On Teaching: Further thoughts about rhythm, part 4

Gavin Black is director of the Princeton Early Keyboard Center, Princeton, New Jersey.

Further thoughts about rhythm, part 4

After writing my last column and in the course of searching for further ideas about rhythm, I came across this quote, which was new to me and I quite like. It is from a review of Rhythm and Tempo by Curt Sachs published in 1953 in Journal of Research in Music Education, written by Theodore F. Newmann:

The essence of all musical expression is that of creating a feeling of organic movement. And that factor, above all else, which contributes to a work’s organic unity, its total feeling, is rhythm. Rhythm is not a matter of divisions of time and accent but rather a relation of tensions—the preparation of new expectations by the resolution of former ones. It is rooted deeply in the framework of all living organisms. It permeates the very body of music—its tempo, melody, harmony, and form. Rhythm is, in fact, the most vital element in music.

I might quibble about the absolute terms in which some of this is expressed. Rather than “the essence” I might say “an essence” or “one essential part.” Despite my focus on rhythm right now, I would not say that it is “the most vital” element in music—certainly one of the most vital, and one that, defined broadly enough, must be present.

What I really like is the part about expectations. This is a powerful way of framing the theoretical background for my second mode (as per last month) of determining rhythm in music. The ebb and flow of tension, the creation of expectations—and then the fulfilling or the subverting of those expectations—is the source of the rhythm of my second sort. And this operates at every level, from a whole piece to the motion from one short note to the next. Newmann’s quote is a beautiful expression of the notion that rhythm can be derived from and described as something other than counting beats. For me, this description also establishes that the expectation-based sense of unfolding rhythm is primary. It is not just a system of departing from the rhythm created by the beat structure. If anything it is the deeper, more essential source of rhythm.

How is it possible to reconcile the simultaneous existence of these two radically different ways of defining, projecting, and perceiving rhythm in one piece of music? I do not know if I have a thorough or systematic answer to that. Both forces being discussed here are real, and therefore they do coexist pretty much regardless of whether we know why. Not having an answer to how or why that works may reflect something about how flexible it is in actual practice. It may also relate to the phenomenon of people responding differently as listeners to the choices made by performers. After all, expectation is a listener’s experience. And if a piece or passage of music creates and then fulfills or subverts expectations, then the actual flow of that experience is very likely determined in large part by the listener’s experience, training, philosophical stance, temperament, and so on. Different listeners will hear what is being performed in a way that is literally different.

If a piece of music is ostensibly measured and structured in a regular meter, then any overt conflict between the two approaches to rhythm would take this form. The intuitive, listening-based approach might well lead to something other than playing the rhythm as written, and playing the rhythm as written might lead to ignoring the ebb and flow of tension and expectation. One axis along which a solution to this conflict might exist is that the sense of pulse is remarkably strong and resilient.

My experiment with Helmut Walcha’s recording of Sweelinck, as described in my December 2021 column, page 11, is a manifestation of this resilience. As I wrote then, my reason for testing out the literal metronome steadiness of the recording in the first place was that the performance came across to me as remarkably and inexorably steady. So it was interesting and telling that the metronome reading varied so much across the piece. I have conducted similar tests on other recorded performances. For example, about eighteen different recordings of the beginning of the first movement of Charles-Marie Widor’s Sixth Symphony yielded similar results. Some of my odd driving-based experiments about time have this same import.

A Chopin anecdote

There is an anecdote about Chopin that I have come across in writing twice over the years, though I cannot remember where exactly. One version framed the story around his mazurkas, the other around his waltzes. It works equally well for either, as they are in ¾ time. The point is that a friend of Chopin’s said to the composer that when he played one of his own waltzes there were four quarter notes in each measure, not the notated three. Chopin denied this. The friend made him sit down and play and kept strict track of the timing, and Chopin was indeed taking four quarter notes worth of time between bar lines. This story implies three things: that Chopin was playing very freely, holding some supposed quarter notes for as much as twice as long as they “should be” held; that there was nonetheless something that could be identified as “the quarter note;” and that Chopin himself intuitively accepted what he was doing as being a manifestation of three quarter notes per measure.

It seems also possible that if the intuitive, expectation-based approach to rhythm is applied thoughtfully in a way that is really derived from careful, involved listening, then it might enhance the sense of regular pulse while being less regular than a purely beat-based rhythm would be. If the sense of momentum that is associated with the concept of regular accent-based meter is as robust and as difficult to destroy as I suspect, and if an expectation-based shaping of rhythm can make the ebb and flow of intensity the most compelling that it can be, then it could make sense that the latter approach would make the accentuation patterns that define regular meter more rather than less convincing.

The actual results of the two approaches do not always have to be very different from one another. For some performers and listeners experiencing repertoire, the listening-based intuitive approach might well yield results that are quite regular in their actual rhythms.

In musing a bit about how to introduce this idea to students, I start by quoting something that I wrote in my column from April 2015, page 17:

I often suggest to students the following practice tool. Once they have identified a spot where they want to make a rhythmic gesture (usually of the sort that might be described as “rubato”) they practice that gesture, in the privacy of the studio, in as exaggerated a manner as possible: take the risk of executing a gesture that is utterly tasteless, mannered, “schmaltzy.” This is to counter the fact that we usually only visit the gestures that we think we want to make “from below” (so to speak), that is, only compared to and judged in comparison to not making such a gesture, or to a modest version of the gesture. This stems from and then reinforces a philosophy that teaches a kind of reluctance about such gestures. If you hear a rhythmic inflection from both sides, you get a different sense of exactly how it might be effective.

When I wrote the above, I invoked the word “rubato.” Now I would be inclined to avoid that, since it is a term that relates the flow of rhythm to the underlying regular meter. The relationship is real, but I would want to encourage the student to move away from considering that relationship to be primary or of overriding importance. In applying this practice suggestion with many students over many years I have observed that almost everyone finds it easy to identify spots where they intuitively want to create rhythmic gestures that are not identical to what a strict reading of the metric notation would suggest. While these gestures fluctuate in number from piece to piece, it is more common for the student to believe that their interpretation of some of them are wrong.

Experimenting with notable works

As an experiment, try giving a rhythmic inflection different degrees of magnitude—including the highly exaggerated, but not limited to that—and try to experience and analyze what is happening with each different shape. For example, consider the very beginning of J. S. Bach’s Toccata in D Minor, BWV 565. Play the opening mordent and then listen for when you think that the following scale passage “wants” to come in. Try taking longer next time. What has changed? Have you lost the moment when the impact would have been greatest, or does the added suspense increase the impact? Does it sound like you caught a sort of wave of expectation when you initiated the scale passage, or did you miss it? If you missed it, could there be another one coming up, different yet also fulfilling?

Questions like these are easy to highlight with a moment in an intrinsically free passage like this, but they are potentially present all the time. Returning to the pieces that I invoked at the beginning of this series of columns, I would say that the first barline in the opening of the first movement of Widor’s Sixth Symphony is also a place where the dynamic of this sort of rhythm is clear. A student can play it metronomically and then try various timings, listening as closely and open-mindedly as possible. What is gained or lost by various different timing possibilities? What does the timing at that barline imply or suggest or even necessitate about playing the second measure itself? Would the answers be different if this were not the opening of the piece? Is the answer to that question different if we stipulate either that it was a recapitulation or, on the other hand, that it was a new idea introduced in mid-piece? Does the use of this motif later in the movement as written affect choices here at the beginning?

