Agency, action, and norms

How do norms influence behavior? More fundamentally, what is a norm?

The question arises for two separate reasons. First, we are interested in knowing why people behave as they do (agency). And second, we are interested in knowing how large social factors (moral and cognitive frameworks, for example) exert influence over individuals (social causation).

The agency question is the more fundamental. Philosophers typically want to answer the question in terms of a model of practical rationality and deliberation. One philosophical answer derives from Aristotle and represents action as the result of rational deliberation. Individuals have a set of goals and values; they have a set of beliefs about the world; and they deliberate about the choices they confront with the aim of achieving their goals consistent with their values, given their beliefs about the world. But philosophers and thinkers as diverse as Nietzsche, Skinner, or Adorno would take issue with this reason-centered theory of action. Other sources of behavior might include unconscious habits or prejudices, instincts, impulses, emotions, and role-playing. A model that incorporates these diverse possible influences on action is unavoidably complex — but human behavior is likewise complex.

Now let’s try to locate the role of norms within a theory of agency. Norms have to do with the reasons and motives that people have for their actions. A norm is a particular kind of influence on action: it is a rule of behavior that leads someone to do something that is otherwise contrary to immediate impulse or interest. Norms get us to do things we don’t want to do.

We might say, then, that a norm is a rule of behavior — for example, “Don’t wear shorts to a business meeting,” “Don’t take coins from the blind man’s cup,” “Give up your seat on the bus to a disabled person.” And a rule can either be internally or externally represented; this means that the the rule may be internalized into the agent’s process of decision-making, or it may influence the agent’s behavior through punishments and rewards.

Even this simple discussion raises questions, however. Do norms have to be consciously accessible to the agent? Is a moral principle such as “Always keep your promises” a norm, or do ethical principles fall in a different category? Do norms have rational justification, or are they simply an accidental social product like tastes or styles?

As for the ways in which norms influence behavior —

It would seem that there are only a few mechanisms through which norms could possibly influence individual and collective behavior, largely distinguished by being external and internal.

First, it may be that there is an effective mechanism of social education through which each individual develops or activates an internally regulative system of norms or rules. This process can be described as “moral education.” The most superficial observation of social behavior indicates that this is so, and social psychologists and sociologists have some ideas about how these systems work. But the bottom line appears fairly clear: individuals who are reared in normal human settings eventually possess action-behavior systems that embody a set of personal norms that influence their conduct. We might draw the analogy to the example of language learning: a normal human child is exposed to the linguistic behavior of others, and arrives at a psychologically realized grammar that guides his/her own language production.

Second, a norm might be embodied in the attitudes, judgment, and behavior of others in such a way that their actions and reactions create incentives and disincentives for the actor. For example, others may possess a set of norms concerning civility in public discourse, and they may punish or reward others according to whether their words are consistent with these norms. In this case the agent conforms to the requirements of the norm out of a calculation of costs and benefits of performance. (It would appear that there is a possibility of circularity here: the externally imposed norm depends upon the internally embodied norm of enforcement of the content of the rule on the part of others.)

Third, it might be the case that there are some norms of inter-personal behavior that are hard-wired. Some norms might have a biological, evolutionary basis. This is the line of thought that sociobiologists have explored with varying levels of success. The emotional responses that adults have to infants and children probably fall in this category — though it is a conceptually interesting question to consider whether these emotional responses are “norms” or simply features of the affective system. This is relevant to the work that Allan Gibbard does in Wise Choices, Apt Feelings: A Theory of Normative Judgment. Gibbard’s fundamental insight seems to be that there must be an evolutionary basis for the “norm-acquisition system” — the features of human psychology that permit them to acquire certain kinds of moral motives (altruism, friendship, fairness).

