Staging the Secession: the revolutionary career of Alfred Roller

[From OPERA, November 2014, pp. 1360-5]

Alfred Roller

Max Mell’s brief 1922 biography of Alfred Roller tells us that when asked as a child what he wanted to be when he grew up, Roller would reply, ‘someone who’s allowed to go backstage’. It seems like an almost comically modest ambition for the artist-turned-stage-designer who would prove to be so significant a figure in early-20th-century Austro-German theatre and opera. Inevitably, though, Roller stands in the shadow of two great musical collaborators: Mahler, with whom he helped create a brief golden age at the Vienna Hofoper at the beginning of the century; and Richard Strauss, for many of whose operas he provided important designs, and with whom he shares a 150th anniversary this year (Roller was born in Brno on 2 October 1864).

In retrospect, it’s easy to see Roller as the figure that the Haus am Ring was in desperate need of to provide a visual complement to the musical reforms Mahler—appointed its director in 1897—had introduced. The son of a distinguished graphic artist and teacher, Roller showed an interest in art from an early age. He moved to Vienna in 1884 and, respecting his father’s desire that he take a suitably professional course in his life, first read law and art history at the university, before being drawn to courses in painting and architecture at the Akademie der bildenden Künste. He was a founder member of the Viennese Secession, and co-editor of its official journal, Ver Sacrum, for which he provided illustrations—including for the cover of its first issue. He joined the staff of the Vienna School for Applied Arts (the Kunstgewerbeschule) in 1900 and succeeded Carl Moll as the Secession’s president in 1902.

The next important chapter of Roller’s artistic life began that same year, when he masterminded the Secession’s famous Beethoven exhibition, built around Max Klinger’s vast Beethoven Monument. Another of the exhibition’s main features was Klimt’s Beethovenfries (still exhibited in the Secession building) and Roller’s own fresco Die sinkende Nacht took up position behind Klinger’s Beethoven, its apparently explicit Wagnerian overtones undermined only slightly by the fact that opposite it was another, less overtly Tristanesque fresco, Der werdende Tag (‘The Dawning Day’), by Adolf Böhm.

Plans to have Mahler contribute by conducting the Vienna Philharmonic and the Hofoper chorus in Beethoven’s Ninth Symphony foundered on what—according to Mahler’s own account—was the mercenary attitude of the orchestra (in typical Viennese fashion, it turned into a scandal, accompanied by much claim, counter-claim, speculation and finger-pointing in the press).

There are contradictory accounts of the first meetings between Mahler and Roller, which took place around this time—one involving Roller sketching designs for Tristan on a café tablecloth, another involving a Tristan design Roller was supposed to have exhibited. Most likely is that the two men met at Moll’s house, and that the conversation led quickly onto Wagner in general, and Tristan in particular. It was the last of the great Wagner works that Roller had got to know, and he had been gripped by it like no other, never missing a performance at the Hofoper, even if, he claimed, he tended to sit through them with his eyes closed.

Roller was critical of the Hofoper’s existing design team of Anton Brioschi and Heinrich Lefler; Mahler seems to have had little admiration for them either. Brioschi, though a fine artist, was an establishment figure who had followed in his father Carlo’s footsteps at the Hofoper and honed his style under Simon Quaglio (responsible for the scenery of the inadequately old-fashioned Munich premieres of Rheingold and Walküre). Brioschi was increasingly out of fashion in a Secessionist Vienna keen to unleash the full power of Wagner’s works; Hermann Bahr, always a reliable barometer of the city’s prevailing artistic attitudes, said that he ‘painted bad designs to perfection’.

Lefler was a co-founder of the Hagenbund (an alternative artistic movement to the Secession, but one with similar aims) and brother-in-law of Joseph Urban, who would later became an important stage designer in New York. But although Lefler certainly had some vitalizing influence after he joined the Hofoper as Brioschi’s assistant in 1900, it seems to have been limited. Mahler gave the responsibility for the designs of the new Tristan to Roller, and did so straight after that first meeting. 

Both Mahler and Roller, it seems, were well versed in the writings of the Swiss theorist Adolphe Appia, who argued for a theatre not of illusion but one of suggestion—an ‘Andeutungsbühne’ rather than a ‘Illusionsbühne’, which would rely on and exploit the viewer’s imagination rather than restrict and restrain it with literalism. As Patrick Carnegy notes in his Wagner and the Art of the Theatre, Mahler reportedly welcomed Roller’s idea of ‘a stage on which everything is only intimated’.

As with many of Roller’s designs, it’s difficult in our seen-it-all-before, post-Regietheater age to appreciate quite how revolutionary the Mahler-Roller Tristan was—each act, after all, showed the settings Wagner’s score specified. But, as Carnegy points out, only two decades after Wagner’s death it was almost unheard-of to create a production that didn’t take as its starting point Wagner’s own Munich production of 1865, or even the more questionably ‘authoritative’ Bayreuth staging of 1886, realized by his widow.

It was also a production that sought at every point to reflect and complement the mood of the score, and one that used lighting as it had never been used before, as an expressive tool—even almost, in the symbolic use of colour, as an expressive language in and of itself, rather than as simply a means of illuminating painted backcloths. Props and scenery were, for once, three-dimensional, a fact that offered, through the possibilities of shadow, yet further expressive opportunities.

Further important productions of the Mahler era included a staging of Wolf’s Der Corregidor in 1903, as well as an architecturally imposing Fidelio in 1904, the massiveness of whose sets (and resultant lengthy scene changes) arguably played a part in Mahler’s decision to insert the Leonore No. 3 overture before the final scene—a tradition that persists in Vienna to this day. The 1906 production of Don Giovanni was, if anything, a bolder experiment, in which four ‘Roller Towers’, as they were dubbed, were fixed, two on each side of the stage, framing the action and serving multiple purposes while the scenery behind was changed.

