Introduction

Research has focused on the impact of forgiveness on the forgiver. Important as he or she is, the forgiver represents only one relevant source of information…. We need to extend the focus of our research beyond the forgiver to the world of the perpetrator.

—Kenneth Pargament and Mark Rye, “Forgiveness as a Method of Religious Coping”

Forgiveness is a universal and timeless feature of human relationships and a centerpiece of most of the world’s major religions (Tsang et al. 2005). Every human being will at some point be faced with the need to be forgiven and the option to forgive an offender. This social fact partially accounts for why scientific inquiry into the nature of forgiveness has gained momentum in recent years. Since 1985, hundreds of studies have deepened our understanding of what was previously known about forgiveness (see Scherer et al. 2005; Worthington 1998), but this literature remains limited in several respects. First, as Pargament and Rye (1998) note, most of the forgiveness literature focuses on forgiveness from the victim’s point of view, which leaves unanswered an important empirical question: What can offenders teach us about forgiveness and its multiple dimensions? Second, the literature offers precious little insight into how people express forgiveness in natural, real-time, situation-specific settings during critical life moments. Most studies of forgiveness draw on information from survey respondents (Mullet et al. 1998) or college students (Maltby et al. 2001), or are retrospective studies in which subjects are asked to recall instances in which they received or offered forgiveness to others (Barrile 2015). Third, little is known about how people prioritize forgiveness at critical life moments, that is, whether in such instances they prefer certain expressions of forgiveness over others. Fourth, attempts to theorize forgiveness have led to little consensus about why people prioritize certain dimensions of forgiveness over others during critical life events. Finally, prior studies of forgiveness have failed to track change over time in how offenders express forgiveness, leaving questions about whether there is a hierarchy of preferred forgiveness expressions made in real-time, situation-specific settings and whether the hierarchy remains stable or changes over time.

This study attempts to fill these gaps in the forgiveness literature. I utilize unique data from the Texas Department of Criminal Justice (TDCJ 2016) to examine, in real time, how offenders (death row inmates) prioritize forgiveness in real time during the situation-specific moment of imminent death. Do offenders prefer certain expressions of forgiveness over others? Is there a definitive hierarchy of forgiveness? I use religious coping theory (RCT) to explain why certain religious expressions of forgiveness are used more frequently than others and I draw on prior literature to theorize about non-religiously based forgiveness expressions. Through an in-depth analysis of the last statements of death row inmates from 1982 to 2016, I demonstrate how and why the preferred hierarchy of forgiveness changes over time.

In the following sections, I first provide a brief discussion of the dimensions of forgiveness. Second, I offer religious coping theory as a plausible explanation foreshadowing why death row inmates might draw on certain forgiveness expressions over others as a means to cope with imminent death. I then home in on prior research on what offenders had to say about forgiveness at the moment of imminent death. Following a discussion of the data, analytical strategy, and results, I summarize the research findings, discuss their contributions to the literature, and explain the limitations of the study.

Dimensions of Forgiveness

While there has been little consensus regarding how forgiveness should be defined, a popular and concise definition describes it as a coping mechanism or a way of dealing “with a hurt or offense that primarily benefits the victim” (Pargament and Rye 1998). It is generally understood that forgiveness is not the same as reconciliation, or condoning or excusing hurtful behavior (Wade and Worthington 2005). In addition to the lack of definitional consensus, there is little agreement on what the most important dimensions of forgiveness are (Krause and Ingersoll-Dayton 2001). What we know thus far is useful for forming expectations regarding how death row inmates frame forgiveness at the moment of imminent death. Drawing on previous research, Toussaint et al. (2001, p. 250), delineated four dimensions of forgiveness: (1) self-forgiveness, the “release of negative affect and self-blame associated with past wrongdoings, mistakes and regrets”; (2) Forgiveness of others, “forgiving another for some harm done”; (3) forgiveness by God, “a belief or perception that one’s transgressions are forgiven by the divine”; and (4) proactive forgiveness, “initiating the process of giving and receiving forgiveness” (250). Using these four dimensions of forgiveness, the authors showed that forgiveness of others and feeling forgiven by God are more pronounced among middle-aged and older adults than younger adults. They also found a stronger association between forgiveness and self-reported mental and physical health among middle-aged and older adults compared with younger adults. In a follow-up study, Toussaint and Williams (2008) distinguished expressions of forgiveness made by survey respondents who self-identified as religious versus having no religious affiliation. A hierarchy of preferred forgiveness dimensions can be gleaned from their data based on the ranking of the mean scores they provide. Mean scores among self-described religious respondents prioritized feeling forgiven by God, followed by forgiving others, seeking forgiveness from God, and self-forgiveness. The hierarchy among nonaffiliates was somewhat different: feeling forgiven by God superseded self-forgiveness, which foreshadowed forgiving others, followed by efforts to seek forgiveness from God.Footnote 1

Before the work of Toussaint et al., most empirical investigations focused on only one dimension: forgiveness of others. The Toussaint research team showed empirically that there are multiple dimensions of forgiveness, and some dimensions are more relied upon than others. However, one self-described limitation of their study is the fact that they examined forgiveness without taking into account the contextual offense that precipitated the need for forgiveness in the first place, what they call a particular act of “wrongdoing, injury, or loss” (Toussaint et al. 2001, p. 256).

Whereas Toussaint et al. relied on a national survey to identify and measure important dimensions of religious forgiveness, Krause and Ingersoll-Dayton (2001) conducted a content analysis of 129 in-depth qualitative interviews of elderly respondents exploring how older Christians practice forgiveness. Three major forgiveness themes emerged from the interviews: a willingness to forgive others, what transgressors must do to be forgiven, and whether elderly respondents could forget as well as forgive. It was not clear whether respondents relied on certain forgiveness themes more than others. A related study by Krause and Ellison (2003) based on a national survey distinguished forgiveness of others from forgiveness by God. They found that forgiveness of others enhances psychological well-being and the positive effects of forgiving others were found to be greater than those associated with forgiveness by God. Krause and Ellison’s chief contribution is their comparison of the effects of forgiving others to the effects of receiving forgiveness from God. Similarly, Toussaint et al. (2001) concluded that forgiveness of others exerts a more beneficial effect on psychological distress and life satisfaction than forgiveness by God. The bulk of information describing the most common sources of forgiveness and how forgiveness should be measured emanate from self-ratings, in which respondents describe their own attitudes or behavior regarding forgiveness via questionnaires or interviews (McCullough et al. 2000). According to Hoyt and McCullough (2005, p. 113), this approach is problematic because the factors that researchers assume to be correlated with forgiveness are also measured by way of self-reports, a methodological limitation of unknown consequence. As a remedy, Hoyt and McCullough encouraged researchers to gather data using alternate methods, including life events approaches, which can offer an objective perspective on forgiveness from a targeted population.Footnote 2 Concurring with this approach, this study employs data drawn from the first-person, situation-specific accounts of the final words of death row inmates moments before their executions. An important contribution of this approach is that the most important dimensions of forgiveness are not predetermined a priori. Instead, the data reveals the most important dimensions of forgiveness, as expressed and prioritized by offenders in the natural setting of the death chamber. Such an inquiry is a significant departure from prior attempts to determine the most important dimensions of forgiveness based on previous literature (Toussaint et al. 2001; Toussaint and Williams 2008), in-depth interviews (Krause and Ingersoll-Dayton 2001), or self-reports (Thompson et al. 2005).

Forgiveness and Religious Coping Theory

A voluminous body of literature has established a remarkable correlation between forgiveness and improved physical health, psychological well-being, and enhanced personal relationships (Lawler-Row et al. 2008; Orcutt 2006; Hannon et al. 2010). Little is known, however, about the positive implications of forgiveness during critical life events. Why, for instance, would death row inmates express forgiveness at the moment of imminent death? Religious coping theory (RCT) offers an answer to this unexplored query. According to Pargament et al. (2000, p. 305), “The meaning of forgiveness and its implications for personal and social functioning can also be sharpened by integrating the construct into well-established theoretical and conceptual frameworks.” In this vein, Pargament et al. (2011) religious coping theory can be extended to incorporate an expanded understanding of forgiveness from the point of view of the offender. They define religious coping as “efforts to understand and deal with life stressors in a way related to the sacred” (2). In a broad range of studies, religious coping has been found to be associated with higher church attendance, lowered depression (Koenig 1995), improved mental and physical health, and diminished mortality rates (Harris et al. 1995; Oxman et al. 1995; Pargament et al. 1994).

For the purposes of this study, I am interested Pargament et al.’s (2000, 521) five methods of religious coping. They theorized that during stressful events, people use religion to (1) find meaning in their circumstance (e.g., redefining the stressor as a punishment from God for their sins), (2) establish control (e.g., active religious surrender—giving up control to God), (3) gain comfort from and closeness to God (e.g., seeking God’s forgiveness), (4) gain intimacy with other people (e.g., looking for spiritual support from others), and (5) achieve life transformation (e.g., religious forgiving—looking to religion to shift to a state of peace from the anger, hurt, and fear associated with an offense). Research shows these are the most comprehensive and the most frequently used measures in the religious coping literature (Pargament et al. 2011).

If RCT is correct, it is reasonable to expect death row inmates to express forgiveness at the moment of imminent death in order to find meaning; gain control, comfort and closeness to God, and intimacy with others; and achieve life transformation. As a corollary goal, I am interested in determining the extent to which offenders rely more heavily on certain coping mechanisms than others. A detailed examination of the qualitative content of preferred religious coping mechanisms can shed light on the reasons why offenders frame forgiveness the way they do at the moment of imminent death.

Forgiveness and the Last Statements of Death Row Inmates

The study of death row inmates’ last statements is an emerging area of empirical investigation. Researchers are attracted to last statements for a number of reasons: the natural settings in which they are elicited, their authentic finality and profound humanity, and sheer existential curiosity about what people say when they know death is moments away. What we know so far is instructive for formulating expectations regarding how death row inmates might prioritize forgiveness expressions. Several of the most noteworthy studies employ content analysis to uncover major themes in the last statement data. Heflick (2005) was among the first scholars to systematically document the presence of forgiveness expressions and other themes in the last statements of death row inmates. Heflick found forgiveness to be an important theme, along with belief in an afterlife and expressions of activism (e.g., promoting social causes and advice to others), appreciation and love, silence, and innocence. Schuck and Ward (2008, 49–50), going a step further, examined the Texas last statement data and ranked the most common themes. Expressions of love or appreciation aimed at family and friends (65%) topped Schuck and Ward’s list, followed by expressions addressing family (including the victim’s) and friends (55%), religious expressions (46%), forgiveness (39%), self-comfort (39%), acknowledgement of guilt or responsibility (18%), declarations of innocence (16%), and political statements (10%) lamenting the death penalty or the lack of fair treatment during their trial.

