In 1978, Robert Alton Harris, a 26-year-old paroled murderer, kidnapped two California teenagers in the parking lot of a fast-food restaurant. He drove to a remote canyon where he killed the two boys, then finished the hamburgers they’d been eating and drove off. Harris was later caught robbing a bank. He confessed to the murders and was sentenced to death in 1979. Yet he managed to delay his execution for 13 years by repeatedly manipulating the appeals system. Harris and his lawyers challenged the quality of his psychiatric evaluation, claimed California’s gas chamber was unconstitutional, and argued that the death penalty discriminated against younger killers, males, and those who killed whites. Each of these claims stalled his execution. In a flurry of last-minute appeals, the Supreme Court overturned four separate stays on the night of Harris’ execution before it took the unprecedented step of forbidding lower courts to issue further stays. By the time he was finally put to death in 1992, Harris had managed to get more than 20 appeals.

Like Harris, Aaron Patterson’s case centered on a murder confession. He was subsequently sentenced to death for the 1986 stabbing deaths of an elderly Chicago couple. But Patterson is one of Illinois’ “Death Row 10,” a group facing execution who claim their confessions were beaten out of them by a notorious Chicago police lieutenant since fired for abuse and torture. No physical evidence connects Patterson to the murders. The knife was never found. A fingerprint at the scene—not a match—has inexplicably disappeared. The lone witness against him, a teenager who claims police coerced her testimony, has tearfully recanted.

While Harris is a metaphor for the conservative case against the current capital justice system, Patterson’s case has been seized upon by abolitionists. Both are justified in their outrage. More than 20 years after its reinstatement, the death penalty is an unwieldy and ineffective compromise between the roughly 70 percent of Americans who favor capital punishment and the minority who fiercely oppose it. The present system is, quite literally, the worst of both worlds. While an ideological battle has been waged in court, a bottleneck has developed in the nation’s prisons. Currently, 3,565 inmates sit on death row, and new arrivals average 300 annually. Yet despite rising numbers of death sentences, executions have never topped 100 in a single year. Everyone is being sentenced to death and no one is dying.

In the landmark 1972 case of Furman vs. Georgia, the Supreme Court struck down the death penalty on four grounds: it was imposed arbitrarily; used unfairly against minorities and the poor; its infrequent use made it an ineffective deterrent; and lastly, state-sanctioned killing was no longer acceptable behavior. The decision resulted in tremendous uncertainty. Each justice filed a separate opinion, an unprecedented move in 20th century jurisprudence, yielding the longest written decision in Supreme Court history. The length is indicative of the decision’s complicated rationale. While the case for abolition earned the five votes necessary, the motivation for each vote varied. This uncertainty left the door open for states to write new death penalty laws that might pass constitutional muster.

Abolition was never very popular. Polls taken in March 1972, shortly before Furman was decided, showed a 50 percent approval rating for capital punishment; that number jumped to 57 percent by November. Within the year, 19 state legislatures passed retooled death penalty statutes they hoped would meet the vague standard of constitutionality hinted at in Furman. By the time the Court allowed the modern death penalty by upholding Gregg vs. Georgia in 1976, 35 states had passed new death penalty laws. They have been in dispute ever since, and the result is today’s beleaguered system.

Shortly after Gregg, the court acceded to death penalty opponents by outlawing mandatory death sentences and instituting complex safeguards. This gave rise to a problem that still hampers the courts: Unable to abolish capital punishment in the legislature, defense lawyers have sought to institute a de facto halt to executions by stalling cases in court. The more savvy, like Harris’ lawyers, quickly mastered the art of “sandbagging,” or using delay tactics to stop an execution. By filing spurious last-minute appeals and repeatedly seeking reversals on obscure technical grounds, activist lawyers (and some judges) could bring the system to its knees. Angered by this end run around the law, conservatives have limited the appeals process at the state and federal levels and expanded the scope of crimes punishable by death. The unfortunate result, as federal prosecutor David Lazarus wrote in this magazine last year (“Mortal Combat,” June 1998), is that abolitionists have clogged the courts with endless litigation and slowed the number of actual executions to a trickle. The average delay from sentencing to execution has grown from six years in 1985, to eight years in 1990, to 11 years today.

