Concise, critical reviews of books, exhibitions, and projects in all areas and periods of art history and visual studies
December 21, 2021
Diana Seave Greenwald Painting by Numbers: Data-Driven Histories of Nineteenth-Century Art Princeton: Princeton University Press, 2021. 256 pp.; 55 color ills.; 9 b/w ills. Cloth $42.00 (9780691192451)

Diana Greenwald is trained as both an art historian and an economist, and in Painting by Numbers: Data-Driven Histories of Nineteenth-Century Art, she aims to bring the methods and explanatory force of both disciplines together in the analysis of nineteenth-century art. As suggested by the subtitle, the book is driven primarily by a methodological call to the field. In this, it is analogous to Matthew Jockers’s Macroanalysis: Digital Methods and Literary History (University of Illinois Press, 2017), which challenged literary scholars to use computational methods to enlarge the scale of literary history from the micro to the macro, moving from close readings of a single text to the analysis of thousands of texts. But despite their shared interest in combining quantitative methods with humanistic inquiry, there are important differences in their approaches. Greenwald’s interest lies not in digital and computational methods per se, but in bringing the economic tools of larger data sets and statistical analysis to the questions of art history, allowing for “analyses that oscillate between focus on individual artworks and general trends” (51). 

The book falls into two main parts. It opens with two chapters on Greenwald’s methodological aims and data sets, making a case for a data-driven approach to art history. The nineteenth century, a century increasingly focused on record keeping and statistics, is fertile ground for this approach, and Greenwald has done tremendous work locating, transcribing (with the necessary help of grant money and paid assistants, whose work she acknowledges), and analyzing three data sets: the Whitley Index to Paris Salon Painting; the Historical American Art Exhibition Database; and the Royal Academy Exhibition Database. The compilation of these data sets is one of the project’s significant intellectual contributions, and Greenwald has made them available online, a form of transparency that is unfortunately rare in the humanities and should be widely emulated. Greenwald is also a gifted interpreter of economic methods, and her appendix on regression analysis is a model of clarity.

The remainder of the book is devoted to three case studies that aim to demonstrate the usefulness of her approach, including a chapter each on the relationship between industrialization and the production of rural genre in French painting; the professional challenges faced by women artists in New York; and the representation of empire at Royal Academy exhibitions in London. The breadth of existing scholarship in each of these research fields brings significant opportunities and risks. This extensive scholarly work ensures that the case studies will be of interest to many researchers in the field, and that there are usable data sets available for analysis. However, it also requires Greenwald to rephrase a large and often contentious body of literature in terms amenable to quantitative measurement. This translation process means that the conclusions Greenwald can draw from her materials often feel adjacent to the art historical literature she engages. 

For example, in the chapter on the links between the painting of rural scenes and the forces of modernization, she describes the work of Robert Herbert, T. J. Clark, Linda Nochlin, Griselda Pollock, and Nicholas Green as sharing the hypothesis that “as France developed, images of rural life and landscape apparently increased in frequency, notoriety, and political importance” (53). Using statistical analysis of data sets of Salon exhibitions with titles and keywords acting as proxies for subject matter, she argues that the numbers do not support the claim that industrialization resulted in the production of more rural-genre scenes and uses regression analysis to show that no particular effect of modernization (e.g., urbanization rates, political unrest) had a significant causal relation to the subject matter of rural genre or landscape. (The development of transportation networks is the one exception: artists were more likely to paint areas to which they could easily and cheaply travel.) Art historians are, in fact, sometimes guilty of making claims about “growing interest in X” without providing quantitative support, and Greenwald’s work is a welcome correction to this tendency to generalize. But of course what those scholars tend to mean is that there is growing interest in a given issue among the artists and artworks that art historians have already identified as important; and the fact that there were not, in fact, significantly higher numbers of rural-genre paintings at the Salon is unlikely to shift these scholars’ underlying presumptions—that the work of Courbet, for example, is responsive to the changing economic and cultural meanings of the urban and the rural in midcentury France. And if the data suggest that he was, in fact, unusual in taking up such subjects, the claim of originality is more likely to solidify his canonical place in art history than to dislodge it. 

