A widespread opinion within the academic community is that art historians have been left behind by the digital turn in the humanities (see Greenhalgh 2004, Zorich 2012, and the essay by Elizabeth Honig in this issue). According to many researchers and educators, the field of digital art history (DAH) — that is, the application of computational tools and analytical techniques to art-historical research questions — has not fulfilled its potential and remains a marginal branch of the discipline (Zorich 2012, 6). As Benjamin Zweig has argued, however, the assumption that art history has lagged behind other humanities disciplines in employing digital technologies is not entirely correct. In a 2015 article, he documents a selection of pioneering DAH projects, initiatives that emerged as early as the 1970s. Examples include the 1983 collaboration between the Getty Trust and the Architectural Drawings Advisory Group to produce a cataloging standard for digitized drawings that would allow scholars to search across all electronic repositories of works on paper as well as manipulate this information in ways that could stimulate new views of the material and thus, new research questions. The ambitious MORELLI project, launched in the mid 1980s by William Vaughan, a professor of art history at Birkbeck, University of London, is another example of an early DAH project. MORELLI, named for the nineteenth-century Italian connoisseur Giovanni Morelli who developed a system of identifying an artist’s “hand” by means of the close examination of a few key details that the artist executes consistently, was a pattern recognition tool that sought to classify and evaluate the formal qualities of images (Zweig 2015).
Unfortunately, these and similar initiatives did not fulfill their potential. Issues of funding, sustainability, archiving and access, copyright, and technological barriers ensured that what few gains were made were quickly undermined. In short, art historians faced challenges sourcing the software, maintaining and supporting collaborations and staff, and building and maintaining the datasets necessary for sustained intellectual engagement with DAH. As these early projects foundered, the field of literary studies advanced in the digital humanities — and with good reason. When the dataset is a text that can be scanned, when the “catalog” — in this case, the set of words included in that text — is readable across all platforms, it is easier to explore research questions using widely available digital tools, such as topic modeling and word embeddings. Unfortunately, these tools are not as useful for art historians, unless they want to analyze the oeuvre of an art historian, critic, or the writings of an important art-historical figure. Such an approach could be useful, but what about working with images? Isn’t that what art history is about?
This is one of the reasons why there remains anxiety among art historians that they have fallen behind their academic peers: art history is different. What defines art history is the study of the visual. Thus, the tools used and methodologies developed for DAH should support visual analysis and learning. Unfortunately, in the rush toward working digitally, many art historians and museum professionals have enthusiastically borrowed theoretical models and technical terms from disciplines and sub-disciplines such as literary theory, computer science, and cultural analytics without careful consideration as to how these methods and terms function within the specificities of the field of art history. Equally troubling is how many art historians have adopted a software developed for other disciplines and industries without due appreciation of how it structures data and thus might privilege certain information and promote certain trends (see Jaskot and Van der Graff 2017). In short, art historians require different tools — such as image-recognition software — tools that allow them to search, organize, catalog, and investigate images effectively, and few tools that fulfill these needs are available. Fortunately, with recent advances in the fields of computer vision and Artificial Intelligence, this will change. The future is looking bright.
But what about the present? Despite the challenges outlined above, we would argue that it is still possible for art historians to benefit from embracing existing current digital tools and computational techniques. As the essays in this special issue attest, there is a place in the discipline right now for the digital.
The issue begins with Alison Langmead’s analysis of art history’s longstanding engagement with image-based reproductions and how these “significant remediations” have affected the development of the discipline. Far from being marginalized by the digital turn, she argues for art history’s central role in the digital humanities as a model for the “mindful” adoption of technology. The discipline, she concludes, “can offer objective lessons on the risks of taking both the technical and social affordances of [digital] tools for granted” and thus, may promote awareness of how technologies impact the research process.
The remaining essays are divided into two groups that analyze DAH at work in the art gallery and the university. C. Richard Johnson, Jr. and William A. Sethares’s “Hunting for Weave Matches: Computation in Art Scholarship” outlines the pioneering application of signal processing to a traditional technique of painting analysis — thread counting. Determining the weave density of the canvas support of a painting offers conservators crucial technical information as well as useful clues regarding the painting’s origin and approximate date of execution. Manual techniques for thread counting, however, are time-consuming and laborious. Johnson, a professor of engineering at Cornell University, realized that this process could be automated by means of algorithmic analysis of x-rays of paintings. Using this method, the canvas supports of different paintings may be matched. These matches suggest that the paintings may have originated from the same bolt of cloth, a discovery that in turn suggests the paintings were in the same place at one point in their history. In their article, the authors outline the procedure for comparing possible weave matches in full, stage by stage, providing increased understanding of this ground-breaking technique and offering a model for future collaborations between art historians and scientists.
“Animated Shadows on Virtual Stone: Ancient Sundials in a Gallery Setting” by Sebastian Heath, Rachel Herschman, and Christine Roughan describes the use of animated displays at a 2016–2017 exhibition held at New York University’s Institute for the Study of the Ancient World (ISAW). The exhibition introduced audiences to more than one hundred time-tracking devices from Greco-Roman antiquity. To illustrate how ancient audiences experienced these devices, the organizers formed collaborations across multiple departments, including the work of graduate students, to develop a series of 3D animations using the open-source animated suite Blender. The resulting digital surrogates were cost-effective and highly informative, not only helping the public to understand how the objects on display functioned but also showing examples that could not be included in the exhibition due to size or location. Happily, these “exhibits” will enjoy an afterlife on the journal’s site.
