Mechanisms and powers

source: William Bechtel, Discovering Cell Mechanisms: The Creation of Modern Cell Biology

The causal-powers approach to the understanding of causation is sometimes presented as an exclusive alternative to both traditional regularity theories and to more recent causal mechanism theories. In an earlier post I discussed Ruth Groff’s contributions to this topic. Here I would like to present a provocative view: that the causal mechanisms and causal powers are complementary rather than contradictory. The causal mechanisms theory benefits by being supplemented by a causal powers theory and the causal powers theory benefits by being supplemented by a causal mechanisms theory. In other words, the two theories are not exclusive alternatives to each other, but rather serve to identify different parts of the whole of causation.

The causal powers theory rests on the claim that causation is conveyed from cause to effect through the active powers and capacities that inhere in the entities making up the cause. The causal mechanisms theory comes down to the idea that cause and effect are mediated by a series of events or interactions that lead (typically) from the occurrence of the cause to the occurrence of the effect. In other words, cause and effect are linked by real underlying causal sequences (often repeatable sequences).

My thesis of the mutual compatibility of powers and mechanisms goes along these lines. If we press down on a putative mechanisms explanation, we are led eventually to postulating a set of causal powers that provide the motive force of the postulated mechanisms. But equally, if we press down on the claim that a certain kind of entity has a specified causal power or disposition, we are led to hypotheses about what mechanisms are set in play be its constituents so as to bring about this disposition.

Begin with a causal mechanism story:

• C => {x happens bringing about y, bringing about z, bringing about u, which is E} => E

How is it that the sub-links of this chain of mechanism pieces happen to work to bring about their consequent? We seem to have two choices: We can look to discover a further underlying mechanism; or we can postulate that the sub-link entity or structure has the power to bring about its consequent. So if we push downward within the terms of a mechanism explanation, one way to close the story is by postulating a causal power at some level.

Now start with a causal power claim. Suppose we assert that:

• Salt has the causal power of making H2O electrically conductive when dissolved.

Is this simply an unanalyzable fact about salt (or saline solution)? It is not; instead, we can look downward to identify the physical mechanisms that are brought into play when salt enters solution in H2O. That mechanism is well understood: the Na+ and Cl- ions created by the dissolution of salt permit free electrons to pass through the solution.

So we can explain the causal power by discovering the causal mechanism that gives rise to it; we explain links in the putative mechanism by alluding to the powers of the entities involved at that stage; and we can explain other things by referring to the causal powers that we have discovered to be associated with various kinds of things and structures.

If we take this set of possibilities seriously, then powers and mechanisms are answering different questions within the causal nexus. The reference to powers answers the question, “What does x do?”, while the reference to mechanisms answers the question, “How does x work?"

From a scientific point of view, it is always legitimate to ask how the powers of an entity or structure come to be in the natural world. What is it about the micro-structure of the thing in virtue of which the thing’s properties are established? In fact, this is one of the key intellectual challenges of the sciences. And this is a request for specification of some of the mechanisms that are at work. But likewise, it is always legitimate to ask what gives force to a given mechanism; and here we are eventually driven back to the answer, “some of the components of the mechanism have X, Y, Z powers to affect other entities” without further analysis within that particular explanation.

One might imagine that there are primitive causal powers — powers attached to primitive particles that have no underlying components or mechanisms.  We might begin to give a list of primitive causal powers: mechanical interactions among physical objects (transfer of momentum from one particle to another); electromagnetic properties inhering in one object and creating forces affecting other objects; gravitational forces among objects possessing mass; the causal interactions that occur within the central nervous system. And we might seek to demonstrate that all causal powers depend on combinations of these sorts of "primitive" causal powers -- a kind of Hobbesian materialism.  But this is needlessly strenuous from a metaphysical point of view. Better is to consider the middle-level range of powers and mechanisms where we are able to move upwards and downwards in our search for underlying causal mechanisms and supervening causal powers.

This line of thought suggests that questions about the metaphysics of causation are perhaps less pressing than they are sometimes made out to be. A thing's powers are not irreducible attributes of the thing; rather, they are the orderly consequence of the composition of the thing and the causal properties of those components and their interactions. It is hard to see that much turns on whether we think of the world as consisting of entities with powers, or as composites with system properties created by their components. The key question seems to be something like this: what is implied when we make a causal assertion? Both CP and CM agree that the core implication is the idea that one event, structure, or condition brought about the occurrence of another event, structure, or condition.  And the languages of both powers and mechanisms do a pretty good job of expressing what we mean in asserting this implication.

(John Dupré takes a similarly ecumenical view about several approaches to the theory of causation in a recent article, "Living Causes", where he advocates for what he calls "causal pluralism"; link. He writes: "I believe that causality is a complex and diverse set of phenomena, and most or all of these accounts provide valuable and complementary perspectives on the topic. Such a pluralistic view is quite a common one among contemporary philosophers; however there are significant differences in the form that such pluralisms can take" (20). On the mechanisms side within the philosophy of biology is William Bechtel's Discovering Cell Mechanisms: The Creation of Modern Cell Biology, who writes: "Beginning in the 1940s an initially small cadre of investigators who were pioneers in the modern discipline of cell biology began to figure out the biochemical mechanisms that enable cells to perform these functions. although miniaturized, the mechanisms they found to be operative in each cell are staggeringly complex" (1 ).)

Causality and metaphysics

Advocates of the causal powers approach attach a great deal of importance to the metaphysics of causation -- the sorts of properties and relations that we attribute to the kinds of things that we want to postulate. The neo-Aristotelian point of view represented by Ruth Groff and others appears to have metaphysical objections to the causal-mechanisms approach: the CM approach postulates the wrong kind of relations among entities, according to this group. So if I want to argue that mechanisms and powers are compatible, as I do, then I need to take into account the metaphysical arguments. It will be necessary to tell a story about the nature of the world that gives a place and meaning to the metaphysical premises of each theory.

The possibility of fundamental metaphysical incompatibility cannot be trivially ruled out. Consider this different kind of example: the distance between the premises of analytical Marxism and the neo-Hegelian theory of internal relations espoused by Bertell Ollman in Alienation: Marx's Conception of Man in Capitalist Society (link). Even if there were the possibility of some degree of convergence in conclusions about capitalism -- e.g. the likelihood of recurring crises -- the two schools of thought differ fundamentally on the nature of social entities and structures. They differ in terms of their social metaphysics. Analytical Marxists take the view that the structures of capitalism are the composite effect of variously motivated individuals; so there is an underlying atomism in the ontology of AM. Causes are fully distinct from the things they affect. Ollman, by contrast, believes that we need to conceive of the structures and social relations of capitalism relationally: the wage labor relation is not an atomistic relation between capitalist and worker, but rather a mutually implicating set of relations between the two that cannot be fully separated. Here is a passage in which Ollman attempts to capture the distinctive features of Marx's social metaphysics:

What is distinctive in Marx's conception of social reality is best approached through the cluster of qualities he ascribes to particular social factors. Taking capital as the example, we find Marx depicting it as "that kind of property which exploits wage-labor, and which cannot increase except on condition of getting a new supply of wage-labor for fresh exploitation" (Marx and Engels, 1945, 33). What requires emphasis is that the relation between capital and labor is treated here as a function of capital itself, and part of the meaning of "capital". This tie is extended to cover the worker as well, where Marx refers to him as "variable capital" (Marx, 1958, 209). The capitalist is incorporated into the same whole: "capital is necessarily at the same time the capitalist . . . the capitalist is contained in the concept of capital" (Marx, 1973, 512). Elsewhere, Marx asserts that "the means of production monopolized by a certain section of society", "the products of laborers turned into independent powers", "money", "commodities" and even "value that sucks up the value creating powers" are also capital (Marx, 1959b, 794-5; Marx, 1958, 153; Marx, 571). What emerges from these diverse characterizations is a conception of many tied facets, whose sense depends upon the relations Marx believes to exist between its components: property, wage-labor, worker, his product, commodities, means of production, capitalist, money, value (the list can be made longer still). (Chapter 2, section ii)

This example demonstrates the possibility of a genuine and deep incompatibility between two social theories at the level of the assumptions they make about the nature of the world -- their metaphysical theories.

