Introduction

Four-part vocal style is a common format used for a range of performance, pedagogical and analytical purposes in Western common-practice music. The four parts in question are derived from the high and low voice ranges for generic “male” and “female” voices. These are, in descending order of pitch: Soprano, Alto, Tenor and Bass. Four-part vocal style involves simultaneous writing for these four voice types, either with one person or many people per part. Although we typically use the word “style” it is better thought of as a “texture.” 

There are a number of other terms used to describe four-part vocal style. Sometimes, it is simply called “four-part harmony.” Sometimes, we use the acronym derived from the four voice types as a substitute term: “SATB,” or “SATB style.” Another common term to describe this format is “chorale style,” which takes its name from the four-part chorale that became a mainstay of Lutheran church music from the seventeenth century. Because this tradition is often regarded as having culminated in the chorales of J. S. Bach, the term “Bach chorale style” is sometimes used. All these terms are largely interchangeable.

Before getting into the basics, it is important to acknowledge that the value of four-part vocal style as a pedagogical tool has been called into question for various reasons. Despite the fact that it has been almost ubiquitous in Western music theory for centuries, it does represent a very specific and stylistically and culturally narrow spectrum. It is often seen as overly fussy and concerned with “rules” (although this can be rather overstated and is usually more a case of poor teaching than something inherent in the format itself). Teaching four-part vocal style is also seen as privileging a hegemonistic (white, Christian, colonial, male, etc.) culture at the expense of other ways of learning about tonal music. It is sometimes considered anachronistic and irrelevant—after all, how many students of books such at this will find careers as eighteenth-century German kapellmeisters or church organists?—with a narrow and highly technical skillset.^[1]

While acknowledging these valid criticisms, we do use four-part vocal style in this book and use it in various ways. Despite its limitations, we contend that it still offers a useful lens through which to distill fundamental concepts and establishing some level of competence in understanding, writing and practicing the style retains intrinsic value for someone wanting to master Western common-practice tonality. We use it, therefore, mindful of its limitations and its biases alongside its residual value. Much the same can be said of other formats we refer to in this text—species counterpoint, keyboard style and figured bass, and so on.

The Basics

Range and Format

Each of the four voices in four-part vocal style has a generic range, as shown in Example 00. As the main notes (with stems) show, each voice generally encompasses a twelfth. Very occasionally notes just outside of these ranges (seldom more than a step) might be used (as shown by the smaller note heads in brackets). Also notice that the voice ranges overlap to a considerable extent, up to an octave, as shown at the bottom of the example.

Depending on the use and context, there are two ways of setting out the musical notation for four-part vocal style: open score and short score. These are sometimes referred to as full score and condensed or reduced score.  Example 00 presents the same phrase is in both layouts. Depending on the context, either the open score or short score layout might be used in practice. For pedagogical, theoretical and analytical purposes the short score layout is generally used.

In open score, each voice part has its own staff.  The tenor voice uses a sub octave treble clef, a treble clef with an 8 dangling from it. This indicates that the notes are sung an octave lower than written.^[2] The four staves are grouped in a system, with a bracket outside the left line. Bar lines are drawn on the staves only. The notes with stems follow the usual practice of upward stems for notes heads below the centre line and downward stems for note heads above the centre line.

In short score, the four voice parts are distributed on the great staff. Soprano and alto parts are written on the upper (treble clef) line; tenor and bass parts on the lower (bass clef) line. The tenor notes are now written at sung pitch.

The stem direction in short score is arranged so that the soprano and tenor notes always have upward stems, irrespective of where the note head is on the staff. Conversely, alto and bass notes always have downward stems.

If the soprano and alto or the tenor and bass happen to sing in unison on the same note value in short score, it is common practice to have them share the note head, which will then be double stemmed—see the third beat of bar 1 and the second beat of bar 2 in the short score version.

Lines and chords

Four-part vocal style very clearly exists on two interdependent axes. Each voice part carries a melody on the “horizontal” axis, which we call a line. Where the notes of each of these lines are heard simultaneously, on the “vertical” axis, they produce a chord. (The soprano line and the opening chord are shown enclosed in the dotted lines in Example 00.)

It is the balance and management of these complementary aspects—the linear and the chordal—that is the main challenge of understanding, writing and practicing this kind of music. It should also be clear from Example 00 that the two different layouts somewhat exaggerate one aspect at the expense of the other. The lines are very clear for each part in open score, whereas the chords are rendered more obviously in short score. Possibly because the vast majority of pedagogical texts present four-part vocal style music in the short score layout, students tend to become adept at writing “chords” in this style fairly quickly, but struggle more with the equally (if not more) important concept of managing lines. For this reason, it is a good idea to cultivate both approaches to notating this kind of music.

