FORGIVENESS AND HEALTH: REVIEW AND REFLECTIONS ON A MATTER OF FAITH, FEELINGS, AND PHYSIOLOGY.

Empirical research on the links between forgiveness and both mental and physical health is burgeoning. This article reviews current research, with reflections on how Christians might engage this literature. It considers Christian and psychological conceptualizations of forgiveness, reviews the published literature on forgiveness and mental and physical health, addresses theoretical and interpretive issues, and reflects on ways that Christians may thoughtfully consider the contributions and limitations of empirical research on forgiveness and health.

While forgiveness has long been at the heart of the Christian faith, empirical research has only recently examined its links with the emotional heart and the cardiovascular system. Several factors have encouraged both therapists and researchers to focus on forgiveness and health. One has been Lewis Smedes' (1984) Forgive and Forget: Healing the Hurts We Don't Deserve. Smedes masterfully made the ethereal concept of forgiveness concrete, framing it in ways that stirred the imaginations of theorists, therapists, and theologians alike. A second has been Robert Enright's 1994 founding of the International Forgiveness Institute at the University of Wisconsin-Madison. Enright and colleagues have published several empirical studies on forgiveness and health (Al-Mabuk, Enright, & Cardis, 1995; Coyle & Enright, 1997; Freedman & Enright, 1996; Hebl & Enright, 1993), as well as books designed to help people forgive (Enright, 2001; Enright & Fitzgibbons, 2000). A concurrent and third influence has been the program of for giveness research conducted by Everett Worthington, Jr., Michael McCullough, Steven Sandage, and colleagues who have published forgiveness intervention research (e.g., McCullough & Worthington, 1995; McCullough, Worthington, & Rachal, 1997; see Worthington, Sandage, & Berry, 2000) and four books that address mental and physical health within broader examinations of forgiveness (McCullough, Sandage, & Worthington, 1997; McCullough, Pargament, & Thoresen, 2000; Worthington, 1998a, 2001). A fourth factor has been the Campaign for Forgiveness Research, initiated by the John Templeton Foundation in 1998. Having funded dozens of proposals, this competitive grants program has the promise to multiply the number of published studies on the links between forgiveness and both mental and physical health related variables.

The burgeoning body of forgiveness and health research raises important questions for Christians to consider. First, how is forgiveness defined? Second, what are the published empirical findings on forgiveness and mental and physical health? Third, in light of this research, what additional theoretical and interpretive issues should be considered? Fourth, how might Christians thoughtfully consider the contributions and limitations of empirical research on forgiveness and health? This article will address these four questions in turn.

COMPLEMENTARY UNDERSTANDINGS OF FORGIVENESS

Christian understandings of forgiveness begin with Scripture. The word forgiveness brings to mind memorable biblical texts, such as the prodigal son, Jesus' command to forgive seventy times seven, and the parable of the unmerciful servant. Christian understandings of forgiveness are rooted in the transforming message of the Hebrew scriptures and the New Testament. Forgiveness is at the center of the gospel message and shapes Christian identity. It permeates practices of piety, and is emphasized in the Lord's Prayer, Christian creeds, and the sacraments. When efforts to embody forgiveness as Christian communities and individuals fail, we repent and ask for God's forgiveness. Forgiveness is a gift of God's grace, even as it simultaneously involves our own choices and responses as granters and receivers of forgiveness. Our choices and responses directly involve the spiritual, social, cognitive, emotional, behavioral, and physiological aspects of our selves (all of which are integrally related). All people bear the Imago Dei and have the capacity to be forgiving. Forgiveness is so necessary because all people are affected by the fall and experience brokenness in their relationships with God and each other.

Whereas religious understandings are broad and evocative, scientific definitions are of necessity circumscribed and concrete. Already, the nascent sub-discipline of forgiveness research includes a variety of definitions of forgiveness. Enright and colleagues have articulated the cognitive, emotional, and behavioral elements of forgiveness in view of its philosophical and theological foundations (e.g., Enright & Coyle, 1998; Enright, Freedman, & Rique, 1998). It is essential to distinguish forgiveness from other concepts. Forgiveness does not involve forgetting, ignoring, denying, overlooking, excusing, minimizing, tolerating, exonerating, or condoning. Further, it is distinct from reconciling [1], and it does not replace the role of justice. Rather, it involves a two-pronged response: releasing the negative feelings toward the offender, and enacting merciful responses toward the wrongdoer. According to North (1987), forgiveness happens when the victim can "view the wrongdoer with compassion, benevolence, and love while recognizing that he has willfully abandoned his right to them" (p. 502). As research related to forgiveness proliferates, we will likely encounter other researchers who base forgiveness definitions solely on a set of conciliatory behaviors, an approach that leaves room for primate and cross-species studies. While we may learn much from animal studies (as we have about the psychophysiology of fear, for example), we need human research to understand the involvement and interaction of the spiritual, cognitive, emotional, overt behavioral, social, and physical dimensions of granting and receiving forgiveness. Indeed, even when we study humans, we will need multiple studies using multiple methodologies and clear definitions to best understand forgiveness. Worthington (1998b) has noted that forgiveness is so complex that it warrants a variety of investigations to assess the effects of single and repetitive offenses, the act of forgiveness and the trait of forgivingness, the experience of forgiveness fro m God, others, and oneself, the psychological and physical correlates of forgiveness, and the ways that forgiving differs from reconciling, exonerating, or condoning. So far, the focus of empirical research on forgiveness and health has been on broad understandings of forgiveness and its embodiment across people with different faith commitments and experiences. Most of this research has focused on the perspective of the person granting forgiveness to other people.

PUBLISHED RESEARCH ON FORGIVENESS AND HEALTH

Although several published sources have included reviews of the broader forgiveness literature (e.g., Enright & Fitzgibbons, 2000; McCullough, Exline, and Baumeister, 1998; McCullough, Pargament, & Thoresen, 2000; Worthington, 1998a; Worthington, Sandage, & Berry, 2000), this article aims to examine the current state of published empirical investigations of forgiveness and both mental and physical health with reflections on how Christians might engage this literature.

What Do We Know About Forgiving and Mental Health?

The empirical forgiveness literature on mental health had modest beginnings in the 1960s. In what appears to be the first reported study addressing forgiveness and mental health, Emerson (1964) used a Q-sort method to examine the associations of feeling forgiven and psychological well-being. Emerson found that when people's scores indicated better current adjustment, their scores also indicated a stronger sense of being forgiven. Likewise, their sense of what they ought to be in terms of adjustment paralleled the forgiveness they believed they ought to feel. Emerson's work suggests that emotional adjustment and perceived forgiveness are closely related. Although this investigation lacked sophistication and modern inferential statistics, it was probably the first scientific inquiry into the association of forgiveness with mental health and well-being, a topic that would not be studied empirically for another three decades.

In 1993, Hebl & Enright published the first in a wave of studies using contemporary statistical methods to test the efficacy of forgiveness interventions for improving mental health. The authors focused on 24 elderly women who identified a particular, painful forgiveness issue (e.g., children who did not visit them as hoped, marital conflict), and were not grieving a major loss at the time. The women were randomly assigned to either an 8-week forgiveness intervention group or a control group, which discussed topics raised by its members. The forgiveness group therapy stimulated higher scores than the control group on measures of forgiveness and willingness to forgive. Interestingly, though, both groups showed improvements in anxiety and depression scores. Thus, better mental health (on these measures) was not solely attributable to the forgiveness intervention. However, when participants from both groups were combined, higher levels of forgiveness were associated with higher levels of self-esteem and lower l evels of anxiety and depression. So, while this research does not show improved mental health only for people in the forgiveness intervention, it does indicate that participants in the forgiveness intervention showed enhanced forgiving. In addition, regardless of which group people were part of, those with higher forgiveness scores also had better mental health in terms of self-esteem, depression, and anxiety scores.

Drawing on this intervention research, Al-Mabuk et al. (1995) conducted two studies that focused on the effects of a group intervention designed to help college students forgive their parents for perceived love deprivation. In Study I, participants in a forgiveness group were compared to those in a human relations group. Forgiveness group participants were brought to the point of deciding and committing to forgive. They showed more hope and willingness to forgive after the 4-session, 2-week program, even though their scores of actual forgiving were not greater than controls. Study II compared a group forgiveness intervention and a human relations group (both met 6 times in 6 weeks). This time, the forgiveness group was brought to the point of granting forgiveness, and they showed significant improvements in forgiveness, hope, and attitude toward parents. Compared to the controls, they also showed higher self-esteem, but did not differ on measures of depression or anxiety. Similar to the approach of Hebl and Enright (1993), these researchers examined the data from all participants across studies. This revealed that higher reports of forgiveness of parents were associated with higher self-esteem, more positive views of parents, and lower anxiety and depression, replicating and extending the forgiveness and mental health connection previously found by Hebl and Enright (1993).

Subsequently, Freedman and Enright (1996) applied a forgiveness intervention to survivors of incest. The 12 adult women in this study had survived incest that included physical contact by a male relative two or more years before. To control for factors that could affect treatment outcome, pairs of women were matched on demographic and abuse history variables. One woman from each pair was randomly assigned to the one-on-one treatment, and the other to a wait-list control group. For those in the treatment condition, therapy was terminated once forgiveness was granted, so the length of therapy differed across women (with an average of 14 months). Compared to pre-treatment scores, treated women showed higher increases on measures of forgiveness and hope, and greater decreases on measures of anxiety and depression, than wait-listed women. After each woman in the experimental condition completed treatment, her wait-listed match entered treatment. When wait-listed women completed the forgiveness intervention, they also showed improvements compared to their own pre-treatment scores on the mental health and self-esteem measures. Benefits of the forgiveness intervention were still evident in the original treatment group one-year after treatment had terminated. This study provided additional support for the efficacy of a forgiveness treatment, highlighting its utility for survivors of incestuous abuse.

More recently, Coyle and Enright (1997) used a similar forgiveness intervention with ten men who identified themselves as feeling hurt by their partners' abortion decisions. They were randomly assigned to a 12-week one-on-one forgiveness intervention or a wait-list control group. Multiple forgiveness and mental health measures were administered to establish a pretest baseline of forgiveness, anger, anxiety, and prenatal grief. These measures of forgiveness and mental health were also administered at the end of the 12-week intervention. Those receiving the forgiveness intervention reported significant increases in forgiveness and significant decreases in grief, anger, and anxiety after treatment. At this time, the wait-listed men began the 12-week treatment. Assessments of forgiveness and mental health were again administered after termination of their treatment; this also served as a 12-week follow-up assessment for the originally treated men. Once treated, the originally wait-listed men also showed signific antly greater improvements in forgiveness, anxiety and grief. And, the originally treated men maintained gains at follow-up testing. These findings suggest that the treatment was effective in promoting improvements in both forgiveness and mental health variables.

In a variation on the forgiveness intervention approach, McCullough and Worthington (1995) compared the effects of group interventions to promote forgiveness that emphasized either interpersonal or self-enhancing reasons to forgive. These two groups were compared with each other and with a wait-list control group. Participants in both forgiveness groups (compared to the control group) experienced reductions in desire for revenge, and increases in positive feelings toward the offender and desire for reconciliation. Interestingly, the forgiveness group with a focus on self-enhancement showed these effects even more strongly than the group with an interpersonal emphasis. This suggests that an emphasis on what benefits forgivers might derive from granting forgiveness reduced their unforgiving responses and facilitated their forgiving and conciliatory desires toward their offenders. While the measures used in this study are not overt indicators of mental health, increased positive feelings and possible reconcilia tion may facilitate mental health, especially given the strong effects of supportive social relationships in promoting health.

The effects of supportive relationships were at the heart of Wuthnow's (2000) survey of 1,379 Americans in religious small groups (e.g., prayer groups, Bible studies, or other religiously oriented small groups). Sixty-one percent of respondents reported that their group had helped them forgive. Furthermore, membership in a group that fostered forgiveness was positively related to members' attempts and successes in overcoming addiction, overcoming guilt, and perceiving encouragement when otherwise discouraged.

Two published studies, originally conducted in order to develop measures of forgiveness, provide information about the relationship of forgiveness and mental health. The first was a study of clients at an outpatient Christian counseling center. Mauger and colleagues (1992) developed and validated the Forgiveness of Others and the Forgiveness of Self scales, which measure distinct dispositional constructs rather than peoples' responses to isolated forgiveness situations. Low forgivingness scores on both scales were correlated with indicators of psychopathology on the Minnesota Multiphasic Personality Inventory (MMPI). Interestingly, the Forgiveness of Self scale was more strongly related to depression, anxiety, and low self-esteem than the Forgiveness of Other scale. It seems that people who latched onto self-punishment and did not accept the gift of mercy that comes with forgiveness were especially prone to experience adverse mental health effects.

