"The Network Community"
An Introduction to
Networks in the Global Village

Barry Wellman

Things Ain't Wot They Used To Be - And They Never Were!

Community has never been lost. Yet since the Industrial Revolution, most people have believed that large-scale technological and social changes had destroyed community in the developed world and were well on their way to killing it in developing countries. Policymakers and pundits echoed and reinforced this belief, and until a generation ago, most social scientists agreed with them.

Wherever they have looked, researchers have found thriving communities. This is so well documented that there is no longer any scholarly need to demonstrate that community ties exist everywhere, although the alarmed public, politicians and pundits need to be constantly reassured and re-educated. But there is a pressing need to understand what kinds of community flourish, what communities do — and do not do — for people, and how communities operate in different social systems.

Back

The Community Question

1. How does the structure of large-scale social systems affect the composition, structure and contents of interpersonal relations within them? For example, do different countries, ethnic groups, gender relations or socioeconomic strata affect the nature of community? The authors in this book focus on this aspect of the Community Question in relation to community networks. But much the same issues pertain to the study of other sorts of interpersonal networks: kinship groups, households and work groups.
2. How does the nature of community networks affect the nature of the large-scale social systems in which they are embedded? This is the reciprocal part of the Community Question. For example, do resources flow freely from one part of society to another? Or are communities isolated from each other in racial or class enclaves (sometimes called "ghettoes")? Speaking to this issue, Mark Granovetter’s "strength of weak ties" argument (1972, 1983, 1995) has shown how weak community ties integrate social systems by linking heterogeneous groups of people; Edward Laumann’s Bonds of Pluralism (1973) has shown how community networks structure relations among ethnic and religious groups in Chicago; while Alexis Ferrand, Lise Mounier and Alain Degenne’s chapter in this book shows how community networks integrate certain socioeconomic strata in France but decouple others.

Thus, the Community Question stands at a crucial nexus between societal and interpersonal social systems. It juxtaposes the problem of the structural integration of a social system and the interpersonal means by which the members of this social system have access to scarce resources. The social capital vested in ties provide interpersonal resources for people to use to deal with daily life, seize opportunities and reduce uncertainties. (Kadushin 1981; Bourdieu 1984; Coleman 1990; Wellman and Wortley 1990; Flap 1995; Putnam 1995; Burt 1997; Lin 1997; Ruan, et al. 1997; Schweizer, Schnegg and Berzborn 1998).

Back

Looking For Community

It is likely that pundits have worried about the impact of social change on community ever since people ventured beyond their caves. The Community Question clearly preoccupied biblical prophets from the 11^th century BCE onward (see Samuel 8), concerned then (as now) that the establishment of an Israeli state would lead to communal disintegration (Leach 1966; Buccellati 1967; Zeidman 1985). As the prophet Jeremiah warned the rapidly-modernizing Israelites in the 6^th century BCE in "all their wickedness" (1:16):

Take ye heed every one of his neighbor, and on any brother place ye no reliance; for every brother will surely supplant, and every neighbor will go about as a talebearer (8:3).¹

Two thousand years later, the Community Question continued to be a major issue to Renaissance intellectuals. Their concerns ranged from Machiavelli's (1532) celebration of the liberation of communal patterns to Hobbes' (1651) fears that the absence of social structures would result in the interpersonal war of all against all. A bit later, the Community Question was a key preoccupation of such 18^th century British philosophers as John Locke and David Hume (see also Wills 1978) as they sought to deduce the social basis of larger-scale societies from their understanding of primordial communal relations. Their student, Thomas Jefferson, gave the question an anti-urban cast -- communal bonds are not viable in industrial, commercial cities -- when he asserted:

The mobs of great cities add just so much to the support of pure government, as sores do to the strength of the human body (1784, p. 86).

In the two centuries since then, many commentators have wrestled to understand the ways in which large-scale social changes associated with the Industrial Revolution may have affected the composition, structure and operations of communities. Their analyses have reflected the ambivalence with which nineteenth-century pundits faced the impacts of industrialization, bureaucratization, capitalism, imperialism and technological developments on interpersonal relations. Where religion, locality and kinship group had some integrative claims on interpersonal relations, the shift to mobile, market societies now had the potential to disconnect individuals from the strengths and constraints of traditional societies (Marx 1964; Smith 1979; White and White 1962; Williams 1973).

This was not only an isolated, nostalgic lament for the supposed loss of the mythical pastoral past where happy villagers knew their place. Many commentators shared Tönnies’ fears about the supposed contemporary loss of community although they offered different reasons for why it was happening, such as industrialization, urbanization, bureaucratization, capitalism, socialism or technological change. Thus the loss of community was a centerpiece of Karl Marx’s (1852) and Friedrich Engels' (1885) communist analyses, asserting that industrial capitalism had created new types of interpersonal exploitation that drove people apart. Capitalism had alienated workers not only from their work but from each other. By contrast, although sociologist Max Weber (1946, 1958) extolled modern rationality, he also feared that bureaucratization and urbanization were weakening communal bonds and traditional authority. Sociologist Émile Durkheim (1897) feared that the loss of solidarity had weakened communal support and fostered social pathology. A generation later, sociologist Georg Simmel celebrated urban liberation but also worried that the new individualism would lead to superficial relationships (1922).

Tönnies’ vision was part of a particularly European debate about the transformation of societies — aristocrats, intellectuals and parvenus coming to terms with the transformation of once-ordered, hierarchal societies of peasants and landowners, workers and merchants. Despite different social conditions, social scientists in the new North American world adapted Tönnies’ concerns, debating whether modern times have occasioned the loss of community in developed Western societies (e.g., Berger 1960; Gans 1962; Gans 1967; Grant 1969; Nisbet 1962; Parsons 1943; Slater 1970; Stein 1960).  Robert Redfield’s (1947) folk-urban continuum was especially influential, asserting that the possibilities for community varied linearly between highly-communal rural villages through towns to cities lacking community.

In confronting their own society, many Americans decried the loss of solidary communities of family, kin and neighbors, bound by custom and tradition. Their analyses reflected the continuing American tension between individualism and communalism originally put forward by the influential historian, Frederick Jackson Turner. Focusing on the populace’s march westward to settle the supposedly-empty frontier, Turner argued (1893) that constant movement left little room for community to develop. He argued that what little there was of community in the West consisted of transient groups of settlers helping each other, with instrumental aid overshadowing emotional support, companionship, or a sense of communal belonging. Even the cities were filled with migrants: floating proletarians who were constantly on the move, seeking work that would push them up the ladder (Thernstrom 1964, 1973 Chudacoff 1972; Katz, Doucet, and Stern 1982). The successful rural settlers and urban migrants embodied the Turnerian spirit of individualism and practicality. They had avoided being trapped in traditional community bonds (Starr 1985, 1990).²

Back

What Could Have Caused Changes in Community?

Contemporary analysts have debated the causes of changes in community almost as much as they have debated whether community has, in fact, changed and what the nature of these changes might be. This is because the strong association between the appearance of industrialization, bureaucratization, urbanization, capitalism, socialism and new transportation and communication technologies has made it difficult to tease out the ultimate cause, if any (see the discussion in Abu-Lughod 1991). Various analysts have pointed to:

1. The increased scale of the nation-state's activities, with a concomitant low level of local community autonomy and solidary (e.g., Tilly 1973, 1975; 1984).
2. The increased scale of the global world-system, with footloose financial capital creating uncertainty in local communities and encouraging workers to uproot themselves and migrate to places with better employment possibilities (e.g., Burawoy 1976).
3. The development of narrowly instrumental bureaucratic institutions for production and reproduction which may have lead to the transformation of former, broadly-supportive community ties to contractually-defined, narrow relations of exchange (Castells 1976; Howard 1988).
4. The large size of cities, with the consequent population and organizational potential for diverse interest groups (Wirth 1938; Fischer 1984).
5. The high social density of interaction among the segments of the population (even where the spatial density is low), with the resulting complexities of organizational and ecological sorting (Abu-Lughod 1969; Freedman 1975; Gillis and Hagan 1982).
6. The diversity of persons with whom urbanites can come into contact under conditions of heightened mobility (Jacobs 1961);
7. The proliferation of widespread networks of cheap and efficient transportation and communication facilities that have allowed contact to be maintained with greater ease and over longer distances: in transportation, from railroads through superhighways and planes; in communication, from overnight mail service to direct long-distance telephone dialing to the Internet and the World Wide Web (Meier 1962; Wellman and Gulia's net surfers chapter).³ The increased velocity of transactions has fostered interactional density. The large-scale metropolis is accessible and links to diverse social networks can be maintained more readily.

