The deterrent impact of Ontario's 12-hour licence suspension law, a law intended to mete out swift punishment is evaluated. A process model of deterrence is included. Time series analyses of the monthly log odds of number of positive B.A.C. fatalities over the number of negative B.A.C. fatalities indicated a small, short-term effect. The intermediate measures of four surveys of media coverage, Toronto residents, Ontario residents, and police forces supported these results, and affirmed the potential validity of the process model of deterrence. The results suggest that laws to increase the celerity and certainty of punishment will have little deterrent impact without enforcement and publicity of the new laws.