The Concept of Social Institution

The Concept of Social Institution

[Citation: Gaurang R. Sahay, The Concept of Social Institution (2012, Unpublished)]

While rarely giving reasons for doing so, social scientists apply the term social institution to an amazing array of phenomena such as marriage, family, kinship, caste, religion, school, panchayat, etc. Such inclusiveness prompts questions about what these phenomena have in common. What makes anything a social institution? Without explicit conceptualization or criteria, it is difficult to tell. According to Hughes (1971:5), “The only idea common to all usages of the term institution is that of some sort of establishment of relative permanence of a distinctly social sort”. This comment, which Hughes made at the 1935 American Sociological Society meetings, is as apt today as it was then. Hughes was acknowledging the failure of sociologists — and psychologists and economists and so on — to specify the meaning(s) of the social institution concept.

The Term Social Institution in Twentieth-Century Sociology

Although this feature is only sometimes explicitly noted, the most universal theme in sociologists’ definition of social institution is endurance: An institution persists, it is not ephemeral. Other typical features that are implied or asserted are institutions’ external, macro, and constraining qualities and their equation with “major” societal realms such as family, religion, education, polity, and economy. Some authors say a particular set of institutions, those that meet “basic” societal “goals” or” needs,” is required for a society to exist; Berger and Luckmann (1966:55) describe society as “an agglomeration of institutions.” Many imply that institutions are harmonious and benevolent (e.g., Searle 1969, 1995) although Balzer (2003) and Nisbet (1953), among others, disagree. Bellah and colleagues (1991) claim that institutions have a moral or ethical quality. Some scholars apply the term institutions to formal organizations, for example, schools, nursing homes, universities. Some focus on what institutions are, others on what they do. Many represent institutions as internally consistent, conflict-free, fixed, and unchanging, yet a growing number focus on conflicts, internal inconsistencies, and change as well as power, inequalities, privilege, and disadvantage as institutional features (see Balzer 2003). Nearly all conceptions depict institutions as controlling, obligating, or inhibiting, although some also note their facilitating and empowering effects (see Berger & Luckmann 1966, Giddens 1984, and March & Olsen 1989 on this point). In the mid-twentieth century, many sociologists equated social institutions with ideas, norms, values, or beliefs with no attention to processes or practices. This narrow and static definition has been under challenge for some time by scholars who assert the centrality of practices in constituting social institutions (Giddens 1984; Schatzki, Knorr-Cetina & Von Savigny 2001).

Relative to practices that constitute institutions, many contemporary scholars focus on rules, procedures, customs, and routines. March and Olsen (1989: 24), for instance, define institutions as sets of rules and routines that are “constructed around clusters of appropriate activities” and “appropriate procedures”. Giddens (1984: 21–23) acknowledges rules, procedures, and “enactment” in saying “Let us regard the rules of social life . . . as techniques or generalizable procedures [emphasis mine] applied in the enactment/reproduction of social practices. . . . The most important aspects of structure are rules and resources recursively involved in institutions”. Similarly, Connell (1987: 140) says social institution “classically signifies custom, routine, and repetition”.

Many sociology encyclopedias, dictionaries, and introductory textbooks define institutions globally, with a vague sentence or two followed by a list of examples. Two sources reflect both this pattern and the older, idea-focused definition. The 2000 edition of the Encyclopedia of Sociology (Borgatta & Montgomery 2000) lacks an entry for institutions. Instead, readers who search for the term are told to “see American Society.” The American Society entry (Williams 2000: 142) defines institutions in one sentence: “Institution here means a definite set of interrelated norms, beliefs, and values centered on important and recurrent social needs and activities.” The Blackwell Dictionary of Sociology (Johnson 2000: 157), an e-dictionary, defines institutions similarly: Institutions are “an enduring set of ideas about how to accomplish goals generally recognized as important in a society”. Johnson’s invocation of “ideas,” like Williams’s of “norms, beliefs, and values,” implies that institutions are subjective (see below) rather than material or behavioral in content. Johnson’s use of the term enduring implies a time element, as does Williams’s use of the term recurrent, although neither discusses time explicitly. Neither says where goals or “recurrent social needs” come from nor how their importance is determined. Both fail to mention practices or acknowledge that institutions are conflicted, dynamic, and changing. Finally, both fail to address the relationship of individuals to institutions.

The examples listed by these sources, also typical, include “family and kinship, social stratification, economic system, the polity, education, and religion” (Williams 2000:142) and “family, religious, economic, educational, healing, and political institutions” (Johnson 2000:157). Giddens (1984) objects to equating institutions with such “substantive” lists because doing so implies that the phenomena are universal, necessary, and unchanging and, equally troublesome, that they stand “outside human agency.” As I explain later, I share Giddens’s concern.

