News your connection to The Boston Globe

Locked in

Twenty years ago, a Massachusetts judge helped draft a set of rules that promised to make criminal sentencing fairer and more rational. Today, will he stand in the way of cleaning up the mess that they've become?

TWENTY YEARS AGO, it looked like a utopian moment in criminal-justice reform was dawning in Boston. In 1984, while serving as a federal appeals judge on the First Circuit here, Stephen Breyer helped design a new set of federal sentencing guidelines that promised to end the power of judges to impose unduly long sentences, or ill-advisedly short ones, for just about any reason they chose. Hailed for their transparency and rationality -- not just sentencing by numbers, but sentencing by science -- the new rules were ushered through Congress by the unusual bipartisan teamwork of Senators Edward Kennedy and Strom Thurmond. The Massachusetts senator likened past sentencing to "roulette" and declared that the passage of the new rules "represents a major step toward a fairer and more effective criminal justice system."

But two decades later, the most uniform thing about the guidelines is the loathing that the federaljudiciary and the defense bar feel for them. To the judges, the guidelinesare a time-consuming straitjacket that have increased the power of prosecutors at the expense of the bench. To defense lawyers, they have resulted in sentences that are not so much uniform as uniformly long. Since the rules went into effect in 1987, the number of federal prisoners has jumped from fewer than 42,000 to more than 177,000 and their average time served has doubled.

In June, five justices of the Supreme Court essentially tolled the death knell for the guidelines with its ruling in Blakely v. Washington. In theory, the court struck down the sentencing rules followed only by the state of Washington. But it didn't take long for federal judges and academics across the country to connect the dots and realize that the decision may well lead to the end of the federal guidelines as well.

Blakely ruled that judges can't increase a sentence above the legally authorized maximum based on facts that a jury hasn't found. Under the federal guidelines, judges do that sort of fact-finding at sentencing all the time -- in 90 percent of cases, according to one estimate. That explains why four of five federal appeals courts that have so far wrestled with Blakely's impact have already found the federal rules unconstitutional -- and why thousands of defendants, from Martha Stewart on up, have tried to take advantage of the ruling.

Meanwhile, the man who helped create the federal guidelines in 1984 finds himself in a strange -- and once again pivotal -- position. In Blakely, Breyer was the sentencing guidelines' lone supporter among the court's left-leaning members. Now that the question is what's next for federal sentencing, in the courts and in Congress, Breyer needs to shift ground -- or risk pushing along a mop-up of the sentencing mess that would actually make the current system even worse.

. . .

It wasn't supposed to be this way. The federal guidelines weren't designed to make sentences more lenient (if anything, Breyer and Kennedy emphasized a tough approach to violent crime). But the rules did aim to make punishments fairer. The guidelines set tighter ranges of punishment for each federal crime, so that judges were no longer free to choose anything from, say, probation to 20 years, or 15 years to life.

At the same time, Breyer and the other architects of the guidelines believed that two defendants convicted of the same crime didn't necessarily deserve precisely the same punishment. For example, they argued, a bank robber who brandished a gun should be punished more harshly than one who didn't raise his gun or didn't have a gun at all. To make room for such shades of difference, the guidelines asked judges to look beyond the charges brought by prosecutors and the facts found by juries. At sentencing, the judges were instructed to determine the defendant's "real offense," by taking into account many offense characteristics and all of the "relevant conduct" connected to his crime.

But this reform has had a perverse effect. Suddenly, a drug dealer convicted of trafficking 500 grams of cocaine could be sentenced for selling two kilos if the testimony at trial -- or, in the far more likely event of a plea bargain, a report by the probation department -- persuaded the judge that the larger drug sale had probably occurred.

To be sure, once a judge makes such a determination, he or she is boxed into a 43-level grid, with each sentence calibrated to a different matrix of crime, relevant conduct, and criminal history. But many of the key factual criteria that add or subtract years from a sentence -- whether a bank robber brandished his gun, the amount embezzled by a white-collar executive -- were up to the judge to decide. And the standard for that decision was not the standard used by juries -- beyond a reasonable doubt -- but the weaker "preponderance of the evidence," which just means more likely than not. The guidelines retained this traditional aspect of the judge's role because Breyer and others feared that the more extensive procedures available under the higher standard, like the rule against hearsay evidence, would prove woefully burdensome for the courts.

But on June 24, the Supreme Court effectively put an end to all this judicial decision-making. In the Washington case, Ralph Blakely pled guilty to second-degree kidnapping, for which the maximum sentence was normally 53 months. The judge found that he'd acted with "deliberate cruelty" and increased the sentence to 90 months. That violated Blakely's right under the Sixth Amendment to be tried by a jury of his peers, said Justice Antonin Scalia.

Scalia wrote for an odd majority: He and Clarence Thomas are both arch conservatives, while Ruth Bader Ginsburg, David Souter, and John Paul Stevens are generally liberal. For different reasons, the five found common ground. Scalia and Thomas cared about the conception of the jury held by the framers of the Constitution. "The jury could not function as circuit breaker in the State's machinery of justice it if were relegated to making a determination that the defendant at some point did something wrong, a mere preliminary to a judicial inquisition," Scalia said in his opinion.

The three liberals haven't written to explain their positions, but it's not hard to imagine their concern: more prisoners sitting behind bars for more years. The National Association of Criminal Defense Lawyers filed a friend-of-the-court brief that urged the result reached in Blakely and calls the decision "a great leap forward." As Scalia drily noted, "It is hard to believe that the National Association of Criminal Defense Lawyers was somehow duped into arguing for the wrong side."

In his dissent, however, Breyer essentially argued just that. He pointed out that Blakely leaves legislatures with two choices.