So what about causal powers and causal mechanisms? The primary metaphysical commitment that the CP theorists advocate derives from their treatment of powers and essences -- two characteristic ideas from Aristotle. A power is thought to inhere in a thing in a particularly deep way; it is not an accidental expression of the empirical properties of the thing but rather an essential and active expression of the nature of the thing. The causal powers theory comes down to the idea that things and structures have an active capacity to bring about certain kinds of effects. In Groff’s terms, things are not passive but rather active.

Here is how Tuukka Kaidesoja introduces the metaphysical framework of critical realism in relation to causality in Naturalizing Critical Realist Social Ontology. Kaidesoja finds that the concept of a causal power is fundamental to critical realism (105). A thing's power is the characteristic of the world through which causal influence arises; without the concept of causal power, we would indeed be stranded in a Humean world of pure constant conjunction. Kaidesoja quotes Harre and Madden in these terms:

"X has the power to A" means "X (will)/(can) do A, in the appropriate conditions, in virtue of its intrinsic nature". (Kaidesoja, 106)

So what about the metaphysics of the causal mechanisms theory? Generally speaking, advocates of the mechanisms approach have not been very interested in the metaphysical issues. They (we) are generally realist, so we postulate that there are real causal interactions. This is indeed a metaphysical position. But this family of thinkers tends to be mid-range realists: they want to understand the necessity of causal relations at one level as deriving from the real workings of the physical or social system a bit lower down; but they generally don't seem to want to pose the ultimate question: how could any event or structure exert causal influence on another? So the causal mechanisms theorists are perhaps better described as scientific realists rather than philosophical or critical realists. They take the view that the world has the properties (approximately) that our best scientific theories attribute to things. (Could we call them "Galilean realists"?)

Curiously enough, this contrast seems to have a lot to do with the quibble I raised for Ruth Groff in the earlier post: whether powers should be thought to be "irreducible". Scientific realists would say they are not irreducible; rather, we can eventually arrive at a theory (molecular, genetic, economic, psychological, rational choice, physical) that displays the processes and mechanisms through which the ascribed power flows from the arrangement and properties of the thing.

On this standpoint, powers are attributions we make to things when we don't know quite enough about their composition to work out the physics (or sociology) of the underlying mechanisms. They do attach to the entity or structure in question, surely enough; but they do so in virtue of the physical or sociological composition of the entity, not because of some inherent metaphysical property.

We might try to reconcile these two perspectives with a few simple ideas:

1. Entities and structures at a range of levels of being have causal powers: active capacities to influence other entities and structures.
2. Whenever we identify a causal power of a thing, it is always open to us to ask how this power is embodied; what it is about the inner constitution of the entity that gives it this power.
3. When we succeed in arriving at a good scientific answer to this question, we will have shown that the power in question is not irreducible; it is rather the consequence of a set of mechanisms set in play by the constitution of the entity.

So the discovery of a given causal power of a thing is not a metaphysical fundamental; it is rather an empirical scientific discovery that invites analysis into its underlying composition.

Causal powers from a metaphysical point of view

A number of scholars who are interested in causation have recently expressed new interest in the concept of causal powers. This makes sense in a very straightforward and commonsensical way. But it also raises some difficult questions about metaphysics: how are we to think about the underlying nature of reality such that things, events, or conditions have "causal powers"? These questions raise issues that a number of talented philosophers are now taking on in a systematic way. Particularly interesting are recent writings by Ruth Groff, who represents a wave of contemporary thinking in metaphysics that aims to revitalize portions of Aristotle's views of causation in opposition to Hume's.

Groff's work on causal powers is sustained over a number of recent works, including especially her 2012 book Ontology Revisited: Metaphysics in Social and Political Philosophy (Ontological Explorations), her introduction and chapter in Greco and Groff, eds., Powers and Capacities in Philosophy: The New Aristotelianism, and her contribution to Illari, Russo, and Williamson, Causality in the Sciences. Groff emphasizes a broad clash of perspectives between a Humean theory of causation ("constant conjunction, no necessary relations among things or events") and a neo-Aristotelian theory ("things have powers, powers underlie causal relations among things and events"). Here is how she and Greco put the perspective of the "New Aristotelianism" in Powers and Capacities:

Humeanism is now under serious pressure within analytic metaphysics. In particular, after having been dismissed for generations as so much antiquated animism, the loosely-Aristotelian theorizing of real causal powers has now come to be a major focus of research within the specialty. (kl 204)

Moreover, Groff believes that American social sciences are still largely in the grip of the Humean metaphysics.  In "Getting Past Hume" in Causality in the Sciences, she writes:

One can't help but wonder what the outcome would actually be, were there to be a floor-fight on the question, i.e., a substantive debate within analytic philosophy and methodology of social science on the merits of Humean anti-realism about causality versus the merits of a powers-based, realist account of causality.

...

What is significant about all of this for my own argument is not so much that Humeanism continues to be the default ontology of especially American, often positivist, social science; but rather that it can be combined with the idea that it is not -- i.e., with the idea that competing versions of regularity theory somehow differ in a deep way, or that it is possible to remain neutral on what causality is, whilst engaging in causal explanation.

This poses a stark contrast; either you are positivist, anti-realist, and Humean or you are anti-positivist, realist, and neo-Aristotelian.  However, it is worth observing that there seem to be two currents of realist thought that reject Humean causation, not just one: the powers ontology that Groff (and Mumford and Anjum in Getting Causes from Powers) advocate; and the causal-mechanisms approach that has been advocated by philosophers and sociologists such as Hedstrom, Elster, and Ylikoski under the broad banner of analytical sociology. One might take the view that the causal-mechanisms approach ultimately requires something like the powers ontology -- "How else are we to account for the fact that sparks cause gasoline to explode?"; but on its face, these are two fairly independent realist responses to Hume. And certainly it is difficult to find a neo-Aristotelian predilection among the causal-mechanism advocates.

Groff takes up causal mechanisms theory in "Getting Past Hume" in  Causality in the Sciences. She concedes that this approach -- in the hands of Jon Elster and in my own writings, for example -- claims to be realist and anti-positivist, in that it rejects the notion that explanation depends on the discovery of general laws. But she doesn't think that the causal-mechanisms approach actually succeeds in presenting a substantive alternative to the Humean framework on causation: "Upon closer examination, the mainstream mechanisms model is more of the same, metaphysically."

Fleshing out her argument, she seems to be arguing that causal-mechanisms theory can either retreat to constant conjunction (at the level of the linkages of individual causal mechanisms) or it can press forward to a causal powers interpretation; there is no third possibility. "As with the other models, nothing on the mainstream mechanisms model is in a position of actually doing anything, in the sense of actively producing an effect. Thus here too, with an extra bit of ironic panache, the explanation-form functions as a delivery mechanism (no pun intended) for a Humean metaphysics." And it is true that most definitions of causal mechanisms make some kind of reference to regularities and repeatability.

My own formulation of the mechanisms theory is one of the targets of Groff's critique. And in fact I can reconstruct my reasons for thinking that mechanisms need to involve some kind of regularities; and I don't think it implies a collapse onto Humean causation. (At one point I wanted to call them "pocket regularities", to distinguish them from the grand social or psychological laws that Hempel and Mill seemed to want to discover.)  I wanted to assert that:

"M [information diffusion] is the mechanism connecting E [police beating] with O [rapid mobilization of an angry crowd]" is a description of a real underlying (perhaps unknown) causal process through which the features of E bring about the occurrence of O.

This is an ontological claim and it is a realist claim. But there is also an epistemic issue: How would we know that M is indeed such an underlying reality? It seems unavoidable that we would need to either produce empirical evidence supporting the conclusion that M frequently conveys these kinds of effects in these kinds of circumstances (the approach Tilly takes) or we need to have a theory of the mechanism which accounts for how it works to bring about the effect. The first boils down to a discovery of a limited set of regularities in a range of circumstances; the latter is a theoretical demonstration of how it works.  So this way of conceptualizing mechanisms does indeed invoke regularities of some sort.  However, it doesn't agree with the Humean idea that causation is nothing but regularities or constant conjunction. The regularities that are invoked are symptoms of the underlying causal mechanism, not criterial replacements for the mechanism.

Moreover, the powers theory seems to be subject to the same possible objection: how do we know that lightning has the causal power of starting barns on fire, unless we have repeatedly observed the chain of events leading from lightning strike to blaze?

