Content uploaded by Nathaniel G Wade
Author content
All content in this area was uploaded by Nathaniel G Wade
Content may be subject to copyright.
Forgivingness, Vengeful Rumination, and
Affective Traits
Jack W. Berry and Everett L. Worthington, Jr.
Virginia Commonwealth University
Lynn E. O’Connor
The Wright Institute
Les Parrott III
Seattle Pacific University
Nathaniel G. Wade
Virginia Commonwealth University
ABSTRACT Trait forgivingness is the disposition to forgive interper-
sonal transgressions over time and across situations. We define forgive-
ness as the replacement of negative unforgiving emotions with positive,
other-oriented emotions. Rumination has been suggested as a mediator
between forgivingness and emotional outcomes; however, we suggest that
different content of rumination leads to different outcomes after trans-
gressions. In four studies of 179, 233, 80, and 66 undergraduate students,
trait forgivingness was negatively correlated with trait anger, hostility,
neuroticism, fear, and vengeful rumination and was positively correlated
Correspondence should be directed to Jack W. Berry, or Everett L. Worthington,
Jr., Department of Psychology, Virginia Commonwealth University, Richmond, VA
23284-2018 ( jwberry@saturn.vcu.edu or eworth@vcu.edu, respectively).
Jack W. Berry, Department of Psychology; Everett L. Worthington, Jr., Depart-
ment of Psychology; Lynn E. O’Connor, The Wright Institute; Les Parrott III, De-
partment of Psychology; Nathaniel G. Wade, Department of Psychology.
This research was supported by grant
#
239 from the John Templeton Foundation to
Worthington. We gratefully acknowledge their support.
We wish to thank one of the anonymous reviewers for suggesting that we locate a
proxy for fearful or depressive rumination. This helpful suggestion is responsible for
our inclusion of Study 4 in this paper.
Journal of Personality 73:1, February 2005
r Blackwell Publishing 2004
DOI: 10.1111/j.1467-6494.2004.00308.x
with agreeableness, extraversion, and trait empathy. The disposition to
ruminate vengefully mediated the relationship between trait forgivingness
and (1) anger-related traits and (2) both revenge motivations and state
anger following a specific recent transgression, but it did not mediate be-
tween forgivingness and (1) fearfulness and (2) avoidance motivations
following a specific transgress ion. Self-hate statements, a proxy for de-
pressive rumination, mediated the relationship between forgivingness and
both depression and fearfulness but not the relationship between forgiv-
ingness and trait anger. Future research should distinguish the contents of
mental rumination following interpersonal transgressions.
With the growth of the field of positive psychology, researchers are
examining psychological processes that have traditionally been
thought of as ethically desirable or psychologically or socially ben-
eficial (Seligman & Csikszentmihalyi, 2000; Snyder & Lopez, 2002).
Personality researchers have suggested that positive psychology
might give renewed life to the classical notion of the virtues, now
sometimes called ‘‘human strengths’’ or ‘‘positive traits’’ (Aspinwall
& Staudinger, 2003; McCullough & Snyder, 2000).
One positive psychological process that has received attention is
forgiveness (Enright & Fitzgibbons, 2000; McCullough, Pargament,
& Thoresen, 2000; McCullough & Witvliet, 2002; Worthington,
1998a). McCullough, Hoyt, and Rachal (2000) classified research
on forgiveness as either (a) offense-specific (aimed at understanding
forgiveness of a specific person for a single transgression), (b) dyadic
(aimed at understanding forgiveness of a specific person for a history
of transgressions), or (c) dispositional (aimed at understanding for-
giveness as an enduring personality trait). Most research on forgive-
ness has focused on offense-specific or dyadic forgiveness. Several
intervention studies have aimed at promoting acts of forgiveness (for
reviews of this intervention research, see Hargrave & Sells, 1997;
Thoresen, Luskin, & Harris, 1998; Wade & Worthington, 2003).
There have also been studies of forgiveness and related constructs in
experimental social psychology aimed at determining variables that
influence willingness to forgive particular transgressions or trans-
gressors (McCullough, Worthington, & Rachal, 1997; McCullough
et al., 1998; Zechmeister & Romero, 2002).
In the present paper, we discuss forgiveness at the dispositional
level, termed ‘‘forgivingness’’ by Roberts (1995) to distinguish the
disposition from individual acts of forgiveness. This term has gained
184 Berry, Worthington, Jr., O’Connor, et al.
currency among researchers (Berry, Worthington, Parrott, O’Con-
nor, & Wade, 2001; Mullet, Houdbine, Laumonier, & Girard, 1998;
Mullet et al., 2003). Forgivingness can be construed as a human
strength or personality trait with positive consequences for individ-
uals and social relationships. However, it is possible to use forgive-
ness for self-serving motives (see Baumeister, Exline, & Sommer,
1998), or fail to pursue justice by too readily forgiving (Exline, Wor-
thington, Hill, & McCullough, 2003). After describing our under-
standing of forgivingness, we present the results of four studies that
support our conceptualization, and we recommend research that will
help answer some unanswered questions that are raised by the
present findings.
Definitions of Forgiveness and Unforgiveness
There is no consensus definition of forgiveness. Enright has cham-
pioned a definition that emphasizes the interplay of cognition, emo-
tion, and behavior (Enright & Fitzgibbons, 2000). McCullough has
emphasized reductions of motivations of revenge and avoidance over
time and increase of conciliation over time (McCullough, Fincham,
& Tsang, 2003). Snyder has emphasized cognition (Thompson &
Snyder, 2003). Worthington has emphasized emotion (Worthington
& Wade, 1999). All agree that forgiving is a complex phenomenon.
Following Worthington and Wade (1999), we explore forgivingness
as based on emotion.
When an interpersonal transgression occurs, the victim can per-
ceive the transgression as hurtful, offensive, or some mixture of both.
These perceptions are often accompanied by immediate emotional
reactions of anger (to the extent that the transgression is perceived as
an offense; Thoresen et al., 1998) or of fear (to the extent that the
transgression is perceived as hurtful; Worthington & Wade, 1999).
Such immediate and highly charged emotional reactions can occur
even when forgiveness is quickly forthcoming. Sometimes, however,
these immediate emotions are transformed into a more enduring
state of unforgiveness. Unforgiveness is emotionally complex in-
volving such affects as resentment, bitterness, hostility, hatred,
residual anger, fearfulness, and depression (Worthington &
Wade, 1999). The development of unforgiveness is hypothesized
to be brought about by rumination about the nature of a transgres-
sion, its consequences for the victim or the relationship, and the
Forgivingness, Vengeful Rumination, Affective Traits 185
attributions about the motivations of the transgressor (Berry et al.,
2001; Fincham, 2000; McCullough, Bellah, Kilpatrick, & Johnson,
2001). When unforgiveness is generalized across time and situations,
we call it dispositional ‘‘unforgivingness,’’ as an adaptation of Robe-
rts’s (1995) ‘‘forgivingness.’’
Unforgiveness is usually experienced as an unpleasant emotional
state, and people are usually motivated to reduce or overcome it.
Worthington (2001) hypothesized that people may reduce unforgiving
emotions by seeking revenge or legal justice, accepting the injustice,
cognitively reframing the event, consequences, or transgressor’s mo-
tives, denying their emotions, or forgiving the transgression.
We have conceptualized forgiveness as the juxtaposition or su-
perimposition of strong, positive, other-oriented emotions over the
negative emotions of unforgiveness (Worthington & Wade, 1999).
The positive emotions associated with forgiveness might include
love, compassion, empathy, or sympathy for the transgressor (Wor-
thington & Wade, 1999). Other positive emotions, such as humility
about one’s own real or potential culpability (see Tangney, 2000), or
gratitude for one’s own experiences of forgiveness (see McCullough,
Emmons, & Tsang, 2002; Witvliet, Ludwig, & Vander Laan, 2001),
might facilitate reducing or replacing the negative emotions of un-
forgiveness. In Worthington’s model, forgiveness can occur after
unforgiveness has developed, but it can also prevent the development
of unforgiveness by replacing immediate negative emotions.
The Need for Research on Disposi tional Constructs Related
to Forgiveness
Several researchers have stressed the need for research on forgiveness
at the dispositional level (Emmons, 2000; Mauger et al., 1992;
McCullough, 2000; Mullet et al., 1998; Worthington & Wade,
1999). In both the secular and religious literatures on forgiveness,
much of the discussion is implicitly about the disposition to forgive.
Religious doctrines often promote forgiveness as a virtue, and reli-
gious teachings often instill beliefs and recommend methods to
facilitate the proneness to forgive interpersonal transgressions
(Mullet et al., 2003; Pargament & Rye, 1998; Rye et al., 2000;
Thoresen et al., 1998). Literatures on forgiveness have developed
around health effects (Toussaint, Williams, Musick, & Everson,
2001) and physiological mechanisms (Berry & Worthington, 2001;
186 Berry, Worthington, Jr., O’Connor, et al.
Farrow et al., 2001; Lawler et al., 2003; Witvliet et al., 2001; see
Worthington & Scherer, 2004, for a review). When researchers ad-
vocate forgiveness as a means of promoting long-term benefits to
physical health, psychological functioning, and better social adjust-
ment, they are often advocating the promotion of the disposition to
forgive (Kaplan, 1992; Witvliet et al., 2001).
At the level of personality dispositions, our perspective on trait
forgivingness has parallels with recent theories of personality. Many
theories include a trait dimension associated with warmth, affection,
and altruism—traits that we propose are characteristic of forgiving-
ness. Examples are agreeableness in the Big Five model of person-
ality (John, 1990); intimacy motivation in McAdams’s (1994) theory;
and affiliation in the interpersonal circumplex model (Kiesler, 1983).
These theories also include a dimension associated with self-protec-
tion or the detection of threats—traits that might be characteristic of
unforgivingness. For example, such dimensions include neuroticism
in the Big Five model, power motivation in McAdams’s theory, and
control in the interpersonal circumplex model.
In the studies described below, we examined the relationships
among trait forgivingness, the Big Five (especially the affective dis-
positions), and vengeful rumination. In the first study, we assumed
that the construct validity of trait forgiveness had been established
by prior researchers (Ashton, Paunonen, Helmes, & Jackson, 1998;
Berry & Worthington, 2001; Berry et al., 2001; Brose, Rye, & Lutz,
2002; Emmons, 2000; Girard & Mullet, 1997; Macaskill, Maltby, &
Day, 2002; Mauger et al., 1992; McCullough et al., 2001; McCul-
lough & Hoyt, 2002; Mullet et al., 2003; Mullet et al., 1998; Neto &
Mullet, 2003; Symington, Walker, & Gorsuch, 2002; Tangney, Fee,
Reinsmith, Boone, & Lee, 1999; Thompson & Snyder, 2003; Thomp-
son et al., 2003; Walker & Gorsuch, 2002). We aimed primarily at
establishing the construct validity of the Trait Forgivingness Scale
(TFS), which is adapted from a longer scale we employed in previous
research (Berry & Worthington, 2001). The TFS, if found to have
sufficient empirical concurrent validity (based on previous research),
would be a quick measure of trait forgivingness because it is com-
posed of only 10 Likert-type items. We emphasized correlations of
the TFS with the Big Five personality traits.
Despite theorizing that rumination is crucial to unforgiveness and
its diminution to forgiveness, to date, no studies have investigated
the mediation of forgivingness and negative affect by rumination. In
Forgivingness, Vengeful Rumination, Affective Traits 187
the second, third, and fourth studies, we focused on rumination and
its relationship with forgiveness and with unforgiving states and traits.
Rumination about interpersonal hurts and offenses is considered a
crucial factor in establishing and maintaining negative affects such as
hostility and anger. Several studies of forgiveness have assessed ru-
mination explicitly (Ashton et al., 1998; Berry et al., 2001; Brooks &
Toussaint, 2003; Emmons, 1992; Mauger et al., 1992; McCullough et
al., 1998; McCullough et al., 2001; Tangney et al., 1999). In Studies 2
and 3, we tested the hypothesis that the disposition to ruminate
vengefully mediates the relationship between trait forgivingness and
(1) dispositional variables related to unforgiveness (such as trait anger
and chronic hostility) and (2) current responses to a recent transgres-
sion (including state anger and motivations to avoid and retaliate
against the transgressor). In Study 4, we examined different types of
rumination and the mediational role of such rumination.
