Rudderless
Before I had a computer, I used to write a lot of personal letters on a typewriter — windy, self-indulgent monologues that rarely took much consideration of who they were addressed to. It was too hard to delete anything, so I would have to just write my way out of whatever half-baked idea I started to describe. I would get carried along instead, and this took me toward what I thought was a more revealing version of myself, as though self-expression were the point. My correspondents probably had different ideas.
When I would be typing out a letter, the more I found myself writing to a specific person, the more the words felt diminished, confused — they ceased to seem self-expressive to me and instead became an exercise in self-performance, or flattery, or condescension. It wasn't that I thought I had some authentic self inside that I had to express in a certain way; it was more that when I was addressing no one in particular, I would discover some self emerging that seemed real. I don't know where that self came from.
Ordinarily I don't think of myself as a "self"; there is just a bunch of anxious reactions and maneuvers. When I use social media, that feeling often intensifies. I feel goaded into having reactions — into selecting the emoji face that best captures my feelings. Staring at the cartoon faces I suddenly feel as though I'd rather have no feelings at all than represent them with those icons.
Social media are sometimes touted as vehicles of self-expression, but they seem more like reaction accumulators. We might take these reactions as representative of our identity, but then all the different platforms extract different reactions, different forms of "expression." The platforms all configure the audience we can reach through them in different ways, which prompts different sorts of selves at different occasions.
Nonetheless, data brokers turn all the different reactions, the different selves, on these platforms into a consolidated identity, aggregating all the data together and drawing conclusions about what it all means and who we really are. This procedure works against our lived experience of improvising for the media we're using at a particular time. All the different media we use and consume fractures and alienates identity, splits our sense of focus, and lets us pass through our days in a fog of distraction, mostly unaware of our scattered selves.
The underlying data aggregation, meanwhile, assembles a singular identity for me, out of data that I can't fully access or change, only add to. What I knowingly add is combined with the data trails I unwittingly generate, gathered through various forms of surveillance tracking my location, my purchases, my friends, the tracks of my eyeballs on a screen. This permanent record — my secret identity — then shapes how information is presented to me and what permissions are granted to me. It recommends things to me. It autocompletes me.
Most of all, this aggregated identity calls my own interior sense of self into question, seeming to refute it or work against it. Who data brokers think I am, and how that affects what I can know, seems to precede who I can think I am or what I can say.
Sometimes it feels good to be guided in this way, to have cues loom in front of you, like when you're playing a video game and a ghostly arrow superimposes itself over the action, telling you which way to go to advance the story. Maybe we should be reassured by the idea that there might be an algorithmic formula to identity — that there could be an external, empirical answer to the question of ourselves. Maybe all that surveillance really is a surreptitious form of care, with the goal of gathering the crucial data needed to complete the identity formula. But of course, the picture is never complete — more data produces a stronger need to interpret the data, which in turn produces a hunger for more data. Kate Crawford has called this "surveillant anxiety."
As reassuring as it sometimes is to be force-fed videos I actually want to watch next, or to be recommend books I actually end up buying, the existence of my external secret identity also leads easily to paranoia: Why won't they tell me what they think of me? What demographics they are lumping me into? Why are they using information about me against me, to change me into a person who is more easily controlled or more easily manipulated or more easily sold on things?
The existence of this shadow self steeps me in a positivist logic against my will: It forces me to wonder if my data and the subsequent analysis of it is more "real" than my inner thoughts about things, which are just epiphenomena. Maybe autocomplete is true, and what I thought I wanted to search just a pretense, a pose. Maybe autocomplete isn't the guess at what I want; maybe I am guessing in my head at what it would behoove me to want. Maybe what I think about myself is merely provisional — ironically too subjective, too self-interested, to be trusted — and what they know about me is the truth.
As a seemingly more concrete and unified identity is developed outside my control, my personal experience of identity becomes more tenuous, doubtful, divided. It seems more "ontologically secure" outside of my head in the data clouds then it does in my own thinking about myself, which seems more other-dependent than ever. Maybe it is because ubiquitous surveillance has turned my entire life into a personal letter addressed to no one in particular.
***
These sorts of concerns about fractured identity made me receptive to R.D. Laing's The Divided Self (1960). That is, when I saw it on the 50¢ cart at Housing Works, I bought it. In the book, Laing offers a definition of "ontological insecurity" that seems to correspond to way I was describing social media above:
The individual in the ordinary circumstances of living may feel more unreal than real ... precariously differentiated from the rest of the world, so that his identity and autonomy are always in question. He may lack the experience of his own temporal continuity. He may not possess an overriding sense of personal consistency or cohesiveness. He may feel more insubstantial than substantial, and unable to assume that the stuff he is made of is genuine, good, valuable. And he may feel his self as partially forced from his body.
With that definition in mind, it seems that social media are an engine of ontological insecurity, even if they sometimes masquerades as its cure, as a fount of self-expression and "meaningful interaction." The creation of an identity archive would seem to ground the self, but it merely creates an incomplete and inadequate double — a “self partially forced from the body” — over which one has little control. It can be out there popping up in conversations I don't even know about, like Dostoyevsky's Double. Whether that "stuff" I am made of in social media feels "genuine, good, valuable" to me depends on how others (and algorithms) respond to it, what sort of outputs it ultimately generates. The metrics attached to social behavior online provide a score for how substantial my self should feel to me.
