Digging Deeper: Singing the Music of Elliott Carter

Elliott Carter Applauds his Birthday Concert
Elliott Carter applauds the performances of his music during his 103rd birthday concert at the 92nd Street Y in New York City on December 8, 2011. (Photo by Cory Weaver, courtesy 92nd Street Y)

From a singer’s perspective, Elliott Carter’s works are challenging. Yes, there is a whole bag of tricks a singer can employ in learning this music, from extensive use of Dr. Beat, to charting out pitch relationships, to practicing against pedal tones—all of which are important techniques to learn. And so I try to impart these tools of the trade to my students, hopefully to positive effect. However, it seems to me that the crux of the matter in grasping this difficult and satisfying music lies not in conquering its inherent and unavoidable technical issues. What’s crucial is finding the broader context in which those challenges can be seen not as obstacles to successful performance, but rather as essential musical materials that upon close investigation reveal important information about the nature of Carter’s music itself, its structure, aesthetic, and intent.

Even amongst skilled performers, a commonly heard knock against Carter’s vocal music (and that of other modernist composers) is that the writing for voice is unidiomatic. (Exactly what constitutes idiomatic vocal writing is a topic for a different article. Suffice it to say that to my way of thinking, Berio’s Sequenza III is more closely aligned with so-called “natural” vocal expression than is Mozart’s “Dove sono.” Perhaps Carter’s vocal lines have more in common with the latter.) But the very notion of idiomatic writing, for any instrument, is usually unconsciously tied to the familiar, historic repertoire for that instrument. Having embodied this repertoire, performers sometimes conflate threads of stylistic musical requirements with those of technical efficiency or “performability.” One conclusion frequently drawn from this conflation is a tenet that there exists an idiomatic paradigm for each instrument. Not surprisingly, these paradigms almost always fall within brackets of one or two narrow styles (that differ according to instrument). All of which prompts me to ask: are the virtuosic piano works of Liszt really “pianistic,” or has the repetition of these works, coupled with Liszt’s cemented place in history, colored our view? Does the bel canto approach to singing really result in the optimal vocal sound? Besides, isn’t bel canto a style? And what does that style have to do with the music of Elliott Carter?

Well, not much. And so in approaching new works (or old ones, for that matter), I find it absolutely necessary to forgo such habitual fixations in order to better examine the essence of musical style from the bottom up, for every composer, and in each discrete piece of music. Ultimately, every work deserves to be addressed on its own terms, and performance as a whole undoubtedly benefits from such a modus operandi.

For example, one broad exploration of stylistic essence encompasses how we performers are asked, by the music itself, to handle the passage of time. Musical time seems to manifest itself on a continuum between objective and subjective—that is, between clock-time and our internal experience of time. An extreme example of music that is constructed so as to depend on the performer’s adherence to objective time would be Conlon Nancarrow’s player piano rolls. At the opposite pole, music that structurally hangs on the performer’s mastery and manipulation of subjective time is the gesture-driven music of György Kurtág.

It could be said, for a variety of reasons, that Elliott Carter’s music is firmly planted on the objective side of the time continuum (more on this soon). Conversely, the canon of so-called idiomatically written vocal music (think Italian opera and late-romantic art song) leans decidedly to the subjective side of the time continuum—that is, it allows for and even depends on tools such as rubato and tenuto for its success in performance. And there’s the rub, when it comes to the singer’s approach! Let me illustrate a bit further.

Upon examination of Carter’s vocal music, certain elements stand out which reveal it to be rooted in objective time. Carter’s pervasive contrapuntal cycling of rhythmic material is emergent in the instrumental complement (be it chamber ensemble or piano). The vocal line participates in this rhythmic play to a lesser extent, but when it does, it almost always cycles at a more protracted rate than do the instruments. Acoustic balance is achieved by the clearing of small spaces, often less than a sixteenth-note’s duration, in which the singer must precisely place text. For this element alone, it follows that an accurate rhythmic performance insures better intelligibility of the sublime poems employed (including those of the great Americans Elizabeth Bishop, Hart Crane, John Hollander, Marianne Moore, and Ezra Pound). The use of rubato in Carter is, in most cases, intrusive to both the progress of the music and to tight ensemble.

