What social science can do

Quite a few postings here emphasize the limits of social science knowledge. Prediction of the behavior of large social wholes is difficult to impossible. There are few strong regularities among social phenomena. Social entities and processes are heterogeneous, plastic, and path-dependent. So the question arises: what can the social sciences do that takes them beyond the realm of description and reportage of the blooming, buzzing confusion of social comings and goings, to something that is more explanatory and generalizable?

I think there is an answer to this, and it has to do with identifying mid-level mechanisms and processes that recur in roughly similar ways in a range of different social settings. The social sciences can identify a fairly large number of these sorts of recurring mechanisms. For example —

  • public goods problems
  • political entrepreneurship
  • principal-agent problems
  • features of ethnic or religious group mobilization
  • market mechanisms and failures
  • rent-seeking behavior
  • the social psychology associated with small groups
  • the moral emotions of family and kinship
  • the dynamics of a transport network
  • the communications characteristics of medium-size social networks
  • the psychology and circumstances of solidarity

Further, the social sciences can attempt to discover the circumstances at the level of individual agents that make these mechanisms robust across social settings. They can model the dynamics and features of aggregation that they possess. And they can attempt to discover the workings of such mechanisms in particular social and historical settings, and work towards explanations of particular features of these events based on their theories of the properties of the mechanisms. Finally, they can attempt to find rigorous ways of attempting to model the effects of aggregating multiple mechanisms in a particular setting.

What this comes down to is the view that the main theoretical and generalizing contribution that the social sciences can make is the discovery and analysis of a wise range of recurring social mechanisms grounded in features of human agency and common institutional and material settings. They can help to constitute a rich tool box for social explanation. And, in a weak and fallible way, they can lay the basis for some limited social generalizations — for example, “In circumstances where a group of independent individuals make private decisions about their actions, the public goods shared by the group will be under-provided.”

This approach affords a degree of explanatory capacity and generalization to the social sciences. What it does not underwrite is the ability to offer general, comprehensive theories about any complex kind of phenomenon — cities, schools, revolutions. And it does not provide a foundation for confidence about large predictions about the future behavior of complex social wholes.

Social surprises

The near meltdown of the US financial system this week came as a surprise to most of us — experts, legislators, and citizens alike. That isn’t to say that the components of the disaster were unknown — the subprime crisis, the earlier financial undoings of Fannie Mae and Bear Stearns this summer, and the sudden collapse of Lehman Brothers last week. But what has come as a surprise is the severity of the warnings by the Federal Reserve and Treasury that the entire financial system is only a few steps from seizure and collapse. This is a catastrophic system failure — and no one would have anticipated its possibility six months ago.

Think of a few other surprises in the past thirty years — the collapse of the Soviet Union, the Iranian Revolution, or the emergence of China as a roaring engine of market-based growth. In each case the event was a discontinuous break from the trajectory of the past, and it surprised experts and citizens alike. (The photo above depicts the surprising Yeltsin standing on a tank in 1991.)

So what is a surprise? It is an event that shouldn’t have happened, given our best understanding of how things work. It is an event that deviates widely from our most informed expectations, given our best beliefs about the causal environment in which it takes place. A surprise is a deviation between our expectations about the world’s behavior, and the events that actually take place. Many of our expectations are based on the idea of continuity: tomorrow will be pretty similar to today; a delta change in the background will create at most an epsilon change in the outcome. A surprise is a circumstance that appears to represent a discontinuity in a historical series.

It would be a major surprise if the sun suddenly stopped shining, because we understand the physics of fusion that sustains the sun’s energy production. It would be a major surprise to discover a population of animals in which acquired traits are passed across generations, given our understanding of the mechanisms of evolution. And it would be a major surprise if a presidential election were decided by a unanimous vote for one candidate, given our understanding of how the voting process works. The natural world doesn’t present us with a large number of surprises; but history and social life are full of them.

