Continuing Bonds Theory: A Modern Framework for Grief

The idea that grief has an endpoint was always a cultural assumption, not a fact

For most of the twentieth century, Western grief advice pointed in one direction: toward letting go. Textbooks talked about "resolution," therapists talked about "closure," friends and family — meaning well — urged the bereaved to "move on." The unspoken assumption was that a healthy person eventually severed the emotional tie to the one who had died, tucked the memory away, and returned to normal life. If you were still talking to your mother a year after her funeral, still keeping your husband's shirts in the closet, still setting a place for your child at Thanksgiving, the culture had a word for you: stuck.

Then, in 1996, three researchers published a book that quietly upended that framework. Dennis Klass, Phyllis Silverman, and Steven Nickman edited a volume titled Continuing Bonds: New Understandings of Grief, and it did something the field had not seriously done before: it listened to what grievers were actually doing. What they found was that people across cultures, religions, and family structures were maintaining relationships with the people they had lost — and that, far from being pathological, these ongoing connections were often exactly what allowed them to live well after loss.

Nothing that follows is prescriptive. There is no single right way to grieve, and there is no schedule your relationship with the person you loved is supposed to keep. What continuing bonds theory offers is not a formula but a wider frame — a way of understanding experiences that many grieving people have quietly wondered whether they were allowed to have. If any of the practices described here feel meaningful, that is a signal to try them. If they do not, that is a signal to trust yourself and leave them alone.

The theory that emerged, now known simply as continuing bonds, has become one of the most influential frameworks in modern bereavement research. It gives permission for something most grieving people already sense: the relationship does not have to end just because the person has died. This article walks through where the theory came from, what the research has learned in the thirty years since, how continuing bonds show up in daily life, and — importantly — how to tell the difference between a bond that is helping you heal and one that may be keeping you stuck.

The 1996 book that changed the field

Before 1996, the reigning models of grief were built on an assumption of detachment. Sigmund Freud's 1917 essay Mourning and Melancholia introduced the idea of "decathexis" — the psychological work of withdrawing emotional energy (cathexis) from the lost person so it could be reinvested elsewhere. Grief, in Freud's account, was labor whose purpose was letting go. John Bowlby's mid-century attachment work refined the picture but kept the basic arc: healthy mourning ended with the survivor emotionally free of the deceased. Elisabeth Kübler-Ross's famous five stages of grief, published in 1969, were originally observations of dying patients rather than the bereaved, but they were quickly repurposed into a popular map of mourning that ended in "acceptance" — often read, culturally, as "done."

Klass, Silverman, and Nickman were not the first to notice that grievers did not actually follow this script. But they were the first to compile the evidence into a serious scholarly challenge to the detachment paradigm. Silverman's work on widows through the Harvard Bereavement Study, Klass's decade-long research with a chapter of The Compassionate Friends (a support organization for bereaved parents), and Nickman's clinical work with adopted children all pointed in the same direction: the people who seemed to be doing best were not the ones who had "let go" but the ones who had found a new way to stay in relationship with the person who died.

The bereaved parents in Klass's research, for example, were not trying to erase their children. They kept bedrooms preserved for a while, then reorganized them; they wore jewelry made from a child's handwriting; they spoke to their child at the cemetery; they described feeling their child's presence at specific moments. These were not signs of pathology. They were the architecture of a changed but continuing relationship.

The book proposed a simple but radical reframe: bereavement is not the severing of a bond but its transformation. The relationship changes form. It moves from a physical, back-and-forth relationship with a living person to an internal, symbolic, remembered relationship with a person who is no longer here. That reframe gave researchers, clinicians, and grievers a new vocabulary — and gave permission to millions of people who had quietly been sustaining these bonds anyway.

What "continuing bonds" actually means

A continuing bond, in the technical sense, is any ongoing inner relationship a bereaved person maintains with the deceased. It is not one behavior but a family of them. Common expressions include:

• Internal representation — carrying an inner sense of the person: what they would say, how they would react, what they valued.
• Ongoing dialogue — talking to the person, silently or aloud, at the graveside, in the car, before big decisions.
• Sensed presence — a felt sense that the person is nearby, particularly at meaningful moments or in familiar places.
• Dreams — vivid dreams in which the person appears, sometimes conveying comfort, sometimes unfinished conversation. Many bereaved people find these dreams of a loved one deeply meaningful.
• Signs and symbolic communication — noticing coincidences, natural phenomena, or objects that feel connected to the person. Research on signs from a deceased loved one suggests these experiences are far more common than most people admit, cutting across religious and secular identities.
• Keeping and using objects — wearing their sweater, cooking with their pan, keeping voicemails, driving their car.
• Living out their values — asking "what would they want?" and letting the answer shape decisions.