Study 1: Forgivingness and Af fective Traits
Introduction
Perhaps the most obvious negative emotion associated with unfor-
giveness is anger (Enright, Gassin, & Wu, 1992; Fizgibbons, 1986).
Empirical research has demonstrated that individual acts of forgive-
ness are associated with a reduction in anger over transgressions
(Huang & Enright, 2000; Rye et al., 2001; Seybold, Hill, Neumann,
& Chi, 2001; Weiner, Graham, Peter, & Zmuidinas, 1991). Many
researchers have proposed that, at the dispositional level, trait for-
givingness is likely to be negatively associated with trait anger and
related stable dispositions such as hostility and resentment (Kaplan,
1992; Williams & Williams, 1993). Empirical investigations have
largely supported these hypotheses. Seybold et al. (2001) found that
the disposition to forgive was negatively correlated with trait anger.
Using a five-item, scenario-based measure of forgivingness, Berry
et al. (2001), in three studies of college students, found negative as-
sociations with measures of both trait anger and chronic hostility.
Berry and Worthington (2001) replicated Berry et al. (2001) with 39
adults, using two different measures of trait forgivingness. Tangney
et al. (1999) assessed 285 undergraduates using her Multidimensional
Forgiveness Inventory (MFI). They found forgiving others to be
negatively correlated with ratings of anger, aggression, and vengeance.
188 Berry, Worthington, Jr., O’Connor, et al.
Thompson et al. (2003) used three measures of trait forgivingness in
studies with over 2800 undergraduates. They found trait anger to be
correlated with the forgiving other (FO) subscales of their own Heart-
land Forgiveness Scale (HFS), Tangney et al.’s (1999) MFI, and
Mauger et al.’s (1992) forgivingness scale. In the present study, we
hypothesized that the TFS would correlate negatively with trait anger.
Worthington and Wade (1999) proposed that fear might also lead
to the development of emotional unforgiveness, perhaps in combi-
nation with anger. Fear (or perhaps anxiety) is likely to be the pre-
dominant emotional response when a transgression is perceived as
more hurtful than offensive, and avoidance of the transgressor, rath-
er than revenge, might be the primary motivation in such cases.
Tangney et al. (1999) found a nonsignificant relationship between
her MFS-FO and trait anxiety. Seybold et al. (2001) found a signif-
icant negative relationship between Mauger et al.’s (1992) FO sub-
scale and trait anxiety. Thompson et al. (2003) found significant
correlations between trait anxiety and the FO subscales of the HFS
and MFI, and with Mauger et al.’s (1992) FO. We hypothesize that
individuals who are elevated in dispositional proneness to fearfulness
will be lower in trait forgivingness.
Several investigators have studied the relationship of the Big Five
to forgivingness (for a summary, see Table 1). Researchers have
suggested that the Big Five personality factor of neuroticism should
be negatively related to forgiveness at the dispositional level (Ashton
et al., 1998; McCullough, 2000). Neuroticism is a disposition to ex-
perience negative affects, such as anxiety and depression, and a
proneness to worry ( John, 1990). Several studies have reported that
neuroticism and related traits (such as anxiety and depression) are
negatively associated with trait forgivingness and with forgiveness of
specific historical transgressions (Ashton et al., 1998; Berry et al.,
2001; McCullough & Hoyt, 2002; Seybold et al., 2001; Symington
et al., 2002; Walker & Gorsuch, 2002). Neto and Mullet (2003)
found that although forgivingness was associated with shyness and
embarrassment, it was not associated with self-esteem or loneliness.
They suggest that not all aspects of neuroticism need be related
(negatively) to forgivingness, but only those factors that have an
inherent interpersonal dimension. By examining Table 1, we see that
neuroticism has been consistently negatively correlated with trait
forgivingness (Berry et al., 2001). Even more, though, neuroticism
has been found to be related to negative motivations of revenge
Forgivingness, Vengeful Rumination, Affective Traits 189
Table 1
Summary of Findings in Prior Investigations of Forgiveness and the Big Five Personality Traits
Author (Date)
Participants
and Method
Agree-
ableness
Neurot-
icism
Extra-
version
Conscien-
tiousness
Open-
ness Others
Emmons
(1992)
420
undergrads
and adults;
Beliefs about
Revenge
Questionnaire
(
nn
)(
nn
)(
n
) (ns) (ns) 3 measures
of rumination
all (
nn
)
Ashton
et al.
(1998)
118
undergrads
(69 F, 49 M);
Two
measures of
forgivingness:
(Empathy/
Attachment
(EA),
Forgiveness/
Non-
retaliation
(F/NR)
(
nn
)(
n
)
a
(ns) (ns) (ns) Fretful
rumination
assoc/w
Hi A/Hi N,
but not
Hi A/Low N
Berry et al.
(2001)
Undergrads: TAS
Study 2 (S2):
N 5 292
S2: (
n
) S2: (
n
) S2: (ns) S2: (ns) S2: (ns) S2: (
nn
)
S3: N 5 466 S3: (
nn
)S3(
nn
) S3: (
n
) S3: (
nn
) S3: (ns) S3: (
nn
)
S3: N 5 62 S4: (
n
) S4: (
n
) S4: (ns) S4: (ns) S4: (ns) AQ
TNTF S2: (
nn
)
S3: (
n
)
S4: (
nn
)
DRS:S4(
n
)
McCullough
et al.
(2001)
S1:
Undergrads
N 5 91
S2:
Undergrads
N 5 192
S2: (
n
) S2: (
n
) S2: (ns) S2: (
n
) S2: (ns) S1: At
baseline,
rumination
corrected:
suppression,
Trait
Vengefulness
(not Trait
Forgivingness)
AandRall(
n
).
Changes t1 to
t2; D rum-
D avoid (
n
),
D rum- D rev
(continued)
Brose et al.
(2002)
275
Undergrads
(195 F, 80 M);
240 Catholic;
259
Caucasian;
Three
measures of
forgivingness:
Absence of
Negative
(AN);
Presence of
Positive (PP);
Likelihood of
Forgiving
(LF)
AN: (
nn
) AN: (
nn
) AN: (
n
) AN: (
n
) AN: (ns) Empathy:
PP: (
nn
) PP: (
nn
) PP: (
n
) PP: (ns) PP: (
n
) AN: (ns)
LF: (
nn
) LF: (
nn
) LF:(
n
) LF: (
n
) LF: (ns) PP: (ns)
LF: (ns)
Hostility:
AN: (
nn
)
PP: (ns)
LF: (ns)
Table 1
(Continued )
Author
(Date)
Participants
and Method
Agree-
ableness
Neurot-
icism
Extra-
version
Conscien-
tiousness
Open-
ness Others
Symington
et al.
(2002)
180
undergrads;
Forgiveness of
a person: Big
Five assessed
by organizing
measures
under Big Five
categories
Agreeable (
n
) Emo Stability (
n
) (ns) (ns) (ns)
Warm (
n
) Emotional (
n
) 6 measures 3 measures 3 measures
Sensitive (
n
) Neurot Anxiety (ns)
Distrust (ns)
Intellectual
Openness (
n
)
McCullough
& Hoyt
(2002)
S1: 137
undergrads;
Multiple
TRIMS with
significant
others, same
sex, and other
sex friends;
historical
and
hypothetical
events
S2: 95
undergrads;
145 peer
raters
Self (S1)
Av (ns) Av (
n
) Av (ns) Av (ns) Av (ns)
Rev (
n
) Rev (ns) Rev (ns) Rev (ns) Rev (ns)
Ben (ns) Ben (
n
) Ben (ns) Ben (ns) Ben (ns)
Self (S2)
A(
n
)A(
n
) A (ns) A (ns) A (ns)
R(
n
) R (ns) R (ns) R (ns) R (ns)
B(
n
)B(
n
) B (ns) B (ns) B (ns)
Peer (S2)
A (ns) A (
n
) A (ns) A (ns)
R(
n
)R(
n
) R (ns) R (ns) R (ns)
B(
n
)B(
n
) B (ns) B (ns) B (ns)
Note.
a
The correlation of the EA scale with N was negative.
n
po.05.
nn
po.01.
(McCullough & Hoyt, 2002) and to a vengeful personality (McCul-
lough et al., 2001). Avoidance motivations were even more strongly
positively related to neuroticism in McCullough and Hoyt than were
revenge motivations. McCullough and Hoyt (2002) also found neu-
roticism to be negatively correlated with benevolence motivations.
These patterns generally held when people rated themselves or were
rated by others who knew them. In the present study, we hypoth-
esized that the TFS would correlate negatively with neuroticism.
Forgiveness has been hypothesized to be facilitated by positive
emotions associated with love and altruism (McCullough et al.,
1998). At the offense-specific level of analysis, both empathy and
relationship closeness have been found to increase forgiveness of
specific transgressions (McCullough et al., 1998; McCullough et al.,
1997). We found only three studies of forgiveness and empathy at the
trait level of analysis. Macaskill et al. (2002) surveyed 324 British
undergraduates. They found Mehrabian’s measure of empathy to be
related to forgiving others (but not to forgiving the self). Tangney
et al. (1999) found her MFI-FO to be related to Davis’s (1983) per-
spective-taking and empathic-concern subscales of the Interpersonal
Reactivity Inventory. Brose et al. (2002) failed to find a significant
correlation between their measure of forgivingness and their measure
of empathy. In the present study, we hypothesized that the TFS
would correlate positively with trait empathy (specifically, with em-
pathic concern, though possibly with cognitive perspective taking).
Based on theorizing by Worthington and Wade (1999) and
McCullough et al. (1998), we also hypothesized that the TFS would
correlate positively with the Big Five personality factor of agreea-
bleness. Agreeableness is related to empathy and traits linked to
positive interpersonal relationships, such as kindness, trust, and the
capacity for intimacy (Asendorpf, 1998; Graziano, Jensen-Campbell,
& Hair, 1996). The trait adjective ‘‘forgiving’’ has been associated
with agreeableness in empirical research ( John, 1990). Several stud-
ies have found that agreeableness is positively correlated with the
disposition to forgive (see Table 1; Ashton et al., 1998; Berry et al.,
2001; Brose et al., 2002; Mauger et al., 1992; McCullough & Hoyt,
2002; Symington et al., 2002). McCullough et al. (2001) found that
agreeableness was negatively associated with vengefulness, which
was negatively correlated with forgiveness of a specific transgression.
The relationship between trait forgivingness and other Big Five
factors is less clear. Although significant positive correlations of trait
194 Berry, Worthington, Jr., O’Connor, et al.
forgivingness with extraversion (Brose et al., 2002; Berry et al., 2001;
Emmons, 1992) and conscientiousness (Brose et al., 2002; Berry
et al., 2001, Study 3; McCullough et al., 2001, Study 2) have been
found, the correlations have been weak and unstable across studies.
For openness, only two significant correlations have been found in
23 tests (see Table 1). In the present study, therefore, we made no
hypotheses about the relationship between forgivingness and other
Big Five factors.
METHOD
Participants
Participants were 179 undergraduate students from a mid-Atlantic state
university. The sample included 151 females (84%) and 28 males with a
mean age of 24.3 years (SD 5 8.3). There were 96 European Americans
(54%); 53 African Americans (30%); 15 Asian Americans (8%); 6 Latino/
Latina Americans (3%); 8 Other (4.5%); and 1 nonrespondent (.6%). All
students in the sample participated voluntarily for class credits.
Instruments
Demographic Questionnaire. Participants reported their age, gender,
ethnicity, and frequency of participation in religious activities.
