Ideas about the Goldberg Variations BWV 988

Arguably the most important composition for the harpsichord (as fundamental  to the art of the harpsichord as Frescobaldi’s seminal Libro Primo of 1615), the so-called Goldberg Variations are one of a very special handful of large scale compositions from the last decade of Bach’s life. This is the decade in which he revised the St Matthew Passion, assembled the B Minor Mass, composed the Art of Fugue, and Canonic Variations on von Himmel Hoch. It is his last and most sophisticated work for the harpsichord, and it seems appropriate that a performer think as deeply as possible about it in order to arrive at a good interpretation. More than that, since Bach’s published works have a pedagogical dimension, how can anyone resist the opportunity to learn as much as possible from it?

As I’ve become better acquainted with BWV 988 (I have to make an effort not to call it the Goldberg Variations, a moniker that has glued itself to the piece and to our minds, but which the musicologists now agree, I think, comes from a bogus story), it’s become harder to conclude what is the most important perspective to have on it. It is as much a lesson by demonstration in the art of composition as a tutor for a very sophisticated harpsichord technique. It is clearly inspired or triggered by the publication of Domenico Scarlatti’s 30 Essercizi in London a few years before, as it integrates the ancient art of counterpoint and polyphony with the new art of dazzling keyboard virtuosity. It also pays homage to Buxtehude and Bach’s other compositional ancestors. And from so many perspectives, it’s very unusual.

I can imagine Bach looking over Scarlatti’s Essercizi the first time, thinking to himself; how intriguing, these figurations, these hand-crossings, textures. Hmm. Lacks form and structure, though, and the themes are weak. Oh, there’s this change in the musical material halfway through the A section, that’s a cool idea I could do something with.

It poses questions and challenges for anyone who chooses to study it. It seems to me that’s who he wrote it for; any connoisseur willing and able to grapple with it and probe its depths. He hides clever little things in it for us to find and marvel at. Did he write it with the intention it be performed at one sitting? That didn’t seem to be a tradition then, so we can guess fairly safely that wasn’t its primary purpose. But of course it was masterfully enough composed, so that a performance could be engaging and fascinating, even for a piece that takes some 80 minutes to play through (with most or all repeats), remains in one tonality throughout, and adheres to the same harmonic progression 32 (64?) times.

It begins with a little riddle; entitled Aria with 30 Variations. The opening Aria seems much like a sarabande, but in using the term Aria he’s telling us it’s not a dance. What does this mean? It might mean not to take it too slowly, or it might mean not to emphasize the second beat as you would in a sarabande. As I play it, at least this week, I try to pay most attention to the bass line, since that is after all the theme of the variation set, not the soprano melodic line. So I treat the first beat as the strong one and I don’t enunciate the meter of a sarabande. But I reserve the right to change my mind. Bach had a habit of leaving little clues; one should pay attention and think about what they might mean. He might also mean in some way, things are not always what they seem. Because, of course what’s very unusual here is that the theme isn’t really overtly stated here. The theme is of course the bass line and its harmonic progression, but it’s obscured by the lovely soprano melody, and the bass line itself is never stated without elaboration.

As a performer, I get to imagine and consider aspects and possibilities which might make for a better, more interesting interpretation. I don’t have to be authentic, whatever that might mean, or “right” but I do feel it’s important to play it in an esthetic consistent with the time in which it was written. So I have to think about the language in which it was written, which is the language of gesture and what they call rhetoric. I can’t claim to be well-read in this, but as I understand this means several key things.

First, the music is intended to convey or elicit emotions. The performer has to find a feeling in each variation that is persuasive, and also works in terms of the piece as a whole and in connections or transitions from one variation to the next. I think the search for those feelings, or affects, are as much of a challenge that he poses for the performer as mastering the notes themselves. While many of the variations are extraordinary physical/mechanical challenges, finding the most persuasive feeling is just as, or more, important. And fascinating, too, as I don’t think the music is that prescriptive; there seem to be often not just room for nuances but very plausible significantly different choices. So it seems, at least in part, an exercise in finding the affect possibilities that the music permits, or better, suggests.

