By Nathan Pell (The Graduate Center, CUNY)
First an anecdote:
At the end of my masters degree, I was required, as all composers were, to pass a jury in which members of the composition faculty comment on a portfolio of recent works. One of these—a spare, quasi-recitative setting, the first in a larger collection of Blake songs—was met with great displeasure from one of my examiners (a composer of renown, whom I respect a great deal) because I had written it with no dynamic markings whatsoever. What could have possibly motivated me to do this?, I was asked. Did I want the performance to be “flat,” that is, lacking dynamic shape?
I gave a twofold answer. First, that I had omitted dynamics in order to restore to the performers some interpretive license (the tempo marking was “Liberamente”): my small antidote to what I regard as an outbreak of over-notation that has pockmarked scores from the past century. As composers have given increasingly specific performance instructions, performers in turn have grown more dependent on these instructions—and, thereby, on the composers who write them. By notating less, I said in my jury, I hoped I could encourage performers to reclaim a more equal share in the creation of the music—even if that resulted in performances I didn’t like. (Nor am I alone in this opinion.) I added that modern performers certainly know how to put across “under-notated” music (Bach, for instance) without sounding “flat,” despite the anachronism of the notation.
Second, I told the jury that I considered all of the dynamics in my piece to be “extrinsic”; the rest of this post will explain what I meant by that. I can’t say that either this argument or the preceding screed against over-notation managed to win over my jurors. But all’s well that ends well: I passed my jury, and the composer in question was kind not to mention the disagreement in the official write-up. I do understand this composer’s points and am glad they were raised, for they have led me to develop my thinking on this subject.
* * *
Charles Rosen writes:
Dynamics [in the eighteenth century]…were one way for the performer to elucidate the structure of pitch and rhythm and make them expressive and even personal. With Haydn and, above all, Beethoven, however, the dynamics are often an integral part of the motif, which has become unthinkable and unintelligible without them. Gradually…dynamics…[were] removed from the process of realization and transferred into the basic process of composition…. In several pieces of Debussy, it would give the music less of a shock to play a wrong note than to play the wrong dynamics or apply the wrong touch (Freedom and The Arts: Essays on Music and Literature).
Continuing this line of thought, I would argue that dynamics are of two sorts: those that are more or less implicit from their musical contexts, and those that are not. Put another way, some passages give the sense that they “belong” at one and only one dynamic, whereas others suggest a wide range of plausible interpretive possibilities. I have sensed this very often while composing. Some dynamic markings write themselves, so much do they feel woven into the fabric of the piece. Others give me pause, and at these moments I feel that I am interpreting more than I am composing: indeed prescribing a single performance solution out of several conceivable ones, rather than describing the dynamic that seems already “built into” the music. These two sorts of dynamics can be said to occur at different stages in the composition process—the interpretive kind at a later, more “post-compositional” point, at which performance markings are added to pitches and rhythms that have already been worked out.
I’ll call “intrinsic” those dynamics that seem “baked into” what music the composer has already written. Those that seem subject to negotiation are the more open “extrinsic” dynamics, written in by the composer-as-interpreter. Of course, most music falls on a spectrum somewhere between these two poles, and sensible musicians will be able to disagree about most cases.
* * *
My first examples will come from music whose dynamics are left unnotated, for these will show what I mean most clearly. As a cellist, my natural gravitation is towards Bach. Who would dream to play the opening of the Third Suite at any dynamic other than forte? Nobody, I’d hope! I surely haven’t heard it any other way, and for good reason. It is expansive, proclamatory, a bit grand; indeed, much could be said about the topical associations of these traits. It is also, intrinsically, forte music, even though no written instruction to that effect is provided. When nineteenth- and twentieth-century musicians edited the suites, they all dutifully wrote in “f” here (unless their editions contained no dynamics at all).
But in doing so, they weren’t telling anyone anything they didn’t already know: they were stating the obvious. By virtue of the culture in which this music came to be and (to some extent) still remains, its forte-ness is understood.