Ambivalence about the consequences of large-scale changes has continued through the twentieth century, with scholars and pundits asking if things have, in fact, fallen apart. Unfortunately, the fundamental concerns of the Community Question have become confounded in many analyses with narrower issues:

1. Some researchers continue the habit of looking for community ties only in local areas, reflecting community sociology's origins in studying neighborhoods (Stein 1960). Seeing community in concrete, bounded neighborhoods is easier than seeing community in far-flung networks whose ties spread almost invisibly through the ether.
2. A general preoccupation with identifying the conditions under which solidary sentiments can be maintained. In so doing, they reflect a continuing worry about whether normative integration and consensus persist. People worry if they can get help from strangers or even from the members of their community; they worry that they will be alone in confronting crime, disease, joblessness or natural catastrophes (Etzioni 1991; Nisbet 1962). The most recent manifestation of this concern has been Robert Putnam’s raising the alarm that Americans are now "bowling alone" (1995); they are much less involved in voluntary organized groups, be they bowling leagues, churches, the Lions Club or unions. In his Tocqueville-like analysis (e.g., 1834), Putnam fears that this lessened organizational participation means less civic involvement in promoting good government and less "social trust" in governments and fellow citizens. He wonders if amorphous community networks can substitute for participation in more bounded and concrete organizations.

Concerns about the persistence of community are frequently projected onto the future in Manichean debates about whether community will die or flourish in cyberspace as Wellman and Gulia’s chapter documents. Science-fiction novels have echoed fears of the loss of community, providing scenarios ranging from alienation in densely-packed (Ballard 1975), hyper-capitalistic (Brunner 1968) mass societies, to post-atomic holocaust returns to tribal solidarities (Atwood 1985; Lessing 1974). However, a more optimistic genre has foretold wired people in wired cities moving easily between interest groups (Brunner 1975; Delaney 1976; Gibson 1986; Stephenson 1992). Similar to the novels, the predominant depiction of the future in films has been of small, scattered and impoverished tribal bands trying to survive in a desolate land filled with marauders, a genre popularized in contemporary times by George Miller’s (and Mel Gibson’s) Mad Max (1979) and The Road Warrior (1981) and by James Cameron’s (and Arnold Schwarzenegger’s) The Terminator (1984). An alternative futuristic vision has been equally as bleak, in a cinematic fashion set by Ridley Scott’s (and Harrison Ford’s) Blade Runner (1982): the squalid, overpopulated, East Asian-influenced landscape of alienated urban masses in a society visually dominated by huge organizations and their equally huge neon advertisements.

With the growth of the Internet and the World-Wide Web, what had once been science-fiction has become a staple for apocalyptic speculation, although with much less analysis. As Wellman and Gulia’s "Net Surfers" chapter recounts, those on either side of this debate assert that the Internet either will create wonderful new forms of community or will destroy community altogether. This latter side of the debate is Tönnies nouveau, warning that meaningful contact will wither without the full bandwidth provided by in-person, in-the flesh contact. This debate has been unscholarly, presentist and parochial. Consistent with the present-oriented ethos of computer users, pundits write as if people and scholars had never worried about community before the Internet arose. Too many analysts treat the Internet as an isolated phenomenon without taking into account how online interactions fit with other aspects of people’s lives.

Back

Finding Community

Rediscovering Traditional Community with Flowers in Its Hair

One intellectual generation ago the watchwords of community sociologists were documentation and description. The profession was preoccupied with proving that community persisted — dare they say "flourished"? Scholars of the First (Western, developed, non-socialist) world wanted to show that supportive communal bonds remained even in allegedly pernicious habitats: inner-city slums (Gans 1962; Liebow 1967; Whyte 1943; Young and Willmott 1957) and middle-class suburbs (Bell 1968; Clark 1966; Gans 1967). Scholars of the Third ("underdeveloped") world battled fears that the migrants flooding into industrializing cities would form communally-disconnected, politically-dangerous hordes.⁴  Although the argument that capitalism had shaped urban communities called for comparative approaches (e.g., Castells 1972; see also Fischer 1978), few scholars tackled the Community Question in the Second (socialist) world of Eastern Europe and China. To have done so, would have been contradictory to the anti-Tocquevillean (1835) communist ethos that saw each person relating individually to the state, without intermediary structures. Hence community network studies (such as Sik and Wellman’s chapter in this book) have only developed in post-communist times.⁵

With hindsight, postwar fears about the "loss of community" came in part from the same sources as some Americans fearing evil creatures from outer space and U.S. Senator Joe McCarthy's searching for Reds under the bed. The fearful saw alien forces believed that the Frankensteinian "machine in the garden" (Marx 1964) had run amok and destroyed traditional communities. Beneath the jingoistic celebration of small-town virtues lurked the fear that people were inherently evil: ready to rob, rape, pillage and turn atheistically communist when communal bonds were loosened.

The developing body of research has shown that communities while may have changed in response to the pressures, opportunities and constraints of large-scale forces, they have not withered away. They buffer households against large-scale forces, provide mutual aid, and serve as secure bases to engage with the outside world (see reviews in Choldin 1985; Fischer 1976; Gordon 1978; Keller 1968; Smith 1979; Warren 1978). They provide Kirkian emotional aid, Spockian information, McCoyesque companionship, and Scottynian instrumental aid: the four archetypes of the original Star Trek television show (Whitfield and Roddenberry 1970). For example, Espinoza’s chapter shows that informal community ties are the keys to daily survival in the impoverished barrios of Santiago, Chile. They provide food, shelter, short-term loans, job leads, and help in dealing with organizations. In this situation, neighbors (who are often kin) provide most everyday support. Yet such neighbors are poor themselves. To get sizable amounts of money or access to good jobs, the residents must rely on their weaker ties to wealthier, better-situated relatives who live outside the barrios. The situation fits well with Granovetter’s (1973) and Wellman and Leighton’s argument that weak, ramifying ties are well-suited for obtaining access to new resources while strong, solidary ties are well-suited for mobilizing and conserving existing resources.

This scholarly rediscovery of community resonated strongly with the political developments of the 1960s. The civil rights movement encouraged more positive evaluations of urban black neighborhoods (e.g., Stack 1974) and, by extension, of lumpenproletariat life everywhere. The neo-Rousseauian student movement preached the inherent goodness of human kind. Students and anthropologists boarded new low-cost charter flights to spend $5 a day discovering that Europe and the Third World were full of enjoyable people in interesting villages and cities. Planners turned away from urban renewal toward the preservation of dense, noisy downtown neighborhoods (as expressed most vividly in Jane Jacobs' 1961 anthem). Instead of bulldozing neighborhoods to encourage suburban growth and metropolitan expressways, planners and politicians started banning large-scale inner-city housing projects and terminating expressways outside city cores. Renovation and gentrification became the buzzwords of the 1970s. There were hard-won battles, fought with demonstrations, sit-ins, court decrees, elections and scholarly articles. Despite much migration to the suburbs, the centers of such cities as Boston, New York, San Francisco and Toronto remained well-populated.

