(Neuro)predictions, Dangerousness, and Retributivism

Through the criminal justice system so-called dangerous offenders are, besides the offence that they are being convicted of and sentenced to, also punished for acts that they have not done but that they are believe to be likely to commit in the future. The aim of this paper is to critically discuss whether some adherents of retributivism give a plausible rationale for punishing offenders more harshly if they, all else being equal, by means of predictions are believed to be more dangerous than other offenders. While consequentialism has no problem, at least in principle, with this use of predictions most retributivists have been opponents of punishing offenders on the basis of predictions. How can an offender deserve to be punished for something that he has not done? But some retributivists like Anthony Duff and Stephen Morse have argued in favor of punishing offenders who are considered to be dangerous in the future more harshly than non-dangerous offenders. After having reconstructed their arguments in detail, it will be argued that both Duff's and Morse's attempts to give a retributivistic justification have several shortcomings.