Guest post by Guus Duindam

Guus Duindam is a J.D./Ph.D. student in philosophy at the University of Michigan. His primary areas of interest are Ethics and Kant. Thanks, Guus, for providing this rigorous treatment of Bhaskar's philosophical argument for critical realism.

Bhaskar contra Kant: Why Critical Realism is not Transcendental Realism

Let me start by thanking Dan Little for inviting me to write this guest-post. I’d like to take the opportunity to examine Roy Bhaskar’s arguments for critical realism, in particular those presented in his A Realist Theory of Science (RTS). The aim of that work is remarkable: to establish by transcendental argument the mind-independence and structured nature of the objects of science.

Bhaskar’s views are explicitly grounded in Kantian arguments. But the rejection of Kantian transcendental idealism is a central feature of Bhaskar’s critical realism. For Bhaskar, critical realism is also transcendental realism, a position he posits as an alternative to both Kantian and (neo-)Humean philosophy of science.

Transcendental idealism is, at minimum, the idea that the conditions on human cognition – especially space and time, the forms of human intuition – in part determine the objects of knowledge. According to transcendental idealism, we cannot know things as they are ‘in themselves’, but rather only as they appear to beings like us. Kant thus distinguishes between things-in-themselves, the epistemically inaccessible noumena, and phenomena, things as they appear to us given the conditions on human cognition. The former are transcendentally real – unknowable but entirely mind-independent. The latter are empirically real – knowable, but in part dependent on the conditions on cognition. For Kant, science can study only the empirically real: to study the transcendentally real would require that we transcend the conditions on our own cognition – that we erase the distinction between the knower and the object of knowledge – a mystical feat of which we are evidently incapable.

Bhaskar makes a different distinction, between the intransitive and the transitive. Intransitive objects do not depend on human activity; they are entirely mind-independent (RTS 21). To say that some object is intransitive is therefore equivalent to saying that it is transcendentally real (this is clear throughout RTS; see also The Possibility of Naturalism 6). Hence, it is Bhaskar’s aim to prove the transcendental reality (intransitivity) of the objects of science and perception. According to Bhaskar, we can know the objects of science as they are in themselves.

Bhaskar defends this ambitious thesis by means of transcendental arguments. An argument is transcendental insofar as it shows that some commonly accepted claim x necessarily presupposes a controversial claim y; where y is the conclusion of the argument. Thus, a transcendental argument claims that its conclusion is the only possible way to account for the uncontroversial phenomenon which it takes as its premise. Unlike other arguments for scientific realism, then, Bhaskar’s make a claim to necessity.

Bhaskar’s analysis of perception contains the first of his transcendental arguments: call it the argument from perception. It has roughly the following form: multiple agents can, at the same time, perceive the same object in different ways (x). This could be possible only given the mind-independence of the object (y). Therefore, given the occurrence of differential perception, the objects of perception must be transcendentally real.

Here’s Bhaskar himself making the argument:

If changing experience of objects is to be possible, objects must have a distinct being in space and time from the experience of which they are the objects. For Kepler to see the rim of the earth drop away, while Tycho Brahe watches the sun rise, we must suppose that there is something they both see. (RTS, 31)

Earlier, he appears to be making the even stronger claim that perception simpliciter presupposes the intransitivity of the perceived:

The intelligibility of sense-perception presupposes the intransitivity of the object perceived. For it is in the independent occurrence or existence of such objects that the meaning of ‘perception’, and the epistemic significance of perception, lies. (Ibid.)

Let’s take the argument from perception to involve the weaker claim that differential experience by different agents necessarily presupposes the intransitive nature of the object perceived. If the argument fails to ground this claim, we know a fortiori that it fails to ground the stronger conclusion.

