Criminal Record Rides Again: Revisiting a 50-year-old Problem

Organised by Dr Gabrielle Watson, Chancellor’s Fellow and Director, Empirical Legal Research Network, University of Edinburgh.

Professor Julian V Roberts KC (Hon) from the University of Oxford will present his paper ‘Criminal Record Rides Again: Revisiting a 50-year-old Problem‘ in person at the University of Edinburgh.

Of all sentencing factors, criminal history affects the greatest proportion of offenders. Around 4/5 of defendants in England and Wales have previous convictions. How should a sentencing court treat the presence or absence of prior crimes? This simple question turns out to be very complex at the levels of both theory and practice.

Utilitarians argue that prior crimes increase the risk of offending and should aggravate sentence. Sentence severity should rise in direct relation to the extent of the offender’s criminal record. Retributive theorists vary greatly in their reactions. Fletcher argued that prior crimes should play no role as they are unrelated to harm or culpability, the twin elements of proportional sentencing. Von Hirsch advocated a Progressive Loss of Mitigation model: first offenders receive mitigation which is progressively withdrawn following subsequent convictions. Once the mitigation is lost, no further increase in severity is imposed.

Other retributivists deem recidivists more culpable and deserving of greater punishment. Most recently, scholars have argued that prior crimes should serve as a ground for mitigating sentence for the recidivist. Sentencing regimes also vary greatly in the way they treat criminal history. Across the US, prior crimes carry great weight in aggravation. In some states, prior crimes are more important than the current conviction in determining sentencing outcomes.

Courts in England and Wales follow the Progressive Loss of Mitigation model. In this presentation, I draw upon 30 years of scholarship to critically review retributive and preventive approaches to the problem. I ultimately oppose a unitary model and advocate a more nuanced and pluralistic one which more closely reflects recent research in the area of desistance.