RESEARCH_FILE
The Science of Grief
Most of what people 'know' about grief — five stages, a deadline, the idea that feeling okay means denial — comes from a 1969 book about dying patients, not from studies of bereaved people.
Fifty years of actual data tells a different story: resilience is the most common response, staying connected to the person you lost is healthy, and switching between grieving and getting on with life is how coping works. Here's how the science changed, and where the old script came from.
How the science changed
- 1969
Elisabeth Kübler-Ross publishes On Death and Dying, describing five stages — denial, anger, bargaining, depression, acceptance — based on interviews with terminally ill patients facing their own death, not with bereaved people. ↗
- 1989
Wortman and Silver's 'The Myths of Coping with Loss' reviews the evidence and finds that core assumptions — depression is inevitable, distress is necessary, loss must be 'worked through' — are unsupported and sometimes contradicted by data. ↗
- 1996
Klass, Silverman and Nickman publish Continuing Bonds, showing that maintaining an inner relationship with the deceased is common across cultures and typically healthy — overturning the idea that grief resolution requires 'letting go'. ↗
- 1999
Stroebe and Schut propose the Dual Process Model: healthy coping oscillates between confronting the loss and attending to life's practical demands — and taking breaks from grief is part of the process, not avoidance. ↗
- 2002
Bonanno and colleagues publish the first large prospective bereavement study (measuring people before and after loss): resilience — low, stable distress — is the single most common trajectory, at 45.9% of the sample. ↗
- 2004
In American Psychologist, Bonanno argues that resilience after loss is not rare or pathological but ordinary — and that its absence of visible grieving had long been misread by clinicians as denial or 'delayed grief'. ↗
- 2007
Maciejewski and colleagues run the first large empirical test of stage theory in JAMA: acceptance — not denial — is the most frequently endorsed response from the very start, and most bereaved people don't move through the stages in sequence. ↗
What people believe vs. what the data shows
The belief“Grief has five stages you must pass through, in order.”
The dataThe stages came from interviews with dying patients, not bereaved people, and were never validated as a required sequence. When tested empirically, most bereaved people didn't follow the order — acceptance was the most common response from the start. ↗
The belief“If you're not devastated, you're in denial — the grief will hit you later, worse.”
The dataResilience — low, stable distress — is the most common grief trajectory (45.9% in the first large prospective study), and prospective data show no support for 'delayed grief' as a common outcome of coping well early on. ↗
The belief“You have to do 'grief work' — fully confront and process the pain, or it comes back to haunt you.”
The dataThe grief work hypothesis was never empirically supported: reviews found no evidence that intensive 'working through' is necessary for recovery, and some evidence it can be associated with worse outcomes. ↗
The belief“Healthy grieving means letting go — cutting the bond and moving on.”
The dataContinuing bonds research shows that keeping an inner relationship with the person who died — talking to them, keeping objects, feeling their presence — is common across cultures and typically part of healthy adaptation, not a failure to grieve. ↗
The belief“Laughing, working, or having a good day means you're grieving wrong.”
The dataThe Dual Process Model describes healthy coping as oscillation: sometimes facing the loss, sometimes attending to daily life and restoration. Time off from grief is a built-in part of adaptive coping, not avoidance. ↗
TEST_YOURSELF · How well do you know this science?
01 Where did the famous 'five stages of grief' actually come from?
Kübler-Ross developed the stages in On Death and Dying (1969) from interviews with dying patients about facing their own death. The model was later applied to bereavement without being validated for it. source ↗
02 In the first large prospective study of bereavement (Bonanno et al., 2002), which trajectory was most common?
Resilience was the most common trajectory at 45.9% of the sample — measured from about three years before the loss through 18 months after. The 'expected' pattern of high distress then recovery fit only about 11%. source ↗
03 According to the Dual Process Model (Stroebe & Schut, 1999), what does healthy coping with loss look like?
The model identifies loss-oriented and restoration-oriented stressors and describes a regulatory process of oscillation: the griever at times confronts, at other times avoids, the different tasks of grieving. Both movements are part of coping. source ↗
04 What did the continuing bonds research (Klass, Silverman & Nickman, 1996) conclude about staying connected to someone who died?
Continuing Bonds (1996) documented that bereaved people across cultures maintain inner relationships with the deceased, and that these bonds are typically a normal, often helpful part of adaptation — not pathology. source ↗
05 When stage theory was finally tested in a large empirical study (JAMA, 2007), which response did bereaved people endorse most frequently from the start?
Maciejewski and colleagues found acceptance was the most frequently endorsed item throughout the study period — including right after the loss — while denial never dominated. Most people did not move through the stages in the theorized sequence. source ↗