Archiving Activism
#ArchivingActivism: Preserving the Web Content of Social Movements and Revolutions
Author: Kyle H. Goedert
Introduction
This bibliography considers how libraries and archives preserve digital content born from social justice movements, protests, and revolutions. This includes tweets, Facebook posts, blogs, viral videos, and other digital-born content, with particular attention paid to those connected to popular hashtags like #metoo, #blacklivesmatter, and #occupy. Relevant literature was found in LIS journals such as the Library Trends and The American Archivist, as well as publications in the fields of communication, media, and information networking.
Because the most written about early examples of viral social movements and revolutions are the Arab Spring and Occupy Wall Street, which both began in 2011, only articles from 2012 to 2020 were considered. All resources selected are peer-reviewed journal articles. Prominent subtopics include how digital content is collected, how it is displayed, working with communities and activists, and what benefits this work produces.
The most successful indexing terms in searches were:
- Activist archives
- Digital archives
- Living archives
- Protest movements
- Social media
- Social movements
- Web archiving
Annotations
Allard and Ferris describe the University of Manitoba’s Digital Archives and Marginalized Communities Project (DAMC), which includes three digital collections: The Missing and Murdered Indigenous Women Database (MMIWD), the Sex Work Database (SWD), and the Post-Apology Residential School Database (PARSD). Stressing the need to work with these communities to set goals for the collections, the archivists capture social media posts, blogs, and websites by sex work activists fighting for decriminalization and legal protections; online news media; digital public records; photos and videos; academic articles; and memorial Facebook groups and webpages. They argue that the Internet “constitutes a key organizing and dissemination space for commemorations posted by grassroots organizations struggling to foreground the concerns of Indigenous women in their antiviolence, antipoverty, and feminist work” (p. 362). Allard and Ferris broach the topic of whether mainstream news, court records, and other official documents that portray these communities derogatorily should be displayed beside first-hand accounts and artifacts from the communities themselves. They also discuss how to deal with who should have access to what information, particularly in relation to activities that police might have interest in surveilling. This article contains valuable information for archivists interested in creating participatory archives, explaining how they created partnerships with activists and community members, ethical issues that may arise, and how to communicate effectively with interdisciplinary partners and community participants.
Arnold and Sampson explain Twitter’s Application Programming Interfaces (APIs) and how they can be used by researchers hoping to examine an ongoing social movement. They suggest that it is best to start by identifying leaders (users being tagged most often by other users), and then use the REST API (REpresentational State Transfer) to aggregate their tweets from the prior week. Popular users and hashtags can then be used as search terms in the STREAM API, which collects tweets as they are generated in real time. They stress that documenting “the tools and commands, as well as the decisions behind their use” (p. 522) must be a top priority to anyone archiving tweets, because the dataset will lose all meaning if the methodology is unknown. The authors provide recommendations of best practices derived from the Open Archive Information System (OAIS) Reference Model and its Archival Information Package (AIP), and their examination of AIP’s metadata concerns – “provenance, reference, fixity, context, and access rights information” (p. 524) – is valuable to anyone planning to aggregate and archive Twitter data.
Atiso and Freeland attempt to identify reasons that the Documenting Ferguson archive received less community participation than archives such as Our Marathon, which commemorates the Boston Marathon bombings. They tested two hypotheses: first, that there may have been a lack of awareness and/or lack of trust that prevented individuals from contributing; and second, that functionality issues with browsing, searching, contributing, and sharing features made it too difficult for many to do so. They also attempted to identify motivations for contributing to the archive. A questionnaire and usability study led them to conclude that most contributors participated in the archive for altruistic reasons, but that lack of awareness and search issues on the website played large roles in lack of participation. The authors end with recommendations that the archives raise awareness, involve the community, make contributing to the site easier, and update privacy policies. Both the questionnaire and usability study provide useful insights into participation for future digital archivists to consider when building their archives.
Foster and Evans examine the Documenting Ferguson project launched by Washington University in St. Louis, as archivists in the library system attempted to respond in real time to the protests erupting after the police shooting of Michael Brown, Jr., on August 9, 2014. The authors highlight the issues faced by the project as they attempted to curate the wave of digital-born content created by activists, most significantly gaining the trust of the community and confronting lack of metadata from user-uploaded content. They note that reliance on activists to submit their digital content also means they must accept as much or as little metadata as the submitters provide in real time; because protestors are not knowledgeable about the importance of administrative and technical metadata, and many fear their information ending up in the hands of law enforcement, only a name and email address were required to upload and was often all that was provided. Here, the archivists accepted that the format of the project fell below traditional standards of archival science and did not attempt to remedy the metadata issue. In order to maintain neutrality, the archivists also employed an Archive-It web crawler that collected news articles and social media postings from all perspectives. The authors conclude that, despite the difficulties the project faced, it provides a framework for future living archives and has the power to change the way these types of events are documented.
