Exploring Ancient Rhythms with High-Tech Tools

by Susan Moke

John Johnson wears an unusual wrist watch--his watch has two faces. Johnson, who is an associate professor of folklore at Indiana University Bloomington, explains that one of his watch faces helps him keep track of the time in Bloomington, and the other helps him keep track of the time in Somalia, where he has done his field research for the last thirty years. But Johnson's double-faced watch tells him more than just the time of day in two disparate time zones. It reflects two different methods of counting time. In the Middle East, according to an older method of timekeeping than our own, the day begins at sundown rather than at midnight. Johnson's two-faced watch is also emblematic of his work, which involves applying the structural analyses of Western poetics to orally composed Somali verse. By investigating the scansion patterns of pastoral Somali poems--that is, by figuring out how the poems "keep time"--Johnson says he hopes to answer a larger question: How do people go about composing oral poetry?

Johnson uses sophisticated technological methods to investigate what in some circles might be seen as primitive poetic artifacts, but which turn out to be sophisticated compositions and performances. In Indiana University's Folklore Institute's Sound and Video Analysis and Instruction Laboratory (SAVAIL), he examines the particular, non-Western prosodic patterns of Somali verse by slowing recitations of Somali poetry down to one one-hundredth of a second. This allows him to isolate and analyze characteristics such as pitch, meter, and the poem's interplay with the musical accompaniment against which it is recited. Many of these characteristics are nearly inaudible to the untrained Western ear. Indeed, without the aid of high-tech computer analysis, Johnson's current conclusions about the formal elements and composition processes of Somali poetry would have been impossible. The multimedia delivery system in SAVAIL allows sound waves to be analyzed as visual spectrograms that chart both the pitch and duration of a stream of speech. Funded by grants from the Research Facilities Fund and the Office for Information Technologies, the laboratory consists of one NeXT and two Macintosh workstations, software, and peripheral devices. "It is," Johnson says, "one of the best computer analysis laboratories on campus, combining sound and visual analysis capabilities."

The oldest assumption about the creation of oral poetry was that people in predominantly oral cultures composed poems without the aid of writing implements, and then they and others memorized the poems. Challenging this theory is a now widely accepted idea that has come to be called "the oral formulaic thesis." Advanced in the late 1950s/early 1960s by Harvard scholars Milman Perry and A. B. Lord, the oral formulaic thesis contended that oral poetry, even oral epics, were composed according to predetermined formulae while they were being recited. We are acquainted with such oral-formulaic poetry in the form of archaic children's rhymes such as "Down by the seashore, Maryann/Down by the seashore sifting sand." Johnson's studies of the various genres of Somali poetry have led him to expand and revise the oral formulaic thesis.

Johnson explains that in Somali culture, where poetry is an all-pervasive part of nearly everyone's daily experience, four criteria--scansion, melody, topic, and function--act in concert to differentiate one genre from another. Students of Somali literature group the various genres of verse into three basic categories: classical poetry, which deals with politics and serious issues such as interclan relations; work poetry, which is specific to particular tasks such as herding camels or churning milk; and recreational or dance poetry, which might deal with such frivolous topics as romantic love--something Somalis view as a "two-year affliction, a disease to be gotten over," according to Johnson. Within these three categories, Johnson estimates there may be more than fifty poetic genres of Somali verse. Examination of the scansion patterns of the various genres has led him to conclude:

Research suggests that Somali classical poetry (gabay) consists of texts composed in private and memorized verbatim for public performance. Work songs, (hees) seem highly formulaic, though some of them are probably composed during performance, as the rhythms of work and poetry join. Finally, poetry in the dance songs (cayaar) is composed and recited simultaneously, but not within the formulaic method. . . . In my view, a more unified and all-encompassing theory of oral composition than has so far been proposed and which will embrace the oral-formulaic thesis, spontaneous composition, and verbatim memorization is called for. The study of the Somali tradition can potentially lead us to that theory.

The methodological demands of Johnson's investigations into the pitch, meter, and rhythms of Somali verse have caused him to return to the classroom as a student. During a leave of absence funded by a Lilly Open Fellowship, Johnson enrolled in music theory courses in IU's School of Music and hired a music tutor so that he could better understand the complicated relationship of the poetic syllables to the musical rhythms that determine how these poems are sung.

For more than a quarter of a century, Johnson has researched the meter of Somali poetry using the tools of the literary scholar. Although every poem in the Somali generic repertoire can be sung to musical accompaniment and usually is in its natural context (even if the accompaniment is only the rhythm of hand clapping), Johnson notes that until three or four years ago, he had spent little time on the musical aspects of this verse. After extensive investigation into the music of Somali prosody, Johnson has concluded that "any research into African poetry that ignores its musical setting is a grave error." Using the musical paradigm, Johnson discovered silences in scansion that literary scholars completely ignore, but that music transcribers account for as "rests." Such silences or rests may be integral to the actual prosody of a given poem. This line of research has led Johnson to conclude that "in Somalia, poetry and music are not separate art forms, but are regarded by their indigenous composers and audiences as one and the same act of creativity. In Somali dual rhythmic verbal art, linguistic meter may be considered only one of the several meters that contribute to the overall performance of any given genre." He cautions that "music theorists and ethnomusicologists who study African poetry ignoring its linguistic and prosodic components make the same error as the literary scholar, but in reverse."

