Contrary to their popular image as secluded private acts, most violent crimes are com-
mitted in the presence of a social audience. For example, data from the National Crime
Victimization Survey (NCVS) in the 1990s suggest that bystanders are present in about
two-thirds of violent victimizations. Bystanders are present in about 70% of assaults, 52%
of robberies, and 29% of the rapes or sexual assaults in these national data (Planty, 2002).
According to NCVS data, an estimated 6.4 million violent crimes are witnessed by third
parties each year. 1
When witnessing a criminal act or any other potentially dangerous situation, bystanders
have several choices. They can ignore the situation and do nothing, offer indirect inter-
vention by summoning the police or other people for help, or directly intervene to assist
the victim in thwarting the attack. Previous research on helping behavior suggests that
bystander inactivity is the predominant response in a variety of potentially dangerous situ-
ations. Field experiments and observational studies reveal that helping behavior is often
the exceptional case when people are seriously injured in accidents, have excessive bleed-
ing, or are involved in an intense verbal altercation with another party (see, for review;
Fisher, Greitemeyer, Pollozek, & Frey, 2006; Howard & Crano, 1974; Latane & Darley,
1970; Smithson, Amato, & Pearce, 1983). Even when helping involves little direct costs
to the bystander, most people do not typically offer assistance to another. The passage of
“Good Samaritan” laws and the public designation of people who help others in selfless
acts of bravery as heroes is also indirect evidence of the relative infrequency of helping
behavior in contemporary American society.
NCVS data do not provide a direct measure of bystander inaction. Instead, victims are
asked survey questions about whether third-party involvement “helped or worsened” the
situation. Among offenses in which the actions of the third party were known, nearly half
the victims reported that the bystander neither helped nor worsened the situation (Planty,
2002). Consistent with general studies of helping behavior, these results also suggest that
inactivity is the typical reaction of bystanders who witness violent crimes.
When bystanders intervene in criminal offenses, their behavior is judged far more
likely to help than hurt (see Planty, 2002). This ratio of helping/hurting is highest among
aggravated assaults (3.5:1) and lowest in cases of rape and sexual assault (2.2:1). The
predominant way in which bystanders help is through the “prevention of injury or further
injury” to the victim. Bystanders are judged as worsening the situation primarily by “mak-
ing the offender angrier.” Third parties may worsen the situation by overreacting or saying
something foolish that escalates the violence. Alternatively, offenders may inflict greater
injury to victims in front of third parties as an immediate public forum for maintaining or
reaffirming one’s masculine identity as a “tough guy” (see Goffman, 1959; Lofland, 1969;
Miethe & Deibert, 2007). It is within these public situations that the presence of bystanders
may hurt more than help
http://