Available via license: CC BY-NC-ND 4.0
Content may be subject to copyright.
Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=wamt20
Journal of Aggression, Maltreatment & Trauma
ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/wamt20
Deny, Attack, and Reverse Victim and Offender
(DARVO): What Is the Influence on Perceived
Perpetrator and Victim Credibility?
Sarah Harsey & Jennifer J. Freyd
To cite this article: Sarah Harsey & Jennifer J. Freyd (2020): Deny, Attack, and Reverse Victim
and Offender (DARVO): What Is the Influence on Perceived Perpetrator and Victim Credibility?,
Journal of Aggression, Maltreatment & Trauma, DOI: 10.1080/10926771.2020.1774695
To link to this article: https://doi.org/10.1080/10926771.2020.1774695
© 2020 The Author(s). Published with
license by Taylor & Francis Group, LLC
Published online: 08 Jun 2020.
Submit your article to this journal
Article views: 730
View related articles
View Crossmark data
Deny, Attack, and Reverse Victim and Oender (DARVO):
What Is the Inuence on Perceived Perpetrator and Victim
Credibility?
Sarah Harsey
a
and Jennifer J. Freyd
b
a
Department of Psychology, University of California, Santa Cruz, California, USA;
b
Department of
Psychology, University of Oregon, Eugene, Oregon, USA
ABSTRACT
Perpetrators of interpersonal violence sometimes use denial,
engage in personal attacks on victim credibility, and assume
a victimized role (Deny, Attack, Reverse Victim and Oender;
DARVO) to deect blame. Two new experimental vignette studies
were conducted to investigate DARVO. Experiment 1 (316 uni-
versity students) aimed to assess the eects of a perpetrator’s use
of DARVO on perceptions of perpetrator and victim credibility,
responsibility, and abusiveness. Participants who were exposed
to DARVO perceived the victim to be less believable, more
responsible for the violence, and more abusive; DARVO also led
participants to judge the perpetrator as less abusive and less
responsible. Experiment 2 (360 university students) examined
whether learning about DARVO could mitigate its eects on
individuals’ perceptions of perpetrators and victims. Results
from Experiment 2 indicate that DARVO-educated participants
perceived the victim as less abusive and more believable, and
rated the perpetrator as less believable. These experiments show
DARVO eectively reinforces the distrust of victims’ narratives,
but education can reduce some of its power. We suggest that
more research and education about this perpetrator tactic is
needed to combat its anti-victim eects.
ARTICLE HISTORY
Received 15 November 2019
Revised 11 February 2020
Accepted 17 April 2020
KEYWORDS
Abuse; victims; perpetrators;
victim credibility;
interpersonal violence;
intimate partner violence;
dating violence; DARVO
DARVO (Deny, Attack, Reverse Victim and Offender roles) describes how
perpetrators of interpersonal violence deflect blame and responsibility when
confronted for their abusive behavior (Freyd, 1997). When used, a perpetrator
denies or minimizes the harms of any wrongdoing, attacks the victim’s cred-
ibility, and reverses victim and offender roles such that the perpetrator
assumes a victimized position and declares the victim to be the true perpe-
trator. To date, there is very little research investigating DARVO. One empiri-
cal study examining the prevalence of DARVO during victim-perpetrator
confrontations revealed that it is a commonly-used response by perpetrators
(Harsey et al., 2017). This research also discovered a relationship between the
perpetrators’ use of DARVO and victims’ feelings of self-blame: the more
CONTACT Sarah Harsey sharsey@ucsc.edu Department of Psychology, University of California, Santa Cruz,
1156 High St., Santa Cruz, CA 95064
JOURNAL OF AGGRESSION, MALTREATMENT & TRAUMA
https://doi.org/10.1080/10926771.2020.1774695
© 2020 The Author(s). Published with license by Taylor & Francis Group, LLC
This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives
License (http://creativecommons.org/licenses/by-nc-nd/4.0/), which permits non-commercial re-use, distribution, and reproduction in
any medium, provided the original work is properly cited, and is not altered, transformed, or built upon in any way.
DARVO that the perpetrator used during the confrontation, the more victims
reported feeling blameworthy for the wrongdoing.
Perpetrator tactics
Previous research has described tactics that perpetrators might use to temper
reactions to their wrongdoings. “Outrage management” is a term that repre-
sents a set of techniques employed by perpetrators that mitigate observers’
negative evaluations of both perpetrators and their objectionable behaviors
(McDonald et al., 2010). The perpetrator, in order to avoid facing conse-
quences, therefore tries to mollify the potential backlash when held accoun-
table for their actions. McDonald et al. (2010) noted that such outrage
management techniques include casting doubt onto the credibility of the
victim and denying the victims’ versions of events or reframing them so that
they appear more innocuous. This closely mirrors the denial and personal
attacks described by DARVO. Similarly, both outrage management and
DARVO represent ways in which perpetrators actively try to explain away
and manipulate bystanders’ understanding of abusive events.
Researchers have also proposed that perpetrators will engage in one of two
strategies in order to deflect blame for wrongdoing: either admit to committing
the wrongdoing but emphasize previous good behavior (play the hero), or
highlight some past suffering (play the victim) (Gray & Wegner, 2011). In
a series of experimental studies, Gray and Wegner examined the impact of these
two roles on observers and discovered that playing the victim – but not playing
the hero – effectively decreases the amount of blame ascribed to the perpetrator.
Although research on DARVO as a perpetrator tactic is limited, previous
studies have examined the components of DARVO individually. For example,
the literature on perpetrators describes how denial, minimization, and victim-
blaming are commonly expressed by individuals who have committed inter-
personal violence, including sexual assault and intimate partner violence
(Henning et al., 2005; Lila et al., 2008; Scott & Straus, 2007). Studies that
have conducted interviews with perpetrators of intimate partner violence have
found that the abusers who minimized the severity of the abuse were also likely
to implicate the victim as the instigator of the violence (Dutton, 1986; Lila
et al., 2008). Recent research on DARVO confirms that the three parts of
DARVO – denial, attacking the victim, and reversal of victim and offender –
are indeed used together when individuals are confronted about a wrongdoing
they committed. In a survey of 138 undergraduates who had confronted
a perpetrator over a wrongdoing, nearly 72% of participants reported that
the perpetrator simultaneously used denial, personal attacks, and attempted to
reverse victim and offender roles (Harsey et al., 2017). These results suggest
that perpetrators tend to employ DARVO as opposed to using a singular
strategy (such as denial alone) to offset blame.
