Mechanisms, singular and general

Let’s think again about the semantics of causal ascriptions. Suppose that we want to know what  caused a building crane to collapse during a windstorm. We might arrive at an account something like this:

  • An unusually heavy gust of wind at 3:20 pm, in the presence of this crane’s specific material and structural properties, with the occurrence of the operator’s effort to adjust the crane’s extension at 3:21 pm, brought about cascading failures of structural elements of the crane, leading to collapse at 3:25 pm.

The process described here proceeds from the “gust of wind striking the crane” through an account of the material and structural properties of the device, incorporating the untimely effort by the operator to readjust the device’s extension, leading to a cascade from small failures to a large failure. And we can identify the features of causal necessity that were operative at the several links of the chain.

Notice that there are few causal regularities or necessary and constant conjunctions in this account. Wind does not usually bring about the collapse of cranes; if the operator’s intervention had occurred a few minutes earlier or later, perhaps the failure would not have occurred; and small failures do not always lead to large failures. Nonetheless, in the circumstances described here there is causal necessity extending from the antecedent situation at 3:15 pm to the full catastrophic collapse at 3:25 pm.

Does this narrative identify a causal mechanism? Are we better off describing this as a sequences of cause-effect sequences, none of which represents a causal mechanism per se? Or, on the contrary, can we look at the whole sequence as a single causal mechanism — though one that is never to be repeated? Does a causal mechanism need to be a recurring and robust chain of events, or can it be a highly unique and contingent chain?

Most mechanisms theorists insist on a degree of repeatability in the sequences that they describe as “mechanisms”. A causal mechanism is the triggering pathway through which one event leads to the production of another event in a range of circumstances in an environment. Fundamentally a causal mechanism is a “molecule” of causal process which can recur in a range of different social settings.

For example:

  • X typically brings about O.

Whenever this sequence of events occurs, in the appropriate timing, the outcome O is produced. This ensemble of events {X, O} is a single mechanism.

And here is the crucial point: to call this a mechanism requires that this sequence recurs in multiple instances across a range of background conditions.

This suggests an answer to the question about the collapsing crane: the sequence from gust to operator error to crane collapse is not a mechanism, but is rather a unique causal sequence. Each part of the sequence has a causal explanation available; each conveys a form of causal necessity in the circumstances. But the aggregation of these cause-effect connections falls short of constituting a causal mechanism because the circumstances in which it works are all but unique. A satisfactory causal explanation of the internal cause-effect pairs will refer to real repeatable mechanisms — for example, “twisting a steel frame leads to a loss of support strength”. But the concatenation does not add up to another, more complex, mechanism.

Contrast this with “stuck valve” accidents in nuclear power reactors. Valves control the flow of cooling fluids around the critical fuel. If the fuel is deprived of coolant it rapidly overheats and melts. A “stuck valve-loss of fluid-critical overheating” sequence is a recognized mechanism of nuclear meltdown, and has been observed in a range of nuclear-plant crises. It is therefore appropriate to describe this sequence as a genuine causal mechanism in the creation of a nuclear plant failure.

(Stuart Glennan takes up a similar question in “Singular and General Causal Relations: A Mechanist Perspective”; link.)

Mechanisms and intellectual movements

I am particularly interested in the idea that we can explain social outcomes by identifying the social mechanisms that (often, typically, occasionally) bring them about. I also find the evolution of science and systems of ideas to be particularly fascinating within contemporary sociology, in that this aspect of human life embraces both rationally directed thought and sociological influences. So it is very interesting to consider what we can discover about the structures, networks, and professional organizations that influence the course that a given discipline or field of research takes.

It is therefore interesting to consider the role that reference to social mechanisms has played in recent works of the sociology of science and the sociology of knowledge. A particularly good example is found in the work of sociologists like Camic, Lamont, Gross, and Frickel, and Frickel and Gross’s “General Theory of Scientific/Intellectual Movements” (2005) is a good place to start (link). Frickel and Gross put their goal in this article in this way:

The theory seeks to answer the question, under what social conditions is any particular scientific/intellectual movement, or SIM (whose nature we clarify shortly), most likely to emerge, gain adherents, win intellectual prestige, and ultimately acquire some level of institutional stability? (205)

This description evokes an explanatory goal with a causal perspective — “conditions” that make “emergence” likely. But on its face this is not a mechanisms-based approach — rather, it is more akin to a “facilitating or necessary conditions” kind of analysis of social causation. This impression is reinforced by the assertion that the theory is inductive, based on an examination of a number of case studies of SIMs aimed at identifying such conditions. (The authors also make a point of giving emphasis to failed SIMs because of the traction offered by such cases for counterfactual analysis.) They emphasize the importance of identifying common features of SIMs, in order to “mark them as objects for sociological study” (208), which implies that a precondition of sociological study is that we need to identify a social kind of entities with reasonably similar properties. This too suggests an underlying causal perspective that looks to regularities and common properties rather than causal mechanisms or causal powers.