The occurrence of major surprises in history and social life is an important reminder that our understanding of the complex processes that are underway in the social world is radically incomplete and inexact. We cannot fully anticipate the behavior of the subsystems that we study — financial systems, political regimes, ensembles of collective behavior — and we especially cannot fully anticipate the interactions that arise when processes and systems intersect. Often we cannot even offer reliable approximations of what the effects are likely to be of a given intervention. This has a major implication: we need to be very modest in the predictions we make about the social world, and we need to be cautious about the efforts at social engineering that we engage in. The likelihood of unforeseen and uncalculated consequences is great.

And in fact commentators are now raising exactly these concerns about the 700 billion dollar rescue plan currently being designed by the Bush administration to save the financial system. “Will it work?” is the headline; “What unforeseen consequences will it produce?” is the subtext; and “Who will benefit?” is the natural followup question.

It is difficult to reconcile this caution about the limits of our rational expectations about the future based on social science knowledge, with the need for action and policy change in times of crisis. If we cannot rely on our expectations about what effects an intervention is likely to have, then we can’t have confidence in the actions and policies that we choose. And yet we must act; if war is looming, if famine is breaking out, if the banking system is teetering, a government needs to adopt policies that are well designed to minimize the bad consequences. It is necessary to make decisions about action that are based on incomplete information and insufficient theory. So it is a major challenge for the theory of public policy, to attempt to incorporate the limits of knowledge about consequences into the design of a policy process. One approach that might be taken is the model of designing for “soft landings” — designing strategies that are likely to do the least harm if they function differently than expected. Another is to emulate a strategy that safety engineers employ when designing complex, dangerous systems: to attempt to de-link the subsystems to the extent possible, in order to minimize the likelihood of unforeseeable interactions. (Nancy Leveson describes some of these strategies in Safeware: System Safety and Computers.) And there are probably other heuristics that could be imagined as well.

What is social scientific knowledge?

The social and behavioral sciences endeavor to describe, explain, and interpret the range of the social and behavioral facts that surround us. To refer to this body of findings as “science” is to claim a set of epistemic values about the nature of the methods of inquiry and evaluation that are used to arrive at and assess the conclusions offered about this domain. The label “science” brings with it a set of presuppositions about rigor, evidence, generalizability, logical analysis, objectivity, cumulativeness, and the likelihood that the assertions that are made are true.

Consider a few assumptions that are often made about scientific knowledge—some valid and some not. Science is based on a set of rationally justified methods of inquiry and testing. Scientific knowledge progresses, in scope, in detail of understanding, and in reliability. Science is performed by specialists, working within equally exacting communities of peers and competitors and subject to a demanding set of standards of evaluation—peer-reviewed journals, university review processes, national laboratories, and international associations and conferences. The result of these processes of testing and evaluation, we expect, is an expanding body of hypotheses, experimental findings, observations, theories, and explanations that have substantial credibility—and substantially higher credibility than the writings of casual observers of a given range of phenomena. We come to know the nature of the world better through the institutions and methods of science.

In addition to these reasonably valid assumptions about scientific knowledge, there is another group of more questionable ideas that derive from assumptions drawn from the natural sciences. Science permits generalizations; it permits us to systematize otherwise apparently separate domains of phenomena (planetary motion, the tides; rational choice theory, behavior of the family) and to demonstrate that apparently heterogeneous sets of phenomena are in fact governed by the same general laws. Science permits predictions; if the fundamentals are thus-and-so, then the compounds will behave thusly. Science aims at unification: the discovery of unitary systems of forces and entities whose aggregate properties represent the whole of nature.

