Will The Supreme Court Compound Its Ex Post Facto Error?
Part Two of a Two-Part Series on Unconstitutional Retroactive Criminal Legislation

In my last column, I wrote about a pending U.S. Supreme Court case - Stogner v. California - that tests the meaning of the Ex Post Facto Clauses of the Constitution, which prohibit certain kinds of retroactive criminal laws. In this column, I will turn to a recent Court decision, Carmell v. Texas, that also addresses the Ex Post Facto Clauses, and which may play a central role in resolving Stogner.

My claim is that the Court made a serious mistake when it decided Carmell, and that very mistake may cause it to decide Stogner wrongly, as well.

A Brief Recap of My Last Column: The Issue Stogner Raises

To understand why the pending ruling is Stogner may be affected by what I believe is the Court's erroneous past ruling in Carmell, it's necessary to briefly summarize the argument made in my prior column.

The Stogner case arose because, in 1994, California changed its statute of limitations for certain child molestation crimes. The earlier version required each case to be brought within three years of the commission of the crime. In contrast, the new version now permits prosecution so long as the case is brought within a single year of the time the child makes the accusation, regardless of how long ago the crime occurred.

No one doubts that the new statute can be applied to crimes committed after 1994. In Stogner, however, the Court must decide whether it is permissible for California to retroactively apply the new statute of limitations to crimes that occurred in the 1950's and 1960's. Obviously, as to these crimes, the three-year statute of limitations in effect back then had expired before charges were brought. The question is, after the statute of limitations has expired, can the California legislature then revive the criminal claims?

As I explained last column, the primary purpose of the Constitution's ban on ex post facto laws is to prevent legislatures from punishing conduct that was innocent when done. If legislatures could rewrite the rules of criminal liability after a person's conduct has already taken place, lawmakers could visit punishments on their political enemies. They could also unreasonably interfere with the reliance interests of individuals who were not placed on clear notice that what they chose to do could subject them to punishment.

In short, the Constitution's prohibition of ex post facto laws is part of a broader requirement that the government be even-handed and fair to all persons, including alleged criminals.

Under these principles, I argued in my last column that the State of California should win its case. Child molestation was wrong and illegal when Mr. Stogner allegedly committed it - indeed, child abuse is one of those so-called malum in se crimes, where the evil is inherent and obvious to all. (By contrast, malum prohibitum crimes are not necessarily evil, but rather made wrongful only by law.)

In essence, I suggested, changes in statutes of limitations seem analogous to changes in rules of evidence: Assume certain kinds of evidence (say fingerprints or DNA tests) are not admissible at the time a crime is committed. That does not mean that such evidence cannot be introduced years later at trial, provided the reliability of such kinds of evidence has been verified in the meantime. Similarly, statutes of limitations may change in response to interim factfinding relevant to the crime at issue.

At the end of my column, however, I noted that even if the Supreme Court accepts the analogy between changes in statutes of limitations and rules of evidence, it may not rule for California. That is because three years ago, in Carmell v. Texas, the Supreme Court suggested that the Ex Post Facto Clauses applied to all significant changes in evidentiary rules. I believe the Carmell ruling is wrong, and that Stogner provides a chance for the Court to fix its mistake.

The Carmell Case: A Slim Majority Sees an Ex Post Facto Clause Violation

Like Stogner, Carmell also involved sexual molestation of a child, a teenage girl with the initials "K.M.," whom Texas prosecutors said was abused by her stepfather, Scott Carmell, during the early 1990's.

At the time, Texas law said that defendants accused of certain sexual assaults could not be convicted merely on the testimony of the victims. Instead, conviction required some corroborating physical evidence, or at least some contemporaneous statement by the victim - termed "outcry" - such as a remark made to a friend or counselor or doctor near the time of the assault.

But in 1993, Texas changed its attitude towards teenage victims of sexual abuse. Rejecting old stereotypes, and recognizing that teenage accusers (ordinarily girls) are not usually liars, Texas rewrote the evidence laws. Now, they no longer require more than a victim's testimony. By itself, such testimony can, under the new law, establish guilt beyond a reasonable doubt if it is convincing enough.

In 1997, Scott Carmell was prosecuted, based on the uncorroborated testimony of K.M. The jury believed her, and convicted Carmell on fifteen counts.

Carmell then sought and was granted Supreme Court review. There, he argued that Texas could not apply its new, 1993, evidence law to his trial, because his crimes were committed earlier in the 1990s, when the old laws were in place.

The Supreme Court, by a bare 5-4 majority, threw out the conviction. The majority was an extremely unusual one, consisting of the so-called "liberal" Justices Stevens, Souter, and Breyer, and the so-called "conservative" Justices Scalia and Thomas. (Dissenting, then, were conservative Chief Justice Rehnquist, Justice Kennedy and Justice O'Connor, and liberal Justice Ginsburg.)

Moreover, Carmell could not be said to have relied in any meaningful way on the prior rules of evidence (For instance, he did not discard any helpful evidence that could have proved him innocent based on assurances that K.M.'s case would not be prosecuted on her testimony alone.) Accordingly, there is simply no unfairness in applying the new rule to him.

