 101 to dynamics, and 206 to expression and articulation. It is striking that, for these two operas alone, Wagner gives the instruction "Do not drag" 36 times and "Without sentimentality" 17 times. There is no room here to analyse all the instructions, but even those that essentially describe a slowing — and were counted in that category — nevertheless imply an acceleration compared with today's practice. Let me illustrate this with two examples: in Act I of Siegfried at Mime's text "Nun tobst du wieder wie toll" Wagner first writes in the score "Very gradually always somewhat slower" and a few bars later "Very moderate and still slower," only to indicate shortly thereafter, with "Andante," a new tempo arrived at through this slowing. Evidently this was already leading, at the time of the premiere, to a tempo that was wrong because too slow. To this Wagner remarked: "Observe all tempo modifications precisely, but in slow passages never go so far that a feeling of lingering repose arises." At the start of Act II of Siegfried the score gives the tempo marking "Sluggish and dragging," and the music is essentially dominated by the "Giant motif" and the "Hoard motif" as the musical expression of the giant-worm Fafner sleeping on the hoard. In recent interpretations this opening is taken extraordinarily slowly. But Wagner described very precisely in his 1876 rehearsal instructions what he meant by this: "The 'sluggish and dragging' is best expressed by a slight holding-back on the 2nd and 4th quarter of the Giant motif; in the intervening bars, again forward in tempo." In Act III, at the "Father-joy motif" which he marks in the score with "very moderate," he remarks: "These first eighth-notes always slightly held, then continuing more flowingly..." The examples cited and the great number of instructions for textual articulation and for expression make clear that Wagner followed, with his tempi, the rhythm of speech and the verbal intelligibility, thereby counteracting every form of stretching out and false sentimentality. Porges writes in his report on the premiere of 1876 (Rheingold, p. 15): "These dialogue passages will produce the right impression only if the tempo in which they are performed is essentially the same as that of spoken speech." It cannot be said more clearly. Elsewhere (p. 14) Porges also points to the dynamic relation between singer and orchestra, which ultimately has consequences not only for dynamics but also for tempo: "During the rehearsals of the Nibelungen Ring it turned out, in fact, to be a necessity to moderate the dynamic markings of sound-strength at many places — often to put a forte in place of a fortissimo, a mezzo forte in place of a forte, and so on. This was done above all in order to allow the singer's word and tone to come through clearly; for we should not forget for a moment that we are attending a dramatic performance, one which is supposed to take effect through the convincing presence of an action modelled on real life, and not, say, taking in a work of purely symphonic art. For the rendering of those symphonic passages in which the performer is at the same time supposed to take effect through the sung word, the rule therefore applies that in such places the strength of tone-production may never reach the utmost degree." Experience with singers of the "heavy" repertoire shows, however, that the bigger the tone, the slower the tempo becomes. Wagner evidently wanted to deliberately counteract this. Porges continues: "This relation of orchestral sound-strength to the singer came up repeatedly in the course of the rehearsals, and the master made repeated and favoured use of the comparison that the orchestra should always carry the singer as a moving sea carries a small boat — but must never bring it into danger of capsizing or even swallow it up. The observance of this rule must, however, just as little tempt the players into a soft or even indifferent manner of performance; rather they must, with the most concentrated attention, take care to bring out the plastic outlines of their melodic and thematic complexes in all their pithiness, through especially clear phrasing of the periods and the utmost definiteness in the execution of the metric and rhythmic accents." If this already held for the orchestra-damping acoustic situation in Bayreuth, how much more does it hold for other opera houses! It is interesting, with this insight and the knowledge of Wagner's original notes (which up to now stood in no score), to follow the performance traditions. On one side there are the surviving performance times, which can of course only reflect tendencies. It becomes clearer in the surviving recordings. There my thesis is confirmed by extant sound recordings.

The HMV-recorded "Potted" Ring (1927–32), with the conductors L. Collingwood, L. Blech, A. Coates, J. Barbirolli, R. Heger, K. Alwin, K. Muck, makes clear that Wagner's instructions lived on, and that each conductor in his own way tried to come as close as possible to Wagner's intentions. These are still to be heard on these recordings, with their differing musical signatures, in flowing tempi and brilliant treatment of the text. The recordings correspond to the situation shown in the chart above: that during the first 70 years of performance practice a "performance tradition" based on Wagner himself still existed. This was interrupted by two factors: the breaking-off of the direct performance tradition, and the rise of fascism together with the simultaneously emerging influence of the media.