“RULES”

Four-part vocal music has a partially deserved reputation for having a great number of stuffy rules. Many students find this aspect of four-part writing discouraging and petty, although others enjoy the challenge of adhering to the rules while still producing music that is interesting and expressive. Like many rules, those applying to four-part vocal style have a reasonable basis in that they help to cultivate a coherent approach to writing in and understanding this kind of music. The rules, such as they are, are intended to help students create textures that manage the competing aspects of the linear and the chordal, as outlined above, and to produce consistent results.

However, this is not to say that the rules should never be broken. The history of music is littered with moments in which “breaking rules” led to new practices, which then became a “new normal”—until these new “rules” were in turn broken. In most cases of rule breaking, however, those who exceeded the existing norms did so with a thorough knowledge of the existing practice—think about Monteverdi’s conception of a prima and second prattica or Debussy’s reported quip: “I don’t write in the fugal style because I know it.”

With this in mind, we present a range of rules and tendencies around four-part vocal style. Some of these will be quite explicit, falling more in the category of a hard and fast “rule” while others will be more open to a flexible approach, being closer to tendencies or “good practice.” If you can go someway towards understanding and applying all of this, then you will have a better appreciation of when and how to break the rules, or a better understanding about why a composer might have chosen to do so in any particular instance.

We start with some basic rules around the distribution of parts.

Spacing and Voicing

Spacing refers to the intervals that separate the notes in the chordal (vertical) dimension. Voicing refers to the arrangement or ordering of notes of a chord.

The spacing and voicing of the notes in four-part vocal style (and many others) cannot be completely understood in isolation, they are dependent on the context of surrounding chords and the linear arrangement of all parts. However, we can use isolated chords to examine the basic principles.

In Example 00 three different voicings of individual chords are shown.  In each case, the bass note (and therefore the inversion) does not change, but the arrangement and order of other notes of the triad or seventh chord are different.

For spacing, as a general rule, it is best to avoid excessively large intervals between adjacent voices.  Common practice tends to avoid exceeding the following limits:

• No more than an octave between soprano and alto
• No more than an octave between alto and tenor
• No more than a twelfth between tenor and bass

Example 00 below shows the same chord progression, first with good spacing and then with poor spacing.  You might also notice, as a by product of contriving the second example, that the linear motion of the parts (especially the alto voice) is quite disjunct in places.  Poor spacing is often related to problems in the linear design of individual parts.

Spacing and voicing are obviously related—(b) is a “revoiced” version of (a). In (a) the interval between adjacent voices never exceeds an octave. In (b), the voicing of every chord involves an interval, either between alto and soprano or tenor and alto, that is more than an octave. Of course you will be able to find occasional examples in four-part vocal music of “excessive” spacing. However, in most cases there will be a clear contextual reason.

Within the “rules” for spacing and voicing, there is obviously a large range of possibilities for distributing the notes of a single chord.  Things get more complex once we encounter the context of moving from one chord to the next. However, it is worth taking a moment to think about the voicing of chords in isolation as the effect of different spacing is clearly audible. The notes of chords in four-part vocal (and many other styles) can be closely bunched together or (within stylistic limits) spread out. Consider the following examples of different spacing option for the same chords.

In (a) all the notes are as close as possible to each other (without doubling). This is often referred to as closed position (or sometimes close position). When sung or played, the effect is quite compact and dense. In (b) and (c) the three higher voices are moved progressively up from the bass voice.  We might say that the tenor, alto and soprano remain in closed position but as a whole, the voicing is no longer strictly in closed position and, despite the proximity of the three higher voices, we would already call this open position. In (d), (e) and (f), all voices are spread increasingly apart (although still within the limits outlines above). The effect as we progress through these positions becomes increasingly transparent.

Note Doubling

You will have probably noticed by now that when we apply triads in four-part vocal style, we have more voices (four) then notes in a chord (three).  This means that two voices will need to sing the same pitch class (either at an octave or double octave, or even as a unison). This is referred to as doubling.

Over the years, a great number of “rules” have developed around which notes in certain triads should and should not be doubled. These are of some value but the frequency with which some of them are broken (often for sound reasons) makes for considerable confusion as students progress in four-part writing. A further complication is encountered in that sometimes it is possible to omit a note altogether and triple another. So here is a disclaimer, before we consider the “rules”:

There are really only two hard and fast rules regarding note doubling in triads:

Rule 1: Never double the leading note

Rule 2: Always double the third of diminished triads.

The remaining “rules” are more in the nature of the things that just tend to happen when you mostly write efficient four-voice textures. The better you get at writing in the style, the less you will need to think consciously about these.

