GRE aims to bring knowledge of Roman garden archaeology out of local archaeological journals and print books into a free open access resource in a consistent format that provides scholars, students, and professionals global access to evidence of all types for ancient garden culture.

The entries range from sites that have been excavated using contemporary techniques to early sites where gardens are suspected but not yet proven. The range of evidence assembled includes up-to-date descriptions, plans, stratigraphic sections, bibliography, and photographs of gardens known archaeologically.

For the beta launch of this project, GRE is publishing a selection of the entries assembled by Wilhelmina Jashemski, Kathryn Gleason, Kim J. Hartswick, Amina-Aïcha Malek, and a team of area editors working between 1988-2010. While the work of publishing these assembled gardens will continue over the course of 2021-2022, the new website has also been engineered in such a way as to allow the GRE team and its contributors to update older entries with new information and bibliography and to add new gardens as they are discovered. In some cases, older entries by distinguished scholars who are now deceased will be updated and the original entry will be archived. We intend for both editions of the garden entry to be visible to interested readers.

In the next phase of development, GRE is planning to add a glossary and thematic entries on topics such as garden representations, inscriptions, archaeobotanical findings, and water features. In assembling all the known garden evidence into a single online corpus, GRE seeks to expand the scope of research on Roman gardens, as well as to stimulate further field research.

The data structure and website was designed by the Library of the Institute for the Study of the Ancient World, in consultation with Sebastian Heath, who suggested the basic strategy. Christian Casey, the CLIR postdoctoral fellow in the ISAW Library, elaborated and realized this strategic approach with significant contributions and a GIS map of the garden sites by Keith Jenkins of the Cornell University Library. Substantial support over the years was also provided by: Cornell University, Archéologie & Philologie d’Orient et d’Occident, Centre national de la recherche scientifique, the Cambridge University Press, and Dumbarton Oaks.

GRE invites all scholars and the general public to explore the new beta site and provide us with constructive feedback as to how we may improve the site over the next year. New editorial teams are currently reaching out to archaeologists and scholars to solicit new information and newly discovered or described gardens. Interested in contributing?

Data structure

The basic unit of the GRE is the “garden.” This is to say that we have defined a template for the “garden” as an object, which is to data structures what genre is to literature. As such, the “garden” is an abstraction made up of a collection of stereotyped types and forms of data, or, one might say that it is a specific aggregation of metadata fields, like a bibliographic record. These fields are in turn themselves abstractions of the messy underlying reality of the excavated and attested gardens of the Roman Empire. So, for example, each garden has a unique ID, a title, a location, a description, maps, plans, bibliography, a date, an author, etc. This is, in essence, no different than how any corpus or collection is assembled (think of the metadata fields associated with catalogs of vases, coins, inscriptions, or manuscripts). Occasionally, we needed to modify or elaborate our basic template, as was the case in the gardens of Pompeii, where the geography is more fully articulated than in many other regions of the Roman world, e,g., with regiones and houses.

The best way to think of the GRE, in other words, is as a corpus of “gardens,” such that each identified garden of the Roman Empire has a unique record or file associated with it. Thus, if a house in Pompeii or a tetrarchic palace in Thessaloniki has two or three gardens, there will be two or three records for that house or palace, one for each garden, as opposed to one record for the house or palace that discusses all of the gardens together. The advantage of this approach is that one gains precision over gardens as an object of study. So, for example, each record has a stable, unique URL, which means that one can cite a specific garden and link to it with confidence. The price of this precision, however, is the loss of a synoptic or synthetic approach to how such gardens may have been experienced in relationship to each other (e.g., in a villa, a town, a region, a province, etc.). Like all corpora, synthesis is largely left to the reader and scholar.

Open linked data

GRE adopts open linked data approaches whenever it can. All of the data associated with the GRE site is free and openly available to the public in our github repository.



GRE has researched copyright requirements and restrictions for all of the materials displayed on this site to the best of its ability and has either secured permission or determined that the image or text is in the public domain or may be served under the doctrine of fair use.

If you believe that you are the copyright owner of any material displayed here, for which you have not granted permission (or is not covered by a copyright exception under applicable copyright laws), please see our take-down policy.


The content on the GRE site is governed by a CC BY-NC-SA 4.0 Attribution-NonCommercial-ShareAlike 4.0 International license.

History of the project

Originally modeled on Gardens of Pompeii (1979-1993), with its book of essays and catalogue of archaeological sites, Wilhelmina Jashemski planned Gardens of the Roman Empire as a far more ambitious project. It has engaged an interdisciplinary team of nearly 200 scholars, students, volunteers, and specialists to assemble a vast body of data into this digital project, which complements, but is independent of, a print volume of essays published by Cambridge University Press, Gardens of the Roman Empire (2018). Ultimately, this website is designed to publish the assembled catalogue of material in the format in which Jashemski first envisioned it, while using contemporary digital tools to make the data readily discoverable, navigable and citable. It takes the project forward into the future with essential search tools, links, and maps. The result is a website that realizes her original vision while transforming the project into a fully independent digital forum for research on Roman gardens and designed landscapes.