Another thing that demands more attention is an assumption about the implications of the "realism" of powers. What follows from the idea that things have real causal powers? Groff puts the view in these terms: to assert that powers are real is to assert that they are irreducible (kl 204). But that seems questionable. We may think that feudalism was real, while at the same time thinking that its properties and dynamics derived from more fundamental social relations that compounded to create the distinctive dynamics of feudalism. So it doesn't seem that realists have to also accept the idea of irreducibility of the things about which they are realist. Or to put the point the other way around: the idea that realism implies irreducibility appears to also imply a fairly strong thesis about emergence. Groff returns to this set of ideas in Ontology Revisited, chapter three: "An emergent phenomenon (property or entity) is one that is not equivalent, ontologically, to the plurality of its parts." And here too she emphasizes irreducibility. But, as Poe Yu-ze Wan shows in "Emergence a la Systems Theory: Epistemological Totalausschluss or Ontological Novelty?" in Philosophy of the Social Sciences, philosophers have differed on the question of whether "emergence" implies "irreducibility" (link). The theory of emergence offered by Mario Bunge does not require irreducibility.

One thing I particularly like about Groff's work on causal powers is her persistence in working through the logical and conceptual implications of this field. She is painstaking in her effort to discover the implications of various parts of the several theories of causation (and freedom of the will in other essays); and this is exactly how we make progress on difficult philosophical issues like these.

Issues about microfoundations

I believe that hypotheses, theories, and explanations in the social sciences need to be subject to the requirement of microfoundationalism. This requirement can be understood in a weak and a strong version, and sometimes people understand the idea as a requirement of reductionism.  In brief, I defend the position in a weak form that does not imply a reductionist theory of social explanation. Recent discussions with Julie Zahle have led me to sharpen my understanding of the requirement of microfoundations in social theorizing and explanation. Here I would like to clarify my own thinking about the role and scope of the principle of microfoundationalism.

A microfoundation is something like this: an account of the mechanisms at the individual actor level (and perhaps at levels intermediate between actors and the current level -- e.g. institutions) that work to create the structural and causal properties that we observe at the meso or macro level. A fully specified microfoundational account of a meso-level feature consists of an account that traces out (1) the features of the actors and (2) the characteristics of the action environment (including norms and institutions) which jointly lead to (3) the social pattern or causal power we are interested in. A microfoundation specifies the individual-level mechanisms that lead to the macro- or meso-level social fact. This is the kind of account that Thomas Schelling illustrates so well in Micromotives and Macrobehavior.

My thinking about the need for microfoundations has changed over the years from a more narrow requirement ("we need to have a pretty good idea of what the individual-level mechanisms are for a macro-property") to a less restrictive requirement ("we need to have reason to believe that there are individual-level mechanisms for a macro-property"). In The Scientific Marx I liked the idea of “aggregative explanations”, which are really explanations that move from features of individual actors and their interactions, to derivations about social and collective behavior. In Varieties of Social ExplanationI relaxed the idea along these lines:

This doctrine [of microfoundationalism] may be put in both a weak and a strong version. Weakly social explanations must be compatible with the existence of microfoundations of the postulated social regularities, which may, however, be entirely unknown. More strongly social explanations must be explicitly grounded on an account of the microfoundations that produce them. I will argue for an intermediate form—that we must have at least an approximate idea of the underlying mechanisms at the individual level if we are to have a credible hypothesis about explanatory social regularities at all, A putative explanation couched in terms of high-level social factors whose underlying individual-level mechanisms are entirely unknown is no explanation at all. (kl 4746)

My adherence to microfoundationalism today is a little bit weaker still. I now advocate a version of microfoundationalism that specifies only that we must be confident (an epistemic concept) that such micro-to-macro relations exist. We must be confident there are such mechanisms but not obliged to specify them. (I also hold that the best ground for having that confidence is being able to gesture plausibly towards roughly how they might work.) Another way to put it is this requirement: "No magical thinking!" That is, we exclude explanations that would only be possible if we assumed action at a distance, blocks of wood that have complicated mental lives, or intelligent beings with infinite computational faculties. A convincing way of discrediting a meso-level assertion is to give an argument that it is unlikely that real human agents would in fact act in ways that lead to this meso-level situation. (Example: Chinese planners who created the collective farming system in the Great Leap Forward assumed that collective farms would be highly productive because a "new socialist man" would emerge. This was unlikely, and therefore the individual behavior to be expected on collective farms would lead to "easy riding" and low productivity.)

Here is an effort to simplify these issues into a series of assertions:

1. All social forces, powers, structures, processes, and laws (social features) are ultimately constituted by mechanisms at the level of individual actors. (ontological principle)
2. When we assert the reality or causal powers of a social entity, we need to be confident that there are microfoundations that cause this social entity to have the properties we attribute to it. (microfoundations principle)
   1. A description of the microfoundations of a social entity S is an account of the circumstances and individual mechanisms that bring about patterns of individual activity resulting in the properties of S.
   2. Strong version: we must provide a credible statement of the microfoundations.
   3. Intermediate version: we must have a back-of-envelope sketch of possible microfoundations.
   4. Weak version: we must have confidence that there are microfoundations, but we don’t have to have any specific ideas about what they are.
3. A "vertical" social explanation of the features of a social entity S is a derivation of S from facts about the individual level. This is equivalent to providing a specification of the microfoundations of S; a derivation of the properties of S from a model of the action situation of the individuals involved; an agent-based model. This is what JZ calls an individualist explanation.
4. A "horizontal" social explanation is one in which we explain a social entity or structure S by referring to the causal properties of other meso-level entities and conditions. This is what we call a meso-level explanation. (The diagram above illustrates these ideas.)
   1. Horizontal explanations are likewise subject to the microfoundations requirement 2: the entities and powers postulated need to be such that we have good reason to believe that there are microfoundations available for these entities and properties. (Epistemic requirement)
   2. Or slightly stronger: we need to be able to offer at least a plausible sketch of the microfoundations / individual-level mechanisms that would support the postulated entities. (Epistemic+ requirement)
5. Providing or hypothesizing about microfoundations always involves modeling the behaviors and interactions of individuals; so it requires assuming a theory of the actor. So when we try to specify or hypothesize about microfoundations for something we are obliged to make use of some theory of the actor.
6. Traditional theories of the actor are generally too abstract and too committed to a rational-choice model.
7. Social scientists will be better able to hypothesize microfoundations when they have richer theories of the actor. (heuristic principle)

So the ontological principle is simply that social entities are wholly fixed by the properties and dynamics of the actions of the actors that constitute them. The requirement of microfoundations simply reproduces the ontological principle, ruling out ontologically impossible relations among social entities. The requirement of microfoundations is not a requirement on what an explanation needs to look like; rather, it is a requirement about certain beliefs we need to be justified in accepting when we advance a claim about social entities. It is what JZ calls a “confirmation” requirement (or perhaps better, a justificatory requirement). A better theory of the actor supports the discovery of microfoundations for social assertions. Further, it provides a richer "sociological imagination" for macro- and meso-level sociologists. So the requirement of microfoundations and the recommendation that social scientists seek out better theories of the actor are also valuable as heuristics for social research: they provide intellectual resources that help social researchers decide where to look for explanatory links, and what kinds of mechanisms might turn out to be relevant.

What is reduction?

The topics of methodological individualism and microfoundationalism unavoidably cross with the idea of reductionism -- the notion that higher level entities and structures need somehow to be "reduced" to facts or properties having to do with lower level structures. In the social sciences, this amounts to something along these lines: the properties and dynamics of social entities need to be explained by the properties and interactions of the individuals who constitute them. Social facts need to reduce to a set of individual-level facts and laws. Similar positions arise in psychology ("psychological properties and dynamics need to reduce to facts about the activities and properties of the central nervous system") and biology ("complex biological systems like genes and cells need to reduce to the biochemistry of the interacting systems of molecules that make them up").

Reductionism has a bad flavor within much of philosophy, but it is worth dwelling on the concept a bit more fully.

Why would the strategy of reduction be appealing within a scientific research tradition? Here is one reason: there is evident explanatory gain that results from showing how the complex properties and functionings of a higher-level entity are the result of the properties and interactions of its lower level constituents. This kind of demonstration serves to explain the upper level system's properties in terms of the entities that make it up. This is the rationale for Peter Hedstrom's metaphor of "dissecting the social" (Dissecting the Social: On the Principles of Analytical Sociology); in his words,

To dissect, as the term is used here, is to decompose a complex totality into its constituent entities and activities and then to bring into focus what is believed to be its most essential elements. (kl 76)

Aggregate or macro-level patterns usually say surprisingly little about why we observe particular aggregate patterns, and our explanations must therefore focus on the micro-level processes that brought them about. (kl 141)

The explanatory strategy illustrated by Thomas Schelling in Micromotives and Macrobehavior proceeds in a similar fashion. Schelling wants to show how a complex social phenomenon (say, residential segregation) can be the result of a set of preferences and beliefs of the independent individuals who make up the relevant population. And this is also the approach that is taken by researchers who develop agent-based models (link).

