(Adapted from my presentation at the Middle East Studies Association (MESA) annual conference)
On October 16 this year, the popular photo hosting service Twitpic announced it would be shutting down. The announcement came after weeks of negotiation, as the company attempted to find a buyer that might save it from obsolescence.
When Twitpic launched in 2008, it quickly became one of the most popular and successful services for sharing photos on Twitter, particularly among citizen journalists documenting live events. But with the announcement of its impending closure, all of those images were in jeopardy.
On the website, Twitpic advised users to download their own photo archives and gave step-by-step directions to aid users in saving and storing their photos before it was too late. Fewer than ten days later, no one—not even the photographers themselves—would have access to the Twitpic records anymore.
The announcement went relatively unnoticed, not least because of previous false alarms. In the end, Twitpic gave us little time to consider what might happen to those photos, where they would go, who owned them, and what we could do about it. Documentary material from nearly every major news event over the past five years was sitting in the digital Twitpic repository. Although most of it had wound up in the public sphere—posted, shared, retweeted—at one point or another, now its survival depended not on the countless digital consumers who had seen and shared the images but on the original producers.
In order to save the public record, it had to be made private again. It had to be actively downloaded from the shared Twitpic servers onto personal computers and storage devices—perhaps to be shared again somewhere new, perhaps to join a burgeoning personal archive reaching unmanageable proportions. We create more now than we can organize and retrieve.
The Twitpic announcement has tangible consequences for how we remember the past on social media because jumbled up with countless cat photos and college selfies are photos from tsunamis, earthquakes, Tahrir Square and the Pearl Roundabout.
The Arab Spring, as it is popularly known, reinforced our collective enchantment with social media as events across the Middle East and North Africa were broadcast on TV and computer screens worldwide. The revolutions were unprecedentedly technologically mediated. They were organized, documented, and transmitted through digital social media, which supported some early and misguided perceptions that social media was the revolution, rather than the longstanding historical cries for political and economic reform.
Still, social media was important, and this is a discussion of how a digital narrative of the Arab Spring is underpinned by the preponderance of digital texts that we—academics, journalists, activists, and others—use to explain it.
In the Twitpic shutdown, some activists, journalists and other Twitter users saw the potential for great losses. They tweeted, texted and posted pleas for Twitpic users to preserve their photos, knowing that at least part of the revolutionary story was stored on Twitpic servers.
The example of Twitpic should make us pause to think—in the midst of unprecedented, vibrant social media production—about what gets lost and forgotten and how the commercial architecture of major internet platforms is implicated in these processes of preservation and annihilation.
Some questions I have begun to ask myself about the digital ‘archive’ of the Arab Spring include:
(1) When a digital platform can no longer host digital ‘artifacts’ for users, where does that archive relocate? What are the political, social, and technological mechanisms by which those public archives are made obsolete, invisible, or inaccessible?
(2) When collections are saved privately, they leave the public domain, if/how/should they be made accessible again? How are the unfolding counter-revolutionary politics changing ethical expectations about how information can and should be shared?
We’re getting to this point now, nearly four years later, when we’re beginning to look retrospectively at the Arab Spring as a historical moment. We’re starting to ask whether these revolutionary moments are over, or whether what we are dealing with is history at all.
It seemed to me that we should interrogate the sources we turn to in order to answer these questions. We should consider the process of history-making and the role that digital artifacts will come to play in this process for all of us — academics, journalists, etc.
Idea for this talk came from my own experiences trying to save, record, archive material for my DPhil, as well as interviews from my field work in Egypt with people who are trying to actively archive their experiences during the revolution. Many people rely on social media repositories of digital artifacts, like photos and videos.
While the unique characteristics of the political and technological moment of revolution present unprecedented opportunities to retain, save, and remember multi-media events, they also encompass a shocking potential for loss, deletion and forgetting. It is so easy to be completely immersed in what is there—photos, eyewitness videos, live tweets, blog posts—that we have to constantly remind ourselves to consider what’s not there.
There is also a tendency to believe that digital technology makes memory easier – and in many ways it does! We could be forgiven for believing that digital artifacts are more permanent, less corruptible, and even perfectly suited for aggregation and categorization. In fact, many platforms help us do this more efficiently through search functions and algorithmic sorting.
But, as I will argue, these assumptions are the product of the same kind of tech-utopianism that called the Egyptian revolution a “Facebook” or “Twitter” revolution. They are reductive. They do not take into account the techno-social relationships that create these kinds of archives — leaving the story to technology first and foremost.