As much of the recent discussion of critical realism makes clear, it is very important to be as explicit as possible about the assumptions we make about causation in the social sciences. So a quick review of the article may be useful in order to shed light on the kinds of causal thinking that Frickel and Gross engage in here.

To begin, what is a SIM?

The most abbreviated definition is this: SIMs are collective efforts to pursue research programs or projects for thought in the face of resistance from others in the scientific or intellectual community. (206)

So one criterion for an ensemble of thinkers and institutions to constitute a SIM in the F/G definition is that their shared intellectual program needs to challenge the status quo, the dominant way of thinking about the subject matter of concern. F/G explicitly model their analysis on the study of social movements; notice the parallel with McAdam, Tarrow, and Tilly’s formulation in Dynamics of Contention of their central question.

Under what conditions will normally apathetic, frightened, or disorganized people explode into the streets, put down their tools, or mount the barricades? How do different actors and identities appear and transform in episodes of contention? Finally, what kinds of trajectories do these processes follow? (chapter 2)

It is interesting that F/G are quite explicit in looking for a “general theory”. What they mean by this, apparently, is an account of a limited set of social conditions whose presence or absence “explains” the success or failure of a candidate SIM at a point in time. And this in turn sounds quite a bit like the comparative method pursued by Theda Skocpol in States and Social Revolutions: A Comparative Analysis of France, Russia and China: through comparative study of cases, discover a background set of social and political conditions that serve as jointly sufficient and/or necessary conditions for the occurrence of social revolution (link). (Like Skocpol, F/G make use of the probabilistic versions of sufficiency and necessity: “makes more likely” and “makes more unlikely”.) Mechanisms come into the story fairly quickly: “Our general theory insists that the precise mechanisms whereby a field’s external environment shapes a SIM must be specified” (209); but in fact, there is very little discussion of concrete mechanisms in the article.

The four premises of the general theory are these:

  • Proposition 1: A SIM is more likely to emerge when high-status intellectual actors harbor complaints against what they understand to be the central intellectual tendencies of the day. (209; italics mine)
  • Proposition 2: SIMs are more likely to be successful when structural conditions provide access to key resources. (213)
  • Proposition 3: The greater a SIM’s access to various micro mobilization contexts, the more likely it is to be successful. (219)
  • Proposition 4: The success of a SIM is contingent upon the work done by movement participants to frame movement ideas in ways that resonate with the concerns of those who inhabit an intellectual field or fields. (221)

For each of these theoretical propositions they offer the sketch of an idea about what the mechanisms are that might support this factor. For example, concerning proposition 1, they maintain that “grievance” is a necessary condition for the emergence of an SIM because it puts potential adherents in a state of psychological readiness for mobilization. Another mechanism they cite for the emergence and mobilization of an SIM is the sudden entry into a field of non-traditional practitioners — for example, women or African-American scholars entering the field of urban studies in the 1960s who found that prevailing wisdom failed to do justice to their own experiences. And on the resources point, F/G refer to the job market, academic organizations, and funding sources, and sketch out how favorable conditions with regard to these structural features can facilitate the success of a SIM. This is, at least in sketch, a mechanisms analysis.

The mechanisms associated with Proposition 3 are encapsulated in the notion of “micromobilization”. Like Tilly in his analysis of the counter-revolution in The Vendee, F/G hold that the success of a SIM is influenced by the strength or weakness of the various organizations and networks through which it is able to spread its message and its mobilization efforts at the grassroots level. They mention laboratories, conferences, research retreats, and academic departments (219). Once one or more advocates of the given SIM has a position of influence in one of these centers, he or she is enabled to influence and mobilize other scholars to the SIM.

