Voice-leading and Composition

Question

This is a follow up to a post I made a couple of months ago regarding the way the four-part harmony voice-leading, taught in most harmony texts, is to be understood in relation to composition. In the post I asserted that I had come away from the texts with the impression that composers were composing around a "fundamental voice-leading framework" like those shown in the reductions in many texts, and that my problem was that I couldn't really imagine how someone could compose in such a way. I was given an excellent answer by Pat Muchmore that at first I thought had settled the matter for me, But after some thought I still find myself unclear on the matter.

If I understand Pat's answer he seems to be saying that composers of the era were not consciously writing music that obeyed these principles. So am I to take it that at no point was a composer thinking "Ok, so this is the fundamental voice-leading note here, and these other notes are not, and this fundamental note connects to this fundamental note, so these parallels in between don't count."? That is they were never thinking in terms of an underlying voice-leading structure?

Assuming this to be the case, and that they were not thinking in a kind of reverse way to voice-leading reduction, the question then becomes how were they thinking about voice-leading? We're they even thinking about it at all? Just how much were they even thinking in terms of voices? What for instance are we to make of some sweeping, 3 octave, rolling, left hand piano figuration? Is it really a collection of voices with octave doublings and the like, or is it actually just a harmonic cushion of sound?

The ultimate question that leads from this is really how are we supposed to think about voice-leading when composing? If composers of the past were unconscious of their voice-leading practice does this too mean that worrying about the intervallic relationships between voices, or even what may or may not be considered a voice, is something that should not really concern us?

Answer 1

If I understand Pat's answer he seems to be saying that composers of the era were not consciously writing music that obeyed these principles. So am I to take it that at no point was a composer thinking "Ok, so this is the fundamental voice-leading note here, and these other notes are not, and this fundamental note connects to this fundamental note, so these parallels in between don't count."? That is they were never thinking in terms of an underlying voice-leading structure?

Pat's point was that they weren't thinking about the musical "grammar" of their music, much like you aren't consciously thinking about your English grammar - you have thoughts and you just type them out as you're thinking about them, and by and large your grammar will make sense. You can do this because you've (presumably) been speaking English your entire life - you've been steeped in it. So too were the composers of old. You also need to remember that they didn't have theory classes like you do. They studied with professional composers who put them through intense contrapuntal exercises.

Assuming this to be the case, and that they were not thinking in a kind of reverse way to voice-leading reduction, the question then becomes how were they thinking about voice-leading? We're they even thinking about it at all? Just how much were they even thinking in terms of voices? What for instance are we to make of some sweeping, 3 octave, rolling, left hand piano figuration? Is it really a collection of voices with octave doublings and the like, or is it actually just a harmonic cushion of sound?

Especially during the Baroque period, composers conceived of their music horizontally - a by-product of their contrapuntal development. The extent of vertical voice relationships then consisted primarily of consonances and dissonances; how / when / why to resolve the dissonances when they sprang up. Composers by and large did not think in terms of chord progressions as many musicians do today. The reason again being that their musical practice stemmed from their musical training. They were trained to think of music polyphony, so that's what they wrote, not homophony.

I think it is important for you to realize there is a distinct difference between parallel motion and doubling a part at the octave. I think it is also important for you to realize that you simply don't find sweeping, three octave piano figurations (apart from the fact that the piano wasn't invented until 1700) in Baroque and most Classic music. Romanticism would not wield its head until the early 1830's.

The ultimate question that leads from this is really how are we supposed to think about voice-leading when composing? If composers of the past were unconscious of their voice-leading practice does this too mean that worrying about the intervallic relationships between voices, or even what may or may not be considered a voice, is something that should not really concern us?

300 years ago, the definition of music was very clear for Western Europeans. For young / new musicians such as yourself, it can be difficult to grapple with where current musical trends fit in on the historical timeline. I think it is important to realize that there is a difference between how music was written then versus how it is written now. Learning about historical practices does not mean that it was the "correct" way and that everything you learn now is in contradiction to it.

The primary difference here is that over the past centuries, composers have demonstrated what is possible. While I'm not saying that all possibilities have been exhausted, it appears that the focus has shifted from the model of: accept -> reject to what is it you're trying to say?