None of these is required. None is a sign of "doing grief right." What continuing bonds theory says is that when these expressions happen, they are usually part of healthy adaptation — not evidence that someone is failing to move on.

The research base since 1996

In the three decades since Continuing Bonds was published, researchers have moved from asking "do these bonds exist?" (they clearly do) to asking a more nuanced question: when do continuing bonds help, and when do they hurt? Because grief researchers noticed something important early on: not every bereaved person who maintained a strong bond was thriving. Some seemed comforted by it. Others seemed tormented.

Field and Filanosky (2010): externalized vs. internalized bonds

A pivotal contribution came from Nigel Field and Charles Filanosky, whose 2010 study in the journal Death Studies distinguished two broad types of continuing bond. Internalized bonds involve using the deceased as a secure inner base — remembering their values, imagining their perspective, feeling their steady influence. Externalized bonds involve experiences that seem to place the deceased outside oneself — hallucinations, illusions, or the persistent sense that the person is still physically present in a way that blurs the reality of the death.

Field and Filanosky found that internalized bonds were generally associated with better adjustment, while externalized bonds — particularly those tied to violent or sudden loss — were more often linked to complicated grief symptoms. Importantly, the researchers were careful not to pathologize the experiences themselves. Sensing a loved one's presence, for example, is extremely common and often comforting. The concern was specifically about experiences that seemed to deny the fact of the death or that intensified suffering rather than easing it.

Root and Exline (2014): comfort bonds vs. distress bonds

Benjamin Root and Julie Exline, writing in Death Studies in 2014, offered another useful distinction. They proposed that continuing bonds can be sorted along a spectrum from comforting to distressing. A comfort bond is one in which the ongoing connection soothes, orients, and supports meaning-making. A distress bond is one in which the connection is dominated by longing, guilt, unresolved conflict, or yearning that resists integration.

Their finding, replicated in later work, is subtle but important: the presence of a continuing bond does not, by itself, predict how someone is doing. What matters is the quality of the bond and the emotional experience it generates. A widow who talks to her husband each morning and feels steadied by it is doing something very different from a widow who talks to her husband each morning and feels the freshness of his absence pierce her every time. Same behavior, very different experience.

The meta-analytic picture

Broader reviews of the continuing bonds literature — including work synthesizing dozens of studies — have converged on a consistent conclusion: continuing bonds are usually adaptive. When they correlate with distress, the distress is more often explained by other variables — the suddenness of the death, unresolved conflict with the person who died, the survivor's own attachment history, or lack of social support — than by the bond itself. Robert Neimeyer, one of the field's most cited researchers, has repeatedly emphasized that grief is fundamentally a process of meaning reconstruction, and continuing bonds are one of the primary ways bereaved people rebuild a coherent sense of self and world after loss.

Put simply: staying connected is not the problem. When there is a problem, the bond is usually the messenger, not the cause.

Healthy bonds versus complicated bonds: how to tell

Because the research now distinguishes helpful from unhelpful expressions of continuing bonds, it can be useful — carefully, non-diagnostically — to recognize the difference. The table below summarizes what clinicians and researchers tend to notice.

This is not a checklist to grade yourself with. Grief is not linear, and any bereaved person may recognize items from both columns at different moments. The point is directional: over time, most people's experience drifts toward the left column. When it does not — when the right column is still describing daily life a year or more after the death, especially with functional impairment — that is a signal that additional support may help.

How continuing bonds show up in ordinary life

One of the reasons the 1996 book resonated so widely is that its examples felt immediately familiar. Continuing bonds are not exotic. They are woven into the small, private rituals of nearly every family that has lost someone.

The everyday textures of connection

• A grown daughter keeps her mother's voicemails on her phone, listening to one on hard days.
• A widower wears his wife's wedding ring on a chain around his neck.
• A son cooks his father's Bolognese every Sunday, using the same wooden spoon.
• A grieving mother writes birthday cards to her son each year and keeps them in a drawer.
• A friend asks "what would she do?" before making a difficult decision at work.
• A family gathers at the grave each spring to clean the stone and share stories.

Some of these expressions accumulate into physical spaces. Many bereaved families create memory boxes — a curated collection of letters, photographs, clothing scraps, jewelry, or objects that hold the person's presence. These containers give a continuing bond a tangible home. They can be brought out on anniversaries, shared with grandchildren, or simply held on a difficult night.

