Saturday, May 31, 2014

Heterogeneity according to Cartwright

Nancy Cartwright is one of the best philosophers of science around, in many people's opinion. I find her work particularly interesting for the new ways she offers of thinking about old ideas like "laws of nature" and the ways things work in the natural world. Much of what she writes about the entities and processes of the natural world is equally perceptive when applied to the social world. Especially interesting is her 1999 book, The Dappled World: A Study of the Boundaries of Science.

In The Dappled World Cartwright focuses on an idea that I've highlighted many times here as well -- the idea of the fundamental heterogeneity of the world. Here are the opening words of the book:
This book supposes that, as appearances suggest, we live in a dappled world, a world rich in different things, with different natures, behaving in different ways. The laws that describe this world are a patchwork, not a pyramid. They do not take after the simple, elegant and abstract structure of a system of axioms and theorems. Rather they look like — and steadfastly stick to looking like — science as we know it: apportioned into disciplines, apparently arbitrarily grown up; governing different sets of properties at levels of abstraction; pockets of great precision; large parcels of qualitative maxims resisting precise formulation; erratic overlaps; here and there, once in a while, corners that line up, but mostly ragged edges; and always the cover of law just loosely attached to the jumbled world of material things. (1)
She is particularly interested in demolishing the quest for scientific unity — a single unifying theory that can be said to represent the whole of a field of natural or social phenomena. She firmly rejects the idealized notion that quantum mechanics deductively encompasses all areas of physics, or that rational choice theory encompasses all areas of the social sciences. Instead, she argues that the “patchwork” nature of the disciplines of the sciences — different definitions of domain, different ideas about methodology and proof — corresponds in a deep way to the patchwork nature of the world. So methodology and ontology are intermingled.
The problem is that our beliefs about the structure of the world go hand-in-hand with the methodologies we adopt to study it. The worry is not so much that we will adopt wrong images with which to represent the world, but rather that we will choose wrong tools with which to change it. (12)
One way that Cartwright chooses to explain her “dappled” notion of the world is to insist that all scientific laws require ceteris paribus conditions. So scientific laws — even supposedly fundamental laws of mechanics like F = ma — do not apply unconditionally; rather, they apply subject to specific statements of boundary conditions and isolation conditions. A scientific experiment is designed in such a way as to exclude the workings of extraneous forces or influences; but Cartwright observes that in the real world of experience, we almost never observe this kind of isolation. Instead, baseballs are conveyed through parabolic arcs by mass, momentum, air currents, humidity, and fluid frictions — leading to a resultant arc which is only approximately described by the mathematical formula of the hyperbola (25-27).

The idea that the laws of nature always bring with them a set of ceteris paribus conditions is one way of pointing to the heterogeneous nature of the world. But a different way of characterizing the overall behavior of the spherical solid baseball is to refer to its nature — the inherent properties of the thing that work to produce its causal powers. The fact that it is a material object gives it a disposition to move in accordance with the laws of inertia and gravitation. The fact that it has a leather skin gives it a disposition to interact with surrounding fluids in ways that create patterns of micro-turbulence. The fact that its center of gravity is not at the geometrical center of the sphere gives it a tendency to wobble in flight. Each of these properties of the thing give separable tendencies to motion when the object is disturbed (hit with a baseball bat).

If we use the language of a thing's “nature” then the laws of nature are doubly secondary: they are approximate; and they derive from something more fundamental, the ensemble of natures that inhere in the objects in question.

Here are the three core ideas that she puts forward (4):
  1. The impressive empirical successes of our best physics theories may argue for the truth of these theories but not for their universality...
  2. Laws' where they do apply, hold only ceteris paribus....
  3. Our most wide-ranging scientific knowledge is not knowledge of laws but knowledge of the natures of things...
One consequence of Cartwright's views here, and her skepticism about the universality and generality of scientific laws, is a principled rejection of reductionism:
Not only do I want to challenge the possibility of downwards reduction but also the possibility of 'cross-wise reduction'. Do the laws of physics that are true of systems ... in the highly contrived environments of a laboratory or inside the housing of a modern technological device, do these laws carry across to systems, even systems of very much the same kind, in different and less regulated settings? (25)
And her answer to these questions is negative.

Cartwright's position has a lot in common with current work on causal powers, and this extends to her appeal to Aristotelian metaphysics as an alternative to Humean theories of regularities.
What kinds of facts (if any) determine the behaviour of a nomological machine? The Humean tradition, which finds nothing in nature except what regularly happens, insists that it must be further regularities. This chapter will continue the argument that laws in the sense of claims about what regularly happens are not our most basic kind of scientific knowledge. More basic is knowledge about capacities, in particular about what capacities are associated with what features. (77)
Where do laws of nature come from? ... It is capacities that are basic, and laws of nature obtain ... on account of the capacities; or more explicitly, on account of the repeated operation of a system of components with stable capacities in particularly fortunate circumstances. (49)
The idea of a nomological machine plays a central role in Cartwright's arguments; so what does this amount to? Cartwright is clear and explicit about this question:
What is a nomological machine? It is a fixed (enough) arrangement of components, or factors, with stable (enough) capacities that in the right sort of stable (enough) environment will, with repeated operation, give rise to the kind of regular behavior that we represent in our scientific laws. (50)
So a nomological machine is an important way of linking capacities and laws in Cartwright's account: it is a bundled set of special circumstances that give rise to the sorts of strong regularities that Humeans are looking for. But Cartwright's key point is that these circumstances are very special indeed: controlled laboratory setups and isolated technical devices are the examples she offers. This gives substance to the three core ideas articulated above: crucially, the construct illustrates the lack of universality and generality for scientific laws that she thinks is ineliminable in the study of nature (and society).

