Prison conversion.

After studying nonviolent drug offenders, a criminologist who once said "Let 'Em Rot" now says "Let 'Em Go."

Denese Calixte, an illiterate Haitian immigrant, managed to support her seven children by picking fruit in Florida for about $60 a day. In 1994, she fell from a ladder and injured her neck. Unable to work, she was desperate for money when a neighborhood crack dealer offered her $200 to keep his supply in her house at night. Early one morning, the police broke into her house and found 69 grams of crack. Convicted of possession with intent to deliver, Calixte was sentenced to 10 years in federal prison.

Loren Pogue, a real estate agent from Missouri, does not fit the stereotype of a drug trafficker. The father of 27 children, 15 of them adopted, and the co-founder of a home for orphans, he is an Air Force veteran, a Shriner, and a member of the Lions Club, the VFW, and the American Legion. Until 1990, Pogue's only trouble with the law was a 14-month sentence for tax evasion in the 1970s. But then he helped a part-time employee sell a plot of land in Costa Rica, where Pogue had moved with his family. As it turned out, the employee was a paid government informant, the buyers were federal agents pretending to be drug dealers, and the sale was a trap. Because the agents had talked about using the land as an airstrip to fly cocaine into the United States (even though the site was not suitable for that purpose), Pogue was charged with conspiracy to import drugs and money laundering. He is serving a 22-year sentence.

Despite a heroin habit that began in high school, Brenda Pearson was leading a stable, productive life. Married for two decades, she had two children and a job at a New York securities firm. She had never sold drugs or stolen anything to support her habit. In 1994, a close friend of Pearson's who was also a heroin user moved from New York to Michigan. She asked Pearson to mail heroin to her because she was having trouble finding a satisfactory supply. Over the course of a year, Pearson sent her friend about 40 packages, each containing less than 1.5 grams of heroin in single-dose bags that she had purchased on the street. When Pearson was caught, New York and federal prosecutors were not interested in the case, since she was simply sharing heroin with a friend. But a Michigan prosecutor had Pearson extradited and charged her with heroin distribution. She plead guilty to 10 counts and received 10 consecutive sentences of five to 20 years each. She will not be eligible for parole until she serves her 50-year minimum.

In 1991, David Ciglar, an Oakland, California, firefighter and paramedic credited with saving more than 100 lives, was training for a new career as an MRI technician, having been disabled by an injury during a rescue. Acting on a tip, agents of the Drug Enforcement Administration found 167 marijuana seedlings in his garage. To prevent prosecutors from charging his wife, which would have sent her to prison and their three children to a foster home, Ciglar pleaded guilty to marijuana cultivation. Because he had a prior conviction for cocaine possession and the DEA found an antique gun replica in his house, Ciglar received a 10-year sentence. The government also confiscated the family's home.

These cases, described in the newsletter of Families Against Mandatory Minimums and in the book Shattered Lives: Portraits From America's Drug War (www.hr95.org), are supposed to make you shake your head. Even people who support the crusade for a drug-free America can see that something is wrong with separating decent, productive individuals from their families and making them serve hard time with violent felons. But while opponents of prohibition tend to see such cases as typical, representing the inherent injustice of the war on drugs, supporters see them as aberrations. According to this view, despite occasional instances of excessive punishment, drug offenders generally deserve the treatment they get: Not only do they contribute to a host of social problems by peddling addictive drugs, they also commit crimes against people and property. In other words, they would belong in prison even if there were no drug laws to break.

Assuming that involvement with illegal substances is a marker for anti-social behavior, the drug laws may very well make it easier to incapacitate predatory criminals. But so might a policy that allowed police to round up young black men, guys with tattoos, or simply people they thought looked suspicious. The fact that such discrimination might help prevent crime does not make it fair or reasonable. Nevertheless, it is much easier to support the status quo if you believe that people imprisoned for drug offenses are, with few exceptions, dangerous thugs. Conversely, if you believe that many drug offenders are people like Denese Calixte, Loren Pogue, Brenda Pearson, and David Ciglar - people who do not represent a threat to others - the argument for keeping them all locked up is far less compelling.

