Cultural Change in the Digital Humanities: Balancing access, participation, and security

I attended this talk at Columbia today (as did Noreen), given by Arienne M. Dwyer, Professor of Linguistic Anthropology and Co-Director of the Institute of Digital Research in the Humanities at the University of Kansas and currently a visiting professor of digital humanities at the CUNY Grad Center.

Dwyer covered many of the issues that came up in our class discussion, particularly how the humanities continue to value publication of analysis (results) over sharing of datasets. Datasets, she said, should be redefined as having value in and of themselves, counted as a type of result, considered as “research products.” Her focus was on sharing of data rather than open-access publication, which is a different aspect of digital humanities work.

Among reasons data is not being shared in the humanities, she noted (1) it sounds too quantitative; (2) it is often the result of, or is a basis for, collaboration, which has not been the way humanists work; (3) repurposing others’ data can seem like replication of their work; and (4) humanities fields do not yet reward it.

A principal example she used, that of the controversy over the publication of the Dead Sea Scrolls, had particular resonance for me, as at the time, in the early 1990s, I was the project editor of an encyclopedia of archaeology in the Near East, out of which a separate encyclopedia of the Dead Sea Scrolls emerged. There was considerable outrage at the time that scholars outside the group who had been given scholarly control over the material published facsimiles of the scrolls, though this was in large part due to the fact that the main group of scholars had been working for 40 years without publishing their findings. The story has considerable more intrigue than this because of the personalities and actions of some of the individual scholars involved, but at the time, the idea that material should be shared, even if the work was ongoing, was deemed radical, and a lawsuit resulted in the affirmation of the original group’s “ownership” of the intellectual content. Even if the humanities has a long way to go in sharing work that is in process, it seems to me that we have already come a long way from that model of 20 years ago.

Another example Dwyer presented focused on the contributions of those outside academia proper: Gunnar Janning, a career diplomat from Sweden, had a longstanding interest in the Uighurs. He was able to publish his work without being dependent on the academic system for his career advancement and standing, acting out of a desire to share with others what he had learned. (His material is now at Lunds University.) A more open digital humanities environment could allow for greater participation by nonscholarly experts like Janning.

Among the ways forward Dwyer put forth are (1) creating incentives for humanists to try self-publishing, which allows for nimble and rapid response to peer and user review; a project she worked on was able to take comments into account to publish a second edition soon after the first. (2) Increase on-campus initiatives in alternative peer review, not just for articles but also for datasets. (3) Create new consortia for establishing discipline-specific data repositories. Sharing costs and workload burdens for ongoing maintenance is a significant factor in making such projects viable after initial publication.

Among the topics raised in the discussion that follows was the obstacle of perfectionism. Scholars are hesitant to publish work in process for fear that they will elicit negative comments from reviewers and that this could have a negative impact on their professional careers. But a positive way of looking at this is that scholars can take such comments into account in improving their manuscripts and addressing criticism prior to formal publication; their own analysis and direction can change in light of such feedback.

I would like to see examples of how datasets that have been shared have then been used by other scholars. Are they using the data for very different purposes, or are they covering similar topics but providing their own analysis and interpretation?