SOCIAL STRUCTURE

"Social structure" is a general term for any collective social circumstance that cannot be altered by isolated actions and thus is fixed or given for the individual. It thus provides a context, environment, or fixed backdrop for action. The size of organizations, the distribution of activities in space, shared language, and the distribution of wealth all might be regarded as social structural circumstances that set limits on feasible activities for individuals.

Social structure is objective in the sense that it is the same for everyone and is beyond the capacity for alteration by any individual. Accordingly, social structure often is spoken of in the singular and as a thing apart, as if there were only one from whose effects no one can escape. This usage masks disagreement about the exact extension of the term but reflects the intention of authors to highlight abstract patterns as an inflexible collective circumstance to which individuals must adapt.

Social structure, or the weaker structural regularities, arises because of the prevalence of social routine. Many social patterns change very slowly either through unmotivated inertia, through willful efforts to renew or reproduce them, or as a collective consequence of individual efforts undertaken for independent reasons. An image or picture, such as a map colored by the linguistic practices of the inhabitants of geographic areas, will lose accuracy slowly and often would remain largely accurate after a century or more. Such substantial durability, along with the accompanying slow continuous change, suggests the possibility of regularities or even scientific laws governing the phenomena that underlie the description.

Routines endure and structural regularities persist for at least three general reasons. Social life is subject to physical constraints such as distance. Thus, most people live close to where they work or do both at one place. For related reasons, many persons maintain stable residences. Furthermore, many people need or desire the company or cooperation of representative social types, such as those who share their religious convictions, or particular work skills. Accordingly, one can associate social attributes with geographic maps. This was a central activity of the Chicago school of sociology (Park and Burgess 1924) and gave rise to the perspective of human ecology (Hawley 1986). The specialization of social types and activities in space is subject to powerful incentives that induce similarity in the face of turnover among individuals. For example, ethnic concentrations result in specialized facilities, such as food shops, that attract replacements that conserve the ethnic character. Such patterns often persist beyond the lifetimes of the people who initiated them.

A second source of routine is limited learning capacity or the complexity of many social activities. Linguistic rules, moral codes, and work skills illustrate social capacities whose acquisition requires considerable time and effort. This socialization often requires extended exposure to others who know the routines well, especially when the delicate skills of interpretation are involved.

The difficulties of acquiring capacity can confound individual wills. Bernstein (1975) described how linguistic conventions acquired in the home reflect the conditions of adult work and render individuals unsuited for occupations that are not similar to those of their parents. In the same way, a New Yorker who wished to speak in Latin would have to make a huge investment in learning a novel linguistic code. However, this would not undo the investment in English by other New Yorkers, and thus Latin would be impractical for directing taxi drivers. Similar reasons impel the adoption of the abrasive social style of New Yorkers by newcomers. The general principle is that most people must adapt to many surrounding ways of doing things because those ways change so slowly.

A third source of structural regularity is laws governing averages. An example is the suicide rates studied by Durkheim. People commit suicide for a variety of personal motives and the act is never repeated by anyone, yet the frequency of the act is fairly stable over time and thus is stably different among different populations. This is the case because variable causes tend to average into stable totals whenever many instances are drawn from constant underlying conditions. Many of the rates that result are sufficiently stable to sustain plans and projections, which in turn can be embedded in routines, even though the underlying activity is very complex in its detailed causation.

The several sources of stable routines underlie the properties that frequently are associated with proposed structural regularities. Structural regularities often are depicted as abstract, enduring, and operative across a large scale of units. These attributes reflect genesis, for many structural regularities ultimately stem from the long historical process of imposing routines that made a large scale feasible. For example, Tilly (1975) shows how the modern European state resulted from parallel decisions by state makers forced by military competition to pursue centralization by reordering the established routines of ordinary people. The history of collective contention (Tilly 1986) can be seen as the efforts of the victims to defend older patterns against intrusions by state agents such as tax collectors and against the vast reorganization of work and fortune implicit in the expansion of capitalism.

Mann (1986) argues that large-scale cooperation rests on enduring patterns of power. For example, shared religious ideology is a form of power because it makes people subject to claims on their activity. Rapid religious change is not infrequently the result of conquest. Once it has been established, religion is often compulsory. Coercion aside, religious conformity can provide insurance against the risks and pains of social isolation. Other large-scale patterns, such as the division of labor, are maintained in the face of considerable shifting by persons among different roles. This often is implemented by powerful actors who are motivated to induce (or coerce) approximate substitutes to fill in for those who withdraw (or die). In such terms, abstract stability, a large scale, and consequent duration often can be seen to be sustained through underlying causal regularities. In human terms, the reproduction of social structure consists of a myriad of modest efforts that sum to a stable result.