In this discussion, I’m going to reference Egypt primarily, as that’s where I do most of my field work, but this discussion is open to other contexts and ramifications. In my thesis, I examine a number of specific archival ‘projects’ that emerged out of the Egyptian revolutionary moment. They all took particular interest in trying to aggregate the digital artifacts of the revolution, recognizing that this was a uniquely digitally mediated event and that its history would be recorded across both physical and virtual domains. I will not go into these cases in any great detail here, but instead, I am going to draw out several themes that emerged from looking at these archival initiatives and their successes and failure. These issues, I think, are relevant to broader theoretical discussions about digital archives, and they are informed by critical archival studies of analogue archives.
Specifically, I will focus on four themes that have emerged for me from my study of digital archive of the Egyptian revolution: curation, deletion, temporality, and re/over-writing.
When I talk about curation of the digital record, I am talking about issues of authority: who gets to contribute to the archive and what that means about its longevity, its resilience, its biases, and its assumptions.
The curatorial process in the digital archives of revolution is often hard to pinpoint. It’s not always clear who or what is doing the curating in a digital space. But curatorial authority is obscured in ways that differ from the obscurity of authority in analogue archives. In archival projects like the ones I am looking at, someone or a group of individuals has taken on the role of collecting and sorting digital artifacts. But those artifacts might have been produced and indeed pre-sorted, pre-categorized, pre-filtered by another set of creators-slash-curators, either the owners/authors of the content themselves or the technological processes acting on that content on certain platforms.
If you search ‘#EgyptianRevolution’ on Twitter, for instance, to find photos, posts, and videos, aren’t you also finding a combination of what has been filtered for you by users (who have labeled their content already) and by the search function of the platform, which sets its own criteria for relevance and importance?
In my own work on digital archiving of the revolution, it is also apparent that many self-identified curators are themselves political activists. They are also digitally fluent, meaning they have high levels of digital literacy, and indeed, often literacy in both English and Arabic. They see their archival projects as acts of witnessing, much like the photography and filming of Tahrir Square during revolution.
But there are other kinds of ‘curators’ in this space, too. NGOs, libraries, and universities have launched archival projects in Egypt, and, of course, the government has also expressed interest in curating digital artifacts. All of these actors understand — quite explicitly — that curating digital artifacts is also about writing a particular narrative.
I think it’s worth mentioning that I’ve grappled with how to classify these archival initiatives or projects so as to distinguish between their different structures and objectives. We might say, for instance, there are ‘official’ and ‘unofficial’ archives. Organizations and the government classify their collection endeavors as archival projects — meaning, they set out to preserve an historical record of the revolution. But the ‘unofficial’ domain of archiving encompasses pretty much everything and anything else. Arguably, we have all become archivists in our production of digital content for social media.
This abundance of content, the diversity of potential ‘archives’ available for aggregation and curation, preserves the Egyptian revolution from myriad perspectives. This is essentially what was lauded about the mediation of the revolution in the early days of 2011: the eye-witness accounts that challenged ‘official’ narratives. We could see the revolution unfold from the ground. But this diversity of perspective does not negate the need to confront the issue of archive neutrality, as critical archival studies would remind us. All archives — including the burgeoning, distributed social media archive of content from the Egyptian revolution — are created in the dialectic between inclusion and exclusion.
One major source of exclusion when we’re talking about digital archives is the digital divide between those with regular, stable internet access and those without. In Egypt, this divide is highly consequential. In 2011, internet penetration in Egypt was just around 25%. To overcome this exclusion, some activists have actively tried to bridge the online/offline divide by digitizing non-digital artifacts, records, and recollections. I talk more about this in my thesis, but it was important for these citizen-archivists to confront and subvert this exclusionary inequality of access in curating digital archives. It was a sort of quiet acknowledgment that born-digital archives would wind up being inherently partial and — arguably — elite.
Another source of exclusion was what we might call “partitioning” of content across multiple platforms and formats. I refer to the “distributed archive” in my doctoral research because of the way digital artifacts are scattered across different proprietary, personal, and public platforms. A key issue related to curatorial authority here is the question of ownership. Who actually owns the hardware (software, servers, etc.) that artifacts are stored on? When a user relies on a proprietary platform, like Twitpic, to store and help to sort some of their personal digital archive, what happens when the company goes bust? The preponderance of digital artifacts that we all produce everyday as regular digital users means that we increasingly rely on intermediaries to store, sort, and publish our content — those intermediaries introduce new kinds of precarity into our digital repositories.
By ‘deletion,’ I am referring to the active and passive removal of content from a particular archival space. One source of deletion is government censorship. Already, the Egyptian government has blocked and removed certain websites, and as we saw during the 18 days of protest in 2011, the government can even shut the internet down entirely. In 2013, Egypt briefly blocked YouTube over the publication of the “Innocence of Muslims” film, and there is currently a lawsuit pending in a Cairo court on banning Twitter and Facebook in Egypt unless the companies obtain a special license to operate.