Cartwright develops these ideas largely in the context of the physical sciences. But she has been an important contributor to the philosophy of economics and the social sciences as well. And her skepticism about governing laws is even more compelling in the latter realms. As I tried to articulate the point in “On the Scope and Limits of Generalizations in the Social Sciences" (1993; link), we should think of regularities in the social world as phenomenal rather than governing. The social regularities we observe are the consequence of the workings of social mechanisms, and we should not imagine that there is an underlying set of governing laws that "generate" the social world.

Thursday, May 22, 2014

Positive social change

The City Year national leadership conference is taking place in Boston this week, celebrating the 25th anniversary of City Year. CY is a strong example of a social-change organization that is in the process of actually making an impact on a key social problem.

CY's goal is specific and audacious: to improve the high school graduation rate in the nation's poorest-performing schools. And it works on the basis of a data-driven plan of work: to place teams of corps members in challenged schools to provide near-peer support for at-risk children. There is good statistical evidence showing that this system works. The strongest implementation of this theory is a program called Diplomas Now (link).

City Year is an organization that is focused on the situation of low-income students and families. And in America this means that CY is also focused on addressing racial inequalities in our country. It is an organization that does a great job of embodying a philosophy and practice of multi-racial inclusiveness, both in the corps and in the organization. The icons of the organization are MLK, Gandhi, Mandela, and John and Robert Kennedy. (I was happy to see a great photo of Malcolm X and MLK in CEO Michael Brown's office in Boston.)

The young people who give a year or more in service are profoundly inspiring when you meet them: multi-racial, multi-class, and all committed to a unique kind of practical idealism. They express a wonderful empathy and connection with the children and young people whom they serve in some of our country's most disadvantaged schools. They live out the CY motto: "Give a year, change the world!"

Two things are unique about the City Year movement. One is its organizational structure. It is a national organization with a highly professional staff in Boston. The national organization provides the research capacity and strategic planning that the organization pursues throughout the country. And there are 25 cities in the US, as well as sites in South Africa and the UK, that have their own local staffs and committed site boards. This structure permits a great combination of national strategy and local adaptation and resource gathering that permits strong impact throughout the country. 

The other distinctive feature of the City Year movement is the sharp focus the leaders of City Year have placed on impact. It is not enough to mobilize several thousand young people in a year of service; it is crucial that their efforts should pay off in terms of important social outcomes. This focus on measurable impact has increased dramatically in the past 15 years. The organization has shifted from the development of the young people in the corps, to measurable improvement in raising the high school graduation rate in the country's most challenged schools. This results-oriented leadership and focus helps the organization move forward in impact and scope every year. 

So congratulations, Michael Brown and all the leaders and founders of this remarkable organization. Twenty-five years on and City Year is well on its way to closing the gap in opportunity and hope for urban young people throughout the United States. 

Saturday, May 17, 2014

Domain of the social sciences

Is the domain of "all social phenomena" a valid subject for scientific study? Is there a place for a purely general sociology, designed to be a theory of the social everything? Sociologists from Comte to Parsons have sometimes put forward this idea, and James Coleman pursued something like this in Foundations of Social Theory. But upon reflection, this seems like the wrong way of thinking about the social world.

Instead, the social world consists of a hodge-podge of actions, rules, organizations, motives, and the like, at a wide range of scales (link, link). A key task for social scientists is to segment social phenomena into related groups of actions or entities, with a scientific goal of making sense of how these various kinds of stuff work. So social scientists should permit themselves to be eclectic and specialized, identifying sub-domains of interest and uncovering the mechanisms and processes that are at work in this area of social life.

These considerations make it understandable that social research focuses on specific groups of social phenomena like these -- contention, organizations, racial discrimination, norm systems, market behavior, voting behavior, families, delinquency, .... These are all selected groups of social action and structures, similar enough to admit of a condensed description of what unifies them and plausibly explained by a similar set of causal mechanisms and processes. This provides a logic to the separation of the discipline of sociology into a large number of sub-disciplines; as of last count the American Sociological Association encompassed 52 sections, each organized around a set of research topics and methods (link).

Consider a possible analogy. Suppose we have just arrived on planet Earth and have noticed that we are surrounded by noises, and we want to understand this blooming, buzzing confusion. More exactly, we "hear" a wide range of phenomena with qualitative properties through our auditory sensory systems. And suppose we wanted to formulate a science of sound. How would we proceed?

We wouldn't begin, most likely, by collecting and cataloging the sounds that we perceive -- birds tweeting, jets roaring, children singing, elephants rumbling, ... No, we would begin by trying to determine what "sound" is (what the physical phenomenon is that underlies the phenomenology of sound through our hearing system). We will answer that question fairly quickly -- "sound is the result of vibrations in a medium in a certain range of frequencies, conveyed through a medium that extends from the vibrating object to the sensory detectors in our inner ears." And then the vast majority of sounds will be disregarded as scientifically uninteresting. The essentials have been identified: a source of vibration and a medium of transmission. More interesting will be the anomalies -- harmonics, differential speeds of propagation, echoes, Doppler effect, ... And we will be interested in the non-obvious mechanisms of sound production and transmission.

I'm not really interested in sounds, but rather in social phenomena. But there is a possible analogy here. The social world that we observe presents a bewildering variety of social phenomena. Where should we start in formulating a science of society? Perhaps the clue from acoustics and sound is this: we can ignore much of the phenomena, identify the surprising bits, and look for the mechanisms that underlie this surprising outcome or that. It is the fact of patterns and recurring surprises that will be of primary scientific interest.

