“To Err is Human, to Forgive, Divine”: Religious Doubt, Psychological Well-Being and the Moderating Role of Divine Forgiveness

Abstract

A significant amount of research to date has been done to study the effects of forgiveness on mental health and well-being, but less research has been conducted on divine forgiveness. The main purpose of the current study is to examine the possible moderating role of divine forgiveness. Drawing on a nationally representative sample of 1,500 older adults, regression results suggest that greater divine forgiveness exacerbated the relationship between high religious doubt and greater depressive symptoms and lower life satisfaction. We did not document similar moderation patterns between religious doubt and forgiveness of others and self-forgiveness. Understanding the impact of divine forgiveness for those experiencing uncertainty in their faith is crucial to gaining a more complete picture of religion’s “dark side, and we hope future research continues to pursue these objectives.

In most Christian traditions, forgiveness is identified as a moral virtue essential for human flourishing. Empirical research on forgiveness emerged during the latter part of the 20^th century, but the turn of the 21^st century heralded a significant increase of scholarly attention to this topic (Woodyatt et al., 2007). Most of this research has studied the effects forgiveness on health and well-being (Touissaint et al., 2015) as well as life satisfaction (Davis et al., 2015). However, more recent discussions of forgiveness have given considerably less attention to divine forgiveness, that is, receiving forgiveness from God or a divine entity, relative to research on self-forgiveness and interpersonal forgiveness (forgiveness of others). Several empirical studies have begun to show that greater perceptions of divine forgiveness are associated with better mental well-being (Krause, 2015; Touissaint et al., 2008). However, researchers have not studied the social context that might shape the relationship between forgiveness by God and well-being. This oversight is important, especially since recent evidence has shown that divine forgiveness accounts for additional variance in explaining health outcomes above and beyond that which is attributed to self or interpersonal forgiveness (Fincham & May, 2022).

In this study, we probe the relationship between religious doubt, divine forgiveness, and mental well-being. Religious doubt is defined as “uncertainties about, and questioning of, religious teachings and beliefs” (Dein, 2013, pg.201). Evidence from both cross-sectional and longitudinal study designs have shown that religious doubts are associated with lower mental well-being (Ellison & Krause, 2009; Upenieks, 2021). Specifically, doubts or struggles with one’s faith have been linked to greater anxiety (Harris et al., 2014), and depression (Ellison & Lee, 2010). Common explanations for these pernicious associations between doubt and well-being are complex, included a loss of psychosocial resources like social support and self-esteem (Hill et al., 2021) or a reduced sense of meaning in life arising from tensions or uncertainty with faith (Upenieks, 2021). Though these mechanisms help to account for why religious doubt can bring about painful mental health consequences, paradoxically, there are other aspects of religion that could help people cope more effectively with religious doubts. There may be elements of religious life that can help sort out the doubts one harbors about their faith. However, there are also reasons to expect divine forgiveness will exacerbate the harmful consequences of religious doubt. The main purpose of this current study is to examine the possible role of divine forgiveness in shielding religious adherents from the negative consequences of religious doubt or heightening their impact.

In what follows, we examine the interlinkages between religious doubt, divine forgiveness, and well-being, measured by both depressive symptoms and life satisfaction, in a nationally representative sample of nearly 1,500 older and middle-aged adults, a group of people with specific characteristics as it pertains to religion and forgiveness.

Religious Doubt and Mental Well-Being

Religious doubts represent the uncertainty and questioning of one’s religious beliefs, and broadly represent conflicts or significant challenges R/S domain of life (Hunsberger et al., 2002). Harboring doubts about one’s faith could be one component of a broader facet involving spiritual struggles, which Pargament (2007, pg.112) defined as “signs of spiritual disorientation, tension, and strain.” Pargament (2007) identified three types of spiritual struggles, one of which encompasses doubt. Pargament (2007) also acknowledged several types of religious doubt. One aspect of doubt is to question whether God is directly involved in one’s life, or to question certain teachings of the faith. As Krause and colleagues (1999) noted, religious doubt may be unsettling because of the indecision stemming from conflicting beliefs. Religious doubt may be considered a form of cognitive dissonance (Festinger, 1957). An inability to resolve cognitive dissonance may contribute to mental health problem. Religious doubt may be especially painful because it is likely accompanied by feelings of shame or guilt, stemming from the belief that one should not question the main tenets of one’s faith (Gauthier et al., 2006). If shame and guilt occur as a result of religious doubt, it could lower feeling of self-esteem, which is an important correlate of well-being.

Religious doubts and spiritual struggles have been the focus of many empirical studies in religion and health over the past two decades (Krause & Ellison, 2009; Pargament, 2007). As we noted above, a host of studies have demonstrated that religious doubts are linked with greater symptoms of depression and anxiety as well as more distress (Ellison & Lee, 2010;; Upenieks, 2021), in addition to shorter lifespans (Pargament et al., 2001). Even though religious doubts have been reliably associated with lower well-being, some scholars have argued that religious doubts are an important and natural component of religious life (Pargament, 2007) and could even lead to personal growth (Desai & Pargament, 2015). As Wilt and colleagues (2016) argue, doubts or struggles in faith can be framed as challenges that can be worked through rather than as something to be “fixed” or even prevented. We argue here that the receipt of divine forgiveness may either help buffer or exacerbate the association between religious doubt and well-being.

A long line of scholarship in Western religious traditions have suggested that individuals experience their relationships to a divine being in ways that are similar to what occurs in human relationships. Indeed, Christian religious traditions emphasize that believers should seek to develop and nurture a personal relationship with God (Kirkpatrick & Shaver, 1992). If divine relationships share common features with human relationships, however, they will also be subject to ups and downs. On some occasions, God may be perceived by believers as providing them with comfort and love (Granqvist et al., 2010), while at other times, they may experience their relationship with God in more negative ways, such as through divine struggles (Exline et al., 2014), either by becoming angry at God, questioning His love for them, fearing God, or believing that God is unresponsive to their prayers, to name just a few. Divine struggles are often accompanied by stress, anxiety, and depression (Wilt et al., 2016). In particular, communication with God may be more difficult under these conditions (Wilt et al., 2020). As research by Wilt and colleagues (2023) has shown, more negative imagined conversations with God tend to come about more frequently for people with a less positive view of God. Such people may also be at a higher likelihood of experiencing religious doubt. Consistent with previous research, then, we would expect the following:

Hypothesis 1.

Greater religious doubt will be associated with higher depressive symptoms and lower life satisfaction.

The Moderating Role of Divine Forgiveness During Religious Doubt

We see two possibilities grounded in existing work as to how divine forgiveness (or lack thereof) may affect those with religious doubts. As one possibility, when reflecting on religious doubt or struggles in faith, some believers assign blame to God for allowing them to undergo a struggle in faith (Exline, Park, Smyth, & Carey, 2011). A hallmark of spiritual struggles are strong feelings of disappointment or anger directed towards God, and in the absence of divine forgiveness, attributing some blame to God may create an additional source of guilt or fear (Wilt et al., 2022). Believers who remain angry at God for a long period of time could feel weighed down by this anger, and at the same time may feel they are cut off from a powerful source who could assist them in overcoming their struggles. It seems plausible, then, that feeling forgiven by God may be especially harmful for those experiencing religious doubt. These people, perhaps because of negative self-views, might people that their current state of struggle is unworthy of God’s forgiveness, which could add to their existing burden of religious doubt if they feel that they do not deserve God’s forgiveness. This may reflect a unique form of religious cognitive dissonance (Mannheimer & Hill, 2015), where perceptions of divine forgiveness are inconsistent with such struggles experienced in one’s faith. This could have a destabilizing influence on mental well-being as people may feel guilty for receiving divine forgiveness when they are in the midst of spiritual conflict. In a way, then, people who perceive themselves to be unworthy of God’s forgiveness may be a different type of religious doubt, which we would expect to exact a mental health toll.

As a second possibility, divine forgiveness may be a useful buffer for those encountering religious doubt. Religious doctrine on both forgiveness as well as redemption may serve as an important source of hope for those caught under the weight of their struggles. Divine forgiveness has long been posited as a spiritual coping mechanism, especially for those who are experiencing negative life events (Pargament, Smith, Koenig, & Perez, 1998). Divine forgiveness may also be salient for those undergoing religious struggles; divine forgiveness could be experienced as a relational transaction with God that could serve to draw believers into a deeper level of closeness with God and help them endure and overcome any doubts they face.

Viewed in this light, divine forgiveness may be a positive form of religious coping. One can look to forgiveness from God as a source of hope to get them through their current struggles. It may seek counterintuitive to scour religiosity for variables that might buffer the effects of problems in the religious domains. Yet, having strong beliefs in divine forgiveness may help those facing religious doubt to experience forgiveness and lessen any guilt felt from allowing the struggle to arise. Divine forgiveness may help bring about a sense of peace in one’s relationship to God and others. The sense of peace that is restored might allow individuals to see their conditional self-worth, even with the doubts they may hold (Lawler-Row, 2010). This is important, because believing that God loves sinners may not be associated with appreciable changes in mental well-being until such a belief is also part of how the individuals see themselves. Existing research by Uecker and colleagues (2016), for instance, found that belief in the sinful nature of humans, a negative religious cognition, was less harmful to mental well-being if accompanied by belief in divine forgiveness. Applied to the current study, when confronted with struggles of faith, feeling forgiven by God may help to restore a sense of self-worth. The perception that one has received forgiveness from God after a commission of sin tends to be associated with positive outcomes, such as greater optimism and less shame, guilt, fear, and anger (Worthington & Wade, 1999). Forgiveness by God, then, may be a resource that can help an individual meet their spiritual needs or fill any gaps or doubts.

While no prior study has examined whether divine forgiveness alters the relationship between religious doubt and well-being, one prior study on the topic has looked at whether a secure attachment to God (i.e., low avoidance and low anxious attachment) was a moderator of the relationship between religious doubt and positive mental health (Love et al., 2022). Drawing from attachment theory, Rowatt and Kirkpatrick (2002) noted that a secure attachment, viewing God as a “secure base” through which to view the world, was predictive of greater confidence. An insecure attachment occurs when an individual is anxious about being abandoned by God or avoids intimacy with God. According to a study by Love and colleagues (2022), a secure attachment to God mitigated the negative relationship between religious doubt and positive mental health. However, an insecure attachment to God worsened the relationship between cognitive doubt and worse mental health.

Altogether, then, receiving forgiveness from God may allow people to view themselves in a more favorable light and have greater self-compassion during moments of doubt in their faith. The person who experiences divine forgiveness may deal more effectively with the negative effects of religious doubt, and thereby protect their well-being. In a paradoxical sense, then, this second approach suggests that religion may be the problem and solution for older adults undergoing divine struggle. We therefore propose a set of two competing hypotheses:

Hypothesis 2.

The association between religious doubt and lower well-being will be exacerbated among older adults who endorse greater divine forgiveness.

Hypothesis 3.

The association between religious doubt and lower well-being will be buffered among older adults who endorse greater divine forgiveness.

Forgiveness in All Its Forms: Self-Forgiveness and Other Forgiveness as Moderators

In order to rule out the possibility that it is forgiveness more generally and not self-forgiveness of others which helps people to cope with religious doubt, we also consider the possible moderating role of two other types of forgiveness: self-forgiveness and other forgiveness.

Self-Forgiveness:

There are several definitions of self-forgiveness that have been offered in the literature.For their part, Hall and Fincham (2005, pg.162) defined self-forgiveness as “motivational changes towards one’s self or stimuli associated with the offense, which can occur in the domains of avoidance, retaliation, or benevolence towards the self.” Webb and colleagues (2017) offered a definition of self-forgiveness as a deliberate process that takes place over time after the commission of a wrongful act that that involves constructive changes in one’s relationship with the self. According to Kim et al. (2021), people self-forgive when they are compassionate toward themselves and abandon any tendencies toward self-resentment. Research has thus far suggested that self-forgiveness is a critical factor underlying psychological well-being (Woodyatt et al., 2017).

Some studies have shown that divine forgiveness may be the first step to self-forgiveness (McConnell & Dixon, 2012). In one study that investigated temporal order, Fincham, May, and Chavez (2020) found that forgiveness by God was associated with greater self-forgiveness seven weeks later, but not vice versa. People may believe that God forgives through an act of grace (i.e., an unearned privilege), but have a harder time forgiving themselves. Though divine forgiveness may be a precursor to self-forgiveness, we are interested here in whether self-forgiveness may offer additional utility in understanding our focal relationship between religious doubt and well-being. On the one hand, forgiving one’s self for any struggles in faith could help to alleviate the impact on well-being. Self-forgiveness has been shown to be negatively associated with depressive symptoms and self-blame, but has been positively linked to better mental health (Davis et al., 2015). On the other hand, however, since religious doubt may be specific to one’s relationship with God or other domains of the religious life, self-forgiveness may be less effective than divine forgiveness for addressing this particular issue. Since the afflicted party in the case of religious doubt is God, then forgiveness may need to also be derived from this source. Self-forgiveness may also lead to the forgiveness of the self in the future (Krause, 2017), even if we do not expect it to moderate (exacerbate or buffer) the effect of religious doubt on well-being.

Other Forgiveness:

The forgiveness of others (i.e., interpersonal forgiveness) is the tendency to forgive others for transgressions committed against the self (Woodyatt et al., 2017). The forgiveness of others appears to associate with better well-being by repairing damaged interpersonal bonds and facilitating the healing process, leading to lower psychological strain (Lawler-Row et al., 2011).

Other forgiveness is likely also linked with divine forgiveness. Divine forgiveness, that God forgives human sins, serves as a model of the how we are to forgive others who transgress against us. As is written in the Lord’s Prayer in Christianity reads, “And forgive us our debts, as we also have forgiven our debtors” (Matthew 6:12). Forgiveness among humans is thus portrayed as a moral imperative. Evidence has also suggested a positive association between the forgiveness of others and divine forgiveness (Chen et al., 2019). It is possible that in the midst of doubting one’s own faith, living the Christian ideal by forgiving others and modeling how one forgives others after God forgives them may also be instrumental in blunting some of the negative well-being consequences of experiencing such struggles. For religious believers, forgiving the transgressions of another involves not only those others who have sinned against a person, but also God (Krause & Ellison, 2003). Therefore, forgiving their transgressor may equip one with the knowledge that they are living up to the standards of their religion and could help alleviate the distress of struggles by acting out their faith in a meaningful way. As with self-forgiveness, we would also expect divine forgiveness to be the most efficacious in reducing the impact of religious doubt, since both occur in the religious domain; yet, we leave open the possibility that forgiveness of others could also be a tool in blunting some of the negative consequences.

Methods

We draw from the Religion, Aging, and Health survey, a nationally representative survey of Whites and African Americans. The population of the study was defined as all household residents who were non-institutionalized, spoke English, and were at least 66 years of age. The study population was also restricted to all eligible persons residing in the coterminous United States (i.e., residents of Hawaii and Alaska were not included). The study population also included only practicing Christians, people who were Christian in the past but do not practice any religion currently, or people who were not affiliated with a faith at any point in their lives.

A total of 1,500 face-to-face interviews took place in 2001. There was an oversampling of African Americans. The response rate for the Wave 1 survey was 62%. Sampling information for Waves 2 to 4 is as follows: Wave 2 (2004), n = 1024 (re-interview rate = 80%), Wave 3 (2007), n = 969, (re-interview rate = 75%), Wave 4 (2008), n = 718 (re-interview rate = 88%).

In June of 2013, a fifth wave of interviews was completed. At the time that Wave 5 interviews were completed, only 229 of the original study participants could be re-interviewed due to mortality (N = 611) or significant health problems. In addition to re-interviewing the 229 original study participants, this group was supplemented by an additional sample of 1,306 respondents who had not participated in the Religion, Aging, and Health Study previously. These individuals were selected to match the 2010 Census proportionately of the population aged 50 and over who were either White or African American. The age eligibility restriction was lowered at Wave 5 from 66 to 50 to make it possible to study religiosity in midlife as well as later life. Altogether, the re-interview rate for prior study respondents was 63%, and the response rate for new respondents was 45%. In total, 1,535 respondents participated in the Wave 5 interviews. The analyses presented below are based on the Wave 5 data only to increase sample size and to lower the age range so that our work can speak to a wider segment of the population. After eliminating those respondents who had missing data on our study variables, we are left with an analytic sample of n = 1497.

Measures

Dependent Variables

Depressive Symptoms:

There were four items taken from the Center for Epidemiologic Studies Depression Scale (Radloff, 1977) and used to measure depressive symptoms in the current study: (1) “I felt like I could not shake off the blues, even with the help of my family and friends,” (2) “I felt depressed,” (3) “I had crying spells,” and (4) “I felt sad. Responses were scored where 1 = “Strongly Disagree,” 2 = “Disagree,” 3 = “Agree,” and 4 = “Strongly Agree” and summed across all four items, creating an index that ranged from 4–16. A higher score on these items denotes a more depressed affect (α = .85).

Life Satisfaction:

Three items were adopted from the scale devised by Neugarten et al. (1961) to measure life satisfaction as well as one additional item. The items were: (1) “These are the best years of my life,” (2) “As I look back on my life, I am fairly well satisfied,” and (3) I would not change the past even if I could. These first three items were scored where 1 = “Strongly Disagree,” 2 = “Disagree,” 3 = “Agree,” and 4 = “Strongly Agree.” A final item asked, (4) “Now please think about your life as a whole. How satisfied are you with it?” Responses to this last item were scored in the following manner: 1 = “not very satisfied,” 2 = “not very satisfied,” 3 = “somewhat satisfied,” 4 = “very satisfied,” and 5 = “completely satisfied.” Scores were summed, creating a variable ranging from 4–17 where high scores denote greater life satisfaction (α = .73).

Focal Independent Variables

Religious Doubts:

These were measured as a sum of the following five items: (1) How often do you have doubts about your religious or spiritual beliefs?” (2) “How often do you have doubts about the things you’ve been taught in church?” (3) “How often do you doubt whether solutions to your problems can be found in the Bible?” (4) “How often do you doubt whether your prayers make a difference in your life?” (5)” How often do you doubt that God is directly involved in your life?” Response options for each question were coded 1 = “Never,” 2 = “Once in a while,” 3 = “Fairly often,” and 4 = “Very often,” where higher scores indicate greater R/S struggles. Scores on the R/S struggles scale ranged from 5–20 (α = .83).

Divine Forgiveness:

Beginning first with divine forgiveness, this was measured as the sum of the following two items, which respondents were asked to rate their level of agreement with. (1) “I have done some things that even God may not forgive” and (2) I believe God forgives me for the things I have done.” Both items were coded where 1 = “Strongly disagree,” 2 = “Disagree, 3 = “Neither agree nor disagree, 4 = “Agree,” and 5 = “Strongly agree.” Scores ranges from 0–10, where higher scores indicated greater divine forgiveness (α = .78).

Self-Forgiveness:

Two items comprised the scale of “self-forgiveness.” The two items were: (1) “I forgive myself for the thing I have done (1 = “Strongly disagree,” 2 = “Disagree,” 3 = “Neither agree nor disagree,” 4 = “Agree,” and 5 = “Strongly agree.”) (2) “How hard is it for you to forgive yourself?” (1 = “Just a little hard,” 2 = “Extremely hard,” 3 = “Somewhat hard,” 4 = “Just a little hard,” and 5 = I forgive others easily.”). Responses to these two items were summed to create a score ranging from 0–10, where higher scores indicate greater self-forgiveness (α = .74).

Other Forgiveness:

Two items comprised the “other forgiveness” scale. Respondents were asked to respond to the following prompts: (1) “How often do you forgive others for the things they have done to you?” (1 = “Not sure,” 2 = “Never,” 3 = “Once in a while,” 4 = “Fairly often,” and 5 = “Very often”). (2) “How hard is it for you to forgive others?” (1 = “Just a little hard,” 2 = “Extremely hard,” 3 = “Somewhat hard,” 4 = “Just a little hard,” and 5 = I forgive others easily.”). Responses to these two items were summed to create a score ranging from 0–10, where higher scores indicate greater forgiveness of others (α = .71).

Control Variables

A number of additional factors were included in all analyses. All analyses adjust for the gender of respondent (male = 1) and marital status (married = 1; 0 = otherwise) of the respondent. A measure of education is also included (coded continuously in years).

Additional controls adjust for several indicators of religiosity, including a measure of religious attendance (“How often do you attend religious services?”) Responses were initially coded into nine categories, but we ultimately coded attendance into a four-category variable, with (1) Never attends, (2) Attends yearly, (3) Attends monthly, and (4) Attends weekly or more. We also adjusted for religious denomination at Wave 5. This was coded where (1) Protestant, (2) Catholic, and (3) Other Christian.

Analysis

A series of regression models with robust standard errors were used to test our hypotheses. Listwise deletion was used to handle missing data, yielding a final sample size of n = 1497.

We considered a series of four models for each outcome of depression and life satisfaction. Model 1 regressed religious doubt on each dependent variable, controlling for divine, self, and other forgiveness as well as all demographic controls. This serves as a test of Hypothesis 1. The next three series of models tested a series of interaction terms. Model 2 tests our focal relationship, that testing whether divine forgiveness exacerbates (Hypothesis 2) or buffers (Hypothesis 3) the relationship between religious doubt and each indicator of well-being. Models 3 and 4 test whether forgiveness of others and self-forgiveness, respectively, moderate the relationship between religious doubt and both depression and life satisfaction. In cases where a significant interaction is detected, we use the margins command in Stata 15 to show predicted depressive symptoms or life satisfaction scores at three levels of religious doubt, low doubt (−1 SD below the mean), mean doubt, and high doubt (+1 SD above the mean) at three levels of divine forgiveness: low divine forgiveness (−1 SD below the mean), mean divine forgiveness, and high divine forgiveness (+1 SD above the mean).

Results

Table 1 presents descriptive statistics for study variables. Respondents reported an average depressive symptom score of 6.11 (SD = 2.72) on a 16-point scale. The mean of life satisfaction scores was 12.56 (SD = 2.56) on a 17-point scale. Our sample also reported average religious doubt scores of 8.09, but with a fair amount of variation (SD = 3.81) on a scale ranging from 5–20. Divine forgiveness was also found to be high (average 8.09 on a 10-point scale, SD = 1.65). Other forgiveness had a lower mean of 7.41 (SD = 1.32), while the mean of self-forgiveness was 8.59 (SD = 1.78).

Table 1.

Descriptive Statistics, 2013 Religion, Aging, and Health Survey (n = 1497)

	Mean/%	S.D.	Minimum	Maximum
Dependent Variables
Depressive Symptoms	6.11	2.72	4	16
Life Satisfaction	12.56	2.56	4	17
Focal Independent Variables
Religious Doubt	8.09	3.81	5	20
Divine Forgiveness	8.23	1.65	2	10
Other Forgiveness	7.41	1.32	2	10
Self-Forgiveness	8.59	1.78	2	10
Control Variables
Education (years)	13.08	2.24	3	20
Age	63.28	11.65	50	102
Married	41.55
Male	39.35
Black	35.58
Religious Affiliation
Mainline Protestant	40.79
Catholic	16.58
Other	42.64
Religious Attendance
Never Attends	11.69
Attends Yearly	25.61
Attends Monthly	13.20
Attends Weekly or More	49.51

Note. Standard deviations are omitted for categorical variables

Speaking to demographic characteristics, our sample had an average of 13.08 years of education (SD = 2.24), and respondents had an average age of 63.28 years. About 42% of the sample was currently married, and 40% of our sample was male was 36% of our sample was comprised of Black older adults.

Multivariable Regression Results

Depressive Symptoms

Table 2 presents results from a series of models predicting depressive symptoms. As shown in Model 1, greater religious doubt was associated with greater depressive symptoms (b = 0.09, p < .05, 𝛽 = 0.07), net of all three forgiveness types and all study control variables. This provides evidence in support of Hypothesis 1. As for the three types of forgiveness, we see in Model 1 that forgiveness of others is associated with lower depressive symptoms (b = −0.31, p < .001, 𝛽 = −0.14), as is greater self-forgiveness (b = −0.33, p < .001, 𝛽 = −0.11). Of note, divine forgiveness was not associated with depressive symptoms in the presence of these other two types of forgiveness. In a separate model where only divine forgiveness was included, in the absence of forgiveness of others and self-forgiveness, we note that divine forgiveness was associated with lower depressive symptoms (b = −0.22, p = 0.047), almost falling outside the conventional standards for significance.

Table 2.

Religious Doubt and Depressive Symptoms: Forgiveness Type as Moderators 2013 Religion, Health, and Aging Survey (n= 1497)

	Model 1	Model 2	Model 3	Model 4
Religious Doubt	0.09* (0.04)	−0.17 (0.21)	0.24 (0.17)	0.07 (0.20)
Focal Independent Variables
Divine Forgiveness	−0.11 (0.11)	−0.60* (0.25)
Other Forgiveness	−0.31*** (0.08)		−0.31 (0.16)
Self-Forgiveness	−0.33** (0.11)			−0.54* (0.22)
Religious Doubt*Divine Forgiveness		0.05* (0.02)
Religious Doubt *Other Forgiveness			−0.01 (0.02)
Religious Doubt*Self Forgiveness				0.01 (0.02)
Control Variables
Education (years)	−0.23** (0.08)	−0.21** (0.08)	−0.20** (0.07)	−0.25** (0.08)
Age	−0.01 (0.02)	−0.03 (0.02)	−0.02 (0.02)	−0.01 (0.02)
Married	−1.09** (0.36)	−1.21** (0.35)	−1.11** (0.34)	−1.31*** (0.35)
Male	−0.84* (0.36)	−0.60 (0.35)	−0.80* (0.35)	−0.67 (0.35)
Black	−0.86* (0.37)	−0.85* (0.36)	−0.95** (0.36)	−0.80* (0.37)
Religious Affiliation (ref. = Protestant)
Catholic	0.03 (0.05)	0.04 (0.05)	0.04 (0.05)	0.03 (0.05)
Other Christian	−0.01 (0.02)	−0.02 (0.02)	−0.02 (0.02)	−0.02 (0.02)
Religious Attendance (ref. = Never attends)
Attends Yearly	0.07 (0.06)	0.09 (0.06)	0.09 (0.07)	0.08 (0.07)
Attends Monthly	0.09 (0.06)	0.08 (0.06)	0.09 (0.06)	0.08 (0.06)
Attends Weekly or More	−0.11* (0.05)	−0.10* (0.05)	−0.11* (0.05)	−0.10* (0.05)
R2	0.07	0.10	0.08	0.09

Notes.

^*p < .05

^**p < .01

^***p < .001. Unstandardized coefficients and robust standard errors shown in parentheses

Model 2 tests the focal relationship of our study, that between religious doubt and divine forgiveness. We see in Model 2 the presence of a positive interaction term between religious doubt and divine forgiveness (b = 0.05, p < .05). This suggests an exacerbating pattern, where those with greater religious doubt who also report higher divine forgiveness report worse depressive symptoms, supporting Hypothesis 2. This can be seen clearly in Figure 1. Drawing attention to the third set of bars in Figure 1, we see that high divine forgiveness in the face of R/S struggles is associated with average depressive symptoms of 13.55. It is worthwhile to mention that for respondents in the other two groups, low religious doubt and mean religious doubt, high divine forgiveness is associated with lower depressive symptoms than those perceiving low or moderate divine forgiveness (11.27 and 11.41 depressive symptoms, respectively). On the basis of this evidence, then, we see support for Hypothesis 2: the association between greater religious doubt and greater depressive symptoms is exacerbated for those perceiving greater divine forgiveness.

Figure 1.

Religious Doubt and Depressive Symptoms: The Moderating Role of Divine Forgiveness

Note. Estimates are derived from Model 2 of Table 2. All other covariates are held at their respective means.

It is also notable that neither other forgiveness (Model 3) nor self-forgiveness (Model 4) significantly moderated the association between religious doubt and depressive symptoms.

Life Satisfaction

Table 3 presents results from an identical series of four models, this time taking life satisfaction as the outcome of interest. As shown in Model 1, greater religious doubt is associated with lower life satisfaction (b = −0.04, p < .05, 𝛽 = −0.05), again providing support for Hypothesis 1. In addition, perceiving greater divine forgiveness (b = 0.12, p < .05, 𝛽 = 0.08), other forgiveness (b = 0.19, p < .001, 𝛽 = 0.18) and self-forgiveness (b = 0.21, p < .001, 𝛽 = 0.15) are associated with greater life satisfaction. While the latter two associations were observed for depressive symptoms, greater divine forgiveness also predicts greater life satisfaction.

Table 3.

Religious Doubt and Life Satisfaction: Forgiveness Type as Moderators 2013 Religion, Health, and Aging Survey (n= 1497)

	Model 1	Model 2	Model 3	Model 4
Religious Doubt	−0.04* (0.02)	0.08 (0.10)	−0.10 (0.08)	0.11 (0.09)
Focal Independent Variables
Divine Forgiveness	0.12* (0.05)	0.38** (0.12)
Other Forgiveness	0.19*** (0.04)		0.21** (0.08)
Self-Forgiveness	0.21*** (0.05)			0.45*** (0.11)
Religious Doubt*Divine Forgiveness		−0.02* (0.01)
Religious Doubt *Other Forgiveness			0.01 (0.01)
Religious Doubt*Self Forgiveness				−0.02 (0.02)
Control Variables
Education (years)	0.04 (0.04)	0.07 (0.04)	0.06 (0.04)	0.06 (0.04)
Age	0.01 (0.01)	0.01 (0.01)	0.01 (0.01)	0.01 (0.01)
Married	0.84*** (0.17)	0.87*** (0.17)	0.89*** (0.17)	0.99*** (0.17)
Male	−0.02 (0.17)	−0.19 (0.18)	−0.07 (0.17)	−0.19 (0.17)
Black	0.73*** (0.18)	0.76*** (0.18)	0.83*** (0.18)	0.66*** (0.18)
Religious Affiliation (ref. = Protestant)
Catholic	0.10 (0.08)	0.11 (0.09)	0.12 (0.09)	0.10 (0.09)
Other Christian	0.14 (0.11)	0.13 (0.11)	0.12 (0.10)	0.12 (0.10)
Religious Attendance (ref. = Never attends)
Attends Yearly	0.17 (0.13)	0.16 (0.13)	0.15 (0.12)	0.16 (0.13)
Attends Monthly	0.16 (0.14)	0.15 (0.14)	0.15 (0.14)	0.14 (0.14)
Attends Weekly or More	0.20* (0.09)	0.21 (0.10)	0.20* (0.10)	0.22* (0.11)
R2	0.12	0.16	0.14	0.13

Notes.

^*p < .05

^**p < .01

^***p < .001. Unstandardized coefficients and robust standard errors shown.

In Model 2, we see evidence of a negative interaction term between religious doubt and divine forgiveness (b = −0.02, p < .05). Because life satisfaction scores are coded where higher scores indicate greater life satisfaction, this coefficient also indicates an exacerbating moderation pattern of divine forgiveness for respondents experiencing religious doubt. Figure 2 illustrates this relationship graphically. Indeed, as was observed for depression, those experiencing high religious doubt have lower life satisfaction scores (average = 12.29) compared with respondents who have low religious doubt and high divine forgiveness (13.27) and mean religious doubt (12.78). An identical conclusion is reached for life satisfaction as for depressive symptoms and Hypothesis 2 is supported: the association between greater religious doubt and lower life satisfaction is exacerbated for those perceiving greater divine forgiveness.

