“Did he jump or was he pushed?” Jumping is an action. Being pushed is an event. Action theory is an approach to the study of social life that is based on the ontological premise that people jump. For example, the flow of traffic on a busy street differs from the flow of electrons on a copper wire. Electrons are pushed, drivers are not. From a structural perspective, we can learn a great deal about the flow of traffic by focusing on exogenous determinants, without ever knowing much about what drives human behavior. While few action theorists would disagree with the value of structural analysis, they also see the need to look beyond the constraints on action, to the intentions, purposes, and goals that motivate efforts to push back. Action theory has roots in Max Weber’s interpretative method and in Talcott Parsons’s effort to integrate this with E´mile Durkheim’s macrosocial approach. In “The Place of Ultimate Values in Sociological Theory,” Parsons insisted that “man is essentially an active, creative, evaluating creature” whose behavior must be understood in terms of the ends of action, and not “in terms of ‘causes’ and ‘conditions’” (1935). His “voluntaristic theory of action” opposed the deterministic account of human behavior as “pushed,” whether by Sigmund Freud’s “unconscious” or Pavlov’s bell. Action theory informs a diverse range of contemporary sociological theorizing, including rational action, symbolic interactionism, conflict theory, and hermeneutics. Conceptually, there are two main branches – one based on interests, the other on identity. Rational-action theory posits instrumental pursuit of self-interest, which can include an interest in public as well as private goods and an interest in social approval and avoidance of sanctions. Using mathematical formalism, the theory can generate testable predictions from a relatively small number of assumptions. However, the scope of the theory is limited by heroic assumptions about perfect information and unlimited calculating ability. Even versions based on “bounded rationality” are limited to actions intended to maximize utility, which excludes expressive and enthusiastic behavior and actions motivated by normative obligation and moral righteousness.
That void has been addressed by theories of action based on identity rather than interest. For identity theorists, “interests are only the surface of things. What is beneath the surface is a strong emotion, a feeling of a group of people that they are alike and belong together,” according to Randall Collins in Sociological Insight (1992: 28). Individuals order the social world by carving out cognitive categories through interaction with others, leading to stereotyping, in-group favoritism, and out-group prejudice. Social and moral boundaries are defined and affirmed by punishing deviants. Punishment is not calibrated to deter deviance; rather, it is unleashed as an expression of indignation at the violation of normative boundaries, even when this may excite opposition rather than suppress it.
Interest and identity theories of action both emphasize the dynamics of interaction among autonomous but interdependent agents. However, they differ in how this interdependence is understood. Interest theory posits strategic interdependence, in which the consequences of individual choices depend in part on the choices of others. Game theorists (see game theory) model this interdependence as a payoff matrix defined by the intersection of all possible choices of the players, with individual payoffs assigned to each cell. For example, the payoff for providing favors depends on whether the partner reciprocates. Peer pressure is also an example of strategic interdependence created by the application of sanctions conditional upon compliance with expected behavior.
Identity theorists point instead to the cognitive interdependence of agents who influence one another in response to the influences they receive, through processes like communication, persuasion, instruction, and imitation. Action theory poses three related and perplexing puzzles: the problem of social order, the tension between structure and action, and the problem of free will and determinism. Contemporary research on complex dynamical systems has enriched action theory by providing plausible solutions to each of these puzzles, based, in turn, on the principles of selforganization, emergence, and deterministic chaos.
Macrosocial theories of social order posit a structured system of institutions and norms that shape individual behavior from the top down. In contrast, action theories assume that much of social life emerges from the bottom up, more like improvisational jazz than a symphony orchestra. People do not simply play roles written by elites and directed by managers. We each chart our own course, on the fly. How then is social order possible? If every musician is free to play as they choose, why do we not end up with a nasty and brutish cacophony, a noisy war of all against all?
Parsons addressed the “Hobbesian problem of order” by positing a set of shared norms and
values that secure the cultural consensus necessary for social systems to function. Yet this is not a satisfactory solution. In effect, society remains a symphony orchestra in which the musicians must still learn their parts, except that now the Leviathan needs to carry only a thin baton, and not a lethal weapon.
An alternative solution was anticipated by Parsons’s student, Niklas Luhmann. Luhmann bridged the gap between action theory and systems theory by placing individual actors in a web of communicative interaction with others. His rather abstruse ideas on autopoietic systems of interaction find clearer expression in complexity theory. The emergence of order out of local interaction in complex systems has come to be known as “selforganization” according to S. Kaufman in Origins of Order (1993). The archetype is biological evolution, but there are parallels across the sciences, cases in which surprising (and often quite exquisite) global patterns emerge from interactions among relatively simple but interdependent processes, in the absence of central coordination, direction, or planning. These include flocks of birds, traffic jams, fads, forest fires, riots, and residential segregation. There is no leader bird who choreographs the dance-like movement of a flock of geese. There is no supervisor in charge of a riot. There is no conspiracy of banks and realtors who are assigning people to ethnically homogeneous neighborhoods. These processes are examples of complex systems in which global order emerges spontaneously out of a web of local interactions among large numbers of autonomous yet interdependent agents. Emergence is a defining feature of complex systems and is ultimately responsible for the self-organization we find beneath the apparent chaos of nature (Coveney and Highfield, Frontiers of Complexity, 1995).