There is a unifying feature of all sound phenomena — vibrations in a medium. And likewise, there is a unifying feature of all social phenomena — real human beings interacting with each other on the basis of their own mental maps of the world around them and conceptions of where they are trying to go, subject to constraints created by the natural environment and the behaviors and practices of individuals around them. But the real substance of research in both fields takes place at a more refined level, in which researchers seek to identify mechanisms and explanations for apparently anomalous outcomes.

What this thought experiment suggests is that we should not think of the subject matter of social science as the domain of all social action and social phenomena. Rather, much of what we observe in social life can be put in the category of noise, best filtered out as we focus our attention on surprising outcomes and mechanisms that permit of substantial illumination and explanation. The real subject matter is various bundles of related phenomena where we have succeeded in finding some important structural similarities and some relevantly similar causal mechanisms.

And we shouldn’t hold out the chimera of a unified social theory that explains all social phenomena — whether rational choice theory, world systems theory, or psychoanalysis.

*              *              *

This is the conclusion that I am inclined to support. But there is a different way of taking this analogy which comes back to giving greater support to James Coleman's aspirations in Foundations of Social Theory. If it is agreed that there is a unifying feature underlying all social behavior --
real human beings interacting with each other on the basis of their own mental maps of the world around them and conceptions of where they are trying to go, subject to constraints created by the natural environment and the behaviors and practices of individuals around them
-- then we might argue after all that "general sociology" is possible after all. It would consist of (a) discovery of the premises that govern the actors, (b) exploration of the most important ways in which actors interact, and (c) exploration of some of the specific boundary conditions within which (a) and (b) play out in the actual social world. Much of the variation represented in the list of the sections of the ASA above takes the form of enumeration of special cases in category (c). But the general science of sociology is contained in (a) and (b).

This gloss on the story also conforms to some extent to the science of acoustics: a common set of principles defining the physics and mathematics of the production and propagation of vibration through a medium, and a set of special cases where these principles produce odd results.

*           *            *

These contradictory conclusions point to a central tension within sociological theory between greater generalization and greater attention to context and variation. (This tension is highlighted in the recent post on Peggy Somers' rebuttal to Kiser and Hechter; link.) But I'm forced to acknowledge that this is also a tension in my philosophy of social science as well. The question of the availability of general theories in sociology relates to my own advocacy for the idea of methodological localism -- the idea that all social phenomena derive from the actions and thoughts of locally situated and locally constructed individual actors in proximate relations to other actors. "The “molecule” of all social life is the socially constructed and socially situated individual, who lives, acts, and develops within a set of local social relationships, institutions, norms, and rules" (link). But this formulation seems to leave it open that we might aspire to having general theories of the actor that permit us to explain social outcomes in terms of the features of action that are identified. And that seems to be the program that Coleman is exploring in Foundations.

Sunday, May 11, 2014

Are there meso-level social mechanisms?

It is fairly well accepted that there are social mechanisms underlying various patterns of the social world — free-rider problems, communications networks, etc. But the examples that come readily to mind are generally specified at the level of individuals. The new institutionalists, for example, describe numerous social mechanisms that explain social outcomes; but these mechanisms typically have to do with the actions that purposive individuals take within a given set of rules and incentives.

The question here is whether we can also make sense of the notion of a mechanism that takes place at the social level. Are there meso-level social mechanisms? (As always, it is acknowledged that social stuff depends on the actions of the actors.)

This question is analogous to two other similar issues in other special sciences:
  • Are there information-system level causal mechanisms in human cognition?
  • Are there cellular-level causal mechanisms in biological systems?
Or, to the contrary, are all mechanisms in sociology, cognition science, and biology properly understood to be carried out at the level of individuals, neurons, and biochemistry?

Here is my version of a definition of a causal mechanism (link):
A causal mechanism is (i) a particular configuration of conditions and processes that (ii) always or normally leads from one set of conditions C to an outcome O (iii) through the properties and powers of the events and entities in the domain of concern. 
And here is the definition offered by Doug McAdam, Sidney Tarrow, and Chuck Tilly in Dynamics of Contention:
Mechanisms are a delimited class of events that alter relations among specified sets of elements in identical or closely similar ways over a variety of situations. (kl 354)
We should begin by asking what it is that we are looking for. What would a meso-level mechanism look like?

Here is a start: it would be a linkage between two conditions or entities, each of which is itself a meso-level structure or entity. So a meso-level causal mechanism is one in which both C and O are meso-level entities or conditions and where C leads to O "always or normally".

Earlier I argued that meso-level entities possess causal powers: regular dispositions to produce specified effects, grounded in the substrate of social activity (link). If some of those effects Oi are themselves meso-level outcomes or structures, then our question here is answered. Any pair {C,Oi} is itself a meso-level causal mechanism. If, on the other hand, the causal powers of meso-level entities are restricted to changes in the behavior of individuals, then meso-level mechanisms do not exist.

