Performing Bach's Well-Tempered Clavier: modern harpsichordists, Gustav Leonhardt, and the "48"
By Bernard D. Sherman, reprinted from Early Music America (Emag), November, 2000. Revised for the Web.
Bernard D. Sherman's books include Inside Early Music (Oxford University Press) and Performing Brahms (Cambridge University Press). His articles appear regularly in The New York Times, Early Music, Emag, and elsewhere. A number of them are about Bach.
at left: Gustav Leonhardt
Gustav Leonhardt | Bob Van Asperen | Ton Koopman | Masaaki Suzuki | Colin Tilney | Glen Wilson | Davitt Moroney | Leon Bérben | Edward Parmentier
Pablo Casals called Bach's Well-Tempered Clavier "the foundation of all music." Narrow that to "Western art music since Bach" and it's not as hyperbolic as it sounds. No music has influenced how Western composers write so much as the "48" (a nickname that counts the number of prelude/fugue pairs). Years before the first published editions came out, manuscripts circulated. From them, Mozart arranged fugues for chamber groups - the influence changed how Mozart wrote his own music - and Beethoven grew up playing the complete collection. Countless composers have followed suit.
But if everyone's been playing the WTC, few have agreed on how to play it. Bach wrote out few of the performance markings - the allegros and adagios, fortes and pianos - that later became standard.Musicians have been disagreeing ever since about what those markings would have been.
By Beethoven's day a new way of playing the fugues had taken hold. As Matthew Dirst shows in his forthcoming The Iconic Bach, this Beethovenian style suited concert performance - something Bach probably didn't envision for the WTC - by emphasizing the main entries of fugue themes in a dramatic way (as Bach probably didn't). Beethoven used the full range of expressive devices available on then-modern fortepianos. Later in the 19th century, however, pianists who indulged in up-to-date expressive nuances had opposition: other pianists who favored the "classic" style of playing Bach. By this they meant staying "true to the score" and avoiding expressive additions.
It would be a mistake to compare that dialectic to the current one between pianists and harpsichordists. Today's harpsichordists, unlike the "classicists" of a century ago, have often been more active, sometimes outlandish, in their interpretations than many pianists. These harpsichordists, as we will see, have created another new way of playing the WTC.
But is it also the oldest
way of playing the work - the way Bach himself played it? Harpsichord pioneers
may have aimed for historical fidelity, but few musicians today would claim
For one thing, while period instruments are crucial to a lot of Baroque keyboard music they're not as obviously necessary to the Well-tempered Clavier. Bach's title means "the Appropriately Tuned Keyboard Instrument," and gives no clue to which instrument he means. Some of the music itself seems designed to be playable on whatever keyboard happens to be handy. In 1722, when he finished Book 1, that would have been a harpsichord or organ, but by the time he compiled Book 2, 20 years later, it might also have been an unfretted clavichord or perhaps a primitive fortepiano. Some pieces do seem to work best on one period instrument or another, but many work just fine regardless of which instrument you use.
So can you make a case against a sensitive performer using a modern piano? You can, according to the Dutch harpsichordist Gustav Leonhardt, who has influenced the current style of playing more than anyone else. He told me that modern Steinways are designed to sing, whereas Bach's music needs an instrument that's designed to speak, a description that covers any period instrument. The English harpsichordist Davitt Moroney believes Bach tailored the WTC to German harpsichords, which work most "fully in tandem" with this music. But - or so I will argue later - the modern piano brings its own strengths (as well as weaknesses) to Bach.