The second study validated the Enright Forgiveness Inventory, a measure of interpersonal forgiving and pseudo-forgiving, such as condoning and denial (Subkoviak et al., 1995). Adolescents and their same-gender parents were assessed (as long as they did not score high on pseudo-forgiveness), and the relationship between forgiveness and mental health measures was examined. The results showed that higher forgiveness scores were associated with lower anxiety scores in adolescents who reported any degree of hurt associated with a male-female relationship, in adolescents who reported deep hurt in this area, and in parents who experienced some degree of hurt in relation to their spouses. Depression scores did not significantly relate to forgiveness scores, possibly because of a restricted range of depression scores in these samples.

Not all studies have found a strong link between measures relevant to forgiveness and well-being. McCullough and colleagues conducted an eight-week longitudinal study of forgiveness, vengefulness, revenge, avoidance, rumination, suppression, and satisfaction with life in student volunteers who reported having been hurt by another person within the preceding two months (McCullough, Bellah, Kilpatrick, & Johnson, 2001). They found that volunteers who were higher in trait vengefulness at the beginning of the study continued to have higher revenge motivations against their perpetrators eight weeks later. Volunteers who became more forgiving over the eight-week period ruminated less about their interpersonal offense. However, changes in forgiveness toward specific offenders were uncorrelated with change in life satisfaction, which was a stable rating (test-retest r(58) = .79, p [less than].001). The latter finding raises questions about the relationship between forgiveness and well-being, especially when measures of forgiveness focus on an isolated event and measures of well-being focus on a global assessment such as satisfaction with life.

These published studies provide a foundation for building a literature of research that investigates the efficacy of forgiveness interventions on mental health. In general, the findings suggest that forgiveness interventions promote forgiveness, may foster reductions in unforgiveness and adverse mental health factors, and prompt increases in hope and self-esteem. In general, levels of forgiveness are positively correlated with indicators of mental health and negatively correlated with indicators of dysfunction or distress. However, McCullough et al. (2001) did not find a correlation between forgiving a specific offender and satisfaction with life in general. In the future, the forgiveness intervention literature could be strengthened by comparing forgiveness treatments with other treatments that are known to be effective for reducing anger, anxiety and depression, for example. This would allow us to determine whether forgiveness interventions promote particular types of mental health benefits or if such bene fits occur across a variety of interventions, and whether this occurs through similar or different means.

WHAT DO WE KNOW ABOUT FORGIVING, NOT FORGIVING, AND PHYSICAL HEALTH?

Given increasing evidence of the mind-body connection, forgiveness may well be related to physical --as well as mental--health. Yet, few published papers actually examine forgiveness and physical benefits, instead highlighting the physical costs of unforgiving responses. Given that the literature lacks controlled outcome studies, it is no surprise that Thoresen, Harris, and Luskin (2000) dubbed the issue of forgiveness and health an unanswered question.

In general, forgiveness-related physical health studies have focused primarily on the adverse cardiovascular effects of one type of unforgiving response: hostility. Forgiveness and hostility generally have an inverse relationship: as one goes up, the other comes down. For example, if people cultivate genuine forgiving responses to their offenders, then their hostility toward them is reduced. If people instead nurse grudges, harbor revenge, and stoke the heat of hostility, genuine forgiveness is impeded. (Although it is possible to have a generally forgiving person who is hostile towards a particular offender; or a generally hostile person who is forgiving in a particular circumstance.) In any case, considerable evidence points to hostility as a risk factor for coronary heart disease. The following is an overview of findings relevant to the topic of unforgiveness, forgiveness, and physical health.

A focus on hostility emerged from research on the Type A and B personality constructs by Friedman and Rosenman in the now-famous Western Collaborative Group Study (WCGS). They found that Type A features--highly competitive, ambitious, rushed, easily angered and hostile--predicted coronary heart disease (CHD). Twice as many Type A men developed heart disease in an eight and a half-year time frame as those without these features (labeled Type Bs). (See Rosenman et al., 1975; Friedman & Rosenman, 1974)

Importantly, one particular aspect of the Type A personality construct-hostility--has been found to independently predict the CHD. An accessible review of the research on hostility and health is presented in Williams and Williams' Anger Kills (1993), which documents numerous studies examining hostility. Although nor every study has found a strong connection with CHD, most using widely accepted measures of hostility have found a relationship. One meta-analysis of 45 published empirical studies on hostility and physical health strongly implicated hostility as an independent risk factor for CHD and premature death (Miller, Smith, Turner, Guijarro, & Hallet, 1996).

Why might hostility predispose people to coronary problems? Highly hostile people seem to be more physiologically reactive to interpersonal offenses than low-hostile people (Suarez & Williams, 1991). Highly hostile people experience exaggerated fight-or-flight responses of the sympathetic nervous system (SNS; see Williams & Williams, 1993). The exaggerated release of stress hormones during this reaction is a culprit in coronary heart disease. Furthermore, the calming parasympathetic nervous system (PNS) appears to be weaker in Type As than type Bs. According to Williams and Williams (1993), The net effects of this relatively strong SNS and weak PNS in hostile persons-increased cardiovascular activation, increased mobilization of cholesterol into the blood, increased clumping of platelets, and decreased immune system functions, to mention but a few-are quite capable of starting in motion pathological processes that would account for the higher death rates observed among hostile persons....(p. 52)

As if this were nor enough, highly hostile people also tend to engage in more risky behaviors, including smoking and greater consumption of food and alcohol, all of which exacerbate the risk for disease. They also tend to experience less social support, which puts them at greater risk for both mental and physical problems (Williams &Williams, 1993).

If hostility is physically dangerous, then reducing hostility ought to reduce coronary problems. Interestingly, Friedman et al. (1986) examined this possibility in the Recurrent Coronary Prevention Project, a controlled clinical trial. They found that for Type As at risk for recurring heart attacks, those randomly assigned to a behavioral modification program (vs. standard treatment from a cardiologist) showed a greater reduction in hostile behavior and a reduction in heart problems. A consultant for this research, Kaplan (1992) was especially interested in the neglected construct of the Type B personality, which seems to emphasize forgiving and freedom. Kaplan noticed that the behavioral intervention, which was effective for treating Type As, emphasized the development of these Type B features. Strikingly, when these men were assessed after treatment, they often commented on the importance of learning how to cultivate the forgiving heart (p. 6). Indeed, these patients seemed to effectively reduce their host ility by learning to be more forgiving (p.8). It seems that being forgiving may promote coronary health, in part by reducing the negative coronary effects of anger and hostility.

Data from psychophysiological research complement this view (Witvliet, Ludwig, & Vander Laan, 2001). Witvliet and colleagues asked participants to identify a real-life offender and to subsequently imagine responding to that particular offender in a variety of unforgiving and forgiving ways, using a within-subjects repeated measures design. They assessed the emotional and physiological responses participants generated when they rehearsed memories of being hurt and nursed grudges (i.e., were unforgiving) compared to when they cultivated empathic perspective taking and imagined actually granting forgiveness to the offender (i.e., were forgiving). Across multiple counterbalanced imagery trials, participants showed significantly greater reactivity in the cardiovascular (heart rate, blood pressure) and sympathetic nervous system (skin conductance level) measures, as well as greater brow muscle (corrugator) tension during the unforgiving imagery trials compared to the forgiving imagery trials. Furthermore, the hear t rate, sweat, and brow muscle effects persisted after imagery into relaxing recovery periods, suggesting that the effects of unforgiving thoughts were difficult to quell. Participants also reported significantly higher levels of negative emotion (e.g., anger, sadness) and lower levels of perceived control during the unforgiving imagery trials. In contrast, during the forgiving imagery conditions participants experienced less physiological stress, lower levels of negative emotion, higher levels of positive emotion, and greater perceived control. These results suggest that when people harbor unforgiving responses toward their offenders, they may incur emotional and physiological costs. Instead, when they adopt forgiving responses, they may reduce these costs and accrue psychophysiological benefits, at least in the short term.

Another published study of forgiveness included physiological and behavioral measures. Huang and Enright (2000) assessed participants' motivations for forgiving and compared the anger-related expressions of people who forgave because of obligation to the expressions of people who forgave because of unconditional love. They found that during descriptions of a past experience with conflict, participants who forgave due to obligation expressed more anger-related responses, such as masking smiles and downcast eyes. They also found that in the first minute of describing this interpersonal hurt (vs. describing a typical day), individuals who had forgiven because of obligation showed greater blood pressure increases compared with those who had forgiven because of unconditional love. These data suggest that even when people forgive, their motivations for forgiving-and what forgiveness means to them-may influence their anger-related behavioral and physiological expressions.

In a third published study, Seybold, Hill, Neumann, and Chi (in press) investigated the correlations between dispositional measures of forgiveness and single measures of corrugator (brow) EMG, blood pressure, and blood and plasma assays 30 minutes after insertion of the needle, No significant correlations were observed between trait forgivingness of oneself or of others and one-time measures of corrugator EMG, blood pressure, heart rate, plasma protein, cholesterol, high density lipoprotein, nonesterified fatty acids, triglycerides, or total levels of lymphocytes, neutrophils, monocytes, T-cells, or T-activated cells. Lower trait forgivingness was associated with higher blood viscosity, which the authors identify as a possible cardiovascular risk factor; however, hematocrit scores were largely within normal limits (mean = 42.9%, range = 33-58; Neumann, personal communication, May, 2001). Higher trait forgiveness was positively associated with TxPA levels (i.e., very low density lipoprotein toxicity-preventin g activity). Although TxPA is not currently widely accepted as a standard measure, Arbogast and colleagues' prior research suggests that TxPA may help neutralize the adverse effects of very low density lipoproteins on endothelial cells (Arbogast & Dreher, 1987; Arbogast, Gill, & Schwertner, 1985). Finally, lower levels of self-forgiveness and overall forgivingness were associated with higher ratios of Thelper/T-cytotoxic cells. Because the authors associate higher ratios with better immune system functioning, this correlation suggests that less forgiving persons may have better immune system performance; "however, the ratio is not used clinically independent of other indices and the absolute numbers" (Neumann, personal communication, May, 2001). Overall, this exploratory correlational study addresses a variety of measures that have not been considered previously in forgiveness research. Because the methodology involved correlating dispositional forgivingness with numerous one-time baseline measures of physiol ogical variables, and because some physiological measures (e.g., hematocrit) were largely within normal limits, additional controlled research will be important in order to replicate and determine the importance of these findings.

Although few studies have examined forgiveness and health, the experimental research suggests that forgiving and unforgiving responses could have long term effects on health if they are sufficiently frequent, intense, and enduring. When physiological systems remain activated, they can influence cardiovascular health through changes in allostasis and allostatic load. Allostasis involves the changes in multiple physiological systems which allow people to survive the demands of both internal and external stressors (McEwen, 1998). Although allostasis is needed for survival, extended physiological stress responses brought about by psychosocial factors such as anxiety and hostility can result in allostatic load, eventually leading to physical breakdown. Interpersonal transgressions and people's adverse reactions to them may contribute to allostatic load and health risk through sympathetic nervous system (SNS), endocrine, and immune system changes (cf. Kiecolt-Glaser, 1999). In contrast,

increased frequency of forgiving others ... could function to reduce the chronicity of distress (e.g., anger, blame, and vengeful thoughts and feelings) that has prospectively been shown to alter brain, coronary, and immune functioning. Such reductions could encourage diminished SNS arousal in frequency, magnitude and duration, resulting over time in less physical disease risk. (Thoresen et al., 2000, p. 259)

ISSUES FOR FUTURE FORGIVENESS AND HEALTH RESEARCH

Overall, the small published literature of empirical investigations suggests that as forgiveness increases, so do indicators of mental health; and as unforgiveness increases, so do indicators of physiological stress and coronary heart disease (and vice versa). This emerging literature is beginning to suggest that forgiveness may benefit the heart in both senses of the word. Yet, the work on forgiveness has only begun. As empirical research on forgiveness and health continues to mature, we will encounter many additional issues, including the following three.