If it is possible for Brahe and Kepler to have different perceptions of the same object, there must be an object which they both see: this much seems clear. But the inference from this to the object’s intransitivity is fallacious, for the presupposition that the objects of sense-perception are empirically real is sufficient to explain differential perception. For the transcendental idealist, there is something which Brahe and Kepler both see: they both see the sun. The sun is empirically real, i.e., it partially depends on the conditions on human cognition. But Brahe and Kepler, being human, share the conditions on cognition and interact with the same mind-independent reality. Thus, there is nothing unintelligible about their different perceptions under the assumption that what they perceive is empirically real (partially mind-dependent). Bhaskar supposes that we must assume it is also transcendentally real (i.e., that Brahe and Kepler see the sun ‘as it is in-itself’) but does nothing to establish this. The argument from perception does not show that the objects of knowledge must be intransitive given the occurrence of (differential) perception. It fails as a transcendental argument for critical realism.

Bhaskar’s second argument is much more central to the critical realist endeavor, and it is presented in his analysis of experimental activity. Call it the argument from experimentation. For Bhaskar, “two essential functions” are involved in an experiment:

First, [the experimental scientist] must trigger the mechanism under study to ensure that it is active; and secondly he must prevent any interference with the operation of the mechanism. […] Both involve changing or being prepared to change the ‘course of nature’, i.e. the sequence of events that would otherwise have occurred. […] Only if the mechanism is active and the system in which it operates is closed can scientists in general record a unique relationship between the antecedent and consequent of a lawlike statement. (RTS, 53)

Bhaskar notes that the experimenter who sets up a causally closed system thereby becomes causally responsible for a constant conjunction of events, but not for the underlying causal mechanism. Contra Humean accounts of law, Bhaskar’s account of experimentation entails an ontological distinction between constant conjunctions and causal mechanisms.

For Bhaskar, the intelligibility of such experimental activity can be used to transcendentally establish the intransitivity of the objects of science. “As a piece of philosophy,” he claims, “we can say (given that science occurs) that some real things and generative mechanisms must exist (and act),” where by ‘real’ Bhaskar means ‘intransitive’ (RTS 52). In “Transcendental Realisms in the Philosophy of Science: On Bhaskar and Cartwright,” Stephen Clarke provides the following helpful gloss on the argument:

Premise 1: Scientific explanatory practice (in particular the practice of exporting explanations from laboratory circumstances to general circumstances) is experienced by us as intelligible. 

Premise 2: Scientific explanatory practice could not be experienced by us as intelligible unless causal powers exist and those causal powers are governed by universal laws of nature.

______________________________________________________ 

Conclusion: causal powers exist and are governed by universal laws of nature. (Clarke 302)

Clarke calls this an “attack on idealism” (303) but Bhaskar explicitly frames it as an attack on transcendental idealism (RTS 27). Clarke’s gloss is telling, for it is indeed unclear how the argument could work as an attack on the latter view.

Bhaskar argues that we must suppose the world to be intransitively ordered if scientific explanatory practice is to be intelligible. But, he claims, “transcendental idealism maintains that this order is actually imposed by men in their cognitive activity” (RTS 27). And if order were imposed in cognitive activity, all experience would be ordered, eliminating the need for explanatory export from the closed causal systems of experimentation to the open causal systems of uncontrolled experience (RTS 27, Clarke 303).

This argument is invalid. It does not follow from the premise that all experience is ordered that there is no need for explanatory export from closed to open causal systems. To the contrary: the very occurrence of such export presupposes that experience is ordered. After all, the aim of experimentation is to discover causal mechanisms and universal laws of nature. But to suppose that the causal mechanism discovered in a replicable scientific experiment generalizes to open causal systems is to suppose that the same laws operate in open causal systems, even if other mechanisms sometimes obscure them. And to presuppose that there are such things as knowable universal laws of nature – operative in closed and open causal systems alike – just is to presuppose that all experience is ordered. The ordered nature of experience is, therefore, a necessary presupposition for experimentation.

