Roll over Beethoven

Schuller and Zander try to revolutionize the Fifth

by Jefrey Gantz

If there's one piece of Western classical music that's universal, it's Beethoven's Fifth Symphony. Even if you've never heard the entire piece, you're familiar with the "da-da-da-daaaah" first four notes. "Fate knocks at the door," Ludwig is reported to have said of this theme; once Samuel Morse invented his Morse code, it came to represent "V" for victory. It was the basis for Walter Murphy's 1976 disco hit "A Fifth of Beethoven," which millions heard in Saturday Night Fever. Asking someone to identify these notes would be about right for the $100 question on Who Wants To Be a Millionaire.

All the same, we're being told we've never really heard Beethoven's Fifth. Regular classical-music listeners know that over the past 15 years the period-instrument movement has given us recordings of Beethoven's symphonies (by the likes of Christopher Hogwood and Roger Norrington) that are utterly unlike the "mainstream" recordings of Arturo Toscanini, Leonard Bernstein, Herbert von Karajan, Georg Solti, et al.: clean, bright, and above all fast. But not, it seems, fast enough. At issue are Beethoven's metronome marks, which he provided for every movement of his nine symphonies and which in most cases call for tempos appreciably (in some cases alarmingly) quicker than exist on record.

Leading this charge to dethrone Establishment Beethoven and recrown Revolutionary Beethoven are two of Boston's own: composer, conductor, author, and former New England Conservatory president Gunther Schuller and Boston Philharmonic director Benjamin Zander. Schuller's massive The Compleat Conductor (Oxford University Press) appeared in 1997; after setting out essays on the philosophy and the history of conducting, the author subjects eight symphonic works to close scrutiny, and his conclusion in the case of Beethoven's Fifth is that the 90-odd recordings he listened to all ride roughshod over the score, ignoring not only Beethoven's metronome numbers but also his dynamic markings and his phrasings and bowings. What's more, Schuller has put his baton where his mouth is, collecting an all-star orchestra in New York City to record the Fifth and Brahms's First on his own GM Recordings label. Ben Zander, meanwhile, has a contract with Telarc to record all of Beethoven's symphonies with the prestigious Philharmonia of London; the Fifth and the Seventh came out earlier this year, and on the back of the CD the New Yorker's Andrew Porter is quoted as saying, "If Mr. Zander is right, we have been hearing the music of the greatest composer only in misrepresentation." Porter is referring to Beethoven's metronome marks: Zander, like Schuller, means to adhere to them, and now those two have been joined by David Zinman in his set of the symphonies with the Tonhalle Orchester of Zürich on Arte Nova.

So . . . is this a musical revolution? You might wonder whether any music from 1800 can sound revolutionary almost 200 years later, in the wake of the Tristan chord, the Mahler chord (the nine-note dissonance from the Adagio of the Tenth), Le sacre du printemps, serialism, swing, Elvis, the Beatles, Last Exit, and Metallica. But that isn't the biggest controversy Schuller's book and these new recordings raise. At issue is the extent to which a composer can define a piece of music -- and that in turn raises the question: "What is music?" Is it "complete" on paper, in the score? What, if anything, does the performer add? Are the notes in any way changeable? What about dynamic markings and phrasings? And tempos? Where does the composition leave off and the performance begin? Is it legitimate for a composer to dictate in every detail how his composition must be played? Is a score a "prescription" for the perfect performance? Or does there have to be some kind of synergy between composer and performer, a meeting of the minds?

IN THE COMPLEAT CONDUCTOR, Gunther Schuller (who in 1994 won a Pulitzer Prize for his composition Of Reminiscences and Reflections) comes down squarely on the side of the composer. In his preface, he says that in the course of his own conducting career, "the score became a kind of sacred document for me." His opening chapter, "A Philosophy of Conducting," lists among the qualities of the ideal conductor "an unquenchable curiosity about the miracle of the creative process; a profound reverence and respect for the document -- the printed score -- that embodies and reflects that creation; the intellectual capacity to analyze a score in all of its myriad internal details and relationships; a lively musical, aural imagination that can translate the abstract musical notations of a score into an inspired, vibrant performance; and on a more practical level, a keen, discerning ear and mind; a versatile, disciplined, expressive baton technique; an efficient rehearsal technique." Schuller allows that "given human fallibility and variability, absolute perfection is probably not achievable. But it is certainly the goal that conductors must strive for -- in order to have the right to interpret, to realize, the works of the great masters, whose genius is many, many times greater than their own."

