Anomalous Presumptions

In The Social Theory of Practices: Tradition, Tacit Knowledge, and Presuppositions (1994), Stephen Turner mounts a sophisticated attack on the idea of “social practices” as some kind of supra-individual entities. (I will call these “coarse grained entities” below, to avoid the value laden implications of “higher level” or similar locutions.) This attack is part of a broad effort to support methodological individualism and attack any theory or evidence that contradicts it.

This problem is important in its own right, but it gains additional significance in the context of population thinking. If “only the population is real” then should we regard claims about coarse grained entities as fictitious? Of course my answer is “no”, but this requires careful analysis. We want to develop a firm distinction that allows us to adopt a realistic ontology for these coarse grained entities, but reject any treatment of them as abstract entities that somehow exist independent of the population.

Turner’s book is worth a response because it is a relatively clear and thoughtful examination of the argument against supra-individual entities. Analyzing Turner can help us figure out why he (and the methodological individualists in general) are wrong, but also can bring into clearer focus the nature, dynamics, and importance of coarse grained entities.

The major flaws in Turner’s argument can be grouped under three headings: he doesn’t recognize the possibility of multi-scale ontologies; he can’t explain the maintenance of social coherence; and he can’t explain the creative elaboration of novel performances that build on existing practices.

Doesn’t recognize multi-grain ontologies

Coarse grained entities are are exhaustively comprised of a set of finer grained sub-entities (e.g. bodies and cells), but in general the coarser grained entity cannot simply be reduced to the set of sub-entities, because to constitute the coarser grained entity, the sub-entities have to have correlated states. If the states of the sub-entities were de-correlated, but the individual sub-entities were left intact, the coarser grained entity would cease to exist. For example, if all the cells in my body were de-correlated by independently setting each to some “normal” state at random, my heart and brain (at least) would cease to function, and I would probably die. In contrast, if we performed the same operation on all the cell-sized pieces of a rock, it would have no perceptible effect.

Coarser grained entities have causal power in two (related) senses. First, the entity can be the recipient of and source of causal chains that cannot be reduced to a set of causal chains affecting some of its sub-entities. For example, when I see a chair, the stimulus array causes me to see by stimulating a set of photo-receptive cells in my retina, but the causal process cannot be reduced to a set of causal interactions distributed over these photo-receptive cells. My perception of a chair is a function of the correlations in the stimulus array, combined with the perceptual context— my posture, angle of gaze, expectations, etc.— which are constituted by correlations among many other cells in my body. Only when all these correlations fit together in the appropriate way can a stimulus array cause me to see a chair, and my seeing a chair is itself a change in these correlations. Thus the whole correlated set of cells necessary for the perception is the target of the causal chain, not the individual cells in the set.

Second, a coarse grained entity maintains itself by causing appropriate patterns of correlations to be produced in its sub-entities, and also by causing the production or recruitment of new sub-entities. My body maintains itself by monitoring and managing its cells, and sometimes by destroying them— for example if they have become infected or cancerous. It also constantly grows new cells and discards older ones. These processes are carried out entirely by activity in the cells of my body, since my body is exhaustively constituted by my cells. However without the correlations between these cells they would not carry out these tasks in a way that would maintain my body rather than destroy it, so we cannot attribute the causal power to maintain my body to the individual cells, but only to the body as a coarser grained entity.

Turner, in attributing social activities to “habit”, makes the mistake of believing that since he has identified the individual source of all the actions that make up these social activities, no higher level entity is needed, and indeed that none can exist. However he completely fails to explain the persistent and causally crucial correlations between the individual “habits” that constitute social order. In the same way that a body cannot be reduced to its cells, social activities cannot be reduced to habits of members of that society, even though the exercise of those habits exhaustively constitutes social activities.

Doesn’t explain maintenance of coherence

In talking about habits as the exhaustive explanation for social activity, Turner ignores the problem of maintaining social coherence over time. He views habits as learned through participation in social observances, and asserts (I think correctly) that they may differ widely between individuals and typically consist largely of tacit, inarticulable knowledge.

However given this model, and no coarser grained entity with causal power, the continued maintenance of social coherence is simply inexplicable. Learning any habit involves some “noise”— the habit as learned will lead the practitioner to conduct observances somewhat differently than those she learned it from. For example, each new English speaker will use words somewhat differently from both previous speakers and from each other. Iterated for a few generations, the result will be the dissolution of English as a mutually intelligible language.