Now there are at least two ways in which experience could be thus ordered: because order is imposed on it in cognitive activity, or because the order is intransitive. Bhaskar supposes the former would render experimentation superfluous. This is a flummoxing claim to make. Surely Bhaskar does not mean to accuse the transcendental idealist of the view that the projection of order onto the world is somehow a conscious activity – that we already know every scientific truth. That would render experimentation superfluous, but I don’t think it is a view anybody defends. Science is as much a process of gradual discovery for the Kantian as it is for everyone else.

Maybe confusion arises from the fact that for Kantians genuinely universal scientific laws must be synthetic a-priori. Perhaps Bhaskar supposes that, because positing a universal law involves making a claim to synthetic a-priori knowledge, we should be able to derive the laws of nature by a-priori deduction, rendering experimentation superfluous. But this would be a misunderstanding of transcendental idealism. Suppose that because my perceptions of sparks and wood are frequently followed by perceptions of conflagration, I come to associate sparks and wood with fire. I can ask whether this association is subjective or objective. To claim that it is objective is, for the Kantian, to apply one of the Categories. For instance, one way of taking my association of sparks and dry wood with fire to be objective is to make a claim like “sparks and wood cause fire,” applying the Category of causation. This claim is a-priori insofar as it involves the application of an a-priori (pure) concept, a-posteriori insofar as it is about the objects of experience.

Transcendental idealism entails we are entitled to make causal claims, but it does not entail the empirical truth of our claims. Experimentation with sparks and wood may lead me to modify my claim. For instance, I may discover that sparks and wet wood do not jointly give rise to fire, and adjust my claim to “sparks and dry wood cause fire.” Further experimentation may lead to further refinements. I could not have deduced any of these conclusions about sparks and wood a-priori. The thesis that scientific claims have an a-priori component does not render experimentation either superfluous or unintelligible.

As it turns out, Bhaskar supposes that, for the Kantian, causal mechanisms are mere “figment[s] of the imagination” (RTS 45). If true, this would provide an independent argument against the intelligibility of experimentation on a transcendentally idealist account. But, as should by now be clear, this is an incorrect characterization of transcendental idealism. It is only for skeptics and solipsistic idealists that causal mechanisms are figments of the imagination. Kantians and transcendental realists agree causal mechanisms exist: they disagree only about whether they are transcendentally or empirically real.

Bhaskar’s transcendental arguments for critical realism fail, and the Kantian view to which Bhaskar opposes his own is frequently misinterpreted. Most problematically, the meaning of the Kantian distinction between the transcendentally and empirically real is ignored, and the latter category is treated as if it contained only figments of our imagination. Bhaskar maintains that epistemic access to the transcendentally real is a necessary condition for science and perception. But, as we have seen, it is merely epistemic access to the empirically real that is necessary. Bhaskar does not prove that we have knowledge of things as they are in-themselves. Critical realism is not transcendental realism.

Sociology of life expectations

Each individual has a distinctive personality and orienting set of values. It is intriguing to wonder how these features take shape in the individual's development through the experiences of childhood, adolescence, and early adulthood. But we can also ask whether there are patterns of mentality and orienting values across many or most individuals in a cohort. Are there commonalities in the definition of a good life across a cohort? Is there such a thing as the millennial generation or the sixties generation, in possession of distinctive and broadly shared sets of values, frameworks, and dispositions?

These are questions that sociologists have attempted to probe using a range of tools of inquiry. It is possible to use survey methodology to observe shifts in attitudes over time, thereby pinpointing some important cohort differences. But qualitative tools seem the most appropriate for this question, and in fact sociologists have conducted extensive interviews with a selected group of individuals from the indicated group, and have used qualitative methods to analyze and understand the results.

A very interesting example of this kind of research is Jennifer Silva's Coming Up Short: Working-Class Adulthood in an Age of Uncertainty. Silva is interested in studying the other half of the millennial generation -- the unemployed and underemployed young people, mostly working class, whom the past fifteen years have treated harshly. What she finds in this segment of the cohort born in the late 1970s and early 1980s is an insecure and precarious set of life circumstances, and new modes of transition to adulthood that don't look very much like the standard progress of family formation, career progress, and rising affluence that was perhaps characteristic of this same social segment in the 1950s.

