Abstract
Professionals must at times make snap judgments that have profound consequences. Does a doctor perform an otherwise forbidden abortion to preserve a patient’s failing health? Does a police officer fire at a suspect pointing an unidentified metal object? The criminal law tells these professionals: Don’t intervene unless the danger is serious and the risk is imminent. But it offers little to guide that determination beyond: Be reasonable. Yet the “right” choice—the one that’s legally safe—often resembles the “wrong” one: a felony punishable by decades in prison. These ambiguous regimes operate as liability cliffs that professionals are forced to traverse at a rapid pace. Professional cliff running is time sensitive and high stakes. It’s also been overlooked by courts and commentators alike. We introduce this neglected phenomenon, trace its surprising operation in socially salient contexts, and examine its far-reaching significance for the law and for democracy.
Professional liability cliffs arise at the fault line of two enduring values that often conflict: deference to expertise and accountability for misconduct. Maintaining equilibrium between these values is vital. Too much deference emboldens professionals to act hastily and violate the law. Too much accountability chills them from meeting the moment under fast-changing circumstances. The reasonableness doctrines that govern cliff running for police officers and abortion providers look almost identical on the books. Yet on the ground, the law strikes starkly different balances between deference and accountability. The divergence has to do with trust between professionals and prosecutors. The confidence that officers and prosecutors tend to have in each other makes cliff running relatively safe for police but treacherous for the victims of their on-duty violence, disproportionately Black and Latino men. Meanwhile, mutual suspicion between doctors and prosecutors predictably chills emergency abortion care, imperiling patients who rely on it, especially low-income women of color who lack insurance and other resources to travel for treatment. These dynamics motivate promising and practical doctrinal and institutional reforms.
“[It’s] like a sword of Damocles hanging over us. If we do something with a good intention, but it turns out to be illegal, the consequences are extremely serious for medical practitioners.”
—Doctor Peter Boylan
“If anybody wants to fight or run, I’m a little trigger-happy, guys. I’m not gonna lie, and I get paid a ton of money in overtime, if I had to shoot somebody.”
—Officer Stephen Barone
A doctor and a police officer must make split-second decisions. Another person’s fate lies in their hands—so does their own. If a prosecutor determines that they chose wrong, they could spend decades in prison.
- In Texas, an obstetrician-gynecologist performs an abortion on a patient whose complicated pregnancy could threaten her life. Texas makes it a felony to provide an abortion except where necessary to treat “serious risk of [death or] substantial impairment of a major bodily function.”[3] If a prosecutor thinks that the physician didn’t apply “reasonable medical judgment” to evaluate the patient’s condition, the doctor faces trial and, if convicted, a prison sentence of up to ninety-nine years.[4]
- In California, a cop shoots and kills a suspect who reached into his waistband after she ordered him to stop. Intentional killing is forbidden except (as relevant here) to “defend against an imminent threat of death or serious bodily injury.”[5] If a prosecutor thinks that the officer didn’t “reasonably believe” that the force was necessary, the cop faces trial and, if convicted, a prison sentence of fifteen years to life.[6]
Policing and obstetrics are strikingly different professions. Yet the Texas doctor and California cop find themselves in situations with almost identical legal structures. Each confronts a high-stakes emergency under the threat of enormous criminal punishment if they get it wrong, and only a vague legal standard to guide their quick decisions.
These professionals run on the edges of “professional liability cliffs.”[7] Liability cliffs exist where small differences in conduct or circumstances result in radically disparate legal outcomes.[8] Scholars have analyzed liability cliffs in many areas of law, from torts to contracts to property, but their use as a form of professional regulation has gone unnoticed in the literature.[9] Yet that is exactly how the legal system regulates professional actions under these circumstances: It forbids the doctor from performing an abortion to preserve a patient’s health if her life’s not at risk, and forbids the officer from shooting a suspect who’s acting erratically and disobeying orders but not threatening anyone’s life.[10] Still, they’re expected to save patients or bystanders whose lives are sufficiently threatened.
Doctors and officers appear to be in very similar legal boats. And yet the law treats them very differently. Criminal charges for on-duty police violence are exceptionally rare, and convictions are rarer still.[11] Even after Derek Chauvin murdered George Floyd, prosecutors rarely second-guess the judgment of officers on the job, contributing to a “shoot first” police culture.[12] By contrast, ob-gyns in the dozen-plus states that prohibit abortion except in response to a medical emergency are understandably terrified that prosecutors will accuse them of treating a patient who wasn’t close enough to dying and charge them with a felony that carries an effective life sentence.[13] So they hesitate to perform emergency abortions that come near the legal line, often to their patients’ detriment.[14]
The cliffs that cops and doctors run on are nearly identical on paper, yet cliff running is reassuringly safe for the police and paralyzingly dangerous for doctors who perform abortions. What gives? Professional liability cliffs sit at an uneasy juncture of two regulatory values: deference to expertise and legal accountability. Professionals with specialized training might be expected to know what the right decision is more than untrained lawmakers, prosecutors, and jurors. At the same time, the law cannot relinquish domains like homicide to the judgments of individuals, or even whole professions. Best case, the liability rules that govern cliff running strike a workable balance between deference and accountability.[15]
But cliff running becomes too safe or too dangerous when liability standards tilt too far toward deference or accountability. Either can have disastrous consequences for professionals, the people they serve, the public, and even the substantive law. The details depend on the nature of the imbalance. When the cliff slopes too far toward deference, professionals are emboldened to act without consideration of the criminal law because of their (relative) impunity. That results in unjustified police shootings.[16] When the cliff slopes too far the other way, professionals are chilled from invoking valid exceptions to criminal laws. That leads to doctors withholding critical medical care from pregnant patients because they fear their own criminal exposure. It may also lead to doctors fleeing the jurisdiction, leaving healthcare deserts in their wake, where whole regions are left without reliable access to essential treatment.[17] To see how cliff running works on the ground and to know what liability rules really govern professionals, you have to understand the cliff’s balance between deference and accountability.
But why would professional liability cliffs slope to one side or the other? This Article closely examines police violence and emergency abortions to show that a critical consideration is the presence or absence of trust between prosecutors and professionals.[18] Do prosecutors trust that professionals will do their best to conform to the criminal law, or do they think they’ll try to sabotage or undermine it? Do members of the profession think that prosecutors will charge them with crimes only when warranted, or do they think they’ll bring bogus cases for political grandstanding? Trust has many sources: among the most important are institutional connections between prosecutors and the profession, their cultural and political alignment, and historical patterns of enforcement.[19] Wherever trust springs from (or fails to), well-functioning professional liability cliffs must optimize bilateral trust between prosecutors and professionals—not maximize or minimize but optimize.[20]
These dynamics call for reform when the liability standards governing professional liability cliffs tilt too much in either direction, as they do in both of our case studies. We propose revisiting doctrinal liability standards to make cliff running sensitive to background conditions of trust.[21] If reasonable person standards over-deter professionals in low-trust environments, lawmakers should consider replacing them with good faith standards.[22] Where trust levels are excessive, underdeterrence is the more serious worry, so good faith standards should be avoided.[23] Lawmakers should adopt policies responsive to the level of trust between prosecutors and professionals.[24]
Other reforms draw on distinctive strengths of professional standards and organizations. For instance, the prosecution of doctors for performing standard-of-care procedures like abortion could require a state medical board to determine as a preliminary matter that the emergency exception didn’t apply. A larger role in the criminal process for an independent and professional board could foster greater trust that prosecutions won’t tread unduly on medical expertise and judgment.
The Article proceeds in four Parts. Part I defines professional liability cliffs, spells out the reasonable person standards that putatively govern them, and describes cliff running’s dynamics and incentives from the perspective of lawmakers, prosecutors, professionals, and third parties. Part II applies and extends this framework through two in-depth case studies: doctors who perform emergency abortions (or abstain from performing them) and police officers who shoot civilians (or abstain from shooting them). Part III confronts broader implications of cliff running for law and democracy. As to law, cliff running reinforces a core lesson of the Legal Realist movement: To know what the law is, studying legal texts isn’t enough. An understanding of institutions, groups, and the people who inhabit them is at least as important. As to democracy, unbalanced professional liability cliffs are dangerous. Not only do they undermine the substantive law that democratic legislatures enacted—they also diffuse responsibility in a way that frustrates political accountability. Part IV offers doctrinal and institutional mechanisms for rebalancing off-kilter liability cliffs.
This Part introduces the liability cliffs that professionals like police and doctors sometimes must run along. The first Section defines the core concept of a professional liability cliff. The second explains why objective “reasonable person” liability standards are pervasive in professional cliff running, at least on paper. The final Section explores the dynamics of cliff-running from four perspectives: lawmakers, prosecutors, professionals, and the public. Together, Part I sets the stage for the detailed examination of police violence and emergency abortions that follows in Part II.
A. Professional Liability Cliffs
Professional liability cliffs, as we’ll use that term, have three defining characteristics. The first two are obvious enough: First, they regulate professional activity. The people who run along these cliffs do so because they have undertaken a particular line of work for which they have relevant training, education, and experience unlikely to be possessed by lawmakers, prosecutors, or the public.[25]
Second, they are liability cliffs. Lee Anne Fennel explains that liability cliffs are scenarios in which law renders an “all-or-nothing” judgment.[26] Either an officer’s fear of imminent harm from a suspect was reasonable, in which her use of force was lawful, or it wasn’t, in which case the force constitutes murder or manslaughter. Either a doctor reasonably believed that a pregnancy threatened a patient’s life, in which case an abortion is justified everywhere, or she didn’t, in which case it is a high-order felony in many states. There are no in-betweens. The professional acted lawfully, or she committed a serious crime. Liability cliffs are what Adam Kolber calls “bumpy” rules that convert ambiguity into binary judgments. That distinguishes them from “smooth” legal rules, where the outputs of penalties or punishments change gradually with incremental changes in inputs, i.e., behaviors or external circumstances.[27]
Finally, as used here, professional liability cliffs feature a specific kind of liability: They involve exposure to serious criminal punishment. As noted, liability cliffs exist in many areas of the law.[28] Criminal professional liability cliffs differ because the stakes of prosecution, conviction, and punishment are vastly greater than professional sanctions or civil fines that might be covered by insurance.[29] Criminal law invokes law’s most serious penalties—incarceration and even death.[30] That makes criminal liability cliffs “higher” than other cliffs in the sense that someone who slips has farther to fall before hitting the ground.[31] Criminal liability cliffs are to liability cliffs that impose fines or licensure sanctions[32] what Mount Thor (4,000-foot drop) is to the White Cliffs of Dover (under 500 feet).[33] When first exploring a phenomenon, it’s useful to begin with the paradigm case. That’s why we limit our scope to criminal cliffs, leaving to future work questions about whether our analysis applies as well to the shorter liability cliffs that professionals may encounter in other sources of law.
B. Legal Reasonableness Standards
The legal topography of professional liability cliffs looks simple enough. Whether professionals run safely or fall typically turns on whether their choices comported with the hypothetical decisions that a “reasonable person” or “reasonable professional” would have made under the circumstances.[34] Liability depends on an “objective” analysis of their conduct, rather than whether they subjectively believed that the conduct was warranted.[35] This Section explores why occupationally disparate liability cliffs share that doctrinal feature.
In criminal law theory terms, professional liability cliffs occur where a justification defense turns on the exercise of professional judgment.[36] Justification defenses apply where a person engages in conduct that is ordinarily prohibited but in this case has a socially beneficial reason.[37] (Contrast that with excuse defenses, which apply where a person engages in prohibited conduct for a reason that is understandable but not good in that sense.[38]) The paradigmatic justification is self-defense.[39] Killing a person ordinarily runs afoul of the laws against homicide. But when someone kills an aggressor who was about to kill or seriously injure them, we deem the act justified and do not punish the actor.[40]
Professional liability cliffs work just that way. On-duty police violence is often governed by self-defense rules directly or by closely aligned justification defenses that are specific to the context of law enforcement.[41] And in states that restrict abortion, the law does not punish the doctor who performs the procedure when an emergency medical exception applies.[42]
Justification defenses lie at the core of professional liability cliffs, helping to explain why “reasonableness” liability standards so frequently govern them. Justification defenses have triggering conditions—“circumstances that must exist before an actor is eligible to act,” as Paul Robinson observes.[43] In evaluating whether a triggering condition existed when an actor did an otherwise criminal act, we must ask: (1) what facts about the world did the actor believe to be true?, and (2) do those facts constitute the condition?[44] Courts typically frame the second question in terms of whether the actor’s belief that the triggering condition existed was objectively “reasonable” under the circumstances.[45] That reasonableness standard provides little concrete guidance, as the case studies will make clear.[46]
Reasonable-person analysis makes the legal structure of professional liability cliffs look very similar on paper. One point of difference is between cliffs where liability turns on negligence alone and cliffs for which “recklessness” is required. Texas’s emergency medical exception to its abortion ban asks how a “reasonably prudent physician, knowledgeable about a case and the treatment possibilities” would have handled a case.[47] That’s a negligence cliff. Other liability standards—because they turn on whether the professional acted recklessly—add a subjective question on top of the objective one: Did the actor understand the (objectively) unreasonable risk and consciously disregard it?[48] The legal structures of professional liability cliffs across different domains aren’t perfectly identical, but they are nonetheless strikingly uniform.[49]
C. Cliff Running from Four Perspectives
A person’s view of a professional liability cliff depends on where they stand. This Section surveys four vantage points—those of legislators who enact liability cliffs; prosecutors who patrol them; regulated professionals who run along them; and third parties who experience their consequences. Note that factfinders—jurors and trial judges—are not a major part of the story. That’s because they’re rarely stable or predictable enough for people to take them into account when deciding on conduct. It seems unlikely, that is, that a doctor in Austin, Texas would factor her jurisdiction’s relatively liberal jury pool into an abortion-or-not decision, or the same in reverse for a doctor in Dallas, and the logic works the same for cops. Besides, we suspect the terrifying event that professionals seek to avoid is being charged with a serious crime, and factfinders enter the picture after that’s happened.[50]
Professional liability cliffs are creations of law. So why would lawmakers—legislators and, to a lesser extent, judges—create them? There are good reasons and bad ones. It is a fraught enterprise to punish individuals, or even to threaten to punish them, for things that they do in a professional capacity. Professionals possess expertise that prosecutors generally lack. Making decisions about professionals under conditions of ignorance runs a risk of arbitrariness.[51] Prosecutors, or for that matter judges or jurors, asked to determine whether a doctor conducted herself as a “reasonably prudent physician” during an obstetrical emergency,[52] for instance, can do little more than guess.[53] Still, there is good reason not to relinquish legal decision-making to individual professionals or professional organizations: doing so would render individual professionals and, indeed, entire professions, unaccountable.[54] Yet professionals and the public and private bodies that regulate them have interests that do not (necessarily) align with the public’s.[55] Further, the judgments of professionals and their regulators lack the democratic pedigree of pronouncements by legislators, judges, and juries.[56] That matters, especially when liability standards turn on value judgments rather than purely technical questions.[57]
So there is a real tension, when using criminal punishment to regulate professionals, between accountability and deference to expertise.[58] All-or-nothing solutions are possible—lawmakers could confer absolute immunity on (certain) professionals for (specific) decisions made in their professional capacities.[59] Alternatively, they could decree that one’s professional status has no bearing on litigation—imagine a new rule of evidence declaring that: “Evidence of professional or occupational training or experience is not admissible to prove whether a person acted negligently or recklessly on a particular occasion.”[60] Corner solutions, however, are rarely optimal.[61] The first alternative wholly discounts professional accountability, while the second entirely dispenses with deference.
The public-minded explanation of professional liability cliffs is that they attempt to accommodate professional accountability and deference. Unlike an immunity regime, a cliff makes liability for on-duty professional conduct possible.[62] Part of the analysis is one’s status as a professional, and thus some measure of deference. But the balance between accountability and deference in a professional liability cliff depends on the details of the liability rule. Does that rule turn on recklessness or mere negligence? How severe are the sanctions if liability is established? The balance depends more still on the attitudes and actions of actors beneath the social planner level, as we’ll see below.[63] Lawmakers can utilize professional liability cliffs to avoid all-or-nothing corner solutions, but their control over the balance of accountability and deference is limited.
One might also justify professional liability cliffs as regulatory devices. While vague exceptions to crushing punishments appear to be confusing and precarious, they might be a normatively fitting way to mediate intractable controversies with large moral stakes on each side.[64] The moral complexity of topics like police use of force or emergency abortion is irreducible. Is it permissible to take one life to save another? That’s how self-defense works.[65] Or to end a nascent life to preserve a pregnant patient’s health? That’s what happens in emergency abortions. Suppressing these underlying stakes could also squeeze the values that animate them, leaving them to reemerge elsewhere in more pernicious forms: concealment that evades unaccountability, or exit from the profession that results in staffing shortages or pressures on health or safety.[66] If this is right, maybe it’s better to write those ethical tensions into the substantive law itself, bringing ambivalence to the surface and out into the open. On this view—which we articulate but do not endorse—forcing officers and doctors to make deeply uncomfortable and legally freighted decisions appropriately reflects that ambivalence.[67]
There is also a more sinister explanation for why legislators and judges may become attracted to professional liability cliffs. Imagine that a legislature dislikes a substantive law implicating the work of professionals but is constrained by politics or law from changing it. Perhaps legislators want police officers to be relieved of the prospect of homicide liability for their on-duty conduct but know that they’d face enormous public backlash for doing so.[68] Or perhaps legislators want to ban abortions without an emergency medical exception but know that a court would rule the law unconstitutional.[69] Acting directly, these legislators cannot achieve their unpopular or unlawful policy goals. Professional liability cliffs offer a nefarious way forward: Combined with liability rules that tilt toward either deference or accountability, they make it possible for legislators to approximate objectives they cannot pursue directly.
Legislators and judges create professional liability cliffs, but prosecutors enforce them. For prosecutors, professional liability cliffs are tools that can be used to accomplish their ends. Unlike some of their European counterparts, American prosecutors have almost limitless discretion to decide what cases to bring.[70] They can use that power to shape a professional liability cliff’s balance between accountability and deference.