1. Early Twentieth-Century Conceptions

In a commentary on a newly discovered and unpublished manuscript by Parsons (written in the 1930s but not published until 1990), Charles Camic (1990) addresses Parsons’s influence in shifting the concept of social institutions away from behavior and practices toward a “subjective only” definition. According to Camic, social institution was a key concept for “anthropologists, political theorists, economists, sociologists and even psychologists” in the early twentieth century (1990:315). However, simultaneous with Parsons’s rising influence on U.S. sociology, the meanings scholars attached to the institution concept became less consistent than they had been earlier.

Among the early twentieth century scholars who touted the concept’s utility for understanding society were economist Walton Hamilton (1932) and sociologists Charles Horton Cooley ([1909] 1962), L.T. Hobhouse ([1924] 1966), and William Graham Sumner ([1906] 1979). Hamilton, one of Parsons’s teachers at Amherst College, defined an institution as a complex phenomenon with multiple facets: “‘a cluster of social usages,’ possessing some ‘prevalence and permanence,’ ‘embedded in the habits of a group of people or the customs of a people’ and accompanied by formal and informal ‘sanctions’ that ‘function’ to ‘fix the confines of and impose form upon the activities of human beings’” (Hamilton 1932:84 as cited in Camic 1990:316). According to Camic, this formulation was “not idiosyncratic” but was widely embraced, by for example Cooley, Hobhouse, and Sumner who viewed institutions as central to sociology’s domain. These scholars focused on the practices (or usages) aspect of institutions, heir endurance over time, and their constraining influence on societal members. That is, institutions were “recognized and established usages governing [social] relations (in conjunction with the ‘principles’ and ‘organization’ of those usages)” (cited in Camic 1990:316 from Hobhouse [1924] 1966:49); “usages that are crystallize[d and] enduring” (cited in Camic 1990:316 from Cooley [1909] 1962:313); and usages that are “endowed with a coercive and inhibitive force” (cited in Camic 1990:316 from Sumner [1906] 1979:67). Institutions were framed as organized, persisting, and “behavioral,” not only ideational or normative.

According to Camic (1990:317), MacIver “codified” these themes when he argued that social institutions are one of two “great classes [of] social fact,” the other being social relations. For MacIver, institutions were “forms of social activity” or “the established forms or conditions [e.g., rules] of procedure characteristic of group activity” (cited in Camic 1990:317 from MacIver 1928:7, 1936:16–17). MacIver said institutions entail activities, or practices, that inform and enable members about how to get things done. They also maintain order by encouraging or requiring some activities and discouraging or forbidding others. He noted also that institutions persist over time, and thus he made their time element explicit. These principles were “virtually axiomatic among sociologists” by the 1930s, Camic (1990:317) says.

Talcott Parsons, whose influence on U.S. sociology in the twentieth century was extensive, rejected MacIver’s inclusion of the “forms of social activity/ procedures” so as to “preserve” the concept for sociology. Parsons was responding to claims by psychologists and behaviorists that “the habitual behavior of individuals should be viewed as units of biology, chemistry and other natural sciences,” not sociology (Camic 1990:316–17).6 Parsons responded by differentiating the notion of institutions into “uniform modes of behavior and forms of relationship” and “the idea of sanction” and, then, relinquishing the former to the behaviorists and retaining the latter for sociology. The former became equated with an “objective” approach to institutions (e.g., MacIver’s and his predecessors) and the latter with the “subjective” approach favored by Parsons. This step equated an institution with “the normative rules that underlie it,” setting “aside modes of behavior and forms of relationship” and “leaving behind much of what otherwise might have been regarded as these thinkers’ institutional analysis” (Camic 1990:317–18).7

[When Parsons] tries to counteract behaviorist reductionism by grounding institutions in common ultimate values, much of what sociologists and institutionalists of his time were pointing to with the term institution — diverse forms of social activity, specific prescribed usages, historically-changing economic, political, religious, familial and other practices, and their variable sources simply disappears. (Camic 1990:318)

Parsons’s actions fostered confusion where many of us labour (and fail) to equate institutions with “norms, beliefs, and ideas” and organizations with “practices and structure,” as if institutions lack practices and structure and organizations lack norms, beliefs, and ideas. While his conception allowed societal members to have agency in shaping the norms, beliefs, and values that constrain them, Parsons’s rejection of behavior and practices removed the dynamics by which change is produced. As a result, social institutions suggested a society where conformity and stasis are usual and conflict and change are unusual.

2. Later Twentieth-Century Conceptions

Not everyone agreed with Parsons. Nisbet (1953: 87–90) critiqued sociologists’ definition of institutions that ignored people; represented each institution as separate from all others; suggested an ahistorical picture; depicted institutions as free of conflict, inconsistencies, and change; and failed to address their origins. He called for more attention to human agency in creating and changing social institutions and for scholars to frame institutions as rife with conflict, incoherence, and change. Conflict within as well as between institutions is pervasive, and these dynamics should be explored (see also Roscigno 2000 on interinstitutional dynamics).

