Due to the role that copying texts played in the process of Mesopotamian scribal education, cuneiform literary compositions are often preserved in multiple (albeit usually fragmentary) copies. Since these copies tend, by and large, to align closely with each other in language, style, and content, the preferred format in the modern scholarly community for editing and publishing such a composition is as a score transliteration, or partitur, in which each "line" of the presumed original or canonical version of the text is published along with all of its individual witnesses. This format not only aids in the reconstruction of the composition, but also highlights those areas where the witnesses disagree, which is the first step in understanding how and why they differ.

While this method of publishing texts has served the Assyriological community very well, there are a few texts (or portions of texts) for which it is less useful, namely those for which the witnesses do not preserve highly consistent versions. When the contents, presence, or order of lines varies from manuscript to manuscript, the composite must necessarily choose or create one "standard" version, against which to compare the others. In these cases, the composite can obscure the very differences that it tries to illuminate, because it overlays the original organization of the witnesses with its own. This can hinder or mislead scholars trying to understand the text's meaning.

One text to which this problem applies is Gilgmesh, Enkidu, and the Netherworld (GEN), the story of Enkidu's descent and imprisonment in the kur, the Mesopotamian underworld. The witnesses of GEN do generally agree in text, line order, and line attestation, almost to the very end; but the composition's concluding segment, Enkidu's "Catalog of the Dead," instead shows substantial variation in line choice, order, and content. Not only do the ghosts recorded by the witnesses differ significantly, but three witnesses actually preserve extended conclusions, completely different continuations to the larger narrative, with different forms and implications for our interpretation of the story. These variations cannot easily be accommodated by the composite model, that I have outlined above, despite numerous scholarly attempts to do so.

Enkidu's catalog, with its detailed image of the afterlife, has potentially great implications for our understanding of Mesopotamian eschatology, if we can understand it. Further, the very existence of these differences suggests that a better understanding of the forces at work in this text can inform us about ancient scribal practice. Therefore, rather than trying to analyze all of the witnesses of the catalog as "one" text, I propose to compare them to each other and see if we can observe any patterns in their content that might let us organize them into separate recensions. Then, I will look at each of those individual recensions and see what they communicate on their own, before comparing the recensions to each other and, finally, to the other literary texts with which GEN circulated in Mesopotamian society. In this way, I hope to better illuminate the meaning of this important but challenging text, as well as the scribal tradition that created it.