Next, the language of this music, really all baroque music and each subtype in its own way, is an integration of bodily gestures (like bowing, or stepping, or leaping) and the elements of speech (consonants, vowels, punctuation, parentheses, paragraphs). These all unfold within the all-important envelope of meter and rhythm.  So there is a co-existence of the irregular, because speech is irregular, and meter is subtly irregular, within the regular overall flow (the fundamental tempo or pulse) and structure. This is the essence of high baroque art. It really is important to remember that this music belongs to an age well before that of rhythmic evenness, well before the metronome and its exact equality of beats and subdivisions within beats. This music requires a masterful grasp of how to enunciate meter and clearly project strong and weak beats, not to mention unequal subdivisions within those beats. Was that a particular challenge in BWV 988, or a baseline expectation that a player should bring to it? Because that is without doubt the language this piece is written in, so if you’re not grappling with these issues in performance you are perhaps missing the point. And grappling with them is very challenging.

It is really impressive, actually astounding, what an elaborate piece Bach creates from such a simple idea, the bass line of the first 8 bars. It’s a very simple stepwise descending line, and then a cadence. That’s its first sentence. Also within its first sentence is a soprano line that in the first bar, ascends one step.

Perhaps it’s clever, maybe just a coincidence, that Bach’s large scale structure is built on every third variation being a canon, starting with a canon at the unison and the ascending one step at a time, until it’s a canon at the ninth, in two parts. I think that’s deliberate, though. He’s demonstrating the relationship of the smallest element to the whole. I see that also in the bass line theme, which has two halves, and the two halves of the whole work as punctuated by variation 16, the French overture. But I won’t go into the numbers thing here, other people have written plenty about that.

Within its formal architecture, there is also an increasing progression of inventiveness and complexity, consistent with what you would expect of an rhetorical argument. In the last few variations, the innovativeness really goes bananas, and is really unlike anything else. The burbling figurations of variation 26, the sustained 32nd note trills in variation 28, the crazy barrages of chords and dazzling runs of variation 29, are all really unique and bring the work to an incredible concluding intensity, which require a cool-down of both the quodlibet and the aria again.

I have a hypothesis; in the first half of the piece it seems to me that each set of three variations which culminate in a canon lead up to the canon in dramatic intensity. That’s most clearly true in the first twelve variations, but it works consistently. In this view, variation 15 is bit of an enigma but that’s fine, Bach is happy with enigmas. Variation 9 is the exception, I think the high point is the Fughetta, Variation 10. Perhaps I’m wrong, but perhaps Bach the composer was teaching us that too much predictability is a bad thing. But in the second half, that doesn’t seem to work, at least for me. It’s more convincing, to me, that the canon that ends each threesome is more, I think, recessed, some kind of anticlimax or has a wistful aspect. There’s even a funeral song, in variation 25.  variation 27 is intriguing, and Bach calls attention to how unusual it is in a few ways; a canon at the ninth, now, that’s really weird. A drop of a seventh in the theme, that’s weird and important. And 27 is 3 times 3 times 3, which we know signifies importance. I wonder if you play it in some way as the voice of counterpoint from beyond the grave (because we already had the funeral song, remember), or even some kind of resurrection, it works? The bravura style finally wins out over the ancient learned art of counterpoint, and variation 29 is its triumphant exultation. This raises a wonderful question of how to play the quodlibet.

That’s a possible dramatic narrative, and it’s possible to make everything fit within that narrative in some way. You get to ask yourself whether it’s a struggle between the old and new, or whether it’s the rise of the new, with its youthful exuberance while the old gets tired, worn out, fades away. An interesting concept for the performer. I don’t know if it’s right, I care if it adds to the performance. But it is hard to imagine that in that decade Bach would compose a strange masterpiece that had less than a profound underlying meaning.

This piece is about many things, and contains a wealth of mysteries. It’s the first real set of etudes, but really for me they are etudes of rhetoric and feelings foremost, disguised as etudes of finger challenges and compositional prowess.



Posted in Uncategorized | Leave a comment

On public performance

Concert-goers might find this of interest. And there’s a bit further down that’s important for harpsichordists.