Berger and Luckmann (1966) challenged prevailing definitions by reinstating the “objective” aspects of institutional phenomena and linking their “subjective” and “objective” qualities. Rejecting the implications of a functionalist view of institutions as “positive” or necessary, they made the issues of power and domination key to institutional dynamics. Focusing on institutions’ coercive powers, Berger and Luckmann (1966) highlighted legitimation dynamics wherein powerful elites claim and justify to a broader social audience the rightness and necessity of institutional arrangements that work to their benefit. Here institutions are actively constructed, as human products, claiming that a repeated action “frequently becomes cast into a pattern, which can then be reproduced with an economy of effort and which, ipso facto, is apprehended by its performer as that pattern. Habitualization [the repetition of a pattern] . . . implies that the action in question may be performed again in the future in the same manner and with the same economical effect” (Berger and Luckmann 1966:52–53). Although not inevitably, processes of habitualization often develop into institutions.

Berger and Luckmann emphasized both historicity and control: “Institutions always have a history, of which they are the products” (1966:54); they do not develop spontaneously in response to “societal needs or goals.” Institutions have a “controlling characteristic.” “To say that a segment of human activity has been institutionalized is . . . to say that this segment of human activity has been subsumed under social control” (Berger and Luckmann 1966: 55). Their emphasis on the controlling effect(s) of institutions resonates with work from decades earlier by Cooley, Hamilton, and MacIver. Berger and Luckmann also insisted that institutions span an extensive amount of time, a point with which contemporary theorists agree.

Giddens (1984) extends this prior work by making recursive human practices key to the constitution of institutions. Social institutions are recursive human practices with the greatest time and space distanciation. Recursive human practices are not repeated identically each time; through recursive human practices, group members constitute and reconstitute social institutions. In Giddens’s view, “Institutions are . . . the more enduring features of social life . . . , the human practices that last longest and extend farthest in geographic space — the most “temporally long-established and spatially widespread” (1984: 23, 301). Only recursive practices that last a long time and extend far in space should be deemed institutions.

Time’s importance to a conception of institutions is reflected in Giddens’s (1984) distinction between individuals’ repetitive daily routines and life span experiences and the long “duree” (endurance) of institutional time.

The events and routines of daily life do not have a one-way flow to them; they are recursive or repetitive. . . . [T]he routines . . . are formed in terms of the interaction of passing (but continually returning) days and seasons. Daily life has a duration, a flow, but it does not lead anywhere; the very adjective “day-to-day” and its synonyms indicate that time here is constituted only in repetition. The life of the individual, by contrast, is not only finite but irreversible, ‘being towards death.’ Time in this case is the time of the body . . . and the life cycle is really a concept that belongs to the succession of generations and thus to a third dimension of temporality. . . . This [third dimension] is the ‘supra-individual’ duree of the long-term existence of institutions, the long duree of institutional time. (35)

As already noted, Giddens objects to defining social institutions substantively (1984:34), that is, as family, education, religion, and so on — and he opposes a strict micro–macro distinction between individuals and institutions, saying it suggests that institutions are detached from or only external to people. Rather, institutions are “internalized” by the human actors who constitute them. While institutions are simultaneously constraining and enabling, Giddens places more emphasis than do many others on internalization and enablement.

Unique among authors reviewed so far, Giddens instates the body and embodiment in institutional dynamics, affirming the significance of bodies that materially exist and consequentially do things, a theme developed further by Connell (1987) in relation to gender and sexuality (see also Acker 1990; Lorber 1999). People have bodies that do things via physical and communicative action and, in acting, constitute themselves and society, with structuration referring to the simultaneous constitution of “agents” and “structures” (Giddens 1984:25–26). Giddens acknowledges people as situated actors who actively constitute and reconstitute social institutions and, in so doing, suggests where institutions “come from” and how they are maintained, resisted, and changed.

Toward (Re)Defining Social Institution

Building on particularly above review, I identify the following features of social institutions:

1. Institutions are profoundly social; they are characteristic of groups. Institutions are constituted by collectivities of people who associate with each other extensively and, through interaction, develop recursive practices and associated meanings.

2. Institutions endure/persist across extensive time and geographic space. Social institutions have a history that can be studied. In accord with Giddens (1984), only phenomena with extensive time and space distanciation are usefully viewed as social institutions.

