in: News & Features

February 14, 2012

Finishing a Short-Order Cook’s Works


Robert Levin and John Harbison are on a Mozart myth-busting mission — or more accurately, an icon-busting mission. The popular image of the 18th-century genius that Levin, a Harvard professor who is a distinguished pianist and Mozart scholar, and Harbison, beloved local composer and MIT professor, are trying to counter is a spun-sugar one of pristine perfection and untouchable sublimity, tinged with the easy-listening slur which two centuries of popularity inevitably brings. On Saturday, February 25th at 8 pm at MIT’s Kresge Auditorium, the two longtime collaborators will engage with Mozart in the most fundamental way: by performing Levin’s completions of three of Mozart’s unfinished works. The “Mozart Marathon,” which is free and open to the public, will include the Divertimento in B-flat Major for Horns and Strings, K. 287; the Trio for Violin, Cello and Piano, K. 442; and the Piano Concerto in D Major, K. 537, as well as a public discussion with Levin and Harbison of the ins and outs of collaborating with their centuries-displaced colleague.

Why tackle the intimidating challenge of creating more Mozart?

Robert Levin (BMInt staff photo)

Levin points out that Mozart left behind the largest proportion of unfinished works of any composer: a staggering one in five of his extant manuscripts were abandoned at some point, and in his later years the ration is nearer to one in three. “Access to work of a genius is bound to be enlightening …there’s always an intriguing enigma about something that stops somewhere in the course of a journey. Completions enable an audience to take acquaintance with pieces they wouldn’t have heard otherwise. It’s like reading a short story that breaks off in the middle. No one wants to read it if they know they won’t be able to find out the ending.” Harbison says that their pre-concert discussion will “try to get the people who are there engaged in the interesting question of this large body of work — why this happens and what’s interesting about it. Some of [the unfinished works] are in genres that are rarely visited; for example, there’s an extraordinary concerto for piano, violin, and orchestra. What an unusual combination!”

So completing what Mozart didn’t — isn’t that heresy? If the composer himself saw fit to abandon a project, then who is Mr. Levin to naysay him? “That’s sort of the cynical view,” Levin says, “that a composer abandons work on something that’s not sufficiently good.” There’s another reason for abandonment, however, and one that’s much more practical. “Basically, Mozart composed in a very economical way, because he was a short-order cook, constantly dependent on sources of money. If he was puttering along with a string quartet, and someone knocked on his door with the offer of an opera commission, he had to abandon the string quartet…. He’s constantly juggling things in his head, so he develops a way of putting things down that is as economical and as swift as possible, allowing him to come back in six months or six years.” Levin points to the strong probability that Mozart meant to finish these abandoned works and cites the pioneering work of British musicologist Alan Tyson as piquing his early interest in Mozart completions. Tyson engaged in CSI-like detective activities, examining Mozart’s papers for indicators such as watermarks and tints of ink; his discoveries resulted in the re-dating of some of the composer’s works. “Numerous pieces that were completed were fragments on his workbench for as long as 10 years,” says Levin. “This suggests that of the remaining fragments, had Mozart lived longer he would have taken them up and completed them.”

Tyson’s investigations also offer some suggestions as to the process by which one might go about attempting to tie up Mozart’s loose ends. “Ink was sold in small quantities at the chemist’s and varied from batch to batch,” Levin explained. “You can see in many Mozart manuscripts as many as four tints of ink on every page.” In an unfinished piece “you can follow his train of thought. When he had a concept crucial to the sound of a piece, all instruments are notated, but elsewhere you can follow the melody like a birdie in a badminton game. There is always something going on somewhere, but in between there are empty places.” So analyzing Mozart’s works in various stages of completion gives insight into his compositional process: “We are able to recreate the entire process of creation — a criminological kind of activity. And every fragment has a different kind of chronology.” The next question, Levin says, becomes “do you want to draft the rest of the piece as he would have done it, or do you want to finish off the missing fragment and then finish the piece? In general, I like working the way Mozart worked.”

What happens when you get down to nuts and bolts, piano and computer keys, though? Both Levin and Harbison stress the dual challenges of structure and detail in Mozart’s music. Levin starts by staring at the unfinished score for great lengths of time and hearing the music in his inner ear. “It’s not for the amateur. It requires a complete immersion in language, vocabulary, narrative style. You have to understand a great deal of his procedures, and then you have to take a risk. You have to know how much of the piece, architecturally, is there. You have to grasp an idea of size and structure.” With his work on the “Coronation” Concerto, which he will perform on the 25th with a reduced string quartet accompaniment, Levin would only test his work at the piano after this rigorous thought process; “sometimes taking it to the piano and trying it out is a sobering thing to do.”

Levin describes Harbison, his colleague of 40-plus years, as the “person that I most consistently take the pieces to for advice.” Harbison, who himself dabbled in Mozart completions during his undergraduate years, says that “the big issues are proportions — how long something should go on, where the key areas ought to go — and then there are a number of questions of detail, which are quite interesting. Mozart is an extremely attentive composer to minute details like voice leading. Often the questions we ask are ones like ‘is this the best voice leading here?’ and ‘how should we lead up to this cadence?’ ” As a composer, Harbison is fascinated as well by the way that Mozart’s pieces came into existence. “He seems to work in serial form, taking them to a certain point, breaking them off and completing them later.”

