The case for voice-leading

It’s 2012. Why would I waste time and effort considering a musical idiom that’s seen its day, as far as most contemporary composers are concerned? The avant-garde writes music in which pitch is either of little importance or absent altogether. Voice-leading is an obviously inapplicable notion in such a context. On the other hand, composers who use pitch more traditionally, and who might even adhere to tonality, seem to feel that ignoring voice-leading is licit simply because of the date. Time marches on, and as it does (the argument goes), it removes, one after another, previously established standards.

In the case of the former, I’d argue that things like this, or even this, are not actually music. Sometimes you see music defined as “organized sound.” I’d refine that a bit and say music is organized pitch. Pitch is an essential element in music. Trying to create music without pitch would be like trying to create food without flavor. Pitch is the substance with which one works. Is the chef not judged by how well or how innovatively she manipulates and combines foods with different flavor profiles? What would you say if you took a bite of something and tasted nothing? Texture (the tactile experience when eating) isn’t enough by itself, just as unpitched sound itself is not enough. If pitch is present, but its temporal location is not explicitly notated by the composer (having been left to chance), then in what meaningful sense can we say that the composer has actually composed the piece?

In the case of the latter, I’ll admit that the march of time and the change attendant to it are unavoidable, and often to be welcomed. But not on the basis of whim, or worse, of incompetence. It sometimes happens that dictionaries must acknowledge technically incorrect word spellings or usages, simply because so many people get it wrong. This is what has happened to voice-leading in contemporary tonal music.

The problem is that many contemporary musicians are unaware of how and why voice-leading guidelines came about. They were not imposed, top-down, by some individual who dreamt them up from whole cloth. They evolved, or perhaps more accurately, they emerged. They emerged from close listening, and consideration of harmonic/contrapuntal implications, by composers across the ages. One can observe how voice-leading guidelines gradually coalesced as monody gave way to organum, which gave way to ars antiqua, then ars nova, stile antico, and eventually culminating in the complex machinations of composers like Johannes Brahms. It was at this point that most composers felt some new means of generating content was necessary. Imagine, however, if scientists were to have said at some point: “the scientific method has been exhausted; we need some new means of producing reliable knowledge.” A keen observer might detect that those campaigning for the “new method,” in either discipline, had simply found the “old method” too challenging, intellectually, and were hoping to institute a more relativistic system with fewer standards. One certainly can’t argue that the “old method” led to scientific or artistic stasis. Just look where the scientific method has gotten us! And then look at the difference between Palestrina and Brahms! Yet the basic principles governing the production of content in both composers’ work are the same! As Schenker often said: “Semper idem, sed non eodem modo.” Not realizing voice-leading’s organic raison d’etre, contemporary composers suppose it can be unthinkingly dismissed as a superficial fashion of centuries past, an unnecessary conceit.

In most disciplines, practitioners usually try to proceed in the best way possible, and usually attempt to derive the best way possible in as objectively sound a manner as possible. Yet, even in mathematics, in order to make any progress, some things must simply be agreed upon, not necessarily because there is absolute proof for them, but because we agree they are more-or-less self-evident. These “first priciples” are called axioms, and in music, voice-leading follows from these two axioms:

1) Redundancy is to be avoided as much as possible.

2) Unqualified ambiguity is to be avoided as much as possible.

Inasmuch as we want to define a composition as skillful and obviously deliberative manipulation of pitched sound, these axioms are necessary to distinguish the composition from unskilled or completely random collections of pitched sound. Most instances of successive parallel perfect intervals breach both of these precepts.

I should hardly need to explain how parallel octaves, in two discrete voices, represent redundancy. Parallel fifths are redundant because the upper voice will really be nothing more than a doubling of the lower voice’s second partial. J. J. Fux had this in mind when he wrote, in Gradus ad Parnassum, that avoiding parallel motion is the primary way to achieve variety amongst the voices.

Parallel fifths are also ambiguous. As Schenker explained, one of the two voices will leave the key established by the other voice. For example, imagine this cantus firmus: C, B, C, E, G, F, E. This is an obvious and strong unfolding of a C-major triad. A voice in parallel fifths would proceed thus: G, F#, G, B, D, C, B. Our sense of C-major has been undone, and C-major is now competing with G-major. Which tonality is it?

