A Case Study in Interpretation: Density 21.5 (Part 3 of 3)

Well, after a too long of a delay, we’re back with Part 3 of our series on Edgard Varese’s Density 21.5, specifically as performed by Laura Pou. As we illustrated in Part 2, a close analysis of the score’s micro-details suggests that, contrary to some opinions, Density 21.5 is music with a high degree of character and affect – a beautifully hazy, shifting, and obscure composition.  On the other hand, the score’s macro-level indications seems to suggest an opposite perspective; the instruction “Always strictly in time – follow metronomic indications” and the absence of any character markings support the mechanistic, unemotive, inhuman view of the piece that was put forward in Part 1 of this series. This contradiction was implicitly noted by one IPAP reader who posted the following comment on my Facebook page:

If one keeps a solid pulse, is there room for some elasticity in the rhythm? If one needs an extra moment to execute a large shift, string crossing, chord, or even a peak of phrase etc. but the time is made up somewhere else in the measure/line, is it kosher if it’s subtle enough that no one could flat out say the rhythm is distorted or the pulse compromised? Bach comes to mind, pulse is paramount, but some push/pull is required to keep it breathing and keep it sounding organic and musical. I didn’t miss the point about the rhythmic expression/details, I’m just going a step further into how a musician (does or should) interprets an instruction like: “stay at X tempo, do not deviate.”

If I understand the reader correctly we can condense her comment into two questions: What level of interpretive discretion (and/or the application of commonly held standards of musicality) is appropriate to this music and, furthermore, are the issues raised by this analysis of Varese applicable to broader swaths of the Modern repertoire?

To answer these questions, it might be helpful to think a little bit about the time in which Varese was living and what might have been his general expectations towards musical performance and interpretation. Born in 1883, Varese spent his late teens and early 20s in Paris at the Schola Cantorum  and the Paris Conservatoire, thereby coming of age surrounded by the performance practices of the late Romantic period.  Below are three of the earliest surviving recordings of the great violinists.  In their similarities, these are likely to be emblematic of the kind of music making practices that Varese grew up with.

Jascha Heifetz (b. 1901) Mozart’s “Gavotte in G” from Idomeneo  (rec. 1912)

Joseph Joachim (b. 1831) Bach’s “Adagio” from the Violin Sonata no. 1 (rec. 1904)

Eugène Ysaÿe (b. 1858) Kreisler’s Caprice Viennois (rec. 1912)

In brief, Heifetz’s recording contains frequent and pronounced variations of tempo within a phrase and distinct changes of tempo for contrasting material (see, for example, 0:40 to 0:47). Joachim’s recording could be considered almost tempoless since each measure is of a significantly different duration (see below).  Ysaÿe’s recording is similarly fluid.

m. 1 – c. 9.25”
m. 2 – c. 7.75”
m. 3 – c. 10.25”
m. 4 – c. 8.66”

Table 1 – seconds per measure in Joachim’s performance of Bach’s Violin Sonata

Furthermore, not much had changed in the twenty-odd years between these recordings and the date of Density 21.5’s composition in 1936. As evidence, consider the wide range of tempos applied to Beethoven by conductors such as Toscanini and Klemperer. See John Rockwell’s very good introductory summary of that rivalry here.

So, Varese was composing in a time when the same universally-known piece of music could be played either very fast or very slow (Toscanini vs. Klemperer) and had been trained in a time when a piece of music written in a single tempo could be played either in several tempos (Heifetz) or in a tempo that fluctuates to such an extent that the actual pulse becomes obscured (Joachim, Ysaÿe). By contrast, we now live in a world where music software can lock in tempos to at least a thousandth of a second and the bulk of the music is militantly anti-fluid and machine-dependant in its rhythmicity. In this light, it makes sense that Varese’s instructions do not mean that the piece should be performed mechanically or inhumanly but only that the tempo should be consistent and clearly recognizable. Neither those physically necessary “extra moments”, organic “push/pull”, nor even deliberate affectations such as a slight retard to mark the end of a phrase will compromise the clarity or consistency called for by Varese.  However, the performative distortions and absolute fluidity of the Romantic approach to tempo would obscure the music’s very subtle details and connections that were analyzed back in part 2.

At this point, I think it safe to say that we can conclude that Laura Pou’s deeply affecting performance of Density 21.5 is by no means a deviation from either the spirit or the letter of Varese’s composition, and that its qualities are directly related to the music’s score.  Although we may never know the exact process through which her interpretation was arrived, at we hope that the methodologies employed here imply ways of working that will be beneficial to you as you construct compelling performance interpretations of whatever repertoire you may be working on.