The break in the direct tradition came after the deaths of Siegfried Wagner and Cosima (1930) and the dying out of the first and second generations of conductors. Siegfried Wagner had failed to build up a successor generation. None of the later conductors except Kaehler had been an assistant at the Festival. It is understandable that from this point on the orally-transmitted and scattered written performance details going back to Wagner himself disappeared in the following years. The performances by conductors who were not fully fluent in German brought additional confusion into performance practice, since musical instructions such as for example "sehr gehalten" (very sustained) were suddenly understood as tempo instructions rather than articulation instructions. Alongside Toscanini — who belonged to that category and was one of the slowest Wagner conductors of all — stood the next great Wagner-conducting personality: W. Furtwängler.

In contrast to the "Bayreuth style," he rejected the equal standing of text, theatre and music and gave clear primacy to the music: "But the 'whole' of the opera, its structure and its meaning, is determined by the music, to which therefore the primacy within the opera also falls."

That the fascist period evidently not only abused Wagner's work ideologically but also led to sentimental, pathetic and thus slower performances cannot, unfortunately, be conclusively proved from the Bayreuth performance times, since the relevant durations have not been (sufficiently completely) preserved.

Notable in this connection, however, is the fact that Furtwängler's interpretation became 40 minutes (!) slower between his first Ring of 1936 — which still roughly corresponded to the premiere times — and his recording of 1953. The recording also clearly shows that the surviving instructions of Wagner were almost no longer taken into account, and in many cases stood in outright opposition to the now-reassembled remarks of Wagner.

That Furtwängler shaped many subsequent conductors is of course beyond doubt. And the majority of later recordings and performances are slower than the premiere or than the tempi of the first 70 years after the work's creation. The deviation of the slowest Ring from the premiere amounts to 1 hour 7 minutes; the deviation of the fastest Ring from the premiere amounts to 1 hour 12 minutes. One must keep in mind that for Wagner's wishes the premiere — especially of Siegfried — was already too slow, so the difference between today's slower performances and Wagner's conception was even greater.

There was, however, another line of tradition that did preserve something of the original Bayreuth style: Richard Strauss, who assisted at Bayreuth in 1898, met — as regards the tempi of performances of his own works by others — a fate similar to Richard Wagner's. (One need only compare the recordings under his direction with newer recordings, which are almost without exception slower.) He held Felix Mottl (assistant for the first Ring and conductor of the 1896 Ring) in the deepest reverence. Although Mottl was often criticised for his "slow" tempi (he took 1 minute (!) longer for the entire Ring than Richter), we can assume that the tempi were still very close to Wagner's intentions. Cosima wrote that "Mottl was a thoroughly stage-oriented conductor, who knew masterfully how to preserve the connection between scene and orchestra." With this he obeyed a central demand of the Bayreuth style. Strauss felt himself a direct successor to Mottl, and in turn found successors such as Clemens Krauss and Karl Böhm, all of whom remained somewhat below the premiere times.

Richard Strauss once said: "It is not I who am faster in Parsifal; it is you in Bayreuth who have become ever slower. Believe me, what you do in Bayreuth is really wrong."

Gustav Mahler also expressed himself to the same effect.

Finally, there is the peculiarity of the "invisible" orchestra, in which, owing to the orchestra being placed especially deep below the stage, direct contact between the individual players and the stage is impossible — which is why a general tendency toward slow tempi is to be observed in Bayreuth, aptly described by Wieland Wagner:

"This is also where, to a great extent, the dragging here in Bayreuth comes from. Each waits, more or less unconsciously, for the other, and only resolves to go on once he believes he hears him." It is generally known that the special and much-praised acoustics of Bayreuth are really fully effective only in Parsifal. In the early works composed for other stages, and also in the Ring — which lives by far denser structures than Parsifal, especially also in Meistersinger — one is aware that the Bayreuth acoustic is by no means the ideal, since it blurs the counterpoint of these works. From remarks and notes one can surmise that Wagner himself might possibly have made changes if he had been able to realise the Ring himself once more in 1896. As it is, however, the orchestra placement at Bayreuth has been declared "sacred," and H. von Karajan ran into great difficulties when he wanted to try to change something. Also an example of how tradition can ossify. Documents that have since been found prove that in the years up to Siegfried Wagner's death there was real experimentation with the acoustics and the orchestra placement.