Figure 2.

Religious Doubt and Life Satisfaction: The Moderating Role of Divine Forgivenes

Note. Estimates are derived from Model 2 of Table 3. All other covariates are held at their respective means.

Mirroring the results for depressive symptoms, we find no evidence that forgiveness of others (Model 3) or self-forgiveness (Model 4) buffers or exacerbates the relationship between religious doubt and life satisfaction. We return in the discussion section to reflect on these patterns.

Supplemental Analyses

In additional analyses, we explored the possibility that religious doubts may be more troubling to the mental health of those who are actively involved in religious communities. Indeed, the exacerbating role of divine forgiveness for older adults experiencing high levels of religious doubts could be even worse for members of religious groups which place a premium on holding unwavering certainty in beliefs. It is possible that religious doubt could be detrimental for those actively involved in a religious community because it alienates them from the group and the positive sense of belonging that may follow. One prior study found that religious doubt was most harmful among those for whom religious is highly salient and a central identity category, who may face greater difficulties in dealing with doubt (Krause and Wulff 2004).

To test this possibility, we conducted a three-way interaction term between religious doubt, divine forgiveness, and religious attendance (as a gauge of religious community involvement). These interaction terms failed to achieve statistical significance for both our outcomes, depressive symptoms and life satisfaction. These results suggest that the relationship between religious doubt, divine forgiveness, and well-being does not further depend on religious participation, and may affect mental well-being on a more interpsychic level as older adults contemplate having greater doubts while still endorsing strong beliefs in a God who forgives.

Discussion

Religious doubt, long acknowledged to generate affective or cognitive distress and represent the “dark side” of religion, have been linked to lower mental well-being (Ellison & Lee, 2010). Existing studies on religious doubt have been instructive for illustrating how the doubting process can affect mental health, but we know less about the factors that can influence how an individual experiences religious doubt and its impact on mental well-being. The central goal of this study was to probe the relationship between religious doubt and two forms of mental well-being (depressive symptoms and life satisfaction) by considering the possible exacerbating or buffering role of divine forgiveness. It is well-established that religious doubts have a pernicious relationship with mental health and are both distressing and can compromise well-being (Ellison & Lee, 2010; Hill et al., 2021; Ellison & Krause, 2009; Upenieks, 2021). We sought to consider how divine forgiveness, a foundational concept in Christian faith, may factor into the overall association between religious doubt and mental well-being.

Our analyses revealed a series of key findings. First, we would note that, in keeping with previous research, greater religious doubt was significantly associated with greater depressive symptoms and lower life satisfaction among older adults. This is in keeping with research that has identified a “dark side” of religion, that is, elements of the religious life that may not be advantageous for mental well-being. In addition, we tested whether three forms of forgiveness (divine forgiveness, other forgiveness, and self-forgiveness) associated with mental well-being. We found that while stronger perceptions of both other forgiveness and self-forgiveness were associated with lower depressive symptoms and greater life satisfaction, divine forgiveness was only associated with greater life satisfaction. One interpretation of this finding is that forgiveness of others and of self may be easier to detect since they occur on the relational level and exist only in the human realm. As such, self and other forgiveness may be more consistently linked with mental well-being compared to divine forgiveness, which may be harder to pick up on and exists with a human-divine relationship, which is inherently more complex.

The most central findings of the study are found in the consistent moderation pattern of exacerbation that we observed between religious doubt, divine forgiveness, and both depressive symptoms and life satisfaction. Indeed, for older adults experiencing high religious doubt, they experienced greater depressive symptoms and lower life satisfaction if they also perceived high levels of divine forgiveness. The consistency of our findings across two outcomes allows us to place greater confidence in this observation.

We put forth two competing possibilities for how divine forgiveness may alter the relationship between religious doubt and mental well-being. Our results indicated no evidence that divine forgiveness was helpful in blunting the negative consequences to well-being for those who were experiencing the highest levels of religious doubt. At lower or moderate levels of religious doubt (at the sample mean or at one standard deviation below the sample mean of religious doubt, greater perceptions of divine forgiveness were indeed beneficial for well-being. It is possible that turning to God’s forgiveness in the midst of low or moderate doubt could be a source of comfort for older adults. Perceptions of divine forgiveness may stem from recalling religious teachings on divine forgiveness, God’s grace, or redemption, and may be an important source of hope for people to release themselves from the weight of their struggle (Abu-Raiya et al., 2016; Pargament et al., 1998). For those with low/moderate levels of doubt, divine forgiveness may be experienced as a positive form of religious coping, allowing them to restore a sense of peace with God and reminding them of the unconditional love and mercy of God even as they experience challenges in their faith (Worthington & Wade, 1999).

At the other end of the continuum, for those experiencing greater divine struggles, greater perceptions of divine forgiveness were associated with worse mental well-being. What might explain this finding? Though religious texts contain numerous references to a forgiving and merciful God, they also contain references to a God that can be anger, saddened, or disappointed by human sin, and whose relationships with humans can vary prior to and after we are forgiven. One feature of religious doubt which stands out is intense faith-based distress, that may be accompanied by disappointment, anger, or tension in one’s relationship with God (Ano & Vasconcelles, 2005). Such emotions may be accompanied by the view that one is unworthy or perhaps even incapable of experiencing forgiveness from God.

