copyright (c) 1996 by Jim Michmerhuizen

This review was first published on in March of 1996

Erich van Bruggen on CD MRS 9310012

Paul Gabler kindly sent me a tape, almost two months ago, of a two-manual pedal clavichord CD, with an invitation to review it for the group.

The catalog number is listed as "M.R.S. 9310012".  From a detail in the liner notes it appears that "M.R.S." stands for "Music Recording Service."

This is Erik van Bruggen performing, on a two-manual pedal clavichord made by Dick Verwolf.  It is patterned after an instrument made by Johann David Gerstenberg in 1760.  According to the liner notes that Mr. Gabler xeroxed and sent with the tape, it has two 8' strings per key on the manuals, and 4 strings per pedal, 2x16' and 2x8'.  There is no indication, in the brief notes or the performances themselves, that these are "stoppable" as they would be on a harpsichord.

What's on it?  Ambitious things:

Prelude in a minor          BWV 569
Trio Sonata in Eb           BWV 525
Prelude and fugue in C      BWV 545
Fugue (gigue) in G          BWV 577
Passacaglia and fugue in c  BWV 582
Fugue in d minor            BWV 539
Fugue in g minor            BWV 542

What does it sound like?

Well, forget what you thought you knew, or could have guessed, about "the sound of a clavichord".  This is a sonorous, weighty, even ponderous instrument.  I've listened to clavichord recordings by Ralph Kirkpatrick, Thurston Dart, and a few others.  Many years ago, I played one.

The a minor prelude is an early work, from that period in Bach's development when he was systematically exploring.  In this prelude, what he systematically explores is a rhythmic pattern.  Or not the pattern exactly -- _that_ he simply repeats ad infinitum -- but all the harmonic and voice-leading changes he can insert into the rhythm.  It's not at all a profound work; if there is
any period in Bach's life where he might legitimately be called "intellectual", this is it.

Erik van Bruggen is a fine and sensitive musician.  His way with this prelude is sensible: he doesn't try to wring things out of it that it could never provide.  But pieces from this period in Bach's compositional life are difficult to perform, not technically but expressively.  It can be very difficult to persuade a listener that he is hearing anything more than a well-organized little catalog of effects.  Other pieces from this period might be, for example, some of the clavier toccata movements, or the organ prelude and fugue in G major, BWV 550.

Van Bruggen succeeds in bringing the piece suitably to life, without exaggeration.  His tempos, here and elsewhere, are on the leisurely side; but the clavichord is not a sprightly instrument, and this instrument, as mentioned earlier, is positively ponderous.

Let's talk about this a little.  You know the mechanism.  It's the simplest imaginable, at the other end of the key is a little metal thingy that comes up and hits a string.  In a harpsichord, the little thingy doesn't hit the string, it plucks it.  The only other difference is the soundboard: on a clavichord it's smaller and (presumably -- I haven't made measurements or read any research) less resonant than on a harpsichord.

These two facts together make a HUGE difference in the way the whole thing sounds.  First of all, when the little thingy on the other end of the key comes up and hits the string, it also FRETS the string; that is, it actually forms one end of the vibrating length of the string.  On harpsichords and pianos, the item that supplies the initial impulse is NOT also responsible for fretting the string.  The builder is free to position it anywhere he wants to along the length of the string.  Just as in playing an acoustic guitar, it turns out that the point where you choose to pick (or hammer, or pluck) the string has a lot to do with the spectrum, especially during the first fraction of a second.

Now on a clavichord you don't have that choice.  The impulse that's going to make the string vibrate occurs exactly AT ONE END OF IT.  (I've seen guitarists play with their left hand alone, using their fingers to bang the strings down against the frets.  Clavichords bang the fret against the string.)  Of all the conventional ways to persuade a string to vibrate, this generates the noisiest attack transient.

And in most clavichord recordings, that really shows.  Many of them have little warnings, somewhere in their liner notes, about keeping the volume down because the clavichord is a delicate soft instrument.  It is, in fact, a good idea to keep the volume down, but the reason is NOT that clavichords are soft and delicate.  It is that they're NOISY: relative to the amount of sound they produce, they're the noisiest instruments you'll ever hear.  And if you turn the volume up on your recording, you will hear that noise.