This interdependence underlies the transcendence of abstract structural regularity over individual will. Generally, one cannot learn more than locally applicable routines and must rely on others for critical needs. Thus, one assumes that the staff members in the emergency room will not all take the day off. This frees accountants from the necessity of acquiring medical skill to meet their own needs. As a result, the details of actual routines are known only locally, and the only possible knowledge of the overall pattern is coarse or abstract. Even accountants cannot count up the details they must count on. A further implication is that the alternatives to enacting the routines with which one is familiar are often limited. It requires time on a historical scale to construct such a pattern. That history has happened, and if the conditions that made narrow, specialized learning practical suddenly came unglued most people would be in a terrible fix.

A special case of social routines consists of those worked out with others, who then are often hard to replace. Replacement is generally more troublesome as the duration is longer, such as many kinship bonds, which can be effectively irreplaceable for many adults. These social relations can be mapped as social networks describing the pattern of the links that surround individuals. The analysis of such patterns is not infrequently (or unreasonably) called "structural analysis," though it hardly exhausts the term.

Elaborate routines, especially social ties, are subject to pressures toward isomorphy, which is defined as a common anatomy or structure. An example is the formation of families, which are different in detail but share common features partly in response to common problems that must be solved within a shared environment. Goode (1970) analyzes the sources and consequences of such regularities. In modern societies, assumptions about such features often are written into administrative procedures such as tax codes, which provide further impetus for individuals to adopt a variant of the pattern defined as normal. Changing and varied individual desires are often in conflict with those pressures to cooperate in the reproduction of the supposedly "normal" pattern.

Emergent properties that apply to wholes but not to parts often are attributed to social structure. Some properties, such as size distributions and complexity, do not have direct individual analogues. Others arise because the net result of many partially independent actions can be different from the intentions of individuals. Thus, markets with many participants can experience crashes in value when many people try to sell in anticipation that others are about to do so, producing a result that no one desires. Kindleberger (1980) describes the recurrence of such crises. Routines are executed by fallible humans and are only locally adapted, somewhat independent, and imperfectly flexible. Many properties of the resulting averages or combinations do not follow from the components in any simple sense.

As the preceding analysis suggests, structural visions are various. One unifying theme is an appeal to abstract, extraindividual patterns that change slowly or not at all. A second unifying theme is that those regularities cause or condition many of the choices and behaviors of individuals. A final common theme is less unifying than divisive. Some structural visions are accompanied by claims of centrality. A particular array of simple elements is proclaimed, often on metatheoretical or philosophic grounds, to be the central deep structure whose inevitable unfolding underlies a vast array of surface appearances. Such comprehensive views have inspired competing, incompatible schools of thought on whose behalf a claim is sometimes made to the structural vision of society or the human condition.

Most of these structural visions are comprehensive worldviews that require detailed study in their own right. Among the most prominent are those of Marx and Freud, but there have been structuralist movements in nearly every field of social studies. Nearly all proceed from some highly abstract characterization of the human mind, laws of thought, or the human condition. All of social or mental life is viewed as a manifestation of the reproduction of such elements, often unfolding dialectically. This is presented as the inevitable underpinnings of individual or collective biographies. Piaget (1970) has provided an unusually concise description of an interdisciplinary structuralism based on mathematical progress; this description parallels his more famous theory of discontinuous advancement in human cognitive development. Originators and their descendants often delight in such subtle and insightful reductions of familiar patterns to the chosen central supports.

The term "structure" is most commonly employed in sociology without these all-encompassing ambitions. In empirical sociology, especially quantitative studies based on random samples of persons, the term is invoked for varied efforts to use the larger and often more durable features of social life as explanatory factors for individual conduct and outcomes. The most common contrast is with individual-level causes, including attitudes and aspirations. Sometimes attributes such as race, gender, and class are labeled structural to imply that the underlying mechanism is an external force imposed on individuals independently of their wills.