3. Institutions entail distinct social practices that recur (Giddens 1984), recycle (Connell 1987), or are repeated (over time) by group members. Through acting or doing, individually and collectively, group members constitute institutions. Distinctive practices differentiate institutions from each other. Barnes (2001) equates social institutions with practices, and Tuomela (2003:123) views social institutions as “norm-governed social practices” (also Searle 1969, 1995). Even Bellah and colleagues (1991:40), who generally favor a “subjective”/Parsonian view, acknowledge that institutions are “patterns of social activity that give shape to collective and individual experience” [emphasis in this and the preceding quote is mine]. Practices that recur over extensive spans of time and geography are defining features of social institutions and the means by which they are constituted (Connell 1987; Giddens 1984).

4. Institutions both constrain and facilitate behavior/actions by societal/group members. Institutions constrain group members by forbidding some alternatives and choices of actions and empower them by making some alternatives and choices of action possible and preferable (March & Olsen 1989).

5. Institutions have social positions and relations that are characterized by particular expectations, rules/norms, and procedures. An institution entails a set of social positions that are interrelated, “make sense,” and are enacted relative to each other. For instance, religion has lay member, pastor/priest/rabbi/ayatollah, acolyte, elder, deacon, worshiper, and so on. The behavior of incumbents to these positions is shaped by widely shared cultural rules or norms.

6. Institutions are constituted and reconstituted by embodied agents. Institutions exist because embodied agents, societal members with material bodies, enact practices to constitute them. Institutions persist because embodied agents continually constitute them, although in varying ways (see points 9 and 10). The material body is a critical element in the social relations and dynamics of institutions and, according to Connell (1987), its influence must be studied without invoking biological reductionism or other essentialist interpretations (also Giddens 1984; Lorber 1994).

7. Institutions are internalized by group members as identities and selves and they are displayed as personalities. Institutions are not only external to individuals. Members’ experiences within — with and in — institutions become incorporated into their identities and selves as members identify with their positions, the practices they enact, and the positions they occupy. Through this dynamic, institutional phenomena acquire personal meaning and significance (Connell 1987).

8. Institutions have a legitimating ideology that proclaims the rightness and necessity of their arrangements, practices, and social relations. Legitimating ideology that justifies institutional practices and social relations is created by elites who benefit from the arrangements and practices they valorize (Berger & Luckmann 1966). Gender or caste ideology is widely known and generally believed.

9. Institutions are inconsistent, contradictory, and rife with conflict. Despite their persistence, institutions are not highly coherent or integrated. They entail many diverse practices, some of which conflict with others. Due to inconsistencies and internal conflicts, struggles among group members over particular practices are common, not rare.

10. Institutions continuously change. Related to the prior criterion, and paradoxically contrary to the second criterion about endurance, institutional relations and practices are in flux. One reason is that present practices modify past practices (Connell 1987) and produce slightly, sometimes vastly, altered practices. Also, the interdependence of institutions means that changes in one institution “unsettle” conditions and practices in other institutions, causing disruption (Nisbet 1953; Roscigno 2000). Finally, over time, old institutions die out and new ones are constituted. For example, slavery is mostly relegated to the past, whereas the mass media institution is a recent creation. Attention to dynamics not only within but between institutions, for example, between gender and the media, is required by this criterion (see Grindstaff 2002).

11. Institutions are organized in accord with and permeated by power. Institutional positions and practices allocate privilege and advantages to incumbents of some social positions and subordination and disadvantages to others. Power differentials are manifest in the recursive practices that orient, constrain, and facilitate members’ behavior. Social positions that are highly valorized provide incumbents with power over incumbents of less valorized positions. I agree with Stephen Lukes (1974) that organization per se creates power (cf. Balzer 2003). Wherever social practices and relations are “organized,” as they are in institutions, power differences and dynamics are at play.

12. Institutions and individuals mutually constitute each other; they are not separable into macro and micro phenomena. Giddens (1984; also Nisbet 1953) rejects the claim that institutions are macro and individuals are micro, arguing that this distinction distorts their mutual constitution and implies that institutions are only external rather than also internalized (criterion 7). It also implies that they are not susceptible to human agency (criterion 3). This criterion rejects the premises that institutions are big and individuals are small and that institutions are separate from individuals.

13. Institutions are interdependent on each other. In other words, no institution is totally separate from others; each links to others, often extensively (Roscigno 2000). For example, gender and sexuality are intertwined — as are gender and family, gender and work/the economy, gender and religion — but so are family and the polity/ state, family and the economy, economy and the polity/state, and education and the polity/state, and so forth (Acker 1992). Assuming that any institution is separate from others will produce flawed understanding (Nisbet 1953). Interinstitutional influences are not only pervasive, they change over time. Thus, changes in gender in accord with second-wave feminism have “unsettled” the family institution, the military, and religion, among other institutions, by challenging the legitimacy of girls’ and women’s subordination (Connell 1987; Gerson 2002; Katzenstein 1998). In other historical eras, these institutions more extensively “unsettled” gender.