Ritual as institutionalized continuing bonds

What Western psychology called "pathological" in 1970, most of the world's cultures have always called ancestor practice. Continuing bonds theory has been especially valuable because it names, in clinical language, what many traditions have long done ritually:

• Día de los Muertos in Mexico and Latin America: families build ofrendas with food, photographs, and marigolds, welcoming the dead home for a night each November.
• Obon in Japan: a summer Buddhist festival during which ancestral spirits are invited back to the family home; lanterns guide them.
• Jesa in Korea: annual memorial rites for parents and grandparents, with specific foods offered on a carefully arranged table.
• Yahrzeit in Judaism: the yearly anniversary of a death, marked by lighting a memorial candle and reciting Kaddish.
• Qingming in China: the tomb-sweeping festival, when families visit graves, offer food and paper offerings, and update their ancestors on family news.
• All Souls' Day in Catholic tradition: a communal remembrance of the faithful departed, with cemetery visits and prayer.

These are not exotic. They are what humans have almost always done. One of the quieter contributions of continuing bonds theory is the recognition that Western late-twentieth-century "closure" culture was the anomaly, not the norm. When Klass, Silverman, and Nickman wrote in 1996, they were, in effect, restoring the field's memory of what most of the human species had always known: the dead stay with us, and that is not a problem to be solved.

Practical prompts for nurturing a healthy bond

These suggestions are not a program. Grief is not something to optimize. But bereaved people often ask what they can do, and the continuing bonds literature offers gentle, evidence-informed possibilities. Try what draws you; ignore what does not.

Write

• Journal to them, not just about them. Address entries as letters — "Dear Mom" — and write what you would say if she could hear you. Some bereaved people do this daily for a season, then set it down.
• Write on the anniversary. The first and subsequent anniversaries of a death often bring a distinct wave of grief. Some families use the day to write a letter, share a meal the person loved, or read aloud from old cards.
• Write a legacy letter inspired by them. A legacy letter, or ethical will, records what a person hoped, believed, or wanted to pass on. Writing one inspired by the values of a loved one who has died is one way of internalizing the bond — making their legacy an active influence rather than a static memory.

Share

• Tell children and grandchildren specific stories: not "she was wonderful" but "she used to hum when she was concentrating."
• Teach the next generation something the person loved — a song, a recipe, a card game, a woodworking joint, a way of pruning tomatoes.
• Say the person's name out loud in ordinary conversation. Many bereaved people report that friends stop saying the name for fear of causing pain; hearing it, in fact, is often a relief.

Mark

• Create a small annual ritual: lighting a candle on their birthday, visiting a place they loved on the death anniversary, cooking a favorite meal.
• Assemble a physical or digital collection — a memory box, a photo album, a playlist of their music.
• Consider a charitable act in their name — a donation, a volunteer shift, a scholarship contribution — that connects their values to your ongoing life.

None of these has to be permanent. Some rituals are helpful for a season and then quietly fall away. That, too, is healthy — the bond is meant to transform, not to be preserved in amber.

When it may be time to seek professional support

Most people integrate a loss without formal treatment, even a devastating one. That said, roughly seven to ten percent of bereaved adults develop what clinicians now call prolonged grief disorder (formerly complicated grief), which was added to the DSM-5-TR in 2022 and to the ICD-11 as prolonged grief disorder. It is characterized by persistent, intense yearning; identity disruption; difficulty accepting the death; emotional numbness; and functional impairment that continues at least twelve months after the loss (six months for children).

Screening tools such as the PG-13-R (Prigerson's revised Prolonged Grief instrument, sometimes referenced alongside the earlier PGDI-R) can help clinicians identify when grief has moved into a form that benefits from targeted treatment. The most evidence-based intervention is Complicated Grief Treatment (CGT), developed by Dr. M. Katherine Shear and colleagues at Columbia University's Center for Prolonged Grief. Randomized controlled trials, including work published in JAMA Psychiatry, have found CGT more effective than standard interpersonal psychotherapy for people who meet criteria for the disorder.

Reasons a bereaved person might seek professional support include:

• Yearning or preoccupation that is not softening after a year or more.
• Inability to accept the reality of the death, even intellectually.
• Avoidance of reminders that is shrinking daily life.
• Distressing hallucinations or intrusive images of the death itself.
• Loss of a sense of self, purpose, or future.
• Thoughts of joining the person who died, or of not wanting to live.