The reasoning behind this is not always explicit, but the usage is justifiable. Generally, the factors labeled structural are alternatives among a differentiated array of possibilities to which individuals are confined for substantial periods. "Structure" then refers to the differentiating average conditions in which people live their lives. At least implicitly, such differences correspond to differences in the routines employed to adapt to local conditions as well as to resources that render routines practical. Classifying people by indicators of the local conditions that surround them reflects the opportunities they have for association and hence for processes such as influence, cooperation, and victimization. Some characterizations also correspond to labels, most notably race and gender, and broadly indicate common tendencies in routines of others to which one is likely to be exposed. Such differences are quite stable, impersonal, and hard to evade. Taken together, these differences in conditions contribute to differences in average responses or individual behaviors.

There is some confusion about the nature of such structural causation, which often is framed as an alternative explanation to individual choice. Persuasive force often comes from stories in which the predominant outcome is made to feel inevitable. This is at odds with the normal empirical result of a difference in tendency or proportion. Rules that hold without exception are rare. This should be expected. Structural abstractions mask much detail that varies. The implicit reference is to averages over multiple executions of complex routines. To take an obvious example, racial discrimination involving job applicants is not invariant but occurs often enough to lead to considerable differences.

Structural causes are not literally the antithesis of individual choice. More precisely, they reflect patterns over which individuals have limited control. The binding force of structural regularity is intrinsically probabilistic. People almost invariably have options, and exceptions to regularities are somewhere in reach. However, established structure—ultimately routines acquired over time, bonds developed to particular others, and the meshed ways of doing that result—exerts a frictional tug. Friction is implicit in the pain of forgone routines and the time required to work out new ones. Such pains may be amplified when those who benefit from regularities exert their power to maintain them. On any large scale, the path of least resistance consists of acting today nearly the same way as one acted yesterday. By no means does this rule out individual exceptions, resistance, or willful alterations to parts of the overall web, but friction is cumulative. For example, the rupture and replacement of one bond are quite different from the rupturing of all bonds at once. Similarly, any single person may change jobs, although in practice only to a very limited range of alternatives, yet if all jobs were randomly reshuffled one day, nearly all would go undone, for every job would be subject to the incompetence of the "first day on the job." In summary, the frictional forces of social structure do not rule out rare and/or modest exceptions but generally ensure that wholesale, simultaneous exceptions are rare to the vanishing point.

Empirical applications generally draw on fragments of social structure that are taken as conditioning factors for particular outcomes. The larger challenge is to translate the impersonal, durable complexity of stable differences in condition into a formal calculus, or a theory of social structure. Parsons's (1951) extensive analysis of the logic of social systems was an early and seminal attempt. His student Merton, under the banner of "theories of the middle range," provided a more easily applied set of general tools for structural analysis. Several of Merton's students, including Boudon, Blau, and Coleman, further developed formal calculi for social structure that benefit from the use of mathematical tools.

Parsons's complex system begins with the conditions for stabilizing interaction or, in current terms, meshing routines. Parsons characterizes the routines that govern choice as extended chains of logic linking means to ends. At their most abstract, those chains are anchored in ultimate ends, or values. Durable stability results from consensus on the values that are installed in individuals by more or less extended socialization.

In Parsons's view, the logical chains governing decision making are morally potent norms, or rules governing social conduct. The durable web that shapes individual choice is therefore the complex of norms animated by the anchoring ultimate values. Parsons imposes on this a logical calculus of the different functions necessary for ensuring that the pattern is resistant to shocks that draw it away from equilibrium. A concomitant of this theme of differentiation is complementary specialization in distinct but interdependent expectations bundled into the social roles enacted by different players.

Parsons's calculus of the functional necessities of meshing differentiated normative specifications proved widely compelling but difficult to apply. His presentation is notoriously hard to read. Applications of the scheme usually consisted of classifying normative elements into taxonomies delimiting functional contributions. These qualitative operations were by no means mechanical or easily communicated as a stable procedure that would steer different investigators to identical results. This rendered moot the possibility of generating conclusions from initial conditions through the application of formal tools. In a similar way, while many were inclined to agree that Parsons's system illuminated how a social system governed by a logic over normative rules might work, it was less than evident that concrete social systems had such logical coherence.

