Bayard, Samuel P. 1950. "Prolegomena to a Study of the Principal Melodic Families of British-American Folk Song." Journal of American Folklore 63/247:1-44.
"a prelude to ... distinguishing of variant forms of different individual folk tunes in tradition" (p. 1) from the perspective of the "student" or "collector". Sees connection between his work and the subjects of "folk memory" and poetics of folk music, defends his approach to the music independent of song texts (p. 4). The bulk of the article is in three parts (p. 5): 1) a contemplation of the hopelessness of tracing influence and development among tune variations in any causal manner, a call to concentrate on description and analysis of internal evidence, and a critique of constructs such as "formula," "type," and "style" used by other folk song analysts; 2) a classification of variations that typically occur, the melodic locations where variations most often occur, a delineation of the process of tune family formation, and a definition of tune family; and 3) an analysis of tune repertories in British Isles, and differences in analogous repertory the U.S. Concludes by praising the glories of British folk music, and by criticizing the proponents of gesunkenes Kulturgut theory who claim that all folk tradition stems from the corruption of cultivated tradition.

In the first section (pp. 5-17), author lays out his assumptions for analysis of tune variation: the most important characteristics for his groupings are "correspondances in melodic line" (contour?) and "order (or recurrence) of stressed notes or tones.... Range, rhythm, and phrase-order are all variable, and less dependable" (6). The goal is "to confirm... relations of a genetic sort.... to ascertain, if possible, whether the resemblances which meet [the] eye are the products of chance or the actual manifestations of some fundamental individual design that endures throughout repeated alterations" (7). Here, but not everywhere, he seems to be talking about similar tunes rather than multiple performances of the same tune. Downplays the role of formula as the fundamental building block of melodies, sees them as more fluid in their identity, and would rather study melodies as wholes. Spends five pages refuting a proposal of Bartok's concerning style: rather than treating each body of similar melodies as a "style," he sees style as a local identity marker that cuts across melodies, and that a melody can appear in multiple localities under the guises of different "styles" (15).

In the second section (pp. 17-33), author presents eight major types of variation: 1) large leaps up or down; 2) change in tempo; 3) change in rhythm, perhaps to accommodate text; 4) change in mode; 5) influence of other melodies; 6) repetition of formula, replacing contrasting material; 7) shifting order of strains; 8) corrupt rendition. With type #5, notes that dance tunes are more likely to mix and match phrases from different tune families than song tunes. With type #8, discusses at length (pp. 19-24) the case of two variants of the same song, one of which is sung to one strain while the other to two, and maintains that it is impossible to know whether this indicates a "building up" of complexity or a "wearing down." Postulates the existence of the "mixed tune" which borrows from two distinct tune families, and which must necessarily be newer that the "clear-cut" tunes from whose families it borrows. A ninth distinct type of variation is the substitution of cadential formulae (p. 28). A tune is most like to vary at the very beginning, the very end, or at the midpoint cadence, and changes at these locations is least likely to affect melodic identity. Outlines 3 types of tunes (29-30): 1) those that belong to large, diverse tune families; 2) those that belong to smaller tune families, some well-defined and some diffuse; and 3) tunes that are "indeterminate," that don't fit at all, that are "chance," "hybrid," "local." Is clearly uncomfortable with the last category, but feels compelled to mention it. Sees individual composers as irrelevant, given the similarities he finds among tunes not all of which could be composed by the same person(30). Gives a 12-step process of formation of a tune family, followed by a definition: "A tune family is a group of melodies showing basic interrelation by means of constant melodic correspondence, and presumably owing their mutual likeness to descent from a single air that has assumed multiple forms through processes of variation, imitation, and assimilation" (33). The 12 steps outline a complex set of possibilities of the variation, imitation, and assimilation to which the definition refers, not summarizable here. One of them is particularly notable with respect to its cog-sci slant, however: "4. This merging of parent and derivative variants would tend to fix the basic outlines and important intervals of a tune firmly in the minds of the singing people--so that the pattern and principal formulae of the tune would become a part of the way in which they habitually thought melodically" (31).

In the final section (pp. 33-42), author applies concepts he has introduced generally to folk music of the British Isles as it exists there and as it traveled to North America. Identifies six repertory groups: English, Lowland Scots, Welsh, Irish, Highland and Hebridean Scots, and sea shanties (marine work-songs). Reinforces his distinction between style and tune repertory by discussing three main styles of British folk music--English, Irish, and Hebridean--as sharing a common repertory of tunes, as well as possessing some tunes unique to themselves. Compares irregular Hebridean melody construction with French. Discusses how British corpus travelled to North America in waves, and how it was also undergoing change on new soil throughout. Emphasizes difficulty of ascertaining provenance of a tune based on style alone, yet still seems compelled to discuss the intricate possibilities of tune history, and to attribute some tunes to particular locales, cf. the following two quotes: "The realization of how untraceable these melodies actually are has finally halted the rather futile controversies over the claims of different lands to various internationally current melodies of merit" (37); and, "we in America have to cope with a pretty complicated set of conditions in our efforts to discover approximately the true history of our individual folk songs and their music" (39). Offers a four-pronged approach for such a study, then eight "facts which must be recognized if we want to appreciate properly the relations between old-world and new-world folk song records," devoting at least a paragraph to each: 1) Scottish Gaelic repertory is not in N. American rep.; 2) Ditto for tunes peculiar to Irish rep.; 3) the common repertory of the British Isles is also dominant in N. America, and tune-versions from opposite sides of the ocean are surprisingly close to each other; 4) although the repertories are shared, the relative popularity of individual tunes varies quite a lot, based on frequency of occurrence in published collections; 5) new "nondescript," hybrid melodies have appeared in N. America; 6) although they have borrowed some from European folk and popular sources, Negro folk songs are by and large an "independent creation"; 7) there are both similarities and differences in text-tune combinations; 8) the known popular broadside tunes of 17th century "bear no relation whatever to the melodies of the British common repertory" (41). Unclear how the last point bears on transatlantic transmission.
Mark DeWitt folklore/folk music/European-American/cognition/ethnomusicology.