Public performance of art music is a very risky thing. While we go to a concert with high expectations, usually, of at least part of it being a lovely experience, there are many bad concerts and that’s just the nature of the thing. Even great performers have bad concerts; memory slips, concentration lapses, clams all over the place, jet lag, illness, anxiety and tension – these are all normal and more frequent than you would think. It’s part of the business. You move on. It’s truly wonderful when it all comes together, and everything is beautiful and fascinating, passionate or relaxed, and if there are mistakes they’re hardly noticeable. It’s wonderful for everyone, everybody gets to go home really happy if not ecstatic. Me, I’m happy if 50% of the concert is like that (as a listener).

In the twentieth century concert goers and presenters worshipped a bit too much at the church of perfection, where accuracy and cleanliness are too prominent and hardly anyone seems to regret a loss of artistic exploration and risk taking. yes, I would like to have both, and everyone would, but that’s the way it is. It’s a challenging life for touring professional solo pianists, who perhaps have the toughest job of all; their craft requires a phenomenal level of education, preparation, and discipline.

There are so many stories about famous artists’ bad concerts; Sviatoslav Richter used to tour in many small towns before he hit the big cities, and I’m told he preferred them. Yes, he gave bad concerts. All the horror stories, I’m sure; I’m guessing that many of the messy ones were where an excess of passion and excitement produced little disasters, but then I’d probably be wrong. Many people I admire have mess ups, and some of them make me cringe and some of them I absolutely don’t care about, because the rest of it was so incredible.

Professional musicians today generally understand that it is very important to be well prepared, ensure your program falls within your abilities, and is musically interesting, to be rested and pace yourself. This and more is what you learn from your teachers. The concert, for your audience, is part emotional experience and part watching NASCAR or a high wire trapeze act. Be ready.

So now about harpsichord concerts. There are not many solo harpsichord concerts in the US, so there are not really any professional harpsichord soloists in this country. Most harpsichordists can’t invest the massive amount of time it requires to do that well. They are busy playing in ensembles and orchestras, maybe teaching – and those are very different things. That’s where the work is, and it doesn’t allow for solo practice of, let’s just toss some number out there, maybe 3 hours a day. They’re not playing one program in 10 cities over 2 months. They simply can’t be adequately prepared. This is true, to a different degree, for most harpsichordists everywhere; many of the fine ones are conducting and have little time for solo concert preparation at the right level. It’s a problem. To an extent, we should be understanding of their challenge; it seems only fair.

And people should understand how nonstandard harpsichords are. The keyboards are different sizes and topologies; the sounds of the attack and the nature of the decay are very different. Some of them don’t work well. I did a concert once, was only allowed to see the instrument the night before, the technician didn’t show up, the instrument was awfully dead and feeble and very hard to hear at the keyboard. It was extremely stressful. Could I have dealt with it better? Perhaps. Did I blame the instrument? No. Should I have cancelled? Probably. Some excellent musicians liked it a lot, one who never tells little white lies; others were disappointed.

So presenters avoid the instrument, concert fees are too low to support a truly high caliber soloist, and opportunities too few. So the harpsichord concerts there are tend to create a dim, sad expectation what a harpsichord concert is expected to be.

There are plenty of musicians who love the instrument and its music, and like to perform in public. But the performances are too often either unspeakably dull or very sloppy, and sometimes – gasp – both. I would ask these people to seriously reconsider what they’re doing because it reinforces abysmally low standards, minimal opportunities and low fees. Do a concert within your abilities – shorter, mix it up with other players, maybe other instruments. Shine within a better judged context. Do a fringe concert. A house concert.

Then there are people who perform and blame the instrument. I once went to a concert where the performer chatted a little between pieces, and said there should be a disclaimer on the lid of a harpsichord saying, excuse the errors this is a harpsichord. So now we have performers, at a paid concert, teaching the audience to expect lots of wrong notes. We’re training them, now, for very low standards. It was, sadly, a very boring concert with many misjudgments. Dull, slow, square, uneven, and full of mistakes. In short, inadequately prepared and the wrong program for this person to take on; and it was their personal instrument, generally one not inclined to play wrong notes. But there is no excuse for lack of integrity; if you commit to perform a program, please have some respect for your audience and the music. Prepare well. Practice adequately for the program. Have a competent technique. Try to make it musical and interesting. If you don’t have time for that, do house concerts. I wish there was a way to have a nice friendly conversation about this, and talk about it in the kindest way possible.