It is interesting to listen for intuition-based timing of quick notes within a texture. In Example 1, in this Mozart Piano Sonata in A Major, K. 331, how can we determine exactly where to play the sixteenth notes?

My own approach to figuring this out would be to leave those notes out at first, making the rhythm of the first half of each of the first two measures the same as the rhythm of the second half. I would play this a few times, just like that, as if it were the piece. Then I would play it a few times hearing the sixteenth notes in my head, but not playing them and then add those notes back. I often do not know and cannot seem to determine as a matter of counting and calculation where I want to place short notes after dotted notes. To use a piece I have played a lot, I have no idea exactly what ratio I give to the pervasive dotted rhythms in Bach’s Prelude in E-flat Major, BWV 552i. I can only hear whether they do or do not seem to do the right sort of job in creating momentum from one long note to the next.

When one asks a student to listen carefully to every moment in a passage and to discern as much as they can about rhythm through that listening, interesting things happen. Some of these are advantages while others are pitfalls. One great advantage is that this tends to keep the student from rushing. Not everyone has a tendency to rush, but most of us do at least some of the time. One reason for rushing is that we pay the most attention to the things that we must do physically—and for organ and harpsichord that means initiating the notes and releasing them. As soon as one note is safely played, there is some pull toward either the moment of release or the moment of initiating the next note. The sort of listening that I am describing here focuses the attention as much on the middle of a note as on the beginning and end.

Listening this way tends to lead to slower playing. That has been true for me, to a fault some of the time, though it is a perfectly natural outcome of this approach. The way to counter this tendency is through awareness and honesty. This listening approach is both a practice technique and something to bear in mind while actually performing. As with any other sort of practice, it is important and good to utilize it at a slow tempo, but as with any slow practice, it should be ramped up to the desired tempo in a systematic way.

It is natural to prepare for the moment when you believe the next note should occur and then play that note. This makes that moment in the music late by definition and leads to slow playing. It is probably acceptable for this to happen early in practicing a passage, as long as you are aware of it and work on evening it out as you get to know the music better. I have also been guilty of this in performance, not recognizing it clearly enough in practicing, and therefore not working to smooth out the process.

This is analogous to something that can happen when using a metronome, when one listens for the metronome beat and then plays the note. Something like this can also happen with releases. If you hear the instant at which you want to release a note and then release it, the release is late. A release that is executed by pushing off—down, then up—will also be late if the downstroke is timed to the desired moment of release. There is a sort of leap of faith when executing something, whether the timing of the beginning of a note, as we are discussing here, or of the end, without waiting to confirm that the looked-for time has arrived.

A few miscellaneous points

What is the meaning of any given note value? We know the answer: a quarter note is equal to two eighth notes or to half a half note; a whole note is four quarter notes, etc. But here’s another definition: a quarter note (or whatever note value) is whatever a listener will hear or accept as a quarter note. This relates directly to the Chopin story above.

Watching a TV game show recently, I heard the host say to a contestant, “Time is ticking away!” Why? Why is time defined as made up of discrete entities, ticks of a clock? Is it just because old clock technologies produced audible ticking? Or is there really something more fundamental about time that requires us to think that way? This vignette from a fraught moment in a game reminds us that thinking of time as ticking tends to create a sense of impatience or even panic.

The beating of the heart is sometimes evoked as a possible source for the notion of beats in music. This makes sense because the heart does beat fairly regularly in the absence of a medical issue. That beating is audible to the one whose heart it is, though not usually in a way that rises to the level of conscious awareness. If you quiet the outside world, you will easily hear your heart beating. But no heart beats as absolutely steadily as the beat we hear from a metronome. If the heart is a model for musical beats, maybe it should confer some flexibility onto those beats. If the heart is the model for a beat-derived sense of musical rhythm, breathing should be the model for a sense of rhythm based on expectation and fulfillment.

Gavin Black is director of the Princeton Early Keyboard Center in Princeton, New Jersey.

Further thoughts about rhythm, part 2

Continuing from my September column, I offer here a few additional anecdotes and ideas relevant to rhythm and the teaching of rhythm. In my next column I will explore the question of how to teach rhythm or “counting” in a way that enables a student to connect those concepts as directly as possible with rhetoric and the student’s interpretive stance.

I begin by revisiting the anecdote from the last column about my colleague who, in the eyes of a particular third party, “couldn’t count.” Since I wrote that, I have had further relevant thoughts. I remember that as I walked into the rehearsal studio after that exchange, I was very sorry to have been told my colleague’s opinion. I did not want to approach my work with this fellow musician with any preconceptions. And while I certainly did not assume that the older colleague was right, I found it hard not to have a question layered on top of my focus during rehearsal. “Was that rhythm correct? Is this where I learn that she indeed cannot count?” That was distracting.

I wondered why he wanted me to focus my attention on her rhythmic deficiencies. Why did it matter in the overall picture? Why tell me as I headed into the rehearsal that “she can’t count” rather than that “she is an amazingly compelling performer?” This seems like a manifestation of bias in favor of that which can be quantified—that which seems more objective or safer to describe.

Since that day decades ago, I have become very aware that there is a strong pull in all sorts of areas of life to focus on things that can be measured and described in a way that is objective to some extent. It is a truism that in evaluating a competition, for example, or even just in reviewing a concert or a recording, it is relatively easier to note that a performance either does or does not get all the right notes, or that the rhythms are or are not all in strict accordance with the notation, or that the tempo is or is not the same as what the composer requested.

It is much more difficult to describe how expressive or moving a performance is. This is not a bad thing, and competition judges and reviewers certainly do not fail to grapple with the elusive side of things. But this has some bearing on teaching rhythm and counting. On paper, rhythm is one of the most objective matters about music—at least with our familiar rhythmic notation. Quarter notes are twice as long as eighth notes, etc. In a way, some of what is non-objective can even be described objectively. Certain dotted-note patterns, for example, are not meant to be what they apparently look like on the page, and we can often describe the ways that they differ from that presentation.

But how do we teach subtle variations in rhythms that look the same on paper? Is it an acceptable part of our understanding of rhythm for there to be such variations? If we think not, is that for musical reasons, or is it because of this pull toward the objective? If a quarter note is really always exactly twice as long as an eighth note, if all the eighth notes are really exactly the same as one another, what does that mean for a student or for how we teach? And if they are not, what does that mean? Is it possible that rhythmic notation means “these notes should be in any rhythmic relationship that would be expressed more closely this way than any other way?”

Another story from many years ago: I was coaching an amateur chamber music group—a violinist, a flautist, and a pianist who was exploring the harpsichord as part of this project. At that time, in my own work as a player I was exploring Sweelinck and beginning to discover some ideas about freedom of rhythm. At one point I played a segment of a Sweelinck toccata for the members of the group and drew their attention to a particular passage. This was one where I felt that the rhetorical force and expressiveness of the music could be enhanced by playing very freely. I had an approach to that passage that involved drawing out the development of harmonic tension and, when it was resolved, using timing to make it seem difficult to go on. That is a trite and inadequate description, but the point is that it called for freedom, and the musicians in the room found it very effective. I also played through the passage as written, and they found that almost embarrassingly boring and pointless, which was exactly what I intended to convey. So far, so good. But when we had finished this and were ready to get back to rehearsal, one of the musicians said, “Of course, you couldn’t do that with Bach.”