2S. HARSEY AND J. J. FREYD
Perceptions of victims and perpetrators
For those who have committed abusive acts, the ability to influence how others
perceive them and their victims is indispensable. Convincing bystanders that no
abusive behavior took place (or that if something did occur it was not harmful)
and that the victim is untrustworthy gives the perpetrator a clear advantage in
both social networks and the legal system. If successful, the perpetrator can avoid
blame and thereby avoid disadvantageous outcomes. The victim’s account is
doubted and ultimately disregarded in favor of the perpetrator’s narrative.
Perceptions of victims and perpetrators can indeed be manipulated, as indi-
cated by experimental research that has identified numerous factors that influence
observers’ attributions of blame, credibility, and responsibility. Characteristics
such as victim and perpetrator gender (Stewart et al., 2012; Van der Bruggen &
Grubb, 2014), age (Bottoms et al., 2014), and race (George & Martinez, 2002)
impact observers’ perceptions of victims and perpetrators. Research on how
behavior affects observers’ perceptions finds that victims’ consumption of intox-
icants (Angelone et al., 2016), lack of resistance during an assault (Angelone et al.,
2015), and flat emotional expression (Ask, 2010) reduce victim credibility.
Research examining perpetrators’ behavior in the context of blame attribution
is relatively sparse and focuses on intoxication, which generally has an exonerat-
ing effect for perpetrators (Cameron & Stritzke, 2003; Stormo et al., 1997).
Studies investigating perceptions of victims and perpetrators seem to over-
look perhaps one of the more straightforward ways such perceptions might be
manipulated: through perpetrators’ response to the abuse. DARVO provides
a useful framework through which the impact of perpetrators’ statements
about their wrongdoing can be experimentally examined.
Some past research has used perpetrators’ statements as stimuli for experi-
ments; sometimes, these statements contain denial but were not manipulated
to examine how the denial affects observes’ perceptions (Goodman-Delahunty
& Graham, 2011; O’Donohue & O’Hare, 1997). Other studies have manipu-
lated written descriptions of the perpetrator’s emotional display in vignettes,
finding that a perpetrator who expressed sadness was rated less harshly than
a perpetrator who showed more neutral emotions (Robinson et al., 1994;
Tsoudis & Smith-Lovin, 1998). However, the actual words said by the perpe-
trator – the statements themselves – were not varied.
The present study
One goal of the present study is to therefore assess the extent to which
a perpetrator’s use of DARVO influences observers’ attributions of victim and
perpetrator credibility, responsibility, and abusiveness. Experiment 1 used a 2 × 2
independent measures experimental design that varied perpetrator/victim gender
and perpetrator’s DARVO use in a series of vignettes describing an incident of
JOURNAL OF AGGRESSION, MALTREATMENT & TRAUMA 3
intimate partner violence. Since the proposed aim of DARVO is to excuse the
actions of the perpetrator and implicate the victim as the wrongdoer (Freyd, 1997),
we hypothesized that participants who read the perpetrator statement character-
ized by DARVO would rate the perpetrator as being less abusive, less responsible
for the abuse, and as more credible than the perpetrator who did not use DARVO.
Further, we predicted that the perpetrator’s use of DARVO would also lead
participants to rate the victim as more abusive, more responsible for the abuse,
and as less credible. By varying the gender of the perpetrator and victim pairs, we
sought out to further explore the relationships between gender and DARVO.
Another aim of the study is to determine whether learning about DARVO
can alter its possible impact on perceptions of victims and perpetrators. More
specifically, we were interested in investigating the extent to which educating
individuals about this perpetrator tactic can minimize its hypothesized influ-
ence. To examine this question, Experiment 2 was conducted using a two-
condition independent measures experiment. Both conditions contained the
same vignette regarding an instance of sexual assault, which included
a statement from the victim and DARVO-identified statement from the perpe-
trator. A general description of DARVO was included in one of the conditions.
We hypothesized that, compared to individuals who did not read about
DARVO, those who read the DARVO description would rate the perpetrator
as less believable, more abusive, and more responsible. Similarly, we predicted
that receiving a brief DARVO education would lead individuals to perceive the
victim to be more believable, less abusive, and less responsible. Experiment 2
also included additional dependent variables measuring whether participants
believed the perpetrator and victim should be punished for their actions.
Experiment 1
Method
Participants
Participants were 316 undergraduate students from a large, public Northwestern
university (215 women and 101 men) ranging in age from 18 to 49-years-old
(M
Age
= 20.71, SD
Age
= 3.77). Consistent with the demographics of the uni-
versity, a majority of the participants identified as Caucasian (69.3%) while
smaller numbers of those who completed the study identified as Asian
(17.4%), “Other” (8.9%), and African American (1.3%). Five participants indi-
cated they were Native American or Pacific Islander, while an additional five
participants elected to not disclose their racial identity.
Materials
The study was comprised of a set of vignettes followed by six questions
regarding participants’ perceptions of the characters described in the vignettes.
4S. HARSEY AND J. J. FREYD
Each vignette described an incident of interpersonal violence between dating
partners recounted as a first-person narrative from either the victim or
perpetrator’s perspective. Before reading the victim and perpetrator’s accounts
of the incident, participants were provided with a brief description that clearly
stated that the perpetrator acted violently toward the victim and that the
victim was left with visible bruising the day following the attack. Half of the
participants read vignettes in which a female was the victim of a male’s abuse
while the other half read vignettes describing an incidence of violence in which
a female character victimized a male.
After reading the brief description, participants then read two vignettes,
with one describing the victim’s point of view and a second containing the
perpetrator’s narrative. Across all four conditions, the victim’s narrative was
identical, with the exception of the gender of the victim and perpetrator. The
perpetrator’s narrative varied according to the condition: in half of the con-
ditions, the perpetrator used DARVO tactics in order to deflect blame and
responsibility for the abuse. In the other, non-DARVO conditions, the perpe-
trator’s narrative more closely approximated the victim’s version of events; the
perpetrator in these conditions also took responsibility for the abuse and
expressed remorse.
