Archiving Sensitive Material

In his analysis of archives, David Zeitlyn underlines the tension which exists between what he calls “sources intended to inform” and “sources never intended to be part of the historical record, but which were nonetheless archived” (Zeitlyn 463). Archiving materials considered to be “sensitive” — that is to say, materials whose contents or themes may compromise their creators or subjects — is challenging from an ethical standpoint, even more so in the context of a digital archive and its double-edged potential for increased accessibility, exhaustiveness, and dissemination. The addition of unofficial, previously unpublished, widely unknown, or personal materials to an archive might be a very important step towards the elaboration of a multifaceted and inclusive point of view, as well as offer invaluable research material. However, it also generates a number of ethical issues in regards to the appropriateness and quality, contemporary/future relevance and accuracy, as well as the potential compromising nature and unforeseen consequences of publishing these items. Such concerns are heightened when one considers that a given community involves members who may be more or less marginalized and more or less vulnerable based on any number of stable and unstable factors including one’s sexuality, race, gender, class, among other matters.

Archiving unpublished material presents an ethical challenge because this content may reflect ideas and points of view that the original creator no longer holds or would not offer without additional context. Carusi and Jirotka explore a comparable matter when they discuss the ethical issues raised by “lost” materials that are rediscovered after a researcher’s disappearance: the sensitivity of these materials and/or the ramifications of their inclusion to an archive may not be well understood or even suspected. It then falls to the archivist to decide whether to use that material or not, by estimating the ethical issues such a publication might entail, and whether it might cause damage to the creator and/or subjects of the item. This leaves a lot of room for miscalculation. Carusi and Jirotka’s solution to this issue is to suggest that researchers leave clear instructions as to what is to be done with their research, whether published or not, such as plans for storage or ultimate disposal (Carusi & Jirotka 7).

Obviously, things are rarely this simple in reality. This is a good reason why present permission should be sought from creators before publishing works in a public digital archive.

A digital archive provides easier, faster, and more direct access to potentially sensitive materials, and may thus lead to the unexpected rise in popularity of an item which was not initially intended for a large public. Nina Rao depicts such an instance while discussing the “unforeseeable uses of all kinds of footage” stored in moving image archives, by using the example of a documentary making extensive use of a family’s home movies: some of the materials used to produce that documentary were also employed against the interests of the family which it depicted, in a court trial, “to support the idea that they were crazy” (Rao 109).

The availability of such material to the public or institutions through digital archives allows for uses that might cause unanticipated moral, social, or legal damages to its subjects or creators. Permission is important; informed permission that explains such risks is even more valuable.

The Zine Librarian Code of Ethics stresses the importance of this very issue in the production of zine archives, as the ability of zines “to give voice to those who are not traditionally represented in libraries/archives” also puts at risk creators who discuss topics considered as sensitive such as sexuality or politics. As underlined by the authors, zinesters “create their work without thought to their work ending up in institutions or being read by large numbers of people.” The later inclusion of zines containing such sensitive material in archives connected to either governmental or academic institutions, for instance, may have damaging consequences for zinesters whose identity was either never concealed or easily guessable.

The importance of establishing a “right of refusal” for the creators and/or subjects of an item that an archive proposes to obtain and make accessible then seems essential for both the preservation of that individual or group’s privacy, but also potentially for the protection of their legal interests and/or social status. Furthermore, considering the fact that societal values, an individual or group’s status, and an item’s impact may vary or fluctuate, it may be necessary for that “right of refusal” to remain active, even after the item’s integration to the archive, for that protection to remain effective. Such a measure, however, presupposes either that the creator/subject of the item will stay updated as to the status of that item, or accountability on the part of the archive in tracking them down to renew their consent.

Both options present their own challenges, ranging from time-management issues (can one expect the creators/subjects to remember the work’s existence, and reconsider or even know its potential impact?), to authority issues (should the right of refusal be given to the creator, the subject, the owner of the rights on that item, or a combination of two or three of these entities?) as well as lineage issues (Can the descendants of a creator/subject still claim this right of refusal? What if several descendants disagree on what should be done? Can the archivists be expected to track descendants down?).