The mechanisms associated with Proposition 4 pick up on the rhetorical side of intellectual work.  We might unsympathetically refer to this aspect of the development of a SIM as the marketing campaign it pursues. In order to influence prospective adherents to an intellectual movement it is necessary to provide “messages” that resonate with them. (Fritz Ringer’s analysis of the German mandarins between the wars in The Decline of the German Mandarins: The German Academic Community, 1890-1933 seems to illustrate this mechanism; a few highly effective reactionary authors caught the wave of pessimism that was present in German culture between the wars, and this seems to have had an important effect on the development of social science thinking in the period.) This factor has to do with effective framing of issues and research questions:

Fundamental to framing, and underlying and connecting to the three other dimensions we describe shortly, is the notion of intellectual identity. We see intellectual identity as one of the crucial links between micro, mess, and macro levels of analysis in the sociology of ideas. (222)

It is possible to take issue with the notion that there is a general theory on offer here. I would rather call the analysis provided here an account of some generalizations about the causal conditions that facilitate or impede intellectual movements. The phrase “general theory” makes the effort seem more comprehensive than it actually aims to be. What this treatment lacks (by design) is a micro- or meso-level account of how specific institutions, identity features, resource sources, and networks have played out in specific instances of intellectual change. (The contributions to Camic, Gross, and Lamont’s Social Knowledge in the Making do this in a variety of ways.)

But consider Chuck Tilly’s frequent critique of a similar effort in contentious politics studies: it is the underlying mechanisms and processes, not the general similarities and common conditions, that provide real insight into the explanation of episodes of contentious action. Tilly argues that there is a great deal of variation across episodes; but we can nonetheless discover some common underlying mechanisms and processes. And this would suggest that a more meso-level might be helpful in the study of SIMs as well. Or putting it in other terms, more attention to mechanisms and less emphasis on general conditions might provide more insight into the phenomena of intellectual movements.

There is one final observation that appears relevant here. The “social mechanisms” approach itself might be classified as a SIM in the making. This intellectual movement involves a relatively small group of practitioners embedded within specific centers of institutional influence; it emerged from dissatisfaction with the received view of causation in the social sciences; and it is involved in a struggle for resources and prestige in the field of the philosophy of social science, both in Europe and North America. (For that matter, much the same could be said for critical realism.)

Finally, I am keeping my eyes open for meso-level social mechanisms in the sociology literature, and so I was curious in reading through this piece again whether any of the mechanisms postulated here were meso-meso. It seems that they are not. Rather, the social mechanisms mentioned generally proceed from a structure or institution to individual behavior (meso-micro) or from individual behavior to a meso- or macro-level outcome (progress of the SIM). But if this is correct, then the explanatory work offered here conforms to the downward and upward struts of Coleman’s boat, not the type 4 causation from meso to meso that Coleman precludes (link). This makes the analysis perhaps more compatible with the strictures of analytical sociology that the authors might have guessed (link).

Modest predictions in history

Image: the owl of Minerva

In spite of their reputations as historical determinists, Hegel and Marx each had their own versions of skepticism about “learning from history” — in particular, the possibility of predicting the future based on historical knowledge. Notwithstanding his view that history embodies reason, Hegel is famous for his idea in the Philosophy of Right: “When philosophy paints its grey in grey then has a shape of life grown old. By philosophy’s grey in grey it cannot be rejuvenated but only understood. The owl of Minerva spreads its wings only with the falling of dusk.” And Marx puts the point more sardonically in the Eighteenth Brumaire: “Hegel remarks somewhere that all great world-historic facts and personages appear, so to speak, twice. He forgot to add: the first time as tragedy, the second time as farce.” Both, then, cast specific doubt on the idea that history presents us with general patterns that can be projected into the future. Marx’s remarks to Vera Zasulich about the prospects for communist revolution in Russia are instructive: “I thus expressly limited the ‘historical inevitability’ of this process to the countries of Western Europe.”

This is a view I agree with very profoundly: history is contingent, there are always alternative pathways that might have been taken, and history has no general plan. So — no grand predictions in history.

But then we have to ask a different sort of question. Specifically — what kinds of predictions or projections are possible in history? And what is the intellectual base of grounded historical predictions? Here are a few predictions that seem to be supportable, drawn from recent postings on UnderstandingSociety:

  • The Alsatian language is likely to disappear as a functioning medium of communication in Alsace within the next fifty years.
  • Labor unrest in China will intensify over the next ten years.
  • Social unrest will continue to occur over the next decade in Thailand, with a gradual increase in influence to dispossessed groups (red shirts).
  • Large and deadly technology failures will occur in Europe and the United States in the next decade.
  • Social movements will arise more frequently and more adaptively as a result of the use of social media (twitter, blogs, facebook, email).
  • Conflicts between Arabs and Jews in East Jerusalem will continue to deepen in the next ten years as a consequence of the practical politics of land use and reclamation in the city.

