This paper develops a framework for architectural research of an applied nature, that is, research that will generate 'operational knowledge' in relation to architecture's principal identifying characteristic, and that element which rests in the hands of the designer, namely the organisation of space. A theoretical framework is developed which insists upon a clear distinction between the characteristics of the space and the 'meanings' constructed upon the spatial object by 'consumers'. The nature of 'conflict' is defined, as that noncorrespondence within the system comprising a society and its architectural activity which propels the evolution of new spatial types forward over real time. Architectural objects are conceived as complex products of three categories of relationships; functional, compositional, and constructive. On the basis of those categories are defined architectural types, activity types, and 'applied' types, each of which may exist in conflict-bearing or canonical forms. The framework for an architectural science is then constructed which distinguishes its object, its subject matter, and its method. Measurements are identified as the fundamental elements of operational knowledge, and they are distinguished as being of two kinds -- 'parameters' of the object itself, and 'properties' describing its functioning in some human context.A methodology of applied architectural research is presented which comprises alternating deductive and inductive cycles. Each has three component operations, which are discussed in some detail. The discussion of parameters and properties leads to the distinguishing of three classes of object, the empirical (actual), the hypothetical (possible), and the type (the required), and a corresponding array of functional, compositional, and constructive models.
