3. Conceptual framework
3.3. The social network
The structure of social interactions affected the long-term development of Australia’s economic history community. The analysis of the social network is based on theoretical explanations for the development of ties, and the effect of these ties on the production of knowledge in a group. Based on these assumptions, some interactions have been
visualised using SNA. Social networks indicate the probability of interactions between scholars based on geographic proximity and collaboration. These are a key vehicle through which communication and the diffusion of ideas may occur.
3.3.1. The development of social networks
Social networks are created through interpersonal interactions. These may be motivated by intrinsic factors, with some arguing that individuals form connections based on self-interest, meaning they aim to maximise their personal preferences and invest in their social capital.98 These decisions are constrained by interdependence with others in the community, and there is a delicate balance between the constraints and resources of each interaction.99 The exchange approach argues that individuals are motivated to minimise
98 R. Burt, 'Structural holes and good ideas', American Journal of Sociology, 110, 2, 2004; Burt, 'Social capital'; Coleman, 'Social capital'; N. Katz, D. Lazer, H. Arrow and N. Contractor, 'Network theory and small groups', Small group research, 35, 3, 2004; Millar and Choi, 'Networks'.
99 Katz, et al., 'Small groups'; Millar and Choi, 'Networks'.
28
their dependence on others, and maximise the dependence of others on them for
resources or contacts.100 Homans, the forebear of this model, and later Blau, have argued that people establish connections with those with whom they can exchange resources. The
‘value’ from each interaction then structures the long-term relationships in the
community.101 The valuation and exchange of knowledge has been argued to drive the creation of intellectual communities.102 The mutual interest model argues that individuals form groups to maximise their collective abilities and the benefits of co-ordinated
action.103 Individuals, in this model, tend to interact more with those that have similar interests and are willing to share ideas and resources. Groups operate similar to a ‘club good’, in which participants must make some sort of contribution in order to share in the benefits of the network. Collective action, particularly through teamwork and
collaboration, has been emphasised for the creation of new knowledge in intellectual communities.104
Networks may also form based on cognitive factors. Transactive memory argues that individuals, with their own set of skills and knowledge, develop connections with those who have complementary skills and knowledge. This facilitates the flow of ideas, and reduces the need for each member to possess knowledge that is available elsewhere in the group.105 Balance theory, conversely argues that ‘cognitive balance’ is required for
100 P. Blau, Exchange and power in social life, New York: Wiley, 1964; R. Cropanzano and M. Mitchell, 'Social exchange theory: An interdisciplinary review', Journal of Management, 31, 6, 2005; R.
Emerson, 'Social exchange theory', Annual Review of Sociology, 2, 1, 1976; A. Gouldner, 'The norm of reciprocity: A preliminary statement', American Sociological Review, 25, 1, 1960; G. Homans, The human group, New York: Harcourt, Brace & World, 1950; J. Thibault and H. Kelley, The social psychology of groups, New York: John Wiley, 1959.
101 Blau, Exchange and power; Homans, Human group.
102 J. Kenway, E. Bullen and S. Robb, 'The knowledge economy, the techno-preneur and the
problematic future of the university', Policy Futures in Education, 2, 2, 2004; Nahapiet and Ghoshal, 'Social capital'.
103 M. Granovetter, 'Threshold models of collective behaviour', American Journal of Sociology, 83, 6, 1978; R. Hardin, Collective action, Baltimore: John Hopkins University Press, 1982; M. J. Olson, The logic of collective action, Cambridge, MA: Harvard University Press, 1965; P. Samuelson, 'The pure theory of public expenditure', Review of Economics and Statistics, 36, 1, 1954.
104 J. Katz and B. Martin, 'What is research collaboration', Research Policy, 26, 1, 1997; Katz, et al., 'Small groups'; Millar and Choi, 'Networks'; Nahapiet and Ghoshal, 'Social capital'; L. Zucker, M.
Darby, M. Brewer and Y. Peng, 'Collaboration structures and information dilemmas in
biotechnology: Organisation boundaries as trust production', in Kramer and Tyler, ed., Trust in organisations: Frontiers of theory and research, Thousand Oaks, CA: Sage, 1996.