Why is the appeal to reduction sometimes frustrating to other scientists and philosophers? Because it often seems to be a way of changing the subject away from our original scientific interest. We started out, let's say, with an interest in motion perception, looking at the perceiver as an information-processing system, and the reductionist keeps insisting that we turn our attention to the organization of a set of nerve cells. But we weren't interested in nerve cells; we were interested in the computational systems associated with motion perception.

Another reason to be frustrated with "methodological reductionism" is the conviction that mid-level entities have stable properties of their own. So it isn't necessary to reduce those properties to their underlying constituents; rather, we can investigate those properties in their own terms, and then make use of this knowledge to explain other things at that level.

Finally, it is often the case that it is simply impossible to reconstruct with any useful precision the micro-level processes that give rise to a given higher-level structure. The mathematical properties of complex systems come in here: even relatively simple physical systems, governed by deterministic mechanical laws, exhibit behavior that cannot be calculated on the basis of information about the starting conditions of the system. A solar system with a massive star at the center and a handful of relatively low-mass planets produces a regular set of elliptical orbits. But a three-body gravitational system creates computational challenges that make it impossible to predict the future state of the system; even small errors of measurement or intruding forces can significantly shift the evolution of the system. (Here is an interesting animation of a three-body gravitational system; the image at the top is a screenshot.)

We might capture part of this set of ideas by noting that we can distinguish broadly between vertical and lateral explanatory strategies. Reduction is a vertical strategy. The discovery of the causal powers of a mid-level entity and use of those properties to explain the behavior of other mid-level entities and processes is a lateral or horizontal strategy. It remains within a given level of structure rather than moving up and down over two or more levels.

William Wimsatt is a philosopher of biology whose writings about reduction have illuminated the topic significantly. His article "Reductionism and its heuristics: Making methodological reductionism honest" is particularly useful (link). Wimsatt distinguishes among three varieties of reductionism in the philosophy of science: inter-level reductive explanations, same-level reductive theory succession, and eliminative reduction (448). He finds that eliminative reduction is a non-starter; virtually no scientists see value in attempting to eliminate references to the higher-level domain in favor of a lower-level domain. Inter-level reduction is essentially what was described above. And theory-succession reduction is a mapping from one theory to the next of the ontologies that they depend upon. Here is his description of "successional reduction":

Successional reductions commonly relate theories or models of entities which are either at the same compositional level or they relate theories that aren't level-specific.... They are relationships between theoretical structures where one theory or model is transformed into another ... to localize similarities and differences between them. (449)

I suppose an example of this kind of reduction is the mapping of the quantum theory of the atom onto the classical theory of the atom.

Here is Wimsatt's description of inter-level reductive explanation:

Inter-level reductions explain phenomena (entities, relations, causal regularities) at one level via operations of often qualitatively different mechanisms at lower levels. (450)

Here is an example he offers of the "reduction" of Mendel's factors in biology:

Mendel's factors are successively localized through mechanistic accounts (1) on chromosomes by the Boveri–Sutton hypothesis (Darden, 1991), (2) relative to other genes in the chromosomes by linkage mapping (Wimsatt, 1992), (3) to bands in the physical chromosomes by deletion mapping (Carlson, 1967), and finally (4) to specific sites in chromosomal DNA thru various methods using PCR (polymerase chain reaction) to amplify the number of copies of targeted segments of DNA to identify and localize them (Waters, 1994).

What I find useful about Wimsatt's approach is the fact that he succeeds in de-dramatizing this issue. He puts aside the comprehensive and general claims that have sometimes been made on behalf of "methodological reductionism" in the past, and considers specific instances in biology where scientists have found it very useful to investigate the vertical relations that exist between higher-level and lower-level structures. This takes reductionism out of the domain of a general philosophical principle and into that of a particular research heuristic.

Meso causes and microfoundations

In earlier posts I've paid attention to the need for microfoundations and the legitimacy of meso-level causation. And I noted that there seems to be a prima facie tension between the two views in the philosophy of social science. I believe the two are compatible if we understand the microfoundations thesis as a claim about social ontology and not about explanation, and if we interpret it in a weak rather than a strong way. Others have also found this tension to be of interest. The September issue of The Philosophy of the Social Sciences" provides a very interesting set of articles on this set of issues.

Particularly interesting is a contribution by Tuukka Kaidesoja, "Overcoming the Biases of Microfoundations: Social Mechanisms and Collective Agents" (link). Here are the four claims advanced in the article:

1. The mechanism approach to social explanation does not presuppose a commitment to the individual-level microfoundationalism.
2. The microfoundationalist requirement that explanatory social mechanisms should always consists of interacting individuals has given rise to problematic methodological biases in social research.
3. It is possible to specify a number of plausible candidates for social macro-mechanisms where interacting collective agents (e.g. formal organizations) form the core actors.
4. The distributed cognition perspective combined with organization studies could provide us with explanatory understanding of the emergent cognitive capacities of collective agents. (abstract)

I agree with many of Kaidesoja's criticisms of what he calls individual-level microfoundationalism (IMF). I also agree with his preference for the weak "rationalist" conception of emergence (along the lines of Mario Bunge) rather than the strong conception associated with Niklas Luhman (link). However, I want to continue to maintain that there is a different version of microfoundationalism that is not vulnerable to the criticisms he offers -- what I call the "weak" version of microfoundations. (This is explicated in several earlier posts; link.) On this approach, claims about higher-level entities need to be plausibly compatible with there being microfoundations at the individual level (an ontological principle), but I deny that we always need to provide those microfoundations when offering a social explanation (an explanatory principle). And in fact, Kaidesoja seems to adopt a very similar position:

By contrast, in many explanatory studies on large-scale macro-phenomena, it is sufficient that we have a general understanding how the collective agents of this kind function (e.g., how collective-decisions are typically made in the organizations that are the components of the relevant macro-mechanism) and empirically grounded reasons to believe that the macro-phenomenon of interest was causally generated by the interactions of this kind of collective agents with emergent powers.... Of course, it is always possible to zoom in to a particular collective agent and study the underlying mechanisms of its emergent causal powers, but this type of research requires the uses of different methods and data from the explanatory studies on large-scale macro- phenomena. (316)

So it is the in-principle availability of lower-level analyses that is important, not the actual provision of those accounts. Or in other words, K is offering a set of arguments designed to establish the explanatory sufficiency of at least some meso- and macro-level causal accounts (horizontal) rather than requiring that explanations should be vertical (rising from lower levels to higher levels). This is what I want to refer to as "relative explanatory autonomy of the meso-level."

Kaidesoja's position is a realist one; he couches his analysis of causation in terms of the idea of causal powers. Here is Kaidesoja's description of the idea of causal powers:

In general terms, causal powers of complex entities include their dispositions, abilities, tendencies, liabilities, capacities, and capabilities to generate specific type of effects in suitable conditions. Each particular entity (or powerful particular) possesses its powers by virtue of its nature, which in turn can typically be explicated in terms of the intrinsic relational structure of the entity. (302)

This position provides an answer to one of the questions recently posed here: are causal powers and causal mechanisms compatible? I think they are, and Kaidesoja appears to as well.

One important nuance concerns the kinds of higher-level social structures that Kaidesoja offers as examples. They all involve collective actors, thus assimilating social causal power to intentional action. But the category of macro social factor that possess causal powers is broader than this. There are credible examples of social powers that do not depend on any kind of intentionality. Most of the examples offered by Charles Perrow, for example, of organizations with causal powers depend on features of operation of the organization, not its functioning as a quasi-intentional agent.

