Recording Notes III
Before my Frescobaldi recording sessions began, I found my thoughts drifting to the golf course, as they often do. For better or worse, we had scheduled the sessions for a time of year when being out on the golf course is at its nicest: early spring, not too hot or cold, no fear of snow or hurricanes. This is not just coincidence. That kind of weather is also best for location recording. Our venue is an old, modernized barn out in the country. It has thick walls—pretty good for keeping out noise and for keeping the climate stable, but not as perfect for either as an actual studio might be. It’s nice to record at a time when we don’t need heat or air conditioning—when, if we turn off the refrigerator in this barn to get rid of its noise, the items in it won’t heat up too promptly, with a good chance there won’t be thunder, and so on. But though it’s nice to be recording then, it’s also tempting not to be recording, not to be indoors.
So back to the golf course. It occurs to me that the concept of the “mulligan” has something to say about the recording process. For those who don’t know, a mulligan is a shot that does something that the golfer doesn’t like—gets the golf ball into a bad situation or causes the ball to be lost altogether—and that the player then decides not to count. It is a “do-over.” It is a violation of the formal rules, a form of cheating, strictly speaking. It is also a common informal practice and one that isn’t necessarily unethical—isn’t really cheating if you have agreed with anyone against whom you are playing that you will let one another do it. I have never allowed myself to take mulligans, not because I am temperamentally devoted to rules—which I am not—but because I find that a commitment to counting every shot helps me focus on my shot-making in the way that I want and strive to do. A little voice in my head telling me that if I don’t like the shot I won’t count it would undermine what I am trying to work on when I play golf.
So, with its theoretically endless possibilities for editing, is the recording process a cornucopia of mulligans? There’s always a chance for a do-over. As I wrote back in June, editing is a defining characteristic of recording. Endless mulligans without penalty. Can I learn anything about performance during recording sessions and about the editing process from my feeling about mulligans or from thinking about what the differences are between those situations? The difference is this: that the legacy of any one trip through the golf course or through a particular hole is the awareness and then the memory of what happened. The legacy of the recording process is the artifact that results from it. If I try a certain golf shot a dozen times and lose the ball on the first eleven, only to finally get it right on the twelfth, then that story is the story of what happened, regardless of whether or not I call most of those shots mulligans and don’t write the big number of strokes on my scorecard. If I try a passage in a piece that I am recording a dozen times and only get it the way I want it on the last of those times, then I have still fully succeeded in getting it the way I want it. No compromise, and no one needs to know how long it took! If I am afraid that on the golf course a willingness to take mulligans would make it hard for me to focus and concentrate the way I want, then in the recording studio, where repeated takes and editing possibilities are useful, good, and necessary, then I must be sure that the opportunity to try things over and over again doesn’t also make me lose focus.
I will now turn to my daily notes.
Day 1
I am heading to a rural spot in Bucks County, Pennsylvania, about an hour from my house, where I will record ninety minutes or so of Frescobaldi keyboard music on a seventeenth-century Italian harpsichord. The venue is the same one where I recorded The Art of the Fugue in a version for two harpsichords with my colleague and former student George Hazelrigg several years ago. George is serving as recording engineer and producer for this project, and he will also help with the tuning and general care of the harpsichord during the sessions.
The instrument is in my car on this drive, and we will begin by getting it inside and setting everything up. Since this instrument is so old and utterly irreplaceable, I am glad it is not raining. It is out of the question to get any rain on this harpsichord, even at the cost of delaying the start of things. I am not saying that I need to be especially careful with this instrument on the grounds that it is particularly fragile. It really isn’t: it is remarkably sturdy and stable. It’s important that it not suffer any damage or stress.
Moving the harpsichord inside goes very smoothly. Since we have used this space before, we have a sense of where it might be best acoustically and logistically to place the harpsichord. The process of setting up recording equipment and generally preparing the room—which is mostly about noise, turning off appliances, and so on—serves to give the instrument time to relax and get used to the room. This is a very stable harpsichord and doesn’t seem to change at all from being driven about or from being placed in a new space. I’m not talking about tuning: every harpsichord is perpetually going out of tune and needing to be tuned, whether it is being moved or not.