Most conductors, of course, rate their genius rather higher than Schuller does. But they might think twice after reading through this 583-page tome, which excoriates in excruciating detail the conducting profession for its shabby treatment of classical music's sacred documents. Detail after detail, we're told, gets ignored: rhythms are butchered, critical lines obscured, dynamics muddied, tempos indulged. And it's not just run-of-the-mill baton wavers who're exposed but the luminaries of the profession: Furtwängler, Toscanini, Klemperer, Bernstein, Karajan, Solti, Boulez -- no one escapes Schuller's Dies Irae wrath. The opening pages on Beethoven's Fifth make for sobering reading: some conductors, it appears, don't just forgo the score during their performances but hardly bother to open it in rehearsal.

Every profession, Schuller acknowledges, has its charlatans, and conducting certainly has more than its share of egomaniacs. But The Compleat Conductor would go down easier with less self-righteousness and a better grip on reality. A working conductor (who's often also a music director with administrative responsibilities) doesn't necessarily have time to reconstruct a score, correct all the mistakes that have crept into the parts, and memorize every detail. And if he somehow did, his orchestra would never have the rehearsal time to put all that into practice. Even the most conscientious conductor (and some of those Schuller scolds were and are) can't get every detail in a recording right. Neither are the indicted individuals -- those who are still alive -- afforded any opportunity for rebuttal.

(Few could be more conscientious than Schuller himself, and yet The Compleat Conductor is by no means error-free. "The Viennese novelist and composer E.T.A. Hoffman" [page 138] spelt his name Hoffmann and never set foot in Vienna -- and Schuller baldly misrepresents his 1810 review of the Fifth, which was a pioneer in music criticism. Nikolaus Harnoncourt's name is misspelled on page 220. On page 137, "measure 64" should read "measure 94"; on page 209, "measure 104" should read "measure 114"; and so on. Even the dust cover gives the date of Schuller's Pulitzer as 1995 rather than 1994. Does Schuller know better? Of course. That's the point.)

But the larger question here has to do with what music is and how it is made. In a seminal 1987 essay, "Beethoven's Symphonies: The New Antiquity" (Schuller cites it favorably, though he somehow misconstrues its praise for Roger Norrington as censure), which appeared in the WCRB magazine Opus, Richard Taruskin writes, "The score is not meant to define the work, only to make its performance possible. Both phylogenetically and ontogenetically, music is anterior to its notation." If that last sentence is too much of a mouthful, just concentrate on the second part: the music comes first, then the writing down of it.

Schuller acknowledges, somewhat grudgingly, that "our musical notation has its limitations" and that "the ultimate, most subtle nuances and personal refinements of interpretation are in fact not, in an absolute sense, notable." The Compleat Conductor, he concedes, "will not be about such subtleties and refinements of interpretation" -- and in his view, it doesn't have to be, since "tempo, tempo modifications, dynamic and timbral indications . . . are more than adequate to achieve an ideal realization of a work. . . . Indeed, the problem in conducting and interpretation is not that our notation is inadequate but that 50 percent of it is ignored by most conductors." This is, I would argue, half right. There is a lot of information in the score, and conductors who ignore it should be taken to task. But if the score were truly all that's needed to achieve an "ideal realization" of Beethoven's Fifth, we could run it through a conducting computer. It's the very thing The Compleat Conductor disregards -- a conductor's "subtleties and refinements" -- that makes the music live.

Music doesn't live on paper. Jazz certainly doesn't -- and no one knows that better than Schuller, who's also the author of Early Jazz: Its Roots and Musical Development (1968) and The Swing Era: The Development of Jazz, 1930-1945 (1989), the first two volumes of a projected trilogy. At the outset of Early Jazz, he describes "swing" as "a force in music that maintains the perfect equilibrium between the horizontal and vertical relationships of musical sounds." He's trying to distinguish jazz from classical music -- but why shouldn't classical "swing" too? Or take what Schuller calls the " `democratization' of rhythmic values," by which he means "that in jazz so-called weak beats (or weak parts of rhythmic units) are not underplayed as in `classical' music." In fact it's the very ambiguity of weak beats that energizes classical music -- all aristocracy makes for a dull performance. Noting "the sing-song emphasis of strong beats" in the last movement of Mozart's Third Horn Concerto, Schuller explains how one can avoid this effect "by consciously emphasizing the eighth notes." Exactly.