Here is how Silva frames the problem she wants to better understand:

What, then, does it mean to “grow up” today? Even just a few decades ago, the transition to adulthood would not have been experienced as a time of confusion, anxiety, or uncertainty. In 1960, the vast majority of women married before they turned twenty-one and had their first child before twenty-three. By thirty, most men and women had moved out of their parents’ homes, completed school, gotten married, and begun having children. Completing these steps was understood as normal and natural, the only path to a complete and respectable adult life: indeed, half of American women at this time believed that people who did not get married were “selfish and peculiar,” and a full 85 percent agreed that women and men should get married and have children (Furstenberg et al. 2004). (6)

Silva is interested in exploring in detail the making of "working class life adulthood" in the early twenty-first century. And her findings are somewhat bleak:

Experiences of powerlessness, confusion, and betrayal within the labor market, institutions such as education and the government, and the family teach young working-class men and women that they are completely alone, responsible for their own fates and dependent on outside help only at their peril. They are learning the hard way that being an adult means trusting no one but yourself. (9)

At its core, this emerging working-class adult self is characterized by low expectations of work, wariness toward romantic commitment, widespread distrust of social institutions, profound isolation from others, and an overriding focus on their emotions and psychic health. Rather than turn to politics to address the obstacles standing in the way of a secure adult life, the majority of the men and women I interviewed crafted deeply personal coming of age stories, grounding their adult identities in recovering from their painful pasts—whether addictions, childhood abuse, family trauma, or abandonment—and forging an emancipated, transformed, and adult self. (10)

Key to Silva's interpretation is the importance and coherence of the meanings that young people create for themselves -- the narratives through which they make sense of the unfolding of their lives and where they are going. She locates the context and origins of these self-stories in the structural circumstances of the American economy of the 1990s; but her real interest is in finding the recurring themes in the stories and descriptions these young people tell about themselves and their lives.

For Silva, the bleakness of this generation of young working class adults has structural causes: economic stagnation, dissolution of safety nets, loss of decent industrial-sector jobs and the rise of insecure service-sector jobs, and neoliberalism as a guiding social philosophy that systematically turns its back on under-class young people. It is sobering that her research is based on interviews carried out in a few cities in the United States, but the findings seem valid for many countries in western Europe as well (Britain, Germany, France). And this in turn may have relevance for the rise of populism in many countries as well.

What is most worrisome about Silva's account is the very limited opportunities for social progress that it implies. As progressives we would like to imagine our democracy has the potential of evolving towards greater social dignity and opportunity for all segments of society. But what Silva describes is unpromising for this hopeful scenario. The avenues of higher education, skills-intensive work, and better life circumstances seem unlikely as a progressive end of this story. And the Sprawl of the grim anti-utopian novels of William Gibson (Neuromancer, Count Zero) seem to fit the world Silva describes better than the usual American optimism about the inevitability of progress. Significantly, the young people whom Silva interviews have very little interest in engagement in politics and supporting candidates who are committed to real change; they do not really believe in the possibility of change.

It is worth noticing the parallel in findings and methodology between Silva's work on young working class men and women and Al Young's studies of inner city black men (The Minds of Marginalized Black Men: Making Sense of Mobility, Opportunity, and Future Life Chances). Both fall within the scope of cultural sociology. Both proceed on the basis of extensive interviews with 50-100 subjects, both make use of valuable tools of qualitative analysis to make sense of the interviews, and both arrive at important new understandings of the mentalities of these groups of young Americans.

(Here is a prior post on cultural sociology and its efforts to "get inside the frame" (link); and here is a post on "disaffected youth" that touches on some of these themes in a different way; link.)

Explanation and critical realism

To explain something is to provide a true account of the causes and circumstances that brought it about. There is of course more to say on the subject, but this is the essential part of the story. And this normative account of explanation should work as well for investigations created within the framework of critical realism as any other scientific framework.

