This was a phenomenological study that used content analysis of in-depth interviews with 20 male batterers and their female partners. The interview data were collected as part of a research project on wife-beating in northern Israel during 1986-89. All of the participating couples had reported at least one episode of man-to-woman physical violence during the 12 months preceding contact by researchers and were thus on record in the public social service agencies. There was no evidence of a history of mental illness, retardation, or chronic drug addiction in any of the subjects. An in-depth semi-structured interview was conducted based on an interview guide that covered 10 themes related to the context and situation of the violence. The data were analyzed by using grounded theory methods (Strauss, 1987). The findings show that the extent of authenticity and assumption of responsibility concerning moral feelings will determine whether these will or will not lead to violence among intimates. It is possible to trace five alternative paths of the relationship between moral feelings and such violence. The paths are nonviolent conflicts, no moral feelings following violence, positive ways of resolving moral feelings, unwillingness to acknowledge moral feelings, and awareness and authenticity toward moral feelings. The degree of one's awareness of feelings, actions, and other people impacts on the direction and extent of such processes as acceptance or denial of the violent event and acceptance or denial of moral feelings following the event. This in turn may affect further developments, such as the authenticity of reaction to moral feelings and the way these are resolved. Future research in this area is suggested. 2 figures and 55 references