Merton's (1968) "theories of the middle range" provided a more readily applicable set of tools. Like Parsons, Merton proposed that the enduring regularities that make up social structure are normatively defined. However, instead of attempting to calculate over extended normative webs, he drew attention to the implications of positions. Thus, he emphasized that roles place individuals in relations with concrete others or that membership in groups, both present and anticipated, provides reference points for calculating comparisons of expectations and outcomes. Unlike Parsons's more elaborate concerns, Merton's lent themselves to the construction and interpretation of surveys and other manageable research projects.

Merton did not assume, as Parsons did, that norms and roles can be divined from an overarching logic. More frequently, he treated contrasting norms as empirical counterparts of lay distinctions among different roles or group memberships. This can be viewed as a central motivation for the common use of the structural concepts outlined above. However, Merton more often used factual (or readily inferred) norms grounded in different stable positions to highlight dilemmas. Concrete people could be understood as facing practical problems of resolving competing and often contrary normative standards. This strategy of framing the practical problem as the resolution of contrary expectations frequently leads to insight into choices that at first seem senseless or even self-defeating.

Merton's analyses rested on qualitative inferences, often turning on the meaning of norms. One of Merton's students, Boudon (1982), provides formulations in which social structure refers to numerically definite distributions so that the implications of such extraindividual constraints emerge from formal calculations. For example, he posits an array of young persons committed to personal advancement who make investments in education. However, when all do what is individually sensible, the collective result illustrates Merton's unintended consequences. If there is a fixed and therefore scarce supply of desired positions that will go to those who have the most education, many of those who invest will discover that their efforts are frustrated by the simultaneous striving of others. Boudon provides many illustrations of the perverse effects that can obtain when individual motive operates against a backdrop of a fixed system of positions.

Another of Merton's students, Blau (1977), presents a deductive structural theory based on the notion that social structure consists of arrays of positions, which he calls parameters. Blau divides differentiation into two types: among unranked or nominal categories such as religion and among continuous arrays of ranked positions that differ in their amounts of a scarce and valued resource. The distribution of individuals over positions gives rise to numerical properties of whole social structures, including the heterogeneity of nominal differences, inequality among ranks, and consolidation intersection, or the degree of correlation independence of positions on separate dimensions.

Blau's concept of social structure leads to differences in the sizes of collections of individuals occupying different positions. Size in turn strongly conditions the rate of interaction, or social association. More differentiated structures result in higher rates of intergroup association, and Blau argues that this leads to the successful meshing of routines, or social integration. The intersection of different dimensions, which results in even smaller subgroups defined by multiple positions, also enhances social integration. Conversely, the consolidation of dimensions, homogeneity rather than heterogeneity, diminishes rates of intergroup contact and hence hinders social integration. Inequality emerges as a special case that illustrates Blau's taste for paradoxical results. Greater inequality leads to smaller strata and fosters intergroup relations, but those relations often take the form of interpersonal conflict, including crime (Blau and Blau 1982).

Blau's notions are particularly suitable for research application because his notion of structure more or less directly corresponds to widely used operationalizations such as gender, race, ethnicity, religion, occupational rank, and wealth. Of course, these are social constructs and in some final analysis are defined by norms and other ideal elements. At the same time, they are for most people most of the time subject to slow or even no change. This sustains the usefulness of a numerical calculus that rests on the notion that size is an objective, impersonal, and durable reality.

Coleman (1990) provides one of the most ambitious attempts to specify social structure as a mathematically tractable map of interdependence. He posits actors with rights of control over their own actions and over tangible things desired by others or resources. His actors maximize the achievement of their desires by exchanging their control in return for that which others control. The result in general is an equilibrium in which initial control in conjunction with the desires of others produces differential power. Within this apparatus, Coleman is able to provide a rigorous analysis of the emergence of larger-scale phenomena, including groups, norms, and corporate actors.

Although all these accounts lie along a single path of intellectual descent, there is a major divide with respect to the elemental nature of social structure. For Parsons, it is an interdependent complex of norms. Unfortunately, there does not exist at present any way to formalize or calculate the mutual implications in a web of symbolic elements. Later analysts who have gone much farther in rendering complexity calculable have done so from "hard" assumptions that take social structure from the outset as a set of objective positions (with objective properties) so that size, distribution, rates of exchange, and so forth, can be treated mathematically.