The "realization" of music -- jazz, classical, or any other -- is a re-creative art, just like the "realization" of a play. For Shakespeare, the play was what went up on stage (and it was different in every performance), not what came out in the quarto volume. He'd add contemporary material; his company would revive a slightly different version for the sake of novelty. The counterpoint between the rhythm of iambic pentameter and the rhythm of his phrases created tension and breathed life into his lines -- that's part of what raises Shakespeare above his Elizabethan and Jacobean peers. And each actor who played Hamlet or King Lear would give the part a different personality. The stage was a conversation between creator and performer.

Some playwrights, like some composers, try to straitjacket the performers by legislating every detail -- George Bernard Shaw, Seán O'Casey, Eugene O'Neill, and Samuel Beckett come to mind. Descriptions of how a character should dress, walk, feel, and think can be helpful, but past a certain point I hear the playwright saying, "You can't play with my play." Maybe it's just a coincidence, but the dramatists who rank highest in Western culture -- Aeschylus, Sophocles, Euripides, Shakespeare -- leave the performance to the performers.

Not all performers play fair, of course. Earlier generations played fast and loose with Shakespeare's texts. In 1681 Nahum Tate rewrote King Lear, eliminating the Fool and giving the play a happy ending; the result ruled the stage for the next century and a half. And in the 19th century symphonic conductors routinely rescored whatever they thought didn't work. Gustav Mahler added dynamic extremes and doubled numerous string and woodwind lines in Beethoven's Ninth, to compensate for the different weights and timbres of modern instruments -- which raises the question whether he'd approve the same treatment of his own music.

Nowadays musicians almost never alter a composer's notes, and there's no reason they shouldn't be solicitous of dynamic markings, phrasings, and relative tempos (i.e., when the composer asks you to speed up or slow down, you should). But absolute tempo is still what separates the performers from the composers. Conductors, in particular, have always reserved the right to march to their own drummer -- perhaps that's why a verbal tempo indication like Andante con moto ("walking quickly") or Lustig in Tempo und keck in ausdruck ("merry in tempo and saucy in expression") has as much to do with character as with speed. It's traditional for composers to complain that performers misconstrue their tempos -- that's why Beethoven, in 1817, provided metronome numbers for the eight symphonies he'd written thus far. Conductors ignored those as well.

There's no question that Beethoven had precise -- mostly fast -- notions about tempo in his symphonies. But does a piece of music have only one tempo that works? Would Beethoven have changed his mind after hearing Furtwängler, or Bernstein? In an interview that accompanies his Beethoven set, Nikolaus Harnoncourt notes, "I also have experience of composers who later changed their own markings: when they heard their own work in someone else's recording, then they not infrequently said that the tempo marking seemed more appropriate than the one they originally indicated." In a letter to Frederick Fennell concerning Fennell's recording of his A Solemn Music, Virgil Thomson writes, "Your tempo, which is more than twice as slow as the one I use, has convinced me completely. I shall adopt it from now on."

Consider the sublime Adagio of Beethoven's Ninth. Christopher Hogwood (10:44), Roger Norrington (11:08), David Zinman (11:31), John Eliot Gardiner (12:06), and Philippe Herreweghe (12:26) have proved it can be played at Beethoven's metronome mark (60 quarter notes to the minute). But every conductor has to find his own equilibrium between vertical and horizontal. Otto Klemperer (15 minutes), Claudio Abbado (17 with the Vienna Philharmonic), Sergiu Celibidache (18-1/2), Wilhelm Furtwängler (19-1/2), and Leonard Bernstein (a full 20 in his historic "An der Freiheit" performance at the fall of the Berlin Wall) sacrifice melodic cogency in the interest of delivering more harmonic information; each "discovers" a different kind of beauty.

In his Cambridge University Press monograph, Nicholas Cook argues that Beethoven's Ninth "achieved lasting success in performance only after the dual leadership system had been swept away and the modern conductor had taken over. It is, in this sense, a work of the Romantic repertory. The attempt to restore it to the Classical Period is in reality an attempt to appropriate it for the late twentieth century, in disguise." No great classical work can belong to just one period or century, just as no great classical work can belong to just one tempo -- otherwise it would be a museum piece. Schuller himself scarcely mentions the instruments of Beethoven's day (they have a clearer, rawer sound than that of a modern orchestra), except to chastise the period-instrument camp (Christopher Hogwood, John Eliot Gardiner, Roger Norrington, et al.) for their dull performances (he's mostly right). And it appears a conductor can be "compleat" without taking into consideration the usual orchestra seating arrangement of Beethoven's time, whereby the first and second violins were divided antiphonally -- in the opening bars of the Fifth this creates a stereophonic effect as those germinal four notes are passed back and forth. If the sound of this symphony doesn't have to be "authentic," why must the tempo?

