Therapists working with couples in the aftermath of a relationship rupture distinguish between two different kinds of wound. One is called an attachment injury. The other is called betrayal. They can look identical from the outside — the same tears, the same sleeplessness, the same fights — and they require meaningfully different repair work. Couples who treat an attachment injury as if it were a betrayal often over-correct in ways that don't fit the wound. Couples who treat a betrayal as if it were "just" an attachment injury under-do the necessary repair and stay stuck for years. The distinction is one of the more useful pieces of clinical vocabulary available to lay readers, and it doesn't get translated out of the journals as often as it should.
The attachment-injury concept was formalised by Sue Johnson and Marlene Williams-Keeler in a 1998 paper that grew out of Johnson's clinical work in Emotionally Focused Therapy. Johnson noticed that some couples in EFT were not progressing through the standard treatment stages — they were stuck on a specific past event, often something the partner who had done the injuring genuinely considered a minor misstep, that had become the load-bearing emotional event in the relationship's recent history. The injured partner kept coming back to it. The injuring partner kept failing to understand why. The framework Johnson developed to address this stuck-point became the attachment-injury treatment model, and it has substantially shaped how EFT-trained therapists work with these couples.
Defining The Two Wounds
An attachment injury, in Johnson's framework, is a moment in which one partner needed the other to be emotionally present — at a moment of real vulnerability, distress, fear, grief, or significant transition — and the partner failed to show up in the way the attachment system demanded. The failure may have been a minor logistical thing from the outside. The injury is not the size of the action; it is the meaning the action took on at a moment of attachment-system need. The injured partner's working model now treats this person as someone who is not reliably there when it matters.
A betrayal, in the related but distinct literature, is a violation of an explicit or strongly-implicit agreement about the relationship — most commonly infidelity, but also undisclosed major financial decisions, secret addictions, ongoing deception about identity or history. The betrayed partner is not only attachment-injured (though they typically are); they have also discovered that the relationship's basic factual premises were inaccurate. The injury overlaps with the attachment frame; the betrayal adds an epistemological layer the attachment frame doesn't cover. (See trust after cheating.)
Why The Distinction Matters Clinically
An attachment injury, in Johnson's treatment model, is repairable through a structured EFT process that focuses on the attachment system's need to update its assessment of the partner's availability. The repair has roughly four phases: the injuring partner hears the injury without minimising; the injured partner expresses the underlying fear and grief without weaponising; the injuring partner acknowledges specifically what happened from the injured partner's frame; the injured partner allows the new responsive behaviour to start updating the working model. The work is hard but the framework is straightforward.
A betrayal, by contrast, demands a more extensive repair architecture. The attachment-system updating still has to happen, but it cannot happen until the factual layer is restored — full disclosure, the end of the deception, often the involvement of a therapist or third party to verify. Esther Perel's clinical work on infidelity has emphasised how betrayal additionally introduces what she calls a "splitting" of the relationship into the pre-discovery and post-discovery versions; the post-discovery couple has to do attachment work the pre-discovery couple did not have to do. The two layers overlap; treating only one stalls the other. (See attachment injury repair.)
The common mistake — confusing the two
Couples (and sometimes therapists who haven't been trained in the framework) often treat attachment injuries as betrayals, which makes the injured partner feel justified in withholding repair indefinitely. Or they treat betrayals as attachment injuries, which underestimates the factual-disclosure work the relationship needs. Both errors are common; both keep couples stuck for years past the point where targeted work would have moved them. Getting the framing right early changes the trajectory.
The Hallmarks Of An Attachment Injury
Attachment injuries have characteristic features. Recognising them in your own relationship's history is useful for orienting the repair work.
Hallmark 1 — A specific identifiable moment. Attachment injuries are typically tied to a particular event or short sequence of events. The injured partner can usually name the moment with specificity. "The night I told you I was bleeding and you went to bed because you had a meeting in the morning." "When your mother spoke to me at the funeral and you didn't say anything." The specificity is part of the diagnosis.
Hallmark 2 — The event happened at a moment of real attachment need. Attachment injuries don't form around routine moments. They form around moments when the injured partner's attachment system was activated by genuine distress, fear, vulnerability, or grief — and the partner's response failed to register the activation. The same behaviour from the partner on a calm Tuesday would not have produced an injury. The timing is part of the diagnosis.
Hallmark 3 — The injured partner returns to it. The hallmark behavioural indicator. Months or years after the event, in arguments about unrelated topics, the injury keeps surfacing. "You're like you were that night." The injured partner is not "bringing it up again"; the working model is. Each surfacing is the attachment system's attempt to get the missing repair. (See repair after conflict.)
