Forget Forgetting: We Can Still Find You. Probably.

Document Actions
Digital embarrassments are not the end of the world; instead, they’re out there and available to be discovered by random searches or profile stalking. But what happens when you actually want to be forgotten?

Kim Lacey

Full disclosure: I’ve posted a ton of stuff online that I regret. (Oh, and you haven’t?) Ok, not racy or inappropriate things, per se, but perhaps a few too many drunken Tweets volunteering myself to be James Franco’s dissertation director. These digital embarrassments are not the end of the world; instead, they’re out there and available to be discovered by random searches or profile stalking. But what happens when you actually want to be forgotten? Do you have the right to renege once you click reply? A recent ruling in the EU thinks so and supports individuals’ requests for privacy. In what’s being called “The Right to be Forgotten,” EU courts have legalized the possibility of deleting oneself from database searches, giving citizens the ability to be remembered as they would like to be instead of how the digital world deems appropriate.

Properly known as the “European Data Protection Law,” it allows one to control what others can access (or more precisely, what one cannot access) in online spaces. According to an article posted on Search Engine Land, the premise for the law:

“[…] arose when a Spanish citizen wanted information on real estate debts (the auction of his house) removed from Google’s Spanish search results. The information was published in a Spanish newspaper. The individual, Mr. Costeja González, asked the newspaper to remove the information because the debt had been resolved and the matter was 16 years old. The paper refused”


González’s request suggests something larger in grand narrative of digital communication—it is the question of whether we have the right to control the ways in which we will be remembered. For González, he wanted outdated material stripped from his record. But what about more severe offenses—should we allow murderers, sex offenders, or felons to digitally restore one’s reputation by omission? More generally, the ways we currently understand forgetting (i.e., the counterpart to biological memory) must be reconsidered if we have the ability to destroy evidence of our past. Such authority over one’s self-presentation (or even self-preservation) is not only worthy of great consideration for anyone who has ever posted anything online, but also for the ways we construct cultural histories.

Ok, let’s say this sounds appetizing and you know of a few items you’d like removed from the internet. How do accomplish this task? Better yet, who’s in charge of “the internet” and thus makes the decision about your presence or absence? In order to request a digital deletion, individuals have to complete a form listing all the URLs requesting to be omitted from searches, identity verification, and at times explanations for the request. For example, Google requests the following for evaluation:

(a) Provide the URL for each link appearing in a Google search for your name that you request to be removed. (The URL can be taken from your browser bar after clicking on the search result in question).

(b) Explain, if not clear, why the linked page is about you (or, if you are submitting this form on behalf of someone else, the person named above).

(c) Explain how this URL in search results is irrelevant, outdated, or otherwise inappropriate. Seems straightforward, but it’s not that easy. Notice this is Google’s request form. Your search results will continue to appear on other search engines like Bing or Yahoo. What’s more is this specific request is only for the EU versions of the searches: your digital mistake might not be among the results of, but will still include the “forgotten” results. In what follows, I trace the conjoined ideas of digital memory and digital forgetting. The EU ruling could ripple throughout other nations, but before that happens, it is helpful to get a grasp on the types of memory and forgetting that are currently built into the systems.

What Are We Talking About, Again?: Exploring Digital Forgetting

Digital forgetting takes into consideration two questions: what do we choose to store and what do we choose to leave out? With digital storage, we have a choice to remember (by requesting our favorite sites be crawled) or to forget (by requesting our slip-ups be deleted). Our biological memory functions a bit differently—sometimes we repress ideas because we want to, and other times we cannot control the forgetting that is brought on by disease, injury, or age. As we begin to think more critically about how memory will be transformed in the digital age, we must also remember to think about how quickly ideas can be “forgotten” and deleted.

Ancient memory networks (shout out to Plato and crew!) were personalized—one orator would create one effective memory system to store, locate, and recall details. Socrates’ distaste of externalization represents a move away from the nature of rhetoric, summed up later in Aristotle’s classic definition of Rhetoric: “the faculty of observing in any given case the available means of persuasion” (Rhetoric 6). This early shift towards the externalization of memory recognizes some of the fears that arise from transferring our biological processes into non-biological mechanisms, one that has not disappeared even in this current digitally-dependent age. For early rhetoricians, being prepared for oration meant collecting and having argumentation organized and readied for any given situation. One of the means to accomplish effective oration was to exercise the memory so that during a speech, one could quickly employ a number of memorized techniques as evidence. The ability to efficiently recall these effective approaches allowed the orator to appear natural; rather than having to create appropriate evidence in the moment, the orator would simply retrieve what he already held in his memory. This ‘storehouse’ of techniques is similar to today’s ideas of external memory storage—digital spaces allow individuals to store an infinite amount of information, while mundane activities, such as Googling, fill the gaps in one’s knowledge. With Google aiding our memory to the point where no one needs to “forget” anything, the modern emphasis on externalized memory sharpens our mental reflexes so that the information we need to retrieve can occur more rapidly. Even though it is too reductive to claim that information technologies simply offer ‘more’ in terms of storing available means, it is helpful to characterize the available means as the method by which the information can be sorted, tagged, and informed by the user in order to be recalled quickly and utilized effectively. I am suggesting that the crucial issue with contemporary memory is not how much we can store, but rather the structures and networks we develop in order to process and use that information efficiently.

