Concept
Decline of the Rule
Year-and-a-day rule (time limit on homicide)
Decline of the Rule
The year-and-a-day rule rested on a factual premise about medicine, and a rule built on a factual premise weakens when the premise fails. The cutoff reflected the view that medical science could not establish causation beyond a reasonable doubt when a great deal of time had passed between the injury and the death.1 Since then, advances in medicine have made causal attribution far more reliable, and life-support machinery has given doctors the ability to prolong life indefinitely.2 Describing the result, the Supreme Court in Rogers v. Tennessee recognized that most legislatures and courts that had recently considered the issue had abandoned the rule as "an outdated relic of the common law."3
The two developments undermine the rule from different directions. Better medicine removes the reason for the cutoff, because the rule answered a proof problem, and where an expert can now trace a death to a wound after a long interval, a categorical bar stops a conviction that the evidence would otherwise support. Life support adds a second difficulty that the old rule never faced. When machines can sustain a gravely injured person past the one-year mark, the length of the interval between injury and death depends on how long the machines ran, and a fixed cutoff then lets a defendant avoid homicide liability by the accident of how long the victim was sustained.
Pennsylvania shows what abandoning the rule can mean in practice. In 1966, William J. Barnes shot and partially paralyzed a Philadelphia police officer during a failed burglary, and he served twenty years in prison for his crimes.4 When the officer died of a urinary tract infection forty-one years after the shooting, a prosecutor asserted that a chain of causation ran from the 1966 shooting to the 2007 death and charged the 72-year-old Barnes with murder.5 Under the common-law rule no such charge could have been brought, because the death fell decades outside the period. Barnes was acquitted of the homicide in 2010.6 The acquittal is a reminder that abolishing the cutoff removes a time bar but leaves the State to prove causation in the ordinary way, which is the work the cutoff once did in advance.
Two questions follow from the trend, and the next two sections take them up in turn. The first is how jurisdictions have actually gotten rid of the rule, by statute or by judicial decision, and what that changes when a victim dies long after the injury. The second is the constitutional question that arose in Rogers when a court abolished the rule and applied the change to a defendant whose act came before the decision. The trend also has exceptions. Abolition is widespread but not universal, and accounts of how many jurisdictions still keep the rule differ, so the first step with any late death is to confirm whether the rule is in force in the jurisdiction the problem names, or to follow the call of the question if it directs a particular approach.
Abolition by Statute and Court
The preceding section explained why the premise behind the year-and-a-day rule weakened. This section turns to how the rule actually leaves a jurisdiction's law. Judges of an earlier era created the cutoff, so it began as a common-law rule, and the great majority of states have since abrogated it either by judicial decision or by legislation, while some jurisdictions still keep it.8 The rule crossed the Atlantic with the colonists, which is why a state that has never addressed it may still have it on the books as part of its inherited common law.9 Because a common-law rule can be removed in two ways, a court can decline to follow it or a legislature can repeal it, the first question for any late death is which of those things has happened in the jurisdiction at issue.
Judicial abolition rests on the ordinary power of a common-law court to revisit a rule when conditions change. Courts exercising that power frequently look to the decisions of other jurisdictions in deciding whether to modify a rule in light of changed circumstances, increased knowledge, and general logic and experience.10 Rogers v. Tennessee is the leading example. The Tennessee Supreme Court refused to apply the rule to a murder conviction resting on a death fifteen months after the stabbing, and it felt free to do so because so many other jurisdictions had already discarded the rule by legislation or by decision.11 When the Supreme Court recounted that holding in Metrish v. Lancaster, it again described the rule as widely viewed as outdated and as abolished, legislatively or judicially, in the vast majority of jurisdictions that had addressed it.12 Each such decision rests on reasons that the next court can borrow, namely medical advances that make causation provable and life support that stretches survival past a year, so abolition tends to feed on itself. The District of Columbia shows a variation in timing: its court of appeals, in a decision called Jackson, abrogated the rule prospectively, meaning going forward.13 Whether a court may apply its abolition to a defendant whose act came earlier is the constitutional question taken up in the next section.
The legislative route is more direct, because a statute can repeal the common-law bar outright. Statutes that deal with the time limit do differ in wording, however. Some carry the old phrase into the text and give it the meaning of one full year, while others drop the phrase and state the period in plain terms.14 Either wording codifies the bar rather than abolishing it, so a statute that merely repeats the phrase has not removed anything, and the counting convention from earlier in this chapter still governs how the period is measured. A statute that abolishes the rule does so by removing the time limit, and the presence or absence of the old phrase is not what signals abolition.