In addition to these micro-rhythmic intricacies, there is another important characteristic that influences the perception of Carter-time on the macro-level. Rarely virtuosic, Carter’s vocal lines are often slowly unfolding arches that stand in contrast to the concurrently bubbling instrumental lines. Taken alone, the vocal line progresses in regular rhythms. Seldom is there a tuplet to be found. Many “football notes” appear; indeed, the text is sometimes sustained to the point of obscuring meaning (intelligibility of text is, in part, dependent on the listener’s ability to discern the rhythmic properties of language). Perhaps this high-flying, sustained approach to setting text can be viewed as Carter finding yet another way to manipulate the experience of time towards the objective pole.

So, here we have two factors in Carter—the limited practicality of rubato and the deliberate distortion of natural speech rhythm—that pose a conundrum to the well-trained singer, who has absorbed the message that idiomatic (i.e. “musical”) compositions allow for a flexibility in rhythm that will assist her in clearly delivering the text. It’s no wonder that in the face of this incongruity, she resists the apparently fatal barriers to expressiveness and instead bends the music to the familiar paradigm! The result of doing which, unfortunately, might be a performance that either is redolent of Brahms or that betrays the discomfort of waking up in a musical straitjacket.

Finally, then, how is the singer to tease “natural” expression out of music anathema to the bending of time?

One answer lies in unpacking the text to expose another facet of espressivo not always explicit in musical notation: timbre. Think of music as it exists on three axes, where x = time, y = pitch, and z = timbre. On the two-dimensional page of printed music, the inflective instructions of the z-axis appear rather sparse and general when compared to the specificity found in the representation of rhythmic and pitch structures. This is true of all notated music; Carter’s is no exception. However, the presence of a text provides the advantage of a limitless supply of timbral tools for all attending musicians to draw upon, in addition to any specified phenomenal accents and dynamic indications. The expectation is that all performers (singing and non-singing) extrapolate from the inherent sonic material of poetry generated by methods—onomatopoeia, alliteration, anaphora and rhyme among them—that variegate timbre and texture, thus animating the z-axis.

Perhaps the most coloristic components of language reside in consonants, which, in the pedagogy of singing, regrettably are regarded as the ugly stepsisters of vowels. Yet deft management of the shading, rhythmic placement, and duration of an initial consonant (so long as the ensuing vowel starts on time) can highlight a word more meaningfully than any fermata (always placed over a vowel!) could ever hope to do.

Clear diction without rhythmic aberration requires quite a bit of dynamic and coloristic flexibility, especially with regard to stressed and unstressed syllables in artificially sustained phrases. Further, encoded in the syntax of the poetic phrases is a dynamic shape that Carter has clearly absorbed and provided the framework for in both his selection of intervals and rhythmic relationships. This dynamic shape is far more detailed than could be represented by hairpins and accents. Take for instance the first stanza of Hollander’s poem “High on our Tower,” the text for the opening song in Carter’s cycle Of Challenge and of Love:

High on our tower
Where the winds were
Did my head turning
Turn yours,
Or were we burning
In the one wind?

In his setting, Carter indicates an accent (>) over 16 of the 25 syllables. Certainly that does not mean 16 equivalent attacks!

Score Excerpt from Elliott Carter's Of Challenge and of Love

Of Challenge and of Love
Elliott Carter and John Hollander
(c) Copyright 1995 by Hendon Music, Inc., a Boosey & Hawkes company
International copyright secured. All rights reserved

Play sample audio: Elliott Carter: Of Challenge and of Love I. “High on Our Tower”

Sung by Tony Arnold, soprano, accompanied by pianist Jacob Greenberg from the CD The Music of Elliott Carter, Volume 5 (Bridge 9128). Courtesy Bridge Records. (Order from Amazon or download from iTunes.)