Notice that these latter expectations are derived from the successes and specific characteristics of certain of the natural sciences. And this marks the first of many opportunities for error in the philosophy of social science. There is no reason to expect that the social domain possesses the underlying nature and orderliness that would make it possible to achieve some of these characteristics (in particular, uniformity, generalizability, unification, simplicity). Consider some other areas of possible empirical research—for example, animal behavior. We should not expect there to be comprehensive theories of animal behavior. Instead, we should expect many threads of research, corresponding to many dimensions of animal behavior: cognition, memory, instinct, social behaviors, migratory behavior. And these many strands of research would reach out to different kinds of causal backgrounds: evolutionary biology, neurophysiology, intra-group learning. Likewise with the domain of social behavior. There is no single unified “theory of human motivation”—whether rational choice theory, social psychology, or any other unified theory. And this is so, because there is no unified reality of motivation and action; rather, there is a heterogeneous range of motives, errors, impulses, commitments, and habits that together constitute “dispositions to behave.”

What underwrites the claim of truthfulness for scientific knowledge? What gives us a rational basis for believing that the results of the socially constructed activities of science lead to true hypotheses about the nature and workings of the phenomena that scientific inquiry considers? There is, first, the basic argument of empiricism: we can observe some features of the world and establish certain statements as being probable. And we can use a collection of tools of inference to establish credibility of other non-observational statements (deductive and inductive logic, statistics, the experimental method, causal modeling).

This simple empiricist epistemology underwrites the strongest claims for veridicality and justification for the social sciences. The discovery of empirical facts about the social world is possible but challenging; this is what much of social science methodology attempts to under-gird. And hypotheses about the causal relationships that exist among social entities and processes can be tested using a variety of methods of inference that themselves possess strong epistemic justification. We have learned from the writings of philosophers of science since the 1960s to emphasize corrigibility and anti-foundationalism in our interpretation of scientific knowledge; but a coherentist epistemology and a perspective of causal realism provides a philosophically powerful grounding for social science knowledge. (See articles in the Stanford Encyclopedia of Philosophy on coherence epistemology and scientific realism by Kvanvig and Boyd).

In addition, in some areas of the natural sciences, there is the fact that cumulative scientific research leads to the invention of technologies that work as they were designed to do: new materials are invented in the electronics industry, new designs are created for large structures (buildings, aircraft, electron microscopes), and these materials and artifacts perform as expected on the basis of the underlying theories. So scientific theories of materials, structures, and natural systems are supported by the effectiveness of the technologies that they give rise to. If the theories and hypotheses were fundamentally untrue about the parts of the natural world that they describe, then we would expect the technologies to fail; the technologies do not fail; so we have some additional reason to believe the scientific theories that underlay these technologies. (This is a version of Richard Boyd’s argument of methodological realism.)

Is there anything analogous to the relationship between the natural sciences and technology, for the social and behavioral sciences? On the whole, there is not. Social predictions are notoriously unreliable; public policies based on social-science theory commonly give rise to unanticipated consequences; and the twists and turns of deliberate social processes (war, alliance, efforts to address global warming) continue to surprise us. This unpredictability in the social world derives from the nature of social action. Human behavior and social processes are plainly subject to an open-ended range of causes, motives, and influences. The construction of various areas of science with the goal of understanding and explaining this multiplicity is therefore a profoundly challenging task.

Here, then, is a very elliptical description of a plausible interpretation of social science epistemology: There are empirical foundations for knowledge in the form of social observation (empiricism); there are social causes that influence social behavior, processes, and outcomes (causal realism); there is no a priori reason to expect strong generalizations across social phenomena, “regulating” the social world; and there is no reason to expect unified master theories of social phenomena, suggesting instead a preference for theories of the middle range.

Realism for the social sciences?

Scientific realism is the idea that scientific theories provide descriptions of the world that are approximately true. This view implies a correspondence theory of truth — the idea that the world is separate from the concepts that we use to describe it. And it implies some sort of theory of scientific rationality — a theory of the grounds that we have for believing or accepting the findings of a given area of science. (See a brief article on the basics of scientific realism including some useful references here.) Realism, objectivity, and facts go together. We can interpret a theory realistically just in case we believe that there is a fact of the matter concerning the assertions contained in the theory. (See earlier postings relevant to this topic, Concepts and the World and Social Construction.)

