Notes on the “data self”
1. Identity is fundamentally multiple, incomplete, provisional, cyborg, contextually contingent, etc. Deleuze et al. are right about that. (See Turkle’s “Multiple Subjectivity and Virtual Community at the End of the Freudian Century” for grudging admission of this.) But consciousness or subjectivity is unitary (we only think we inhabit any one given identity at a time), which can confuse things. We can end up thinking our self is as unified as seamless and consistent as our conscious subjectivity seems to us to be. This tendency to conflate the two can be exploited for ideological ends, as was the case in the consumer capitalist era, in which the idea that we are all unique individuals who have complete control over our identity (which is discovered and revealed rather than constructed) suited consumerist ideas about power and expression through choice in the marketplace. Consumerism exploited the anxiety about lacking a unified self and offered goods as a means to revealing it, communicating it, healing it. Consumerism was invested in the illusion of the single identity and marketing helped sustain it. (For example, subjects were interpellated by ads addressing them as unique individuals, as Judith Williamson argues in Decoding Advertisements.) If we were in fact cycling through multiple identities, consumerism was there to help mask that fact from ourselves to preserve the illusion of power that stemmed from having autonomy over one’s self-concept: the power of self-actualization through consumption.
2. The emergence of networked sociality and social media has changed the calculations about identity. The capability of social media to document more and more of what a given person does and store that data, make it available for processing and redistribution, makes it harder to sustain the illusion of a unified self. That may seem paradoxical, as all the data gets assigned to the same Profile, unifying it in a sense by default. But what happens is that the activities of different “selves” are forced to cohere, making the body of data incoherent and “inauthentic”. By imposing a single persistent identity on users, social media companies inevitably confront them with their own inconsistencies. The data history inevitably makes us inauthentic in our current behavior from some perspective, by virtue of some regression analysis on the data. When we cycle through our selves, we can’t mask that process from ourselves like we could previously. There is no smoothing over the transitions from selves, just dissonance within the imposed unitary profile of Facebook (or Google+ or whatever service is maintaining a “real names” policy). This causes a crisis of authenticity; we are confronted with how phony we are by the old standards that urged us to see ourselves as in control of who we are, acting on the basis of some prior inalienable sense of what we want and who we really are. We are revealed instead as craven suck-ups doing whatever it takes to get attention, as chameleon-like Zeligs who change when the climate calls for it.
3. This crisis leads to an evolution in subjectivity, prompting the broader acceptance of what I’ve been calling the data self. Within this subjectivity, we don’t worry about our selves being coherent or unified, we don’t worry about authenticity, we don’t worry about coming across as calculating or “interested” in the 18th century sense. The documenting on social media would make it hard to sustain those ideas anyway, and the more we use social media the harder it would become. And post-consumer capitalism has figured out that it wants us using social media productively: it generates cheap innovation, supplies exploitable affective and immaterial labor for nothing, allows for microtargeting with advertising and makes the process of fostering demand easier than it was when every product had to seem to speak to some generic “uniqueness” in isolated individuals. Networked subjects are more flexible, can be adapted “virally” to whatever demands are required of them. So the ideological construct of the self-directed unique individual whose pre-existing unique wants need to be “discovered” and expressed through consumerism is out. Instead we have the well-connected influencer, the flexible, sharing, autoconfessional self who never pauses in disclosing information and thus runs ahead of any need to have consistency imposed. The touted pleasures of this sort of subjectivity, which revels in the possibility of having multiple identities to access, are influence, connection, access, open-endedness, limitless possibility. (This as opposed the the pleasures of the individualist subjectivity — independence, autonomy, uniqueness.) Also pleasing is the possibility to be empirically normal. The old Freudian individual self could aspire to a normality, but was always a unique product of its dysfunctions, which became integral. The data self can achieve normality relative to a statistical average profile based on users who share a similar data set within the various networks.
4. Since authenticity is less pertinent, and one isn’t aspiring to true to some mysterious real self, one is more ready to accept the identities that are dictated to us, the ones we readily construct for ourselves on the basis of other people’s whims and platforms, on whatever trends are floating through the zeitgeist or whatever. Authenticity is rejected in favor of constructedness — we aren’t true to some preexisting self, we actively construct the coolest self out of the relevant data for the given moment, the given context or situation. And we are receptive to data being processed to tell us what we should be, how we should think of ourselves. We tolerate a provisional self as we are in the process of producing the data that will eventually give us an identity repertoire. The real self doesn’t precede the process; instead real selves (real in the sense of being influential in networks) emerge through the process of information processing (sharing, being shared, being on a social graph, having recommendations automated, being processed by algorithms, and so on).
Does this make sense?