How the Court in Carmell Got Off the Right Track: Chasing a Bad Lead

In reaching the wrong result, the Court in Carmell made a few all too familiar mistakes.

First, the Carmell majority leaned too much and too unthinkingly on its past precedent. It should have focused on the Ex Post Facto Clauses themselves, and their origins. Instead, the majority focused unduly on Justice Chase's famous analysis of the ex post facto provisions in his separate opinion in the Calder v. Bull decision, issued in 1798.

In his Calder v. Bull opinion, Justice Chase identified four distinct categories of cases that violate ex post facto principles: laws that make criminal conduct that was innocent when done (the bedrock principle I've discussed); laws that retroactively increase the punishment for crimes; laws that retroactively increase the offense-grade of crimes; and all laws that "alter[] the legal rules of evidence, and receive[] less, or different, testimony, than the law required at the time of the commission of the offense, in order to convict the offender."

Thus, in his fourth category, Justice Chase made clear that, in his view, changes in the law of evidence are unconstitutional ex post facto laws. If one accepts this idea, Carmell's result is quite predictable. There, the law of evidence relating to the crime was changed, and in a way that certainly led to Carmell's conviction, which would previously have been impossible.

The problem, however, is that it's not clear that this idea belongs to the Constitution, and not just to Justice Chase - for reasons I will now detail.

The Problems with Justice Chase's Analysis, and Thus With Carmell

First, and importantly, as my brother, Akhil Reed Amar, has pointed out, Justice Chase's four-part definition of ex post facto laws was never advanced by "any leading Federalist speaker or pamphlet before the Constitution's ratification." In light of this, arguing that Chase's definition of ex post facto laws is the same as the Constitution's is dubious at best.

Proponents of the Constitution in 1787 had every incentive to tout the limits the new document placed on Congress in order to win over skeptical Anti-Federalists, and the Ex Post Facto Clauses were just such limits. Yet, as noted above, not one of them offered the broad definition of "ex post facto" that Justice Chase came up with.

Nor was the authority of Justice Chase equivalent to the authority of, say, Justice Marshall or Justice Story. Justice Chase was never regarded as a leading light on the early Supreme Court - nor has history so regarded him. Indeed, Chase was impeached for alleged misconduct.

By focusing too much on past cases and categories, the Carmell Court never asked the basic questions. Has the Texas legislature deprived Carmell of fair notice or otherwise undermined any reasonably reliance on his part?

If it had, the result would have been different. The new law made it easier to convict Carmell - but crucially, it did not do so in an unfair way, because Carmell did not rely, in any justifiable way, on the prior law.

Indeed, the only kind of reliance that seems possible would have been repulsive - a molester's reliance that a conviction on his victim's say-so alone was legally impossible. Certainly that's the kind of reliance we want to defeat, not encourage.

Carmell's Potential Effect on Stogner: One Wrong Makes Another Wrong?

Carmell should be a good case for Mr. Stogner to cite, in his attempt to convince the Court to throw out the charges against him. Indeed, if Texas's change in the rules of evidence unfairly burdened Mr. Carmell, then California's change in the statute of limitations would seem to impose a larger and even more unfair burden on Mr. Stogner.

At least Mr. Stogner has some plausible claim of unfairness, and justifiable reliance: "I thought I was free to move ahead with my life when three years passed since my misdeeds, and the statute of limitations expired.." Mr. Carmell has no such argument.

An analogy to another area of law also suggests that if Mr. Carmell was correct, Mr. Stogner is even more correct. When a federal court is hearing a case that is brought under state (rather than federal) law because the parties live in different states (so-called "diversity" jurisdiction), the federal court applying state law would apply state statutes of limitations but not state rules of evidence. That's because it recognizes that statutes of limitations are more "substantive" than are evidentiary rules. And a point of view that looks towards evidentiary rules as merely procedural will see more harm to Mr. Stogner - who lost the benefit of a substantive statute of limitations - than to Mr. Carmell, who lost, on this view, a mere procedural advantage.

Ironically, though, the Court may retain Carmell but rule against Mr. Stogner. The Court might say: "Texas's law fell within the Calder categories, but changes in statutes of limitation do not, because Justice Chase didn't mention them."

This is one of the primary arguments that has been made by the United States, which filed an amicus ("friend of the court") brief in Stogner. Congress has retroactively eliminated the statutes of limitations for many terrorism crimes in the U.S.A. Patriot Act. As a result, the U.S. has a strong stake in arguing that changing statutes of limitations retroactively is perfectly fine.

If the Court takes this tack, the result may be satisfying, but the reasoning will not. Unthinking application of the Calder doctrine and Chase's boxes is the problem, not the solution.

The Court shouldn't try to squeeze this case into one of the Calder categories. Nor should it simply conclude that it simply cannot so be squeezed. Instead, the Court should take a step back, and ask what the Ex Post Facto Clauses are designed to get at: What purposes do they serve? And, do the Calder categories really capture these purposes in a complete and nuanced way?

In so doing, the Court will be able to get back to the Constitution, and away from a misguided line of precedent. That is originalism in the very best sense.