The other part of Bach's title - "Appropriately tuned" - is also confusing. For a century or so, musicians assumed that it referred to equal temperament, the modern tuning used by pianists, where every harmony is just a little out of tune but is equally usable. Then in the 1970s and 80s, the prevailing view became that it referred to slightly unequal Baroque temperaments. These, like equal temperament, allow you to modulate to distant keys without sounding painfully out of tune, but they also add flavor to the varying keys. Just when a consensus was forming for this viewpoint, at least one scholar began to argue that equal temperament was the stronger candidate after all. But the pendulum seems to have swung away from him. After all, Bach's son Carl Philip Emanuel described his dad as "Anti-Rameau" (in other words, anti-equal-temperament); and Bach was said to tune his own instruments, meaning he didn't trust anyone else to do it; he was also said to have tuned his thirds a little sharp. How sharp? Once again, period practice - and its importance - hasn't been easy to pin down in the WTC.
As with most Baroque music, musicians argue about plenty of performance issues. How did Bach notate tempo? How does he refer to contemporary genres, and how did they affect performance? How much did he ornament the score when he performed? And to name a topic I won't be discussing, how old-fashioned was his fingering, and what - if anything - did it imply about articulation? (My own view: nothing significant.) And how much should we care about all this? Early-music performers have differed as widely on these questions as they have on issues of tuning. It's odd, then, that many of them have a common teacher.
Leonhardt and the Harpsichord
Ernst Bloch said of his fellow composers, "We are all sons of Bach"; similarly, harpsichordists tend now to be "children" of Leonhardt. Such well-known harpsichordists as Bob Van Asperen, Lisa Crawford, Alan Curtis, Richard Egarr, John Gibbons, Pierre Hantaï, Ketil Haugsand, Ton Koopman, Charlotte Mattax, Davitt Moroney, Edward Parmentier - let me pause for breath - Andreas Staier, Skip Sempé, Jeannette Sorrell, Colin Tilney, Glen Wilson and (believe me) many more have all had some training with him.
They, in turn, have trained other harpsichordists, such as Masaaki Suzuki. (Christophe Rousset studied with Asperen, but says a session with Leonhardt "liberated me" at the keyboard.) Some non-Leonhardtian harpsichordists are equally notable, of course. Kenneth Gilbert is "the other most significant figure in harpsichord teaching over the last 30 years," with an approach "complementary" to Leonhardt's, according to Moroney, who learned from both. Plenty of other notable harpsichordists are non-Leonhardtians; but labeling them all that way isn't meaningless. That tells you something. Due to limits of space, and with apologies, this article focuses on artists connected to Leonhardt.
The harpsichord revival did not begin with Leonhardt, of course. Arnold Dolmetsch gave it a push a century ago, and Wanda Landowska, on an unhistorical modern harpsichord, gave it a bigger one. Thanks to her genius and sense of musical drama, her recording of the Well-Tempered Clavier remains one of the greatest, even if its style seems Romantic today. Ralph Kirkpatrick, another fine musician, reacted against Landowska (with whom he had studied); and Helmut Walcha, as Leonhardt told me, "reduced everything to zero," with a style that minimized inflection.
What made Leonhardt's approach to the harpsichord distinctive? A suitcase full of inflections. There's his way of letting the quill rest against the string before striking it, allowing a wider range of subtly varied touches than a modern piano has. There's his tendency to enunciate motifs, rather than to seek a long, smooth melodic line; the idea is to deliver the music as if speaking. There's also the wash of resonance that results from holding keys down after striking them. That wash lets you vary the sonority as the harmonic tension ebbs and rises (listen to Suzuki's rendering of the first prelude for a memorable demonstration).
There's also an emphasis, not so evident in Romantic or modernistic Bach players, on differentiating between strong beats (downbeats) and weak beats. This distinction was central to Bach's language, particularly when he plays against expected beat patterns, and it is harder to bring out on a harpsichord than on a piano. Harpsichordists can't differentiate beats with obvious gradations of loud and soft, so they set off strong beats with rhythmic nuances. To emphasize downbeats they often insert a tiny silence just before them; sometimes they hold downbeats a little longer than written. They also use timing and other devices to emphasize motifs, cadences, and points of arrival. If Leonhardt was by no means the first or only one to explore these devices, it was he who integrated them into what has become the prevailing harpsichord style.
