Conlon Nancarrow, like all artists interesting to read about, was a fount of idiosyncracies. One was the tendency to bring out earlier music, often abandoned works from his early years, as brand new music. His most spectacular instance of this was renumbering his Player Piano Studies Nos. 38 and 39 as Nos. 41 and 48 because he was using them to fulfill commissions, and didn't want his patrons to know that they were paying for works that had already been completed prior to the commission. (No shame in this, by the way; Stravinsky did it as a matter of course, and recommended the practice.) As a result, in the complicated numbering of Nancarrow's 50-odd studies, Nos. 38 and 39 do not exist.

Now this personal quirk is occasioning some interesting wrinkles in Nancarrow scholarship, and I've spent the week unraveling some of the mysteries I wrote about last week. First of all, the Three Movements for Chamber Orchestra, his last work, since it's receiving it's US premiere with Alan Pierson and the Alarm Will Sound ensemble this Saturday, Feburary 19, at Columbia University's Miller Theater. Around 1993 Conlon received a commission from New York's Parnassus ensemble. Having already suffered a stroke, and not feeling up to conceiving a major work from scratch, he was rumored to have gone through some of his abandoned player piano works (of which there are dozens) and orchestrated them. Now, thanks to les Alarm Will Sound has sent me, I've been able to confirm this.

Trimpin labeled all the unknown piano rolls alphabetically, A through Z, then AA through ZZ, up through BBB. Some of these are mere sketches or jokes, some apparently early versions of studies you're familiar with, others entirely completed works he abandoned for some reason, and one of them remarkably Romantic in tonality, kind of Lisztian. [I have since learned from Trimpin that this roll was by another composer who sent it to Conlon.] The Three Movements for Chamber Orchestra is based on three separate rolls, marked R, AA 39 A (because either Trimpin or I once thought this had something to do with #39/48), and "UK Finished A," so labeled by Trimpin - "UK" meaning unknown, and "Finished" to indicate that it was clearly a completed work. (If memory serves, this last roll is one of the ones Trimpin presented at the Kitchen in 1993.)

R is only a rhythmic canon using five pitches, and might have been presumed to have been written for Conlon's abortive experiments with a percussion machine in the 1950s; however, I'm not so sure of that, because the five pitches are spaced out at octaves, and the percussion rolls seem to used a chromatic scale over a much smaller range. Alan Pierson confirms that the first movement is only for percussion, and that the tempo relationships (which I couldn't quite figure out from the piano roll) are 75:96:105:120:126. This doesn't quite correspond to the roll, which seems to be 75:84:96:105:120; some change must have been made while arranging. Also, this is a remarkably complex ratio for Conlon, the kind of number series he used in his last few works, but never in his early music. I wonder if that suggests a more recent time period.

The second movement seems pretty literally based on AA 39A. The third starts off based on UK Finished A, but some notes are missing and altered, and I'm still curious as to how this differs from the original. UK Finished A contains a large canon, but one somewhat obscured by an additive process in which chromatic pitch gestures repeated over and over with one more pitch added each time. I've put up a MIDI version of UK Finished A here, if you'd like to listen and compare it with the third of the Three Movements being premiered this weekend.

As for the named earlier works for player piano cited by Helena Bugallo in her dissertation, she's kindly clarified their provenance for me. The Didactic Studies are all different versions of Study #2a, simply the same tune and ostinatos but with a variety of different tempo relationships. I had found these scores and piano rolls in Nancarrow's studio, and wrote about them, and they are almost certainly a product of the 1950s when he was first starting out. But apparently in a 1980 interview he referred to them by the title Didactic Studies and avowed an intention of publishing them as a set. Perhaps he really did return to this piece as late as 1980, but it seems odd. I think he had abandoned any such intention by 1988, because I looked at those scores then with his supervision, and he said no such thing. The other work, For Ligeti, was apparently presented publicly in 1988 (though he also never mentioned this to me either). This, according to Sacher Foundation archivist Felix Meyer, has been found to be an early study originally intended as Study #3, but withdrawn. I'm eager to get more information.

In addition, two readers wrote to inform me that, while several of the recordings of the Canons for Ursula contain only two canons, there are two recordings that include the third, or rather middle, canon. One was a 1996 recording by Joanna MacGregor which may soon be rereleased on her own label. The other is the brand-new Wergo recording by Helena Bugallo herself, on which, in addition to some player piano study transcriptions that she plays with her duo partner Amy Williams, she also plays all three canons. Sorry I missed this fact.

There are doubtless new Nancarrow works yet to come to light; a few years ago at a Nancarrow conference in Mexico City someone showed me a couple of brief 1940s piano pieces found among his papers, and I myself had discovered a movement for large orchestra apparently intended for an expanded version of his Piece No. 1 for Small Orchestra from about 1942. But I suspect most of what we have to look forward to comes from the piano rolls, and perhaps I'll have time to get some of my MIDI files of them into performance shape.

Copyright 2005 by Kyle Gann

Return to the Kyle Gann Home Page

return to the home page