The Body's Unfinished Business: Trauma, Repetition, and the Physiology of Domestic Violence

Introduction

The relationship between early childhood trauma and adult involvement in domestic violence — whether as the person harmed, the person causing harm, or both — is one of the most consistently replicated findings in the trauma literature. Felitti et al.'s landmark Adverse Childhood Experiences (ACE) study (1998), drawing on a cohort of over 17,000 adults, established a dose-response relationship between the number of adverse childhood experiences and a wide range of adverse adult outcomes, including intimate partner violence. Adults who had experienced four or more categories of childhood adversity were significantly more likely to have experienced or perpetrated domestic violence than those with no ACE history. This finding has since been replicated across multiple populations and methodologies (Norman et al., 2012; Ports et al., 2016).

What the epidemiological literature establishes clearly is the association. What it has been slower to account for is the mechanism — the precise physiological pathway by which early adverse experience shapes the nervous system in ways that manifest in the specific patterns of adult intimate partner violence that practitioners and researchers consistently observe but find difficult to explain within conventional frameworks of choice, agency, and motivation.

That mechanistic account has been developed, in its constituent parts, across several distinct bodies of research. Van der Kolk's neurobiological investigations into trauma and the body (1989, 1994, 2014) demonstrated that traumatic experience is encoded not primarily as declarative memory but as somatic and procedural memory — held in the body's systems of threat response and activation, running below and often independent of conscious awareness. Porges' polyvagal theory (1994, 2011) provided the autonomic architecture: a hierarchical nervous system in which early and sustained experience of threat shapes the baseline functioning of the vagal system in ways that affect every subsequent encounter with stress, intimacy, and perceived danger. Levine's somatic experiencing framework (1997) introduced the concept of the incomplete threat cycle — the observation that the body's mobilisation in response to threat requires completion, and that when completion is blocked or disrupted, the nervous system continues to seek it, sometimes across decades and across relationships. Schore's affect regulation research (1994, 2003) established the developmental conditions under which internal regulatory capacity fails to form — leaving the nervous system dependent on external means, including the regulatory function of an attachment figure, to manage states of high arousal.

Each of these bodies of work illuminates part of the picture. The contribution this essay attempts is their integration — specifically, the application of that integrated framework to the patterns in domestic violence that have most resisted explanation: the escalation of the person harmed towards further harm; the failure of distance and separation to interrupt the cycle; the selection of successive partners who reproduce the dynamic; and the function such partners serve beyond the cycle itself. It further addresses, drawing on Stoller's account of the eroticisation of early trauma (1975) and Dutton and Painter's traumatic bonding research (1981, 1993), the conditions under which the threat cycle becomes fused with arousal — a dimension of the literature that clinical observation has long recognised and that mainstream DV discourse has been slow to name.

---

1. The Neurobiological Foundation: How Early Threat Shapes the Nervous System

The human nervous system does not develop in isolation from the environment in which it matures. Current understanding of neurodevelopment, drawing substantially on the work of Perry (1995, 2006), Schore (1994, 2003), and Siegel (1999, 2012), establishes that the architecture of the developing brain is shaped by experience in ways that are not merely psychological but structural and functional. The brain regions most directly involved in threat detection, stress response, and emotional regulation — the amygdala, the hippocampus, the prefrontal cortex, and the structures of the autonomic nervous system — develop within a relational and environmental context, and their development is sensitive to the nature of that context.

Perry's neurosequential model (1995) documents the sequence in which brain structures develop, and the implications for those that develop under conditions of chronic threat and dysregulation. The survival brain — the brainstem and its associated structures — is the first to mature, and under chronic threat conditions it develops with threat-detection biased towards hyperactivation: faster to fire, more easily triggered, slower to return to baseline. The limbic system, which develops subsequently and which carries the emotional regulatory functions that mediate between threat response and considered action, is shaped by the quality of co-regulation available from early caregivers. Where that co-regulation is itself disrupted — because the caregiver is the source of threat, or is so compromised by their own history that they cannot provide the regulatory contact the child's developing nervous system requires — the limbic structures develop without the experiential scaffolding they need to function reliably under stress. This three-layer architecture — survival brain, feeling brain, and thinking brain — and the way chronic early threat distorts the hierarchy between them is explored in more accessible register in the YoungFamilyLife essay Learning to Survive: How the Human Brain Navigates Opportunity and Danger, which provides the foundational account of threat response patterns used throughout the YFL platform.

Porges' polyvagal theory (1994, 2011) provides the autonomic framework within which this development can be understood. The theory identifies three hierarchical states of the autonomic nervous system, each associated with a distinct evolutionary layer of neural circuitry. The most recently evolved, the ventral vagal complex, supports social engagement — the capacity for calm, connected interaction with others. Below this, the sympathetic nervous system supports fight-or-flight mobilisation in response to perceived threat. At the base, the dorsal vagal complex supports freeze and collapse responses associated with inescapable danger. Under normal developmental conditions, the ventral vagal system is the default operating mode, with the sympathetic and dorsal vagal systems activated only in genuine threat conditions and returning to baseline once the threat has passed. Under conditions of chronic early threat, however, the hierarchy is disrupted: the nervous system is conditioned to operate at or near sympathetic activation as its baseline, with the ventral vagal state experienced not as normal but as abnormal — a temporary and not entirely trusted departure from the body's learned sense of how the world is.

Schore's affect regulation theory (1994, 2003) extends this account to the relational dimension. Schore's research demonstrates that the right hemisphere — which carries the primary functions of emotional processing and autoregulatory capacity — develops in the context of early caregiver attunement. The infant's developing nervous system literally borrows regulatory capacity from the caregiver's more mature system, using the co-regulatory experience to build internal structures capable of managing arousal independently. Where early caregiving is consistently frightening, absent, or emotionally unavailable, this scaffolding is compromised. The child's capacity for internal regulation — the ability to move from high arousal back to settled functioning without external input — develops inadequately. The consequence is a nervous system that, under stress, lacks the internal resources to self-regulate and must seek regulation externally. In adulthood, this dependency on external regulation for managing high arousal states has profound implications for intimate relationships, as will be discussed. The practical dimension of this — how the body reads relational safety before the conscious mind catches up — is the subject of Kate Cairns' work, explored in accessible form in the YFL IOW piece The Body Knows Safety Before the Mind Does.