Obviously this conclusion is absurd. People have overwhelming incentives to make themselves intelligible to others, and plenty of opportunities to test their intelligibility and correct any failures. In other words, there are strong forces maintaining and evolving the correlations between people’s language habits to maintain rough coherence in the community of English speakers.

But this is precisely the point that Turner generally overlooks and sometimes explicitly rejects. We cannot explain the coherence of social activity— the existence of “shared practices” (Turner’s bête noir)— solely at the level of individual habit acquisition. We must also discuss the correlations between habits, and analyze the way that some wide-spread correlations cause the continuing maintenance of those correlations as people acquire new habits or change old ones. In the process we should also explain why other widespread correlations are causally inert and do not actively maintain themselves, and why some correlations change more quickly than others, are more strongly affected by exogenous changes in the society, and so forth. These are the questions about social practices discussed by traditional social (in different language), and these are exactly the questions that Turner tries to sweep into the dustbin of philosophically incoherent and unnecessary explanations.

Doesn’t explain congruent elaboration

Even at the level of individual habits Turner’s account has a gaping hole that cannot be filled without invalidating his conclusions. As I noted above, he claims that individuals learn idiosyncratic, tacit habits through participation in social observances (and I agree). But Turner’s habits simply enable their possessor to produce a sufficiently similar performances at subsequent appropriate occasions (the ability to recognize appropriate occasions is presumably also part of the habit).

The problem is that social actors continually adapt and elaborate their performances in novel ways and deploy them in novel circumstances, and other social actors understand these novel performances and respond appropriately, often without consciously noting the novelty of the performance. For example, people often use words in novel ways or in novel contexts, and others generally understand them well enough, and respond in ways that often elaborate and extend the novel use.

Turner’s emphasis on the acquisition of habits in overt social observance, and their idiosyncratic, individual character bars a response to this problem. If habits are just idiosyncratic ways of adequately producing performances similar to those previously observed or participated in, this ability to produce and respond to appropriate novelty is simply inexplicable.

Again, the solution is to accept the necessity of coarse grained structure that cannot be reduced to individual habits, and to follow out the implications. People do not just, or even primarily, learn habits for participating in observances. They learn a complex set of skills for creating and maintaining correlations between their behavior and others. These correlations are maintained by accurately anticipating others’ actions, and acting in ways that shape others’ anticipations. Such skills are no more mysterious than our ability to domesticate an animal by shaping its anticipations, or to capture prey by anticipating its actions. Of course in a social environment, with mutual anticipation and shaping of behavior, the possibilities are fundamentally richer.

In fully developed social action, our exercise of these skills depends on deep background knowledge of the other actors’ prior relevant experience, based on overlap between our lives. Some of this experience may be idiosyncratic, but much of it can be presumed to be part of the background of anyone who grew up in a given culture, or underwent a given type of professional training, or became an aficionado of a given type of performance, and so forth. Our performances succeed by playing off pre-existing correlations implicit in this background.Conversely, the background itself is constituted by many such performances. The correlations that are stable across long series of such performances, among many different subsets of actors, form the structural elements of a given domain of practice. Note that such correlations have an interesting reflexive stability: they are reinforced in performances where actors are playing off their presence in the background. To get a bit technical, this means that the structure of a given domain of practice is a fixed point of the correlations exhibited by performances in that domain.

Summary

Based on this review, we can see that:

1. Coarse grained entities exist in a strong causal sense in various multi-scale domains.
2. The coarse grained entities are constituted by correlations between the states of finer grained entities that make them up.
3. Coarse grained entities maintain themselves by producing and/or recruiting fine grained entities, destroying or expelling ones that don’t maintain the necessary correlations, and managing the correlations between the ones that are currently constituting the coarse grained entity.
4. Social domains are full of such coarse grained entities, and social life is almost entirely made up of the actions of such entities.
5. The correlations in social practices are fixed points of social performance in which each participant plays off of the correlations in the background available to all the other participants.
6. Thus coarse grained social entities (social practices) while complex and subtle in their causal structure, are not ontologically mysterious or problematic.

Filed by Jed at 4:32 pm under Philosophy, Population thinking
2 Comments