First, our understanding of forgiveness and health may need to become more complex. The forgiveness-health story may have many sub-plots. Rather than a simple theme, such as "forgiveness is health-enhancing, and unforgiveness is health-eroding," nuances are likely to enrich the tale. For example, individual differences may make some people likely to interpret a given behavior as an offense, when another person barely notices it. The more easily hurt person may suffer more health costs, despite having forgiven the offender. Other people may have a highly reactive sympathetic nervous system and slow-to-respond parasympathetic nervous system. This may make them more likely to experience greater costs when they are hurt, and more likely to experience hurt and anger when memories of offenses are aroused. These types of individual differences may well exist regardless of people's beliefs, and such differences may be powerful mediators of the forgiveness-health connection.

We also ought to consider the possibility that low-forgivers may function better than high-forgivers in some situations, such as when the offenses endured by the high-forgivers are more severe. The cost of a severe offense may exceed--or at least obscure--the benefit of forgiving, as assessed by health measures. It may also be that some people relish revenge and derive significant satisfaction from exacting an emotional payback. Such people may show an unforgiveness-health relationship that runs counter to expectations. In other contexts, victims may be surrounded by a strong social support network that encourages begrudging and hostile responses toward offenders in ways that make the victim feel justified, comforted, and satisfied with an unforgiving stance. With sufficient social support for unforgiving responses, victims may not suffer health consequences. These possible scenarios merely illustrate the adage that things often are not as simple as they seem. The forgiveness-health connection is likely to be a complex one. Yet, it is exciting that empirical researchers can examine these types of possibilities.

Second, as we interpret existing and forthcoming research on forgiveness and health, we must attend to the question of causality. A number of intervention studies and psychophysiological studies have already used experimental designs; however, some studies are correlational, leaving us unable to determine causation. For example, a positive correlation may indicate that forgiveness causes better mental health. Or, it may mean that people with better mental health may have more cognitive and emotional resources to do the difficult work of forgiveness. Or, a third factor (such as social support, or divine intervention evident in another form) may be influencing both forgiveness and mental health variables. For example, the person may be surrounded with supportive relationships that simultaneously encourage and reward forgiveness and foster mental health.

Third, we must consider how theorists might interpret forgiveness and health data. One approach to this question is to consider cognitive behavioral perspectives that take into account the dynamic and reciprocal interrelationships among people's thoughts, feelings, behaviors, and physiology. This approach emphasizes that because these systems are connected, changing one aspect of our responding will yield changes in others. In the case of forgiving, granting forgiveness involves relinquishing negative feelings, and also adopting a merciful set of thoughts, feelings, and behaviors toward the offender. Experientially, when faced with alternative responses to an offender, our thoughts and feelings are intertwined. Whether considering revenge or forgiveness, our emotions shift when we actively think about these different responses to our offenders. When we think, or imagine, reacting to an offender in a begrudging manner or in an empathic or forgiving manner, these cognitive acts are also inherently emotional. W hen people respond in unforgiving or forgiving ways, even in the privacy of their own minds, their emotions are inevitably affected. Some might argue that changes in a forgiver's anxiety, depression, anger, or hope scores are simply the natural outcome of a forgiveness process that changes the forgiver's thoughts and feelings toward the offender.

To go even further, each cognitive or emotional event is also a biological event. We are embodied, with physiology involved even in fleeting thoughts and feelings. If we have thoughts of an unforgiving or forgiving nature, emotion is involved and physiological changes are occurring. Even more interesting, when people think about or imagine engaging in emotional events, their physiological responses are powerfully influenced by the valence (negative, positive) and arousal (low, high) level of their emotion. For example, research has shown that when research participants thought about engaging in emotional situations, their heart rate and sweat (i.e., skin conductance) were most responsive to the arousal level of emotion, and particular facial muscles tensed and relaxed depending on whether the emotion was positive or negative (Witvliet & Vrana, 1995). Some theorists could argue that forgiveness (or unforgiveness) and physical health results are due to nothing but basic emotion processes. In such processes, th e negarivity/positivity and arousal involved in different unforgiving and forgiving responses simply guide physiological patterns, which may ultimately influence physical health outcome measures. For example, during repetitive bouts of anger and hostility, highly arousing emotion involves escalating blood pressure and heart rate, which may damage the inner lining of arteries. Simultaneous surges of stress hormones release fat into the bloodstream that is converted to cholesterol unless it is burned with vigorous exercise. The rush of adrenaline and noradrenaline enables platelets to stick to the damaged areas of eroded arterial lining. Together with cholesterol and other chemicals, they form arteriosclerotic plaque, the key culprit in coronary heart disease (see Williams & Williams, 1993).

Such an explanation (very simply sketched here) has the advantages of cleanly accounting for the data, with clear links to a theoretical literature. Still, we have to be aware that the benefits of parsimonious explanations come with the risks of reducing complex constructs to nothing but a simple list of ingredients. As the Nobel Laureate Roger Sperry once stated, "The meaning of the message will not be found in the chemistry of the ink" (Jeeves, 1997). We would do well to realize that the full meaning of forgiveness will not be found in the neurochemistry or psychophysiology of the response. Yet, use of cognitive-behavioral and psychophysiological theories, methods, and data can nevertheless augment our understanding of the complex effects of forgiving.

Forgiveness is multifaceted, has profound spiritual meaning, and involves our social, psychological, and physical faculties. The spiritual aspect of forgiving is not completely separate from or unrelated to the others; neither is it completely subsumed by the social, cognitive, emotional, or physical aspects. They are all connected like facets of the same diamond. The disciplines of theology and philosophy allow us to see the brilliance of particular facets, while the social and natural sciences illuminate others, and the arts and humanities shed light on still others. Despite their distinctness, these facets are ultimately connected. When we view them through complementary cross-disciplinary lenses, we can best see the brilliant unity and dazzling complexity of the spiritual, social, psychological, and physical facets of forgiveness.

REFLECTIONS ON QUESTIONS OF THE CHRISTIAN COMMUNITY

As empirical research on forgiveness and health develops, it will be important for Christians to consider the potential promises and pitfalls in conducting and interpreting this research. All Christians, no matter what their tradition, should strive for a coherent and integrated understanding of forgiveness that is rooted in Scripture and draws on the best current research and reflections. [2] To explore the relationship between matters of faith and empirical research on forgiveness and health, it may be useful to address several questions that are pertinent for Christians.

