The following excerpts are from Judge Mark Wolf’s opinion allowing the federal capital prosecution of Gary Lee Sampson to proceed. In his decision, Judge Wolf of the Federal District Court in Boston expressed reservations about the risks of executing the innocent and appeared to criticize the Justice Department’s zealous approach to seeking capital convictions. The headings for these excerpts, which are not part of the original text, are followed by page numbers that correlate with the referenced section of the full text. Read the full text of the decision.


As the Supreme Court has repeatedly reiterated, whether a penalty constitutes cruel and unusual punishment is not determined by the standards of the eighteenth century when the Eighth Amendment was adopted. Rather, the Eighth Amendment must draw its meaning from “‘the evolving standards of decency that mark the progress of a maturing society.’” Atkins v. Virginia, 536 U.S. 304, 311-12 (2002) (quoting Trop v. Dulles, 356 U.S. 86, 101 (1958) (plurality opinion)). It is, therefore, the duty of the courts to reconsider periodically whether the death penalty offends contemporary standards of decency. See §IV, infra.


[Sampson] has, however, persuaded the court that this is a serious question, that future developments could strengthen this argument, and that courts will have a duty to monitor carefully future legislation and jury verdicts concerning the death penalty in deciding what is likely to be the constantly recurring question of whether the risk of executing innocent individuals renders the death penalty generally, or the FDPA particularly, unconstitutional. See §VII, infra.

More specifically, in 1993, a majority of the Justices of the Supreme Court stated that the execution of an innocent person would violate the Constitution. See Herrera v. Collins, 506 U.S. 390 (1993). This court agrees.

The risk of executing the innocent has long been recognized. However, in the past decade substantial evidence has emerged to demonstrate that innocent individuals are sentenced to death, and undoubtedly executed, much more often than previously understood. In that period, DNA testing has established the actual innocence of at least a dozen inmates who had been sentenced to death. These developments have prompted the reinvestigation of many other capital cases, resulting in the release of more than 100 innocent individuals from the nation’s death rows.

In deciding in 2002 that it is no longer constitutional to execute the mentally retarded, the Supreme Court wrote that “we cannot ignore the fact that in recent years a disturbing number of inmates on death row have been exonerated.” Atkins, 536 U.S. at 320 n.25. The government correctly asserts that the Supreme Court was addressing convictions obtained in state courts, rather than under the FDPA. The government contends that similar errors could not occur in federal courts.

The government’s confidence that the FDPA will never lead to the execution of innocent individuals is not shared by the only federal judge to have conducted the trial of an FDPA case in Massachusetts. Judge Michael Ponsor presided in the trial of Kristen Gilbert, a nurse convicted of murdering four of her patients and attempting to murder three others. After the jury’s 2001 verdict decided that she should be sentenced to life in prison, Judge Ponsor wrote that “[t]he experience left me with one unavoidable conclusion: that a legal regime relying on the death penalty will inevitably execute innocent people ­ not too often, one hopes, but undoubtedly sometimes.” Appendix (“A-“)-90, Michael Ponsor, “Life, Death, and Uncertainty,” Boston Globe, July 8, 2001, at D2.

There are compelling reasons to believe that Judge Ponsor’s prediction is prophetic. Federal judges, like state judges, are human and, therefore, fallible. Jurors in federal cases are essentially the same citizens who serve as jurors in state cases. In addition, many federal cases, including this one, result from investigations conducted primarily, if not exclusively, by state and local law enforcement.


Important errors are, however, not always identified prior to death sentences being imposed, at times because of misconduct by state and federal investigators. It is now clear that in 1967 Joseph Salvati and several other individuals were unfairly convicted because the FBI had withheld information that its informants, rather than the defendants, had murdered Edward Deegan, and had allowed its informants to testify falsely against the innocent men. Several of the defendants, including Peter Limone, were sentenced to death. While those death sentences were reduced to life in prison following the invalidation of the death penalty by Furman, two of the wrongfully convicted men died in prison. Salvati, who was originally sentenced to life in prison, receiveda commutation and was released in 1997. Limone was released in 2001, after his wrongful conviction had been demonstrated. See United States v. Flemmi, 195 F. Supp. 2d 243, 251 (D. Mass. 2001).