This is a clear example of the limits of notation when it comes to timbre. Even when those accents occur in relationship to a specified dynamic shape, it is the text that can singularly elucidate how to execute each discrete accent. Three instances:

1. Exaggerating the [hw] and [w]’s of the alliterative and onomatopoeic line “Where the winds were” immediately evokes wind. Whether it is to be a breeze or a gale is at the discretion of the singer, fully informed by the musical materials that encircle that image.

2. Carter highlights the anaphora of “turning, / Turn” by placing the second “Turn” one half-step above the first. The singer can further reinforce this by intensifying the articulation of [t], especially in the context of the soft dynamic specified in the score.

3. Carter discloses his personal take on how the poem should be read through the sudden rhythmic diminution of the final two lines of the stanza. This coupled with the insertion of both a rest and a large intervallic drop between “one” and “wind” results in a rush of urgent questioning: “or were we burning in the ONE… wind?” Again, the alliteration of two [w]’s appears, but this time requiring a decidedly different affect via both literal meaning and positioning as the final cadence of the stanza. The singer must adjust the energy and texture of those juicy consonants accordingly.

This is but a slice of the rich world revealed by digging deeper into the sonic properties of text. Any singer who continually broadens her timbral palette can free herself of what might otherwise feel like a wooden adherence to objective time. The tightly wound rhythmic structures found in the music of Elliott Carter deserve to be complemented with the utmost creative investment in the oft-neglected parameter of timbre. In doing so, the myriad expressivity embedded in his musical language is unearthed. As in the words of John Hollander, set by Carter in Of Challenge and of Love:

But when true beauty does finally come crashing at us through the stretched paper of the picturesque, we can wonder how we had for so long been able to remain distracted from its absence.

13 thoughts on “Digging Deeper: Singing the Music of Elliott Carter

  1. Joe Barron

    Tony, this is one of the most informative essays on Carter’s music I have read. A couple of comments:

    … how is the singer to tease “natural” expression out of music anathema to the bending of time?

    It is my understanding, based on other reading, that Carter’s precise rhythms and metric modulations are a form of written rubato. In essence, he does the time bending for you – the subjective time element is presented objectively – and perhaps the secret to natural expression is feel the way this time element lies within the notation.

    As for Carter’s unidiomatic vocal writing: I’ve always felt Carter captures spoken English much better than some of his atonalist contemporaries (we won’t mention any names), who seem to force the words into the wide intervals that result from strict adherence to 12-tone technique.

    But then, I’m not a vocalist, so perhaps I can’t judge.

    1. Tony Arnold Post author

      Thanks, Joe.

      You’re absolutely right that “Carter’s precise rhythms and metric modulations are a form of written rubato,” and I think it’s possible to feel it that way in music that is quicker moving (like the example shown, the first mvt. of OCAOL). Perhaps I should have illustrated with another movement as well, such as the 2nd song, “Under the Dome….”

      In it, the text is heroically sustained to the point where the singer loses touch with pulse quite easily and completely. With such stretching, it is nearly impossible to feel a written out rubato — everything just feels timeless (and sometimes this is not a good feeling for the singer who is managing the high tessitura as well.) Incidentally in this case, the singer’s instinct is to rush, not drag.

      Having another parameter to focus on is extremely helpful in technically challenging spots such as these. It can be a game changer for a singer to deliver a WORD, rather than a NOTE.

      In the case of at least some of the other composers to which you have alluded, the same applies! To the above I would add: it is a game changer to deliver a GESTURE rather than a series of notes and rhythms.

      BTW — I happen to love wide-stepping intervals. Very relaxing for the voice, and ultimately much, much easier to sing than the music of say Reich or Glass, which hold the voice in a single register for long periods of time.