Several things are apparent when we consider these predictions. First, they are limited in scope; they are small-scale features of the historical drama. Second, they depend on specific and identifiable social circumstances, along with clear ideas about social mechanisms connecting the present to the fruture. Third, they are at least by implication probabilistic; they indicate likelihoods rather than inevitabilities. Fourth, they imply the existence of ceteris paribus conditions: “Absent intervening factors, such-and-so is likely to occur.” But, finally, they all appear to be intellectually justifiable. They may not be true, but they can be grounded in an empirically and historically justified analysis of the mechanisms that produce social change, and a model projecting the future effects of those mechanisms in combination.

The heart of prediction is our ability to identify dynamic processes and mechanisms that are at work in the present, and our ability to project their effects into the future. Modest predictions are those that single out fairly humdrum current processes in specific detail, and derive some expectations about how these processes will play out in the relatively short run. Grand predictions, on the other hand, purport to discover wide and encompassing patterns of development and then to extrapolate their civilizational consequences over a very long period. A modest prediction about China is the expectation that labor protest will intensify over the next ten years. A grand prediction about China is that it will become the dominant economic and military superpower of the late twenty-first century. We can have a fair degree of confidence in the first type of prediction; whereas there are vastly too many possible branches in history, too many “countervailing tendencies,” too many accidents and contingencies, that may occur to give us any confidence in the latter prediction.

Ceteris paribus conditions are unavoidable in formulating historical expectations about the future, because social change is inherently complex and multi-causal. So even if it is case that a given process, accurately described in the present, creates a tendency for a certain kind of result — it remains the case that there may well be other processes at work that will offset this result. The tendency of powerful agents to seize opportunities for enhancing their wealth through processes of urban development implies a certain kind of urban geography in the future; but this outcome might be offset by a genuinely robust and sustained citizens’ movement at the city council level.

The idea that historical predictions are generally probabilistic is partly a consequence of the fact of the existence of unknown ceteris paribus conditions. But it is also, more fundamentally, a consequence of the fact that social causation itself is almost always probabilistic. If we say that rising conflict over important resources (X) is a cause of inter-group violence (Y), we don’t mean that X is necessarily followed by Y; instead, we mean that X raises the likelihood of the occurrence of Y.

So two conclusions seem justified. First, there is a perfectly valid intellectual role for making historical predictions. But these need to be modest predictions: limited in scope, closely tied to theories of existing social mechanisms, and accompanied by ceteris paribus conditions. And second, grand predictions should be treated with great suspicion. At their best, they depend on identifying a few existing mechanisms and processes; but the fact of multi-causal historical change, the fact of the compounding of uncertainties, and the fact of the unpredictability of complex systems should all make us dubious about large and immodest claims about the future. For the big answers, we really have to wait for the owl of Minerva to spread her wings.

The Perestroika debate in political science

A debate has been raging in the discipline of political science for at least a decade, over the nature of the scientific status and methods of the discipline. Fundamentally, the “dissidents” argue that a narrow and “scientistic” conception of what good political science research ought to look like has reigned and has repressed other, more pluralistic approaches to political science research. The formal methods of rational choice theory, game theory, and statistical analysis prevail, and the more narrative approaches associated with comparative research, area studies, and qualitiative research have been marginalized. And, the critics maintain, the flagship journals of the discipline and the tenure committees of the leading departments converge in maintaining this orthodoxy within the discipline. (Kristen Renwick Monroe has edited a valuable collection that gives the reader a pretty good understanding of the origins and faultlines of the debate; Perestroika!: The Raucous Rebellion in Political Science).

One of the central issues is this: what should a science of politics involve? What form of knowledge should political science produce? What is the role of universal laws or regularities in political science? How important are predictions?

Another key issue, related to the first, is the issue of the methodology of research that ought to be favored. Should quantitative methods be preferred? Should stylized assumptions be offered as the basis for formal rational-choice models of various forms of political behavior? What role should ethnographic research or case-study research play in the discovery of social-science knowledge?