To date, Vollum (2008) and Vollum and Longmire’s (2009) analyses provide one of the most detailed examinations of death row inmates’ last statements, including a hierarchy of forgiveness among offenders. By hierarchy of forgiveness, I mean expressions of forgiveness that range from the most often used to the least often used expressions at the moment of imminent death. Frequency of usage can serve as an indicator of importance from the offenders’ purview. Vollum and Longmire’s analysis of 292 valid statements published by the state of Texas between December 1982 and March 2004 reveal that, more than any other forgiveness expression, offenders sought forgiveness from the victim’s family and friends (11%), followed by forgiving others (9.2%), seeking forgiveness from God (6.8%), an unspecified form of forgiveness (6.2%), and request for forgiveness from their own family and friends (2.7%) (Vollum and Longmire 2009). The present study updates this hierarchy by adding twelve additional years of data and examining whether the hierarchy has changed over time.

Other last statement research employs multivariate techniques to control for relevant background factors while predicting the type of apology offenders express. Eaton and Theuer’s (2009) analysis of the Texas last statement data found that an apology was 3.6 times more likely to be offered by offenders if they asked for forgiveness from the victim’s family.Footnote 3 A similar analysis by Cooney and Philips (2013) showed that death row inmates make expressions that affiliate themselves with God in an effort to raise their status, and this affiliation turns out to be the best predictor that inmates will apologize for their offense. While forgiveness is not an explicit feature of their analysis, it is arguably part of the nuanced understanding of the apology/God affiliation nexus they seek to convey. A study by Rice et al. (2009) on guilt, defiance, and repentance, based on last statement data, highlights the importance of context by showing that death row inmates are more likely to express repentance when victims’ family members and other co-victims are present at the execution. A recent study by Goranson et al. (2017) indirectly implicates the importance of forgiveness. The authors compared the last statements of death row inmates with simulated last statements created by noninmates who were asked to imagine they had been found guilty of a capital crime and would be executed the next day. The authors found the death row inmates’ last statements to be more positive and religious than the noninmates’. Their finding underscored the prominent use of positive religious coping methods under dire conditions and that religion is a useful method for reducing death anxiety.

The publication of death row inmates’ last statements is not confined to Texas. Upton et al.’s exploratory study (2017) examined the last words of inmates from Missouri from 1995 to 2011. While Upton et al.’s findings largely corroborated the six major themes found in Heflick’s (2005) study, forgiveness expressions were more prominently featured in the Texas data than the Missouri data (Upton et al. 2017, 390).

There is a small but growing body of research on the forgiveness expressions of death row inmates. As important as this research is, it lacks a coherent understanding of how offenders frame forgiveness, that is, whether they prefer certain expressions of forgiveness over others at the moment of imminent death. Equally absent in the literature is a clear theory that explains why offenders frame forgiveness the way they do. This omission is important because, as Younger et al. (2004) note, “an understanding of forgiveness motivation may, in turn, inform the definition of forgiveness. What forgiveness is may be heavily influenced by why it is offered” (840). And, finally, the last statement literature is silent on whether the most preferred forgiveness expressions have remained stable or have fluctuated over time.

Data and Plan of Analysis

To fill these gaps in the literature I draw on 537 records of inmates sentenced to death by the State of Texas from December 7, 1982, to April 6, 2016, accessed at the Texas Department of Criminal Justice website (TDCJ 2016). During that time death row inmates made 429 oral last statements, 115 of which included the word “forgive” or forgiveness” in religious (references to a deity, prayer, scripture, afterlife, etc.) and nonreligious narratives.

In contrast to studies of forgiveness that rely on convenience samples, college students, or retrospective interviews and questionnaires of victims, last statements are usually delivered orally moments before offenders are executed by lethal injection (Johnson et al. 2013).Footnote 4 Therefore, last statements offer a unique natural setting for studying forgiveness.

This study draws on data from Texas because it is one of the rare states in which last statement information is archived and made available to the public via the TDCJ (2016) website. In addition, Texas executes more prisoners than any other state even though, paradoxically, it is known for its relatively high religiosity (Pew Research Center 2016), an important conduit of forgiveness.

The analysis unfolds in three steps. First, as part of a larger project, I worked with two external coders to perform a content analysis of the 429 last statements. The goal was to inductively uncover major themes in the last statement data using open coding (Strauss and Corbin 1990). After reading each statement, we independently listed the most frequent expressions. To increase intercoder agreement (which exceeded 90%), we triangulated the three lists and came to a consensus on any ambiguities. Since prior research produced a comprehensive list of the most frequent expressions (Schuck and Ward 2008; Vollum and Longmire 2009), the second phase of the analysis focused only on detecting forgiveness expressions in the last statements. Statements were coded as having expressions of forgiveness only if they specifically included an explicit statement of forgiveness (n = 115).

Forgiveness statements that are embedded in an explicitly religious context are coded as religious forgiveness. If forgiveness is mentioned without an accompanying religious expression anywhere in the narrative, it is coded as a nonreligious form of forgiveness. For example, asking God for forgiveness of sin or to forgive others is coded as religious. Asking others for forgiveness in the absence of a religious or spiritual context is not coded as religious. Similarly, forgiving others in the absence of a religious or spiritual context is not assumed to be religious. As a corollary goal, I sought to determine whether inmates relied on certain forgiveness expressions more than others.

It was not uncommon to find multiple forgiveness expressions in one last statement as some inmates asked God to forgive others, acknowledged their own forgiveness from God, and extended forgiveness to others, as in this statement:

Heavenly Father, I give thanks for this time, for the time that we have been together, the fellowship in your world, the Christian family presented to me (He called the names of the personal witnesses.). Allow your holy spirit to flow as I know your love has been showered upon me. Forgive them for they know not what they do, as I know that you have forgiven me, as I have forgiven them. Lord Jesus, I commit my soul to you, I praise you, and I thank you (emphases added).

After isolating the most common themes of forgiveness and ranking them from most to least often employed, I examined the change in the use of forgiveness expressions over the 34-year period covered by the data (1982–2016). Pargament et al. (2011) raised the possibility that religious coping methods need not be static–-they may vary across time and across situations—yet little is known about how they might vary. To explore this issue I divided the data into three time periods: 1982–1995, 1996–2000, and 2001–2016.Footnote 5

Results

Change over Time in Forgiveness Expressions

I begin with a baseline analysis of change over time in the use of forgiveness expressions among death row inmates. Figure 1 tracks change over time in the percentage of forgiveness expressions in last statements from three perspectives. First depicted is the proportion of all forgiveness statements expressed as a percentage of all religious oral statements (top curve, N = 270)—a measure that provides insight into the extent to which religious inmates made use of the forgiveness motif. Second is the share of forgiveness statements expressed as a percentage of all oral statements (middle curve, N = 429)—an indicator of the degree to which all verbalized last statements (whether religious or not) featured forgiveness expressions. Last, the bottom curve features forgiveness expressions as a percentage of all executions (N = 537)—regardless of whether or not an oral statement was made during the execution process.

Fig. 1
figure 1

Change over time in forgiveness expressions among Texas death row inmates. All data in the present figure and in the figure and table to follow are derived from oral statements only. Omitted for theoretical reasons are written statements, “none” statements, and similar notations by the TDCJ, such as “This offender declined to make a last statement” or “No last statement,” and cases in which the inmate remained silent

Full size image

Several patterns are worth noting. First, from the early time period (1982–1995) to the recent time period (2001–2016), all three curves trend slightly upward. While the increases are not dramatic, the patterns suggest that inmates placed an increasing premium on expressing forgiveness across the 34 years covered by the data. Second, as expected, the lion’s share of forgiveness expressions are disproportionately represented among all religious statements in each time period (top curve), suggesting that offenders largely framed forgiveness within a religious context.

Last, following the implementation of new policies in 1996 and 1998 allowing co-victims, family members, and friends of the victim to witness the execution of the person who killed their loved one, the share of forgiveness expressions as a percentage of all religious statements increased from 39% in the middle period (1996–2000) to 43% in the most recent period. Similarly, forgiveness expressions as a share of all executions increased from 20% in the middle period to 25% in the most recent period. However, forgiveness expressions as a percentage of all oral statements only increased by 1% across the two latter time periods. Overall, these patterns suggest that expressions of forgiveness became increasingly important over time, but whether the increase is a function of the sudden presence of co-victims, family members, and friends at executions is an empirical question not addressed in the present study.

How Do Offenders Prioritize Forgiveness at the Moment of Imminent Death?

The answer to this question is displayed in Table 1 under columns 1 and 2. Table 1 shows that a sizable share (50%) of forgiveness expressions reflect efforts to seek religious forgiveness from others, compared to 40% for seeking forgiveness from God. That offenders would prioritize forgiveness from other people over forgiveness from God may seem paradoxical under the circumstances. However, it makes sense in light of the fact that in most cases, the offender’s last statement in the death chamber is the only opportunity to address co-victims.

Table 1 Hierarchy of forgiveness expressions among death row inmates in relation to religious coping theory
Full size table

While offenders found other dimensions of forgiveness to be important, they were less prominently featured in their last statements. They assigned much less priority to extending forgiveness to others in a religious context (17%), and even less to petitioning God to forgive others (11%).

Importantly, the data also show that forgiveness need not be expressed in a religious context. A full 11% of all forgiveness expressions represent efforts to seek forgiveness from others in a nonreligious context. A small share (3%) of forgiveness expressions were aimed at forgiving others in a nonreligious context. Thus, offenders employed multiple dimensions of forgiveness to express themselves at the moment of imminent death. Expressions were prioritized in a manner that signaled the magnitude of their importance, with expressions aimed at seeking religious forgiveness from others superseding nonreligious forgiveness expressions. Religious dominated nonreligious expressions of forgiveness by a substantial margin, suggesting that offenders viewed forgiveness through a multidimensional religious lens.

Why Do Inmates Frame Forgiveness the Way They Do?

