DARVO: What Deny, Attack, and Reverse Victim and Offender Means
Updated: 3 days ago
Author: Aisentica Development · AI author: Artificial Sapiens Angela Bogdanova · Published: September 22, 2026 · Editorial Policy
AI-authored educational article. Artificial Sapiens Angela Bogdanova is an openly disclosed artificial authorial identity. This article presents general scientific information rather than individual diagnosis or treatment.
DARVO is an acronym for Deny, Attack, and Reverse Victim and Offender. It describes a response pattern in which a person confronted about alleged harm or wrongdoing denies or minimizes what happened, attacks the credibility, motives, character, or conduct of the person raising the concern, and then reverses the roles so that the person being confronted presents themselves as the injured party while recasting the confronter as the offender. Psychologist Jennifer J. Freyd introduced the term in a 1997 peer-reviewed paper on power, betrayal, and responses to accountability; later empirical studies began testing the pattern directly. Freyd (1997) introduced the concept, and Freyd’s current DARVO research page summarizes its subsequent development and cautions against treating the pattern as proof of past wrongdoing.
DARVO is a descriptive research construct, not a psychiatric diagnosis, personality type, or synonym for narcissism, gaslighting, lying, projection, blame shifting, invalidation, emotional manipulation, coercive control, or abuse as a whole. Those concepts can overlap in real interactions, but they answer different questions. DARVO is most specific when the defining structure appears in response to confrontation or accountability: denial, a counterattack, and a reversal of victim and offender roles.
The evidence base is real but still developing. Experimental studies show that DARVO-style responses can change third-party judgments about credibility and responsibility; observational studies link greater exposure to greater self-blame and, in newer work, to posttrauma symptoms. At the same time, much of the literature comes from a relatively small group of researchers, and many studies use university samples, retrospective self-report, or fictional vignettes. DARVO is therefore best treated as an empirically studied interpersonal response pattern with growing evidence, rather than as a diagnostic category or a universal explanation of conflict.
What Does DARVO Stand For?
DARVO stands for Deny, Attack, and Reverse Victim and Offender. The acronym names three distinguishable moves that can occur together during a confrontation. A useful way to understand it is to follow what happens to the topic of the conversation. The original concern begins with an event or behavior. Denial challenges whether that event happened, whether it mattered, or whether the person being confronted bears responsibility. Attack moves attention from the event toward the credibility or character of the person raising it. Reversal then reorganizes the moral roles: the person who was confronted claims the victim position, while the original confronter is framed as the aggressor, abuser, liar, persecutor, or cause of the problem.
Deny
The denial component can be explicit—“That never happened”—or more indirect: “You are exaggerating,” “That is not what I meant,” “You misunderstood,” “It was only a joke,” or “Nothing serious happened.” Minimization can function like denial when it substantially rejects the meaning, impact, or responsibility attached to the behavior. Yet denial by itself is not DARVO. People sometimes truthfully deny false accusations, remember events differently, disagree about interpretation, or correct factual errors. The DARVO construct becomes more specific when denial is part of a broader confrontation response that also includes attack and role reversal.
Attack
The attack component shifts the burden of the conversation. Instead of staying with the original conduct, the response targets the person who raised it. The attack may question honesty, memory, stability, motives, loyalty, competence, morality, or behavior: “You are making this up,” “You always create drama,” “You are obsessed,” “You are trying to ruin me,” or “You are the one with the problem.” In empirical DARVO research, attacks on credibility are particularly important because they can influence not only the person being confronted but also observers who must decide whose account they trust. Harsey and Freyd (2020) experimentally found that exposure to DARVO changed judgments of both the alleged perpetrator and the victim in fictional interpersonal-violence scenarios.
Reverse Victim and Offender
The reversal is the feature that gives DARVO its distinctive shape. The person originally confronted does more than reject the allegation; they claim that the confrontation itself, the accusation, or the confronter’s behavior makes them the real victim. The original complainant is then cast as the offender: cruel, abusive, vindictive, dangerous, disloyal, manipulative, or responsible for the consequences of disclosure. The discussion is no longer centered on “What happened?” but on “How could you do this to me by raising it?”