Trait Forgivingness Scale (TFS). The TFS consists of 10 items aimed at
assessing a respondent’s self-appraisal of his or her proneness to forgive
interpersonal transgressions. The TFS was adapted from a 15-item scale
used in an earlier study of trait forgivingness, relationship quality, and
cortisol stress responses (Berry & Worthington, 2001). The items of the
TFS, with directions and scoring procedures, are given in Appendix A. We
conducted a pilot study to examine the psychometric performance of our
10 forgivingness items. The items were analyzed using both classical sta-
tistics and Rasch item-response procedures. The Rasch model employed
was Andrich’s (1978) rating scale model. Details of the psychometric prop-
erties of the TFS obtained in the pilot study and in all studies reported in
the present paper are discussed in Appendix B along with norms for college
students of different gender and ethnicity (see Table B1).
The Trait Anger Scale (TAS; Spielberger, Jacobs, Russell, & Crane,
1983). The TAS is a 15-item self-report scale for assessing anger as a
personality trait in terms of the frequency of angry states experienced over
time. Alpha coefficients among college students were reported as .87 for
Forgivingness, Vengeful Rumination, Affective Traits 195
males and .87 for fema les. Among Navy recruits, alpha coefficients were
.87 for males and .84 for females.
The Big Five Personality Inventory, V44 (BFI-44; John, Donahue, &
Kentle, 1992). The BFI-44 is a 44-item measure of neuroticism, extra-
version, agreeableness, conscientiousness, and openness to experience
( John, 1990). Short phrases are rated on a Likert-type scale according to
how descriptive the phrases are of the respondent. Each subscale consists
of 8 to 10 items. John et al. (1992) estimated internal consistencies for the
subscales ranging from .75 to .88 for self- and peer reports. The subscales
were further validated by peer-peer and peer-self co rrelations that ranged
from .21 for agreeableness to .63 for extraversion.
Interpersonal Reactivity Inventory (IRI; Davis, 1983). The IRI is a 28-
item self-report instrument measuring cognitive and emotional aspects of
empathy. The IRI has four subscales assessing perspective taking (ability
to adopt the perspectives of other people), fantasy (a tendency to identify
with fictional characters in novels, plays, and films), empathic concern
(feelings of warmth and compassion for others), and personal distress
(feelings of anxiety and distress when observing other peoples’ negative
experiences). Internal consistencies of the IRI subscales ranged from
.71 to .77.
The Fear Questionnaire (FQ; Marks & Mathews, 1978). The FQ is a 24-
item self-report measure using a Likert-type format to assess the degree to
which a respondent avoids situations due to fear or other unpleasant
feelings. Factor analyses have shown that items of the FQ can be grouped
into agoraphobia, social phobia, and blood/injury phobia subscales. The
FQ also includes items relating to anxiety and depression. The FQ can
differentiate phobic patients from nonphobics and has been shown to be
sensitive to phobia treatment interventions. The instrument has good one-
week test-retest reliability.
Procedure
Questionnaires were distributed in classes. Students return ed the packets
to designated drop-off areas within a week after distribution. Of 200
packets distributed, 179 (90%) were returned.
RESULTS
In Table 2, we present correlations between the TFS and affective
trait measures. We used a conservative alpha level of .01 because
196 Berry, Worthington, Jr., O’Connor, et al.
of the number of theory-relevant hypotheses we were testing. All
hypotheses were supported. In addition, although neither was hy-
pothesized, extraversion and conscientiousness were positively relat-
ed to trait forgivingness.
DISCUSSION
Trait forgivingness was negatively associated with a variety of emo-
tional traits that have been hypothesized to reflect unforgivingness,
such as anger (Kaplan, 1992), neuroticism (McCullough, 2000), and
fear (Worthington & Wade, 1999). Consistent with our definition of
trait forgivingness, the TFS was correlated with a variety of positive
emotional traits, including agreeableness (Worthington and Wade,
1999), empathic concern (McCullough, 2000), and empathic per-
spective taking (McCullough, 2000). Although we made no predic-
tion about conscientiousness, it too was correlated with
forgivingness. In 22 previous tests, across studies and laboratories
(see Table 1), four correlations (all correlating conscientiousness and
forgivingness) were positive. This additional finding in the present
article might suggest that a weak correlation in fact exists. Exline et
al. (2003) hypothesized that some people might decide to forgive
because, for them, forgiveness is an act that is consistent with their
moral code. Often this moral code is religiously motivated (Enright,
Santos, & Al-Mabuk, 1989; Exline, 2002; Gorsuch & Hao, 1993;
Krause & Ellison, 2003; Krause & Ingersoll-Dayton, 2001; Mullet et
al., 1998; Tsang & McCullough, 2003; Wuthnow, 2000). In such de-
cision-based forgiveness, forgiveness is perhaps motivated by con-
scientiousness rather than by empathy and positive affect. Because
relatively few people treat forgiveness as an exercise in conscien-
tiousness, the correlations in samples of students (as most research to
date has investigated) are likely to be small and unstable.
Interestingly, the TFS was also associated with extraversion in the
present research. Research has shown extraversion to be linked to
happiness and subjective well-being (Pavot, Diener, & Fujita, 1990).
Five of 25 tests in Table 1 have yielded a positive correlation between
forgivingness and extraversion. Our findings in the present study
strengthen the case for a stable, true correlation. Given our defini-
tion of forgivingness as a propensity to overcome the negative emo-
tions of unforgiveness with positive emotions, it is worth considering
Forgivingness, Vengeful Rumination, Affective Traits 197
Table 2
Correlations Between the Trait Forgivingness Scale (TFS) and Affective Trait Scales (Means and Standard
Deviations in Parentheses)
Study 1 Study 2 Study 3
Trait Anger Scale
a
.48
nn
(29.8; 7.8) .45
nn
(30.1; 7.8) .49
nn
(28.3; 7.0)
Big Five Inventory
Agreeableness
a
.52
nn
(3.8; .66) .55
nn
(3.8; .63) .60
nn
(3.9; .56)
Neuroticism
a
.41
nn
(2.9; .76) .25
n
(2.9; .75) .49
nn
(2.8; .73)
Conscientiousness .41
nn
(3.7; .63) .10 (3.5; .62) .23
+
(3.9; .59)
Extraversion .24
n
(3.4; .79) .03 (3.3; .79) .26
+
(3.4; .93)
Openness .16
+
(3.6; .67) .08 (3.6; .60) .01 (3.8; .62)
Interpersonal Reactivity Inventory
Empathic Concern
a
.38
nn
(27.7; 4.3) .24
nn
(27.5; 4.2) .27
n
(22.4; 4.1)
Perspective-taking
a
.36
nn
(23.8; 4.6) .18
n
(21.2; 3.5) .32
n
(20.8; 4.5)
Fear Questionnaire .28
nn
(46.4; 21.3) .44
nn
(49.3; 19.4)
Dissipation-Rumination Scale
a
.60
nn
(34.1; 8.6) .69
nn
(23.5; 10.7)
b
Aggression Questionnaire
Anger
a
.40
nn
(13.9; 4.4)
Hostility
a
.35
n
(18.8; 6.0)
Verbal Aggression .10 (13.1; 3.7)
Physical Aggression .17 (15.2; 6.0)
Note. Study 1 N ranges from 174 to 179; Study 2 N ranges from 228 to 233; Study 3 N ranges from 77 to 80 (except, see note b below).
Means and standard deviations for the TFS are as follows: for Study 1, M 5 33.5, SD 5 7.1; for Study 2, M 5 34.6, SD 5 6.5; for Study 3,
phase 1, M 5 37.6, SD 5 6.2; and for Study 3, phase 2, M 5 38.6, SD 5 5.4.
a
Theory-relevent hypothesized correlations.
b
The correlation between the TFS and DRS is based on 62 participants who completed both scales concurrently in the second phase of Study 3.
+
po.05, ns.
n
po.0 1.
nn
po.001.
whether forgiveness is one mechanism partly responsible for the rel-
ative happiness of extraverted individuals over time.
Study 2: Forgivingness and Vengeful Rumination
Introduction
We have proposed that emotional unforgiveness following a trans-
gression is often the consequence of ruminating negatively about the
transgression. By extension, the disposition to be unforgiving is like-
ly to be associated with a disposition to negative rumination. How-
ever, there are many types of negative rumination. Rumination is
repetitive, affect-laden cognition that is automatic, intrusive, and
usually not productive of adaptive behavior. Rumination has been
associated with post-traumatic stress disorder (PTSD; Horowitz &
Solomon, 1975; Horowitz, Wilner, Kaltreider, & Alvarez, 1980).
Rumination has also been linked to depression (Nolen-Hoeksema,
1987, 1991). Thompson et al. (2003) found depressive rumination,
using a measure by Nolen-Hoeksema and Morrow (1991), to be re-
lated negatively to the HFS FO scale (r 5 .24). Rumination has
long been tied to anxiety ( Janis, 1958; Segerstrom, Tsao, Alden, &
Craske, 2000) and to normal worry (Kroll, Johnson, Egan, Carey, &
Erickson, 2002). Rumination has been found to interfere with
healthy coping and to exacerbate cancer (Baider & De-Nour, 1997;
Epping-Jordan et al., 1999). Rumination has also been implicated in
obsessive-compulsive disorder (Hodgson & Rachman, 1977).
In the present studies, we are most interested in the relationship
between forgivingness and angry rumination (Caprara, 1986; Cap-
rara, Manzi, & Perugini, 1992; Collins & Bell, 1987). A few studies
have examined forgivingness and other types of rumination. For in-
stance, Brooks and Toussaint (2003) studied 283 undergraduates.
Trait forgivingness was weakly and negatively correlated with de-
pression. Depressive rumination was negatively correlated with trait
forgivingness. In a hierarchical multiple regression predicting de-
pression, trait forgiveness, when controlling for forgiveness of self
and state forgiveness, was mediated by depressive rumination and
generic intrusive thoughts, but not by angry rumination. Tangney
et al. (1999) found low dispositional forgivingness of others to be
related to paranoid ideation. Ashton et al. (1998) studied fretful ru-
mination in 118 students. Students high in agreeableness and low in
Forgivingness, Vengeful Rumination, Affective Traits 199
neuroticism had low fretful rumination. Emmons (1992) found that
beliefs about vengeance were positively correlated with three meas-
ures of rumination—the Padua Inventory (Sanavio, 1988), the Maud-
sley Obsessional Thinking Inventory (Hodgson & Rachman, 1977),
and the Rehearsal Scale of the Emotional Control Questionnaire
(Roger & Nesshoever, 1987). McCullough et al. (1998) examined
transgressions in 187 undergraduates. They measured rumination us-
ing the Impact of Events Scale (IES). In their fixed structural model,
rumination was predicted by the closeness of the relationship before
the transgression. Rumination predicted revenge (but not avoidance)
motivations. McCullough et al. (2001) have suggested that some peo-
ple are dispositionally more vengeful than others, and this desire for
vengeance is motivated by ruminative thinking about offenses.
McCullough et al. (2001) studied rumination in 91 students using
the IES. At baseline, rumination was positively related to vengefulness
(r 5 .31), efforts to suppress thoughts (r 5 .62), avoidance motivations
(r 5 .39), and revenge motivations (r 5 .40). At time 2, eight weeks
later, reductions in rumination were related to reductions in thought
suppression, avoidance motivations, and revenge motivations.
Berry et al. (2001) assessed vengeful angry rumination using Cap-
rara’s Dissipation-Rumination Scale (DRS) in 62 undergraduates.
Using a scenario-based measure of trait forgivingness, they found
that forgivingness was negatively correlated with the DRS in an 8-
week prospective study (r 5 .49).
