"Centonization and Concordance in the American Southern Uplands
 Folksong Melody: a Study of the Musical Generative and
Transmittive Processes of an Oral Tradition"


J. Marshall Bevil, Ph.D.

Ann Arbor: University Microfilms, International, 1985 (UMI No. 8423854, "Dissertation Services"); Microfiche/-film of Ph.D. - Musicology dissertation, University of North Texas, 1984. Known locations of copies: author; University of North Texas; University of Texas, Austin; University of California (Los Angeles and Berkeley campuses); University of Washington (State); Brown University; Princeton University; Welsh Folk Museum, Sain Ffagan, Glamorganshire, U.K.; U.S. Library of Congress.



       This study presents a theory of melodic creation, transmission, memory, and recall within the Anglo- and Celtic-American cultures of lower Appalachia, from the time of the earliest European settlers until around 1935 and, on a more limited basis, from the mid-thirties to the present. The various aspects of this theory and its attendant hypotheses are based upon 1) consideration of earlier published ideas, 2) currently accepted theories of perception, storage, and recall of information, and 3) the results of applying an analytical system based in part on indigenous concepts of scale and formal structure in the folk tune. The analytical procedure, which is supported by a set of computer programs written by the investigator, yields information concerning both the behavior of the individual melody and the relationships between tunes. Basic principles and practices of generative-transformational linguistics were adapted and melded with recognized, established  views of oral-aural musical transmission in the formulation of the present corpus of theories and hypotheses.
Sources utilized in the research for the dissertation included published studies of folksong melody, reliable tune collections, and previous examinations of the the psychological bases of learning and recall. Also important was an early treatise pertaining to scale in traditional Celtic music, with a modern interpretation of that document. A final body of source material was a small group of sound-recordings that were transcribed by the investigator.

                                     Theses and Conclusions


        The primary thesis of this study is that the centonate, or oral transmittive, process in the species of folksong under consideration is controlled by both an overall  melodic morphological norm and a gamut of five overlapping anhemitonic octave divisions. Most melodies within the genres examined in this study (mainly ballads and folk hymns, with some examples of other folksong species) are essentially variants of the broad pattern formed by the interaction of those two phenomena.
        A secondary thesis is that the creation, transmission, and refashioning of a tune is subject to the Gestalt Law of Closure, an important axiom of cognitive psychology. In sum, that theorem holds that the memory trace of a perceived entity (in the present instance, a complete melody or dual-phrase melodic strain, or period) is most accurate at its beginning and ending points. Other cognitive concepts, including that of functional mapping, underlie much of the present theory of melodic fashioning, transmission and perception, assimilation and recall, preservation, and decay.
   Another secondary thesis is that melodic transmission, preservation, and change within a musical oral tradition parallels in many respects the processes of oral language in a pre-, non-, or minimally literate culture. For that reason, the analytical procedure developed for this investigation  draws on linguistic principles, especially those of the generative-transformational school, as cited above ("Introduction").

Motivic Similitude

        Within the behavioral constraints outlined above, both variation and preservation occur in two interdependent spheres, the first of which comprises the formulaic motifs that open and close strains and specifically define melodic contours. Many of those patterns, known as primary cells, are organized around blocks of mostly adjacent tones in the anhemitonic gamut. Both direct and mutated anticipations and restatements of the formulas within the melody often provide much of the material that fills the spaces between the initial and terminal points. Every transcribed melody that was examined in this study had the pitch, rhythmic, and durational sequences of each of its primary cells encoded and placed in an array that was stored as a textfile. That file was read later by a programmed computer as part of the mechanically-assisted phase of the comparative analysis.

Overall Contour Correspondence

        The second area of preservational and variational activity is overall melodic contour, which includes and is in part defined by the primary cells. Contour exists at a large number of structural and perceptual levels. For convenience and in keeping with established principles of layer analysis, they are grouped into three strata. The first of those, known as the elemental level, is based on the opening and closing notes of strains and the primary stressed beats of all intervening measures. The second, or broadly-detailed, level encompasses the first and last three different notes of each phrase (except in a few relatively rare instances), together with primary and secondary stressed beats of internal measures. The third, or specifically detailed, level also includes unstressed beats and the sounding afterbeats of stressed notes and rests, along with the three-note openings and closings of internal phrases and the primary cells that open and close the strains. An important difference between the specific details and the broad details is that the former group consistently includes repeated notes that occur within primary cells while the latter group normally does not. Like the primary cells, each of the three layers was encoded, and the codes for its sequence of sounding events were placed in an array that was stored as a textfile and later accessed by the computer.


                                  Analytical Procedures Employed

       Seventeen sets of parameters were formulated prior to the actual testing process by performing numerous comparisons of both actual and contrived melodies representing the full range of kinship and disparity. Following the breakdown of transcribed tunes into their structural-pereptual layers, preliminary manual comparisons were made of both primary-cell and contour sequences in order to estimate the nature of the melodic relationship in each individual case.   Kinship then was measured by the computer according to tonal and rhythmic correspondence between primary cells alone and in combination, contour correspondence at both corresponding and non-corresponding levels, and various combinations of those concordances.  The pre-determined seventeen sets of parameters were applied near the end of each computer-assisted phase of melodic comparison, and one or more conclusions were printed along with a data table of primary-cell and contour concordances.*   Both preliminary analyses and final determinations were made independently of the computer, which was used to expedite the handling of the large volume of data and to reduce clerical errors. A second set of computer programs was used in the same way to process the results of all the comparisons in a group. That procedure was performed both to assess the idiosyncratic differences between individual singers and to determine the ways in which those individuals adhered to prevailing regional norms of melodic preservation and change.

* Near the end of the dissertation research, a new program was created by the investigator to generate comparative graphs based
on the data in the stored contour arrays. No computer-generated graphs appear in the dissertation, but they have been used in
several subsequent investigations.


Abstract of "Scale in Southern Appalachian Folksong: a Reexamination"

Abstract of "A Paradigm of Folktune Preservation and Change within the Oral Tradition of a Southern Appalachian Community"

Abstract and Full Text, "And the Band Played On: Hypotheses Concerning What Music Was Performed
Near the Climax of the Titanic Disaster" (application of theories of oral-aural musical processes)



Dr. J. Marshall Bevil is a native of Houston, where he also currently lives. He is both a string music educator and a musicologist (B.Mus. with honors, Oklahoma Baptist University, 1970; M.Mus. - Musicology, University of North Texas, 1973; Ph.D. - Musicology, University of North Texas, 1984) with specialization in the history of bowed string instruments, oral-aural musical transmission, British and British-American folk music, and British academic music of the late nineteenth and early twentieth centuries. His doctoral dissertation has been published by University Microfilms, International (UMI No. 8423854, "Dissertation Services"), and he has published post-doctoral studies in professional journals and presented papers in his areas of specialization at regional, national, and international academic convocations in both the United States and Great Britain. He also is the author of encyclopedia articles on John Avery Lomax, Alan Lomax, and Percy Aldridge Grainger; and he has published on the Internet. In addition to his pedagogic and academic pursuits, he is a performer on the Welsh crwth, a composer and arranger for string and vocal ensembles (publications on sibeliusmusic.com, from December of 2004), and a forensic musicological consultant and expert witness  in copyright and intellectual property misappropriation disputes ( links:  1    2  )

                                                                                                     E-Mail Link