The deliberate misconduct by federal investigators that was so belatedly revealed with regard to the Deegan murder is neither ancient history nor unique to Boston. Daniel Bright was, in 1996, convicted of murder by the state of Louisiana and sentenced to death. Several months ago, a federal judge found that the FBI had evidence that another person had claimed to have committed the murder, but the FBI violated the government’s constitutional duty to disclose that evidence to Bright before his trial, and later lied to the federal judge about its existence. See Bright v. Ashcroft, 259 F. Supp. 2d 494 (E.D. La. 2003) and 259 F. Supp. 2d 502 (E.D. La. 2003).

The government misconduct concerning Salvati and Bright are not isolated occurrences. A recent study of capital cases from 1973 to 1995 reported that one of the two most common errors prompting the reversal of state convictions in which the defendant was sentenced to death was the improper failure of police or prosecutors to disclose “important evidence that the defendant was innocent or did not deserve to die.” James S. Liebman, et al., A Broken System: Error Rates in Capital Cases, 1973-1995 at ii (2000). As indicated earlier, the performance of state and local police is important to the operation of the FDPA because many cases, including this one, have initially been investigated by them and later brought in federal court, at times in an effort to achieve a death sentence that is not available under state law.

Serious errors appear to be common in capital cases. After analyzing more than 4500 appeals of capital cases, the same study found that “the overall rate of prejudicial error in the American capital punishment system was 68%. ” Id. at i (emphasis in original). As the authors later wrote:

For cases whose outcomes are known, an astonishing 82% of retried death row inmates turned out not to deserve the death penalty; 7% were not guilty. The process took nine years on average. Put simply, most death verdicts are too flawed to carry out, and most flawed ones are scrapped for good. One in 20 death row inmates is later found not guilty.

A-284, James Liebman, et al., “Technical Errors Can Kill,” Nat’l L.J., Sept. 4, 2000, at A16.

In view of the foregoing, this court agrees with Judge Ponsor, among others, that the FDPA, like the state death penalty statutes, will inevitably result in the execution of innocent people. Since a majority of the Supreme Court stated in 1993 that the execution of an innocent person would be unconstitutional, the critical question is how many of those who will be executed must be innocent to offend contemporary standards of decency and, therefore, render the FDPA unconstitutional.


As described earlier, courts are required to discern contemporary standards of decency from objective factors to the maximum possible extent. Those factors demonstrate the following.In 1791, the concept of “cruel and unusual punishment” incorporated in the Eighth Amendment was imported from English law. England and other nations that share our heritage have now abolished capital punishment.

Recent opinion polls show that 73% of Americans believe that our nation’s death penalty statutes have resulted in the execution of an innocent person in the past five years. Nevertheless, 74% say they support the death penalty. However, only a slight majority (53%) prefer it to life in prison without parole (44%) for convicted murderers.

The decisions of juries in recent FDPA cases indicate that there is a definite disparity between the attitudes of Americans toward the death penalty in general and their willingness to impose it in particular cases. In sixteen of the last seventeen penalty phase verdicts returned by juries in FDPA cases the defendant was not sentenced to death. In fifteen of those sixteen cases the defendant had been convicted of a federal crime involving murder. Therefore, juries have recently been regularly disagreeing with the Attorney General’s contention that the death penalty is justified in the most egregious federal cases involving murder.


In Atkins the Supreme Court essentially held that because Virginia diverged from the substantial consensus that had emerged in legislation, decisions of prosecutors, and jury verdicts in many other states, it was arbitrary and capricious and, therefore, cruel and unusual for a retarded person in Virginia to face execution when a similarly situated individual in another jurisdiction would not. If the evolution of events concerning the general imposition of the death penalty parallels the developments described in Atkins concerning the execution of the retarded, the day may come when courts properly can and should declare the ultimate sanction to be unconstitutional in all cases.

However, that day has not come yet. There is not now sufficient objective evidence to establish that the death penalty offends contemporary standards of decency to permit a court to end political debate and democratic decisionmaking concerning its propriety.

Nevertheless, “the Clause forbidding ‘cruel and unusual’ punishments … ‘may acquire meaning as public opinion becomes enlightened by a humane justice.’” Gregg, 428 U.S. at 171 (quoting Weems v. United States, 217 U.S. 349, 378 (1910)). Judges seek to administer humane justice. Judicial decisions are part of a colloquy with citizens and those they elect to make and execute our laws. Those decisions have the potential to influence contemporary standards of decency and, therefore, the current meaning of the Eighth Amendment.

While this court does not find that the risk of executing the innocent now renders the FDPA unconstitutional, the record regarding this issue raises profound questions. Those questions are not hypothetical. Rather, as demonstrated by the experiences of Salvati and Bright, among others, those questions are real and recurring.