More recent treatments have built on the rich, although eclectic, tradition of taking relations or social ties that are knitted into networks as fundamental. One point of departure is Granovetter's (1973) observation that weak ties are surprisingly efficacious in securing resources, notably access to better jobs. Weak ties are most likely to form bridges between clusters of interconnected and thus redundant strong ties. Burt (1992) generalizes this, suggesting that "structural holes," or gaps spanned by positions whose ties unite the otherwise disconnected, are a potent source of advantage. He was able to display supporting evidence from contexts as diverse as executives competing for promotion and sectors of an industrial economy. Burt's concepts are derivative in the best sense; that is, they are a conceptual refinement that moves on to novel terrain, building on what he can take as an established view of social structure.

Tilly (1998) has distilled from network concerns a potent challenge to much received thinking about stratification. He proposes that "durable inequality" reverberates from underlying schemas governing how networks are formed. Categorical divisions such as race, gender, nationality, and citizenship are embedded in widely shared, deeply learned propensities for action, or routines. Recurrent organizational problems, such as assigning work and dividing rewards, are most easily and durably resolved when they are consonant with widely shared assumptions about categorical differences. Somewhat like Burt, Tilly focuses less on origins and more on implications. He examines how relational considerations secure inequalities through persistent configurations of exploitation and resource hoarding that are diffused by emulation and ultimately underpinned by adaptation. In this view, social structure is not globally coherent or uniform but is, somewhat like DNA, a complex melange constructed from varying combinations of a few very simple elements.

A noteworthy gap here is that the proponents of formal theory (and those proposing building blocks) tend to posit or assume "hard" properties, giving limited attention to how or why the hypothesized elementary patterns emerged or became predominant. This leaves open issues of variability and interpretive options (or meaning) that others see as fundamental. Indeed, some authors believe that human judgment is distinctive and that no mechanical analogue or simulation of human society (Habermas 1987) or human cognition, (Penrose 1989) will ever be possible.

In summary, there are no widely accepted sets of notions that capture all the properties that have been seen as fundamental to the concept of social structure. The huge catalogue of demonstrated effects of structural regularities cannot be organized in a tidy way. Enthusiasm for the different attempts to represent the concept in compact terms varies widely. Sufficiently close attention to the details of competing claims could convince one that no shared subject is really at issue. As in the analysis of social structure itself, it is necessary to carefully select the right degree of abstraction and appropriate pattern of highlighting to discern any common pattern in the competing pictures, but there is nevertheless a pattern to be found.

references

Bernstein, Basil 1975 Class, Codes, and Control. New York: Schocken.

Blau, Peter 1977 Inequality and Heterogeneity. New York: Academic Press.

Blau, Judith, and Peter Blau 1982 "The Cost of Inequality: Metropolitan Structure and Violent Crime." American Sociological Review 47:114–129.

Boudon, Raymond 1982 The Unintended Consequences of Social Action. New York: St. Martin's.

Burt, Ronald S. 1992 Structural Holes. Cambridge, Mass.: Harvard University Press.

Coleman, James S. 1990 Foundations of Social Theory. Cambridge, Mass.: Harvard University Press.

Goode, William J. 1970 World Revolution in Family Patterns. New York: Free Press.

Granovetter, Mark 1973 "The Strength of Weak Ties." American Journal of Sociology 78:1360–1380.

Habermas, Jurgen 1987 The Theory of Communicative Action, vol. 2: Lifeworld and System: A Critique of Functionalist Reason. Boston: Beacon Press.

Hawley, Amos 1986 Human Ecology. Chicago: University of Chicago Press.

Kindleberger, Charles 1980 Panics, Manias, and Crashes. New York: Basic Books.

Mann, Michael 1986 The Sources of Social Power, vol. 1: A History of Social Power from the Beginning to A.D. 1760. New York: Cambridge University Press.

Merton, Robert 1968 Social Theory and Social Structure. New York: Free Press.

Park, Robert, and Ernest Burgess 1924 Introduction to the Science of Sociology. Chicago: University of Chicago Press.

Parsons, Talcott 1951 The Social System. New York: Free Press.

Penrose, Roger 1989 The Emperor's New Mind. New York: Oxford University Press.

Piaget, Jean 1970 Structuralism. New York: Basic Books.

Tilly, Charles 1975 The Formation of Nation States in Western Europe. Princeton, N.J.: Princeton University Press.

—— 1986 The Contentious French. Cambridge, Mass.: Harvard University Press.

—— 1998 Durable Inequality. Berkeley: University of California Press.

Steven L. Rytina