255 The complexity of evidence notwithstanding: A reply to Swartz Ashraf Kagee Department of Psychology, Stellenbosch University, Private Bag X1, Matieland, 7602, South Africa skagee@sun.ac.za This article is a response to the caveats raised by Swartz (this issue) regarding the complexi- ties of emphasising evidence-based interventions in psychology. It is acknowledged that the process of amassing evidence to support psychological interventions requires overcoming considerable logistical and contextual difficulties. Yet, the dangers of not taking into account the vast body of data on the effectiveness of psychotherapy are likely to have serious conse- quences for consumers of psychological service and for the profession of psychology itself. The challenge for South African psychology is to address the complexities raised by Swartz in ways that minimise leaps of faith regarding the effectiveness of interventions so that thera- peutic decisions are based on sound data rather than unsubstantiated theoretical conjecture. Professor Leslie Swartz’s ‘The complexity of evidence or the evidence of complexity?’ (Swartz, 2006), is an elegant and nuanced appraisal of the evidence-based movement as it applies to psychology. It is apparent that he and I share similar concerns about some aspects of the practice of psychology in South Africa. There is a vast body of research in clinical psychology, as can be seen from the several million cited references in the PsychInfo database (PsychInfo, 2006). Yet, it appears that activities such as training clinicians and providing treatment to clients do not always take into account this body of knowledge. I agree with Swartz that we do not always know how we know. My concern remains, however, that if vague epistemology goes unchecked and if we adopt a laissez-faire attitude towards what constitutes psychological knowledge, it may be a slippery slope to South African psychology journals joining the ranks of Odyssey and other similar publications. Such a state of affairs will surely be a source of consternation for many in the profession. Epistemological criteria that require theoretical conjecture to be open to refutation may serve to prevent such intellectual decay. © Psychological Society of South Africa All rights reserved South African Journal of Psychology, 36(2) 2006, pp 255–258 ISSN 0081-2463