The information in the archive of our personality can be accessed in ways that we never would be able to access one another in life — details from vastly different times can be juxtaposed, contradictions that would be trivial in the ordinary flow of time can be made to seem damning. The archive has the always dormant possibility of confronting me with my incohesiveness, my lack of consistency, my incoherence. But of course, I can always turn to the immediacy of social media to assuage these feelings. I can post something that feels corrective, I can accumulate tokens that remind me I exist.
This means social media set up a cycle of dependency: They systematically impose a sense of insubstantiality on users, which generates a more intense demand for the small reassurances they also afford. But the reassurance doesn't last and can always be put in an unflattering perspective: someone else's selfie got more likes than mine. That picture was before, what about now?
For the ontologically insecure, according to Laing, “the world of his experience comes to be one he can no longer share with other people.” If that is true, then in social media terms, it means that “sharing” on platforms increases as one comes to feel that no one shares their world. The more I post, the more isolated I feel. The more I mediate my experience to offer it you, the more I make concrete my feeling that you don’t know or share what I experience or even acknowledge it, and that I have to keep shoving examples of it at you. Algorithmic sorting makes this drive more intense. It"s the inverse of Crawford's surveillant anxiety: a kind of exhibitionist anxiety that no matter how much one reveals, it is insufficient to convince anyone you are real.
Social media invert what they make explicit: They turn identity into incoherence by flattening what we do into a false archival unity. They purport to contain identity, but they render it into a blinking cursor, an untaken photo, a vacuum that never fills, no matter how much is poured in. They turn sharing into isolation, making the attention of others measurable, storable, transferable, something that seems to come only at someone else’s expense, obscuring the idea that attention can vary in form and intensity, that it can be given without being surrendered, that it can harmonize with the attention of others into something immeasurably greater.
Laing suggests that "what the individual variously terms his ‘own’ ‘inner’ ‘true’ ‘real’ self is experienced as divorced from all activity that is observable by another.” That is, it is the opposite of the social media landscape, or the quasi-empirical data self. Anything that can be measured or tracked won't feel real to me, part of my real self. The more I post online, the less real I'll feel.
One of my basic intuitions about social media is that they let us make things “observable” about ourselves, which means that we get to discard them from being a part of our "true" self. Boris Groys argues something similar in In the Flow: "The Internet is a space in which the subject is originally constituted as something transparent, observable," he writes, "only afterwards does he or she take steps to be technically protected, to conceal the originally revealed secret." We try to negotiate this subjectivation, revise it by reclaiming some of original opacity.
This suggests another spin to exhibitionist anxiety: the more we reveal about ourselves, the more we discover what it is we haven't confessed yet, how each revelation opens new avenues for further exposure. If Laing is right, this stuff we come to realize we haven't shared yet will feel like our inner true real self. Confessing structures other self-observations as unconfessed, as secrets, as the truth. Posting things online allows us to discover and feel that true self, which consists precisely in what we haven't said. The more one uses social media, the sharper the image of the self comes into focus, not as an amalgam of what is shared but as a remainder, a residue. We speak not because we have something to say but because we want know ourselves as something unspeakable.
The “self-expression” online, then, is more of a self-purgation, to see what is truly inexpressible (and thus essential) about the self, the part that can’t be separated from your embodiment. This inverts the commonsense idea that we want people to acknowledge and like things we post because we see them as representations of ourselves — as standing in for us. Instead, when people “observe” our online behavior, it confirms that material has been seen and expelled from the "real" self, which doesn’t circulate. (When a parent posts a picture of their child, the child’s circulating image becomes a kind of talisman warding off the sense that the parent’s self is leaking away, becoming superseded.)
But when a second self, a more objective-seeming self that is nonetheless foreign to us, is made out of our data, it can replace the need for that process of revealing oneself to get to the inexpressible essence. Rather than post things and thereby purge them, we can engage with our mysterious data double and let the algorithms continue to produce new and surprising truths about ourselves. We become real to ourselves because this feed of content about us comes not from within but without: it seems like an objective fact; it is as real to us as other people are. So rather than seek a true self within, in aspects of our consciousness that seem inexpressible, we can seek it outside ourselves, in aspects of our data that can be recombined in ways that are novel to us — in ways that we didn't intend, and which therefore must be authentic and spontaneous.
***
When I started writing letters on a word processor, I started to cut and paste passages from a letter to one person and send them to someone else. This seemed wrong, but rather than resist that possibility I stopped writing personal letters altogether. Email started to seem more appropriate anyway; it couldn't fail to convey immediacy back then, which seemed like the most important quality you could convey to someone. The content was superfluous.
If I had access to all the personal letters I wrote in the past, I have no doubt I would have re-edited them, pillaged them for other purposes, made essays out of them. But the fact that I don't have access to them anymore makes them feel inherently embarrassing. Back then, that danger never would have occurred to me; relinquishing control was simply part of the bargain involved in having an audience for my tedious textual performances. Now I feel like I never really have to let anything go like that. I can post something, go back and change it, and eventually delete it. They don't arrive at any final state of completion; in my mind, they just perpetuate as a draft. This lets me feel like I have been the only one reading after all, that my yearning for a particular audience was just a phase.