In addition, commercial platforms like Facebook and Twitter practice their own forms of intentional and unintentional censorship. The “reporting” feature that allows users to flag inappropriate content often results in the removal of politicized content, resulting in at least temporary censorship of some digital records that certain activists might argue constitute an important archive of political events. There are many examples of this kind of erasure occurring on Syrian revolutionary pages, where graphic images have fallen foul of platform community standards.
By contrast, there are also non-ideological and non-politicized erasure and removal of content due to expiry, where online repositories of digital artifacts simply cease to exist. Because even personal websites and blogs need to be maintained to stay active (and sometimes this involves paying hosting fees or maintaining servers), repositories can disappear.
And of course, there is also self-censorship and active deletion by the creators of digital content. Dramatic changes in the political climate in Egypt and elsewhere have prompted many activists and citizen-journalists to remove content once posted online or to refrain from posting new content.
Another theme that has emerged from looking at archival initiatives seeking to preserve the digital artifacts of revolution is the issue of temporality — the interaction of both offline (political/social/human) time and platform time.
Most social media platforms are built on the logic of the “feed,” in which the most recent posts appear first. Older posts are pushed back, and eventually out of sight. With changes to the application programming interfaces (APIs) used to interact with platforms like Facebook and Twitter, it is becoming more difficult to search and query the past. Older posts (though permanent, in a sense) are made inaccessible without technical knowledge or money to access them through an API. Moreover, older content is replaced quickly with new content; the speed at which this replacement occurs results in certain kinds of erasure — annihilation by sheer quantity.
To search such large stores of data, distributed across the web, we rely on search engines as intermediaries, whose algorithmic processes of sorting also influence what we are likely to turn up.
And the viral effect of certain media content, which receives a large, rapid response, has a tendency to eclipse other content and events. What is the effect of vitality on our definition of (or ascribing of meaning to) certain events?
If everything can potentially be documented and shared, how much do we look for what isn’t?
As we enter the long tail of revolution now, we are seeing more and more counter-revolutionary narratives emerging and competing with revolutionary ones. Misinformation campaigns actively seek to re-write or over-write history, and the curatorial diversity discussed above means that many competing “authorities” can make claims to represent the true or complete record of the revolution — a highly politically contested event.
Changes in the political landscape can seriously affect the trajectory of digital “archives” — and which stakeholders are interested in its preservation. In Egypt, as elsewhere following the Arab Spring, there is widespread suspicion among revolutionary participants of “official” archives and narratives, meaning that energy and interest has turned strongly toward unofficial, personal, and dispersed archival practices. And when archiving is such a political endeavor in the present, the loss of political will or a shift in political energies can halt the archiving process, leading to abandoned projects and defunct platforms.
As new platforms emerge and old platforms die, how can digital archives of the future preserve not only the artifacts (in the form of text or videos or images) but also the context in which they appeared? Does it matter that a video first appeared on Facebook? Well, maybe.
Preserving digital artifacts by format alone would risk flattening this crucial context — the cultures, norms, and logics that influenced its creation, interpretation, or dissemination. These are the kinds of considerations that digital archives of such a highly politicized moment bring to light very starkly.
I’d like to close by pointing out that these potential opportunities for erasure, deletion, or — at its more esoteric — forgetting through the digital archive — are also all potentially very positive and powerful opportunities for change and resistance.
As scholars, we will of course turn to digital artifacts of the Egyptian revolution in making sense of what happened, but in doing so, we run the risk of privileging technologically mediated ‘texts’ and ‘artifacts’ in our analyses. It is obvious why we might rely too heavily on the digital record: it is relatively easy to aggregate, locate, and search. Digital artifacts lend themselves easily to storage and organization.
But we also need to be aware that the overwhelming amount of digital content will populate our archives of the Arab Spring might eclipse other artifacts — offline artifacts. The sheer quantity of public digital content engenders an erasure, a deletion, of that which is not there — that which is few, private, off the network.
Social media content may be abundant, but it is not representative of the full spectrum of participants in or experiences of revolution. The same inequalities that create the digital divide in access between ICT users and non-users act upon the digital texts that tell the story of revolution. Even online, where a seemingly limitless multiplicity of voices can contribute, we must always consider which voices are not heard, which artifacts have been excluded and why.
In Egypt, new state restrictions on expression and protest have not only constrained activity on the streets, they have stifled certain digital practices as well. Digital content has tremendous representational power; it can (and does) define an entire moment. Videos, images, and blog posts streaming out of Tahrir Square helped the entire world make sense of Egypt’s revolutionary moment. The digital age has rendered archiving a practice we must engage with in the present. For the reasons discussed above—the fracturing of curation, the explicit and implicit practices of deletion, the ephemeral nature of online content, and the processes of re- and over-writing—archiving has to be part of the unfolding revolutionary moment. We must consider how to incorporate digital artifacts into our practices of remembering, or like the fallible, perishable, vulnerable documents of the past, this archive may too be lost to new forces of forgetting.