The present study differs from that of McCullough et al. (2001) in
two ways. First, all of our scales measure dispositional traits; second,
our measure of rumination is explicitly concerned with vengeful
rumination. McCullough et al. (2001) used the Impact of Events
Scale (IES; Horowitz, Wilner, & Alvarez, 1979) to assess rumination
about a specific offense. The IES assesses the frequency of intrusive
content (thoughts, images, affect) and attempts to suppress this
content. The precise content of the intrusive thoughts (e.g., vengeful,
fearful, sad) is not assessed. We used the Dissipation-Rumination
Scale (DRS; Caprara, 1986) as a measure of the disposition
to experience ruminative thinking explicitly related to revenge,
retaliation, and the desire to ‘‘get even.’’ McCullough et al. (2001)
used the Transgression-Related Motivations Inventory (TRIM;
McCullough et al., 1998) to assess a state of forgiveness of a specif-
ic transgression (which they conceptualized as a reduction in revenge
and avoidance motivations). In the present study, we used the TFS
200 Berry, Worthington, Jr., O’Connor, et al.
as a measure of the disposition to forgive. We differ from Brooks
and Toussaint (2003) in that, although they used the DRS, they at-
tempted to predict depression using the DRS. We differed from
Emmons (1992) in predicting trait forgivingness instead of beliefs
about vengeance.
In Study 2, we hypothesized that the disposition for vengeful ru-
mination would be positively correlated with trait anger and nega-
tively correlated with trait forgivingness. Furthermore, we
hypothesized—based on Worthington and Wade (1999)—that the
negative association between forgivingness and trait anger would be
mediated by the disposition to vengeful rumination.
METHOD
Participants
Participants were 233 undergraduate students from the same mid-Atlan-
tic state university as in Study 1. The sample included 155 females
(66.5%) and 78 males (33.5%) with a mean age of 19.2 years
(SD 5 2.5). There were 143 European Americans (61.4%); 54 African
Americans (23.2%); 17 Asian Americans (7.3%); 10 Latino/Latina Amer-
icans (4.3%); 5 Other (2.2%); and 4 nonrespondents (1.7%). All students
participated voluntarily for class credits.
Instruments
Participants completed all scales used in Study 1 except the Fear Ques-
tionnaire (inadvertently left out of the packets). In addition, they complet-
ed the Dissipation-Rumination Scale (DRS; Caprara, 1986). The DRS is a
20-item, self-report scale used to assess an individual’s tendency to think
and act aggressively following an interpersonal offense. Of the 20 Likert-
type items, 5 are ‘‘control’’ items that are not scored. Dissipation and ru-
mination are conceptualized as opposite ends of a single continuum. Dis-
sipation reflects the tendency to ‘‘shrug off’’ insults or offenses with little
rumination. Rumination reflects the tendency toward an increasing desire
for retaliation over time. Items on the DRS include ‘‘When I am outraged,
the more I think about it, the angrier I become,’’ ‘‘I can remember very well
the last time I was insulted,’’ ‘‘Sometimes I can’t sleep because of a wrong
done to me,’’ and ‘‘I still remember the offenses I have suffered, even after
many years.’’ Caprara (1986) infers that an increasing tendency to retaliate
must mean that the person is ruminating over the transgression. The in-
ternal consistency of the English version of the scale was estimated to be
.87. The scale has been validated using structural models and correlations
Forgivingness, Vengeful Rumination, Affective Traits 201
with such variables as tolerance toward violence, irritability, emotional
susceptibility, and fear of punishment (Caprara et al., 1992). Validity has
also been supported in experimental studies in which high and low scorers
were given the opportunity to retaliate against research confederates who
insulted them (Caprara, 1986).
Procedure
Questionnaires were distributed in classes. Students return ed the packets
to designated drop-off areas within a week after distribution. Of 300
packets distributed, 233 (78%) were returned.
RESULTS
Means and standard deviations, as well as bivariate correlations be-
tween the TFS and all other dispositional scales, are shown in Table
2. All a priori hypotheses from Study 1 were replicated in Study 2. As
hypothesized, the TFS was significantly negatively correlated with
the DRS. Also as hypothesized, the DRS was significantly positively
correlated with trait anger, r (228) 5 .60, po.001.
To test our central hypothesis that vengeful rumination would
mediate the relationship between trait forgivingness and trait anger,
we used the regression analysis strategy described by Baron and
Kenny (1986). The negative correlation between trait forgivingness
(independent variable) and trait anger (dependent variable) was sta-
tistically significant (as shown previously in Table 2). The negative
correlation between trait forgivingness and the DRS (potential
mediator) was statistically significant (also shown in Table 2). In
Table 3, we present the results of the hierarchical regression predict-
ing trait anger from forgivingness alone at Step 1, and from forgiv-
ingness and the DRS together at Step 2. The change in R-squared
from Step 1 to Step 2 was statistically significant, suggesting that the
DRS added predictive variance above that of trait forgivingness
alone. In addition, the negative association between trait forgiving-
ness and trait anger dropped substantially when controlling
for vengeful rumination, suggesting a mediating role for vengeful
rumination. Finally, we calculated a Sobel (1982) statistic to test
whether the DRS was a potential mediator of the relationship be-
tween forgivingness and trait anger (see Table 3). The Sobel test was
statistically significant.
202 Berry, Worthington, Jr., O’Connor, et al.
Table 3
Mediation Analyses for Studies 2, 3, and 4: Step 1 and Step 2 of
Hierarchical Regressions Predicting Negative Trait Variables and
Responses to a Specific Transgression (Means and Standard
Deviations of State Variables in Parentheses)
Criterion
Variables: Step 1 R
2
Step 2 Partial r
Change
in R
2
Sobel
Test
Study 2
Trait Anger
Scale
TFS .20
nnn
TFS .12
DRS .53
nnn
.18
nnn
6.42
nnn
Study 3 (Trait Variables)
Trait Anger
Scale
TFS .24
nnn
TFS .06
DRS .55
nn
.14
nn
3.07
nn
AQ-Anger TFS .16
nnn
TFS .06
DRS .62
nnn
.18
nnn
3.23
nn
AQ-Hostility TFS .12
nn
TFS .24
DRS .26
.08
n
1.97
n
Study 3 (State Variables)
State Anger
Scale (20.9; 7.69)
TFS .10
nn
TFS .07
DRS .36
n
.07
n
2.18
n
TRIM Revenge
(7.3; 3.88)
TFS .13
nn
TFS .06
DRS .44
nn
.10
nn
2.84
nn
TRIM Avoidance
(16.2; 9.07)
TFS .10
nn
TFS .32
n
DRS .01
.01 .07
Study 4
BDI TFS .06
n
TFS .01
Self-hate .62
nnn
.32
nnn
3.02
nn
(continued)
Forgivingness, Vengeful Rumination, Affective Traits
203
DISCUSSION
As in Study 1, and consistent with the seven studies reviewed in Ta-
ble 1, the results of Study 2 found that dispositional forgivingness
was negatively associated with negative affective traits (trait anger
and neuroticism) and positively associated with positive affective
traits (empathy and agreeableness). Unlike Study 2, there was no
association between forgivingness and either extraversion or consci-
entiousness. The instability of these associations is similar to those
reported in Table 1 in a variety of studies.
The association of trait forgivingness with vengeful rumination is
especially important for our theoretical perspective on forgiveness.
There is evidence that rumination about negative events can perpet-
uate negative emotions triggered by those events (Greenberg, 1995).
Angry, vengeful rumination about transgressions has been linked to
many cognitive and emotional variables tied to aggression (Caprara
et al., 1992; Collins & Bell, 1997). Berry et al. (2001) found a negative
correlation between vengeful rumination and trait forgivingness. The
most important finding of Study 2 is the strong mediating role of
vengeful rumination on the association between trait forgivingness
Table 3
(Continued )
Criterion
Variables: Step 1 R
2
Step 2 Partial r
Change
in R
2
Sobel
Test
Trait Anger Scale TFS .33
nnn
TFS .43
nnn
Self-hate .37
nn
.11
nn
1.89
Fear Questionnaire TFS .12
nn
TFS .26
n
Self-hate .33
nn
.10
nn
2.64
nn
Note. The correlations between the DRS (potential mediator) and dependent var-
iables in the studies are as follows. In Study 2, the correlation with trait anger was
r(218) 5 .59, po.001. In Study 3, the correlation with the TAS was r(59) 5 .59,
po.001; with AQ-Anger, r(60) 5 .57, po.001; with AQ-Hostility, r(59) 5 .44,
po.001; with State Anger, r(61) 5 .41, po.01; with Revenge, r(60) 5 .48, po.001;
and with Avoidance, r(60) 5 .25, po.05. In Study 4, the correlation of self-hate
(potential mediator) with the BDI was r(64) 5 .62, po.001; with trait anger,
r(64) 5 .52, po.001; and with the Fear Questionnaire, r(64) 5 .54, po.001.
n
po.05.
nn
po.01.
nnn
po.001.
204 Berry, Worthington, Jr., O’Connor, et al.
and trait anger. We have argued that vengeful rumination plays a
causal role in promoting unforgiveness and in inhibiting forgiveness
(Worthington & Wade, 1999). The results of Study 2 are consistent
with our hypothesis that forgivingness acts to inhibit trait anger at
least partly through its association with lower levels of vengeful ru-
mination (Worthington, 2000; Worthington & Wade, 1999). These
findings can also be interpreted as an extension of the findings of
McCullough et al. (2001) to the level of dispositional traits. If this
turns out to be a stable finding upon replication, it would imply that
future researchers might need to differentiate the roles of angry
rumination, depressive rumination (Brooks & Toussaint, 2003;
Thompson et al., 2003), anxious rumination or worry (Sergerstrom
et al., 2000), obsessional thinking (Emmons, 1992), intrusive
thoughts (McCullough et al., 2001), and other types of rumination.
Each type of rumination might produce different outcomes. For ex-
ample, rumination might lead to vengeance for people with vengeful
dispositions (Emmons, 1992; McCullough et al., 2001), but might
lead to depression in people who are prone to self-condemnation
(Brooks & Toussaint, 2003; Mauger et al., 1992; Tangney et al.,
1999; Thompson et al., 2003). Rumination that interferes with peo-
ple’s ability to cope with cancer might be of a completely different
type (Epping-Jordan et al., 1999). We thus turn to determining
whether the current findings replicate and to extending the reasoning
to forgiveness of discrete transgressions.
Study 3: Forgivingness, Vengeful Rumination, and Responses
to a Specific Transgression
Introduction
We had two main aims in Study 3. First, we attempted to replicate
the trait-level findings of Study 2. Second, we wanted to determine
whether trait forgivingness and the disposition to ruminate venge-
fully would predict responses to a recent specific transgression
(McCullough et al., 1998).
We also included a measure of chronic hostility to test its asso-
ciation with trait forgivingness and vengeful rumination. Hostility is
a generalized negative attitude toward others and an aggressive pre-
paredness. In previous research using a scenario-based measure of
forgivingness, Berry et al. (2001) found that hostility was correlated
Forgivingness, Vengeful Rumination, Affective Traits 205
weakly with trait forgivingness in only two of their three studies.
Hostility was not as strongly or consistently related to trait forgiv-
ingness as was trait anger. That result was unexpected. Hostility
seems intuitively to be the kind of negative attitude that we attribute
to unforgiveness, and chronically hostile individuals (including those
with Type A personality patterns) have often been characterized as
unforgiving (Kaplan, 1992; Williams, 1989). Seybold et al. (2001)
found a stronger correlation between trait forgivingness (using the
measure by Mauger et al., 1992) and hostility. In two currently un-
published studies, Brose et al. (2002) and Tangney et al. (1999) found
that hostility was negatively correlated with the absence of negativity
in their forgivingness measures.
In the present study, we tested the hypothesis that chronic hostility
would correlate negatively with trait forgivingness and positively with
vengeful rumination. We hypothesized that vengeful rumination
would mediate the relationship between forgivingness and hostility
(as it would the relationship between forgivingness and trait anger).
We also hypothesized that unforgiving responses to a recent
transgression (state anger, revenge motivations, and avoidance moti-
vations) would be negatively correlated with trait forgivingness and
positively correlated with the disposition to vengeful rumination.
This would constitute a near replication of the results of McCul-
lough et al. (2001), except that in the present study, rumination is
assessed as a dispositional construct, and the content of the rumi-
nation is explicitly related to revenge and retaliation. McCullough
et al. (1998) found that rumination about a specific transgression
was related to revenge but not avoidance on the TRIM. Finally, we
hypothesized that the negative associations between trait forgiving-
ness and current unforgiving responses to a specific, real transgres-
sion would be mediated by the disposition to vengeful rumination.