Error is, of course, possible in any criminal case. While our system promises everyone a fair trial, it does not pretend to perform perfectly. However, as the Supreme Court has repeatedly reiterated, “[t]he penalty of death differs from all other forms of criminal punishment not in degree but in kind. It is unique in its total irrevocabality.” Furman, 408 U.S. at 306 (Stewart, J., concurring);5 see also Ring, 536 U.S. at 605-06; Harmelin v. Michigan, 501 U.S. 957, 994 (1991); Gardner v. Florida, 430 U.S. 349, 357 (1977) (Stevens, J.) (plurality opinion). Among other things, an execution eliminates the opportunity to end any injustice, even belatedly. Thus, this court will strive to provide the government and Sampson as fair a trial as possible.


This court agrees that “executing the innocent is inconsistent with the Constitution.” Id. at 419 (O’Connor, J. and Kennedy, J., concurring). The open issues in this case are whether the FDPA will inevitably result in the execution of innocent individuals and, if so, whether this renders the statute unconstitutional, and inapplicable to Sampson because it is an invalid law. For the reasons described below, the court finds that: the FDPA will inevitably result in the execution of innocent individuals; there is not now, however, a proper basis to declare the FDPA unconstitutional for this reason; and, therefore, it is not necessary to decide Sampson’s claim that he has a right not to be tried under an unconstitutional statute.

EVIDENCE OF THE RISK (pages 56-57)

Since 1973, more than 100 innocent people have been released from death rows. See A-27, Death Penalty Information Center, “Facts About the Death Penalty”; A-97, Affidavit of Richard Dieter. In January 2003, the Governor of Illinois commuted the death sentences of more than 150 individuals awaiting execution in that state after previously determining that seventeen people on Illinois’ death row were actually innocent. See A-288; see also Alex Kotlowitz, “In the Face of Death,” N.Y. Times Sunday Magazine, July 6, 2003, at 32, 34. One of those individuals was “Anthony Porter, who spent no less than 16 years on death row until prosecutors decided they had made a mistake (upon which determination they then brought murder charges against a different suspect, who confessed).” Quinones, 205 F. Supp. 2d at 265.


In 2003, District Judge James Jones of Virginia stated that:

After having recently spent several weeks … individually interviewing in voir dire nearly two hundred prospective jurors on their attitudes toward the death penalty, I am convinced that our fellow citizens are largely conflicted about the death penalty. Many favor it in principle in the appropriate case, but are concerned about it in practice.

Church, 217 F. Supp. 2d at 702-03; see also Kotlowitz, supra (“Faced with the decision to execute or not, pro-death penalty jurors are increasingly sparing lives.”). There is evidence to validate Judge Jones’ insight.

The decisions of juries in recent FDPA cases indicate that there is a definite disparity between attitudes toward the death penalty in principle and the willingness of federal jurors to impose it when they are fully informed about a particular case, and their decision will have real and serious consequences. The death penalty is sought only in those eligible federal cases in which the Attorney General has personally decided that it is justified. DOJ Study at 5, 26. Jurors who express an unyielding general unwillingness to impose the death penalty may not serve in a capital case. See Witt, supra; Witherspoon, supra.

Nevertheless, in sixteen of the last seventeen penalty phase verdicts returned by juries in FDPA cases the defendant has not been sentenced to death.16 A-38, A-41, A-66 to A-67; Def.’s Supp. Mem.; Def.’s Second Supp. Mem.; Def.’s Fourth Supp. Mem. In fifteen of those sixteen cases the defendant was convicted of a crime involving murder. Id.17

As described earlier, the Supreme Court has repeatedly reiterated that the decisions of citizens as jurors are ” ‘a significant and reliable index of contemporary values.’” Atkins, 536 U.S. at 323 (Rehnquist, C.J., dissenting) (quoting Coker, 433 U.S. at 596 (plurality opinion) and Gregg, 428 U.S. at 181). In finding the death sentence to be cruel and unusual punishment for rape in Coker, the Supreme Court “credited data showing that ‘at least 9 out of 10’ juries in Georgia did not impose the death sentence for rape convictions.” Id. (quoting Coker, 433 U.S. at 596-97). The statistical sample may now be too small to draw any definite conclusions from the most recent FDPA jury verdicts. However, if juries continue to reject the death penalty in the most egregious federal cases, the courts will have significant objective evidence that the ultimate sanction is not compatible with contemporary standards of decency.