Religious coping theory provides a useful framework to understand why condemned offenders’ last statements draw on certain themes of forgiveness. Using the forgiveness hierarchy depicted in Table 1 (column 1) below, I show how each dimension of forgiveness is linked to one or more of the five methods of religious coping (column 3) discussed earlier and I provide examples of each coping strategy as gleaned from the data.

Seeking Religious Forgiveness from Others (Religious Context)

That offenders place the highest premium on seeking forgiveness from others within a religious context (50%) is quite informative. In Pargament et al.’s (2000) scheme, these efforts are consistent with attempts to gain intimacy with people who are present as witnesses to the execution (family and friends of the victims). Typical expressions of this type of interpersonal forgiveness feature sentiments of hope, sorrow, prayer, and peace:

I hope that you can all forgive me. I pray that you can all forgive me.

I want to say I am sorry and I say a prayer today for you so you can have peace and I hope that you can forgive me.

Other attempts to gain intimacy with others feature afterlife expressions that inform witnesses that even beyond the grave it is possible for the offender to seek forgiveness in the next life from the individual(s) they have been convicted of murdering:

Please forgive me and I hope you find it in your heart to forgive me. He will be waiting in heaven for me. I will be able to talk to him and ask him for forgiveness personally. Forgive me for the pain.

These expressions of religious forgiveness at the moment of imminent death are noteworthy for their depiction of the primacy of the need for interpersonal forgiveness with other people, especially co-victims. Not surprisingly, such expressions occur within a religious context because religion (especially Christianity in this context) provides a language and narrative of forgiveness.

Seeking Forgiveness from God for Self

The second modal category of forgiveness, seeking forgiveness from God for oneself, is consistent with two religious coping methods: efforts to gain comfort from and closeness to God and efforts to establish control (Pargament et al. 2000). While praying to God for forgiveness as an attempt to gain comfort from and closeness to God is self-evident, asking God for forgiveness to establish control is better understood within religious coping theory as a type of active religious surrender where the inmate gives up control to God. This may sound peculiar given the existential reality of imminent death—the state has literal control over the inmate’s body and is now, as the inmate lies on a gurney, exacting punishment for the inmate’s crimes. But religious surrender in this context has little to do with what the state is about to do to the offender’s body. Instead, it is all about the soul at this point, and where the inmate believes he will spend eternity apart from the body. In fact, an often quoted Christian verse teaches that “to be absent from the body is to be present with the Lord” (Corinthians 5:8).

The lower proportion of forgiveness expressions aimed at seeking forgiveness from God for oneself relative to seeking forgiveness from others begs another interpretation. It is possible that some offenders have already sought forgiveness from God and believe it has been granted prior to entering the death chamber, as one offender seemed to imply:

Lord Jesus Christ [is] in my life, I know he has forgiven me, I have accepted his forgiveness.

Other offenders, however, take the occasion to explicitly broach the issue in front of witnesses, whether they have privately asked God for forgiveness prior to entering the death chamber or not:

I pray the Lord grant me forgiveness.

Please Lord forgive me. I ask the Lord to please forgive me. Father God, I ask you too for forgiveness. I ask you for forgiveness, Lord.

When viewed together, these data lend credence to Pargament and Rye’s (1998) contention that forgiveness is a method of religious coping. When offenders are given the opportunity for the first time to address their victims’ family and friends and, in a separate viewing area, their own family and friends—all of whom serve as witnesses to the execution (Johnson et al. 2013)—they do so in an environment that provides a unique and informative natural laboratory to study forgiveness.

Seeking to Forgive Others (Religious Context)

The third highest frequency of forgiveness expressions, seeking to forgive others in a religious context, can also be interpreted as an attempt to gain intimacy with others. Some of the most noteworthy expressions of this construct are distinguished because they are joined with other expressions of forgiveness that are simultaneously admonishing and conciliatory in tone:

What is about to transpire in a few moments is wrong! However, we as human beings do make mistakes and errors. This execution is one of those wrongs yet doesn’t mean our whole system of justice is wrong. Therefore, I would forgive all who have taken part in any way in my death. Also, to anyone I have offended in any way during my 39 years, I pray and ask your forgiveness, just as I forgive anyone who offended me in any way. And I pray and ask God’s forgiveness for all of us respectively as human beings.

Drawing on the Christian notion that you must forgive in order to be forgiven, one offender put the matter this way:

I want everybody to know that I hold nothing against them. I [have] forgiven them all…. I’ve been praying for [my victim’s] wife to drive bitterness from her heart because that bitterness that’s in her heart will send her to Hell just as surely as any other sin.

Although hell rarely appears in last statements, its strategic use in the above statement is noteworthy mainly because instead of being a feared outcome for the inmate, it is presented as a potential destination for a relative of one of the inmate’s victims, not because of any crime she has committed, but because of her perceived lack of forgiveness (“bitterness in her heart”) toward the offender. This may seem peculiar and ironic given the offender’s circumstances, but it is consistent with the biblical injunction to forgive: “For if you forgive other people when they sin against you, your heavenly Father will also forgive you. But if you do not forgive others their sins, your Father will not forgive your sins” (Matthew 6:14–15).

Seeking Forgiveness from God for Others

When refracted through the prism of religious coping theory, seeking forgiveness from God on behalf of others is consistent with efforts to gain comfort from and closeness to God and to gain intimacy with others. Consider the following last statement:

Lord, be merciful with those who are actively involved with the taking of my life, forgive them as I am forgiving them.

Expressing the same sentiment, one inmate echoed the last words of Jesus found in Luke 23:34:

God Forgive them. God forgive them for they know not what they do.

Another inmate offered a more succinct thought on the matter:

I hope God forgives all of you too.

In this context, it’s clear that offenders who implore God to forgive those involved in the execution process are, either directly or indirectly, attempting to level the playing field between themselves and co-victims with God as mediator. Thus, expressing forgiveness in this context is also consistent with the religious coping desire to achieve life transformation, an act of religious forgiving that helps the offender shift from the anger, hurt, and fear associated with the offense (Pargament et al. 2000) to one of hope in an afterlife made possible by forgiveness.

Nonreligious Forgiveness

Seeking Forgiveness from Others (Nonreligious)

We learned from Table 1 that a nontrivial share of forgiveness expressions were framed within a nonreligious context, such as efforts to seek forgiveness from others (11%). By definition these expressions do not fit within religious coping theory. However, these utterances, like the others above, are directly aimed at witnesses to the execution. While the systematic study of seeking forgiveness has garnered very little theoretical and empirical attention, forgiveness has been defined as “a motivation to accept moral responsibility and to attempt interpersonal reparation following relational injury in which one is morally culpable (Sandage et al. 2000). Even if forgiveness is treated as a nonreligious construct, offenders who seek forgiveness offer compelling and interpersonal acknowledgements of guilt:

I hope you can find it in your heart to forgive me. I hope they forgive me.

I know I am wrong but I am asking y’all to forgive me.

I would like you to reach down in your hearts and forgive me. Please forgive me.

Other nonreligious attempts to seek forgiveness from others strike a deep and personal chord aimed at a member of the offender’s own family:

Mother, I am sorry for all the pain I’ve caused you. Please forgive me.

Mike Graczyk, a recently retired AP reporter who has witnessed over 429 executions in Texas and written stories on over 400 death row inmates, described the moments when the inmates appeal to their mothers as the most gripping experience he has ever witnessed. According to Graczyk, during such moments inmates tended to be in a “very bad way” and “real emotional.”Footnote 6

Seeking to Forgive Others (Nonreligious)

As with religiously motivated efforts to forgive others, the very small share (3%) of inmates who offered forgiveness to others in a nonreligious context did so by indicting the legal system or calling out individuals they believed to be responsible their predicament:

I forgive everyone for this process which seems to be wrong.

Mindy, I’m with you, honey. I do not know why, Mindy, you are doing this, but I will still forgive you.

Other nonreligious statements of forgiving others target specific co-victims by name and offer clear evidence that some inmates saw themselves as both offender and victim:

Mr. Bryant, I have wronged you and your family and for that I am truly sorry. I forgive and I have been forgiven.

This finding provides a window into the offender/victim dichotomy that can inhabit a single person. That this dualism occurs sometimes in a nonreligious context belies the notion that such expressions are attempts to gain intimacy with others, as religious coping theory would predict. Instead, this begs the question, what, if anything, does the inmate gain from nonreligious efforts to forgive others? Drawing on prior forgiveness literature framed from the victim’s point of view, it is reasonable to assume that when inmates see themselves as victims, the pronouncement of forgiving others enhances their psychological well-being (Krause and Ellison 2003; Toussaint et al. 2001), even though certain death is only moments away.

Tracking Change in the Hierarchy of Forgiveness over Time

Having established a clear hierarchy of forgiveness among death row inmates in an effort to delineate how offenders frame forgiveness, and having employed religious coping theory to understand why offenders frame forgiveness the way they do, I now consider whether the hierarchy of forgiveness depicted in Table 1 remains constant or changes over time. While such an exercise in this context is largely descriptive, uncovering a constant hierarchy of forgiveness provides insight into the level of reliance placed on specific forgiveness expressions across time and across offenders. Conversely, fluctuation over time in the hierarchy of forgiveness may signal a shift in forgiveness priorities as a method of coping with imminent death. To adjudicate between these two possible outcomes, I subdivided forgiveness expressions into three time periods, displayed in Fig. 2. Several noteworthy patterns emerge. While the aggregated data displayed in Table 1 shows that efforts to seek forgiveness from others lay at the top of the forgiveness hierarchy, Fig. 2 shows that this was not always the case. In fact, in the first time period (1982–1995), a larger share of offenders sought forgiveness from God for themselves (29%) than from others (18%). However, by the middle period (1996–2000), the two modal categories reversed in order and a larger proportion of offenders (35%) sought forgiveness from others in a religious context than sought forgiveness from God for themselves (26%).

Fig. 2
figure 2

Change over time in the hierarchy of forgiveness expressions. The sixth category in Table 1, seeking to forgive others in a nonreligious context, was omitted due to excessively small cell counts

Full size image

This pattern continued into the recent period (2001–2016), with the proportion of inmates seeking forgiveness from others substantially increasing from 35% (middle period) to 43% (recent period), along with inmates seeking forgiveness from God for themselves, which increased from 26% in the middle period to 32% in the recent period. Overall, these strongly suggest that offenders became increasingly interested in seeking forgiveness from others (over seeking forgiveness for themselves from God) immediately following the implementation of the new witness policy (1996 and 1998) and well beyond its implementation (2001–2016).

The proportion of inmates who sought to forgive others (religious context) precipitously declined over the three time periods from 24% in the first period, to 19% and 10%, in the second and third periods. Efforts to seek forgiveness from others in a nonreligious context changed little from the first period (12%) to the second period (13%), but a substantial decline occurred by the recent period (7%). Finally, efforts to seek forgiveness on behalf of others substantially declined from the first period (18%) to the second period (6%), with a slight increase by the recent period (8%).

Two big takeaways from Fig. 2 are worth underscoring. First, the 34-year period covered by the data witnessed a steady increase in death row inmates’ need to seek religious forgiveness from others or, as RCT would suggest, offenders increased their efforts to gain intimacy with others (e.g., co-victims) and the increase continued well after the implementation of the 1996 and 1998 witness policies. Second, efforts to seek personal forgiveness from God (i.e., establish control and gain comfort from and closeness to God, in RCT terms) waxed and waned over time even as it remained the second highest modal category in each time period.

Discussion and Conclusion

My empirical analysis provides a more nuanced understanding of forgiveness from the offender’s point of view than previously reported. Three previously unanswered questions framed my analysis (1) how do death row inmates prioritize forgiveness expressions at the moment of imminent death? (2) Why are certain expressions of forgiveness used more frequently than others? (3) Does the hierarchy of forgiveness change over time? With regard to the first question, offenders prioritize forgiveness by seeking religious forgiveness from others, followed by efforts to seek forgiveness from God for themselves. From there, a declining share of offenders employ forgiveness expressions in an effort to forgive others (religious context) and seek forgiveness from others (religious context). The data also showed that not all expressions of forgiveness are framed within a religious context. Offenders did seek forgiveness from others and, to a much lesser degree, sought to forgive others within a nonreligious context, but these efforts represented a relatively small share of forgiveness expressions. By and large, the offenders in this study viewed forgiveness through the lens of their religion (i.e., Christianity).

Religious coping theory offers a useful but incomplete explanation of why offenders prioritized certain expressions of forgiveness over others. Offenders who expressed themselves in forgiveness terms within a religious context did so, first and foremost, to gain intimacy with others. This need to gain intimacy with others (via requesting their forgiveness) is likely a function of the fact that the last statement represents the only time (in most cases) offenders have the opportunity to speak directly to the family and friends of their victims. Thus, it stands to reason that seeking the forgiveness of co-victims would rank highest on their list of forgiveness expressions. With a little less primacy, offenders are also mindful of their need to seek forgiveness from God, which religious coping theory regards as an effort to gain comfort from and closeness to God but also to gain control over the imminent death process (i.e., seeking divine forgiveness relinquishes control to God). Religious coping theory also provides a useful explanation for the two remaining religious forgiveness expressions: seeking to forgive others and seeking forgiveness from God on behalf of others. Both expressions are consistent with religious coping efforts to gain intimacy with others, such as the witnesses to the execution.

But when it comes to explaining forgiveness expressions found outside an explicit religious context, other conceptual frameworks must be employed. The data show that a full 11% of forgiveness expressions are efforts to seek forgiveness from others in an explicitly nonreligious context. One major consequence of ignoring what offenders have to say about forgiveness is that we miss opportunities to inform theory and empirical research about the nuanced contours of seeking forgiveness (Sandage et al. 2000).

Sandage et al. (2000) view seeking forgiveness as multidimensional, involving, among other things, acceptance of moral responsibility, empathy with one’s victims, adaptive guilt, and behaviors of reparative action (e.g., apology, confession, and restitution). Thus, at the moment of imminent death, offenders who seek forgiveness within a nonreligious framework may be empathizing with their victims’ family and friends, and accepting guilt and moral responsibility for their crimes. Additional research is needed to further untangle the underlying coping mechanisms that might be at play here.

That just 3% of offenders sought to forgive others in a nonreligious context further suggests that they largely conceptualized forgiveness in religious terms. Nonreligious expressions of forgiveness can be viewed as a type of coping strategy to enhance psychological well-being (Krause and Ellison 2003; Toussaint et al. 2001) at the moment of imminent death.

Finally, analysis of change over time shows that following the implementation of policies that allow the family and friends of murder victims to witness an offender’s execution, the share of religious efforts to seek forgiveness from others surpassed efforts to seek forgiveness from God, with the former continuing to outpace all other forgiveness expressions in recent years (2001–2016). Viewed through the lens of religious coping theory, the change over time analysis demonstrates that offenders grew increasingly interested in gaining intimacy with others (e.g., witnesses in the execution chamber) via the pursuit of interpersonal forgiveness at a pace that superseded efforts to seek forgiveness from God—although the prominence of both forms of forgiveness increased over time while other forgiveness expressions declined.

This study contributes to the extant literature in several ways. First, this study confirms that forgiveness is a way of religious coping (Pargament and Rye 1998). However, while religious coping theory is useful for explaining why offenders might want to draw on forgiveness expressions at the moment of imminent death, it does not further theorize why such expressions emerge in a nonreligious context or postulate the form nonreligious expressions of forgiveness might take. By uncovering the spontaneous use of both religious and nonreligious expressions of forgiveness at the moment of imminent death, I demonstrate that when given the opportunity to express forgiveness in natural settings, offenders conceptualize forgiveness in mostly sacred but also secular terms.

