Limits on the Rule

Limits on the Rule

The breadth of the unlawful-act predicate is the source of the rule's danger, and the limits taught here exist because courts recognized that danger. Ordinary involuntary manslaughter requires recklessness or criminal negligence, but where misdemeanor manslaughter applies, the defendant's intentional commission of the misdemeanor stands in for that culpability.¹ The punishment, however, remains the punishment for homicide. If the fault requirement disappeared and every misdemeanor stayed available as a predicate, the rule would convict anyone whose minor violation happened to coincide with a death, and the exposure would keep growing because misdemeanors are far more numerous and varied now than in the early common-law era.² Courts accordingly came to regard the rule as too harsh in its original form and began to confine it, and each confinement works by restoring some connection between the defendant's wrongful act and the risk of death.³

There are three tools for restoring that connection, and the first is inherent wrongfulness. Courts divide unlawful acts into those that are malum in se, meaning inherently wrong by common moral standards, such as battery or larceny, and those that are malum prohibitum, meaning wrong only because a law forbids them, such as many licensing and regulatory offenses.⁴ Where the predicate is malum in se, the usual rule imposes involuntary manslaughter liability for an unintended death without asking whether the death was foreseeable.⁵ The working assumption is that a defendant who is already doing something wrong by ordinary moral standards has less ground to complain when the resulting harm is charged as homicide, whereas the same charge for a regulatory slip looks like punishment for coincidence.

The second tool, foreseeability, governs the offenses that are not malum in se. Under the majority rule, a death during a misdemeanor that is not malum in se is manslaughter only if death was a foreseeable result of committing that misdemeanor.⁶ A minority of courts take the stricter course of restricting misdemeanor manslaughter to predicates that are malum in se, so that a regulatory offense cannot qualify however foreseeable the death.⁷ The foreseeability test is objective: it asks what a reasonable person would have anticipated, and the defendant need not have foreseen death personally.⁸ Its function is to separate deaths that fall within the risk the violation created from deaths that merely happened while the violation was underway.

The third tool is dangerousness, and it replaces the moral label with a question about risk. The District of Columbia limits the doctrine to misdemeanors dangerous in and of themselves, a category that Comber v. United States described as covering misdemeanors that bear an inherent danger of physical injury, with simple assault as the example.⁹ In United States v. Walker, the same jurisdiction treated carrying a pistol without a license as dangerous in itself where the offense ended in a bystander's death.¹⁰ Kansas takes a related but distinct route by confining its statute to misdemeanors enacted for the protection of human life or safety, so that the legislature's own purpose in creating the offense does the filtering.¹¹ Because jurisdictions choose among these formulations instead of stacking them, the first task on any problem is to identify which one governs, and the usual error is to import the majority foreseeability test into a jurisdiction that asks only about danger.

Classification is the soft spot in the first tool. Courts do not always say whether the act at issue is malum in se or malum prohibitum, and some decisions on unlicensed driving and similar offenses leave the label unstated even though most jurisdictions would treat the offense as malum prohibitum.¹² Traffic offenses make the difficulty concrete, because one label can cover conduct of very different danger: a lapsed registration adds almost nothing to the risk on the road, while driving far above the limit through a crowded crosswalk adds a great deal. How dangerous the conduct was as committed therefore often drives the classification in practice, and the sections below give working examples instead of a fixed list.

Step one does the real work, because steps two through four are alternative filters and a jurisdiction applies the one its own law adopts. The next section begins with the oldest filter, inherent wrongfulness, and shows how courts sort the clear cases of battery and larceny from the harder ones.

Malum in Se Offenses

The first limit asks what kind of wrong the predicate offense is, and the answer sorts unlawful acts into two groups.¹⁵ An act is malum in se when it is inherently wrong by ordinary moral standards, so that it would be wrong even if no statute addressed it, as with battery or larceny. An act is malum prohibitum when it is wrong merely because the legislature says it is, as with driving on an expired license.¹⁶ The grade of the offense does not settle the question, since a misdemeanor can land on either side of the line. What matters is whether the blame comes from the conduct itself or from the prohibition alone.

