The 1981 study asked witnesses whose cases had been decided in mid-1980 if they had been intimidated and then conducted telephone interviews with those who admitted being threatened, realizing that persons who had been most frightened were unlikely to agree to an interview. Of the respondents, 72 percent were both victims and eyewitnesses to the crime, and 59 percent of the cases involved violent crimes. Almost half the victims knew the defendant before the crime. Threats ranged from ominous looks or gestures to rumors and physical confrontation, with verbal threats being most common. Moreover, 72 percent were threatened in their homes, neighborhoods, schools, or workplaces where the criminal justice system can exert little control. The survey found that 69 percent of the witnesses reported threats to criminal justice officials whose primary response was to warn the defendant not to harass the witness. Few arrests were made, probably due to insufficient evidence. While 58 percent felt that reporting the threat helped, 22 percent were bothered again by defendants. Witnesses frequently complained that the case outcome was not strong enough or that the defendant should not have been released on bail. The criminal justice officials interviewed tended to overstate the kinds of assistance normally given to intimidated witnesses, but agreed that intimidation was a significant problem. Measures to reduce intimidation include separating victims and defendants immediately upon arrest, providing a separate waiting area for victims at police precincts and witness reception centers in courthouses, expediting cases where threats are reported, informing witnesses when defendants are released, and making smaller concessions in plea bargains when intimidation has been reported. The article includes tables and nine footnotes.