Also interesting in the article is Kaidesoja's gloss on the idea of distributed cognition. I'm not receptive to the idea of collective social actors is a strongly intentionalist sense (link), but K makes use of the idea of distributed cognition in a sense that seems unobjectionable to people who think that social entities ultimately depend on individual actors. K's interpretation doesn't imply commitment to collective thoughts or intentions. Here is a clear statement of the idea:

An important implication of the above perspectives is that they enable one to ascribe emergent cognitive capacities to social groups and to study the underlying mechanisms of these capacities empirically (e.g., Hutchins 1995; Theiner and O’Connor 2010). This nevertheless requires that we reconsider our received concept of cognition that ties all cognitive capacities to individual organisms (e.g., human beings), since groups obviously lack system-level consciousness or brains as distinct from those of their individual members.  (317)

Now, drawing on organization studies (e.g., Scott and Davis 2003), I suggest that formal organizations (in short, organizations) can be understood as social groups that are designed to accomplish some (more or less clearly specified) goal or goals, and whose activities are planned, administrated, and managed by their members (or some subgroup of their members such as managers). Examples of organizations include schools, business firms, universities, hospitals, political parties, and governments. (318)

This is a conception of "cognition" that doesn't imply anything like "collective minds" or group intentions, and seems unobjectionable from an ontological point of view.

This is a very nice piece of work in the philosophy of social science, and it suggests that it will be worthwhile to spend time reading Kaidesoja's recent book, Naturalizing Critical Realist Social Ontology (Ontological Explorations), as well.

 

Social mechanisms and meso-level causes

(This post summarizes a paper I presented at the British Society for the Philosophy of Science Annual Meeting in 2012.)

Here and elsewhere I want to defend the theoretical possibility of attributing causal powers to meso-level social entities and structures.  In this I follow a number of philosophers and sociologists, including many critical realists (e.g. Roy Bhaskar, A Realist Theory of Science and Margaret Archer, Realist Social Theory: The Morphogenetic Approach) and also the recent thinking of Dave Elder-Vass (The Causal Power of Social Structures).  But I also defend the idea of an actor-centered sociology, according to which the substance of social phenomena is entirely made up of the actions, interactions, and states of mind of socially constituted individual actors.  Making out both positions, and demonstrating their consistency, is the work of this paper. I refer to this position as “relative explanatory autonomy” of the meso-level. This topic is of renewed interest because of the current influence and progress of analytical sociology (Peter Hedström, Dissecting the Social: On the Principles of Analytical Sociology; Hedström and Bearman, The Oxford Handbook of Analytical Sociology ; Peter Demeulenaere, Analytical Sociology and Social Mechanisms), which offers an emphatic “no” to the question; whereas critical realists are equally firm in defending an affirmative answer to the question.

My defense of meso-level causation is based on four ideas. First, the practice of sociologists justifies this claim, since sociologists do in fact make use of meso-meso claims.  They often do not attempt to provide vertical explanations from circumstances of the actor to meso- and macro-level outcomes; instead, they often provide horizontal explanations that explain one set of meso and macro outcomes on the basis of the causal powers of another set of meso and macro conditions or structures.  Second, sociology is a “special science” analogous to cognitive science, dependent on a set of causally linked entities at a lower level. Arguments offered for the relative explanatory autonomy of the higher-level theories are applicable to sociology as well. The basis for rejecting reductionism is well established here. Third, meso entities (organizations, institutions, normative systems) often have stable characteristics with regular behavioral consequences. This is illustrated with the example of organizations. Fourth, those entities must have microfoundations; we must be confident that there are individual behaviors at lower levels that support these macro characteristics.  But it is legitimate to draw out the macro-level effects of the macro-circumstance under investigation, without tracing out the way that effect works in detail on the swarms of actors encompassed by the case.   The requirement of microfoundations is not a requirement on explanation; it does not require that our explanations proceed through the microfoundational level. It is an ontological principle but not a methodological principle.  Rather, it is a condition that must be satisfied on prima facie grounds, prior to offering the explanation. (I refer to this as the "weak" requirement of microfoundations; link.)  In short, we are not obliged to trace out the struts of Coleman’s boat in order to provide a satisfactory macro- or meso-level explanation or mechanism.

Does the extended social world have causal powers?  Here are some reasons for thinking that it does.

First, working sociologists offer explanations that postulate meso-meso causal connections on a regular basis. They identify what they take to be causal properties of social structures and institutions, and then draw out causal chains involving those causal properties. And often they are able to answer the follow-on question: how does that causal power work, in approximate terms, at the micro level? But answering that question is not an essential part of their argument. They do not in fact attempt to work through the agent-based simulation that would validate their general view about how the processes work at the lower level.

This explanatory framework seems entirely reasonable in the social sciences.  It does not seem necessary to disaggregate every claim like “organizational deficiencies at the Bhopal chemical plant caused the devastating chemical spill” onto specific individual-level activities. We understand pretty well, in a generic way, what the microfoundations of organizations are, and it isn’t necessary to provide a detailed account in order to have a satisfactory explanation.  In other words, we can make careful statements about macro-macro and macro-meso causal relations without proceeding according to the logic of Coleman’s boat—up and down the struts. So one argument for the relative autonomy of meso-level causal claims is precisely the fact that good sociologists do in fact make credible use of such claims.

Second, there is a more general reason within the philosophy of the social sciences for being receptive to the idea of meso-meso social causes.  This derives from the arguments against reductionism in a range of the special sciences.  The idea of relative explanatory autonomy has been invoked by cognitive scientists against the reductionist claims of neuro-scientists. Of course cognitive mechanisms must be grounded in neurophysiological processes. But this doesn’t entail that cognitive theories need to be reduced to neurophysiological statements. Jerry Fodor introduced highly influential arguments against reductionism in “Special sciences and the disunity of science as a working hypothesis” (link).
Once we have reason to accept something like the idea of relative explanatory autonomy in the social sciences, we also have a strong basis for rejecting the exclusive validity of one particular approach to social explanation, the reductionist approach associated with methodological individualism, analytical sociology, and Coleman’s boat. Rather, social scientists can legitimately use explanations that call upon meso-level causal linkages without needing to reduce these to derivations from facts about individuals. And this implies the legitimacy of a fairly broad conception of methodological pluralism in the social sciences, constrained always by the requirement of weak microfoundations.

Third, we have good research-based reasons to maintain that meso-level social structures have causal powers. Consider organizations as paradigm examples of meso-level social structures. An organization is a social entity which possesses a degree of stability in functioning that can be studied empirically and theoretically. An organization consists of a structured group of individuals, often hierarchically organized, pursuing a relatively clearly defined set of tasks.  In the abstract, it is a set of rules and procedures that regulate and motive the behavior of the individuals who function within the organization. There are also informal practices within an organization that are not codified that have significant effects on the functioning of the organization (for example, the coffee room as a medium of informal communication, or the norm of covering for a co-worker’s absence).  Some of those individuals have responsibilities of oversight, which is a primary way in which the abstract rules of the organization are transformed into concrete patterns of activity by other individuals. Another behavioral characteristic of an organization is the set of incentives and rewards that it creates for participants in the organization. Often the incentives that exist were planned and designed to have specific effects on behavior of participants; by offering rewards for behaviors X, Y, Z, the organization is expected to produce a lot of X, Y, and Z. Sometimes, though, the incentives are unintended, created perhaps by the intersection of two rules of operation that lead to a perverse incentive leading to W.

Now we are in a position to address the central question here: do organizations have causal powers? It seems to me that the answer is yes, in fairly specific ways. The most obvious causal properties of an organization are bound up in the function of the organization. An organization is invented and developed in order to bring about certain social effects: operate and maintain a complex technology, reduce pollution or crime, distribute goods throughout a population, provide services to individuals, seize and hold territory, disseminate information. But the specifics of an organization also give rise to unintended consequences; these too contribute to the causal powers of the organization. All these effects occur as a result of the coordinated activities of people within the organization. When organizations work correctly they bring about one set of effects; when they break down they bring about another set of effects. Here we can think about organizations in analogy with technology components like amplifiers, thermostats, stabilizers, or surge protectors.

Finally, I too believe that there is a burden of proof that must be met in asserting a causal power or disposition for a social entity -- something like “the entity demonstrates an empirical regularity in behaving in such and such a way” or “we have good theoretical reasons for believing that X social arrangements will have Y effects.” And some macro concepts (e.g. State, Islam, market economy) are likely cast at too high a level to admit of such regularities. That is why I favor “meso” social entities as the bearers of social powers. As new institutionalists demonstrate all the time, one property regime elicits very different collective behavior from its highly similar cousin. And this gives the relevant causal stability criterion. Good examples include Robert Ellickson’s new-institutionalist treatment of Shasta County and liability norms and Charles Perrow’s treatment of the operating characteristics of technology organizations. In each case the microfoundations are easy to provide. What is more challenging is to show how these social causal properties interact in cases to create outcomes we want to explain.

