Missing the social in on-line social life

Clay Shirky’s 2008 book, Here Comes Everybody, provides a useful comparison for thinking about social life on-line. It’s a scale, essentially of ever denser interaction and commitment from sharing to conversation to collaborating to collective action. Pretty good sociology for media studies, but not quite good enough as sociology. His scale of involvement is more about connections between persons that may be transferred or forged on-line than about connections of persons to an on-line environment. Shirky seems to recognise this in framing IT as permitting people to do things they couldn’t do before – what some call “affordance.”

But the problem is deeper andcan be seen in Shirky’s examples, which are good and plentiful for sharing, conversing, collaborating. Those readily match technologies such as electronic bulletin boards earlier and Flikr/YouTube/Facebook or blogs now for sharing, or listservs earlier and IM-ing now for conversing, file-sharing earlier and wikis now for collaborating. These are also all activities that display what he calls “power law distributions,” elsewhere the so-called 80/20 rule, that a few sites get most of the views, a few users make most of the contributions, a few topics get most of the attention. This is the media-world measure of sociality as traffic; it has all the social density of crowds and falls apart with collective action on-line.

Shirky’s aware of this, and attributes it to the relative recentness of the technologies and of on-line life they sustain; but that’s still the problem. His is a view implicitly that social life grows from primitives, that these primitives are exchanges or transactions of, essentially, selflessness. This is the positive Putnam, the urban migrant who has to build a social life one relationship at a time, aware that this is work and that such relations require constant tending. There is no society out there: the apodictic social actor is a migrant. This is also a very special kind of migrant, not the Mexican or Salvadoran with dense ties to a home village or province, enmeshed in networks of kin and patronage relations who brings, gone north for work not available at home. Instead it is the classic American migrant to the city to make good, to start a new life, leaving behind family, starting anew or starting over, the Self-Made Man. It’s a masculine figure, the relationship-maker, not the female-inflected relationship-maintainer. Interestingly, Shirky’s best example of collective action is women using MeetUp.

What’s missing in this vision of sociality as interaction that grows on-line with tools that facilitate interaction and particularly projects is much sense for the social as phenomena beyond mere aggregation. That is what classical sociological theory held to be the additional data (actually, additional ontology) about social phenomena that are not reducible to facts about individuals. This is basic sociology; but the case can be made empirically. Two examples:

First, a recent story, “Virtual Worlds Get Real About Punishment,” in the Washington Post discusses sims and games where what can only be described as social institutions develop to deal with bad or disruptive behavior. These include courts, even governments, bordering on vigilante committees, that apply and enforce classic social punishments of exclusion – in this case, situation-specific ones of shunning, recognising that boredom is punishment for users seeking engagement, and filth-filters that exclude communictation. Sounds a lot like a small community, because that’s where these social behaviors work. At an individual level, they could be said to be about commitment, but only in the most stripped-down objective (tautological?) sense. They are better seen for what they are, which is institutions or bodies of rules that actors follow not out of choice but to be in the game.

This essential discovery of the social sciences – of social reality or phenomena independent of or not reducible to facts about individuals – is weakly developed in media studies. With its ties to advertising (and to journalism as information service), media studies’ sociality is met as audiences, preferences, profiles that are essentially aggregations, ultimately of choices. In this field, sense for the social in itself is typically framed as critique of macro-systematics: industrial capitalism produces a particular kind of actor, post-industrial capitalism another, to fit their respective “demands” or, less provocatively but no less tautologically, “requirements” or what engineers might call “system parameters” perhaps. The problem with this division of interpretive labor is that microsociology makes actors seem like schemers while macrosociology makes them seem like dopes (“over-socialized” as a classic sociology article put it) and so comes across as critical, and negative at that.

The empirical shortcoming of this division of interpretive labor is to leave out the middle range social phenomena where actors meet social phenomena that are both intractable and optional: reference groups, for instance, networks, institutions that are there without us. Where do we find this on-line? One answer is in “meta-data,” or data about data, what in anthropology would be called a classification. In computer programming, these would be field names. Meta-data abound in detritus of on-line activities, computer searches, links, and other connectives; cookies report meta-data such as sites visited, for instance. These are user-contributed content but not user-defined content.

The anthropologist Michael Wesch, doing “digital ethnography,” zeros in on such meta-data of what he calls “user-generated organization” or, with specific reference to YouTube videos (but could also be applied to blogs), “user-generated distribution” (An Anthropological Introduction to YouTube). This is real social science: it looks beyond the media studies focus on content, and the notion of “user-generated content,” and also beyond interaction, to to their organisational contexts or frames. He begins with remix as an example, tracing the spread of a bit of music that included various re-uses and re-mixtures with other music, visuals, stories, relationships. Those traces, which he was able to retrieve partly from computer meta-data and partly thorugh off-line ethnography, are also “user-generated content” but what is user-generated is not content like the videos themselves but connections: organization and distribution.