105 A. Hollingshead, 'Retrieval processes in transactive memory systems', Journal of personality and social psychology, 74, 1, 1998; A. Hollingshead, J. Fulk and P. Monge, 'Fostering intranet knowledge-sharing: An integration of transactive memory and public goods approaches', in Hinds and Kiesler, ed., Distributed work, MA: MIT Press, 2002; Katz, et al., 'Small groups'; D. Wegner, 'Transactive memory: A contemporary analysis of the group mind', in Mullen and Goethals, ed., Theories of group behaviour, New York: Springer-Verlag, 1987.
29
interaction between actors.106 If two individuals do not consistently evaluate a third person, they experience a state of discomfort and strive to either re-evaluate their impression of the third person, or their existing friendship. Individuals are thus more likely to interact with those whose friends are also friends with one another.107 Homophily argues that individuals are likely to form networks with those that have similar ideas and personal characteristics.108 Similar values, age, gender, occupation, or academic discipline are thought to “ease communication, increase predictability of behaviour and foster trust and reciprocity”.109 Homophily has been applied to intellectual communities, with some arguing that ‘sameness’ between scholars creates networks where other motivations for interaction (such as geographic proximity) fail.110
These internal motivations argue that elements of the knowledge network structure social interactions. Scholars may choose to connect with others based on similar backgrounds, the value of their intellectual contribution, complementary skills, or the compatibility of their approach.
Social networks may also form based on macro-factors such as geographic space, common institutions, or joint activities. The social networks in this thesis are informed by Feld’s foci theory, which argues that contextual entities are an important determinant of networks.111 Figure 3.2 presents Feld’s formulation of the process of network formation through contextual factors. Two individuals who share a focus (such as a common
workplace or neighbourhood) are more likely to share joint activities than two individuals who do not share that focus. Joint activities lead to interactions between individuals, and if there is a positive outcome from these interactions, individuals will try to develop new foci
106 D. Bloor, Knowledge and social imagery, Chicago: University of Chicago Press, 1976; F. Harary, R.
Norman and D. Cartwright, Structural models, New York: Wiley, 1965; F. Heider, 'Attitiudes and cognitive organisation', Journal of Psychology, 21, 1, 1946; F. Heider, The psychology of interpersonal relations, New York: Wiley, 1958; Homans, Human group; T. Newcomb, The acquaintance process, New York: Holt, Rinehart & Winston, 1961.
107 See D. Krackhardt and M. Kilduff, 'Friendship patterns and culture: The control of organizational diversity', American Anthropologist, 92, 1, 1990 for an applied example.
108 D. Brass, 'A social network perspective on human resources management', Research in Personnel and Human Resources Management, 13, 1, 1995; P. Lazarsfeld and R. Merton, 'Friendship as a social process: a substantive and methodological analysis', in Berger, ed., Freedom and control in modern society, New York: Van Nostrand, 1954; M. McPherson, L. Smith-Lovin and J. Cook, 'Birds of a feather: Homophily in social networks', Annual Review of Sociology, 27, 1, 2001.
109 Brass, 'Human resources management', p.51.
110 S. Aral, L. Muchnik and A. Sundarararjan, 'Distinguishing influence-based contagion from homphily-driven diffusion in social networks', Proceedings of the National Academy of Sciences of the United States of America, 106, 51, 2009; C. Rawlings, D. McFarland, L. Dahlander and D. Wang, 'Streams of thought: Knowledge flows and intellectual cohesion in a multidisciplinary era', Social Forces, 93, 4, 2015.
111 Feld, 'Social ties'.
30
to organise joint activities.112 This emphasises the effect of context, though ongoing interactions are also dependent on the internal sentiments that individuals derive from each interaction. This approach has been used to explain intellectual networks, with institutions, conferences, journals and professional societies argued to increase interaction and the diffusion of knowledge between scholars.113
Figure 3.2: Dynamic model of foci and group development
Source: Feld, ‘Social ties’, p.1026.
