Twentieth-century African American choreographer Katherine Dunham is familiar to international dance audiences; however, a granular understanding of how Dunham successfully managed her company transnationally over an eighty-year career span is less well-known. Dunham's Data: Katherine Dunham and Digital Methods for Dance Historical Inquiry, created by Kate Elswit and Harmony Bench, along with postdoctoral research assistants Antonio Jimenez-Mavillard, Tia-Monique Uzor, and Takiyah Nur Amin, explores three primary datasets manually curated from a large body of undigitized primary source print materials comprised mostly of what Dunham herself chose to save. The three datasets document Dunham's daily travels from 1947 to 1960, the nearly two hundred performers who joined and left her company during that time, and the nearly two hundred repertory numbers they performed. The datasets speak to and support one another, resulting in many multimodal outputs such as interactive and static visualizations, essays, dataset user guides, and the datasets themselves. Dunham's Data uses digital methods more commonly applied in humanities fields to offer new perspective on Dunham's already heavily researched historical canon. The meticulous, manual curation and analysis of the primary datasets differentiates Dunham's Data from similar computational projects navigating large quantities of data using machine learning techniques. Consequently, Dunham's Data leverages the team's expertise in complex analysis and grapples with the messiness of large-scale data analysis through manual assessment and thoughtful curation rather than automating data toward discrete, generic chunks. The entire project is available in a comprehensive online portfolio (https://www.dunhamsdata.org/) that acts as a wayfinding mechanism between various outputs. The portfolio provides four navigation tabs based on the following output types: interactive visualization, dataset, essay, and static visualization. The outputs within each tab are thematically cross-referenced so viewers can access related outputs by clicking on hyperlinks to connected items. For example, the web page data visualization of the “Interactive Network of Dunham Company Repertory” also comes with a hyperlinked list of topically related essays, media, datasets, images, and other visualizations. Collectively, the project offers a thorough analysis of Dunham's career and explores the questions and problems that make the analysis and visualization of data meaningful to dance history.
Dunham's Data expands upon existing scholarship (including the work of VèVè Amasasa Clark, Sara E. Johnson, Dr. Halifu Osumare, Joanna Dee Das, Joyce Aschenbrenner, and Ojeya Cruz Banks, among others) by applying an intersectional feminist analytic approach to the examination of small company exchanges and movements. The project moves between and through canonically driven research to reveal the persistent, daily wear and tear of managing a modern dance company as an African American woman in the twentieth century. Existing scholarship on Dunham's travels, for example, focuses on several key cities, three of the most prominently featured being Paris, Port-au-Prince, and New York. By tracking Dunham's daily location from 1947 to 1960, Dunham's Data fills in travel gaps between prominent cities to tell a different story of persistent travel—a pattern the project team refers to as “movement on the move.” (Bench and Elswit Reference Bench and Elswit2019) The analysis of her travels reveals the median stay in a city is five nights, which when correlated with the other datasets, brings forth a transitory narrative that correlates short stays with the corporeal fatigue of persistently pursuing financial stability. The daily tracking of Dunham's whereabouts also exposes her regular performance at nonconcert venues such as nightclubs, highlighting the nontraditional venue's importance to company solvency. Beyond travel and what it reveals about dance and labor at midcentury, Dunham's Data also builds upon canonical narratives. As Bench, Elswit, and Jimenez-Mavillard point out in the project essay “Connectivity vs. Canonicity: Data Science and Dance Studies in Historical Dialogue,” “the repertory that historians have identified as the most significant are not necessarily the ones performed most frequently” (Reference Bench, Elswit and Jimenez-Mavillard2022, 5). The discrepancy here reinforces the risks Dunham would have incurred by regularly bringing more evocative and challenging works on tour. As the project's data amplifies, nightclubs and concerts in cities with populations of fewer than 1.6 million financially supported the company when public funding and philanthropic donors were not available. Collectively, the project's findings connect to related scholarly work that argues, critical to understanding the circulation of Black dance are deeper understandings of intimate communal relationships (Edwards, DeFrantz, Johnson) and small exchanges of knowledge (Hamera, Noland and Ness). In this context, Dunham's Data offers a unique, data-driven approach to studying the everyday toward feminist, anti-racist examinations of dance history.