Appropriately, many of the articles in this issue examine the use of digital tools and new methodologies in the classroom. Elizabeth Honig provides an honest and enlightening analysis of two courses she designed that disrupt the art history lecture, the cornerstone of art-historical instruction. As she notes, transitioning from the lecture format, reliant on the projection of images in a darkened room as well as the instructor’s expertise, to an interactive model that requires students to engage with data collection, digital technologies, and each other is not a simple process. Professors of art history will have to gain new skills in computer science and statistics; what will be more difficult is learning to embrace the serendipitous encounters, unplanned epiphanies, and failures of project-based learning. The professor “cannot control either process or outcomes” as in a lecture course, Honig observes. Reworking the traditional art history class, however, will ultimately enable students to develop new skills in critical analysis and knowledge production.
“3D Modeling in the Urban Classroom: Using Photogrammetry for the Study of Historic Architecture in Coral Gables, Florida” presents an additional, highly innovative approach to rehabilitating the traditional art history course. Instead of focusing on the formal qualities and construction history of the architectural monuments in lectures and emphasizing art-historical research and writing in assignments, Karen Mathews, Assistant Professor of Art and Art History at the University of Miami, and her team seize the opportunity to teach students about data collection and curation, collaboration, online exhibitions, digital tools and storytelling, and public art history as well. The materials, textual and visual, that are produced during the semester are published on an interactive website, a project that not only increases the students’ but also the public’s engagement, creating “bridges between the university and the community around it [and] promoting awareness of cultural heritage sites and their preservation.” The authors’ thorough explanation of their process generously allows other educators to investigate this innovative pedagogical paradigm.
Two articles in this special issue focus on Pre-Columbian topics; this may not be coincidental but instead indicative of how specialists of non-Western cultures have often been critical of conventional research methods and pedagogical strategies and are more willing to explore novel approaches in their scholarship and instruction. Ellen Hoobler’s article, produced in collaboration with three of her students, presents an additional example of how 3D modeling can offer increased understanding of the material past. Hoobler’s original plan was to expand on her dissertation research examining the tombs of Monte Albán in Oaxaca, Mexico, built by the Zapotec peoples in ca. 500 BCE–850 CE, by producing an interactive model of one of these tombs, including all of its contents. During the course of two summers, her team created dozens of 3D models, visualizations, and videos as well as a website documenting their work. While the completion of her original vision proved a much more difficult process than anticipated, the experience of Hoobler and her students yields instructive insight into issues of collaboration, sustainability, funding, and community engagement. Hoobler notes that perhaps the most important lessons her students gained from this DAH project were the “byproducts” of their contributions: increased skills in visual analysis and archival research as well as a deeper comprehension of the process of contextualization.
Outlining the incorporation of DAH methods and tools into a Pre-Columbian art history survey course, Lauren G. Kilroy-Ewbank’s essay expands on several of the themes developed throughout this special issue, including the instructive byproducts of the student-centered DAH project. The centerpiece of her course is the creation of a collaborative online exhibition using Omeka, a popular open-source content management system for the production of online exhibitions, focused on the Mixtec Codex Zouche-Nuttall (ca. 1450). The assignment requires students to source and annotate images, create metadata, and assess online resources as well as undertake traditional art-historical research; it also encourages students to consider issues of digital visual culture, digital storytelling, and the possibilities of public art history. Like Honig and Hoobler, Kilroy-Ewbank finds that student engagement with DAH is not always a straightforward process. Yet, she argues, students receive valuable lessons regarding responsibility, both to their team and the discipline, from the experience. Helpfully, Kilroy-Ewbank includes her course syllabus as an appendix, making it easy for similar-minded pedagogues to integrate these types of assignment in their own classrooms successfully.
The issue concludes with two non-peer-reviewed pieces that record experiences of working within the field of DAH. John B. Henry documents his evolving research on David Wojnarowicz and how information resources created by the Artist Archives Project that guide researchers through the collections held by New York University’s Fales Library and Special Collections affected his understanding of the artist. Henry argues that this tool not only facilitates research but also offers the opportunity for additional knowledge creation through the incorporation of online comments and the institution of an online community. More than a digitized collection, it is in effect an “ambitious project in digital art history” that offers new models for academic study.
Finally, Naraelle Hohensee presents her 360-degree video of Seattle’s Olmsted Brothers-designed Volunteer Park (1904–1910), an educational resource that offers an immersive environmental experience as well as presentations of historical documentation. Her literal exploration of this public space fuses design history with urban studies and offers an embodied understanding of the site, thus — like many of the DAH projects introduced in this issue — providing increased comprehension of how people interact with the built environment. The result is both an informative and entertaining mediation on the designers’ accomplishment and a subtle analysis of the evolving role of the historian in the digital age that raises questions regarding the use of digital storytelling in the academy and across cultural institutions.
Technology that allows art historians to work with images rapidly and effectively will be available in the near future, and art historians must take an active role in determining how this technology is developed. After all, who better than art historians to direct these advancements? Or, as Kilroy-Ewbank observes: “Art historians (and our students) are […] positioned to think critically about digital visuality and analyze how digital visual environments encode ideas.” Art historians should remain alert to these new developments in computer vision, stay informed, and, when possible, collaborate with technologists, computer scientists, and other digital humanists to ensure that these professionals and their projects benefit from their expertise. In this way art historians will once again become significant contributors to the digital humanities.