Part 3 - Identity and the Internet

Social Networks


As we have seen, community has typically been difficult to define. The criteria for definition seem aqueous and lack a firm empirical ground. According to Fischer (1982: 271), "The first problem is deciding how to delimit the basic unit, the locality. This is a persistent problem in the sociology of community." Early empirically oriented sociologists (e.g., Chicago School sociologists) tended to define such units ecologically on the basis of a purely spatial notion (neighborhoods), whereas more recently, as Fischer points out, locality is egocentric, or located with the relationships between people and not solely in the space itself (cf. Wellman, 1983:1-2 on "personal communities"). In this way, communities tend to shade imperceptibly from one into another, making the objective definition of any given community quite difficult.


It is precisely this egocentric perspective on community that arose in the 1970s around the problem of empirical definition. A new direction was found in the renewed interest 24 in social network analysis, a style of investigation into relationships that is based roughly upon the sociometry developed by Moreno (1934). Most network analysts 25 look at the objective pattern of ties linking the members of society. Some important concepts that are central to describing this objective structure in network analysis are density, clustering, multiplexity and intensity. Density is the extent of interlinkage among the actors, usually expressed as the ratio of the number of existing links to the number of possible links. Clustering is the extent to which the total network is divided into distinguishable "cliques" or groups. These concepts describe the network as a whole, while the latter two describe the dyadic links within the networks. Multiplexity, also known as multistrandedness, is the number of relations within a given link (i.e., the number of ways in which two people know each other or the number of role associations). Intensity is the intimacy of the relationship as represented by the degree of commitment 26 in a link. (Fischer, 1977: 36).


As I have mentioned, there has been great interest in developing and applying such concepts to studies of social structure in the past thirty years. These applications have spanned the range of sociological investigative techniques, whether quantitative or qualitative (although most have been quantitatively oriented) because of their usefulness at conducting empirical investigations.


The methods of quantitative social network analysts rely largely upon the development of computer models of aggregate data. They are most stringent in maintaining the intellectual position that the structure of social networks is wholly objective in character and they go about studying society in terms of these objective positional linkages. Very often, the computer models used to represent networks can get quite complex. These can range from graph-theoretical models (e.g., blockmodels) and clustering analyses (such as CONCOR) developed by mathematicians to "a number of powerful programs intended for chemists and biologists which can be used to produce useful images of social structure" (Freeman - net cite). A great number of programs have been developed since the 1970s, when "Davis, Holland, and Leinhardt led the way out of the ‘statistical darkness’ and used random graph distributions to study clusterability and clique formation" (Galaskiewicz & Wasserman, 1993). A large contingent of this type of analysis is represented in Social Networks, the major journal of the field. The analysts typically focus on the mathematical properties, rather than the human qualities, of social networks, in a positivistic attempt to discover rules of . Although they subscribe to a variety of different ideas pertaining to social networks, the sociologists mentioned above, as well as those such as Harrison White, Ron Burt, Stanley Wasserman, Joseph Galaskiewicz, Phillip Bonacich, Karen Cook and many others generally fall into the category of quantitative social network analysis.


Some researchers rely less on mathematical and structural models in lieu of more traditional statistical and investigative research methods. They employ network concepts, but use them in the context of a focus upon the relations between actors in the networks (as opposed to traditional sociology which focuses on individuals as independent actors), rather than solely the positions they may fill. This type of social network analysis is more aligned with a symbolic interactionist standpoint, due to the focus on behavior. They may use survey techniques to obtain data and then apply regression models or some other standard statistical technique to help in explaining the findings. But the chief characteristic that makes these researchers network analysts and not traditionally oriented is their commitment to the important contribution of the network idea that "integrates a structural analysis of society with a viable model of the individual, one in which he or she is a participant in the construction of the social world" (Fischer, 1977: 29) (a notion that, as we have seen, is also foremost in symbolic interactionist sociology). The use of concepts such as density and multistrandedness allow for the creation of this network idea. Furthermore, these sociologists tend to lend more credence to the age old question of community that has been the focus of so much attention. Many researchers in this category have come from schools of urban sociology (e.g. Fischer, 1977, 1982; Wellman, 1978, 1983, 1989).


On the other side of the spectrum of network research is the cultural analysis of social networks. This kind of research has generally appeared much more infrequently than the prior two types, but certainly has credence in contemporary sociology. This research is often descriptive, relying on the network metaphor in a looser context than the former two sub-disciplines. It may involve more theorizing, more historical review and less of a focus on the interpretation of raw data collected by standard methodologies. The focus is often upon the cultural and economic institutions which have increasingly organized society itself around a network model. The perspective is not new; for example, Simmel’s The Web of Group Affiliations (1955) is written in this vein, as is Wiener’s (1954) The Human Use of Human Beings: Cybernetics and Society. However, with the very recent expansion of information and communication technologies en masse, there is a renewed interest in this type of analysis. Perhaps the most recent large work 27 on the subject is Castells' The Rise of the Network Society (1996) , in which he notes,

"Networks constitute the new social morphology of our societies, and the diffusion of networking logic substantially modifies the operation and outcomes in processes of production, experience, power, and culture. While the networking form of social organization has existed in other times and spaces, the new information technology paradigm provides the material basis for its pervasive expansion throughout the entire social structure" (1996:469).


This change in the character of society is recognized by earlier social theorists, such as Arendt and Mills,28 but becomes really meaningful in today’s world, as Castells says.


The relationship of this "morphology" to changes upon individuals is also addressed by cultural network analysts. Specific to this interest is the construction and maintenance of identity; in fact, the second volume of Castells’ The Rise of the Network Society (1996) is titled The Power of Identity (1997). As he explains (1996: 24,4), "elements of an interpretive framework to explain the rising power of identity must be found at a broad level, in relationship to macroprocesses of institutional change, to a large extent connected to the emergence of a new global system… We should keep in mind that the search for identity is as powerful as techno-economic change in charting the new history." This type of cultural/individual analysis helps to more clearly define the theoretical proposition of bridging the micro-macro gap that is important to the other types of network analysis.29 The meaning of the relationships between macro-processes of institutions and organizations and the construction of identity in a society that is becoming intensely connected is generally more thoroughly approached in such work. Mulgan’s (1997) ideas about "connexity," Lynch’s (1998) ideas about "memes," Habermas’ (1987) ideas about "communicative action," and Gergen’s ideas about the "saturated self" are other examples of cultural network analysis.


The goal of the following research proposal is to analyze specifically the symbolic interactionist implications of computer-mediated communications media upon identity, status and social support. Although the network metaphor is useful in all of the aforementioned respects, from formal quantitative to informal cultural analysis, and while the lines between them are certainly not cut and dry, it is the network theories of the informal network researchers and cultural analysts that fit this project most adequately; the theories and questions that these perspectives arrive at are most germane to this project. While much of network analysis focuses upon quantitatively deduced models of social structures, the current research fits into a more relational approach, by virtue of its being informed by a symbolic interactionist perspective and by virtue that its focus lay in the development and maintenance of identity. The need for this type of research is clear. As Jones (1995) says,

"It would be far easier to understand the physical, hardwired, connections [of CMC] than to understand the symbolic connections that emerge from interaction. Much of our energy has been directed toward understanding the speed and volume with which computers can be used as communication tools. Conspicuously absent is an understanding of how computers are used as tools for connection and community."

To Next Section - Connections and Research Directions

To Previous Section - Identity
Back to Title Page


Back to the Infospace Page