One can pore over Mozart’s manuscripts and voice-lead until the end of time, but a major point that both Levin and Harbison make is that in Mozart’s music, all is not preconceived. Levin is well known as a champion of the improvised cadenza, and Harbison is a noted jazz musician. Both see the relationship between the methodology of composition and the spontaneity of performance as crucial and mutually enriching. Levin describes himself as primarily a performer, not a composer, and he says that one of the main effects that working on completions has had on him is the realization, through the constant choices he makes in composition, of the constant choices he has onstage as a pianist. “The sense of volatility of the language and the different possibilities has a fantastic effect on performing the music. At any point there are options.” Instead of viewing the music as perfect or unalterable — a criticism he has long leveled at the recording industry — Levin says that, after working on a completion, “I start to play the piece in a much more risky and unsettled way, aware of the options that went through Mozart’s head at any point in the piece. I realize things that are lurking below the surface.”

Harbison takes up the viola for the performance on the 25th, playing in the Divertimento and in the accompanying quartet for the Coronation Concerto. He joins Mozart in the distinguished club of violist-composers. (Is there a special significance here?) Harbison agrees that Mozart holds a special place in violists’ hearts because “he wrote a lot of great viola notes, which sort of fits with his particular acuteness of part-writing,” but submits that Mozart’s instrumental choices as a player raise a larger issue. “It’s been clearly documented that these two very flashy divertimentos [of which the B-flat is the second; Harbison’s wife Rose Mary will take over the soloistic violin part] were sort of his final words as a violin virtuoso… He wrote this piece as a demonstration of just what a fancy player he was.” Mozart then turned to the viola as his main instrument for the latter part of his career, playing in a string quartet of which Haydn was first violinist. “In general, it’s a reminder that composers played and performed in public a lot, which is currently not a norm — a large change in the role of composers since Mozart’s time.”

John Harbison (BMInt staff photo)

The fact that Mozart was a performer meant that he felt no need to set every detail of his works in stone, especially if he were to be the one performing them. Says Levin, “when pieces became marketable commodities, that is when he wrote down cadenzas on separate pieces of paper.” Harbison points the interested musician to Mozart’s letters to his sister, which include suggestions in varying degrees of outrageousness as to embellishments she might make in performing his piano concerti.

The topics of embellishment and improvisation inevitably bring to mind Harbison’s experience as a jazz performer — and he does indeed see a connection between the role of improvisation in 20th– as well as 18th-century music. He describes improvisation as a “skill or element of music that diminished as the 19th-century came on, and American jazz was a rebirth of that impulse in a way.” Levin also brings up the rise of aleatoric music within the more academic establishment in the 20th century, which he describes as “the same challenge that you can bring to music that is 250 years old. But with Mozart, most tend not to think about the flexibility of the language, but only the topiary matter of polishing the surface.” Says Harbison, “one of the pleasures of accompanying when Bob is playing is that it’s not the same experience each time. I feel strongly about those kinds of connections between musicians.” And although he points out that jazz and Mozart have “different vocabularies and different boundaries,…the possibilities of those kinds of performance situations are exhilarating.” Harbison describes first hearing the recordings of Viennese pianist Friedrich Gulda in the 50s; the eccentric but admired Gulda had a long career as both a classical and jazz musician and, coincidentally, was the first to record improvised cadenzas to Mozart’s concerti.

When I ask Harbison what, as a composer, he considers to be “completion,” he again brings up jazz. He says that the composition process, for him, is kind of like “later performances of standards by Thelonious Monk, which would become less and less of improvisation and more of a composition. That’s the way I feel composition happens — better versions serve a certain self.” It seems likely that the beauty and satisfaction of Mozart’s music might also be served in a similar way, with continued experimentation, examination, and performance. Interested parties should go investigate on February 25th. Levin, Harbison, and Rose Mary Harbison will be joined by Heidi Braun-Hill, violin; Rhonda Rider, cello; Liz Foulser, bass; and Richard Menaul and Alyssa Daly, horns.

See related review here.

Zoe Kemmerling is a recent graduate of the Boston Conservatory and a freelance violist, baroque violinist, writer, and string instructor.


  1. When John Harbison says, “In general, it’s a reminder that composers played and performed in public a lot, which is currently not a norm — a large change in the role of composers since Mozart’s time,”  of course he’s not talking about rock bands beginning with the Beatles. I wonder if there is a relation between the composer-performer dichotomy and the composition — by Beethoven and his successors — of works which are beyond the reach of the amateur musician.

    Comment by Joe Whipple — February 16, 2012 at 12:57 pm

  2. BTW, I can’t be there on the 25th, but I’d love to hear the completed works and what Levin and Harbison have to say.

    Hopefully Benjamin Roe will send out a crew to record it for later broadcast on CNE.

    Comment by Joe Whipple — February 16, 2012 at 1:02 pm

  3. I guess that I’m one of a small number of people who saw the bio-pic “Copying Beethoven”.  I even wonder how many people saw it in theaters thinking it was another in the succession of Beethoven-the-St. Bernard dog pictures.  One of the scenes in the film showed Beethoven (played very well by Ed Harris who also played an astronaut in the film version of  “The Right Stuff”) walking into his publisher’s office with a new score, and the exasperation of that publisher with LvB’s subsequent compositions after such initial popular successes as the “Moonlight” sonata and the Septet.  It’s amazing that the “Hammerklavier” sonata got published at all!

    Comment by Laurence Glavin — February 21, 2012 at 5:42 pm

RSS feed for comments on this post.

Sorry, this comment forum is now closed.