Delays in death penalty cases are caused in part by the tacit acceptance of such maneuvers by liberal judges. The most famous examples are former Supreme Court justices William Brennan, Jr. and Thurgood Marshall, who voted against every death sentence that came before them. Recent limits to federal appeals should eventually block some avenues for delay. But as old methods are thwarted, new ones crop up. Kent Scheidegger, legal director for the conservative Criminal Justice Legal Foundation, cites as an example an increase in the use of “Ford claims”—assertions that a prisoner has gone insane since sentencing and therefore can’t be executed under the Eighth Amendment. (This is a particularly savvy ploy, because arguing that insanity occurred after the trial phase means the claim isn’t subject to a limit on appeals.)

Other recent tools that have been used for manipulation are Marsden and Faretta motions, which a defendant can use to fire his lawyer and represent himself. In a notorious California capital case, the defendant filed numerous such motions, going so far as to sue two court-appointed defenders for $1 million in an effort to delay his sentence. Common sense needn’t be a requirement for a claim—a Montana prisoner, who’d held off his execution through two decades of appeals, finally argued that the length of time he’d had to endure on death row was unconstitutional.

Though liberals are commonly faulted for delays, conservatives also slowed the process by making access to court-appointed attorneys more difficult for the poor. In 1995 more than a quarter of California’s capital defendants didn’t have a lawyer and their cases couldn’t begin until one was assigned. And, with conservative anger buttressing support for prosecutors’ win-at-all-costs attitude, unscrupulous prosecutors regularly tie up capital cases by fighting defense requests for information.

The unfortunate response to dilatory tactics has been to curtail the judicial process by implementing strict time limits. Thirty-six of the 38 death penalty states impose deadlines for presenting new evidence following a conviction. The strictest such law is Virginia’s notorious “21-Day Rule,” which limits the introduction of new evidence to 21 days after trial. Other states have similar limits that range from 30 days to a year. These laws were intended to establish finality in all criminal cases and most predate the modern death penalty. But they have recently been championed by conservatives as a way to combat delays in execution, despite the obvious risk they carry of killing or imprisoning innocent people.

In recent years, the law’s shortcomings have become apparent. Virginia’s 21-Day Rule, for example, forbids new DNA evidence from being introduced more than three weeks after trial. And, amazingly, the law doesn’t even make provisions for illegally suppressed evidence—such evidence is considered “new” in Virginia and therefore is inadmissible. (A prisoner would instead have to file a federal appeal.) The problem was ably summed up in a defense attorney’s testimony before a Virginia legislative committee: “After 21 days, no court in this state, nor any judicial forum afterwards, can save you from execution based on the proof that you are innocent.”

The circuit court that oversees this system—referred to locally as “the rocket docket”—has succeeded in speeding the processing time of death penalty defendants. Capital cases in Virginia move much more quickly than they do elsewhere. But the results come at a price. One of the most controversial examples is the case of Earl Washington, Jr., a mildly retarded man sentenced to death for the 1982 rape-murder of a Virginia woman. A concerned prosecutor later ordered DNA tests, unavailable at the time, that provided strong evidence Washington is innocent. But because the evidence appeared after the 21-day deadline, Washington had no avenue for appeal. The only recourse for a prisoner with exculpatory evidence is gubernatorial clemency. But governors aren’t part of the judiciary and can’t order a new trial. And of course they’re heavily subject to political pressure. On the last day of his term, Virginia Governor Douglas Wilder commuted Washington’s sentence to life in prison. But with no judicial outlet to order a new case, Washington remains in prison.

Such problems are exacerbated by the fact that many prisoners can’t find attorneys for post-conviction hearings—in 1995 Congress eliminated the resource centers that once provided them. Without an attorney, the possibility of correcting a mistake like the one in Washington’s case all but disappears. State post-conviction review is the first opportunity to present claims of ineffective counsel and the last chance to investigate claims like actual innocence and prosecutorial misconduct. Ironically, the Supreme Court has affirmed that indigent death row inmates have a right to counsel at federal appeals. But without a lawyer at the state level, their claims will doubtless never make it that far.

