The “Ancient Lives” project is an international collaboration between multiple departments and institutions: The Adler Planetarium in Chicago, Illinois, the Departments of Classics and Astrophysics at the University of Oxford, and the Departments of Classical and Near Eastern Studies, Minnesota Supercomputing Institute, and Physics and Astronomy at the University of Minnesota. Through the Ancient Lives website, the project continues to amass an immense database of crowd-sourced transcribed texts in order to assist in the classification, cataloguing, and identification of the tens of thousands of papyrus fragments housed in the Sackler Library of the Ashmolean Museum. Consequently, a digital database of both previously edited and unedited Greek papyri texts exists for the first time. In addition to the computerized assistance in cataloguing and identifying documents and known literary texts, the aim of this project is to build a computational interface to assist in the critical editing of Greek texts. Two of the principal goals are: (1) to refine and implement a consensus algorithm, adapted from those used in the study of DNA sequencing, that collates multiple transcriptions, producing a variety of meaningful statistical data and digitally searchable transcriptions that will be accessible to scholars through a unique graphical user interface (GUI); (2) based on the extant strings of Greek characters, to implement algorithmic meaning extraction to assist in the contextualization and identification of unknown literary works through an automated projection of possible linguistic/word scenarios. We are also aiming at computationally repairing gaps/holes in papyri through an automated projection of characters that are not only dimensionally suitable but also contextually sound through linguistic parallels elicited from the searchable online databases of Greek texts. The purpose of this paper is to report initial project results and to outline in greater detail our methodology for building this computational interface. In sum, we are merging human and machine intelligence – automated algorithmic methods – to increase the accuracy with which Greek texts are edited. We are transforming image data from Oxyrhynchus papyri into meaningful information that scholars can use – information that once took generations to produce.