So, the question I had, and still have, is, “Why not?” Anyone may or may not appreciate the rhythmic choices performers exercise. In theory, there could be a plausible analysis of a Bach piece that concludes that free rhythm, or any particular approach to rhythm, might not be effective. And the question of what the composer himself would have thought always looms over our thinking about an issue like this. But none of that is what was meant by that remark that day. We discussed it, and the ensemble member specifically meant that since listeners’ expectations about Bach were pretty well formed already, unlike their expectations of Sweelinck, it would be imprudent to go too far in violating those expectations. To some extent, the way that music had been played and heard in then-recent decades had become part of the actual identity of that music. This seems to be another way in which something objective can gain a kind of privilege or priority that it might not have earned in any substantive way.

Another question that I will delve into more next month is, why rhythm? That is, what is the goal of having music organized into regular or somewhat regular micro units of time? Is it to create a sense of pulse or momentum? Is it something about comparability of experience throughout the duration of a piece or a movement? Is it very specifically about creating the palpable sense of a regular beat in the listener’s ear?

I mention it here to introduce an experiment I once conducted that was predicated on the observation that many people are of the opinion that Helmut Walcha’s recordings are rhythmically conservative. I obtained an interesting result.

I had always reacted to Walcha’s recording of Sweelinck’s Fantasia Chromatica as having both extreme steadiness and inexorable forward momentum. At the time that I did the experiment, I took it for granted that the way to achieve those qualities was to keep the beat very steady. But I must have suspected something that led me to investigate. I put on the LP and measured the beat at the opening. I then picked the needle up and dropped it a bit farther in. I measured again. I did it a third time, maybe a fourth. They were all quite different. This led to the intriguing notion that maybe inexorable forward momentum, and even the very sense of steadiness itself, might sometimes come from something other than regimented sameness of beat.

I have now repeated this exercise using more modern methods: the piece playing on my computer and an online tempo tool. The beat at the beginning hovered around 118, and later on it was over sustained passages as high as about 126 and as low as the mid 90s. That is very much like what I measured thirty years ago. I am intensely interested in the relationship between literal sameness of beat—or departures from that—and a subjective sense of steadiness, momentum, and pulse. Is it possible that sometimes a performance that features a doggedly steady beat comes across as uneven to listeners? If so, how can this be?

There are two games that I have played while riding in a car that both have to do with the use of time in music. First, if you are riding along a fairly busy two-way street or highway, pay attention to the sound of the cars passing on the opposite side. On a busy road cars will space themselves out almost regularly, since all else being equal, people pay at least a little bit of attention to following distance in front and behind. But there’s an emphasis on “almost.” The line of cars is never spaced out exactly evenly. So as the “whoosh” of each successive car goes by, see how far from even those sounds can be and if you can still accept them as conceptually even. How far apart can sounds 2 and 3 be—compared to the time between sounds 1 and 2—and can you still hear that timing as rubato or agogic accent rather than just discontinuity? For me there is a wider range of timings that I can assimilate to evenness than I would have expected.

If you can accept a stream of not quite even cars as conceptually even, is there anything interesting about the shape of the unevenness? Does it have any rhetoric to it, groupings or patterns of weak and strong beats, or impulses? One fascinating feature of this exercise is that each instance of it is ephemeral: you can hear it only once, never again the same way.

Another exercise I enjoy while driving is this: when going under an overpass, I try to experience the time in the shadow of that overpass as lasting forever. Since it will come to an end, usually in seconds or less than a second, I attempt to experience it as simultaneously brief and infinite. This feels even more intense if it is raining. It seems to me this has implications for rhythm in music. The ostensible rhythm, flow, motion of any increment of music is always about the next thing: when will the next beat come? But the state of being of each note, harmony, sonority, or beat, is also a thing that exists for as long as it exists, and that has identity and importance.

(It is probably best to do the above exercises as a passenger rather than as the driver, unless you are very sure that none of it will distract you from driving. I wonder how different any of this will seem if we have nothing but self-driving cars on the road!)

Are there other little slices of everyday life that might illuminate aspects of rhythm and the role of time in music? Rain is an interesting one. There are pieces of music that are inspired by or attempt to depict rain. What is the rhythm of rain like? It is more regular when it is fairly light. What about walking? How regular is that, and how does walking respond to outside conditions? Can one walk while hearing in one’s head a passage of music that moves at a pace different from the footsteps?

It occurred to me a few years ago that I often experience trills as containing some of that simultaneously brief and infinite quality that I get out of my second driving exercise. I sometimes suspect that that is what trills are essentially about, though that is speculative and unknowable. The fast and unmeasured quality of (usually) the middle portion of a trill seems to take it out of time, while, like all music, it is in the end bounded and defined by time.

I will pick this up again next month.

Further thoughts on
counterpoint II: a miscellany

I mentioned at the end of last month’s column that I have been rethinking some of my ideas and practices about teaching rhythm and counting. This is the result of recent experiences teaching students contrapuntal voices. I have noticed—more than I previously observed—that moving from an understanding of the rhythm of individual voices to an understanding of the way the rhythms of those voices fits together is not always easy or natural for everyone. It is helpful to address it directly, and I developed some new ideas about how to do so.

As outlined in my April 2012 column, my approach to teaching rhythm and counting involves ignoring time signatures, bar lines, and the “one-and-two-and” structure that we often employ. Rhythm is about ratios, and my experience has suggested that grappling with ratios directly—understanding that a dotted quarter note lasts as long as three eighth notes, that a half note lasts twice as long as a quarter note, and so on—is the best way for a student to become comfortable with rhythm. This is true for a beginner learning to understand rhythmic notation for the first time and also for a more experienced student who is having trouble with a particular rhythm or needs a refresher course. This has the advantage of being simple and completely accurate. I have seen much more trouble arise when students worry about whether a particular note is “on the ‘and’ of two” than I have seen clarity arising out of this way of describing rhythm.

This is an over-simplified summary of an approach that differs from other approaches in emphasis, since the learning of note ratios is always fundamental. This is not the place for a complete restatement of what I wrote six years ago, or to explain why I believe that it is a good approach. You can find it in the above-mentioned issue of The Diapason, which I have posted at http://gavinblack-baroque.com/Diapason column April 2012.pdf.

When a student needs to work out the rhythm of a passage, I suggest counting individual notes in relation to a suitable short note value. For example, if that short unit is the eighth note, then each eighth note is counted as “one,” each quarter note is counted “one-two,” each half note “one-two-three-four,” and so on. This does not involve denying that those beats also relate to positions in a measure, but it just deemphasizes that at first. In learning the rhythm of a single line or passage, the texture of which involves all of the notes, voices, chords, etc., being in the same rhythm as one another (say in the manner of many hymns), this approach is very successful.

Therefore, it is also successful when learning individual voices of what will be a contrapuntal texture, regardless of its eventual level of complexity. So the following two “voices” (the pitches are not important) could each be counted as indicated in Examples 1 and 2. As with any rhythmic counting, the numbers have to come at an even pace.

Each of these examples contains nineteen eighth notes, and the two of them fit together. But the sort of counting that I have suggested would be extraordinarily hard, if not impossible, to follow in Example 3.

The first-order solution is that the rhythm of each voice should be internalized before the voices are put together. This is a goal worth striving for in the practicing process. However it seems sensible to readmit traditional measure-based counting at this point.

I have purposely created an example that is not divided into measures. But it could be, like Example 4, which then allows for the regular (“one-and-two-and”) approach to counting.