Oh boy will you ever.  At fast tempos, it can almost swamp the sound.  You see, the noise is eventually converted to a sounding pitch.  "Eventually" here means "within a few hundredths of a second".  (I'm guessing, very roughly, and will welcome correction from anybody out there who really has measured this.)  In pianos and harpsichords this conversion time is much shorter, and the noise isn't nearly as bad to begin with.

But this is where much of that "ponderousness" I keep mentioning comes from. The faster you try to play a clavichord, the shorter (obviously) the notes get; and the shorter the notes get, the less chance there is for the note to "evolve" to a definite pitch out of its initial noise.

This in turn sets an upper limit on tempos.  In the gigue fugue BWV 577, van Bruggen does about mm90 for a dotted quarter.  Peter Hurford, in his organ performance, does mm118, almost one-third faster, without sounding at all rushed.  Similarly, a quick comparison of the rest of the works show van Bruggen invariably slower by anywhere from 5% to 20% than Hurford.  Not that there's anything sacred about Hurford -- his CD's lay close to hand, and his tempos seem good middle-of-the-road choices.

Overall, musical as these performances are, they don't strike me as coming from somebody who's entirely _comfortable_ with his instrument.  I listened, for a comparison, to some of Ralph Kirkpatrick's WTC performances on clavichord.  I hadn't done that for several years, and after living with van Bruggen's recording for several weeks found Kirkpatrick to be an astounding virtuoso.  He said, several times, that he preferred clavichord over harpsichord.  Certainly, his performances establish him at a level of fluency and brilliance that's just not there for van Bruggen.  But that's another essay, and a different one.

There are some interesting extra-musical qualities to the recording.  I hear environmental noises in the background: scraping chairs, birds, page-turning, and some tape hiss.  (The latter is not from Mr. Gabler's duplicating recorder: it vanishes in the gaps between movements, and returns just a fraction of a second before the music.  I deduce from this that it is on the master.)  The other noises are occasionally obtrusive but not, to my jaded and hyper-purified synthesist's ears, objectionable.  I like to know when I'm in the real world and when I'm not, and in this recording I am unquestionably in our common acoustical environment, not a studio.  (Or maybe it's just that I brought myself up on Glenn Gould.)  According to the liner notes, the recording was made at Waalse Kerk in Leiden.  The acoustical ambience -- reverb and whatnot -- suggests more an outdoor cloister or courtyard than a sanctuary.

Never mind the noises.  Sonically, this is one of the best clavichord recordings I've heard.  At normal listening levels, it is natural and realistic, and I can even turn it up little above natural volume without irritation.

In many respects these are very satisfying performances.  It's nice to be able to hear the voice-leading; many organ performances, too heavily registered, obscure this.  The Passacaglia and Fugue gathers a very satisfying momentum in its closing pages, not at all delicate.

Now, finally, let's talk about these clavichord performances as "alternative" Bach.  Let's do some idea play, with intermittent nods in the direction of Historical Fact.

Suppose we didn't have any citations showing that organists used pedal clavichords for practice.  Would this be a "deviant" performance?

Well, they did.  But did they ever PERFORM on pedal clavichords?  For audiences?  At home, for visitors?  If they didn't, then isn't this a deviant performance anyway, not because it's a pedal clavichord but because we're an audience, and van Bruggen isn't "practicing"?

Well, suppose for a moment that we're in a science-fiction "alternate history", in which organists did NOT, in the 18th century, acquire pedal clavichords for practice purposes.  That would make this performance provably counter-historical.  It would not, however, change the auditory experience by one iota; it would only place that experience in a different historical and cultural context.  In that context, performing any of these pieces on a pedal clavichord would be "radical".

But the auditory experience would still be the same.

I can't resist taking this one further step.  Suppose that the present recording is not in fact a clavichord at all?  Suppose that it was in fact entirely synthesized, right down to the chair-scraping and the birds?  What then?  

Well, I'll not prolong such twilight-zone speculations.  Paul Gabler sent me an interesting clavichord tape, and I've enjoyed it, and enjoyed telling everybody about it.

Copyright © Jim Michmerhuizen 1996