Today there is one memory network, and it must be accessible and searchable by many. There is one critical distinction between the two that will drive this essay: the type of system that is created to maintain memory and promote what information architect Peter Morville calls “findability.” Morville defines findability in three ways: “a. the quality of being locatable or navigable; b. the degree to which a particular object is easy to discover or locate; c. the degree to which a system or environment supports navigation and retrieval” (4). I suggest that contemporary memory networks both benefit and falter from a wide user base. On the one hand, the growing number of users allows people to create their own memory spaces within the larger system. On the other hand, the downfall is that at the same time users input material to be stored or shared, the network itself also creates and outputs new information about the recently uploaded data. By participating in what communication theorist Mark Andrejevic calls the digital enclosure, “the creation of an interactive realm wherein every action and transaction generates information about itself,” uploaded digital memories both retain information about personally relevant events while at the same time adding to the existing body of knowledge by creating new information in which others will participate (2).

The critical issue with contemporary memory is quickly becoming how our memories are recollected. If we do not “tag” our pictures on Instagram, others will be unable to find them through search words alone. With digital memory, we must be as aware of the programs we create to organize and access our stored material as we are with the content we place in external systems. Therefore, control of the networks (even small, personalized networks like tagging systems) becomes critical to our understanding of how memory functions. More and more, digital memory is replacing natural, biological memory. As a result, by relinquishing control of our biological systems to the techno-neurological swarm, we must introduce some form of control apparatus or system in order to retain access to our own memories. Even if our biological memory can at times seem inaccessible, with thoughts buried so deeply in our subconscious that we do not know they exist, we have the capability to overcome the limits of biological mechanisms through digital organizational systems and control units.

Consider the Wayback Machine. As the project operated by the Internet Archive which strives to collect everything that has ever appeared on the internet, The Wayback Machine desires to hold onto everything (this might be the first digitally diagnosed case of Obsessive Compulsive Disorder). While “hoarding” has recently become grossly intriguing because of reality television shows, digital hoarding is popular for a different reason. It is exciting to think that everything we have seen on the internet can be viewed again, that we have a place to return to, and that our memory will be saved for us.

There are, however, many skeptics of digital preservation. Digital doomsayers, like Viktor Mayer-Schoenberger cringe at the idea of saving everything and suggest that we purge our hard drives on a regular basis. The downside of such massive collections is organization and ease of locating the desire information. If we cannot locate material quickly and regularly, then it might as well be forgotten. One of the issues with such massive collections of stuff is not how much we can store, but the structures we must develop in order to process and use the information efficiently. In the first issue of the journal Memory Studies, Paul Connerton suggested “Seven Types of Forgetting” in his article of the same name. The seven types all describe forgetting as a purposeful action occurring as a result of the individual, as opposed to involuntary forgetting as a result of a disorder or brain injury. Connerton contends that while we usually view memory as a triumph, forgetting is most often a type of failure: forgetting where we placed our keys, forgetting we have a committee meeting, forgetting to grab milk on multiple trips to the grocery store. He argues that no matter what kind of forgetting occurs, they all have one similar feature: “they imply an obligation on my part to remember something and my failure to discharge that obligation” (59). Subsequently, digital memory shifts the obligation away from the user into the system itself—the failure of forgetting is no longer mine, but is instead a lapse of technological memory. But aren’t we the ones who actually place our faith and our memories in digital systems like The Wayback Machine? Who’s to blame then? Characterized by its functionality, external memory’s most striking property is not merely its ability to supplement biological memory, but that it “remembers” events exactly as they occurred and retains knowledge without the fear of our brain failing us, or forgetting.

However, becoming digitally forgotten is not a result of anti-participatory, Luddite behavior. Aside from requesting deletion, another way of being digitally forgotten is the result of being left behind on what Wired magazine editor-in-chief Chris Anderson characterizes with the term “the long tail.” Anderson uses the long tail as an optimistic business model—it is beneficial to offer more products to fewer people than it is to have fewer products available to everyone. Statistically speaking, the long tail represents the probability of a select few popular “hits” receiving the most attention while the remainder are forgotten about. The long tail is derived from the image of these results: the most popular and frequent hits have very high points on the y-axis of a graph and the numerous other results trail out in a thin line across the x-axis. The results along the x-axis look like “a long tail”—think of a rat’s skinny, long tail. Digital forgetting occurs because the majority of information is lost among the long tail results, thus a select few popular ideas remain at the helm of digital memory. The higher the search ranking, the more preferential treatment the result will receive. Unfortunately, the opposite holds true, too—because there are so many results tucked away in the long tail, there are only a selected handful of results that receive regular attention for any given search term combination.