---

2. The Incomplete Threat Cycle

Central to the argument of this essay is Peter Levine's concept of the incomplete threat cycle, developed in his somatic experiencing framework and most fully articulated in Waking the Tiger (1997) and In an Unspoken Voice (2010).

Levine's observation — grounded in ethological research as well as clinical practice — is that mammals exposed to threatening situations undergo a predictable physiological sequence. Threat is detected; the nervous system mobilises, flooding the body with the neurochemical resources appropriate to fighting or fleeing; and the threat is either resolved through action or, if action is impossible, the system shifts into freeze or collapse as a survival strategy. Critically, Levine argues that this mobilisation is not self-resolving. It requires a completion — a physical discharge of the energy mobilised — in order for the nervous system to return to its baseline. In animals, this completion typically occurs: the predator is outrun, the threat is repelled, or the freeze response completes and the animal shakes and trembles as the mobilised energy discharges. In humans, particularly in conditions of complex and relational trauma, this completion is frequently blocked — by the social context, by the impossibility of action, by the ongoing nature of the threat that allows neither resolution nor escape. The mobilised energy remains in the body, held in states of chronic tension and readiness, and the nervous system remains, functionally, in a threat cycle that has not completed.

Van der Kolk's neurobiological research (1989, 1994, 2014) provides the evidence base for the somatic reality of this incomplete state. His investigations of trauma and the body demonstrated that people with trauma histories show measurable differences in the functioning of the amygdala, the hippocampus, and the prefrontal cortex — differences that affect both the threshold at which threat responses are activated and the capacity to modulate those responses once activated. Crucially, van der Kolk's work showed that traumatic memory is stored differently from ordinary declarative memory: not as a narrative that can be recalled, examined, and contextualised, but as somatic and sensory fragments — physical states, body feelings, and autonomic activations that can be triggered by stimuli reminiscent of the original threat, often without any conscious recognition of the connection. The IOW companion to this essay, How a Brain Builds Itself — and What That Has to Do With Relationships, addresses how these somatic encodings develop and why they outlast the circumstances that created them.

The implications for domestic violence are significant. A person who grew up in a home where violence or serious threat was a regular feature of evening hours does not merely carry memories of those evenings. They carry the physiological programme the nervous system built around them — including the programme of what completion felt like. The tension of the build-up, the peak of the threat, and the exhausted discharge afterwards are encoded together as a cycle, not as separate events. The settlement that followed the worst of the night is not stored as something that came after the danger. It is stored as the endpoint of the danger — the physiological moment when the body finally received permission to rest. As an adult, when the nervous system moves into the threat activation it learned in that home, it is not merely building towards fear. It is building towards the only completion it has ever known.

This is the mechanism that underlies the clinical observation — puzzling from outside, but coherent once the framework is applied — of a person in a DV relationship who appears to escalate towards violence rather than away from it. The escalation is not self-destructive in intent. It is, from the nervous system's perspective, the completion of an unfinished cycle.

---

3. Repetition Compulsion: From Freud to Neurological Restatement

The clinical phenomenon of trauma repetition — the pattern by which people appear compelled to reproduce, in adult life, the traumatic experiences of their earlier years — has a long theoretical history. Freud named it the Wiederholungszwang, the compulsion to repeat, in Beyond the Pleasure Principle (1920), where he observed that patients in analysis appeared to re-enact traumatic experiences rather than simply remembering them, and that this re-enactment seemed to operate against the pleasure principle he had previously taken as primary. He proposed the concept as evidence for a death drive — a speculative formulation that later theorists have largely set aside in favour of more tractable explanations.

The most clinically and empirically productive reframing of repetition compulsion is van der Kolk's (1989), in his paper The Compulsion to Repeat the Trauma. Van der Kolk repositions repetition not as a drive towards self-destruction but as the nervous system's attempt to master an experience that it could not process at the time of its occurrence. The re-enactment, in this reading, is not a failure of self-protection. It is an attempt at completion — the nervous system seeking, in each new iteration, the resolution it could not find the first time. The tragedy of the pattern is that re-enactment rarely produces resolution. The cycle repeats not because each repetition brings the completion closer, but because the body knows no other route to it.

Object relations theory offers a complementary account at the relational level. Fairbairn's (1952) concept of the libidinal ego — the part of the self that remains fixated on the unmet needs of early experience and continues to seek their resolution in adult relationships — and Guntrip's (1969) extension of this work both address the way in which early relational templates are not simply remembered but lived forward. The adult does not consciously seek to recreate their early experience. They seek what their deepest relational programming recognises as the possibility of need-meeting — and that programming was calibrated in conditions that may bear very little resemblance to what an outside observer would identify as safe or loving.

Winnicott's distinction between the true self and the false self (1960) is also relevant here. The false self — the adaptive, socially presentable self that develops in response to caregiving that was conditional, threatening, or inadequate — is the self that most partners initially encounter. The true self, in Winnicott's formulation, is the more primitive and authentic self that emerges in conditions of genuine intimacy and domestic regularity — conditions that, as Scharff and Scharff (1991) and Johnson (2004) have noted in the context of couple relationships, can activate relational templates that were entirely suppressed during the early period of a relationship. The partner who appeared, in the pub or at the gym, to be confident and easy-going is not performing. They are genuinely operating from a regulated, ventral-vagal state in which the older patterns are held in check. The domestic context — with its familiarity, its lowered guard, its resemblance to the conditions in which the original patterns were formed — activates what was always there. This mechanism — the gap between the self presented in optimal relational conditions and the survival-brain self that emerges in sustained domestic intimacy — is the subject of the YFL Repositorium essay Authentic and Inauthentic Behaviour: What Childhood Wires In, and What Adulthood Inherits. Berne's transactional analysis framework, which maps the ego state architecture underlying these shifts, is explored in the YFL essay Eric Berne's Transactional Analysis: From Freudian Theory to Observable Interaction, and its specific application to child protection contexts in Transactional Analysis and Child Protection. The HWTK piece Why Rational Adults Can Respond With Childish Venom addresses the same mechanism from the lived experience angle — what it looks and feels like when the Child ego state takes over in an adult.

