**3.4 'Social networks' and 'Quasi-groups'**

Network analysis draws out the everyday point that one way of locating yourself in relation to other people is, not just in terms of what characteristics you have (e.g. gender, age), but 'who you know', or more generally what sort of people you associate with. Although others have used this term in different ways (notably Dahrendorf 1968), I portray networks as 'quasi-groups': that is, as a form of social organisation that links people but which need not be as formally organised and clearly bounded as 'proper' groups are.

The root metaphor in this approach is that of webs and graphs. Fischer (cited in Wilson, 1983: 54) puts it well:

'Society affects us largely through tugs on the strands of our networks - shaping our attitudes, providing opportunities, making demands on us, and so forth. And it is by




Similarly, Merton has suggested cognate mechanisms they may handle stress in status-sets



A major sociological theme has been that stress arises from awkward combinations of statuses that a person holds. Lenski introduced the notion of 'status inconsistency' which hypothesised that those people occupying 'incongruent' status-sets might suffer increased social stress - or that there might be other consequences that flow from their 'cross-pressure' situation. There are a variety of effects which might follow from 'minority' or 'unusual'

Rose Coser (1991) has moved beyond this stress or conflict view to emphasise the positive opportunities opened up by more complex status-sets. She argues that it is within the very interstices opened-up by complex status-sets that wider degrees of individual freedom can come to be realised. One aspect of this is that people learn more sophisticated social skills including linguistic flexibility - as they learn to handle role complexity. It may also be that more energy is generated as a result of the interplay between statuses. There are also

Network analysis draws out the everyday point that one way of locating yourself in relation to other people is, not just in terms of what characteristics you have (e.g. gender, age), but 'who you know', or more generally what sort of people you associate with. Although others have used this term in different ways (notably Dahrendorf 1968), I portray networks as 'quasi-groups': that is, as a form of social organisation that links people but which need not

The root metaphor in this approach is that of webs and graphs. Fischer (cited in Wilson,

'Society affects us largely through tugs on the strands of our networks - shaping our attitudes, providing opportunities, making demands on us, and so forth. And it is by

be as formally organised and clearly bounded as 'proper' groups are.



of unawareness of the extent to which values are in fact shared);


earlier-held statuses and those pertaining in later statuses; - self-selection of statuses which are 'neutral' to one another.

relationships are central and others peripheral); - differences in power amongst those involved in a role-set;

coping with an un-integrated role-set;

degree recognised to have families);

possibilities for integration and for innovation.

**3.4 'Social networks' and 'Quasi-groups'** 

(Crothers 1987:94):

situations.

1983: 54) puts it well:

tugging at those same strands that we make our individual impact on society - influencing other people's opinions, obtaining favours from 'insiders', forming action groups'.

Another, more aggregated, way of conceptualising network linkages is in terms of Bourdieu's concept of 'social capital' (which has also been picked up network analysts such as Coleman 1990: see also Bourdieu, & Wacquant, 1992, Lin, 2001). Social capital is seen by Bourdieu as, in effect, the 'linkage reach' of people, and especially the extent to which they can convert other forms of capital into effective use.

One strength of network approaches is that they detect patterns of social life operating beneath and around more formal structures. For example, working class residential communities may not be studded with links through formal organisations, and, therefore, may appear to the casual observer to be devoid of social structure. Whereas, in fact, they may be quite tightly interlaced by informal social links. Another strength is that network analysis can probe behind surface patterns of links to show indirect paths of contact, mediated through other people or collective units. Yet another emphasis in network analysis is on actual, concrete links between actual units, rather than rather more vague pictures of expectations and possibilities, which is where role analysis often leaves matters.

Network analysts vary in the vigour and exclusiveness of their stance: the most radical denigrate any attention to people's opinions and views, seeing these as emanations of their network position. The **form** of relationships is often stressed over their **content.**

A very important distinction is that between 'network cohesion' and 'structural equivalence'. The two ideas posit quite different ways of examining nodes and their linkages. The network cohesion concept links those who interact with each other: for example, in a medical centre each set of patients, receptionists, practise nurses and doctor form a network based around each particular doctor. However, each of these four types of position are the basis for network links based on the 'structural equivalence' of the people concerned. That is, each plays an equivalent role in 'their' network, and analysis can be built around this similarity. Often these positions are, in fact, also socially prescribed status-roles, but they need not be. Nodes can occupy 'structurally equivalent' positions without this being formally recognised by the culture.