IN THE END, of course, the proof -- and the music -- is in the performance. "The sad irony here," Schuller writes, "is that anyone who has not experienced a true `realization' (as opposed to an `interpretation') can have no idea what an incredible experience and revelation that can be." So . . . how revelatory is Schuller's own Fifth? And Ben Zander's? I went through The Compleat Conductor and marked my brand-new $55 Bärenreiter score as to every point at which conductors, according to Schuller, do it wrong. Then I checked that score against a some of my favorite recordings (and a few not-so-favorite): Wilhelm Furtwängler and the Berlin Philharmonic (1947, 1943, 1954); Otto Klemperer with the Vienna Symphony (1950) and the Philharmonia (1955 and 1959); Carlos Kleiber and the Vienna Philharmonic (1976); Nikolaus Harnoncourt and the Chamber Orchestra of Europe (1991); Roger Norrington and the London Classical Players (1989), Frans Brüggen and the Orchestra of the Eighteenth Century (1991), and John Eliot Gardiner and the Orchestra Révolutionnaire et Romantique (1994) all representing the period-instrument camp; and David Zinman and his Tonhalle Orchester of Zürich (1997).

For the most part, Schuller is right. Even among my test group of distinguished conductors, Beethoven is not in the details. Dynamic markings are ignored, balances are out of whack, critical lines are obscured, rhythms are not always accurate. Only the criticisms of Roger Norrington are puzzling: I don't hear that, in bar 5, the second fermata is shorter than the first (it sounds a hair longer, though not clearly longer, as it should); what follows seems reasonably close Beethoven's p marking; the timpani at bar 354 is hardly too loud, and so on -- it's as if Schuller were trying too hard to fault a recording of genuine merit.

And Schuller's own recording? It leaves the box awkwardly: the clarinets that accompany the opening four notes are as inaudible here as on every other recording, and I doubt many listeners would hear his second fermata as longer than the first. Thereafter, however, he delivers: the clarity of the oboes, clarinets, and French horns at bar 32 of the Allegro con brio; the pizzicato violins beginning at bar 49 of the Andante, and the strings at bar 81; the glorious long cello phrase that stretches from bar 114 to 130 in the Scherzo, and the long line of the first violins from bar 339 on; the alto trombone at bar 7 of the Finale, the second violins and the violas at bar 132, and the piccolo and contrabassoon throughout (the latter appears to have its own microphone, which still begs the question of whether we have Beethoven's sound picture right). The occasional slip-up -- in the Trio of the Scherzo he slips from the 96 metronome mark back to 88, and at bar 273 of the Finale the winds and brass don't really sound dolce -- merely reminds us that even a perfectionist like Schuller can't get every detail right in an actual recording.

The catch, unfortunately, is that revolutionary detail doesn't necessarily make for a revolutionary performance. Many of Schuller's nuances (like the French horns at bar 34 of the Allegro con brio and the oboes and bassoons at bar 36) barely register; without a score, you'd hardly know they're there. Where's the audience that could detect all these subtleties? Worse, the micro-managing comes at the cost of macro-meaning -- sometimes Schuller doesn't see the forest for the trees. There's no tension building in the call-and-response between strings and winds that begins at bar 196 of the Allegro con brio (try Norrington for comparison); there's no sweep in the clarinets and bassoons beginning at bar 71 of the Andante con moto, or the lower-string figurations at bar 98, and at bars 131 and 166, the winds don't swing. If precise execution of Beethoven's notation could produce the perfect performance, even Walter Murphy could be a compleat conductor: "A Fifth of Beethoven" nails the opening anacrusis of the Allegro con brio (i.e., the real beat is on the fourth note, not the first), grasps the way it moves in four-bar units, and hits Beethoven's 108 BPM right on the nose.