I believe that it is a good idea for this sort of counting to be secondary. To the extent that working out the rhythm of each voice separately is something that must be approached consciously and deliberately, I think that my per-note way of counting is useful. In addition to being sensible and effective in itself, it can promote an awareness of the importance of the middle and ends of notes, whereas “one-and-two-and” counting, especially with counterpoint, can focus attention disproportionately on the beginnings of notes. As soon as the measure-based number for a note has been spoken or thought, the next step is to see what notes in other voices correspond to the next counting numbers. This can shift a player’s overall focus toward a more homophonic feeling.

Thinking about applying the normal measure-based templates to counting multi-voiced pieces has made me more open to other ways of keeping track of the rhythm of pieces that are not derived from a strict awareness of the voice structure. For example, there is net or overall rhythm. Sometimes, while the rhythm of each voice in a contrapuntal passage is different from all the others, the surface rhythm is quite simple. That rhythm might be thought of as a new note every eighth note.

Here are a few examples of writing in which this concept might make things easier. Example 5 is from Bach, Art of the Fugue, Contrapunctus 8. Example 6 is from Mendelssohn, Prelude and Fugue, op. 37, no. 1 (fugue). Example 7 is taken from Bach, Well-Tempered Clavier, Book II, Fugue #2.

In the last of these, the net rhythm is more complicated, but still simpler than each individual rhythm. It is interesting that the first three measures of each piece have the same overall surface rhythm, which in turn is the rhythm of the fugue subject. In the opening measure, the fugue subject is alone, in the second it is combined with a countersubject that does not change the surface rhythm, and in the third measure only the surface rhythm alludes to or quotes that subject.

If the printing is neatly aligned, it is possible to derive all of the rhythm from the surface rhythm by how things line up. Sometimes it is useful to picture a cursor line going steadily from left to right and to identify whatever it touches as what is to be played at that moment. But this should all be secondary, understood to be forms of assistance, not a primary way of understanding rhythm.

Some observations

Here are a few more observations about counterpoint, some of which may be a bit abstract:

It is natural to think of counterpoint as being something that is easily differentiable from harmony. After all, there is counterpoint that exists outside the norms and expectations of tonality or functional harmony. But for most of the music that we play on the organ and other keyboard instruments, pieces that are rigorously contrapuntal are also completely embedded in harmony. This is something essential about the meaning of the counterpoint itself: that is, the way that the flow of contrapuntal voices/melodies comes across to us, and therefore the way that the directionality of the different voices interact with one another function, is determined crucially by where the voices are moving harmonically.

For this discussion, I have invented an exercise. Try playing a piece of contrapuntal music with all but one of the voices transposed, and not each transposed by the same amount. The simplest case would be something like Example 8, from the opening of the first Bach Two-Part Invention.

This is going to sound discordant, to put it mildly. But that reaction or judgment is not necessary. The point is that it sounds radically different from the “real” version, although, analytically, a good deal of the counterpoint itself remains unchanged. A little bit of this exercise, possibly even just the amount that I have written out here can be eye-opening for a student. (It is not bad as an odd sort of sight-reading exercise either, and it is much harder with more than two voices.)

A bit of exposure to this oddity can also make the ears more sensitive to the ebb and flow of harmony-based motion or pressure in contrapuntal works. Even though the interaction of voices is always shaped in some way by the separate voices’ harmonic direction, the evident force of that phenomenon is stronger in some places, yet weaker in others. It is also more divergent in some places and more in sync in others. The places where the force of the harmonic direction of different voices is most in sync are cadences and moments that feel cadential.

Many years ago, probably when I was still in high school, an older friend of mine recounted something that a friend of hers had said. This friend once removed was an extremely erudite and thoughtful music listener, thinker about music, and keyboard player. The comment that was relayed to me was: “Helmut Walcha’s problem is that he is too focused on the counterpoint.” I immediately felt that I was learning something—not so much about music or playing or performance, but about attitudes toward musical work. The first thought that occurred to me was that Helmut Walcha was not the one with the “problem!” The phrase “Helmut Walcha’s problem is . . .” really meant, “The reason that I do not like Walcha’s playing is . . . .” That comment and my reaction to it at the time taught me that many people take their own reactions for objective truth, and that it made me uncomfortable. That in turn has shaped a great deal of my own approach to my development as a player and teacher.

What does that comment tell us about counterpoint? I have never thought that it was an apt description of anything I heard in Walcha’s playing. That may be in part because I grew up listening to him, and, at that time, his approach was my own default. But it has made me sensitive to the question of what it would mean to pay too much attention to counterpoint and too little to other things.

My guess is that this commentator was probably referring to Walcha’s relatively steady approach to rhythm, and also, just as importantly, to his focus on consistently phrasing subjects. (For what it is worth, he was also reacting to hearing recordings, not to any conversations with Walcha or to any direct knowledge of the roots of his approach.)

This leads to a host of questions:

Is it correct that relatively steady metronomic playing is associated with bringing out counterpoint clearly, whereas playing that is rhythmically free is associated with harmonically derived music, or with an approach to playing music that is more focused on harmony?

What about the business of “subject phrasing?” If a phrase, theme, or motif happens more than once in a piece, is it important to articulate and phrase it consistently? Does doing that enhance the extent to which the counterpoint seems like counterpoint?

Is it important, or is it good or bad, for the notes of different voices that are written ostensibly at the same time to happen at exactly the same time? Could there be something good or necessary about notes being somewhat staggered at times? The reason that I associate this question with the Walcha-related anecdote is that in his playing, notes do line up rhythmically to a great extent. Furthermore, the above-mentioned commenter believed that Walcha was too focused on counterpoint. But there are many players and theorists who believe that counterpoint is enhanced by staggering: it is difficult for the ears to follow separate lines if the notes of those lines coincide with one another exactly. Which of these is true? Does it depend in part on habits of listening? Is it really about clarity of counterpoint, or is it about some other aspect of style?

How does this relate to thoughts about basic counting in counterpoint expressed above? I suppose that I feel most interested in the notion that any melody including a contrapuntal voice has some logic by which it might be stretched out or otherwise inflected rhythmically. Also, two voices that occur at the same time might have different logic. They might not line up all the time, but not because the phenomenon of their not lining up should be sought out, rather, because the internal logic of each has them manifesting slightly different timing. This may be why I am more interested in counting individual voices that are somewhat removed from each other and from the template of measures. But then all voices have to be together enough in the end so that the piece is coherent.

I present all of this as questions or ideas that are interesting to think about or, more to the point, to engage with in conjunction with students. None of it is hard and fast; none of it represents answers or conclusions.

One last topic: notes that we play have overtones. On the organ some of those are the inherent overtones of the pipes, while some are the higher stops added for a particular purpose. Especially in the former category, there are often overtones that, while they blend in enough to make the sound coherent, can also be heard as faint separate notes. And the rather cool thing is this: different overtones are relatively audible on different notes. So in playing a theme you might hear the twelfth over the first note, the fifteenth over the second note, the twelfth again, then the seventeenth, and so on. In Example 9, in this melody, you might happen to hear the higher (gray) notes as very faint, audible accompaniment to the “official” theme.

This is a kind of shadow overtone counterpoint that is always present, at a greater or lesser degree of audibility, and is present separately in multiple contrapuntal voices when they combine.

Gavin Black is director of the Princeton Early Keyboard Center, Princeton, New Jersey. He can be reached by email at [email protected].