As a case study, Dunham's Data serves as a provocation toward integrating computational methods in dance history research. The project builds upon the early work of Christena Schlundt and Jane Pritchard, who both aggregated exhaustive accounts of early modern dance choreographers’ performance venues and repertory. By integrating computational methods from the field of digital humanities, Dunham's Data contemporizes Schlundt and Pritchard's data aggregation practices by managing, analyzing, and displaying the data as digital media. As pointed out by digital humanist Tara McPherson, the digital humanities is an expansive field that trends toward more hegemonic, technophilic practices (Reference McPherson and Gold2000). However, through a critical review of digital humanities research, Dunham's Data intentionally aligns with research that retains ontological complexity and data messiness (Katie Rawson and Trevor Muñoz); employs feminist, anti-racist methods (D'Ignazio and Klein); advocates for data to retain a sense of embodiment (Marie Johnson, Parham); and amplifies the power and precarity of Black intellect in data contexts (Nur Amin, McKittrick). This is exemplified in the authors’ concept of “movement on the move,” which grapples with the relentless negative impact Dunham's transitory lifestyle had on her health and well-being. The desire for the data to retain its visceral sense of embodiment led to what Bench and Elswit define as “visceral data,” or data that actively contends with the very real embodied labor that the data represents (Reference Bench and Elswit2022). Also, worth noting here is the very intentional labor of making all three datasets both publicly available and accessible. Each dataset is accompanied by a detailed user guide that describes the dataset, contextualizes data categories, and provides helpful navigation tips.
In addition to the written scholarship most often considered in a review such as this, it is also worth highlighting the data visualizations, which play a key role in the portfolio's accessibility and legibility. The two visualizations I find most intuitive are “Katherine Dunham's Global Travel” and “Interactive Inspiration Map for Katherine Dunham's Repertory.” Both visualizations use a global map to situate the data, which offers a familiar framework for those wishing to explore the datasets without a particular scholarly motive or question. The play function in “Katherine Dunham's Global Travel” is particularly provocative. The slow, chronological movement through the data and across the map clearly and methodically moves back and forth from city to city, which amplifies the projects recurring discussion of Dunham's relentless global travels and corporeal exhaustion. Also playing with visceral experience is the “Interactive Network of Dunham Company Repertory,” which examines the connections between Dunham's repertory. When I pull or move one node, the entire connected network responds with tenuous, indirect movements that resemble the view of small particles gliding along a microscope slide. The network's entanglement and refusal to give way to extraction builds upon the complex and relational approach to the entire online portfolio. The “Interactive Chord Diagram of Katherine Dunham's Dancers,” “Interactive Flow of Katherine Dunham's Dancers, Drummers, and Singers,” and “Interactive Timeline of Katherine Dunham's Travel” visualizations provide more abstract representational structures that are beautiful as static visualizations but more cumbersome to playfully navigate when attempting to gain a global sense of the affiliated data. All three appear more compelling when applied toward the examination of specific research questions.
As a mixed method, multimodal project, I desire more of the content to engage me beyond my sense of vision. The focus on visual media is not surprising, given how important broad online access is to the project's legibility. However, the recurring theme of “visceral data,” which nods toward Kelly Dobson's use of “data visceralization” (Reference Dobson2012), both attempts to elude ocularcentrism while also depending upon vision-based media for transmission. The profound effort to draw out visceral experience within and across the project's outputs is very clear. However, my most visceral data experiences came from reading the stories told in the project's essays—not the data visualizations themselves. I think growing research in the aesthetics of data sonification (data represented as digital sound [Vickers, Hogg, and Worrall Reference Vickers, Hogg, Worrall and Wöllner2017]) or even data haptification (data represented as digital touch [Hayes and Rajko Reference Hayes and Rajko2017]) could augment the data visualizations and enrich visceral, felt experiences of the project's complex datasets.
Dunham's Data: Katherine Dunham and Digital Methods for Dance Historical Inquiry significantly documents Dunham's career and provides a compelling case study for examining the kinds of questions and problems that make the analysis and visualization of data meaningful for dance history. The mixed-method approach prioritizes the precise manual curation of undigitized primary source print materials comprised mostly of Dunham's personal archive. Given the project's focus on expanding an understanding of Dunham's work beyond the existing canon of written scholarship, it is only fitting that this project also expand the canon of materials reviewed in Dance Research Journal. This gesture initiates more nuanced conversations about how digital methods and multimodal outputs might be recognized as impactful dance studies research. As a meticulously designed online portfolio, Dunham's Data offers students and researchers new data on Dunham's career, novel computational approaches to dance history, and an accessible tool for exploring archival data.
