We applaud David Crighton’s challenge to the idea that somehow clinicians are able to predict and prevent violence.1 However, we would like to make these additional points. The author divides uncertainty in risk assessment into epistemological uncertainty from a lack of knowledge and aleatory uncertainty, which is likened to the ‘throw of a dice, the toss of a coin’. In fact there is little uncertainty in the throw of a dice or the toss of a coin because with sufficient trials the result is certain. In his treatise ‘risk, uncertainty and profit’, Frank Knight used the term uncertainty to describe much less predictable events than those described in probability distributions and those resulting from ignorance.2 The subsequent development of chaos theory shows us that complex systems, particularly those with multiple feedback loops, exhibit a higher ‘uncertainty’ than the probability derived from the throw of a…