Microphone placement: this is important, but until the day’s work was over, I had no idea how important. To make a long story less long, it took the whole day. To start with, we placed the microphones that we had decided to use in a position that made sense. (I wanted the sound to be a bit less dry than the sound of The Art of the Fugue recording, so our starting point had the microphones a bit farther out from the instrument than what we had done in that session.) I played a bit, and we listened to the results. It sounded good, but maybe not quite exactly what we wanted. We moved the mics around, changing distance and angle, playing, listening, conferring. After a while the more professional recording-oriented ears in the room began to feel that what we felt lacking in these various samples was caused by an over-sensitivity of the microphones to a somewhat bass-heavy quality in the room. So we switched to a pair of microphones that we had earlier considered for the project, but initially decided against. We then spent another long while placing them in many different ways. We even tried unconventional placements in which the mics were quite far apart or on different sides of the instrument.
This was all fascinating and fun. It gave me a chance to practice the same short passage over and over again and to get used to playing in that room. It also allowed the nervous side of my character to fantasize that this would actually go on forever, and that I would never have to buckle down to the business of playing and recording for real.
However, the most interesting thing was this: the instrument sounded really different depending on which microphones we used and how we placed them. One of the criteria that we used in trying to decide what was best was how much we thought that what we heard through the speakers with each attempt reminded us of what the instrument sounded like in person. (And I think that in the end we did a good job of that.) But it was made vivid to me that the recorded sound of an instrument is in part a construct, not an objective fact. This ties in with the set of questions about whether a recording is a document or “record” of something in the world outside that recording or an object of its own, an artistic artifact to be understood on its own terms. There is no conflict between these things philosophically for me with respect to this project in particular, since the authentic, accurate sound of this particular harpsichord is magnificent. I take it for granted that the more accurately we represent the sound, the better it is likely to come across artistically for most listeners. But it is quite telling to be reminded and to have it demonstrated so vividly that the sound as it comes across on the recording is something that we have very purposefully constructed and taken quite a while to construct.
Day 2
This day I am thinking about the nature of the seclusion that I need to record. When I am teaching or writing, I can and do take breaks in which I interact with the world. In between lessons I will check my phone or, if there is enough time, go for a walk or do an errand, or whatever presents itself. If someone I know is present in another part of the church where the Princeton Early Keyboard Center studio is located, I might go chat. In between bursts of writing a column I will orient my computer to the outside world and check the news, my e-mail, or Facebook. These breaks recharge my concentration, and I never have any trouble shifting back from such things to the focus on teaching or on writing. For the recording project, however, I find that I want to feel sequestered or secluded. I want the hour of driving to the venue to feel like it is taking me away from all of the interactive electronics and even from interactions with people. I want to be heading towards a sort of cloister. I am not sure why this difference exists, but I feel it very vividly.
I have not brought my computer, and I find that I can detach myself enough from the world to restrict checking my phone to about twice during the seven-hour recording session. No one has called.
Today we are doing the pieces that I want to play on two 8′ stops. This instrument has two 8s, like most Italian harpsichords. However, there is no working stop mechanism. In concert it is effectively impossible to change stops. Every time that I have used this instrument in live performance I have had to decide in advance whether to use both 8s or only one of them for the whole event. For the recording we can remove and replace jacks between pieces. Of course for efficiency we don’t want to do that more often than is necessary. So part of my preparation has been to decide in advance on the registration for each piece and to use that in determining recording order: 2x8′ pieces first, then the pieces for one of the 8′ stops, then the pieces for the other. This does misrepresent what the instrument can and can’t do. It does not misrepresent what an instrument of this sort might possibly do, or what this instrument could be made to do if I were willing to alter it. This ties in with the questions about the documentary nature of a project like this.
