The ethical principles outlined apply to 'program' experiments, which involve alteration in the operation of the justice system, distinguished from simulation or laboratory experiments. The overriding theme of the committee's ethical guidance is that decisions about program experiments are linked to justice system administrators' decisions about how their institutions ought to function; therefore ethical considerations in experimentation must derive from the same ethics that guides all administrative decisions about institutional operations. Implicit in this central theme is the recommendation that when an experiment is needed to improve the justice system operation, not to experiment is wrong, so long as the effort is designed to provide the required information. Concomitantly, a program experiment should be undertaken only if it is needed. The committee justifies the use of random assignment research methodology on the grounds that reliable information may not be obtained by any other means. Further, informed consent to be involved in the experiment is not an ethical consideration unless the subject has a general moral privilege to preclude the researcher from performing the proposed task. The principles outlined are applied to a hypothetical police experiment. Five notes are provided.