Still, how insistently you should emphasize strong beats and bring out motifs is a matter of dispute. The pianist Dady Mehta recently wrote about "egregious practices often called 'authentic,' but which could more aptly be called agogic idiosyncrasies." Whether Leonhardtian inflections are "authentic" in the historical sense is hard to know; but that they can sometimes be idiosyncratic and overdone is beyond doubt. A couple of recordings by Leonhardt students are rife with examples (I won't cover these two recordings in this survey). One of them ruins the lyrical A-Major Prelude in Book 2 by sitting on most of the downbeats, thus interfering with the flow of the piece. Emphasizing every strong beat can also obscure the music's larger shapes and make it sound dry. And "speaking" rhetorically sometimes yields seasick tempo fluctuations. Such instances demonstrate a common problem in the arts: a new approach to style tends to be exaggerated before it is fully digested.
On the other hand, when harpsichordists just ignore the "metrical hierarchy" of strong and weak beats, they can sound mechanical. And when the Leonhardtian approach has been lived with sufficiently, it can produce rich results.
Leonhardt himself is a master musician, and his own recording of the WTC, about three decades old, has some unsurpassed performances (such as the great Eb-major prelude in Book 1). Still, for all its moments of insight and imagination, few of his admirers would count it among his best recordings. Too often it seems sober. In my book Inside Early Music, Leonhardt said that in the recording studio he tries to play "neatly," whereas in concert he tries to play "beautifully." In concert, Leonhardt has struck me as the greatest living harpsichordist, but recordings often fail to him justice.
Of course, the same can be said of most musicians: they tend to become either too strict or too extreme in front of microphones and editors. If every recording of the WTC is flawed in one way or another, the very process of recording may be part of the reason.
Besides, the Leonhardtian approach fosters quite a bit of creative re-thinking from one performance to the next - more than usually happens among pianists - and such variation doesn't suit the studio.
That openness to variety may be part of why Leonhardt has avoided creating an orthodoxy about how to play the WTC. His pupils have often played the pieces quite differently from him. In an essay in Bach Perspectives 4, John Butt suggested that some students actively try to sound "Not Leonhardt."
Ornamentation: Should we add notes?
Butt made the point with respect to Ton Koopman's "Goldberg Variations," and it applies to Koopman's WTC. His most obvious "Not Leonhardt" feature is his added ornamentation. He lays it on thick, smothering the great C#-minor fugue in Book 1 (for example) under a mound of ornaments. Pamela Kamatani, a student of Butt, compared Koopman's playing to "adding buttercream rosettes to whipped-cream decorations."
The cake, as her metaphor implies, is already decorated. Bach no doubt added ornaments when he played; but it seems likely, David Schulenberg notes, that in mature works Bach was increasingly writing out the ornamentation rather than leaving it mostly to the performer. The result is music that integrates the underlying idea closely with the surface details. Such well-crafted detail is hard to equal with one's own ornaments, even if Bach didn't mean for all of the details to be absolutely definitive. Besides, Bach's ornaments decorate an underlying musical idea, which careless new ornaments can obscure.
Still, adding ornaments doesn't always detract from the WTC. Masaaki Suzuki, a student of Koopman, adds ornaments in his recording of Book 1 with more discrimination than his teacher. With some exceptions (such as the Eb-major prelude, in which the twiddles and arpeggiations go overboard), the results sound natural. Suzuki also makes liberal use of the other expressive nuances of modern harpsichordists, but makes convincing sense of them. That's partly because he maintains a natural forward motion amidst the nuance. (One of Bach's feats is keeping a piece aloft, not letting it touch ground until the end. Overly inflected playing can sabotage him.) Suzuki's playing is often emotionally moving as well. As in all recordings, some performances are misconceived; but overall, his Book 1 is masterly.
My favorite added ornament on disc is by Bob Van Asperen in the C#-Major Fugue in Book 1. In the continuation of the first entry, he throws in a twiddle that feels like a spontaneous overflow of joy, rather than a considered application of orthodoxy. Baroque ornamentation should aim for such improvisatory emotional rightness.