So I'm not yet ready to throw out my Otto Klemperer CDs. I don't need to hear that da-da-da-dah exactly as Beethoven wrote it -- I like the way Otto leans on those first four notes, creating the kind of rhythmic ambiguity you'd expect from a great jazz or pop performer like Satchmo or Sinatra (you won't find Louis's "West End Blues" or Frank's "Strangers in the Night" on any printed page). Many conductors, I grant, simply haven't understand Beethoven's rhythms; but I hear the likes of Furtwängler and Klemperer playing with them, exploring their contours. And if Otto indulges in a portentous unmarked ritardando at bars 378-382, when the "V" motif appears for the last time, it's because he understands that we're not hearing those notes the same way we did at the outset, that the movement has changed them -- and us. I like the unauthorized way Furtwängler, in his 1954 recording, brings up the French horns at bar 423 of the Allegro con brio. I like the way, in the Scherzo, the excellent Carlos Kleiber (even Schuller has kind words for him) lets the horns continue to be heard after bar 27 and again after bar 79; I like the way Roger Norrington's natural horns sound raw at bar 457 of the Allegro con brio and spooky at bar 286 of the Scherzo.

Then there's David Zinman, a journeyman conductor vaulted into the Beethoven spotlight in 1997 thanks to his Arte Nova box of all nine symphonies, with its kaleidoscopic sound, observance of the metronome markings, and flabbergasting price ($24 for the five-disc set). His Fifth is brisk, crisp, and colorful, with gorgeous details particularly in the Allegro con brio: the oboes and French horns at bar 32, the horns at 34, the oboes and bassoons at 36, the winds and horns at 110, the bassoons at 256-57 and 260-61, and at 268, the most poignant and imaginative one-bar oboe adagio I've ever heard. In the Finale, beginning at bar 234, there's even a hint of natural, unmiked contrabassoon. What's missing, along with the pizzicato violins beginning at bar 49 of the Andante, is a real sense of urgency and forward movement -- in terms of light, this performance is too much particle and not enough wave. But what character, what personality, the Tonhalle Orchester of Zürich displays, especially in its winds and French horns, and in its three-dimensional spacing. And in his Eroica, Zinman's presentation of the first 100 bars of the Marcia funebre (this has to be the passage that, in an unforgettable Peanuts strip, gives Schroeder goosebumps) does in fact start the revolution: unbelievably rapid, unbelievably romantic. If you want to check out the "new Beethoven" without cashing in your IRA, this set is for you.

Finally, Benjamin Zander. His Fifth is a call to arms, with heavy percussion and kinetic forward motion. He doesn't enjoy the kind of three-dimensional sound picture that Arte Nova has given Zinman, but here too there are winsome details: the cogent phrasing of the violas, cellos, and horns in bars 143-45 of the Allegro con brio and again in the violins (firsts and seconds divided, as they should be) in bars 151-53; the rising horns, à la Furtwängler, starting at bar 423; the Klemperer-like winds in bars 10-15 of the Andante, the pizzicato violins beginning at 49, the clarity of the string figurations at 81-85 and again at 114-120, and the model ritard, heartfelt but not disruptive, at bar 226; the Hoffmann-like "ghost" dance that begins at bar 244 of the Scherzo (no one does this better); the brass flare at bar 12 of the Finale, the clarity of the underpinning strings everywhere, and the care given to the bassoon-and-horn call-and-response at 317-21. Zander says he not only read The Compleat Conductor but consulted Schuller, and it shows. What's more, his Fifth, which is coupled with the Seventh, comes with a bonus disc in which he discusses Beethoven's tempos (you'll never hear the opening movement of the Moonlight Sonata the same way). Zander's discussion -- equal parts evangelistic fervor and refreshing humility -- would by itself be worth the $17.

The perplexing thing is that this Fifth, like Zinman's, seems almost too slow, and at times, like the second subject of the Allegro con brio, it becomes mechanical. That's the general reaction I've had to Zander's live performances of Beethoven's Third, Fifth, Seventh, and Ninth: at these fast tempos, it's hard to swing. Maybe that's why, after recording the Eroica with the Philharmonia, Zander says he asked Telarc not to release the performance. Maybe it didn't have the freedom and expressiveness of the demonstration he gave me in his Cambridge home, where his piano "performance" of sections of the Eroica had Ludwig's lightning and Satchmo's swing. If Zander -- or any conductor -- can get an orchestra to play this fast and this free, Beethoven's metronome markings will be vindicated. Ludwig would probably say it's about time. But compleat conducting -- or any kind of compleat musicmaking -- isn't about time. It's about what's in the heart, and the mind, and the soul. And the good news is, it can happen at any time.