Sanford Schram identifies some of the strands of the Perestroika critique in these terms: “Some focus on the overly abstract nature of much of the research done today, some on the lack of nuance in decontextualized, large-sample empirical studies, others on the inhumaneness of thinking about social relations in causal terms, and still others on the ways in which contemporary social science all too often fails to produce the kind of knowledge that can meaningfully inform social life” (Monroe : 103).

One of the most useful contributions to the Monroe book mentioned above is an essay by David Laitin. He takes issue with Bent Flyvbjerg’s book, Making Social Science Matter: Why Social Inquiry Fails and How it Can Succeed Again, and his advocacy of “phronesis”. Laitin characterizes the method of phronesis as one that is sensitive to context and that pays close attention to the singular and specific features of a particular social process — for example, the positioning that occurs as a city decides on its economic development strategy. So the method of phronesis is intentionally not aiming to discover regularities across a set of instances, but rather to uncover some specific features of a particular ongoing process.

Laitin argues that this approach is too narrow a foundation for social-science knowledge. He assimilates the phronesis method to what he calls a “narrative” approach; and he argues that good social science needs to use a three-fold methodology. Investigators should make use of the tools of narrative analysis; but they also need to use statistical methods (quantitative analysis across cases) and formal modeling (models of complex social situations based on assumptions along the lines of rational choice theory). Laitin refers to this approach as a “tripartite” method of comparative research.

Where does the philosophy of social science fit into this debate? I suppose that the philosophy of social science I have advocated has quite a bit in common with the criticisms raised by the Perestroikans. My views emphasize the contingency of social processes, lack of social regularities, multiple conjunctural causes at work, plasticity of social institutions, the value of ethnographic work, and the need for a plurality of methods of inquiry and explanation in the social sciences. And these views are at odds with the natural-science assumptions about how social phenomena ought to be investigated that the Perestroika group is criticizing. And some of the researchers whom I admire most deeply — James Scott, Charles Tilly, Benedict Anderson, Theda Skocpol, or Susanne Rudolph — are cited in the original Perestroika manifesto! At the same time, I am committed to the idea of empirical rigor, causal explanation, and making a connection between social science knowledge and practical social problems — a set of views that are post-positivist but still in the tradition of enlightened empiricism, and opposed to the currents of post-modern jargon that are sometimes mixed into the debate.

So the task is clear: to formulate a conception of social-science research and knowledge that preserves the values of empirical rigor and theoretial clarity, while embracing a pluralism that will permit the formulation of social-science knowledge adequate to the social world and social problems we find around ourselves. The Perestroika debate is an important one, and can help us better in the task of understanding society.

Chaos and coordination in social life

Much social behavior is chaotic, in that it simply emerges from the independent choices of numerous agents during a period of time. It is analogous to Brownian motion — particles in a liquid moving in random motions as a result of innumerable bumps and pushes at the molecular level.

However, there are also many patterns that become visible in social behavior — examples of what I would like to call “coordinated social action”: stock market panic selling, holiday travel, rumors, style, riots, pickpockets in train stations. And we can identify many causes of coordination of individual behavior into larger patterns: commands, regulations, institutions, customs, conventions, collective plans, shared beliefs about social behavior, common sources of information, and common changes in the environment of choice, for example.

What I mean by “coordination” here is the opposite of chaos — something analogous to the coherence of photons associated with the laser effect. In a laser a set of photons are stimulated to fire coherently with each other, resulting in a beam of light that possesses focus and parallel propagation that is different in kind from the scattered diffusion of photons from an incandescent bulb. “Coordinated” social action is a set of actions that possess synchronicity or regularity in their occurrence, resulting in an observable regularity of behavior over time and space. A crucial problem for social inquiry is to provide an explanation of the mechanisms that underlie the instances of coordinated social action that we can identify.

Examples of coordinated social action can easily be offered, and specific mechanisms can be identified that produce these forms of social coordination. An army moves in concert across a landscape (command). People drive on the right in North America (regulation). People send their children to school (institution). People greet each other with a polite “good morning” (custom). Villagers come together to fish as a group in the morning (convention). People discuss a spontaneous demonstration in front of the mayor’s office on Wednesday, and many appear (collective plans). Drivers choose Route 3 rather than Route 1 because they expect a lot of traffic on Route 1 (shared beliefs). People buy a large number of batteries and chocolate, anticipating an approaching hurricane (common environmental change).