In Feld’s model, the characteristics of each focus affects the type of network that develops.
If the focus has greater constraint – meaning greater restrictions on time, effort and emotion – it is more likely that two individuals will develop a relationship. For example, a family unit is highly constrained because individuals are forced to interact heavily and often. This means that all individuals associated with the family will have a relationship. A city neighbourhood, on the other hand, is much less constrained, meaning only a slightly
112 Feld, 'Social ties'.
113 R. Boschma, 'Proximity and innovation: a critical assessment', Regional Studies, 39, 1, 2005;
Crane, Invisible colleges; A. Jaffe, M. Trajtenberg and R. Henderson, 'Geographic localization of knowledge spillovers as evidenced by patient citations', The Quarterly Journal of Economics, 108, 3, 1993; J. Katz, 'Geographical proximity and scientific collaboration', Scientometrics, 31, 1, 1994;
Kuhn, Structure of scientific revolutions; Mullins, Theory and theory groups; R. Ponds, F. van Oort and K. Frenken, 'The geographical and institutional proximity of research collaboration', Papers in Regional Science, 86, 3, 2007; Price, Little science; Wright, 'A network analysis'.
31
higher proportion of individuals will be tied than otherwise. The size of the foci also has an effect, with larger foci generally less constrained as it is more difficult to arrange for many people to engage in frequent joint activities. Thus, as the size of the foci increases, the probability of a strong relationship between two individuals decreases. Easier co-ordination means that smaller foci are generally more highly constrained, with a greater chance of strong ties.114
Feld’s foci model provides a dynamic specification of the development of social networks.
Ties are maintained over time, provided the foci remain consistent. Ties may also be reinforced over time, with positive outcomes from an interaction encouraging individuals to find new foci around which to organise activities. By a similar token, negative outcomes from an interaction may mean that individuals remove themselves from the particular foci, contributing to the dissipation of the tie over time.
Social networks may also be affected over time through technological change. Social ties are generated and maintained through interpersonal communication, and so
improvements in communication and travel technology may affect the type and intensity of ties in a community.115 Technological progress tends to ease communication with distant actors, with individuals able to form connections on the basis of common interests rather than simply convenience.116 ‘Revolutionary’ changes in technology, such as the invention of the phone, the internet, or widespread air travel, has increased the
geographic reach of social networks over time.117 However, contemporary analysis finds that there is still a substantial positive effect for geographic proximity between scholars.118 Technological improvements may improve local social ties as well, with expanded
communication options found to contribute to deepening contact between close friends.119
114 Feld, 'Social ties'.
115 G. Boyce, 'Communicating infrastructures', in Boyce, Macintyre and Ville, ed., How organisations connect, Melbourne: Melbourne University Press, 2006
116 R. Kraut, M. Patterson, V. Lundmark, S. Kiesler, T. Mukophadhyay and W. Scherlis, 'Internet paradox: A social technology that reduces social involvement and psychological well-being?', American psychologist, 53, 9, 1998; H. Rheingold, The virtual community: Homesteading on the electronic frontier, Cambridge: MIT press, 2000.
117 Rheingold, The virtual community; I. Shklovski, S. Kiesler and R. Kraut, 'The internet and social interaction', Computers, phones, and the Internet, 2006.
118 F. Van Oort, Urban growth and innovation: Spatially bounded externalities in the Netherlands, Ashgate: Aldershot, 2004; Ponds, et al., 'Research collaboration'; C. Hussler and P. Ronde, 'The impact of cognitive communities on the diffusion of academic knowledge: Evidence from the networks of inventors of a French university', Research Policy, 36, 2, 2007.
119 C. Licoppe and Z. Smoreda, 'Are social networks technologically embedded?: How networks are changing today with changes in communication technology', Social Networks, 27, 4, 2005; C.
Haythornthwaite, 'Social networks and Internet connectivity effects', Information, Communication &
32
Reconciling these trends and Feld’s model of network formation, technological change may lead to the maintenance of pre-existing relationships. Improved technology may mean that two scholars can maintain a social tie in between periods of face-to-face contact.120 For this community, the time period of interest did not contain any
revolutionary changes in communication or travel technology.121 There were incremental improvements, with cheaper phone calls, more accessible overseas travel, and improved transport infrastructure between the main cities. These factors may have slightly
improved the ability of scholars to maintain geographically disparate links over time.
