OUSTON, June 9 - Texas juries in capital cases must make a prediction. They may impose a death sentence only if they find that the defendant will probably commit more violent acts.

Other states look backward, asking juries to consider the moral blameworthiness of the crime. Texas, which leads the nation in executions, wants to know the future: Will the killer kill again?

	Advertisement

"The fact is," said David R. Dow, a law professor at the University of Houston, "you're being punished for something that you haven't done."

In making their predictions, juries rely on expert testimony. In 1986, for instance, Dr. Edward Gripon, a psychiatrist, testified that David Harris, then 25 and freshly convicted of murder, posed a substantial risk of further violent acts. Dr. Gripon, who had never examined or even met Mr. Harris, based his conclusion on a prosecutor's description of the defendant's past conduct.

Mr. Harris, now 43, is to be executed on June 30. On Wednesday, his lawyers submitted a petition to a state appeals court. It says Dr. Gripon's prediction 18 years ago has turned out to be wrong: Mr. Harris's years in prison have been marred by only minor infractions, like having too many postage stamps or hanging a clothesline in his cell. His most serious offense, according to the authorities, was kicking a guard's boot while wearing shower slippers; Mr. Harris says he slipped.

Mr. Harris's case is not unique. A recent study by the Texas Defender Service, a group that represents defendants in capital cases, examined 155 such cases in which prosecution experts had predicted, often with a claim of scientific certainty, that the defendants would commit more violent crimes. "These experts," the report concluded, "were wrong 95 percent of the time." Though the 155 inmates in question generally served at least a decade on death row, none of them killed again. Eight committed serious assaults, all against prison employees or other inmates; two were prosecuted.

Maintaining that he was sentenced to death on the basis of junk science, Mr. Harris says he deserves a stay of execution and a resentencing hearing at which evidence of how the future turned out can be presented.

A spokesman for the state attorney general's office declined to comment, citing the pending litigation.

But Shannon Edmonds, a lawyer with the Texas District and County Attorneys Association, said objections to the future-dangerousness standard were a smokescreen.

"There is no issue or litmus test that would be accurate enough to satisfy the Texas Defender Service," Mr. Edmonds said. "They basically don't think anyone should get the death penalty."

Gena Bunn, chief of the Texas attorney general's capital litigation unit, defended the state's future-dangerousness requirement in a law review article in 2000.

Ms. Bunn said abstract arguments must fail in the face of the reality of vicious killers like Aaron Fuller, who raped and killed an elderly woman in 1989.

"Although the use of psychiatric testimony to predict future dangerousness is roundly condemned in the scientific community," she wrote with a co-author in The Texas Review of Law & Politics, "the reader need only make a common-sense inquiry to see the logic of the system. Would the reader want to share a jail cell with Aaron Fuller?"

Texas law makes the death penalty available for about a dozen categories of murder, including those committed during a robbery or a sexual assault, those done for hire and those in which the victim is a child or a police officer.

But conviction of such a crime is not enough. To impose a death sentence, a jury must find, unanimously and beyond a reasonable doubt, that "there is a probability that the defendant would commit criminal acts of violence that would constitute a continuing threat to society." Because those convicted of capital crimes must serve at least 40 years, as a practical matter that threat is to prison guards and other inmates.

Only Oregon joins Texas in having death sentences turn solely on predictions of future dangerousness. Six other states allow the factor to play a role in juries' decisions, along with many others. A ninth state, Virginia, which trails only Texas in the number of people it executes, requires a jury either to find future dangerousness or to determine that the crime was "outrageously or wantonly vile, horrible or inhuman."

"What is a significant but not determinative factor in other states," Professor Dow of the University of Houston said, "basically in Texas is the whole ball of wax."

(In the 29 other states with the death penalty, future dangerousness plays no role at all.)

State Representative Jim Dunnam, a Democrat from Waco, has introduced legislation to conform capital sentencing procedures in Texas to those in some other states.

"I was trying to address the problem of unreliable testimony by experts in capital cases," Mr. Dunnam said. "My proposal was that we have more of a list of factors, both aggravating and mitigating, for the jury to balance in making their decisions."

Mr. Harris, the inmate now appealing his sentence by contesting the future-dangerousness standard, was convicted in 1986 of killing Mark Mays in a Beaumont gunfight as Mr. Harris was trying to kidnap Roxanne Lockard, Mr. Mays's girlfriend. Mr. Harris later gained fame in the 1988 documentary "The Thin Blue Line," in which he acknowledged knowingly giving false testimony about the 1976 murder of a Dallas police officer, evidence that almost resulted in the execution of an innocent man.

Solace K. Southwick, one of Mr. Harris's lawyers, discussed his case in her office at a big corporate law firm here. Ms. Southwick has represented him since 1990.

"The physical difference in David Harris isn't just that he's aged and gotten a little fatter," she said. "He's a completely different person."

She showed a visitor an elaborate crafts display Mr. Harris had made for her. It included the images of a bald eagle and a flag, a clock and the words "God Bless America."

The prison kitsch has no place in her sleek office, and she keeps it tucked in a closet. But she shook her head to look at it. "Isn't it amazing that he would have that sentiment?" she said.

In 1983, the United States Supreme Court declined to bar expert testimony concerning future dangerousness under the Texas law. The American Psychiatric Association had filed a supporting brief in that case, saying expert opinions on the subject were useless. "The unreliability of psychiatric predictions of long-term future dangerousness is by now established fact within the profession," the association told the court.

But Justice Byron R. White, writing for the majority, said, "We are not persuaded that such testimony is almost entirely unreliable." In the petition filed Wednesday, Mr. Harris contends that recent studies demonstrate that expert predictions in this area are just that - almost entirely unreliable.

A decade after that Supreme Court decision, the justices held in a civil case that trial judges must carefully scrutinize scientific expert testimony before presenting it to a jury. It is unclear whether that principle applies at capital sentencing hearings.

But one federal appeals court judge, Emilio M. Garza, has written that the later case, known as Daubert, suggests that expert testimony about future dangerousness should be rejected.

"The scientific community virtually unanimously agrees that psychiatric testimony on future dangerousness is, to put it bluntly, unreliable and unscientific," Judge Garza wrote in a 2000 concurrence to a decision upholding a death sentence.

Some prosecution experts have been criticized for the uniformity and severity of their opinions, others for unorthodox methodology.

One, Dr. Walter Quijano, testified in at least seven cases in Texas that being black or Hispanic correlated positively to future dangerousness. State officials have told courts that it was a mistake to present the testimony, but local prosecutors, in a rare bit of infighting among law enforcement officials, insist the death sentences should stand.

The courts have so far agreed that new sentencing hearings are required in those cases, and the Texas Legislature recently enacted a law barring testimony about future dangerousness based on the defendant's race.

But the future-dangerousness requirement itself stands, frustrating critics.

"I see the Texas statute as dishonest to the true purpose of capital punishment," said Andrea Keilen, the deputy director of the Texas Defender Service. "It doesn't reliably sort who deserves to live from who deserves to die."

Home Delivery of The Times from $2.90/week - Act Now!