A Mid-Republican House from Gabii

This first major publication from the Gabii Project is an e-publication, but it is not a standard-format e-book, as it includes many interactive elements. In addition to the editors, other contributors are Tyler Johnson, Antonio F. Ferrandes, Laura Banducci, Francesca Alhaique, Laura Motta, Shannon Ness, Jason Farr, Sam Lash, and Matt Naglak.

The open-access home page includes an introductory paragraph to the project, the site, and this mid-Republican house, known as Area B or Tincu House, excavated between 2009 and 2012. It includes a static image of the three-dimensional interactive reconstruction model of the house.

The home page provides links to an additional open-access and illustrated introductory page, which outlines the organization of the main report (not freely available), explains Gabii’s importance for the history of Roman urbanism, gives a history of the project, and describes the excavation dating and recording methods, the latter moving from paper to digital recording during the life of the project and including spatial mapping of the site and finds. This open-access section finishes with the rationale for the three-dimensional model and the bibliography for the section. Missing are plans showing the location of five excavated city blocks and of this particular house within them. Also, while the icons used for the three-dimensional model are introduced here, other features are described on the “Help” page, and the rationale for the separation is unclear.

The main report and the three-dimensional module are mutually interactive and both available only after purchase. The main report is organized as “three layers” on a continuous page, with links to the three-dimensional model. The first layer tells the “story of the house,” and the second layer, “More,” provides information on how this story was reached. These two layers, written for both general readers and scholars, summarize the house’s construction and occupation history as interpreted through the excavation results, the different phases of this construction and its contribution to recent studies on the later (third-century B.C.E.) date for the atrium-style house, and the results of artifact and ecofact analyses as chronological indicators. The third layer, “Details,” provides information on the stratigraphic contexts and specialist reports.

In the first layer of the narrative, the reader is introduced to the history of the house through its excavation. However, the authors have not explained, at this stage or indeed later in the narrative, how to use the interactive model to follow the living activities described here. In the second layer, the phases of change and adaptation are summarized, although these are not signposted until well into the layer, where it becomes easier to follow the narrative of the house phases on the three-dimensional model. In this second layer, the first partial plan of the site is provided. However, the location of the excavation area, or this part of it, is not clearly shown on the only map in the publication (the results of the geophysical survey in 2008, fig. 4 on the introductory page). More information on the location of the house and more explanation of the interactivity of the model would have been useful for reading these first two layers. For example, it would have been helpful if the “?” on the model led to the “Help” page. However, there you will not learn that the “stratigraphic units” and “3D Models” buttons do not toggle but can be all turned on at the same time, irrespective of any “Reconstruction” toggle, and that the “3D Models” buttons actually show excavated structural remains.

The third layer, “Details,” is in the form of a traditional excavation report with detailed stratigraphic analyses that provide excellent links between the stratigraphic contexts and information about these contexts on the three-dimensional model. The specialist contributions are short reports on the building materials and the archaeobotanical remains and more detailed reports and analyses of the 3,891 zooarchaeological remains and the 14,000 ceramic remains. The zooarchaeological analyses characterize the faunal remains and assess their evidence for food sources within each phase. The ceramic analyses (with drawings of diagnostic types) chart the ceramic assemblages phase by phase and assess their evidence for the house’s chronology, along with any evidence for vessel use and discard patterns. These two long reports are followed by two further short reports, on 46 notable finds (especially cloth-working implements and apparel) and on coins. Both of the latter reports link to entries in the special finds database (with some photographs). Clearer illustrations of the coins appear at the end of this second report, on a very long narrative page comprising these three layers, the final section of which is an “Apologia” highlighting problems with the data and their interpretative limitations and any inconsistences resulting from changing recording methods.

Each stratigraphic context mentioned in this third layer links to the three-dimensional model to show its location, a photograph, and related find types. However, the finds described in the narrative do not always concur with the display in the model. For example, according to Alhaique’s report, phase B-0 stratigraphic context SU 1416 had six bone specimens of a “medium mammal,” but there is no “fauna” symbol for these on the three-dimensional model for this context. For the ceramics, a more detailed system of “activities” is tabulated but not mapped on the model.