Social ties may thus form through a combination of internal cognitive motivations, or external contextual factors. The knowledge and social networks for the Australian
economic history community were intertwined, with both similar ideas and joint activities prompting collaboration between scholars.
3.3.1.1 Strong and weak ties
Social interactions are generally formed as the result of activities between scholars associated with a common focus. The constraint of that focus determines the type of interaction, with highly constrained foci producing dense networks with a higher
proportion of strong ties, and loosely constrained foci producing less-connected networks with weaker ties. The nature of ties in a network then affects the ease with which
individuals communicate and the type of knowledge that is produced.
Strong ties generally facilitate trust, norms, accountability and common values, improving co-ordinated action and increasing social capital. For intellectual networks, social capital exists through the ability of individuals to have access to the knowledge through
interactions with others. Social capital means that those with strong ties are more likely to exchange knowledge, and are more likely to seek out and offer help to others in the
network.122 Strong ties also mean that members of the group are more aware of the
Society, 8, 2, 2005; C. L. Coyle and H. Vaughn, 'Social networking: Communication revolution or evolution?', Bell Labs Technical Journal, 13, 2, 2008; Shklovski, et al., 'internet and social interaction'.
120 Kraut, et al., 'Internet paradox'; Coyle and Vaughn, 'Social networking'
121 Except for perhaps the widespread use of the fax machine.
122 R. Burt, Structural holes, Cambridge, MA: Harvard University Press, 1992; Burt, 'Social capital';
Coleman, 'Social capital'; J. Coleman, Foundations of social theory, Cambridge: Harvard University Press, 1990; N. Friedkin, 'A test of structural features of Granovetter's strength of weak ties theory', Social Networks, 2, 1, 1980; Hu and Racherla, 'Knowledge networks'; A. McWilliams, A. Lockett, J.
Katz and D. Van Fleet, 'Who is talking to whom? The network of intellectual influence in
management research', Journal of Applied Management and Entrepreneurship, 14, 2, 2009; Millar and Choi, 'Networks'; Nahapiet and Ghoshal, 'Social capital'; Nieves and Osorio, 'Role of social
33
‘relationship-specific heuristics’ that may affect knowledge diffusion.123 There is often a compatibility in language, theoretical background and methodology, meaning information is communicated more effectively by the source and understood more easily by the recipient.124 Strong ties are particularly beneficial for diffusing the tacit aspects of knowledge, and for recognising the value of new knowledge.125 However, a dense social structure may lead to ‘mechanisms of control’ that route communication inwards and exclude potentially innovative non-members.126 This can lead to informational inertia, as members of the community have access to similar contacts, information and ideas.127 Weak ties, while associated with lower trust and fewer common values, are argued to increase the diversity of knowledge.128 Weak ties mean there is a lower chance that an individual’s connections are also acquainted. This lower redundancy of ties means that weak ties lead to contacts who have diverse backgrounds and distinct ideas. Weak ties are seen as the source of ‘bridges’ between different domains of knowledge, as fewer
redundant ties means that it is more likely that a particular individual is the only path between two clusters.129 Innovative knowledge and interdisciplinary research is argued to emerge from the synthesis of ideas across these bridges.130 Weak ties can thus contribute to the diversity and overall knowledge output in intellectual communities.
As with the dichotomy between disciplinary and interdisciplinary knowledge, these network structures can be considered complementary rather than mutually exclusive. The
networks'; K. Siler, 'Citation choice and innovation in science studies', Scientometrics, 95, 1, 2013; O.
Sorenson, J. Rivkin and L. Fleming, 'Complexity, networks and knowledge flow', Research Policy, 35, 1, 2006; Whitley, Organisation of the sciences.
123 Coleman, 'Social capital'; Nieves and Osorio, 'Role of social networks'; B. Uzzi, 'Social structure and competition in interfirm networks: The paradox of embeddedness', Administrative Science Quarterly, 42, 1, 1997.