The harpsichord: an introduction, part 2

At the end of my previous column (August 2024, page 8), I noted that I would next turn my attention to a detailed discussion of harpsichord sonority. In thinking about that subject recently, I have realized that I should start by returning to the question of variety or non-standardization. This can affect our understanding of what is and is not a harpsichord, and what is or is not harpsichord sonority. Some of this variety is about setup—keyboards, stops, compass—and I will address these issues in my next column, including discussion of the mechanism of the harpsichord and what that implies about touch and technique. But some of the variety has to do with the sound itself.

Sound variety

It can be useful to think of sound variety as existing along two separate axes. For over 300 years, the harpsichord was the predominant domestic keyboard instrument, and its sounds varied from both region and historical period. Surviving antique instruments are one principal source of knowledge we have about different geographic schools of harpsichord sound and how that sound evolved over the years. Examples built in modern times along the lines of the older, original instruments are another source of information. There are all sorts of interesting complexities about the conclusions we can draw from any of these instruments. For example, it is difficult to ascertain how the sounds of existing antique instruments have changed over the years, and it is also difficult to know how successful the sound of a given modern instrument is in recreating the sound of the harpsichord as it was in the Baroque period, assuming the modern builder is trying to do that.

This brings us to the other axis along which harpsichords vary in their sonority. As the modern rediscovery of the harpsichord began in the very late nineteenth century and especially as it became popular in the early twentieth century and thereafter, harpsichord builders made different choices about how to approach the rediscovery or reinvention of this long-dormant instrument. No one building a harpsichord in modern times has been working from an ongoing, living tradition of instrument building. The arc of this learning process was and still is very long. Some of the choices that builders have made have been motivated specifically by concepts relating to sonority. For example, some harpsichord builders, especially in the mid-twentieth century, were interested in tackling questions about tuning stability and mechanical reliability in a world with central heating and air conditioning. The gist is that there are many harpsichords out there that reflect different stages in their builders’ learning processes, and that manifests different choices or preferences about harpsichord sound as well as mechanics.

Much has been written about this modern history. Since there are many very different-sounding instruments out there that are identified as harpsichords, two important concepts should be considered. First, that one should not assume that one’s concept of what a harpsichord “should” sound like is necessarily indicative of all harpsichords. Second, the best way to know what is going on with the sound of a given harpsichord is to listen to that sound carefully, closely, and with as few preconceptions as possible.

There are two traps that are easy to fall into. One of these is hearing a harpsichord whose sound you do not like and deciding that you do not like the harpsichord, and the other is getting attached to the sound of the first harpsichord that you hear and thereafter never listening open-mindedly to other sorts of harpsichords. The first of these was very common indeed in the mid-twentieth century when there were many harpsichords around that were experimental and largely unsuccessful. The second, interestingly, is a trap that I fell into early in my harpsichord life. I was lucky enough that the first harpsichord I ever owned and spent a lot of time with was an instrument with a compelling, gorgeous sound. At the time I did not think that I understood how strongly that instrument shaped my sense of what harpsichord should be and thereby limited my ability to appreciate many of the very different beautiful and effective sounds that were made by other harpsichords. I eventually coaxed myself out of this by increasingly remembering to pay attention.

What are some of the salient features of the sound of a harpsichord? What if anything is universal, what are common threads to look out for, what varies, and how can one best understand the sound of a given harpsichord with a view to using it most effectively?

For me the starting place is something that seems more technical than aesthetic, though in the end it is crucially about aesthetics and performance. The sound of each note on every harpsichord has an intrinsic behavior. From the instant the sound starts, from the moment when the plectrum lets go of the string, the sound does what it is going to do without any input from the player until it ends. This may be more purely true of harpsichord than of any other instrument. With violins, voices, flutes, etc., the performer can change a lot about the sound while performing; with plucked string instruments, there are measures of control, more or less subtle in nature, over ongoing notes that can be exercised by the player. Even with organ and piano there are some small such things: damper pedal actions that affect the overall sonority in a way that changes an existing sound picture; swell pedals; on some organs, manipulations of the wind during sustained notes by the playing of shorter notes. Nothing like any of this exists with the harpsichord. On almost all instruments, to some degree the player can at least partially create the sound; with harpsichord, the player works with a given sound. This sounds a bit inflexible, cold, and limited. And it could be if the sounds themselves were not extraordinary both in what they seem spontaneously to sound like and in what possibilities they create.

But if every harpsichord note has some sort of intrinsic behavior, that behavior is very different from one harpsichord to another. These differences pertain sometimes to the simple quality of sustaining. Interestingly, there is a tradition of believing that the harpsichord does not sustain as well or long as the piano. This is not necessarily true. It is true of some harpsichords, yet definitely not true of others. What is the shape of the first tiny fraction of a second of sound, the “attack?” Does the sound then die away in more or less a straight line or according to some other shape? Is the attack the loudest instant of the sound? Does the timbre of the sound remain constant as it dies away? Do all the partial tones die away at the same rate, or does the timbre change as the note grows quieter and eventually ceases to be heard?

Answers to these questions can be found on some harpsichords out there. The only way to answer the questions is to listen, and I will discuss some listening techniques here, as well as ways to think about relating this analysis of the sound to playing music. There are some generalizations that are interesting and helpful if we remember that they might not always apply. Surviving, well-restored harpsichords from the sixteenth through eighteenth centuries tend to share certain acoustic traits, offering similar answers to the questions above and to other questions about acoustics. They tend to have a crisp attack—not all according to exactly the same pattern, and not necessarily loud, but all with immediacy and clarity, often seeming either bell-like or “plucky.” They tend to sustain longer than we sometimes expect with a harpsichord, long enough that in most music, all but the longest notes are vividly present for as long as they are notated. The issue of harpsichord notes dying away so promptly that they do not fill the space allotted is a false problem, at least on instruments of this sort. On many older instruments, the sound dies away not in straight line, but in gentle but perceptible waves, which create a subtle pulse to the sound. At the same time, the sound also changes in timbre over the length of the note.

There are two concrete and compelling effects that the above combination of traits has on the ways in which sonority relates to playing music. The first has to do with tempo and timing. If the sound has an intrinsic pulse to it, that exists in relation to time as such and is not created, shaped, or determined by the rhythmic picture of a piece being played. That means that since choices about tempo are also choices about how much of the lifespan of each note to use, those choices affect what the actual shape of the notes will be. A shorter note is not just shorter. Since it also uses a smaller part of the lifespan of each note, it has a different shape from what a longer note would have had. For example, if there is going to be a peak of a (gentle) wave 0.6 seconds after the beginning of a note, then if your tempo for that note is 120, you will never get to that peak; if your tempo is 90, you will.

On an instrument with a sound such as this, one can record a passage at two different tempos and then use computer magic to adjust them to being at the same tempo as each other, and they will not actually sound the same. The sonority-scape will be very clearly different.

The second effect concerns the relationships among notes that are occurring at the same time. If there are longer and shorter notes happening in juxtaposition to one another, then the longer notes will be heard more prominently while the quicker notes are going by. This is like the situation in a choral piece where one section has longer notes—and the conductor reminds the singers to do something through the duration of those notes: something as active, if subtle, with volume, timbre, vibrato, or whatever else. This can keep the effect of those longer notes from being static or from dragging down the momentum of the music. The intrinsic sound of a harpsichord can serve this same purpose to similar effect. This is one reason that the presence of lots of long notes in a keyboard piece does not necessarily mean that it should be performed on an organ rather than a harpsichord. The organ can sustain long notes indefinitely, but those sounds can also defeat momentum in a way that the sound of a harpsichord never does.

