Description of Pleiades (2019)

Like any good gazetteer, Pleiades (https://pleiades.stoa.org) is an organized spatial reference work for use in research, publication, and teaching. Jointly operated by the Institute for the Study of the Ancient World at New York University and by the Ancient World Mapping Center at the University of North Carolina at Chapel Hill, it constitutes the most accurate and comprehensive geographic dataset for the ancient Mediterranean world, identifying and describing over 36,865 records for places and spaces that include information about 32,784 associated toponyms and 39,985 associated spatial location geometries. It serves scholars and students alike as a ready reference and a guide to associated bibliography for places encountered in primary sources and secondary literature. As a database fronted by a web application, Pleiades transcends the familiar structure of print gazetteers (alphabetical lists of placenames), to provide its users with multiple modes of discovery: thematic browsing, hyperlinks, and a search engine. Pleiades publishes not just for individual human users, but also for search engines and for the burgeoning array of computational research and visualization tools that support work in fields like computational linguistics, digital text encoding, computational text analysis, natural language processing, named entity recognition, and Geographic Information Systems (GIS).

The present project has its origins in the Classical Atlas Project (CAP), a 12-year effort that produced the Barrington Atlas of the Greek and Roman World (R. Talbert, ed., Princeton) in the year 2000 (see retrospective in: Talbert, Richard J. A. “Barrington Atlas of the Greek and Roman World: The Cartographic Fundamentals in Retrospect.” Cartographic Perspectives, no. 46 (January 9, 2003): 4–27, 72–76: https://doi.org/10/gnrdkq). In acknowledgement of this heritage, Pleiades is named after the daughters of Atlas in Greek mythology. CAP, which had received funding from the National Endowment for the Humanities and other sources, redressed a critical gap in research tools for the Classics that had been identified in the early 1980s: historical cartography was so neglected as a subdiscipline of ancient studies that there existed no comprehensive, up-to-date scholarly reference atlas for Greek and Roman civilization (Roger S Bagnall, ed. Research Tools for the Classics: The Report of the American Philological Association’s Ad Hoc Committee on Basic Research Tools. Chico, Calif.: Scholars Press. 1980).

An NEH Preservation and Access Research and Development grant (PA-51873-06, PI: Talbert) launched Pleiades in Chapel Hill in 2006 (final reports from all referenced grants available at: https://pleiades.stoa.org/docs/reports-proposals). It culminated in 2008 having funded conceptual and experimental work to develop the data model now used, proving our hypothesis that off-the-shelf GIS data models were inadequate for the task and that a custom database was required (Sean Gillies. “What’s an Un-GIS?” And a Laser in My Shoe. November 1, 2010).

From 2008-2010, a JISC/NEH Transatlantic Digitization Collaboration Grant (PX-50003-08; PI Bagnall) with a team at King's College, London used the Pleiades prototype to explore modes of cross-project geographic linking. Its findings, adapted to the more suitable Linked Open Data (LOD) model, formed the conceptual foundation for the Pelagios Project, now known as the Pelagios Network.

From 2010 to 2014, Pleiades moved into its implementation phase, supported by a combination of institutional funds and an NEH Preservation and Access Humanities Collections and Reference Resources grant (PW-50557-10, PI: Elliott). This grant facilitated the complete digitization of the Barrington Atlas materials, incorporating work already done on digitizing the atlas' coordinates with separate funds by Prof. Michael McCormick and his students for Harvard's Digital Atlas of Roman and Medieval Civilization.

Pleiades content is updated collaboratively by volunteers around the world who work under the supervision of a volunteer Editorial College (see https://pleiades.stoa.org/credits), filling a gap in the scholarly communications fabric for the typically small, highly technical notes and bits of geographic data that are otherwise unlikely to see the light of day in conventional, narrative-focused publications. Through Pleiades these volunteers also provide authority control for ancient places. Just as the Library of Congress issues “authority files” to help improve discoverability in library catalogs by standardizing subject terms, names, and titles, Pleiades assigns a stable identifying number to each place and to each record about a place (a “Place ID” or PID). The use of numbers, rather than “standard” or “preferred” placenames, permits Pleiades to itemize places that never had a formal name in antiquity (e.g., a particular courtyard, road segment, or bridge) or whose ancient name has been lost from our sources over the intervening centuries. The PIDs are incorporated into the page address in a uniform way that is easy to remember and to encode, thereby forming for each place a Uniform Resource Identifier (URI), a standard mechanism for identifying documents and objects on the world-wide web (more specifically, Pleiades uses HTTP URIs: Leo Sauermann and Richard Cyganiak, eds. Cool URIs for the Semantic Web. W3C Interest Group Note. 2008. World-Wide Web Consortium (W3C). http://www.w3.org/TR/cooluris/). So, for example, Pleiades assigns the PID 29573 to an ancient region known as “Gedrosia”. The URI for the Pleiades page for this Gedrosia is https://pleiades.stoa.org/places/29573. This practice facilitates citation and linking for individual users and establishes a simple, permanent reference number for each place that can be used anywhere such references are valuable.

In fact, by establishing an authoritative gazetteer and associated open information services, Pleiades helped start a revolution: a constellation of online publications treating the histories, languages, texts, and artifacts of antiquity that make use of Pleiades and other similarly constituted digital gazetteers to contextualize their holdings interoperably across the Web. For example, operating under the aegis of the UK-based Pelagios Network, over 40 teams from 8 different countries have incorporated Pleiades URIs into their datasets, thereby identifying places mentioned in texts and the origins and findspots of artifacts. Several of these projects use Pleiades data to provide their users with links, dynamic maps, and other services. All of them publish the pairwise matches between their own records and Pleiades place records (over a million so far) as open data for other parties to use and other systems to harvest. Pleiades uses this data too, creating lists of related content on every place page by comparing the URI for the place in question against the data published by Pelagios partners.