Until recently, it was impossible to say with much confidence which view was closer to the truth. But during the last five years or so, evidence from various sources has indicated that a sizable percentage of convicted drug offenders, and perhaps a majority of those now being sentenced in some jurisdictions, are not predatory criminals. These "drug-only offenders," as researchers have dubbed them, represent a substantial and growing share of the total prison population, largely because of state and federal laws that impose mandatory minimum sentences without parole for certain drug crimes. The troubling implications of this trend for both justice and public safety have made sentencing reform a cause embraced by hard-headed conservatives as well as soft-hearted liberals.

John J. DiIulio Jr., a criminologist who has little patience with what he calls "the soft-in-the-head anti-incarceration left," is emblematic of this shift. DiIulio was initially skeptical of the notion that the drug laws imprison large numbers of people who are not menaces to society. But he recently completed research that provides some of the strongest evidence yet for this claim. Though he still supports prohibition - not surprising for a political scientist who trained with a prominent defender of the drug laws, James Q. Wilson, and co-authored a book on crime with former drug czar William Bennett - DiIulio has become an outspoken critic of current sentencing policy. "Basically, what you're getting from me is coerced by the latest data," he says. "It seems to me that with respect to these drug offenders, the mandatory minimums have begun to go haywire."

A 41-year-old Harvard Ph.D., DiIulio is a former professor of politics and public policy at Princeton and former director of the Brookings Institution's Center for Public Management. He is now a senior fellow of the Manhattan Institute and the Fox Leadership Professor of Politics, Religion, and Civil Society at the University of Pennsylvania. An expert on prison management who published influential books on that subject in 1987 and 1991, he later turned his attention to the cost-effectiveness of incarceration. Based on surveys of prisoners, he concluded that it costs less to keep the average predatory criminal locked up than it does to let him go, given the crimes he would commit if he were free. Publishing articles with titles like "The Value of Prisons," "Prisons Are a Bargain, by Any Measure," and "Let 'Em Rot," DiIulio was often cited by conservatives advocating more prisons and longer sentences. "No one - at least, no one in elite policy-wonk circles - is a bigger fan of incarcerating known, adjudicated adult and juvenile criminals than me," he wrote in a 1996 article for Slate.

But DiIulio's argument for the cost-effectiveness of prisons hinged on the ability of the criminal justice system to distinguish between offenders who would prey on their fellow citizens if released and offenders who would not pose a threat to others. Locking up people in the latter category would not only waste money; it would waste prison space that could otherwise have been used to incapacitate predatory criminals. Given this consideration, the question of how to deal with drug offenders was bound to come up.

DiIulio took an early crack at the issue of whether prison is cost-effective in a 1991 article for The Brookings Review co-authored by Anne Morrison Piehl, then a Princeton doctoral candidate and now an economist at Harvard's John F. Kennedy School of Government. There DiIulio and Piehl simply excluded drug crimes from their calculations, following the standard practice of other researchers in this area. "This is not the place to referee the drug debate," they wrote.

In a Wall Street Journal op-ed published the next year, DiIulio downplayed the distinction between drug offenders and predatory criminals. "About 90 percent of prisoners are serious (two or more felony convictions) or violent offenders," he wrote, citing what became one of his favorite statistics. That figure, based on a 1991 survey of state prisoners by the Bureau of Justice Statistics, puts people arrested twice for drug offenses (such as David Ciglar, the California firefighter) in the same category as serial bank robbers, rapists, and murderers. In the opening paragraph of the Journal article, DiIulio referred in one breath to "criminals who assault, rape, rob, murder and deal drugs."

When DiIulio directly addressed the treatment of drug offenders in a 1993 book review for The New Republic, he implied that they were getting off too lightly. "It is not unreasonable to argue," he wrote, "that the problem with the 'get-tough' approach of the last twenty-five years is that it hasn't actually been followed. Despite mandatory sentencing laws, most drug offenders and other felons continue to spend only a fraction of their sentences behind bars."