These are all mechanisms that create a degree of coordination or synchronization of behavior among independent agents. There seem to be several large categories of mechanisms here: hierarchical coordination (command, regulation); common response coordination (each individually responds to the same signal); communications and network coordination (individuals exchange messages to secure coordination); and strategic coordination (each intends to behave in a way that will be desirable given his/her expectation of actions by others). Might we try out the thought that all forms of regularities of social behavior derive from one or another of these forms of coordination? This thought is probably somewhat too strong a claim. For example, there are probably social regularities that derive from our biology and evolutionary histories — limitations of memory, bonds of intra-group loyalty, kin altruism. But the impulse is a sound one: when we are able to observe patterns of social behavior, there must be a cause of those regularities that works its way through influence on the individual actors who constitute the domain of action. And there are only so many mechanisms that might serve.

These sorts of regularities and mechanisms constitute part of the regularity of social life, but perhaps only part. It may be that they don’t capture other kinds of more “structural” regularities — for example, “racial discrimination increases health disparities,” “feudal political systems are slow to respond to external aggression”, “capitalist market systems are more innovaative than planned economies”. But there is an important aspect of social explanation that centers exactly on this question: what are the social mechanisms that bring a degree of coherence and coordination among the actions of a population of independent actors?

The flea market analogy

Is the flea market a helpful analogy for understanding the social world (“The Dis-unity of Science”)? Does it serve to provide a different mental model in terms of which to consider the nature of social phenomena?

What it has going for it is heterogeneity and contingency, and an obvious share of agent-dependency. The people who show up on a given Saturday are a contingent and largely disorganized mix of humanity. And the products that wind up on the jumble tables too are highly disorderly and random. Each has its own unique story for how it got there. There is no overall guiding design.

But there is also a degree of order underlying the apparent chaos of the jumble tables. All is not random in a flea market. The participants, for example: there are regular vendors, street people, police officers, health inspectors, jugglers, and pickpockets — as well as regular shoppers, tourists, school children, and occasional shoppers looking for a used toaster or a single kitchen chair. In most cases there are reasons they are there — and the reasons are socially interesting. Moreover, the ethnographer of the flea market is likely enough to spot some seasonal or social patterns in the products and people present in a certain month or time of year. So — a blend of chaos and order.

But the order that can be discerned is the result of a large number of overlapping, independent conditions and processes — not the manifestation of a few simple forces or a guiding system of laws.

Both accident and order are characteristic of the larger social world as well. The helter-skelter of the flea market is in fact highly analogous to many aspects of social phenomena — army recruitment, incidents of crime, mortgage defaults. But it is also true that there are other social phenomena that aren’t so accidental. So the jumble sale is perhaps less good as an analogy for highly organized and managed social processes — a tight administrative hierarchy, an orchestrated campaign event, or a coordinated attack in battle.

This addresses the “accidental conjunction” part of the analogy. What about the “composite order” part of the analogy? This element too works pretty well for many examples of social phenomena. When students of the professions discover that there are interesting patterns of recruitment into accountancy or the officer corps, or discover that there are similarities in the organizations of pharmacists and psychotherapists — they also recognize that these patterns result from complex, intertwined patterns of strategic positioning, organizational learning, and economic circumstances. In other words, the patterns and regularities are themselves the result of multiple social mechanisms, motives, and processes. And these processes are in no way analogous to laws of nature.

So, all considered, the analogy of the flea market works pretty well as a mental model for what we should expect of social phenomena: a degree of accident and conjunction, a degree of emerging pattern and order that results from many independent but converging social processes, and an inescapable dimension of agent-dependency that refutes any hope of discovering an underlying, law-governed system.

The "dis"-unity of social science

One of the central goals of Vienna Circle philosophy of science was the idea of the unity of science. The idea included at least two separable parts: methodological unity and unity of content under a single system of laws. On the methodological side there was the idea that the logic of explanation and confirmation should be the same in all the empirical sciences. If there were to be differences across disciplines, these should be heuristic rather than epistemic differences. (Jordi Cat provides an extensive discussion of the unity of science doctrine in an article in the Stanford Encyclopedia of Philosophy.)

The more basic goal for unity was the idea of a single comprehensive theory that would, in principle, provide the foundation for the theories of all the special sciences. Physics was the intended foundation, and the goal was to demonstrate that all the fields of the natural sciences could be derived in principle from the laws of physics. For example, the hope was that the properties and laws of chemical elements and molecules should be derivable from fundamental physics.

