Many of the songs in the 1844 edition of The Sacred Harp are original, representing composers including B.F. White and E.J. King, the compilers of the book. How did the songs that are not original enter the book? About half of the songs in The Sacred Harp match songs in William Walker’s The Southern Harmony (1840 ed.) exactly. What was the origin of the songs in this book?
My thesis is that neither William Walker, B.F. White, nor E.J. King had access to enormous libraries, so most of the pieces in The Sacred Harp or The Southern Harmony are either original or copies or edited versions of songs from one of a handful of books. In short, the principal sources of The Sacred Harp (1844) are The Southern Harmony (1840 ed.), The Easy Instructor (ed. c.1815), The Missouri Harmony, and A Supplement to the Kentucky Harmony (eds. 2, 3). The principal sources of The Southern Harmony are Carden’s Missouri Harmony (1830s ed.), Moore’s Columbian Harmony, Davisson’s A Supplement to the Kentucky Harmony (ed. 2), Carden’s Western Harmony, and Wyeth’s Repository, Parts I and II.
Method
My method is to identify all possible sources for a song, using the Temperley ’s Hymn Tune Index (HTI) and Pappas’s Southern and Western American Sacred Music and Influential Sources (SWASMIS), two databases containing over 100,000 indexed tunes. Then I compare the versions to find exact or near-exact matches, if they exist. I have compared entire arrangements, not just the tenor parts, including the text, attribution, and typography of the page.
I’ve summarized my findings in the accompanying tables. Many songs have a unique match in previously published tunebooks, producing a list of “known sources” that include The Missouri Harmony, The Easy Instructor, and others. Where a song has a match in multiple books, I attribute it to a known source if possible, rather than introducing a new source to my list. The number of songs assigned to each source is somewhat fluid, since there may be identical versions in two or more books. For example, Moore’s Columbian Harmony and Davisson’s A Supplement to the Kentucky Harmony, ed.2, have songs in common that match exactly. When making a tally of songs in each source, songs that match multiple “known sources” are attributed to the book that appears to be the most used source. In the preceding example, since significantly more songs are attributed to Columbian Harmony than A Supplement, I attribute a song that matches both books to Columbian Harmony.
Results
In short, the principal sources of The Sacred Harp (1844) are The Southern Harmony (1840 ed.), The Easy Instructor (ed. c.1815), The Missouri Harmony, and A Supplement to the Kentucky Harmony (eds. 2, 3). (The 1850 edition introduced The Hesperian Harp as a principal source.) The principal sources of The Southern Harmony are Carden’s Missouri Harmony (1830s ed.), Moore’s Columbian Harmony, Davisson’s A Supplement to the Kentucky Harmony (ed. 2), Carden’s Western Harmony, and Wyeth’s Repository, Parts I and II. Other shape-note sources used include Carrell’s Songs of Zion, Hickok and Fleming’s Evangelical Musick, Leavitt’s The Christian Lyre, Clayton and Carrell’s Virginia Harmony, Caldwell’s Union Harmony, and Johnson’s Tennessee Harmony (ed.2). These last three contribute only one song apiece. Either Davisson’s Kentucky Harmony was not used, or its contributions are indistinguishable from The Missouri Harmony, which drew heavily on The Kentucky Harmony. I think the former is more likely, because The Missouri Harmony was widely available and remained in print into the 1840s, while the final edition of The Kentucky Harmony dates to 1821.
Walker mentions in the introduction to The Southern Harmony that he sometimes combined parts from different arrangements. Eighteen songs in The Southern Harmony are what I have called “composite” arrangements. They do not match published versions, but aren’t completely new. They combine parts from two or more sources and/or have been edited (by Walker or unknown others). Some composite pieces (like THE TRUMPET) have a new treble added to a published two-part song. White & King added trebles to two songs but only minimally altered or subtracted existing parts in published songs.
I had previously assumed that Walker omitted alto parts from New England pieces because he favored a three-part texture. While it is true that all of Walker’s own compositions or arrangements of this time period, except ALABAMA, lack an alto, the deciding factor between omitting or retaining the alto appears to be the use of the alto clef. Walker retained only one alto in the alto clef (FAREWELL ANTHEM, where the alto has solos), while including almost all of the alto parts that were written in the treble or bass clef in his source tunebooks. He changed the clef of the alto from treble to bass in two songs. He never rewrote alto clef parts using the treble or bass clef. In contrast, White and King did not omit alto parts from songs in their source tunebooks. Subsequent nineteenth-century revisions of The Sacred Harp even contain newly composed songs or added altos using the alto clef.
