What Forgiveness Is Not
The most important clarification in forgiveness research is what forgiveness is not. Forgiveness is not condoning — it does not mean the transgression was acceptable. It is not forgetting — the memory of the event remains. It is not reconciliation — you can forgive someone without continuing the relationship. It is not a single decision — it is a process that unfolds over time and often involves setbacks. And it is not weakness — research consistently shows that the capacity to forgive requires significant emotional strength and is associated with higher psychological resilience.
What Forgiveness Actually Is
The most widely accepted scientific definition comes from Worthington (2006): forgiveness is the replacement of unforgiving emotions — resentment, bitterness, hostility — with positive or neutral emotions toward the transgressor, motivated by empathy, compassion, or a recognition of shared humanity. It is fundamentally an emotional and motivational shift, not a cognitive decision.
This distinction matters because many people attempt to forgive by deciding to forgive — telling themselves they should let it go — without doing the emotional work that produces genuine forgiveness. Research by McCullough, Worthington, and Rachal (1997) found that the strongest predictor of forgiveness is empathy for the transgressor: the ability to understand (not excuse) their perspective and motivations. Empathy cannot be willed into existence; it requires genuine engagement with the other person's experience.
"The strongest predictor of forgiveness is empathy for the transgressor — not the severity of the transgression, not the length of the relationship."— McCullough, M.E., Worthington, E.L. & Rachal, K.C. (1997). Interpersonal forgiving in close relationships. Journal of Personality and Social Psychology.
The Health Benefits
The physical health benefits of forgiveness are among the most robust findings in relationship psychology. A 2001 study by Witvliet, Ludwig, and Vander Laan found that when people ruminated on a transgression, they showed elevated heart rate, blood pressure, and skin conductance — markers of physiological stress. When they imagined forgiving the transgressor, these markers returned to baseline. The cardiovascular reactivity reduction was 23% — a clinically meaningful effect.
Longitudinal research has linked higher dispositional forgiveness (the general tendency to forgive) to lower rates of depression, anxiety, and cardiovascular disease, and to higher life satisfaction and longevity. A 2005 meta-analysis by Toussaint and Webb found that forgiveness was more strongly associated with mental health outcomes than any other prosocial emotion, including gratitude and compassion.
The REACH Model
Everett Worthington developed the most empirically validated forgiveness intervention, the REACH model, which has been tested in over 20 randomised controlled trials. REACH stands for: Recall the hurt (without minimising), Empathise with the transgressor, Altruistic gift of forgiveness, Commit to the forgiveness publicly, and Hold onto forgiveness when doubt arises.
Meta-analyses of REACH-based interventions find an average effect size of 0.56 on forgiveness outcomes — a large effect by psychological standards. The model typically requires 6–8 sessions to produce significant progress, though the effects continue to develop after the intervention ends. Crucially, the model is designed to produce genuine emotional forgiveness rather than decisional forgiveness — the difference between actually releasing resentment and deciding to act as if you have.
What This Means for Your Relationships
If you are carrying resentment from a past or current relationship, the research suggests that the cost of not forgiving is primarily borne by you — in elevated , cardiovascular reactivity, and reduced wellbeing. This is not an argument for forgiving people who continue to harm you; it is an argument for doing the emotional work of forgiveness for your own sake, regardless of whether the relationship continues.
The research also suggests that forgiveness is a skill that can be developed. People who have engaged in structured forgiveness work — through therapy, the REACH model, or other evidence-based approaches — show lasting increases in their capacity to forgive across all relationships, not just the specific transgression they worked on. Forgiveness, like most psychological capacities, improves with practice.
Forgiveness is not condoning, forgetting, or reconciliation — it is an emotional process of releasing resentment. It has robust health benefits including 23% cardiovascular reactivity reduction. The REACH model is the most empirically validated intervention, requiring 6–8 sessions for significant progress.