This transformation in thinking became the academic orthodoxy of the late 1960s and the 1970s. Scholars, planners and some politicians and members of the public no longer thought of cities as evil, permeated with Original Sin. Their Jacobsean cum Rousseauesque celebrations of community had the lingering aroma of the 1960s, seeing urbanites as permeated with Original Good and happily maintaining mutually supportive ties. The rest of the populace was slower to catch on: many policymakers, the media and the public at large continued to fear the urban, yearn for the pastoral, and settle for suburbia.

Back

How Green Were the Valleys?

In saying that communities are not as local as they used to be, analysts must avoid committing the pastoralist fallacy of thinking that contemporary cities and suburbs are inferior to the villages or pestilent cities of yesteryear, with their pestilence, crime and insecurity. At the same time that sociologists were discovering the existence of contemporary communities, historical analysts started using similar research methods to study preindustrial villages, towns and cities. Until their work became known, analysts had contrasted the disorderly urban present with the pastoral ideal of bucolic, solidary villages (Poggioli 1975). They assumed that such communities were socially cohesive and stable, with little movement in or out. Yet the supposed communalism of the preindustrial world has turned out to be an artifact of how earlier commentators thought about it. Pre-industrial communities may never have been as locally-bounded as tradition has maintained. Whenever scholars have looked for non-local ties, they have found far-ranging networks. For example, radioactive analyses of obsidian have found Neolithic spear points and choppers more than one thousand miles from their origin (Dixon Cann and Renfrew 1968).  By looking for community in localities and not in networks, analysts had focused on local phenomena and stability rather than on long distances and mobility. For example, Emmanuel LeRoy Ladurie's (1975) rich account of medieval village life in southern France reveals a good deal of geographical mobility as early as the 1300s. To trace networks of Albigensian heretics, Catholic investigators asked all residents of the village of Montaillou to report who their friends were, who had influence, and how they spent their days. They used this information to build up detailed accounts of the village community. These accounts reveal that many villagers travelled widely. Some were shepherds following their flocks over the Pyrenees, some were itinerant soldiers, while others travelled south to the Spanish coast or west along the Mediterranean to northern Italy. The people of Montaillou had frequent contact with other villages, and passing travellers often gave them news of the outside world. With such contact came new ideas, intermarriage, and new alliances.

These studies suggest that the average preindustrial household was quite small, For example, at the turn of the 19^th century, the typical adult inhabitant of the Latvian village of Pinkenhof had only three kin and five friends/neighbors/coworkers in their personal communities (Wetherell, Plakans and Wellman 1994). Contrary to contemporary pastoralist myth of immutable villages, many families were socially and spatially mobile. They often worked in the city when they were young adults, but retained ties with their rural villages. Artisans and soldiers were frequently on the road. Women married and moved, geographically and socially. Servants' ties to their distant families concurrently linked their masters' families to the servants' rural homes. And, as all readers of Jane Austen know, these complex connections linked far-flung networks of community ties. For example, in Sense and Sensibility (1811), the Misses Dashwood made long journeys between their original Sussex home, their new Devonshire house, and the London social milieu. Even while residing at "Barton Cottage," Devonshire, they — and their network members — were forever jaunting long distances to visit each other, apparently oblivious to the many other homes that they rode past. They maintained far-flung kinship and friendship networks throughout southern and central England while seemingly not being connected to most of their neighbors.

Back

Neighbourhood or Community?

Despite these cautionary tales from the past, the fundamentally structural Community Question has often been a search for local solidarity rather than a search for supportive ties, wherever located and however solidary. As a result of the continuing sociological and public fixation on communities as solidary neighborhoods, community studies have usually been neighborhood studies, be they the "symbiotic" communities of Robert Park’s treatises (1925) or the empirical studies of street life by Whyte (1943), Liebow (1967) and Anderson (1990). Definitions of community have usually included three ingredients:

a. Interpersonal networks that provide sociability, social support and social capital to their members;

b. Residence in a common locality, such as a village or neighborhood;

c. Solidary sentiments and activities (see Hillery 1955).

It is principally the emphasis on common locality, and to a lesser extent the emphasis on solidarity, that has encouraged the identification of "community" with "neighborhood". There are several reasons why the concept of "neighborhood" has been almost synonymous with the concept of "community":

1. Community researchers have to start somewhere. The neighborhood is an easily identifiable research site, while the street corner is an obvious, visible and accessible place for observing interpersonal interactions. Indeed, Lee and Campbell's and Espinoza's chapters in this book start by drawing their samples from one or more neighborhoods.
2. Many urban scholars have seen the neighborhood as the microcosm of the city, and the city as an aggregate of neighborhoods. They have emphasized the local rather than the cosmopolitan (Merton 1957) in a building block approach to analysis that has given scant attention to the interpersonal and interorganizational relationships that form large-scale social structures.
3. Administrative officials have imposed their own definitions of neighborhood boundaries upon urban maps in attempts to create bureaucratic units. Politicians are even more neighborhood-oriented, in part because they usually have to be elected from local constituencies. Spatial areas, labeled and treated as coherent neighborhoods, have come to be regarded as natural phenomena - by politicians, the public, researchers and even by the people who live there. In Toronto, downtown street signs proclaim "Little Italy" in a neighborhood that has, ironically, become filled with Portuguese-Canadians after the former Italian-Canadian residents moved to the suburbs. Citizens fight to keep unwanted garbage dumps outside of their municipality, even if they live far from its proposed location (Michelson 1997). In Chicago, politicians, administrators and bank officials are forever coping with urban problems by announcing neighborhood development programs (Taub, et al. 1977).
4. Urban sociology's particular concern with spatial distributions of social phenomena (e.g., Schwirian and Mesch 1993) has tended to be translated into local area concerns. Census data, originally designed to enumerate populations in electoral districts, provide large quantities of demographic and social data organized in (too)-convenient, territorially-defined census tracts and enumeration areas. The easy availability of these data has encouraged researchers to think in terms of spatial patterns. Territory has come to be seen as the inherently most important organizing factor in urban social relations rather than as just one potentially important factor.
5. Many sociologists have been preoccupied with the conditions under which solidary sentiments can be maintained in cities and societies. Their preoccupation reflects a persistent overarching public, political and scholarly concern with achieving normative consensus and social solidarity. The neighborhood has been widely seen and studied as an apparently obvious container of normative solidarity in "the community".

This concentration on the neighborhood has had a strong impact on definitions of, research into, and theorizing about community. Neighborhood studies have produced many finely wrought depictions of urban life, and they have given us powerful ideas about how interpersonal relations operate in a variety of social contexts (see review in Fischer 1976). Analyses have taken mappings of local area boundaries as their staring points and then looked into the extent of communal interaction and sentiment within these boundaries. They have thus assumed, a priori, that a significant portion of a person’s interpersonal ties are organized by locality. Such a territorial perspective, searching for answers to the Community Question only within bounded population aggregates, has been especially sensitive to the evaluation of community solidarity in terms of shared values and social integration. Consequently, when observers cannot find much solidary local behavior and sentiments, they have too-often concluded that "community" has disappeared.

But does the concept of "neighborhood" equal the concept of "community"? Are the two terms synonymous? The contemporary milieu of frequent residential mobility, spatially-dispersed relationships and activities, and the movement of interactions from public spaces to private homes have all limited the amount of observable interactions in neighborhoods. This does not mean that community has been lost but that it is much less likely now to be locally based and locally observed.

The paramount concerns of sociologists are social structures and social processes — and not spatial groupings. Concerns about the spatial location of social structures and processes must necessarily occupy secondary positions. To sociologists, unlike geographers, spatial distributions are not inherently important variables. They assume importance only as they affect such social structural questions as the formation, composition and structure of interpersonal networks; the flow of resources through such networks; and the interplay of such community networks with the division of labor and the organization of power within larger-scale social systems.