McAdam, Tarrow, and Tilly address a very similar question in Dynamics of Contention, and they argue for what they call "relational mechanisms":
Relational mechanisms alter connections among people, groups, and interpersonal networks. Brokerage, a mechanism that recurs throughout Parts II and III of the book, we define as the linking of two or more previously unconnected social sites by a unit that mediates their relations with one another and/or with yet other sites. Most analysts see brokerage as a mechanism relating groups and individuals to one another in stable sites, but it can also become a relational mechanism for mobilization during periods of contentious politics, as new groups are thrown together by increased interaction and uncertainty, thus discovering their common interests. (kl 376)
Having formulated the question in these terms, it seems that we can provide a credible affirmative answer: it is possible to identify a raft of social explanations in sociology that represent causal assertions of social mechanisms linking one meso-level condition to another. Here are a few examples:
  • Al Young: decreasing social isolation causes rising inter-group hostility (link)
  • Michael Mann: the presence of paramilitary organizations makes fascist mobilization more likely (link)
  • Robert Sampson: features of neighborhoods influence crime rates (link)
  • Chuck Tilly: the availability of trust networks makes political mobilization more likely (link)
  • Robert Brenner: the divided sovereignty system of French feudalism impeded agricultural modernization (link)
  • Charles Perrow: legislative control of regulatory agencies causes poor enforcement performance (link)
We might also consider the possibility of compound meso-level mechanisms, in which M1 produces M2 which in turn produces M3. Does the sequence also qualify as a mechanism? That depends on the strength of the relationships that exist at each link; if the conditional probabilities of the links fall low enough, then the compound probability of the chain is no longer sufficient to satisfy condition (ii) above ("initial condition normally leads to the outcome").

Essentially this question comes down to the tightness of the linkages that exist among the sub-components of social systems. If there are sub-components within bureaucracies that maintain their properties and are tightly linked to specified outcomes, then these can play a role within meso-level causal mechanism narratives. If, on the other hand, the effects of a given subcomponent of a social system vary widely over time and space, then that type of component does not play a useful role in a causal mechanisms analysis. So the question of how extensive meso-level causal mechanisms are is itself an empirical one; it depends on the specific features of the social world.

So it seems as though we can offer two related conclusions about the causal reality of meso-level entities: meso-level structures possess causal powers, and there are causal mechanisms that invoke meso-level entities as both input and output.

Saturday, May 10, 2014

Peggy Somers' contribution to realism

Peggy Somers is an important contributor to the active field of sociological theory. And she identifies as a critical realist when it comes to understanding the logic and epistemology of the historical social sciences. Her views were extensively developed in “We’re no angels” (1998; link). The title picks up on the epistemology that she favors: non-perfectionist, anti-foundationalist, historically situated. In the essay she moves back and forth between post-Kuhnian philosophy of science and specific controversies in the methodology of the historical social sciences.

The essay takes its origin as a rebuttal to a critical review of the methodologies of historical sociology offered by Kiser and Hechter in “The role of general theory in comparative-historical sociology” (1991; link). A central target of Somers' arguments here is the idea defended by Kiser and Hechter that rational choice theory is the preferred theoretical framework for historical social science. Somers believes — as do most historical sociologists — that rational choice theory (RCT) is a legitimate but partial contribution to a pluralistic approach to historical sociology. Emphatically, it is too narrow a basis for constructing explanations of important large-scale historical movements and outcomes. So RCT advocates like Kiser and Hechter make the mistake of “theoretical monism” — imagining that a single theoretical premise might be sufficient to explain a large, complex domain of social phenomena.

A key theme in Somers’ treatment here is a contrast among several kinds of realism. Here is Somers' brief description:
All versions of realism accept that causal mechanisms—despite being unobservable—must be used as the basis of explanatory theoretical accounts; but only rational choice realism generates those mechanisms using on “ontic methodology” (Salmon 1984) in which the causal mechanisms of social explanation are postulated a priori from the same general theory that “guides” their research. (726)
And here are some key examples of what she means by causal mechanisms in the social world: "price mechanisms, maximizing preferences, class consciousness, value-driven intentionality, or domination" (726).

Somers identifies at least two kinds of realism -- what she calls "theoretical realism" and "relational and pragmatic realism". She favors the latter:
Relational realism posits that belief in the causal power of unobservables—such as states, markets, or social classes—does not depend on the rationality or truth of any given theory but upon practical evidence of its causal impact on the relationships in which it is embedded.... Relational realists believe that, while it is justifiable to theorize about unobservables, any particular theory entailing theoretical phenomena is historically provisional. For relational realism that means one can believe in the reality of a phenomenon without necessarily believing in the absolute truth or ultimate reality of any single theory that claims to explain it. (743-44)
And she believes that the two realisms have very different epistemological backgrounds -- deductivist and pragmatist:
Where the two realisms differ, then, is that while theoretical realism attributes an ontological truth to the theoretical phenomenon (e.g., the theory of electrons or the theory of market equilibrium), relational realism focuses on the relational effect of the phenomenon itself (e.g., the impact of the hypothesized electron on its environment or of the hypothesized market forces on an observable outcome). (745)
The most basic criticism that Somers offers of Kiser and Hechter is their mono-theoretic deductivism -- their claim that rigorous social science requires deductive derivation of a given social outcome from a theoretical premise. It is the theory that is at the heart of the explanation, according to this view of methodology. But for Somers -- and in the relational, pragmatist version of realism that she favors -- the ontology comes first. We may not know exactly what an electron is in detail; but we know the reality of electricity by the effects and causal properties we can probe practically and experimentally. This is the pragmatic aspect of her favored realism:
Social phenomena endure; but the “theoretical entities” that have purported to explain them are socially constructed—some more convincingly than others because they are more pragmatic and relational.
Somers faults what she calls theoretical realism for its commitment to explanation and confirmation through the hypothetic-deductive method. So what are the chief characteristics of her preferred alternative?

