Risk assessment in offenders with mental disorders: relative efficacy of personal demographic, criminal history, and clinical variables

Following the meta-analysis by Bonta, Law, and Hanson, (1998) this study examined the ability of personal demographic, criminal history, and clinical variables to predict reoffending in offenders in the United Kingdom who had mental disorders. The efficacy of each variable in predicting rate of general reoffending and violent reoffending was investigated. Age on admission, number of days hospitalized, and number of previous offenses were the most effective variables in predicting re-offending, with number of previous offenses being the strongest predictor. Clinical diagnosis was not predictive of reoffending when the variance attributable to these other predictors was controlled for. None of the variables were able to discriminate between general offenders and violent offenders indicating that the same variables predict both types of reoffending. The results showed that reconviction in offenders with mental disorders can be predicted using the same criminogenic variables that are predictive in offenders without mental disorders.     

Soldiers as Public Officials: A Moral Justification for Combatant Immunity

How can we make moral sense of the international humanitarian law doctrine of combatant immunity? The doctrine is morally shocking to many: It holds soldiers on both sides of a war immune from criminal prosecution for their otherwise criminal acts of killing, maiming, destroying property, etc., carried out as part of their country's war effort. That is, soldiers who kill as part of an attack benefit from the immunity just as much as those defending their country. Traditionally, just war theorists have tried to provide situation‐specific arguments to show that soldiers on both sides had a good moral justification for their actions. Recently, self‐styled “revisionist just war theorists” have suggested that the doctrine of combatant immunity is just a convention designed to minimize harm. In this article, I suggest that the moral foundation of the doctrine lies in the status of soldiers as public officials in the service of their country. The reason why we hold them immune from prosecution for their war‐making acts is that such acts are properly thought of as acts of a state, rather than as acts of a particular individual. And the reason why states are immune from prosecution for their acts is one of moral standing: No other state has the moral standing to tell another how to carry out the matters that define its jurisdiction. So as long as a country deems (however implausibly) that it must use force to defend itself from aggression, then it may do what is required to defend itself. No other state has the standing to prohibit such acts or to punish those who carry them out. This argument is rooted in an understanding of how individuals may interact as free and equal under law. It does not aim at the perfection of human action, but it does serve to eliminate the worst forms of tyranny.     

Comparison of UK and China's nuclear force developments

Introduction In many ways, the UK and China could not be more different.Not only are they geographically on opposite sides of the world.They also have profoundly different recent histories. The central roles played by collective security and liberalised trade in the post-1945 order have made relative power less important than it was in previous eras.And the UK has taken full advantage of these opportunities, building new partnerships to ensure its security and prosperity despite the decline in its relative status.     

A Social Democratic Future? Political and Institutional Hurdles in Scotland

The Nordic model has long been admired in Scotland, and has featured prominently in aspects of the Scottish independence referendum debate. This article explores the difficulties in instituting a similar system here, identifying two significant barriers: the institutional setting (the powers available to Scottish politicians) and the partisan nature of competition between the two parties that might be able to deliver upon such a commitment. It concludes that the prospects of moving towards a Nordic‐style social investment model are slight, given the political, institutional and attitudinal barriers in place.     

Thoughts on Juvenile Justice Systems and Research

This article updates comments published in this journal in 1994 about the nature of the American juvenile justice system, which laid out reasons that it might not serve as a useful model for other nations. Since that time, the US system has moved further right towards the justice model and away from the welfare model. Individualistic philosophies and political conservatism have combined to produce a more adult-like and punitive juvenile system, applied to increasing numbers of minors and to increasingly younger minors. Successful demonstrations of community absorption or treatment have been too few to balance the rightward direction. European scholars are urged to undertake increased studies in two critical areas: (1) the nature, functions, and comparative differences in their juvenile justice systems, and (2) the nature of local communities and their contributions to patterns of delinquency.     

A Comparison of Logistic Regression,Classification and Regression Tree,and Neural Networks Models in Predicting Violent Re-Offending

Previous studies that have compared logistic regression (LR), classification and regression tree (CART), and neural networks (NNs) models for their predictive validity have shown inconsistent results in demonstrating superiority of any one model. The three models were tested in a prospective sample of 1225 UK male prisoners followed up for a mean of 3.31 years after release. Items in a widely-used risk assessment instrument (the Historical, Clinical, Risk Management-20, or HCR-20) were used as predictors and violent reconvictions as outcome. Multi-validation procedure was used to reduce sampling error in reporting the predictive accuracy. The low base rate was controlled by using different measures in the three models to minimize prediction error and achieve a more balanced classification. Overall accuracy of the three models varied between 0.59 and 0.67, with an overall AUC range of 0.65–0.72. Although the performance of NNs was slightly better than that of LR and CART models, it did not demonstrate a significant improvement.