Back

The Network Analytic Approach to Studying Community

The authors in this book examine the Community Question from a network analytic perspective. Social network analysis provides a useful way to study community without presuming that it is confined to a local area. The essence of social network analysis is its focus on social relations and social structures - wherever they may be located and whoever they may be with. Social network analysis does not assume that the world is always composed of normatively-guided individuals aggregated into bounded groups or areas. Rather, it starts with a set of network members (sometimes called nodes) and a set of ties that connect some or all nodes (Wasserman and Faust 1993).⁷ Social network analysis conceives of social structure as the patterned organization of these network members and their relationships (Wellman 1988b). The utility of the network approach is that it does not take as its starting point putative neighborhood solidarities nor does it seek primarily to find and explain the persistence (or absence) of solidary sentiments. Thus the network approach attempts to avoid individual-level research perspectives, with their inherently social psychological explanatory bases that see internalized attitudes as determining community relations.

The social network approach provides ways for analysts to think about social relationships that are neither groups nor isolated duets. Instead of the either/or distinction between group membership and social isolation characteristic of those fearing the alleged loss of community, network analysts can study a more diversified set of structural phenomena, such as:

• The density and clustering of a network;
• How tightly it is bounded;
• Whether it is variegated or constricted in its size and heterogeneity;
• How narrowly specialized or broadly multiplex are its relationships;
• How indirect connections and structural positions affect behavior.

Although all studies have to start somewhere with some populations, most social network analyses do not treat officially-defined group or neighborhood boundaries as truly social boundaries, be they departments in organizations or neighborhoods in cities. Instead they trace the social relationships of the persons they are studying, wherever these relationships go and whomever they are with. Only then do network analysts look to see if such relationships cross officially defined boundaries. In this way, formal boundaries become important analytic variables rather than a priori analytic constraints.

The social network approach does not preclude finding that communities are urban villages where everyone knows each other and provides the abundant, broadly-based support, that Tönnies thought only to be a nostalgic relic of vanishing villages. Nor does it preclude finding that organizations really function as Weberian hierarchical bureaucracies. But the social network approach allows the discovery of other forms of community — perhaps sparsely-knit and spatially-dispersed — and other forms of organization — perhaps loosely-coupled or virtual.

Back

Social Networks of Community

Social network analysis has freed the community question from its traditional preoccupation with solidarity and neighborhood.⁸ It provides a new way to study community that is based on the community relationships that people actually have rather than on the places where they live or the solidary sentiments they have. It offers three advantages:

1. It avoids the assumption that people necessarily interact in neighborhoods, kinship groups or other bounded solidarities. This facilitates the study of a wide range of relationships, wherever located and however structured. Look in this book at how Ohtani (in Japan), Wellman’s research group (in Toronto) and especially, Salaff, Fong and Wong (in Hong Kong) find that residential proximity is at most, only one dimension of community (see also Fischer 1982b; Nozawa 1997). Yet, as Campbell and Lee’s and Espinoza’s chapters also show, the network approach also supports the analysis of those community ties that do remain in neighborhoods. Thus the social network approach is not anti-neighborhood — the traditional stuff of community studies — but allows neighborhood ties to be discovered without an a priori assumption of their importance.
2. Its ability to study linkages at all scales, ranging from interpersonal relations to world systems, facilitates the analytic linkage of everyday lives with large-scale social change. For example, Espinoza’s chapter in this book relates massive sociopolitical upheavals in Chile to the kinds of supportive relations that poor people must maintain, Bian and Sik and Wellman trace community relations during the transition from communism to post-communism, and Ohtani shows how Japanese personal communities have come to resemble North American communities. Moreover, Ferrand, Mounier and Degenne show how interpersonal linkages help to structure linkages and cleavages between French social classes.
3. Its established body of techniques, both qualitative and quantitative, for discovering, describing and analyzing the presence, composition, structure and operations of interpersonal networks (Scott 1991; Wasserman and Faust 1993; Wellman 1992a).

By using the social network approach, analysts discovered that community had not disappeared. Instead, community had moved out of its traditional neighborhood base as the constraints of space weakened. Except in situations of ethnic or racial segregation (e.g., Lee and Campbell’s chapter), contemporary Western communities are rarely tightly-bounded, densely-knit groups of broadly-based ties. They are usually loosely-bounded, sparsely-knit, ramifying networks of specialized ties. Therefore, analysts should be able to find community wherever it exists: in neighborhoods, in family solidarities, or in networks that reach farther out and include many friends and acquaintances (Oliver 1988; Wellman 1979; Wellman and Leighton 1979; Fischer 1982).

Back

Community Networks as Personal Communities

There are two ways to look at community networks (or at any social networks, for that matter): as whole networks or as personal communities. Many analysts view social networks much as aliens might view the earth’s people: as outside observers hovering above, observing the relationships linking all members of the population. This alien’s-eye (or Copernican) view of an entire social system is the study of whole networks, describing the comprehensive structure of role relationships in a complete population. Analysts can have simultaneous views of the social system as a whole and of the parts that make up the system. Through manipulating matrices, they can find patterns of connectivity and cleavage within social systems, structurally equivalent role relationships among social system members, changes in network structures over time, and the ways in which system members are directly and indirectly connected. For example, analysts can trace horizontal and vertical flows of resources and detect structural constraints operating on flows of resources. They can find densely-knit clusters, structural holes (Burt 1992), areas of high interaction or social isolation (Scott 1991; Wasserman and Faust 1993).

Therefore many community network analysts — including the authors in this book — have concentrated on studying smaller personal (or ego-centered) networks defined from the standpoint of focal persons: a sample of individuals at the centers of their own networks. Rather than showing the universe as it is viewed by an outside observer, personal network studies provide Ptolemaic views of networks as they may be viewed by the individuals at their centers: the world we each see revolving around us. Figure 1, for example, shows the significant interpersonal ties of a typical North American.⁹  She is directly tied with each network member (by definition), and many network members are also significantly tied with each other. (For the sake of clarity, Figure 1 omits the direct ties between the focal person and her network members.) She has a densely-knit cluster of kin — three of whom are her socially-close intimates — and more sparsely-knit relations among a half-dozen friends and neighbors. One workmate stands apart, his isolation reflecting a separation of work and social life in this focal person’s life.

Figure 1: Personal Network Diagram (click on thumbnail for larger image)

Personal network studies enable researchers to study community ties, whoever with, wherever located, and however structured. They focus on the inherently social nature of community and avoid the trap of looking for community only in spatially-defined areas. These personal community studies have meshed well with mainstream survey research techniques. Researchers have typically interviewed an (often large) sample of focal persons, asking about the composition, relational patters, and contents of "their" networks. To measure network density, they typically ask the focal persons in their samples to report about relationships among the members of their networks. Such studies, began in Detroit (Laumann 1969a; Laumann 1969b; Laumann 1973) and Toronto (Casey 1995; Coates 1966; Coates and Wellman 1969; Craven and Wellman 1973; Wellman 1968) in the 1960s and have flourished ever since. Reflecting the substantial amount of funding available for studying health, many psychologists, sociologists and social workers have concentrated on studying the social support that community networks provide: the supportive resources that community ties convey and their consequences for mental and physical well-being and longevity (see reviews in Fischer 1984; Wellman 1990a; Wellman 1992c; Wellman 1993). For example, researchers have found that people with larger, more diversified personal communities were less susceptible to common colds and produced less mucus (Cohen et al., 1997).

By framing analyses in network analytic terms, researchers have been able to show that the fears of a former generation about the loss of community were incorrect. Community, network analysts argue, has rarely disappeared from societies. It has been transformed. Community network analyses — including those represented in this book — have shown the continuing abundance and vitality of interpersonal relationships even as they have been affected by capitalism, socialism, urbanization, industrialization, bureaucratization, and new transportation and communication technology. New forms of community have come into being to replace older ones. The demonstration of the pervasiveness and importance of personal communities has rebutted contentions that large-scale social transformations have produced widespread social isolation in an alienated "mass society" (e.g., Kornhauser 1959). If analysts focus on social ties and systems of informal resource exchange than on people living in neighborhoods and villages, community can be seen. The discovery that most community ties extend well beyond the neighborhood and the village has redressed the common tendency to identify communities with neighborhoods. In the Western world and perhaps elsewhere, most community ties stretch across a metropolitan region, with many extending across the nation or to another continent.