Realism raises all kinds of interesting questions when we consider applying it to the social sciences. For one thing, it requires a useable distinction between the world and the knower. This raises the question: is there an objective social world independent from the perceptions and concepts of observers? And this also is a complicated question, because the persons who make up social processes at the micro-level are themselves “knowers” of the social world. So there is a question about the objectivity of the social world and a corresponding question about social construction of social reality. If all social phenomena are socially constructed, then how can it be the case that some statements about social phenomena are objective and independent from the conceptual schemes of the observer?

Scientific realism got its impetus from the fact that physical theories invoke theoretical concepts that are not themselves directly observational — muon, gravity wave, gene (at an early stage of biology). So the question arose, what is the status of the reference and truth of scientific sentences that include non-observational concepts — for example, “muons have a negative electric charge and a spin of -1/2”? Since we can’t directly inspect muons and measure their charge and spin, sentences like this depend for their empirical confirmation on their logical relationships to larger bits of physical theory — and ultimately upon a measure of the overall degree to which this physical theory issues true experimental and observational predictions. And the empirical confirmation of the theory as a whole, the story goes, provides a rational basis for assigning a reference and truth value to its constituent sentences. So the fact that “muon” is embedded within a mathematical theory of subatomic reality and the theory is well confirmed by experimental means, gives us reason to believe that muons exist and possess approximately the characteristics attributed to them by muon theory.

But all of this has to do with esoteric physical theory. Is there any relevant application of realism in the social sciences? Here’s one important difference: the social sciences are barely “theoretical” at all in the sense associated with the natural sciences. The concepts that play central roles in social theories — charisma, bureaucratic state, class, power — aren’t exactly “theoretical” in the sense of being non-observational. And social concepts aren’t defined implicitly, in terms of the role that they play in an extended formal theoretical structure. Rather, we can give a pretty good definition of social concepts in terms of behavior and common-sense attributes of social entities. In the social sciences we don’t find the conceptual holism that Duhem and Quine attributed to the natural sciences (Pierre Duhem, The Aim and Structure of Physical Theory; W. V. O. Quine, Word and Object). Instead, both meaning and confirmation can proceed piecemeal. So if realism were primarily a doctrine about the interpretation of theoretical terms, there wouldn’t be much need for it in the social sciences.

But here are several specific ways in which scientific realism is useful in the social sciences, I think. And they all have to do with the kinds of statements in the social sciences that we think can be interpreted as expressing facts about the world, independent of our theories and concepts.

Causal realism. We can be realist about the meaning of assertions about causation and causal mechanisms. We can take the position that there is a fact of the matter as to whether X caused Y in the circumstances, and we can assert the objective reality of social causal mechanisms. On the realist interpretation, social causal mechanisms exist in the social world — they are not simply constructs of the observer’s conceptual scheme. And the statement that “Q is the process through which X causes Y” makes a purportedly objective and observer-independent claim about Q; it is an objective social process, and it conveys causation from X to Y. Q is the causal mechanism underlying the causal relationship between X and Y.

Structure realism. We can be realist about the existence of extended social entities and structures — for example, “the working class,” “the American Congress,” “the movement for racial equality.” These social entities and structures have some curious ontological characteristics — it is difficult to draw boundaries between members of the working class and the artisan class, so the distinctness of the respective classes is at risk; institutions like the Congress change over time; a social movement may be characterized in multiple and sometimes incompatible ways; and social entities don’t fall into “kinds” that are uniform across settings. But surely it is compelling to judge that the Civil Rights movement was an objective fact in the 1960s or that the Congress exists and is a partisan environment. And this is a version of social realism.

