These two dimensions of sociality have been named group
and grid. The exact nature of the two dimensions has
been the topic of many discussions, especially relating to methodological
consequences. The two dimensions should be seen as polythetic scales (Douglas,
1978, p. 15; Hampton, 1982; Gross and Rayner, 1985, chapter 4). That is:
they include a series of aspects, but those are not necessarily present
in each case observed.
A most detailed discussion of the dimensions can be found
in Measuring Culture in which
Steve Rayner and Jonathan Gross present a mathematical model for grid-group
analysis (Gross and Rayner, 1985). Their book is an attempt to formalise
and operationalise the two dimensions and it includes a computer program
to ‘calculate the values of grid/group predicates from experimental data’
(Gross and Rayner, 1985, p.117 ff).
group
Group stands for incorporation into a bounded group: it is strong
when the individual is a member of one corporate group, weak when individuals
do not belong to such a group (Douglas, 1978, p. 8, or 1982a, p. 192).
It gives the answer to the fundamental question Who am I?, or
more precisely Who am I with? (Wildavsky, 1987, p.6).
Gross and Rayner include five group predicates in their model (1985,
pp 73 ff):
-
proximity (closeness of members to each other),
-
transitivity (of relations),
-
frequency (proportion of the allocable time of members during which they
interact with other members of the unit), scope (activities with members
as proportion of all activities inside and outside the relevant social
unit), and
-
impermeability (which measures how easily eligible non-members who want
to join actually join the unit).
They also mention other predicates such as a commonality of experience
and life-support dependency.
grid
Grid stands for ‘the cross-hatch of rule to which individuals are subject
in the course of their interaction’ (Douglas, 1978, p. 8, or 1982a, p.
192).
It gives the answer to the fundamental question How should I behave?
(Wildavsky,
1987, p.6).
Gross and Rayner include four grid predicates in their model (1985,
pp 79 ff):
-
specialisation (operationalised as the proportion of all possible roles
that a member assumes),
-
asymmetry (lack of symmetry in role exchange),
-
entitlement (operationalised as the proportion of all roles that are ascribed
to certain categories of persons, as opposed to roles open to all that
can be attained by achievement), and
-
accountability (operationalised as the proportion of all roles that are
accountability interactions, which are interactions in which there is one
dominant role and one subordinate role, the former can put sanctions on
the later).