It was an idea that allowed unprecedentedly swift scene changes but which was also in part motivated by an ‘authentic’ desire to reconstitute the 18th-century idea of the two-part stage—all but banished in modern theatres of the time—with the ‘proscenium’ the domain of the principals, the stage itself reserved for the ballet, the chorus, the special effects and the extras.

There were further innovations for a new Rheingold and Walküre—the former performed, for once, without an interval, the latter uncut and with a design for the Walkürenfels that was strongly reminiscent, Carnegy has suggested, of Appia’s own design for Act 3 of that opera. The simple choreography Roller encouraged for Iphigénie en Aulide in 1907—the final Mahler-Roller production—showed once more the concern shared by both men that stage action should reflect primarily the music, creating opera performance not as a disconnected agglomeration of elements but, rather, as something approaching the ideal of the Gesamtkunstwerk.

The Ring project never reached full fruition, with Mahler resigning at the start of the 1907-8 season; Roller, without Mahler there, lost control over the remaining instalments. Still, he produced designs for several new productions, including the first Vienna Elektra in 1909, the year that Roller returned to teaching, taking up the directorship of the Kunstgewerbeschule.

But as his close working relationship with Mahler was coming to an end, so began another chapter—that with the triumvirate that would, a decade later, set up the Salzburg Festival: Strauss, Hofmannsthal and Reinhardt. Roller’s activities had for some time also included work in the spoken theatre, and he had designed important pre-war productions for Reinhardt, including the director’s massive 1911 productions of Oedipus the King, The Oresteia and Jedermann in Berlin’s Zirkus Schumann. It was in that same year that Roller provided the designs for the 1911 Dresden premiere of Der Rosenkavalier, which Reinhardt, joining the creative team late in the day, had been instrumental in successfully bringing to the stage in the face of the internal politics of the Semperoper.

Unusually, those iconic, much-reproduced designs also became part of the work itself, part of the publisher’s property and—in an interesting development for a designer whose work is often seen as a precursor to Regietheater—an essentially inalterable element for anyone wishing to perform the work themselves, at least in the first few years of the opera’s existence. But if Roller’s work on Rosenkavalier was deemed a success, Hofmannsthal, in particular, was dissatisfied with his designs for Die Frau ohne Schatten, premiered in 1919 in an Operntheater (the Viennese house was briefly without either of its prefixes, ‘Hof-’ or ‘Staats-’) under the control of Strauss and the conductor Franz Schalk. It lacked the necessary Baroque theatrical magic, he said, even if, as with Der Rosenkavalier, the librettist and the designer had worked in detail on a Regiebuch—a ‘direction book’—to help fix the idea of how the work should be performed.

The team behind the first performance of Der Rosenkavalier

Roller had helped create the aesthetic languages that defined these two Strauss-Hofmannsthal operas, and was involved in important, technically brilliant postwar Viennese stagings of Palestrina (1918), Die Gezeichneten (1920) and Die tote Stadt (1921). He also provided costumes for the magnificent first Viennese performances (glamorously double-cast and conducted by Toscanini) of Turandot in 1928. The ‘director’—inasmuch as modern parlance can be applied to the differently defined roles of the time—on that occasion was Lothar Wallenstein, with whom Roller also collaborated on important Viennese Verdi stagings in the late 1920s.

Maria Nemeth as Turandot in the Roller-designed
production in Vienna

A brief sketch like this of so long a career inevitably misses out some important achievements: a Shakespeare cycle at Vienna’s Burgtheater in the early ’20s, for example, the first Viennese performance of Salome (at the Volksoper in 1910), and extensive work at the Salzburg Festival. But one final production, at a rival festival, deserves mention: the 1934 Parsifal at Bayreuth, which, after much wrangling, replaced Wagner’s own 1882 staging. Politics bubbled around the new production with wholly predictable ferocity, with hard-line Wagnerians, determined to preserve the Master’s staging, coming up against Hitler’s own desire for a new production; and Roller—despite, in the words of Frederic Spotts, being ‘ill and in his cups’—was given the responsibility. In the end, he produced a staging ‘little different from the sacrosanct original’.

Hitler’s reasons for choosing Roller were complex, but had their roots in the designer’s time in Mahler’s Hofoper. The 19-year-old Hitler, a budding artist who greatly admired Roller’s Wagner productions, had, through a series of acquaintances, managed to gain an invitation to visit the great designer, to show him some of his work. However, as Brigitte Hamann has outlined (in Hitler’s Vienna: A Dictator’s Apprenticeship, Oxford University Press, 1999), when the day came, the young man bottled it.

That meeting that never happened offers a tantalizing suggestion of a path that history never took, and it’s unclear—without burrowing into the archives—what Roller thought of that young artist’s subsequent choice of career. It was also lucky, in a way, that he didn’t live to see the horrors that would come of it. Roller himself died in 1935; his son Ulrich, also an important designer, who worked on, among other things, the first Viennese performance of Strauss’s Daphne in 1940, wasn’t so lucky, perishing on the Eastern Front in late 1942.

Much of Roller’s achievement is difficult to gauge today: the impressions of his productions that can be pieced together only from eye-witness accounts and extant sketches and photographs are inevitably imperfect and unsatisfactory, and the importance of design and the visual aspect of opera was not, a century ago, deemed as central as it is now. Some, too, might regret the developments he helped pioneer, and the boundaries he helped break down. In the anniversary year that he shares with a composer with whom he worked so closely, though, it’s perhaps time to remember how important he was in revolutionizing the visual side of the multi-faceted art form that is opera.