Moreover, CR is well equipped with intellectual resources to produce explanations of social outcomes based on this understanding. In particular, CR emphasizes the reality of causal mechanisms in the social world. To explain a social outcome, then -- perhaps the rise of Trumpism -- we are instructed to identify the causal mechanisms and conditions that were in play such that a novice from reality television would gain the support of millions of voters and win the presidency. So far, so good.

But a good explanation of an outcome is not just a story about mechanisms that might have produced the outcome; instead, we need a true story: these mechanisms existed and occurred, they brought about the outcome, and the outcome would not have occurred in the absence of this combination of mechanisms. Therefore we need to have empirical methods to allow us to evaluate the truth of these hypotheses.

There is also the important and interesting point that Bhaskar makes to the effect that the social world involves open causal configurations, not closed causal configurations. This appears to me to be an important insight into the social world; but it makes the problem of validating causal explanations even more challenging.

This brings us to a point of contact with the theme of much current work in critical realism: a firm opposition to positivism and an allegiance to post-positivism. Because a central thrust of positivism was the demand for substantive empirical confirmation or verification of substantive claims; and that is precisely where we have arrived in this rapid analysis of explanation as well. In fact, it is quite obvious that CR theories and explanations require empirical validation no less than positivistic theories. We cannot dispense with empirical validation and continue to believe we are involved in science.

Put the point another way: there is no possible avenue of validation of substantive explanatory hypotheses that proceeds through purely intuitive or theoretical avenues. At some point a good explanation requires empirical assessment.

For example, it is appealing in the case of Trumpism to attribute Trump's rise to the latent xenophobia of the disaffected lower working class. But is this true? And if true, is it critical as a causal factor in his rise? How would we confirm or disconfirm this hypothetical mechanism? Once again, this brings us into proximity to a few core commitments of empiricism and positivism -- confirmation theory and falsifiability. And yet, a rational adherence to the importance of empirical validation takes us in this direction ineluctably.

It is worth pointing out that the social and historical sciences have indeed developed empirical methods that are both rigorous and distinctive to the domain of the social: process tracing, single-case and small-N studies, comparative analysis, paired comparisons, and the like. So the demand for empirical methods does not imply standard (and simplistic) models of confirmation like the H-D model. What it does imply is that it is imperative to use careful reasoning, detailed observation, and discovery of obscure historical facts to validate one's hypotheses and claims.

Bhaskar addresses these issues in his appendix on the philosophy of science in RTS. He clearly presupposes two things: that rigorous evidence must be used in assessment of explanatory hypotheses in social science; and flat-footed positivism fails in providing an appropriate account of what that empirical reasoning ought to look like. And, as indicated above, the open character of social causation presents the greatest barrier to the positivist approach. Positivism assets that the task of confirmation and refutation concerns only the empirical correspondence between hypothesis and observation.

Elsewhere I have argued for the piecemeal validation of social theories and hypotheses (link). This is possible because we are not forced to adopt the assumption of holism that generally guides philosophy in the consideration of physical theory. Instead, hypotheses about mechanisms and processes can be evaluated and confirmed through numerous independent lines of investigation. Duhem may have been right about physics, but he is not right about our knowledge of the social world.

Cacophony of the social

Take a typical day in a major city -- a busy street with a subway stop, a park, a coffee bar, and a large consumer financial office. There are several thousand people in view, mostly in ones and twos. Some people are rushing to an appointment with a doctor, a job interview, a drug dealer in the park. A group of young men and women are beginning to chant in a demonstration in the park against a particularly egregious announcement of government policy on contraception.

There is a blooming, buzzing confusion to the scene. And yet there are overlapping forms of order -- pedestrians crossing streets at the crosswalks, surges of suits and ties at certain times of day, snatch and grab artists looking for an unguarded cell phone. The brokers in the financial office are more coordinated in their actions, tasked to generate sales with customers who walk in for service. The demonstrators have assembled from many parts of the city, arriving by subway in the previous hour. Their presence is, of course, coordinated; they were alerted to the demo by a group text from the activist organization they belong to. 