The first is to revert to giving judges wide discretion over punishment -- an option Breyer bemoaned as a step back to the era when lawyers quipped that a sentence could depend on "what a judge ate for breakfast." The second is for legislatures to turn over to juries the expanded fact-finding role at sentencing that the 1984 guidelines gave to judges. Breyer argues that this would disadvantage the 95 percent of federal defendants who plead guilty. He claims that if the defendants know they've lost the opportunity to beg a judge for a lower sentence, they'll plead to a higher charge than they would otherwise accept.

It's difficult to predict the effect of Blakely on plea bargaining -- difficult enough to have sparked a heated and unresolved academic debate. But it's easy to refute Breyer's further claim, which is that for the small minority of defendants who go to trial, jury fact-finding will require long and unwieldy sentencing hearings that will be too costly for the states.

In fact, the evidence points in the opposite direction. In 2001, the Kansas Supreme Court struck down that state's sentencing guidelines, citing the same reason the majority gave in Blakely. The sky didn't fall. When they want a heavy sentence, prosecutors make their case to a jury in a post-conviction proceeding that generally takes about an hour. The state counted only three such cases in 2002 and 2003. Kevin Reitz, a sentencing expert at the University of Colorado law school, said in a recent interview that the number is an undercount, but not by much.

In Washington, Kansas, and the dozen other states with sentencing guidelines, Blakely isn't a major upheaval because it affects fewer than 10 percent of criminal cases, Reitz added. That's because the states never fully embraced the ideas of Breyer and his fellow reformers to begin with. "No state system ever adopted anything like the relevant conduct standard that has led to such horrendous results in the federal system," Reitz says. "The guidelines states are going to be able to adapt to Blakely in ways that don't derail them."

But the federal courts are another matter. There, as the tumult since Blakely illustrates, judicial fact-finding is uniquely pervasive. That's one reason that lower courts have decided that the federal guidelines are now unconstitutional. "While Blakely has gone a long way to make the sentencing system more fair, and to reinvigorate the role of juries in the process, it is inconceivable that the system now required by the decision is at all consistent with anything contemplated" by the guidelines, wrote federal district judge Nancy Gertner in Boston in a thoughtful decision issued last week.

. . .

If juries can't easily apply the federal guidelines, why not simplify them? In fact, an alternative to the complex federal rules can be found in the more straightforward guidelines adopted by several states. Led by Minnesota in 1980, nine states have enacted guidelines based on the crime the defendant committed and his criminal history, and also abolished parole so that defendants aren't released early on the back end. These states have increased the average sentence for a serious violent offense. Yet eight of the nine have slower rates of prison growth than the national average, according to a recent study by Reitz.

The states have achieved this balance by explicitly considering the cost of prisons and directing judges to give alternative sentences to some nonviolent offenders. A guidelines proposal in Massachusetts pending in the legislature adopts many of the successful elements of these systems. It's been stalled for years, however, over whether to loosen mandatory drug sentences for nonviolent offenders, and has been shelved for another year as a result of the uncertainties created by Blakely. Meanwhile, Massachusetts state prisons and county jails are overflowing with more than 22,000 inmates, compared to fewer than 6,000 in 1980.

But as Congress decides whether and how to address Blakely, it's not talking about following the states' example. Federal spending on prisons has risen steeply but remains a tiny fraction of the overall federal budget, which means that Congress faces little of the cost-control pressure that state legislatures do.

Instead, Congress is feeling political heat as defendants sentenced in the wake of Blakely get shorter terms than they otherwise would have -- six and a half years instead of 19 years for a crack dealer in Maine sentenced at the end of June, for example. Likely appeals from high-profile white-collar defendants like former Dynegy executive Jamie Olis, sentenced to 24 years in April, have also spurred action.

The Republican-led Senate Judiciary Committee is floating two proposals in response to Blakely -- both of which would probably lead to more long sentences. The first would increase judges' freedom to sentence above the current guidelines, by replacing the top of that range with the higher statutory maximum. The second would enact more mandatory minimum sentences for federal drug crimes.

For Breyer and Kennedy, the onslaught of mandatory minimums has been a real blow. Since their enactment in the early 1990s, both men have vehemently opposed mandatory minimums for undermining the guidelines' edifice of carefully calibrated punishments. Such blanket punishments "rarely reflect an effort to achieve sentencing proportionality -- a key element of sentencing fairness that demands that the law punish a drug `kingpin' and a `mule' differently," Breyer wrote in the 2002 case United States v. Harris.

Yet in Harris, Breyer had the chance to kick some mandatory minimums out the constitutional door -- and he didn't take it. Instead, he provided the key fifth vote in upholding the authority of judges, rather than juries, to find the facts that trigger a mandatory minimum sentence. Clinging to his first principles -- transparent and proportional punishment -- the justice was trying to stave off a frontal assault on the guidelines.

But that battle has now been lost with Blakely. A big question among the academics and judges who avidly track sentencing developments (check out the action at is whether Breyer will now reconsider his defensive -- but now ineffectual -- vote in Harris.

Court watchers may not have long to wait. Prosecutors seeking clarity have already brought Blakely test cases to the Supreme Court, urging a fast review. These cases come from Wisconsin and Maine; in addition, the court could in its next term hear an appeal of a decision by Utah judge Paul Cassell, a conservative Republican. Faced with sentencing a drug offender to 62 years in prison because of mandatory minimums, Cassell invited the man's lawyers to challenge them. The case could become a vehicle for the Supreme Court to reconsider Harris. If that happens, Breyer will have to decide whether to concede that his vision of federal sentencing is dead -- and whether to cast his future votes in a way that makes the best of the new reality. 

Today (free)
Yesterday (free)
Past 30 days
Last 12 months