Bach’s Goldberg Variations

December 12, 2020

According to an anecdote in an 1802 biography of J.S. Bach, the maestro had a harpsichord student named Johann Gottfried Goldberg who worked for a certain Count Kaiserling, an insomniac. Goldberg’s job was to play pleasant music to soothe his sleepless boss in the middle of night. Having run out of tunes, Goldberg asked his teacher to write him some more, and thus the 32-movement Goldberg Variations were created. The count was delighted, Bach was handsomely paid, and today’s keyboard players still adore this 1741 work; there were many new recordings of it in 2020 alone.

Part of the work’s modern popularity comes from the surprising boost it received in 1955, when 23-year-old Canadian pianist Glenn Gould, handsome and mysterious, released the variations on Columbia Records, attracting the attention of everyone from seasoned music critics to dreamy-eyed teen girls who’d never heard of Bach. Gould’s contribution, besides the sheer gall of choosing this work for his debut album, was a rhythmic precision and clarity of phrasing unheard-of in piano versions of the work, with the exception of Rosalyn Tureck’s recordings. Of course there were harpsichordists (notably Ralph Kirkpatrick and Gustav Leonhardt) who were already focused on Baroque “authenticity” before it was the norm, but pianists tended toward a lusher, stretchier concept of time in Bach, influenced by the Romantic period.

It may seem like ancient history, but Gould’s 1955 recording – far more than his strident 1981 re-conception – looms over the playing of every subsequent pianist. Some have risen to the challenge, while some have tried too hard to leave their own individual mark. In the latter category is celebrated Chinese virtuoso Lang Lang, who released his first-ever recording of the Goldberg just this year on Deutsche Grammophon. While his Chopin and Tchaikovsky are breathtaking and his Mozart stately and charming, Bach seems outside his comfort zone. And the trouble, as Gould pointed out all those decades ago, lies in the rhythm. Mess with Bach’s underlying heartbeat, and the counterpoint falls apart like a house of cards.

You can hear in Lang Lang’s Variation 3 that his undulating, unreliable pulse washes out any chance of forward motion. Bach designed the accented and unaccented beats of his counterpoint like pistons and gears; the design elements just aren’t connecting here, so the wheels stop halfway up the hill.

It’s useful to compare that Lang Lang slow movement with one by Geoffrey Madge, whose Goldberg recently came out from Zefir Records. Madge does employ a very slight rubato in Variation 13, yet there are long passages of 16^th notes played with clocklike precision, so he never founders and is always moving toward a cadential point. My personal preference is for less pedal than Madge uses, but the overall interpretation is thoughtful and convincing.

For the more up-tempo Variation 1, Madge has a determined buoyancy and a clear understanding of the movement’s harmony (which, as is true in every variation, is based on the chordal structure of the opening “Aria”), even if his rhythm is far from metronomic.

For a piano recording with rhythmic precision in the Gould tradition, I recommend the new release by South Korean pianist Jimin Oh-Havenith on Musicaphon. Her playing is confident and percussive; she’s not interested in turning these tightly wrought Baroque variations into late 19^th-century melodramas. This version of Variation 8, which Bach wrote in cut time (i.e., with a strong feeling of two), sounds like a march:

The similarities to Gould are even more striking in the vivace movements such as Variation 5. While Oh-Havenith’s playing is not as magically quicksilver as Gould’s, her commitment to rhythmic stability is appreciated:

There is always a steady trickle of new Goldberg recordings on harpsichord, and 2020 is no exception. Rubicon Classics has released an interesting offering by Marcin Świątkiewicz, a Polish keyboardist in his 30s. One of his particular strengths is being able to tease out the inner voices from Bach’s complex polyphony. This is nicely demonstrated in Variation 3, where the mid-range melody, surrounded by moving fingers of both hands, sings out clearly.

Świątkiewicz’s Variation 7 is a virtuosic juggernaut, as sensitive as it is powerful. His ability to express the counterpoint clearly yet musically turns this canon into a meaningful interplay between voices without diminishing the wonder of the puzzle-like structure that Bach created. Recordings like this one remind us just how great a work the Goldberg is.

As is true of most of Bach’s works, everybody wants a chance to play it, no matter what instrument it was originally intended for. Parker Ramsay is an early-music harpist and organist; surely his familiarity with keyboards was crucial as he arranged the Goldberg Variations for solo harp. His new recording, released by Kings College, Cambridge, captures a performance in the acoustically marvelous Kings College Chapel, venue for many a fine choir recording.

There’s no denying that the sound is glittering, even angelic, and the playing exquisite. But is this still the Goldberg Variations? Only sort of. Consider Ramsay’s version of Variation 11, so heavily pedaled that the notes seem to swim in a richly sweet caramel pulled with a relentless rubato that obscures the natural accents. It’s lovely but unrecognizable.

Ramsay’s languid approach is far more successful for Variation 25, marked Adagio. He glories in Bach’s aching melody line, while under it the chordal accompaniment shimmers like crystalline pillars.

By way of comparison in this sparsely populated field, it’s worth mentioning the 2009 recording (on Deutsche Grammophon) by Catrin Finch of her own arrangement of the Goldberg for solo harp. Her version has the more expected Baroque rhythm and precision and far less dependence on the sustain pedal.

With so much polyphony crammed into each Goldberg movement, it’s hardly surprising that musicians are tempted to arrange these solo variations for multiple instruments. The ensemble Parnassi Musici, directed by Helene Lerch, recently recorded a new version for chamber orchestra on the label MV Cremona.

This ensemble is skilled in the historically informed performance of Baroque music and uses period instruments. The result is a completely believable work, something Bach might have written himself: He often re-used his own compositions to create new pieces, and this is what he would likely have made if he needed, say, some new orchestral movements for a cantata.

The arrangement is credited to the entire ensemble, so it must have been an invigorating group project to decide how to distribute the counterpoint among the instrumental timbres. For an example of their excellent work, here is Variation 10, a short fugue.