The reasons for wanting to see a unified physical theory were a preference for parsimony and simplicity and a metaphysical conviction that all of nature must really derive from a single set of fundamental laws.

The fate of the unity of science doctrine can be pursued elsewhere. Here the question is whether there is a similar aspiration for the social sciences. The parallel principle could be stated along these lines: there should be some set of basic facts about individuals in social interactions that is sufficient to permit one to derive all varieties of social behavior, given relevant knowledge about context and boundary conditions.

The attractions of such a unified social science are the same as in the natural science case: parsimony, simplicity, and comprehensiveness. And, in fact, unifying theories for social explanation are sometimes advanced. The most thorough-going attempt is the effort by rational choice theory and microeconomics to unify all social action as the consequence of preference-maximizing individual rationality within constraints.

Attractive as this effort might be from an abstract or aesthetic perspective, it is profoundly misguided when if comes to understanding society. Social phenomena are not the law-governed consequences of a few underlying facts and features of individuals. Rather, they are the contingent and mixed results of an inherently heterogeneous set of motives, psychologies, and institutions. The fundamental problem is that the social world is not a system at all in the natural science sense. Instead, it is the contingent and dynamic sum of a variety of shifting processes and contexts.

A better metaphor for the social world — better than the metaphor of a table of billiard balls governed by the laws of mechanics — is a large urban flea market. The wares on sale on a particular Saturday are simply the sum of the accidents of circumstance that led a collection of sellers to converge on that particular day. There are some interesting regularities that emerge over time — in the spring one finds more used lawnmowers and in times of dearth one finds more family treasures. These regularities require explanation. But they do not derive from some governing “law of flea markets” that might be discovered. Instead, the flea market and the larger society are, alike, simply the aggregate result of large numbers of actions, motives, circumstances, and structures that turn kaleidoscopically and produce patterned but non-lawlike outcomes.

So where does this take us with regard to “unified social science”? It leads us to expect something else entirely: rather than unity, we should expect eclectic theories, piecemeal explanations, and a patchwork of inquiries at a range of levels of description. Some explanatory theories will turn out to be more portable than others. But none will be comprehensive, and the social sciences will always remain open-ended and extensible. Instead of theoretical unification we might rather look for a more and more satisfactory coverage, through a range of disciplines and methods, of the aspects of the social world we judge most interesting and important. And these judgments can be trusted to shift over time. And this means that we should be skeptical about the appropriateness of the goal of creating a unified social science.

(See an earlier posting on “Coverage of the Social Sciences” for more relevant comments on this topic.)

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.

Social "laws" and causal mechanisms

Are there social regularities? Is there anything like a “law of nature” that governs or describes social phenomena?

My view is that this is a question that needs to be approached very carefully. As a bottom line, I take the view that there are no “social laws” analogous to “laws of nature”, even though there are some mid-level regularities that can be discovered across a variety of kinds of social phenomena. But care is needed because of the constant temptation of naturalism — the idea that the social world should be understood in strong analogy with the natural world. If natural phenomena are governed by laws of nature, then social phenomena should be governed by “laws of society.” But the analogy is false.

Of course there are observable regularities among social phenomena. Urban geographers have noticed a similar mathematical relationship in the size distribution of cities in a wide range of countries. Durkheim noticed similar suicide rates among Catholic countries — rates that differ consistently from those found in Protestant countries. Political economists notice that there is a negative correlation between state spending on social goods and the infant mortality rate. And we could extend the list indefinitely.

But what does this fact demonstrate? Not, I want to argue, that social phenomena are “law-governed”. Instead, it results from two related facts. First, there are social-causal mechanisms; and second, there is some recurrence of common causes across social settings.
Take the mechanism of “collective action failures in the presence of public goods.” Here the heart of the mechanism is the analytical point that rationally self-interested decision-makers will take account of private goods but not public goods; so they will tend to avoid investments in activities that produce public goods. They will tend to become “free riders” or “easy riders.” The social regularity that corresponds to this mechanism is a “soft” generalization — that situations that involve a strong component of collective opportunities for creating public goods will tend to demonstrate low contribution levels from members of affected groups. So public radio fundraising will receive contributions only from a small minority of listeners; boycotts and strikes will be difficult to maintain over time; fishing resources will tend to be over-fished. And in fact, these regularities can be identified in a range of historical and social settings.

