Tuesday, September 27, 2022

OFF the BEATen path

I was recently researching various musical settings of William Blake's poem The Lamb. One of my favorites is by the great American song composer Lee Hoiby, and I came across this recording of Hoiby himself accompanying the song.

It's a fine recording of a song which is successful in part because of its simplicity and directness, but there's something un-simple right at the start. Having accompanied and coached this song many times, I know well that Hoiby has laid a subtle trap for the singer (and pianist) in the one-bar piano introduction. As you can see below in bar 2, the left hand is syncopated one sixteenth note off from the right hand melody and vocal line, but Hoiby chose to begin the song with that left hand figure syncopated against...well, against nothing. Or, more specifically, against some unheard downbeat felt by the pianist and perhaps by the singer - but inaudible. So, if the notation is followed precisely, it will likely sound like the singer comes in early. We hear four eighth notes (suggesting something steady) but the voice enters halfway through the final one.

The curious thing about the recording above - with the composer at the piano - is that the singer basically comes in after four full eighth notes as if it had been notated this way:
It would be tempting to say that the performers are mis-reading the pianist's own music. In fact, that is literally true, but in a broader sense, this is a nice example of how notation can be both limiting and - maybe - freeing. Although one could plausibly interpret this situation as the composer simply changing his mind about how the song should start, one could also think of the syncopated notation as more suggestive than literal. I love the idea of suggestive notation. So what is Hoiby suggesting?

When I see something like this with a missing downbeat, it somehow communicates that the music should be a bit off the ground. It suggests a floating sensation which probably means my wrists will come up as the fingers go down. This is as opposed to a grounded sensation in which I'd let my arm weight sink into the keys more. What difference does that make? Well, pianists have been arguing for ages about how touch and weight affect sound, and I don't want to get into the physics of whether any variable other than velocity is in play. But I do think the imagination of this floating sensation can have an effect on how the music comes out...somehow.

It's possible Hoiby also wants a sense that the sound doesn't really have a strong beginning; we're just tuning into a figure which has been going on imperceptibly until it sneaks in pianissimo. Unfortunately, Hoiby's performance, at least as it comes across in recording, doesn't quite fulfill that ideal as the low E-flat is actually quite rich and present. This recording, also featuring the composer at the piano, achieves this gently tuning in effect a bit better, though in this case I hear Hoiby's rhythm as closer to this (it's an important reality that performances of clearly notated music won't and shouldn't always come out mathematically precise):
That version also feels different because it is transposed a full major 3rd lower, which makes a big difference in this low piano register. The intro bar here is also simply slower than what follows so it has an out of time feeling, which perhaps is what Hoiby hoped to suggest? Or, more realistically, what he came to want. A very important consideration here is that sometimes composers don't know exactly what they want - or what they want changes after going to press.

I suppose Hoiby might be astonished that I've spent even this much time thinking about it, but he's asking for....something... with this kind of notation. I've now spent too many years thinking about that unmoored syncopation to be satisfied with something that doesn't approximate the rhythm shown. For me, it should suggest a very gentle rocking motion over which a melody comes floating in from above. [In addition to rocking, which fits with the lullaby character, another image that comes to mind is of a shepherd lightly plucking harp strings nonchalantly against a melody; this image goes well with the pastoral character of Blake's poem.]

The fact that the melody and rocking motion are out of sync should not create tension but rather suggest an equilibrium between two complementary gestures. Below is a recording I made years ago of my arrangement for solo piano. While I can't say definitively that the first low E-flat doesn't sound like a downbeat, I do like the continuity of that figure into the more obviously offbeat left-hand as the song continues:

I also couldn't resist doing a little audioshopping of the first Hoiby recording posted above. Here, I've compressed the space (by about a sixteenth) right before the singer enters, and I prefer it to Hoiby's unaltered performance. So for all my speculation about Hoiby wanting something less literal than he wrote, I still want something less pedestrian than what he settled on. I wonder if he simply found that the ideal of an imagined downbeat works better for the performer than the listener.

Of course I'm always interested in the connections this kind of situation sparks in my memory. In this case, as I thought about the notation, I had a strong sense I'd experienced a similar rhythmic challenge in another well-known art song. I kept thinking it was probably something by Schumann because...well, because that was my intuition and I know that Schumann loved to write in suggestive ways. But I also wondered if I was remembering something by Fauré or Debussy. It took a good night's sleep and a good hour's more thinking about it before I finally remembered the opening to this little song from Schumann's Op. 39 Liederkreis:

You may hear it sung here. I would not say this is one of my favorite Schumann songs, but there's no missing that suspended gesture in the very short intro. I'm sure my mind made this Hoiby  Schumann association because coaching a singer to find the downbeat in one is closely analogous to the other. The Schumann intro is ultimately easier and more natural both because 1) the notes aren't jumping around and 2) coming in on an upbeat doesn't feel quite so off-kilter as the out-of-the-blue downbeat in Hoiby. (The Schumann has still bewildered more than a few students.) Hoiby's choice to begin with such a low note also makes it sound more like a downbeat to begin with than Schumann's mid-range, second inversion harmonies. But I can't help wonder if Hoiby had Schumann's accompaniment in mind (consciously or not) and if he hoped to achieve a similar effect. 

However, I think Schumann was in a class by himself when it comes to this sort of floating above ground rhythmic/metrical gesture. The opening song and final postlude of Dichterliebe achieve this beautifully, even if downbeats are easy enough to find. You can hear the first song here and the impossibly beautiful postlude (see below) beginning here at 29:34. Charles Rosen wrote memorably about Schumann's love for the fragment. Beginning a song with no clear center of gravity is a wonderful expression of that Romantic impulse to break away from rigid frames.

This final example from Schumann doesn't cause any ensemble problems for the performers, but the way the incandescent third movement of his Piano Quartet begins with an accented diminished chord on beat 2 achieves the same suspended effect. Steady downbeats soon follow, but after the gorgeous main tune has been explored, the piano right hand begins a wandering melody at 1:40 which is gently syncopated against the left hand. Soon it's not clear what's a beat and what's an offbeat and we are simply floating. 

As a sort of postscript, I'll mention one other song which came to mind when I was doing a mental search for what turned out to be Schumann. The opening of Hugo Wolf's Ich hab' in Penna has confused many singers-in-training for similar reasons as Schumann and Hoiby, though it is quite different in character. Here, the frantic piano part begins with constant eighth notes starting on the "and of 1" and the singer should begin on the "and of 1" in the next bar. Though this isn't really a syncopation because there's no rhythmic stress on an offbeat, the absence of an initial downbeat helps propel the song's headlong energy. However, some singers can't help but hear the first note as a downbeat (even when the pianist tries to put a subtle accent on beat 3) so this misleads them into starting one eighth note too late. It's not so hard for the prepared pianist to right the ship but that makes for an uncomfortable mental gear shift! Hear here. (No pianist needs the stress of an unsettled start knowing that terrifying postlude is just around the corner.)

No comments: