Tucked away in a footnote in Robin D. G. Kelley's Thelonious Monk: The Life and Times of an American Original is a telling anecdote that deserves serious consideration:
Although it appears to be apocryphal, a story has been circulating for many years among jazz musicians whereby Horowitz played his variation of "Tea for Two" for Art Tatum, whom he admired very much, and then when Tatum responded in kind with his variation, Horowitz was astounded. He asked Tatum, "How long did it take for you to make that up?" Tatum replied, "I don’t know, how long was it?" Horowitz found it amazing that he could improvise such an elaborate variation of the theme and, legend has it, he never played "Tea for Two" in public again. What is true is that the only recording we have of Horowitz playing "Tea for Two" took place that November 1962, and it was clearly not intended for release.
Whether or not this story is actually true, it makes an important point that extends beyond the domain of music into the nature of expertise and the behavior of those who claim to be experts.
My reading of this account begins with the premise that Horowitz saw himself as an expert pianist. That expertise included mastering the performance of major works, such as Modest Mussorgsky's Pictures at an Exhibition and Franz Liszt's B minor piano sonata. While critics may differ in their opinions of how well Horowitz performed these compositions, they would probably agree that these are the sorts of pieces that establish the credibility of a piano virtuoso. From this point of view, while Horowitz may have admired Tatum "very much," he probably also felt superior to any pianist who could not pull off a performance lasting much more than three minutes. (For the record, according to my Complete Capitol Recordings CD, Tatum's recording of "Tea for Two" for Capitol lasted three minutes and five seconds.) There is thus a good chance that Horowitz figured he could impress Tatum by "pulling a Liszt" on "Tea for Two" while putting Tatum in his place at the same time.
Instead, it was Tatum who put Horowitz in his place; and I think the lesson to be learned from this is that those who are locked into the discipline of preparing performances from notated scores simply cannot conceive that a performance can be prepared in any other way. I have no idea to what extent Tatum worked with music notation. In the course of my travels, I picked up a Belwin publication entitled The Genius of Art Tatum, which purports to be transcriptions of Tatum's solos; and it gives absolutely no indication of whether or not those transcriptions were by Tatum himself or whether they were taken by ear from his recordings. On the other hand it is clear from the evidence in Kelley's book that both Monk and Bud Powell learned to play Tatum's interpretations by ear; and the interviews that Kelley cites indicate that Monk expected his sidemen to learn his own music exactly the same way. Kelley's evidence further indicates that most of the complexity in Monk's music was initially worked out by Monk at the keyboard, again by ear. In other words notation never served any purpose other than providing a documentary record that was, at best, incidental to the practice itself.
During Sunday's Day of Exploration With Midori at the San Francisco Conservatory of Music, John Adams made a similar observation about recordings, saying something to the effect (since I cannot remember his exact words) that a recording was "just a document" that could never substitute for an actual performance. Unfortunately, he then went on to say some things about improvisation that suggested that he was no better informed about how jazz musicians can work (almost) entirely by ear than Horowitz had been. This was what first suggested to me that there was a broader lesson about expertise to be learned, which is that experts tend to fall into the trap of assuming that their own area of expertise is the only one that matters. This is a lesson that extends far beyond the domain of musical performance. As I wrote last year, "Stephen Jay Gould used to joke that cognitive scientists interested in whether or not non-human animals exhibited linguistic behavior spent too much time reading other cognitive scientists and not enough time studying animal trainers." At that time my target was economists who "assume that only other economists are worth reading." It does not matter which area of specialization you choose; this kind of narrow-mindedness is endemic to all of them!
Can it be avoided? Once upon a time there were those who believed that the purpose of a liberal education was to open the mind to the diversity of points of view, but the very idea of a liberal education has pretty much died of attrition. As a result, I see that the last time I cited this perspective was in a response to a Truthdig column by Chris Hedges entitled "The Best and the Brightest Led America Off a Cliff." As the old Monty Python sketch put it, say no more. So it does not matter whether we are talking about sitting at a keyboard to play or trying to puzzle out economic models. We have established a culture in which we approach complex problems from a position of confining ignorance, and then we wonder why we never arrive at any effective solutions!
One last thought: That 1962 recording that Horowitz made of "Tea for Two" took place over half a decade after Tatum's death in 1956!