124 Coleman, 'Social capital'; R. Reagans and B. McEvily, 'Network structure and knowledge transfer:
The effects of cohesion and range', Administrative Science Quarterly, 48 2, 2003; Sorenson, et al., 'Complexity'.
125 Burt, 'Social capital'; Rost, 'Strength of strong ties'; Sorenson, et al., 'Complexity'.
126 Coleman, Social theory; Whitley, Organisation of the sciences; Frodeman and Mitcham, 'Interdisciplinarity'; Jacobs and Frickel, 'Interdisciplinarity'; J. Klein, Interdisciplinarity: History theory, and practice, Detroit: Wayne State University, 1990; Klein, Crossing boundaries.
127 Katz and Allen, 'NIH syndrome'; M. Granovetter, 'The strength of weak ties', American Journal of Sociology, 78, 6, 1973; Millar and Choi, 'Networks'.
128 Granovetter’s work in this area is seminal, Granovetter, 'Strength of weak ties'. See also Burt, 'Structural holes'; Harary, et al., Structural models; J. Podolny and J. Baron, 'Resources and
relationships: Social networks and mobility in the workplace', American Sociological Review, 62, 5, 1997; C. Wang, S. Rodan, M. Fruin and X. Xu, 'Knowledge networks, collaboration networks and exploratory innovation', Academy of Management Journal, 57, 2, 2014.
129 Granovetter, 'Strength of weak ties'.
130 Burt, Structural holes; Burt, 'Structural holes'; Granovetter, 'Strength of weak ties'; Reagans and McEvily, 'Network structure'.
34
strong ties paradigm emphasises the relational interpretation of social capital, highlighting solidarity and the ease of knowledge diffusion between scholars. The weak ties model emphasises the structural interpretation of social capital, advocating the benefits of diverse knowledge in an intellectual community.131 These two types of ties can co-exist happily in a network, with different types of knowledge requiring different connections.132 Complex knowledge with a large tacit component is best diffused through strong ties. Tacit knowledge is not easily codified, and so requires strong interpersonal connections to diffuse effectively. On the other hand, simple knowledge diffuses equally to those with strong or weak connections. An unaided search of published material adequately substitutes for any gaps made by an imperfect knowledge transfer.133
In consideration of this, some have recommended intellectual networks in which tightly-knit clusters are complemented by areas of sparse connections. Small-world research, first articulated by Stanley Milgram, advocates for teams with strong ties and a dense network structure in the local group, but with a large number of weak bridging ties to other clusters.134 The generation of new knowledge first requires a weak network structure, as the presence of structural holes gives the network access to ideas from different areas.
However, strong ties are crucial to recognising the value of these innovations, as they allow ideas to be tested and refined by members of the group. Weak networks thus
provide the ‘bridges’ over which new innovations travel, with final decision-making on the usefulness and value of these innovations made through the credibility strategies of tight-knit clusters.135 The social networks visualised in this thesis have various constraints, with some scholars involved in very large, loosely connected foci, and some in smaller and more densely-connected groups. The ties between scholars, and the knowledge produced in each group, reflected the size and constraint of these foci.
131 Rost, 'Strength of strong ties'.
132 Burt, 'Social capital'; M. Hansen, 'The search-transfer problem: The role of weak ties in sharing knowledge across organisational subunits', Administrative Science Quarterly, 44, 1, 1999; A.
Montanari and A. Saberi, 'The spread of innovations in social networks', Proceedings of the National Academy of Sciences of the United States of America, 107, 47, 2010; Nieves and Osorio, 'Role of social networks'; Reagans and McEvily, 'Network structure'.
133 Lynch, Art and artifact in laboratory science: A study of shop work and shop talk in a research laboratory, London: Routledge, 1985; Sorenson, et al., 'Complexity'.
134 S. Milgram, 'The small-world problem', Psychology Today, 1, 5, 1967.
134 S. Milgram, 'The small-world problem', Psychology Today, 1, 5, 1967.