Jay L. Lemke
City University of New York
How do moments add up to lives?
How do our shared moments together add up to social life as such?
Every human action, all human activity takes place on one or more characteristic timescales. A heartbeat, a breath, a step, a spoken word takes but a moment; a stroll, a conversation extends over many such moments; and an education or a relationship may be a lifetime project. The great cathedrals of Europe were built over many human lifetimes, and the languages and discourse patterns of our communities have developed over still longer times. And yet a conversation consists of many momentary utterances, a relationship may be built of many strolls and conversations together, a building or a social institution is erected by the sum of many individual actions in a community.
How? How do actions or events on one timescale come to add up to more than just a series of isolated happenings? How does a language emerge from many utterances? How does a community emerge from many people-in-action? On how many different timescales is our social life organized? How does persistent organization on longer timescales constrain the likelihood of events on shorter timescales? How do organizational units and processes on shorter timescales make possible the emergent patternings we recognize at longer timescales?
Why time? Our material world is organized on many scales: space, time, matter, energy, information transfer. In many natural systems there is a strong correlation among these: the quick is also small and light and weak and simple; but in more complex systems, especially those where signs and meaning play a role in behavior and system dynamics, these simple correlations break down.
Classical systems theory is rooted in spatial metaphors and the reductionist project: large systems are to be understood by analyzing them in terms of interactions among smaller component subsystems. Molecules are understood in terms of interactions of atoms, atoms through interactions of smaller particles. Organisms are analyzed by a hierarchy of interactions among units at progressively smaller spatial scales: organ systems, tissues, cells, organelles, and macromolecules. Ecosystems are modeled as interactions among species and abiotic elements; galaxies as interactions among star clusters and individual suns. In all these cases there is a fundamental assumption: units nearer in space are more likely to interact and to interact more strongly (i.e. with greater effect on one another). This assumption imposes a ‘spherical’ topology on the system: relative to any center, items at the same distance scale (i.e. in the same spherical shell) are equally likely to be interaction partners, with the closer ones interacting more and the further ones less.
In many complex systems, however, this assumption fails. Two distant points along the same stream may interact more than two nearer points not linked by the stream. Two distant cells may communicate chemically via the bloodstream; two distant neurons may interact more than closer ones not in the same neural network of pathways, sensitized to the same neurotransmitters or neuromodulators. In a pond or an ocean, two species in the same layer of water, at the same depth may be more likely to interact over wide (horizontal) distances than they are to encounter a species nearer in vertical distance, but separated ecologically by depth-dependent differences of light, temperature, salinity, or pressure. Species roam far in the rainforest canopy without ever venturing a few meters down. In our human ecosocial systems, which are just a specialized kind of ecosystem, people who are linked by the same river, the same railroad, the same phone network, the same chat room on the internet may interact far more than they do with spatially nearer neighbors who are off these social transport and communication networks. In a modern city, spatial proximity may have little relevance to probability or intensity of interaction.
In addition to the ‘spherical’ topology for strength of interaction, there are clearly at least these two others: the laminar topology (horizontal layers) and the network topology (lines of connectivity). All three of these principles are at work in the spatial organization of human ecosocial systems, but I generally agree with Bruno Latour that ‘sociotechnical networks’ are critically important for answering the questions with which I began this article (cf. Latour 1994, 1996; Lemke 1995, 1997). Many people interpret Latour’s arguments, or the somewhat similar arguments proposed by ethnomethodologists (e.g. Garfinkel & Sacks 1970, Schegloff 19..), as leading us to a ‘flat’ view of social systems: that there are only local interactions of people and things, and all the rest (families, institutions, languages, social communities, class conflicts) are contingent and epiphenomenal, essentially unreal figments of our overly fervid sociological imagination. The ‘flat’ view sees only the human scale, indeed only the scale of the moment and the event, privileging that scale in relation to all others. It does not ask how and why events widely separated in time and space seem to re-enact the same patterns; it does not recognize that there are emergent phenomena unique to every level of organization in a complex dynamical system: recurring and typical patterns of interaction that cannot be explained or predicted from analysis of the interacting units.
Neither of these essentially ‘spatial’ views of complex
ecosocial systems is satisfactory. The spatial hierarchy model ignores the
important role of network toplogies of interaction. The ‘flat’
interpretation of network models cannot account for regularities at higher-scale
levels. The ‘spatial’ view is incomplete, and indeed is not, I believe, the
fundamental view needed to understand complex systems, especially human
ecosocial systems. I want to argue here for the usefulness of an alternative,
more dynamical view.
In dynamical theories of complex systems, the fundamental
unit of analysis is a process. It is in relation to the process that its
participants are defined, as filling roles in that process. Things, or organisms,
or persons, or institutions, as usually defined, are not dynamical
notions: they are ordinarily defined in terms of their stable and persistent, or
invariant, properties. They are not about dynamics; not about change and doing,
but about being what they are. Every process, or action, or social practice, or
activity, occurs on some timescale (in complex cases more than one timescale).
In a dynamical theory, an ecosocial system is a system of interdependent
processes; an ecosocial or sociotechnical network is described by saying
what’s going on, what’s participating and how, and how one going-on is
interdependent with another. Each scale of organization in an ecosocial system
is an integration of faster, more local processes (activities, practices,
doings, happenings) into longer timescale, more global or extended networks. It
is relative timescale that determines the probability and intensity of
interdependence (according to what I will call below the adiabatic principle),
and it is the circulation through the network of semiotic artifacts
(books, buildings, bodies) that enables coordination between processes at
radically different timescales.
In this view the two fundamental questions for analyzing the dynamics of ecosocial systems – and human activities within them – are: What processes, what kinds of change or doing, are characteristic of each relevant timescale of organization of the system/network? and How are processes integrated across different timescales? In the sections that follow I want to develop in more detail the implications of multiple-timescale analysis for the study of meaningful human activity and to raise a host of research questions generated by this perspective. My principal example will be schooling in relation to identity development and cultural continuity. I will move from a brief consideration of the basic dynamics of complex systems in general, to the case of ecosystems in which meanings matter, and finally to the conclusion that ‘it takes a village’ to study a village.