Emergent properties are not reducible to the properties of the individual agents. The idea of emergence was anticipated by one of the founders of sociology, who established this as a fundamental rule of the sociological method. “The hardness of bronze is not in the copper, the tin, or the lead, which are its ingredients and which are soft and malleable bodies,” E´mile Durkheim wrote in The Rules of the Sociological Method, “it is in their mixture.” “Let us apply this principle to sociology,” he continued; “[Social facts] reside exclusively in the very society itself which produces them, and not in its parts, i.e., its members” (1986: xlvii). Structuralists have reified Durkheim’s theory of social facts as emergent properties, leaving individual actors as little more than the incumbents of social locations and the carriers of structural imperatives. Heterogeneity in preferences and beliefs affects only which individuals will fill which “empty slots,” the origin of which lies in processes that operate at the societal level. In The Structure of Social Action (1937), Parsons also argued for the emergent properties of social systems, but believed Durkheim went too far in concluding that these “social facts” are entirely independent of individual consciousness. Parsons corrects the hyperstructuralist interpretation of Durkheim by incorporating an essential insight of Joseph Schumpeter’s “methodological individualism,” the idea that societal patterns emerge from motivated choices and not from social facts external to individuals. Methodological individualism can be taken to imply that social facts are but the aggregated expression of individual goals and intentions. For example, residential segregation reflects the preferences of individuals for living among people similar to themselves. In contrast, structuralists assume that individual differences in ethnic identity affect who will live where in segregated neighborhoods but are not the cause of neighborhood segregation, which emanates from societal processes like red-lining and patterns of urban development.
Action theory is often most effective when it steers between these extremes. A classic example is Thomas Schelling’s model of neighborhood segregation in his “Dynamic Model of Segregation” in the Journal of Mathematical Sociology 1971 (1). Schelling challenged the macrosocial assumption that segregation is imposed from the top down, through institutional means like “red-lining.” At the same time, his famous experiment also challenged the microsocial assumption that segregation floats from the bottom up, through the aggregation of individual prejudices against ethnic minorities and outsiders. Schelling randomly distributed red and green chips on a large checkerboard and moved individual chips to empty locations if the number of in-group neighbors fell below an individual’s threshold of tolerance. He discovered that extreme segregation can emerge even in a population that tolerates diversity, as agents relocate to avoid being in the minority. This surprisingly strong tendency towards neighborhood segregation is an emergent property of the population, generated by local interactions among large numbers of interdependent but autonomous agents, even when every individual is tolerant of diversity.
Action theory explains social life by identifying the reasons for action (whether instrumental interests or symbolic meanings). As Anthony Giddens put it in The Constitution of Society, “I propose simply to declare that reasons are causes” (1984: 345). Yet most people now accept that everything in the universe is physically determined. How can this determinism be reconciled with a voluntaristic theory of action? Consider a sunbather who moves his/her towel to fend off a late afternoon shadow. Meanwhile, next to the towel, a heliotropic plant turns to follow the sun’s trajectory, thereby maximizing its access to an essential resource. Even the most dedicated Cartesian would not suggest that a sunflower is a purposive agent whose actions can be explained by the plant’s need for photosynthesis. How do we know that the sunbather is any different? One answer is that the sunbather could have chosen to remain in the shadow, while the sunflower could not. However, it is trivial to construct a stochastic sunflower that “chooses” to move, based on a probability distribution given by the location of the sun. A better answer is that the sunbather can tell you that the desire for sunlight is the reason for the action, while the sunflower will tell you nothing of the kind. Plants cannot provide reasons for their behavior, humans can. But does this mean that the sunbather is right? Is it possible that the sunbather, like the sunflower, is simply responding to physical stimuli that induce heliotropic movement, and, unlike for the sunflower, this movement is accompanied by the epiphenomenal feeling of choosing?
There is mounting evidence from neuroscientists and experimental psychologists that supports that possibility. In 1983, Benjamin Libet found that “cerebral neural activity (‘readiness potential’) precedes the subject’s awareness of his/her intention or wish to act by at least 350 msec” (“Commentary on ‘Free Will in the Light of Neuropsychiatry,’” 1996). More recently, in The Illusion of Conscious Will (2002), Daniel Wegner reported substantial evidence to support the hypothesis that “conscious will” is largely an illusion, useful to help us remember our authorship of actions whose causes lie elsewhere. These and other studies point to the possibility that our intentions are formed in the course of initiating action, but in a separate cognitive subsystem that assigns authorship after the fact. If so, then perhaps humans are unique in the ability to provide rational accounts for our actions, but we have no more free will than does a sunflower.
The theory of complex systems suggests an alternative possibility – that free will is compatible with determinism. Even relatively simple dynamical systems can require exponential amounts of computing power for every additional input into the system, until the number of bits required to predict system behavior, even in the near term, can exceed the number of particles in the universe. Thus, a highly nonlinear deterministic system like the brain can be indeterminable, which leaves open a window for intentional choice that is not reducible to system determinants (James P. Crutchfield, “Complexity: Order Contra Chaos,” 1989). Meanwhile, a growing interest in complex adaptive systems has opened up action theory to “backward-looking” approaches in which intentionality is empirically variable rather than presupposed. In backward-looking models, the ends of action attract the behaviors that produce them, whether or not the agent intended the outcome or is even aware of its existence. From a forward-looking perspective, this idea appears hopelessly teleological since the ends of action are located in the future and cannot reach back through time to attract the choices needed to bring them about. Models of complex adaptive systems avoid this problem by pointing backward, not forward – attributing action to outcomes that have already occurred. In agent-based evolutionary models, outcomes of a given action alter the population distribution of agents who engage in that action. In learning models, outcomes of a given action alter the probability distribution of actions within the repertoire of any given agent. Either way, the link between action and outcome is a set of experiences, not intentions. Agents look forward by holding a mirror to the past. They jump when they are pushed.
Tidak ada komentar:
Posting Komentar