The rudiments of The Sacred Harp and other shape-note tunebooks have traditionally categorized the intervals between any two degrees of the scale, or dyads, as either concords, “which produce harmony when sounded together,” or discords, “which, when sounded together, produce a disagreeable sound to the ear.”1 The concords were fifths and octaves (or unisons), which were termed “perfect” concords, and thirds and sixths, termed “imperfect” concords, while seconds, fourths,2 and sevenths were considered discords.
Most commenters have agreed that discords are not categorically barred from Sacred Harp music, but that they can, by alternation with or resolution into “perfect” chords, make the latter all the more sweet by contrast. Or, by allowing for more attractive or convincing melodic motion in an individual part, discords can be used “in such a manner and place as to show more fully the power and beauty of music.”3 The proper use of discords was regarded as an advanced and subjective element of composition, however, and little has been written regarding when and how a discord could be “justified.”
One striking discord occurs a number of times in The Sacred Harp, in works of major composers from all periods of the book’s history. Perhaps they all simply liked the way it sounded, but it may also suggest something more general regarding the stylistic treatment of discords in Sacred Harp part-writing. This “canonical discord” is a chord consisting of the first, fourth, and seventh degrees (1-4-7) of the minor scale—a “stacked fourths” chord creating a dissonant seventh between the outer notes. It usually occurs as the parts pass from one voicing of the tonic minor chord to another. Perhaps its most straightforward appearance is in D. P. White’s “Song to the Lamb”4 (p. 138t in The Sacred Harp, Fourth Edition, 1870), where it is used twice:
In this progression, the dissonant notes serve a passing or neighboring function between consonant intervals, comparable to the use of non-chordal accessory tones in unaccented positions (for example, on the second of a pair of eighth notes). However, composers clearly felt that the logic of the progression was so strong that the 1-4-7 chord could be used even in accented (on-beat) positions, as in “Logan” (p. 302 in The Sacred Harp: 1991 Edition) and “The Morning Trumpet” (p. 85):
Further examples of the 1-4-7 chord show how variations—such as elaboration of the melodic lines or displacement of one of the bass notes—may be introduced while the underlying progression remains the same:
What is happening in this progression? Can this discord’s recurrence be explained in terms of some kind of underlying logic or intuitive procedure?
My interpretation of the “justification” for this discord is that the composers felt that if two voices (parts) were bound together in a sufficiently strong relationship, one of the voices could be brought into dissonance with a third part. To understand how this “binding together” can occur, it may help to use terminology from classical counterpoint or voice leading theory, describing the possible relationships between two voices in two successive dyads:
- parallel motion, where the voices move the same distance (interval) in the same direction;
- similar motion, where the voices move in the same direction but different intervals;
- contrary motion, where the voices move in opposite directions; and
- oblique motion, where one voice remains on the same note while the other moves.6
In a three- or four-voice polyphonic texture like that of Sacred Harp music, combinations or compounds of two or more of these types of motion may occur in the succession of any two chords. Varying and contrasting different types or combinations of motion is an important element of Sacred Harp composition. Parallel motion has long been noted in Sacred Harp music (especially its free use of “forbidden” parallel fifths and octaves), but all four types are used in distinctive and characteristic ways; for example, the voice-crossing which characterizes dispersed harmony implies the use of contrary motion.
If we reduce the harmonic progression from the above tunes to a sort of basic form as a passing sonority between two different voicings of the tonic minor chord, we can see how the tenor and treble/alto are bound together strongly through parallel motion in perfect fourths, while oblique motion is used between the tenor and bass. When these treatments—both, in themselves, common and straightforward—are combined, a 1-4-7 chord arises:
Another 1-4-7 chord occurs in “Praise God” (p. 328), in the context of a different underlying harmonic progression (towards the dominant), but through the same procedure. In this case the lower voices, moving upwards, are bound together with strong parallel motion, while oblique motion occurs against the upper voice (treble):
This analysis, whatever its merit or relevance as a theoretical explanation, describes why in practice this discord, unlike some, can generally be rendered by a class with relatively little difficulty.7 The fact that one of the voices in the discord has a strong harmonic relationship with the tenor, while the other (the bass) usually has merely to repeat its note from the previous chord, facilitates its accurate execution.
