In this article, I present a rotulus from the Genizah that contains three she’iltot from the beginning of the She’iltot book by R. Ahai. The rotulus is in the handwriting of R. Ephraim Ben Shemarya, the head of the Palestinian community in Fustat during the first half of the eleventh century. The order of the she’iltot in this rotulus diverges from their order in the common edition of the book. I propose that the she’iltot were rearranged thus because this copy was made by R. Ephraim for his sermon at the Ben-Ezra synagogue of the Palestinian community, on the shabbat when the first seder according to the triennial reading cycle of the Torah is read. This seder comprises only the creation story (1:1–2:3), and Ephraim’s copy reorganizes these three she’iltot to fit this seder precisely, using other verses as a source for the commandments, instead of those used in the She’iltot book. It is well-known that in the Ben-Ezra synagogue the reading of the Torah was practiced as a ‘double reading’: every shabbat the parasha (according to the annual cycle) was read from personal codices, and then the seder (according to the triennial cycle) was read from a Torah scroll. Evidence of the double reading practice dates as far as the beginning of the 13th century. In the article I review the various testimonies of the Torah reading practice in the Ben-Ezra synagogue, and the different opinions regarding its origin. If my theory is correct, R. Ephraim’s rotulus is the earliest evidence of reading the Torah according to the triennial cycle in the Ben-Ezra synagogue. Moreover, we can learn from this that during this period the sermon on shabbat was concerned with the seder that was read from the Torah scroll, and not with the Babylonian parasha that was read from codices. This, although the liturgy of the shabbat service was probably accompanying the parasha and not the seder. The assumption that R. Ephraim’s rotulus presents a Palestinian adaptation of the Babylonian She’iltot, which rearranges it according to the triennial reading cycle, is supported by two additional evidences, one external and one internal. Besides R. Ephraim’s rotulus, I present in the article another manuscript of the She’iltot from the Genizah where every she’ilta is titled with the corresponding seder instead of the corresponding parasha as is customary in this book. Furthermore, many of the she’iltot in this manuscript are not associated with the parasha to which they belong in the common book, but with a seder which is located in a different parasha. That is, this manuscript did not just associate the she’iltot with the sedarim, but also rearranged them in a different order. Apparently, R. Ephraim's rotulus also belongs to the same Palestinian tradition of this book. This argument is reinforced by internal evidence: The structure of the she’ilta that deals with shabbat in R. Ephraim’s copy is different from all other manuscripts of the book, but similar to the presentation of this she’ilta in Midrash Tanchuma. In other words, Ephraim copied the she’iltot from a manuscript that belonged to the Palestinian textual tradition of the book, the same tradition which is presented in the Tanchuma. Another evidence in the support of the Palestinian origin of R. Ephraim’s copy is its orthography, which has distinctly Palestinian characteristics. This article thus reveals the Palestinian textual tradition of the She’iltot book to which the rotulus of R. Ephraim Ben Shemarya belongs, as does the other manuscript described above. The she’iltot fragments which are found in Midrash Tanchuma emerge from the same tradition, and apparently this is also the source for the she’iltot that were embedded in other books from the Palestinian tradition as Sefer veHizhir.
Read full abstract