A TRIAL COURT RULING THAT THE FEDERAL DEATH PENALTY IS UNCONSTITUTIONAL LEAVES A KEY QUESTION UNANSWERED

By MICHAEL C. DORF

Wednesday, Jul. 10, 2002

Last week, in United States v. Quinones, federal district judge Jed Rakoff ruled the federal death penalty unconstitutional. The decision is far more sweeping than the recent U.S. Supreme Court decisions in Atkins v. Virginia and Ring v. Arizona. Those cases decided, respectively, that the death penalty cannot be meted out to the mentally retarded, or by a judge acting alone. But they left intact the existing systems of capital punishment in most states and at the federal level.

Judge Rakoff's ruling, by contrast, categorically invalidates the federal death penalty. In addition, though the ruling does not expressly address state death penalties, there is no doubt that its logic applies to state systems of capital punishment as well.

Judge Rakoff found that the death penalty, as carried out in the United States, imposes an unacceptably high risk that innocent defendants will be executed. He is certainly correct that our justice system does not perfectly distinguish guilt from innocence in capital cases. But that leaves us with a profound question: How close to perfect would a system of capital punishment have to be in order to survive constitutional scrutiny?

The Quinones Case

Judge Rakoff issued his ruling in a narcotics and homicide case in which the federal government had indicted eight defendants. Six pleaded guilty, but two - Alan Quinones and Diego Rodriguez - did not. Subsequently, the government made clear its intent to seek the death penalty if the jury convicted Quinones and Rodriguez.

Federal law provides for special jury selection procedures in capital cases. Accordingly, before trial, Quinones and Rodriguez moved to have the capital element of their indictments ruled unconstitutional. In a preliminary opinion and order in April of this year, Judge Rakoff indicated that he was inclined to grant the defendants' motion.

Later, after considering the government's arguments in support of the death penalty, Judge Rakoff adhered to his original disposition, finding it unconstitutional.

A Ruling Based on Due Process, Not the "Cruel and Unusual Punishments" Ban

In categorically barring the federal government from seeking the death penalty, Judge Rakoff did not find that capital punishment is cruel and unusual punishment in violation of the Eighth Amendment. (By comparison, the Eighth Amendment was the basis for the Supreme Court's ruling in Atkins barring the states from executing mentally retarded convicts.) Rather, Judge Rakoff relied on the Fifth Amendment's bar on deprivations of "life . . . without due process of law."

Execution deprives people of life. When the criminal justice system kills innocent people without affording them sufficient opportunity to show that they do not deserve to die, it deprives them of life without due process of law.

Studies over the last decade reveal that in a substantial number of cases, incontrovertible evidence of innocence emerges many years after a defendant has been convicted of a capital offense - and, sometimes, after he has been executed. Thus, Judge Rakoff reasoned that, even given the presumption of innocence, the right of appeal, the availability of post-conviction relief, and ultimately the opportunity for executive clemency, there remains a substantial possibility that an innocent person will be executed before he can obtain judicial or executive relief. And that substantial possibility, he concluded, renders the federal death penalty unconstitutional.

The Statistics: Proof That Innocents Have Been Executed

How did Judge Rakoff know that innocent people are sentenced to death?

For one thing, he noted that since 1993, twelve people who had been convicted and sentenced to death, and who had had their subsequent court challenges to their convictions rejected, were exonerated by DNA evidence.

The government argued, however, that this sort of injustice cannot recur. Now, it suggested, the availability of DNA evidence reduces the risk of executing an innocent person.

But Judge Rakoff noted in response that, in many cases, there is no use for DNA evidence. He also noted that more people who were sentenced to die had been exonerated by conventional means than by DNA evidence.

On the whole, Judge Rakoff concluded, the fact that so many people who are scheduled for execution are subsequently exonerated strongly suggests that other innocent people are not exonerated.

How error-prone is the death penalty? Judge Rakoff identified thirty-two death row inmates who were subsequently exonerated out of "a relevant pool of anywhere from around 800 to around 3,700 death row inmates." That means that somewhere between one in twenty-five and one in 116 people who are sentenced to die is innocent.

The Unanswered Question in Quinones: Is Any Death Penalty Error Rate Acceptable?

Judge Rakoff's ruling assumes that even if the smaller error rate is the true rate, that is still unacceptable. Is that right?

Surely no one would say that an extremely high error rate, such as one in two, could be acceptable. But why is a rate of one in a hundred unacceptable? Proverbial wisdom holds that it is better that ten, not one hundred, guilty persons go free than that one innocent person be unjustly punished.

Of course, Judge Rakoff's ruling does not require any guilty persons to go free. The government will still be permitted to imprison convicted murderers for life, or for at least as long as it takes to exonerate the innocent among them.

But the availability of life imprisonment as an alternative to capital punishment does not fully answer the objection. Suppose the error rate were only one in a thousand or one in ten thousand. If life imprisonment is a completely satisfactory alternative, so long as the error rate were anything greater than zero, why would we ever permit the government to take the small but non-zero chance that it is executing an innocent person?

Given human fallibility, we know that we can never have a perfect criminal justice system. Witnesses do not always see what they think they see. Their memories are faulty. Sometimes they lie. Physical evidence does not present flaws of perception, memory or veracity, but it is only as reliable as the people who collect, assemble, and present it. Juries and judges make their own mistakes.

And there's the rub. For if the non-zero risk of error in capital cases is unconstitutional, then it is unconstitutional regardless of what the studies show. The statistics cited by Judge Rakoff merely confirm an inevitable conclusion.

Our capital sentencing system has never been perfect, and it never will be. To require it to be perfect is to render it unconstitutional always and for all time. Looking to technological developments such as the advent of DNA evidence is thus, in at least some respects, beside the point.

The Impact of Quinones: If Followed, It Will Affect State Systems, Too

The Quinones decision is both broader and narrower than it appears. It is broader because its logic fully applies to state as well as federal executions. Were Quinones a ruling of the U.S. Supreme Court, it would render the death penalty unconstitutional throughout the United States.

However, the decision is also much narrower. A ruling by a single federal judge sets no binding precedent on anybody, not even the same federal judge in a different case. Such a ruling has only persuasive force; other judges ought to follow it, that is, only if they agree with it. Accordingly, unless Judge Rakoff's ruling is affirmed on appeal, state and federal courts will remain free to impose the death penalty.

The next step will be an appeal to the United States Court of Appeals for the Second Circuit, and after that, possibly the United States Supreme Court. What risk of executing an innocent person will these courts deem acceptable? Can they say that one in a hundred is too high but one in a thousand would be acceptable?

And if they do strike down the death penalty on these grounds, how could they ever permit the government to resume executions? To do so would require a showing that the risk of executing the innocent had been reduced. But that cannot be shown without first running a presumptively unconstitutional capital system for a period of years, and a court could never countenance the due process violations - the takings of life - that would occur in the interim.

These complexities suggest that there are really only two options. Here is the first: The courts can say that due process is a matter of fair procedures, so that a system that uses procedures that are fair in themselves does not violate due process simply because those procedures do not prevent every possible miscarriage of justice.

Alternatively, the courts can take the truly radical path that Judge Rakoff purports to avoid, and say that no system of capital punishment administered by human beings can satisfy due process. On this account, a set of procedures is fair if and only if it is calculated to produce just outcomes, and that is a standard our capital systems can never satisfy.


Michael C. Dorf is a Professor of Law at Columbia University.

FindLaw Career Center


      Post a Job  |  View More Jobs

    View More