Social-relations realism. If we say that “Pierre is actively involved in a network of retired French military officers”, we refer to a set of social relations encapsulated under the concept of a social network and composed of many pair-wise social relations. Here too we can take the perspective of social realism. It seems unproblematic to postulate the objective reality of both the pair-wise social relations and the aggregate network that these constitute. Each level of social relationship can be investigated empirically (we can discover that Pierre has regular interactions with Jean but not with Claude), and it seems unproblematic to judge that there is a fact of the matter about the existence and properties of the network — independent of the assumptions and concepts of the observer.

Meaning realism. Now, how about the hardest case: meanings and the objectivity of interpretation. Can we say that there is ever a fact of the matter about the interpretation of an action or thought? When Thaksin offends Charat by exposing the bottoms of his feet to him — can we say that “Charat’s angry reaction is the result of the meaning of this insulting gesture in Thai culture”? Even here, it is credible to me that there is a basis for saying that this judgment expresses an objective fact (even if it is a fact about subjective experience); and therefore, we can interpret this sentence along realist lines: “Thaksin’s gesture was objectively offensive to Charat in the setting of Thai culture.” It is evident that many of our interpretations of behavior and action are substantially underdetermined by context and evidence; so it may be that much interpretation of meaning does not constitute a “fact of the matter.” But this seems to be a fact about particular judgments rather than a universal feature of the interpretation of meanings.

So it seems that it is feasible and useful to take a social realist perspective on many of the assertions and theories of the social sciences; and what this says, is that we can interpret social science statements as being approximately true of a domain of social phenomena that have objective properties (i.e. properties that are independent from our conceptualization of them).

Social science and social problems

Several of the interviews that I’ve conducted in recent weeks have agreed on an important point: that the social sciences ought to be directed towards addressing important social problems, and that the research agenda for social science ought to be influenced or shaped by the constituencies in society who are most affected by these social problems. At bottom – the social sciences ought to be engaged in a serious way in improving the quality of life for the people of the globe. They can best do this, it would seem, by discovering some of the causes of persistent social problems and providing a sound basis for designing policies that have a chance of ameliorating them. And they can focus their research agendas by working closely with practitioners and the ordinary people who experience these social problems.

What is the reality, however? To what extent do the social sciences conform to this ideal? As David Featherman expresses in his interview, the social sciences in the United States in the 1950s and 1960s took a turn away from goals of social engagement and problem solving, and they really haven’t yet turned back. Instead, the major social science disciplines took on the model of disinterested, academic theory formation. The natural sciences provided the role model, and the driving goals were quantifiability, theoretical parsimony, and formalizability. “Applied” research was devalued.

Re-establishing the connection between social science and social problems should be a high priority for all of us — social scientists and citizens alike. The social problems we face are crucially important, they are intractable, and they are often trending in the wrong direction. Consider this partial list of particularly pressing problems facing our society and the world.

  • United States
    • Endemic urban poverty
    • Racial segregation
    • Urban decline and despair
    • Rising inequalities of income, wealth, and quality of life
    • Lack of universal provision of health services
    • Rising social cost of health care
    • Failing delivery of education for children and adolescents
  • International —
  • deepening poverty in many countries
  • deepening inequalities of wealth, income, and quality of life
  • Violence against individuals and groups
    • Ethnic violence
    • Genocide
    • Crime
    • Thuggery
    • Oppressive states
  • Oppression of women and girls
  • Global environmental crisis
    • Climate change
    • Resource exhaustion
    • Environment degradation
  • Political regimes
    • Persistent authoritarian regimes
    • Imperfect democracies
    • Corruption
    • Inadequate systemic response to disaster

These are all problems with massive consequences for human wellbeing. Each of them is itself the manifestation of complex social and behavioral forces. And solutions will require the artful design of new institutions and new ways of coordinating social behavior. In short — these are problems that are much more challenging, intellectually and practically, than decoding the human genome or controlling a nuclear reactor or putting a human on Mars. The best efforts of talented and committed researchers will be needed in order to understand and change these conditions.