At this point, it’s important to remember that avoiding parallel motion only in the surface-structure of a piece, that is, concerning oneself only with the motion from one note to the immediately subsequent note, is not sufficient. One of the primal functions of our brains is to seek pattern. Those who pay close attention when listening to music will relate not only immediately adjacent events, but also important events separated by spans of varying size. These important pitch-events, these harmonic and contrapuntal pillars (what Schenker would term the “middleground”) should also proceed according to the two axioms above.

In addition to the proscription against parallel perfect intervals, the axioms also prescribe a manner of writing that involves the setting-up and fulfilling of goals. Schenker famously wrote: “without a goal, there can be no content.” Indeed, a piece with no discernible direction, that isn’t about attaining harmonic or pitch goals that have been intentionally established – a piece that isn’t heading somewhere – doesn’t compel the listener to keep listening. Aimless meandering will disintegrate into redundancy, often accompanied by a decent helping of bad and unintentional ambiguity. Of course, the ability to establish goals rests primarily on the harmonic implications of the overtone series and the major triad which is derived from it. But that is a topic for another post.

Now, all this having been written, there’s no need to worry that the composer is locked into proceeding in just one, very specific, unimaginative way. On the contrary, there are many ways one can write legitimate parallel perfect intervals without transgressing our two axioms. And the reason the second axiom calls for the avoidance of “unqualified” ambiguity is that not all ambiguity is bad. Indeed, one of the tests for compositional skill is the ingenuity with which a composer can seemingly flout “the rules,” while beneath it all actually maintaining them.

The first “workaround” is that a composer may write parallel octaves and/or fifths with the understanding that the parallel notes all represent one voice. This is easiest to see in the left hand (bass) of almost any piano composition. The left hand is constantly required to play parallel octaves; but those octaves represent one voice – the bass. A similar phenomenon occurs with a style known as “planing”, for which Debussy was famous. For instance, the parallel octaves and fifths, for both hands, in the Sarabande from his suite Pour le Piano are not intended as individual voices. The injunction against parallel motion applies only to voices which are intended to be discrete. This is why organum was comprised entirely of parallel perfect intervals. Composers writing in that style were still thinking in terms of monody. The addition of a perfect fifth was not intended as an addition of a separate voice. No, the problem, which is all too common in contemporary tonal music, is when two or more voices proceed appropriately for a while, then suddenly engage in parallel motion which can’t be explained as a unisono or a tutti. 99 times out of 100 it’s because the composer was not in control of what he or she was putting on the paper.

Ambiguity can actually be a source of great beauty, provided the context makes it evident that a) it was intentional, and b) it doesn’t obfuscate the (hopefully present) perspicuity of the composer’s other intentions. The opening of the Adagietto from Mahler’s 5th symphony is an example of good ambiguity. The harp and strings sound only “A” and “C” for several seconds. We aren’t sure if we’re being fed the tonic and third of an A-minor sonority, or the third and fifth of an F-major sonority. When the violins ascend the tetrachord beginning on “C” and wind up asserting the tonic pitch “F”, via a heart-wrenching retardation on “E”, we melt. The ambiguity was brought into sharp relief and made all the more effective because it was dealt with: it was resolved. Ambiguities that are ignored, that are not resolved, are indistinguishable from ambiguities that are unintentional. (I feel I should add, here, that I’m not a particular admirer of Mahler. Much of his music is more gimmick/effect-driven than content-driven. But the opening of the Adagietto was undeniably a good and beautiful idea.)

I’ll close with another analogy. Even if you want to achieve a fresh sound, it behooves you to understand the medium with which you’re working as thoroughly as possible. Composers who don’t know how and why voice-leading guidelines came about, or worse, don’t know what those guidelines are, do not embark auspiciously on their careers. Here’s the analogy: world-renowned scholar and virtuoso organist Jacques van Oortmerssen remarked in a master-class that freedom with tempo means nothing if you, the performer, aren’t thinking of what the unfree, metronomic pulse is. The listener must be able to discern that you are approximating freedom from something. Carte blanche whim is no freedom at all. As philosophical free will deniers and incompatibilists respond, when confronted with the argument that quantum indeterminacy might salvage the classical notion of free will: randomness is not freedom. You have to know where music has been if you want to be involved in where it’s going.


0 Responses to “The case for voice-leading”

  1. Leave a Comment

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: