I grew up during the “space race.” Some of my best memories during grade school involved watching CBS News for the latest on a Gemini mission. This science playing out on the screen helped to shape my own interest in science, which I now practice in archaeology of the Levant.

Since my interest is in the Late Antique (Byzantine) Period, to which I have been contributing at the excavation at Hippos/Sussita, I find myself captivated again by the capabilities of Gemini as I go about my research. This time, the action is not in space but in the cloud, specifically through Google’s Gemini 2.5 Pro AI tool.

My post-excavation research centers on teasing out the contribution of theology to the diverse forms of the early church basilicas. Put another way, what is the relationship between spirit and stone, between buildings and belief?

Since some have misread my initial forays as advocating a direct link between theology and an architectural feature (“this church is monophysite because it has x”), I am working at framing my hypothesis in a more nuanced way. Put simply, I am convinced that “the Early Byzantine basilica was a dynamic entity, a place where stone and spirit met. The forms it took were shaped by many hands and minds, but the theological convictions of the age were undeniably a vital force in breathing meaning and sacred purpose into its diverse architectural expressions” (from an early draft of a forthcoming paper).

While I continue my work, I have used AI to create an informal expression of the direction my research is taking me. The product is an audio conversion running about six and a half minutes. I produced it using Google Gemini AI. It is conversational in form and accessible by a general audience. Take a listen…

Direct link.

I have delayed research on this project because of other duties in the spring of 2024. I was teaching, helping with a church vacancy, and preparing to teach in Ethiopia in the fall while working on a paper for ASOR. All was interrupted when I was diagnosed with cancer in August. So far, the treatment is going well. As I will be kept close to home for treatment over the next year, I will be returning to this project.

I do have more data to add to the charts on the project website. But I have also been looking for a better analytical approach. During this year’s ASOR meeting, I heard a paper applying correspondence analysis to identify relationships in a complex table.

Correspondence analysis is a multivariate statistical technique. It applies to categorical rather than continuous data. In a similar manner to principal component analysis, it provides a means of displaying or summarizing a set of data in two-dimensional graphical form. Its aim is to display in a biplot any structure hidden in the multivariate setting of the data table. So says Wikipedia.

I have much to learn about this method. But, if it can help me tease out structural relationships among the categories in my table, that may be a step toward uncovering the rationale behind the structural variety among churches of the late Antique period in the area surrounding Hippos Palaestinēs.

Our friends at the Hippos (Sussita) Excavations Project have made available schematics of the churches excavated at Hippos to date. Our thanks for these and permission to use them. They allow side-by-side comparison of the churches and can be the basis for comparative research.

As with any schematic drawings, details are adjusted for the sake of uniformity and in view of common usages. So, published drawings must always be consulted on minor points. For example, the schematic for the Northeast Church shows five columns for both aisles. What we found in the excavation was a highly disturbed site with six columns (one missing) on the north and four on the south. The posited fifth column on the south accommodates a standard spanning distance between the columns; but there is no evidence for the western-most column in the field (at least, not yet).

As to the “Archaeology of Liturgy” project, work has slowed as I am teaching this semester and have taken on some other temporary professional duties. The “difficulties” in Israel are also a factor. I am looking forward to resuming work more fully in the summer of 2024.

With the addition of church sites from the Syrian Hauran and from Southern Phoenicia, the dataset for the project is now complete. When I went to the Albright in the spring of 2023, my goal was to develop a regional dataset of basilical churches through the comparative study of which I hoped to understand more fully the diversity of floor plans and annexes in the churches at Hippos. By the time I left Jerusalem in May, I had an initial list of 104 churches, mostly basilicas, which would form the dataset for my comparative analysis. I had also gathered preliminary and/or final excavations reports on each and had begun to summarize them in a common format.

Subsequently, I have worked through all the materials (translating as needed) and have summarized them on a map with links to web pages for each church. Of the initial group, 97 different structures are in the dataset all within 100 km of Hippos. Phase 1 of the project is complete.

After a literature review of theories about the development of the forms of Late Antique churches, I will collate all this data into charts and from those charts seek to develop an analytical protocol for understanding the particular layout of a church within the regional context defined by the dataset.

It had been my intent to do the rest of this work at the Albright in Jerusalem starting next week. However, the security situation is such that I have cancelled the trip. Instead, I will do the work here in the states over the next months and propose a paper outlining my conclusions for the next meeting of ASOR in November of 2024.