First, relational realism is “minimalist” --
[Relational realism is] minimalist because it recognizes that the partial concept- dependence of social life puts limits on the general realist premise of the absolute mind-independent status of the social world; yet realist nonetheless, in contrast to hermeneutics or radical constructivism in that some degree of concept-dependence does not in any way subvert the premise of a social world that exists independently of our beliefs about it. (766)
This amounts to an anti-foundationalist epistemology: we cannot establish the truth of all the premises and presuppositions of an explanation.

Second, relational realism is pluralistic; it encourages the discovery of multiple causal factors within a complex circumstance. This is in opposition to the theoretical monism of RCT supporters and is consistent with Robert Merton’s advocacy of a social science based on a search for theories of the middle range (link).

Third, relational realism is anti-essentialist; it recommends that the researcher should look at the social world as consisting of shifting configurations of social actors and institutions.
A relational ontology thus follows Popper’s rejection of essentialism and instead looks at the basic units of the social world as relational identities constituted in relational configurations. In place of a language of essences and inherent causal properties, a relational realism substitutes a language of networks and relationships that are not predetermined but made the indeterminate objects of investigation. (767)
An earlier post here raised a rather similar question about several kinds of realism, and the conclusions I reached were somewhat parallel as well. I offered support for scientific realism over critical realism. Here is the crucial passage:

So when we postulate that "class" is an important entity or structure in the modern world, our evidence for this claim is not largely based on inference to the best explanation and the overall success of class theory; it is instead the bundle of concrete researches that have been performed to identify, specify, and investigate the workings of class. Conceptual specification is more important that theoretical articulation and deduction: we need to know what a given researcher means to encompass in his or her use of the term "class structure". To take the photo above of Eton boys as an example -- what inferences can we draw about class from the photo? And what do we mean when we say that it illustrates an important social reality in the Britain of the 1930s, the reality of class? Is it a fact about attitudes; about the mechanisms of opportunity and selection; about the differential assignment of privilege; about modes of speech and thought?

My own philosophy of social science has several key features:
  1. I look at social science as inherently eclectic and pluralistic. There is no "best" method or "most fundamental" theory.
  2. I strongly suspect that social causation is fundamentally heterogeneous over multiple kinds of mechanisms and multiple temporalities. Outcomes are conjunctural, compositional, and contingent.
  3. I place a great deal of importance on empirical research and discovery. I am in that particular regard an enlightened "empiricist" about social and historical knowledge.
  4. I think there is an important place for theory and hypotheses in the social sciences. These need to be "theories of the middle range."
  5. I take an actor-centered approach to social theorizing. The substrate of the social world is individuals doing and thinking a range of things in various social settings.
  6. I am realist about a raft of social things: institutions, practices, value communities, social networks. All these social entities and structures exist as embodied in the thinking and acting of the socially constructed individuals who make them up, but they often have persistent and knowable properties that do not call for reduction to the micro level.
  7. I am realist about social causation, and I understand causation in terms of mechanisms.
  8. I am realist about the causal properties of at least some social entities -- structures, organizations, knowledge systems.
  9. I think ontology is important, but primarily at the level of the ontological assumptions implicated in various areas of scientific and historical research. Universal or philosophical ontology does not seem so important to me.
These commitments add up to a form of realism; but it isn't critical realism in the technical or substantive senses. It is a realism of a different stripe -- a pragmatic realism, a galilean realism, a scientific realism.

I wonder which of these premises Somers would endorse, and which she would criticize? I suspect that premise (5) will make her uneasy, given her desire to emphasize relationality in the social world; but that is certainly not ruled out in an actor-centered approach to social research. (This was also a contrast that Chuck Tilly drew between his approach to the social world and mine: "Dan, your approach is more individualistic than mine. I prefer to emphasize relations among the actors!")

Saturday, May 3, 2014

Liquid modernity?

image: Len Stomski, River Flow

Zygmunt Bauman advocates for the idea of "liquid modernity" (Liquid Modernity). This view emphasizes the fact of change within society; and it argues that change is occurring more and more rapidly in the "modern" world. Here is an observation by Bauman about the modern world:

Forms of modern life may differ in quite a few respects – but what unites them all is precisely their fragility, temporariness, vulnerability and inclination to constant change. To ‘be modern’ means to modernize – compulsively, obsessively; not so much just ‘to be’, let alone to keep its identity intact, but forever ‘becoming’, avoiding completion, staying underdefined. Each new structure which replaces the previous one as soon as it is declared old-fashioned and past its use-by date is only another momentary settlement – acknowledged as temporary and ‘until further notice’. Being always, at any stage and at all times, ‘post-something’ is also an undetachable feature of modernity. As time flows on, ‘modernity’ changes its forms in the manner of the legendary Proteus . . . What was some time ago dubbed (erroneously) 'post-modernity' and what I've chosen to call, more to the point, 'liquid modernity', is the growing conviction that change is the only permanence, and uncertainty the only certainty. A hundred years ago 'to be modern' meant to chase 'the final state of perfection' -- now it means an infinity of improvement, with no 'final state' in sight and none desired. (kl 82)

This is an evocative paragraph, but it is important to be specific in reading what it does and does not assert. Plainly Bauman underscores the fact of change in modern social life. "Change is permanent." But Bauman makes no assertion here about discontinuity, randomness, or complete absence of "structure". Indeed, the quotation explicitly allows that there are structures -- which implies that they persist over some period of time; and it observes, reasonably enough, that structures change and extinguish. As he puts the point a few pages later:

The original cause of the solids melting was not resentment against solidity as such, but dissatisfaction with the degree of solidity of the extant and inherited solids: purely and simply, the bequeathed solids were found not to be solid enough (insufficiently resistant or immunized to change) by the standards of the order-obsessed and compulsively order-building modern powers.... Flexibility has replaced solidity as the ideal condition to be pursued of things and affairs. (kl 97)

