It is proposed that some of the difficulties of the research-design problem are only apparent - are in fact a result of the paradigm in which the research-design relationship is too often posed. Within this paradigm, it is conventionally presumed that the result of environmental research is an increase in objective knowledge about the environment which then needs to be incorporated into the process of designing, replacing or reducing the intuitive component and thus supposedly improving the outcome of that process. Far from trying to eliminate 'subjectivity' in the form of the preconceptions of researchers or designers, we need in fact to recognise that the intelligibility of the results of both scientific research and the design process are dependent on the preconceptions and prestructurings of the problem in the minds of researchers and designers. The cognitive schemes by which we interpret the world and prestructure our observations are in fact an integral part of the field of science and need to be seen as an essential part of the subject matter of science. (Hillier et al 1972)
The idea of a science which produces factual knowledge or information which can be assimilated into design, along with a view of design which proceeds by decomposing a problem into its elements, adding the information derived from scientific work, and then synthesising (inducing) a solution by means of a set of logical or procedural rules, therefore needs explicitly to be replaced by a view that has it that both science and design operate reflexively between prestucturings of problems (or more generally of parts of the world), and those parts of the world as they show themselves to be when examined in those terms.
The question then of what the purpose is of research in the design field, and in the field of the built environment, is simply answered: it is to affect the preconceptions and prestructurings of the shape and nature of the built environment in the minds of the designers themselves, so that they may approach problems with better, more complete and more workable preconceptions and prestructurings. These prestructurings are in the form of more or less general models that can structure an understanding of real world relations and dynamics and an approach to particular cases.