Second, very few studies conceptualize and empirically assess forgiveness from the offender’s perspective (Pargament and Rye 1998). By drawing attention to how offenders prioritize forgiveness at the moment of imminent death, I demonstrate that not all dimensions of forgiveness are created equally. At least with regard to the imminent death process, the urgency of the moment privileges religious forgiveness strategies (over nonreligious strategies) aimed at seeking forgiveness from others and personal forgiveness from God. Along with the theoretical contribution discussed above, this provides a more nuanced understanding of what we previously knew about forgiveness in the last statement literature (Vollum and Longmire 2009).

Third, previous studies do not seek to explain why certain expressions of forgiveness are prioritized over others at the moment of imminent death. I contribute to the existing literature by suggesting via religious coping theory and prior research that forgiveness expressions are largely coping mechanisms aimed at gaining intimacy with others and closeness to God, and an attempt to establish a sense of control (via God) during a situation in which the offender lacks control. In a nonreligious context, offenders are motivated to express empathy and guilt with their victims and enhance their own psychological well-being.

Finally, prior research has largely neglected the study of change over time in forgiveness expressions.Footnote 7 By examining change over time in the hierarchical dimensions of forgiveness employed by offenders, I demonstrated that forgiveness, in addition to being multidimensional, is also nonstatic. It is reasonable to assume that as the conditions under which death row inmates were executed in Texas (via lethal injection) changed between the first and second time periods covered by the data, so did the prioritization and content of forgiveness expressions. While it was beyond the scope of this study to control for potential confounders in a multivariate framework, the qualitative analysis employed here hints at the underlying causal mechanisms linking expressions of forgiveness to the implementation of the new witness policy. Indeed, the descriptive patterns found in the data can inform the expectations of future research in this nascent area of forgiveness studies.

In conclusion, this study underscores that offenders and victims often inhabit the same person. As prior research has observed, a single individual can simultaneously possess both attributes (Exline and Baumeister 2000), in that offenders can often “usurp” victim status (Zechmeister and Romero 2002), such as when they seek to forgive others in both religious and nonreligious contexts.

That offenders made sure to enshrine most of their forgiveness expressions in a religious context may partially be a function of the influential role that chaplains and other religious leaders play in the weeks and days leading up to the execution (Purdum and Paredes 1989). In this context, religious leaders, including inmate ministers assigned to death row (Hallett et al. 2016), have the time and opportunity to share with offenders the Christian conceptions of forgiveness. It stands to reason that the promise of an afterlife coupled with particular Christian precepts—such as that God forgives for the sake of the offender (Maier 2017) and that no sin except blasphemy (Matthew 12:31) is unforgiveable—offers a certain sense of security for the religious offender.Footnote 8 It is easy to see how these important tenets can furnish the religious inmate with a “language and framework for forgiveness” (Hallett et al. 2016, 89). Hallett et al. (2016) note how many inmates experience an “existential crises” that “might also motivate offenders to search for answers about existence and the meaning and purpose of life by adopting a new interpretative system that offers guidance, meaning and forgiveness” (90).

One forgiveness dimension discussed in prior studies is conspicuous for its absence in the hierarchy of forgiveness reported here: self-forgiveness (Thompson et al. 2005; Zechmeister and Romero 2002), or what Toussaint et al. (2001) refer to as “the release of negative affect and self-blame associated with past wrong-doings, mistakes, and regrets” (250). This may be because the vast majority of forgiveness expressions are offered within a Christian context. Scholars note that while self-forgiveness may be popular in the psychotherapeutic and counseling literature, it is not found in the Judeo-Christian tradition as it is believed that only God or the victim of an offense can offer forgiveness (Vitz and Meade 2011). Thus, self-forgiveness is usually not situated within Christian theology, which is not the same as saying that Christians don’t engage in self-forgiveness when primed by researchers to do so (Sandage et al. 2000).

This study has certain limitations. Many offenders who provided a last statement failed to express themselves in forgiveness terms (religious or otherwise) despite a situational context that favors such expressions. Also, a total of 108 offenders gave no statement at all, leaving us wondering how they may have framed forgiveness. The data are limited to the state of Texas, and it is unknown whether the findings are generalizable to other states (Rice et al. 2009; Vollum 2008). An analysis of last statement data from Missouri yields results similar Texas (Upton et al. 2017). Finally, despite discrepancies found between last statements reported by journalists who attended the execution and those reported by the TDCJ (Vollum 2008), the number of such cases has been found to be negligible (Lester and Gunn 2013).

This study showed that the last statements of death row inmates offer a nuanced and novel source of data demonstrating how offenders conceptualize and experience forgiveness at the moment of imminent death. Hopefully, the hierarchy of forgiveness reported here, and the changes it endures over time, will inform future attempts to theorize and empirically investigate forgiveness. Meanwhile, the findings suggest that our conventional understanding of forgiveness is incomplete in that we ignore the forgiveness perspective of offenders or disregard the possibility that forgiveness priorities may change over time.