Today was the first of three days that violinist Stefan Jackiw and I will spend recording the 3 Brahms Sonatas. We are making the recording at SUNY Purchase (about 45 minutes outside New York City). The CD is supposed to be released first in Korea (the label is Sony) before December, and then - we hope - in other countries after that.
It has been quite a while since I've made a recording, and it has given me cause to reflect on the recording process. I think some people have a kind of ethical objection to the fact that recordings can be "spliced" together, unlike the live performances for which the Brahms Sonatas were presumably intended. I don't have this objection to the recording process - in this I am much influenced by Glenn Gould, who famously gave up concertizing entirely and instead focused only on making records. Gould (who wrote quite a lot, in a witty and sometimes obtuse way) expressed his belief that the best service we could pay both to composer and listener was to put forth the "ideal" performance of a piece, a performance which only could be achieved by eliminating "accidents" like wrong notes or other performance flaws, and also eliminating extraneous issues like audience members coughing, etc.
The question arises of how to recreate the sense of spontaneity and excitement that generally (but not always) comes in a live performance. In a way, you can actually play with *more* abandon in a recording session, since any flaws or rough spots can be corrected in later takes. On the other hand, something resembling perfection is expected in a recording, so it is hard not to listen to oneself very critically, under an aural microscope. This can inhibit being expressive, but fortunately our wonderfully helpful producer, Steve Epstein, reminded us on a few occasions just to play the music, and not to worry about the details. (In fact worrying about the details can, in many cases, cause problems - it's like a baseball player who starts "overthinking" his swing).
The other problem in making a recording can be the lack of continuity in a performance. What I mean by that is that when we perform live, we hear the thread of a piece continuing from first note to last, developing, telling a story, etc. At least, that is what we are trying to do. In a CD, all the starting and stopping *can* mean that we lose a sense of the whole. What Steve had us doing today was about four complete takes for each movement (we did the 3rd Sonata today - the plan is to do #1 tomorrow, and #2 the next day). Then we were "covered", so to speak, for 99% of the piece in terms of accidental wrong notes, squeaks, notes that didn't come out, bad voicing, balance problems, etc. Only after that would we spend some time on the remaining short passages that needed to be "fixed" in some way. I believe that what he will do is base the final recording on one complete take, with various splices from other takes - but the basis is one cohesive "performance." That helps, I think, to alleviate the potential choppiness of a recording session.
Naturally one of the complications in making a recording with 2 people (I can only imagine a whole orchestra!) is that we have to both be playing our best at the same time. I can see why in some kinds of music, musicians are recorded in separate rooms, hearing each other on headphones, so that one person's problem does not create problems for everyone else. We are not quite so "artificial" in that sense - that would be a little like those terrible scenes in the Star Wars movie "Phantom Menace" where the live actors had to interact with the CGI "actor" Jar Jar Binks - it seems artificial because it was artificial.
Listening to recordings is also notably different from listening to live performances, and it does affect our approach to making a recording, I think. In a live performance, audience members give full attention to the performer, for 30 minutes or 2 hours or however long the performance is. We naturally forgive certain flaws, up to a certain point at least, in exchange for the thrill of a live performance (a part of which has some resemblance to a tight-rope act). On a recording, we expect perfection, basically, and also I personally almost never sit and listen to a whole CD, say, of the Brahms Violin Sonatas, from beginning to end. I might listen to a movement, or maybe 2. Or else I will listen to it in the car, where I am frankly not able to concentrate on every measure as I would in a live performance (since after all I am trying to drive my car). Many people listen to music while exercising or doing the laundry or as background music. While I certainly wish people would give my recordings their full and undivided attention, I know that the probably won't - at least not all the time.
The one way in which this has affected our musical approach is that we don't worry as much about ending each piece with a "bang" so to speak - we don't need to encourage applause (not an issue with the 3rd Sonata, but the 1st Sonata, for example, doesn't end with a grand gesture that tells the audience to start clapping). We also don't need to try, necessarily, to highlight the differences in each Sonata (which we definitely did in the various live performances we gave over the last year of the 3 Sonatas played in one concert). This is because, as I said, people are not so likely to listen to it as an "album" (and in these days of Itunes, people may not even download all three Sonatas anyway!).
This brings me back to Glenn Gould's assertion that recording was somehow more "pure" than a live performance - we are playing each piece in the way we believe it ought to be played, without regard for any other issues (like "wowing" the audience in a superficial way - I seem to remember that Gould wrote an article entitled "Let's ban applause", but I know I wouldn't want that to happen in live performance!).
Brahms was born in time to see the very beginning stages of the recording business - there is a recording circulating of Brahms playing the piano (http://www-ccrma.stanford.edu/~brg/brahms2.html) since Edison had just invented the technology to record sound. I doubt that he would have foreseen the technology that would allow the recording process as we know now it, or the remarkable way in which his music can now be disseminated to the listening public. But I do remember a quote attributed to Brahms - he said the best performances he ever heard of his music were the ones he heard sitting in his comfortable chair, with the score in his hand, *imagining* the music in his mind. Perhaps we can do similarly by taking a good recording of his music, sitting in a comfy chair and putting on the headphones.