Invariably, you must be clearest with your intent. It is one thing if you ignore specific rules to convey your intent, it is another if you ignore them due to ignorance. Do not worry, musicians can tell the difference.

When I write music, I think about my intent, my goals. Do my voices need to be smooth or disjunct? Do I have voices or sound masses? Do I have individual textures? Music does not need to comprise voices. If intervallic relationships are important for your musical language, then yes, you should worry about them. If not, then don't.

Contemporary music is very situational. Because we musicians have proven what is possible, listeners generally accept the music at face value - their experience of the piece is contingent upon the rules (see, language and grammar) that you prescribe. Contemporary audiences don't listen with the same ears that you or I listen with. They will not slap their knee at how clever you are for starting a recapitulation in the parallel minor sub-dominant. They will not be upset if your [0,2,5,6] set is inverted the wrong way. They really don't care.

If you're going to write music, and I mean seriously write music, there are only two things you really need to think about:

• 1.) What am I trying to say?
• 2.) How am I trying to say it?

Once you answer those questions, the music writes itself.

Hope that helps.

Answer 2

am I to take it that at no point was a composer thinking... in terms of an underlying voice-leading structure?

I want to zoom in on this question about whether, historically, composers thought in terms of a higher-level framework. The answer is yes they did think in such ways, from the earliest reaches of the Common Practice period, and even before. Late Renaissance and Early Baroque treatises abound on the art of performing "diminutions" (or, "divisions") as a necessary way to elaborate a simple line.

In the page below, you can see such a Rennaissance treatise, containing pairs of measures that alternate between a simple pair of notes outlining an interval (descending 2nd, 3rd, 4th, and 5th), and an example of an elaborated version of that same interval. The idea is that when you saw that interval in a piece, you could replace it with the elaborated version.

Originally, these were done in improvisatory style, as a performance practice. But later, it became more common for composers to write these figurations directly into the parts, leading to the style of music being called "Baroque", meaning heavily ornamented, or extravagantly detailed. In as much as later styles of music can be seen as an evolution of Baroque music, this technique is a fundamental, and typically subconscious, part of the composition process.

For an understanding of how composers applied voice leading principles in such melodic lines, you should study Species Counterpoint, in which a simple cantus firmus is accompanied by an increasingly florid counterpoint.

As for keyboard and lute music specifically, much of the style was derived from the practice of basso continuo, in which a keyboard part (or other chordal instrument parts, such as lutes, guitars, and theorbos) could be improvised ("realized") from a bass line with numerical indications of which chords should be played. There is evidence that, when extemporaneously realizing a continuo part, voice-leading rules were often applied less strictly. Specifically, sometimes only the relationship between the bass voice and the top-most voice of the right hand was scrutinized -- in this case, the inner voices were viewed as less important "filler" and need not strictly follow all the rules in detail.

To support this claim, the following extended quote is from the opening commentary to Francesca Caccini's Il primo libro delle musiche of 1618: A Modern Critical Edition (2004). Emphasis is mine.

In his 1983 study John Walter Hill notes that the majority of seventeenth-century lute tabllature accompaniments are primarily chordal in nature and emphasize a strict homorhythm above the given bass line. They also display a strong disregard for both the smooth melodic line of the top voice and so-called "proper" voice leading. In particular, the harmonies are full of "consecutive perfect consonances."

Such parallelisms were clearly acceptable within the context of early Baroque basso continuo accompaniment, contrary to that of the contemporary polyphonic practice. This "pre-tonal" phenomenon is verified repeatedly in the writings of Viadana, Guidotti, Galilei, and Giulio Caccino himself. The main concerns of these "realizations" was clearly the production of as full a chordal accompaniment as possible within the limits of the instrument...

A similar harmonic paractice is also found in collections calling for the chitarra... This parallel practice suggests that lutenists and guitarists both used standardized chord forms or shapes that were easily played and remembered while accompanying solo songs, and that these shapes were employed with little or no attention to voice leading beyond occasional suspensions and candential patterns.

It is significant that keyboard realizations examined in the same study do not differ drastically from the various lute tablatures, revealing similar chordal characteristics...