Patrick McGowan has recently attempted to examine the relationship between ‘Economic Dependence and Economic Performance in Black Africa’ in this Journal, Vol. xiv, No. I, March 1976, pp. 25–40. His article reemphasised the need for concrete analysis since the generalities of ‘dependence’ have been more extensively studied than their concrete expressions. However, a fundamental problem of his analysis is that the criteria for verifying the theory of dependence are unrelated to the theory itself: the ‘test’ devised is a series of correlations between measures of dependence and indicators of economic performance, since ‘the theory [of dependence] predicts that dependence is negatively associated with indicators of economic growth and development’ (p. 27). Part of the problem is that McGowan does not really define ‘dependence’, but in addition it is not at all clear why this is expected to be negatively associated with indicators of economic growth.