While state legislatures have failed to improve the system, reforms at the federal level have also been woefully misguided. In the mid-1980s federal judges couldn’t find enough qualified lawyers to handle death penalty cases. Attorneys assigned by the court were expensive and didn’t necessarily provide competent counsel. So in 1988, at the behest of federal judges, Congress created legal resource centers to process and defend the poor. By most accounts, these centers served their purpose.

But in the tough-on-crime environment of the early ’90s, the centers became a political football. Congressman Bob Ingis (R-South Carolina) branded them “think tanks for legal theories that would frustrate the implementation of all death sentences.” In 1996, citing fiscal priorities, Congress eliminated federal funding for resource centers—$20 million for 21 centers nationwide. Budget hawks figured they’d won a double victory. They reasoned that by choking off agencies that represent death row inmates they’d rid themselves of an impediment to executions, while saving taxpayers millions in the process.

That thinking backfired. A critical function of resource centers—one intended by federal judges—was to keep the appellate system functioning smoothly. With organizational support, full-time lawyers kept tabs on death row cases, filed appeals promptly and steered capital defendants through the complicated legal system. Qualified representation meant indigent defendants didn’t require the court’s help. The infusion of lawyers trained in capital cases helped offset the costly and often inexperienced court-appointed lawyers who otherwise handled capital cases. Resource centers were a government program that worked. So their elimination had the opposite of the intended effect—it slowed the pace of appeals and increased the likelihood of error.

Other measures were similarly misguided. In 1994, by threatening to block the federal crime bill, congressional Republicans forced passage of the Federal Death Penalty Act, making 54 new crimes death-penalty eligible. The law was largely a symbolic measure—there hadn’t been a federal execution in 31 years. Yet death-penalty offenses suddenly came to include drug dealing crimes, conspiracy to commit murder even if no murder was committed, and such rarely committed crimes as murder of a federal poultry inspector. The broadened scope only added to the courts’ burden. Federal death sentences soared from 28 the year before the law, to 153 in 1997. Even conservatives thought the law frivolous. “It was an easy way to appear tough,” said Scheidegger. “Congress enacted a long string of federal death penalty crimes, most of which are entirely unnecessary, so that they could go back and claim to be tough on the death penalty. It was a rather cynical ploy.”

Cynical but costly. Last fall, a judicial conference report determined the average cost of prosecuting a federal death penalty case to be $365,000. The same case cost $218,112 to defend. But if a sentence other than death were sought, the defense costs dropped to $55,772. Factored out for the additional convictions, that’s a $73 million price tag on a measure that did little more than let congressmen flex their muscles.

Congress responded to the Oklahoma City bombing with its most ambitious reforms to date. The Antiterrorism and Effective Death Penalty Act of 1996 limited to one the number of appeals a prisoner could file in federal court, barring exceptional circumstances, and placed a one-year time limit on their ability to do so. Conservatives believed this would eliminate sandbagging and streamline the appellate process. “No delay in disposition shall be permissible because of general congestion of the court’s calendar,” the law read.

But like earlier efforts, the Act’s aim was largely misdirected—a disingenuous marriage of death penalty and terrorism law that represented a large political compromise. Sponsors touted a connection between terrorists like Timothy McVeigh and speedier laws which, they suggested, would bring criminals to justice. The bill was rushed to passage before the first anniversary of the bombings. But the death penalty “reform” on which the law hinged mainly affected state court cases, where the overwhelming majority of death row inmates are sentenced. Terrorists like McVeigh are prosecuted in federal court.

If one source of the problem is in congress, another is at the local level where most death penalty cases originate. Here, too, there is need for reform. The rate and effectiveness with which the death penalty is used varies drastically from states like Ohio, which has executed just one person since reinstatement, to Virginia and Texas, which together have executed 247 people (44 percent of all executions). In recent years, the scrutiny surrounding cases like Aaron Patterson’s has made Illinois ground zero for the death penalty debate. The state’s 12 exonerations are symptomatic of the problems in the way cases are prosecuted and defended.