However, the “free rider” mechanism is only one of several that affect collective action. There are other social mechanisms that have the effect of enhancing collective action rather than undermining it. For example, the presence of competent organizations makes a big difference in eliciting voluntary contributions to public goods; the fact that many decision-makers appear to be “conditional altruists” rather than “rationally self-interested maximizers” makes a difference; and the fact that people can be mobilized to exercise sanctions against free riders affects the level of contribution to public goods. (If your neighbors complain bitterly about your smoky fireplace, you may be incentivized to purchase a cleaner-burning wood or coal.) The result is that the free-rider mechanism rarely operates by itself — so the expected regularities may be diminished or even extinguished.

What I draw from this is pretty simple. It is that social regularities are “phenomenal” rather than “governing”: they emerge as the result of the workings of common social-causal mechanisms, and social causation is generally conjunctural and contingent. So the regularities that become manifest are weak and exception-laden — and they are descriptive of outcomes rather than expressive of underlying “laws of motion” of social circumstances.

And there is a research heuristic that emerges from this discussion as well. This is the importance of searching out the concrete social-causal mechanisms through which social phenomena take shape. We do a poor job of understanding industrial strikes if we simply collect a thousand instances and perform statistical analysis on the features we’ve measured against the outcome variables. We do a much better job of understanding them if we put together a set of theories about the features of structure and agency through which a strike emerges and through which individuals make decisions about participation. Analysis of the common “agent/structure” factors that are relevant to mobilization will permit us to understand individual instances of mobilization, explain the soft regularities that we discover, and account for the negative instances as well.

A non-naturalistic approach to social science

The most basic error that is conveyed by the naturalist framework into the premises of sociology—the folk epistemology—that was shared by Durkheim, Mill, and Comte, is the assumption that all phenomena are subject to laws; that the relevant laws are abstract and obscure; and that there is an orderly relationship between gross phenomena and a rising level of natural laws that embrace those observable phenomena. The task of scientific study is to discover this rising pyramid of regularities and laws (motions of planets ENCOMPASSED BY elliptical orbits ENCOMPASSED BY gravitational attraction ENCOMPASSING other phenomena such as tides). This model is then used to frame the sociologist’s expectations of the orderliness of sociological observations and regularities. The social world is assumed to be a system of phenomena governed by hidden regularities and causal laws; the task of social science research is to discover these governing regularities and laws. . However, this conception of the world does not fit the domain of the social at all.

We can provide an alternative social ontology—a better grounding for sociological research. The social sciences could have begun with a greater degree of agnosticism about the orderliness of social phenomena. We could have started with the observations that—

  • Social phenomena are created by human beings (deliberately, intentionally, or unknowingly)
  • Human beings behave as a result of their socially constructed beliefs, values, goals, attitudes, modes of reasoning, emotions, …
  • There is a wide range of variation that is visible among social arrangements and institutions, across cultures, across space, and across time (long duration and short duration)
  • Social institutions, organizations, and structures have a degree of observable stability across cohorts and generations of the human beings who make them up
  • There are social causes, and they are ordinary, observable, and mundane. They are variants of the agent-structure nexus.

These initial ontological observations would have led us to some framing expectations about the social and about the likely results of social science inquiry:

  • contingency of social outcomes
  • Variation of social trajectory
  • Plasticity of social institutions
  • Heterogeneity among instances of a “type” of social thing
  • No “laws of motion” for development or modernization

And we might have set several research objectives for the social sciences:

  • To study in some detail how various institutions work in different social settings (empirical, fact-driven observation and analysis)
  • To study human behavior, motivation, and action – again, with sensibility to variation, without the assumption that there is one ultimate human nature or governing mode of behavior.
  • To be as aware of variation and plasticity as we are attentive to the discovery of social regularities
  • To discover and theorize some of the causal mechanisms that can be observed within social processes
  • To identify weak regularities of behavior and institution through observation
  • To theorize these regularities in terms of agent-structure dynamics; aggregation of features of decision-making; unintended consequences. For example, free rider phenomena (economists) and self-regulating commons (common-property resource institutions)

We then might have arrived at a different conception of what a “finished” social science might involve: not a deductive theory with a few high-level generalizations and laws, but rather an “agent-based simulation” that embodies as many of the characteristics and varieties of behavior as possible into the simulation, and then projects different possible scenarios. The ideal might have been “sim-society” rather than deductive-nomological theory.

%d bloggers like this: