General theory of Inventions - 1. Coordination problem
In the most general terms, the problem of coordination in interaction can be described as the need to align individual perceptions when agents interpret social order. This is a basic condition of any collective action. In this sense it is hardly distinguishable from various formulations of the problem of social order and especially from the well‑known problem of double contingency (Parsons, Luhmann, Garfinkel). Explicit attempts to solve this problem make up a large part of twentieth‑century sociology.
problem of double contingency: we need to explain how interaction becomes coordinated when each participant’s actions depend on the other’s unpredictable alternatives and reactions, which renders the outcome of interaction fundamentally uncertain.
Even so, the problem of coordination in interaction is a more radical and elementary version of the problem of double contingency.
The basic way of tackling double contingency has been to appeal to structures that are given prior to any particular interaction and that make agreement between two individual agents possible. The main route here consisted in describing the conditions under which relatively determinate structures or institutions shape their intervention in the character and logic of communication between Alter and Ego, thereby eliminating the double contingency of communication/interaction. Solutions of this kind can be called external solutions, insofar as they presuppose the existence of an agreed‑upon and relatively determinate unity that lies outside the situation of interaction and removes its original uncertainty.
external solutions: interaction can be coordinated thanks to an instance external to the interaction that does the work of coordination.
The classical alternative to this approach, above all in Garfinkel’s hypothesis of the social plenum, assumes that agents are capable of situationally constructing local ways of reducing uncertainty: such local ways – ethnomethods – trace the contours along which the initial uncertainty of interaction is distributed. However, carrying ethnomethods over into other contexts and scaling them up is itself a separate research problem, one that could only be solved satisfactorily by resorting to hypotheses similar to external solutions, which would in turn undermine the basic tenets of the hypothesis of the social plenum. Such solutions can be called internal solutions.
internal solutions: interaction can be coordinated thanks to the immanent capacity of social agents to create instances of coordination that are internal to their interactions.
problem of internal solutions: transferring and scaling up such agent‑constructed coordination instances into other contexts is problematic.
For a long time social theory paid little attention to a more basic tension which, in its most explicit form, is the problem of coordination in interaction: the need to solve the problem of double contingency precisely in cases where the very existence of mechanisms that under normal conditions reduce the uncertainty of interaction is called into question.
problem of coordination in interaction: we need to explain how interaction becomes coordinated when each participant’s actions depend on the other’s unpredictable alternatives and reactions, and when instances of coordination are no longer able to reduce uncertainty.
Stating the problem this way explicitly rules out external solutions, since now the source of coordination has to be located “inside” the situation of interaction.
condition for solving the problem of coordination in interaction 1: it must be shown how the coordination of interaction results from the actions of agents themselves, rather than from mechanisms external to the interaction that guarantee certainty.
At the same time this makes an internal strategy more demanding, because from a sociological point of view the empirically interesting condition for solving the problem of coordination in interaction will also include the requirement to explain how particular instances of coordination are scaled up beyond a concrete situation, and how they “work” in new contexts and at new orders of magnitude.
condition for solving the problem of coordination in interaction 2: it must be shown how instances of coordination can be scaled up.
This is the central clarification: what is at stake is the scaling and unfolding of particular historical ways of managing uncertainty. In other words, the task is not just to analytically describe the parameters of social interaction, but to investigate historical regimes of managing uncertainty as phenomena of interaction.
A variation of this problem, applied to Bourdieusian crisis theory, was already presented here as an account of the impossibility of a logical transition from individual experiences of the destabilization of coordination instances to an analytics of social crisis. In that text I also proposed a solution linked to the analysis of middle‑range operators as historically specific mechanisms of managing uncertainty. When we move from the local problems of reflexive sociology to the problem of coordination in interaction, the status of these social objects needs to be specified more clearly: what is at issue are historically specific inventions constructed to solve particular problems, endowed with a local rationality tied to the trajectory of the inventor‑group, and easily interiorized as “commonplaces” of social interaction.