This paper is an assessment of the attempts by Ackoff, Beer and Churchman to provide a justification for the claim that OR is a science. These writers tackle the problem using the philosophies of positivism, pragmatism and conventionalism/idealism in an attempt to build an ideal scientific methodology for OR. The conclusions reached are that no such methodology can exist and that philosophy confuses rather than clarifies the nature of OR as a science. The weakness of the philosophical approach to some of the problems faced by OR is demonstrated through a discussion of Ackoff and Emery's analysis 'On Purposeful Systems'. An outline is given of an alternative framework through which the problems of practising OR can more fruitfully be viewed.