[Note: This post has been edited for clarity, and to correct certain infelicities of expression as of 2:19 AM Eastern on 27 Nov.]
I've now had a chance to audition properly the Sergey Schepkin reading of Bach's Goldberg Variations, and that reading plays perfectly into the more general post I'd begun writing concerning Bach's keyboard works performed on the modern piano, which post I stayed when I determined to first give this reading a listen before committing my general thoughts to print. That intended more general post is now rendered superfluous as I can make all my principal points writing about this one reading alone.*
As I noted here, from the Amazon clips I'd listened to, Schepkin's reading promised to be superb. As it turned out, it's merely yet another thoughtful reading of the Goldberg notwithstanding that, throughout, the reading displays a secure and highly accomplished pianism, and an often pianistic awareness of the possibilities and limitations of Bach's original instrument; an awareness largely disregarded in most pianists' readings of this work. My snap observation in my above referenced squib that the clips I'd heard indicated Schepkin owed an obvious debt to Glenn Gould was borne out by my audition of the complete reading, and was, I'm pleased to report, even corroborated by Schepkin himself in the enclosed liner notes. Despite that debt, however, I found this reading to be an ultimately disappointing one.
First, it's my feeling this reading would have benefited greatly had Schepkin not chosen to take all the repeats (or any of the repeats, for that matter); a practice that makes little sense for any modern performer of this work as those repeats are there not because the music demands them to complete or fulfill itself musically, but because the repeats provided the contemporary performer of the work the occasion for applying at will his own invented embellishment** as was the common, even obligatory, practice of the time, most especially when the performer was a genuine keyboard virtuoso as was the young Johann Gottlieb Goldberg for whom the Variations were written. As the great Donald Francis Tovey, writing in the 1930's, put it concerning the repeats in the Goldberg (and I paraphrase a bit only because I cannot immediately lay my hands on the source for the verbatim quote), "Anyone who today takes all the repeats in the Goldberg is but an academic pedant." To which I say, Amen.
Needless to say or rather, it ought to be needless to say no modern performer, no matter how much the virtuoso, or how much the scholar, has either the Baroque sensibility, the Baroque training, or immersion in the living Baroque culture and tradition necessary to permit him to supply his own invented embellishment at will and have it sound as authentic Baroque ears would have expected it to sound (especially true of one Schepkin-invented embellishment, several times used, that sounds like the musical equivalent of a hiccup). Consequently, any self-invented embellishment freely applied by a modern performer (and Schepkin does so in profusion, and at times applies his invented embellishment in the first statement as well as the repeat) is little more than a greater or lesser informed conceit on his part, and a practice that ought not to be tolerated, much less encouraged.
So, beyond the business of the repeats and the self-invented embellishment, what was it I found so ultimately disappointing about Schepkin's reading? The very thing that makes all readings of my experience of this work performed on the modern piano ultimately disappointing, Gould's 1955 reading most singularly excepted: the performing pianists simply cannot forget they're pianists as they must when performing any Baroque keyboard work by Bach especially, and the Goldberg most especially of all.
And by that I do not mean they should forget how to work their instrument as pianists; indeed, all their pianistic skills will be called upon in extreme measure in order to overcome the impediment of their instrument in realizing this music as it needs, even demands, to be realized. What these pianists must, but seemingly cannot, forget as pianists are the various standard pianistic techniques employed to produce what has universally come to be accepted as beautiful and expressive piano playing and sonority, among them the techniques of the pianistic legato and cantabile achieved by fingering alone, or in combination with, or alone by, discrete use of pedal and a certain lightly tripping staccato; effects impossible on the instrument for which these variations were written, and therefore inimical to this music.
But perhaps the most damaging pianistic device of all one universally employed by all pianists for all keyboard music, the keyboard music of the Baroque not excluded (again, Gould singularly excepted) is the pianistic realization of the notion that all music is made up of melody and accompaniment, with the melody always expressed, to greater or lesser degree, in some measure of relief in terms of loud-soft (forte-piano) dynamics. No pianistic device is more destructive fundamentally destructive of the essential contrapuntal structure, the polyphony, of any Bach keyboard work. In short, it's musically the very kiss of death for a Bach keyboard work, and nothing, no matter how otherwise salutary, can overcome or compensate for its employment (or, rather, misemployment).
Where the Goldberg is concerned, one cannot argue, as one can with some plausibility argue with, say, the preludes and fugues of the Well Tempered Clavier, that Bach wrote the work to be performed on any keyboard instrument, presumably, by extension, including even the unknown to him modern piano; wrote it to be performed on the feeble-voiced clavichord (an instrument thought by some to be Bach's favorite) with its possibilities for a limited range of forte-piano dynamic and true legato, as much as he wrote it to be performed on the harpsichord. The argument fails with the Goldberg as not only was the work written for the harpsichord specifically, but for a double-manual harpsichord in particular. That means the harpsichord, with all its possibilities and limitations, actually shaped to a meaningful degree the very structure of the music itself. And it does no good to argue an argument typically attempted as justification for the pianistic excesses of most modern pianists that had Bach access to a modern piano he would have utilized all its possibilities. Had Bach access to a modern piano the Goldberg would have been a very different piece of music from what it is, for in the same way that the harpsichord to meaningful degree shaped the musical structure of the Goldberg, so would have the modern piano had Bach access to it. The argument, then, is both specious and moot, and cannot stand.
I've spoken above of Schepkin's pianistic awareness of the possibilities and limitations of Bach's intended instrument, that awareness reflective, in part, of Schepkin's debt to Gould. As a consequence, Schepkin doesn't quite make the pianist's typical argument for Bach played on the modern piano. Instead, he makes his plea for the piano more subtly (if one were inclined to be cynical about it, one might say more insidiously) thusly (this from the liner notes):
I've encountered people who would sniff at the very idea of Baroque music being performed on the modern grand. It all depends on what you do, and how you do it. [...] Under the fingers of an imaginative player ... [the piano] assumes an uncanny power to produce illusory sonorities of every possible kind. So when we play Baroque music on the piano, why not try to create an illusion of a harpsichord while enjoying the absence of the preset, inflexible sonority that plagues harpsichords? My idea is to play the piano as if it were a "superharpsichord": an instrument with clear and crisp sound, but one that allows for literally millions of degrees of touch and subtle change[s] in sonority.
In his reading of the Goldberg, Schepkin, unhappily, does just that, and more pianistically-grounded malaprops all. In short, this is yet another example of a pianist wanting to eat his Baroque cake and have it too, and as with all such attempts at cake eating, he cannot escape the predictable and inexorable consequence.
A damn shame, too, because in many of the variations Schepkin gets it right which means he knows what's right and when he does, the results are truly first-rate. But the Goldberg is not a work experienced variation by variation, but as a single organic unity notwithstanding that its creator almost surely never expected the work to be performed beginning to end in a single performance; a unity shattered by the failure of any one of its parts. Perhaps Schepkin, as another expression of his debt to Gould, will one day give the Goldberg a second recorded reading, but unlike Gould, get all of it right the second time around.
* For an introduction to the subject of Bach played on the piano versus the harpsichord, see this earlier post.
** By invented embellishment is meant invented as to the embellishment itself, and/or in its placement, as the case may be.