Where the predicate is malum in se, the usual rule treats an unintended death caused in committing or attempting it as involuntary manslaughter, with no inquiry into whether the death was foreseeable.¹⁷ The reason lies in what the rule substitutes. Misdemeanor manslaughter lets the defendant's intentional commission of the predicate stand in for the recklessness or criminal negligence that ordinary involuntary manslaughter demands, and that substitution is credible only when the predicate itself carries blame. A person who commits a battery or a theft knows the conduct is wrong whatever the statute book says, so a homicide charge for a resulting death does not catch that person unaware. A person who lets a license lapse has violated a regulation, but the violation says nothing about willingness to risk another's life, which is why the label does the screening that a fault requirement would otherwise do.

Intentional battery is the clean example, and the doctrine reaches far. A defendant who strikes another with a moderate blow of the fist, intending only to hurt, is almost universally held guilty of involuntary manslaughter when the victim, unknown to the attacker, has a weak heart or a thin skull and dies.¹⁸ The same reasoning extends to assault: a defendant who approaches intending to strike but not to kill, and whose target suffers a fatal heart seizure from fright, is guilty although the defendant never touched the victim.¹⁹ In Commonwealth v. Mink, a Massachusetts court in 1877 applied the label to an attempted suicide. The defendant accidentally killed her fiancee, who had intervened to stop her, and because the court classed suicide as an unlawful act malum in se, it found her guilty of manslaughter.²⁰ Mink also shows that the person who dies need not be the person at whom the predicate act was aimed.

Traffic offenses show the classification at work, because one field of law supplies predicates on both sides of the line. A Maine court in State v. Budge stated the division directly: misadventure, meaning an accidental death, does not excuse a malum in se act, while an act that is only malum prohibitum must be shown to be the proximate cause of the death.²¹ In State v. Davis, an Iowa court affirmed a manslaughter conviction against a drunk driver who caused a death, reasoning that driving while intoxicated is malum in se.²² In Keller v. State, the Tennessee court affirmed a manslaughter conviction against a drunk driver who killed a pedestrian crossing mid-block, a pedestrian whom even a sober driver could not have avoided.²³ That outcome illustrates how little the prosecution must tie the wrongful feature of the act to the particular death when the predicate is malum in se.²⁴

The contrast case comes from the other side of the line. In Commonwealth v. Williams, the driver's only unlawful act was failing to renew a license, and the Pennsylvania court reversed a vehicular manslaughter conviction because the lapse had no causal connection to a collision that resulted from another driver's carelessness.²⁵ The label is not always easy to apply, and some courts have criticized it on the ground that many malum prohibitum statutes, traffic regulations among them, exist to protect the public from serious injury or death.²⁶ Those courts ask instead whether the statute was designed to protect the public from that kind of risk, a test taken up later in this chapter. For traffic offenses, then, the useful question is how dangerous the conduct was as committed, since drunk driving and a lapsed license share a field but not a risk.

Jurisdictions also differ on how much work the label does once it is applied. Most courts require either that the misdemeanor be malum in se or, if it is malum prohibitum, that the resulting death be a foreseeable or natural consequence of the unlawful conduct.²⁷ Under that majority approach, malum in se is one route to qualification, and a malum in se predicate needs nothing further, while a malum prohibitum predicate gets a second chance through foreseeability, which the next section develops. A minority of courts use the label as a gate instead and restrict misdemeanor manslaughter to cases in which the underlying misdemeanor is malum in se.²⁸ Michigan, for example, has confined the unlawful-act doctrine to malum in se misdemeanors rather than regulatory ones.²⁹ In such a jurisdiction the analysis of a regulatory predicate ends at classification, no showing of foreseeability revives it, and the prosecution must prove criminal negligence under ordinary involuntary manslaughter. The minority takes the reasoning about blame to its end: if culpability must come from the act itself, a regulatory violation never supplies it, however foreseeable the death. Both approaches treat a malum in se predicate as sufficient, so the usual error is to demand proof of foreseeability for a malum in se predicate.³⁰