Six questions followed the victim and perpetrator vignettes. Half of the
questions pertained to the perceived qualities of the victim, while the remaining
three questions regarded the perceived qualities of the perpetrator. These
qualities are believability, responsibility for the incident of interpersonal vio-
lence, and abusiveness. The questions are: (1) “How believable do you think [the
victim’s] side of the story is?,” (2) “How responsible do you think [the victim] is for
what happened?,” (3) “How abusive do you think [the victim’s] behavior is?,” (4)
“How believable do you think [perpetrator’s] side of the story is?” (5) “How
responsible do you think [the perpetrator] is for what happened?,” (6) “How
abusive do you think [the perpetrator’s] behavior is?” Participants responded to
each of the questions on a 4-point Likert scale ranging from 0 to 3. For example,
participants could indicate whether they thought the victim’s behavior was 0 –
Not at all, abusive, 1 – Not very abusive, 2 – Somewhat abusive, or 3 – Very
abusive.
Procedure
This study was administered online to participants as part of a larger depart-
ment-wide survey consisting of a battery of brief measures from psychology
researchers at the university. This general survey has approval by the univer-
sity’s institutional review board and takes participants no longer than one hour
to complete. To participate in this survey, students accessed the university’s
research study sign-up webpage and elected to take part in an online survey for
course credit – no information about the content of the measures contained
within the survey was provided as to prevent self-selection.
JOURNAL OF AGGRESSION, MALTREATMENT & TRAUMA 5
Participants in this study were randomly assigned to one of four vignette
conditions (2x2 of gender by perpetrator DARVO use): (1) male victim, female
perpetrator who uses DARVO, (2) female victim, male perpetrator who uses
DARVO, (3) male victim, female perpetrator who does not use DARVO
(instead, accepts responsibility and is remorseful), (4) female victim, male
perpetrator who does not use DARVO. In each condition, the victim’s vignette
was positioned before the perpetrator’s, with each vignette directly followed by
questions regarding the perceived believability, responsibility, and abusiveness
of the character (either the victim or perpetrator) narrating the vignette.
A total of 162 participants were randomly assigned to the DARVO conditions
(87 in the male victim condition, and 75 in the female victim condition), and
an additional 154 completed the non-DARVO conditions (81 in the male
victim condition, and 73 in the female victim condition).
Results
DARVO vs. non-DARVO
A one-way MANOVA revealed an overall significant effect of DARVO on the
dependent variables, F(1, 314) = 25.80, p <.001. Tests of between-subjects
effects revealed that participants who read perpetrator accounts characterized
by DARVO rated victims as more responsible for the abuse (F(1, 314) = 13.84,
p < .001) and their actions as more abusive (F(1, 314) = 13.68, p < .001)
compared to those who read the perpetrator vignettes that did not contain
DARVO. Similarly, participants exposed to the perpetrator’s DARVO use
perceived the perpetrator as less responsible for the abuse (F(1, 314) = 5.63,
p = .018) and as less abusive (F(1, 314) = 26.43, p < .001). When comparing
believability between the DARVO and non-DARVO conditions, we discov-
ered that those in the DARVO condition rated both the victim (F(1,
314) = 25.91, p < .001) and the perpetrator as less believable (F(1,
314) = 93.85, p < .001) See Figure 1 for an illustration of these findings.
Perpetrator and victim gender
A MANOVA was conducted to examine the effects of perpetrator and victim
gender on individuals’ perceptions, which produced a significant test, F(1,
314) = 5.48, p < .001. Out of the six dependent variables, four were found to
produce statistically significant differences in this comparison: in contrast to
the male victim, the female victim (victimized by a male perpetrator) was rated
as being less responsible for the abuse (F(1, 314) = 19.89, p < .001). The female
perpetrator was viewed as slightly more believable (F(1, 314) = 2.45, p = .047),
less responsible (F(1, 314) = 2.79, p = .033), and as less abusive (F(1,
314) = 5.07, p = .005). Female and male victims were not rated differently in
terms of believability and abusiveness.
6S. HARSEY AND J. J. FREYD
0
0.5
1
1.5
2
2.5
Victim Believability Perpetrator Believability
Average Rating
Panel A: Perceived Believability
Perpetrator DARVO No Perpetrator DARVO
0
0.5
1
1.5
2
2.5
Victim
Responsibility
Victim
Abusiveness
Perpetrator
Responsibility
Perpetrator
Abusiveness
Average Rating
Panel B: Perceived Responsibility and Abusiveness
Perpetrator DARVO No Perpetrator DARVO
Figure 1. Bar graphs for Experiment 1 containing mean ratings for victim and perpetrator
believability (Panel A, favorable evaluations), responsibility, and abusiveness (Panel B, unfavorable
evaluations) by condition.
JOURNAL OF AGGRESSION, MALTREATMENT & TRAUMA 7
Participant gender
A MANOVA resulted in a significant effect of participant gender on perceptions
of the victim and perpetrator, F(1, 314) = 2.64, p = .016. Tests of between-subjects
effects revealed that women, compared to men, rated victims as more believable
(F(1, 314) = 4.60, p = .004), less responsible for the abuse (F(1, 314) = 5.39,
p = .004), and viewed the vignette perpetrators as more abusive (F(1, 314) = 4.74,
p = .006). There were no differences between men and women’s ratings for victim
abusiveness, perpetrator believability, and perpetrator responsibility.
Interactions between DARVO, victim/perpetrator gender, and participant gender
When testing for the two-way interactions between DARVO condition and vign-
ette victim/perpetrator gender, only victim believability resulted in a significant
interaction (F(1, 312) = 6.30, p = .013). Male victims were rated as less believable in
the DARVO condition (M = 1.90, SE = .648) compared to the No DARVO
condition (M = 2.49, SE = .691), F(1, 312) = 29.72, p < .001. This difference was
not found for the female victims in the DARVO (M = 2.13, SE = .082) and No
DARVO (M = 2.33, SE = .083) conditions, F(1, 312) = 2.81, p = .095.