---

4. When the Body Finds Its Own Endings: Eroticisation, Risk, and the Completion Cycle

The threat cycle described in the preceding sections does not seek its completion exclusively through intimate partner violence. For many people whose nervous systems were shaped by early and chronic threat, the body finds routes to the neurochemical endpoint of the cycle across a wide range of behaviours — some within relationships, some entirely outside them. Understanding this breadth is essential to understanding the full clinical picture, both in DV contexts and in the wider treatment of trauma.

Before proceeding, a necessary qualification must be stated clearly. The behaviours discussed in this section — self-harm, addiction, risk-taking, extreme physical exercise, body modification, eating disorders — are each, in themselves, complex phenomena with multiple aetiologies. Many people engage in extreme exercise for reasons entirely unrelated to trauma history. Body modification and tattooing carry cultural, aesthetic, and identity dimensions that have nothing to do with the completion cycle described here. Risky behaviour, including criminal activity, has its own sociological and developmental literature. This section does not offer a unified trauma explanation for any of these behaviours, and the framework presented here should not be used in isolation to pathologise, diagnose, or interpret any individual's conduct. What it offers is one possible lens — physiologically grounded and clinically observed — for understanding a subset of presentations in which these behaviours appear to be driven, at least in part, by the nervous system's search for completion.

The Neurochemical Foundation

The common thread across the behaviours discussed here is neurochemical. The body's endogenous opioid system — activated by physical pain, by extreme physical exertion, and by the relief that follows high arousal — produces endorphin release that provides genuine physiological settling. This is not metaphorical. It is measurable. Sprouse-Blum et al. (2010) document the endorphin response to physical pain and exercise; van der Kolk (2014) extends this to the specific context of self-harm and trauma, noting that many people who self-harm report the same physiological relief — a settling, a discharge, a return from unbearable tension to something manageable — that Levine describes as the completion of the threat cycle. The cortisol-to-calm arc of high-risk behaviour; the dopaminergic anticipation-completion cycle of addiction; the adrenaline peak and subsequent crash of dangerous activity — each of these represents a different biochemical route to the same physiological destination: the endpoint the nervous system learned to associate with safety.

The broader role of pain as physiological signal and regulator is explored in the YFL essay Pain as Evolutionary Communication, which provides supporting context for understanding why the body's pain-relief sequence carries such neurological weight in the formation of both attachment and behavioural patterns under adverse developmental conditions.

Eroticisation: The Most Fused Expression

Among the routes to completion, the eroticisation of the threat cycle represents the most fully fused — the condition in which the original pattern of threat, arousal, and relief has become so thoroughly integrated with sexual response that the two can no longer be cleanly separated. Robert Stoller's Perversion: The Erotic Form of Hatred (1975) provides the foundational theoretical account. Stoller's central argument is that sadomasochistic sexual arousal patterns are frequently the adult nervous system's attempt to master, through eroticisation, what was originally experienced as overwhelming and uncontrollable. The childhood experience of threat, pain, or humiliation — which the child's nervous system was unable to process — is transformed, through what Stoller describes as the conversion of trauma to triumph, into something over which the adult now exercises apparent agency. The overwhelming affect of the original experience is channelled into arousal; the original passivity is replaced, in fantasy or in practice, by orchestration. The pleasure, such as it is, is not in the pain itself but in the illusion of mastery over what was once inescapable.

The cortisol surge of high arousal, followed by the oxytocin release associated with physical contact in the aftermath of a violent episode, produces a neurochemical sequence that can, under specific developmental conditions, become associated with attachment and relief. Dutton and Painter's traumatic bonding research (1981, 1993) describes the intermittent reinforcement dynamic that consolidates this association: the alternation of threat and relief, harm and comfort, activates the same neurological reward pathways as other forms of intermittent reinforcement, producing an attachment that is, in neurological terms, indistinguishable from the bonds formed in less harmful relationships. The HWTK piece Why Caring Parents Get Short Tempered With Their Children illustrates the same survival-brain override in a domestic parenting context — a less extreme but structurally identical instance of the threat response displacing considered behaviour.

Other Routes to Completion

Self-harm is one of the most directly documented expressions of the completion cycle outside of intimate relationships. Van der Kolk (2014) describes self-harm in trauma survivors as frequently serving a regulatory function — the physical pain triggering the endorphin response that provides genuine, if temporary, relief from the chronic tension of the incomplete cycle. Many people who self-harm report that the act itself is less significant than the settling that follows it: not the wound, but the quiet afterwards. Understanding this does not diminish the seriousness of self-harm or suggest it is an acceptable coping strategy. It explains the physiological logic that makes it, for some nervous systems in some circumstances, effective in the immediate term — and therefore resistant to simple prohibition or cognitive redirection.

Addiction — whether to substances, alcohol, or behaviours — addresses the same underlying dysregulation through different biochemical mechanisms. Where the nervous system lacks the internal regulatory capacity to move from high arousal to settled functioning, substances provide that transition externally. Alcohol sedates the build-up; stimulants accelerate the arousal phase; opioids provide the relief endpoint directly. The addiction is not simply to the substance. It is, in many trauma-informed accounts (Maté, 2008; van der Kolk, 2014), to the regulation the substance provides — the only reliable route the nervous system has found to the settled state it cannot reach on its own.

High-risk behaviour, including in some cases criminal activity, provides adrenaline mobilisation that mimics the original threat activation, with the outcome — escape, success, arrest, consequence — functioning as a completion. The physiological logic is the same as that which drives escalation in the DV relationship: the body needs the peak in order to access the descent. The specific form the behaviour takes — whether within a relationship or outside it — is shaped by availability, circumstance, and individual history, not by the underlying mechanism, which is consistent across all expressions.