The definition of "the community" in community network studies is a matter of how investigators define ties, where they draw boundaries, and how high they raise the level of analytic magnification to take into account internal links within clusters.

1. Do analysts look at all types of ties or only those that provide specific types of support? For example, Espinoza’s chapter in this book looks only at ties that provide material aid - goods and services. Emotional support and sheer sociability are not important for his tale of survival in poor Chilean neighbourhoods.
2. Where do analysts draw boundaries? Although researchers may draw samples from a national population (see Ferrand, Mounier and Degenne’s study of France in this book), studying the total personal networks of many respondents is impractical. Each respondent would have to be questioned about more than 1,000 ties. Therefore analysts almost always look only at relations with either:

• a few close confidants (Laumann1973; Fischer, et al., 1977; Burt 1984, 1986; Marsden 1987);
• a handful of socially-close intimates (e.g., the chapters by Lee and Campbell; Ohtani; Ferrand, Mounier and Degenne);
• Or, at most, a score or so of active ties (e.g., the chapters by Espinoza; Wellman and Potter; Wellman and Gulia [social support]).

Moreover, researchers have to start somewhere. In community network studies, they typically select a random sample from a neighborhood or metropolitan area, even though they trace the residents’ network relationships to wherever they may be found.

3. How detailed are the analyses? Several studies have looked only at "community ties," dropping all the ties of all the respondents into one tiewise data set. Such studies are useful for showing that few ties stay within the neighborhood (Wellman, Carrington and Hall 1988), parents and adult children disproportionately exchange emotional aid (Wellman 1979; Wellman and Wortley 1989; Wellman and Wortley 1990), and Japanese community networks resemble North American ones (see Ohtani’s chapter in this book; see also Nozawa 1997). Other studies treat the personal community as the unit of analysis. For example, Wellman and Gulia’s support chapter here shows the effect of network size on the provision of social support, while other research using the same data shows that women exchange more emotional support than men (Wellman 1988a). Few studies have looked at internal structural variation within the personal community, although there are often densely-knit clusters within sparsely-knit communities. For example, our research group (Wellman, et al. 1991) found married people to have densely-knit clusters of their own kin within communities that were generally sparsely-knit. This is because in-laws rarely interact with the other side of the family.

Thinking of communities as personal communities does have its costs.

• It concentrates only on strong ties - and sometimes only on strong, supportive ties - neglecting the weaker ties that Mark Granovetter has argued (1973, 1982, 1995) transmit new information between groups and integrate social systems.
• It ignores the ecological juxtapositions with which all people must deal in their residential and social spaces. Even if they are not in my network, I am disturbed by the young men who party and play drums every night, and I am aided by the daytime residents who keep an eye on my street (Jacobs 1961).
• Analyzing the network structure of each personal community is procedurally difficult. This is because software for social network analysis such as UCINet is designed to analyze only one network at a time. Although each personal community can be treated as a whole network, the lack of provision for batch processing means that the data-crunching of hundreds of personal networks must be undertaken one at a time.¹⁰
• In a sample survey of any size, interviewing the members of a person’s personal community is impractical because such an approach would increase the sample size enormously. (For example a sample of 300 focal persons each with an average of 20 network members would require 6,000 interviews.) Hence community network studies usually rely on surveyed respondents’ reports about their network members. This hinders reliability (Bernard, et al. 1984) although no more so than the respondents’ reports about other aspects of their behavior. The least reliable and valid survey data are the respondents’ reports about the nature of the relationships among the members of their personal communities. Many people just do not know what Cousin Betty feels about Uncle Henry.

Back

The Nature of Community Networks

The authors in this book - along with other scholars - have already discovered much about the composition, structure, dynamics and operation of community networks. As an introduction to the chapters in this book, this section reviews what we now know.

1. Community Ties are Narrow, Specialized Relationships, not Broadly Supportive.

Both scholars and the public have traditionally thought of communities as composed of broadly-based relationships in which each community member felt securely able to obtain a variety of help. Yet a good deal of research (including the chapters by Ohtani; Bian; Ferrand, Mounier and Degenne; Wellman and Gulia [support]; Espinoza; Lee and Campbell) has shown that most community ties are specialized, with community network members usually supplying only a few kinds of social support (see also the reviews in Wellman 1988, 1992c). In France, kin and neighbors engage in mutual aid, but friends and neighbors are the confidants (Ferrand, Mounier and Degenne). In California, there are differences between trouble-shooting kin and companionable friends (Fischer 1982; Schweizer, Schnegg and Berzborn 1997). In Toronto, active community members usually supply only one or two out of the five types of social support, for example, small services and emotional aid but not large services, companionship or financial aid (Hall and Wellman 1985; Wellman and Wortley 1988; Wellman and Wortley 1990). (By contrast, Toronto spouses supply each other with all types of social support [Wellman and Wellman 1992]). Those network members who provide small services or emotional aid rarely provide large services, companionship or financial aid (Wellman, Carrington and Hall 1988; Wellman and Wortley 1989; Wellman and Wortley 1990). Parents and adult children provide the widest range of support although they rarely supply sociable companionship. Accessible ties — people living or working near-by, or otherwise in frequent in-person or telecommunications contact — provide important goods and services (Wellman and Wortley 1990). The strength of ties is important, with socially-close voluntary and multiple-role ties providing high levels of support. Yet Granovetter (1973, 1982) has cogently argued the importance of weak ties for linking sparsely-knit communities and providing people with a wider range of information.

The specialized provision of support in communities means that people must maintain differentiated portfolios of ties to obtain a variety of resources. They can no longer assume that any or all of their relationships will help them, no matter what is the problem. In market terms, people must shop at specialized boutiques for needed resources instead of casually dropping in at a general store. Like boutique shoppers, people who only have a few network members supplying one kind of support have insecure sources of supply. If the relationship ends — if the boutique closes — the supply of that particular type of support may disappear.

2. People are Not Wrapped Up in Traditional Densely-Knit, Tightly-Bounded Communities But are Manuevering in Sparsely-Knit, Loosely-Bounded, Frequently-Changing Networks.

In reality, personal communities are usually sparsely knit and loosely bounded. For example, the density of 0.33 we found in two Toronto studies means that only one-third of a person's active community members have active ties with each other. Moreover, these networks become even more sparsely-knit as people age and their networks get more complex: Mean network density declined from 0.33 to 0.13 over a decade (Wellman Wong Tindall, et al. 1997). As Wellman and Potter and Wellman and Gulia (support) show in this book, variation in the composition and structure of these community networks is more complex than the traditional Tönniesian dichotomy of communal versus contractual organization.

The complex and specialized nature of personal communities means that these are fragmented networks. People must actively maintain each supportive relationship rather than relying on solidary communities to do their maintenance work. In both Japan (Ohtani’s chapter, Nozawa 1997) and North America (Wellman and Gulia’s support chapter; Wellman 1979; Wellman and Wortley 1989, 1990; Fischer 1982), the kinship system as such does not supply much social support: Extended kin are rarely supportive, although a few immediate kin — parents, children and siblings — are quite supportive. Moreover, the tendency of computer-mediated communication to emphasize relationships based on shared interests rather than relationships based on kinship or neighborhood may mean that most on-line ties will also be specialized, based on a single shared interest, and transitory, as interests change (Wellman and Gulia’s surfers chapter).