As part of a "radical experiment" in "the serious enforcement of existing laws," DiIulio said, "prosecutors and judges would need to tighten the screws on drug-law violators who are used to copping pleas to lesser charges and spending little or no time locked up." Denying that "the war on drugs has placed tens of thousands of petty first-time drug-law violators behind bars," he said "the vast majority of 'drug' criminals behind bars were convicted of sale or manufacture, not simple possession, and had prior criminal records. Many of them would be and should be incarcerated even if the antidrug laws did not exist."

Meanwhile, an increasing number of researchers and judges were concluding that the sentencing policies set by Congress in the 1980s, at the height of the Reagan-Bush drug war fever, were senselessly rigid and excessively harsh. Supposedly aimed at "drug kingpins," the mandatory minimum sentences that Congress began to enact in 1986 - which were tied to drug weight, without regard to the extent of the defendant's culpability or the level of his involvement - mainly affected small-time operators and flunkies. According to a 1994 report from the Federal Judicial Center, less than 15 percent of the defendants convicted under mandatory minimum statutes in fiscal year 1992 were drug trafficking managers or supervisors. Not surprisingly, the report concluded that "mandatory minimums have had no observable effect" on crime or drug availability.

Also in 1994, a report from the U.S. Department of Justice concluded that "a substantial number of drug law violators who are sentenced to incarceration in Bureau of Prisons custody can be classified as 'low-level.' "Under one set of criteria used in the report - no record of violence, no involvement in sophisticated criminal activity, and no prior imprisonment - there were 16,316 such prisoners in June 1993, representing 36 percent of drug offenders and 21 percent of all federal inmates. Even when the DOJ restricted the category to prisoners with no prior arrests in their records, there were 9,673 low-level offenders, 21 percent of the people serving time for drug crimes. Average sentences were nearly seven years for the first group, more than five years for the second. The report noted that sentences for drug offenders with minimal or no criminal histories had more than doubled since 1985. "The relative harm of drug trafficking," it said, "has been elevated above that of almost every serious crime other than murder."

By the early 1990s, federal district court judges such as Jose Cabranes in Connecticut, Stanley Sporkin in the District of Columbia, and Jack B. Weinstein in New York were openly criticizing the laws that forced them to give sentences they considered unjust. They were joined by appellate judges and Supreme Court justices. Testifying before Congress in 1994, Justice Anthony Kennedy said, "I think I'm in agreement with most judges in the federal system that mandatory minimums are an imprudent, unwise, and often unjust mechanism for sentencing." In a 1996 survey by the Federal Judicial Center, four-fifths of district and circuit court judges said the guidelines written by the U.S. Sentencing Commission should not be linked to the penalties set by Congress, an approach that has made sentences longer even for offenders not covered by the statutory minimums.

The federal experience with mandatory minimums was mirrored at the state level, where sentences could be even more onerous. Michigan had its notorious "650 lifer" law, under which 650 grams of cocaine or heroin, less than a pound and a half, triggered a life sentence without parole. But the real leader in mandatory minimums was New York, which had been punishing drug offenders with unusual severity since 1973. That was the year that Gov. Nelson Rockefeller pushed the state legislature to pass two laws, the first establishing mandatory minimums for drug offenses, the second raising sentences for repeat offenders. These laws have been modified over the years, mainly to reduce penalties for marijuana offenses, but they remain exceptionally harsh. A first-time offender convicted of selling two ounces or possessing four ounces of heroin or cocaine receives a mandatory sentence of 15 years to life - the same as the penalty for murder. Possessing as little as half a gram of pure cocaine or heroin triggers a minimum sentence of a year. If it's a second offense, the sentence is doubled.

A 1997 report from Human Rights Watch noted that drug offenders represented more than a third of New York's prison population, up from less than a tenth in 1980. This group was not limited to predatory criminals: Forty-four percent had never been incarcerated before, and 17 percent had no prior arrests. Even the repeat offenders did not look so dangerous upon closer examination: Among the drug offenders sentenced as second-time felons in 1995, 55 percent had been convicted of another drug crime the first time around; only 20 percent had been convicted of a violent felony.