Johnson's research methodology has been productively transformed by multimedia technology. He received entre into the world of sound and video computer analysis through the technological tutelage of one of his doctoral students. Nina Fales, who was Johnson's technology mentor, is now the director of the Folklore Institute's Sound and Video Analysis and Instruction Laboratory. Because of the research Johnson was able to conduct on computers, the chairperson of his department asked him to write a grant to duplicate his workstation for the Folklore Institute and to teach the methodology he had learned in his research to students in the folklore department. Johnson wrote the grant and the laboratory was funded. The SAVAIL multimedia delivery system allows Johnson and his students to answer questions they could not otherwise effectively investigate.

The laboratory consists of two workstations--one for sound analysis and one for video analysis. The sound analysis workstation, which is dedicated to recording and analyzing speech, song, and music, includes a Power Macintosh 7100 capable of digitally recording mono or stereo digital-quality sound at a 16 bit resolution 44.1 k/Hz sampling rate. It runs SoundScope, a software tool that can perform a wide range of analytical and statistical operations on recorded sounds and display the results of these operations graphically or numerically. The Power Macintosh also runs Finale, a program designed to perform all functions of musical notation and publishing. In addition to the Power Macintosh, SAVAIL's sound analysis workstation includes a NeXT computer that supports NeXTstep and the following applications: SpecDraw, software designed to allow extremely fine filtering of sound -using the graphic capabilities of MacDraw, SpecDraw allows the user to draw in frequential and temporal areas of pass- or stop-band filtering; Oedipe, a program for the editing, calculating, and displaying of two- (frequency-by-amplitude or frequency-by-time) and three-dimensional (frequency-by-amplitude-by-time) representations of sound; and Spectro-3.0, a program useful for specifying peak frequencies with greater accuracy than most analysis software can provide.

SAVAIL'S video analysis and postproduction workstation allows users to manipulate video, still images, and sound. It enables precise examinations of the relationships between speech and gesture, music and dance, and other aural/visual phenomena, and it provides editing capabilities. The video analysis workstation consists of a Macintosh Quadra 840 AV. The software used includes Media 100, a video postproduction system that allows users to convert video and audio signals recorded in various formats (Hi8, 8mm, SVHS, VHS, and audiocassette) into digitized information that can be stored for later use or comparative analysis. Media 100 also possesses sophisticated editing, titling, and visual effect capabilities that let users prepare finished programs for presentations. In addition to Media 100, the Macintosh Quadra 840 AV runs PhotoShop, software that allows users to manipulate images created in a variety of formats and that can be used in conjunction with Media 100 to extract and refine still images from video sources, and Aldus Fetch, a multimedia database manager that allows users to catalog, search, and access files containing sound, still images, and video created in a variety of formats. It also has a high-performance AV drive specifically designed for video applications; a high-capacity, removable DAT tape backup drive; two videotape recorders; and a video/audio monitor.

In a class titled Computer-Assisted Performance Analysis, Johnson uses his own research and methodology as an inquiry model for his students. The course requires students to learn the hardware and software in SAVAIL and to define research questions that may be explored on the available equipment. SAVAIL Director Nina Fales notes that students have used the laboratory's sound workstation to explore interesting and unusual areas such as "acoustic correlates to the quality of heaviness in heavy metal music as it has evolved since the early 1970s." Video analysis projects have included investigations of the methods video documentary narrators use to establish their authority. The journalism student who did this SAVAIL project used Connie Chung as an example. In the class, Johnson and his students spend time talking about the range of research possibilities opened up by this technology. One of his former students in the course comments: "The thing that was interesting about the course was that Dr. Johnson was learning while we were. What was wonderful about his teaching was being able to actually see the progression of someone's research, because he did take us, step by step through some of the breakthroughs that he made with this technological methodology. In that sense, it was very hands on, and that was refreshing."

Johnson says that his own mentor, Bogamil Witalis ("Goosh") Andrzejewski, late professor emeritus in the School of Oriental and African Studies at the University of London and the benefactor of valuable Somali research materials to the IUB Main Library and the Archives of Traditional Music, believed that computers are doing for us what the printing press did for our forbears. The gap between what one can record in books and the information one can store and analyze on a computer is now as great or greater than the gap between what one can remember and what one can print and store in a library. High-tech tools such as SAVAIL's multimedia delivery systems transform scholars' approaches to research problems by giving them new ways to analyze and understand the cultural artifacts of various times and places.

Back to the Table of Contents