Imagine, for instance, that a county prosecutor or a state attorney general wants professionals to feel comfortable running on a liability cliff without fear of punishment. That is, the prosecutor wants to emphasize deference to professional expertise over accountability. Either the prosecutor can require that any decision to charge a professional undergo extra layers of internal office review,[71] or she can order her line prosecutors to apply an exacting evidentiary standard to those cases.[72] Further, she could adopt an implicit or even explicit policy of simply not enforcing the underlying criminal prohibition in cases involving professionals.[73]
Now imagine that a prosecutor wishes to maximize accountability instead of deference. The prosecutor might make threatening public statements directed at members of the profession.[74] Or she might adopt a policy allowing subordinates to charge on the basis of mere probable cause.[75] (Although the law in most jurisdictions permits prosecutors to file charges based on probable cause,[76] the ABA recommends that prosecutors proceed only if “admissible evidence will be sufficient to support conviction beyond a reasonable doubt.”[77]) Or she might tell subordinates that they lack discretion in cases involving the professionals—if the charging standard is met, they must charge.[78]
Prosecutors thus have mechanisms through which they can impact a professional liability cliff’s balance of deference and accountability to achieve their ends. But what exactly are their ends? Some are surely idiosyncratic, depending on quirks of the prosecutor’s values, interests, and experiences.[79] Systematically, two variables—or really, sets of variables—are likely important. The first is a prosecutor’s stance vis-à-vis the substantive law. The logic is similar to the “sinister” explanation of why legislators create professional liability cliffs in the first place.[80] If a prosecutor opposes the justification defense, she will probably tilt the professional liability cliff in the direction of accountability. On the other hand, if she opposes the underlying criminal prohibition, we should expect a pro-deference posture.[81]
The second systemically important consideration is whether the prosecutor trusts the regulated profession(als).[82] Does she believe that members of the profession are (in general) likely to comply with the underlying criminal prohibition? Or does she think they intend to sabotage or circumvent it? Trust, in that sense, is a reason for a prosecutor to angle the professional liability cliff in the deference direction, while distrust is a reason to tilt it toward accountability. Why would prosecutors trust or distrust profession(al)s? Trust is likely to emerge when institutional connections exist between the prosecutor (and her office) and members of the profession.[83] Culture and politics matter too. If a prosecutor believes that members of the profession or even the profession itself oppose the underlying prohibition, that is likely to arouse prosecutorial suspicion, and thus distrust.[84] We will see these dynamics at play in the case studies below.[85]
What does cliff running mean for the professionals who engage in it? The threat of criminal prosecution fundamentally changes the calculus for individuals otherwise regulated by professional discipline and tort liability.[86] That said, the legal standards that govern professional liability cliffs provide little concrete guidance to police officers or doctors trying to decide—under enormous stress and time pressure—whether self-defense or an emergency medical exception applies to a particular situation.[87] Given the law’s ambiguity, is cliff running implacably unsafe from the professional’s perspective? Must professionals do this work perpetually anxious that their split-second decisions could land them in prison?[88] It depends, and the critical consideration is, once again, trust.
Consider first professionals who trust that prosecutors will defer to their professional judgments. If professionals believe that prosecutors will give them a wide berth and resolve uncertainties in their favor, cliff running is a relatively safe activity, notwithstanding the ambiguity of formal law.[89] That leaves them free to exercise their expertise independently. If the realistic threat of prosecution is off the table, however, it may also embolden them to invoke the justification defense (e.g., self-defense) in situations where it does not apply. That invites professionals to violate underlying prohibitions (e.g., homicide).[90]
The situation is very different for professionals who don’t trust prosecutors to defer to their professional judgments. For them, underspecified reasonableness standards beneath the specter of whopping punishments chill justification defenses. That can manifest itself in several ways. The most extreme is that people might exit the profession entirely, deciding that the risks are not worth the rewards. Or they may stay on the job, but refuse to perform the service that triggers the justification.[91] Doctors, for instance, might refuse to provide abortions under even extreme circumstances.[92] Or they might hesitate, delay, or display what courts have described in the policing context as “unwarranted timidity.”[93] Interference with independent professional judgment has obvious consequences for patients, who we will turn to shortly.[94] It affects the doctor’s quality of life, too.[95] The professional who feels powerless to carry out the action that she’s trained and charged to do can lead that person to experience weariness and exasperation.[96] This moral distress can leave regulated professionals feeling disaffected, defensive, and ridden by angst.[97]
What determines whether professionals trust prosecutors? As with the inverse question—whether prosecutors trust professionals—institutional connections and political and cultural alignment are likely salient.[98] To those, we add two more. The history of criminal prosecution surely matters. A long history of non-enforcement almost certainly inspires trust where a history of aggressive enforcement (or the absence of historical practice in either direction) would not. Finally, professionals’ perception of whether prosecutors trust them seems likely to be important. Again, we’ll find these dynamics in the case studies below. [99]
The way in which professionals respond to liability cliffs has consequences for third parties too. Professionals’ responses also impact their clients and others with whom professionals engage. This Section considers cliff running’s externalities. These externalities flow directly from the dynamics discussed in the last Section, and they vary depending on whether a liability cliff slopes too far toward accountability or deference. We saw that in cliffs where too little trust yields too much accountability (and not enough deference), professionals might cease providing services (like emergency abortions) that could expose them to liability, or they might hesitate too long before providing them.
The costs of that decision are borne as much if not more by a professional’s client than by the professional herself. When a liability cliff slopes decidedly away from deference, doctors may decide to stop performing abortions under any circumstances, or they may perform emergency abortions only when their patient is at death’s door—in either case imperiling pregnant patients with life-threatening complications. But the downstream consequences go further. Ob‑gyns nervous about criminal exposure for providing emergency abortion care might leave a jurisdiction altogether, and newly trained specialists might steer clear, leaving the jurisdiction with a deficit of qualified doctors. Patients would thereafter find it difficult or impossible to obtain obstetrical and gynecological care unrelated to abortion.
Cliffs that feature too much trust and too little deference pose different externalities. The professionals who run on such cliffs, we’ve seen, are emboldened not to worry too much about core prohibitions. While freeing for the professional, that liberation jeopardizes those whom the core prohibition was meant to protect from harm. In the context of police use of violence, for instance, a sloped liability cliff tempts officers to “shoot first” in the face of questionable threats to themselves or others. These shootings can take the lives of individuals and devastate their communities. And all of these externalities are likely to be disproportionately carried by the underprivileged and racial minorities. We explore those disparities in the case studies below.
Having defined cliff running and explored its dynamics in the abstract, we turn to examining how it works in the real world. This Part explores two paradigm cases of professionals running on criminal liability cliffs, both of which are cultural flashpoints: doctors performing emergency abortions and police officers using deadly force.
Doctors in many states run on professional liability cliffs when they perform abortions on patients experiencing medical emergencies. Nearly three years into the post-Roe era, abortion is illegal under most circumstances in at least twelve states without regard to gestational age.[100] In four more, quasi-bans make it illegal at or around six weeks.[101] Physicians who perform unlawful abortions in these states face not merely licensing consequences, but criminal punishment.[102] The amount varies by state. In Alabama and Texas, the maximum penalty is ninety-nine years in prison.[103] It’s fifteen years in Tennessee,[104] and ten in Arkansas, Georgia, Louisiana, Mississippi, and Oklahoma.[105] In at least eight states, doctors cannot even hope that merciful judges will spare them from prison. A doctor convicted of performing an illegal abortion faces a mandatory minimum sentence of ten years in Alabama, five in Texas, three in Tennessee, two in Idaho, and one in Georgia, Kentucky, Louisiana, and Mississippi.[106]
Every ban makes an exception for abortions performed in response to certain medical emergencies.[107] In some states, the exception applies only when the procedure is necessary to save the patient’s life. In Idaho, for instance, an abortion is allowed only where “necessary to prevent the death of the pregnant woman.”[108] The exceptions in Arkansas, Iowa, Mississippi, Oklahoma, and South Dakota are similarly circumscribed to situations in which the doctor’s patient may die without an abortion.[109] Other abortion-ban states expand their exceptions to scenarios where an abortion is necessary “to prevent a serious health risk” (Alabama),[110] “permanent impairment of a life-sustaining organ” (Kentucky and Louisiana),[111] or “substantial” and/or “irreversible” “impairment of a major bodily function” (Florida, Georgia, Indiana, South Carolina, Tennessee, and Texas).[112]
Both the narrow and broad versions of emergency exceptions turn on inherently uncertain predictions about how bad a patient’s condition might get and how fast. The Dobbs dissenters put the point like this: “Suppose a patient with pulmonary hypertension has a 30-to-50 percent risk of dying with ongoing pregnancy; is that enough?”[113] The broad versions, moreover, don’t specify in any detail just how damaging a harm must be to qualify, let alone how likely.[114] What if a delay in abortion care causes an unavoidable hysterectomy that keeps someone from getting pregnant again? Or if the timely cancer treatment someone needs isn’t compatible with her pregnancy? Or if complications from her diabetes or seizure disorder would risk blindness or brain damage? What about severe preeclampsia or premature rupture of the amniotic sac that risks infection and hemorrhaging. At what point is a patient’s fever too high? Her blood pressure too low? These laws don’t say.[115]
What they do say is that a doctor’s medical judgment triggers the exception only if it’s “reasonable.”[116] Statutes in at least eleven of the states with abortion bans or quasi-bans provide that emergency exceptions apply only when doctors exercise “reasonable medical judgment.”[117] That’s an objective determination. Statutes in three states declare that a reasonable medical judgment “means a medical judgment that would be made by a reasonably prudent physician, knowledgeable about a case and the treatment possibilities with respect to the medical conditions involved.”[118] It would appear that if prosecutors, judges, and juries later find those medical judgments to be unreasonable, then they are thereby disqualified as medically reasonable, even if they were rendered in good faith.[119]
This trio of qualities about abortion bans—harsh punishment, unrelenting vagueness, and ex post reasonableness review—renders cliff running perilous for doctors.[120] Elected officials turn the danger up to eleven. Prosecutors decide whether to charge doctors for performing abortions.[121] Doctors have so far only been charged for performing medication abortions, not surgical ones.[122] Still, elected prosecutors—especially red state attorneys general—have made it abundantly clear that prosecuting doctors for abortion is a high priority.[123] In 2022, for instance, Indiana’s Attorney General threatened to prosecute a doctor who’d performed an abortion for a ten-year-old rape survivor.[124] The same year the Alabama Attorney General threatened to prosecute health care providers if they helped their patients travel to another state for legal abortion care.[125] In the context of emergency abortions, Texas Attorney General Ken Paxton has been the most aggressive. After a Texas doctor, Damla Karsan, won a temporary restraining order barring officials from going after her for ending a pregnancy under the law’s emergency exception,[126] Paxton warned hospitals and doctors that, notwithstanding the order, they would still be prosecuted if they helped a patient get an abortion.[127]
Not every local prosecutor in abortion-ban states agrees. But when any have appeared reticent to enforce abortion bans, state officials have moved to force their hands.[128] Florida Governor Ron DeSantis suspended an elected district attorney who opposed criminalizing abortion.[129] The Texas legislature, meanwhile, enacted a law to remove prosecutors who refuse to bring charges against abortion providers.[130]
Between the abortion bans themselves and the expressed attitudes of prosecutors and other elected officials, it is little wonder that physicians are terrified of being prosecuted for performing emergency abortions. Three in five obstetrician-gynecologists surveyed in states that make it a crime to provide an abortion report fear of prosecution.[131] In another study, a doctor described the fear that pervades her everyday practice: “I feel like there’s a politician in the room with me with patients . . . just waiting to send me to jail, to make an example out of me . . . .”[132]
There’s nothing doctors in abortion-ban states can do to alleviate that fear. They can consult with risk managers, ethics committees, and referral services, but none of those sources can avert the possibility that abortion care they see as life-preserving might look like a crime to an aggressive prosecutor.[133] Any “case that to a medical provider falls in a ‘gray area’ might appear very much black and white to a prosecutor,” one doctor explained in the New York Times.[134] That leaves physicians, another doctor told a reporter, “plagued by fears that what [they] see[] as an emergency won’t be.”[135]
Why do legislators, prosecutors, and other elected officials make running on this cliff so dangerous? Perhaps individuals like Paxton oppose abortion wholesale and seek to undermine emergency exceptions they don’t like in the first place.[136] That may well be part of the story, but we lack evidence one way or the other, so set the possibility to one side.
The key to understanding why cliff running is so dangerous for abortion providers is that (some) prosecutors don’t trust obstetricians and (some) obstetricians don’t trust prosecutors.[137] The prosecutors don’t trust the doctors for a couple reasons. First, there are no meaningful institutional connections between them.[138] Second, they are often far apart politically and culturally: Red state prosecutors may think ob-gyns want to sabotage the prohibition on abortion, and they might be right.[139] The doctors don’t trust the prosecutors for those two reasons plus two others: the newness of abortion bans adds to the uncertainty and suspicion; and doctors know that prosecutors don’t trust them.[140]
The lack of trust in both directions yields an unbalanced liability cliff with too much accountability and too little deference. The consequence, as we have seen, is doctors who are petrified of being prosecuted for performing emergency abortions. That fear has terrible repercussions for doctors and, even more importantly, for their patients.[141]
What happens when doctors fear that performing a medical procedure may result in their being incarcerated? They hesitate, even when an abortion is medically and legally appropriate. Doctors might struggle to work within the law’s constraints to keep those under their care safe when continued pregnancy poses a serious but uncertain threat to their life or health.[142] But the specter of prison cannot help but influence their clinical decision-making. Lori Freedman explains that “the threat of criminalization looms so large for doctors” that they “wait as long as possible to intervene to ensure that the threat to the patient’s life is evident, so that no staff member, family member, or colleague will accuse them of performing an abortion for which they might be prosecuted.”[143]
There’s clear evidence that doctors are waiting too long before they intervene in dangerous pregnancies, or defensively declining patients whose miscarriage management or treatment for complication might be mistaken for an abortion.[144] In a recent survey of obstetricians practicing in states that restrict abortion, nearly half (forty-two percent) reported that “delays in delivering care have put a pregnant patient’s life or health at risk.”[145]
Clinicians are holding off as things get riskier for their patients, notwithstanding any legal exemptions to save a woman’s life, “because they’re not dying yet.”[146] That waiting and second-guessing chills care.[147] In academic studies and news stories, doctors have said as much. An Alabama maternal-fetal medicine doctor (a subspecialty within obstetrics) explained the impossible choice that doctors face when treating a pregnant patient as her condition deteriorates:
Will I have to wait until she gets a fever, so I can check off that box, that she’s in danger? Is that sufficient or will it require her heart rate to go up or her blood pressure to go down? Will she have to wait until she’s unstable to have this option offered to her? At what point, exactly, will I be risking jail for helping my patient through this, unharmed?[148]
Another doctor recalled a patient who faced dangerous complications:
She broke down and said, ‘I don’t want to die.’ . . . I thought, ‘I’m really worried about this patient. She’s incredibly sick, but did I do all the paperwork correctly? . . . What kind of trouble am I going to get into if something . . . happens when the patient is already extremely unstable.’[149]
Sometimes the hesitation comes from a doctor’s colleagues. As one doctor recalled a case in which a patient at nineteen to twenty weeks of pregnancy presented in an emergency room with “severe pain and advanced labor”:
Anesthesiology colleagues refused to provide an epidural for pain. They believed that providing an epidural could be considered [a crime] under the new law. . . . I will never forget this case because I overheard the primary provider say to a nurse that so much as offering a helping hand to a patient getting onto the gurney while in the throes of a miscarriage could be construed as ‘aiding and abetting an abortion.’ Best not to so much as touch the patient who is miscarrying.[150]
Even if doctors are willing to take on legal risk themselves, lawyers may counsel them to hesitate. As one doctor explained:
The way our legal teams interpreted it, until they became septic and started hemorrhaging, we couldn’t proceed . . . . [T]he legal people seem to have a different definition that also just feels horrible, to say until you’re at a greater than likely chance of dying, [I] can’t make a decision.[151]
The consequences of hesitant medicine are most obvious and direct for patients.[152] When doctors don’t perform medically necessary procedures, patients suffer—some die.[153] Since Texas sharply limited abortion access in 2021, sepsis rates have soared by more than fifty percent for women hospitalized when they lost their pregnancies after twelve weeks.[154] At least three Texas women—Josseli Barnica, Nevaeh Crain, and Porsha Ngumezi—have died because doctors waited too long to treat their miscarriages.[155] In Georgia, two women who had taken abortion medication, Candi Miller and Amber Thurman, died after doctors hesitated to treat emergency complications until it was too late—in Thurman’s case, declining abortion care for over twenty hours, despite “acute severe sepsis,” as her condition steadily deteriorated until her organs failed and her heart stopped.[156]
For more systematic evidence about the costs of hesitation, consider a study that followed twenty-eight women who had been admitted with pregnancy complications to two Dallas hospitals.[157] This was over the nine months before Dobbs after Texas’s civil ban went into effect, forbidding abortion starting at six weeks unless “the mother’s life is in danger.”[158] Researchers found that under that civil ban, patients had to wait an average of nine extra days for their status to be considered life-threatening enough to justify abortion.[159] This state-induced delay caused roughly twice as many of the women to experience preventable health problems so serious that they required intensive care and readmission.[160] Their resulting hemorrhaging, sepsis, and hysterectomies all could have been avoided by the immediate intervention that was routine before the prohibition.[161]
In many situations, wealthy patients can travel to access care in time for it to be effective. But undocumented immigrants and poor women of color disproportionately lack the insurance, savings, or social support required to travel far distances for treatment elsewhere without imperiling their jobs or caretaking responsibilities at home.[162] Whether they end up going untreated or pursuing dangerous self-help from unlicensed or unapproved underground alternatives, it’s patients who bear the brunt when clinicians who practice under the low-trust threat of liability cliffs feel hamstrung from providing them with emergency medical care.
Hesitant medicine harms doctors too. The fear of prison impedes their ability to do the work they trained for years to do. In one study, ninety-three percent of ob-gyns reported situations in which they or their colleagues could not follow clinical standards due to legal constraints, using descriptors like “muzzled,” “handcuffed,” and “straitjacketed.”[163] Their patients, moreover, stop trusting them, as they can tell that their professional judgment has been distorted.[164]
Preventing doctors from responding to the needs of people who come to them can estrange doctors from their work.[165] This moral vertigo takes a psychological toll on clinicians, making it hard for providers to forgive themselves or come to terms with becoming the kind of doctor who turns away patients seeking evidence-based care that they’re qualified to provide.[166] Here’s how one doctor described the inner turmoil:
I’m in the [operating room] dry heaving. I’m not dry heaving because of this surgery. I know how to do this surgery. . . . I trained for the stress of treating an unstable . . . ectopic pregnancy. [What] I did not train for, I am not ready for thinking about, “Is this the case that’s gonna make me a felon?”[167]
It’s not just that doctors hesitate. When they’re scared and miserable, some stop performing any abortions (regardless of the clinical circumstances), move to a jurisdiction with less officious laws, or quit practicing medicine altogether.[168] One maternal-fetal medicine physician explained that anxiety and fear drove her to uproot her family from their home in Idaho after the state banned abortion: “[N]ot being able to help [patients] the way that I should was just terrifying.”[169]
Departing doctors at least have a measure of choice about that decision, even if they see no good options. The patients they leave behind have no say in the matter at all. When an obstetrician-gynecologist leaves town, patients obviously lose access to a provider who can perform an emergency abortion. But ob-gyn practice includes much more than abortions, its scope extending to the “integrated medical and surgical care of women’s health throughout the lifespan.”[170] So patients also lose obstetrical and gynecological care having nothing to do with abortion.[171] An exodus of reproductive care providers reduces access to every kind of procedure that falls within the specialty.[172] That’s yet another way in which patients suffer the consequences of doctors’ cliff running.