Like federal mandatory minimums, the Rockefeller drug laws ' come down hard on people who are not exactly kingpins. In a 1993 case, an appeals court reviewed the sentence of Jesus Portilla, an asbestos remover with a wife and small child. A first-time offender, he had received a sentence of eight and a third to 25 years for a $30 cocaine sale. In 1997, Gov. George Pataki granted clemency to Angela Thompson, a first-time offender who had served eight years of her 15-year sentence. When she was 17, she had done the bidding of her uncle and legal guardian by trying to sell cocaine to an undercover cop.

As with the federal sentences, judges were among the first to recognize the injustices created by the Rockefeller drug laws. In a 1976 dissent, three judges on the New York Court of Appeals, the state's highest court, called a sentence of 15 years to life for heroin possession "unconscionable and barbaric." State trial judges publicly expressed their anguish at being forced to impose palpably unfair punishments. By 1995, George Pataki, another ambitious Republican at least as anxious to maintain a tough-on-crime image as Nelson Rockefeller, was conceding that harsh penalties for drug offenders "have filled New York's prisons and have not increased public safety."

Surprisingly, critics of the Rockefeller drug laws also included John DiIulio, who called for their abolition at an October 1995 governor's forum in Albany. His decision to oppose mandatory minimums came around the same time as a shift in his rhetoric regarding drug offenders. Five months before, at a conference in Berkeley, he had observed that "some small but as yet undetermined fraction of imprisoned state and federal drug law violators are neither major drug traffickers nor persons who have committed lots of serious non-drug felonies. At present, it is impossible to know how many low-level 'drug-only' offenders - first-time or repeat criminals whose only crimes, or most serious crimes, have been low-level drug crimes or mere possession - are behind bars in America today. One recent survey of state prisoners suggests that the figure could be as high as 15 percent."

DiIulio was referring to a 1993 study that he and Anne Morrison Piehl had conducted in New Jersey, where 30 percent of the respondents reported that drug offenses were the only crimes they had committed in the four months before they were locked up. "If even half of the inmates who report that their only crime was selling drugs are telling the truth," they wrote in the Winter 1995 Brookings Review, "then 15 percent of New Jersey's spending on prisons is being devoted to 'sending a message' about drug dealing." In the New Jersey study, DiIulio says, "you start to get a hint of this drug-only population being a not huge but nonetheless significant part of the mix."

DiIulio and Piehl emphasized that drug dealers are easily replaced, so "the best estimate of the incapacitation effect (number of drug sales prevented by incarcerating a drug dealer) is zero." DiIulio says this conclusion was strongly influenced by the views of his mentor, Wilson, who had written in Commentary the previous year that prison terms for crack dealers "do not have the same incapacitative effect as sentences for robbery. ... A drug dealer sent away is replaced by a new one because an opportunity has opened up." While cautioning that "we do not yet have a definitive estimate of the fraction of the prison population that consists of drug-only offenders," DiIulio and Piehl suggested that locking up such offenders was a waste of money.

DiIulio continued this theme in a 1996 New York Times op-ed. "Prison definitely pays," he wrote, "but there's one class of criminal that is an arguable exception: low-level, first-time drug offenders. ... Though the numbers of petty drug offenders may prove small, it makes no sense to lock away even one drug offender whose case could be adjudicated in special drug courts and handled less expensively through intensively supervised probation featuring no-nonsense drug treatment and community service." He repeated his call for repealing the Rockefeller drug laws, describing their abolition as part of the effort "to keep violent and repeat criminals where they can't harm the rest of us."

Two months later, however, DiIulio was again emphasizing that drug offenders are dangerous characters who belong in prison. "Almost all imprisoned drug traffickers, federal and state, have long criminal records, adult and juvenile," he wrote in "No Angels Fill Those Cells," a Washington Post op-ed. "Only their latest or most serious adult conviction is for a drug offense. There are exceptional cases of truly first-time, non-violent, low-level drug offenders who end up behind bars. But they are the exceptions that prove the rule. ... Rather than accepting the usual bogus generalizations about drug offenders or non-violent offenders behind bars, demand to know...the totality of the adult and juvenile crimes committed by imprisoned felons against life, liberty, or property... In the end, the truth about prisoners' complete criminal histories will prevail - and the truth will set very few prisoners free."

