Variation within Limits: An Evolutionary Approach to the Structure and Dynamics of the Multiform

One¹ of Albert Lord’s more surprising discoveries about living oral traditions was that no two oral performances of the South Slavic epic that he collected were precisely the same (1953).² Scholars before Milman Parry and Lord seem to have assumed that any long performance must have been of a text that had been memorized verbatim. It took the Theory of Oral Composition as originally developed by Parry and Lord and elaborated and refined in the decades since³ to explain why in many traditions no two performances are the same at the level of the word and sentence even though the audiences and the performers state that the “same” story is being performed (cf. Foley 2002:12–20). Recent work has shown how the influence of the “performance arena” and the differing skill sets and tendencies of individual singers contribute to the significant variation in the exact words used in any given performance of a specific song (Foley 1995:8–11). But although individual performances vary, they do not vary infinitely, for if they did, there would be no tradition. Oral Theory has used the term “multiform” to describe verbal or textual entities that display this “variation within limits” (Foley 1991:6–8, 1998:149).

Although use of “multiform” both as adjective and noun is widespread in scholarship, it remains difficult to find an agreed-upon definition. Lauri Honko’s description of multiforms as “repeatable and artistic expressions of variable length which are constitutive for narration and function as generic markers” (Honko 1998:100–05; cf. Foley 1995:102) is probably as close to a consensus as one can find, but the problem that Lord noted in The Singer of Tales remains: “unlike the oral poet, we are not accustomed to thinking in terms of fluidity. We find it difficult to grasp anything that is multiform. It seems to us necessary to construct an ideal text or seek an original, and we remain dissatisfied with an ever-changing phenomenon” (1960:100). Despite the efforts of many scholars to explain the phenomenon of multiformity (perhaps epitomized by Foley’s How to Read an Oral Poem [2002]), it remains difficult to think and talk about the multiform without collapsing it to a single, textual entity.⁴ Scholars do not even agree completely on the size of multiforms: Parry and Lord’s original approach limited varying formulas to circumstances with identical metrical conditions (although Lord also discussed “themes” that were groupings of ideas [1960:68]), but more recent work has identified much larger multiforms that extend well beyond sentence length (Honko 1998:102–14), and the scholarship seems to be moving in the direction of identifying as multiforms even complete songs (Foley 1998).⁵ The work of Gregory Nagy in developing an “evolutionary” model has been influential in this area, in particular his view that the multiform should be understood in relative rather than absolute terms, so that any particular composition could be more or less multiform “along a graded continuum” (Nagy 1996Nagy 2001:109–10). That multiforms vary at different levels from the micro to the macro is borne out by studies such as Honko’s of Siri epic or Foley’s work with the variants collected by Parry and Lord (Foley 1998). Nevertheless, significant disagreements among researchers remain, both in theoretical terms and, more specifically, about the relative multiformity—and attendant orality—of particular works (for example, the Homeric Iliad versus the Cypria [Finkelberg 2000; Nagy 2001]).

This state of difficulty is not confined to purely oral, or even primarily oral, traditions, either. In medieval studies Paul Zumthor’s discussion of mouvance—most succinctly expressed by Bernard Cerquiglini’s assertion that (1989:111) “l’écriture médiévale ne produit pas de variantes, elle est variance” (“medieval writing does not produce variants, it is variance”)—brought about significant changes in both theoretical approaches and editorial practices (Zumthor 1972Zumthor 1987).⁶ Although Cerquiglini’s position is seen as extreme in contemporary medieval studies, the variable nature of texts—even when they are not considered to be particularly close to an...