Neither the interactions between victim/perpetrator gender and participant
gender nor between DARVO and participant gender were found to be statistically
significant for any of the dependent variables. An examination of the three-way
interactions between DARVO, victim/perpetrator gender, and participant gender
resulted in one marginally significant interaction for perpetrator believability (F(1,
308) = 3.88, p = .050). Further investigation of this three-way interaction revealed
that, for men, the female perpetrator was rated as less believable when using
DARVO (M = 1.43, SE = .113) compared to when she did not use DARVO
(M = 2.19, SE = .131), F(1, 97) = 19.49, p < .001. The believability of male
perpetrators between the DARVO (1.45, .149) and No DARVO conditions
(M = 1.75, SE = .149) did not differ amongst male participants, F(1, 97) = 2.02,
p = .159. However, women rated both female and male perpetrators as less
believable in the when DARVO was used. More specifically, women gave female
perpetrators lower believability scores in the DARVO condition (M = 1.52,
SE = .096) than in the No DARVO condition (M = 2.26, SE = .093), F(1,
211) = 30.25, p < .001; similarly, women also gave male perpetrators lower
believability scores in the DARVO condition (M = 1.24, SE = .093) than in the
No DARVO condition (M = 2.17, SE = .095), F(2, 211) = 49.22, p < .001.
Experiment 2
Method
Participants
A total of 360 undergraduate students took part in Experiment 2. Consistent
with the participant demographics in Experiment 1, the majority were women
8S. HARSEY AND J. J. FREYD
(70%), identified as Caucasian (59.2%), heterosexual (87.2%), and on average
were 19.66 years old (SD
Age
= 2.28). Three participants indicated their gender
identity as either nonbinary or questioning; seven participants identified as
lesbian or gay, 34 identified as bisexual, and five indicated they were asexual or
“other” (e.g., pansexual). Among the participants who did not identify as
Caucasian, 53 were East Asian, 23 were Latinx or Chicanx, 20 were biracial
or multiracial, 16 were Southeast Asian, 13 were Black/African American, 7
were Pacific Islander, and 3 identified as Middle Eastern or North African.
Eleven participants reported their racial identity as “other,” and one partici-
pant opted to not report their racial identity.
Materials
The vignettes used in this study were comprised of two fictional, first-person
narratives describing a sexual assault from a female victim (Sophia) and male
perpetrator’s (Jacob) perspective. Before reading the statements, participants
were provided with the following information: “The following are statements
about a sexual assault that occurred at a party. Jacob is accused of sexually
assaulting Sophia.” Sophia’s statement included a description of the assault by
Jacob, which included being forcibly kissed and grabbed without her consent.
Jacob’s statement did not admit to the assault but instead employed DARVO;
his version of events denied committing any assault, claimed Sophia had too
much to drink and fabricated the assault to protect her reputation, and that his
own reputation had been damaged by her accusation. After reading the
statements from Sophia and Jacob, participants responded to the same set of
dependent variables included in Experiment 1: victim and perpetrator believ-
ability, responsibility, and abusiveness. Participants responded to these ques-
tions on the 4-point Likert scale described in Experiment 1.
Two additional items were included in Experiment 2 that assessed partici-
pants’ beliefs about victim and perpetrator punishment. Participants
responded to the questions, “Do you think Sophia should be disciplined or
punished for her actions?” and “Do you think Jacob should be disciplined or
punished for his actions” with either “Yes,” “No,” or “Not Sure.” The experi-
mental condition contained a description of DARVO that accompanied the
statements from Sophia and Jacob. The description included in this condition
is as follows:
DARVO stands for Deny, Attack, Reverse Victim and Offender. It describes how some
people may react when they are accused of or held responsible for bad behavior. People
may use DARVO to deflect blame and responsibility for the wrongdoing. Deny: the
person will deny that they did anything wrong. Sometimes they will acknowledge
something happened, but that whatever happened wasn’t that bad and that it didn’t
cause any harm. Attack: some people will attack the credibility of their accusers, making
it seem like the accusers are untrustworthy and should therefore not be believed. People
may say that their accusers are liars, mentally ill, or have ulterior motives. Reverse Victim
JOURNAL OF AGGRESSION, MALTREATMENT & TRAUMA 9
and Offender: finally, some people will try to convince others that they are the “true”
victim, and that their accuser is actually the guilty one.
After learning about DARVO, participants were asked if they had ever experi-
enced someone using DARVO on them or on someone they knew and
responded with “Yes,” “No,” or “Not Sure.” Additionally, using the same
response scale, participants indicated whether they believed Sophia and
Jacob used DARVO in their narratives. The participants who did not read
the DARVO description did not respond to these questions about DARVO.
Procedure
The procedures for this experiment were largely identical to those used in
Experiment 1; unlike Experiment 1, however, participants in Experiment 2
were randomly assigned to one of two conditions: perpetrator and victim
statements only (control condition), or the same statements plus the
DARVO description (experimental condition). There were 241 individuals
in the experimental condition and 119 in the control condition. As with
Experiment 1, participants did not know about the content of the study at
the time of sign-up to minimize self-selection.
Results
DARVO items
Among the participants who read the description of DARVO, 56.9% (N = 136)
reported that they had experienced someone using DARVO on them or on
someone they know. An additional 45 (19%) participants were “Not Sure” if
they had experienced DARVO in this way, while the remaining 58 participants
(24.1%) responded that they had not experienced DARVO. Most participants
in this condition (73.9%) judged Jacob, the perpetrator, to have used DARVO
in his statement. In contrast, only 14.9% believed that the victim had used
DARVO in her statement. When judging both Jacob and Sophia’s statements,
participants expressed a fair amount of uncertainty: 21.2% of individuals
reported being unsure if Sophia had used DARVO and 16.2% were unsure if
Jacob had used DARVO. The other participants in this condition indicated
that the victim (64%) and the perpetrator (10%) did not use DARVO.
Gender
As in Experiment 1, we found significant differences in how men and women
perceived the perpetrator and victim. Women generally found the victim to be
more believable (F(1, 352) = 30.74, p < .001), less responsible (F(1, 352) = 9.35,
p = .002), and less abusive (F(1, 352) = 7.20, p = .008). Men rated the
perpetrator as more believable (F(1, 352) = 4.64, p = .032) and less abusive
(F(1, 352) = 4.43, p = .036. We did not find a difference between men and
10 S. HARSEY AND J. J. FREYD
women in their perceptions of perpetrator responsibility. Table 1 contains the
means for this set of analyses.
