Performance Olympics
I’ve been asked for recording advice by fellow musicians ranging from gifted 12-year-olds to young professionals to performers in mid-career. I no sooner mention recording in concert than they say it’s such a good idea!— but won’t work for them.
If you’re doing concerts anyway (I point out), it’s not much more trouble to record. If you’re streaming a concert, it’s no more trouble. In either case, why not give it a try? It’s fun! It’s exciting! It may even result in a better recording! After all, we expect a skater’s performance to peak in the Olympics, before an excited crowd, not in an empty practice rink; and it’s natural for a musician’s performance to peak in the concert hall, before an involved audience, not in an empty room.
Or so I think, with the greatest respect for the views and results of those who record in a different fashion.
Of course “nerves” are part of the picture, helping us as much as they help athletes. In both fields, we hear people say, “I was nervous; but as soon as I (exited the gate / played the first notes) I felt great!” That’s when nerves change from distraction to inspiration; when errors become less likely to occur. Contrast this with a famous producer telling me she’d love to record her artists in concert, but, “They’re not comfortable that way.” Well, my goodness! Since when is “comfortable” an optimum condition for high achievement?
Recording Styles
Concert recording, broadly speaking, aims at the excellence that comes from the excitement of performer and audience. The audience is essential, and little editing is possible (or, we hope, needed).
Studio recording, by contrast, aims to achieve excellence via editing. To provide raw material for that editing, it needs the freedom to record endless takes—sometimes hundreds! An audience would soon be bored, and bored listeners are noisy ones. If they leave, the acoustics would change and edits not match; if they stay, their presence will be a drag on the engineer-performer interaction and on the recording process (the opposite of their effect in a concert).
True, once in a blue moon, you hear of someone entering the studio and recording some gigantic piece, like the Goldberg or Diabelli Variations, in one take: no editing needed! But we can’t say this is the aim of the process; rather, no matter how welcome, it’s a very unusual occurrence!
As clearly different as they are, the two recording styles are not rigidly defined or mutually exclusive. At one extreme, you can do a pure concert recording. Or you might record in concert and come back the next day to “cover” wrong notes with short takes, as was likely done with Horowitz’s 1965 “Carnegie Hall return” album. It was released as “Horowitz Live and Unedited”; but a musician friend who heard wrong notes at that concert reported with amusement that they didn’t appear on the album.
Nearer the “studio-recording” end of the spectrum was a pianist-producer friend I ran into one day. I asked how things were going with his studio recording of another pianist, with a famous name; and he exclaimed, “I could tear my hair out! He won’t play more than four bars at a time, and wants my opinion about each take—as though you could have an opinion about four bars! And he expects me to edit this mess into a coherent performance.”
Before tape-recording existed (pre-1948 or so), we used 78 RPM records and no editing was possible: the performer had to play a whole record side—about five minutes—well enough to release—or do it again. A “now or never” psychology may have made for more exciting records. As suggested in part 2, this psychology was probably in play with “direct-to-disc” LPs.
Signs & Symptoms
To look at editing more closely, consider this illustration:
The top line represents a portion of an unedited recording with note-errors of different durations (X’s). We replace them with error-free takes (check-marks), as discussed in Part 2; and the recording is now note-perfect (second line), but with the following risks:
Discontinuities in background noise at edit-points can cause low-pitched thunks easily overlooked in editing but annoying in careful listening.
Mismatch of dynamics between an insert and the surrounding material will stand out at the edit-point. Can they be made to match? Sure! But if you shaped the dynamics differently in the insert from what you did in the main performance, what matches at the one point may not match throughout.
“Cross-phrasing”: Note that phrases 3 and 5 come from the main performance; phrase 4, mostly from the insert. Imagine the three phrases carry one continuous melody played détaché. Imagine also that in the performance from which phrase 4 is taken, your détaché was more staccato than in the other performance. Will the listener notice that the staccato-ish playing is flanked by legato-ish playing? Maybe not, if you’re lucky; on the other hand, who wishes for an insensitive listener? Or maybe the contrast in articulation will seem meaningful to the listener, though the meaning won’t be intended by you or the composer. If you’re altogether unlucky, the inconsistency may undermine the meaning of the whole piece.
What Are We Learning?
Subtleties matter, in other words; and I fear that young musicians may be learning all the wrong things about recording and performing from growing up with “perfection” that comes at a cost of problems like these. Such studio products have been, not perfected, but perfectionized. As mastering engineer Doug Sax says, “The perception of a performance as opposed to a perfect rendition is for the few and unfortunately far between.”
I would wish all musicians to learn that even the best editing puts at risk the integrity of a performance; and therefore we should minimize the amount of editing. And that’s done most easily by recording in concert.
And because so many of us agonize—”Will this performance be worthy of recording?”—which wastes time and energy, I would wish for everyone to record always. Make it not an issue! You can always not use a recording you don’t like. 