That conclusion was based largely on a study for the Wisconsin Policy Research Institute that DiIulio and WPRI analyst George A. Mitchell had just completed. In a sample of 170 inmates from Milwaukee County, they found that 12, or 7 percent, had been sentenced for drug offenses. Within this group, 75 percent had a record of violence. Of the remaining three, one had been convicted of various property crimes, including burglary and auto theft. So just 17 percent of the drug offenders, and less than 3 percent of the total prison sample, were guilty only of drug crimes.

DiIulio concedes that the drug offender subgroup was "a teeny, teeny sample"; the study was not designed to get at the issue of drug-only offenders. Still, "it gave me a little bit of pause," reinforcing his initial inclination to believe that drug-only offenders are a rare breed. He reiterated that impression at a talk he gave at the Lindesmith Center, a Manhattan-based drug policy think tank, a few days before his essay appeared in the Post. But he said he was undertaking a study of New York prisoners in which he planned to look at a larger sample.

While DiIulio was working on that project, evidence from Massachusetts suggested that drug-only offenders are not so unusual after all. A 1997 study by Piehl, DiIulio's collaborator on the New Jersey survey, found that two-thirds of the state's recently incarcerated drug offenders had never been convicted of a violent crime. Data obtained by The Boston Globe from the Massachusetts Department of Corrections in 1998 showed that 84 percent of prisoners serving mandatory sentences for drug crimes were first-time offenders in the state.

From late 1996 through September 1997, DiIulio, Piehl, and University of New Mexico sociologist Bert Useem surveyed about 1,500 incoming prisoners in New York, Arizona, and New Mexico. They used anonymous questionnaires to ask the prisoners detailed questions about their backgrounds and criminal histories. They were not able to obtain access to the prisoners' records for confirmation, but research by the RAND Corporation in the 1970s had shown a good correspondence between inmates' self-reports and their rap sheets, once the data were adjusted to exclude apparent fakers (those claiming to have committed thousands of crimes in a year, for example).

DiIulio and his collaborators found that drug-only offenders accounted for 28 percent of incoming male prisoners in New York, 18 percent in Arizona, and 15 percent in New Mexico. Among female prisoners, the figures were 49 percent, 20 percent, and 14 percent, respectively. In New York, drug offenders represented 47 percent of new prisoners in 1997, so DiIulio et al.'s findings indicate that most of them were not predatory criminals. "There is a large chunk of genuinely drug-only offenders," Useem says. "That was a surprise to me, and I think to John." He says the percentage of prisoners in this category "was strikingly high to me - and worrisome."

These results had a dramatic impact on DiIulio's public posture vis-a-vis drug offenders.

"If it turns out that we have reached the point of diminishing returns with respect to somewhere between a third and half of the people who are now being sentenced to prison under mandatory minimum and kindred laws as drug offenders," he says, "even though [sentencing reform] isn't going to free up a gazillion beds, it's going to free up a certain amount of space, it's going to relieve a certain amount of drain on the public purse, and it's going to make the system more effective at delivering public safety for the marginal tax dollar."

DiIulio laid out this argument in a March 12, 1999, op-ed for The Wall Street Journal. "Current laws put too many nonviolent drug offenders in prison," he declared, citing highlights from the three-state study, which was published in early summer by the Manhattan Institute. Among other things, he recommended abolition of mandatory minimums, reform of the federal sentencing guidelines, release of nonviolent drug offenders, and an end to the federalization of crime policy. "Such changes would undoubtedly reduce the number of drug-only offenders in federal prisons by tens of thousands," he wrote.

DiIulio followed that up with a May 17 National Review article making "a conservative crime-control case" for repealing mandatory minimums. "To continue to imprison drug-only offenders mandatorily," he wrote, "is to hamstring further a justice system that controls crime in a daily war of inches, not miles, and that has among its main beneficiaries low-income urban dwellers." He explains: "I view the criminal justice system as a sorting machine. ... For me, it's about one thing, essentially: How do you improve at the margin the system's capacity to sort in ways that increase public safety?"