When looking at gender differences in victim and perpetrator punishment
beliefs, we found that 13.6% men but only 5.2% of women endorsed punish-
ment for the victim; a chi-squared test indicated that this difference in
proportions was significant, χ
2
(1, N = 360) = 7.50, p = .006. However, we
did not find a significant association between gender and perpetrator punish-
ment, χ
2
(2, N = 359) = 5.45, p = .07.
DARVO education vs. no education
A MANOVA was conducted to compare responses on the six dependent
variables regarding perceptions of the victim and perpetrator between condi-
tions. This omnibus resulted in significance (F(1, 347) = 3.82, p = .001), which
prompted individual F-tests to be conducted for each of the dependent vari-
ables. Results of these tests revealed a significant difference between means for
perpetrator believability (F(1, 347) = 18.75, p < .001), victim abusiveness (F(1,
347) = 8.47, p = .004), and victim believability (F(1, 347) = 6.31, p = .012).
More specifically, we found that DARVO-educated individuals rated the
perpetrator as less believable and more abusive. Participants who read about
DARVO also rated the victim as less abusive and more believable. Perpetrator
responsibility (F(1, 347) = 2.59, p = .11) and victim responsibility (F(1,
347) = 3.55, p = .06) produced insignificant results. Figure 2 below illustrates
the significant results of this analysis.
Victim and perpetrator punishment
Pearson chi-squares tests were computed for the two questions regarding
victim and perpetrator punishment. The tests showed that responses for
victim punishment (χ
2
(2, N = 360) = 6.59, p = .037) and perpetrator punish-
ment (χ
2
(2, N = 359) = 6.86, p = .032) differed between the two conditions.
More specifically, 58% of participants in the DARVO education condition
believed the perpetrator should be punished or disciplined, whereas a smaller
proportion (43.7%) of individuals who did not read about DARVO agreed
with this assessment, χ
2
(1, N = 360) = 6.53, p = .01. Tests also revealed that
fewer individuals in the DARVO explanation condition (5.4%) endorsed
punishment for the victim compared to participants in the control condition
(12.6%) (χ
2
(1, N = 360) = 5.74, p = .02.
Discussion
DARVO describes how perpetrators of interpersonal violence deny their
abusive actions, attack the credibility of their victims, and reverse victim and
offender roles such that the victim is portrayed as the true agent of abuse. The
aim of this study was to investigate how perpetrators’ use of DARVO
JOURNAL OF AGGRESSION, MALTREATMENT & TRAUMA 11
Table 1. Means, standard deviations, significance levels, and partial eta-squared effect sizes for tests of between-subjects effects for Experiment 1 and Experiment 2.
Experiment 1 Perpetrator DARVO No Perpetrator DARVO
Female
Participants Male Participants Female Victim, Male Perpetrator Male Victim, Female Perpetrator
Victim M(SD)M(SD) η
p2
M(SD)M(SD) η
p2
M(SD)M(SD) η
p2
Believability 2.01(.70) 2.42(.73)*** .076 2.29(.72) 2.03(.77)** .026 2.23(.76) 2.18(.73) –
Responsibility 1.52(.78) 1.19(.79)*** .042 1.27(.80) 1.55(.78)** .027 1.16(.85) 1.55(.72)*** .060
Abusiveness 1.31(.64) 1.02(.75)*** .042 1.13(.68) 1.24(.76) – 1.18(.73) 1.16(.70) –
Perpetrator
Believability 1.40(.62) 2.15(.76)*** .230 1.80(.82) 1.69(.73) – 1.67(.83) 1.85(.74)* .013
Responsibility 2.13(.76) 2.34(.80)* .018 2.28(.77) 2.13(.81) – 2.33(.79) 2.14(.77)* .014
Abusiveness 2.08(.80) 2.53(.73)*** .078 2.38(.76) 2.12(.85)** .023 2.43(.79) 2.18(.79)** .025
Experiment 2 No DARVO Education DARVO Education Female Participants Male Participants – – –
Victim M(SD)M(SD) η
p2
M(SD)M(SD) η
p2
– – –
Believability 3.30(.72) 3.49(.63)* .018 3.55(.59) 3.14(.74)*** .08 – – –
Responsibility 2.07(.91) 1.88(.86) – 1.85(.88) 2.15(.84)** .026 – – –
Abusiveness 1.83(.88) 1.57(.78)* .023 1.58(.78) 1.85(.88)** .020 – – –
Perpetrator – – –
Believability 2.64(.67) 2.29(.75)** .051 2.35(.74) 2.53(.74)* .013 – – –
Responsibility 3.18(.74) 3.31(.78) – 3.29(.76) 3.19(.76) – – – –
Abusiveness 2.87(.79) 3.07(.81)* .014 3.06(.78) 2.88(.86)* .012 – – –
*p <.05, **p <.01, ***p <.001.
12 S. HARSEY AND J. J. FREYD
0
0.5
1
1.5
2
2.5
3
3.5
Victim Believability Perpetrator Believability
Average Rating
Panel A: Perceived Believability
No DARVO Education DARVO Education
0
0.5
1
1.5
2
2.5
3
Victim Abusiveness Perpetrator Abusiveness
Average Rating
Panel B: Perceived Abusiveness
No DARVO Education DARVO Education
Figure 2. Bar graphs for Experiment 2 containing mean ratings for victim and perpetrator
believability (Panel A, favorable evaluation) and abusiveness (Panel B, unfavorable evaluations)
by condition.
JOURNAL OF AGGRESSION, MALTREATMENT & TRAUMA 13
influences perceptions of victims and perpetrators, and whether learning
about DARVO could reduce these influences. The results of our experiment
revealed that a perpetrator’s use of DARVO does influence observers’ percep-
tions, leading individuals to view victims as more responsible for the abuse
perpetrated against them and as more abusive. Conversely, DARVO causes
perpetrators to be seen as less responsible for the abuse they committed and as
less abusive in general. We found that some of these effects, however, can be
mitigated when the observer has some knowledge of DARVO.
These findings confirm our predictions that exposure to a perpetrator’s use
of DARVO measurably displaces at least some of the blame for the abusive
behavior from the perpetrator to the victim. This fulfills a major aim of
DARVO, one in which the perception of victims as blameless targets of
abuse is suppressed in favor of the belief that victims play a culpable role in
their victimization. However, the results did not support our hypothesis in
regard to perpetrator believability: instead of supporting the prediction that
DARVO would increase perpetrator believability, analyses revealed that
DARVO decreases the extent to which perpetrators are viewed as believable.