The fragmentation, specialization and low density suggests that the nature of individual relationships may be more important than the nature of the networks for the provision of social support. This means that to receive support people must actively maintain each relationship rather than rely on solidary communities to do this for them. It also means that tie characteristics may have more effect than network characteristics on the provision of social support. Although tie characteristics are important (Wellman 1992c), Wellman and Gulia’s support chapter shows that the characteristics of community networks are also important. Larger, more heterogeneous and denser networks provide more support. A network is more than the sum of its ties: The composition and structure of community networks affect the provision of support beyond the effects of the characteristics of the specific ties in these networks. Emergent properties are alive and well and living in Toronto.

It is not that people's communities are disintegrating, but that they are in flux. Rather than locking people into one tightly-bounded social circle, 1,000 or so community ties ramify across changing, fragmented communities to connect people to the diverse resources of multiple social arenas (Kochen 1989). Many or the chapters in this book show how people make use of these ramified connections. They are useful for getting jobs in China (Bian; see also Lin 1997; Lin, Ye and Chen 1997) and Chile (Espinoza), finding financial capital in Hungary (Sik and Wellman 1997), and helping Hong Kong immigrants to settle into Canada (Salaff and Wong 1995). Indeed Stanley Milgram's (1967) and Harrison White's (1970) observations that the entire world is linked by paths of five or fewer indirect ties are the basis for a recent play and movie, Six Degrees of Separation (Guare 1990, 1993).

Just because community networks ramify, that does not mean that they connect all persons randomly. "Birds of a feather flock together" whether they flock by gender, socioeconomic status, ethnicity or race (see Ferrand, Mounier and Degenne’s chapter, Lee and Campbell’s chapter, Laumann 1966; Laumann 1973; Wellman 1992b). These clusters organize flows of resources and norms. Even when ties connect people with different social characteristics, they do so unevenly. Moreover, high rates of social mobility leave in their wake cross-cutting ties between people with different social characteristics. Low rates of mobility foster more tightly-bounded clusters (see Ferrand, Mounier and Degenne’s chapter; Herting, Grusky and Van Rompaey 1997).

Sparsely-knit, fragmentary, loosely-bounded communities make it possible to reach many people through short chains of "friends of friends" (Boissevain 1974). Yet in such sparsely-knit and loosely-bounded networks, people cannot depend on the goodwill or social control of a solidary community. Instead, they must actively search and manipulate their ramifying ties, one-by-one, to deal with their affairs. Indeed, Bian’s chapter shows this to be true even in reputedly solidary China (see also Freeman and Ruan 1997; Ruan, et al., 1997).

3. Communities Have Moved Out of Neighborhoods to be Dispersed Networks that Continue to be Supportive and Sociable.

As well as contemporary communities being fragmentary, sparsely-knit and loosely-bounded, they are rarely local groupings of neighbors and kin. The residents of developed societies usually know few neighbors, and most members of their personal communities do not live in the same neighborhood (Wellman 1990b; Wellman 1992b). People easily maintain far-flung relationships by telecommunications (with telephones recently being joined by faxes, electronic mail and the Web) and transportation (based on cars, expressways and airplanes). In Toronto, being within one hour's drive or within the local telephone zone — and not being in the same neighborhood — is the effective boundary for high levels of face-to-face contact and social support (Wellman, Carrington and Hall 1988; Wellman and Tindall 1993). Many relationships stretch even further than the metropolitan area, with an appreciable number spanning the continent or the ocean. This lack of local ties and the presence of community members living elsewhere weakens local commitment and encourages people to vote with their feet: leaving when conditions are bad rather than staying to improve things. For example, the Hong Kong emigrants studied by Salaff, Fong and Wong rely heavily on trans-Pacific ties to make their moves to Canada¹¹. 

However, communities have not totally lost their domestic roots. Although the community networks of Torontonians are far-flung, most of Torontonians’ face-to-face interactions are with people who live or work near them. Torontonians even have much of their telephone contact with neighbors (Wellman Wong Tindall, et al. 1997). Thus, even spatially liberated people cannot avoid neighbors. Local relationships are necessary for domestic safety, controlling actual land-use, and quickly getting goods and services, as Jane Jacobs (1961) has pointed out for North America in the 1950s and Lee and Campbell’s chapter and Wellman and Gulia’s support chapter reaffirm for contemporary North America. Moreover, when transportation and communication resources are scarce, local ties assume more importance as Charles Tilly (1973) has argued for portions of pre-industrial Europe and Vicente Espinoza’s chapter shows for impoverished Chileans.  In saying that communities are not as local as they used to be, we need to avoid committing the pastoralist fallacy of thinking our cities and suburbs are inferior to the pestilent, crime-ridden and insecure villages or cities of yore. Pre-industrial communities may never have been as locally-bounded as tradition has maintained. Whenever scholars have looked for non-local ties, they have found far-ranging networks. For example, radioactive analyses of obsidian have found Neolithic spear points and choppers more than one thousand miles from their origin (Dixon, Cann and Renfrew 1968). Moreover, as noted above, LeRoy Ladurie (1975, 1997), Natalie Davis (1983), among others, have described far-flung, mobile networks in Medieval and Renaissance Europe.

Consider, also, the fruits of the unlikely comparison of communities in 20^th-century Toronto and 18^th-century rural Latvia (Wetherell, Plakans and Wellman 1994). By contrast to the mythical kinship-ridden past, we found that this rural Latvian community did not have enough kin to construct the kinds of social networks that exist today. As these farmers do not appear to have had many friends living beyond the local area, it seems that half the myth was true: Although these groups were local, they only had small clusters of kin at their core. Closer to home, many guests at mid-19th century New York City weddings -- presumably the heart of the marital family’s social networks — came from other parts of the city, and often from other counties or states (Scherzer 1992).

4. Private Intimacy has Replaced Public Sociability.

Rather than operating out of public neighborhood spaces, contemporary communities usually operate out of private homes. Yet until well into this century, men customarily gathered in communal, quasi-public milieus, such as pubs, cafes, parks and village greens. Take for example this description of eighteenth century Paris:

The whole neighborhood overflowed into the street from nearby houses, workshops, shops and taverns. Around every inhabitant a quartier took on its shape, made up of daily contacts and changing reputations. Individuals worked round the corner from where they lived. (Roche 1981, p. 246)

Although the men generally went out to enjoy themselves, they also used these public communities to organize politically, to accomplish collective tasks, and to deal with larger organizations. In colonial New England, "neighbors assumed not only the right but the duty to supervise one another's lives" (Wall 1990). This public community was largely a man's game. A woman who went alone to a Parisian wine shop risked being mistaken for a prostitute (Garrioch 1986). 

Community has moved inside now, into private homes. The separation of work from residential localities means that co-workers commute from different neighborhoods and no longer come home from work in solidary sociable groups. While men now spend more time at home instead of at bars or cafes, the high percentage of women engaged in paid work outside their homes means that women spend less time at home. Thus husbands and wives are now apt to be at home when both are available to each other. They stay home too, for they are in no mood to go out and socialize after their weary trip home from work. In any event, zoning regulations in North America often place commercial areas for recreation far from home. Domestic pursuits dominate, with husbands and wives spending evenings and weekends together instead of the men going off to pubs and street corners, and few women being home during the day. Workaholics bring their computer disks home; couch potatoes rent videos.

Yet the easy accessibility of local relationships means that those local ties that do exist are significant. Although neighbors (living within one mile) comprise only 22% of the Torontonians' active ties, these neighbors engage in fully 42% of all interactions with active network members (Wellman 1996).  The neo-conservative privatization of Western societies, the withering of collective public services for general well-being, is reflected in the movement indoors of community life. Even in Toronto, the safest North American metropolis, 36% of the residents report that they feel unsafe walking alone in their neighborhoods at nights (Duffy 1991). Yet the usual flight to safety — driving a car or staying home and using the telephone or e-mail — offers little opportunity en route for the casual contact and new encounters that can diversify lives. Cars leave garages as sealed units, opened only on reaching the other's home; telephones and modems stay indoors, sustaining closed duets with already-known others.