  2. Joe Barron

    As a listener, the problem that I have with”Under the Dome” is that Carter makes the echo effects louder rather softer with each repetition – for example, in “belie … lie … lie.” I would think the way to portray and an echo musically is to make each repetition softer, as though the sound is fading away. Instead, Carter emphasizes each successive word, sustain the line heroically, as you say. He’s a sensitive reader, and he must have understood what Hollander was going for, so I must ask myself why he made the choice he did. But this is beyond the scope of your post, I guess.

    1. Tony Arnold Post author


      it’s an interesting point that you bring up about the echoes. You’ll notice that in the piano writing, the musical echoes generally follow the scheme you have described: fading away with each repetition.

      In the vocal line, the two instances I believe you are referring to are below in Hollander’s text, with his italics….

      “What do echoes do when they reply?
      Lie, lie lie about what we cried out, about their own
      Helplessness in the face of silence. What do they do
      To the clear call that they make reverberate? Berate,
      it for its faults, its frangible syllables.”

      The occurrences of LIE and BERATE are at those points, I believe, subversions of the ultimate expansive feeling of the text at the summation of the poem:

      “Hearing and overhearing
      My own voice, startled, appalled, instructed, I rejoice.”

      With that in mind, it is perhaps appropriate that Carter chose to go against a natural “echoing” effect. What results is, perhaps, a momentary flash of the unremitting voice of the superego, relentless and hectoring by definition. Almost nightmarish, for a moment, until it recedes into the shadowy sound world of the cavern.

  3. danvisconti

    Thanks for the great article, Tony!

    I’m curious how often you have performed any of Carter’s works from memory. I imagine that an understanding of the music’s underlying structure would be crucial to memorization, perhaps more so than with traditional repetoire.

    I recall an essay by Charles Rosen in which he descibes going onstage to play the Carter piano sonata one evening, only to be acosted by a gentleman who exclaimed, “My god, you’re playing *that* piece from memory?!” (after which Rosen suffered his only memory slip with the piece). Carter’s supposed “difficulty” has become so much an article of faith that often we’re distracted from the beauty, elegance, and simplicity that is equally at home in his many compositions.

    1. Tony Arnold Post author

      Thanks, Dan!

      I have never done any of Carter’s works by memory in full, except for his very early songs. I have performed the first of OCAOL from memory.

      It would be completely reasonable to perform any of the conducted works without a score. But following the stick is by no means the same as internalizing the instrumental parts, which takes a lot of time, study, and copious repetition (which with a single pianist is possible, but with a larger chamber group rarely happens).

      Memorizing a solo score is an entirely different thing from memorizing chamber music…. I wonder if the Pacifica Quartet does any of Carter’s quartets off-book… they certainly have the familiarity with the music. But when 4 people are embodying this objective rhythmic accuracy, and depending on each other all the while, it is certainly living on a razor’s edge. It can make for exciting music making!

  4. John Goodman

    Thanks for this insightful article. A longtime admirer of Carter’s vocal writing since the Elizabeth Bishop cycle (1975), I have often wondered about the challenges these settings pose for the vocalist. Your text clarifies this beautifully, and I am grateful for your comments–surely very much to the point–about the elusive category of ‘idiomatic’ writing. Listening to these song cycles, I often find myself referencing the J.S. Bach cantatas, which Carter studied with Nadia Boulanger in Paris. It is rare for Carter to opt for transparent musical illustration (‘The Sandpiper’ being the exception that makes the rule), which Bach frequently does, but I still sense a deep affinity. Having studied these Carter scores closely, do you have any thoughts about this, specifically about how Bach may or may not have influenced his vocal idiom?

  5. John Goodman

    Tony, on reflection I realize that there is more in the way of illustrative accompaniment in Carter’s vocal settings that is generally thought. Still, I would appreciate anything you might have to say about this, and about the relationship of his settings to those of J.S. Bach.