In other words, although DARVO appears to benefit perpetrators and harm
victims on attributions of responsibility and abusiveness, DARVO appears to
make both victims and perpetrators less believable.
This result suggests that DARVO comes at some cost to its users as it
diminishes perpetrator believability, which may be due to its socially antag-
onistic nature. Relational aggression describes verbal and behavioral actions
that harm others by targeting a victim’s social relationships (Crick &
Grotpeter, 1995). It is associated with higher social status and perceived
popularity (Rose et al., 2004), but those who employ it are generally disliked
by their peers due to its antisocial characteristics (Cillessen & Mayeux, 2004).
DARVO itself could likely be considered a form of relational aggression, given
that it too is characterized by antisocial elements and functions to preserve
one’s social standing. The data presented in this study suggest that DARVO,
like relational aggression, provides perpetrators with some social advantage
(i.e., a reduction in negative attributions), yet simultaneously penalizes those
who use it by reducing perceived credibility. As noted by Cillessen and
Mayeux (2004), the penalties associated with relational aggression do not
appear to outweigh its benefits; in other words, despite its drawbacks, those
who engage in such aggression still are able to obtain overall favorable out-
comes for themselves. It is possible, then, that DARVO operates in a very
similar way for perpetrators.
DARVO’s effect on perceptions of victim believability has clear implications
in the legal system. In the absence of physical evidence, a victim’s perceived
credibility is considered by investigators to be one of the most important
factors in legal proceedings of sexual assault cases (Campbell et al., 2015)
and has long been the target of defense lawyers (Zydervelt et al., 2017).
14 S. HARSEY AND J. J. FREYD
Moreover, characteristics that impact sexual assault victim credibility are
linked with law enforcements’ judgment of a rape report’s legitimacy.
A study examining sexual assault cases reported to the LAPD identified that
reports from victims who had a history of mental health issues were ten times
more likely to be declared unfounded, while reports from victims whose moral
character was called into question were three times more likely to be judged to
be unfounded (Spohn et al., 2014). Due to its capacity to diminish victim
credibility and implicate the victim as being abusive and at least partially
responsible for their victimization, it is possible that DARVO discourages
victims from speaking about abuse and provides perpetrators with an advan-
tage in legal settings.
The current study also revealed that, in addition to increasing negative
perceptions of victim responsibility and abusiveness, DARVO causes obser-
vers to evaluate perpetrators less harshly. By leading observers to see
a perpetrator as less abusive and as bearing a smaller portion of responsibility
for their harmful actions, DARVO helps deflect some of the blame away from
the perpetrator. This illusion of shared responsibility is associated with
a willingness to treat a perpetrator more leniently. In a study examining the
effect of exculpatory writing styles on perceptions of male perpetrators of
domestic violence (Lamb & Keon, 1995), researchers found that participants
who read newspaper articles implicating a shared responsibility for the abuse
assigned more lenient punishments to the perpetrator.
Some of the DARVO effects discovered in the first experiment were found
to be reduced by learning about this perpetrator tactic. Moreover, having
some knowledge about this tactic produced more pro-victim assessments:
compared to those who made victim and perpetrator judgments without
reading about DARVO, a smaller proportion of individuals who received
information about DARVO agreed that the victim should be punished.
DARVO-educated individuals were also more likely to agree that the perpe-
trator should be punished. Although this educational intervention in
Experiment 2 was very brief, it still produced measurable effects – this
perhaps suggests that a more extensive educational intervention, such as
an interactive activity with extensive examples, would result in additional
inoculation against DARVO.
Gender played a meaningful role in perpetrator and victim ratings in the
current study. Compared to the female participants, men rated the victim as
less believable, more abusive, and more responsible for the abuse. The men in
this study also rated the perpetrator as less abusive and as more believable.
Moreover, in Experiment 2, men were more likely to endorse punishment for
the victim. These are unsurprising findings given that past research reliably
shows that men are more likely to exhibit more negative views of victims and
less punitive views of perpetrators (Bryant & Spencer, 2003; Ferrão &
Gonçalves, 2015; Rogers & Davies, 2007; Van der Bruggen & Grubb, 2014).
JOURNAL OF AGGRESSION, MALTREATMENT & TRAUMA 15
Regarding the perpetrator and victim gender in the vignettes from Experiment
1, we found that the male victim was rated as overall more responsible for the
abuse and that the male perpetrator was rated as less believable, more respon-
sible, and more abusive; further, we found an interaction effect suggesting that
the male victim’s believability is disproportionately affected by DARVO. In
general, male-perpetrated physical violence against women is considered to be
more serious than female-perpetrated violence against a man (Williams et al.,
2012). This perception has been attributed to the difference in men and
women’s physical size. For example, individuals asked to evaluate instances
of domestic violence take into consideration how men’s larger and stronger
bodies have the potential to cause more physical harm (Hamby & Jackson,
2010). While women are more likely to be targets of domestic violence and
sustain injuries from their partners (Tjaden & Thoennes, 1998), it is critical
that men who are physically abused by women are believed and regarded as
legitimate victims.
By experimentally manipulating a perpetrator’s use of DARVO through
vignettes, we can draw causal conclusions regarding the effect of DARVO on
victim and perpetrator believability, responsibility, and abusiveness. We can
further conclude that having a rudimentary understanding of DARVO can
produce more favorable views of victims. The predictions of this study were
largely supported by analyses. This study therefore provides proof-of-concept
evidence that DARVO can be effective when used by a perpetrator and that
some of its effects can be mitigated by a brief DARVO education.
Limitations
While the experimental methodology offers value in determining causality, the
present study had several limitations. The sample used in this study was fairly
homogeneous in terms of age and race, with all participants recruited from
a university and the majority of respondents identifying as Caucasian.
Although this particular sample was typical of the students attending the
university where this research was conducted, future studies on DARVO
should include samples that are representative of communities more diverse
in education level, age, and racial identity.