Not so for the Second Suite, whose opening has been played and understood at many different dynamics:
Of particular interest are the editions by two tremendously influential cellists: Julius Klengel and Hugo Becker. Klengel instructs that the D and A be played as double stops, resulting in quite a loud execution.
In contrast, Becker requests that the opening be played up the rather muffled-sounding G string: a much more subdued affect.
By clicking here, you can hear me play the opening of the Suite in these two very different editions. Because this passage suggests no obvious dynamic level, the dynamics are more extrinsic here than in the Third Suite.
* * *
The distinction I am making applies just as much in music whose dynamics are notated. The finale of Beethoven’s Sixth concludes with two orchestral waves, each beginning piano, attaining fortissimo, and ebbing back down again—intrinsic dynamics. The second, larger swell ends at a hushed pp sotto voce (a later composer might have added “religioso”):
But how is the passage’s crescendo (fifth bar) to be taken? Does it effect a linear increase from pianissimo to piano, or should the “p” be taken subito (the crescendo exceeding piano)?
The question has divided musicians ever since the ink dried. Transcriptions by Hummel (below) and Liszt (played here by Gould) adopt the linear interpretation, marking the downbeat fp:
I’ve made a compilation video in which you can hear many orchestras playing the passage over the years. With few exceptions, the performances preserved on early recordings seem to have favored the subito version (Weingartner); only after WWII does the linear option begin to take hold (Barshai). Toscanini’s career encapsulates the changing performance practice: having long performed the “p” subito, he can be heard switching camps in his final recording of the piece. Nowadays, you can hear both in concert (Järvi, Merrill). As these performances show, both versions work at a range of tempos. I strongly prefer one way over the other (I won’t say which!), but must admit that both can yield beautiful performances.
Beethoven could have notated the passage more clearly: “subito p,” Hummel and Liszt’s “fp,” or even omitting the “p” entirely. However, he would not have used mp or mf, which do not appear in the symphonies. But this raises an interesting question: should nothing in the Beethoven symphonies, then, ever be played mezzo forte? Is something about the symphonies intrinsically not mezzo forte? The answer, of course, is no; for no dynamic marking (or in this case, lack thereof) is ever entirely intrinsic—nor do dynamics merely tell one how loudly to play.
* * *
In the C minor Nocturne, Op. 48, No. 1, Chopin marks the A’ section (where the minore returns) pp agitato.
Emil Gilels follows Chopin’s instructions almost to the letter. Most pianists do similarly. Because the A section begins rather quietly (mezza voce) and the A’ very quietly, it seems that Chopin designed these (intrinsic) dynamics to buttress the piece’s ternary form.
But we must reassess upon hearing Myra Hess’s shattering account of the work (a performance I cherish), in which she plays ff agitato at the same spot; the dynamics must not be as intrinsic to this passage as we thought. Her fortissimo makes sense because of the added triplet figuration, carried over from the B section; it’s as if the B section has caused the A’ section to get louder. She’s well aware that playing from here to the end of the piece at this loud dynamic would be onerous on the ear, so she uses the written diminuendo in bar 4 of the minore to get down to Chopin’s marked dynamic, or thereabouts. This approach allows her to dovetail somewhat the join between sections; she helps us recognize that the A’ section blends elements from the A and B sections.
* * *
I am aware that my thoughts on dynamics—particularly their separability or inseparability from “the music”—may tell us far more about how I myself conceive of them, and little if anything about “music itself.” And I have not had time here to address much that I want to, including “workhood,” non-score-based traditions, and changes in performance culture. Furthermore, for many friends of mine, composers who write a dynamic level first (or a texture, or governing instrumental technique) and fill it out with pitches and rhythms second, perhaps all dynamics are necessarily intrinsic. We must ask, then, whether we have been right in calling such parameters “secondary”; for dynamics can be as central to the creation of musical structure and affect as pitch, even when these dynamics are devised by the performer rather than the composer. At the very least, I hope the view from where I sit—as a performer, composer, and (apparently) a theorist—is found interesting if only for the questions it raises.