I feel that I have done a medium-level job at best of relaxing while I play. I am trying too hard to get everything right: notes, ornaments, and interpretive gestures. I am letting a focus on those things cause me to tense up a bit. I think that this becomes less of a problem with successive takes of the same passage. Will the later takes indeed tend to sound the best? Will this slight tension also get better over the longer scale on successive days?
Day 3
On the way to the sessions today I am thinking about what to listen to in the car. Is there anything I can do with the car sound system that will get me to the barn in the right frame of mind to play? Probably not the radio (in keeping with the cloister idea). Music? What kind? Feeling unsure about that, I am trying the nothing solution, leaving the sound off and trying to hear my own music in my head.
I am thinking about a question that is of importance to the outcome of this whole project: the matter of evenness of voicing. Should every note feel and sound the same as to dynamics? It is a theory of mine that, especially with harpsichords that have a crisp and robust sound as this one does, a bit of unevenness in voicing is actually good, that it gives extra life and fluidity to the overall effect. It has to be kept within certain bounds: there is such a thing as too uneven. But I like more unevenness than some people would. I have definitely decided to treat the instrument this way for this project. (I did the voicing and regulation myself.) Essentially I voice the instrument to the point where all it needs is a final refinement. Then I don’t do that refinement. How will that sound for the microphones? There certainly won’t be time to make wholesale changes. That is the approach to which we are committed.
Day 4
Continuing about voicing: when I am going to play on the two 8′ stops together I like to achieve the net sound and feel of each note with somewhat of a random difference in the balance of the two stops. That is, one note may be 55% one stop and 45% the other, another note may be 52%/48%, or exactly even. This creates a pleasing variety of color up and down the keyboard for the 2x8′ sound. However, as we take the jacks of one 8′ stop out to record on the other, I am reminded that for use alone that stop is a bit too uneven, even for me. So a small amount of voicing is needed. I have to be quite aware of getting back into the sequestered player mode once I have dealt with the voicing.
At one point today, April 21, 2016, George looks up from his computer and tells me that Prince has died. The world is there regardless of how sequestered I want to feel. The thought comes to me that he is another person about whose work I will probably find out more now that he is gone than I knew while he was alive. (I certainly had no hostility towards his work. I just had never happened to get into it much.) And that puts me in a similar relationship to him that I am in perforce with the Baroque composers whose music I play.
Today I suspect that a couple of the pieces will work as single takes. However, that brings to mind an old question. It seems fine to me, normal, probably necessary, to use editing to lower the level of tension during sessions, so that any little glitch need not feel like an emergency. However, how should editing then be used to create the final result? If I have a take that went well and that I like, should I look for other takes that have even more effective versions of some passages and build up a sort of super-performance made up of the very best bits of different takes? Or should I just use editing to get around real problems, but fundamentally let performances be what they were on the day as much as possible? Is one of those better philosophically? Will one of them lead to a better finished product?
Day 5
We have had a lot of noise over the last few days, mostly airplanes. That, plus other issues like time spent tuning and the decision to do lots of takes—partly of necessity when things have gone wrong, partly for safety and to provide more choice—is going to lead us to leave out a couple of short pieces. My starting list was very long: much too long for a CD, and about 50% more music than I have recorded for any of my previous projects. The remaining music is still too much for a CD.
Today I feel noticeably more relaxed, better able to play as I would in a normal concert for which I felt well prepared. This has probably been sneaking up on me throughout the week. I am presumably noticing it because we are almost done. I would quite like to go back and re-do the rest of it with this feeling. Will pieces recorded today sound more relaxed?
Conclusion
There is no conclusion. That’s all for now about this experience. I will edit the pieces over the summer, and we will see what sort of dissemination seems best. You can find a couple of pieces (lightly edited, not necessarily identical to what I will consider the final release) at https://www.youtube.com/watch?v=ELHjLVh1hlk.
Next month I will return to the directly pedagogic!