Greenwald tackles the question of the canon head on in her chapter on women artists, demonstrating that while women contributed 12 percent of the works on view at the National Academy of Design, they produced only 6 percent of the nineteenth-century artwork acquired by the Metropolitan Museum of Art. She posits that this disparity is the result of constraints on women’s time that led them to work in “lesser” genres, concluding that, “unfortunately, those same genres and media that are quicker to work in are also the ones that have traditionally been neglected in museum collections” (102). Her discussion of canon formation thus stops at the empirical, which feels like a missed opportunity to theorize more explicitly the places art history and economics seem to speak past one another. There is more than “fortune” at work in these disparities. Much of the social history of art the author engages is explicitly concerned with inequities of power and their effects. There are certainly, as Greenwald notes, economic approaches to questions of inequity, but the underlying claim of the primacy of the quantitative is foundational for economics, and Greenwald does not address the specific ways that kind of data intersects with the primary concerns of art history. How precisely does knowing more about the numbers of rural-genre works exhibited at the Academy relate to the question of what those works meant to their artists and viewers? What cultural values underpinned the devaluation of still life and watercolor?

In making visible these tensions, Greenwald’s book does important work in highlighting the vexed question of “value,” a genuine intellectual problem for the field and one that has still not been adequately theorized. But her economic lens means that she sometimes fails to engage its full difficulty. Most art historians working today would agree that the proper object of our study is decidedly not “the canon,” the relatively small number of artists and artworks that have been deemed “great” and have accrued cultural and financial value over the years in a process fundamentally shaped by power. Greenwald describes the problem as “sample bias,” a research error that “emerges when the group of people, objects, or other entities that a scholar analyzes is both limited and misrepresentative of the entire population that the scholar is interested in studying” (5). But using the term “sample bias” to describe the canon, and suggesting that more data can correct it, sidesteps the core challenge her method makes to the discipline: in a sense, sample bias is the ultimate goal of the long history of connoisseurship and canon building—finding the unique and exceptional among the ordinary, the noise amid the signal. Many of the most significant scholarly contributions of the past fifty years have analyzed the formation of that canon and the values it embodies, and strategically used archival sources (data) to reshape it, both by adding to the body of artwork preserved and analyzed and by expanding the boundaries of “art” itself to encompass other forms of creativity. However, there are still underlying theories of value—examined and unexamined, aesthetic and political—driving those choices. Is the proper study of art history the examination of all objects ever produced? And if not, on what grounds are decisions made about exclusion and inclusion? 

The real power of the book is not, in the end, its conclusions about nineteenth-century art, though there are thought-provoking tabulations of trends and important corrections to some current assumptions in each case study that will be useful to scholars of those specific areas. But Greenwald’s aims and achievements are of a different kind; in her words, the book is intended as “the beginning of a methodological conversation” (22). At the macro level, prompting art historians to ask questions about the values underpinning their definition of their objects of study is an important call to the field at a moment in which crucial challenges to the canon require rigorous and explicit theorization. At the microlevel, Greenwald makes suggestions throughout the text for new data sets and research projects. The sense of possibility that guides her approach to research is generative and intellectually generous, and open-minded readers will be prompted to reimagine their own research questions in a new light. For example, for scholars of exhibition culture, the “metadata” of titles might be the subject of study in their own right: artists used titles to do far more than describe a work’s content, and textual analysis might reveal different kinds of patterns, shedding light on relations between the literary and the visual or on artists’ efforts to shape audience response. My point here is not to make the (annoying) suggestion that Greenwald should have written a book other than the one she wrote; it is to highlight the productive inquiry that the book and the methods it models have the potential to spark. If Greenwald cannot fully resolve all the questions that she raises by bringing art history and economics into closer conversation, she has done a valuable service to the field in asking us to rethink our fundamental categories of disciplinary concern and our responsibilities to the vast range of visual and material culture that might fall within their purview.

Pamela Fletcher
Art Department, Bowdoin College