Hostname: page-component-78c5997874-s2hrs Total loading time: 0 Render date: 2024-11-08T08:24:16.868Z Has data issue: false hasContentIssue false
  • English
  • Français

Computer Applications to Bartók's Serbo-Croatian Material1

Published online by Cambridge University Press:  04 February 2010

Benjamin Suchoff
Get access Rights & Permissions [Opens in a new window]

Extract

In the Introduction to his study of the Parry Collection of Yugoslav folk songs Béla Bartók contrasts ‘lexicographical’ and ‘grammatical’ principles in the systematic grouping of folk melodies. Although the former makes it possible to locate a melody by comparatively simple means, Bartók states, the latter is to be favoured; it is of higher importance “to get a clear idea of the relationship of melodies than to be able to locate them easily”. He then enumerates the components of his own grammatical system which, in order of weight, include: section structure, metric (syllabic) structure, rhythmical character, final notes of sections (caesurae), range, and so forth (cf. Plate 1 column headings).

Type
Research Article
Information
Tempo , Issue 80 , Spring 1967 , pp. 15 - 19
Copyright
Copyright © Cambridge University Press 1967

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

2. Bartok, Bela and Lord, Albert B., Serbo-Croatian Folk Songs, New York, 1951 Google Scholar.

3. Ibid, p. 15.

4. The fair copy, on master sheets, is deposited in the Music Library of Columbia University. The final copy, corrected Ozalid prints of the master sheets (Plate 1) is contained in the New York Bartok Archives. The Tabulation, originally intended by Bartok to be incorporated in Serbo-Croatian Folk Songs, is being prepared by this writer for publication by the Archives.

5. See Bator, Victor, The Bela Bartok Archives: History and Catalog (New York, 1963, pp. 32–33)Google Scholar for a description of the materials.

6. The appearance of omitted variants as the second half of certain four-section melodies (whose classification is based on structural aspects of the first half) further complicated the search (cf. Remarks to ‘Tab. of Mat.’ Nos. 1439 and 1444 in Plate 1).

7. The choice was made on the basis of the comparatively small number of melodies (75), with few exceptions also in skeleton form (that is, stripped of ornamental notes. See Plate 2, measures 1–6, small notation under stanza 1.), which enabled the writer to organize and encode the material for data processing during the Harpur College (SUNY Binghamton) seminar on ‘Music Research and the Computer’ August 14–26, 1966 Google Scholar.

8. It should be noted that content structure was determined by Bartok himself (‘Tab. of Mat.’) pp. 136–138). Thus, each melody section bearing a different capital letter designation (A, B, etc.) was considered as a separate entity for purposes of the project. Cases of disagreement with the author's analysis were limited to strings numbered 6, 12, 105, 118, 130, 147, 148, and 151. These melody sections are considcred by Bartok to be slight variants of preceding ones, although to the writer, at least, they appear to be of different contentstructure. Furthermore, strings 36 and 37 were designated as A B content structure; to the writer they seem almost identical. (It is interesting to note that the computer agrees with Bartok on but two items: strings 146 and 148, 150 and 151).

9. The Ford-Columbia code was selected for three cogent reasons: (a) its graphic-oriented and mnemonic features make it possible even for the non-musician to quickly master the encoding of folk tune incipits, (b) it had been successfully used as data for an already operative computer program along similar lines to the writer's project, and (c) the possibility, in the near future, of obtaining computer output in the form of music notation reflective of the highest quality of the engraver's art.

10. Programmed in FORTRAN by Dennis P. Geller, for use with the IBM 1130 Computer. Operation time was approximately N/9 minutes (about seventeen minutes for the one hundred and fifty-six punchcards). Use of the later model IBM Systems 360 would probably reduce the time by a large factor.

11. Department of Music, State University of New York at Binghamton. A more detailed description of this is presented by Professor Lincoln, in Computers and Humanistic Research (Bowles, Edmund A., Editor), Prentice-Hall (New York), 1967 Google Scholar. See also Bowles's, survey ‘Computerized Research in the Humanities,’ ACLS Newsletter (New York), p. 38 Google Scholar, for additional information about the Ford-Columbia code.

12. Repeated notes, intervallic quality, and rhythmical character are omitted from consideration in this indexing system (although they can be programmed, of course, for more refined comparative purposes). The reader may be interested to note that Zoltan Kodaly concurred with the writer's opinion that the system thus delimited is valid for the lexicographical indexing of Bartok's folk music collections. Professor Kodaly, who reviewed the project at a conference with the writer and Stefan Bauer-Mengelberg (at the New York Bartok Archives on September 16, 1966), added that rhythmical aspects would have to be considered when treating the Hungarian material.

13. It will be recalled—as described in footnote 8, above—that the writer (and the computer!) considers string 105 to be more a B than an Av content structure.

14. In the note to melody No. 37 (SCFS, p. 241), Bartok states that this melody has no variants, and in that of No. 1 no Parry melody is listed as a variant (Ibid., p. 231).

15. SCFS, p. 37.

16. See SCFS, Notes to the Melodies, p. 242.

17. Bartok has established that, unlike West European four-section melodies in which the imperfect cadence occurs at the end of the second section, Serbo-Croatian melodies frequently end with an imperfect cadence which functions as a perfect one. The tonus finalis (G1), to which all melodies are set in Bartok's system of transposition, actually represents here the second degree of an F major or minor (or other variety) scale fragment (SCFS, pp. 59–61; see also Suchoff, B., Guide to the Mikrokosmos of Bela Bartok [Boosey & Hawkes, New York] pp. 16, 45, 59Google Scholar, for Bartok's comments on the so-called Yugoslav cadence).