The survey provided respondents with hypothetical stories, and they were asked to assess responsibility and assign punishment for the parties in the stories. The initial survey in Detroit was a 1977 probability sample of the Detroit Metropolitan Statistical Area (678). Detroit respondents were also asked to complete a mailback questionnaire that included additional vignettes as well as attitudes about such issues as reasons for imprisonment. The questionnaire was completed and returned by 50 percent of the overall sample (339). The first Japanese survey was a 1978 probability sample of 600 Yokohama residents. An additional probability sample survey was conducted in Kanazawa with 640 residents in 1979. The study found that decisions about justice were influenced by whether or not there was an apparent social relationship between the offender and victim. The American tendency was to see actors in isolation, and the Japanese tendency was to view them in relation to others. The Japanese, who emphasized the importance of role obligations and social ties, meted out punishment with the intent of restoring the offender to the social network. Americans, who acknowledge fewer social ties and have firmer convictions that evil resides in individuals, punished wrongdoers by isolating them from the community. The book explores the implications of these diverse views of justice. Appended questionnaires, chapter notes, 460 references, and author and subject indexes