METHOD
Participants
The study was conducted in two phases. In the first phase, participants
were 80 undergraduate students, 62 females (77.5%) and 18 males
(22.5%), from a private university in the Pacific Northwest. This sample
was previously reported in Berry et al. (2001), but data on the TFS and
responses to a specific transgression were not used at that time. The mean
206 Berry, Worthington, Jr., O’Connor, et al.
age of the parti cipants was 19.2 years (SD 5 1.5). The sample included 66
European Americans (82.5%); 5 Asian Americans (6.3%); 3 Latina/La-
tino Americans (3.8%); and 6 Others (7.5%). All participated voluntarily
for class credits.
In the second phase, participants were 62 (of 80) students from the first
phase who completed a packet of questionnaires 2 months later. There
were 50 females (81%) and 12 males (19%). The mean age of the par-
ticipants was 18.9 years (SD 5 .83). The sample included 53 European
Americans (84%); 4 Asian Americans (6.5%); 2 Latino/Latina Americans
(3%); and 4 Others (6.5%).
Instruments
At the beginning of the semester, all participants completed the TFS,
TAS, BFI, IRI, and a demographic form, plus the Fear Questionnaire
used in Study 1. Participants also completed the Aggression Question-
naire (AQ; Buss & Perry, 1992). The AQ is a 29-item self-report ques-
tionnaire to assess four dimensions of aggression: anger, hostility, verbal
aggression, and physical aggression. The validity of the AQ subscales has
been supported by correlations with such variables as competition, as-
sertiveness, and impulsiveness, and by correlations of peer ratings of ag-
gressiveness. The internal consistency of the AQ total scale has been
estimated to be .89. The internal consistencies for the subscales ranged
from .72 to .85. Buss and Perry (1992) report 9-week test-retest reliabil-
ities for the subscale scores between .72 to .80, and for the total scale
scores at .80.
Near the end of the semester, a subset of 62 participants also com-
pleted the DRS, the TAS (for the second time), and a questionnaire con-
cerning a recent, self-identified transgression. The Specific Transgressions
Survey asked the respondent to describe, in an open-ended format, a sit-
uation in which someone did something to upset them greatly, something
that was offensive or hurtful to them, since the beginning of the semester.
After describing the transgression, participants completed two scales that
assessed their current feelings about the transgressor and their motivat-
ions to avoid the person and seek revenge against the person. The State
Anger Sc ale (SAS; Spielberger et al., 1983) is a widely used, valid measure
of the intensity of current feelings of anger. In the present study, partic-
ipants were instructed to complete the scale with reference to how they
felt as they reflected on the recent transgression they had described. The
TRIM (McCullough et al., 1998) is a 12-item, self-report scale for
assessing avoidance of and revenge toward a transgressor. The TRIM
has adequate internal consistency and moderate 3-week and 9-week test-
retest reliabilities. Construct validity of the TRIM has been supported by
Forgivingness, Vengeful Rumination, Affective Traits 207
correlations with a variety of relationship variables (e.g., closeness and
relationship satisfaction), offense-specific variables (degree of apology
from a transgressor), empathy, rumination about an offense, and for-
giveness. In the present study, participants were instructed to complete
the scale with reference to the person involved in the recent transgression
they described.
Procedure
Questionnaires were distributed in classes and returned to the experi-
menter the following class meeting, two days later. At the beginning of the
semester, of 100 questionnaire packets distributed, 80 were returned. Two
months later, of the 80 participants who completed the first questionnaire
packet, 62 (78 %) completed the second packet, returning them 2 days
after distribution. All participated voluntarily for class credits.
RESULTS
In Table 2, we present the means and standard deviations for all trait
variables and the concurrent bivariate correlations between the TFS,
affective trait scales, and vengeful rumination (DRS). Results were
similar to those in the previous studies for the affective traits (see
Table 2). As hypothesized, the TFS was negatively and significantly
correlated with the Hostility subscale of the AQ (as well as with the
Anger subscale). Replicating Study 1, the scores on the TFS were
negatively correlated with scores on the Fear Questionnaire. The
TFS and DRS were significantly correlated both concurrently (both
scales administered in the second phase of the study, r 5 .69,
shown in Table 2) and predictively (TFS from phase 1, DRS from
phase 2, two months later), r(60) 5 .77, po.001.
We used regression analyses to test whether the DRS mediated the
relationship between the TFS and affective trait scales (TAS and the
Anger and Hostility subscales of the AQ). The correlations between
the TFS (independent variable) and the dependent variables were all
statistically significant (as shown in Table 2). The correlation be-
tween the TFS and DRS (potential mediator) was statistically sig-
nificant (see Table 2). In Table 3, we present the results of
hierarchical regressions predicting the dependent variables from
the TFS alone (Step 1) and from the TFS and DRS together (Step
2). For each of the dependent variables, there was a significant
change in the R-squared from Step 1 to Step 2. Furthermore, in each
208 Berry, Worthington, Jr., O’Connor, et al.
case, the relationship between the TFS and the criterion was greatly
reduced (and no longer statistically significant) when controlling for
the DRS, suggesting mediation by vengeful rumination. In addition,
the Sobel tests for mediation were statistically significant (see Table
3). We should emphasize that the criterion variables (TAS, AQ-An-
ger, and AQ-Hostility) had been administered two months before the
predictor variables (TFS and DRS, which had been completed in
phase 2 of the study). These results support the predictive stability of
the trait scales we employed.
We also used regression analyses to test whether the DRS medi-
ated the relationships between the TFS and the three unforgiving
responses to a specific transgression (State Anger, Revenge, and
Avoidance). The correlations between the TFS (independent varia-
ble) and dependent variables were r(60) 5 .32, po.05, for State
Anger; r(60) 5 .36, po.01, for Revenge; and r(60) 5 .32, po.05,
for Avoidance. As noted above, the negative correlation between
the TFS and DRS was statistically significant. In Table 3 we present
the results of the hierarchical regressions predicting the dependent
variables from the TFS alone (Step 1) and from the TFS and DRS
(Step 2). For State Anger and Revenge, but not for Avoidance,
the partial correlations with the TFS were reduced substantially.
In addition, the Sobel tests for mediation were statistically sig-
nificant for State Anger and Revenge but not for Avoidance (see
Table 3).
We included the Fear Questionnaire (FQ) among our measures of
affective traits. We explored whether vengeful rumination might
mediate the relationship between forgivingness and fear (as it did
with anger and hostility), though this was not part of our original set
of hypotheses. As shown in Table 2, trait forgivingness (independent
variable) was significantly negatively correlated with the FQ (de-
pendent variable) and the DRS (potential mediator). The FQ was
positively correlated with vengeful rumination, r(60) 5 .25, po.05.
We used a hierarchical regression predicting fear from trait forgiv-
ingness at step 1 and from both forgivingness and vengeful rumina-
tion at step 2. Vengeful rumination added little to the prediction of
fear over that predicted by forgivingness alone (R-squared
change 5 .004, p 5 .63). Furthermore, the partial correlation of
forgivingness with fear ( .43) changed little, while the partial
correlation with vengeful rumination ( .09) was not statistically
significant. The Sobel test ( .50) was not statistically significant.
Forgivingness, Vengeful Rumination, Affective Traits 209
DISCUSSION
Our hypothesized relationships between the TFS and affective trait
variables were supported, including the negative correlation between
the TFS and chronic hostility. The hypotheses that vengeful rumi-
nation would mediate the relationships between the TFS and the
unforgiving dispositions of trait anger and chronic hostility were also
supported. The results also indicated that vengeful rumination me-
diated the relationship between trait forgivingness and two current
unforgiving responses to a specific transgression: state anger and
motivations for revenge.
Vengeful rumination did not mediate the relationship between
forgivingness and motivations to avoid the transgressor. This is con-
sistent with McCullough et al. (1998), who found a similar result at
the state level using structural equation modeling to test a series of
alternative models.
We also found (in an unplanned analysis) that vengeful rumina-
tion did not mediate the relationship between forgivingness and fear.
In retrospect, we believe that these latter results are consistent with
Worthington’s (1998b) proposal that responses to a transgression
can be either fearful or depressive (when a transgression is perceived
as hurtful) or angry (when a transgression is perceived as morally
offensive) and that motivations for revenge or avoidance might de-
pend on the relative magnitudes of these responses. In future re-
search, fearful rumination might be investigated as a potential
mediator of the relationship between forgivingness and fear.
Study 4: Forgivingn ess, Depression, and Self-Hate
The previous studies have provided evidence that vengeful rumina-
tion mediates the relationship between forgivingness and both (a)
anger-related affective traits and (b) current anger and revenge moti-
vations stemming from a recent specific transgression. Vengeful ru-
mination, however, did not appear to mediate the relationship of
forgivingness with (a) fearfulness or (b) avoidance motivations
(which we have hypothesized to be related to fearful and/or depres-
sive rumination). These findings suggest that the content of rumi-
nation about transgressions might determine the affective and
motivational responses to transgressions. Using a hierarchical
multiple regression, Brooks and Toussaint (2003) found that the
210 Berry, Worthington, Jr., O’Connor, et al.
disposition to depressive rumination did not significantly alter the
relationship between trait forgivingness and depression or between
trait self-forgiveness and depression. They found that the relation-
ship between forgivingness and depression was reduced only when
controlling for state forgiveness, trait forgiveness of self, and both
depressive rumination and generic intrusive thoughts.
In the present study, we used a sample previously described in
Berry et al. (2001). In that study, we administered a pool of 48 ex-
ploratory items that included the TFS as a subset. In addition, par-
ticipants completed the Trait Anger Scale, the Fear Questionnaire,
the Beck Depression Inventory, and the Interpersonal Guilt Ques-
tionnaire-67 (O’Connor, Berry, Weiss, Bush, & Sampson, 1997). The
IGQ-67 includes a subscale of self-hate, which, for the present study,
we considered a potential proxy for depressive rumination.
We hypothesized that forgivingness would be negatively associ-
ated with depression. Furthermore, we hypothesized that this rela-
tionship would be mediated by the propensity to ruminate with
negative self-statements associated with depression. We also exam-
ined whether self-hate statements mediated the relationship between
forgivingness and (a) trait anger and (b) fearfulness. We did not ex-
pect self-hate to mediate these relationships.
METHOD
Participants
Participants wer e 66 students—49 females (74%) and 17 males (26%)—
from an introductory psychology course at a state university in the San
Francisco Bay Area. The mean age of participants was 19.9 years
(SD 5 3.7). There were 33 Asian Americans (50%); 9 European Ameri-
cans (13.6%); 13 Latino/Latina Americans (19.7%); 2 African Americans
(3%); 7 Others (10.6%); and 2 nonresponses (3%).
Instruments
In addition to the TFS, Trait Anger Scale, and Fear Questionnair e
(described above), participants completed the Interpersonal Guilt
Questionnaire-67 (IGQ-67; O’Connor et al., 1997). The IGQ-67 is a
67-item, Likert-type scale that assesses three types of guilt related to wor-
ry about harming others: survivor guilt (guilt over being better off than
others), separation guilt (guilt over being different from others), and om-
nipotent responsibility for the well-being of others. The IGQ -67 also
Forgivingness, Vengeful Rumination, Affective Traits 211
includes a 16-item self-hate scale that includes such statements as ‘‘I al-
ways assume I am at fault when something goes wrong,’’ ‘‘If I fail at
something I condemn myself and want to harm myself,’’ and ‘‘If some-
thing bad happens to me I feel I must have deserved it.’’ In previous re-
search, the self-hate scale has correlated highly with depression
(O’Connor, Berry, & Weiss, 1999; O’Connor, Berry, Weiss, & Gilbert,
2002) and with depressive automatic thoughts (O’Connor, Berry, Weiss,
Schweitzer, & Sevier, 2000). Participants also completed the Beck De-
pression Inventory (BDI: Beck, 1972), a wid ely used, 21-item, self-report
depression inventory representing cognitive, affective, and vegetative
symptoms of depression. The reliability and validity of the BDI have
been studied extensively (Beck, Steer & Garbin, 1988).
