Policy networks and the GM crops issue: assessing the utility of a dialectical model of policy networks
| Date | 01 June 2003 |
| Published date | 01 June 2003 |
| Author | David Toke,David Marsh |
| DOI | http://doi.org/10.1111/1467-9299.00344 |
Public Administration Vol. 81 No. 2, 2003 (229–251)
© Blackwell Publishing Ltd. 2003, 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street,
Malden, MA 02148, USA.
POLICY NETWORKS AND THE GM CROPS
ISSUE: ASSESSING THE UTILITY OF A
DIALECTICAL MODEL OF POLICY NETWORKS
DAVID TOKE AND DAVID MARSH
A dialectical model of policy networks is deployed to analyse policy change in the
area of GM crops in the UK. The model uses an analysis of the interaction between
agents and structure, network and context and network and outcomes to understand
and explain how policy change has occurred. A key advantage of the model is that it
increases understanding of network transformation, explanation of which has been
an alleged weakness of the policy network approach. However, this case study does
throw up some weaknesses with the model, including the tendency of the model to
emphasize the role of ‘insider’ agents and downplay the role of ‘outsiders’ in the policy
process.
The Genetically Modified (GM) food and crops case study is interesting for
policy analysts for a variety of reasons. First, it offers an example of a conflict
between powerful economic interests and the environmental lobby and,
what is more, one in which the normally dominant economic interests have
been under a great deal of pressure. Second, in the case of GM foods and
crops public opinion, and even ‘consumer power’, appear to have played a
significant role. Third, in this area, as in some others, the role of ‘scientific
experts’ has been increasingly questioned. As such, the issue raises very
important questions about the power of interests, experts and the public.
Our aim here is to examine the changes that have occurred in policy on
GM crops in the UK in the recent period. More specifically, we use the
dialectical model of policy networks, developed by Marsh and Smith, to
analyse those changes (2000). Of course, there is a lively debate in the policy
analysis literature on whether policy networks affect policy outcomes (see
particularly Dowding 1995 and 2000; Marsh and Smith 2000 and 2001). To
an extent, the different positions protagonists adopt in the debate reflect
their differing approaches to social science; in particular their different
epistemological positions (see Marsh and Smith 2001; Marsh and Furlong
2002). Not surprisingly, we share the Marsh and Smith position, rejecting
Dowding’s positivism, and acknowledging that there are different ways to
study political science and, thus, different ways to approach the role of
David Toke is a Research Fellow at the Department of Political Science and International Studies,
University of Birmingham and David Marsh is Professor of Politics and Head of the Department of
Sociology, University of Birmingham.
230 DAVID TOKE AND DAVID MARSH
© Blackwell Publishing Ltd. 2003
networks. More specifically, we adopt a realist position. There is no space
here to develop our position at length (see Marsh and Furlong 2002). Never-
theless, it is important briefly to explain our position.
Like positivists we do attempt to explain, but unlike positivists, we
believe that some relationships between social phenomena are not directly
observable and that, for example, the relationship between networks and
outcomes is mediated by actors’ understandings of them. When the realist
talks of explanation, however, they are appealing to a different notion of
explanation than that used by positivists. To a realist, not every relationship
between variables can be directly observed; one cannot merely generate
hypotheses from a theory which one then tests. However, the realist argues
that one can support the view that these unobservable relationships do exist.
We can observe other relationships between variables that, in a sense,
follow from the unobserved relationships. So, while we may not be able to
observe patriarchy or race, we can observe the consequences of these, for
example, the fact that so few key women are members of the key policy
networks. We cannot ‘prove’ the existence of patriarchy; clearly some obser-
vers might argue that men and women are essentially different (men are from
Mars and women are from Venus) and that men are best-fitted and have
more of the requisite skills to be in the network. To the realist it is their
theoretical position that underpins their argument that patriarchy is operat-
ing here; so the theorist utilizing a broadly feminist theoretical position
explains the empirical observation, the unequal access of women to key
policy networks, as a reflection of patriarchy.
The realist also recognizes the double hermeneutic. So, we stress that any
explanation of behaviour has to take into account how the actor understands
his or her behaviour recognizing that actors interpret the contexts in which
they operate (the first level of the hermeneutic). At the same time, as obser-
vers, our understanding of what we observe is partial; crucially it is mediated
(a) through the lens of our theory, and (b) that the observers interpret the
actions of agents (the second level of the hermeneutic). Hence, any explanation
we offer is theory-dependent.
One big advantage a realist has is that they can offer work which is inter-
esting to both positivists and interpretivists. Realists do attempt to explain
and a positivist can assess the explanation in their own terms. However, the
realist also acknowledges the importance of the double hermeneutic. As
such, interpretivists can relate to that aspect of the work of realists, while
rejecting the aim of developing explanations.
Raab (2001) argues that the dialectical approach takes us no further than
existing network theory. In contrast, Dowding (1995, 2000) contends that the
policy network approach lacks a theoretical basis. In particular, he argues
(1995) that ‘the major problem’ is that ‘whilst the (network) metaphors are
heuristically useful (. ..) they are incapable of explaining network transform-
ation’ (Dowding, 1995, p. 139). Our argument against both these critics is
that the dialectical approach gives us a more developed conceptual framework
To continue reading
Request your trial