Fortunately, there are some signs that mainstream social scientists are beginning to turn their gaze back in the direction of concrete social problems. There is significant, sustained work going on in sociology and political science on the topics of poverty, inequality, racial segregation, and social disaffection, and this work is taking on some of the urgency and relevance that was displayed in the research of the Chicago school of sociology seventy-five years ago. The Center for Advancing Research and Social Solutions at the University of Michigan is an example of a group of researchers coming together with a commitment to bringing social science research to bear on social problems. (See the Featherman interview for a description of CARSS.) A recent symposium published in The Yard, Harvard’s Faculty of Arts and Sciences magazine, features a group of Harvard social scientists under the heading, “The New Social Science,” and the discussion focuses almost entirely on these social problems and some of the methods that can be used to address them. Featured in the article are Edward Glaeser, William Julius Wilson, Mary Waters, Claudia Gay, David Cutler, and Robert Putnam. (Unfortunately this publication doesn’t have a web presence.) It is very good to see research at this level of empirical detail and practical focus coming into the spotlight.

There seem to be two large meta-goals that the social sciences should have in confronting social problems. One is the problem of understanding these problems in detail – both the empirical details of how the problem is distributed and evolving, and the causal issue of discovering some of the factors that produce and reproduce the problem. What are the trends in urban and suburban social evolution? Why is urban poverty so intractable over multiple generations? Why have urban schools been unsuccessful in providing a high-quality education to all the children that they serve?

The second large meta-goal for the social sciences is to be able to provide a basis for policies and interventions that have a meaningful probability of solving the problems that we care about. Policies should be driven by the best possible understanding of the social and behavioral dynamics of the problems they are designed to address. And the social sciences should endeavor to provide sober assessments of the likely consequences of various proposed policies.

But nothing is simple in social life – and it is clear enough that there are complex interactive causal processes at work in the creation and sustenance of most social problems. The scope of prediction in the social sciences is limited, and this means that it is rarely possible to provide a categorical prescription such as this: “do this, and such-and-so will result.” Instead, the social sciences are perhaps most useful when they help to identify some of the behavioral complexities that might turn into “unforeseen consequences” – and thereby help to design policies that are more fault-tolerant.

None of this is simple. But there is no doubt that our society needs the knowledge and methods that the social sciences can provide, if we are to have a good chance of solving the problems we face. And this means that the social sciences need to take on the task of practical engagement with seriousness and commitment.

Agent-based modeling as social explanation

Logical positivism favored a theory of scientific explanation that focused on subsumption under general laws. We explain an outcome by identifying one or more general laws, a set of boundary conditions, and a derivation of the outcome from these statements. A second and competing theory of scientific explanation can be called “causal realism.” On this approach, we explain an outcome by identifying the causal processes and mechanisms that give rise to it. And we explain a pattern of outcomes by identifying common causal mechanisms that tend to produce outcomes of this sort in circumstances like these. (If we observe that patterns of reciprocity tend to break down as villages become towns, we may identify the causal mechanism at work as the erosion of the face-to-face relationships that are a necessary condition for reciprocity.)

But there are other approaches we might take to social explanation and prediction. And one particularly promising avenue of approach is “agent-based simulation.” Here the basic idea is that we want to explain how a certain kind of social process unfolds. We can take our lead from the general insight that social processes depend on microfoundations at the level of socially situated individuals. Social outcomes are the aggregate result of intentional, strategic interactions among large numbers of agents. And we can attempt to implement a computer simulation that represents the decision-making processes and the structural constraints that characterize a large number of interacting agents.

Thomas Schelling’s writings give the clearest exposition to the logic of this approach Micromotives and Macrobehavior. Schelling demonstrates in a large number of convincing cases, how we can explain large and complex social outcomes, as the aggregate consequence of behavior by purposive agents pursuing their goals within constraints. He offers a simple model of residential segregation, for example, by modeling the consequences of assuming that blue residents prefer neighborhoods that are at least 50% blue, and red residents prefer neighborhoods at least 25% red. The consequence — a randomly distributed residential patterns becomes highly segregated in an extended series of iterations of individual moves.