Asperen adds relatively few ornaments, but he generally plays with engaging spontaneity. He uses Leonhardt-style inflections as subtly as his teacher. In the C Major prelude of Book 2, for example, he far surpasses some other Leonhardt followers, the ones who underline too many cadences and arrivals with slowings or hesitations. Asperen finds the right mix of forward flow and improvisatory inflection here and, for that matter, in the A-major Prelude of Book 2. In the WTC, Asperen rarely overinflects the tempo; if anything, he is sometimes too strict. His F-major Prelude in Book 2 (as in many recordings) is faster than I think the 3/2 signature indicates, and sounds mechanical.
Asperen differs most obviously from his teacher in his tempos, which are often slower or faster than those of other harpsichordists. Anyone who believes that you can't take a truly slow tempo on a harpsichord (since its tones die away quickly) has never heard Asperen play the Eb-minor fugue in Book 1. He takes it far more deliberately than any evidence suggests is plausible, but he never lets it sag. Admittedly, a few of his tempos do drag, as in the B Minor Fugue in Book 1. As for his unusually fast movements, the Prelude in D major in Book 2 comes close to my own ideal of how it should go. Its lively tempo responds to the time signature (12/8 combined with ¢ - the latter sign suggests speed). And Asperen makes the controversial choice to not assimilate the eighth-note pairs to the prevailing triplet rhythm. What makes his realization special is that he brings out an emotional contrast between the eighth-note pairs and the opening flourish. Ton Koopman does not assimilate the pairs either, but he doesn't convey the affective contrast.
In Prelude 24 in Book 1, on the other hand, Asperen plays way too fast and too staccato (his articulation is sometimes too dry elsewhere as well). His tempo follows a tradition started by Leonhardt - and, independently, by Glenn Gould - and maintained by many harpsichordists. The idea is presumably that the "Andante" Bach wrote over the score had no implications for tempo. Yet relevant contemporary sources make it clear that in Bach's time and place, the word "Andante" slowed the tempo somewhat. Among the players surveyed here, only Suzuki and Colin Tilney get this piece right. Their unhurried approach allows the expressive dissonances of the Prelude to have their full effect.
But, then, Tilney is not given to hurry. His WTC is the slowest overall on period instruments. While several of his tempos seem glacial (try the A-Major prelude in Book 1), he usually makes his tempos come alive. And a few of Tilney's tempos are unusually lively. He is one of a handful of performers who play particularly fast when the time signature has a 16th-note denominator (e.g., 24/16). Bach's student J. P. Kirnberger first argued that such signatures indicate extra speed in the WTC, and his report fits with other principles of period notation. I find the results musically right (it is also observed by Koopman, Scott Ross, Léon Berben, and Glen Wilson).
As this example suggests, Bach notated tempo differently than we do. He wrote, for example, only nine tempo words in the entire WTC, where we would write them on every piece. Controversy continues over the tempo implications of these words and of Bach's time signatures. No resolution will ever be final. After all, even those of us with decided views on these matters admit that notation (whether Bach's or our own) is not an algorithm that always leads performers to the right tempo. Both Bb-minor preludes, for example, feel slower to me than the time-signature (as I read it) suggests.
Besides, plenty of other influences affect any tempo, such as the emotional tone of the piece, how fast the harmonies and figurations move, and the specifics of each performance. Consider Tilney's slow tempos. His gift of phrasing helps make them work. He is a master of tempo inflection, letting the music breathe in a communicative way while maintaining forward momentum. And Tilney seems to savor every turn of Bach's thought, particularly in the realm of harmony. The tempos let Tilney communicate his delight in Bach's thought processes. They would work in few other performances, but here they are part of a whole musical experience.