So — what can we say about norms? Human beings act on the basis of deliberation, norm, impulse, and emotion. So our theory of practical rationality and action must make a place for the workings of norms. Second, norms are transmitted to individuals through concrete social processes — family experiences, schooling, religious institutions, etc. Our theories of social life must incorporate an account of the processes of normative education through which individuals come to possess a particular normative structure. These experiences are the counterpart to the exposure to language on the part of the infant. And third, norms are socially enforced through the actions of others. So norms are socially embodied — in the institutions of enforcement, the institutions and practices of moral education, and in the practical cognition of the individuals who make up the society.

Prejudice and social framing


People bring highly contingent assumptions, beliefs, and frames to their reading of their social worlds. These framing assumptions are presumably the effect of prior life experiences and learning — this is what we can refer to as the social psychology of social perception.

(It is possible there is some degree of biology here as well; we can’t exclude the possibility that there is a natural-selection basis to a neurophysiology of social perception, as argued by the sociobiologists. The case is not resolved at present. Are there any social impulses that are hard-wired through our evolutionary history?)

Another thing we know about social cognition is that human beings are great storytellers. We can take a small detail and weave it into an orderly narrative. And we are likely to tell stories that play out our expectations, fears, or hopes. We interpret the events and behavior around us in ways that go vastly beyond the slender facts that we observe.

(There is probably a developed area of research on this particular feature of social cognition, analogous to the study of reading or pattern recognition, though I am unaware of such research. But it would go something like this: assemble a set of video clips of people acting and interacting without much explicit context, and ask the subjects to briefly describe what is going on. Insert various social cues and see how that changes the stories subjects construct — for example, change the actor’s clothing or adornments slightly.)

Now let’s see what the point is. I suggest that these features of human social cognition make prejudice and discriminiation a very common feature of social cognition. Take a small dimension of mistrust of strangers; add to this a slight propensity for being uncomfortable with difference; add the usual fact of the information sparsity available in most social interactions; and fold in the degree of fictionalizing and narrative construction that social cognition normally involves — and what are you likely to get? It seems credible that the resulting stories will often enough represent the other in terms that support prejudice, discrimination and fear. And it seems credible that these internalized stories, and the actions and consequences they produce, will reinforce and proliferate the prejudicial stories and behaviors.

This suggests a basis for expecting mechanisms of social cognition that are xenophobic, racist, homophobic, and sexist. It is an unpleasant possibility.

It also suggests that when we advocate for a society based on assumptions of trust, equality, and mutual respect — that we need to be considering as well how to create a learning environment that creates these cognitive habits. We shouldn’t assume that trust and equality are “natural” states of mind, but rather a set of cognitive habits that need to be specifically cultivated.

If this has some credibility, it probably gives some indications of what a non-xenophobic pedagogy ought to look like. It ought to work to provide more background knowledge about human differences — to fill in part of the data gap. It ought to work specifically to defuse the
origins of “stranger anxiety” — to work against the background of fear that structures many human interactions. And it ought to affirmatively make the case for equality among persons — to counteract a tendency for the group superiority stories to emerge.

In other words, a just and equalitarian society needs to be created. It isn’t an accident.

"Folk" sociology


All of us are sociologists, at some level. We have social concepts in terms of which we analyze the social world around us — “boss,” “working class guy,” “politician,” “evangelical”, “millennial generation”. (Stereotypes of groups defined in terms of race and class probably fall in that category.) We operate on the basis of stylized schemata about social causes — what sorts of things influence what other things. And we operate with some stylized social facts. (“Bad economic times make people more suspicious,” “Big cities are more unsafe than towns,” “Elections are decided by big campaign contributions,” “Midwestern people are more socially conservative than Californians.”) Putting all these sorts of assumptions together, we can say that we possess a conceptual framework and causal theory of the social world, which helps us to navigate the social relationships, conflicts, and needs that we have in ordinary life. Action proceeds on the basis of a representation of the world.