Patterson’s case illustrates many of these problems. To begin with, he was represented by public defenders and was therefore an easier target. Numerous studies have shown that public defenders often lack the skill, resources, and commitment to defend capital cases properly. Not surprisingly, 90 percent of current death row inmates didn’t have a private attorney. As prosecutor David Lazarus has noted, “Clients of the world’s great defense attorneys (and even the good ones) don’t receive death sentences. Almost without exception, a prerequisite for receiving a death sentence is the inability to hire a lawyer sufficiently talented or motivated to mount a credible defense.” Public defenders, he continues, will “substantially increase the probability that a defendant will be convicted of capital murder as opposed to some appropriate lesser offense.”

Overzealous prosecutors are another problem. Tenet number one for the career-minded prosecutor is “get tough on crime.” While ambitious prosecutors often tout their death penalty convictions, they’re far less likely to question whether such convictions are cost effective or necessary. The number of death sentences later reversed suggests that in many cases they’re grossly unfair: 40 percent of death row cases are vacated at some point in the appeals process. This can occur ten or even 15 years down the line, at enormous cost to taxpayers and tremendous injustice to defendants. But as a rule, prosecutors don’t abandon death penalty cases for fear of appearing soft on crime. Federal judge Alex Kozinski has pointed out that while 80 to 90 percent of all criminal cases end in plea bargains, capital cases almost always go to trial. There is little justification for this discrepancy, given that plea-bargaining death penalty cases would limit costs and nullify the issue of protracted appeals.

The tough-on-crime mentality also infects state lawmakers who, like congress, can broaden statutes to include more death penalty crimes. Illinois, which in 1977 had seven crimes that qualified for death, today boasts 19. Often, new additions are an ill-considered reaction to high-profile crimes like last year’s murder of a neighborhood activist in Chicago, which prompted legislators to make killing an activist a death penalty offense. Such moves are well intentioned but only hamper an overburdened system.

Prosecutorial excesses also run up costs. While defense lawyers are commonly faulted for delays, prosecutors share the blame. Because of their never-say-die attitude toward upholding convictions, they often drag their feet when exculpatory evidence raises questions of innocence. Prosecutors fought to prevent DNA testing that ultimately freed Ronald Jones, the last man to walk off Illinois’ death row. Jones spent nearly two years on death row after tests proved his innocence because prosecutors were unwilling to abandon their case.

In fairness, Illinois has taken steps to address the most serious shortcomings, and other states would do well to follow its lead. Recognizing the need for adequate representation of the poor, the state legislature stepped in to save Illinois’ legal resource center when federal funding ended. The result is the Capital Litigation Division of the Illinois State Appellate Defender’s Office, essentially a state-funded resource center.

Illinois also joined New York in passing a provision for post-conviction DNA testing. The law acts as a safeguard against mistaken convictions by allowing inmates to petition for DNA analysis of evidence used to convict them. No other state has used DNA to uncover more wrongly convicted defendants than Illinois. Perhaps most noteworthy in the near term is Illinois’ activist community. It has played a part in freeing all of the 12 released death row inmates and forced many of the positive changes in the criminal justice system.

Finally, most of the state’s major media outlets have given nuanced coverage to death penalty law and have provided detailed reporting on flaws and abuses. This is in marked contrast to the national media, which eagerly covers high-profile exonerations but shies away from stories that don’t rise to the level of a 60 Minutes piece. Consider, for example, the sloppiness of the May 31 Newsweek feature on Aaron Patterson and Northwestern University professor David Protess’ efforts to free him. “The Newsweek article had by my count and my students’ count 21 factual errors in three pages,” Protess says. “In the entire time that I’ve been involved in this, I’ve never seen an article more off-base than that.” The author wrote that Patterson “consented to an oral confession,” which he did not. She also wrote that the jury “found it hard to believe that Patterson, the son of a Chicago police lieutenant, would be mistreated.” In fact, the jury never heard any evidence of physical abuse, thanks to a pretrial ruling. If journalists were to cover the death penalty as thoroughly as they’ve covered welfare reform, such carelessness would be less common, and the shortcomings of the death penalty system would be a far more prominent public concern.

Pragmatism should eclipse the battle between the extremes of liberalism and conservatism that have reigned on the death penalty. Since courts have been unable to speed processing time adequately, the solution is to reduce the number of death sentences handed down. The first step is to provide resources for the system to function properly, then to enact laws that ensure it does so in a timely manner. Improvements should begin with the resurrection of legal resource centers. Qualified resource center attorneys would prevent prosecutors from running up capital convictions against unprepared or unqualified public defenders. In fact, the motivation for ambitious prosecutors would shift from quantity of convictions to quality. They’d need to be certain of an airtight case before seeking the death penalty, since failing to win a conviction is a political embarrassment. This might even have the secondary effect of removing the stigma attached to settling dubious capital cases.

Resource centers would have a trickle-down effect that would alleviate problems later in the justice cycle: only the most deserving killers would be executed; more convictions would be upheld; having competent lawyers at trial would eliminate the claims of incompetent counsel that delay so many cases today. Cases would proceed more quickly because defendants wouldn’t have to wait years for the court to assign them an attorney. And fewer convictions and appeals would curtail the staggering cost to taxpayers.

Even the most conservative estimate of savings is substantial. Federal Judge Kozinski estimates each death penalty case (in state court) costs taxpayers $1 million. If each case were thought of as a million-dollar tax increase, prosecutors would have to be much more discriminating. Cutting the number of death sentences in half would save $150 million each year. To put that figure in perspective, resource centers would need to prevent just 20 cases a year to cover the funding Congress cut for them in 1995.

States need to institute minimum qualifications for defense lawyers. Studies show that often it’s the worst lawyers, not the worst criminals, who draw death sentences—a fact illustrated in in the death penalty case of George McFarland in Texas, whose lawyer repeatedly fell asleep in court (“It’s boring,” he explained to the Houston Chronicle). This reform too would cut time and cost. The justice department concluded last fall: “Assuring appropriate resources for the defense at the trial stage minimizes the risk of time-consuming and expensive post-conviction litigation.” States could also place time limits on appeals (some already do) to force delay-minded defendants to air their claims early on, rather than withhold them for time-consuming appeals. Appeals could also be bundled and made at one time, rather than sequentially. An exception would of course be made for evidence discovered after conviction. Conservatives, no matter how justified their frustration over delays, should not perpetuate a system that refuses to hear exculpatory evidence. By amending laws like Virginia’s 21-Day Rule to allow a judge to rule on credible new evidence, lawmakers would remove a major incentive for delay. The amendment should mirror new programs like one in Illinois that allows post-conviction DNA testing. As a final safeguard, Illinois’ attorney general has proposed a board to review capital cases before execution—an idea endorsed by both the state bar association and conservatives like Scheidegger.

None of these improvements will come without political leadership. So politicians sincere about being tough on crime need to be held accountable for the problem. This spring the Texas legislature unanimously passed a bill that would have created an agency to match indigent capital defendants with qualified lawyers. (That Texas, which leads the nation in executions, could pass such a bill unanimously is an indication of how badly it is needed). Nevertheless, Governor Bush vetoed it in June. Undoubtedly, he intended to send a message to voters. But his veto killed a practical, bipartisan agreement that would have hastened justice and likely saved lives. Voters should send their own message right back.

The way to repair the “machinery of death” is to shift resources to points earlier in the criminal justice cycle. Only then, with the chance for error greatly diminished, would faster processing be justified. Proponents of capital punishment would have an efficient and equitable system, while opponents could argue abolition purely on its merits. With violent crime at its lowest rate in 25 years, there is no better time for sensible reform.

Our ideas can save democracy... But we need your help! Donate Now!

Joshua Green is a national correspondent at Bloomberg Businessweek, a CNN political analyst, and the author of Devil’s Bargain: Steve Bannon, Donald Trump, and the Nationalist Uprising. He was an editor at the Washington Monthly from 2001 to 2003.