Among other influences on one's tempo are the venue (resonant concert hall? dry living room?) and the instrument. In the WTC harpsichordists tend to play slow pieces (like the serious fugues) faster than pianists, while pianists play the virtuoso pieces much quicker. Perhaps one reason why many harpsichordists play virtuoso pieces more slowly is the Leonhardtian style: to bring out a strong beat on a harpsichord requires tiny delays in timing, which can slow the tempo somewhat. Pianists can thus play faster without sounding clattery. Perhaps, also, harpsichordists are thinking at times of Baroque dance genres, to which some preludes and fugues refer. In some virtuoso pieces these dance genres could cause one to play a little more slowly than typical pianists.
WTC is chock full of references to Baroque genres, from the old-fashioned stile antico to the up-to-date style galant. In principle, knowing period genres should help us understand how to play the pieces. But again, with the WTC nothing is that simple. For one thing, Laurence Dreyfus (in his landmark Bach and the Patterns of Invention) points out that Bach liked to "compose against the grain" of genre conventions; as a result his genre references may not imply as much about performance as we would hope.
Besides, it can be hard for us moderns to figure out what genres Bach is actually referring to. Sometimes we have no trouble: the F major fugue in Book 1, for example, clearly refers to a dance, the passepied. But what do we make of the G minor prelude in Book 2, which the musicologist Peter Schleuning confidently calls a French overture - when David Schulenberg concludes that it is not a French overture? The disagreement is not unusual.
In this case and many others I find Schulenberg's book The Keyboard Music of J. S. Bach a reliable guide. When artists play the piece as a French overture (like Asperen, for example), the "double-dotting" effect associated with the genre seems a little dogged. I am more moved by Moroney or Tilney, who play it as a normally dotted rhythm, and create a more natural sense of pulse.
Tilney is the only Leonhardt pupil to have recorded Book 1 complete on a clavichord. Not since Ralph Kirkpatrick in 1959 had anyone else recorded a full book of the WTC on that sensitive instrument with its subtle gradations of loud and soft. Some period sources suggest that Bach preferred the clavichord to the harpsichord. Harpsichordists are too quick to dismiss these sources (which could plausibly refer to Bach's last decade, when he might well have had access to "unfretted" clavichords suitable to his music). Tilney's musicianship demonstrates the attractions of the instrument.
So far, no one has recorded the piece on a harpsichord strung not with the usual iron alloy but with brass, as some scholars believe Bach preferred. Their argument has come under question, but so have the standard Flemish- or French-style harpsichords favored by almost all present-day harpsichordists. (Suzuki's harpsichord is an example of the shortcomings - a rather dry sound, albeit enriched by his outstanding playing.) Recent arguments by John Koster and others suggest that Bach's time and place may have favored more colorful instruments, sometimes with 16' stops that can double the lower line an octave below, and upper stops that create special sonorities. None of the Leonhardtian harpsichordists reviewed here use such an instrument, and most are demure with respect to colorful registration. Leonhardt himself does more with registrations, without going to unmusical extremes. The same holds for Glen Wilson.
Wilson is one of the few to use a German harpsichord, if not the kind just described. (He uses a modern copy; Asperen and [in Book 2] Tilney use historical German harpsichords.) As in any recording, some of the movements on Wilson's strike me as overinflected or too fast or slow; but his overall quality is unusually high, and in some movements unsurpassed. Only Tilney equals him, for example, in conveying the depths of the Bb-Minor Prelude in Book 2.
One inflection Wilson uses, like all of his colleagues, is chord-spreading. On a harpsichord, hitting all the notes in a chord at once can sometimes sound harsh. Spreading the chord a bit can create rich effects, which can be used for expressive or structural emphasis. In the lyrical Eb Major prelude in Book 2, when the opening is recapitulated, Wilson underlines the return with a beautifully spread chord. Many harpsichordists do the same, but his realization is especially well timed. Wilson is sometimes criticized for overdoing this device (if guilty, he is hardly alone), but to my ears he often makes it work.