When an on-duty police officer uses force, especially deadly force, she runs on a professional liability cliff.[173] If the force is deemed justified, she’s in the legal clear, though post-traumatic stress and psychological scars may endure.[174] If it isn’t, she faces the prospect of prosecution and conviction for assault, manslaughter, or murder.[175] In most states, a use-of-force statute defines the cliff’s shape.[176] Generally speaking, these statutes provide that officers are justified in using force, including deadly force when necessary, in three distinct situations—to defend themselves or others, to effectuate an arrest, and to stop certain dangerous felons from escaping.[177] While the legal details of these law enforcement justifications vary between states, a “reasonable person” standard usually backstops all three.[178] As Cynthia Lee explains, use-of-force statutes authorize violence when officers “reasonably believe” that the conditions of the justification exist.[179] That, Lee observes, is “such an open-ended standard” that it “provides little-to-no guidance” in “deciding whether an officer’s use of force was justified.”[180]
And yet, the vague liability standard notwithstanding, cliff running is remarkably safe for police. Officers are almost never charged with crimes for on-duty force.[181] Exact numbers are elusive,[182] but exhaustive reporting by the Washington Post in 2015 opened a window into the frequency and character of police prosecutions.[183] The Post found that between 2005 and 2015, a time period during which thousands of people were shot and killed by police, fifty-four officers were charged.[184] The vast majority of those cases (about eighty percent) included what the Post described as an “exceptional” circumstance, such as “a video recording of the incident, incriminating testimony from other officers or allegations of a coverup.”[185] More than a third of the cases included two such circumstances. Of course, a charge is not a conviction. Of the fifty-four officers charged, thirty-five had their cases resolved by the time the story went to press, and only fourteen had been convicted. The rest, the Post reported, “were acquitted or saw their charges dropped.”[186] The Post data—the most recent and reliable available to follow these police-shooting cases to their conclusion—reveals how remote the chances are that an officer will fall off of her professional liability cliff.[187]
This Section explores the features of the police professional liability cliff that makes running on it so safe. We start with the mechanics—the legal and institutional reasons why police are rarely charged. We then go in search of the motivations underlying those mechanisms. Next, we examine how police perceive the liability cliff, a perception that differs vastly, we will see, from how that cliff actually operates against officers in practice. Finally, we consider the cliff’s consequences for third parties: most significantly, direct victims of excessive police violence, who are disproportionately people of color.
How do police so often avoid charges for on-duty violence, including killings? Law is part of the story. Law enforcement justification defenses are somewhat more robust than the justification defenses available to civilians. Police officers may stand their ground even in states where civilians must retreat from physical threat posed by another civilian (if safely possible) rather than use force to defend against that danger.[188] Moreover, civilian justification defenses feature strict proportionality and necessity requirements that are minimized or absent in cases involving law enforcement.[189] A handful of jurisdictions even swap reasonable person analysis for a liability standard that inquires only whether the officer believed in good faith that the force was warranted.[190] Special procedural rules also apply, especially in the interrogation context. The Constitution (as the Supreme Court construes it) imposes few limits on the tactics that investigators may use when questioning suspects.[191] When interrogating civilian suspects, police make the most of their constitutional leeway, applying psychologically manipulative techniques that elicit confessions both true and false.[192] It’s different when police interrogate police. Kate Levine, the leading scholar on police as criminal suspects and defendants, summarizes the protections afforded officers under “LEOBORs” (Law Enforcement Officers’ Bills of Rights) enacted by statute in some states and incorporated into police union contracts in others:
LEOBORs often provide that police suspects may be questioned only during the day; that they may be questioned only by a limited number of interrogators; that they must be given time to attend to their personal needs; that they may not be threatened, subjected to abusive language, or induced to confess through untrue promises of leniency; and that their choice to inculpate themselves must not be conditioned on losing their job or benefits.[193]
This is a “set of rights,” Levine writes, “that, for all intents and purposes, applies only to police officers.”[194]
Formal legal rules, both substantive and procedural, thus play some role in making cliff running safe for police officers. The policies and practices of prosecutors are more important. As legions of commentators observe, prosecutors are reluctant to charge officers.[195] That doesn’t mean that prosecutors can ignore cases in which police officers kill suspects under questionable circumstances. Especially in recent years, when police killings have been attended by community protest, that’s untenable.[196] Prosecutors can, however, engage in robust investigations of the factual, legal, or normative strengths and weaknesses of a case prior to charging. As Levine explains, “prosecutors use their full arsenal of procedural discretion when a police-suspect’s liberty and reputation is on the line” including by conducting “thorough precharge investigations” and by crediting “an officer-suspect’s version of events.”[197] They might even present all the facts, inculpatory and exculpatory, to a grand jury and let it take responsibility for the charging decision.[198] That’s a stark contrast to prosecutorial precharging practices in standard cases, where the analysis of a case is often perfunctory and where the grand jury, if one is even employed, is a rubber stamp.[199] It turns out that when prosecutors investigate fully before charging, they often don’t charge at all.
The question is why prosecutors construct procedural barriers to prosecution only for police suspects. In other words, why do prosecutors make cliff running safe for the police? In Part I, we hypothesized that prosecutors might tilt a professional liability cliff toward deference (and away from accountability) based on their view of the substantive law—either because they think the underlying criminal prohibition goes too far or because they think the justification defense doesn’t go far enough.[200] Obviously, prosecutors do not believe that the criminal law’s prohibitions on homicide and assault go too far. It’s more plausible that they believe that the law enforcement justifications aren’t sufficiently robust, but we know of no data on that, so we set the possibility aside.[201]
We also hypothesized in Part I that prosecutors would prioritize deference when they trust the regulated professionals to comply with the underlying criminal prohibition rather than attempt to sabotage it.[202] And we identified two sources from which such trust might spring—institutional connections between prosecutors and professionals, and politics and culture. For prosecutors and police, both sources (usually) point in the direction of trust.
The institutional connections between prosecutors and police run deep. Police officers bring prosecutors most of the cases they prosecute.[203] In grand juries, preliminary hearings, and the occasional trial, police are often prosecutors’ principal witnesses.[204] Without cooperation from the police, prosecutors could not function, at least not in any recognizable form. Likewise, without prosecutors, the police could not credibly threaten anyone with criminal conviction. Prosecutors and police “depend on each other and work as a team,” as Kami Simmons explains.[205] They are, as Somil Trivedi and Nicole Gonzalez Van Cleve put it, part of “interdependent institutions.”[206] It is hardly surprising that relationships of trust would emerge between the representatives of such deeply entwined institutions.[207] Nor is it unexpected that a prosecutor would think twice—or thrice—before prosecuting an officer that the “prosecutor knows and may be friendly with,” as Levine notes.[208]
Beyond their institutional and personal connections, police and prosecutors share cultural norms. To be sure, there are important cultural differences between them. Prosecutors are lawyers, and law enforcement officers (aside from FBI agents) usually are not.[209] Their career trajectories are different, with the typical law enforcement officer staying in a job far longer than the typical prosecutor.[210] At the same time, as Bruce Green and Rebecca Roiphe observe, they “serve the same law enforcement interests.”[211] That can contribute to prosecutors identifying with officers. When they interviewed prosecutors in Chicago, Trivedi and Van Cleve found that “part of what it meant to be a prosecutor was to align with the police at all costs.”[212] As Dan Richman explains, “one ought not underestimate the unifying influence of a shared commitment to ‘getting the bad guys,’ hardened by the adversarial process, nurtured by mutual respect and need, and on occasion lubricated by alcohol.”[213] “[A]s in any other organizational setting,” he adds, “the social relationships that can arise out of constant and routine contacts will provide a solid foundation for trust.”[214]
Given all of that, and the longstanding pattern of non-prosecution,[215] one might expect that police officers wouldn’t worry much about being criminally prosecuted for what they do on the job. Not so, at least not today. A 2023 survey by Police1, a leading news website catering to law enforcement, found that 21% of officers identified the “risk of prosecution for on-duty actions” as their greatest challenge, up from 17% the year before.[216] A 2024 survey of officers by the Minnesota Police and Peace Officers Association, meanwhile, reported that 98% of officers in that state were “somewhat, very, or extremely concerned about risk of prosecution for on-duty actions.”[217]
These fears are not empirically grounded. While there has been an increase in police homicide cases in recent years,[218] the trend is profoundly modest. Criminologist Philip Stinson reports that the post-2014 growth is “just a few cases” a year, going from “maybe eight or nine” to “15 or 20 instead.”[219] Besides, he notes, often several officers are charged in a single incident, “mak[ing] it look like there’s more going on there.”[220]
If prosecution remains extremely rare, why are officers so anxious about it? That is, why do they increasingly say that they distrust prosecutors? We suspect that two recent events contribute, and both reinforce the centrality of culture in the construction of professional liability cliffs. The first is the progressive prosecutor movement,[221] which brought to office elected prosecutors with values and norms diverging from those traditionally associated with police officers.[222] Some of these progressive prosecutors talked about prosecuting police officers,[223] though, as we have seen, they rarely followed through.[224] The second is the unprecedented wave of protests in 2020 following the killing of George Floyd,[225] and the successful murder prosecution of Minneapolis police officer Derek Chauvin.[226] The protests revealed widespread doubt about whether police use force appropriately.[227] While we joined the many who saw Chauvin’s conviction as amply justified, it sparked resentment in many police officers.[228] “It’s disheartening to hear the prosecution throw cops under the bus and leave the defense to build them up, which is the opposite of what normally happens,” a Florida detective told NBC News during the trial.[229] Police anxiety about being prosecuted for on-duty conduct is, we suspect, partially the product of an availability bias stemming from the Chauvin case.[230] It may be no accident that the survey we noted above, in which the vast majority of police respondents claimed to fear prosecution, is from Minnesota.[231]
Notwithstanding officers’ self-reported fear of prosecution, cliff running remains, for them, extremely safe. This is still, in other words, a liability cliff in which deference trumps accountability. What consequences does that have for third parties? In Part I, we hypothesized that in cliffs that slope toward deference, the underlying criminal prohibition may be undermined.[232] Here, the underlying criminal prohibitions are laws against homicide and lesser forms of assault. Have those been undermined? In other words, are police killing (and otherwise injuring) people without justification? Certainly, many scholars and activists think so.[233] And that was the central premise underlying the extraordinary protests of 2020, so millions of ordinary people apparently agree.[234]
A counternarrative is possible. Perhaps the reason police are almost never charged with homicide for on-duty shootings is that they almost always get it right. They shoot suspects who pose a threat and abstain from shooting suspects who do not. Maybe fifty-four police officers were charged for shooting suspects between 2005 and 2015 because fifty-four police officers shot suspects without justification during that time.[235] While that narrative is theoretically conceivable, it’s unpersuasive. The best evidence against it is that police gunfire is disparately distributed by race.
If police officers were preternaturally accurate in evaluating the threat posed by suspects, then once one controls for those threats, they should be equally likely to shoot Black suspects and suspects of other races. They are not. In a sophisticated empirical study utilizing the Washington Post’s database of police shootings, which allowed them to control for suspects’ threat levels and whether they fled, Jeffrey Fagan and Alexis Campbell found that “Black suspects are more than twice as likely to be killed by police than are suspects from other racial or ethnic groups, including shootings where there are no obvious reasonable circumstances.”[236]
That finding is hard to reconcile with a narrative that police rarely make unreasonable mistakes when deploying deadly force. It does, however, help explain why researchers find that “[f]or young men of color, police use of force is among the leading causes of death.”[237] At least when it comes to Black suspects—especially Black male suspects—cliff running is so safe that police officers can sometimes get away with murder. That is terrible in and of itself, but it also disserves the goal of public safety, breeding fear and distrust among those citizens whose cooperation law enforcement rely on to warn them about criminal conduct. Cops emboldened to run hastily along cliffs may come to be seen as trigger-happy by the affected friends and neighbors who become reluctant to call on law enforcement or cooperate with them. Whatever benefits such encounters promise to reap may be outweighed by the anticipated peril of turning unjustifiably violent. Reduced collaboration between communities and police in turn tends to reduce public safety.[238]
The foregoing Parts have shown how the dynamics of cliff running play out for those involved with and impacted by them, often for the worse. Cliff running has additional implications for law and democracy. This Part explores them.
In a 2024 case, State v. Zurawski, patients and doctors asked the Texas Supreme Court to clarify the legal standard governing when physicians can lawfully perform abortions under the state’s medical emergency exception.[239] They wanted the court to apply a “good faith” standard.[240] That would have meant that when a doctor determines in good faith that the patient has a life-threatening condition that risks death or serious impairment without an abortion, the exception applies, full stop. If the court had adopted that standard—which it did not—doctors would not have had to worry about a prosecutor or court examining their decision for “reasonableness” after the fact. Instead, doctors could be prosecuted only if the state could prove that they hadn’t really believed that the exception applied, a difficult showing.
At oral argument in the case, Texas Supreme Court Justice Evan Young pressed the plaintiffs’ lawyer to explain why her physician clients should have a more favorable standard than police officers when they use force on the job. That is, why should doctors—who Justice Young acknowledged must make “grave decisions” without “being sure what the law allows and being subject to serious consequences if they are wrong”—have it easier than police officers who make at least as precarious decisions with “much less time” to think about them?[241] On one level, Young’s question makes sense. Both doctors and lawyers make split-second decisions in the line of duty, carrying enormous legal risks—the phenomenon that we called cliff running. Why should doctors avoid ex post reasonableness analysis of their decisions when cops must endure it for theirs?
On closer inspection, however, Young’s question was misguided. While the legal standards governing cops and doctors are similar in the books, the laws that actually govern these professionals are very different in action.[242] Reasonableness standards across professional contexts don’t vary all that much on their face.[243] In each, the doctrine calls for objective measures of conduct or judgment. That legal treatment of cliff running remains strikingly consistent at the doctrinal level. Nonetheless, as we saw in Part II, police officers can run cliffs with relative abandon, while obstetrician-gynecologists shy away in fear.[244] The former cliff is remarkably safe while the latter is intensely dangerous.
The general point is that superficially similar liability rules can mask profoundly different modes of regulation. If all we know about a professional liability cliff is that it’s governed by a reasonableness standard, we don’t know much about it. Liability rules are mediated by institutions that are staffed by people. To understand a liability cliff, we have to know about those relationships, and specifically whether prosecutors trust the regulated professionals and whether the regulated professionals trust prosecutors. If cliff runners are confident that they’ll be given the benefit of the doubt, they can observe, interpret, and respond using their best professional judgment without worrying that they’ll fall. That deference vanishes without trust. Sensing mistrust and exposure to catastrophic risks to their career and freedom, professionals will tend to adopt a more defensive posture and stay far away from the blurry line for fear of crossing it unwittingly.
The presence of trust can even, in effect, change the legal standard. A trusted professional is a reasonable one. When prosecutors trust police officers, the question of whether an on-duty shooting was reasonable becomes almost circular. Was it reasonable for Officer Smith to shoot the suspect holding a knife? Well, Officer Smith is a trustworthy officer and was in a better position than anyone to judge. And he thought it was reasonable! Ergo it was. The logic breaks down if Officer Smith didn’t think the shooting was reasonable—i.e., if he acted in bad faith—but now we’ve swapped an objective inquiry for a subjective one. Trust collapses the distinction between reasonableness and good faith.[245]
Professional liability cliffs are thus a useful reminder of an old lesson, that “the law” depends at least as much on institutions and the people inhabiting them as the words printed in statute books and judicial opinions.[246] Focusing too much on the latter obscures the real nature of legal regulation and leads to false equivalences. That was the deep flaw in Justice Young’s question comparing cops and doctors at the Zurawski argument.
Part II focused on the dangers of running on unbalanced professional liability cliffs for professionals, their clients, and the public. Lopsided cliffs also jeopardize democratic governance. We saw in Part I that liability cliffs arise when legislatures create two substantive legal rules: a core prohibition (like laws against homicide or abortion) and a justification defense.[247] Those legislative choices are displaced when the balance between accountability and deference to expertise slopes too far to one side or the other in a way that fails to optimize trust between professionals and prosecutors.
The details, of course, depend on the direction of the imbalance. When trust tilts a professional liability cliff to deference, the core prohibition weakens. If police officers are too confident that they won’t be prosecuted for on-duty violence, they may be emboldened to shoot people unjustifiably, undermining the legislature’s decision to criminalize homicide. When a lack of trust tilts a professional liability cliff in the other direction, the justification defense goes inactive. Doctors who are apprehensive about going to prison are chilled from performing emergency abortions that the law allows, and that professional duty compels. Again, the legislature’s decision is thwarted. To the extent that core prohibitions and justification defenses represent the outcome of democratic legislative processes, both outcomes are problematic.
Unbalanced liability cliffs presage a subtler democratic deficit too. We have seen that underenforcement of homicide laws against police[248] and overdeterrence of abortion laws against ob-gyns[249] lead to resentment and mistrust by citizens and patients whose safety and health these professionals and professions are meant to serve. These tensions generate democratic pressure for change. The structure of professional liability cliffs and the nature of cliff running, however, make reform more difficult. That’s because it disperses responsibility among legislators, prosecutors, and the professionals themselves.
The politics of emergency abortion illustrate this well. Imagine you are a Texan with a high-risk pregnancy. You hope for a successful pregnancy and birth, but because of your history of preeclampsia, there is a real prospect that you’ll need an abortion. You’re nervous that if you do, you won’t be able to get one in a safe and timely way. You discuss the concern with your ob-gyn, who tells you that per advice she got from her lawyer, she can’t perform an abortion based on preeclampsia unless the patient (you) has had three seizures or begun organ failure. You’re livid. Your doctor suggests that you complain to your state senator. When you do, he tells you that Texas law does allow life-saving abortions and nothing in state law requires doctors to wait for multiple seizures or organ failure. He says to call your local prosecutor for more information. You do. As long as your doctor uses reasonable medical judgment, the prosecutor says, she has nothing to worry about. Is it medically reasonable to make you wait until your organs are failing, you ask. That’s between you and your doctor, the prosecutor says, explaining that he’d look at all the facts and circumstances after the procedure when deciding whether to prosecute. At your next checkup, you tell the ob-gyn what the legislator and prosecutor told you. The doctor shows you a press release from the state attorney general threatening to charge doctors who perform emergency abortions.[250] Then she points to a picture of her children on the wall and tells you that she can’t risk going to prison for the rest of her life.[251]
Whom should you be angry with? You’ve clearly been put at risk, but not by any one actor or institution acting alone. You might blame the legislator for enacting a vague law with harsh penalties. You might blame the local prosecutor for refusing to provide real guidance. You might blame the state attorney general for grandstanding. You might even blame your doctor for putting her own well-being over yours. Meanwhile, the legislator, prosecutor, attorney general, and doctor can all point fingers at each other. The Texas Supreme Court itself did a version of that in Zurawski, cynically blaming doctors for not performing abortions that the state’s emergency medical exception would permit.[252] Dispersed blame obfuscates accountability, making it hard for an aggrieved public to know precisely where to focus finite political energies.[253] Lopsided liability cliffs don’t just endanger the public and undermine democratic legislative outcomes. They also make it harder for that same public to demand democratic change.
Considering the perils and pitfalls of professional liability cliffs, what’s a well-meaning lawmaker to do? The italicized qualifier is important. We have little to offer legislators or judges who enact or execute professional liability cliffs as a means of doing indirectly what they are unable to do directly, whether because of law, popular sentiment, or governing majorities.[254] Our recommendations are aimed at officials who seek to do the public’s bidding within those constraints.[255] What should such lawmakers worry about when it comes to the professional liability cliffs in their jurisdictions? We have suggested three worthy objectives: maintaining a balance between legal accountability and deference to professional expertise, preserving substantive legislative choices, and avoiding needless externalities. Above all, lawmakers should ensure that prosecutors are faithful to the substantive compromises that they embedded in the law.
One option—the most straightforward—is to eliminate professional liability cliffs by taking the prospect of severe punishment off the table.[256] That might mean retaining criminal sanctions but moderating them.[257] Texas, for instance, could subject doctors who perform unauthorized abortions to a criminal fine instead of decades in prison.[258] Liability cliffs could also be smoothed into sloping hills via incremental penalty steps that distinguish among violations that are bad-faith (worst), reckless (medium), and negligent (least bad).[259]
Alternatively, legislators could abandon criminal sanctions altogether in favor of civil remedies, internal discipline, or licensing sanctions. If a police officer shoots a suspect without justification, she might lose her ability to be a cop and/or pay a monetary judgment to her victim’s heirs, but she would not be subject to criminal punishment. Civil justice, internal discipline, and occupational licensing regimes have strengths and weaknesses of their own that lie beyond our scope. But it’s worth noting that in the context of police violence, they have a very poor track record.[260]
Beyond such considerations, the viability of replacing a liability cliff with a moderated-carceral or non-carceral regime depends significantly on the gravity of the harm that the underlying prohibition is meant to prevent. The less serious the harm, the easier it is for lawmakers to eliminate a liability cliff.[261] On the other hand, few would consider decriminalizing homicide, the underlying prohibition in police killings.[262] Red-state legislators likely see unauthorized abortion in a similar light.[263]
While eliminating liability cliffs is possible in principle and sometimes justified, in the high-salience scenarios with which we’ve engaged, it may be ethically or politically nonviable. When liability cliffs cannot be eliminated, how can lawmakers ameliorate their negative consequences? We have seen that serious problems accrue for professionals and third parties when prosecutors tilt liability standards too far in the direction of either deference or accountability.[264] When that happens, well-intentioned lawmakers need to apply countermeasures, which come in both doctrinal and institutional forms.[265]
In Part II, we argued that institutions and politics/culture matter more than doctrine. But that doesn’t mean that one ought to ignore doctrine, and it’s better that doctrine be calibrated to reality than not. Doctrinally, lawmakers should consider second-best liability standards. The general theory of second best posits that when one of the conditions required for an optimal ordering (economic, institutional, legal, etc.) isn’t available, it’s usually a bad idea to set the other conditions at the levels they would have in that optimum.[266] What that theory means, as one of us has explained elsewhere, is that a “social policy (or law, institution, etc.) that is undesirable on first principles might be usefully checked by a second policy that is also, on first principles, undesirable.”[267] In our context, lawmakers can use otherwise undesirable liability rules to check undesirable prosecutorial priorities and practices.[268]
Imagine that a lawmaker thinks that for some professional liability cliff, a recklessness standard optimally balances legal accountability and deference to professional expertise. That substantive determination is based on several factors, including the esteem (or lack thereof) in which the lawmaker holds the profession, the degree to which the lawmaker thinks that non-criminal sources of regulation (licensing boards, civil remedies, etc.) keep abuses in check, and the nature of the harm the lawmaker wants to avoid. Holding all else equal, recklessness would be the right liability rule for the lawmaker to choose.
But what if the lawmaker knows that prosecutors will tip the standard toward accountability in practice? Now the lawmaker needs a counterweight. One strategy is to dial the liability rule down: Instead of recklessness, she might adopt a good-faith standard. For instance, if red-state legislators know that their prosecutors will tilt liability rules for obstetricians to accountability (and the legislators are “well-intentioned” within our meaning), they might adopt a standard providing a defense to their abortion ban if a physician establishes their good faith belief that the procedure was medically necessary.[269] Or they could go even further with a liability rule requiring the government to prove beyond a reasonable doubt that the physician actually knew or intended that the emergency exception didn’t apply. The Supreme Court recently did something like that in Ruan v. United States, ruling (with good sense but not much textual support) that doctors violate the Controlled Substances Act by dispensing prescriptions only if they knowingly or intentionally lacked prescribing authorization.[270]
Instead of worrying about prosecutors tilting a liability standard toward accountability and away from deference, lawmakers might fear the opposite. Now the lawmaker might want to adopt a supra-optimal liability standard, like ordinary negligence, instead of recklessness. For instance, if lawmakers believe that prosecutors are finding excuses not to charge blameworthy police officers for on-duty killings, they could consider changing the liability standard for manslaughter in police cases to mere negligence, even if they think recklessness is ordinarily right.[271] The point is that the liability standards that are optimal in the abstract aren’t necessarily the right choice in a second-best world.[272] Rather, legislators should make liability rules narrower or broader to counter distortions by prosecutors.[273]
Doctrine is not the only tool available to well-intentioned lawmakers. They can also change the rules that govern professional liability cliffs by reworking the institutional design of prosecution.[274] Again, the specifics depend on the nature of the problem. If prosecutors are too close to those they regulate, producing a surplus of bilateral trust and deference, the obvious reform is to use different prosecutors.[275] So, for example, if local prosecutors are personally and professionally aligned too closely to local cops, then police killings could be handled by prosecutors from other counties, the state attorney general’s office, or elsewhere.[276]
Stabilizing liability cliffs sloped toward accountability is more difficult, but creative solutions are possible. The key is to integrate the profession itself into the decision to prosecute one of its members. Prosecutors ordinarily have plenary control over the decision to charge someone with a crime.[277] But that need not be the case: Well-intentioned lawmakers could adopt a rule precluding prosecutors from bringing a criminal case against a licensed professional for specific kinds of on-the-job conduct unless the regulator with jurisdiction over the professional’s license first refers the matter for prosecution.
For instance, state medical boards are charged with overseeing physician licensing and, in many states, the practice of medicine.[278] A state that bans abortions except during medical emergencies might make a referral from its medical board a condition precedent to the prosecution of a licensed physician.[279] The state medical board, in turn, could apply its expertise to determine whether the emergency medical exception applied.[280] That determination wouldn’t be conclusive in cases referred for prosecution—per the Sixth Amendment, that would be up to the jury. But the medical board’s judgment would be final in cases that it declined to send.[281]
The state boards would use a version of whatever process they have for investigating and adjudicating other kinds of complaints.[282] So imagine that someone files a complaint saying that Dr. Jones performed an unlawful abortion. Initially, the board’s staff would investigate. If the staff thinks that there wasn’t a valid medical emergency justification for the procedure, there could be a hearing in which Dr. Jones argues that there was. The hearing might be before a panel of the board, perhaps, though an administrative law judge could be the first step, with appeal to the board. The board would then make a determination, with the judge’s help or not. If it finds that the emergency exception does not apply, it could refer the case to the local district attorney for prosecution, take action against Dr. Jones’s license, or both.
There may be reasons to doubt that state medical boards are independent and/or that they apply genuine expertise.[283] But to the extent that a board’s work reflects objective decision-making, its determination could interpose a constructive layer of professional judgment into charging decisions. And while it would be unusual (indeed, perhaps novel) to formally require a regulatory referral prior to prosecution, there are analogues. It would give doctors an immunity like that enjoyed by diplomats, who can only be prosecuted if their home country allows it.[284] As a practical matter, moreover, many categories of federal crimes are typically prosecuted when a regulatory agency refers a matter to the Department of Justice.[285] The rule we envision differs only in that it makes referral a formal precondition for prosecution.[286]
This Article introduces the theory and practice of professional liability cliffs, where small differences in conduct or circumstances result in drastically different legal outcomes for on-the-job professionals. Running along these cliffs is time sensitive, high-stakes, and, for some, legally precarious. Cliff running has gone conspicuously unrecognized in existing scholarship. We have critically analyzed this phenomenon’s operation in the contexts of policing and medicine, arguing that most important to its appraisal is balancing professional accountability alongside deference to expertise. We’ve shown how varying levels of trust in institutional relationships play a decisive role in maintaining that equilibrium.
We hope that this first word on professional liability cliffs is not the last. Future research could profitably extend our analysis in at least three dimensions. First, we examined the most extreme form of professional liability cliffs—criminal cliffs where the consequences of falling are disastrous. Outside the binary extremes of criminal liability, tort law and professional licensure likely form more modest liability cliffs for professionals.[287] Further scholarship could explore whether the dynamics we found in criminal cliffs exist there as well.
Second, we focused on the legal consequences on the steep side of the cliff—that is, criminal prosecution. A new project could investigate how professionals’ incentive structures are complicated by legal consequences on the other side of the cliff. For instance, obstetricians face criminal exposure for performing emergency abortions, as we have explored in detail. But do they face a meaningful risk of malpractice liability for not performing emergency abortions?[288] That is beyond our scope but would be a fruitful avenue for next inquiries.
Finally, researchers could explore how our analysis applies to different kinds of professionals: pilots and air traffic controllers, firefighters and sea captains, lawyers and accountants, journalists and therapists. All of them at times find themselves caught between the duties that they owe to the individuals or organizations they serve on the one hand, and, on the other, to the various social interests in public safety or national security that their work inevitably implicates.
-
* Herzog Research Professor of Law, University of San Diego. ↑
-
** Professor of Law, University of Georgia School of Law. We owe a special debt to David Cohen, Kevin Cole, Don Dripps, Avlana Eisenberg, Miranda Fleischer, Lee Fennell, Natalie Jacewicz, Adam Kolber, Michelle Layser, David McGowan, Tracey Meares, Caley Petrucci, Robert Schapiro, and Seth Stoughton. We are also grateful to the editors of the Washington University Law Review. ↑
-
. Tex. Health & Safety Code Ann. § 170A.002(b)(2) (West 2023). ↑
-
. Id.; see Tex. Penal Code Ann. § 12.32 (West 2023). ↑
-
. Cal. Penal Code § 835a(c) (West 2025). ↑
-
. Id.; see also id. § 190. ↑
-
. To be clear about the metaphor, doctors and officers run along the edge of these cliffs—they don’t dash straight off them, like Wile E. Coyote and the Road Runner. For video illustration of runners who sprint along actual cliffs, see Sébastien Montaz-Rosset, The Break, YouTube, at 0:45–0:58 (May 10, 2016), https://youtu.be/KIBbIWycPAY (last visited July 19, 2025). ↑
-
. On liability cliffs generally, see Lee Anne Fennell, Slices and Lumps: Division and Aggregation in Law and Life 191 (2019). ↑
-
. See infra Section I.A. ↑
-
. And they’re not the only professionals that must run on professional liability cliffs. Palliative care and ICU doctors who prescribe pain medications that could hasten a terminal patient’s death operate in a similar legal regime. See infra note 46. ↑
-
. See Amelia Thompson-DeVeaux, Nathaniel Rakich & Likhitha Butchireddygari, Why It’s So Rare for Police Officers to Face Legal Consequences, FiveThirtyEight (Sept. 23, 2020, 4:53 AM), https://fivethirtyeight.com/features/why-its-still-so-rare-for-police-officers-to-face-legal-consequences -for-misconduct/ [https://perma.cc/G5F5-AFCE]. For available data and best estimates about the relative frequency of such prosecutions, see infra text accompanying notes 179–85. ↑
-
. See Avidan Y. Cover, Reconstructing the Right Against Excessive Force, 68 Fla. L. Rev. 1773, 1777 (2016) (describing “strands [of law that] have evolved into a judicial apology for—if not an endorsement of—a ‘shoot first, think later’ police culture”). ↑
-
. Erika L. Sabbath, Samantha M. McKetchnie, Kavita S. Arora & Mara Buchbinder, US Obstetrician-Gynecologists’ Perceived Impacts of Post–Dobbs v. Jackson State Abortion Bans, 7 JAMA Network Open, Jan. 2024, art. no. e2352109. ↑
-
. See Anjali Nambiar, Shivani Patel, Patricia Santiago-Munoz, Catherine Y. Spong & David B. Nelson, Maternal Morbidity and Fetal Outcomes Among Pregnant Women at 22 Weeks’ Gestation or Less with Complications in 2 Texas Hospitals After Legislation on Abortion, 227 Am. J. Obstetrics & Gynecology 648 (2022). ↑
-
. Questions of accountability and standards of reasonableness can also govern professional liability when someone else runs too close to a cliff in the real world. See Peter Kozera, Falling off a Cliff: Who is to Blame?, 36 Law Soc’y J. 42, 42 (1998). ↑
-
. See infra notes 230–36 and accompanying text. ↑
-
. See Sabbath et al., supra note 11, at 5; see also infra notes 166–70 and accompanying text. ↑
-
. See infra notes 80–83 and accompanying text. ↑
-
. See infra notes 96–97 and accompanying text. ↑
-
. See infra notes 245–51 and accompanying text. ↑
-
. See infra Part IV. ↑
-
. See infra Part IV. ↑
-
. See John E. Calfee & Richard Craswell, Some Effects of Uncertainty on Compliance with Legal Standards, 70 Va. L. Rev. 965 (1984) (describing overcompliance and undercompliance with rules based on uncertainty about liability). ↑
-
. See infra Part IV. ↑
-
. While few would question medicine as a “profession,” debates about whether policing qualifies as one are longstanding. Compare Richard A. Posner, Professionalisms, 40 Ariz. L. Rev. 1, 2 (1998) (“Occupations that are not commonly referred to as professions include . . . police and detective work . . . .”), with Mancini v. City of Tacoma, 479 P.3d 656, 671 (Wash. 2021) (Madsen, J., dissenting) (“I would hold that police officers are professionals just as engineers and attorneys are, and they must be held to the standard of care for their profession.”). For that matter, so are debates about how to define a “profession” in the first place. See Morris L. Cogan, The Problem of Defining a Profession, 297 Annals Am. Acad. Pol. & Soc. Sci. 105, 105 (1955) (“To define ‘profession’ is to invite controversy.”). We do not mean to wade into those debates. We use the term “profession” capaciously. For our purposes, a profession is a line of work that involves specialized training. Under that approach, policing qualifies. See generally Anna Lvovsky, Rethinking Police Expertise, 131 Yale L.J. 475, 484 (2021) (describing “police expertise” as “that broad constellation of insights, training, and experience that makes officers especially adept at what the courts take to be their core professional tasks”). We express no view about whether policing is a profession applying narrower definitions of the term. ↑
-
. Fennell, supra note 6, at 191. ↑
-
. Adam J. Kolber, Smooth and Bumpy Laws, 102 Cal. L. Rev. 655, 657, 667 (2014); see also Avlana K. Eisenberg, Discontinuities in Criminal Law, 22 Theoretical Inquiries L. 137, 138 (2021) (discussing punishments with “bumpy” or “discontinuous” inputs and outputs). ↑
-
. Fennell, supra note 6, at 190 (noting presence of liability cliffs in property, torts, criminal law, and contracts). ↑
-
. See Erin Stuart, Comment, Delayed Emergency Care: How Professional Liability Insurance Affects Doctors’ Decisions After Dobbs and What Needs to Change, 107 Marq. L. Rev. 1097, 1100–01 (2024). ↑
-
. See Guyora Binder & Robert Weisberg, Response, What Is Criminal Law About?, 114 Mich. L. Rev. 1173, 1175 (2016) (“Punishment is the strongest manifestation of government power . . . .”). ↑
-
. Some criminal cliffs are of course steeper than others: for example, the punishment for providing an abortion is ninety-nine years in prison in a state like Texas as compared with one year in Georgia. Despite these disparate sentences, both share in common the identical threshold threat of prosecution, conviction, and incarceration—as distinct from civil liability—that is enough to deter most people and drive them to assiduously avoid. ↑
-
. See infra note 285 and accompanying text. ↑
-
. Compare Ken Jennings, Meet Canada’s Mount Thor: The World’s Steepest, Tallest Cliff, Condé Nast Traveler (June 10, 2023), https://www.cntraveler.com/stories/2013-06-10/mount-thor-canada-maphead-ken-jennings [https://perma.cc/VKY3-V2EB], with Jim Hargan, The White Cliffs of Dover–Everything You Need to Know, Brit. Heritage Travel (July 6, 2024), https://britishheritage. com/travel/facts-white-cliffs-dover [https://perma.cc/5MV6-UA2Q]. ↑
-
. See supra notes 1–13 and accompanying text. ↑
-
. For a skeptical overview of the line between objective and subjective liability standards, see R. George Wright, Objective and Subjective Tests in the Law, 16 U.N.H. L. Rev. 121 (2017). ↑
-
. For discussion of “public authority justification[]” defenses, which are “often limited to certain persons whose position or training makes them particularly appropriate protectors of the interest at stake,” see Paul H. Robinson, Criminal Law Defenses: A Systematic Analysis, 82 Colum. L. Rev. 199, 215–16 (1982). ↑
-
. See Paul H. Robinson, Structure and Function in Criminal Law 95 (1997) (explaining that the existence of “justifying circumstances means that, while the harm prohibited by the offence does occur, it is outweighed by the avoidance of a greater harm or by the advancement of a greater good”). ↑
-
. See id. at 96 (distinguishing justification defenses from excuse defenses); see also Janet C. Hoeffel, Deconstructing the Cultural Evidence Debate, 17 U. Fla. J.L. & Pub. Pol’y 303, 311 n.49 (2006) (“All excuse defenses–insanity, duress, intoxication–are built on the notion that what the defendant did was not desirable, but understandable.”). ↑
-
. See Mitchell N. Berman & Ian P. Farrell, Provocation Manslaughter as Partial Justification and Partial Excuse, 52 Wm. & Mary L. Rev. 1027, 1031 (2011). ↑
-
. See Wayne R. LaFave, 2 Subst. Crim. L. § 5.7(a) (3d ed. 2000) (“It is only just that one who is unlawfully attacked by another, and who has no opportunity to resort to the law for his defense, should be able to take reasonable steps to defend himself from physical harm.”). ↑
-
. See, e.g., Kelly M. Hogue, When an Officer Kills: Turning Legal Police Conduct into Illegal Police Misconduct, 98 Tex. L. Rev. 601, 607–08 (2020). ↑
-
. See infra notes 105–11 and accompanying text. To be clear, “reasonable” isn’t the same thing as “correct.” Reasonable people get things wrong, so allowance for normal accidents is compatible with reasonable person standards. For example, doctors who act reasonably and in good faith will still get some calls wrong in these situations. When the doctor performs the abortion, whether she was right in assessing the emergency is an unknowable counterfactual, in the sense that we usually can’t say with certainty that the patient would otherwise have died. We can more easily know if a doctor was wrong when she doesn’t perform the abortion, though—if the doctor refuses to perform an abortion and the patient dies or has organ failure as a result, the doctor made an error. See, e.g., infra text accompanying notes 153–59. ↑
-
. See Robinson, supra note 35, at 99. ↑
-
. Some criminal law theorists would rephrase the first question with: “what facts were actually true about the world?” See, e.g., id. at 101. This is the long-standing debate in criminal law theory over whether justification is properly understood as an “objective” or “subjective” defense. Mitchel Berman summarizes the debate:
Roughly, the subjectivists argue that an actor who believes that circumstances exist which would confer a justification is legally justified (at least if the belief is reasonable, and perhaps even if not), whereas one who acts in ignorance of potentially justificatory circumstances that do exist is fully inculpated. Objectivists, in contrast, contend that the actor who mistakenly believes his conduct justified is, at most, excused, and that one who commits an offense unaware of circumstances sufficient to confer a justification is at least partially exculpated.