So how does the micro-macro link look when we attempt to provide the idea of meso explanations with microfoundations?  The various versions of methodological individualism—microeconomics, analytical sociology, Elster’s theories of explanation, and the model of Coleman’s boat—presume that explanation needs to invoke the story of the micro level events as part of the explanation. The perspective offered here requires something quite different.  This position requires that we be confident that these micro-level events exist and work to compose the meso level; but it does not require that the causal argument incorporates a reconstruction of the pathway through the individual level in order to have a satisfactory explanation.  This account suggests an alternative diagram to Coleman’s boat.


The diagram represents each of the causal linkages represented in the Coleman boat. But it also calls out the meso-meso causal connection that Coleman prohibits in his analysis. And it replaces the idea that causation proceeds through the individual level, with the idea that each meso level factor has a set of actor-level microfoundations. But this is an ontological fact, not a prescription on explanation.
(Here is the full paper as presented at the British Society for the Philosophy of Science Annual Meeting in July 2012 at the University of Stirling.)

Social structures and causal powers

The idea of a causal power has been appealing to the realist tradition within the philosophy of science, and especially so for the philosophy of social science. Proponents of this idea include Nancy Cartwright (Nature's Capacities and Their Measurements), Margaret Archer (Realist Social Theory: The Morphogenetic Approach), and Dave Elder-Vass (The Causal Power of Social Structures). Elder-Vass provides a succinct description of the tradition:

Bhaskar offers us an alternative way of understanding causality, a causal powers theory. This draws on a different, realist, tradition of thinking about cause, one that goes back at least as far as Aristotle, but one that has been less influential than the covering law model in twentieth-century social science. As [Ruth] Groff puts it, 'realists about causality think, contra Hume, that causal relations are relations of natural or metaphysical necessity, rather than of contingent sequence' -- and that this necessity arises from the nature of the objects involved in those causal relations (Groff 2008:2-3). (43)

Stephen Mumford and Rani Lill Anjum's Getting Causes from Powers is an important contribution to this debate. Here are several useful comments from Mumford's contribution to the The Oxford Handbook of Causation, "Causal Powers and Capacities":

Where it is most radical, the powers ontology proposes a major reconceptualization of causation. Hume, as traditionally interpreted, understood the world to consist of distinct and discrete, unconnected existences. If this is accepted, then the best that can be made of causation is that it is a contingent and external relation between such existences. The powers ontology accepts necessary connections in nature, in which the causal interactions of a thing, in virtue of its properties, can be essential to it. Instead of contingently related cause and effect, we have power and its manifestation, which remain distinct existences but with a necessary connection between.

One such tradition was based in Britain and came from the work of Rom Harré (1970; 2001, and with Madden 1973; 1975), which seems to have been an influence on Roy Bhaskar (1975) and Nancy Cartwright (1983; 1989; 1999). In Cartwright, the commitment is to capacities, which in her account differ from dispositions in that they ‘are not restricted to any single kind of manifestation ... [but] can behave very differently in different circumstances’. (1999: 59)

Putting the point simply, the assertion that an entity has a causal power comes down to a claim about the nature of the entity and the strong dispositional properties that this nature gives rise to. Sugar has the causal power to stimulate the taste of sweetness in typical human subjects; this power derives from the chemical structure of the sugar molecule and the micro-organization and functioning of taste receptor neurons. A magnet has a power to attract a piece of iron, in virtue of its microstructure. In each case we have identified a real feature of the entity, and this feature is a consequence of real properties of its microstructure.

This approach makes sense with regard to social structures and institutions as well. If paramilitary organizations have a propensity to create young adherents who are easily mobilized in support of fascist politics (as argued by Michael Mann in Fascists), then we can make reference to this causal power in our explanation of the rise of Italian fascism. University X's tenure system produces a teaching environment in which students get little attention from their faculty, as a consequence of the incentives and habits it cultivates in young faculty. This means something fairly straightforward: given the specific arrangements associated with this tenure system, the interactions that individuals have within this institution inculcates patterns of behavior that bring about the consequence. On this story, "producing a faculty climate that gives little priority to undergraduate students" is a causal power of this institutional arrangement. Change the internal arrangements and you get different causal properties.

In the case of the social world, however, the fundamental constituents of social powers are the constrained and developed actions of persons who act within the context of a given set of institutions and structures. Unlike the iron magnet, whose powers derive from identical iron atoms arranged in certain geometries, a tenure institution or a safety organization derives its properties from the structured actions of the individuals who compose it.

The rationale for asserting necessity in either the natural or the social realm -- the idea that the power is a real property of the thing -- is the theory of scientific realism: things actually have the causal powers we observe because they have an inner constitution that propels their interactions with other entities. So the causal relation is a kind of necessary relation, not just a brute fact about regularities. Metals conduct electricity because of the chemical-physical structure of the copper wire. And universities have the properties they possess because of the institutional arrangements they embody and the actions of individuals within those arrangements.

So the theory of causal powers doesn't have to presuppose an objectionable form of metaphysical essentialism. Instead, it can be a defensible framework for embodying the idea of causal realism: things have the causal properties and dispositions they have in virtue of their micro-composition.

Why is it useful to use the language of causal powers? Because we can encapsulate a large amount of the pertinent causal properties of an entity into a fairly simple set of expectations. If iron is magnetic (a causal power) we can derive a large number of expectations about its behavior in a variety of circumstances; and we can explain those circumstances based on the powers we have empirically or theoretically established. If a certain kind of regulatory organization is observed to have the causal power of "contributing to an abnormal number of accidents" -- then one part of an explanation of a particular accident may be the fact that it occurred within the scope of that kind of regulatory organization. (Charles Perrow offers an argument along these lines in Normal Accidents: Living with High-Risk Technologies.)

Causal concepts

source: D. Little, “Causal Explanation in the Social Sciences,” Southern Journal of Philosophy (1995) (link)

It may be useful to provide a brief account of some of the key ideas that are often invoked in causal explanations in the social sciences. (Here is an earlier post that summarized some current issues in causation research; link. And here are several earlier articles on causal explanation; link, link, link.)

The general idea of a social cause (X causes Y) goes along these lines: X is a structure or feature of social life that varies across social settings and whose presence increases the likelihood of occurrence of Y. The presence of X (perhaps in the presence of Y and Z as well) contributes to processes leading to Y.

This simple formulation contains several hidden assumptions -- most importantly, that outcomes have causes, that causes retain their characteristics over time and across instances, and that there are processes or dynamics within the domain of things and processes that convey with some form of necessity one set of circumstances and events onto another.

An example

For example, consider this hypothetical narrative describing a riot in a European city with a large community of impoverished immigrant people:

• (C1) simmering resentment by immigrant youth of joblessness and low social esteem
• (C2) heat wave creating discomfort and misery in crowded neighborhoods
• (C3) chronic disrespectful and rough police treatment of immigrant youth
• (I) forceful arrest of mis-identified young person in a city park, leading to serious injury of the youth
• (O) several days of rioting occur

The associated causal hypothesis goes along these lines: In the context of simmering resentment by immigrant youth and a pattern of mistreatment by police, feelings in the community were unusually elevated by the heat wave. When the arrest occurred a small protest began in the park, which spread to other blocks in the city and eventuated in the burning of cars, smashing of shop windows, and multiple further arrests.

Conditions Ci are standing conditions that played a causal role in the occurrence of the riot. The arrest incident was the instigating event, the match that ignited the social "gasoline". If any of C1, C2, C3 had been changed six months earlier, it is unlikely that O would have occurred. Each was necessary for I leading to O in the circumstances of the day.  If C1, C2, C3 are present, it is likely that some instigating event will occur in the normal hustle-bustle of urban life. I was the instigating condition. For researchers seeking general explanations of urban unrest, C1 and C3 appear to be strong candidates for common causes across many examples of urban riots. Two mechanisms are invoked here: a mechanism having to do with the individual's propensity to engage in protest ("resentment and mistreatment elevates propensity to protest") and a mechanism having to do with the spread of protest ("a small disturbance between a few teenagers and the police escalates through direct contact with other disaffected individuals through the neighborhood").

Here are brief discussions of many of the concepts that are commonly invoked in discussions of social causation.