Procedure
Questionnaire packets were distributed to 100 students after class. Of the
100 distributed, 66 questionnaires were returned one week later at the
next class meeting. Participants received class credits for returning their
completed packets.
RESULTS
The TFS (independent variable) was correlated significantly (nega-
tively) with the dependent variables: trait anger, r(64) 5 .58,
po.001; fearfulness, r(64) 5 .35, po.01; and depression, r(64) 5
.25, po.05. The TFS was also significantly correlated with the
potential mediator, Self-hate, r(64) 5 .40, po.01.
Table 3 presents the results of the hierarchical regressions pre-
dicting the dependent variables from the TFS alone (Step 1) and
from the TFS and Self-hate together (Step 2). Self-hate contributed
to a significant increase in the R-squared in predicting all three de-
pendent variables. The partial correlation of forgivingness with de-
pression was reduced to .01 when controlling for self-hate, and the
Sobel test was statistically significant, suggesting, as hypothesized,
that self-hate statements mediated the relationship between forgiv-
ingness and depression. The partial correlations between forgiving-
ness and both trait anger and fearfulness remained moderate and
statistically significant. The Sobel test for trait anger was not statis-
tically significant; the Sobel test for fearfulness, however, was sta-
tistically significant.
212 Berry, Worthington, Jr., O’Connor, et al.
DISCUSSION
Study 4 provides evidence that the propensity to make self-hate state-
ments, which we used as a proxy for depressive rumination, mediates
the negative association between trait forgivingness and depression.
This finding is contrary to the results of Brooks and Toussaint (2003),
who used a different measure of depressive rumination. The present
study also suggests that self-hate statements did not mediate the neg-
ative relationship between forgivingness and trait anger, but it might
at least partially mediate the relationship between forgivingness and
fearfulness. These results add to our hypothesis that differential con-
tent of rumination following transgressions plays a role in determin-
ing the affective and motivational responses to transgressions.
GENERAL DISCUSSION
Consistent with our emotional replacement hypothesis, the disposition
to forgive was positively correlated with several traits linked to positive
and pro-social affects (agreeableness, empathic concern, perspective-
taking, and extraversion) and was negatively correlated with a variety
of variables related to negative affects (neuroticism, trait anger, hostil-
ity, depression, and fear). There were also strong negative associations
(concurrently and prospectively) between forgivingness and the ten-
dency to ruminate vengefu lly following transgressions.
These studies also provide evidence that vengeful rumination has
a mediating role in the relationships between trait forgivingness and
anger-related unforgiving dispositional traits (hostility and trait an-
ger). This mediation was also shown for anger and revenge moti-
vations surrounding a recent specific transgression.
However, we also found that vengeful rumination did not mediate
the relationship between forgivingness and (1) fearfulness or (2) cur-
rent avoidance motivations following a recent transgression. More-
over, we found that self-hate statements (employed as a proxy for
depressive rumination) appeared to mediate the relationship between
forgivingness and depression, and perhaps fearfulness, but not trait
anger. Worthington (1998b) has emphasized that transgressions can
be perceived as hurtful or offensive to various degrees. Perhaps the
more hurtful the transgression, the more fearful or depressive the
emotional reaction and the stronger the motivation to avoid a trans-
gressor. The more offensive the transgression, the more angry the
Forgivingness, Vengeful Rumination, Affective Traits 213
affective reaction and the stronger the desire for revenge. The present
studies suggest that the content of rumination (determined by the
nature of the transgression and individual differences in affective
tendencies) could be an important factor in shaping subsequent af-
fects and motivations following a transgression.
Even though the results of the present research were stable across
several studies, there are potential limitations to the generalizability
of the findings and the theoretical inferences that can be drawn from
the studies. All participants were college students, and only the sam-
ple in Study 4 could be considered ethnically diverse (and it was rel-
atively small). Furthermore, the studies were correlational in nature,
and all except one were based on concurrent predictions of self-re-
port scales. We also acknowledge the limitations in drawing firm
conclusions about mediation with disposition-level variables. Medi-
ation is a complex causal process, and with dispositional measures, it
is difficult to capture this process very well.
Although the TFS has demonstrated adequate psychometric
properties across four studies, it has not been assessed for its rela-
tionship with social desirability. Our scenario-based measure of trait
forgivingness (TNTF) had a negligible association with social desir-
ability (Berry et al., 2001). However, the TFS, though correlated
with the TNTF (see Appendix B), uses a different test format, and it
is therefore unknown to what extent socially desirable response sets
have influenced the present research results.
Another limitation is the sampling we employed for measures.
Namely, there might be some degree of conceptual overlap between
the DRS and measures of vengeance. Some of the items on the DRS
resemble items that tap vengeance, and although the TFS does not
inquire about vengeance, the TRIM has a revenge subscale. In ad-
dition, our measure of self-hate only approximated depressive ru-
mination. These studies merit replication with other instruments
before firm conclusions are possible.
Despite these limitations, we believe the present research has
raised interesting questions. We hope in future studies to extend
these results, examining more directly the nature of ruminative
tendencies that might contribute to depressive and fearful affect
and avoidance motivations following transgressions.
In addition, we believe more research is needed around the
precise role of rumination in determining affect. Does vengeful rumi-
nation simply increase negative affect? Or might it also focus the vic-
214 Berry, Worthington, Jr., O’Connor, et al.
tim on the transgression and the negative qualities of the
offender, thus inhibiting empathy for the offender? McCullough
et al. (1997) and McCullough et al. (1998) have demonstrated
empirically the important role of empathy in forgiving (see also Maca-
skill et al., 2002; Tangney et al., 1999). McCullough et al. (1998) found
that empathy was not affected substantially when a more general, less
affect-focused measure of state rumination was used. But would that
result hold if vengeful rumination were considered?
We also think it would be worthwhile to study the spontaneous
post-transgression coping responses of highly forgiving individuals
(see Worthington & Scherer, 2004). Does forgiveness simply involve
a lack of rumination following transgressions? Is this process simply
one of inhibiting thoughts that lead to unforgiveness?
We have focused on the content of rumination throughout the
present article. Potentially, the controllability of rumination might
also affect whether unforgiveness develops.
If rumination is broadly understood to mean increased effortful
mental work following a negative event, we think it more likely that
forgiving individuals engage in their own form of rumination fol-
lowing transgressions. The content of this rumination would be
aimed not only at thwarting the development of hostile and cold
attitudes, but at reestablishing or maintaining positive and loving
attitudes toward transgressors. In a meta-analysis of forgiveness in-
tervention research, Worthington et al. (2000) found a strong rela-
tionship between intervention effect sizes and the amount of time
spent in trying to promote forgiveness. Most forgiveness interven-
tions not only attempt to suppress negative rumination, they also
promote empathy, gratitude for past forgiveness of one’s own trans-
gressions, and other efforts, which we might call ‘‘forgiving rumi-
nation,’’ that support the kinds of positive emotional replacement
described by Worthington and Wade (1999). We suggest that in fu-
ture research and theorizing it could be important to make distinc-
tions among a variety of forms of rumination (broadly understood):
vengeful rumination, fearful rumination, depressive rumination, and
even forgiving rumination.
REFERENCES
Andrich, D. (1978). A rating formulation for ordered response categories. Psy-
chometrika, 43, 561–573.
Forgivingness, Vengeful Rumination, Affective Traits 215
Asendorpf, J. B. (1998). Personality effects on social relationships. Journal of
Personality and Social Psychology, 74, 1531–1544.
Ashton, M. C., Paunonen, S. V., Helmes, E., & Jackson, D. N. (1998). Kin al-
truism, reciprocal altruism, and the Big Five personality factors. Evolution and
Human Behavior, 19, 243–255.
Aspinwall, L. G., & Staudinger, U. M. (2003). A psychology of human strengths:
Some central issues of an emerging field. In L. G. Aspinwall & U. M.
Staudinger (Eds.), A psychology of human strengths: Fundamental questions
and future directions for a positive psychology (pp. 9–22). Washington, DC:
APA Books.
Baider, L., & De-Nour, A. K. (1997). Psychological distress and intrusive thoughts
in cancer patients. Journal of Nervous and Mental Disease, 185, 346–348.
Baron, R., & Kenny, D. (1986). The moderator—mediator variable distinction in
social psychological research: Conceptual, strategic and statistical considera-
tions. Journal of Personality and Social Psychology, 51, 1173–1181.
Baumeister, R. F., Exline, J. J., & Sommer, K. L. (1998). The victim role, grudge
theory, and two dimensions of forgiveness. In E. L. Worthington Jr. (Ed.),
Dimensions of forgiveness: Psychological research and theological perspectives
(pp. 79–104). Philadelphia: Templeton Foundation Press.
Beck, A. T. (1972). Measuring depression: The depression inventory. In T. A.
Williams, M. M. Katz & J. A. Shields (Eds.), Recent advances in the psycho-
biology of the depressive illnesses (pp. 299–302). Washington, DC: U. S. Gov-
ernment Printing Office.
Beck, A. T., Steer, R. A., & Garbin, M. G. (1988). Psychometric properties of the
Beck Depression Inventory: Twenty-five years of evaluation. Clinical Psychol-
ogy Review, 8, 77–100.
Berry, J. W., & Worthington, E. L., Jr. (2001). Forgivingness, relationship quality,
stress while imagining relationship events, and physical and mental health.
Journal of Counseling Psychology, 48, 447–455.
Berry, J. W., Worthington, E. L., Jr, Parrott, L., O’Connor, L. E., & Wade, N. G.
(2001). Dispositional forgivingness: Development and construct validity of the
Transgression Narrative Test of Forgivingness (TNTF). Personality and Social
Psychology Bulletin, 27, 1277–1290.
Brooks C. W., & Toussaint, L. (2003). The relationship between forgiveness and de-
pression: Rumination as a link. Unpublished manuscript, Idaho State University.
Brose, L. A., Rye, M. S., & Lutz, C. J. (2002, August). Forgiveness and the five-
factor model of personality Poster presented at the 110th Annual Convention of
the American Psychological Association, Chicago.
Buss, A. H., & Perry, M. (1992). The Aggression Questionnaire. Journal of Per-
sonality and Social Psychology, 63, 452–459.
Caprara, G. V. (1986). Indicators of aggression: The Dissipation-Rumination
Scale. Personality and Individual Differences, 7, 763–769.
Caprara, G. V., Manzi, J., & Perugini, M. (1992). Investigating guilt in relation
to emotionality and aggression. Personality and Individual Differences, 13,
519–532.
Collins, K., & Bell, R. (1997). Personality and aggression: The Dissipation-Ru-
mination Scale. Personality and Individual Differences, 22, 751–755.
216 Berry, Worthington, Jr., O’Connor, et al.
Davis, M. H. (1983). Measuring individual differences in empathy: Evidence for a
multidimensional approach. Journal of Personality and Social Psychology, 44,
113–126.
Emmons, R. A. (1992, August). Revenge: Individual differences in correlates.
Poster presented at the annual meeting of the American Psychological Asso-
ciation, Washington, D.C.
Emmons, R. A. (2000). Personality and forgiveness. In M. E. McCullough, K. I.
Pargament, & C. E. Thoresen (Eds.), Forgiveness: Theory, research, and prac-
tice (pp. 156–175). New York: Guilford.
Enright, R. D., & Fitzgibbons, R. P. (2000). Helping clients forgive: An empirical
guide for resolving anger and restoring hope. Washington, DC: APA Books.
Enright, R. D., Gassin, E. A., & Wu, C. (1992). Forgiveness: A developmental
view. Journal of Moral Education, 21, 99–114.
Enright, R. D., Santos, M. J., & Al-Mabuk, R. (1989). The adolescent as forgiver.
Journal of Adolescence, 12, 95–110.