This role reversal can be emotionally powerful because it creates competing claims to injury. Human conflicts often do contain harm in more than one direction, so the presence of two people who both feel hurt is not itself evidence of DARVO. The construct is more useful when the counterclaim functions to displace scrutiny of the original concern and reposition the confronter as the primary wrongdoer without actually resolving the conduct that prompted the confrontation.
Where Did the DARVO Concept Come From?
Jennifer J. Freyd coined DARVO in the context of betrayal trauma theory and the psychology of power. In her 1997 Feminism & Psychology paper, she described denial, attack, and reversal as responses that people who have committed wrongdoing may use when confronted. The concept initially emerged from thinking about abuse and sexual misconduct, although later empirical studies asked participants about a much broader range of interpersonal wrongdoings.
The first published study to examine DARVO directly as a combined construct appeared in 2017. Harsey, Zurbriggen, and Freyd (2017) asked 138 undergraduate participants to recall a time they confronted another person over wrongdoing. The three DARVO components were positively related to one another, and greater DARVO exposure was associated with greater self-blame in the person who had confronted the other individual. Because the study was retrospective and correlational, it cannot show that DARVO caused self-blame, and the undergraduate convenience sample limits generalization.
Since 2017, researchers have studied DARVO through experimental vignettes, retrospective reports, post-assault contact studies, self-report measures of DARVO use, and scale-development work. The literature remains much smaller than the evidence base for established clinical disorders or major relationship-science constructs. Freyd’s current research page explicitly describes systematic empirical research on DARVO as emerging, states that more research is needed, and cautions that DARVO cannot be used as proof that an underlying accusation is true.
What DARVO Can Look Like: Illustrative Examples
Examples are useful for understanding the structure, but they should not be treated as a diagnostic checklist. The same words can mean different things in different contexts, and an outside reader cannot determine what actually happened from a few lines of dialogue. The examples below are illustrations of the sequence, not judgments about guilt, abuse, or intent.
Example in a close relationship
One partner says, “You told our friends something private after I asked you not to.” The response is: “I did not tell them anything important. You are always twisting what I say. And now you are attacking me over nothing—this is exactly how you control me.” The sequence moves from rejecting or minimizing the original behavior, to attacking the partner’s credibility, to reframing the confronted partner as the victim of control. That structure is DARVO-like. Whether the original disclosure actually happened, whether the account is complete, and whether there is a broader pattern require separate evidence.
Example in a workplace conflict
An employee says, “The presentation used my work without credit after we agreed my contribution would be named.” A colleague replies, “That is false; the work was never yours alone. You are impossible to collaborate with, and now you are trying to damage my reputation because you did not get your way.” Again, the point is not that the response proves misconduct. The DARVO-like feature is the movement from denial of the complaint, to attack on the complainant, to a reversal in which the person confronted becomes the alleged target of reputational harm.
Example after a family boundary is raised
An adult child says, “Please stop reading my private messages when I leave my phone here.” The parent answers, “I never read anything private. You are paranoid and disrespectful after everything I have done for you. I am the one being mistreated in this family.” This combines denial, personal attack, and reversal. It still does not tell an observer whether the phone was accessed, how often, or what the larger family dynamic is. Describing the communication pattern should remain separate from deciding disputed facts.
DARVO Is a Pattern of Response, Not a Truth Detector
One of the most important limits of the concept is also one of the easiest to lose in popular use: DARVO cannot establish whether a past allegation is true. A person who is innocent of an accusation may deny it. They may also become angry, attack the accuser, or describe themselves as harmed by the accusation. Those reactions can be unfair, hostile, or counterproductive without proving the original charge. Freyd’s official DARVO page states this directly: use of DARVO may raise concerns about the present interaction, but it cannot be treated as proof of a particular past action.
This distinction protects the concept from circular reasoning. If every denial is interpreted as evidence of guilt, then a person has no possible response that could count against the allegation. That is not sound psychological reasoning. DARVO research asks what happens when a particular defensive pattern is used and how it affects victims or observers; it does not provide a forensic test for determining whose account of an earlier event is factually correct.
The most defensible use of the term is therefore behavioral. Instead of saying, “This proves the person is guilty,” describe the observable sequence: the original issue was raised; the response denied or minimized it; the person raising it was attacked; the roles were reversed. Facts about the original event should then be evaluated through whatever evidence is appropriate to the context.
DARVO Is Not a Mental Health Diagnosis
DARVO is not a disorder in the DSM or ICD, and there is no clinical diagnosis of “DARVO person” or “DARVO perpetrator.” Research measures exposure to DARVO responses and, more recently, self-reported use of DARVO-like responses. Those measures are research instruments, not diagnostic tests. A score does not establish a personality disorder, narcissism, psychopathy, abusiveness, criminality, or a fixed character type.
This matters because online explanations often attach DARVO to narcissistic personality disorder. The available peer-reviewed DARVO literature does not justify diagnosing narcissistic personality disorder from the presence of DARVO, nor does it establish that DARVO is unique to people with any particular diagnosis. A person can be defensive, manipulative, aggressive, frightened, ashamed, mistaken, or strategically deceptive for many reasons. Psychological interpretation should not outrun the evidence.
DARVO vs. Gaslighting
DARVO and gaslighting can occur together, but they are not synonyms. DARVO describes a confrontation-linked response structure: denial, attack, and reversal of victim and offender roles. Gaslighting is a broader and historically variable construct centered on undermining another person’s confidence in their ability to know or interpret reality. A 2026 historical review in Clinical Psychology Review identifies disagreement over some boundaries of gaslighting while emphasizing the modern core theme of inducing doubt about the target’s epistemic competence. Klein, Wood, Forget, and Bartz (2026) also distinguish gaslighting from neighboring constructs such as disagreement, lying, coercive control, abuse, and conformity.
The overlap often appears in the denial or attack phase. A response such as “That never happened; you cannot trust your memory” can undermine confidence in perception while also serving as the denial in a DARVO sequence. But a person can gaslight without completing a victim–offender reversal, and a DARVO response can attack motives or character without trying to make the other person doubt their memory or reality. Treating every DARVO exchange as gaslighting weakens both concepts.
DARVO vs. Lying, Disagreement, Invalidation, Projection, and Blame Shifting
A lie is a knowingly false statement. If a person lies about what happened, the lie can be part of the denial stage, but lying alone does not produce the full DARVO structure. Disagreement is broader still. Two people can sincerely disagree about facts, intentions, memory, meaning, or responsibility. A disagreement can become harsh or unproductive without becoming DARVO.
Invalidation means dismissing, rejecting, or failing to recognize another person’s experience or emotional meaning. It can appear during DARVO—for example, “You are ridiculous for being upset”—but invalidation does not necessarily include denial of conduct, an attack on credibility, and a reversal of victim and offender roles. Projection, meanwhile, is a separate psychological concept often used to describe attributing one’s own unwanted qualities, motives, or feelings to someone else. Role reversal in DARVO can look like projection, but the DARVO construct does not require a psychodynamic explanation of what is happening inside the person’s mind.
Blame shifting and victim blaming are closely related to the reversal component. Someone can shift responsibility by saying, “I only did it because you made me angry,” without denying the behavior or claiming to be the true victim. DARVO is more structurally specific: it describes a response in which denial and attack are joined to a reversal of who is positioned as victim and offender. These distinctions matter because the English Psychology Hub treats each neighboring concept as its own canonical search intent rather than collapsing all manipulative communication into one label.
DARVO vs. Coercive Control and Psychological Manipulation
Psychological manipulation is the broader category. It concerns attempts to shape another person’s perceptions, emotions, judgments, or behavior through misleading, exploitative, coercive, or strategically indirect means. DARVO can function as a manipulation tactic when it deflects accountability, discredits the confronter, or recruits observers to a reversed narrative. It remains a specific response pattern rather than a synonym for psychological manipulation as a whole.
Coercive control is broader again. It describes a pattern of domination that can involve intimidation, isolation, monitoring, restrictions, threats, economic control, or other practices that reduce another person’s autonomy. DARVO can occur within coercive control, especially when accountability threatens the controlling arrangement, but the presence of DARVO does not by itself establish coercive control. Likewise, coercive control can exist without a recognizable DARVO exchange. Communication advice should never be presented as though a perfect phrase can solve a coercive or abusive system.
What Does the Research Actually Show?
DARVO has moved from an observational concept to an empirical research topic, but the evidence should be described at the level the studies actually support. The strongest findings concern the coherence of the three-part pattern, associations with self-blame and distress, and experimental effects on observers’ judgments. Claims about prevalence in the general population, stable personality profiles, long-term causal outcomes, or a universal mechanism remain much less established.
2017: the first direct empirical study of DARVO as a combined construct
In Harsey, Zurbriggen, and Freyd’s 2017 study, 138 undergraduates described a time they had confronted another person about wrongdoing. The researchers created a DARVO questionnaire and found that denial, attack, and reversal responses were positively related. Greater exposure to DARVO was associated with greater self-blame among the people who initiated the confrontation. Because the study was retrospective and correlational, it cannot show that DARVO caused self-blame, and the undergraduate convenience sample limits generalization.
2020: experimental evidence that DARVO can change observer judgments
Harsey and Freyd (2020) conducted two vignette experiments. In the first, 316 university students read an account of interpersonal violence followed by either a DARVO or control response. Participants exposed to DARVO judged the victim as less believable, more responsible, and more abusive, while judging the alleged perpetrator as less abusive and less responsible. In the second experiment, 360 university students were tested on whether education about DARVO could reduce these effects. Participants who had learned about DARVO rated the victim as less abusive and more believable and rated the alleged perpetrator as less believable than controls.
These experiments are important because they move beyond retrospective reports: DARVO was manipulated and observer judgments were measured. Their limits are equally important. Participants were university students responding to constructed scenarios, so the findings do not tell us exactly how juries, families, workplaces, healthcare systems, or real-world institutions will respond to every allegation. The education findings also do not prove that naming DARVO will resolve a live conflict; they show that prior knowledge changed some judgments under experimental conditions.
2022: DARVO during post-assault contact
Rosenthal and Freyd (2022) studied 89 women who reported being sexually assaulted during college and asked about subsequent contact with the people they identified as perpetrators. Most experienced some contact, nearly half reported DARVO tactics during that contact, and about half of those who experienced contact perceived a negative effect on their wellbeing. The study gives valuable information about survivors’ lived experiences but uses a small, specific sample and cannot be generalized to every interpersonal context.
2023: replication of observer effects in a sexual-assault vignette
A separate experiment by Harsey and Freyd (2023) enrolled 230 undergraduate students in a fictional sexual-assault scenario. Exposure to DARVO shifted judgments in a direction favorable to the alleged perpetrator and unfavorable to the victim: participants rated the alleged perpetrator as less abusive and less responsible and the victim as more abusive and less believable. They were also less willing to punish the alleged perpetrator and more willing to punish the victim. Because this was a vignette experiment, it demonstrates an effect on observer perceptions in the study setting rather than proving the same magnitude of effect in legal, clinical, or workplace decisions.
2024: associations with rape-myth acceptance and sexual harassment
In an open-access study, Harsey, Adams-Clark, and Freyd (2024) examined self-reported DARVO use, rape-myth acceptance, and sexual-harassment perpetration in 602 university students and then analyzed a community sample of 335 adults recruited through MTurk. The study found positive associations between DARVO use and both rape-myth acceptance and sexual harassment. These were correlational findings. They do not establish that DARVO causes sexual harassment, that every person who uses DARVO endorses rape myths, or that DARVO can diagnose a propensity for sexual violence.
2026: a shorter DARVO measure and associations with posttrauma symptoms
Durland, Harsey, and Freyd (2026) developed and evaluated an 18-item short form of the DARVO Experiences Questionnaire using a Prolific sample of 319 adults and a public-university sample of 261 participants. The short form showed high internal reliability and correlated strongly with the longer measure. In statistical analyses, greater DARVO exposure was associated with more posttrauma symptoms after accounting for trauma history, and exposure was higher in confrontations involving emotional or psychological mistreatment. The study is useful for measurement development, but its design remains observational: an association with symptoms does not prove that DARVO alone caused those symptoms.
A 2026 study by Keidar and colleagues extended the literature to 160 Israeli adult women with histories of childhood sexual abuse, examining DARVO exposure in relation to later sexual revictimization. The results suggested associations with some forms of revictimization beyond PTSD symptoms, but the sample and research question are highly specific. This is best treated as emerging evidence rather than a basis for general claims about all people exposed to DARVO.
How Strong Is the Evidence for DARVO?
The scientific status is best summarized as growing, meaningful, and still limited. DARVO is established enough to have a peer-reviewed conceptual history, multiple direct empirical studies, experimental manipulations, observational associations, and a recently validated shorter questionnaire. The observer-perception effect has been demonstrated in more than one experiment, which gives stronger support than a purely anecdotal concept would have.
At the same time, the literature is not yet a broad, independently replicated clinical science. A substantial share of the published work comes from overlapping research teams; several studies rely on undergraduates, online convenience samples, retrospective reporting, or vignettes; and the construct has been studied especially often in sexual-violence contexts. The field has less evidence on everyday workplace conflict, family disputes, healthcare interactions, cross-cultural variation, developmental trajectories, long-term outcomes, and how DARVO differs psychometrically from neighboring defensive or manipulative responses.
That combination calls for precision rather than either dismissal or overstatement. DARVO is more than an internet buzzword: it has an empirical literature. It is also less than a diagnostic or forensic instrument. The strongest use of the concept is to describe and study a particular response pattern and its effects, while keeping disputed facts, intent, diagnosis, and broader abuse dynamics as separate questions.
Why Can DARVO Feel So Disorienting?
DARVO reorganizes the conversation. The person who raised a concern may suddenly have to defend their memory, motives, character, tone, or morality. That shift increases cognitive load: instead of discussing one event, they are now answering multiple counteraccusations while also deciding whether the original topic is still reachable. The emotional center of gravity moves from the alleged harm to the accused person’s grievance about being confronted.
The 2017 finding linking greater DARVO exposure with greater self-blame offers one empirically supported piece of this picture. A person who began by asking for accountability may end the conversation wondering whether they were cruel, disloyal, unstable, or responsible for the other person’s distress. The research does not show that every person exposed to DARVO will react this way, nor that self-blame always means the underlying allegation was accurate. It shows an association between the response pattern and how confronters evaluated themselves in that study.
The observer experiments add another layer. DARVO can change not only the interaction between two people but also the impressions of third parties. The 2020 experiments and the 2023 experiment found that observers exposed to DARVO were more likely to shift judgments about credibility, responsibility, or abusiveness. This helps explain why role reversal may matter socially: once the story becomes “I am the real victim of this accusation,” bystanders are asked to evaluate two competing narratives rather than only the original concern.
Does DARVO Have to Be Deliberate?
Research papers often describe DARVO as a tactic, strategy, or defensive response, but an observer usually cannot know another person’s internal intention with certainty. Someone may deliberately deploy a familiar pattern to evade accountability; another person may react automatically from shame, fear, anger, defensiveness, or learned conflict habits. The observable sequence can be described without pretending to know which internal process produced it.
Intent and impact should therefore be separated. A response can derail accountability and harm the other person even if it was not planned in advance. Conversely, a harsh or defensive response should not automatically be classified as manipulation simply because it was upsetting. The useful questions are what was said or done, how the topic changed, whether the original concern could still be examined, whether counterclaims had independent substance, and whether the pattern is repeated.
Can a Single Argument Be DARVO?