I have argued quite a few times in this blog for the idea that social phenomena are plastic, heterogeneous, and contingent. And these ideas seem to converge to some extent with the idea of a "liquid" society. Here are a few statements of these ideas:

On the core idea of plasticity of the social (link):

I maintain that virtually all social entities are "plastic": their properties change significantly over time, as a result of the purposive and unintentional behavior of the socially constructed individuals who make up a society. Organizations, labor unions, universities, churches, and social identities all show a substantial degree of flexibility and fluidity over time, and this fact leads to a substantial degree of heterogeneity among groups of similar social organizations and institutions. This points to a general and important observation about the constitution of the social world: The properties of a social entity or practice can change over time; they are not rigid, fixed, or timeless. They are not bound into consistent and unchanging categories of entities, such as "bureaucratic state," "Islamic society," or "leftist labor organization." Molecules of water preserve their physical characteristics no matter what. But in contrast to natural substances such as gold or water, social things can change their properties indefinitely. This interpretation interprets “plastic” as the contrary to “static and fixed”. A second way in which an entity might be unchanging is as a dynamic equilibrium. A social structure might be a self-correcting system that restores its equilibrium characteristics in the face of disturbing influences. The temperature in this room is subject to external influences that would result in change; but the thermostat provides cool or warm air as needed to bring the office temperature back to the equilibrium value. When I say that social entities are plastic, I also mean to say that they are not generally determined within a dynamic equilibrium (as sociological functionalism maintains, perhaps), with powerful homeostatic mechanisms that correct for disturbing influences. There is no “essential” form to which the structure tends to return in equilibrium.

On gradual institutional change (link):

The issue of the internal processes of change within an institution is of interest here for several reasons. But central among them is the idea of plasticity that has been described in earlier posts (link, link, link). The basic idea of plasticity is that institutions and organizations are the product of various kinds of structured human action, and that they can change over time. So we shouldn't think of institutions as having fixed characteristics, or as though they were equilibrium systems that tend to return to their original states after perturbances. Mahoney and Thelen's volume demonstrates some of the ways in which this plasticity emerges; they prove an account of the mechanisms of gradual institutional change. And this approach makes plain the high degree of path-dependency that institutions display.

On slow institutional change (link):

So what happens when institutions change? Mahoney and Thelen categorize gradual change into four types: displacement, layering, drift, and conversion (kl 444). And they argue that these categories are significant given the different roles that actors and strategies play in each of them. (This categorization seems to have something in common with the way geneticists and ecologists might characterize different modalities of adaptation within a changing environment.) They provide a 2x2 table that predicts the kind of adaptation that will occur, depending on combinations of strong/weak veto possibilities and low/high levels of discretion in interpretation of rules. For example, they assert that strong veto associated with high discretion produces drift rather than layering or conversion. They offer a similar analysis of different types of change agents, and attribute different kinds of strategies to the different categories of change agents. How does this framework relate to the topics of "actor-centered" social science and "meso-level causation" that have been considered in earlier posts? The theoretical framework Mahoney and Thelen describe is clearly actor-centered. They are focused on identifying the ways in which different categories of actors are empowered to interact with various features of a set of institutional rules. This picture seems to correspond to the ascending and descending links of the macro-micro analysis proposed by Coleman's boat.

On the contingencies of economic development identified by Sabel and Zeitlin (link):

One of Sabel and Zeitlin's most basic arguments is the idea that firms are strategic and adaptive as they deal with a current set of business challenges. Rather than an inevitable logic of new technologies and their organizational needs, we see a highly adaptive and selective process in which firms pick and choose among alternatives, often mixing the choices to hedge against failure. They consider carefully a range of possible changes on the horizon, a set of possible strategic adaptations that might be selected; and they frequently hedge their bets by investing in both the old and the new technology. "Economic agents, we found again and again in the course of the seminar's work, do not maximize so much as they strategize" (5).

On the plasticity of technical practices (link):

A particularly interesting question is the degree to which technical practices are “plastic” over time and space. How readily do they morph over time and space (akin to the way in which messages morph in the game of “telephone”)? Is there an analogy between a practice and a gene, in which the gene encodes instructions for the phenotype—producing a next-generation genotype? The stability of species through biological evolution depends on the fact that gene transcription is a highly accurate process, so the offspring is highly likely to encode the same bits of information as the parent. Is there the requisite stability within the domain of practices, or are we more likely to find significant differences in ostensibly similar practices across the villages of a region? The stability question turns on the mechanisms of replication that social practices embody. Traditional social practices are not embodied in standard "handbooks" of best practice; instead, they are transmitted through networks of training and imitation. So changes are likely to occur during the replication of the practice at the local level. Innovation occurs as local illiterate but intelligent farmers or builders discover enhancements. These innovations are imitated and reproduced by neighbors and changes accumulate. Naturally, there is nothing inherently optimal or progressive about such a process. Good ideas and innovations die out; mediocre practices persist; and sometimes genuine advances occur.

And on continuity (link):

It is evident that this expectation of gradual, continuous change is not always a valid guide to events. Abrupt, unexpected events occur -- revolutions, mass cultural changes like the 1960s, sweeping political and legislative changes along the lines of the Reagan revolution. And of course we have the current example of abrupt declines in financial markets -- see the graph of the Dow Jones Industrial Average for the week of September 23-30, 2008 below. So the expectation of continuity sometimes leads us astray. But continuity is probably among our most basic heuristic assumptions about the future when it comes to our expectations about the social world and our plans for the future.
The deeper question is an ontological one: what features of social causation and processes would either support or undermine the expectation of continuity? We can say quite a bit about the features of continuity and discontinuity in physical systems; famously, "non-linearities" occur in some physical systems that lead to singularities and discontinuities, but many physical systems are safely linear and continuous all the way down. And these mathematical features follow from the fundamental physical mechanisms that underlie physical systems. But what about the social world?