Extreme physical exercise occupies a more complex position. For many people, vigorous exercise is a genuinely adaptive route to the neurochemical settling the nervous system needs — one with substantial health benefits and no inherent harm. But for a subset of people, exercise becomes compulsive and driven in ways that mirror other completion-seeking behaviours: not chosen because it is enjoyable or health-promoting, but pursued with an urgency that suggests the nervous system is using it to manage something it cannot manage by other means. The distinction between adaptive use of exercise as a regulatory tool and driven completion-seeking through exercise is not always visible from the outside, and should not be assumed from the behaviour alone.

Body modification and tattooing — for some, though again not all — provide a context of controlled, chosen, and bounded pain with a certain endpoint. The pain is self-selected; its duration is known; the relief that follows is guaranteed. For a nervous system that grew up with unpredictable and uncontrollable pain, this inversion — pain on one's own terms — can carry a genuine and coherent significance. Tattooing in particular has been noted in clinical literature as serving, for some trauma survivors, a function of reclaiming the body as one's own territory (van der Kolk, 2014). This is not a universal or even common explanation for body modification; it is one possible meaning among many.

Eating disorders represent perhaps the most complex expression in this group, carrying their own substantial and distinct literature. The restriction, purging, and bingeing cycles of anorexia and bulimia each have physiological components consistent with the completion-cycle framework: restriction produces a physiological state of alertness and control that can function as a proxy for the threat-activation phase; purging provides a physical discharge with its own neurochemical arc; bingeing can function as self-medication of the build-up. Treasure et al. (2010) and Schmidt and Treasure (2006) address the regulatory functions of eating disorder behaviour in terms broadly consistent with the affect dysregulation model described here, though the eating disorders literature is appropriately cautious about single-mechanism accounts of conditions with complex multi-factorial aetiologies.

The Clinical and Ethical Frame

Across all of these expressions, the same ethical and clinical caution applies as was noted in relation to the eroticisation dimension specifically. None of these behaviours is diagnostic of trauma history. None should be used in isolation to construct a clinical formulation. The framework offered here is one lens among many — appropriate for considering alongside other evidence, not for substituting the careful individual assessment that any complex presentation requires.

What the framework does offer is a coherent physiological account of why these behaviours can be so resistant to cognitive and behavioural intervention alone: because the nervous system is not making a logical error that better information will correct. It is following the only map it has to the only destination it knows. Replacing the behaviour requires — eventually, carefully, with sustained support — helping the nervous system find a different map to a different destination. That is the work of trauma-informed, body-level therapeutic intervention, as discussed in Section 8.

---

5. Partner Selection: The Nervous System Finding What It Recognises

The pattern of successive relationships reproducing the same dynamic — observed clinically and documented in the research literature on revictimisation (Messman-Moore & Long, 2003; Classen et al., 2005) — is not adequately explained by accounts that invoke poor judgement, low self-esteem, or failure to recognise warning signs. Such accounts locate the explanation at the level of conscious appraisal, which is precisely the level at which it does not primarily operate.

The nervous system that was shaped by a particular kind of early relational environment develops a pattern-recognition system calibrated to that environment. This is not a pathological feature. It is adaptive: the child who learned to read minute changes in a caregiver's affect, who could detect the rising tension before any overt sign of threat appeared, who survived by anticipating the unpredictable — that child's pattern-recognition system has been finely tuned by years of practice. The difficulty is that this same system continues to operate in adult relational contexts, reading potential partners for the qualities it has learned to recognise. A particular quality of intensity, a specific kind of emotional charge, a pattern of relating that feels, below the level of conscious appraisal, familiar — these register not as warning signs but as recognition. And recognition, to a nervous system calibrated in the conditions described above, is experienced as safety: not because the situation is safe, but because the body knows this territory and has a programme for navigating it.

Bowlby's attachment theory (1969, 1973, 1980) provides the relational framework within which this selection operates. Bowlby's internal working models — the representational schemas of self, other, and the expected dynamic between them, built from early attachment experience — function as templates against which new relational possibilities are evaluated. Where the early attachment environment was organised around threat and intermittent care, the internal working model encodes that pattern as the expected form of relationship. New relationships that conform to the template feel familiar; those that depart significantly from it may feel uncomfortable, boring, or unreal, regardless of their objective safety or quality.

Hazan and Shaver's application of attachment theory to adult romantic relationships (1987) extended this understanding to partner selection specifically, demonstrating that adults' attachment styles — themselves products of early attachment experience — predict patterns of partner choice and relationship dynamics in ways consistent with the theoretical framework. The adult with a disorganised or fearful attachment style — the style most commonly associated with early experiences of threat from caregivers — shows specific patterns in partner selection that reflect the fundamental disorganisation of their attachment system: drawn towards intensity and familiarity, uncertain of safety, unable to consistently distinguish between the felt sense of attachment and the reality of threat. The YFL IOW piece Everyone Has an Attachment Style — and It Started as a Survival Skill offers an accessible account of how these adult attachment styles develop from early relational experience and what they look like in adult relationships. The foundational YFL treatment of Bowlby's contribution to understanding family relationships is available in A Conversation with Bowlby: Attachment Theory and Family Life.

---

6. The Protection Calculation: The Territorial Function of the Known Danger

A dimension of domestic violence that is rarely addressed in the literature, but which clinical observation consistently surfaces, is the function the partner serves beyond the relational and cycle-completion needs already described. For a person who grew up in an environment in which threat was not confined to one person but was a general feature of the world — in which predation, in various forms, was a possibility that required constant vigilance — the absence of a partner does not simply remove the threat. It removes a form of protection.

This observation does not appear explicitly in the major theoretical frameworks, though it is consistent with their logic. A partner who is controlling, intimidating, or capable of violence occupies what might be described as territorial space in the person's social environment. Their visible presence — the fact of the relationship, the signal it communicates to others — functions as a deterrent to other potential sources of threat. The nervous system's calculation, operating below conscious awareness and drawing on the logic of the environment in which it was formed, identifies this deterrent function as a form of safety. Not safety from the partner — the partner is a source of harm — but safety from the broader threat landscape in which being unpartnered and unprotected is experienced as more dangerous than being harmed within a known relationship. The known and navigable danger is preferred to the unpredictable and uncontained danger of the unguarded position.