Everything from the last few paragraphs is just one set of possible examples of what one might hear in the sound of a harpsichord. The overall points are: one can tell by careful listening what the sound of a given harpsichord is like, especially in that it is largely set and does not change in performance; and what one hears in the sound may well have implications for performance choices and for how the instrument is likely to come across in performance.

There are some points that are more straightforward and that are common to just about all harpsichords. One of these is that when there are two 8′ stops, they differ from each other in timbre. Inevitably, one stop is flutier and one reedier: one gedeckt-like, one with at least a bit of the quintadena about it. (I will discuss this further in my next column, as it arises out of the mechanism and physical setup of a harpsichord.) Each individual stop tends to change in timbre from bottom to top: reedier at the bottom, flutier at the top. This also arises out of the physical setup.

Getting to know a harpsichord

I recommend the following as a set of starting points for getting to know a new harpsichord. First, move the music desk out of the way, as it blocks much of the sound from the player’s ears. (This is especially true in a wing-shaped harpsichord, less true with virginals, spinets, and upright harpsichords.) Make sure there is only one stop on, and know which one it is. Then play some individual notes, starting in the middle of the compass, moving up and down. One might do some of this standing up and leaning out over the soundboard. Hold notes for a long time, and listen to the whole span of the length of each note. Then play some bits of scales, chords, and perhaps passages from pieces. If possible, have someone else play while listening. One should start near the curve in the side of the instrument and then move out a bit farther along that same line. Slower pieces are better for this sort of close listening than faster pieces. Faster pieces certainly demonstrate some of what the instrument can do, but give you less of a direct line to what the sonority is like. When listening to someone else play, one should be slightly less analytical than what I am describing. Are you reacting to the sound as beautiful, compelling, loud enough that you do not have to strain, so loud that it is annoying, such a pleasure that you are reluctant to ask your colleague to stop and go back to playing yourself? All of these could well happen along with an infinite number of other reactions.

Next, go back and play some individual notes on each stop. Do you hear them at all differently now from when you did the same thing a few minutes before? What is the last thing that you hear before a note has fully died away: it might be an overtone; the answer might be different for different notes. The next step is to play a passage, trying to focus on listening to different parts of the texture: lowest notes, highest notes, and most importantly, notes in the inner part of the texture.

Try to do this with as few preconceptions as possible. Do some of the above with your eyes closed. This can help focus on listening, but it can also help banish preconceptions. Since there are interconnections between mechanical aspects of the harpsichord and sonority, I will continue to build on these ideas next time.

Stories and conversation

In mid-March, when I last sat down to write a column, the current health crisis was at a relatively early and very uncertain stage. I wrote that I hoped that by the time that column appeared in The Diapason things would be much better. I sit here writing now a week or two after that last column appeared, and this one will not be read for nearly another six weeks. It seems accurate to say that the situation remains dire and that the sense of uncertainty remains as high as it was then. While society is slowly starting to reopen, we will not know the effects of this action for quite some time. This very morning there are hopeful headlines about a vaccine, but we have no idea whether that hope will pan out or, if so, what sort of timeframe this will take.

I still cannot consider it prudent to schedule concerts. I wrote in my March column (written in mid-January) that I did not have any concerts scheduled at all, a first in nearly thirty-five years. I stated that that was “odd: simultaneously peaceful and eerie.” Today it feels more eerie than peaceful: the latter has been partially replaced by impatience and the fear that it will never seem right to schedule events. Looking back, as of a couple of weeks ago I have not played in public for over a year. That arises out of a chain of mostly unrelated circumstances: first I kept my schedule clear for several months so that I could practice for planned performances of J. S. Bach’s The Art of the Fugue; then I had to deal with my shoulder surgery and recovery; then the current phase in the history of the world set in. The last time that I went more than a year without playing in public was prior to 1980. 

Over the last several weeks there has been a lot of discussion, much of it deeply anguished, about choirs and choral singing. This does not affect me directly at this point in my life except as a listener, though I know it is deeply affecting many of my friends and colleagues. In fact, it may be two years before widespread choral singing will be possible again. I hope very much that by the time you are reading this that hypothesis will have turned out to be overly pessimistic.

There is a lot of variation in how people react to this uncertainty when it comes to the parts of their lives and daily activities that are subject to discretion. Some colleagues are using their extra free time to learn new music or new skills—the technique required to work on new and unfamiliar repertoire or even a new instrument. Some are taking up new activities or hobbies—perhaps ones that they have always meant to pursue. So far, I have done none of the above. My reaction to the situation has been to put much of my motivation to tackling preexisting projects. I mentioned in my previous column that I needed “to take a deep breath.” At that point, early in this whole scenario, I felt that my students needed that as well, and that it was a good thing for all of us. Shortly after writing that, I did start to offer various forms of remote lessons or consultations to my students. However, I have not felt my own motivation returning, either to plunge back into practicing or to explore anything new. Most of what I have been doing has been “comfort food,” as we have been baking a fair amount of bread and cooking a bit more elaborately.

I am not certain why this is. It may be partly a direct reaction to the sadness and difficulty of what is happening. If so, it is not necessarily entirely a depressive reaction or a reaction of feeling indifferent. I suspect that in the face of so much tragedy right around me I am afraid that I will find the music that I might normally be playing too intense. That has been my reaction to the little bits of playing that I have done and also to much of the music that I have heard. Also, I have always had better practice habits when I have performances coming up. That impetus is gone for now. I do feel certain that the motivation will come back. But the main point is this: that any such reaction is okay. I am overjoyed that so many of my colleagues are, for example, posting videos of performances from their homes. That is generous and helpful. I have been an avid viewer and listener, and that is helping me get through certain days. However, I believe it is important that no one feels pressure to cope in ways that are unnatural. In general getting a lot done is more admired in society than not getting anything done. And I am confessing to embracing the latter, though just for now, and claiming that I am within my rights to do so. 

But if it is self-serving, it is not selfish since I hope very much to help persuade everyone to give themselves the same leeway, as much as they need. Doing the things that we have to do is enough as far as fulfilling obligations is concerned. 

At the same time, I have been thinking about counterpoint and The Art of the Fugue. It feels like the odd times in which we live are encouraging me to engage in ever more speculative thinking. Rather than indulging in the technical aspects of counterpoint, I have been pondering more about images and ideas around the concept of counterpoint. Ideally the images and ideas will inform the way that I think about the technicalities. 

One very powerful idea about counterpoint is that it is related to conversation. If two musical entities are engaged with one another, doing different things at the same time, it is natural for us to hear what is happening as analogous to human verbal conversation. This is not an idea of mine, but has been the subject of articles and books as well as informal discussion. It is intuitively convincing. When counterpoint is being produced by separate instruments the conversational aspect is enhanced by the visual and the conceptual: we see and are aware of a different source for each musical line, just as we see and are aware of each different speaker in a conversation. In vocal counterpoint, we see and hear something that is remarkably similar to conversation, down to the humanity of the sources of the sound and the expressions and gestures. At a keyboard instrument the conversational aspect is something that presumably arises solely from the sound. Visually, and often spatially, everything comes from pretty much the same place. The extent to which it is up to the performer and to performance choices to make the conversational aspect of the music convincing is not necessarily very different from the parallel concerns with ensemble counterpoint.

For the performer, one of the great strengths of conceptualizing counterpoint as conversation is that it brings home the need to make each line in and of itself an effective piece of communication—something that has “meaning” though not dictionary or visual-image based meaning. At a minimum this is psychologically helpful, even inspiring, for many performers. For me it serves as a reminder to behave as if every note matters. In conversation every word matters, in that it can be heard by someone and may affect that person. That does not mean that every word is serious, solemn, or weighty. Some are funny, light-hearted, rhetorical rather than meaning-laden. But they are all there and all have an effect.