On a problem, the first move is to ask whether the predicate would be wrong with no statute on the books. Battery, assault, and larceny answer yes, and under the usual rule a death they cause qualifies without any foreseeability showing. A regulatory violation answers no, and it qualifies only where the jurisdiction also admits malum prohibitum predicates on proof that death was foreseeable. The classification step comes before the foreseeability step because the label decides whether foreseeability is asked at all.

Malum Prohibitum and Foreseeability

A predicate misdemeanor that is not malum in se, meaning a regulatory offense that is wrong only because the legislature has forbidden it, does not qualify automatically in most jurisdictions. Under the majority rule, a death occurring during such a misdemeanor is manslaughter only if death was a foreseeable result of committing the misdemeanor.³³ The same requirement is often phrased as a death that is the foreseeable or natural consequence of the unlawful conduct.³⁴ A malum in se predicate needs no such showing, so the foreseeability limit is the route by which a regulatory violation can still carry a homicide charge.

The majority chose foreseeability over a flat bar on regulatory predicates because two facts about regulatory offenses pull in opposite directions. The first is that these offenses are numerous and often technical. The court in Comber v. United States observed that misdemeanors have multiplied enormously in number and form since the early common law, which makes unreasonable application of the rule a real danger.³⁵ The danger is sharper because the rule lets the intentional commission of the misdemeanor stand in for the recklessness that ordinary involuntary manslaughter requires, and a technical violation supplies very little culpability to stand in.³⁶ The second fact is that a flat bar would be too blunt, because many regulatory statutes, traffic laws above all, exist to prevent serious injury or death, and courts have criticized the malum in se line for ignoring that.³⁷ Foreseeability reconciles the two pressures by tying homicide liability to the risk the prohibition addresses. The defendant answers for a death that the violation made likely, while a death that merely coincided with the violation falls outside the rule.

The test is objective. The question is what a reasonable person would have anticipated, so the defendant need not have foreseen death personally.³⁸ The inquiry is generous about detail: death to this particular victim in this particular manner need not have been foreseeable, so long as the victim belonged to the class the violation endangered and death came about in an ordinary rather than an extraordinary way.³⁹ The victim's own carelessness is no defense as such, but it may bear on whether a reasonable person would have foreseen death at all.⁴⁰ The risk threshold is also lower than in ordinary involuntary manslaughter, where the conduct must involve a high degree of risk of death or serious bodily injury and perhaps awareness of that risk; here, because the defendant was already breaking the law, an unreasonable risk suffices and no subjective awareness is required.⁴¹ That low threshold is why the filter must be applied with care, since it screens out only deaths that no reasonable person would have connected to the violation.

The stop-sign driver is liable and the slightly speeding driver is not for one reason: the test examines the violation as actually committed and asks whether a reasonable person would connect it to death. A stop sign is a device for preventing precisely the collision that occurred, so running it makes a crosswalk death a natural result. A few miles per hour over the limit on a clear, dry road adds almost nothing to the danger, and a pedestrian emerging from behind a bush is the extraordinary event the test screens out. The school-zone variation shows why the statute's label cannot decide foreseeability: the same speeding offense yields a different answer when the speed, the place, and the hour make death something any reasonable driver would expect.