What are the opportunities for social science investigation here? What possibilities exist for explanation of some of the phenomena on display?

For one thing there is an interesting opportunity for ethnographic study presented here. A micro-sociologist or urban anthropologist may find it very interesting to look closely to see what details of dress and behavior are on display. This is the kind of work that sociologists inspired by Erving Goffman have pursued.

Another interesting possibility is to see what coordinated patterns of behavior can be observed. Do people establish eye contact as they pass? Are the suits more visibly comfortable with other suits than with the street people and panhandlers with whom they cross paths? Is there a subtle racial etiquette at work among these urban strangers?

These considerations fall at the "micro" end of the spectrum. But it is clear enough that the snapshots we gain from a few hours on the street also illustrate a number of background features of social structure. There is differentiation among actors in these scenes that reflects various kinds of social inequalities. There are visible inequalities of income and quality of life that can be observed. These inequalities in turn can be associated with current activities -- where the various actors work, how much education they have, what schools they attended, their overall state of health. There are spatial indicators of interest as well -- what kinds of neighborhoods, in what parts of the city, did these various actors wake up in this morning?

And for all of these structural differentiators we can ask the question, what were the social mechanisms and processes that performed the sorting of new-borns into affluent/poor, healthy/sick, well educated/poorly educated, and so forth? In other words, how did social structure impose a stamp on this heterogeneous group of people through their own distinctive histories?

We can also ask a series of questions about social networks and social data about these actors. How large are their personal social networks? What are the characteristics of other individuals within various individual networks? How deep do we need to go before we begin to find overlap across the networks of individuals on the street? This is where big data comes in; Amazon, credit agencies, and Verizon know vastly more about these individuals, their habits, and their networks than a social science researcher is likely to discover through a few hundred interviews. 

I'd like to think this disorderly ensemble of purposive but uncoordinated action by several thousand people is highly representative of the realities of the social world. And this picture in turn gives support to the ontology of heterogeneity and contingency that is a core theme here. 

Organizational learning

 

I've posed the question of organizational learning several times in recent months: are there forces that push organizations towards changes leading to improvements in performance over time? Is there a process of organizational evolution in the social world? So where do we stand on this question?

There are only two general theories that would lead us to conclude affirmatively. One is a selection theory. According to this approach, organizations undergo random changes over time, and the environment of action favors those organizations whose changes are functional with respect to performance. The selection theory itself has two variants, depending on how we think about the unit of selection. It might be hypothesized that the firm itself is the unit of selection, so firms survive or fail based on their own fitness. Over time the average level of performance rises through the extinction of low-performance organizations. Or it might be maintained that the unit is at a lower level -- the individual alternative arrangements for performing various kinds of work, which are evaluated and selected on the basis of some metric of performance. On this approach, individual innovations are the object of selection. 

The other large mechanism of organizational learning is quasi-intentional. We postulate that intelligent actors control various aspects of the functioning of an organization; these actors have a set of interests that drive their behavior; and actors fine-tune the arrangements of the organization so as to serve their interests. This is a process I describe as quasi-intentional to convey that the organization itself has no intentionality, but its behavior and arrangements are under the control of a loosely connected set of actors who are individually intentional and purposive. 

In a highly idealized representation of organizations at work, these quasi-intentional processes may indeed push the organization towards higher functioning. Governance processes -- boards of directors, executives -- have a degree of influence over the activities of other actors within and adjacent to the organization, and they are able to push some subordinate behavior in the direction of higher performance and innovation if they have an interest in doing so. And sometimes these governance actors do in fact have an interest in higher performance -- more revenue, less environmental harm, greater safety, gender and racial equity. Under these circumstances it is reasonable to expect that arrangements will be modified to improve performance, and the organization will "evolve".