I have a few caveats about counterpoint as conversation. For one thing, it seems important to me to remember that, as I just mentioned, music in itself does not have dictionary meaning, semantic, idea-based meaning, and that it does not mean anything that can be encapsulated in a visual image. It is liberating and powerful to accept that Example 1 means exactly what it says and nothing else. This freedom from word-like meaning gives a line of music the ability to do things that words cannot do and the flexibility to be used in ways that words are not used. 

Related to that is the first major difference between verbal conversation and musical contrapuntal conversation. In the latter, we not only allow but expect material to be used multiple times. Although the essence of counterpoint is found in two different things happening at the same time, it is habitual for identical or similar things to happen at different times. This can be recurrence, repetition, echoing, answering, returning, and so on. But all of these techniques play a minor role in anything like normal conversation and a limited though sometimes important role in poetry, drama, and literary narrative. They are pervasive and important in music.

In verbal conversation, we do not expect many voices to be sounding at the same time. We expect them to take turns and occasionally overlap, which is fascinating in verbal conversation. Sometimes, it functions to create continuity and an overall arc. At times it is an interruption, which can be a sign of enthusiasm and can constitute rudeness. It is common and normal for interruption to take the form of one person’s finishing another’s thought—not necessarily in the way that the first speaker would have finished it. It is not normal for two or more people simply to talk steadily at the same time as one another for a substantial amount of time. This would cease to be conversation. But it is the norm for musical contrapuntal conversation. 

With words, we do not expect to be able to follow even two let alone three or more lines of thought at the same time. With counterpoint, that is exactly what we expect to do: it is a major concept of the exercise. It is not necessarily easy, and it is not necessarily something at which we always fully succeed. It is almost certainly both common and unproblematic for some of that following to be subconscious or subliminal. People differ in the extent to which they are consciously, specifically aware of following and really parsing the separate lines of counterpoint as it goes by. And, of course, different performances of the same piece or passage can seem to make it easier or harder to follow in that way. (And interestingly different performances can seem different in that respect to different listeners.) I think that it is a pitfall of the counterpoint-as-conversation idea that it can tempt us to try to make the analogy fit even more closely than it naturally does. This might involve downplaying the significance of the simultaneity of lines or even denying that following multiple lines at once is possible. I have heard people suggest that the way we listen to counterpoint should fundamentally involve switching focus from one line to another, as we would presumably have to do if we were trying to listen to two or more people talk at the very same time. 

Questions of how many lines we can listen to simultaneously are complex. Does it vary from one person to another? If so, is that somehow intrinsic—or of life-long standing—or does it arise specifically from music-based training? Can almost everyone follow two voices? Can anyone really follow six? eight? forty? Do people mainly listen to or notice the beginnings of notes, or are the sustained portions of notes important as well? In counterpoint is one line ever more “important” than another, and, if so, what does that mean and what should a performer do about it? Whatever these questions are, I believe it is important not to let the speech analogy influence our answers to them, or how we frame them, more than it should.

Another concern about the conversation analogy is that musical conversational counterpoint is mostly experienced by listeners, whereas verbal conversation is fundamentally experienced by those who participate in it. We who love counterpoint love playing it. It is interesting to contemplate how much we function as listeners while we play and how much of our experience is the pure experience of playing. But the vast majority of music listening is done by listeners. Listening to a spoken conversation in which you have no part happens and is perfectly normal, but not the most usual or common.

The completely different model of counterpoint that has come to interest or even preoccupy me over the last few years is one that is harder to encapsulate in words: counterpoint in music is a model for the whole phenomenon of the existence of the universe. This model was suggested to me by some of my experience as a theater attendee.

Over the last several years I have attended quite a few theater events that are organized in what amounts to contrapuntal layers: different parts of the story going on in different or overlapping spaces, perhaps threads sometimes coming together in one space or passing near one another, sometimes remaining separate. Together they all add up to the complete story. Some such pieces that I have experienced are Sleep No More, Then She Fell, The Grand Paradise, Ghost Light, Here, Seeing You, and versions of Hamlet and A Midsummer Night’s Dream. 

I was initially puzzled by why I found this sort of story-telling so powerful. Events of this sort seem very much like closed worlds: nothing from the outside gets in or interferes. This helps the audience to concentrate and stay committed. It also means that the world built up inside the walls of the event has the chance to feel complete—it is temporarily defined as being all that there is, and it is structured according to its own content.

I realized after a while that the structure always felt, through a number of different styles and each time with a different story, like an analogy to the “real” world: layered and complex enough for that analogy to seem valid and emotionally convincing. 

At some point I realized that the experience of being at this sort of show reminded me strongly of closing my eyes and becoming totally absorbed in a piece of contrapuntal music. In such a piece of music there might be only three or four component lines; in a show such as the ones that I am talking about there might be any number of component storylines weaving their ways around one another. In the universe as a whole there are infinite numbers. But the analogy still seems to hold.

This image neither contradicts nor directly complements the conversation analogy. It is simply another angle and one that I along with some of my students have found particularly interesting and powerful.

It is my intention—uncertainties aside, for the moment—to return next month to some nitty-gritty motivic analysis of The Art of the Fugue, not without some speculation about the role of memory in creating structure.

Further thoughts about rhythm

In the very late 1960s and early 1970s, when I was first interested in the organ and listening to a lot of organ music, I had an LP recording of late-nineteenth-century French organ selections, pieces and excerpts of pieces by several composers, played by several different organists. It was a miscellany, a real sampler. All of the pieces were new to me then, as were the organs and the performers. The piece that impressed me most was the first movement of Charles-Marie Widor’s Sixth Symphony, played by Xavier Darasse. I should probably say that the passage that impressed me was the opening of that movement (Example 1). I remember quite clearly, fifty years later, that I listened to the opening of that track over and over again. I probably destroyed that part of the LP, but it helped to solidify my love for the organ; thus it was worth it!

Several weeks ago, as I decided to write a column or two focusing on various aspects of rhythm, that passage started going through my head again. It has scarcely been out of my head since, except when I have been listening to something else. Every passage of music has some relationship to the concept of rhythm. In spite of my early love for it, the first movement of that Widor symphony is not one that I have performed or analyzed. But thinking about it now and finally analyzing it a little bit, I think that there are all sorts of interesting things occurring with the rhythm of those first few measures. In particular, there are fascinating relationships between the rhetorical and theoretical aspects of rhythm, and this is part of what I address in this column.

The rhythm of the striking opening chords of the movement is demonstrated in Example 2. The rhythm of those powerful chords, and of the melody that is their treble line, is treated as a motive throughout the movement. But the surface rhythm, the rhythm of new notes, whatever voice or part of the texture they come from, is demonstrated in Example 3. That extra quarter note is a passing note in the pedal part, the bass line.

This is all straightforward, just a description of what is in the score. But it is fascinating to me that I hear two things going on at once, two different ways to describe the rhythm of this passage, both valid and meaningful. This seems to be a wrinkle in the relationship between counterpoint and rhythm. Lines of counterpoint, in the way that we usually conceive of them, are almost always characterized in part by having at least some rhythmic difference one to another. But here we have a passage that is seemingly homophonic. But the fact that the rhythm of that last quarter note of the second measure comes across as being on a lower level of importance rhythmically injects an element of counterpoint. This is subjective, but it is an interesting confirmation that the bass/pedal line is used over the next few measures to open the passage into more and more palpable counterpoint, as shown in Example 4.

