In a 2004 paper, Greene and Cohen predicted that neuroscience would revolutionise criminal justice by presenting a mechanistic view of human agency that would change people’s intuitions about retributive punishment. According to their theory, this change in intuitions would in turn lead to the demise of retributivism within criminal justice systems. Their influential paper has been challenged, most notably by Morse, who has argued that it is unlikely that there will be major changes to criminal justice systems in response to neuroscience. In this paper we commence a tentative empirical enquiry into the claims of these theorists, focusing on Australian criminal justice. Our analysis of Australian cases is not supportive of claims about the demise of retributive justice, and instead suggests the possibility that neuroscience may be used by the courts to calibrate retributive desert. It is thus more consistent with the predictive claims of Morse than of Greene and Cohen. We also consider evidence derived from interviews with judges, and this leads us to consider the possibility of a backlash against evidence of brain impairment. Finally we note that change in penal aims may be occurring that is unrelated to developments in neuroscience.