The Body Keeps the Score, But Story Helps It Let Go

A woman in her forties couldn't explain why the sound of a specific kind of argument — not shouting, just a particular low, tense exchange between two people in the same room — would cause her chest to tighten, her breath to shallow, her body to flood with a readiness for something that wasn't happening. Intellectually, she knew she was safe. Her body didn't know that.

It took her years, and eventually a therapist, to connect that response to a childhood spent listening through walls to her parents' late-night conflicts. The memory had never been lost. What had been stored, in a way that no act of remembering could quite touch, was the physical state the child had lived in: hypervigilant, braced, small. The body had kept the score — and it was still tallying.

This is the central finding that psychiatrist and trauma researcher Bessel van der Kolk synthesized in a lifetime of clinical work: trauma doesn't live primarily in conscious narrative memory. It lives in the body's regulatory systems. And understanding that changes everything about how healing actually works.

What the body holds#

For most of the twentieth century, the dominant model of trauma treatment was cognitive and narrative: help people understand what happened, process it in words, revise the meaning they had made of it. This model worked for some people and for some kinds of trauma. For many others, something stubbornly remained — not in thought, but below thought.

What the body holds from traumatic experience is, in the language of neuroscience, an encoded stress response. The amygdala — the brain's threat-detection system — logs the sensory signature of danger without consulting the cortex. The nervous system mobilizes to survive. Heart rate, respiration, muscle tension, and hormonal release all shift in coordinated ways designed to maximize the chance of getting through whatever is happening.

The problem is that this system doesn't automatically reset when the danger passes. In acute, discrete traumatic events, it sometimes does — the stress cycle completes, the activation discharges, the body returns to baseline. But in chronic, relational, or early-life trauma, the cycle frequently doesn't complete. The body remains in a state of modified alertness, scanning for danger that may not be there, responding to echoes of past threat as if they were present ones.

This is why a grown woman's chest tightens at the sound of a particular tonal exchange. The body is not confused. It is doing exactly what it learned to do — it just learned in a context that no longer applies.

Why memory doesn't always arrive in words#

Van der Kolk's foundational contribution was demonstrating that trauma memory is not stored like ordinary autobiographical memory. Ordinary memory is reconstructive: each time we remember something, we rebuild it, adding context, revising interpretation, incorporating what we now know that we didn't know then. This is why our stories about ourselves change over time in healthy development.

Traumatic memory is often encoded differently — as sensory fragments, as emotional states, as physical responses — without the narrative context that lets ordinary memory be revised. The body "remembers" in a way that bypasses the language centers of the brain entirely. This is why people can have detailed intellectual knowledge of a traumatic history and still find themselves flooded by physical responses with no apparent narrative trigger.

It also explains why purely verbal, cognitive approaches sometimes reach a ceiling. The memory isn't primarily where the words are.

The somatic pathway: completing the cycle#

Peter Levine, the developer of Somatic Experiencing, approached trauma from a biological angle that began with an observation about animals in the wild. Animals are regularly exposed to life-threatening situations — prey animals especially — but they don't develop chronic post-traumatic stress the way humans do. Levine noticed that wild animals, when they survive a threat, complete a physical discharge process: shaking, trembling, a full-body release that appears to complete the interrupted stress cycle and return the animal to regulation.

His insight was that humans, with our more complex nervous systems and our cultural conditioning against uncontrolled physical expression, often interrupt this completion. We hold the activation in. We manage it, suppress it, contain it — and in doing so, leave the body in a state of unresolved mobilization that can persist for years or decades.

Somatic Experiencing works directly with this incomplete cycle, using guided attention to bodily sensation to help the nervous system finish what it started. The focus is not on the narrative of what happened but on what the body is doing right now: the tension in the jaw, the constriction in the throat, the impulse in the legs that wants to run. By attending to these directly, in carefully titrated amounts, the body can complete the interrupted response and return toward baseline.

This is bottom-up processing: starting in the body and moving toward the mind.

The window of tolerance#

Stephen Porges' Polyvagal Theory gave trauma researchers and clinicians a more precise neurological map of why all of this matters for any kind of processing — including narrative processing.

The polyvagal framework describes three basic states of the nervous system. The first is a state of safe social engagement — relaxed, curious, able to perceive nuance and complexity, capable of connection. The second is sympathetic activation — mobilized for fight or flight, hypervigilant, attentive primarily to threat. The third is a dorsal vagal shutdown — the freeze or collapse state, characterized by numbness, disconnection, and a kind of profound flatness.

Meaningful narrative work — making sense of what happened, revising meaning, integrating experience — can only occur in the first state. The cortex, which handles language and meaning-making, is effectively taken offline in the other two. When the nervous system is dysregulated, a person cannot process a story; they can only re-experience it or dissociate from it.

This is what clinicians call the window of tolerance: the zone of nervous system activation in which a person has enough access to reflective function to actually work with difficult material. Too much activation, and the story becomes a trigger. Too little (shutdown), and the person can't access the material at all.

The implication for any narrative approach to trauma is significant: the nervous system must be regulated enough to engage with story before story can do its integrating work. Somatic work — whether formal therapy or the informal but real regulation that comes from safe relational contact — is often what makes the window accessible.

How they work together#

The most effective contemporary trauma approaches tend to integrate both directions. EMDR (Eye Movement Desensitization and Reprocessing) combines bilateral sensory stimulation with narrative processing. Narrative Exposure Therapy uses careful, structured telling of traumatic experience while maintaining grounding in present-moment body state. Trauma-informed yoga and movement practices reduce activation levels that make narrative processing possible.

The insight common to all of them is that neither somatic work nor narrative work is sufficient on its own. Somatic work can release activation, restore regulation, and help the body complete what it was prevented from completing — but without a language and a story, the experience remains fragmented, unintegrated, hard to carry forward or pass on. Narrative work can build meaning and identity and coherence — but without a regulated nervous system, the story gets told from inside the trauma rather than from a safe distance from it.

Together, they address what trauma actually is: an experience that is held in both the body and the mind, simultaneously, and that requires both to be genuinely integrated.

What this means for family memory#

For families, this has implications that extend well beyond clinical trauma work.

When people tell stories together — not in therapy, but around a kitchen table, or sitting with a grandparent, or going through old photographs — their nervous systems co-regulate. We are wired, through the polyvagal pathways of the social engagement system, to attune to the people around us: to read their tone, their pace, the quality of their presence. In safe relational contact, this attunement is itself regulating. It shifts us toward the state in which memory can be integrated rather than re-experienced.

This is one reason why collective family storytelling has protective effects that go beyond the informational content of what's shared. The act of remembering together, in a context of genuine safety and connection, is a somatic practice as much as a cognitive one. It exercises exactly the capacity — holding difficult experience while remaining regulated — that Levine and Porges and van der Kolk have shown is at the heart of healing.

The body keeps the score. But the score is not a fixed accounting. The body is also the site of release, completion, and return to regulation — and narrative, when it meets the body in the right conditions, is one of the most powerful means we have for helping that happen.