This calculation has specific implications for intervention. Removing the partner addresses the immediate harm and removes the immediate source of the cycle. But it does not remove the nervous system's experience of the protective function the partner served, or its assessment of the relative risks of the unpartnered state. Without an adequate replacement for that protective function — whether through genuine environmental safety, through social support structures, or through the slow therapeutic work of reorienting the nervous system's threat assessment — the person is left in a state that feels, physiologically, more exposed rather than less. The sanctuary, in this framework, addresses one dimension of what the relationship was providing. It does not address the other.

---

7. The Child in the System: Intergenerational Transmission

The child who grows up in a home where domestic violence is present does not simply witness harm. They develop within an environment in which the nervous system's threat architecture is being shaped in real time — an environment in which the climate of danger, the physiological states of the adults around them, and the quality of the caregiving available to them are all affected by the violence and by what drives it.

Holt, Buckley and Whelan's systematic review (2008) of research into children's exposure to domestic violence found consistent evidence of elevated rates of emotional and behavioural difficulties, including anxiety, depression, aggression, and post-traumatic symptoms, across all age groups. Critically, the review found that the effects were not limited to children who had directly witnessed violent incidents: children who lived in households where violence occurred showed elevated symptomatology even where they reported no direct witnessing, consistent with the understanding that it is the climate of the home, not the specific incidents, that constitutes the primary developmental environment.

The intergenerational dimension — the pathway by which children exposed to domestic violence are at elevated risk of involvement in DV relationships in adulthood — is documented in the meta-analytic literature. Stith et al.'s meta-analysis (2000) found witnessing parental violence to be one of the more consistent predictors of perpetrating or experiencing intimate partner violence in adulthood, though effect sizes were modest, consistent with the understanding that exposure is a risk factor operating within a complex system of mediating and protective variables rather than a direct determinant. The child's experience within a DV home — what the nervous system absorbs from the climate, rather than from specific witnessed incidents, and the particular bind of needing the source of fear — is explored in accessible form in the companion YFL IOW piece What Children Carry: Growing Up in a Home Where Violence Is Present.

The mechanism proposed in this essay provides a more specific account of one pathway through which that intergenerational transmission may operate. The child who grew up with the threat cycle as the background of their developmental environment does not need to have been directly harmed to develop a nervous system calibrated to that cycle. The stress physiology of a chronically frightened or dysregulated caregiver is transmitted, through the co-regulatory relationship, to the child's developing nervous system. The child's attachment system is formed in relation to adults whose own attachment and regulatory systems are operating under the conditions described above. The internal working models built in this environment encode the patterns of threat, intensity, and intermittent relief that the environment provided. And those internal working models become, in adulthood, the templates against which intimate relationships are evaluated and selected.

Critically, this transmission begins before birth. Research into prenatal stress — most substantially developed by Glover (2011, 2014) in her work on foetal programming and maternal stress — has established that the foetus is sensitive to the mother's physiological state from the second trimester onwards. Maternal cortisol and stress hormones cross the placental barrier, and the foetal hypothalamic-pituitary-adrenal (HPA) axis — the stress response system that will govern threat reactivity throughout the child's life — is being calibrated in real time by the mother's own stress physiology. A mother whose nervous system is running the patterns described in this essay — chronically threat-activated, physiologically primed for the cycle's build and peak — is transmitting the hormonal and physiological signature of that state to the developing nervous system of the child she is carrying. The foetus registers the sustained biochemical environment of maternal stress: the cortisol surges, the adrenaline, the autonomic activation of a nervous system that has never fully settled. The child's nervous system begins its work of deciding what kind of world it is being born into before birth — already orienting towards vigilance, already calibrating for a high-threat environment, on the basis of evidence gathered in utero. The intergenerational transmission of the threat cycle, in this light, begins not at birth but at conception — or rather, at whatever point in gestation the foetal stress response system begins to develop and receive its first inputs from the maternal environment.

Bowlby's (1980) concept of the transmission of attachment patterns across generations — extended and evidenced by Main and Hesse (1990) in their research on disorganised attachment — describes the relational mechanism by which this occurs. Main and Hesse's finding that parental unresolved trauma predicted disorganised attachment in infants — mediated by the parent's frightened or frightening caregiving behaviour — provides the empirical basis for understanding how the physiological inheritance described in this essay passes from one generation to the next, not through genetics alone but through the lived relational environment.

---

8. Intervention: What the Evidence Supports

The framework developed in this essay has significant implications for the design of effective intervention in domestic violence. If the patterns described — the completion cycle, the eroticisation of threat, partner selection, the protection calculation — are primarily physiological rather than cognitive, then interventions that operate primarily at the cognitive level are unlikely to be sufficient, however well designed and delivered.

Before examining formal therapeutic options, it is important to establish something the essay's emphasis on physiological mechanism might otherwise obscure: the nervous system is not a closed system with no exit except through specialist therapy. The same neuroplasticity that allowed the threat cycle to be written in can allow something different to be written over it — given the right conditions, over sufficient time. Formal therapeutic intervention is one route. It is not the only one, and for many people it is not the first one.

Natural Healing: What Healthy Living and Healthy Relationships Can Do

The YoungFamilyLife essay Natural Healing: Understanding Recovery Across Physical, Psychological, and Social Dimensions draws a direct parallel between physical and psychological healing that is pertinent here. When a bone fractures, the body's natural healing capacity does not require instruction. It requires conditions: safety, rest, adequate nourishment, protection from re-injury, and time. Under those conditions, repair happens without any need for the person to understand the cellular mechanisms involved. The same principle applies to the nervous system's held threat patterns, though the timescale and conditions required are more complex.