Epping-Jordan, J. E., Compas, B. E., Osowiecki, D. M., Oppedisano, G., Ger-
hardt, C., & Primo, K., et al. (1999). Psychological adjustment in breast cancer:
Processes of emotional distress. Health Psychology, 18, 315–326.
Exline, J. J. (2002). Stumbling blocks on the religious road: Fractured relation-
ships, nagging vices, and the inner struggle to believe. Psychological Inquiry ,
13, 182–189.
Exline, J. J., Worthington, E. L. Jr., Hill, P. C., & McCullough, M. E. (in
press). Forgiveness and perceived justice. Personality and Social Psychology
Review.
Farrow, T. F. D., Zheng, Y., Wilkinson, I. D., Spence, S. A., Deakin, J. F. W.,
Tarrier, N., et al. (2001). Investigating the functional anatomy of empathy and
forgiveness. Neuroreport, 12, 2433–2438.
Fincham, F. D. (2000). The kiss of the porcupines: From attributing responsibility
to forgiving. Personal Relationships, 7, 1–23.
Fitzgibbons, R. P. (1986). The cognitive and emotive uses of forgiveness in the
treatment of anger. Psychotherapy, 23, 629–633.
Girard, M., & Mullet, E. (1997). Propensity to forgive in adolescents, young
adults, older adults, and elderly people. Journal of Adult Development, 4,
209–220.
Gorsuch, R. L., & Hao, J. Y. (1993). Forgiveness: an exploratory factor analysis
and its relationships to religious variables. Review of Religious Research, 34,
333–347.
Graziano, W. G., Jensen-Campbell, L. A., & Hair, E. C. (1996). Perceiving in-
terpersonal conflict and reacting to it: The case for agreeableness. Journal of
Personality and Social Psychology, 70, 820–835.
Greenberg, M. A. (1995). Cognitive processing of traumas: The role of intrusive
thoughts and reappraisals. Journal of Applied Social Psychology, 25,
1262–1296.
Hargrave, T. D., & Sells, J. N. (1997). The development of a forgiveness scale.
Journal of Marital and Family Therapy, 23, 41–62.
Hodgson, R. J., & Rachman, S. (1977). Obsessional-compulsive complaints. Be-
havioural Research and Therapy, 15, 389–395.
Forgivingness, Vengeful Rumination, Affective Traits 217
Hope, D. (1987). The healing paradox of forgiveness. Psychotherapy, 24,
240–244.
Horowitz, M. J., & Solomon, G. F. (1975). A prediction of delayed stress response
syndromes in Vietnam veterans. Journal of Social Issues, 31, 67–80.
Horowitz, M. J., Wilner, N., & Alvarez, W. (1979). Impact of Event Scale: A
measure of subjective stress. Psychosomatic Medicine, 41, 209–218.
Horowitz, M. J., Wilner, N., Kaltreider, N., & Alvarez, W. (1980). Signs and
symptoms of posttraumatic stress disorder. Archives of General Psychiatry, 37,
85–92.
Huang, S. T., & Enright, R. D. (2000). Forgiveness and anger-related emotions in
Taiwan: Implications for therapy. Psychotherapy, 37, 71–79.
Janis, I. (1958). Psychological stress. New York: Wiley and Sons.
John, O. P. (1990). The ‘‘Big Five’’ factor taxonomy: Dimensions of personality
in the natural language and in questionnaires. In L. A. Pervin (Ed.), Handbook
of personality: Theory and research (pp. 66–100). New York: Guilford.
John, O. P., Donahue, E. M., & Kentle, R. (1992). The Big-Five Inventory-Ver-
sions 4a and 54. Technical Report, Institute of Personality and Social Research,
University of California, Berkeley, CA.
Kaplan, B. H. (1992). Social health and the forgiving heart: The Type B story.
Journal of Behavior Medicine, 15, 3–14.
Kiesler, D. J. (1983). The 1982 Interpersonal Circle: A taxonomy for comple-
mentarity in human transactions. Psychological Review, 90, 185–214.
Krause, N., & Ellison, C. G. (2003). Forgiveness by God, forgiveness of others,
and psychological well-being in late life. Journal for the Scientific Study of
Religion, 42, 77–93.
Krause, N., & Ingersoll-Dayton, B. (2001). Religion and the process of forgive-
ness in late life. Review of Religious Research, 42, 252–276.
Kroll, J., Johnson, M., Egan, E., Carey, K., & Erickson, P. (2002). Moral conflict
as a component of ordinary worry. Psychological Reports, 90, 997–1006.
Lawler, K. A., Younger, J. W., Pferi, R. L., Billington, E., Jobe, R., Edmondson,
K., et al. (2003). A change of heart: Cardiovascular correlates of forgiveness
in response to interpersonal conflicts. Journal of Behavioral Medicine,
26, 373–393.
Linacre, J. M. (2003). Size vs. significance: Standardized chi-square fit statistic.
Rasch Measurement Transactions, 17, 918.
Macaskill, A., Maltby, J., & Day, L. (2002). Forgiveness of self and others and
emotional empathy. Journal of Social Psychology, 142, 663–665.
Marks, I. M., & Mathews, A. M. (1978). Brief standard self-rating for phobic
patients. Behaviour Research and Therapy, 17, 263–267.
Mauger, P. A., Perry, J. E., Freeman, T., Grove, D. C., McBride, A. G., &
McKinney, K. E. (1992). The measurement of forgiveness: Preliminary re-
search. Journal of Psychology and Christianity, 11, 170–180.
McAdams, D. P. (1994). The person: An introduction to personality psychology
(2nd ed.). Fort Worth, TX: Harcourt, Brace.
McCullough, M. E. (2000). Forgiveness as human strength: Theory, measure-
ment, and links to well-being. Journal of Social and Clinical Psychology, 19,
43–55.
218 Berry, Worthington, Jr., O’Connor, et al.
McCullough, M. E., Bellah, C. G., Kilpatrick, S. D., & Johnson, J. L. (2001).
Vengefulness: Relationships with forgiveness, rumination, well-being, and the
Big Five. Personality and Social Psychology Bulletin, 27, 601–610.
McCullough, M. E., Emmons, R. A., & Tsang, J. (2002). The grateful disposition:
A conceptual and empirical topography. Journal of Personality and Social
Psychology, 82, 112–127.
McCullough, M. E., Fincham, F. D., & Tsang, J. (2003). Forgiveness,
forbearance, and time: The temporal unfolding of transgression-related inter-
personal motivations. Journal of Personality and Social Psychology, 84,
540–557.
McCullough, M. E., & Hoyt, W. T. (2002). Transgression-related motivational
dispositions: Personality substrates of forgiveness and their links to the Big
Five. Personality and Social Psychology Bulletin, 28, 1556–1573.
McCullough, M. E., Hoyt, W. T., & Rachal, K. C. (2000). What we know (and
need to know) about assessing forgiveness constructs. In M. E. McCullough,
K. I. Pargament, & C. E. Thoresen (Eds.), Forgiveness: Theory, research, and
practice (pp. 65–88). New York: Guilford.
McCullough, M. E., Pargament, K. I., & Thoresen, C. E. (2000). Forgiveness:
Theory, research, and practice. New York: Guilford.
McCullough, M. E., Rachal, K. C., Sandage, S. J., Worthington, E. L., Jr, Brown,
S. W., & Hight, T. L. (1998). Interpersonal forgiving in close relationships II:
Theoretical elaboration and measurement. Journal of Personality and Social
Psychology, 75, 1586–1603.
McCullough, M. E., & Snyder, C. R. (2000). Classical source of human strength:
Revisiting an old home and building a new one. Journal of Social and Clinical
Psychology, 19, 1–10.
McCullough, M. E., & Witvliet, C. V. O. (2002). The psychology of forgiveness.
In C. R. Snyder & S. Lopez (Eds.), Handbook of Positive Psychology (pp. 446–
458). New York: Oxford University Press.
McCullough, M. E., Worthington, E. L., Jr., & Rachal, K. C. (1997). Interper-
sonal forgiveness in close relationships. Journal of Personality and Social Psy-
chology, 75, 321–326.
Mullet, E., Houdbine, A., Laumonier, S., & Girard, M. (1998). ‘‘Forgivingness’’:
Factor structure in a sample of young, middle-aged, and elderly adults. Eu-
ropean Psychologist, 3, 289–297.
Mullet, E., Barros, J., Frongia, L., Usai, V., Neto, F., & Shafighi, S. R. (2003).
Religious involvement and the forgiving personality. Journal of Personality, 71,
1–19.
Neto, F., & Mullet, E. (2003). Personality, self-esteem, and self-construal as cor-
relates of forgivingness. European Journal of Personality, 18, 15–30.
Nolen-Hoeksema, S. (1987). Sex differences in unipolar depression: Evidence and
theory. Psychological Bulletin, 101, 259–282.
Nolen-Hoeksema, S. (1991). Responses to depression and their effects on the du-
ration of depressive episodes. Journal of Abnormal Psychology, 100, 569–582.
Nolen-Hoeksema, S., & Morrow, J. (1991). A prospective study of depression and
posttraumatic stress symptoms after a natural disaster: The 1989 Loma Prieta
earthquake. Journal of Personality and Social Psychology, 61, 115–121.
Forgivingness, Vengeful Rumination, Affective Traits 219
O’Connor, L., Berry, J., & Weiss, J. (1999). Interpersonal guilt, shame and psy-
chological problems. Journal of Social and Clinical Psychology, 18, 181–203.
O’Connor, L. E., Berry, J. W., Weiss, J., Bush, M., & Sampson, H. (1997). In-
terpersonal guilt: The development of a new measure. Journal of Clinical Psy-
chology, 53, 73–89.
O’Connor, L. E., Berry, J. W., Weiss, J., Schweitzer, D., & Sevier, M. (2001). Sur-
vivor guilt, submissive behavior and evolutionary theory: the down-side to win-
ning in social comparison. British Journal of Medical Psychology, 73, 519–530.
O’Connor, L. E., Berry, J. W., Weiss, J., & Gilbert, P. (2002). Guilt, fear,
submission, and empathy in depression. Journal of Affective Disorders, 71,
19–27.
Pargament, K. I., & Rye, M. S. (1998). Forgiveness as a method of religious
coping. In E. L. Worthington, Jr. (Ed.), Dimensions of forgiveness: Psycholog-
ical research and theological perspectives (pp. 59–78). Philadelphia: Templeton
Foundation Press.
Pavot, W., Diener, E., & Fujita, F. (1990). Extraversion and happiness. Person-
ality and Individual Differences, 11, 1299–1306.
Rasch, G. (1960). Probabilistic models for some intelligence and attainment tests.
Chicago: MESA Press.
Roberts, R. C. (1995). Forgivingness. American Philosophical Quarterly , 32, 289–
306.
Roger, D., & Nesshoever, W. (1987). The construction and preliminary validation
of a scale for measuring emotional control. Personality and Individual Differ-
ences, 8, 527–534.
Rye, M. S., Pargament, K. I., Ali, M. A., Beck, G. L., Dorff, E. N., & Hallisey, C.,
et al. (2000). Religious perspectives on forgiveness. In M. E. McCullough, K. I.
Pargament, & C. E. Thoresen (Eds.), Forgiveness: Theory, research, and prac-
tice (pp. 17–40). New York: Guilford.
Rye, M. S., Loiacono, D. M., Folck, C. D., Olszewski, B. T., Heim, T. A., &
Madia, B. P. (2001). Evaluation of the psychometric properties of two for-
giveness scales. Current Psychology: Developmental, Learning, Personality, So-
cial, 20, 260–277.
Sanavio, E. (1988). Obsessions and compulsions: The Padua Inventory. Behaviour
Research and Therapy, 26, 169–177.
Segerstrom, S. C., Tsao, J. C. I., Alden, L. E., & Craske, M. G. (2000). Worry and
rumination: Repetitive thought as a concomitant and predictor of negative
mood. Cognitive Therapy and Research, 24, 671–688.
Seligman, M. E. P., & Csikszentmihalyi, M. (2000). Positive psychology: An in-
troduction. American Psychologist, 55, 5–14.