    1. Tony Arnold Post author

      Thanks for your comments, John. I had never considered that there might be a close connection between Bach and Carter’s vocal writing, and I’d have to defer to the composer whether he ever used Bach as a direct model. That being said, here are a few observations:

      Interesting that Bach comes up in connection with this article, which addresses idiomatic writing a bit… amongst singers, it is generally thought that Bach’s vocal writing is “instrumental” (read: unidiomatic). Handel is held up as the great vocal craftsman of the high Baroque. Maybe this has something to do with genre, since Handel wrote operas and Bach did not. More likely it stems from the complaint that Bach’s vocal lines are continuous, uninterrupted, and often provide little space to breathe in melismatic writing. Carter shares this characteristic to a certain extent. It may be because of the counterpoint in both Bach and Carter –- counterpoint tends to have life of its own, and wants to be spun out, sometimes superseding the constraints of text. Handel, of course, tends much more towards homophony, as would become the convention for most vocal music of the late 18th and 19th centuries.

      In terms of text-painting, quite frankly, I have yet to see a composer who avoids it, even those who set out to deliberately eschew it. Further, singers are always looking for ways to highlight the text, however subtly. The fact is that language is, by its very nature, descriptive – I point to onomatopoeia as an example. Even textless music has a subtext, if there is a human in front of an audience communicating something (intelligible or not) by means of voice. This doesn’t even begin to account for the illustrative role(s) of the ensemble.

      The issue is that music is a closed system in which the singer is a participant, whether fused with the ensemble, commenting on the ensemble, in opposition to the ensemble, or indifferent to the ensemble. Here’s where the options for text-painting arise, compositionally speaking. Bach, often using counterpoint as his primary means, weaves the singing into the texture of set pieces. Text painting therefore often appears as a kind of general mood or attitude, sustaining single moods for longer durations. Carter does this too, for instance in the first song of A Mirror on Which to Dwell, where he employs fixed register to create a sound world reminiscent of a carillon.

      Bach’s use of obligato instruments color texts throughout his works. You have aptly noted Sandpiper from AMOWTD as an analogous piece. Although, here I think what Carter does is much more extreme and intense — yes, there is dialogue between singer and oboe, but I believe the singer is even urging the oboe on (and further, the metric modulations provide a cinematic element, serving as a kind of telescoping focus, depending on whether the text describes the minute details of the sandpiper’s exploits or the backdrop of the vast beach and ocean.) The text-painting occurs much more on a moment-by-moment basis here than in Bach (although there is certainly surface-level text decoration everywhere there, however generally subservient to the constraints of the style).

      Except for the convention of recitative (ubiquitous in the Baroque), Bach, as far as I can tell, rarely sets the singer up in a narration or commentary role in set pieces. But Carter does this often, case in point A View of the Capitol from AMOWTD, wherein all of the action (including a passing marching band) occurs inside the ensemble, and the singer is quite clearly commenting from the outside. (N.B. I do not know well Carter’s opera, What Next. But of the vocal chamber works I can say that he rarely, if ever, alludes to the texture associated with traditional recitative.)

      I’ll have to think on this more, it seems there is much to investigate! On a personal note, I don’t get to sing enough Bach. Whenever I do, it is a sublime experience. My next cantata, I’ll be thinking of Carter all the time, I’m sure! Thank you for the comment and the chance to explore this rich topic. I’m sure this only barely scratches the surface!

  6. Pingback: Tony Arnold on Carter « Secret Geometry – James Primosch's blog

Leave a Reply

Your email address will not be published. Required fields are marked *

Conversation and respectful debate is vital to the NewMusicBox community. However, please remember to keep comments constructive and on-topic. Avoid personal attacks and defamatory language. We reserve the right to remove any comment that the community reports as abusive or that the staff determines is inappropriate.