There is absolutely no reason why the harpsichord should produce more wrong notes than a piano, there are just excuses. But to tell an audience it’s the nature of the instrument is to set the instrument back 50 years, and that’s inexcusable.

Posted in Uncategorized | Leave a comment

Some thoughts on learning the Goldberg Variations

Learning the “Goldberg” Variations

There’s a practical/tactical aspect of learning this work; not to be ignored since it’s 80 minutes plus or minus of music, depending on how many repeats you observe. And very importantly there’s an artistic dimension as well that has to be approached strategically and with clear intention.

I’ll start with the practical side. Everyone will come to this with their own facility of playing and learning, so I’m just going to go from my own experience. First of all, learn it once and then come back to it a few years later. Now, there’s a first phase to it which involves practicing the variations that come more readily, for most people that’s going to include most of the first half, probably with few if any of the hand-crossing variations, and some of the second half, again without the hand-crossing pieces and with or without the overture. The purpose is to realistically segment the work, getting some variations fairly comfortable in the fingers and mind, while reserving enough time for more deliberate work on the challenging pieces, which take most people a lot of time to learn. I’ve found a good strategy is to spend perhaps half your practice time improving and honing what’s been learned, and half your time on learning additional variations, and I’d suggest no more than three at a time. There’s generally an ascending degree of difficulty, so starting with the hand-crossing pieces in the first half makes good sense. My experience was that it takes about a week for a set of three very difficult variations to progress from the very slow learning stage to relatively comfortable and ready to move to the ‘improving/honing’ stage. So that means it will take about 4 to 5 weeks of pretty intensive work to get everything into the honing stage. I won’t venture an opinion on how long the work should be honed before it’s ready to perform. That’s a personal issue but really I would encourage a long time.

It’s important to spend extra time and attention on the difficult ones at the end, 26 through 29. It’s typical to save these for last, and because of the fatigue factor in performance they need to become second nature, very well learned and reliable.

Variation 20 requires special attention and so does 23.

A friend advised me to spend time practicing transitions; endings and beginnings of sequential variations. Good advice, to really ingrain how to start each variation with the right tempo and character, since the change of tempo and character from one to the other is sometimes very difficult. Most people put more creative thought into the beginning of a piece; the ending is just as important and a critical interpretive opportunity.

You have to be very good at feeling the flow of the whole work from beginning to end, yet not suffer from energy deficit near the end. So I recommend, as time goes on, to practice them in reverse order one day, and in proper order the next. Also, mix it up and jump around. I nearly always practice the entire work every day, believing it’s a bad idea to neglect any variation no matter how apparently easy it might seem.

A few thoughts about the artistic or interpretive element; first of all understand what this work is about. It’s a fusion of the ‘old’ style of the baroque, with its canons and intricate counterpoint, and the incredible, really amazing, elaboration above an elegant bass line, with the new style, inspired certainly by Domenico Scarlatti whose 30 Essercizi were published in London just 2 years before, of dazzling keyboard virtuosity. Two principles of creating astonishment are joined, the intellectual and the virtuosic. So Bach must have sought to amaze and surprise, and it’s in this vein that the player should be thinking. And the way one amazes and surprises on the harpsichord is through, not speed necessarily, but texture and character. I think it’s rewarding to see each variation as a study in texture and character; in practice this means exploring the many interpretive possibilities and choices in each variation, finding convincing and interesting possibilities. Then you can make some well-informed choices about progression and juxtaposition, and find details that can surprise and amaze the listener, whether those are small details hidden within a variation or larger interpretive decisions.

A piece that takes this much practice time and repetition carries with it the risk of becoming boring through a kind of sameness; sameness that can result from too much repetition. Practice that involves dramatically different choices and character is a way of maintaining a kind of freshness, and also of understanding and internalizing the music much more deeply that you ordinarily would.

