CRIMINOLOGISTS AGREE ON THE LACK OF DEMONSTRATED EFFECTIVENESS IN OFFENDER REHABILITATION PROGRAMS, AND EVEN A PANEL FROM THE NATIONAL RESEARCH COUNCIL, IN A REPORT ON DETERRENCE AND INCAPACITATION, ARGUES THAT AVAILABLE RESEARCH CONCERNING THE IMPACT OF VARIOUS TREATMENT STRATEGIES ON OFFENDERS SEEMS TO INDICATE THAT THERE ARE NO STATISTICALLY SIGNIFICANT DIFFERENCES BETWEEN THE RECIDIVISM OF TREATMENT AND NONTREATMENT GROUPS. OF METHODS FOR CHALLENGING REHABILITATION APPROACHES, CONTAMINATING THE TREATMENT IS MOST UNIVERSALLY APPLICABLE; I.E.,, RAISING PLAUSIBLE ALTERNATIVE INTERPRETATIONS FOR THE REPORTED EFFECTS, SUCH AS SUGGESTING THAT A STUDY'S RESULTS ARE DUE TO SOME TREATMENT OTHER THAN THAT CLAIMED BY THE AUTHORS. ANOTHER APPROACH, STRESSING CRITERION PROBLEMS, FOCUSES ON MEASUREMENT ISSUES. SINCE MANY TREATMENT STRATEGIES AIM TO REDUCE THE PROBABILITY OF RECIDIVISM, THEY REQUIRE AS A CRITERION MEASURE SOME INDICATOR OF ILLEGAL BEHAVIOR. IT IS IMPOSSIBLE TO MEASURE SUCH CONDUCT ACCURATELY, AND EVERY STUDY IS VULNERABLE TO ATTACK ON THE BASIS OF THE CRITERION PROBLEM. STILL ANOTHER APPROACH INVOLVES SHOWING THAT MASSIVE EFFORTS HAVE FAILED. USING TIME OR MONEY AS UNITS OF ANALYSIS, CRITICS APPEAL TO COMMON SENSE BY ARGUING THAT NO FUTURE EXPERIMENTATION IS WARRANTED BECAUSE PAST MASSIVE INVESTMENTS IN PROGRAMS AND PERSONNEL HAVE FAILED TO PRODUCE UNAMBIGUOUS DEMONSTRATIONS OF EFFECTIVENESS. PROGRAMS CAN ALSO BE CHALLENGED ON THE GROUNDS THAT THEIR TREATMENT STRATEGIES ARE BASED ON FAULTY THEORY. IF IT CAN BE DEMONSTRATED THAT THE ASSUMPTIONS UPON WHICH REHABILITATION IS BUILT ARE FALSE, THEN THE EMPIRICAL CLAIMS OF ALL TREATMENT PROGRAMS MUST ALSO BE FALSE. THE FINAL AND BEST KNOWN DESTRUCTIVE TECHNIQUE INVOLVES SEEKING UNIVERSALS; THAT IS, TO RAISE SERIOUS DOUBTS ABOUT A PROGRAM IT IS NECESSARY ONLY TO SHOW EITHER THAT ALTHOUGH THE TREATMENT METHOD HAS BEEN FOUND TO WORK WITH SOME OFFENDERS, IT IS INEFFECTIVE WITH OTHERS, OR, MORE SIMPLY, THAT IT HAS NOT EVEN BEEN TRIED ON ANYONE. REFERENCES ARE PROVIDED. (KBL)