Where North Americans a generation ago often spent Saturday night going out for a movie and pizza, they now invite a few friends over to their homes to watch videos and order a pizza to be delivered. In 1992, the average Canadian household spent $101 for buying and renting videos compared with $99 for going to live theatre, concerts and movies. It costs $3 per household to rent a video in Toronto, but $8 per person to go to the movies and about $30 to attend a play or concert (Film Canada 1990; Strike 1990). This means that people watch videos at home an average of thirty times per year but go out for entertainment only three or four times a year. The telephone number for Toronto's largest pizza delivery service, 967-11-11, has become so well known that Canadian immigration officers use it as a test to see if border crossers are bona fide Canadian residents. As Toronto pundit Marshall McLuhan observed (1973), North Americans go out to be private — in streets where no one greets each another — but they stay inside to be public — to meet their friends and relatives.

Public spaces have become residual places to pass through or to shop in. Rather than participating in clubs or organizations, when they do go out, North Americans usually go out alone, in couples or in small, informal groups (Putnam 1995). North American church attendance is declining, and Canadian movie attendance declined from eighteen times per year in 1952 to three times per year in 1993. When Torontonians do go out to the movies, most (55%) go alone or in pairs (Oh 1991). The community of the pub in the recent television show, Cheers, was appealing because it is rare. In reality, only 10% of adult Canadians go to a pub once a week or more.¹²  Suburban shopping malls have become residual agoras — for consumption purposes only but not for discussion. Their cafes mock the name, deliberately using tiny tables and uncomfortable chairs to discourage lingering sociability. They provide little opportunity for casual contact or the expansion of networks. This trend is most marked in North America, where "fast food" restaurants tell their patrons to "have a nice day" and expect them to stay less than a half-hour.

As community has become private, people feel responsible for their "own" — the members of their community networks with whom they have strong ties — but not for the many acquaintances and strangers with whom they rub shoulders but are not otherwise connected. Private contact with familiar friends and relatives has so replaced public gregariousness that people pass each other unsmiling on streets. This privatization may be responsible for the lack of informal help for strangers who are in trouble in public spaces (Latané and Darley 1976). It is probably also a reason why people feel that they lack friends and are surrounded by strangers even when their networks are abundantly supportive (Lofland 1973).

Unfortunately, social network analysis has been better at studying the strong ties of personal, private community than at studying the weak ties and ecological juxtapositions of public community. Network analysts have found out only about strong active ties (Campbell and Lee 1991; Marsden and Campbell 1984) by asking about to whom people feel close — as I did (Wellman 1979; Wellman 1982) and as the U.S. General Social Survey did in 1985 (Burt 1984; Marsden 1987) and the Canadian equivalent did in 1985 (Statistics Canada 1987, Stone 1988) — or by asking from whom they get various kinds of social support, as both American (e.g., Fischer 1982b) and Dutch social scientists have done (e.g., Knipscheer and Antonucci 1990). Network analysts have been useful and accurate in saying that strong personal communities continue to exist, but they have neglected to look at what is happening all around these networks.

5. Communities have become Domesticated and Feminized

Home is now the base for relationships that are more voluntary and selective than the public communities of the past. Despite the importance of neighborhood ties portrayed in Lee and Campbell’s and Espinoza’s chapter, only a minority of community ties in the Western world operates in the public contexts of the neighborhood, formal organizations or work. Community networks now contain high proportions of people who enjoy each other and low proportions of people who are forced to interact with each other because they are juxtaposed in the same neighborhood, kinship group, organization or workplace (Feld 1981). Friends and relatives get together as small sets of singles or couples, but rarely as communal groups (Wellman 1992b). This voluntary selectivity means that communities have become homogeneous networks of people with similar attitudes and life-styles. Wellman and Gulia’s surfer chapter suggests that the proliferation of computer-mediated communication will only accelerate this trend.

Where once-public communities had been men's worlds, now home-based community networks bring husbands and wives together. Men's community ties are now tucked away in homes just as women's ties have usually been. As community has moved into the home, homes have become less private. Previous generations had confined visitors to ground-floor parlors and dining rooms, but network members now roam all floors.

In their domestic headquarters, Toronto couples operate their networks jointly (Wellman and Wellman 1992). It is a far different scene from the segregated networks that Elizabeth Bott (1957) described in the 1950s for England where husband and wife each had their separate circles of kin and friends. Usually it is the household that exchanges support rather than the person: for example, our Toronto research found in-laws to be as supportive as blood relatives (Wellman and Wortley 1989). By contrast to the specialized support that community members exchange, spouses supply each other with almost all types of social support (Wellman and Wellman 1992). Hence unmarried adults obtain much less social support domestically and do not have access to the networks (and their resources) that accompany spouses to marriage.

In the current situation, married women not only participate in community, they dominate the practice of it in their households. Women have historically been the "kinkeepers" of western society: mothers and sisters keeping relatives connected for themselves, their husbands and their children. They continue to be the preeminent suppliers of emotional support in community networks as well as the major suppliers of domestic services to households (Wellman 1992c; Wright 1989). With the privatization and domestication of community, community-keeping has become an extension of kinkeeping, with both linked to domestic management. No longer do husbands and wives have many separate friendships. As men now usually stay at home during their leisure time, the informal ties of their wives form the basis for relations between married couples. Women define the nature of friendship and help maintain many of their husbands' friendships. Women bear more than the "double-load" of domestic work and paid work; their "triple load" now includes community "net work".

Thus the privatization and domestication of relationships have transformed the nature of community. The domesticated community ties interact in small groups in private homes rather than in larger groups in public spaces. This makes it more difficult for people to form new community ties with friends of their friends, and it focuses the concerns of relationships on dealing with household problems (Wellman 1992b). Women's ties, which dominate community networks, provide important support for dealing with domestic work. Community members help with daily hassles and crises; neighbors mind each other's children; sisters and friends provide emotional support for child, husband and elder care. Because women are the community-keepers and are pressed for time caring for homes and doing paid work, men have become even more cut off from male friendship groups (Wellman 1992c). North American men rarely use their community ties to accomplish collective projects of work, politics or leisure. Their relationships have largely become sociable ties, either as part of the relationship between two married couples or as disconnected relations with a few male "buddies".

This domestication helps explain the contemporary intellectual shift to seeing community and friendship as something that women do better than men. Just as husbands and wives are more involved with each other at home, the focus of couples and male friends is on private, domestic relations. Men's community ties have come to be defined as women's have been: relations of emotional support, companionship, and domestic aid. Thus the nature and success of community are now being defined in domestic, women's terms. Concurrently, the growing dominance of the service sector in the economy means that the manipulation of people and ideas has acquired more cultural importance than the industrial and resource-extraction sectors' manipulation of material goods. With developed economies having more managers and professionals than blue-collar workers (Statistics Canada and Canada 1993), the workplace has shifted to the very emphasis on social relationships that women have traditionally practiced at home.

Contemporary discussions of community often reverse the traditional sexist discourse that has seen women as inadequate men. Now it is men who are seen as unable to sustain meaningful community relationships, especially when such relationships are defined only in terms of socioemotional support. This socioemotional definition has almost totally replaced the traditional definition of community as also including instrumental aid. Patriarchical arguments for male superiority in getting things done are being replaced by celebrations of female superiority in knitting together social networks. As "feminist author" Maggie Scarf (Scarf 1987) said on the Oprah Winfrey television show, "Men just don't have friends the way women have friends. Men just don't like to make themselves vulnerable to other men."Clitoris-envy, the alleged longing for empathy among men, has become the new-age replacement for penis-envy among the not-so Iron Johns (Bly 1990).