The first question is whether empirical research can or should be used to study forgiveness and health. Christians from a variety of traditions may question whether forgiveness ought to be the subject of empirical study rather than strictly theological inquiry. Empirical researchers may question whether forgiveness is too ethereal and spiritual to be an appropriate topic for study. Yet--depending on the types of health benefits one wants to consider, and the fit between those questions and empirical research methods--such research may be particularly well suited to the task of addressing such questions. Empirical methodologies and statistical tools may be used to observe and analyze the connections among aspects of forgiveness, and social, psychological, and physical functioning. While empirical research need not displace theology and philosophy in studies of forgiveness, it may offer unique insights that illumine our understanding of the benefits and costs of forgiveness. This kind of research is a tangible example of faith seeking understanding.

Scripture communicates God's call for us to forgive, describes God's forgiveness of us, provides models of interpersonal forgiveness, and gives us a vision of the wedding of grace and responsibility in forgiveness. But the Bible does not systematically outline the factors that facilitate or impede forgiveness, or the relationship of forgiveness to current understandings of mental and physical health. Scripture also does not negate the value of understanding these aspects of forgiveness. A biblical understanding of forgiveness may powerfully motivate us to better understand both its experiential components and its emotional and physical health effects.

Still, some might prefer to let the moral philosophers and theologians tackle forgiveness without the input of empirical researchers. After all, the emphasis on quantification can seem reductionistic when approaching the richness of forgiveness. How might we think about this? Researchers need to define forgiveness in their particular studies, formulate ways to measure variables, and make interpretations based on the particular interpretations and manipulations they use. The boundaries of these ideas and measures are important for researchers to determine and for their readers to understand. The intent of using empirical approaches is not to reduce forgiveness to a thimble-sized construct or bleach meaning from its rich fabric. Rather, the purpose and value of empirical research is to clearly understand certain aspects of forgiveness and their relationship to other variables, such as mental and physical health. As stewards of God's gifts, we can use empirical tools to illuminate the experience of forgiving and being forgiven, to clarify the relationships among our thoughts, feelings, and behaviors related to forgiveness, and to elucidate the connections between forgiveness and mental and physical health. These studies can complement, not replace, the message of Scripture.

A second question is whether empirical research should serve as a new apologetic that could speak to our health-focused society. This question includes the implicit--and incorrect--assumption that empirical investigations even have the power to test or to prove theological truth claims. Although empirical research offers a valuable way of knowing based on careful measurement and analysis, it cannot usurp the role of rheology or philosophy. Certain claims cannot be tested empirically. For example, while research may be used to assess whether historical artifacts are consistent with biblical accounts, scientific study is incapable of testing whether Jesus is indeed "the way and the truth and the life." Likewise, while empirical research can help us understand the relationships among thoughts, feelings, physiology, and behavior in forgiveness, this research cannot tell us when or if we should forgive. Relying on forgiveness and health data as an apologetic for Christianity may be inappropriate for a variety of reasons. One is that Christians as well as non-Christians practice forgiveness, and the empirical data base already includes participants from a variety of religious and non-religious backgrounds. Another is that we may ultimately find that forgiving is not significantly more beneficial for health than avoidance, tolerance, or minimizing the severity or significance of the offense. Even if research fails to show an irrefutable link between forgiveness and health, this cannot undermine the truth clams of Christianity. Empirical research and theology offer different epistemologies and typically address different questions. On this matter of spiritual truth and empirical data, C. S. Lewis offered a valuable perspective: If Christianity is untrue, no honest person will want to believe it, however helpful it might be; if true, every honest person will want to believe it, even if it offers no measurable help (cf. Lewis, 1970, pp. 108-109).

A third question is whether empirical research on forgiveness and health should serve as our motivation to forgive. We need to be particularly careful about moving from descriptions of data to prescriptions and proscriptions of behavior based on the results. The current data on forgiving and health may pose a special challenge in our self advancing and health focused culture in that some people might choose to forgive because they themselves will benefit. Rather--in the best case--Christians should be motivated to forgive because Scripture includes clear mandates to forgive (e.g., Matthew 6:14; Matthew 18:2122; Colossians 3:13) as we imitate Christ (e.g., Ephesians 4:32). Still, the current data about forgiveness and health may augment our appreciation of the biblical emphasis on the unity of personhood, help us understand better the experience of granting forgiveness, and deepen our understanding of what it means to be embodied.

A fourth important question is whether we can assume that forgiveness and other virtuous behaviors will result in positive health effects. It seems reasonable that something that is good should also be good for us. However, this is not inevitable. Some people who have taken a stand for social justice or have practiced their Christian faith have been killed for it (clearly not a positive health outcome). Others who minister to the sick and suffering take on the burden of grief and the pain of depression, and some contract the illnesses of those they serve. Being faithful and doing what is good does not inevitably secure good physical and mental health. Faithful discipleship often comes at a cost (cf. Bonhoeffer, 1959).

A final question is whether forgiving ought to yield greater health benefits for believers compared to unbelievers. One might think that if the gospel is true, and forgiveness is central in the gospel, then those who believe the gospel should have more health benefits when they forgive. In Human Nature at the Millennium, the neuroscientist Jeeves (1997) offers a valuable perspective to consider. He stresses that the basic processes of cognition, emotion, and physiology between Christians and non-Christians will not differ (see Jeeves, 1997; p. 18). Everything from conditioned responses, to the physiology of fear, to depth perception, to the connection between thoughts and feelings is likely to operate similarly across people of different beliefs. This may also be the case for forgiveness. If forgiveness of others reduces stress, depression, anger, and anxiety in forgivers, then we might anticipate that this will be the case for both believers and unbelievers. We are all fundamentally human, with interrelated physiological, emotional, and cognitive systems.

But how do we account for the spiritual and grace-filled nature of forgiveness? Won't these aspects of forgiveness differ across those who truly believe the gospel message and those who do not? How can empirical research address this? While empirical methods are not designed to measure grace or spirituality per se, they can be used to measure people's forgiving behaviors, physiological responses, and self-reported experiences and emotions. So, for example, although the involvement of physiological responses during emotion is likely to be similar across people, the meaning of forgiveness is likely to differ for Christians and non-Christians. These differences in meaning and spirituality will likely influence their thoughts, feelings, and behaviors. This means that different physiological responses might also occur, in ways that could generate different health effects for people with different beliefs. Other differences in the forgiving responses of Christians may also occur in ways that impact health. On the positive side, when Christians live out the biblical mandate to forgive, they ought to be more likely to forgive, and hence experience health benefits as by-products of their faithful responses. On the potentially negative side, Christians who feel they ought to forgive, but have trouble enacting the practice of forgiveness, may experience more adverse effects because of the guilt they feel as a result of not carrying out this mandate. [3] In any case, these questions can be studied empirically. Rather than simply speculating about the possibilities, we can put them to the test.