What is its musical effect? In itself, a “stacked fourths” chord has, to my mind, a stark, open, dark-hued sound.8 As a passing chord between two tonic minor chords, in a larger context of at least temporary harmonic stasis, it has the effect of a sudden (however brief) change of color, a certain wildness, straining against the bounds of the tonic chord. In the language of the 1991 Edition rudiments, it can very much be heard as creating a “tension” that is “resolved” with the return to the tonic.
I hope that this study of a “canonical discord,” besides attempting to shed light on a little-discussed topic, and providing a point of introduction to the types of contrapuntal motion and their application in Sacred Harp music, may serve as a humble plea on behalf of this useful and expressive sonority, which has been subjected to well-meaning emendation in recent Sacred Harps. “The Morning Trumpet” has been edited in later printings of the 1991 Edition, and both instances of the chord in “Song to the Lamb” were altered in the 2012 Cooper book.9 I believe that surveying the range and variety of occurrences of this chord will make clear that it is not an error in need of correction, but a characteristic manifestation of the richness, the capacity for complexity and subtlety, of Sacred Harp harmony, showing more fully indeed the power and beauty of music.
- B. F. White, “Rudiments of Music,” in The Sacred Harp, ed. B. F. White and E. J. King (Philadelphia: T. K. and P. G. Collins, 1844), 13. These rudiments were substantially adapted from William Walker’s Southern Harmony (1835); the sections on concords and discords, “Of Harmony and Composition” and “Of Intervals,” are repeated almost verbatim (including a parenthetical aside at one point from “the writer of these extracts”). The Southern Harmony rudiments, in turn, contain a certain amount of borrowing from William Moore’s Columbian Harmony (1829). [↩]
- The rudiments were often rather ambiguous about the fourth, traditionally classifying it as a discord, yet acknowledging that it was nonetheless used extensively. William Moore summed up the general attitude in his Columbian Harmony, classifying the interval as a discord but adding, “For my part, I cannot see why composers call the flat [perfect] fourths discords, when they all make use of them, and I consider the flat fourth one of the most agreeable sounds in nature.” The 1991 Edition rudiments reclassified the perfect fourth as a concord, with considerable justification in musical practice if not historical theory. [↩]
- White, “Rudiments of Music,” 14. [↩]
- The 1-4-7 chord occurs between two relative minor (submediant) chords in this major-key tune. The piece was removed in the 1911 Original Sacred Harp, the James edition, but retained in the Cooper revisions. White’s arrangement was in four parts—the alto, which creates the discord in m. 4, is not a later addition. [↩]
- Compare m. 8 of “The Fountain” (p. 397), m. 9 of “Morning Prayer,” and m. 9 of “Phoebus” (p. 173). [↩]
- The rudiments do not use this terminology, except for some references to parallel motion. However, Paine Denson’s description of “mutual tones” (“a tone belonging to both chords used in making a progression”) in his rudiments to the 1936 edition, where he advises that “the part that has the mutual tone in the beginning of a progression retain in into the next chord, thus binding the two chords together,” represents a clear formulation of the principle of oblique motion. [↩]
- As can be heard, for example, in Alan Lomax’s 1959 recordings of “Logan” and “The Morning Trumpet.” [↩]
- “Stacked fourths” (not necessarily 1-4-7) chords can occasionally be heard in other contexts, notably in m. 5 of R. B. Helms’ haunting “My Native Land” (p. 416t in The Sacred Harp: Revised Cooper Edition). [↩]
- In all three cases the bass note was changed from the 1-la to the 7-sol, substituting simple parallel motion for the original parallel-oblique compound. [↩]