Hallmark 4 — The injuring partner often considers the moment minor. This asymmetry is one of the most reliable diagnostic features. The injured partner experiences the event as significant; the injuring partner often genuinely remembers it as a small thing or doesn't remember the specifics at all. The asymmetry is not a sign of bad faith; it is a sign that the meaning of the event was about the attachment-system context, which only one partner was inside.
The Hallmarks Of A Betrayal
Betrayals share the attachment-injury features but add several distinctive ones.
Hallmark 5 — A violation of an explicit or strongly-implicit agreement. Betrayals involve breaking something the couple had agreed on. Sexual exclusivity. Financial transparency. Honesty about identity, history, addiction, contact with an ex-partner. The agreement may not have been spelled out in a contract, but it was part of the relationship's foundational structure.
Hallmark 6 — Concealment over time. Betrayals typically involve sustained concealment — not a single mistake but a pattern of behaviour the betrayer chose not to disclose. The concealment is the part that produces the epistemological injury: the betrayed partner now knows that what they thought was true wasn't, and has to reassess all the months or years during which the concealment was active.
Hallmark 7 — A "before" and "after." Betrayals split the relationship into the version that existed before the discovery and the version that exists after. The "before" version cannot be returned to. The "after" version is a different relationship that may or may not be tenable. This splitting is what Esther Perel has called the "two relationships" frame — the work after a betrayal is the building of a new relationship with the same partner, not the repair of the existing one.
Most ruptures are attachment injuries, not betrayals.
The slow steady repair work — the kind LoveCertain matches are pointed toward — is built for the everyday injuries that show up in any long relationship. Matching on attachment style at 20% means fewer of these happen in the first place. £49 once. Full refund if no relationship in 90 days. £99 bonus if there is.
The Two Different Repair Architectures
Repairing an attachment injury — the EFT-derived framework
Phase 1 — Acknowledgement without minimisation
The injuring partner has to first acknowledge that something significant happened, on the injured partner's terms, before any other repair work is possible. This is the phase most couples try to skip. The injuring partner says "I'm sorry that happened to you" rather than "I'm sorry I did that to you" — the grammatical distinction matters more than it sounds. The acknowledgement has to be specific about what happened and avoid the qualifying construction ("I'm sorry but...") that defends the injuring partner's intent at the cost of the injured partner's experience.
Phase 2 — The injured partner expresses the underlying fear
Beneath the anger about the event is usually a deeper fear — typically about the injuring partner's reliability, availability, or commitment. The injured partner has to be able to express that fear, which they often haven't done because the anger is more accessible than the fear. The expression of the fear is what gives the injuring partner something to respond to that isn't the surface complaint. The fear underneath "you weren't there" is "I'm not sure you will be there when I need you next time." (See expressing needs without a fight.)
Phase 3 — The injuring partner offers the responsive response
The injuring partner now responds to the underlying fear rather than the surface event. "I hear that you weren't sure I would be there. I want to tell you what I want to do differently when you need me." The response addresses the working-model question, which is the actual question the attachment injury is asking. The verbal response is not enough by itself; it has to be paired with behavioural follow-through in the weeks that follow.
Phase 4 — The behavioural updating
The next several months are the period in which the working model gets updated by direct evidence. The injuring partner shows up differently in the situations that would previously have produced the injury pattern. The injured partner's attachment system has to be allowed to update — which means tolerating the gap between the verbal repair and the felt repair. The verbal often comes within weeks. The felt typically takes months. Both are necessary.
Repairing a betrayal — the additional architecture
Additional Phase A — Full factual disclosure
The betrayer has to disclose the relevant facts — the duration, the people involved (where relevant), the lies that were told, the moments when the betrayed partner asked and was misled. The disclosure should be complete enough that no later revelation can produce a second discovery wound. This is the part most betrayers want to phase in slowly; the slow phasing tends to produce serial re-injuries that prevent repair. (See trust issues in relationships.)
Additional Phase B — The end of the betraying behaviour
Repair cannot meaningfully begin while the betrayal is ongoing. The affair has to end. The hidden bank account has to be closed. The addiction has to be in active treatment. The end of the behaviour is the precondition for the rest of the architecture; couples who try to repair while the betrayal continues do not in fact repair.
Additional Phase C — Restoration of transparency
The betrayed partner often needs, for an extended period (typically months to years), a level of factual transparency that would be invasive in a healthy relationship. Read the messages. Check the location. See the bank statements. Most couples therapists support this provisionally — it is the rebuilding scaffolding, not the new normal. The scaffolding comes down gradually as the working model rebuilds. The couple's eventual stability does not require permanent surveillance; it requires the working model to update past needing it.
Additional Phase D — A coherent shared narrative
The couple has to develop, eventually, a shared understanding of what the betrayal was, why it happened, what it meant, and what changed afterwards. This is not the same as both partners agreeing on every detail; it is the development of a coherent account that both can live with. Couples who arrive here are usually three to five years past the discovery. The work is slow. The pace is unavoidable.
What Doesn't Work — For Either Wound
The "just move on" demand
Some injuring partners (and some betrayers) ask the injured partner to "just move on" or "stop bringing it up." The demand is not just unfair; it is structurally impossible. The working model cannot update on demand. The repair work, when done, produces the move-on as a side effect of the underlying updating. Asking for the side effect without doing the work prevents the work. The injured partner is not bringing it up to punish; the working model is bringing it up to get the missing repair. (See arguing without destroying the relationship.)
The over-apology without behavioural change
The injuring partner who apologises eloquently but doesn't change behaviour produces what therapists call a "repair simulation" — the appearance of repair without the substance. The working model recognises the simulation, often before the conscious mind does, and continues to register the partner as unreliable despite the apologies. Verbal repair has to be paired with sustained behavioural follow-through to do any real work.
The premature reconciliation
Some couples, particularly after a betrayal, rush through the repair architecture because the rupture is unbearable to sit with. They forgive too fast, return to normal too soon, and discover months later that the working model never got the updating it needed. The injured partner becomes hypervigilant or quietly resentful; the relationship that looked repaired begins to deteriorate from the inside. The repair architecture has to be allowed to take its time. The time itself is part of the medicine.
The diagnostic move when something feels stuck
If a particular event keeps coming back in arguments about unrelated topics, name it as a possible attachment injury rather than a recurring complaint. The shift in framing — from "you keep bringing this up" to "this is an injury that hasn't yet been repaired" — usually opens the conversation. From there, the four-phase EFT repair work is available. If the event involved a violation of an explicit or implicit agreement, treat it as a possible betrayal as well, and add the disclosure-and-transparency layer to the repair architecture. The two frames complement each other when the wound is both. (See repair attempts in couples.)
The wider research
Sue Johnson and Marlene Williams-Keeler's 1998 paper introducing the attachment-injury concept (Journal of Marital and Family Therapy) is the foundational source. Sue Johnson's Hold Me Tight (2008) translates the model for lay readers. Esther Perel's The State of Affairs (2017) is the most-cited contemporary treatment of infidelity. John Gottman's repair-attempts research, summarised across his published work, gives the smaller-scale repair vocabulary that complements the larger-scale attachment-injury repair architecture. The two frameworks are compatible and often used together in practice.
The Certain Letter
Weekly relationship-science briefings. 4-minute read.
Why This Matters Earlier — In Dating
Two implications for adults still in dating life. First: the way a partner handles small attachment injuries in the early months of a relationship is one of the most predictive signals about how they will handle larger ones later. A partner who, on a small inadvertent slight, listens, hears the underlying point, acknowledges without minimising, and adjusts — that is a partner who can be expected to do the larger repair work when it eventually arrives. A partner who, on the same small slight, defends, minimises, or escalates — that is a partner who will struggle with the larger work. The small repair is the rehearsal. (See the first fight in a new relationship.)
Second: betrayal is not the only failure mode. Many long relationships end through accumulated unrepaired attachment injuries rather than through any single discoverable betrayal. The honest framing is that both kinds of repair need to be available to a couple over years. The selection of a partner who is capable of the smaller repair makes the larger repair, if it ever needs to happen, more likely to succeed. The LoveCertain matching weights communication style at 15% partly for this reason. (See 12 communication skills that actually work.)
For an external authoritative overview of the Gottman repair literature that interlocks with the attachment-injury framework, see the Gottman Institute's primary-source summary of repair attempts.
The Encouragement
Most long relationships will encounter at least one attachment injury, often several across the decades. Most will not encounter a betrayal. The framework for the smaller wound is well-developed, well-evidenced, and available. The framework for the larger wound is more demanding but also exists. Naming what kind of wound a couple is facing is the start of the repair architecture rather than its conclusion. The naming is also the thing that often allows the conversation to begin moving for the first time in months or years. Use the vocabulary. The clinical literature has done useful work; the work can come into your house.