What this comes down to is the obvious point that humans are cognitive beings who undertake to conceptualize and explain the world around them; they come up with conceptual schemes and causal hypotheses about how things work, and they construct their plans and actions around these frameworks. We are “cognitive” — we undertake to represent the world around us, based on observation and the creation of organizing concepts. And, of course, many of those concepts and hypotheses are badly grounded; they don’t divide the world in a way that is really illuminating, or they offer stereotypes about how things work that aren’t actually true. (“Don’t bet on red — it’s come up four times in a row, so it’s not likely to come up next time.” That’s a false statement about a series of randomly generated red and black events, and the player who follows this rule will lose to the player who is guided by probability theory.)

This sort of everyday social cognition is similar to what philosophers of psychology call “folk psychology” — the ordinary categories of thought and action that we attribute to each other in order to describe and explain each other’s behavior (intention, belief, pain, anger, …). And philosophers have asked whether there is any relation at all between folk psychology and scientific psychology. (Ian Ravenscroft treats this issue in the philosophy of psychology in his article in the Stanford Encyclopedia of Philosophy.) Some philosophers have argued that the most fundamental and scientifically satisfactory explanations of individual behavior will be couched in terms that bear no relation at all to the concepts of ordinary mentalistic psychology.

So what is the status of folk sociology? We can ask several questions about this common sense framework of social cognition and expectation. First, where does it come from? What are the social processes of learning through which we arrive at the specifics of the social assumptions and concepts that we employ? Second, to what extent are there important differences across individuals with respect to the features of their social frameworks? (For example, we can explore whether there are cultural and national differences, gender and race differences, or generational differences across different groups and cohorts.) Third, we can examine the degree to which these categories and assumptions are rigid, or whether they are open to modification through additional experience — “learning”.

A different question, though, is also important: What is the relationship between these ordinary sociological frameworks and scientific sociology? Is there a relationship at all? Can scientific sociology learn from common sense? And can common sense improve its grasp of the social world through interaction with scientific sociology? Might we speculate that ordinary common sense does a fairly good job of picking out the salient features of the social world? Or, on the contrary, might we judge that the categories of “folk” sociology are about as misleading as pre-modern, magical concepts of nature? Or perhaps, might we say that rigorous scientific sociology can serve to refine and improve upon our “folk” concepts of the social world — lead us to abandon categories such as race, for example, in our efforts to understand Obama, Michael Jordan, and DuBois?

The example of the natural sciences would lead us to one set of answers on these questions: “folk” knowledge of the natural world was not in fact a good guide to scientific physics, and the concepts of modern physics bear little intelligible relationship to common sense concepts of ordinary experience of tables and chairs. One way of putting this is to say that physics concepts are “theoretical”, whereas common sense concepts are “phenomenological” (based on immediate experience).

Whether that is a valid distinction or not in physics, it probably is not a valid distinction in the social sciences. Social life is more transparent than the physical world; so our best scientific understanding of the social ought to bear some understandable relationship to the categories of ordinary social cognition. Common sense may not be highly specific in theorizing the concept of “power” in social life; but the phenomena of power are in fact fairly visible, and ordinary common sense captures these phenomena reasonably well. It is possible to paraphrase virtually any esoteric sociological thesis about power, in terms that are understandable in ordinary social experience. And likewise for exploitation, alienation, disaffection, racism, prejudice, discrimination, and affinity groups (to list a grab bag of sociological concepts): each of these concepts can be related to ordinary experiences and ordinary, common sense categories of social interaction.

So here is a possible answer to our original question — how do ordinary social concepts relate to those of scientific sociology? We can say that there ought to be a critical but intelligible relationship between the two sets of concepts. Scientific sociology can point out the limitations and blind spots of ordinary ways of representing the social world. But ordinary social observation and conceptualization constitute the real content of sociological hypothesis and theory. So both systems of social knowledge fruitfully interact with each other, and — ideally — lead to a rising level of competence in cognizing and understanding society.

%d bloggers like this: