The Tarasoff case is the landmark 1976 California Supreme Court ruling that established a mental health professional’s legal duty to protect identifiable third parties from serious harm threatened by a patient. The case, formally *Tarasoff v. Regents of the University of California*, arose from the 1969 murder of Tatiana Tarasoff by Prosenjit Poddar, a graduate student who had confided his intent to kill her to a university psychologist. The psychologist alerted campus police but never warned Tarasoff or her family. Nearly five decades later, the ruling’s core tension between patient confidentiality and public safety continues to shape clinical practice, ethics training, and state law across the country.
What Happened
In the fall of 1968, Prosenjit Poddar, a graduate student at the University of California, Berkeley, became romantically fixated on Tatiana Tarasoff, a fellow student. After Tarasoff rebuffed him, Poddar’s emotional state deteriorated sharply. He sought treatment at the university’s counseling center, where he told his therapist, Dr. Lawrence Moore, that he intended to kill Tarasoff, who was then traveling abroad in Brazil.
Moore took the threat seriously. He notified campus police both verbally and in writing, requesting that Poddar be detained. Campus officers briefly held Poddar but released him after he appeared rational and promised to stay away from Tarasoff. Moore’s supervisor, Dr. Harvey Powelson, then directed that no further action be taken and ordered Moore’s therapy notes destroyed. No one contacted Tarasoff or her family. When Tarasoff returned from Brazil in October 1969, Poddar went to her home and killed her.
Tarasoff’s parents sued the university, the therapists, and the campus police, alleging that they had failed to warn Tatiana of the danger. The case wound through the courts for years before reaching the California Supreme Court, which issued two separate opinions that reshaped mental health law.
Tarasoff I and Tarasoff II
The California Supreme Court first ruled on the case in 1974, in what is commonly called Tarasoff I. That opinion held that a therapist who determines, or should determine, that a patient presents a serious danger to another person bears a duty to warn the intended victim. The court framed the obligation narrowly: you must warn the person at risk.
The decision drew intense criticism from the mental health community, which argued that a rigid warning requirement would discourage patients from being honest with their therapists and ultimately make the public less safe, not more. The court took the unusual step of rehearing the case, and in 1976 issued its revised and more famous ruling, Tarasoff II. This time the court broadened the obligation from a duty to warn to a duty to protect. Justice Mathew Tobriner, who wrote both opinions, declared that therapists must “take reasonable steps to protect potential victims,” which could include warning the victim, notifying law enforcement, or pursuing civil commitment of the patient. The phrase that became the decision’s shorthand: “the protective privilege ends where the public peril begins.”1PubMed Central. The Ethical Dilemma of Balancing Confidentiality and Duty to Protect: A Case Report of Comorbid Schizophrenia and Cannabis Use Disorder with Homicidal Thoughts
The shift from “warn” to “protect” was more than semantic. By giving clinicians a range of protective options rather than a single mandate to call the victim, the court acknowledged that clinical situations vary. In some cases, warning the potential victim is the best move. In others, hospitalizing the patient or intensifying treatment might be more effective. The broader standard gave therapists some flexibility while still holding them accountable for doing something.
Duty to Warn Versus Duty to Protect
Before the Tarasoff decisions, two distinct legal threads existed: a duty to warn (to inform someone of danger) and a duty to control (to prevent a dangerous person from acting). These operated on separate legal footings and required different things from professionals. The Tarasoff ruling effectively merged them under a single “duty to protect” umbrella. For much of the following decades, courts treated warnings as just one method of fulfilling this broader obligation. More recently, some courts have begun separating the two duties again, recognizing that each requires different legal standards to trigger.2PubMed Central. To warn and to control: two distinct legal obligations or variations of a single duty to protect?
This distinction matters in practice. A duty to warn typically requires only that the therapist notify the identifiable victim or law enforcement. A duty to control can involve taking more direct action, such as seeking emergency psychiatric hospitalization for the patient. Some states specify one, some the other, and some combine them. A therapist crossing state lines or practicing via telehealth in multiple jurisdictions needs to know which version applies where they are licensed, because getting it wrong could mean liability on one side or a confidentiality violation on the other.
A Patchwork of State Laws
There is no federal Tarasoff law. What exists is a patchwork of state statutes and court decisions that vary widely. A review of all fifty states and the District of Columbia found they fall into roughly four categories: states that mandate a duty to warn or protect by statute (about 23 states); states where the duty exists through court rulings but has not been formally written into law (about 10); states that neither mandate nor prohibit a breach of confidentiality when threats are present but permit it (about 11); and states with essentially no guidance on the issue at all (about 6).3FOCUS The Journal of Lifelong Learning in Psychiatry. The Tarasoff Rule: The Implications of Interstate Variation and Gaps in Professional Training
The practical consequences of this patchwork are real. In a mandatory-duty state, a therapist who fails to act on a credible threat can be sued for negligence. In a permissive state, the therapist has legal cover to breach confidentiality but is not required to. In a state with no guidance, a therapist facing a threatening patient is essentially flying blind, relying on ethical codes and professional judgment without clear legal backing for whatever decision they make. The same clinical scenario, the same threatening words from a patient, can create vastly different legal obligations depending on geography.
Some states have also tailored their laws in ways that depart from the original Tarasoff framework. New Jersey, for example, specifies that licensed psychologists, psychiatrists, physicians, nurses, clinical social workers, and marriage counselors must report when a patient communicates “a threat of imminent, serious physical violence against a readily identifiable individual.”4Elsevier. Continuing Education Involuntary Civil Commitment of Minors Other states use broader language or apply the duty to a narrower range of practitioners. Some only require action when the threat is against an identifiable person; others extend the duty to threats against property or unidentified groups.
The Problem of Predicting Violence
At the heart of the Tarasoff obligation is an assumption that therapists can tell when a patient is actually dangerous. Research consistently suggests this is far harder than it sounds. Studies have shown that psychiatrists and psychiatric nurses struggle to accurately predict which patients will become violent, even in controlled inpatient settings where they have extensive contact with the patient.5PubMed. Predictions made by psychiatrists and psychiatric nurses of violence by patients
Threats are common in clinical settings. Many patients who express violent ideation never follow through. Meanwhile, some of the most dangerous patients never telegraph their intentions at all. Clinicians regularly face situations where a patient says something alarming but the actual risk of violence is ambiguous. The Tarasoff framework forces a binary decision—act or don’t act—on what is often a deeply uncertain clinical picture. Overreacting leads to unnecessary breaches of confidentiality, damaged therapeutic relationships, and potential violations of patient rights. Underreacting can lead to tragedy.
Critics have argued that current case law places an unreasonable expectation on psychotherapists to protect all at-risk parties or to foresee that a patient will follow through on threats. This expectation, some scholars contend, has “largely failed to guarantee psychotherapy patients unlimited access to care, while potentially inhibiting future honest communication between patients and health professionals and endangering the safety of others.”6PubMed. Patient confidentiality, the duty to protect, and psychotherapeutic care: perspectives from the philosophy of ubuntu In other words, the rule designed to make people safer may, in some scenarios, make them less safe by discouraging patients from disclosing dangerous thoughts.
What Happens After a Warning Is Made
One aspect of the Tarasoff framework that rarely gets discussed is what actually happens once a therapist picks up the phone and calls the police. A follow-up survey of police officers who had received Tarasoff-type warnings found that the vast majority, about 96%, took at least one concrete action in response. The most common step was notifying other officers about the threat. Over half of the officers surveyed said they would take action to remove a firearm from the threatening individual if one was present.7PubMed Central. A 20-year follow-up survey of police officers’ experience with Tarasoff warnings: How law enforcement reacts to clinicians’ duty to protect
That high response rate is reassuring, but the interaction between mental health professionals and law enforcement is not always smooth. Therapists and police officers operate under different frameworks, different training, and different legal authorities. A therapist’s clinical concern may not translate easily into grounds for police action, especially if the patient has not committed a crime. Police may feel ill-equipped to assess the seriousness of a psychiatric threat, while therapists may feel uncertain about what information they can legally share. These handoff problems are a known gap in the system, and they echo the original Tarasoff case itself, where campus police did detain Poddar but released him after a brief evaluation.
Effects on the Therapeutic Relationship
Confidentiality is not just a legal technicality; it is the foundation that allows therapy to work. Patients disclose their most disturbing, shameful, and socially unacceptable thoughts to their therapists because they trust that those disclosures stay in the room. The Tarasoff duty introduces a limit to that trust. Therapists in most states are required to inform patients at the outset of treatment about the limits of confidentiality, including the duty to warn or protect. But knowing that your therapist might call the police if you say the wrong thing inevitably changes the conversation.
The concern is not theoretical. Some clinicians and ethicists worry that the duty to protect creates a chilling effect, discouraging patients with violent ideation from seeking or continuing treatment. A patient who is thinking about hurting someone is arguably the patient who most needs therapy, and who most needs to be able to talk honestly about those thoughts. If that patient knows that a specific threat could trigger a warning to the intended victim or police involvement, they may self-censor, avoid treatment entirely, or lie to their therapist. The very mechanism designed to prevent violence could, paradoxically, remove the professional best positioned to intervene.
Therapists manage this tension daily. Many clinicians develop careful language around informed consent, explaining the limits of confidentiality in a way that preserves therapeutic trust while fulfilling legal obligations. Some frame it as collaborative: “If you tell me something that makes me worried about your safety or someone else’s, I’ll talk with you about what we need to do.” The goal is to avoid making the therapy room feel like a surveillance environment while still meeting the standard the law requires.
When the Duty to Protect Extends Beyond Mental Health
The logic behind Tarasoff—that a professional with knowledge of a foreseeable harm has an obligation to act—has not stayed confined to psychotherapy. Courts and ethicists have explored whether similar duties apply in other medical contexts. One prominent area is genetics: if a physician discovers that a patient carries a hereditary disease risk that could affect the patient’s relatives, does the physician have a duty to warn those family members?
The question has generated significant debate. Some courts have recognized that a duty to warn relatives about risks from certain infectious agents already exists through state and local health agencies, and that the broader principle of breaching confidentiality to warn of imminent harm has been established in case law.8PubMed. The “duty to warn” a patient’s family members about hereditary disease risks Hereditary disease is trickier than a direct violent threat, though. The “harm” is not imminent in the same way, the relatives may not want to know, and the patient’s right to control their own genetic information is a distinct privacy interest. Nevertheless, the Tarasoff principle has served as a conceptual anchor for these discussions, pushing courts and medical ethics boards to weigh confidentiality against the potential for preventable harm in contexts far removed from the therapist’s office.
Common Misconceptions About the Ruling
Several misunderstandings about Tarasoff persist, even among people who have heard of the case. One is that Tarasoff created a universal, nationwide obligation. It did not. The ruling is a California Supreme Court decision, and while its reasoning has been enormously influential, each state has adopted its own version of the duty (or declined to adopt one at all). A therapist in Texas operates under different rules than one in New York.
Another misconception is that the duty applies to any patient who expresses anger or hostility. The Tarasoff framework generally requires a specific and credible threat against an identifiable person. A patient who says “I’m so angry I could kill someone” during a session is not the same as a patient who says “I am going to kill my neighbor when I leave here today.” The former is a clinical concern worth exploring; the latter may trigger a legal obligation. Where exactly the line falls is one of the hardest judgment calls in mental health practice, and it is precisely the kind of question that makes violence prediction so difficult.
A third misunderstanding is that the therapist’s only option is to call the potential victim. As the Tarasoff II ruling made clear, warning the victim is one of several possible protective actions. Depending on the clinical situation and the applicable state law, a therapist might contact law enforcement, adjust the patient’s treatment plan, increase session frequency, seek voluntary or involuntary hospitalization, or take other steps reasonably calculated to prevent harm. The ruling was deliberately flexible on the “how” of protection, even as it was firm on the “whether.”
Telehealth and Jurisdictional Complications
The rise of telehealth has added a new layer of complexity to Tarasoff obligations. When a therapist in one state treats a patient in another state via video call, which state’s law applies? If the patient threatens a person in a third state, does the therapist have to determine the law there as well? These questions were barely imaginable when the Tarasoff ruling was handed down in 1976, and the law has not fully caught up.
Most licensing boards default to the rule that the law of the state where the patient is physically located at the time of the session governs the encounter. But this creates practical absurdities. A patient who moves or travels across state lines during the course of therapy could shift the applicable legal framework from session to session. A therapist practicing via telehealth in multiple states may be simultaneously subject to mandatory-duty rules in one jurisdiction and permissive-duty rules in another. The lack of federal guidance means that each clinician, often without formal legal training, must navigate this on their own or with the help of professional liability consultants.
Professional training has not kept pace with these challenges. The same review that mapped the state-by-state variation found significant gaps in how clinicians are educated about their Tarasoff-related obligations, particularly regarding differences across jurisdictions.9FOCUS The Journal of Lifelong Learning in Psychiatry. The Tarasoff Rule: The Implications of Interstate Variation and Gaps in Professional Training Many training programs teach Tarasoff as a single rule rather than a fragmented landscape, leaving new clinicians unprepared for the reality of practice across state lines.
How the Case Changed Informed Consent
Before Tarasoff, informed consent in psychotherapy focused mostly on treatment procedures. After Tarasoff, it became standard practice to discuss confidentiality and its limits at the start of therapy. Today, virtually every intake form for mental health treatment includes language about the circumstances under which a therapist may be required to break confidentiality: danger to self, danger to others, and suspected child or elder abuse.
This shift has been broadly positive for transparency, even if it complicates the therapeutic relationship. Patients now enter treatment with at least a formal understanding that confidentiality is not absolute. Whether patients truly absorb this information is another question. Research on informed consent in other medical contexts suggests that patients often sign forms without fully processing their contents. A patient in acute distress, seeking help for the first time, may not register the legal caveats buried in an intake packet. Therapists who take the time to discuss these limits verbally, in plain terms, give their patients a genuine opportunity to understand the boundaries of the relationship, which serves everyone better when a crisis does arise.