However, two forms of counter-intentionality arise. The interests of the governing actors are not perfectly aligned with increasing performance. Substantial opportunities for conflict of interest exist at every level, including the executive level (e.g. Enron). So the actions of executives are not always in concert with the goal of improving performance. Second, other actors within the organization are often beyond control of executive actors and are motivated by interests that are quite separate from the goal of increasing performance. Their actions may often lead to status quo performance or even degradation of performance. 

So the question of whether a given organization will change in the direction of higher performance is highly sensitive to (i) the alignment of mission interest and personal interest for executive actors, (ii) the scope of control executive actors are able to exercise over subordinates, and (iii) the strength and pervasiveness of personal interests among subordinates within the organization and the capacity these subordinates have to select and maintain arrangements that favor their interests.

This represents a highly contingent and unpredictable situation for the question of organizational learning. We might regard the question as an ongoing struggle between local private interest and the embodiment of mission-defined interest. And there is no reason at all to believe that this struggle is biased in the direction of enhancement of performance. Some organizations will progress, others will be static, and yet others will decline over time. There is no process of evolution, guided or invisible, that leads inexorably towards improvement of arrangements and performance.

So we might formulate this conclusion in a fairly stark way. If organizations improve in capacity and performance over time in a changing environment, this is entirely the result of intelligent actors undertaking to implement innovations that will lead to these outcomes, at a variety of levels of action within the organization. There is no hidden process that can be expected to generate an evolutionary tendency towards higher organizational performance. 

(The images above are of NASA headquarters and Enron headquarters -- two organizations whose histories reflect the kinds of dysfunctions mentioned here.)

Social change and leadership

Historians pay a lot of attention to important periods of social change -- the emergence of new political movements, the development of a great city, the end of Jim Crow segregation. There is an inclination to give a lot of weight to the importance of leaders, visionaries, and change-makers in driving these processes to successful outcomes. And, indeed, history correctly records the impact of charismatic and visionary leaders. But consider the larger question: are large social changes amenable to design by a small number of actors?

My inclination is to think that the capacity of calculated design for large, complex social changes is very much more limited than we often imagine. Instead, change more often emerges from the independent strategies and actions of numerous actors, only loosely coordinated with others, and proceeding from their own interests and framing assumptions. The large outcome -- the emergence of Chicago as the major metropolis of the Midwest, the forging of the EU and the monetary union, the coalescence of nationalist movements in France and Germany -- are the resultant of multiple actors and causes. Big outcomes are contingent outcomes of multiple streams of action, mobilization, business decisions, political parties, etc.

There are exceptions, of course. Italy's political history would have been radically different without Mussolini, and the American Civil War would probably have had a different course if Douglas had won the 1860 presidential election. 

But these are exceptions, I believe. More common is the history of Chicago, the surge of right-wing nationalism, or the collapse of the USSR. These are all multi-causal and multi-actor outcomes, and there is no single, unified process of development. And there is no author, no architect, of the outcome. 

So what does this imply about individual leaders and organizations who want to change the social and political environment facing them? Are their aspirations for creating change simply illusions? I don't think so. To deny that single visionaries cannot write the future does not imply they cannot nudge it in a desirable direction. And these effects can indeed alter the future, sometimes in the desired direction. An anti-racist politician can influence voters and institutions in ways that inflect the arc of his or her society in a less racist way. This doesn't permanently solve the problem, but it helps. And with good fortune, other actors will have made similar efforts, and gradually the situation of racism changes. 

This framework for thinking about large social change raises large questions about how we should think about improving the world around us. It seems to imply the importance of local and decentralized social change. We should perhaps adjust our aspirations for social progress around the idea of slow, incremental change through many actors, organizations, and coalitions. As Marx once wrote, "men make their own history, but not in circumstances of their own choosing." And we can add a qualification Marx would not have appreciated: change makers are best advised to construct their plans around long, slow, and incremental change instead of blueprints for unified, utopian change.