Shifting the Archival Gaze: A Case for Leveraging Computational Methods to Uncover Media History Narratives

Abstract

Through an interdisciplinary examination of a bourgeoning technology, this essay grapples with the opportunities and challenges of using a type of artificial intelligence called machine learning to catalog and make sense of the unpreceded number of digital materials now available to media historians. The authors—a journalism professor and an AI researcher—describe their recent interdisciplinary efforts to use machine learning to explore a collection of roughly 5.7 million photographic negatives donated to their institution by the Boston Globe. Their work illuminates the tremendous power of emerging computational methods to uncover more complete histories, but it also serves as a reminder that such tools should be used with caution.

Notes

1 Laura Mulvey and Scott MacKenzie, “Visual Pleasure and Narrative Cinema,” in Film Manifestos and Global Cinema Cultures, 1st ed., A Critical Anthology (Berkeley: University of California Press, 2014), 359–70, http://www.jstor.org/stable/10.1525/j.ctt5vk01n.109; bell hooks, “The Oppositional Gaze: Black Female Spectators,” in Feminist Film Theory, edited by Sue Thornham, A Reader (Scotland: Edinburgh University Press, 1999), 307–20, http://www.jstor.org/stable/10.3366/j.ctvxcrtm8.30; E. Ann Kaplan, Looking for the Other: Feminism, Film, and the Imperial Gaze (London: Psychology Press, 1997); Gallen 2008; Jacqueline Z. Wilson and Frank Golding, “Latent Scrutiny: Personal Archives as Perpetual Mementos of the Official Gaze,” Archival Science 16, no. 1 (March 2016): 93–109, https://doi.org/10.1007/s10502-015-9255-3; Kimberly Anderson and Harrison W. Inefuku, “Focusing the Archival Gaze: A Preliminary Definition and Model.”

2 For more on the collection, see Boston Globe, Northeastern University Libraries Archives and Special Collections, Boston, MA, https://globe.library.northeastern.edu/collection-description/.

3 Tze-I Yang, Andrew Torget, and Rada Mihalcea, “Topic Modeling on Historical Newspapers,” in Proceedings of the 5th ACL-HLT Workshop on Language Technology for Cultural Heritage, Social Sciences, and Humanities (Portland, OR: Association for Computational Linguistics, 2011), 96–104, https://aclanthology.org/W11-1513.

4 Marcel Broersma and Frank Harbers, “Exploring Machine Learning to Study the Long-Term Transformation of News,” Digital Journalism 6, no. 9 (October 2018): 1150–64, https://doi.org/10.1080/21670811.2018.1513337.

5 “Exploring Machine Learning,” 1150–64.

6 Lassie Nordhal, et al., “Using Computer Vision to Create A More Accurate Digital Archive,” R&D at the New York Times, July 21, 2021, https://rd.nytimes.com/projects/using-computer-vision-to-create-a-more-accurate-digital-archive (accessed August 1, 2022).

7 A team of librarians, curators, archivists shared their insights on both the physical and digital collection, providing the necessary human expertise to this machine-based project. In particular, the Library’s Patrick Yott, Kim Kennedy, and Sarah Sweeney have provided knowledge on the collection, the digitization pipeline, and the feasibility of the project.

8 Kate Holterhoff, “From Disclaimer to Critique: Race and the Digital Image Archivist,” Digital Humanities Quarterly 11, no. 3. http://www.digitalhumanities.org/dhq/vol/11/3/000324/000324.html (accessed August 28, 2017).

9 Dorothy Berry, “Digitizing and Enhancing Description Across Collections to Make African American Materials More Discoverable on Umbra Search African American History,” (August 2, 2018). The Design for Diversity Learning Toolkit, https://des4div.library.northeastern.edu/digitizing-and-enhancing-description-across-collections-to-make-african-american-materials-more-discoverable-on-umbra-search-african-american-history/ (accessed March 20, 2023).

Additional information

Notes on contributors

Meg Heckman

Meg Heckman is a journalist, author, and educator working to solve two of journalism’s biggest problems: a lack of gender equity in news and the decline of the local information ecosystem. She leverages feminist media history to better understand the contours of modern journalism and enjoys collaborating with computer scientists and digital humanists. Heckman also studies digital news production and dissemination. She is an assistant professor in Northeastern University’s School of Journalism, a faculty affiliate of the NULab for Texts, Maps, and Networks, and an executive committee member for Northeastern’s Women’s, Gender, and Sexuality Studies program.

Giulia Taurino

Giulia Taurino is a Postdoctoral Research Fellow at The Institute for Experiential AI at Northeastern University. Her research focuses on forms of content organization on online platforms and digital archives, cultural implications of algorithmic technologies, and applications of artificial intelligence in the arts, heritage, and museum sectors. She is a member of the NULab for Texts, Maps, and Networks, and the MIT Data + Feminism Lab. Past affiliations include Brown University’s Virtual Humanities Lab, MIT’s Open Documentary Lab, and metaLAB (at) Harvard. Giulia holds a doctoral degree in Media Studies and Visual Arts from the University of Bologna and the University of Montreal.