MOVING FORWARD: RESEARCH APPROACHES TO FORGIVENESS

This is an exciting time for Christian scholars interested in studying forgiveness and other matters of faith. Some of this scholarship will no doubt proceed with commonly used methods, including careful examinations of the presuppositions undergirding interpretations and methodologies, as well as critiques that point out the potential pitfalls and possibilities for empirical research approaches. But, some scholarship will proceed using mainstream empirical methodologies. There is no reason not to use all of the resources in our disciplines that will enhance our understanding of forgiveness. We might ask: What are the best theories, research designs, and statistical methods available? How could they be used to address certain types of questions about forgiveness and health? Christians don't need to re-invent the wheel when studying spiritually relevant topics. We can consider what existing tools can be used effectively to shed light on topics of spiritual relevance. We may need to refinish or sharpen the too ls, but we should consider whether existing theoretical frameworks and methodologies may be well suited to the task. For example, when researching spiritual topics such as forgiveness, we should learn from mainstream psychology and go beyond self-report data. While self-reports are rich and important, behavior and physiology are also potent measures and offer valuable insights about experience. By using a fuller repertoire of dependent measures and by employing a range of methodologies, we will better understand the nuances of the forgiveness and health relationship and produce research that will be more widely embraced in the academic community.

We should also feel confident about applying what is useful from our disciplines, extending extant theories and methods to address new topics of spiritual significance. This approach increases the likelihood that our methods will be well tested, ensuring that our work will make stronger connections with the existing literature and receive a better hearing by mainstream academics.

The topic of forgiveness and health provides an opportunity for Christians to actively engage in hands-on empirical research that is both of vital interest to the Christian community and is respected by mainstream research and clinical psychologists. Empirical explorations of forgiveness and health may even have many benefits beyond expanding our knowledge base. People who are open to empirical research, but skeptical of faith may recognize the importance of forgiveness as a matter of heart, soul, and mind. And people who are open to faith, but skeptical of empirical research on spiritual topics may gain an enriched understanding of forgiveness, feelings, and physiology that complements the biblical view of the unity of personhood.

AUTHOR

WITVLIET, CHARLOTTE vanOYEN: Address: Psychology Department, Hope College, Holland, MI 49422-9000. Title: Assistant Professor of Psychology. Degrees: BA, Psychology and Music, Calvin College; MS and PhD, Clinical Psychology, Purdue University. Specializations: Emotion and psychophysiology, trauma, forgiveness.

This work was supported by a grant from the John Templeton Foundation for research on forgiveness and by a Towsley Research Scholar Award from Hope College.

(1.) While reconciling and forgiveness are close cousins, they are not identical twins. Sometimes, the people we need to forgive have died, or we lose contact with them. Then, there are situations in which it would be unwise for victims and perpetrators to reconcile, such as in the case of repeated abuse or neglect. Still, survivors of maltreatment who retreat from dangerous relationships do not have to remain prisoners of past offenses. They can experience the paradoxical freedom that comes when they muster the courage and resolve to grant the gift of forgiveness, by God's grace.

(2.) The broader relationship of theology and psychology has already received significant scholarly attention (see, for example chapter one of Jones & Butman, 1991; Johnson & Jones, 2000; Tan, 2001 for overviews). The goal of this article, however, is not to survey different views of integration, but rather to consider several issues concerning the Christian faith and empirical research on forgiveness and health in particular. The perspective advanced here is that it is important to steer between the extremes of 1) disregarding or devaluing the contributions of empirical investigations of forgiveness and health and of 2) overlooking or minimizing the theological significance of forgiveness as well as the ways in which one's beliefs and values concerning forgiveness may guide empirical investigations and interpretations of the data (cf. Myers, 1990-1991).

(3.) For those who struggle with the practical "how" of forgiving, two books are especially helpful: Smedes' (1996) The Art of Forgiving: When You Need to Forgive and Don't Know How and McCullough, Sandage, and Worthington's (1997) To Forgive is Human: How to Put Your Past in the Past.

REFERENCES

AI-Mabuk, R.H., Enright, R.D., & Cardis, PA. (1995). Forgiveness education with parentally love-deprived late adolescents. Journal of Moral Education, 24,427444.

Arbogast, B.W., & Dreher, N.J. (1987). Coronary disease prediction using a new atherogenic index. Atherosclerosis, 66, 55.62.

Arbogast, B.W., Gill, LR, & Schwertner, H.A. (1985). A new protective factor in coronary artery disease: Very-low-density lipoprotein toxicity preventing activity. Atherosclerosis, 57,75-86.

Bonhoeffer, D. (1959). The cost of discipleship. New York: Macmillan.

Coyle, C.T., & Enright, R.D. (1997). Forgiveness intervention with posrabortion men. Journal of Consulting and Clinical Psychology 65, 1042-1046.

Emerson, J.G. (1964). Dynamics of forgiveness. Philadelphia, PA:Westminster Press.

Enright, R.D., (2001). Forgiveness is a choice. Washington D.C.: American Psychological Association.

Enright, R.D., & Coyle, CT. (1998). Researching the process model of forgiveness within psychological interventions. In E.L. Worthington, Jr. (Ed.), Dimensions of Forgiveness (pp. 139- 161). Philadelphia, PA: Templeton Foundation Press.

Enright, R.D., & Fitzgibbons, RP. (2000). Helping clients forgive: An empirical guide for resolving anger and restoring hope. Washington D.C.: American Psychological Association.

Enright, R.D., Freedman, S., & Rique, J. (1998). The psychology of forgiveness, in RD. Enright and J. North, (Eds.), Exploring Forgiveness (pp. 46-62). Madison, WI: The University of Wisconsin Press.

Freedman, S.R, & Enright, RD. (1996). Forgiveness as an intervention goal with incest survivors, Journal of Consulting and Clinical Psychology, 64,983-992.

Friedman, M., & Rosenman, R.H. (1974). Type A behavior and your heart. New York: Knopf.