Decided cases show the filter working in both directions. In Reed v. State, a Texas court found a defendant guilty of manslaughter after he committed adultery, then a misdemeanor in Texas, and killed the wife's husband in defending against a murderous attack, because that attack was a foreseeable reaction to the offense.⁴² In Votre v. State, an Indiana court refused to convict a defendant who gave whiskey to a minor contrary to statute, after which the minor died of a heart attack, because the death must follow as a natural and probable consequence of the unlawful act as well as in the course of it.⁴³ The difference lies in the link between violation and death. In Reed the offense itself provoked the violence that ended in a death, while nothing about handing a minor a drink would lead a reasonable person to expect a fatal heart attack. That reasoning also separates regulatory predicates from malum in se battery, where a defendant who intentionally injures another is treated as taking the victim as found and answers even for an unforeseeable death.⁴⁴

The same inquiry explains the gun and licensing cases. In State v. Ayers, an Iowa court reversed an involuntary manslaughter conviction of a defendant who sold a gun to a minor lacking a permit, because the minor's later accidental shooting of a girlfriend while showing off the gun was not a foreseeable consequence of the sale.⁴⁵ In Commonwealth v. Williams, the defendant's only unlawful act was failing to renew a driver's license, and the Pennsylvania court reversed the manslaughter conviction because the expiration had no causal connection to a collision that resulted from another driver's carelessness.⁴⁶ Driving without a license is a crime that neither is a felony nor tends to cause death or serious bodily harm, and manslaughter liability requires that the death be the natural result or probable consequence of the unlawful act.⁴⁷ A lapse that the driver was entitled to cure by renewal adds nothing to the risk on the road, since a driver with a current license would have been in the same collision. The foreseeability limit and the proximate-cause label therefore describe one inquiry, and the next section takes up causation directly.⁴⁸

Step two does the real work, because two drivers charged with the same traffic offense can land on opposite sides of step three depending on what they actually did. Step four preserves the prosecution's other route, since failing the foreseeability limit only removes the shortcut and leaves open a charge built on criminal negligence.

Dangerousness and Related Tests

The malum in se label and the foreseeability test both ask whether a death should be attributed to a particular misdemeanor, and neither asks how risky the misdemeanor was in the first place. A third family of limits asks exactly that, and the reason is visible in what the doctrine does. Where the unlawful-act rule applies, the defendant's intentional commission of a misdemeanor stands in for the recklessness that ordinary involuntary manslaughter requires.⁵⁰ That substitution is easy to defend when the misdemeanor itself signals danger to other people and hard to defend when the misdemeanor is a lapse in paperwork. Dangerousness tests therefore make the predicate show some risk of injury before it can carry a homicide charge, which moves the doctrine back toward the ordinary fault requirement. The formulations vary by state, so what follows is a family of alternatives, and the controlling version on any problem is the one the jurisdiction in the facts has adopted.

The best-known version comes from the District of Columbia. In Comber v. United States the court warned that the traditional rule can cast too wide a net, a danger sharpened by the enormous growth in the number and kinds of misdemeanors since the early common law, and it confined the doctrine to misdemeanors that are dangerous in and of themselves.⁵¹ The court treated that category as covering misdemeanors that bear an inherent danger of physical injury, with simple assault as its example.⁵² Its formulation also has a situational side, asking whether the misdemeanor was committed in a way that is dangerous under the particular circumstances of the case.⁵³ The risk required is modest. A reasonably foreseeable risk of appreciable physical injury is enough, a threshold the court justified by observing that the risk needed for reckless involuntary manslaughter in the District is unusually high.⁵⁴

The facts of United States v. Walker, which Comber discusses, show how the category operates. The defendant carried a pistol without a license, a misdemeanor, and dropped it in an apartment stairwell; the gun discharged and killed a bystander, and the D.C. court held that unlicensed carrying that results in a shooting death is a misdemeanor dangerous in and of itself, reinstating the involuntary manslaughter indictment.⁵⁵ The defense answered that two people walking down a street with holstered pistols present the same danger to others whether or not either holds a license, so the missing license adds no risk.⁵⁶ The court replied that an unlicensed person may lack the mental health, sobriety, or safety training that licensing screens for.⁵⁷ That reply shows how elastic the label can be, because it locates the danger in what the license would have screened out rather than in anything the defendant did on the street.