Another strategy I’ve found helpful is to play the Aria’s bass line alone at first; then just the left hand parts before adding the soprano melody. Since this bass line is the underlying, unifying principle of the work it’s worth getting to know it well on its own and be able to always hear it, and know when to draw attention to it over the course of the whole piece.

We’re very fortunate that there are excellent and very interesting performances on recordings now, and there are several that are very instructive and thought-provoking. Both of Pierre Hantai’s, the original on opus 11/Naïve, and then the more developed, detail intensive one on Mirare; Richard Egarr, and Carole Cerasi. All are worth listening closely to, and not trying to copy, but to understand a sampling of the choices that can be made, and the importance of so many very small details.

Posted in Uncategorized | 1 Comment

Warming up routine

I’ve developed a daily warm-up method; it’s not long, maybe 15 or 20 minutes. I believe it’s an important process. I’m not sure what other players do, though I am curious. Ton Koopman uses 4 Cramer studies, and those are very good indeed. Mine are very simple, deceptively so. They serve to relax the fingers, arms and body, connect the fingers, ears and brain, and focus mostly on small non-simultaneous plucks as well as exactly when the note is released/damped.

I begin with do-re-mi-re-do, back and forth in parallel 6ths maybe 4 times, starting on c below middle c in the left hand, the right hand on a. Using only fingers 2, 3 and 4. All notes held down, only raised to repeat that note when it comes around again. Then working up the degrees of the C major scale. Same thing, next on the D major scale; Eb, E, F, G, A and then Bb major. Kind of like a meditation, become aware of how relaxed the arms and hands are, how well you’re sitting, the quality of the sound. Be aware of tiny non horizontal plucks between the hands, and vary them. Notice where your wrists are, how curved or horizontal your fingers are, how close the keys they stay. I found this exercise improves the ease and quality of trills immensely.

The next one is similar, but you only over-hold two notes at at time. The pattern is do-re-mi-do, re-mi-fa-re, mi-fa-sol-mi, et cetera. Starting with the C major scale, again in parallel sixths, going on to the other scales as in exercise 1. Notice how the left hand plucks just a little bit ahead of the right, then switch them. The key is releasing do just when you pluck mi, and when you release mi you release re at the same time. Koopman showed me this one, and it’s very helpful for achieving precision in the timing of note releases, as well as making 2-note over-holding an easy instinctive thing. I do this one up the scale, and then in reverse down the scale.

My third pattern is an ascending up the scale and then descending triplet pattern: do-re-mi, re-mi-fa, mi-fa-sol, et cetera, with the three notes overheld. Fairly quickly, but never faster than you can control the non horizontal plucking, and never so fast that you lose the suppleness and total relaxation of the hand and arms. Then, I start over again, but with the left hand staccato (the fingers scooping inward) while the right hand is completely overlegato, and then reversed (the right hand staccato and the left over legato), and then both hands staccato. All the time, maintaining awareness of how subtly separated the plucks are between the two hands.

Finally I progress to using all four fingers (not the thumb though), the pattern in C major being do-re-mi-fa, re-mi-fa-sol, et cetera ; rising an octave and then descending back to your starting point. Both hands overlegato, then the right over legato while the left is staccato, then reversed. This one I just do in C major, because I think that’s just enough, most days.

Of course this isn’t a comprehensive method, it’s just a brief warmup. Yes, a few arpeggios are sometimes helpful to add. But the point of this is to connect the brain, ear and fingers to the fine details of over holding and non simultaneous plucking before diving into the day’s music and it seems to be very effective for that.

Posted in Uncategorized | 1 Comment


Geez, I feel bad; haven’t written anything here for quite a while. Have been playing a lot of early english music; Byrd, Tomkins, Bull and the like. Amazing music, masterful polyphony and melody. There are so many masterpieces, little ones and big long ones. It’s great music for improving musicianship and expression, the fine pieces support a lot of exploration of possibilities.