The role of round-note sources is mysterious. Only a handful of pieces in each book had been published exclusively in round notes prior to their inclusion in The Southern Harmony or The Sacred Harp. Most published round-note pieces have been significantly altered from their original versions. For example, the bass fuge entrance in MORNING doesn’t match the original in Pilsbury’s United States Sacred Harmony and the alto is omitted. It is possible that the compilers found these pieces in manuscript collections, rather than books. HANOVER appears in print only once – in Russ & Poor’s obscure Uranian Harmony (1791) – but was included in Amzi Chapin’s manuscript collection (c.1798), which draws heavily on Uranian Harmony. Chapin’s is the only instance of HANOVER indexed in SWASMIS prior to the song’s appearance in The Southern Harmony, in a different key, with different words, without an alto, and with a few other alterations. It’s far more likely that the piece was circulated in manuscript, perhaps by Chapin himself, than that Walker had a copy of Uranian Harmony. Alternately, melodies published in round notes, and even harmony parts, may have circulated through oral tradition. This may account for the garbled versions of round-note tunes like MOUNTVILLE (a parody of ROMAINE or DUNKIRK) and INVITATION (arr. King from Smith’s PSALM 119) and even for Walker’s altered version of HANOVER.
Both Walker and White & King claim that their books include “nearly one hundred pieces never before published.” This is a gross exaggeration only if we count pieces that are attributed to local composers. However, if we generously interpret “never before published” to include original songs and arrangements, “composite” arrangements, and pieces that were only available in round-note collections (or perhaps found in manuscripts or the oral tradition), the number is almost 90 for both The Sacred Harp and the 1835 edition of The Southern Harmony and 100 for the 1840 Southern Harmony.
One of my motivations in studying the sources of The Sacred Harp is to understand how songs (and, in particular, alto parts) change across time. There’s a striking contrast between Walker, White & King, and Davisson. White & King made almost no changes to songs that they imported into The Sacred Harp from other sources. As mentioned before, Walker omitted a number of altos, made edits to some songs, and added or combined parts from several sources. Davisson’s changes are mostly to the alto part. He typically altered altos to make the parts more interesting and higher. He refers in his rudiments to women singing the alto part, and perhaps that is the reason for making the part higher. In addition, he rewrote alto clef parts in treble or bass clef. As The Missouri Harmony is chiefly derived from Davisson’s books, a number of Davisson altos are in The Sacred Harp today.
| SH via SoH | in SH only | Total | “never before published” | ||||
| SH original or original arr | n/a | 82 | 82 | 33.88% | 88 | 36.36% | |
| SoH original, original arr, or composite* | 53 | n/a | 53 | 21.90% | |||
| Missouri Harmony | 36 | 6 | 42 | 17.36% | |||
| Moore | 13 | 0 | 13 | 5.37% | |||
| Easy Instructor | 0 | 13 | 13 | 5.37% | |||
| Western Harmony | 9 | 0 | 9 | 3.72% | |||
| Supplement to the Kentucky Harmony | 2 | 6 | 8 | 3.31% | |||
| Wyeth I or II | 5 | 3 | 8 | 3.31% | |||
| misc shapenote** | 4 | 4 | 8 | 3.31% | |||
| round note or manuscript | 1 | 5 | 6 | 2.48% | |||
| 123 | 119 | 242 | |||||
| 50.83% | 49.17% |
*there are `40 SoH original or original arrangements in the SH and 13 SoH “composites,” meaning arrangements that combine more than one different published version or where additional parts were added to published songs
**miscellaneous shape-note sources
| also in SH | unique SoH | total | pct | “never before published” in 1840 ed | |||
| SoH original, original arr, or composite* | 53 | 44 | 97 | 43.89% | 101 | 43.89% | |
| Missouri Harmony | 36 | 27 | 63 | 28.51% | |||
| Moore | 13 | 8 | 21 | 9.50% | |||
| Western Harmony | 9 | 1 | 10 | 4.52% | |||
| Wyeth I or II | 5 | 4 | 9 | 4.07% | |||
| Supplement to the Kentucky Harmony | 2 | 5 | 7 | 3.17% | |||
| misc shapenote** | 4 | 6 | 10 | 4.52% | |||
| round note or manuscript | 1 | 3 | 4 | 1.81% | |||
| 123 | 98 | 221 | |||||
| 55.66% | 44.34% |
Miscellaneous Sources
- Carrell’s Songs of Zion
- Caldwell’s Union Harmony
- Johnson’s Tennessee Harmony
- Hickok & Fleming’s Evangelical Musick
- Cumberland Harmony or Sweets of Music
- Virginia Harmony
- Christian Lyre
- Carden’s United States Harmony