Abolition is not universal, and Alabama's experience shows how a retaining jurisdiction reasons. In Ex parte Key, a man struck by a car died about eighteen months later, and at the State's request the trial court abrogated the rule so that murder charges against the defendant, Key, could go forward.15 The Alabama Supreme Court reversed the manslaughter conviction that followed. A statute there continues the English common law as the rule of decision unless the Legislature alters or repeals it, and the Legislature had not expressly abolished the year-and-a-day rule when it reenacted the Criminal Code, so the rule remained part of the State's common law.16 The court also declined the State's invitation to abolish the rule by judicial decision.17 What separates this decision from Rogers is the court's view of its own power. Tennessee treated the rule as a common-law doctrine its courts could reassess, while Alabama treated the statute continuing the common law as keeping the rule alive until the Legislature said otherwise, and presumed that a statute does not change the common law beyond what it expressly declares.18 At the time of that decision Alabama kept the rule, and a late death there ended the homicide inquiry no matter how the medical evidence looked.
What abolition changes in practice is the question the court reaches. The time bar disappears, and a death long after the injury returns to the ordinary questions of causation: the State must prove that the defendant's act was a but-for cause of the death and also its proximate cause, beyond a reasonable doubt. A long gap does not by itself break the causal chain, because complications that the injury or the treatment it required would normally bring on leave the defendant answerable for the death.19 The contest therefore shifts to whether the fatal condition was such a complication or an independent cause. The reason the trade makes sense follows from the rule's own structure. The cutoff was a stand-in for a proof problem, and once an expert can trace a death through a long interval, the stand-in over-protects, since it bars a charge in the very cases where the State could prove causation. Abolition hands the work back to the usual tools, so that a long gap becomes evidence a jury weighs instead of a bar a judge applies. The cost is that a homicide charge can arrive many years after the act, which places the entire burden of protection on those usual tools.
The Pennsylvania prosecution described in the preceding section shows both halves of that trade. William Barnes had shot and partially paralyzed a Philadelphia police officer in 1966, and when the officer died of a urinary tract infection forty-one years later, the prosecutor charged Barnes with murder on the theory that a chain of causation ran from the shooting to the death.20 The charge was available only because the common-law bar was no longer in force. It also carried a heavy practical burden for the defendant, because state law barred pretrial release of persons charged with murder, so Barnes was incarcerated while the charge was pending.21 With the time bar gone, what stood between the State and a conviction was its burden of proving causation, and Barnes was acquitted of the homicide in 2010.22 Students lose points here by reasoning that abolition does not mean automatic liability, in reverse: a jurisdiction that has dropped the bar has not presumed causation, and a defendant there can still win on the proof.
Because both routes of abolition lead to the same practical place, a problem rarely turns on which route a jurisdiction used. What matters is whether the rule is in force when the case is analyzed. If the facts say the jurisdiction abolished the rule by statute or by decision, the late death raises only ordinary causation. If the facts say the jurisdiction retains the rule, or quote a statute that states the time limit, the bar applies first and the medical proof never comes into play. If the facts are silent, follow the call of the question and state which approach you are taking.
Rogers and Retroactivity
The previous section left open what happens when a court abolishes the year-and-a-day rule in the same case in which the defendant relies on it. The defendant's act came before the decision, so applying the new rule to that defendant is retroactive, meaning it reaches conduct that took place before the rule was announced. In Rogers v. Tennessee the Supreme Court held that Tennessee could do this without denying the defendant due process, and it framed the test as one of unfair surprise.24 The case is best learned in order, beginning with the facts, because the rule's thin history in Tennessee carries most of the reasoning.
Wilbert Rogers stabbed James Bowdery with a butcher knife on May 6, 1994, and one of the wounds penetrated Bowdery's heart.25 During surgery to repair the wound Bowdery went into cardiac arrest and was resuscitated, but the loss of oxygen to his brain ended his higher brain functions, and he remained in a coma until he died of a kidney infection on August 7, 1995.26 A kidney infection is a common complication for comatose patients, and the county medical examiner gave undisputed testimony that the death was caused by cerebral hypoxia secondary to the stab wound, so causation itself was not contested.27 About fifteen months separated the stabbing from the death, which placed the death outside the common-law period of a year and a day.