I offer an update on harpsichord touch. Recently I sliced a wee piece of my right thumb off during a disagreement with a mandoline in the kitchen. All healed now, nothing serious. But for a week or so, no functional thumb. That’s all right, the early music really doesn’t need much thumb. So I took the opportunity to develop a thumb-deprived touch and hand position. Very interesting. My solution was very flat extended fingers – a neutral plane from the elbow to fingertip, fingers parallel to the keys most of the time, but diagonal when more comfortable. Total relaxation. I mean, a total absence of any perceptible joint tension anywhere from finger to head (not easy or trivial). The fingertips touch the naturals, most of the time, on the very edge of the keys. The hand sort of floats there and as you play, as best I can describe it, it’s a combination of stroking and letting the weight of the fingers depress the keys. Very minimal movement. It seems to me there are two kinds of motion going on; a very little bit of the fingertip inward stroke motion and also a hinging motion at the very top joint between finger and palm. And what I found was a huge improvement in sound quality, more control over every aspect of touch – overholding and releases, and in general a much more supple musical result. As I practice this in, I find that so many things are easier and come out better and more fluid. As long as you retain the body’s sense of effortlessness everything works better. I have read, heard and seen plenty about ‘early’ fingerings without the thumb, but I haven’t seen it applied this way, with the flat fingers. It seems so natural and comfortable I wonder if this was done back then. Have to hop in the wayback machine and see. Back in a min


OK, I’m back. The only thing I’d modify is to gently raise the palm of the hand, letting the fingers hang or maybe droop some. Not too much. Make sure you’re sitting at a height comfortable for this, elbows a bit higher than the keyboard so it’s not an effort to raise the hand a little. Now, the fingers stroke inwards a bit, with very little effort. This is hard to describe in a way that’s accurate enough. The other thing I’d mention is that all the experimenting with touch and finger ergonomics results in all the joints of the fingers being activated, so that in the end the fingers move in a pretty complex way. This seems ideal.

Posted in Uncategorized | Leave a comment


Time is the most important dimension of music. It contains rhythm, tempo, meter, memory, and their close relatives anticipation, predictability and surprise. A deep understanding of time as an expressive element is essential to an interpreter. Looking at a score doesn’t tell you much about where time plays an interpretive role. Baroque scores don’t give very much information about that at all; perhaps that’s why so many players treat time as an inflexible and precisely subdivided thing and their playing sounds dry and lifeless. Of course, it’s not. In music, time breathes, and contains worlds of potential.

Many of the expressive opportunities of time are very subtle; for the harpsichordist it’s not just about well-judged use of rubato. Minute lengthening of a beat can make that beat louder; separating notes that in the score line up vertically creates a hierarchy of louder and quieter, which can also create a kind of three dimensional texture. The shape of the pace of an arpeggiated chord define the kind of energy that is conveyed. All these, and more, are expressive aspects of time. You can’t ever master this; it’s an endless exploration of possibilities. Creativity and imaginative use of time, when done well, is the hallmark of a fine player. Its absence defines a dull one.

I haven’t seen much written about this, and I wonder why. The old school great pianists understood these concepts very well, but it’s less in evidence in most of today’s players.

Posted in Uncategorized | Leave a comment

They’re not hands

You should not imagine them as fingers; Skip Sempe likes to say that harpsichordists don’t have fingers, they have hands because pianists can make do with fingers since they have the sustain pedal, while a harpsichordist has to play with the whole hand.  I have an image that works for me; they’re tentacles.  It’s not to be funny. They have independent nervous systems but can work together, they slither and ooze around, they have no bones, just supple rubbery gracefulness; they move in arabesques, and they can independently play notes and hold down others.  That’s what a harpsichordist has instead of fingers.


Touch is such an important thing to focus on for a musician, because it’s crucial to beauty of tone and gracefulness of phrasing.  At the keyboard everything I do is to improve touch, because from there the musical ideas flow more naturally.  The first thing to do is breathe out any tension from the hands, arms, shoulders and head – as if you’re floating in warm water. Feel the stillness. Then imagine the sounds that will follow – is it a chorus of lutes? A choir? An orchestra? A viola da gamba solo? With an excellent sense of touch, and clarity of musical thought, those first notes can be magical.

Posted in Uncategorized | 1 Comment