A therapist trained in grief-specific approaches — CGT, meaning reconstruction, or acceptance and commitment approaches — can help without asking anyone to "get over it." The goal of good grief therapy is not severing the bond. It is restoring the bond's capacity to comfort rather than wound. If thoughts of suicide are present, call or text 988 in the United States, or contact local emergency services.

A closing note on cultural humility

The continuing bonds framework has been called revolutionary, but in another light it is deeply humble: it acknowledges that mainstream Western psychology got something important wrong for most of a century, and that the correction came in part from listening to grievers and looking outward to other cultures. The bereaved parents of The Compassionate Friends, the widows Silverman interviewed, the adopted children in Nickman's caseload — they were teaching the field what most of the world's mourning traditions already knew.

This matters because grief in the United States is still often lived under old assumptions. Employers offer three bereavement days. Well-meaning friends stop mentioning the person's name after a month. The cultural clock says the mourner should be recognizable again in a season. Continuing bonds theory offers a quiet counterweight: your relationship with the person you loved is allowed to continue. It is allowed to change. It is allowed to comfort you. It is allowed to surprise you decades later at a wedding or a birth. None of that means you are stuck.

The most useful thing this framework offers is not a technique. It is a permission. Permission to keep the voicemail. Permission to speak to them at the grave. Permission to cook the recipe and cry a little and laugh a little and set the pot back on the stove. Permission for grief to become something you carry rather than something you finish.

The relationship changes form. It does not end. And nothing in the research, the clinical literature, or the accumulated wisdom of the world's mourning traditions says it has to.

Sources:
Klass, D., Silverman, P. R., & Nickman, S. L. (Eds.) (1996). Continuing Bonds: New Understandings of Grief. Taylor & Francis. — https://www.taylorfrancis.com/books/edit/10.4324/9781315800790/continuing-bonds-dennis-klass-phyllis-silverman-steven-nickman
Field, N. P., & Filanosky, C. (2010). Continuing bonds, risk factors for complicated grief, and adjustment to bereavement. Death Studies, 34(1), 1–29. — https://www.tandfonline.com/doi/abs/10.1080/07481180903372269
Root, B. L., & Exline, J. J. (2014). The role of continuing bonds in coping with grief: Overview and future directions. Death Studies, 38(1), 1–8. — https://www.tandfonline.com/doi/abs/10.1080/07481187.2012.712608
Neimeyer, R. A. (2001). Meaning reconstruction and the experience of loss. American Psychological Association. — https://psycnet.apa.org/record/2001-16569-000
Neimeyer, R. A., Klass, D., & Dennis, M. R. (2014). A social constructionist account of grief: Loss and the narration of meaning. Death Studies, 38(6-10), 485–498. — https://www.tandfonline.com/doi/abs/10.1080/07481187.2014.913454
Shear, M. K., Reynolds, C. F., Simon, N. M., et al. (2016). Optimizing treatment of complicated grief: A randomized clinical trial. JAMA Psychiatry, 73(7), 685–694. — https://jamanetwork.com/journals/jamapsychiatry/fullarticle/2526304
Prigerson, H. G., Boelen, P. A., Xu, J., Smith, K. V., & Maciejewski, P. K. (2021). Validation of the new DSM-5-TR criteria for prolonged grief disorder and the PG-13-Revised scale. World Psychiatry, 20(1), 96–106. — https://onlinelibrary.wiley.com/doi/10.1002/wps.20823
American Psychological Association — Grief resources and public education. — https://www.apa.org/topics/grief
Columbia University Center for Prolonged Grief — Overview of complicated grief and CGT. — https://prolongedgrief.columbia.edu/
Freud, S. (1917). Mourning and melancholia. Standard Edition, 14, 237–258. — https://www.pep-web.org/document.php?id=se.014.0237a
Bowlby, J. (1980). Attachment and Loss, Vol. 3: Loss, Sadness and Depression. Basic Books. — https://www.basicbooks.com/titles/john-bowlby/loss-sadness-and-depression/9780465042371/
World Health Organization — ICD-11 Prolonged Grief Disorder (6B42). — https://icd.who.int/browse11/l-m/en#/http%3a%2f%2fid.who.int%2ficd%2fentity%2f1183832314
American Psychiatric Association — Prolonged Grief Disorder added to DSM-5-TR (2022). — https://www.psychiatry.org/patients-families/prolonged-grief-disorder
Steffen, E. M., & Coyle, A. (2011). Sense of presence experiences in bereavement and their relationship to mental health. Mental Health, Religion & Culture, 14(6), 579–609. — https://www.tandfonline.com/doi/abs/10.1080/13674676.2010.487229