If we do not know where notes are coming from rhetorically, then our sense of what the rhythm of a passage is can only be the surface rhythm. Turning that the other way around, if we notice hierarchies of rhythmic importance in different notes within a texture, that may be a clue as to some of what is going on rhetorically and contrapuntally in that passage.

For me, this is a new and slightly different way of looking at the relationship between rhythm and counterpoint. This means that I have not yet worked out how to help students apply it to pieces they are working on. There is a lot of music—from Byrd fantasias through Beatles songs, and including a lot of organ repertoire from all time and places—that is clearly not fully, rigorously contrapuntal, but in which counterpoint keeps breaking through. It is probably true that the vast majority of tonal music falls into this category. But nonetheless I have always had a problem feeling comfortable with it conceptually. Is it counterpoint or is it not? I understand that this is just the imposition of a rigid framework. But still, the concept that I am sketching out here seems to be able to help me get more comfortable with counterpoint flowing in and out of a piece or a passage.

My second answer to why this passage started going through my head has to do with the relationship between rhythm and rhythm as rhetoric. The rhythm of the notes at the beginning of this piece is well-defined and clear. But what is that rhythm doing? In a lot of circumstances, the rhythm of the first measure—just two half notes—would suggest a downbeat and an upbeat. But the way that I hear this first measure is something different and harder to describe. I hear each of those chords as a kind of world in itself, neither coming from anywhere rhythmically nor leading anywhere, but just rather insistently being. It is as if the second chord has so much gravity and weight that it refuses to be in a hierarchical relationship to the first chord or to the downbeat of the next measure. This is subjective, my way of hearing it. Assuming for the moment that this is correct or at least meaningful, is it about rhythm? A plausible and normal answer to that would be no. The rhythm is what it is, and everything else is a different aspect: affect, aesthetic, sonority, the push and pull of the harmony, interpretive choices, and so on. But it strikes me that it might be more interesting to expand the concept of rhythm to include more about what the rhythmic pulses and impulses are doing.

I believe that this concept or image could be interesting and helpful to students. It might provide a way of broadening the comfort zone of some students who are making choices about how to precisely execute rhythms on the page. If so, that would probably be through allowing choices about freedom of rhythm, bending and stretching the notated rhythm to feel more like an essential part of the rhythm itself. It might also provide me or any teacher with a way of helping the student to think about interpretive rhythmic choices without simply suggesting details of those choices to the student. I am now eager to work with a student on this piece!

In a way, I have put the cart before the horse, describing some ideas that occurred to me once I decided to write a column about a certain subject. As a result of one conversation with colleagues and a few interactions with students, I revisited a few of my ideas about rhythm, the relationship between notated rhythm and rhythm in practice, and certain practical matters about teaching rhythm. The latter include metronome use, counting, how to approach counting during slow practice, and a few other matters. The core of what I plan to write about is a concept that intrigues me: the possibility of deriving rhythm fundamentally, though only in part, from something other than the notation. I will talk about this at length in my next column.

I finish this column with a few more circumscribed yet fascinating points. The first is an anecdote from well over thirty years ago that has stayed with me all this time. I was then beginning to look for ways to participate in chamber music, and I connected with various colleagues as best I could. There was one player with whom I had a session or two of running through pieces and with whom I started to talk about giving concerts. As I came out of one such session, another colleague caught sight of me and said concerning my rehearsal partner, “You know, she really can’t count.” I did not know what to say, and I ended up with something like “Umm . . . ok,” and I did not let that comment affect my decision to go ahead and work with this fine and interesting musician. That comment was false in experience. This player had no more tendency to misread a rhythm or waver in rhythm or tempo than anyone else. But she was someone who often played purposely and quite freely. I learned a lot from her in this respect. I was left wondering what the real source of the uncalled-for carping criticism was. Did that other colleague have a bad experience with the person with whom I was working? Or misremembered or mistaken her for someone else? Or had it been a case of mistaking intentional, interpretive rhythmic freedom for an inability to count?

This latter idea always intrigues me and can be confounding. How do we know whether something that we hear (as to rhythm, for the purposes of this discussion) that departs from the most literally accurate is a mistake or a purposeful gesture? What different attitudes do we bring to such an event if it is one or the other? Is there a gray area in between? The attitude that we bring to mistakes that our students make is pretty clear—it is part of our job to point them out and help the student to understand what the problem is, how to correct it or to avoid similar things, and so on.

But what if the student says, “No, I meant to do that?” There is a strong pull to ask why and to accept that any deviation from what seems to be on the page is all right if there is a good reason. The discussions that arise from grappling with situations like this can be very fruitful indeed, but I have always thought that it is too restrictive. I hope that the concepts I will discuss next time can be used to help students understand what is going on rhythmically when they feel the pull to do something other than what seems to be the literal meaning of the notation but cannot express why. Furthermore, I hope that this can also help teachers address this situation with students without simply dictating outcomes.

I am reminded of a review that I read once of Joseph Szigeti’s recording of the Bach sonatas and partitas for violin. I apologize, as I have been unable to find this review, an unusual issue for the internet era. I am certain that it was written by B. H. Haggin. As best I remember, Haggin liked the recording and got a lot out of the playing and the pieces. But he also felt that the pieces themselves were not very interesting, that Szigeti, with his rhythmically free and idiosyncratic approach to Bach interpretation, had made great music out of pieces that were intrinsically dull exercises. I believe that this assessment was not as unexpected at that point in history as it would be now, though I do not agree with it. It seems to cast an interesting light on rhythm in performance. The reviewer’s perspective was that the player’s striking rhythmic choices were what we might call “wrong” in the sense that they were not really based on anything intrinsic to the music. They were imposed upon the music and thereby made the music great when it really was not. Is this a good situation or a bad one? How would we react if we thought we heard this happening with our students?

The last item that I mention here is an observation that I made while driving recently. I rolled down the window and shortly thereafter heard the sound of two cars passing me in quick succession going the other way. There were two whooshing sounds probably about 0.7 seconds apart. There had not been any cars ahead of them for a while, nor were there any following behind. So the two sounds were isolated. Although translated into the terms of musical rhythm this was just two notes out of any context, and it immediately evoked for me a very specific moment, namely the opening of Beethoven’s Sonata in E Minor for piano, opus 90 (Example 5).

The two car sounds seemed to deliver the rhythm of this opening gesture. Why and how? Two notes in a row is so commonplace in music that it is almost silly to evoke a specific instance of it as a thing in itself at all! Even if the two cars sounded clearly like an upbeat and a downbeat, that hardly narrows things down at all. And it cannot be harmony or sonority—what I heard had no pitch or harmony, and the sonority was that of a couple of cars. So what made that passage come into my head? I do not have an answer, but it adds to my sense that there is something more to rhythm than what we see notated on the page or can describe in words.

A special note: I will be playing selections from J. S. Bach’s The Art of the Fugue on harpsichord as part of the New York City concert series Midtown Concerts on Thursday, October 28, at 1:15 p.m. at the Church of the Transfiguration, 1 East 29th Street, New York, New York. I take the liberty of mentioning this since I have written extensively in these pages about my Art of the Fugue project. This will be the first public manifestation of that project and my first public concert in two-and-a-half years. If any readers can make it, I would of course be overjoyed to see you there