We cannot conclude much about real-world effect sizes from this research
given the experimental interventions in both experiments were minimal
compared to the typical deployment and potential education of DARVO in
the real world. For instance, it is likely that the effectiveness of DARVO was
diminished in our study by the written presentation of this tactic. DARVO
may produce a more profound impact when delivered by, for example,
a powerful man with a lot of self-confidence giving an oral presentation.
A fair amount of research has identified various participant attitudes and
beliefs that influence individuals’ perceptions of perpetrators and victims,
16 S. HARSEY AND J. J. FREYD
including participants’ rape myth acceptance and gender ideology (Angelone
et al., 2015; Davies et al., 2012), sexist beliefs (Abrams et al., 2003), and belief in
a just world (Stromwall et al., 2014). While the present study did not include
such attitudinal variables, it is possible that such participant characteristics
would interact with the effects of DARVO. Future research experimentally
examining DARVO would benefit from including measures of participant
attitudes and beliefs.
There are many aspects of DARVO that have yet to be investigated. This
includes DARVO as it relates to social class, race, sexuality, and other identities.
In the current study, the vignette characters are presumably heterosexual and
may also perhaps be as perceived, by default, as White. Additional research on
DARVO should examine its effects in the context of nonwhite, non-
heterosexual victims and perpetrators of varying social class. Moreover, future
studies should investigate DARVO and its association with perpetrator guilt.
Previous research confirms that perpetrators sometimes use DARVO on the
individuals they have victimized (Harsey et al., 2017), but it is not clear whether
DARVO is used primarily by those guilty of committing wrongdoing or if it is
also regularly employed by individuals who are innocent. Future research
should determine the relationship between DARVO use and culpability.
In conclusion, this study demonstrates that a perpetrator’s use of denial,
attacks, and reversal of victim and offender roles impacts individuals’ judgments
of victims and perpetrators. DARVO foments skepticism of victims’ trustworthi-
ness and blamelessness and encourages observers to see perpetrators as less
harmful and as playing a smaller role in the abuse they commit. Although it also
casts doubt on perpetrators’ narratives, this manipulation tactic targets and
damages the integrity of victims. DARVO, because it is employed in a society
steeped in commonly-believed myths about interpersonal violence, expertly
plays into and exacerbates the preexisting doubts surrounding the innocence
of victims and culpability of perpetrators. It is critical to recognize this perpe-
trator tactic and provide education about its purpose and effects. By identifying
and calling out DARVO when it occurs, we may be able to effectively interrupt
perpetrators’ attempts to discredit and silence their victims.
ORCID
Sarah Harsey http://orcid.org/0000-0001-9215-9753
References
Abrams, D., Viki, G. T., Masser, B., & Bohner, G. (2003). Perceptions of stranger and
acquaintance rape: The role of benevolent and hostile sexism in victim blame and rape
proclivity. Journal of Personality and Social Psychology, 84(1), 111–125. https://doi.org/10.
1037/0022-3514.84.1.111
JOURNAL OF AGGRESSION, MALTREATMENT & TRAUMA 17
Angelone, D. J., Mitchell, D., & Grossi, L. (2015). Men’s perceptions of an acquaintance rape:
The role of relationship length, victim resistance, and gender role attitudes. Journal of
Interpersonal Violence, 30(13), 2278–2303. https://doi.org/10.1177/0886260514552448
Angelone, D. J., Mitchell, D., & Smith, D. (2016). The influence of gender ideology, victim
resistance, and spiking a drink on acquaintance rape attributions. Journal of Interpersonal
Violence, 33(20), 3186–3210. Advance Online Publication. https://doi.org/10.1177/
0886260516635318
Ask, K. (2010). A survey of police officers’ and prosecutors’ beliefs about crime victims and
behavior. Journal of Interpersonal Violence, 25(6), 1132–1149. https://doi.org/10.1177/
0886260509340535
Bottoms, B. L., Peter-Hagene, L. C., Stevenson, M. C., Wiley, T. R. A., Mitchell, T. S., &
Goodman, G. S. (2014). Explaining gender differences in Jurors‘ reactions to child sexual
assault cases. Behavioral Sciences and the Law, 32(6), 789–812. https://doi.org/10.1002/bsl.
2147
Bryant, S. A., & Spencer, G. A. (2003). University students’ attitudes about attributing blame in
domestic violence. Journal of Family Violence, 18(6), 369–376. https://doi.org/10.1023/
A:102620581
Cameron, C. A., & Stritzke, W. G. (2003). Alcohol and acquaintance rape in Australia: Testing
the presupposition model of attributions about responsibility and blame. Journal of Applied
Social Psychology, 33(5), 983–1008. https://doi.org/10.1111/j.1559-1816.2003.tb01935.x
Campbell, B. A., Menaker, T. A., & King, W. R. (2015). The determination of victim credibility
by adult and juvenile sexual assault investigators. Journal of Criminal Justice, 43(1), 29–39.
https://doi.org/10.1016/j.jcrimjus.2014.12.001
Cillessen, A. H. N., & Mayeux, L. (2004). From censure to reinforcement: Developmental
changes in the association between aggression and social status. Child Development, 75(1),
147–163. https://doi.org/10.1111/j.1467-8624.2004.00660.x
Crick, N. R., & Grotpeter, J. K. (1995). Relational aggression, gender, and social-psychological
adjustment. Child Development, 66(3), 710–722. https://doi.org/10.1111/j.1467-8624.1995.
tb00900.x
Davies, M., Gilston, J., & Rogers, P. (2012). Examining the relationship between male rape
myth acceptance, female rape myth acceptance, victim blame, homophobia, gender roles,
and ambivalent sexism. Journal of Interpersonal Violence, 27(14), 2807–2823. https://doi.
org/10.1177/0886260512438281
Dutton, D. G. (1986). Wife assaulter’s explanations for assault: The neutralization of self-
punishment. Canadian Journal of Behavioural Science, 18(4), 381–390. https://doi.org/10.
1037/h0079964
Ferrão, M. C., & Gonçalves, G. (2015). Rape crimes reviewed: The role of observer variables in
female victim blaming. Psychological Thought, 8(1), 47–67. https://doi.org/10.5964/psyct.
v8i1.131
Freyd, J. J. (1997). Violations of power, adaptive blindness, and betrayal trauma theory.