Causal narrative

An organized and temporally directed account of the occurrence of an event or change, identifying the conditions, circumstances, and events that were causally relevant to its occurrence. A narrative needs to provide empirical evidence for its empirical claims and theoretical justification for the causal mechanisms and processes it postulates.

Standing condition

A condition or circumstance that persists through an extended period of time and that serves as part of the necessary causal background of a given causal process or mechanism. Persistent racial isolation is a standing condition in many explanations of the effects of inner city poverty.

Instigating event

An instigating event is an occurrence, including change of state of some background property, that triggers a change in some other property or process. The early-morning arrest by patrons of a blind pig (unlicensed tavern) in Detroit was the instigating event of the 1967 Detroit riot/uprising.

Necessary condition

A condition that must be present in order for a given causal interaction to occur. "If X had not been present, the outcome O would not have occurred."

Sufficient condition (conjunction of conditions)

A condition (or conjunction of conditions) whose presence suffices to bring about the outcome. "If X&Y&Z were present, then O would have occurred."

Counterfactual statements

It is worth underlining the point that necessary and sufficient conditions invoke counterfactual statements: If X had not occurred, Y would not have occurred. The logic of counterfactuals (modal logic) has a controversial and unresolved history. But given that causal language always implies some kind of necessity, we cannot dispense with counterfactuals and still have an adequate causal vocabulary.

INUS condition (J. L. Mackie)

J.L. Mackie's work on causation in The Cement of the Universe: A Study of Causation brought to closure a long line of thought about the logic of causal relations, culminating in his concept of INUS conditions. Consider this complex causal statement about the circumstances causing P:

'All (ABC or DGH or JKL) are followed by P' and 'All P are preceded by (ABC or DGH or JKL)' (Mackie, 62)

Mackie then defines an INUS condition:

Then in the case described above the complex formula '(ABC or DGH or JKL)' represents a condition which is both necessary and sufficient for P: each conjunction, such as 'ABC', represents a condition which is sufficient but not necessary for P. Besides, ABC is a minimal sufficient condition:  none of its conjuncts is redundant: no part of it, such as AB, is itself sufficient for P. But each single factor, such as A, is neither a necessary nor a sufficient condition for P. Yet it is clearly related to P in an important way: it is an insufficient but non-redundant part of an unnecessary but sufficient condition: it will be convenient to call this … an inus condition. (62)

To simplify:

A is an INUS condition for P if for some X and Y, (AX v Y) is a necessary and sufficient condition for P, but A is not sufficient for P and X is not sufficient for P.

Causal mechanism

An interlocked series of events and processes that, once initiated by some set of conditions, [usually] brings about a given outcome O. The idea that there are real mechanisms embodied in the "stuff" of a given domain of phenomena provides a way of presenting causal relations that serves as a powerful alternative to the "regularity" view associated with Hume. "Poor performance on standardized tests by specific groups is caused by the mechanism of stereotype threat" (Claude Steele, Whistling Vivaldi: How Stereotypes Affect Us and What We Can Do (Issues of Our Time)). This mechanism is a hypothesized process within the cognitive-emotional system of the subjects of the test. (James Mahoney's survey article on the mechanisms literature is a good introduction to the debate; link.)

Causal powers

The idea that certain kinds of things (metals, gases, military bureaucracies) have internal characteristics that lead them to interact causally with the world in specific and knowable ways. This means that we can sometimes identify dispositional properties that attach to kinds of things. Metals conduct electricity; gases expand when heated; military bureaucracies centralize command functions. (Harre and Madden, Causal Powers: Theory of Natural Necessity)

Probabilistic causal relation

A relationship between A and O such that the occurrence of A increases/decreases the likelihood of the occurrence of O. This can be stated in terms of conditional probabilities: P(O|A) ≠ P(O) [the probability of O given A is not equal to the probability of O]. For a causal realist, the definition is extended by a hypothesis about an underlying causal mechanism. [Smoking is a probabilistic cause of lung cancer [working through physiological mechanisms X,Y,Z]. This is equivalent to Wesley Salmon's criterion of causal relevance (Scientific Explanation and the Causal Structure of the World).

Causal explanation of a singular event

When we are interested in the explanation of a single event, a causal narrative leading up to that event is generally what we are looking for. What led to the outbreak of World War I? Why did Khomeini come to power in Iran in 1979? There are generally two difficult problems facing a proposed causal-narrative explanation of a singular event. First, we need to somehow empirically validate the claims about causal mechanisms and processes that are invoked in the narrative. But since this is a singular event, we do not have the option of using experimental methods to empirically test the claim "X leads by mechanism M to Y" that the narrative proposes. This is one important reason why mechanism theorists have generally required that specified mechanisms have roughly similar causal properties in a range of circumstances. Circumstances embodying the core features of a public goods problem usually lead to elevated levels of free riding -- whether in public radio fundraising, strikes, classroom discussions, or rebellions. Second, there is the problem of alternative realizability and multiple causal pathways leading to the same outcome. If the conditions leading to World War I were sufficiently ominous, then whether the assassination of the Archduke or some other event brought it about is of less explanatory importance. Given that potential instigating events occur with a certain probability, some event would have occurred within those few months that led to war. So it is better to identify the standing conditions that made war likely as the causes, rather than the assassination of the Archduke.

Generalizations about the causes of a kind of social entity or event

We are often interested in answering causal questions about classes of events: Why do peasant rebellions occur? Why does corruption rise to such high levels in many cities? Why do democracies not wage war against each other? Here we are looking for common conjunctions of causal factors that can be shown to be causally relevant in many such events. It is possible that we will discover that peasant rebellions do not have a single set of causal antecedents; rather there are multiple profiles of peasant rebellions, each with a set of causal conditions significantly different from the other profiles.

Methods of causal inquiry

How can social researchers identify causal relations among social events and structures? There are several groups of methods that social scientists and historians have employed: statistical-causal models, small-N models based on Mill's methods of similarity and difference (link, link), and case studies and process-tracing methods through which researchers seek to identify and confirm causal relations in individual cases. In each case the method derives from fundamental ideas about the nature of causation: the idea that causal relations between several factors give rise to statistical regularities when we have a large number of cases; the idea that we can use the features of necessary and sufficient conditions to select cases in order to include or exclude certain factors as causally related to the outcome; and the idea that causal mechanisms and processes can often be observed fairly directly in the historical record (Alexander George and Andrew Bennett, Case Studies and Theory Development in the Social Sciences).

Social causation

The idea of social causation is a difficult one, as we dig more deeply into it. What does it mean to say that "poor education causes increased risk of delinquency" or "population growth causes technology change" or "the existence of paramilitary organizations contributed to the rise of German fascism"? What sorts of things can function as "social causes" -- events, structures, actions, forces, other? What social interactions extend over time in the social world to establish the links between cause and effect? What kinds of evidence are available to support the claim that "social factor X causes a change in social factor Y"?

Helen Beebee, Christopher Hitchcock, and Peter Menzies' The Oxford Handbook of Causation is a valuable resource on topics involving the philosophy of causation, and several of the contributions are immediately relevant to current debates within the philosophy of social science.

Harold Kincaid considers a number of the hard questions about social causation in his contribution to the Handbook, "Causation in the Social Sciences". Perhaps most relevant to my ongoing concerns is his defense of non-reductionist claims about social causation. It is often maintained (by methodological individualists) that causal relations exist only among individuals, not among higher-level social entities or structures. (Elster and Hedstrom make claims along these lines in multiple places.) Kincaid rejects this view and affirms the legitimacy of macro- or meso-level causal assertions.

When a particular corporation acts in a market, it has causal influence. The influence of that specific entity is realized by the actions of the individuals composing it just as the influence of the baseball on the breaking window is realized by the sum of particles composing it. The social level causal claims pick out real causal patterns as types that may not be captured by individual kinds because multiple realizability is real. (kl 16102)

These arguments are a valuable antidote to the tendency towards reductionism to the level of individual activity that has often guided philosophers when they have considered the nature of social causation.

Phil Dowe's discussion of causal process theories is useful for the social sciences ("Causal Process Theories"). It is hard to think of the social world as an amalgam of discrete events; it is easier to think of a variety of processes unfolding, subject to a range of forces and obstacles.  Dowe gives much of the credit for current interest in causal process language to Wesley Salmon (along with resurrection of the idea of a "rope of causation" to replace that of a "chain of causation").