Davitt Moroney's playing is often supple and moving. His Book 2 conveys the character of each piece with unusual consistency. His Book 1 is a bit less consistent: the Prelude in C# is a joy, for example, but the Fugue's first entry is given an unsuccessful little rubato at the first change of harmony. But mannerism is not a common problem in Moroney's renditions, and hardly any recordings of this work are consistent. Moroney's WTC as a whole is excellent.
For harpsichordist Leon Bérben mannerism is the name of the game. Consider his alteration of notes and even harmonies from what Bach wrote; in principle Bach may have done the same when he played, to be sure, but see if you like the results here. Or consider Bérben's sometimes extreme rubatos and tempos; he leaves everyone else in the dust in the C#-Minor Prelude in Book 1. No one better illustrates the observation about harpsichordists trying to be "not-Leonhardt," and in this case, "not-Koopman" and "not-Asperen" (he has studied with all three of them).
Why the WTC Will Never Be Performed Historically
But there's room for eccentric WTCs; no music is more familiar, and hearing it played in a novel way can be bracing as well as maddening. Besides, given the WTC's broad range of emotions and styles, no one performer can do justice to all of it. My preferred CD would be custom-made, with selected tracks from various recordings. Pushed to a wall, my favorite Book 1 is a new one by yet another Leonhardt pupil - Edward Parmentier (on the Wildboar label), who has the most beautifully recorded harpsichord I've ever heard, a supple sense of timing, and a subtle, searching sensibility. For Book 2, I'd turn to Moroney, or to clavichordist Ralph Kirkpatrick on DG.
If I were going to be really purist, I'd note that the one indispensable WTC is a copy of the score. After all, this collection best reveals itself when you have the notes in front of you, because Bach wrote the fugues not for CD listening but for playing - by people who are interested in how to compose preludes and fugues. While the listener Bach imagined was responding to the emotion and beauty of the music, as listeners might today, he or she was also following the combinations and transformations of musical ideas. Some of these processes are nearly (or completely) impossible to hear just through listening. That's not because Bach miscalculated. He expected the WTC's listeners to also be the performers, engaging with the music through the ears, the eyes and the fingers.
But let's face it: listening is bliss, and playing is increasingly rare. Thus historically oriented performers today face a conundrum. How much effort do they make to project a fugue's musical events to the person sitting in the last row, when Bach or his contemporaries, playing fugues for themselves, would have made no effort at all?
In my experience, the best pianists project simultaneous strands of counterpoint to the peanut gallery better than most harpsichordists. Schulenberg suggests that one reason may be the pianist's tendency to "think about how dynamics can bring out particular lines," although he adds that there is usually no reason to "emphasize a particular line in a fugue." Indeed, Peter Williams argues that because the piano "cannot help but play expressively...one can easily miss the desired evenness between the parts" of a fugue. Many pianists overemphasize the later entries of the opening theme of a fugue (a classic example is Rosalyn Tureck). Other pianists avoid such distortions only by playing with what Williams calls "a kind of neutral blandness."
But the best pianists avoid both pitfalls, projecting each line in what pianist Charles Rosen calls "low relief." And Schulenberg is right: subtle dynamic shaping of a line can help, giving an edge here to pianists (and clavichordists). Another arguable plus for the piano, as Schulenberg also observes, is that it lets one use dynamics to project "suspensions" - sustained notes that turn dissonant when other notes change. Suspensions play a central role in Bach's music. Thus, to be honest, my desert island disc would probably be a piano set: my current favorite is the first complete recording ever made: Edwin Fischer (monaural, on Naxos). I also admire such pianists as Till Fellner in Book 1 (on ECM) and Andras Schiff (London) and Angela Hewitt (Hyperion) in Book 2.
Still, if the piano has virtues - I'd also mention its wide palette of tone colors - they shine mainly in the concert realm. When you play a fugue for yourself, as Bach and his peers did, there's no need to project suspensions, inner voices, or anything else. Of course, modern harpsichordists do try to project these things to the public, sometimes quite successfully, using all the means at their disposal. They must. Truly historical playing would be too self-involved for concert life. Once again, the WTC by its nature challenges the ideal of historical performance.