The pistol and the store license are both licensing violations, so the label cannot decide the question. What separates them is what the license regulates. A firearm license screens for fitness to carry a device that kills when mishandled, while a business license screens for nothing that bears on whether a customer is physically hurt. A dangerousness test therefore sorts offenses by the risk they carry, and a regulatory offense can land on either side of the line.

The D.C. test can be run in two ways, and jurisdictions differ on which they use. One looks to the offense in the abstract, so that every simple assault qualifies and every paperwork violation fails. The other looks to the particular violation. Under the Wells rule in California, as discussed in People v. Cox, the offense must be dangerous under the circumstances of its commission instead of in its inherent or abstract nature, and the rule is not limited by the malum in se and malum prohibitum distinction.⁵⁸ Both tests treat assault and the lapsed business license the same way. They diverge on offenses such as speeding or unlicensed weapon handling, where the statutory definition and the actual conduct can point in different directions, since the as-committed test lets one speeding violation qualify while another does not.

Some states demand more than physical danger and look to the defendant's fault directly. West Virginia limits the doctrine to misdemeanors that rise to the level of criminal negligence, and Utah limits it to violations that evince a marked disregard for the safety of others.⁵⁹ A misdemeanor that itself shows gross carelessness about other people's safety does not substitute for fault; it is evidence of fault. A Florida appellate court surveying English law noted the same move there, where the unlawful act must be one that all sober and reasonable people would recognize as exposing another to at least some risk of harm, because dangerousness supplies the blameworthiness a homicide conviction needs.⁶⁰ Pushed to its end, the approach requires proof of criminal negligence in addition to the misdemeanor, and courts that do so effectively reject the rule.⁶¹ The reason is mechanical. In a jurisdiction that keeps the doctrine, the prosecution ordinarily has two theories, criminal negligence and the unlawful-act theory that dispenses with it.⁶² Once the second theory also demands criminal negligence, the two collapse into one, and the misdemeanor adds nothing the ordinary charge did not already require.

A related limit concerns the predicate's own mental state. Some courts refuse to rest involuntary manslaughter on a public welfare offense that requires no mens rea; a Nebraska court held that such an offense, though enough for misdemeanor vehicle homicide, does not suffice for involuntary manslaughter.⁶³ The doctrine borrows culpability from the predicate, and a predicate that requires no culpable mental state has none to lend. Using it as a base would stack homicide liability on constructive fault that itself rests on no fault. Because the limit is judge-made, a legislature can displace it: a Michigan statute allows a driver who is one mile per hour over the limit in a construction zone and who kills a worker to be held strictly liable for manslaughter.⁶⁴

Other courts tie the inquiry to what the statute was for. Some ask whether the statute was designed to protect the public from the risk of serious injury or death, which is the test used by courts critical of the malum in se line.⁶⁵ Kansas writes the idea into its statute by reaching killings in the commission of a misdemeanor enacted for the protection of human life or safety.⁶⁶ The legislature's purpose does the work that the facts do in the D.C. approach. A stop-sign law is a safety measure whose violation is dangerous by design, while a business-registration requirement protects no one from physical harm.

Students who know the felony-murder rule often assume that misdemeanor manslaughter carries a matching requirement that the predicate be inherently dangerous. The error is assuming a dangerousness requirement exists in every jurisdiction. Many courts have not imposed one, and even misdemeanors that are not inherently dangerous in any real sense can serve as the predicate, perhaps because the doctrine carries no possibility of the death penalty and courts built fewer safeguards around it.⁶⁷ Dangerousness is a brake that some jurisdictions add, and the problem's facts or the governing statute must supply it.

Step three is where jurisdictions diverge, since an abstract test reads the offense definition and an as-committed test reads the facts. Step four preserves the prosecution's other route, because failing a dangerousness gate removes only the shortcut. Step five points to the next section, which takes up how closely the violation must be tied to the death.