Friedman, M., Thoresen, C.E., Gill, J., Ulmer, D., Powell, L.H., Price, V.A., Brown, B., Thompson, L., Rabin, D.D., Breall, WS., Bourg, W., Levy, R, & Dixon, T. (1986). Alteration of Type A Behavior and its effect on cardiac recurrences in pose myocardial infarction patients: Summary results of the Recurrent Coronary Prevention Project. American Heart Journal, 112, 653-665.

Hebel, J.H., & Enrighr, R.D. (1993). Forgiveness as a psychotherapeutic goal with elderly females. Psychotherapy, 30,658-667.

Huang, S.-T. T., & Enright, RD. (2000). Forgiveness and anger-related emotions in Taiwan: Implications for therapy. Psychotherapy, 37, 71-79.

Jeeves, M.A. (1997). Human nature at the millennium: Reflections on the integration of psychology and Christianity. Grand Rapids, Ml: Baker Books.

Johnson, EL, & Jones. S.L (2000). Psychology & Christianity: 4 views. Downers Grove, IL: Inter Varsity Press.

Jones, S.L., & Burman, RE. (1991). Modern psychotherapies. Downers Grove, IL Inter Varsity Press.

Kaplan, B.H. (1992). Social health and the forgiving heart: The Type B story. Journal of Behavioral Medicine, 15,3.14.

Kiecolt-Glaser, J.K. (1999). Stress, personal relationships, and immune function: Health implications. Brain, Behavior, and Immunity, 13, 61-72.

Mauger, P.A., Perry, J.E., Freedman, T., Grove, D.C., McBride, KG., &McKinney, K.E. (1992). The measurement of forgiveness: Preliminary research. Journal of Psychology and Christianity 11, 170-180.

Lewis, CS. (1970). God in the dock. Grand Rapids, MI: Eerdmans.

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 & Social Psychology Bulletin, 2 7(5), 601-610.

McCullough, M.E., Exline,J.J., & Baumeisrer, R.F. (1998). An annotated bibliography of research and related concepts. In E.L. Worthington, Jr. (Ed.), Dimensions of forgiveness: Psychological research and theological perspectives (pp. 193-317). Philadelphia: Templeton Foundation Press.

McCullough, M.E., Pargament, K.I., & Thoresen, C.E. (2000). Forgiveness: Theory, research, and practice. New York: Guilford Press.

McCullough, M.E., Sandage, S. J., & Worthington, E.L., Jr. (1997). To forgive is human: How to put your past in the past. Downers Grove, IL Inter Varsity Press.

McCullough, M.E., & Worthington, E.L., Jr. (1995). Promoting forgiveness: The comparison of two brief psychoeducational interventions with a waiting-list control. Counseling and Values, 40, 55-68.

MeEwen, B.S. (1998). Protective and damaging effects of stress mediators. New England Journal of Medicine, 338, 171-179.

McCullough, M.E., Worthington, E.L, Jr., & Rachal, K.C. (1997). Interpersonal forgiving in close relationships. Journal of Personality and Social Psychology, 73,321-336.

Miller, T.Q., Smith, L.W., Turner, C.W., Guijarro, M.L., & Haller, A.J. (1996). A meta-analytic review of research on hostility and physical health. Psychological Bulletin, 119,322-348.

Myers, D.C. (1990.1991). Steering between the extremes: On being a Christian scholar within psychology. Christian Scholar's Review, 20,376-383.

North, J. (1987). Wrongdoing and forgiveness. Philosophy, 62, 499-508.

Rosenman, R.H., et al. (1975). Coronary heart disease in the Western Collaborative Groups Study: Final follow-up experience of 81/2 years. Journal of the American Medical Association, 233, 872-877.

Seybold, K.S., Hill, P.C., Neumann, J.K., & Chi, D.S. (in press). Physiological and psychological correlates of forgiveness. Journal of Psychology and Christianity.

Smedes, L.B. (1984). Forgive and forget: Healing the hurts we don't deserve. San Francisco, CA: Harper & Row.

Smedes, L.B. (1996). The art of forgiving: When you need to forgive and don't know how. Nashville, TN: Moorings.

Suarez, E.C., & Williams, R.B. (1991). Situational determinants of cardiovascular and emotional reactivity in high and low hostile men. Psychosomatic Medicine, 51,404-418.

Subkoviak, M.J., Enright, R.D., Wu, C., Gassin, E.A., Freedman, S., Olson, L.M., & Sarinopoulos, I. (1995). Measuring interpersonal forgiveness in late adolescence and middle adulthood. Journal of Adolescence, 18, 641-655.

Tan, S.-Y. (2001). Integration and beyond: Principled, professional, and personal. Journal of Psychology and Christianity, 20, 18-28.

Thoresen, G.E., Harris, A.H.S., & Luskin, F. (2000). Forgiveness and health: An unanswered question. In M.E. McCullough, K.I. Pargament, & CE Thoresen (Eds.), Forgiveness: Theory, research, and practice (pp. 254-280). New York: Guilford Press.

Williams, R., & Williams, V. (1993). Anger kills: Seventeen strategies for controlling the hostility that can harm your health. New York: Times Books.

Witvliet, C.V.O., & Vrana, S.R. (1995). Psychophysiological responses as indices of affective dimensions. Psychophysiology, 32, 436-443.

Witvliet, C.V.O., Ludwig, T., & Vander Laan, K. (2001). Granting forgiveness or harboring grudges: Implications for emotion, physiology, and health. Psychological Science, 12, 117-123.

Worthington, E.L., Jr. (1998a). Dimensions of forgiveness: Psychological research and theological perspectives. Philadelphia: Templeton Foundation Press.

Worthington, E.L. Jr. (1998b). Empirical research in forgiveness: Looking backward, looking forward. In E.L. Worthington, Jr. (Ed.), Dimensions of Forgiveness (pp. 321-339). Philadelphia, PA: Templeton Foundation Press.

Worthington, E.L., Jr. (2001). Five steps to forgiveness: The heart and science of forgiving. New York: Crown Publishers.

Worthington, E.L., Jr., Sandage, S.J., & Berry, J. (2000). Group interventions to promote forgiveness: What researchers and clinicians ought to know. In M.E. McCullough, K.I. Pargament, & C.E. Thoresen (Eds.), Forgiveness: Theory, research, and practice. (pp. 228-253). New York: Guilford Press.

Wuthnow, R (2000). How religious groups promote forgiving: A national study. Journal for the Scientific Study of Religion, 36, 124-137.