Source: (2002) Theoretical Criminology. 6(4): 411-431.Increasingly, criminologists have made reference to Mary Douglas’s writings on risk. Less frequently is it noted that Douglas’s work on risk is dependent upon a more fundamental classificatory scheme, what she calls grid-group theory. This theory measures cultural biases in terms of their level of prescriptiveness of norms (grid) and affiliation between individuals (group). This theory gives four distinctive cultural types, each of which has distinctive attitudes to risk, crime and blame. These types reflect the institutional relations of a particular collectivity within a society. This paper maps contemporary debates on the transformation of punishments onto grid-group theory and argues that criminological theorizing has effectively restricted itself to only two of Douglas’s four cultural types. Expanding the parameters of the debate would not only allow a more accurate representation of what is going on, but may also tentatively sketch a map toward a more optimistic future. Includes a discussion of restorative justice and paradigm shift. (author's abstract).