• For major and minor triads in either root position or first inversion, the default is to double the root. Sometimes doubling the fifth will be a better option. It is much more infrequent to double the third (with some exceptions).
• For second inversion triads it is almost always the best option to double the bass note (which will be the fifth of the chord).
• Try to have all the notes of the triad present in the chord—in some isolated instances it is possible (even desirable) to omit the fifth and treble the root.

In the example below, a four-part vocal progression is shown in both short and open score.  In the short score, each instance of note doubling is shown. The open score is there to give less cluttered view of the four lines of each part.

The vast majority of chords have their roots doubled.  The doubling of the fifth in the third chord of bar 1 does provide a more conjunct motion in both the alto voice, which would otherwise have had to leap to B. That would not be a fatal error, but the line is more efficient and smooth this way. In the submediant triad (iv) at the start of bar 3, we have doubled the third instead of the root.  We cold certainly have doubled the root instead; however doubling the of vi in such contexts is quite common. Moreover, the note in question (B) is the tonic degree.  This leads us to another way of thinking about note doubling.

If we think about the relationship of triads to the scale, we can take note of which triads contain the tonic, subdominant and dominant degrees. This is shown in Example 00 for a major key (D) and a minor key (G), omitting triads that are diminished.

Every triad has at least one member that is scale degree , or . Because these degrees are often considered the most stable, there is a case that doubling these notes within triads is also a good practice.  Like the “rules” above, however, this is not universal.

Voice crossing and voice overlapping

Voice crossing and voice overlapping are two related by slightly different phenomena.  Both are best avoided, more so the former; however, once again, there will be times when either is quite justified.  It is best to begin by avoiding them so, like for all “rules,” you develop a sense of when it is reasonable to “break” them.

Voice crossing occurs where a higher voice is heard below a lower voice (and vice versa). Voice overlapping occurs where a higher voice moves below a note held by a lower voice in the previous chord or vice versa). Examples of both are shown in Example 00.

In Example 00(a) the alto’s C4 is below the tenor’s E4 (or, putting the same thing another way, the tenor’s E4 is above the alto’s C4)—the parts have “crossed” and the chord should be revoiced with alto singing E4 and tenor signing C4.  A similar situation existing in (b) regarding the voicing of G4 and C5 (soprano and alto). Again, they should be revoiced.

In (c) the voicing of the middle chord (I) creates an overlapping between alto and tenor. The alto’s E4 enters below the tenor’s F4 in the previous chord and then, as we go to the final chord, the tenor’s F4 enters above the alto’s E4 in the previous chord. The fix for this is to revioce the middle chord, giving the E4 (the third of I) to the tenor and doubling the fifth of I, G, in the alto (rather than the tenor doubling the root).

In (d) the large and awkward downward leap of a minor seventh in the alto (A4–B3) in the context of relatively tight spacing creates a both an overlap with the tenor’s previous E4 as well as a crossing because the tenor stays on E4. Additionally, the soprano’s leap from C3 to G4 overlaps the alto’s A4 in the first chord. The problem is fixed by simply exchanging the B3 and the G4 between the alto and soprano.

Parallel prohibitions

The dreaded “parallel fifths” and “parallel octaves” have become an infamous bane in the life of every student of Western music theory at some point or other. As far as “rules” are concerned, these come under the most strictly adhered to category. While we could get into an argument around the case for this prohibition, and many will cite some famous examples of composers (including Bach) violating these rules (very occasionally), for now, we will outline them and show you what to look out for and avoid. Also, note that these rules are much the same as for two-part counterpoint, the challenge now being to detect them within the context of four parts.

The rules apply to the combination of any two voices as you go from one chord to the next (by “part” we mean voice or instrumental line equally):

• Do not write consecutive parallel perfect fifths between any two parts
• Do note write consecutive parallel octaves between any two parts
• Likewise, do not write consecutive contrary motion perfect fifths or octaves between any two parts
• These rules apply also to unisons and to compound intervals (twelfths, fifteenths)

There is one caveat to the above, and it occurs when one of a pair of fifths is diminished—that is, you may proceed in parallel from perfect fifth to a diminished fifth—see (e) in Example 00.

Additionally, for a parallel fifth or octave error to occur, the voices concerns have to move—repeating a perfect fifth or octave is not an error. This is because we could simply understand the repeated pair of notes as a single longer value—see (f) in Example 00.

Here are some examples to start with:

Exposed Fifths and Octaves

An extension of the restrictions on parallel fifths and octaves is found in the concept of “exposed” perfect fifths and octaves. This rule relates only to the bass and soprano voices (or to the outer voices of any vocal or instrumental textures). It is as follows: Avoid similar motion between the bass and the soprano to a perfect fifth or octave (including compounds) when the soprano is not moving by step. Some examples are given below: