Even if everything the plaintiffs claimed was true, the motion to dismiss argued, a therapist owed no duty to a third party at risk from his patient. Tort law had long recognized that “negligence is not actionable unless it involves the invasion of a legally protected interest, the violation of a right. Proof of negligence in the air, so to speak, will not do.”11 Applying well-settled law (an application of the “no duty to rescue” rule), the trial court held that the psychotherapist did not owe a duty to a third party and dismissed the plaintiffs’ claim without trial. The case then wound its way to the Supreme Court of California. After the unusual step of withdrawing its first published effort to sort this out, in 1976 the Supreme Court of California issued its second opinion, which recognized a duty to protect third-party nonpatients under certain circumstances. “When a therapist determines, or pursuant to the standards of his profession should determine, that his patient presents a serious danger of violence to another, he incurs an obligation to use reasonable care to protect the intended victim against such danger.”12
It then remanded the case to the trial court to determine the facts of the 1969 allegations and to apply the 1976 newly recognized duty. However, the case settled before the trial took place. This precipitated a boisterous debate and a 30-year, 50-arm, social science experiment. “It is one of the happy incidents of the federal system that a single courageous state may, if its citizens choose, serve as a laboratory, and try novel social and economic experiments without risk to the rest of the country.”13 Tarasoff dealt with issues that have traditionally been the province of the states; each was free to fashion its own solution and to learn from the efforts of others (or not).
The majority opinion in Tarasoff is at once quite remarkable and rather unremarkable. In our postmodern world, few of us live in a splendid isolation. Our lives are interdependent and, save for instances of willful ignorance, we are cognizant that what we do invariably affects others. “Our current crowded and computerized society compels the interdependence of its members.”14 Living in such proximity entitles society to expect each of us to do more than simply keep our promises. Contract law may turn on relationships and arm’s-length bargaining, but negligence law developed around foreseeability. As the famous Palsgraf opinion noted, “The risk reasonably to be perceived defines the duty to be obeyed, and risk imports relation; it is risk to another or to others within the range of apprehension.”15 Tarasoff fits, unremarkably, in the mainstream of American tort law.
Tarasoff is also quite remarkable in its inconsistency. The therapist-patient relationship is a private relationship sanctioned by the state. The Supreme Court of California had been as sensitive to privacy issues as any state high court. The duty recognized permits a third party to question the conduct of a therapist who was not then permitted to acknowledge that he was treating Poddar, the patient who allegedly divulged to his therapist his intention to harm Tatiana Tarasoff. At the time of the psychotherapist’s questioned conduct, no Tarasoff duty had been recognized. Remarkably, the court is untroubled by a retroactive application of this new duty, finding that a psychotherapist may be liable for doing what the law then required him to do—keep patient confidences. In an opinion by a court known for judicial craftsmanship, this opinion is remarkable for circular reasoning. A good example is the court’s use of the therapist-patient relationship as a special relationship for recognition of duty:
[W]hen the avoidance of foreseeable harm requires a defendant to control the conduct of another person, or to warn of such conduct, the common law has traditionally imposed liability only if the defendant bears some special relationship to the dangerous person or to the potential victim. Since the relationship between a therapist and his patient satisfies this requirement, we need not here decide whether foreseeability alone is sufficient to create a duty to exercise reasonable care to protect a potential victim of another’s conduct. (Tarasoff at 435)
The neutral principle
Tarasoff addressed personal violence issues that have traditionally been left to the states, setting up a natural social science experiment. As if on cue from an offstage Busby Berkeley–like director, “OK kids, walk in pairs, first sunbelt, then rust belt, now east, now west . . . ,” Florida, Illinois, Virginia, and Texas opted to reject Tarasoff; Arizona, Michigan, New Jersey, and California opted to accept it. Most chose legislation that recognized a duty but raised the bar for triggering it, eg, threat to a reasonably identifiable victim, and among other things clarified what would satisfy that duty, eg, notify police or victim (Table). Correlatively, some decisions went beyond Tarasoff and lowered the bar, imposing liability even though the target was not directly threatened16; others did not.17
Psychiatrists and psychologists feared that 20-20 hindsight would impose liability even if there was no science to support that risk assessment. But, if the reported cases are an accurate reflection of claim resolution, liability has principally been imposed in cases where the information about the risk was known and perceived as credible but ignored, eg, nurses notes documented minor son’s threats but not communicated to parents he later murdered; duty to notify of revelation that a psychiatric resident was a pedophile.18-20 Allegations that the defendants should have foreseen the risk are less likely to be recognized as viable unless couched in a specific act or threat, eg, uphold-ing dismissal of complaint that alleged that attack on plaintiff should have been foreseen because involuntary patients have tended to be more violent.21
The American Psychiatric Association’s amicus brief, which captured mainstream psychiatric thinking, had predicted that a loss of confidentiality would follow recognition of a duty to protect a third party, which “will deter violence-prone persons from seeking therapy, and hamper the treatment of other patients.”22 The court saw things differently:
Our current crowded and computerized society compels the interdependence of its members. In this risk-infested society we can hardly tolerate the further exposure to danger that would result from a concealed knowledge of the therapist that his patient was lethal. If the exercise of reasonable care to protect the threatened victim requires the thera-pist to warn the endangered party or those who can reasonably be expected to notify him, we see no sufficient societal interest that would protect and justify concealment. The containment of such risks lies in the public interest.23
Does the 30-year social science experiment provide support for either Tarasoff’s proponents or opponents? It is difficult to imagine what might be the measure of success. Since most violent acts are committed by persons who are not mentally ill, crime statistics and other broad measures of societal violence are not to the point. I do not know what is. And so, Tarasoff remains part of the legal landscape, tolerated but not embraced.