Author Archives: Joe Sheppard

Opening the Gates to the Ivory Tower

Image source: Wikimedia commons.

Lately I’ve been surprised by how much I’ve had to learn about the morals and politics of web publishing, in addition to the technical skills like xml coding, or the protocols for marking up Greek and Latin inscriptions and other documents. But as I prepare data about the ancient world for publication online, it’s perfectly understandable that I’d need to bone up on the humanities side of my digital humanities work, as well as learning the digital aspects.

In recent months the public discussion about free and open access to scholarship has increased and sharpened following the tragic story of the young activist and highly gifted programmer Aaron Swartz. Due to face trial this April for allegedly downloading millions of articles from JSTOR (available only by subscription), Swartz committed suicide in January and was mourned widely and very publicly by advocates for freely available information, who only seem to have become even more committed to ensuring that scientific and cultural breakthroughs (especially tax-payer funded research) are accessible to the public at no cost.

The humanities are yet to fully embrace open access scholarly publications, although a few publications like the digital humanities quarterly and the ISAW Papers are blazing the trail in my field, classical studies. (JSTOR has recently made openly available their “early content” that is now outside copyright.) The question over the next years will be how academics meet the costs of publishing research online that still meets their peer-review standards, and the short answer is that the costs might simply be transferred to the researchers, as happens currently with many scientific publications like PLOS. With any luck the specialized European periodicals for Greek and Latin epigraphy will also open up online too.

But there is also the question of making publicly available the source information. In my case, I want to publish online the texts and information from all the unpublished tombstones in Butler’s Rare Book and Manuscript Library (RBML). Traditionally these kinds of data sets have been hoarded by scholars, protective of the efforts they have put into gathering and curating information. But it is only once all the data is freely available and in an organized form that it can be analyzed comprehensively. With epigraphy the tension has been between a quick and dirty digitization of the vast reference volumes of the 19th and 20th centuries and a more thorough and careful publication of new editions of texts, with accompanying images and metadata, from Roman Italy and the provinces, as well as Christian inscriptions.

One low-cost, high-visibility platform for publishing data has been blogs. Here scientists can published their source information for public re-use, but still receive credit for their work. In a similar way, I’ve been able to encode a log of my activities inside the source code for each inscription I’ve edited, so everyone can read exactly who worked on the Columbia inscriptions, and when. There’s really no reason to be protective of this information, and the RBML are only too willing to get their information out there.

Open source information has also been invaluable though because I’ve basically been able to teach myself how to code simply by observing at what other editors have written before me. I can see precisely what level and amount of information previous coders have supplied. This transparency means any can also check the quality of the information before deciding how to re-use it.

The whole point of my project is to connect the ancient data kept at Columbia University to the rest of the world — just one small link in a chain that will only grow longer and stronger as other collections also digitize their holdings to the same standards and join up with each other. The only way to do this is through openly publishing texts and information whose source can be inspected freely.

It drives me crazy whenever I find the gates on 114th St. or at 117th St. are closed, because it seems like a waste of time having to go the long way around to the College Walk. The very least we can do is take the bolt cutters to the treasure troves of information we have here at Columbia because quite frankly they’ll end up open anyway.

A lot of digital ink has already been spilled on these topics, but the EFF is a good place to start and the ProfHacker blog regularly touches on these issues. You can search the Directory of Open Access Journals for public research, and check out Columbia’s online medical journal Tremor.

I See Dead People

Buried beneath never-ending lines of code all day, I find it easy to forget that I am dealing with a whole bunch of dead people. In a simple sense, the ancient Roman tombstones I'm studying obviously marked the end of an individual's life. They sought to commemorate someone in a more tangible and reliable way than the human memory. But as I record the data from these objects, I am occasionally struck by an immediate connection to a person who died nearly two thousand years ago on the other side of the Atlantic.

Take poor Julius Hagnus, for example, who buried his daughter Pallas a month shy of her fourteenth birthday.

Sacred to the Shades of the Dead.

To Julia Pallas

Who lived 13 years and 11 months.

Julius Hagnus, a father, made

this for his very sweet

and most dearly missed daughter.

It was common practice for the Romans to record the age at death, down to the month and even day in many cases. Scholars have often remarked on the formulaic nature of these monuments, since the deceased is often simply described as "well-deserving" or, as here, "very sweet". But perhaps we begin to glimpse the father's bereavement in the unusual adjective "most dearly missed"(desiderantissima). Strictly speaking Pallas was no longer a child—she would probably have been married within the next year. And since no mother is mentioned, we are left with the impression of a poor father left completely on his own.

Roman tombstones have provided historians with good evidence for family structures, burial practices and patterns of commemoration. But they are also vital sources for dead people in another sense, because they are our best evidence for people who were otherwise socially dead. These are individuals such as household slaves, who were alienated from society and have left to us very few other traces of their identities. Similarly, in Butler library I have come across many small–time workers who chose to record their lives as hairdressers, smiths or charioteers, with dignity if not pride. They lived a world apart from the great politicians, writers and generals of our history books.

Hylas, a charioteer for the Blue faction, lived 25 years and won 7 races in the two-horse chariot.

Image CC License, courtesy of PicSwiss.

I want to bring these unpublished documents to life by organizing tagged versions of these tombstones online. These dead people deserve a place in the social history of everyday Romans, which will only become easier to write as more and more Latin inscriptions are systematically resurrected and added to a comprehensive and searchable corpus.

Long live the zombie Romans!

Reporting for Digital Duty


Today begins my project of publishing online the 121 Roman tombstones that are housed in Columbia University's Rare Book and Manuscript Library (RBML). These stone inscriptions — mostly Latin and carved in the first two centuries AD — were donated to the university by Prof. George N. Olcott in 1912, and were only uncovered during recent research towards a centennial exhibition commemorating his bequest.

This is what a Latin inscription looks like! The text is often damaged, incomplete or difficult to read.

By publishing these inscriptions online, scholars around the world will have fast and ready access to these important historical documents at Columbia, fifty of which have never been published before. My work also contributes to a larger project that seeks to bring online all of the ancient Roman inscriptions in the U.S., with accurate texts, high-quality images and searchable, linked metadata for each piece. To achieve this we are using EpiDoc, a collection of guidelines for capturing the information from these particular kinds of documents and marking that up in XML.

I am excited about this project because I will learn new skills while making a real contribution to my field. My aim with this blog is to document the progress I make.

As a graduate student in Classical Studies at Columbia University, however, I look forward to collaborating with my colleagues also working in digital humanities and social sciences at Columbia. For example, we could probably use digital technology to streamline how we record spatial information during the archaeological field work each summer at the APAHA project at the Villa San Marco in Castellammare di Stabia, Italy. And after the stone inscriptions have been published, there are over 3,500 coins in the RBML that need to be catalogued properly before scholars can access them for study — thus I hope to apply the lessons I learn this year for larger projects in the future.

Some of the challenging spaces and architectural features we have to record quickly and accurately during excavations.

Please get in touch with me ( if you would like more information — or if you want to collaborate with me on this project or learn more about digital publishing with me over the 2012-13 year. Apparently I've already learned how to use WordPress! For now though I have to wade through those EpiDoc guidelines and figure out how to track different edited versions of coded texts. Till next time!