Finally let it be pointed out that the predominant slowing-down of tempi in other works of Wagner was equally pronounced, or even more so, as for example in Parsifal, which contains only a few fast tempi that — as set out above — behave largely "tempo-neutrally": the 1882 premiere under H. Levi lasted 4/04, in 1888 under F. Mottl 4/15, in 1897 under A. Seidl 4/19, in 1901 under Karl Muck 4/27; in 1909 under S. Wagner there was a small correction of this tendency at 4/22; in 1931 under A. Toscanini a record of 4/42 (38 minutes slower than the premiere), after which, under the influence of C. Krauss, the opposite tendency followed in 1953 (3/44), only to be reversed again with J. Levine in 1990 to the other extreme of 4/33. Astonishingly, tempo differences of nearly an hour are conceivable within a single opera. Compared with the entire Ring, where the differences come to "only" just under one and three-quarter hours (calculated over about 14 hours of music), the extremes for a 4-hour work are truly extreme — but they constitute clear support for the views I have set out as to the reasons for the slowing.

A few detailed comparisons make the tendency toward slowing of Wagner's music even clearer than the total times: the first recording of the Meistersinger Prelude (1905) with Myrte Elvyn lasts only 7´55 min. R. Strauss took 9.25 min in 1944, and Bruno Walter 10´10 min in 1959. In 1905 E. Paur, in his recording of "Siegfried's Death" on the Welte-Mignon piano, took 8´32 min. A. Toscanini takes nearly half that tempo for the same piece and needs 13´57 min. Olga Samaroff played the "Ride of the Valkyries" (concert version) in 3´45 min in 1905. W. Furtwängler takes 4´30 and A. Toscanini 5´25 min for it.

But let us go back to the sources: in a letter before the first Bayreuth Ring in 1876, Wagner wrote to his premiere conductor Hans Richter: "Friend! It is indispensable that you attend the piano rehearsals carefully; otherwise you will not get to know my tempo, and then it is more than burdensome — to the detriment of the whole — to make this up in the orchestra rehearsals, where I do not gladly enter into a discussion of tempo with you for the first time. Yesterday we could not get out of the dragging, especially with Betz (the Wotan), whom at the piano I have always had sing in the most fiery tempo. ... I really do also believe that throughout you stick too much to beating quarters, which always hinders the swing of a tempo, particularly with long notes such as occur frequently in Wotan's Wrath. By all means beat eighths even, where precision is served by it: only, one will never sustain the character of a lively allegro throughout by quarters." Elsewhere he writes: "It was only in those moments so humiliating to confess what so utterly distressed me, and herewith to declare that it was my horror at perceiving how my Kapellmeister, although I consider him the best I yet know, could not hold to the right time-measure — though it had often come off — because, yes, because he was simply incapable of knowing why it must be conceived thus and not otherwise." Cosima writes in her diary on 20 November 1878: "Richard exclaims again:

"I leave behind not a single person who knows my tempo."

From these passages too it becomes clear that what mattered to him above all was that the tempi should not become too slow, and he even gave directions on how to handle particular places conductorially. The same is confirmed by his loyal helper and assistant at the rehearsals for the premiere, H. Porges (see Contemporaries). In his notes (p. 31) he records: "Nowhere may an unmotivated hesitation or lingering, not dictated by the peculiar nature of the situation, take place," and shortly thereafter he reports of Wagner "that he is averse to every merely individual caprice, even should it manifest itself in a brilliant manner." In his description of the rehearsals for "Walküre" he writes of the scene between Wotan and Fricka: "In the whole dialogue, overflowing with engaging detail, he renewed an often-repeated admonition not to allow any hesitation in the tempo to occur, into which singers and players, particularly at deeply-felt places, are so easily inclined." On one of his visits to Venice as well he spoke about tempo: "Now in the evening I heard from him (the Kapellmeister H.H.) the Tannhäuser March, and the dragging tempo annoyed me; so I had it conveyed to him that, if he ever again did anything of mine, he should tell me, so that I could show him the right tempo, etc."

What I have characterised above as the "Bayreuth style" emerges also from a letter to the theatre director Angelo Neumann, in which he extravagantly praised his "pupil" Seidl: "None of all the conductors knows my tempi and the agreement of the music with the action. Seidl I have instructed."

Most impressive are two somewhat pointed rehearsal-remarks of Richard Wagner from the year 1876, which have come back to light in the course of the investigations into the sources of performance practice:

"If you weren't all such boring fellows, the Rheingold would have to be over in two hours." And "Mood is nothing at all. The main thing is and remains knowledge."

(Computer translated from German)