Mitchell N. Berman, Justification and Excuse, Law and Morality, 53 Duke L.J. 1, 39 (2003). The debate is about whether the “reasonable person”—against whose judgment the real-world actor is judged—is omniscient. For “objectivists,” she is. For “subjectivists,” on the other hand the reasonable person “knows” (or believes) what the real-world actor knew (or believed), even if the real-world actor was mistaken. Then the question is whether, assuming that state of the world, her conduct was objectively justified. ↑
-
. As a leading criminal law treatise explains about self-defense, the law generally requires “the defendant’s belief in the necessity of using force to prevent harm to himself be a reasonable one, so that one who honestly though unreasonably believes in the necessity of using force in self-protection loses the defense.” That’s an “objective requirement,” the treatise notes. LaFave, supra note 38, § 5.7(c); see also id. § 5.8(b) (similar for the defense of another justification defense); id. § 5.4 (necessity defense). ↑
-
. See infra Part II. We should also say that open-ended standards are probably inevitable given the multiplicity of distinct circumstances police and doctors will face that resist straightforward classification into neat and predictable categories. See Anne Drapkin Lyerly, Ruth R. Faden & Michelle M. Mello, Beneath the Sword of Damocles: Moral Obligations of Physicians in a Post-Dobbs Landscape, Hastings Ctr. Rep., May–June 2024, at 15, 16–18. Alternatively, lawmakers could seek to detail eligible facts or conditions as a kind of preclearance to act legally. See, e.g., La. Admin. Code tit. 48, pt. I, § 401 (2024). That might reassure tentative officers or physicians in certain specific scenarios. The accompanying risk is that such enumeration would operate to suppress the case-by-case discretion that we want professionals to be able to exercise to respond appropriately in time-sensitive situations. See Dov Fox, The Abortion Double Bind, 113 Am. J. Pub. Health 1068, 1070 (2023). For an overview of empirical evidence that more clear and certain legal rules have the effect of constraining conduct more than standards (that apply the law in more contextual, individualized ways, adapted to changing circumstances), see Brian Sheppard, Calculating the Standard Error: Just How Much Should Empirical Studies Curb Our Enthusiasm for Legal Standards?, 123 Harv. L. Rev. F. 92, 99–104 (2011). ↑
-
. Tex. Health & Safety Code § 170A.001(4) (West 2023). ↑
-
. That standard governs how palliative care doctors run on professional liability cliffs. Imagine that a palliative care doctor increases the opioid dosage of a patient with a terminal illness, knowing that the drugs might hasten death. See Kelly K. Dineen & James M. DuBois, Between a Rock and a Hard Place: Can Physicians Prescribe Opioids to Treat Pain Adequately While Avoiding Legal Sanction?, 42 Am. J.L. & Med. 7, 35 (2016). If a prosecutor decides that the increase was reckless, the doctor can be charged with manslaughter. But the legal structure of the palliative care doctor’s liability cliff differs in one respect to those involving police violence and emergency abortions. If a person other than a medical provider gives someone a risky quantity of opioids while disregarding the danger, and the person dies, they’ll have a hard time defending against homicide charges. E.g., People v. Gaworecki, 175 N.E.3d 915, 919 (N.Y. 2021) (“[A] person who, ‘with the requisite mens rea, engages in conduct through the sale or provision of dangerous drugs that directly causes the death of a person may be prosecuted for manslaughter in the second degree or criminally negligent homicide.”). The palliative care doctor stands in a different position only because she can argue that the risk was medically justified. That’s not technically a justification defense—it goes to whether the state can prove the recklessness element of manslaughter—but it’s closely aligned: a justification-adjacent failure-of-proof defense. See LaFave, supra note 38, § 9.1(a)(1) (“A failure of proof defense is one in which the defendant has introduced evidence at his criminal trial showing that some essential element of the crime charged has not been proved beyond a reasonable doubt.”). ↑
-
. Beyond the distinction between negligence and recklessness mens rea requirements, some professional liability cliffs involve special procedural mechanisms. For instance, and as explained in detail below, in many jurisdictions police officers enjoy unique protections during investigations of their on-duty use of force due to “Law Enforcement Officers’ Bill of Rights” enacted by state legislatures. See infra notes 186–92 and accompanying text. ↑
-
. Additionally, as Part II shows, cops are rarely charged in the first place, and doctors even less often, so we could only speculate about how jury dynamics would play out there. See infra notes 119–33, 179–85 and accompanying text. ↑
-
. See Jay Alexander Gold, Wiser Than the Laws?: The Legal Accountability of the Medical Profession, 7 Am. J.L. & Med. 145, 152–53 (1981). ↑
-
. See supra note 45 and accompanying text. ↑
-
. See Edward K. Cheng, The Consensus Rule: A New Approach to Scientific Evidence, 75 Vand. L. Rev. 407, 432 (2022) (“The layperson is epistemically incompetent to judge the expert opinion substantively and has neither the time nor the resources to gain such expertise.”). ↑
-
. Cf. Robert Dingwall & Paul Fenn, “A Respectable Profession”? Sociological and Economic Perspectives on the Regulation of Professional Services, 7 Int’l Rev. L. & Econ. 51, 61 (1987) (positing that professions’ “contract[s] with the state” renders them “indirectly accountable to the state rather than simply being the simple servants of their clients”). ↑
-
. See Edward Shinnick & Frank H. Stephen, Professional Cartels and Scale Fees: Chiselling on the Celtic Fringe?, 20 Int’l. Rev. L. & Econ. 407, 407–08 (2000) (explaining that “[e]conomists have been, traditionally, highly critical of many aspects of professional self-regulation” because it has “the effect of a cartel”); see also Richard L. Abel, American Lawyers 35 (1989) (observing that “the extent to which the ‘independent’ professional actually pursues client rather than personal interest and elevates the public interest above both [] is rarely investigated empirically”). ↑
-
. See Cheng, supra note 51, at 450. ↑
-
. See id. at 431 (“[V]alue-laden questions involving policy trade-offs and morality are assumed to be within the competency of a lay decisionmaker and not areas where experts have specialized knowledge.”). ↑
-
. Specifically, we refer to deference to the individual professional’s time-sensitive decision, not the worldview of the profession they are part of. Similar tensions between expertise and accountability appear in many legal domains. See, e.g., K. Sabeel Rahman, Note, Envisioning the Regulatory State: Technocracy, Democracy, and Institutional Experimentation in the 2010 Financial Reform and Oil Spill Statutes, 48 Harv. J. on Legis. 555, 570 (2011) (observing that judicial “[d]eference to [administrative] agency expertise . . . is inevitably in tension with values of political accountability and democratic legitimacy”); Paul L. Regan, Great Expectations? A Contract Law Analysis for Preclusive Corporate Lock-Ups, 21 Cardozo L. Rev. 1, 94 (1999) (noting the “deference-accountability tension” in reviewing corporate directors’ activities during merger talks). ↑
-
. The Supreme Court recently conferred such immunity on the U.S. President, after all, at least for certain conduct. See Trump v. United States, 603 U.S. 593 (2024). ↑
-
. The rule would fit neatly into the specialized relevance rules in Article IV of the Federal Rules of Evidence. ↑
-
. See Brett H. McDonnell, Sticky Defaults and Altering Rules in Corporate Law, 60 SMU L. Rev. 383, 401 (2007) (“Sometimes corner solutions are optimal, but usually they are not.”). ↑
-
. See, e.g., Katherine Mims Crocker, Qualified Immunity, Sovereign Immunity, and Systemic Reform, 71 Duke L.J. 1701, 1713 (2022) (explaining that “many critics believe that qualified immunity . . . contributes to a broad feedback loop between legal unaccountability and police brutality”). ↑
-
. See infra Section I.C.2. ↑
-
. See Robert A. Burt, Death is That Man Taking Names: Intersections of American Medicine, Law, and Culture 157–59, 163 (2002). ↑
-
. See Restatement (Third) of Torts: Intentional Torts to Persons § 23 (Am. L. Inst., Tentative Draft No. 6, 2021). ↑
-
. See, e.g., Poppy Noor, The Doctors Leaving Anti-Abortion States: ‘I Couldn’t Do My Job At All,’ The Guardian (Oct. 26, 2022, 6:00 AM), https://www.theguardian.com/world/2022/oct/26/us-abortion-ban-providers-doctors-leaving-states [https://perma.cc/5GDR-6GT4]; Elise Schmelzer & Seth Klamann, 3 Years After Colorado’s Landmark Police Accountability Bill, What’s Changed? And Has Push for Further Reform Slowed?, Denv. Post (July 2, 2023, 6:03 AM), https://www.denverpost.com/ 2023/07/02/colorado-police-reform-body-cameras-george-floyd/ [https://perma.cc/AM6E-H4JJ] (Colorado police officers leaving their jobs after SB20-217 ended qualified immunity in the state). ↑
-
. In the legal field, tolerance of this type of discomfort can be viewed as a skill that young lawyers must hone to be successful. See Andrew B. Ayers, What if Legal Ethics Can’t Be Reduced to a Maxim?, 26 Geo. J. Legal Ethics 1 (2013); see also Seana Valentine Shiffrin, Essay, Inducing Moral Deliberation: On the Occasional Virtues of Fog, 123 Harv. L. Rev. 1214 (2010) (discussing the salutary impact that superficial opacity may have on citizens’ moral deliberation and on robust democratic engagement with the law). ↑
-
. See, e.g., Stephen Neukam, Graham Floats Potential Compromise on Qualified Immunity, The Hill (Jan. 30, 2023, 11:34 AM), https://thehill.com/homenews/senate/3835986-graham-floats-potential-compromise-on-qualified-immunity/ [https://perma.cc/Q3AV-MJQU]. ↑
-
. See Dobbs v. Jackson Women’s Health Org., 597 U.S. 215, 339 n.2 (2022) (Kavanaugh, J., concurring) (observing, with apparent approval, that “[i]n his dissent in Roe, Justice Rehnquist indicated that an exception to a State’s restriction on abortion would be constitutionally required when an abortion is necessary to save the life of the mother”); State v. Zurawski, 690 S.W.3d 644, 671 (Tex. 2024) (Lehrmann, J., concurring) (“Certainly, a woman’s right to a life-saving abortion is one such limit [on legislative authority to regulate abortions].”). ↑
-
. See William T. Pizzi, Understanding Prosecutorial Discretion in the United States: The Limits of Comparative Criminal Procedure as an Instrument of Reform, 54 Ohio St. L.J. 1325, 1337 (1993) (“American prosecutors have far more discretion than their continental counterparts,” such that when “an American prosecutor decides not to file a criminal charge, there usually is no mechanism that would permit judicial review of that decision.”). ↑
-
. See Russell M. Gold, “Clientless” Lawyers, 92 Wash. L. Rev. 87, 117 (2017) (“Because of the ineffectiveness of the electoral check, scholars have urged internal review processes for prosecutors to provide meaningful input and check each other.”). ↑
-
. We will see an example of this below when we consider prosecutorial charging policies for police shootings. See infra notes 193–212 and accompanying text. ↑
-
. See, e.g., Milan Markovic, Charging Abortion, 92 Fordham L. Rev. 1519 (2024) (suggesting that ethically conflicted prosecutors in antiabortion states use their discretion to prioritize cases where other harms like unsafe care or coercion are at play); Robert A. Mikos, The Evolving Federal Response to State Marijuana Reforms, 26 Widener L. Rev. 1 (2020) (discussing marijuana laws and the uneasy truce between legalizing states and federal prosecutors). ↑
-
. See infra notes 121–25 and accompanying text. ↑
-
. Probable cause is not an exacting standard. See William Ortman, Probable Cause Revisited, 68 Stan. L. Rev. 511, 559–60 (2016) (explaining that in federal courts and many states“grand juries and magistrates may legitimately find probable cause even if they do not believe that guilt is more likely than not”). ↑
-
. Id. at 546 (noting that “the vast majority of states that still routinely require grand jury indictments have adopted probable cause as the charging standard”). ↑
-
. Crim. Just. Standards for the Prosecution Function § 3-4.3(a) (Am. Bar Ass’n 2017). ↑
-
. As we alluded to above, mandatory prosecution is a feature of some European prosecution systems. See Pizzi, supra note 68, at 1337; see also Darryl K. Brown, How Criminal Law Dictates Rules of Criminal Procedure, 70 Rutgers U. L. Rev. 1093, 1098 (2018) (“Some civil law jurisdictions have long had a mandatory or compulsory prosecution rule for both police and prosecutors.”). ↑
-
. Cf. Stephanos Bibas, The Need for Prosecutorial Discretion, 19 Temple Pol. & C.R.L. Rev. 369, 371 (2010) (“What we should fear is not prosecutorial discretion but idiosyncratic prosecutorial discretion.”). ↑
-
. See supra notes 66–67 and accompanying text. ↑
-
. A prosecutor’s stance on the substantive law is itself determined by many factors, including the prosecutor’s ideological commitments, the views of the voting public (in the case of elected prosecutors) or of other government officials (for appointed prosecutors), and even internal power dynamics within state government and other bureaucracies. See generally Daniel Epps, Adversarial Asymmetry in the Criminal Process, 91 N.Y.U. L. Rev. 762, 816 (2016) (noting a “familiar pattern” in which a “legislature passes a criminal statute” and “prosecutors refuse to enforce the law vigorously” and providing plausible reasons for such non-enforcement). ↑
-
. We define trust as “a psychological state comprising the intention to accept vulnerability based upon positive expectations of the intentions or behavior of another.” We borrow this definition from Denise M. Rousseau, Sim B. Sitkin, Ronald S. Burt, & Colin Camerer, Not So Different After All: A Cross-Discipline View of Trust, 23 Acad. Mgmt. Rev. 393, 395 (1998), who also provide an excellent (albeit dated) overview of an enormous literature on trust that spans multiple disciplines. See id. at 395. ↑
-
. See generally Nicholas R. Parrillo, Federal Agency Guidance and the Power to Bind: An Empirical Study of Agencies and Industries, 36 Yale J. on Reg. 165, 179 n.44 (2019) (recognizing “presumptive mutual trust” when a “regulator has a continuing series of interactions with a regulated party”). ↑
-
. See Rousseau et al., supra note 80, at 396–97 (observing that trust “can also be seen as the result of attributes of the other party, such as that party’s competence, concern, openness, and reliability”). ↑
-
. See infra Part II. ↑
-
. See Calfee & Craswell, supra note 21, at 966–67. ↑
-
. See infra notes 105–17, 171–78 and accompanying text. ↑
-
. Cf. Barnes v. Felix, 145 S. Ct. 1353, 1363 (2025) (Kavanaugh, J., concurring) (“[W]hen a driver abruptly pulls away during a traffic stop, an officer has no particularly good or safe options. None of the options available to the officer avoids danger to the community, and all of them require life-or-death decisions that must be made in a few seconds in highly stressful and unpredictable circumstances.”). ↑
-
. See, e.g., Erwin Chemerinsky, The Role of Prosecutors in Dealing with Police Abuse: The Lessons of Los Angeles, 8 Va. J. Soc. Pol’y & L. 305, 305 (2001) (“Prosecutors inevitably must work closely with police officers on literally a daily basis. . . . [P]rosecutors are reluctant to alienate the very officers that they must work with and rely on in their cases.”). ↑
-
. See, e.g., id. at 306, 306 n.7 (suggesting the Rampart scandal, in which officers allegedly “shot un-armed suspects,” was attributable to the lack of prosecutorial accountability). ↑
-
. Rebecca B. Reingold, Lawrence O. Gostin & Michele Bratcher Goodwin, Opinion, Legal Risks and Ethical Dilemmas for Clinicians in the Aftermath of Dobbs, 328 JAMA 1695, 1695 (2022). ↑
-
. See infra text accompanying notes 153–59. ↑
-
. Richardson v. McKnight, 521 U.S. 399, 408 (1997). ↑
-
. See infra Section I.C.4. ↑
-
. See infra notes 162–67 and accompanying text. ↑
-
. Sabbath et al., supra note 11, at 4–5. ↑
-
. Id. at 5–6. ↑
-
. See infra notes 134–38 and accompanying text. ↑
-
. See infra Part II. ↑
-
. See After Roe Fell: Abortion Laws by State, Ctr. for Reprod. Rts., https://reproductive rights.org/maps/abortion-laws-by-state/ [https://perma.cc/UJ2G-KAB7]. ↑
-
. Id. ↑
-
. We’re aware of two exceptions. In Iowa and West Virginia, it appears that doctors who perform illegal abortions are subject to licensing sanctions, but not criminal prosecution. See W. Va. Code § 61-2-8 (2025) (providing criminal punishment for persons who perform abortions other than currently-licensed physicians); id. § 16-2R-7 (requiring license revocation for physician who performs illegal abortion); Planned Parenthood of the Heartland, Inc. v. Reynolds ex rel. State, 9 N.W.3d 37, 53 (Iowa 2024) (“Violations of the fetal heartbeat statute could result in revocation of abortion providers’ medical licenses and fines of up to $10,000.”). ↑
-
. Ala. Code § 26-23H-6(a) (2025) (defining abortion as Class A felony); id. § 13A-5-6(a)(1) (2023) (providing range of ten to ninety-nine years for Class A felonies); Tex. Health & Safety Code Ann. § 170A.004 (West 2023) (defining abortion as a Class 1 felony); Tex. Penal Code Ann. § 12.32(a) (West 2023) (providing range of five to ninety-nine years for Class 1 felonies). ↑
-
. Tenn. Code Ann. § 39-15-213(b) (2025) (defining abortion as a Class C felony); id. § 40-35-111(b)(3) (providing range of three to fifteen years for Class C felonies). ↑
-
. Ark. Code Ann. § 5-61-304(b) (2025) (providing maximum sentence of ten years); Ga. Code Ann. § 16-12-140(b) (2025) (providing range of one to ten years); La. Stat. Ann. § 14:87.7(C) (2024) (providing range of one to ten years); Miss. Code Ann. § 41-41-45(4) (2025) (providing range of one to ten years); Okla. Stat. tit. 63, § 1-731.4(B)(2) (2024) (providing maximum sentence of ten years). ↑
-
. See sources cited supra notes 103–03 (providing the relevant statutes for Alabama, Texas, Tennessee, Georgia, Louisiana, and Mississippi); see also Ky. Rev. Stat. Ann. § 311.772(3) (West 2024) (making abortion a Class D felony); id. § 532.020(1)(a) (providing range of one to five years for Class D felonies). ↑
-
. See Mabel Felix, Laurie Sobel & Alina Salganicoff, KFF, A Review of Exceptions in State Abortion Bans: Implications for the Provision of Abortion Services (2024), https://www.kff.org/womens-health-policy/issue-brief/a-review-of-exceptions-in-abortions-bans-implications-for-the-provision-of-abortion-services/ %5Bhttps://perma.cc/Z2VA-GZ6M%5D (“All of these bans have an exception to prevent the death of the pregnant person and some bans include [an exception] . . . when there is risk to the health of the pregnant person . . . .”). ↑
-
. Idaho Code § 18-622(2)(a)(i) (2025). ↑
-
. Ark. Code Ann. § 5-61-304(a) (2025); Iowa Code §§ 146A.1(2), (6)(a) (2025); Miss. Code Ann. § 41-41-45(2) (2025); S.D. Codified Laws § 22-17-5.1 (2025); Okla. Call for Reprod. Just. v. Drummond, 526 P.3d 1123, 1130 (Okla. 2023) (“The law in Oklahoma has long recognized a woman’s right to obtain an abortion in order to preserve her life . . . . This right can be viewed as protected by the Oklahoma due process section.”). ↑
-
. Ala. Code § 26-23H-4(b) (2025). ↑
-
. Ky. Rev. Stat. Ann. § 311.772(4)(a) (West 2025); La. Stat. Ann. § 14:87.1(1)(b)(v) (2024). ↑
-
. Fla. Stat. § 390.0111(3)(a)(1)(b)(IV) (2024); Ind. Code § 16-34-2-1.2(2) (2025); Tenn. Code Ann. § 39-15-213(c) (2025); Ga. Code Ann. § 16-12-141(a)(3) (2025) (stating specifically, “substantial or irreversible physical impairment of a major bodily function”); S.C. Code Ann. § 44-41-610(9) (2024) (stating specifically, “substantial or irreversible physical impairment of a major bodily function”); Tex. Health & Safety Code Ann. § 170A.002(b)(2) (West 2023) (omitting irreversible and stating only “substantial impairment of a major bodily function”). West Virginia’s exception uses similar but not identical language—an abortion is permitted to “avert . . . serious risk of substantial life-threatening physical impairment of a major bodily function.” W. Va. Code § 16-2R-2 (2025). For close analysis of these phrases and similar language that’s contained in several state abortion bans, see Brian G. Slocum & Nadia Banteka, Fair Notice and Criminalizing Abortions, 113 J. Crim. L. & Criminology 747, 760–69 (2024). ↑
-
. Dobbs v. Jackson Women’s Health Org., 597 U.S. 215, 393 (2022); see also Lisa H. Harris, Navigating Loss of Abortion Services–A Large Academic Medical Center Prepares for the Overturn of Roe v. Wade, 386 New Eng. J. Med. 2061 (2022). ↑
-
. See, e.g., Medical Exceptions to State Abortion Bans: Idaho, Tennessee, Oklahoma, Stories: Patients and Physicians, Ctr. For Reprod. Rts. (Sept. 12, 2023), https://reproductiverights .org/case/emergency-exceptions-abortion-bans-idaho-tennesseeoklahoma/medical-exceptions-id-tn-ok-patients-doctors-stories/ [https://perma.cc/4A3Q-3RUQ]. ↑
-
. Lawmakers who support abortion bans have resisted efforts to clear up these ambiguities. See, e.g., Kavitha Surana, “We Need to Defend This Law”: Inside an Anti-Abortion Meeting with Tennessee’s GOP Lawmakers, ProPublica (Nov. 15, 2022, 12:00 PM), https://www.propublica.org/ article/inside-anti-abortion-meeting-with-tennessee-republican-lawmakers [https://perma.cc/MRZ2-YN5L] (describing a call organized by the National Right to Life Committee urging Tennessee GOP legislators to hold the line on the country’s most restrictive emergency exception language, urging lawmakers not to revise the law). Abortion opponents say the problem isn’t harsh punishments or ambiguous exemptions but “abortion alarmism” and “baseless fears.” Hadley Heath Manning, Abortion Alarmism–Not Dobbs–Is Going to Hurt Women, RealClear Health (Oct. 6, 2022), https:// http://www.realclearhealth.com/articles/2022/10/06/abortion_alarmism__not_dobbs__is_going_to_hurt_women_111420.html [https://perma.cc/UHX7-2HCF]. Medical boards in states that ban abortion have declined invitations to clarify emergency exceptions either. See, e.g., Elizabeth Sepper, Karl White & Anitra Beasley, Opinion, The Texas Medical Board and the Futility of Medical Exceptions to Abortion Bans, 331 JAMA 2073 (2024). And hospitals have struggled to provide any guidance to interpret and apply these medical exceptions either. See Jen Christensen, Amid Contradictory Laws, Hospitals in One State Were Unable to Explain Policies on Emergency Abortion Care, Study Finds, CNN (Apr. 25, 2023, 10:01 AM), https://www.cnn.com/2023/04/25/health/emergency-abortion-confusion-okahoma/index .html [https://perma.cc/2MCU-PHPM]. Meanwhile, the American College of Obstetricians and Gynecologists maintains that medicine is too individualized, case-specific, and time-sensitive to enumerate exceptions in advance: “[I]t is impossible to create an inclusive list of conditions that qualify” as emergencies under an abortion ban carveout and “it is dangerous to attempt to create a finite list of conditions to guide the practice of clinicians attempting to navigate their state’s abortion restrictions.” Understanding and Navigating Medical Emergency Exceptions in Abortion Bans and Restrictions, ACOG (Aug. 15, 2022), https://www.acog.org/news/news-articles/2022/08/understanding-medical-emergency-exceptions-in-abortion-bans-restrictions [https://perma.cc/D8NQ-N9BV]. ↑
-
. The countervailing legal risk—for not performing a medically necessary abortion that results in injury or death—imposes only civil liability and even then is limited to emergency staff at the federally funded hospitals covered by the Emergency Medical Treatment and Labor Act. Hospitals could lose their Medicare funding for failing to stabilize any patient in active labor or whose symptoms are so acute that “the absence of immediate medical attention could reasonably be expected” to place that person’s health “in serious jeopardy.” 42 U.S.C. § 1395dd. ↑
-
. See Ala. Code § 26-23H-3 (2025); Fla. Stat. § 390.0111(1) (2024); Ga. Code Ann. § 16-12-141(b) (2025); Ind. Code § 16-34-2-1 (2025); see also Iowa Code § 146E.2(2) (2025); Ky. Rev. Stat. Ann. § 311.772(4)(a) (West 2025); S.D. Codified Laws § 22-17-5.1 (2025); Tenn. Code Ann. § 39-15-213(c)(1) (2025); Tex. Health & Safety Code Ann. § 170A.002(b) (West 2023); W. Va. Code § 16-2R-3(a) (2025). ↑
-
. Tex. Health & Safety Code Ann. § 170A.001(4) (West 2023); see also Fla. Stat. § 390.011(12) (2024); Iowa Code § 146E.1(6) (2025). ↑
-
. We know of two exceptions. Idaho applies a good faith standard to its emergency exception. See Idaho Code § 18-622(2)(a)(i) (2025) (no liability if the “physician determined, in his good faith medical judgment and based on the facts known to the physician at the time, that the abortion was necessary to prevent the death of the pregnant woman”). For its part, Louisiana lawmakers apparently couldn’t decide between a good faith or reasonableness standard. See La. Stat. Ann. § 14:87.1(b)(v) (2024) (no liability if procedure was “necessary in good faith medical judgment or reasonable medical judgment”). ↑
-
. Scholars have argued that these characteristic features of abortion bans are among those that render them unconstitutionally vague. See Greer Donley & Caroline Kelly, Abortion Disorientation, 74 Duke L.J. 1, 82–88 (2024); Maxine Eichner, Mara Buchbinder, Abby Schultz, Cambray Smith & Amy Bryant, The Inevitable Vagueness of Medical Exceptions to Abortion Bans, 16 UC Irvine L. Rev. (forthcoming 2026), https://ssrn.com/abstract=5224731 [https://perma.cc/ NDG3 -TUMP]. Some lower courts have already vindicated such vagueness challenges by invalidating abortion restrictions on that ground. See, e.g., Planned Parenthood S. Atl. v. Stein, 680 F. Supp. 3d 595, 599 (M.D.N.C. 2023); Isaacson v. Brnovich, 610 F. Supp. 3d 1243, 1255 (D. Ariz. 2022). In fact, whereas Roe struck down Texas’s emergency-exempted-abortion-ban because it violated due process rights, Justice Blackmun’s initial draft of the majority decision in that case reasoned that the statute was void for vagueness because it imposed indeterminate criminal punishment against physicians for providing abortion to individual patients whose medical circumstances are complex, uncertain, and evolving. See Bernard Schwartz, The Unpublished Opinions of the Burger Court 103–19 (1988). ↑
-
. See Markovic, supra note 71, at 1519. ↑
-
. See Jamie Stengle, New York Doctor Is Fined in Texas, Charged in Louisiana over Abortion Pills in Tests of Shield Laws, AP News (Feb. 14, 2025, 2:57 PM), https://apnews.com/article/abortion-doctor-maggie-carpenter-pills-847112cde026e29333c3481310593 582[https://perma.cc/ B825-SAEY]. ↑
-
. See Markovic, supra note 71, at 1532–35. ↑
-
. See Alice Miranda Ollstein, Indiana AG Eyes Criminal Prosecution of 10-year-old Rape Victim’s Abortion Doc, Politico (July 14, 2022, 3:37 PM), https://www.politico.com/news/2022/07/ 14/indiana-abortion-rape-ohio-00045899 [https://perma.cc/EPZ3-R39E]. ↑
-
. See Alander Rocha, Alabama Attorney General Doubles Down on Threats to Prosecute Out-of-State Abortion Care, Ala. Reflector (Aug. 31, 2023, 12:21 PM), https://alabamareflector.com /2023/08/31/alabama-attorney-general-doubles-down-on-threats-to-prosecute-out-of-state-abortion-care/ [https://perma.cc/99PG-8GGR]. ↑
-
. Temporary Restraining Order, Cox v. Texas, No. D-1-GN-23-008611, 2023 WL 8628762 (Tex. Dist. Ct. Dec. 7, 2023). ↑
-
. Brendan Pierson, Texas AG Threatens to Prosecute Doctors in Emergency Abortion, Reuters (Dec. 7, 2023, 4:46 PM), https://www.reuters.com/legal/texas-judge-allows-woman-get-emergency-abortion-despite-state-ban-2023-12-07/ %5Bhttps://perma.cc/HUU62BBX%5D. The Texas Supreme Court ultimately ruled in favor of the state, blocking the previously authorized abortion. In re State, 682 S.W.3d 890, 894–95 (Tex. 2023) (per curiam). ↑
-
. See Markovic, supra note 71, at 1534 (“Antiabortion politicians have not stood idly by as district attorneys stake out positions on the enforcement of antiabortion laws.”). ↑
-
. Id.; see also Elise Catrion Gregg, Abortion Debate Spurs New Efforts To Restrict Prosecutorial Discretion, Cronkite News (Oct. 6, 2023), https://cronkitenews.azpbs.org/2023/10/06/ abortion-debate-prosecutorial-discretion-laws-executive-orders/ [https://perma.cc/S5E3-ND3H]. ↑
-
. See H.B. 17, 88th Leg., Reg. Sess. (Tex. 2023). ↑
-
. See Brittni Frederiksen, Usha Ranji, Ivette Gomez & Alina Salganicoff, KFF, A National Survey of OBGYNs’ Experiences After Dobbs (2023), https://www.kff.org/ report-section/a-national-survey-of-obgyns-experiences-after-dobbs-report/ [https://perma.cc/ 2MAG-GKGP]. ↑
-
. Sabbath et al., supra note 11, at 5. ↑
-
. See, e.g., Fund Tex. Choice v. Deski, No. 22-CV-859, 2023 WL 8856052, at *6–7 (W.D. Tex. Dec. 21, 2023); J. David Goodman, Abortion Ruling Keeps Texas Doctors Afraid of Prosecution, N.Y. Times (Dec. 14, 2023), https://www.nytimes.com/2023/12/13/us/texas-abortion-doctor-prosec
ution.html [https://perma.cc/3R9B-W8Y3]. ↑ -
. Mark Joy, Letter to the Editor, Women’s Lives Put at Risk Under State Abortion Bans, N.Y. Times (Sept. 24, 2022), https://www.nytimes.com/2022/09/24/opinion/letters/abortion-bans.html [https://perma.cc/HQ5J-P4R3]. ↑
-
. Marty Schladen, Ohio Docs Say New Abortion Law Has Them Working Against Oaths to Do No Harm, Ohio Cap. J. (Sept. 7, 2022, 4:00 AM), https://ohiocapitaljournal.com/2022/09/07/ ohio-docs-say-new-abortion-law-has-them-working-against-oaths-to-do-no-harm [https://perma.cc/44UD-USE5]. ↑
-
. The Office of the Texas Attorney General celebrating Dobbs with an agency holiday helps to illuminate Paxton’s position regarding abortion. See Press Release, Tex. Off. of the Att’y Gen., Office of the Attorney General Celebrates One Year Anniversary of the Overturning of Roe v. Wade in Observation of Sanctity of Life Day (June 23, 2023), https://www.texasattorneygeneral.gov/ news/releases/office-attorney-general-celebrates-one-year-anniversary-overturning-roe-v-wade-observation-sanctity [https://perma.cc/XAN4-X8U7]. ↑
-
. See Randye Retkin, Julie Brandfield, Ellen Lawton, Barry Zuckerman & Deanna DeFrancesco, Lawyers and Doctors Working Together – A Formidable Team, 20 Health Law. 33, 33 (2007) (describing a longstanding and deep-seated “mutual distrust and antagonism between professions”). ↑
-
. See Benjamin J. Naitove, Medicolegal Education and the Crisis in Interprofessional Relations, 8 Am. J.L. & Med. 293, 298 (1982) (discussing the failure of legal and medical professionals, respectively, to understand “each other’s methods and concepts”). ↑
-
. Strongly pro-abortion doctors could have a bias in determining the existence of a medical emergency—that’s why prosecutorial distrust of doctors isn’t necessarily off-base. And take that bias far enough and it starts looking like bad faith. But the thing is, prosecutors can’t easily tell the difference between a doctor acting in (or approaching) bad faith to resist a law she disagrees with and a doctor trying to get the law “right” even if she dislikes the law. In the cultural and political climate of states like Texas and Montana, skeptical prosecutors probably assume the former, and that’s how you get an unbalanced cliff. See Kelly K. Dineen Gillespie, Ruan v. United States: “Bad Doctors,” Bad Law, and the Promise of Decriminalizing Medical Care, 2021–2022 Cato Sup. Ct. Rev. 271, 273–74. ↑
-
. See Selena Simmons-Duffin, Doctors Who Want to Defy Abortion Laws Say It’s Too Risky, Nat’l Pub. Radio (Nov. 23, 2022, 5:01 AM), https://www.npr.org/sections/health-shots/2022/11/23/1 137756183/doctors-who-want-to-defy-abortion-laws-say-its-too-risky [https://perma.cc/3Q63-F6AY]. ↑
-
. See MiQuel Davies & Meera Rajput, Nat’l P’ship for Women & Fams. & Physicians for Reprod. Health, Dobbs’ Erosion of the Health Care Workforce: Harms to Providers and Patients (2024), https://prh.org/wp-content/uploads/2024/03/dobbs-erosion-health-care-work force.pdf [https://perma.cc/C9SG-TP9J]; Offs. of Sen. Maria Cantwell et al., Two Years Post-Dobbs: The Nationwide Impact of Abortion Bans (2024), https://www.cantwell.senate.gov/imo/ media/doc/2024_dobbs_anniversary_national_report_final.pdf [https://perma.cc/GF4K-M3AC]. ↑
-
. See Melissa Suran, As Laws Restricting Health Care Surge, Some US Physicians Choose Between Fight or Flight, 329 JAMA 1899, 1901–03 (2023) (quoting one interviewee on how healthcare workers “figure out how to get work done that’s important to them, no matter where they are, but the way they approach it might have to be different”). ↑
-
. Lori Freedman, Bishops and Bodies: Reproductive Care in American Catholic Hospitals 157 (2023). ↑
-
. See Whitney Arey et al., A Preview of the Dangerous Future of Abortion Bans—Texas Senate Bill 8, 387 New Eng. J. Med. 388, 389 (2022). ↑
-
. Rachael Robertson & Kristina Fiore, Docs in Abortion Ban States Find Workarounds to Deliver Care, MedPage Today (June 24, 2024), https://www.medpagetoday.com/special-reports/ exclusives/110801 [https://perma.cc/8FE8-QM6Z]. ↑
-
. Nilo Tabrizy, Kassie Bracken, Mark Boyer & Mariam Dwedar, “Do No Harm”: OB-GYNs Weigh the Legal Impact of Abortion Bans, N.Y. Times, at 04:57 (Sept. 10, 2022), https://www.