It is possible to model various kinds of social situations by attributing a range of sets of preferences and beliefs across a hypothetical set of agents — and then run their interactions forward over a period of time. SimCity is a “toy” version of this idea — what happens when a region is developed by a set of players with a given range of goals and resources? By running the simulation multiple times it is possible to investigate whether there are patterned outcomes that recur across numerous timelines — or, sometimes, whether there are multiple equilibria that can result, depending on more or less random events early in the simulation.

Robert Axelrod’s repeated prisoners’ dilemma tournaments represent another such example of agent-based simulations. (Axelrod demonstrates that reciprocity, or tit-for-tat, is the winning strategy for a population of agents who are engaged in a continuing series of prisoners’ dilemma games with each other.) The most ambitious examples of this kind of modeling (and predicting and explaining) are to be found in the Santa Fe Institute’s research paradigm involving agent-based modeling and the modeling of complex systems. Interdisciplinary researchers at the University of Michigan pursue this approach to explanation at the Center for the Study of Complex Systems. (Mathematician John Casti describes a number of these sorts of experiments and simulations in Would-Be Worlds: How Simulation is Changing the Frontiers of Science and other books.)

This approach to social analysis is profoundly different from the “subsumption under theoretical principles” approach, the covering-law model of explanation. It doesn’t work on the assumption that there are laws or governing regularities pertaining to the social outcomes or complex systems at all. Instead, it attempts to derive descriptions of the outcomes as the aggregate result of the purposive and interactive actions of the many individuals who make up the social interaction over time. It is analogous to the simulation of swarms of insects, birds, or fish, in which we attribute very basic “navigational” rules to the individual organisms, and then run forward the behavior of the group as the compound of the interactive decisions made by the individuals. (Here is a brief account of studies of swarming behavior.)

How would this model of the explanation of group behavior be applied to real problems of social explanation? Consider one example: an effort to tease out the relationships between transportation networks and habitation patterns. We might begin with a compact urban population of a certain size. We might then postulate several things:

  • The preferences that each individual has concerning housing costs, transportation time and expense, and social and environmental environmental amenities.
  • The postulation of a new light rail system extending through the urban center into lightly populated farm land northeast and southwest
  • The postulation of a set of prices and amenities associated with possible housing sites throughout the region to a distance of 25 miles
  • The postulation of a rate of relocation for urban dwellers and a rate of immigration of new residents

Now run this set of assumptions forward through multiple generations, with individuals choosing location based on their preferences, and observe the patterns of habitation that result.

This description of a simulation of urban-suburban residential distribution over time falls within the field of economic geography. It has a lot in common with the nineteenth-century von Thunen’s Isolated State analysis of a city’s reach into the farm land surrounding it. (Click here for an interesting description of von Thunen’s method written in 1920.) What agent-based modeling adds to the analysis is the ability to use plentiful computational power to run models forward that include thousands of hypothetical agents; and to do this repeatedly so that it is possible to observe whether there are groups of patterns that result in different iterations. The results are then the aggregate consequence of the assumptions we make about large numbers of social agents — rather than being the expression of some set of general laws about “urbanization”.

And, most importantly, some of the results of the agent-based modeling and modeling of complexity performed by scholars associated with the Santa Fe Institute demonstrate the understandable novelty that can emerge from this kind of simulation. So an important theme of novelty and contingency is confirmed by this approach to social analysis.

There are powerful software packages that can provide a platform for implementing agent-based simulations; for example, NetLogo. Here is a screen shot from an implementation called “comsumer behavior” by Yudi Limbar Yasik. The simulation has been configured to allow the user to adjust the parameters of agents’ behavior; the software then runs forward in time through a number of iterations. The graphs provide aggregate information about the results.

What does rational choice theory explain?