What the research literature on trauma recovery consistently finds is that the most reliable precondition for change is not insight or formal intervention but relational safety sustained over time. Masten (2001) and Bonanno (2004), in their research on resilience, document that recovery from significant adverse experience occurs, for many people, not through therapeutic processes but through the gradual accumulation of experience in safe, warm, predictable relational environments — the kind that give the nervous system repeated evidence that people can be trusted, that closeness does not inevitably mean danger, that the cycle does not have to run to its learned endpoint in order for the body to settle.

Cozolino (2014), in his work on the neuroscience of human relationships, argues that the therapeutic relationship is itself a vehicle for neural change precisely because it provides what healthy relationships generally provide: consistent attunement, genuine interest, non-threatening presence, and the experience of being regulated by another person's calm. The specific mechanisms are those Schore (1994, 2003) identified in early caregiving — co-regulation building internal regulation — now operating in adulthood. This means that a sustained friendship, a warm and consistent family relationship, a stable community context, or any ongoing relational environment that offers genuine safety and attunement can, over time, begin to rewrite the patterns the nervous system built under threat. Not through talking about them, and not through understanding them, but through providing repeated counter-experience at the level where the patterns live.

This is not a counsel of passive waiting. It is a recognition that the primary conditions for healing — safety, connection, warmth, stability, time — are relational and environmental before they are therapeutic. Where those conditions exist or can be built, the nervous system's capacity for change is genuine and does not require professional facilitation to activate.

Kate Cairns, whose work as a therapeutic foster carer and trainer in trauma-informed practice has been influential in UK children's services, provides one of the most striking practical illustrations of this principle. Cairns did not use formal therapy with the traumatised children placed in her care. What she provided was herself — her physical presence, her bodily calm, her consistent and predictable warmth, her capacity to remain regulated when a child's nervous system was anything but. Communicated not primarily through words but through the body's own signals — tone of voice, facial expression, the quality of attention, the felt sense of safety in the room — her presence gave those children's nervous systems repeated evidence of something they had never reliably experienced: that another person's proximity meant safety rather than danger. Over time, and it took time, that accumulated evidence began to rewrite what their nervous systems had been built to expect. Not through explanation or insight. Through experience, delivered consistently, non-verbally, at the level where the patterns lived. The YFL IOW piece The Body Knows Safety Before the Mind Does explores this dimension of Cairns' contribution — and the broader principle that the body registers safety through channels that precede and run deeper than conscious thought. In the language of this essay, Cairns was feeding the solution — providing the relational counter-experience the survival brain needed — while starving the problem of the threat-activation that had kept the cycle running.

The same logic applies to physical health, movement, and the body's own regulatory systems. Vigorous physical activity — exercise as genuine discharge rather than as driven completion-seeking, to return to the distinction made in Section 4 — activates the same endorphin and cortisol-settling mechanisms the nervous system learned to access through threat. Sleep, adequate nutrition, and reduced chronic stress all support the parasympathetic functioning that Porges (2011) identifies as the prerequisite for nervous system repair. These are not supplements to the real work. For many people, they are a substantial part of it.

When Formal Therapy Is Part of the Picture: The Risk of Re-Traumatisation

Where formal therapeutic support is sought or offered, the choice of approach is not neutral. A poorly chosen or poorly timed therapeutic intervention does not simply fail to help. It can actively harm — consolidating the patterns it attempts to address, or adding new layers of distress to those already present. The concept of secondary trauma and therapeutic re-traumatisation is well established in the clinical literature (Figley, 1995; Herman, 1992), and the mechanisms are precisely those this essay has described.

Approaches that ask the person to narrate their trauma in detail — to tell the story, examine the sequence, articulate what happened and how it felt — are working from the premise that the problem is cognitive and that insight and verbal processing will resolve it. This is a reasonable premise for many forms of psychological difficulty. For the specific presentations described in this essay, it is frequently wrong. Van der Kolk (2014) is explicit on this point: asking a person whose trauma is held somatically to narrate it verbally activates the threat response without providing the somatic discharge through which the cycle can complete. The telling does not discharge the body's held energy. It mobilises it. If the therapeutic context cannot contain what is mobilised — if the therapist's training and skill are insufficient to work with what the body produces — the person leaves more activated than they arrived, with their patterns more deeply entrenched rather than less.

Cognitive-behavioural approaches present a related risk. The evidence base for cognitive-behavioural interventions in domestic violence is modest: a Cochrane review by Feder et al. (2009) found limited evidence of effectiveness, with effect sizes small and longer-term benefit weak. This is consistent with the theoretical framework here. Cognitive approaches ask the thinking brain to override the survival brain's patterns. But the patterns do not run through the thinking brain. They run through the subcortical structures — the amygdala, the brainstem, the autonomic nervous system — that were shaped before language, before cognition, and that operate at speeds that make conscious override functionally impossible in the moment of activation. Understanding the cycle intellectually, in the therapeutic hour, does not interrupt it when the body moves towards its learned endpoint at ten o'clock in the evening.

The risk of the wrong therapy is not merely ineffectiveness. It is that the person concludes therapy itself is the problem — that they are too broken to be helped, or that help makes things worse. Both conclusions impede the recovery that might still be possible through other means. And in the most difficult cases, a therapeutic experience that activates without resolving can consolidate the person's nervous system in precisely the patterns it was intended to address.

When Formal Therapy Is the Right Choice: The Case for Somatic and Non-Verbal Approaches

Where formal therapeutic intervention is appropriate — and it is appropriate for many people, particularly where the patterns are severe, entrenched, or significantly limiting daily functioning — the alignment between the patterns described in this essay and specific therapeutic modalities is clear. The body encoded the pattern. The body must be the primary site of the work.

Levine's somatic experiencing approach (1997) was specifically designed for exactly this purpose — guiding clients to track their body's activation states, to approach the edges of the incomplete cycle without retraumatising, and to facilitate the physiological discharge the original experience could not complete. The approach works with body sensation, movement, and the gradual titration of activation — bringing the nervous system towards the edge of the cycle in small, manageable increments, allowing discharge to occur in doses the person can tolerate, building the body's capacity for completion through routes other than the original one. The process is largely non-verbal. Understanding what is happening is less important than experiencing something different happening in the body.