Seybold, K. S., Hill, P. C., Neumann, J. K., & Chi, D. S. (2001). Physiological and
psychological correlates of forgiveness. Journal of Psychology and Christianity,
20, 250–259.
Snyder, C. R., & Lopez, S. J. (2002). Handbook of positive psychology. New York:
Oxford University Press.
Sobel, M. E. (1982). Asymptotic intervals for indirect effects in structural equa-
tions models. In S. Leinhart (Ed.), Sociological methodology 1982 (pp. 290–
312). San Francisco: Jossey-Bass.
220 Berry, Worthington, Jr., O’Connor, et al.
Spielberger, C. D., Jacobs, G., Russell, S., & Crane, R. S. (1983). Assessment of
anger: The State-Trait Anger Scale. In J. N. Butcher & C. D. Spielberger
(Eds.), Advances in personality assessment (Vol. 2, pp. 161–189). Hillsdale, NJ:
Lawrence Erlbaum Associates.
Symington, S. H., Walker, D. F., & Gorsuch, R. L. (2002). The relationship of
forgiveness and reconciliation to five and sixteen factors of personality. Journal
of Psychology and Christianity, 21, 141–150.
Tangney, J. P. (2000). Humility: Theoretical perspectives, empirical findings,
and directions of future research. Journal of Social and Clinical Psychology, 19,
70–82.
Tangney, J., Fee, R., Reinsmith, C., Boone, A. L., & Lee, N. (1999, August).
Assessing individual differences in the propensity to forgive. Paper presented at
the annual meeting of the American Psychological Association, Boston, MA.
Thompson, L. Y., & Synder, C. R. (2003). Measuring forgiveness. In S. J. Lopez
& C. R. Snyder (Eds.), Positive psychological assessment: A handbook of models
and measures (pp. 301–312). Washington, DC: American Psychological Asso-
ciation.
Thompson, Y. L., Snyder, C. R., Hoffman, L., Michael, S. T., Rasmussen, H. N.,
& Billings, L. S., et al. (2003). Dispositional forgiveness of self, others, and
situations. Unpublished manuscript, University of Kansas-Lawrence.
Thoresen, C. E., Luskin, F., & Harris, A. H. S. (1998). Science and forgiveness
interventions: Reflections and recommendations. In E. L. Worthington, Jr.
(Ed.), Dimensions of forgiveness: Psychological research and theological spec-
ulations. (pp. 163–192). Philadelphia: Templeton Foundation Press.
Toussaint, L. L., Williams, D. R., Musick, M. A., & Everson, S . A. (2001). For-
giveness and health: Age differences in a U.S. probability sample. Journal of
Adult Development, 8, 249–257.
Tsang, J., & McCullough, M. E. (2003). Measuring religious constructs: A hier-
archical approach to construct organization and scale selection. In S. J. Lopez
& C. R. Snyder (Eds.), Positive psychological assessment: A handbook of models
and measures (pp. 345–360). Washington, DC: American Psychological Asso-
ciation.
Wade, N. G., & Worthington, E. L., Jr. (2003). Overcoming interpersonal off-
enses: Is forgiveness the only way to deal with unforgiveness? Journal of Coun-
seling and Development, 81, 343–353.
Walker, D. F., & Gorsuch, R. L. (2002). Forgiveness within the Big Five per-
sonality model. Personality and Individual Differences, 32, 1127–1138.
Weiner, B., Graham, S., Peter, O., & Zmuidinas, M. (1991). Public confession and
forgiveness. Journal of Personality, 59, 281–312.
Williams, R. (1989). The trusting heart. New York: Times Books.
Williams, R., & Williams, V. (1993). Anger kills. New York: Random House.
Witvliet, C., van, O., Ludwig, T. E., & Vander Laan, K. L. (2001). Granting for-
giveness or harboring grudges: Implications for emotion, physiology, and
health. Psychological Science, 12, 117–123.
Worthington, E. L., Jr. (1998a). (Ed.). Dimensions of forgiveness: Psychological
research and theological perspectives. Philadelphia: Templeton Foundation
Press.
Forgivingness, Vengeful Rumination, Affective Traits 221
Worthington, E. L., Jr. (1998b). The Pyramid Model of Forgiveness: Some inter-
disciplinary speculations about unforgiveness and the promotion of forgive-
ness. In E. L. Worthington, Jr. (Ed.), Dimensions of forgiveness: Psychological
research and theological perspectives (pp. 107–137). Philadelphia: Templeton
Foundation Press.
Worthington, E. L., Jr. (2000). Is there a place for forgiveness in the justice sys-
tem? Fordham Urban Law Journal, 27, 1721–1734.
Worthington, E. L., Jr. (2001). Unforgiveness, forgiveness, and reconciliation in
societies. In Raymond G. Helmick & Rodney L. Petersen (Eds.), Forgiveness
and reconciliation: Religion, public policy, and conflict transformation (pp. 161–
182). Philadelphia: Templeton Foundation Press.
Worthington, E. L., Jr., Kurusu, T. A., Collins, W., Berry, J. W., Ripley, J. S., &
Baier, S. N. (2000). Forgiving usually takes time: A lesson learned by studying
interventions to promote forgiveness. Journal of Psychology and Theology, 28,
3–20.
Worthington, E. L., Jr., & Scherer, M. (2004). Forgiveness as an emotion focused
coping strategy that can reduce health risks and promote health resilience:
Theory, review and hypotheses. Psychology and Health, 19, 385–405.
Worthington, E. L., Jr., & Wade, N. G. (1999). The social psychology of unfor-
giveness and forgiveness and implications for clinical practice. Journal of Social
and Clinical Psychology, 18, 385–418.
Wuthnow, R. (2000). How religious groups promote forgiving: A national study.
Journal for the Scientific Study of Religion, 39, 125–139.
Zechmeister, J. S., & Romero, C. (2002). Victim and offender accounts of inter-
personal conflict: Autobiographical narratives of forgiveness and unforgive-
ness. Journal of Personality and Social Psychology, 82, 675–686.
Appendix A
Trait Forgivingness Scale
Directions: Indicate the degree to which you agree or disagree with
each statement below by using the following scale:
1=strongly disagree
2=mildly disagree
3=agree and disagree equally
4=mildly agree
5=strongly agree
——— 1. People close to me probably think I hold a grudge too
long.
——— 2. I can forgive a friend for almost anything.
222 Berry, Worthington, Jr., O’Connor, et al.
——— 3. If someone treats me badly, I treat him or her the
same.
——— 4. I try to forgive others even when they don’t feel guilty
for what they did.
——— 5. I can usually forgive and forget an insult.
——— 6. I feel bitter about many of my relationships.
——— 7. Even after I forgive someone, things often come back
to me that I resent.
——— 8. There are some things for which I could never forgive
even a loved one.
——— 9. I have always forgiven those who have hurt me.
——— 10. I am a forgiving person.
Scoring
To score the TFS such that higher scores reflect higher trait forgiv-
ingness, first reverse score items 1, 3, 6, 7, and 8. After items are
reverse scored, add the 10 items to get the total score.
Appendix B
Psychometric Properties of the TFS
The 10-item TFS is a subset of a 15-items scale used in Berry and
Worthington (2001). In the present studies, items were analyzed us-
ing both classical item statistics and Rasch scaling procedures (Ra-
sch, 1960). Cronbach’s alpha coefficients were .80, .78, .79, and .74
for Studies, 1, 2, 3, and 4, respectively. Corrected item-total corre-
lations for all items, across all studies, ranged from .30 to .63.
The Rasch measurement model employed was Andrich’s (1978)
rating scale model. Rasch scaling is used to estimate a person’s
probable response to a test item based on both (a) the degree to
which the person possesses the latent trait being measured (in this
case, forgivingness) and (b) the location or ‘‘difficulty’’ of the test
item, from easy to endorse to hard to endorse. Rasch scaling locates
each test item along a linear continuum and uses this continuum as a
‘‘yardstick’’ with which to measure test respondents on the variable
of interest. A set of items that fit the Rasch model can be considered
to form a reasonably unidimensional, linear measurement scale. The
Forgivingness, Vengeful Rumination, Affective Traits 223
Rasch scaling yields three useful indices for assessing item and test
quality: (a) person separation reliability, (b) item separation relia-
bility, and (c) mean square fit statistics.
The Rasch person separation reliability is an upper limit of the
proportion of person trait variance not attributable to measurement
error (i.e., the proportion of ‘‘true’’ variance), generally interpreted
by the same standards as Cronbach’s alpha coefficient. Rasch person
separation reliabilities were .81, .79, .79, and .76 across the studies.
The Rasch item separation reliability is an estimate of the pro-
portion of true item variance, which reflects both the spread of item
calibrations and their standard errors. A large item reliability (great-
er than .90) indicates that test items are separated sufficiently in dif-
ficulties, with acceptably small estimation errors, to construct a
useful scale for measuring the variable of interest. Rasch items sep-
aration reliabilities were .95, .97, .96, and .90 across the studies.
Rasch weighted mean-square fit statistics estimate the degree to
which each item is contributing to the construction of a unidimen-
sional scale. These fit statistics have an expected value of 1.00. Linacre
(2003) suggests that values less than 1.50 indicate a productive item.
Across the four studies, item mean-square fit statistics ranged
from .60 to 1.38. We also calculated all pairwise correlations between
the item location (‘‘difficulty’’) estimates obtained in each of the
separate studies. The correlations ranged from .81 to .95 (all po.01),
suggesting that the TFS items form a relatively stable continuum of
item difficulties. In Study 3, we administered the TFS on two occa-
sions, separated by 8 weeks. The test-retest reliability estimate was
r(60) 5 .78, po.001. We also calculated the Rasch item difficulties
separately based on scores obtained at both test times. The correla-
tion between item difficulty estimates on the two occasions was .98,
po.001.
To provide further evidence of validity for the TFS, we asked
partners in romantic relationships to complete the TFS as it applied
to themselves and to their relationship partners. We could thus cal-
culate the correlation between self and other ratings on trait forgiv-
ingness as measured by the TFS. We also administered a brief,
scenario-based scale of forgivingness, the Transgression Narrative
Test of Forgivingness (TNTF; Berry et al., 2001) as an additional
check on the concurrent construct validity of the TFS. Partners
also completed a survey about a specific offense or hurtful event
in their relationship that included questions about the nature and
224 Berry, Worthington, Jr., O’Connor, et al.
severity of the offense, the transgressor’s actions, and their own re-
sponses to the transgression, including forgiveness. In Berry et al.
(2001), we reported only the self-partner correlation for the TNTF
(r 5 .60). Participants were 54 undergraduates (28 males, 26 females)
at an urban, mid-Atlantic state university. Mean age of participants
was 24.4 years (SD 5 5.5). Participants were couples in romantic re-
lationships, with 4 of the couples married and 24 in dating relation-
ships (at least three dates). The correlation between the self-ratings
on the TFS and the other-ratings (made by romantic partners) was
also statistically significant, r(51) 5 .35, po.01.
The Pearson correlation between the self-rating on the TFS and
the TNTF, which presumably measures the same construct by sam-
pling five scenarios, was moderate and statistically significant,
r(49) 5 .50, po.001. In the sample for Study 4 reported above, the
correlation between the TFS and TNTF was r(64) 5 .48, po.001.
Table B1 presents normative TFS total raw score data by gender
and ethnicity. Detailed results of the Rasch scaling procedures, in-
cluding item difficulty estimates and logit measures corresponding to
each total raw score, are available from the first author upon request.
Table B1 (Appendix B)
(Appendix B) Norms for College Students Based on Five Samples
at Public Mid-Atlantic, Public Bay-Area, and Private Religious
Universities
Males Females
n M SD n M SD
European Americans 90 36.3 7.03 244 34.9 7.47
African Americans 33 34.2 5.04 98 36.0 6.34
Asian Americans 17 33.2 6.26 51 31.4 6.55
Latino/Latin Americans 11 30.4 6.84 24 31.3 5.37
Forgivingness, Vengeful Rumination, Affective Traits
225
226
