PSYCHOLOGICAL and sociological theorists now concur with custodial staff in stressing the negative influence of the inmate peer culture in correctional institutions. It is difficult to pinpoint the somewhat reserved origin of professional assent, but sociologists (e.g. Redl and Wineman, 1952;Miller, 1958;Polsky and Kohn, 1959) have pointed to the informal peer group culture as the primary antagonist to organizational goals for quite some time. Professional recognition of the controlling power of the peer group was at first based on informal observation of relatively subtle group transactions. Occasiorial coups d'etat provided dramatic emphasis to these observations. Theoretical expositions remained largely impressionistic and speculative. Recent advances (Buehler, Patterson, and Furniss, 1966;Patterson and Ebner, 1965; for a brief review) in social behavioural research, however, have provided impressive data indicating that, in a correctional institution, delinquent behaviour was rewarded by the peer group significantly more than socially conforming behaviour. Most of the rewards for delinquent behaviour occurred on the non-verbal level of communication. The staff also supported delinquent behaviour by indiscriminately rewarding and punishing it. Patterson, Littman, and Bricker (1967), in taking an additional step toward a theory of aggression, have provided empirical data depicting the effect of reinforcers on aggressive behaviour. In a nursery school setting they have graphically supported the hypothesis that the victim's reactions-e.g. compliance or counter-attack-are positively and negatively reinforcing and govern the rate of the aggressor's aggressiveassertive behaviour. Their speculations about the acquisition of aggressive-assertive behaviour are particularly apposite in thinking about how a delinquent youth gains and upholds the throne of dukedom.Correctional institutions, in coping with aggressive inmate behaviour, typically defer action until the behaviour is quite disruptive. The inmates test limits and the staff must contend with considerable aggressive-disruptive behaviour. When action is taken, or punishment administered, it is frequently carried out with no consideration of the contingency between the aggressive behaviour and the "consequence".Preliminary work (Buchard and Tyler, 1965;Tyler and Brown, 1967) suggests that systematic use of social isolation-time out from reinforcement-is effective in