nytimes.com/video/us/100000008489880/abortion-bans-maternal-health.html (last visited July 19, 2025) (quoting Dr. Rebecca Cohen). ↑ -
. See Kate Zernike, Medical Impact of Roe Reversal Goes Well Beyond Abortion Clinics, Doctors Say, N.Y. Times (Sept. 10, 2022), https://www.nytimes.com/2022/09/10/us/abortion-bans-medical-care-women.html [https://perma.cc/V5QN-AMMT] (“‘There’s such confusion,’ said Dr. Allison Linton, an obstetrician in Milwaukee, ‘and when doctors are hearing this risk of a felony charge, they’re erring on the side of fear.’”). ↑
-
. Chavi Eve Karkowsky, Opinion, New Abortion Laws Are Especially Cruel to My Patients with High-Risk Pregnancies, Macomb Daily, (June 17, 2021, 5:12 AM), https://www. macombdaily.com/2019/05/25/new-abortion-laws-are-especially-cruel-to-my-patients-with-high-risk-pregnancies [https://perma.cc/RDA8-87TA]. ↑
-
. Schladen, supra note 133. ↑
-
. Daniel Grossman et al., Advancing New Standards in Reprod. Health & Tex. Pol’y Evaluation Project, Care Post-Roe: Documenting Cases of Poor-Quality Care Since the Dobbs Decision (2023), https://www.ansirh.org/sites/default/files/2023-05/Care%20Post-Roe%20Preliminary%20Findings.pdf [https://perma.cc/KRN5-6JX6]. ↑
-
. Sabbath et al., supra note 11, at 3. ↑
-
. We borrow the term “hesitant medicine” from Anna-Grace Lilly, Isabelle P. Newman & Sophie Bjork-James, Our Hands Are Tied: Abortion Bans and Hesitant Medicine, 350 Social Sci. & Med., June 2024, art. no. 116912. ↑
-
. See, e.g., Ariana Eunjung Cha, Physicians Face Confusion and Fear in Post-Roe World, Wash. Post (June 28, 2022), https://www.washingtonpost.com/health/2022/06/28/abortion-ban-roe-doctors-confusion/ [https://perma.cc/XZH7-2P6D]; Julia Luchetta, Idaho’s Biggest Hospital Says Emergency Flights for Pregnant Patients Up Sharply, Nat’l Pub. Radio (Apr. 26, 2024, 8:33 AM), https://www.npr.org/2024/04/25/1246990306/more-emergencyflights-for-pregnant-patients–in-idaho [https://perma.cc/U3A9-8BDR]; Nadine El-Bawab, Tess Scott, Christina Ng & Acacia Nunes, Delayed and Denied: Women Pushed to Death’s Door for Abortion Care in Post-Roe America, ABC News (Dec. 14, 2023, 6:09 AM), https://abcnews.go.com/US/delayed-denied-women-pusheddeaths-door-abortion-care/story?id=105563255 [https://perma.cc/GQ2Z-CWUN]. ↑
-
. See Lizzie Presser, Andrea Suozzo, Sophie Chou & Kavitha Surana, Texas Banned Abortion. Then Sepsis Rates Soared, ProPublica (Feb. 20, 2025, 5:00 AM), https://www.propublica. org/article/texas-abortion-ban-sepsis-maternal-mortality-analysis [https://perma.cc/ 6DWL-M72E]. ↑
-
. See Cassandra Jaramillo, Kavitha Surana, Lizzie Presser & Ziva Branstetter, Texas Lawmakers Push for New Exceptions to State’s Strict Abortion Ban After the Deaths of Two Women, ProPublica (Nov. 20, 2024, 5:00 AM), https://www.propublica.org/article/texas-abortion-ban-exceptions-deaths [https://perma.cc/2AAR-5WS6]; Lizzie Presser & Kavitha Surana, A Third Woman Died Under Texas’ Abortion Ban. Doctors Are Avoiding D&Cs and Reaching for Riskier Miscarriage Treatments, ProPublica (Nov. 25, 2024, 6:00 AM), https://www.propublica.org/article/porsha-ngumezi-miscarriage-death-texas-abortion-ban [https://perma.cc/H8YG-L5LB]. ↑
-
. Kavitha Surana, Abortion Bans Have Delayed Emergency Medical Care. In Georgia, Experts Say This Mother’s Death Was Preventable., ProPublica (Sept. 16, 2024, 5:00 AM), https://www.propublica.org/article/georgia-abortion-ban-amber-thurman-death [https://perma.cc/7EKF
-9DBX]; see also Kavitha Surana, Afraid to Seek Care amid Georgia’s Abortion Ban, She Stayed at Home and Died, ProPublica (Sept. 18, 2024, 6:00 AM), https://www.propublica.org/article/candi-miller-abortion-ban-death-georgia [https://perma.cc/KH9S-F6FB]. ↑ -
. Nambiar et al., supra note 12, at 648–49. ↑
-
. Texas Heartbeat Act, 2021 Tex. Gen. Laws ch. 62 (codified in Tex. Health & Safety Code Ann. §§ 171.201–.212 (West 2023)). ↑
-
. Nambiar et al., supra note 12, at 649. ↑
-
. Id. ↑
-
. See id. ↑
-
. See Dan Keating, Tim Meko & Danielle Rindler, Abortion Access Is More Difficult for Women in Poverty, Wash. Post (July 10, 2019), https://www.washingtonpost.com/national/2019/07/ 10/abortion-access-is-more-difficult-women-poverty [https://perma.cc/MJP2-X4F3]; Taylor Johnson & Kelsey Butler, Abortion Desert in US South Is Hurting Black Women the Most, Bloomberg News (Aug. 23, 2022, 6:00 AM), https://www.bloomberg.com/news/articles/2022-08-23/black-women-are-hardest-hit-by-abortion-restrictions-sweeping-the-deep-south [https://perma.cc/9AVN-MZA7]; Nada Hassanein, Inequities in Maternal Health Care Access Are Not New. They Have Deep Roots in History, USA Today (Dec. 16, 2022, 11:22 AM), https://www.usatoday.com/in-depth/news/health/2022/08/ 11/prenatal-maternal-health-care-people-color-history/10084878002 [https://perma.cc/NA9L-ZYEV]. ↑
-
. Sabbath et al., supra note 11, at 5. ↑
-
. Cf. Davies & Rajput, supra note 139, at 3; Cantwell et al., supra note 139, at 20. ↑
-
. See Frederiksen et al., supra note 129, at 13. ↑
-
. See Sabbath et al., supra note 11, at 3–6. ↑
-
. Id. at 5. ↑
-
. Id. at 3–6. ↑
-
. Laura Ungar, Why Some Doctors Stay in US States with Restrictive Abortion Laws and Others Leave, ABC News (June 22, 2023, 6:16 PM), https://abcnews.go.com/HealthwireStory/doctors-stay-us-states-restrictive-abortion-laws-leave-100329329 [https://perma.cc/7PAR-SHDN]. ↑
-
. About: The Scope of Practice of Obstetrics and Gynecology, ACOG, https://www. acog.org/about [https://perma.cc/C9DN-ZZLA]. ↑
-
. See Hillary J. Gyuras, Meredith P. Field, Olivia Thornton, Danielle Bessett & Michelle L. McGowan, The Double-Edged Sword of Abortion Regulations: Decreasing Training Opportunities While Increasing Knowledge Requirements, 28 Med. Educ. Online, issue no. 1, 2023, art. no. 2145104; Rosemary Westwood, Standard Pregnancy Care Is Now Dangerously Disrupted in Louisiana, Report Reveals, Nat’l Pub. Radio (Mar. 19, 2024, 5:01 AM), https://www.npr.org/sections/health-shots/
2024/03/19/1239376395/louisiana-abortion-ban-dangerously-disrupting-pregnancy-miscarriage-care [https://perma.cc/FAC5-B3A5]. ↑ -
. See Preetha Nandi, Danielle M. Roncari, Erika F. Werner, Allison L. Gilbert & Sebastian Z. Ramos, Navigating Miscarriage Management Post-Dobbs: Health Risks and Ethical Dilemmas, 34 Women’s Health Issues 449, 451–52 (2024); Cantwell et al., supra note 139, at 18. ↑
-
. See supra notes 3–4 and accompanying text. If an officer shoots someone not because she feels threatened but, say, for refusing to pay her a bribe, that’s murder but not cliff running. ↑
-
. See Thomas Owen Baker, Opinion, As a Cop, I Killed Someone. Then I Found Out It Happens More Often Than We Know, The Guardian (July 31, 2020, 6:30 AM), https://www.the guardian.com/commentisfree/2020/jul/31/as-a-cop-i-killed-someone-then-i-found-out-it-happens-more -often-than-we-know [https://perma.cc/3M9E-VY6R]. ↑
-
. To be sure, it is only an officer’s criminal liability that is binary in the sense we describe. Officers may also face disciplinary and civil liability consequences for using force, sanctions which can be graduated rather than all-or-nothing. Note also that there’s hardly any legal risk for failing to use force when it was warranted, for example, by freezing rather than disabling a suspect who goes on to injure or kill a bystander or fellow cop. The reason is not so much because of qualified immunity but rather because there’s not a cause of action in the first place. See, e.g., Town of Castle Rock v. Gonzalez, 545 U.S. 748, 760 (2005). ↑
-
. See Seth W. Stoughton, Jeffrey J. Noble & Geoffrey P. Alpert, Evaluating Police Uses of Force 71 (2020) (identifying forty-two states with police use of force statutes); see also Cynthia Lee, Officer-Created Jeopardy: Broadening the Time Frame for Assessing a Police Officer’s Use of Deadly Force, 89 Geo. Wash. L. Rev. 1362, 1378 (2021). ↑
-
. See Lee, supra note 174, at 1383–84 (“The vast majority of state statutes on police use of force allow an officer to use deadly force against a civil if the officer reasonably believed deadly force was necessary to effectuate an arrest, prevent the escape of a felon, or protect the officer or others.”); see also Stoughton et al., supra note 174, at 73–79. We focus on the standards that govern criminal cases. A separate standard governs constitutional tort claims. See Tennessee v. Garner, 471 U. S. 1 (1985); Graham v. Connor, 490 U.S. 386, 396 (1989). While the criminal and tort standards are often discussed interchangeably, they are distinct. See Kate Stith-Cabranes, Criminal Law and the Supreme Court: An Essay on the Jurisprudence of Byron White, 74 U. Colo. L. Rev. 1523, 1554–56 (2003) (explaining the distinction); see also Stoughton et al., supra note 174, at 68–71 (same). ↑
-
. See Stoughton et al., supra note 174, at 86–87 (“Thirty-four states have statutes that adopt a ‘subjective objectivity’ approach by authorizing lethal force when it is reasonably necessary or reasonably perceived as necessary for at least some statutory purpose, or when the officer reasonably believes that certain statutory justifications are present . . . .”); see also Cynthia Lee, Reforming the Law on Police Use of Deadly Force: De-Escalation, Pre-Seizure Conduct, and Imperfect Self-Defense, 2018 U. Ill. L. Rev. 629, 654–55 (2018). ↑
-
. See Lee, supra note 174, at 1383–84. ↑
-
. Lee, supra note 176, at 654–55. Although Lee was discussing the standard’s ambiguity for purposes of jurors’ ex-post evaluations, it is just as ambiguous to the officer in the moment. ↑
-
. See Rachel Moran, Police Go to Court: Police Officers as Witnesses/Defendants, 19 Ann. Rev. L. & Soc. Sci. 93, 100 (2023) (“[P]rosecutors historically have been extremely reluctant to charge police officers with crimes[.]”); Brandon Garrett & Christopher Slobogin, The Law on Police Use of Force in the United States, 21 German L.J. 1526, 1532 (2020) (“[T]he imposition of criminal sanctions on police officers remains rare.”); Kate Levine, How We Prosecute the Police, 104 Geo. L.J. 745, 763–64 (2016) (observing that “police are rarely charged for criminal behavior that most know is commonplace”). ↑
-
. Wesley G. Skogan & Tracey L. Meares, Lawful Policing, 593 Annals Am. Acad. Pol. & Soc. Sci. 66, 73 (2004) (“This is a difficult research area. There is no national repository of data on police use of force, and access to local records is difficult.”). ↑
-
. See Kimberly Kindy & Kimbriell Kelly, Thousands Dead, Few Prosecuted, Wash. Post (Apr. 11, 2015), https://www.washingtonpost.com/sf/investigative/2015/04/11/thousands-dead-few-prosecuted/ [https://perma.cc/NW6X-L47D]. ↑
-
. Id. Below, we (skeptically) consider the hypothesis that fifty-four officers were charged in this time period because that’s how many unjustified police killings occurred. See infra notes 233–36 and accompanying text. ↑
-
. Id. ↑
-
. Id. ↑
-
. We can very roughly calculate the odds. Around 1,000 people are fatally shot by police each year. See Fatal Force, Wash. Post. (Dec. 31, 2024), https://www.washingtonpost.com/ graphics/investigations/police-shootings-database/ [https://perma.cc/Y27U-MBF8]. That means that in the ten-year period included in the Post story, in the neighborhood of 10,000 people were killed by police gunfire. Of the 54 charges found by the Post, 19 cases were still unresolved. For sake of discussion, assume that half of those cases resulted in convictions. Putting them together with the 14 convictions reported by the Post, that makes for 24 convictions out of 10,000 shootings, or 0.24%. Thus, a police officer is more likely to flip a coin ‘heads’ eight times in a row than to be convicted after an on-duty shooting. See Maciej Kowalski, Coin Flip Probability Calculator, Omni Calculator (July 25, 2024), https://www.omnicalculator.com/statistics/coin-flip-probability [https://perma.cc/XPP7-9MUC]. ↑
-
. See Rachel Harmon, Law and Orders, 123 Colum. L. Rev. 1, 18 n.92 (2023) (“Police-specific justification statutes authorize defensive force more expansively than most self-defense and defense-of-others statutes because police statutes permit officers to initiate confrontations and do not require retreat.”). ↑
-
. See Lee, supra note 176, at 656–61. ↑
-
. See Lee, supra note 174, at 1383–84 (“A few states allow an officer to use deadly force based solely on the officer’s honest belief that deadly force was necessary without also requiring that the officer’s belief was objectively reasonable.”). ↑
-
. See William Ortman, Confession and Confrontation, 113 Calif. L. Rev. 377, 407–14 (2025) (describing inadequacies in existing constitutional law governing interrogations). ↑
-
. See id. at 399–404 (describing the “Reid technique” of interrogation). ↑
-
. Kate Levine, Police Suspects, 116 Colum. L. Rev. 1197, 1200 (2016). ↑
-
. Id. at 1227. ↑
-
. See, e.g., Chesa Boudin, Prosecutors, Power, and Justice: Building an Anti-Racist Prosecutorial System, 73 Rutgers U. L. Rev. 1325, 1331 (2021) (“Too often, and for generations across the country, prosecutors have been reluctant or refused outright to seriously investigate or consider filing criminal charges against police officers who killed Black people or Brown people or Asian people, no matter how egregious the conduct.”); Jason Mazzone & Stephen Rushin, From Selma to Ferguson: The Voting Rights Act as a Blueprint for Police Reform, 105 Calif. L. Rev. 263, 278 (2017) (“Prosecutors, dependent upon their own relationships with local law enforcement, are often reluctant to pursue criminal charges against the police.”); Robert M. Myers, Code of Silence: Police Shootings and the Right to Remain Silent, 26 Golden Gate U. L. Rev. 497, 505 (1996) (“[I]t is widely recognized that prosecutors are reluctant to bring charges against police officers.”). ↑
-
. See Levine, supra note 179, at 763 (explaining that in “high-stakes cases” prosecutors “must consider criminal charges against police suspects, either because of their own belief that a crime may have occurred or because of public scrutiny”). ↑
-
. Id. at 755–56. ↑
-
. See id. at 766–67 (describing grand jury process employed by St. Louis prosecutor Robert McCullough after Michael Brown was killed by Darren Wilson); see also Epps, supra note 79, at 812–13 (describing such uses of the grand jury process as “troubling”). ↑
-
. See Andrew D. Leipold, Why Grand Juries Do Not (and Cannot) Protect the Accused, 80 Cornell L. Rev. 260, 263 (1995) (“The grand jury is frequently criticized for failing to act as a meaningful check on the prosecutor’s charging decisions; according to the clichés it is a ‘rubber stamp,’ perfectly willing to ‘indict a ham sandwich’ if asked to do so by the government.”). ↑
-
. See supra notes 78–79 and accompanying text. ↑
-
. It’s also possible that prosecutors are simply bowing to political pressure from police unions not to prosecute their members. See Kate Levine, Who Shouldn’t Prosecute the Police, 101 Iowa L. Rev. 1447, 1465 (2016) (“Almost no criminal case exists without the police as the first contact point.”). ↑
-
. See supra notes 80–83 and accompanying text. ↑
-
. See Levine, supra note 199, at 1465 (“Almost no criminal case exists without the police as the first contact point.”). ↑
-
. See Somil Trivedi & Nicole Gonzalez Van Cleve, To Serve and Protect Each Other: How Police-Prosecutor Codependence Enables Police Misconduct, 100 B.U. L. Rev. 895, 909 (2020) (“From a structural standpoint, police officers are prosecutors’ star witnesses . . . .”). ↑
-
. Kami Chavis Simmons, Increasing Police Accountability: Restoring Trust and Legitimacy Through the Appointment of Independent Prosecutors, 49 Wash. U. J.L. & Pol’y 137, 145 (2015). ↑
-
. Trivedi & Van Cleve, supra note 202, at 900–01. ↑
-
. See Rosseau et al., supra note 80, at 396 (noting that “trust can be the result of deep dependence and identity formation”). ↑
-
. Levine, supra note 199, at 1470. ↑
-
. See Daniel C. Richman, Prosecutors and Their Agents, Agents and Their Prosecutors, 103 Colum. L. Rev. 749, 786–87 (2003). For an account of how their membership in the legal profession affects prosecutors, see William Ortman, The Prosecution Bar, 101 Wash. U. L. Rev. 123, 136–50 (2023). ↑
-
. See Richman, supra note 207, at 787–88. ↑
-
. Bruce A. Green & Rebecca Roiphe, Rethinking Prosecutors’ Conflicts of Interest, 58 B.C. L. Rev. 463, 508 (2017). ↑
-
. Trivedi & Van Cleve, supra note 202, at 905. ↑
-
. Richman, supra note 207, at 792. ↑
-
. Id. ↑
-
. See supra notes 179–85 and accompanying text. ↑
-
. See Joel F. Shults, What Officers Said Were the Biggest Challenges of 2023, Police1 (Dec. 15, 2023, 4:12 PM), https://www.police1.com/year-in-review/what-officers-said-were-the-biggest-challenges-of-2023 [https://perma.cc/6GBJ-S4D7]. ↑
-
. See Press Release, Minn. Police & Peace Officers Ass’n, MPPOA Releases 2024 Membership Survey Results (Mar. 27, 2024), https://www.mppoa.com/images/Survey%20-%20March %202024.pdf [https://perma.cc/RR2T-TTCQ]. ↑
-
. See Moran, supra note 179, at 93 (noting that police have “increasingly found themselves as defendants facing criminal charges”). ↑
-
. See Martin Kaste, Are More Police Officers Facing Prosecution?, Nat’l Pub. Radio (Sept. 25, 2023, 4:36 PM), https://www.npr.org/2023/09/25/1201620935/are-more-police-officers-facing-prosecution-as-the-data-shows-its-complicated [https://perma.cc/KP7A-XR74]. ↑
-
. Id.; see also Eliza Fawcett & Tim Arango, Liberal Prosecutors Are Revisiting Police Killings but Charging Few Officers So Far, N.Y. Times (June 8, 2023), https://www.nytimes.com/2023/06/08/ us/police-killings-charges.html [https://perma.cc/Y73D-5G5M](“[L]iberal prosecutors campaigned on the promise that they would review cases that they felt were hastily closed without charges. . . . [b]ut the re-examinations so far have rarely led to criminal charges.”). ↑
-
. For an overview of the movement, see generally Benjamin Levin, Essay, Imagining the Progressive Prosecutor, 105 Minn. L. Rev. 1415 (2021). ↑
-
. See Bruce A. Green & Rebecca Roiphe, A Fiduciary Theory of Progressive Prosecution, 60 Am. Crim. L. Rev. 1431, 1432 (2023) (“Many progressive prosecutors have faced pushback from judges, other public officials, police, the media, constituents, and subordinate prosecutors.”); Brooks Holland & Steven Zeidman, Progressive Prosecutors or Zealous Defenders, from Coast-to-Coast, 60 Am. Crim. L. Rev. 1467, 1479 (2023) (“In addition to opposition from office staff, local police, and the judiciary, progressive prosecutors have faced an onslaught of hostility from elected officials.”). ↑
-
. See Vida B. Johnson, Whom Do Prosecutors Protect?, 104 B.U. L. Rev. 289, 337 (2024) (“While most traditional prosecutors protect police at almost all costs, some ‘progressive prosecutors’ have run on platforms of police accountability and other reformist goals.”); Levin, supra note 219, at 1440 (“[M]any progressive prosecutors have sought to make their name or to stake their political claim by adopting a tough line against police officers accused of using excessive force against civilians.”); Paul Butler, Progressive Prosecutors Are Not Trying to Dismantle the Master’s House, and the Master Wouldn’t Let Them Anyway, 90 Fordham L. Rev. 1983, 1988–89 (2022) (“Progressive prosecutors push for reform from within the criminal legal system, including by making commitments to reduce incarceration, hold police officers accountable, and reallocate funds to public services.”). ↑