Sensorimotor psychotherapy (Ogden et al., 2006) addresses the same territory through the integration of body-based work with relational and cognitive processing, sequenced carefully to ensure the body's held patterns are approached from a position of sufficient safety. EMDR (Eye Movement Desensitisation and Reprocessing; Shapiro, 1995), though it involves bilateral stimulation rather than primarily somatic technique, similarly targets the subcortical encoding of traumatic memory rather than its cognitive content, and has accumulated a substantial evidence base for trauma presentations including those associated with intimate partner violence (Bisson et al., 2013).

Porges' polyvagal-informed approaches (2011) address the autonomic hierarchy directly — using the therapeutic relationship as a vehicle for repeated ventral vagal activation, providing the nervous system with accumulated experience of social engagement as a viable and genuinely safe state. In this framework, the therapeutic relationship itself is not simply the vehicle for delivering technique. It is the primary intervention: a sustained, consistent, safe relational environment providing the nervous system with repeated counter-evidence — that settling is possible without the cycle having to peak, that another person's proximity can mean safety rather than danger, that the body does not have to stay ready.

What all of these approaches share, and what distinguishes them from the verbal and cognitive approaches likely to re-traumatise rather than resolve, is this: they require practitioners of high skill, substantial specialist training, and significant clinical experience in trauma specifically. The practitioner who has trained in somatic experiencing or sensorimotor psychotherapy as an additional qualification alongside primarily cognitive or humanistic training is a different resource from the practitioner whose primary formation and ongoing practice is in somatic trauma work. The specificity of skill required is not a minor qualification. It is the difference between an intervention that helps the nervous system find a different ending and one that mobilises the cycle without completing it.

Herman's stages of trauma treatment (1992) — safety, mourning, and reconnection — provide the clinical sequencing framework within which any of these approaches must be embedded. Herman's insistence that safety must be established before any processing work begins is consistent with everything this essay has argued: a nervous system that is not yet physiologically safe cannot engage with body-level work without activation overwhelming the capacity for discharge. Sanctuary remains the necessary first stage. The work described here is what must eventually follow it — carefully, with the right practitioner, at the right pace, in the right sequence.

The specific complexity introduced by the completion-seeking behaviours described in Section 4 — the eroticisation dimension, the addiction, the self-harm — requires acknowledgement in any intervention framework. These are not incidental complications to be addressed separately. They are expressions of the same underlying physiology, and any therapeutic approach that addresses the DV pattern without attending to them risks leaving the nervous system still seeking its learned endpoint through other available routes.

---

Conclusion

The patterns in domestic violence that have most resisted conventional explanation — escalation towards harm, the failure of separation to interrupt the cycle, successive reproduction of the same dynamic, the inadequacy of sanctuary alone — are coherent when understood through the integrated neurobiological, developmental, and relational framework this essay has assembled. They are not failures of reason, failures of self-protection, or failures of love in any simple sense. They are the expressions of a nervous system doing precisely what it was built to do, in conditions that no longer match those in which it was built.

The ACE literature establishes the association between early adversity and adult DV involvement with consistent force. The framework offered here — drawing on Levine, van der Kolk, Porges, Schore, Stoller, Dutton and Painter, Bowlby, and the developmental trauma literature — proposes the mechanism: the incomplete threat cycle, formed in early experience, pursued in adult relationship; the eroticisation of that cycle under specific developmental conditions; the selection of partners by a nervous system calibrated to recognise what it knows; and the territorial protection calculation that gives the harmful partner a function beyond the cycle itself.

No single framework describes all experiences of domestic violence. The patterns identified here are specific and clinically significant. They are not universal, and the absence of these patterns in a given DV situation does not in any way reduce the reality or severity of the harm being done. What this framework offers is not a typology of victims, or a distribution of responsibility, but a more accurate account of what the nervous system is doing — and therefore a more useful basis for designing the intervention it actually requires.

The question that evidence-informed practice must ask, in the light of this account, is not why can't they just stop? It is: what does this person's nervous system need in order to find a different ending? That is a harder question to answer and a more demanding standard to meet. It is also, the evidence suggests, the right one.

This essay sits alongside two YoungFamilyLife IOW companion pieces that address the same territory in accessible register: What Children Carry: Growing Up in a Home Where Violence Is Present, which explores the child's experience within the DV home; and The Body's Unfinished Business (IOW), which addresses the adult physiological patterns explored here in plain-language IOW format.

---

References

Bisson, J. I., Roberts, N. P., Andrew, M., Cooper, R., & Lewis, C. (2013). Psychological therapies for chronic post-traumatic stress disorder (PTSD) in adults. Cochrane Database of Systematic Reviews, 12, CD003388.

Bonanno, G. A. (2004). Loss, trauma, and human resilience: Have we underestimated the human capacity to thrive after extremely aversive events? American Psychologist, 59(1), 20–28.

Bowlby, J. (1969). Attachment and Loss, Vol. 1: Attachment. Hogarth Press.

Bowlby, J. (1973). Attachment and Loss, Vol. 2: Separation. Hogarth Press.

Bowlby, J. (1980). Attachment and Loss, Vol. 3: Loss. Hogarth Press.

Classen, C. C., Palesh, O. G., & Aggarwal, R. (2005). Sexual revictimization: A review of the empirical literature. Trauma, Violence, & Abuse, 6(2), 103–129.

Cozolino, L. (2014). The Neuroscience of Human Relationships: Attachment and the Developing Social Brain (2nd ed.). Norton.

Dutton, D. G., & Painter, S. L. (1981). Traumatic bonding: The development of emotional attachments in battered women and other relationships of intermittent abuse. Victimology: An International Journal, 6(1–4), 139–155.

Dutton, D. G., & Painter, S. (1993). Emotional attachments in abusive relationships: A test of traumatic bonding theory. Violence and Victims, 8(2), 105–120.

Fairbairn, W. R. D. (1952). Psychoanalytic Studies of the Personality. Tavistock.