These arguments and others interspersed over the past eight years add up to a view of the realm of the social that attempts to bridge between stability and change, continuity and discontinuity, and to understand both structure and agency as the effects of social actors pursuing their plans. States, property systems, constitutions, religious institutions -- all these persisting social frames are themselves subject to strategic interventions by the actors who inhabit them, and their properties change over time.

So how can we resolve the apparent conundrum presented here: is the social world solid or liquid? Do structures constitute a more-or-less stable context for action, or is the social order simply a shifting play of forces and powers that resolves differently at different moments?

We know that the social world changes, and that these changes extend to deep structures as well as more superficial characteristics. Department stores change owners, so my local Marshall Fields becomes a Macy's. More fundamentally, shopping patterns change and downtown department stores close for good in favor of suburban shopping malls. And in the end Amazon eventually replaces them all. So both superficial and deep-structure characteristics of the social world change over time. Is this "liquid modernity" or is it "change overlaying structural persistence"?

The key to the answer lies in formulating a more nuanced understanding of the temporality of change. It is important not to assume all-or-nothing liquidity: either the social world is wholly labile, from moment to moment; or else there are "determinate" structures that are impervious to change. A better answer might be couched in terms of different temporal scales (along the lines of Paul Pierson's ideas; link). Rather than taking social structures or cultural systems as "static", we are better off thinking of these structural components as having a tempo of change that is many times the tempo of actors' choices. An investor considers purchasing a property in a neighborhood. He or she determines the current market price of the property but also takes into account the fact that market conditions change over time. So the real estate values of this neighborhood may rise or fall significantly in the medium term -- say, thirty years. This is not "permanence," but it is enough temporal range to admit of confident action.

If we took this point of view, then the relevant opposition is not between abiding structure and changing circumstance, but rather between long-duration change and short-duration change.

Introducing this change of language casts a bit of doubt on the validity of the metaphor of "liquid". Bauman emphasizes the instantaneous malleability of a liquid:
What all these features of fluids amount to, in simple language, is that liquids, unlike solids, cannot easily hold their shape. Fluids, so to speak, neither fix space nor bind time. While solids have clear spatial dimensions but neutralize the impact, and thus downgrade the significance, of time (effectively resist its flow or render it irrelevant), fluids do not keep to any shape for long and are constantly ready (and prone) to change it; and so for them it is the flow of time that counts, more than the space they happen to occupy: that space, after all, they fill but ‘for a moment’. (lc 311)
But this isn't quite right in application to the social world either -- any more than the idea that social institutions are "solid" and unchanging.  Instead, what we have is differing degrees of rigidity and different tempos of change attaching to different social structures. (Perhaps a more accurate metaphor is "stiff molasses" -- both liquid and solid!) And the work by institutional sociologists like Thelen and Pierson is specifically aimed at discovering the forces that lead to the stabilization of an institution or organization -- for a while.
There is another important area of uncertainty here as well. I don't yet have a clear opinion about Bauman's idea that the modern world is more plastic than other periods of social life. The arguments that I have offered for concluding that institutions, organizations, and value systems are plastic are as compelling for the Roman world as they are on Wall Street; so there is nothing in that argument that suggests that the modern world is more "liquid" than the ancient or medieval worlds. Perhaps it is so; but this actually seems to be a question for detailed historical research rather than philosophical speculation.


Thursday, May 1, 2014

Morphogenesis and realist meta-theory

Margaret Archer's contribution to critical realism has been an important part of the recent progress of the field, and her theory of morphogenesis is key to this progress. Her recent volume, Social Morphogenesis, represents a rigorous and serious step forward in the project of articulating this theory as both a meta-theory for the social sciences and a potential contribution to sociological theory. The volume includes two good essays by Archer, as well as contributions by Douglas Porpora, Andrea Maccarini, Tony Lawson, Colin Wight, Kate Forbes-Pitt, Wolfgang Hofkirchner, Emmanuel Lazega, Ismael Al-Amoudi, and Pierpaolo Donati.

Archer's philosophy of social science is intended to be a constructive contribution to the theory of critical realism. The central themes are profound social change and the generative mechanisms that produce it:
In concentrating upon morphogenesis we have elected to deal with 'those processes which tend to elaborate or change a system's given form, structure, or state' in preference to morphostatic processes 'that tend to preserve or maintain a system's form, organization, or state'. (2)
There seems to be general agreement that for any process to merit consideration as a generator of social change it must necessarily incorporate structured human relations (context-dependence), human action (activity-dependence) and human ideas (concept-dependence). Necessarily, the three make social theorising non-naturalistic. (4)
She writes about the kind of explanation of the distinctive rapidity of modern social change that she and her collaborators are looking for:
We agree that satisfactory explanation cannot be at the level of experience (the empirical level) or at the level of events (the actual level) but needs to identify a real mechanism whose exercise, even in the open system that is the social order, is responsible for the intensification of social change. (2)
The volume emphasizes one specific aspect of the social world, what Archer highlights as the increasing speed of social change.
This book is about theorising a possible transition from the social order of late modernity. What we examine is the generative mechanism of 'social morphogenesis', held to account for the increasing rapidity of social change. (1)
Archer and other contributors point towards a novel emerging kind of society -- the Morphogenic Society, but their primary emphasis is on the process of morphogenesis rather than the outcome of that process.