The jury convicted Rogers of second degree murder under Tennessee's homicide statute, which defines criminal homicide simply as the unlawful killing of another person and says nothing about the year-and-a-day rule.28 Rogers argued on appeal that the rule survived as part of Tennessee common law and barred his conviction despite the statute's silence. The intermediate appellate court rejected that argument on the ground that a 1989 sentencing act abolishing common-law defenses had already eliminated the rule, and it added that no ex post facto problem arose because the act took effect five years before the crime.29 The Tennessee Supreme Court affirmed on a different ground. It accepted that the rule was part of the state's common law and concluded that the 1989 act had not abolished it, but after reviewing the reasons for the rule it found that those reasons no longer existed, so it abolished the rule as it had existed at common law in Tennessee and applied that decision to Rogers.30
That sequence created the federal question. The Tennessee court observed that the state and federal Ex Post Facto provisions refer only to legislative acts, and it relied on the due process principle of Bouie v. City of Columbia, which forbids retroactive application of a judicial construction of a criminal statute that is unexpected and indefensible by reference to prior law, in concluding that its decision did not offend that principle.31 Rogers pressed the point further. He argued that the retroactive abolition should be tested by the same standards that would govern under the Ex Post Facto Clause had the legislature abolished the rule, and that Bouie required reversal.32 The Supreme Court rejected that premise. It accepted for the sake of argument that the same change by a legislature would have been ex post facto, but it held that Bouie did not carry all of ex post facto doctrine over to judicial changes in the common law, and that the test was whether the defendant was unfairly surprised.33
The Court stated that test as follows: a judicial alteration of a common-law doctrine of criminal law violates fair warning, and so must not be applied retroactively, only where it is "unexpected and indefensible by reference to the law which had been expressed prior to the conduct in issue."34 Each part of that standard has its own logic. The value at stake is fair warning, since a person is entitled to learn what conduct the law punishes before acting, which makes foreseeability the measure of whether a judicial change may reach back. The Court refused to import strict ex post facto rules for a structural reason: the common law presupposes evolution, and strict application of those principles would unduly impair the incremental and reasoned development of precedent, whether a court's act is described as making or finding the law.35 A rule that forced every correction to operate only on later conduct would require a court that had recognized an outdated rule to apply it once more to the parties in front of it. Students lose this issue by measuring the court's decision against the Ex Post Facto Clause, because that Clause by its text limits legislatures and does not of its own force apply to the judicial branch, so a judicial change is tested under due process instead.36
Applying the standard, the Court held that Tennessee's abolition of the rule was not unexpected and indefensible, and it gave two reasons.37 The first concerned the rule's reputation elsewhere. It was widely viewed as an outdated relic of the common law, its main justification was that thirteenth-century medicine could not establish causation beyond a reasonable doubt after a long delay, and the vast majority of jurisdictions to consider it had abolished it by statute or decision.38 The second reason the Court called perhaps the most important: when Rogers acted, the rule had only the most tenuous foothold in Tennessee law. It was absent from the criminal code, it had never once been the ground of decision in a Tennessee murder prosecution, and the reported cases mentioned it only three times, each time in dicta, meaning in passing remarks that decided nothing.39 Even the one early decision that quoted the rule reversed the conviction on other grounds and did not rest on the rule.40
Both reasons go to reliance, which is what fair warning protects. A defendant can count on a limit only if a reasonable person would have expected the courts to enforce it, and a rule that Tennessee courts had never used to decide a murder case, and that most of the country had discarded, gave Rogers little to count on. Abolishing it therefore did not catch him unaware, and the Court treated the decision as a routine exercise of common law decisionmaking that laid an archaic rule to rest.41
Bouie supplies the contrasting pole, where the same standard barred a retroactive judicial change. The South Carolina statute prohibited entry upon another's land after notice from the owner forbidding entry, and the state court applied it to sit-in demonstrators who entered a drug store without any such notice and then refused to leave when asked.42 Earlier South Carolina decisions had uniformly required notice before entry and had given no hint that remaining after being told to leave would satisfy the statute.43 The Supreme Court held that applying the new construction to those defendants violated due process, stressing that it was clearly at variance with the statutory language and had no support in prior decisions.44 What separates the two cases is foreseeability. Bouie stretched narrow and precise statutory words to reach conduct they did not cover, which no defendant could have anticipated, while Rogers removed an obsolete judge-made limit that the state's own decisions had barely used and most jurisdictions had abandoned. Both changes enlarged liability after the act, yet only the first came without warning.45
The Court was closely divided, and the dissent deserves a fair statement because it is the strongest objection to the holding.46 Justice Scalia objected that the Court approved a murder conviction for conduct that was not murder when committed, which offends the ancient principle that there can be no punishment without prior law, and that the result yields a constitution in which legislatures cannot retroactively make a killing murder but judges can.47 He rested the objection on the traditional view of the common law, under which judges declare existing law and do not create new law, so that a real change in the law belongs to the legislature.48 On that view the Tennessee court's own explanation was damaging, because it never claimed the rule had been bad law from the start. It said the need for the rule had disappeared with time, and that is the kind of change the Framers would have left to the legislature.49 The majority answered in practical terms: whether judicial alteration of the common law is called making or finding law, it is a necessary part of judging in states whose criminal law keeps common-law elements, and the fair-warning test supplies the constraint.50
Rogers does not let courts apply every new rule backward, and later decisions mark the edges of the holding. In Metrish v. Lancaster the Supreme Court observed that the argument against retroactivity was arguably less weak than in Rogers, because the diminished-capacity defense was not an outdated relic widely rejected by modern courts and had been repeatedly followed in Michigan. It still denied relief, because Rogers never held that retroactive application is allowed only when the replaced rule was an outdated relic, and because the federal habeas standard for reviewing state convictions is demanding.51 The Connecticut Supreme Court in State v. Brocuglio drew the line the other way, applying the Rogers principle to refuse retroactive use of a new exception to the exclusionary rule because it departed from numerous decisions upholding a common-law right to resist an unlawful entry, a departure it called "both marked and unpredictable."52