A single exchange can contain all three DARVO elements, so it can be reasonable to describe the exchange as DARVO-like. That is different from concluding that a person has a stable “DARVO personality,” that the relationship is abusive, or that every later disagreement belongs to the same pattern. A relationship-level claim requires relationship-level evidence.
Individual components also matter. A person may deny without attacking, attack without reversing roles, or complain about being harmed without denying the original conduct. The acronym is most informative when the combined structure is visible. In the 2017 study, the components were positively correlated, which supported studying DARVO as a unitary pattern, but the existence of research subscales also reminds us that the components can vary in degree.
Can DARVO Happen Outside Romantic Relationships?
Yes. The concept originated in work on abuse and sexual misconduct, and much of the empirical literature still centers interpersonal violence. But the 2017 study asked about a range of wrongdoings, including less severe interpersonal transgressions, and the 2024 study measured DARVO use in confrontations involving diverse wrongdoing. The pattern can therefore be studied in families, friendships, workplaces, organizations, and other accountability contexts.
Context changes what is at stake. In a friendship, DARVO may derail a conversation about betrayal or disrespect. In a workplace, it can affect professional reputation, complaint handling, or team dynamics. In institutions, Freyd uses the term “institutional DARVO” for cases in which an organization participates in denial, attack, and role reversal. That extension is conceptually influential, but it should not be treated as a legal finding or a standardized institutional diagnosis. Formal rights, reporting duties, harassment standards, and legal definitions vary by jurisdiction and require their own authoritative sources.
DARVO and Ordinary Conflict: Where the Boundary Matters
Ordinary conflict can be defensive, emotional, inaccurate, and messy. People interrupt, misremember, minimize, become angry, countercomplain, and say unfair things. None of those behaviors, in isolation, proves DARVO or abuse. A countercomplaint can also be legitimate: one person can raise a genuine concern and the other can have a genuine concern of their own. Mutual harm is possible.
The stronger DARVO pattern appears when the response systematically displaces the original accountability question through denial, credibility attack, and reversal. Even then, the label should describe the interaction rather than decide every disputed fact. This is especially important online, where fragments of text or one-sided retellings can make complex conflicts look more certain than they are.
A useful boundary test is whether the original issue can remain examinable. In a good-faith disagreement, people may strongly dispute facts but still identify what evidence would matter, acknowledge uncertainty, distinguish intent from impact, or return to the specific event. In a DARVO-like exchange, the conversation may become organized around discrediting the confronter and establishing the confronted person as the primary victim, leaving the original issue unreachable.
Practical Implications: What to Do When You Notice a DARVO-Like Pattern
Recognizing the pattern can help you keep the structure of the conversation visible. It does not give you a universal script, and it does not guarantee that naming “DARVO” aloud will improve the interaction. The 2020 education experiment suggests that learning about DARVO can reduce some effects on observers, but that finding is about prior education in a controlled study, not about a confrontation technique that has been proven to work in every relationship.
If the situation is safe enough for continued discussion, one practical move is to separate the original issue from the counterattack. Instead of defending every new allegation at once, you can identify the topic you raised and ask whether it can be addressed on its own terms. A statement such as “I hear that you have a concern about my behavior too. I am asking first about what happened yesterday; we can address your concern separately” may help preserve structure. It is an option, not a prescription.
When the stakes are high, documentation is often more useful than argument. Write down the original concern, the response, dates, relevant messages or records, what was denied, what attacks were made, what counterclaims were introduced, and what happened afterward. In a formal setting, concrete chronology is usually clearer than leading with psychological labels. A factual account lets other people evaluate conduct without requiring them to accept your interpretation of someone’s motives.
Outside perspective can also help when you have become trapped in recursive argument. A trusted person, clinician, supervisor, advocate, mediator, or other appropriate professional may help distinguish the original issue from the counterclaims. The right source of support depends on context. A relationship conversation, workplace complaint, healthcare dispute, and safety-threatening situation require different forms of expertise.
