No one better understands the arbitrariness of the death penalty than Juan Raul Garza. A Hispanic man sentenced to death in Texas for three drug-related murders, he is scheduled to be executed on June 19. This past December, President Clinton put off Garza's execution date because a Justice Department study had raised serious questions about racial, ethnic and geographical disparities in the administration of the federal death penalty. Garza's case raised these questions in a particularly direct way, because the study found that the federal death penalty is disproportionately sought and obtained against black and Hispanic defendants, and disproportionately meted out in a handful of states, Texas the leader among them. Had Garza been white or had he committed his crimes somewhere other than Texas, he probably would not be on death row today.

Earlier this month, however, Attorney General John Ashcroft announced that there is no evidence of bias in the federal death penalty, and the government is now preparing to execute Garza. So what changed? Not the evidence. Ashcroft issued a supplementary report, but it contained no new relevant information and failed to undertake the analysis required to determine whether the federal death-penalty disparities were attributable to bias. Ashcroft said that such a study would take too long and might not be definitive in any event. In its place, his report offers only speculation as to why racial or ethnic disparities might exist in a handful of particular jurisdictions.

In a hearing on June 14, Ashcroft backpedaled and agreed that further study was needed. But in the meantime, he insists on going forward with executing Garza.

The facts are unchanged, and they remain deeply troubling. Blacks and Hispanics, who make up about one-quarter of the general population, are more than 80 percent of those on federal death row. Federal prosecutors are twice as likely to reach plea agreements avoiding death sentences with white than with black and Hispanic capital defendants. And Texas, a single state, accounts for nearly a third of all federal death-row inmates.

Ashcroft stresses that of all federal cases that could have been tried as capital cases, the Attorney General actually authorized the death penalty more often against white defendants than minority defendants. But since over 90 percent of white murder defendants killed white victims, that statistic may simply reflect favoritism toward white victims. At the federal level, the Attorney General has authorized the death penalty twice as often when victims are white than when they are black. Yet the Attorney General's latest report fails even to mention this fact.

Ashcroft's report also emphasizes the many procedural safeguards the federal government has put in place to ensure that race does not affect the imposition of death, including instructions to jurors, appointment of competent defense counsel and centralized review in the Justice Department of all federal death-penalty cases. But the bottom line is that despite these safeguards, the federal system has resulted in substantial disparities.

The root of the problem is that there simply is no fair and objective way to isolate, from among the approximately 20,000 homicides that take place each year, the 100 or so that will lead to the ultimate punishment of death. In the end, race, class, ethnicity, geography or the political inclinations of the Attorney General are as likely to make the difference as anything else. Just ask Juan Garza.

• Get The Nation at home (and online!) for 75 cents a week!
• If you like this article, consider making a donation to The Nation.