The publication includes an openly available database with three linked categories (stratigraphic units, special finds, and spot dates), each made up of two levels. They can be searched or browsed either directly or through “collection access.” However, the latter, though labeled “openly available,” leads back to the three category options after a “no results” window and seems redundant, given that all pathways in the database are openly available. The entries for “spot dates” provide very brief records of some of the ceramic remains used to provide dates for particular contexts.

The interactive reconstruction three-dimensional model provides easy access to most of the visual evidence for the excavation, with other evidence and interpretations represented by symbols. As such, it is a useful, and indeed fascinating, tool to introduce scholars, students, and the public to the excavation of this house. An intriguing aspect here is the use of symbols for the various phases of the house and for the different types of evidence. The former are used on the three-dimensional model to indicate activities that describe the life cycle of the house from initial construction to abandonment. Each activity is represented by the respective icon in the relevant display box overlaid on the model. The keys to these icons are on the model and are listed in figure 1 on the introductory page (where, however, they appear in alphabetical order rather than in a taphonomically logical sequence, and with an icon for “dumping” that is different from that used in the three-dimensional model). The different types of artifacts and ecofacts are also represented on the model by five different symbols, with a key in figure 2 on the introductory page. These provide a quick reference on the three-dimensional model to the finds from each context. However, as noted above, these do not always agree with the narrative. Also, while these symbols are clearly explained, other aspects of the functionality of the model are not, and no compass orientation is indicated. As the main source of illustration for the text, this model indeed demands interaction and is a novel, useful, and potentially important approach to the presentation of a site and its excavation, but it needs some refining.

The model is accessed by sliding it across the top of the narrative text. However, using browsers accessible through my university server, downloading can take more than 10 minutes. This would not be a major problem except that the model needs to be downloaded each time the reader navigates to another part of the website, using the dropdown menu and table of contents or the side navigation arrows. It is also difficult to see the downloading process taking place.

Like the introduction, the main report is in a fairly traditional format made up of one very long page, with embedded images that can be enlarged. Links within the narrative do not always take one to a clearly identified place, and there is no easy way to return to one’s place to continue reading. Much scrolling backward and forward is needed as the reader follows links within the narrative. Using the “Full Contents” or “Database” buttons in the dropdown menu requires redownloading the three-dimensional model. The narrative might have been more effectively formatted as short interlinked pages, with a roadmap that is more easily navigable while the three-dimensional model remains downloaded. It also seems impossible to return to the narrative from the database using the dropdown menu, so that having carried out several searches on the database, the reader needs to go back through numerous windows or log in again.

The website “Help” section is designed to assist with accessing the three-dimensional model and with its interoperability. It is not so useful for understanding how its various components work or what the database comprises. For example, the database currently includes only 53 of the 14,000 ceramic finds and none of the bioarchaeological finds collected from Area B. For this reason, presumably, the “direct links to entries from the project’s online database” do not generally work for most artifact categories, although they also do not seem to work for all stratigraphic units. The “Help” section for the database leads to a generic “Help” page for the University of Michigan Library Image Collections. A larger glossary of abbreviations, terminology, and conventions would also have been useful.

In summary, the content of this publication consists, in large part, of an appropriately rigorous and detailed excavation report with a clear and novel introduction to the site. The main lacunae here are the lack of adequate site plans and incomplete record of finds. Other inconsistencies with data recording and presentation may result from changing and varied approaches to digital recording by different contributors. The concept of the publication format indeed provides an innovative approach to presenting the complexity of an archaeological report, but there is still room for improvement. Many of the shortcomings in the interoperability outlined above are the kinds of things experienced with a new approach that aims to address a wide audience. Unlike in a printed publication, though, these can be fixed, modified, and improved after initial publication. The introduction indicates that this publication is a “beta” version and that as new areas are excavated they will be made available in the same format. One assumes that there will be ongoing updates and improvements, some covered by the original purchase price and some requiring an additional purchase.

Penelope M. Allison
School of Archaeology and Ancient History
University of Leicester
pma9@le.ac.uk