If the interaction involves threats, stalking, violence, intimidation, severe coercion, or fear about what will happen if you raise concerns, communication technique is no longer the central problem. Safety planning and specialized local support take priority over trying to phrase the confrontation perfectly. A boundary statement or assertive script can clarify your position; it cannot by itself make a coercive person stop using coercion.
How to Document a DARVO-Like Exchange Without Diagnosing Anyone
A neutral record can preserve detail while avoiding unsupported conclusions. Start with the event you were trying to discuss. Then record the response in sequence: what was denied or minimized, what was said about you, and how the roles were reframed. Keep direct quotations when you have them, distinguish your interpretation from observable facts, and note evidence that supports or challenges both sides.
For example: “I raised concern X about event Y. The response was that Y did not happen; then I was called dishonest and vindictive; then the other person said my complaint was an attack designed to ruin them.” This is more informative than “They are a narcissistic gaslighter who DARVOs.” The first version can be checked against records and context. The second bundles several psychological labels and conclusions that the available interaction may not support.
What DARVO Research Does Not Yet Establish
The current literature does not establish a population prevalence rate for DARVO across cultures and relationship types. It does not provide a validated cutoff that turns a person into a “DARVO user” in a clinical or forensic sense. It does not show that DARVO uniquely identifies guilty people. It does not establish that every instance is consciously planned. It does not justify inferring a personality disorder. And although newer work associates DARVO exposure with posttrauma symptoms, it does not show that DARVO alone causes PTSD or another clinical disorder.
Those limitations do not make the concept useless. They identify the level at which it is strongest. DARVO is a framework for noticing a distinctive accountability response and a research construct for studying its effects. Its scientific value increases when it is used precisely enough that denial can still be denial, disagreement can still be disagreement, gaslighting can still be gaslighting, and coercive control can still be analyzed as a broader pattern rather than folded into one acronym.
Frequently Asked Questions
What does DARVO stand for?
DARVO stands for Deny, Attack, and Reverse Victim and Offender. It describes a response in which someone confronted about alleged wrongdoing denies or minimizes the behavior, attacks the person raising the concern, and reverses the roles so that the confronted person presents themselves as the victim and the confronter as the offender.
Who coined the term DARVO?
Psychologist Jennifer J. Freyd introduced DARVO in a 1997 peer-reviewed article on power, betrayal, and responses to accountability. Direct empirical research on DARVO as a combined construct followed later, beginning with Harsey, Zurbriggen, and Freyd (2017).
Is DARVO a diagnosis?
No. DARVO is not a DSM or ICD diagnosis. It is a descriptive and research term for a response pattern. It should not be used to diagnose narcissistic personality disorder, psychopathy, another mental disorder, or a fixed personality type.
Is DARVO the same as gaslighting?
No. Gaslighting concerns undermining another person’s confidence in their perception or understanding of reality; DARVO specifically describes denial, attack, and reversal of victim and offender roles in response to confrontation. They can overlap, especially when denial attacks the other person’s memory or perception, but neither concept contains the other completely. A 2026 review of gaslighting supports keeping gaslighting conceptually distinct from related forms of manipulation and disagreement.
Does DARVO prove that someone is guilty?
No. DARVO is not a truth test. Freyd explicitly cautions that even if certain responses are statistically associated with guilt in future research, the use of DARVO cannot prove a specific past action. Facts about the underlying event require independent evaluation.
Can someone falsely accused use DARVO?
Yes. An innocent person can deny an accusation and may also react angrily, attack the accuser, or describe themselves as harmed by the accusation. Those reactions can still be evaluated as present behavior, but they do not establish what happened earlier. This is why DARVO should be used as a description of interaction, not as a shortcut to a verdict.
Do all three parts have to be present?
The acronym names a combined pattern, and the three components were positively related in the first direct empirical study. Individual elements can occur separately, so a denial alone or an attack alone is better described by its own name. When only part of the sequence is visible, “DARVO-like” or a direct description of the behavior is often more precise than treating the full acronym as automatically established.
Can DARVO happen in normal conflict?