One key idea is the importance of 'weak ties'. As opposed to the 'strong ties' which bind groups together, the much more extended range of 'friends of friends' may be particularly important on some matters. (Network analysis incorporates nodes connected by strong ties, too, but is particularly effective in picking up the looser and lighter web of more extended linkages.) In several studies of how people obtain services (e.g. an abortionist, a job) it has been found that weak ties have been more effective than strong ties. This is because only a limited stock of information circulates within a closed group, whereas the surveillance range of a whole slew of weak ties is far wider. Thus, more widely-flung contacts are likely to hold a much greater stock of information, even if this web of weak ties is not very systematic or efficient in passing that information on.

Another key idea is that of indirect 'connectivity'. Formally separated social units may in fact be coordinated or controlled behind the scenes by a web of interconnections. Indeed, analysts of the economic power elite which is considered to run the business world have developed a variety of models of how interconnectedness is achieved behind the backs of

Analysing Social Structures 25

collective organisations, their purposes in life are bent to the wishes of these structures. The intense web of face-to-face social linkages that formerly pertained is now reduced, and subject to severe intrusion from collective persons: e.g. schools, advertisers. The relation between collective entities and natural persons is asymmetrical. Organisations are obtrusive and intrusive, and difficult to gain information about or to control. Perhaps the final irony is, that to obtain some leverage over corporations, natural people may resort to agencies such as the state or to trade unions: but these too can be very distant from and unresponsive

There has been much discussion across many areas of sociology about how people loosely aggregated within social categories may become more tightly welded into collectivities or organisations. The classic discussion was that of Marx concerning the revolutionary consciousness of the working class. To enable collective revolutionary action, the working


Merton's views are more general (Crothers 1987: 97, Merton 1968). He distinguishes between categories, collectivities and groups. Members of categories share statuses, and thereby similar interests and values although not necessarily through shared interaction or a common and distinctive body of norms. Collectivities share norms and have a sense of solidarity, while members of groups interact with each other and share a common identity, which is also attributed to them by others. But he does not then go on to provide sociological explanation of how groupings might move up (or for that matter down) this

Each organisation is in some part unique, but also shares similarities in its attributes with other organisations. They interact with other organisations and can bunch together to form further, higher-level (meta-) organisations. They persist, they change, they are born, they die. However, the metaphor does not carry over exactly, as unlike people, organisations can have major bits break off, or be added to, and can interact with people as well as other collectivities. A further, and central, discontinuity with this individualistic analogy is that organisations tend to be multi-layered. Any organisation can be a veritable 'Russian doll' of nested sub-organisations, and there can also be layers of people who are affected beyond the usual organisational boundaries. Social patterns can also crosscut the layers and boundaries

In analysing an organisation, the major independent variables are the formal institutions in terms of which social conduct is organised: the division of labour, the hierarchy of offices, control and sanctioning mechanisms, production methods, official rules and regulations, personnel practises and so on. The major dependent variables are the results accomplished by operations and the attachment of its members to the organisation, as indicated by productive efficiency, changes effected in the community (say, a decline in crime rates), turnover, satisfaction with work, and various other effect criteria. To explain the relationship between these two sets of abstract variables, it is necessary to investigate the processes of social interaction and the interpersonal

relations and group structures (Blau cited in Calhoun 1990:17).


to citizens' or members' wishes.


class requires:

hierarchy of levels.

of organisations.

markets which are apparently populated by a host of independent businesses. It has been shown that there are:


Such links can be measured and their patterns modelled.

Another important idea is that of 'structural balance'. From examining triads of relations among three people (or nodes) it can be readily seen that some triads are balanced whereas others are unbalanced. For example, if A is dominant over B and B dominant over C, the triad is balanced, if then A is dominant over C. Indeed, one might expect this to occur naturally anyway, although empirically there are exceptions which are unbalanced. This type of analysis is interesting in providing predictions about the longer-term stability of groups, based on the characteristics of their constituent triads.

'Structural holes' (Burt 1992) are the gaps in a network pattern, and they provide entrepreneurial opportunities for those in the existing pattern to move into to exploit. This is part of a sociological contribution to understanding the links between firms in markets, although such structural holes can occur in a wide variety of social structures.