Still, historical or not, those performers who have been thinking about period fingering, tempo, phrasing, articulation, and genres, and who have revived instruments of the composer's day, have changed the way even many pianists play the work. They have revealed much about the music. And their best playing is its own justification. The WTC may challenge the early-music approach, but for the finest musicians, challenges can lead to compelling results.
Selective English-language Bibliography:
An essay on Book One at Yo Tomita's Website: http://www.music.qub.ac.uk/~tomita/essay/wtc1.html#title
J. S. Bach: The Well-Tempered Clavier, edited by Richard D. P. Jones. (The Associated Board of the Royal Colleges of Music, 1994). The best English-language edition so far, in a paperback version that any serious WTC fancier should own. Contains very useful material on manuscript sources, variants, and performance questions. A bonus is the inclusion of Donald Francis Tovey's classic analyses of each prelude and fugue, along with the editor's helpful caveats about the Tovey commentary.
The Keyboard Music of J. S. Bach by David Schulenberg (Schirmer, 1992). The finest book of its kind, with two excellent chapters on the WTC.
Bach and the Patterns of Invention by Laurence Dreyfus (Harvard University Press, 1997). Two chapters of this important book address the WTC, with exceptional insight.
Bach Perspectives 4 (ed. D. Schulenberg, U of Nebraska Press, 1999). The essays by Schulenberg, John Butt, John Koster (on Bach's harpsichords), and Paul Walker are important for anyone interested in Bach keyboard performance; the other essays on Bach's style are equally valuable.
Rudolph Rasch, "Does 'well-tempered' mean 'equal tempered'?" - he argues yes in BHSTercentenaryEssays (1985) 293-310; vs. Bradley Lehman, who argues no in Early Music, February 2005
Interpreting Bach's Well-Tempered Clavier by Ralph Kirkpatrick (Yale University Press, 1984).
Recordings covered (I've left out numerous important recordings, largely because I am focusing on Leonhardt "school" players in this article):
Bob Van Asperen, hpchd, Virgin Classics 5 61711 2 (recorded 1987-89)
Leon Bérben, hpchd, Brilliant Classics 99362/1-4 (rec. 1999)
Edwin Fischer, piano, EMI Classics - #67214 (recorded 1933-36)
Kenneth Gilbert, hpchd, Archiv 413 439-2AH4 (rec. 1984)
Glenn Gould, piano, Sony Classics - book 1, SM2K 52 600 (rec. 1963-65); book 2, #52603 (rec. 1975)
Angela Hewitt, piano, Hyperion - Book 1, CDA67301/2 (rec. 1997); Book 2 #67303 (recorded 1998-99)
Ralph Kirkpatrick, clavichord - Book 1, Archiv 289 463 601-2 (rec. 1959); Book 2, Archiv 289 463 623-2 (rec. 1967)
Ton Koopman, hpchd, Book 1 Erato 2292-45428-2, (rec 1983); book 2 Erato 0630 16175 2
Wanda Landowska, hpchd, RCA Gold Seal GD87825(3) (3CDs) (rec. 1948-54)
Gustav Leonhardt, hpchd, Deutsche Harmonia Mundi Editio Classica GD 77011, 77012 (rec. 1967, 1973)
Davitt Moroney, hpchd,Harmonia Mundi 2908084 (recorded 1988)
Edward Parmentier, hpschd - Book 1, Wild Boar (rec. ?; released 2004)
Andras Schiff (piano) Book one Uni/London Classics - #14388 (rec. 1984) 471236-2 (rec. 1987)
Masaaki Suzuki, hpchd, Book 1, Bis 813/14 (recorded 1996)
Colin Tilney (clavichord and hpschd.), Hyperion 66351 (recorded 1988/89)
Glen Wilson, hpchd, (recorded 1989) Teldec 3894-25714-2 (11 Cds, also containing the Partitas, Inventions, French and English Suites played by Scott Ross, Alan Curtis, and others)