Back

Seeing Community Networks in Context

In less materially comfortable parts of the world, community members do more for each other than being privately sociable and emotionally supportive. Consider how people elsewhere use friends for economic, political and social survival. Greek men argue and plan projects in cafes, poor Chileans help barrio neighbors to survive and find jobs for kin (Espinoza’s chapter), Chinese job seekers rely heavily on networks (Bian’s chapter; Lin, Ye and Chen 1997), Hungarians help each other build new homes (Sik and Wellman’s chapter), while Hong Kong networks help people to leave their homes (Salaff, Fong and Wong’s chapter). Even in more affluent Britain, people value getting services and information from community members as much as they value getting esteem and affection (Argyle 1990). To put matters more broadly, communities do not function in isolation but in political, economic and social milieus that affect their composition, structure and operations. The nature of different societies strongly affects the opportunities and insecurities with which individuals and households must deal, the supportive resources they seek, and the ways in which markets, institutions and networks structure access to these resources.

In many societies, communities are not just ways in which people spend some of their leisure time but key mechanisms by which people and households obtain resources. Yet most North American research has ignored the broader implications of community ties and looked only at "social support": the effects of community ties on maintaining physical and mental health. Although this is an important matter, it is unfortunate how the high level of funding for health-care research has focused attention so narrowly. A broader view would see community as an essential component of society, one of the five principal ways by which people gain access to resources:¹³

• Market Exchanges (as purchases, barter or informal exchanges). Seeing this as the only means of access to resources is in line with the neo-conservative belief in the loss of community. [Liberty]
• Institutional Distributions (by the state or other bureaucracies as citizenship rights, organizational benefits or charitable aid). Such access to resources is in line with those who have traditionally seen society as a moral community writ large, as in the current American debate about whether health care is a community obligation or a market decision. However, the use of the term "community" to describe such institutional distributions can be a subterfuge for bureaucratic privilege, as was the case in communist eastern Europe [Equality]
• Community Exchanges. If informal, interpersonal access to resources occurs within neighborhood or kinship solidarities, then it fits traditional notions of community. However, the two chapters that portray this pattern describe community among impoverished, new in-migrants (see Espinoza’s chapter describing social support among poor Chileans) and among a relatively poor, segregated American population (see Lee and Campbell’s chapter describing neighboring in Nashville. If the exchanges are less-bounded (and so less normatively-enforceable), then it fits the ramified networks that Wellman and Gulia’s support article describes among materially comfortable Canadians. [Fraternity]
• Coercive Appropriations Direct predatory behavior by interpersonal (robbery) or institutional bullies (expropriation). Involuntary appropriations usually occur under the legitimating guise of imbalanced market exchanges or state extractions for unequal institutional distributions (as in governments forcing farmers to sell produce to urbanites at low prices (Tilly 1975)). More extreme instances of the loss of community are common in societies where institutional and communal mechanisms of social control have broken down, such as in Bosnia or Rwanda. [Robbery]
• Self-Provisioning Making and growing things in one's household. Self-provisioning is used even in market societies (see Pahl's [1984] discussion of growing food in England) and in socialist-institutional ones (see Sik's [1988] discussion of Hungarian home building). Such self-provisioning rests on an infrastructure of market and community exchanges that provide advice, skills and materials. [Peasantry]

The comparatively low importance of economic and political concerns in western societies distinguishes the communities in them from those in societies that are less economically or politically secure. Most westerners rely on market exchanges for almost all of their production and much of their consumption. Institutional benefits such as schooling and medical care are abundantly available as citizenship rights. Westerners do not pay as much attention as the inhabitants of statist societies (such as the former East European socialist states) to having community members who can make and fix things (such as home building) or who have connections to strategic institutional circles (see Sik and Wellman’s chapter). To make another contrast, because westerners rarely have urgent cares about daily survival, they can manage domestic resources with less apprehension than third-worlders living on the margins.

Back

Networks in the Global Village

Malvina Reynolds (1963?) sang satirically in the 1960s about supposedly buttoned-up, carefree North American life. She described it as:

"Little boxes made of ticky-tacky
Little boxes, little boxes, little boxes
All the same.
There’s a green one, and a pink one, and a blue one, and a yellow one.
And they’re all made out of ticky-tacky
And they all look just the same."

Although Ms. Reynolds was giving her dystopian vision of North American suburban homes, she also was critiquing American society as a set of little boxes. The chapters in this book show, fortunately, that the little boxes are only the homes and not the social reality. Wherever possible — across the global village — people have reached out and transcended their little neighborhood and kinship boxes. They are involved in complex community networks stretching across their cities, regions, nations, and even the oceans. The multiple clusters and limited social control in these networks give people room to maneuver, even if the cost is that they must actively maintain their ties and scan their networks for help. The cost of escaping these little boxes is that people think that they and the world are not well-connected. The advantage is that they have much autonomy to connect where they will.

In the bad old days, before the 1960s, people feared that community had disappeared.

In the good old days of the 1960s and 1970s, people thought community was thriving naturally, as a combined group love-in and support-in.

In the entrepreneurial days of present time, the product of a neo-conservative zeitgeist, people think that community flourishes only if they go out and pull its strings.

But community is not a bad black hole, it is not a solidary, all-loving group, and it is probably not even a set of exchange freaks playing "let’s make a deal!"

It is a network - nebulous, far-flung and sparsely-knit, but real and supportive.

Back

References

Back

Notes

1. I thank Mark Chapman and Reena Zeidman for advice in Biblical matters, and Abraham Friedman who gave Bev Wellman and me the Bible used here ("translated in accordance with Jewish tradition," 1947) upon our marriage in 1965, inscribed with the blessing, "May you be blessed with Love, Contentment and Devotion for each other").
   Back

2. For further details of this paragraph's argument, see Wellman and Wetherell (1996).
   Back

3. To distinguish in this introduction between the two chapters by Wellman and Gulia in this book, where necessary I refer to the chapter titled about the social network basis of support as "support" and to the chapter about virtual communities as "surfers".
   Back

4. For a summary of mass society fears, see Kornhauser (1968). Key Third World community studies from this period include Mayer (India, 1966), Cohen (Nigeria, 1969), Mayer with Mayer (South Africa, 1974), Mitchell (Rhodesia, 1956) and Peattie (Venezuela, 1968) are.
   Back

5. During the communist era, there were rural village studies such as Hinton's study of Fanshen in China (1967) and also studies of work organizations as intermediary units, such as Burawoy's study of a Hungarian factory (1985). See also Kennedy and Galtz' recent review (1996).With the exception of Radoeva's Bulgarian analysis (1988), I confine myself to works in English.
   Back

6. Similar rural-urban mobility often occurs in contemporary Third World societies, with low-cost buses and letter-writers helping to maintain connectivity (In addition to Espinoza's chapter in this book, see, for example, Mayer and with Mayer 1974; Doudou 1967; Roberts 1973, 1978).
   Back

7. The network members in community studies are persons but in other network analyses they could be organizations, states, etc.
   Back

8. In addition to the discussion below, see also Fischer 1982a; Wellman 1988a; Wellman and Leighton 1979.
   Back

9. The specifics are drawn from the Toronto studies described in this book; see also Wellman, Carrington and Hall 1988.
   Back

10. Haythornthwaite and Wellman (1996) have created a procedure using SAS software for decomposing whole networks into ego-centered networks so that each network member's world can be analyzed separately.
    Back

11. Some, especially businessmen, retain important ties with Hong Kong. They are called ?astronauts? in Hong Kong slang, because they fly back and forth frequently on long, grueling trips. I am not arguing that local ties are unimportant, only that they usually comprise a minority of important community ties.
    Back

12. Special analysis by Scot Wortley of the 1989 Canadian National Alcohol and Other Drug Survey.
    Back

13. French revolutionaries may have realized three-fifths of this with their demand for ?Liberty, Equality and Fraternity.? Perhaps their revolutionary sentiments for a new order led them to deny both Robbery and Peasantry.
    Back