Feminism & Psychology, 7(1), 22–32. https://doi.org/10.1177/0959353597071004
George, W. H., & Martinez, L. J. (2002). Victim blaming in rape: Effects of victim and
perpetrator race, type of rape, and participant racism. Psychology of Women Quarterly, 26
(2), 110–119. https://doi.org/10.1111/1471-6402.00049
Goodman-Delahunty, J., & Graham, K. (2011). The influence of victim intoxication and victim
attire on police responses to sexual assault. Journal of Investigative Psychology and Oender
Profiling, 8(1), 22–40. https://doi.org/10.1002/jip.127
Gray, K., & Wegner, D. M. (2011). To escape blame, don’t be a hero – Be a victim. Journal
of Experimental Social Psychology, 47(2), 516–519. https://doi.org/10.1016/j.jesp.2010.12.
012
18 S. HARSEY AND J. J. FREYD
Hamby, S., & Jackson, A. (2010). Size does matter: The effects of gender on perceptions of
dating violence. Sex Roles, 63(5–6), 324–331. https://doi.org/10.1007/s11199-010-9816-0
Harsey, S. J., Zurbriggen, E. L., & Freyd, J. J. (2017). Perpetrator responses to victim confronta-
tion: DARVO and victim self-blame. Journal of Aggression, Maltreatment & Trauma, 26(6),
644–663. https://doi.org/10.1080/10926771.2017.1320777
Henning, K., Jones, A. R., & Holdford, R. (2005). “I didn’t do it, but if I did I had a good
reason”: Minimization, denial, and attributions of blame among male and female domestic
violence offenders. Journal of Family Violence, 20(3), 131–139. https://doi.org/10.1007/
s10896-005-3647-8
Lamb, S., & Keon, S. (1995). Blaming the perpetrator: Language that distorts reality in news-
paper articles on men battering women. Psychology of Women Quarterly, 19(2), 209–220.
https://doi.org/10.1111/j.1471-6402.1995.tb00288.x
Lila, M., Herrero, J., & Gracia, E. (2008). Evaluating attribution of responsibility and mini-
mization by male batterers: Implications for batterer programs. The Open Criminology
Journal, 1(1), 4–11. https://doi.org/10.2174/1874917800801010004
McDonald, P., Graham, T., & Martin, B. (2010). Outrage management in cases of sexual
harassment as revealed in judicial decisions. Psychology of Women Quarterly, 34(2),
165–180. https://doi.org/10.1111/j.1471-6402.2010.01559.x
O’Donohue, W., & O’Hare, E. (1997). The credibility of sexual abuse allegations: Child sexual
abuse, adult rape, and sexual harassment. Journal of Psychopathology and Behavioral
Assessment, 19(4), 273–279. https://doi.org/10.1007/BF02229021
Robinson, D. T., Smith-Lovin, L., & Tsoudis, O. (1994). Heinous crime or unfortunate
accident? The effects of remorse on responses to mock criminal confessions. Social Forces,
73(1), 175–190. https://doi.org/10.1093/sf/73.1.175
Rogers, P., & Davies, M. (2007). Perceptions of victims and perpetrators in a depicted child
sexual abuse case: Gender and age factors. Journal of Interpersonal Violence, 22(5), 566–584.
https://doi.org/10.1177/0886260506298827
Rose, A. J., Swenson, L. P., & Waller, E. M. (2004). Overt and relational aggression and
perceived popularity: Developmental differences in concurrent and prospective relations.
Developmental Psychology, 40(3), 378–387. https://doi.org/10.1037/0012-1649.40.3.378
Scott, K., & Straus, M. (2007). Denial, minimization, partner blaming, and intimate aggression
in dating partners. Journal of Interpersonal Violence, 22(7), 851–871. https://doi.org/10.
1177/0886260507301227
Spohn, C., White, C., & Tellis, K. (2014). Unfounding sexual assault: Examining the decision to
unfound and identifying false reports. Law & Society Review, 48(1), 161–192. https://doi.org/
10.1111/lasr.12060
Stewart, C., Moore, T., Crone, T., DeFreitas, S. C., & Rhatigan, D. (2012). Who gets blamed for
intimate partner violence? The contributions of perpetrator sex category, victim confronta-
tion, and observer attitudes. Journal of Interpersonal Violence, 27(18), 3739–3754. https://
doi.org/10.1177/0886260512447571
Stormo, K. J., Lang, A. R., & Stritzke, W. G. K. (1997). Attributions about acquaintance rape:
The role of alcohol and individual differences. Journal of Applied Social Psychology, 27(4),
279–305. https://doi.org/10.1111/j.1559-1816.1997.tb00633.x
Stromwall, L. A., Landstrom, S., & Alfredsson, H. (2014). Perpetrator characteristics and blame
attributions in a stranger rape situation. The European Journal of Psychology Applied to Legal
Context, 6(2), 63–67. https://doi.org/10.1016/j.ejpal.2014.06.002
Tjaden, P., & Thoennes, N. (1998). Prevalence, Incidence, and consequences of violence against
women: Findings from the national violence against women survey. DC National Institute of
Justice and the Centers for Disease Control and Prevention.
JOURNAL OF AGGRESSION, MALTREATMENT & TRAUMA 19
Tsoudis, O., & Smith-Lovin, L. (1998). How bad was it? The effects of victim and perpetrator
emotion on responses to criminal court vignettes. Social Forces, 77(2), 695–722. https://doi.
org/10.1093/sf/77.2.695
Van der Bruggen, M., & Grubb, A. (2014). A review of the literature relating to rape victim
blaming: An analysis of the impact of observer and victim characteristics on attribution of
blame in rape cases. Aggression and Violent Behavior, 19(5), 523–531. https://doi.org/10.
1016/j.avb.2014.07.008
Williams, C., Richardson, D. S., Hammock, G. S., & Janit, A. S. (2012). Perceptions of physical
and psychological aggression in close relationships: A review. Aggression and Violent
Behavior, 17(6), 489–494. https://doi.org/10.1016/j.avb.2012.06.005
Zydervelt, S., Zajac, R., Kaladelfos, A., & Westera, N. (2017). Lawyers’ strategies for cross-
examining rape complainants: Have we moved beyond the 1950s? British Journal of
Criminology, 57(3), 551–569. https://doi.org/10.1093/bjc/azw023
20 S. HARSEY AND J. J. FREYD