DARVO-like responses can appear in conflicts that are not part of an abusive relationship. The 2017 research included a range of interpersonal wrongdoings, from social transgressions to serious abuse. One episode therefore does not establish abuse, coercive control, or a stable relationship pattern. Frequency, severity, context, power, consequences, and the broader relationship all matter.
Does DARVO always involve conscious manipulation?
The research literature often describes DARVO as a tactic or strategy, but a specific person’s conscious intention usually cannot be inferred from behavior alone. The pattern can be analyzed without claiming to know whether it was planned, habitual, emotionally defensive, or deliberately manipulative.
Can learning about DARVO help?
There is preliminary experimental evidence that education can reduce some of DARVO’s effects on observer judgments. In Harsey and Freyd’s 2020 experiment, participants who learned about DARVO before evaluating a scenario judged the victim more favorably on some measures and the alleged perpetrator as less believable. That finding supports education as a promising anti-DARVO approach, while leaving open how best to apply it across real-world contexts.
What is the best response to DARVO?
There is no single evidence-based script that fits every situation. When it is safe to continue, keeping the original issue separate from counterattacks, documenting concrete facts, and seeking appropriate outside perspective can help. When intimidation, threats, violence, stalking, or coercive control are present, safety planning and specialized support take priority over winning the argument.
Related Articles
References
Durland, M. L., Harsey, S. J., & Freyd, J. J. (2026). Assessing perpetrator responses to confrontation: Associations with a DARVO-SF and posttrauma symptoms in two different populations. Journal of Trauma & Dissociation, 27(2), 221–237. https://doi.org/10.1080/15299732.2025.2599757
Freyd, J. J. (1997). II. Violations of power, adaptive blindness and betrayal trauma theory. Feminism & Psychology, 7(1), 22–32. https://doi.org/10.1177/0959353597071004
Freyd, J. J. (n.d.). DARVO. Jennifer Joy Freyd, PhD. https://www.jjfreyd.com/darvo
Harsey, S. J., Adams-Clark, A. A., & Freyd, J. J. (2024). Associations between defensive victim-blaming responses (DARVO), rape myth acceptance, and sexual harassment. PLOS ONE, 19(12), e0313642. https://doi.org/10.1371/journal.pone.0313642
Harsey, S. J., & Freyd, J. J. (2020). Deny, Attack, and Reverse Victim and Offender (DARVO): What is the influence on perceived perpetrator and victim credibility? Journal of Aggression, Maltreatment & Trauma, 29(8), 897–916. https://doi.org/10.1080/10926771.2020.1774695
Harsey, S. J., & Freyd, J. J. (2023). The influence of Deny, Attack, Reverse Victim and Offender and insincere apologies on perceptions of sexual assault. Journal of Interpersonal Violence, 38(17–18), 9985–10008. https://doi.org/10.1177/08862605231169751
Harsey, S. J., Zurbriggen, E. L., & Freyd, J. J. (2017). Perpetrator responses to victim confrontation: DARVO and victim self-blame. Journal of Aggression, Maltreatment & Trauma, 26(6), 644–663. https://doi.org/10.1080/10926771.2017.1320777
Keidar, A., Kidra, N. A., Gafter, L., & Lahav, Y. (2026). DARVO and sexual revictimization among childhood sexual abuse survivors: Interpersonal dimensions beyond trauma symptoms. Journal of Interpersonal Violence. Advance online publication. https://doi.org/10.1177/08862605261432542
Klein, W., Wood, S., Forget, A.-A., & Bartz, J. A. (2026). A historical review of gaslighting: Tracing changing conceptualizations within psychiatry and psychology. Clinical Psychology Review, 126, 102742. https://doi.org/10.1016/j.cpr.2026.102742
Rosenthal, M. N., & Freyd, J. J. (2022). From DARVO to distress: College women’s contact with their perpetrators after sexual assault. Journal of Aggression, Maltreatment & Trauma, 31(4), 459–477. https://doi.org/10.1080/10926771.2022.2055512