Here is how Emmanuel Lazega characterizes the central ideas of the project:
The goal of the Morphogenetic Society project is to develop an account of social stability and change at the macro-level in late modernity. It is thus different from the Morphogenetic Approach, as an explanatory framework presented as appropriate for analysis at all levels from the micro- to the macro-level and at all times. According to this perspective, three elements are always involved in any social transformation--big or small: 'structure', 'culture', and 'agency'. The challenge is always to specify their interplay as the basis of explanation for the stability or change of any social phenomenon chosen by the investigator, when using the Morphogenetic Approach or in exploring the notion of Morphogenetic Society. (167)
Key here is the idea of seeking out "generative mechanisms" of social change. What would be an example of such a mechanism? Archer refers to "struggles for domination and control" (7) as a generative mechanism, and later she refers to "conflicting pressures of primary and corporate agency" (14). In each instance structures, rules, and organizations are understood as being malleable and subject to the pushes and pulls of actors within current circumstances. Here is her summary of three major generative mechanisms:
At the macroscopic level (third-order), the generative mechanism is held to derive from 'Contingent Compatibilities' coming to predominate societally for the first time.... At the institutional (second-order) level, however, we confront the paradox of various institutions seeking to take advantage of such synergy whilst also retaining the situational logic of competition.... At the (first-order) level, agents (individual or collective) and actors confront rapidly changing structural and cultural contexts in daily life and across generations. (20)
Mechanisms like these are deeply indeterminate -- an advantage of Archer's theory, in my view; so social outcomes are unpredictable. She underlines that indeterminacy in the final words of the introduction:
However, the transition to and stabilisation of a new Morphogenic social formation ultimately hangs upon system integration and social integration not only increasing but coming into a relationship of mutual regulation -- and that is the most problematic condition of all for transformation. (21)
Emmanuel Lazega's abstract for his 2014 article "‘Morphogenesis Unbound’ from the Dynamics of Multilevel Networks: A Neo-structural Perspective" is helpful in coming to better understand the thrust of the morphogenesis approach:
One way to understand the notion of Morphogenesis Unbound is to focus on the meso level of society, i.e. to look at society as an ‘organizational society’ and to think about the co-evolution of structure, agency and culture – the three dimensions of Archer’s sociology, analytically speaking – in that context. This co-evolutionary vision happens to be very close to the research program of neo-structural sociology. To illustrate this insight, one neo-structural method, multilevel network analysis through linked design, is applied to a set of empirical data so as to propose a network translation of Morphogenesis Unbound and observe its outcome. This chapter reports results in which actors create new relationships beyond the boundaries of the organization with which they are affiliated, thus reshaping/expanding their own personal opportunity structure beyond the limitations imposed upon them by pre-existing structures. Half the population of the innovators observed (here: highly competitive scientists) deploy ‘independentist’ strategies, i.e. all the new personal ties that they develop in their network among the elite of colleagues of their profession are beyond the constraining perimeter predefined by their organization’s inter-organizational network. The kind of organization that they might create would not establish inter-organizational ties with their current organization. Over time, measurements suggest that this independence takes them close to Nowhere in terms of further achievements. Slightly more pedestrian forms of Morphogenesis, i.e. perhaps less Unbound, based on a relational strategy called here ‘individualist’, in which actors keep a strong foot in the organization in which they are affiliated so as to use its resources to create a new set of ties – and eventually a new organization – outside their current organization’s perimeter, seem to be of a more rewarding kind of networks to Somewhere closer to the “prizes [that] go to those who will explore and can manipulate contingent cultural compatibilities to their advantage” (Archer 2012). In this latter case, even if some of the opportunities that they could create for themselves are hoarded by their current organization (or boss). Such neo-structural measurements of Morphogenesis are used to start thinking about situations in which the two generative mechanisms identified by Archer (2012), competition and opportunity, coexist; as differentiated from the situations in which the latter would replace the former. Indeed creating new ties with heterogeneous actors, beyond one’s current position and sometimes even new kinds of organizations, is a highly cultural form of agency. Breiger’s notion of ‘weak culture’ helps speculate about actors’ capacity to reshape opportunity structures by reaching heterogeneous alters in spite of resistance from a rather stable, change-averse, tightly-connected organizational society promoting ordinary incremental innovation that will not challenge pre-existing entrenched interests. (link)
This brief description highlights several generative mechanisms of morphogenesis. Lazega offers several more examples in his analysis of social networks in the current volume (chapter 9); for example, he analyzes the effects of "advice networks" among the members of the Commercial Court of Paris, leading to transformation of the structure of the relational networks that exist among these experts over time (171 ff.).

The contributors to this effort have generally chosen to take a highly abstract perspective on the issues they address, conforming to the idea that morphogenesis is intended to be a meta-theory rather than a theory. But it seems to me that philosophical theorizing about the social world and about the social sciences need to be linked more closely to actual problems of social research. (Lazega's contribution does in fact make these more direct connections.) Is it possible to say how the morphogenetic approach might be thought to shed light on concrete problems of historical sociology -- for example, why the American Civil Rights movement took the shape that it did in the 1950s and 1960s, why European fascism developed as it did, why the Green Movement seems to have stalled in Germany, or how it is that anarchist mobilization against globalization has been as successful as it was for a period of 15 years? Are there particular problems of sociological research and explanation that we can better solve by immersing ourselves in the theories of morphogenesis?