In recent criminological discourse, considerable emphasis has been placed on initiatives for deinstitutionalization, deformalization and liberalization of the regime of criminal justice institutions and policies. This initiative, which is largely contemporary, follows an earlier trend, which insisted on segregation of "deviants", on the centralization of the criminal justice apparatus, and on a certain "cold realism" regarding the real possibilities for inducing positive behavioural changes in the populations of "deviants" by restrictive and particularly penal measures. This older view thus implied that, although punishment and other restrictive practices did not in fact "reform" offenders, or encourage them in any way to become more legitimate members of society, it was the best and safest way for handling criminal deviance society could possibly come up with. These two paradigms, or views, are sometimes labelled "the first" and "the second correctional change". The first correctional change characterized the nineteen and the beginning of the twentieth century. This is the older view mentioned above. The second correctional change, insisting on liberalization and deformalization, is a product of this day. Apart from presenting the two theoretical paradigms and some elements of the background of labelling them "correctional changes", this paper attempts to show that the real constructive potential of the "second correctional change" depends on traditional values and virtues nurtured in the informal community, and that it does not appear to stand in any positive correlation with contemporary values and organization of the system. The reason for this is that, even if institutions could be significantly reduced, which is far from being uncontroversially acceptable, the real impact of informal procedures in place of formal ones will depend on the status of inter-relationships of trust in the informal community, namely the trust between the two sides of penal policy, including the consensual majority of the democratic public on the one, and the dissensual minority of "deviants", on the other hand. The paper argues that the de facto level of this trust is unsatisfactory, given that in liberal social arrangements interest is what governs social policy, and the interests of the majority of the consensual public stand largely opposed to the interests of the dissensual minority. This means that there is likely to be a great deal of reluctance on the two sides of the "control divide", and particularly on the two sides of the penal divide (the convicted and penalized offenders and the administrators of criminal justice) to invest trust in the belief that this structural opposition which leads to confrontation and "the control effort" can be eliminated and informal mechanisms of overriding cooperation and optimization put in its place. In any particular case, with any particular society, the paper argues, the success of informal strategics which many abolitionists are proposing these days, within the paradigm of "second correctional change", always depends on how vital the older, informal virtues and values are within the institutional texture of the system, and how easily and quickly they can be brought back into operation. This question, the paper concludes, is to a considerable extent independent of the question of the virtues and vices of the system itself.