The Mandaic magic scroll Zarazta d-Hibil Ziua: a possible scenario of literary evolution.

• Author: Notarius, Tania
• Position: Critical essay

This paper deals with different textual versions of the important Mandaic magic scroll called Zarazta d-Hibil Ziua. There are around thirteen manuscripts in different Western collections that attest to the text of this talisman. These manuscripts reveal great diversity. The general textual scope and literary structure of each manuscript are investigated. This allows an initial grouping of the texts into several text-types. The textual groupings are evaluated in light of various sorts of meta-textual data: the headings of the manuscripts, editorial comments, and colophons. As a result, a working hypothesis about the possible relationship among different texts-types and the possible literary evolution of the work is suggested.

1. INTRODUCTION

   The text of the Mandaic magic scroll called Zarazta d-Hibil Ziua (ZHZ) first became known to the scholarly community as part of the fifth volume of the renowned Mission scientifique en Perse, published by Jacques de Morgan in 1905. (1) The text of this long talisman was printed by de Morgan in Mandaic script, preceded by a brief introduction, but accompanied by no translation, transliteration, or commentary (Morg). The introduction explains the particular importance of the talisman, permitted to be worn by only the most devoted persons. (2) The great importance of the talisman is the reason for the remarkably large number of manuscripts of this composition now known to Western scholarship. As is apparent from many parallel quotations in the Mandaic Dictionary (MD), this talisman is attested in manuscript 44 of the Drower Collection (DC) in the Bodleian Library in Oxford (DC 44). (3) However, the Mandaic Dictionary (MD) seems to have made less use of other textual witnesses of the same work in the Drower collection, DC 13, DC 14, DC 15, and DC 52. (4) Recently, the pool of manuscripts of this composition was considerably enlarged with the launch of the Rbai Rafid Collection project at the Tel Aviv University. (5) Seven manuscripts from the Rbai Rafid Collection (RRC) provide additional witnesses to the text of ZHZ: RRC IK, RRC 1M, RRC 1S, RRC 2F, RRC 3X, RRC 3Y, and RRC 4A. Thus we find ourselves with a total of thirteen manuscripts that attest the text of all or parts of this talisman. (6)

   The manuscripts differ greatly: some are much shorter than the others, the order of the compositional parts may be different, and on occasion some individual sections are missing. In what follows, I shall first briefly characterize each manuscript, concentrating on its general textual scope and literary structure. This will allow the initial grouping of the texts into several text-types (part II). This initial grouping will then be evaluated in light of different sorts of meta-textual data: the headings of the manuscripts produced by the manuscript owners and by the scribes themselves, or implied by the text of some incantation; editorial comments; and colophons (part III). As a result, a working hypothesis about the possible relationship between different manuscripts and text-types will be suggested; potentially this hypothesis will elucidate several phenomena in the literary evolution of this text (part IV).

2. ZARAZTA D-HIBIL ZIUA: GENERAL LITERARY STRUCTURE

   ZHZ belongs to the genre of Mandaic magical literature attested in long eclectic cycles (or series) written on paper scrolls (called "compilations" in the present work). (7) A compilation compiles several literary "compositions," called in Mandaic z(a)razta 'talisman' or qmaha 'amulet'. (8) A composition is delimited by corresponding discursive and graphic devices: it begins with a marai msaba "blessed is my Lord" formula and an ABaGaDa (the Mandaic alphabetic list), and is commonly closed by ABaGaDa as well.

   Compositions are integral literary works that consist of several passages--units of different length and character. A passage is routinely marked by a bsumaihun "in the name of" formula and/or an eye symbol at the beginning and by the amin amin sala formula, a s_______a line, or just a simple line at the end.

   For the purposes of the present analysis, I suggest dividing the passages that make up a composition into three types: invocation, incantation, and rubric. An invocation will normally open a composition, pronounce the name of the client, and provide a general blessing in the name of Manda d-Hiia. An incantation, a magical spell in the proper sense of the word, commonly called baba (pl. babia) in Mandaic, has a complex literary structure, the details of which are beyond the scope of this paper; it may include demon expulsion formulas, a summoning of angels, as well as historiolae and other literary elements typical of magic literature. A "rubric" is a secondary meta-textual editorial insertion of an instructive character that explains how to write or use the incantation, provides headings, or explains the provenance of a particular incantation. (9) Colophons, which may provide very important meta-textual data about the manuscripts, will also be characterized as a kind of rubric for the needs of the present literary discussion (10) (see Figure 1).