Feder, G., Davies, R. A., Baird, K., Dunne, D., Eldridge, S., Griffiths, C., Gregory, A., Howell, A., Johnson, M., Ramsay, J., Rutterford, C., & Sharp, D. (2009). Identification and referral to improve safety (IRIS) of women experiencing domestic violence with a primary care training and support programme. Lancet, 375(9732), 2160–2167.

Felitti, V. J., Anda, R. F., Nordenberg, D., Williamson, D. F., Spitz, A. M., Edwards, V., Koss, M. P., & Marks, J. S. (1998). Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults. American Journal of Preventive Medicine, 14(4), 245–258.

Figley, C. R. (1995). Compassion Fatigue: Coping with Secondary Traumatic Stress Disorder in Those Who Treat the Traumatized. Brunner/Mazel.

Freud, S. (1920). Beyond the Pleasure Principle. Standard Edition, Vol. 18. Hogarth Press.

Glover, V. (2011). Annual research review: Prenatal stress and the origins of psychopathology: An evolutionary perspective. Journal of Child Psychology and Psychiatry, 52(4), 356–367.

Glover, V. (2014). Maternal depression, anxiety and stress during pregnancy and child outcome: What needs to be done. Best Practice & Research Clinical Obstetrics & Gynaecology, 28(1), 25–35.

Guntrip, H. (1969). Schizoid Phenomena, Object Relations and the Self. Hogarth Press.

Hazan, C., & Shaver, P. (1987). Romantic love conceptualized as an attachment process. Journal of Personality and Social Psychology, 52(3), 511–524.

Herman, J. L. (1992). Trauma and Recovery. Basic Books.

Holt, S., Buckley, H., & Whelan, S. (2008). The impact of exposure to domestic violence on children and young people: A review of the literature. Child Abuse & Neglect, 32(8), 797–810.

Johnson, S. M. (2004). The Practice of Emotionally Focused Couple Therapy: Creating Connection (2nd ed.). Brunner-Routledge.

Levine, P. A. (1997). Waking the Tiger: Healing Trauma. North Atlantic Books.

Levine, P. A. (2010). In an Unspoken Voice: How the Body Releases Trauma and Restores Goodness. North Atlantic Books.

Main, M., & Hesse, E. (1990). Parents' unresolved traumatic experiences are related to infant disorganized attachment status. In M. T. Greenberg, D. Cicchetti, & E. M. Cummings (Eds.), Attachment in the Preschool Years (pp. 161–182). University of Chicago Press.

Maté, G. (2008). In the Realm of Hungry Ghosts: Close Encounters with Addiction. Knopf Canada.

Masten, A. S. (2001). Ordinary magic: Resilience processes in development. American Psychologist, 56(3), 227–238.

Messman-Moore, T. L., & Long, P. J. (2003). The role of childhood sexual abuse sequelae in the sexual revictimization of women. Clinical Psychology Review, 23(4), 537–571.

Norman, R. E., Byambaa, M., De, R., Butchart, A., Scott, J., & Vos, T. (2012). The long-term health consequences of child physical abuse, emotional abuse, and neglect. PLOS Medicine, 9(11), e1001349.

Perry, B. D. (1995). Childhood trauma, the neurobiology of adaptation and use-dependent development of the brain: How states become traits. Infant Mental Health Journal, 16(4), 271–291.

Perry, B. D., & Szalavitz, M. (2006). The Boy Who Was Raised as a Dog. Basic Books.

Porges, S. W. (1994). The polyvagal theory: Phylogenetic substrates of a social nervous system. International Journal of Psychophysiology, 42(2), 123–146.

Porges, S. W. (2011). The Polyvagal Theory: Neurophysiological Foundations of Emotions, Attachment, Communication, and Self-Regulation. Norton.

Ports, K. A., Ford, D. C., & Merrick, M. T. (2016). Adverse childhood experiences and sexual victimization in adulthood. Child Abuse & Neglect, 51, 313–322.

Scharff, D. E., & Scharff, J. S. (1991). Object Relations Couple Therapy. Jason Aronson.

Shapiro, F. (1995). Eye Movement Desensitization and Reprocessing: Basic Principles, Protocols, and Procedures. Guilford Press.

Schmidt, U., & Treasure, J. (2006). Anorexia nervosa: Valued and visible. A cognitive-interpersonal maintenance model and its implications for research and practice. British Journal of Clinical Psychology, 45(3), 343–366.

Ogden, P., Minton, K., & Pain, C. (2006). Trauma and the Body: A Sensorimotor Approach to Psychotherapy. Norton.

Schore, A. N. (1994). Affect Regulation and the Origin of the Self. Lawrence Erlbaum.

Schore, A. N. (2003). Affect Dysregulation and Disorders of the Self. Norton.

Siegel, D. J. (1999). The Developing Mind. Guilford Press.

Siegel, D. J. (2012). The Developing Mind (2nd ed.). Guilford Press.

Sprouse-Blum, A. S., Smith, G., Sugai, D., & Parsa, F. D. (2010). Understanding endorphins and their importance in pain management. Hawaii Medical Journal, 69(3), 70–71.

Stith, S. M., Rosen, K. H., Middleton, K. A., Busch, A. L., Lundeberg, K., & Carlton, R. P. (2000). The intergenerational transmission of spouse abuse: A meta-analysis. Journal of Marriage and Family, 62(3), 640–654.

Stoller, R. J. (1975). Perversion: The Erotic Form of Hatred. Pantheon Books.

Treasure, J., Claudino, A. M., & Zucker, N. (2010). Eating disorders. Lancet, 375(9714), 583–593.

van der Kolk, B. A. (1989). The compulsion to repeat the trauma: Re-enactment, revictimization, and masochism. Psychiatric Clinics of North America, 12(2), 389–411.

van der Kolk, B. A. (1994). The body keeps the score: Memory and the evolving psychobiology of posttraumatic stress. Harvard Review of Psychiatry, 1(5), 253–265.

van der Kolk, B. A. (2014). The Body Keeps the Score: Brain, Mind, and Body in the Healing of Trauma. Viking.

Winnicott, D. W. (1960). Ego distortion in terms of true and false self. In The Maturational Process and the Facilitating Environment (pp. 140–152). Hogarth Press.