-
. See sources cited supra note 218. ↑
-
. See Maneesh Arora, How the Coronavirus Pandemic Helped the Floyd Protests Become the Biggest in U.S. History, Wash. Post. (Aug. 5, 2020), https://www.washingtonpost.com/politics/ 2020/08/05/how-coronavirus-pandemic-helped-floyd-protests-become-biggest-us-history/ [https:// perma.cc/P3BE-LTCD] (noting that “an estimated 15 million to 26 million Americans have taken to the streets to protest police violence and advocate for Black lives”). ↑
-
. See Mark Berman, How Derek Chauvin Became the Rare Police Officer Convicted of Murder, Wash. Post (Apr. 20, 2021), https://www.washingtonpost.com/nation/2021/04/20/chauvin-police-officer/ [https://perma.cc/J3WB-XZS5]. ↑
-
. In a June 2020 Pew poll, only thirty-five percent of respondents said that police did an excellent or good job of “using the right amount of force for each situation.” See Hannah Hartig & Andrew Daniller, Before Release of Video Showing Tyre Nichols’ Beating, Public Views of Police Conduct Had Improved Modestly, Pew Rsch. Ctr. (Feb. 3, 2023), https://www.pewresearch.org/short-reads/2023/02/03/before-release-of-video-showing-tyre-nichols-beating-public-views-of-police-conduct-had-improved-modestly/ [https://perma.cc/K47Y-49RC]. ↑
-
. See John Eligon & Shaila Dewan, Chauvin Verdict Brings the Police Relief and Some Resentment, N.Y. Times (Apr. 22, 2021), https://www.nytimes.com/2021/04/21/us/george-floyd-verdict-police-reaction.html [https://perma.cc/4WEZ-WSFP] (explaining that “rank-and-file” officers had “complicated” feelings about the verdict: “a mix of relief, resentment at being vilified alongside Mr. Chauvin and unsettling thoughts of themselves in his shoes”). ↑
-
. Jon Schuppe, Police Across U.S. Respond to Derek Chauvin Trial: ‘Our American Way of Policing Is on Trial’, NBC News (Apr. 15, 2021, 4:37 PM), https://www.nbcnews.com/news/us-news/police-across-u-s-respond-derek-chauvin-trial-our-american-n1264224 [https://perma.cc/449Z-8CZZ]. ↑
-
. To be sure, the progressive prosecutor movement and the protests of 2020 are hardly the only explanations for why police distrust prosecutors. Another is that police culture, as Seth Stoughton explains, “inculcate[s] an ‘us versus them’ mentality that separates officers from non-officers.” Seth Stoughton, Evidentiary Rulings as Police Reform, 69 U. Miami L. Rev. 429, 451 (2015). While we have emphasized connections between police and prosecutors, the groups remain culturally differentiated. See Richman, supra note 207, at 786 (“At first blush, it would seem that, whatever their legal and institutional obligations to work together, prosecutors and agents would be pushed apart by their membership in distinct, even antagonistic, professional cultures.”). We are grateful to Stoughton for pushing us on this point. ↑
-
. See supra note 215 and accompanying text. ↑
-
. See supra notes 87–88 and accompanying text. ↑
-
. See, e.g., Deborah Tuerkheimer, Criminal Justice and the Mattering of Lives, 116 Mich. L. Rev. 1145, 1151 (2018) (book review) (observing that the police killings of Philando Castile, Tamir Rice, Natasha McKenna, Sean Bell, Eric Garner and others “have sparked protest locally and around the nation”); Jelani Jefferson Exum & David Niven, Where Black Lives Matter Less: Understanding the Impact of Black Victims on Sentencing Outcomes in Texas Capital Murder Cases from 1973 to 2018, 66 St. Louis U. L.J. 677, 680 (2022) (“With regard to police killings, the lesson [from George Floyd’s murder] that police officers disproportionately kill Black people in this country with impunity because our system of policing encourages such violence, and our legal jurisprudence protects that use of violence.”); see also Kate Levine, Police Prosecutions and Punitive Instincts, 98 Wash. U. L. Rev. 997, 1009 (2021) (“Failures to secure charges and convictions [against police officers] have led reform activists to question whether prosecutors are living up to their duty as the ‘guardians of public safety,’ to their duty to do justice, and to their ability to carry out their mandate to charge fairly.”). ↑
-
. See sources cited supra notes 223–24. ↑
-
. See supra text accompanying notes 181–84. ↑
-
. Jeffrey Fagan & Alexis D. Campbell, Race and Reasonableness in Police Killings, 100 B.U. L. Rev. 951, 961 (2020). There could be slippage between the Washington Post data and the conclusion that shootings are unjustified. For one thing, geographical patterns of police patrol activity may skew data. See id. at 1003–04. Beyond that, a skeptic might think that the racial evidence may not prove that cops routinely or often use deadly force when it isn’t justified by theorizing that cops might often avoid deadly force even when the law would permit it, and that the reasons they do so are correlated with race. Just because cops shoot black people more often than identically situated white people doesn’t necessarily show that the shooting of many black people is unjustified. Instead, it could suggest that police more often don’t shoot white people even when the law would permit it. That’s possible though not more plausible than the alternative. At any rate, the Washington Post data and Fagan and Campbell’s analysis of it are the most reliable evidence and interpretation available. ↑
-
. Frank Edwards, Hedwig Lee & Michael Esposito, Risk of Being Killed by Police Use of Force in the United States by Age, Race–Ethnicity, and Sex, 116 Proc. Nat’l Acad. Sci. 16793, 16793 (2019). ↑
-
. See Curtis Bunn, Report: Black People Are Still Killed by Police at a Higher Rate Than Other Groups, NBC News (Mar. 3, 2022, 7:14 PM), https://www.nbcnews.com/news/nbcblk/ report-black-people-are-still-killed-police-higher-rate-groups-rcna17169 [https://perma.cc/DXF3-ZDNP]; Neil Gross, 3 Years After George Floyd’s Murder, Cop Culture Still Hasn’t Changed, Time (May 25, 2023, 7:00 AM), https://time.com/6282369/george-floyds-murder-cop-culture-hasnt-changed/ [https://perma.
cc/5TGP-YLZW]; Emily Washburn, America Less Confident in Police Than Ever Before: A Look at the Numbers, Forbes (Feb. 3, 2023, 4:04 PM), https://www.forbes.com/sites/emilywashburn/2023/02
/03/america-less-confident-in-police-than-ever-before-a-look-at-the-numbers/ [https://perma.cc/8UXE-DUVT]. ↑ -
. See Plaintiffs-Appellees’ Response Brief, State v. Zurawski, 690 S.W.3d 644 (Tex. 2024) (No. 23-0629), 2023 WL 7105779. ↑
-
. Id. at 33. ↑
-
. Transcript of Oral Argument, State v. Zurawski, 690 S.W.3d 644 (Tex. 2024) (No. 23-0629), 2023 WL 8360124. A unanimous court declined to respond and refused to offer any more detail to doctors whose patients have medically complicated pregnancies. ↑
-
. The distinction between law in books and law in action comes from Roscoe Pound, Law in Books and Law in Action, 44 Am. L. Rev. 12, 34 (1910). ↑
-
. See supra notes 32–47 and accompanying text. ↑
-
. See Sections II.A–B. ↑
-
. One could imagine the same logic applying to doctors performing abortions for sick patients. If Dr. Jones is trustworthy, her decision to perform an emergency abortion must be reasonable, or she wouldn’t have done it. Needless to say, doctors in abortion-ban states do not expect such trust. See supra notes 135–38 and accompanying text. ↑
-
. That proposition was a tenet of the Legal Realist movement, which “insisted that accurate description of the ‘real rules’ that govern society cannot be captured by parsing the words of statutes or the written opinions of appellate judges, but must include study of the way that ‘paper rules’ are interpreted, implemented, applied, or ignored as they are carried into practice.” Katherine R. Kruse, The Jurisprudential Turn in Legal Ethics, 53 Ariz. L. Rev. 493, 498 (2011). ↑
-
. See supra notes 34–38 and accompanying text. ↑
-
. See supra Section II.B. ↑
-
. See supra Section II.A. ↑
-
. See Pierson, supra note 125. ↑
-
. The policing parallel may be more familiar: a questionable officer killing that goes unprosecuted, where family members of the deceased grow frustrated by inaction by lawmakers or others who point the blame at other forces or victims themselves. ↑
-
. State v. Zurawski, 690 S.W.3d 644, 653 (Tex. 2024) (“A physician who tells a patient, ‘Your life is threatened by a complication that has arisen during your pregnancy, and you may die, or there is a serious risk you will suffer substantial physical impairment unless an abortion is performed,’ and in the same breath states ‘but the law won’t allow me to provide an abortion in these circumstances’ is simply wrong in that legal assessment.”). ↑
-
. Cf. The Federalist No. 70 (Alexander Hamilton) (“It often becomes impossible, amidst mutual accusations, to determine on whom the blame or the punishment of a pernicious measure . . . ought really to fall.”). ↑
-
. See supra notes 66–67 and accompanying text. ↑
-
. In making them, we remain agnostic about lawmakers’ substantive legislative choices. That is not to say that we lack views about the desirability of underlying prohibitions or justification defenses. Rather, it is that our substantive views are immaterial here. ↑
-
. See Dov Fox, Medical Disobedience, 136 Harv. L. Rev. 1030, 1083–84 (2023). ↑
-
. See id.; see also Jacob Schuman, Sentencing Rules and Standards: How We Decide Criminal Punishment, 83 Tenn. L. Rev. 1, 60 (2015). ↑
-
. See Tex. Health & Safety Code Ann. § 170A.004 (West 2023); Tex. Penal Code Ann. §§ 12.32–.33 (West 2023). ↑
-
. See, e.g., Eisenberg, supra note 25, at 151; Adam J. Kolber, The Bumpiness of Criminal Law, 67 Ala. L. Rev. 855, 875–76 (2016). To be sure, criminal homicide laws typically do have incremental steps, from manslaughter up through first-degree murder. See Joshua Dressler, Understanding Criminal Law 506–07 (7th ed. 2015). As even involuntary manslaughter convictions carry the potential for years in prison, the starting point is already a very high cliff. ↑
-
. See generally Joanna Schwartz, Shielded: How the Police Became Untouchable, at xii (2023) (“Internal affairs divisions in departments across the country have been criticized for conducting shoddy investigations—failing to interview witnesses, collect and preserve evidence, reconcile inconsistent statements, or question when officers submit essentially verbatim statements that suggest collusion.”); id. at xiii (“Courts, legislatures, and officials at every level of government have created so many protections for police officers and other government officials, at every stage of litigation, that a person who has had their life shaken to the very core by government misconduct can have the courthouse doors shut in their face.”). ↑
-
. See supra notes 26–31 and accompanying text. ↑
-
. See Levine, supra note 231, at 1004 (“[I]t is hard to find anyone, other than police advocates, and a few of the most dedicated prison abolitionists, vocally opposing the prosecution of police.”); see also Thomas Ward Frampton, Essay, The Dangerous Few: Taking Seriously Prison Abolition and Its Skeptics, 135 Harv. L. Rev. 2013 (2022) (cataloging and analyzing abolitionist responses to the “problem of the dangerous few”). ↑
-
. See Melanie Kalmanson, Death After Dobbs: Addressing the Viability of Capital Punishment for Abortion, 29 Wm. & Mary J. Race, Gender & Soc. Just. 545, 551 (2023) (“Several states have proposed legislation making abortion punishable as a homicide—which is consistent with the anti-abortion movement’s narrative that abortion is murder and the movement’s end goal of establishing personhood for fetuses.”). ↑
-
. See supra Sections I.C.3–4. ↑
-
. See Jacob E. Gersen & Adrian Vermeule, Chevron as a Voting Rule, 116 Yale L.J. 676, 680–81 (2007). ↑
-
. For an overview of the theory of second-best, see William Ortman, Second-Best Criminal Justice, 96 Wash. U. L. Rev. 1061, 1093–96 (2019). ↑
-
. Id. at 1094. ↑
-
. See id. at 1095. ↑
-
. See Fox, supra note 254, at 1082. ↑
-
. 597 U.S. 450 (2022). In an incisive student note, Mary Claire Bartlett argues that state supreme courts should apply Ruan’s guidance to emergency medical exceptions to abortion bans by likewise imposing subjective mens rea requirements. See Mary Claire Bartlett, Note, Physician Mens Rea: Applying United States v. Ruan to State Abortion Statutes, 123 Colum. L. Rev. 1699, 1732–34 (2023). ↑
-
. This is not just theoretical. In 2019, California responded to the killing of an unarmed person by enacting legislation adjusting the liability rule for police officers invoking self-defense. See Levine, supra note 231, at 1019–20 (describing the legislation). To be sure, there are reasons why legislators might be wise to avoid creating special procedures and liability rules that apply only to police. See id. at 1006 (“In order to re-envision the criminal legal system, we must seek alternatives, even when fairness or racial justice appears to demand a criminal legal recourse,” including for “those who have historically been privileged by criminal law and procedure.”). ↑
-
. See Ortman, supra note 264, at 1064. ↑
-
. For evidence that even modest changes to liability rules can have major real-world effects, consider New Zealand homicide law. When lawmakers in New Zealand decided that their negligence standard for manslaughter made it too easy to prosecute doctors whose mistakes resulted in a patient’s death, they changed the law to require a “‘major departure’ from the standard of care expected of a reasonable person.” Michelle Robson, Jon Maskill & Warren Brookbanks, Doctors Are Aggrieved—Should They Be? Gross Negligence Manslaughter and the Culpable Doctor, 84 J. Crim. L. 312, 323 (2020); see also P.D.G. Skegg, Criminal Prosecutions of Negligent Health Professionals: The New Zealand Experience, 6 Med. L. Rev. 220, 221 (1998) (explaining that pre-reform standard required only ordinary negligence). While that may not appear to be a seismic change in liability standards on paper, it was in practice. After the reform, medical manslaughter prosecutions in New Zealand “all but ceased.” Robson et al., supra, at 323 (quoting P.D.G. Skegg, Medical Acts Hastening Death, in Health Law in New Zealand 616 (2015)). ↑
-
. See Gersen & Vermeule, supra note 263, at 681. ↑
-
. See Levine, supra note 199, at 1487. ↑
-
. See Stoughton et al., supra note 174, at 59 (observing that a “prosecutor may decide to handle [a] case in-house, or the close working relationship between the local prosecutor and the police agency may lead the prosecutor to refer the case to the state attorney general or to a ‘special prosecutor’ or an ‘independent prosecutor’ to avoid the appearance of a conflict of interest”); see also Levine, supra note 196, at 1487–96 (analyzing a variety of mechanisms for prosecution of police by others than local prosecutors). ↑
-
. See supra note 68 and accompanying text. ↑
-
. James N. Thompson & Lisa A. Robin, State Medical Boards: Future Challenges for Regulation and Quality Enhancement of Medical Care, 33 J. Legal Med. 93, 96 (2012). ↑
-
. For discussion of this model in the adjacent context of investigating doctors accused of inappropriately prescribing opioids, see Michael C. Barnes, Taylor J. Kelly & Christopher M. Piemonte, Demanding Better: A Case for Increased Funding and Involvement of State Medical Boards in Response to America’s Drug Abuse Crisis, 106 J. Med. Regul. 6 (2020). ↑
-
. Michael Barnes, A More Sensible Surge: Ending DOJ’s Indiscriminate Raids of Healthcare Providers, 8 Legis. & Pol’y Brief 7, 26–27 (2019) (detailing why state medical licensing boards are so uniquely well-equipped and oriented to make these decisions). ↑
-
. Id. at 28. Kentucky has an analogous procedural mechanism on its books. The state has a pre-Dobbs statutory ban on so-called “partial-birth” abortions, subject to an exception “if the partial-birth abortion was necessary to save the life of the mother whose life was endangered by a physical disorder, illness, or injury.” Ky. Rev. Stat. Ann. § 311.990(11)(a)(1) (West 2025). The statute provides that a “physician may seek a hearing before the State Board of Medical Licensure” about whether the exception applied, and, if the Board makes findings, they “shall be admissible at the trial of the physician.” Id. § 311.990(11)(a)(2). The Board’s decision in favor of a doctor thus apparently isn’t de jure “final,” as it would be in our proposal. But it seems unlikely that prosecutors would pursue cases in which a defendant-doctor could admit the Board’s findings in her favor. If that’s right, the Board’s decision would be final de facto. ↑
-
. See, e.g., Complaint Review Process, Med. Bd. of Cal., https://www.mbc.ca.gov/ Consumers/File-a-Complaint/complaint-process.aspx [https://perma.cc/3JVY-JT7J]. ↑
-
. On the role of state medical boards in superintending the medical profession, see David Alan Johnson, Past Imperfect: Revisiting the History of the Federation of State Medical Boards, 110 J. Med. Regul. 20, 22 (2024). ↑
-
. Medical boards, in this analogy, are the equivalent of foreign states, while individual doctors are equivalent to their diplomats. On diplomatic immunity generally, see Jens David Ohlin, The Combatant’s Privilege in Asymmetric and Covert Conflicts, 40 Yale J. Int’l L. 337, 384 (2015) (explaining that diplomatic immunity “must be asserted by a state on behalf of the individual, and in theory can be waived by the state if so desired (even to the detriment of the individual criminal defendant)”). ↑
-
. See Neal Devins & Michael Herz, The Uneasy Case for Department of Justice Control of Federal Litigation, 5 U. Pa. J. Const. L. 558, 561 (2003) (“Most agencies have criminal investigators, but if the agency decides a criminal prosecution is warranted, it must refer the case to DOJ.”). ↑
-
. Another option would be “abortion committees” that give certain clinicians a measure of legal cover to perform abortions with hospital approval. The case of Doe v. Bolton struck down a Georgia statute requiring advance approval by a hospital abortion committee. 410 U.S. 179 (1973 ), abrogated by Dobbs v. Jackson Women’s Health Org., 597 U.S. 215 (2022). Pre-Roe committees often operated to deny permission and restrict access, especially for poor women of color. Committees were composed primarily of the conservative white men who dominated the ranks of internists, psychiatrists, and ob-gyns at major hospitals. See Rickie Solinger, “A Complete Disaster”: Abortion and the Politics of Hospital Abortion Committees, 1950–1970, 19 Feminist Stud. 241, 250 (1993). ↑
-
. See supra note 31 and accompanying text. ↑
-
. A forthcoming piece argues that they do not. See Yvonne Lindgren & Michelle Oberman, Recalibrating Risk Under Dobbs, 94 Fordham L. Rev. (forthcoming 2025) https://papers.ssrn.com/ sol3/papers.cfm?abstract_id=5003888 [https://perma.cc/TSJ3-YBYZ]. ↑