Rational choice theory could be advanced as a pure set of axioms embodying a formal representation of individual choice under circumstances of uncertainty and strategic interaction. Decision theory incorporates the idea of maximizing utility under circumstances of uncertainty and risk. The basic rule is that the decision-maker could collect information about the utility and probability of each feasible choice, and choose the option that affords the maximum expected utility. (Here the decision-maker is playing against nature.) Game theory expands the range of decision-making situations by giving a representation of strategic interaction: situations in which the actor’s outcome depends upon the choices or strategies made by one or more rational opponents. Mathematical game theorists have demonstrated that this problem too admits of rational solution. The actor needs to discover the choices available to him/her and each other player and he/she needs to assign utilities for each possible outcome for each player. It is then demonstrable, for both zero-sum and non-zero-sum games, that there are one or more equilibrium sets of strategies for each player. This means that there is a single strategy or a mixed set of strategies with the property that, given rational choices by the opponent, there is no other strategy available to self that would produce a higher utility for self. (It may be observed that for games with many strategies for each player, discovery of the equilibrium set may not be practical.) Certain two-person games have received a great deal of analysis, including the prisoners’ dilemma, the game of chicken, and a range of cooperative games.

So much for the formal theory. In what sense does rational choice theory or game theory provide a basis for an explanation of real social outcomes?

We might begin a response to this question by saying that the axioms have empirical content as descriptions of real human decision-making. Real decision-makers do consider alternative choices in terms of the value and probability of the outcome. Moreover, actors who systematically break the expected utility rule will do less well than those who act according to the rule. So non-expected-utility actors will either learn or disappepear if the stakes are high. Likewise, in strategic situations (those depending on the independent actions of other deliberative agents), actors try to choose their strategy based on an analysis of the future choices of other players. So, once again, rational choice theory appears to capture an important dimension of real human decision making.

As noted in a prior posting, it goes without saying that human reasoners are not entirely conformant to the axioms — norms intervene, computational limits interfere, and the assesment of risky situations appears to be systematically divergent from the expected-utility model. In fact, these deviations create the subject matter for experimental or behavioral economics. (Daniel Kahneman and Amos Tversky, Judgment under Uncertainty: Heuristics and Biases.)

But given that there is some degree of correspondence between RC axioms and real human reasoning, we might say that RC theory provides an empirically grounded way of modeling certain real situations of decision making and strategic interaction. The pressing question is whether the empirical failures of the axioms as a description of real actors are sufficient to thoroughly invalidate the models. And it would appear that there are specific social settings that are likely to represent something like the pure case of rational decision-making as hypothesized by RC theory.

This implies in turn that there is a realistic basis for treating RC theory as a possible source of real empirically grounded explananations of observed social behavior. For example, suppose we are interested in the distributive features of WTO treaties or the incidence of peasant rebellions in late imperial China. We might model the WTO problem by treating nations as rational actors, attributing a set of utilities and probabilities to them, and using the findings of bargaining theory to predict the nature of the distribution of benefits and burdens in a series of agreements. Or we might regard the occurrence of a rebellion as a choice situation for each potential rebel that involves multiple outcomes, each with a utility and a probability. And we might hypothesize that rebellions will be most frequent in situations where the costs of participation are lowest and the rewards are greatest (James Tong, Disorder Under Heaven: Collective Violence in the Ming Dynasty).

In each case we have taken an abstract mathematical system, used it to create a model of the actual social situation of interest, and have then solved the terms of the model. This solution can then be interpreted as a prediction or post-diction of the actual situation, and we can compare the model’s results with empirical and hisistorical facts.

(A very negative assessment of the empirical utility of rational choice theory is offered in Donald Green and Ian Shapiro, Pathologies of Rational Choice Theory: A Critique of Applications in Political Science. Their skepticism is taken on by many authors in Jeffrey Friedman’s edited volume, The Rational Choice Controversy: Economic Models of Politics Reconsidered.)

%d bloggers like this: