Sub Groups

The Research Approach of CSEC, Lancaster
A note by Brian Wynne, Jan 7th 2001

Our approach to understanding what shapes public responses to technologies like GMOs is of course a collective enterprise. Thus it is always discursively evolving and not at all rigidly unified. Nevertheless the following features might be regarded as a common set of theoretical, methodological and 'political' principles which guide our work. This approach has been developed since the 1970s through work on public attitudes in several different fields like nuclear power, chemical pesticides, major-hazard industrial plants, chemical and nuclear wastes, as well as - more recently - GMOs. For reasons outlined below our 'public attitudes' work has also usually been accompanied by work analysing the corresponding institutional constructions of risk, risk issues and public perceptions. It has also been informed by a fairly close and active involvement of its senior researchers in policy debates, policy analytic processes and advice - reflected in the origins of the present meeting. This policy involvement is a deliberate part of our research-learning model, and not just seen as post-research dissemination.

Philosophically-speaking we start from a decidedly non-positivistic stance towards social science and its possibilities. We are interested in meanings, and we are oriented towards cultural processes, which requires us to be interested in tacit, unconscious processes as well as explicit reasoned ones.

It has been recognised for a long time - since at least the 1920s - that in principle, natural science observation involves interference with the object(s) being studied (though of course in many situations the energy transfers involved in the act of observation are so relatively small as to mean there is no significant disruption of the object). In social science this is true with bells on. The questions we ask, and the way we ask them, inevitably affect the 'answers' we receive from human actors. Moreover social research involves human actors who are actively constructing meanings for themselves. We can connect this with attitudes by noting that the assumed or identified object of a human subject's attitude indicates the meaning of that issue for that human subject. Thus as researchers interested in the formation and expression of public attitudes we cannot presume to know a priori what the 'object' of any person's attitude is - their meanings may legitimately differ from those of policy experts, or researchers. That is, as researchers we should not presume what the meaning of public issues is to members of the public - this must first itself be researched. For example in daily public policy discourse and in some social science research it is presumed that the object of public attitudes is 'risk', and what is more, 'risk' as defined by particular scientific institutions or disciplinary cultures. Hence the misleading language is used, of these issues as 'risk issues'. 'Risk' (as defined in a particular way) is being inserted and imposed as the presumed object of attitudes, or the presumed meaning of 'the' issue, without first open-endedly exploring whether this is the case. Thus we are interested in using methods which allow this relatively open-ended exploration of autonomous framings or meanings, as a frequently neglected yet crucial first stage in public attitudes research which might enlighten policy.

This point raises the further one that we should not presume to know what it is that publics are responding to, when we try to understand these responses. Although this might sound an obvious point, it is surprising and disappointing to see how often it is ignored in practice. Thus public attitudes and responses might be to:

• Risk magnitudes
  
• Risk-qualities (eg voluntariness, distribution, trend, etc as in psychometrics 'attributes')
  
• The technology as a whole social experience and projection
  
• Institutional (mis)management of those risks,
  
• Dominant institutional definitions of the issue as imposed in official approaches (eg neglect of dimensions and variables which are salient to the public - autocratic imposition of meanings)
  
• Dominant definitions of the public (eg as ignorant, prone to hysteria, instrumental only, individualistic) implicit in expert discourses of the issues (and maybe in some research approaches?)
  (These of course are not mutually exclusive options)
  This underlines the reasons for our commitment to understanding public attitudes in relation to expert-institutional behaviour, including how they construct public knowledge. It is artificial to take such attitude formation processes out of context, even if in practice this always calls for compromises.

- Our research is necessarily interpretive. This is true of all research on public attitudes, and the key question is where, and how explicitly and reflexively, are such interpretive commitments made, and how are they to be critically evaluated by peers and by those they aspire to represent;

- Much of the valuable understanding comes from addressing what is not being said as well as what is being said;

- Observation methods for this approach are best which allow interactive expression of meanings and attitudes. Thus discussion-groups have a particular value, though like any other method they cannot claim to provide unique

- Validation and amendment of our interpretive efforts has to be from peers and as far as possible also from the public which these representations claim to stand for. It cannot claim to come from data alone, and any research method which does make such a claim must ipso-facto be suspect;

- Our efforts to construct authentically-representative interpretations of public attitudes and their shaping factors cannot be validated just by measuring them against supposedly 'true, objective' public attitudes as one might hope for these to be expressed in public-domain validation. In addition to this kind of validation there is an irreducible element of interference by the 'research finding' on the attitude it is supposed to be representing and testing itself against. This cannot be avoided, and our research should acknowledge this role and the responsibilities which come with it.