Pennell explains that, while the documents of the Gaddafi regime collected after his death provide one source of Libyan history, there are two born-digital sources that offer other important narratives of the revolution: the Libya Uprising Archive, a collection of tweets sent during the revolution by fighters, citizens, and journalists; and the depositions given to asylum tribunals by Libyans attempting to flee the country. His analysis of the Libya Uprising Archive from the day that Gaddafi was killed shows the chaos, rumor, fear, and celebration of the revolution and, he explains, “they represent a record of activities, and a means of bringing about change. But they are more, because they originate outside the structures of the state and are directly instrumental in the events themselves and the collapse of that structure” (p. 70). The asylum records, which can be found in legal databases of the governments of the United Kingdom, Australia, Ireland, New Zealand, and other countries, provide first-person accounts of Libyans who supported Gaddafi and found themselves in danger once his regime fell. Pennell’s assessment of these records is that, in order to ensure their safety for future studies of the Libyan revolution, asylum records should be archived in a central place, and the Libya Uprising Archive needs a more permanent home. Due to the inherent instability of digital records, these lessons can be applied to all digital-born content; as Pennell reminds us, “what is born digital may die digital as well, and the circumstances of both birth and death are themselves part of the historical momentum” (p. 71).
Rhodes lays the groundwork for libraries to incorporate living archives into displays, using the New School’s living archives project about the Occupy Wall Street movement, #searchunderoccupy. What makes a “living” archive, she explains, is that content is curated while an event is occurring, members of the community are key contributors, and the collection maintains fluidity in order to respond to changing events. She argues that OWS was ideally suited for this type of project because of the massive amounts of digital data it spawned, including social media posts, videos, survey data, and more, and many students at the New School participated in the demonstrations. She concludes that libraries, as trusted institutions in their communities, are uniquely positioned to support these types of projects and offer support and technical assistance. This article provides several innovative ideas the New School employed that may be of interest to libraries and archives curating displays of social movements.
Rollason-Cass and Reed demonstrate the importance of fluidity in living archives by tracing the evolution of the #blacklivesmatter Web Archive. What began as an attempt to collect digital data created by the protests in Ferguson, Missouri in 2014, evolved to include other protests against police shootings, and eventually adopted the name “#blacklivesmatter.” The authors show the importance of web archives by examining the availability of sites in the #occupy web archive; of 582 #occupy movement websites archived in 2011 – 2012, only 41% were still available live in 2015. They argue that technical specifications must be documented with web captures because websites may be incomplete depending on technology used to capture, i.e., embedded video may not appear in archive. However, they conclude that archives of this type will be incomplete until there is a comprehensive and permanent way to collect this digital data, as most of these living archives are currently reactionary and created as the events progress.
Schafer et al. discuss the various types of web archiving undertaken by the French web legal deposit at the National Library of France (BNF) and the National Audio Visual Institute (Ina) in the wake of the 2015 and 2016 terrorist attacks. Using emergency collection methods, the institutions archived relevant websites and Twitter hashtags such as #jesuischarlie (‘I am Charlie,’ a reference to the attack on the Charlie Hebdo building). The authors highlight the issues involved in these projects, including volume of data, definition of parameters, different languages, and when to stop collecting. Additionally, they point out that counter-discourse must be identified and collected if it is to be used by researchers (for example, the hashtag #jenesuispascharlie, or ‘I am not Charlie’), and this is usually not done in real time. They stress that appearance of web pages and social media posts (such as use of emojis) should be maintained, and that the context and importance of this type of content in the time it is collected needs to be noted for future scholars. They provide several excellent examples of analytic tools provided by the archives to researchers. This article offers a glimpse into a well-rounded archive that has continued to evolve and analyze data in order to make its information valuable to academics.
Uimonen discusses the digital archiving of what she calls “hashtag visuality,” when the line between text and image in a social media post is blurred due to the use of an iconic hashtag such as #metoo. Responding to events in real time, as when the Nordic Museum requested submissions on their website asking for stories and digital images from those who took part in the #metoo movement, can lead to “unruly archives” (p. 10); in this case, they received many more submissions about personal experiences of sexual violence rather than participation in the movement, forcing them to recalibrate the intention of the project. She notes that, in her studies of social media from an anthropologic perspective, as well as her interviews with the director of the Nordic Museum regarding this collection, many hashtag campaigns have a distinct lack of imagery. Rather, the hashtags and the personal narratives illustrate the story. She concludes that this reimagining of social media imagery challenges digital archivists and the way in which they typically curate social media images, but that in cases such as #metoo, they must recognize that the hashtag itself is the iconic image.
Velte explains that archiving activist social media posts is crucial to the narrative of protests and social movements, because they are often used to coordinate events and relay the perspective of activists on the ground. However, there are many challenges associated with social media archives, including the platforms’ privacy policies and terms of service, whether to seek permission from owners of websites and social media accounts before harvesting data, and how to ensure confidentiality to activists. A survey conducted by the author revealed that most archivists collecting this type of digital data find legal and ethical issues to be their biggest concern, particularly whether to get consent from activists before making their social media widely available in archives. Velte concludes that, because there is a lack of professional understanding with these issues, it will require collaboration with researchers and activists in order to determine the best way to curate and display these digital items. She addresses several ethical issues for archivists to consider when curating a social media collection.