Based on these arguments, perceiving divine forgiveness under these conditions may represent a form of religious cognitive dissonance, where one perceives divine forgiveness even while feeling unworthy of it because they are questioning or struggling with core aspects of their faith. Some theologians have posited religious doubt as a temporary stumbling block on the road to greater faith, but this pattern of findings suggests that religious doubt alongside greater perceptions of (perhaps unearned) divine forgiveness is the least favorable combination for well-being. Perceiving divine forgiveness when faith is not also strong likely creates a scenario where the individual believes God is disappointed with them for having such doubts and uncertainties and may feel excess guilt in drawing from the wells of divine forgiveness in this state of confusion. Religious doubt accompanied by greater divine forgiveness could indicate that a person feels like they are not living up to their end of the bargain since they faith remains shaky, which may undermine self-worth and contribute negatively to well-being. While some have posited, “religion as problem, religion as solution,” (Abu-Raiya et al., 2016), our results unequivocally show that if religious doubt are a major problem for older adults, the solution is not likely to be found by drawing from divine forgiveness.

With the above in mind, it is also noteworthy that we did not document similar moderation patterns between religious doubt and forgiveness of others and self. Indeed, stronger reports of other forgiveness and self-forgiveness were not found to exacerbate nor buffer the pernicious association between religious doubt and depression and life satisfaction. While we posited that forgiveness of others (via modeling our forgiveness to reflect the way that God) and forgiveness of self (via self-compassion and seeing oneself worthy of divine forgiveness) might help those undergoing religious doubt, we observed no such evidence. This suggests that although parallels exist, divine forgiveness and forgiveness of self and others (earthly forms of forgiveness) should not be equated. God’s forgiveness, at least among Christian believers, has implications both in this world and for eternal life. Indeed, the consideration of eschatological beliefs could be important for understanding divine forgiveness fully. Divine forgiveness is also free from thoughts of unforgiveness of the harboring of ill will towards the offender, which may be difficult for humans to obtain even if they believe they have fully forgiven their offender, whether self or others (e.g., Wade & Worthington, 2003). In putting forth these unique dimensions of divine forgiveness, it becomes even more clear why divine forgiveness exacerbated the association between religious doubt and mental well-being. The bar for receiving divine forgiveness appears to be higher, yet may not be fully grasped or internalized for those experiencing a shaky faith foundation. Thus, perceiving divine forgiveness and the full restoration of one’s relationship with God in the midst of sin with these doubts in the forefront of one’s mind is a recipe for mental turmoil. That divine forgiveness was the only form that moderated the religious doubt → well-being association also suggests something important about the specificity of type of forgiveness, since both religious doubt and divine forgiveness occur in the religious domain. At least when it comes to a tensions and conflict about sacred matters and well-being, only divine forgiveness was pertinent.

There are several limitations which readers must bear in mind. One obvious concern is that our sample was comprised entirely of middle-age and older adults who were either currently or previously affiliated Christians. To advance the study of religious doubt, forgiveness, and well-being forward, future research should consider data from other religious denominations, perhaps ones where religious doubt and divine forgiveness occupy less central roles in the teachings of the faith than the majority Catholic and Protestant sample we utilized in the current study.

As a second concern, we drew from cross-sectional data in this study, which makes it impossible to establish causal order. It is plausible that perceptions of divine forgiveness (or a lack thereof) may predict varying levels of religious doubt, or that those with lower mental well-being would be more susceptible to greater religious doubt. While we believe the theoretical arguments outlined above have merit, they should be interpreted with caution until longitudinal research is done which replicates our findings.

Third, speaking to our measure of divine forgiveness, it would be worthwhile to know in greater detail what this is explicitly measuring. We lack answers to fundamental questions in the two items that we used. For instance, what is the process by which believers determine they have been forgiven by God? And if divine forgiveness is perceived to occur, how might it be manifested in the affect, cognition, or even behavior of a believer. All told, we lack a sophisticated understanding of how divine forgiveness may be experienced, especially in the context of broader religious doubt and in the latter years of the life span. However, a strength of our study is that we utilized a multi-item measure of divine forgiveness.

Despite these limitations, our study makes a valuable contribution to the literature, as we are the first to test whether divine forgiveness fundamentally alters the relationship between religious doubt and well-being in a nationally representative sample of older adults. This has been a neglected area of research, even as we experience rapid growth in empirical research on the “dark side” of religion (Ellison & Lee, 2010; Upenieks et al., 2022), which has identified scope conditions under which religious doubt negatively impacts well-being. Our study highlights the importance of divine forgiveness, while on its own associated with a positive boost in life satisfaction, tends to worsen the pernicious associations between religious doubt and mental well-being for older and middle-aged adults. Altogether, understanding the impact of divine forgiveness for those experiencing uncertainty in their faith is integral to gaining a more complete picture of religion’s “dark side,” as experiencing religious doubt might make one feel that they are unworthy of experiencing divine forgiveness. Research on the dark side of religion will remain incomplete without further consideration of the interplay of religious doubt and the uncertainty that accompanies them and perceptions of forgiveness by a divine entity, and we hope future research continues to pursue these objectives.

Data Availability Statement

The data that were used in this study are available from the first-listed author upon request.