Being forgotten in the long tail arouses some interesting concerns with digital rhetorical memory, in particular the availability and accessibility of the means of persuasion. SEO not only increases the likelihood of certain links appearing more prominently in a search, but because these favored results have been paid for, less important (and arguably less funded) hits will never receive as much attention as their optimized counterparts. For memory, SEO has the capability to force certain memories into the long tail. Smaller events, or even alternative perspectives, could be suffocated by popular search terms or corporate sponsorship of a single track of memory. Take, for example, online shopping trend suggestions like Amazon or iTunes recommendations. These recommendation-bots are some of the most popular examples (and sometimes the most off-putting because of their precision) of SEO and the digital enclosure. The more popular an item becomes, the more likely it is to appear in one of the first pages of suggested search results, leaving the least popular links to be lost and forgotten among the results. Our participation in the digital enclosure happens automatically whenever we send an e-mail, purchase an item on Amazon, or place money in our mutual funds. The information we are providing is collected in order to “serve us better”—or, at least, to increase our desire to purchase a tailor-made (optimized) suggestion based on the trail of digital information we leave behind. All of our digital communication is tracked, sorted, and compiled to automatically offer ‘premium’ choices based on our preferences. The digital enclosure works by the user providing information in order to receive (without asking) more information about herself. There is one catch—as Andrejevic points out, the digital enclosure is asymmetrical. As users, we provide the content but we do not have access to exactly what content is being collected or, for that matter, how the information is being put to use (41).

But another, more symmetrical, way to consider digital enclosures and memory collectives is via folksonomies. A folksonomy is a digital system that allows users to sort, identify, and group similar items in order to facilitate findability. There are several well-known examples and many collaborative websites, such as Twitter, Delicious, or Instagram, that use a well-known folksonomy—hashtagging. On the social bookmarking site Delicious, users have the option of adding a “tag,” a short (often one word) description of the article’s topic. Delicious often provides “suggested tags”—frequently used words in the article itself or other popular tags used by members who have saved the same article. Delicious users can also embed “tag clouds” in their blogs or personal web pages. Tag clouds are a collection of word tags that represent frequency and popularity of a topic by the size of the word itself—the larger the font, the more popular the topic. Contributing to a tag cloud or other tagging system is not an isolated process, and participants must select popular, even generalized, reference tags in order for the saved links to be connected to those similarly tagged. In the article “‘Folksonomy’ and the Restructuring of Writing Space,” Jodie Nicotra notes that in order to locate related tags, as well as have our saved links recognized by others, we must use the popular tags rather than ones personally meaningful to us. For instance, Nicotra states that Delicious might suggest the tag “marathon” for an article about training for a race (W272). Using the tag “marathon” will automatically link it to others searching for articles about marathons or running (W272). If, however, I tag it with “26.2 mile run,” then the likelihood of someone using that obscure tag decreases, and the memory is not as accessible as it could be with a popular tag like “marathon” (W272). How might memory studies falter because of standardized participation, like using popular, recommended tags? Our memories must be stored in a very specific fashion in order to be recalled, linked to, or recommended later. If I use “26.2 mile run” rather than “marathon,” then the memory runs the risk of being forgotten and left in the long tail. However, selecting popular tags over less useful ones increases the possibility of retrieval while also fortifying the continued recollection of properly classified digital memories.

As a result, folksonomies are significant to memory studies for a few reasons. First, Nicotra indicates that a folksonomy, “moves away from traditional hierarchies and classification systems” (W260). Rather than relying on SEO or corporatized sponsored results, a folksonomy is a collectively organized system that has been compiled entirely by the users’ identification tags. By “disrupting the idea of single authorship,” a folksonomy shows “how multiple, collective subjectivities ‘write,’ enabling possibilities for configurations and systems to emerge as a result of activity of the so-called hive mind that could not have been anticipated or conceived of by an individual author working alone” (Nicotra W260). Next, because a folksonomy can only occur by purposeful, collaborative effort (think of the “marathon” v. “26.2 mile run” example), these collections reshape the ways we think about writing, too: “‘writing” […] certainly has an expansive, performative aspect—not only is it ‘shared,’ as in produced by multiple users, but it is conceived of as the building of a space rather than the production of a text (Nicotra W263). Finally, all objects must contain properties or characteristics that promote findability. Nicotra rightly argues that, “if your ideas of what the site is about don’t match up to others’ ideas, it is essentially useless, a rhetorical failure (W266, emphasis mine). A rhetorical memory failure is nearly identical—if a site never receives attention and is lost in the long tail, it is a rhetorical memory failure because of the doubtful probability of being recollected.

Neurologist and author Oliver Sacks notes that neurology’s favorite word is “deficit,” largely because neurological research always involves the loss of something – loss of memory, loss of speech, loss of identity (3). Memory, as we know it, is always subjected to the ultimate deficit—“forgetting”. Biologically, our personal affective responses will limit, or censor, what we can recall. And although we believed digital memory did not speak to the limits of time, the right to be forgotten has shaken up that conversation. We want our fifteen minutes to expire when we say they do.

Kim Lacey is Assistant Professor, English at Saginaw Valley State University. Graphics from Rosicrucian advertisements in Mechanix Illustrated magazine, circa 1950.

Copyright © Kim Lacey. All rights reserved.

Personal tools