This dissertation proposes a poetics and semiotics of the Bavli (Babylonian Talmud)--how the Bavli, through a complex network of linguistic signs, acts on its implied reader's attempt to find meaning in the text. In doing so, I advance a new understanding of how the Bavli was composed, namely as a book written by its own readers in the act of transmission. In the latter half of the twentieth century, Bavli scholarship focused on the role of the Stam (the collective term for those people responsible for the anonymous voice of the Bavli) in the construction of individual Bavli passages (sugyot). Stam theory details how sugyot were crafted out of pre-existing sources and how the Stam works to control those sources in the service of a particular worldview. This dissertation locates a different force at work in the construction of the Bavli as a single unified book, an authorship that is above and against the work of the Stam--a Superstam.

By examining the effect of the Bavli's use of rare and ambiguous terminology, I expand the unit of inquiry from the individual Bavli passage (sugya) to the Bavli in its entirety. I argue that, for the Bavli's implied reader, meaning is not found in the work of the Stam. While the Stam conveys meaning for a local reader, the global reader I explore does not artificially divide the Bavli into its constituent parts. For this reader, the Superstam acts to subvert the controlling work of the Stam through the placement of key words throughout the book. These words, when rare or ambiguous, direct the reader to other sugyot in which they appear. These sugyot, when read simultaneously, work to convey, for the reader, an expression of ambivalence on the part of the Superstam toward those moments of Stammaitic certainty. I conclude that Super-stammaitic activity itself is the result and product of readers who are trained how to read by the Bavli's own expectations. In this way, the Bavli is a text authored by its own readers who, in transmitting the text, become writers again and again.