He calls this a “community,” but it is a weak sort – objectively a community in the sense of a bunch of people interacting, but subjectively not quite up to a village or tribe, such as he studied in New Guinea. But New Guinea is a good place to start for such communities, which there are notoriously built out as exchange systems dominated by “big men.” But these are systems independent of the actors, whose actions a system frames and toward which those actions – gaining fame, establishing identity as an exemplary actor – are directed. And this is good enough for on-line social life, which resembles nothing so much as a potlatch or the classic Melanesian kula system.

Kula was – is – a system of interisland trade in ritual objects by the region’s “biggest” men. It is also much more, as Marcel Mauss pointed out: kula covers other exchanges and gives them meaning. Mauss recast the whole (including the littler exchanges of lesser valuables down to mundane utilitarian objects and produce) as a vast scheme of relations, of which kula was the most solemn version, gathering up and giving meaning to others and at once political, economic, domestic, religious.

YouTube is a lot like this. The motives of individual content contributors are various but all are in that context expressions (mostly of identity) and action (exchange within a social circle), including their extension. The key meta-datum, and one underdeveloped in Shirky’s scale of social commitment, is reference groups: for whom does one perform/post? Usually to less than a hundred people, Wesch suggests. (While he doesn’t cite evidence for this claim, it accords with the observation of sociologist danah boyd that teens on Facebook or MySpace are not, subjectively, broadcasting to the world but to their specific friends also there.) In other words, this is not quite the intentional “community” Shirky imagines as united by or dedicated to collective action (a sort of super-committed crowd, but nevertheless a crowd). It is a community in the sense of displaying a set of institutions not reducible to individual motives or intentions but, instead, shaping those.

Anthropologists have liked these sorts of communities, for they can be found everywhere. They extensively populate the range between crowds, on the one hand, and on the other the strong-form communities to which sociologists contrasted modern life among strangers – that is, the commune, Gemeinschaft, small town or, in the term made famous by Ervin Goffman “total institutions.” Communities do not have to be corporate to have institutions; but also social actors don’t have to be either dopes encased in a “cake of custom” or free agents “negotiating” contingent individual realities.

In pointing to meta-data for doing “digital ethnography,” Wesch goes beyond modeling social life in terms of the properties of audiences (united by otherwise unacccounted “like-mindedness”) brought by media studies’ updating of crowd sociology (as “smart mobs,” for example), to which the only alternative in critical cultural studies has been macro-systemic features of political economy. Sociologically, he is pointing to the same sort of phenomena that have been conceptualised as “network effects” or, in economists’ terms, as “externalities” or relegated in the language of critical cultural studies to sub-texts. Anthropologists call these, perhaps too blandly, “contexts” that surround and are points of reference conferring meanings that are not, sensu strictu, in the overt texts of behavior. They are not so much “readings” (interpretations of interpretations, in Geertz’s phrase) as they are what actors “read” (interpret) with. Call them “scripts” or “narratives;” as they settle into recognised patterns, they constitute “institutions,” which Durkheim identified as the social phenomena not reducible to individual participation.

There is a lot else that can be said about them – from how they instantiate and serve a larger system, such as in exemplifying the “flexibility” so much more adapted to an economy of out-sourcing and freelancing, to how they appear to respond directly to, or instantiate, “fragmentation” or “decline of meta-narratives.” But it is necessary to appreciate how they are institutoinal forms of on-line social life above and beyond its crowd phenomena, no matter how “smart” the actors in them may be, that at base are Shirky’s data points – and the data from which notions such as “smart mobs” or “network effects” or the “wisdom of crowds” have been generalised. Those generalisations are limited by limiting data to content, overlooking the additional order of data in “user-contributed organisation.” Indeed, as he puts it, what Shirky focuses on, and exclusively, is organization without organizations; but the case is not made that this is what – all – that social software “affords.” I’d guess this is why Shirky doesn’t have much to say about games (or sims), which are explicitly rule-dense

The question for responsible analysis is whether this is all the social organization there is on-line, or whether we just haven’t located and described the organizations. Wesch’s use of meta-data is a step toward resolving whether the matter is analytical (a creature of method) or ontological (a feature of the creature).

Where to go next: (1) are social software institutions? (2) is there cyber-culture?

Explore posts in the same categories: online social life, socialnets