For Salmon the causal structure of the world consists in the nexus of causal processes and interactions. A process is anything with constancy of structure over time. (kl 4924)

The language of causal processes seems to fit the nature of social causation better than that of events and systems of billiard balls. And we have the makings of a metaphysics of process available in the social sciences, in the form of a stream of actions and reactions of individuals aggregating to recognizable social patterns. So when we say that "population increase stimulates technology innovation", we can picture the swarming series of interactions, demands, and opportunities that flows from greater population density, to rewards for innovation, to a more rapid rate of innovation.

Another useful contribution in the Handbook with special relevance to the social world is Stephen Mumford's contribution, "Causal Powers and Capacities."

The powers ontology accepts necessary connections in nature, in which the causal interactions of a thing, in virtue of its properties, can be essential to it. Instead of contingently related cause and effect, we have power and its manifestation, which remain distinct existences but with a necessary connection between. (kl 5971)

The language of causal powers allows us to incorporate a number of typical causal assertions in the social sciences: "Organizations of type X produce lower rates of industrial accidents," "paramilitary organizations promote fascist mobilization," "tenure systems in research universities promote higher levels of faculty research productivity." In each case we are asserting that a certain kind of social organization possesses, in light of the specifics of its rules and functioning, a disposition to stimulate certain kinds of participant behavior and certain kinds of aggregate outcomes. This is to attribute a specific causal power to species of organizations and institutions.

Stuart Glennan's "Mechanisms" is also highly relevant to causation in the social realm. Here is how Glennan puts the mechanisms theory (quoting his own earlier formulations):

Glennan ... characterizes mechanisms in this way: 'A mechanism underlying a behavior is a complex system which produces that behavior by the interaction of a number of parts according to direct causal laws (Glennan 1996: 52). Glennan then suggests that two events are causally related when and only when they are connected by an intervening mechanism. (kl 7069)

This definition works pretty well with typical examples of social mechanisms, with one important exception -- the reference to "direct causal laws". When we say that Organization X works to minimize accidents, the sub-transactions that are involved in the workings of the overall process are not typically "direct causal laws," but rather the intelligible results of individual actors performing their actions within the rules, incentives, and sanctions of the organization. We can tell a mechanism story along these lines: "Organization X embodies a set of protocols of operation, a training regime, a supervisory regime, and and enforcement regime. The protocols have the result that, when followed consistently, accidents are rare. Employees are 'programmed' to perform their tasks according to this set of protocols. Supervisors are trained to observe and measure employee performance against the protocols. Enforcement provides sanctions and incentives for bad and good performance." The complex mechanism of the organization works to implement and maintain the smooth functioning of the guiding protocols. So the organization embodies just the kind of complex system that Glennan describes as a mechanism.

So the Handbook is a good resource for all of us who are interested in working through a more satisfactory account of what it means to look at social phenomena as embodying causal relations in ways that support explanations.

(The photo of ice forming on glass included above is a metaphorical reference to social causation. Intricate patterns have emerged from a causal process; but it is a process that reflects a high degree of contingency and path-dependency across the expanse of the scene. And there is no overall order to the multiple patterns that emerge; each location is independent of other locations, and there is no answer to questions like this: "Why is Structure A located in the particular position and orientation that it is found to be?" Patterns coalesce and they do so as a result of locally operative causal processes, but there is no overall guiding hand or teleology to the process. The greatest disanalogy I can see here is the fact that the ice-formation process is much simpler than typical social-causal systems. Instead of a single causal mechanism at work in the ice case, there are dozens of overlapping and interactive causal mechanisms at work in most social processes.)

Causal inference and random trials

image: Tamil Nadu nutrition study

Nancy Cartwright has spent much of her career probing the assumptions scientists make about causation. She has helped to demonstrate that the Humean assumptions about causation that philosophers (including Carl Hempel) carried into twentieth century philosophy of science don't come close to answering the question correctly, and she has provided many reasons to take seriously the ideas of causal powers and mechanisms rather than governing causal regularities. How the Laws of Physics Lie is an important contribution to the philosophy of science and to realist theory.

Her current book Evidence-Based Policy: A Practical Guide to Doing It Better (with Jeremy Hardie) provides a different critical perspective on causal inference, this time in the context of social policy reasoning. The design and implementation of public policies rest upon a fundamental premise: that we can have evidence-based reasons for predicting what the effects of the policy tool will likely be. But what kind of evidence might that be? The dominant form of evidence favored in the policy science field is random controlled trials: specify the policy intervention P, choose a domain of cases to apply the intervention P to, randomly select cases to receive the intervention (versus the control group that does not), and measure the value of the outcome of interest. If there is a significant difference in the value of the outcome between test group and control group, then we have evidence that P had an effect.

In a nutshell, C&H take issue with the conviction that random controlled trials  (RCT) -- the gold standard of causal inference and experiment in clinical medicine -- provide a basis for expecting that a given policy intervention will have similar effects in the future. Their book can be read as a critique of an excessively statistical understanding of social causality, without realistic analysis of the underlying mechanisms and processes. As Cartwright and Hardie state repeatedly, RCT evidence shows only that the policy worked on the circumstances tested in the study. Instead, they argue that we need to offer evidence about two additional considerations: whether the "causal principle" associated with P will remain the same in new circumstances; and whether the associated conditions necessary for the operation of this principle will be present in the new circumstances.

Here is a fundamental statement of what they mean by a causal principle:

We suppose that causes do not produce their effects by accident, at least not if you are to be able to make reliable predictions about what will happen if you intervene. Rather, if a cause produces an effect, it does so because there is a reliable, systematic connection between the two, a connection that is described in a causal principle. (22)

The statement, "El Nino causes wet winters in North America," is a causal principle. But causal principles are neither universal nor exceptionless:

The fact that causal principles can differ from locale to locale means that you cannot read off that a policy will work here from even very solid evidence that it worked somewhere else. (23)

Here is a more extensive description of this idea:

Causal principles are not universal. They differ from place to place and from time to time. That means that it is not enough for you to know that the policy worked somewhere or even that it has worked at some time here. “It worked there”; it played a positive causal role there. So it was one of the factors from a causal principle that holds there. To predict that it will work here, you need to know that it is one of the factors from a causal principle that holds here. That is what ensures that it can play a positive causal role for you. (50)

Cartwright and Hardie look at causation along the lines of J. L. Mackie's analysis of INUS conditions in The Cement of the Universe: a factor is a cause if it is an "Insufficient but Necessary part of an Unnecessary but Sufficient condition for producing a contribution to the effect" (23). The evidence of favorable CRT studies for a given policy intervention doesn't show that this policy will work in the new circumstances of the new proposed application. In order to draw this inference we need to have confidence that the treatment will play the same causal role in the new setting, and that the necessary conditions will be present in that setting. In other words, we need a more detailed causal analysis of the past and the proposed future.

Here is a sketch of the argument that C&H suggest we need to provide in order to project favorable RCT studies onto future applications:

1. x works there (i.e., x genuinely appears in the causal principle that governs the production of y there post-implementation). 
2. Here and there share that causal principle post-implementation. 
3. The support factors necessary for x to contribute under that principle are present for at least some individuals here post-implementation. 
4. Conclusion. x works here (i.e., x genuinely appears in the causal principle that governs the production of y here post-implementation and the support factors necessary for it to contribute to y are present for at least some individuals here post-implementation). (41)

One way of offering support for premise 2 is to engage in the method of process tracing:

This method confirms the existence of a causal connection between start and finish by confirming, one-by-one, a series of smaller causal steps in between. (38)

Cartwright doesn't put her case in these terms, but I would say that the heart of her intuition is that social outcomes are different from medical outcomes because of their inherent causal heterogeneity. In the social world outcomes like teen pregnancy rates or high school dropout rates are the result of a bundle of conjunctural causal processes. So projecting the results of past random controlled trials into the future requires that we first confirm that the same causal influences and important background conditions are at work. And this is rarely the case. So the fundamental underlying prescription is a pragmatic causal realism about social processes: in order to design and implement policies, we need to have a well developed map of the real causal processes and mechanisms that are underway in the production of the effect we would like to change. In other words, we need to be causal realists if we are to be effective policy makers.

(It is worth observing that this book is deliberately different in tone and specialization from Cartwright's other monographs in the philosophy of causation. The book is designed to be useful for real practitioners of public policy, and it offers clear advice about how to gain the understandings needed in order to validate the idea that a given policy will have desired effects in a novel setting.)