Source: Open data for art history research

Although sharing research data has been a huge concern for researchers in both social science and hard science disciplines for quite some time, the issue certainly came to the fore around 2010, when the National Science Foundation required that all grant proposals include a Data Management Plan, which required considerations for standardizing, archiving, and sharing data.1 As the NSF is one of the major providers of funding for scientific research in the United States, this requirement has had huge implications for researchers, who now not only need to worry about collecting and analyzing data, but also have to develop strategies for preserving this data and making it accessible. As data collection is often difficult and resource intensive, the benefits for sharing research data are potentially huge, enabling future researchers to build upon existing data sets in ways the original researchers never imagined. However, as Christine Borgman discusses, the discourse surrounding data sharing is dense:2

[There] are thick layers of complexity about the nature of data, research, innovation, and scholarship, incentives and rewards, economics and intellectual property, and public policy. Sharing research data is thus an intricate and difficult problem—in other words, a conundrum.

An engaging body of literature about the challenges and possibilities of sharing scientific data has begun to grow, especially in the past several years, with contributions coming from academic librarians, data curators and archivists, and the scientists themselves. Some of the key issues from a library science perspective include the negotiating the role of librarians and archivists in helping to manage research data3 and understanding how to incorporate good data management practices into researchers’ existing workflows.4 While this rich discourse continues to evolve around data sharing in the sciences, I wonder about the possibility for similar discussions in the humanities, and particularly for (digital) art history research. What are there venues for this discourse in the humanities? What are the benefits of sharing art historical research data and what are the potential issues? There is clear motivation and a high level of importance for this discourse in the sciences (not least because of the huge amounts of funding money at stake), but is this discourse just as critical for art historians?

As Borgman suggests, scientists have a plethora of concerns about sharing data, such as how to ensure that re-used data is properly attributed, how data sharing might help to contribute to a faculty member’s tenure considerations, or how to guard against potential misuses of data. Art historians would likely have to negotiate many of these issues as well if data sharing were to become widespread across the discipline. However, data sharing in practice would look very different in art history than it does in the sciences, and the unique nature of art history research as it has traditionally been practiced would raise some serious hurdles.

For one, art history is a far more individualized discipline than any of the social or hard sciences. Art history research is often characterized as long and intense contemplation of images, previous scholarship, and other historical documents by a lone scholar, who deliberately writes up her findings. In this traditional research model, there is not a lot of opportunity for sharing data. Art historians are also notorious perfectionists, who dislike sharing anything before it is completely squared away and ready to be published. Another difficulty is that there is very little standardized data in ‘traditional’ art historical research. While many different social scientists would all be able to (hopefully) make sense of another researchers dataset from a large survey, any two given art historians may employ radically different and idiosyncratic methods of analyzing images, with perhaps a great deal of of that analysis remaining internalized and never explicitly or systematically recorded. Art historians working on living artists or studying current cultural practices also might have incredibly sensitive data, and may not want to share the raw data out of respect for their subjects.

Given these difficulties, many art historians may question if it’s worthwhile at all to even think about sharing data. If other art historians will be able to read and build upon the research once it’s published in a journal or as a monograph, then what’s the point? Although the kind of widespread data sharing common in the sciences will perhaps never catch on with ‘traditional’ art historians, the increasing importance of the digital in art history is rendering that kind of ‘traditional’ research more and more outmoded. To take full advantage of the opportunities of digital art history research, then art historians will have to get comfortable with sharing their datasets, as well as their tools and methods.

There is already real evidence of this shift in the art history disciplinary culture with the working practices of digital scholars especially. For example, in Thomas Padilla’s interview with digital scholar Matthew Lincoln, Lincoln talks about how he rigorously documents his data in order to facilitate sharing the data and having it be used by a wider set of researchers.5 He modeled his own data habits off of other scholars that he admired in the field, taking up these practices as a kind of de facto standard. While Lincoln’s willingness to share his data—and the steps he takes to make his data usable—is exemplary, these practices were taken up on his own volition, and not learned in a methods course or another educational or professional development setting. Digital art history can develop at an accelerated pace if researchers share their data, but they have to learn best practices for standardizing, cleaning up, and making that data accessible. What are the best venues for art history as a discipline to negotiate and articulate those best practices? How should those best practices be taught to students, as well as scholars already established in the field? These are some of the questions that art history has to tackle.

For my own part, I would also be very excited to see more datasets released by museums, similar to what was recently released by the Tate.6 In addition to cataloging data, I would be interested to see more datasets released containing information on conservation and preservation actions artworks have received over time. As one of my main research interests is contemporary art preservation, it would be quite intriguing to see how conservation practices have evolved over time, as well as how conservation is recorded differently for different kinds of art. What language is used in the conservation records for Renaissance paintings versus contemporary sculpture? It might be interesting to visualize these differences through a word cloud, for instance.

Although I’ve mostly been talking about actions that scholars need to take to share their data, museums clearly need to be a part of this discussion as well. Prominent institutions like the Tate and the Getty are setting the example by releasing more and more information for researchers, but how might these same practices trickle down to smaller institutions. What role should these institutions play in establishing best practices for how data should be shared?


[1] National Science Foundation. (2010). NSF data management plans.

[2] Christine L. Borgman, “The Conundrum of Sharing Research Data,” Journal of the American Society for Information Science and Technology 63, no. 6 (2012): 1059–1078. doi:10.1002/asi.22634.

[3] Sheila Corrall, “Roles and Responsibilities: Libraries, Librarians and Data,” in Pryor (ed.), Research Data Management (London: Facet, 2012), 105-133.

[4] Jillian C. Wallis, Christine L. Borgman, Matthew S. Mayernik, and Alberto Pepe. “Moving Archival Practices Upstream: An Exploration of the Life Cycle of Ecological Sensing Data in Collaborative Field Research,” International Journal of Digital Curation 3/1 (2008) 114-126.

[5] Thomas Padilla, “Data-Driven Art History: Framing, Adapting, Documenting,” DH+lib (27 October 2015).