A further option is to limit access to these materials to times when their sensitivity has either decreased or entirely ceased. This approach is in accordance to the “passage of time principle”, defined by Brown as “assum[ing] that reasons for and appropriateness of denying access diminish over time, [and] the public interest […] increases over time” (McKee & Porter 71). In other words, it presupposes that the sensitivity of some issues/items is greater when still relatively recent —in particular when the people concerned, whether authors or subjects, are still living— but decreases as time passes. Many archives thus implement measures to seal access to the personal documents of people who have just died for fifty to seventy years (McKee & Porter 72). Who is granted the authority to make that decision? Does the sensitivity of the document really die with its creator or subject? Can’t it have repercussions on their descendants, their community, or their legacy, and if it does, shouldn’t they themselves be asked for permission?

An example of the particular care that may be given to the question of the inclusion and accessibility of sensitive contents (in this particular case, explicit material) in the elaboration of an archive is found in the Organization for Transformative Works (OTW)’s Archive of Our Own, which hosts and provides access to fan-generated works, such as fanfiction. The archive commits to an inclusive stance, stating in its terms of service that it “will not remove content from the Archive because it contains explicit material, as long as it doesn’t violate any other parts of the content policy (e.g, the harassment policy).” This stance responds to heated debates as to what is and should be considered acceptable for publication within a fanworks archive. Indeed, the publication of fanworks, especially when explicit or graphic, has raised a number of questions in regards to the ethical or even legal nature of certain works containing sensitive material. As such, debates are still ongoing in regards to the appropriateness and/or legality of posting fictional works (whether texts or illustrations) depicting such topics as rape, incest, or relationships between minors and adults.

The complexity of the issue lies in the unreliable nature of community policing, as the cultural values, interests and positions of the various members of the community contributing to and accessing an open archive may either be in conflict or fluctuate. The archive’s inclusive stance enables the preservation of materials that may be considered inappropriate or worthless by some of its users, while simultaneously working to ensure that such users can avoid material which may cause them distress or offense. Works may thus be marked as sensitive through the use of content warnings (for graphic depictions of violence, major character death, rape/non-consensual sex, or underage sex), content ratings (ranking from “general audiences” to “explicit”), as well as more specific tagging. These warnings, ratings, and tags can be used as filters in an archive search, while a warning is given to users in regards to their potential absence in the archive Terms of Service: “Our policy aims to enable creators to choose appropriate labels or to opt not to use ratings and warnings, with the understanding that some users will avoid unrated or unwarned content.” Contributors are also given the option to “lock” their works and profile pages on the archive, in order to prevent them from appearing on search engine results, as well as on the website itself for any user who is not logged in. Furthermore, if a user feels uncomfortable with their association with the sensitive material they have published or wish to publish on the Archive, the Archive offers two alternatives: total deletion from the archive or “orphaning,” which the Archive’s FAQ depicts as the permanent and irreversible elimination of “all [the user’s] identifying data from the selected work(s).” Orphaning a work thus enables its preservation while dissociating it from its creator.

The principles guiding Archive of Our Own provide a multidimensional approach to the preservation of material that might be considered as sensitive in order to ascertain the welfare and comfort of both the creator and the user. However, it is not entirely foolproof either. The digital medium of the archive allows for the downloading of its contents in .pdf, .mobi, .epub, or .html format, and even through a basic copy-pasting of a work’s contents. As such, even if a creator later decides to orphan or entirely wipe work containing sensitive material off the archive, some users may still possess original copies, as well as copies that contain the identifying information wiped from an “orphaned” version. While the creators who decide to delete their work can attempt to leave instructions for users who possess digital or printed copies to prevent its dissemination by not sharing it, they cannot be given any insurance that their wishes will be respected.

The example of Archive of Our Own’s approach to the preservation and publication of sensitive material in digital archives thus provides answers to some of the ethical issues it raises, as well as opens the door to new issues and questions. An underlying problem in the elaboration of a digital archive is the inability to truly and effectively control the dissemination of sensitive information once it has been uploaded, whether or not it has been deleted from the archive. Current methods often rely on the archivists’ judgement, the creator’s foresight, and the user’s honesty or good intentions, which leaves room for miscalculation, negligence, and potentially damaging consequences, whether accidental or deliberate. While there may be no ideal or foolproof way to both preserve and make accessible sensitive material through digital archives without potential harm to its creators and/or subjects, it seems obvious that any future alternative cannot simply rely on the ethical authority of a single actor —archivist, creator, subject, researcher, institution, or public consumer­— but will depend on an adaptive, flexible, and coordinated approach taking into account the creator’s wishes, the subject’s approval, the archive’s purpose and configuration, the institutions and researcher’s objectives, and the evolution of these actors’ statuses as well as values.

By Anne Claret