   II.I General Literary Structure in the DC Manuscripts and the de Morgan Edition

   DC 44 presents the longest version of the compilation. (11) It is a 2140-line scroll, and contains a considerable amount of material that has no parallels in other versions (particularly different kinds of rubrics; see the analysis below). It has two colophons. The number of compositions, separated by the regular discourse devices, is four, but as will be shown below, DC 44, comparable to other long textual witnesses, contains the text of five compositions, while the separation between two of the compositions is omitted. (12)

   Both Morg and DC 15 contain approximately the same range of text as DC 44, but are somewhat shorter. (13) Moreover, Morg and DC 15 represent a perfect literary match; all the passages of these two versions are parallel except one short passage and some headings in DC 15 that are missing from Morg; Morg has two colophons, while DC 15 does not have any. Morg and DC 15 commonly agree textually against the other versions, but there are also notable textual differences between them, and Morg regularly represents a slightly longer version. The number of the compositions in Morg and DC 15 is five. (14)

   Three other manuscripts from the Drower Collection contain a part of the whole literary work. DC 14 and DC 52 correspond to the text of the first composition of DC 44, Morg, and DC 15, itself consisting of seven passages. (15) Though there are no compositional differences between DC 14 and DC 52, the text of DC 14 is usually slightly longer. Neither manuscript has any colophon.

   Another partial attestation of the compilation is represented by DC 13, (16) which starts where DC 14 and DC 52 end, namely with the second composition of the entire compilation (as represented in DC 44, Morg, and DC 15), but the beginning of the manuscript is missing and the text starts with a few remaining lines of the opening invocation. It is not possible to determine whether the first composition was ever there or has been lost. But interestingly DC 44 itself is comprised of two glued manuscripts and the second is attached exactly at the place where DC 14 and DC 52 end and DC 13 starts, namely between the first and the second compositions. (17)

   Altogether, DC 13 and the corresponding part of DC 44 demonstrate some other striking similarities: there are several passages in DC 44 that are missing from Morg and DC 15, but are present in DC 13 (particularly rubrics, although there is still material that is unique to DC 44); moreover DC 44 and DC 13 usually agree textually, differing from Morg and DC 15. However, DC 13 reveals some unique literary characteristics. In DC 13 the same four compositions which comprise part of DC 44, Morg, and DC 15 are arranged in a different order than in those long compilations (cf. also the discussion below). It does not have a colophon, but is completed with a date. (18)

   These remarks allow a preliminary classification of the manuscripts. ZHZ as transmitted in its longest variants is a compilation of five compositions. The merger of five compositions and also their order are not strictly canonical: some compositions (e.g., the first) could exist independently; some compilations could vary the order of the constituents. There are good grounds for dividing the existing manuscripts into four text-types (see Table 1). Two text-types present a shorter version: text-type (A) includes DC 14 and DC 52 and contains one complete well-structured literary composition also present in the long compilations of Morg, DC 15, and DC 44 as its first composition, but absent from another shorter compilation, DC 13. Text-type (B) includes DC 13, itself a literary compilation, which is made up of four compositions absent from text-type A but present in long compilations, although in a different order. (Notice also that the second manuscript of DC 44 might be an indirect confirmation of this text-type, if the lack of the first composition is regarded as significant.)

   Two text-types present long compilations: text-type (C) includes Morg and DC 15 and represents a shorter variant of the final long compilation; text-type (D), DC 44, is the longest variant of the final compilation full of editorial rubrics:

   II.2. General Literary Structure as Evidenced in the Rbai Rafid Collection

   The textual data from RRC strikingly conform to the conclusions based on the Morg and DC manuscripts, but also add some new details. To start with, text-type A is represented by several other manuscripts. RRC 1M contains only one composition, which corresponds to the very first composition of the long compilation. (19) In general, the text of RRC 1M is very close to that of DC 14 and DC 52, but there are some interesting changes in rubrics and a caption is added to drawing 1. (20) There is no colophon.

   RRC 3X also...