DiIulio says "a number of people were surprised by my National Review article. 'Surprised' is one way to put it. ... Another way to put it would be, 'He's been running around in the streets too much, and he finally got hit on the head.'" The critics included his father, a former Philadelphia sheriff's deputy whose reaction DiIulio describes this way: "Spend enough time in these academic places, eventually you'll turn stupid."

The evolution of DiIulio's position partly reflects better-focused research with larger samples. It also reflects a changing reality. The number of people incarcerated for drug offenses has been increasing since 1980, in both absolute terms and as a share of the total prison population. The 1991 Bureau of Justice Statistics survey of state prisoners found that 21 percent were drug offenders, compared to 6 percent in 1979. Judging from DiIulio et al.'s figures for recently incarcerated prisoners, that percentage has increased since then. In 1994, 61 percent of federal prisoners were drug offenders, up from 25 percent in 1980. The proportion of local jail inmates who are drug offenders rose from 9 percent in 1983 to 23 percent in 1989 and now appears to have leveled off.

Last year there were 1.8 million people behind bars in the United States, compared to some 500,000 in 1980. About one in four - more than 400,000 - were there for drug law violations, compared to about one in 10, or 50,000, in 1980. As more and more drug offenders have been imprisoned, it is likely that the percentage who are not predatory criminals has risen.

DiIulio says he has seen anecdotal evidence of such a trend: "I've had prison wardens whom I've known for years say, 'I'm getting guys in now who, when I review their sheets, I can't believe they're here.'" The wardens acknowledge that there always were a few such prisoners, he says, "but when the outlier becomes the central tendency, it's time to re-examine policy."

DiIulio says another important influence has been Ethan Nadelmann, a close friend from graduate school who directs the Lindesmith Center and is a leading advocate of drug policy reform. "He's a constant pain in the ass," DiIulio says with a chuckle. "He was constantly saying, 'John, you're a reactionary prohibitionist.'"

For his part, Nadelmann says, "I used to feel that I would win John over two or three days a week, and Wilson would win him over the rest of the time. The problem was that he would do his writing on the Wilson days." But he adds, "If you look closely at John's writings, you can detect that he's never been a vigorous drug warrior, someone who believed that there was an inherent immorality to drug use that merited a draconian government response."

Some observers speculate that religion also had something to do with DiIulio's change of heart regarding drug offenders. A self-described "born-again Catholic," DiIulio has in recent years emphasized the importance of "the faith factor" in preventing crime and dealing with drug addiction. "I think it's been both a personal, spiritual revolution for him and an analytical change," says Julie Stewart, president of Families Against Mandatory Minimums (FAMM). "He's become softer and more forgiving as a result of his religious experience. But he also backs it up with data."

DiIulio discounts the spiritual take on his policy conversion. "I get a lot of that lately on a number of fronts," he says. "The religion thing, I think, might explain a little bit of the compassion." But he notes that his views on drug offenders began changing "when I was still just a good old-fashioned heathen." Stewart concedes that DiIulio has been criticizing mandatory minimums for several years, "but I guess I didn't really hear him until now. We're thrilled.... I look at him as an ally."

Even with scholars like DiIulio advocating the repeal of mandatory minimums, the prospects for fundamental reform are iffy. Progress so far has come in small pieces. The U.S. Sentencing Commission, for instance, has changed the way that LSD and marijuana weights are calculated under the guidelines, leaving out part of the carrier (usually paper) for LSD and making the presumed yield per marijuana plant more realistic. Because these changes do not affect the way drugs are weighed under the mandatory minimum statute, the upshot is sentencing "cliffs": If a first-time offender is caught with 99 marijuana plants, he is subject to Sentencing Commission guidelines; each plant counts as 100 grams of marijuana, corresponding to a sentence range of 15 to 21 months. If he has one more plant, however, he is subject to the congressional weight standards; each plant counts as a kilogram, triggering a mandatory sentence of five years.

A "safety valve" provision approved by Congress as part of the 1994 Omnibus Crime Bill allows judges to sentence certain nonviolent drug offenders under the guidelines instead of the statutory minimums. The safety valve applies to people with minimal or no criminal histories whose offenses did not involve violence, weapons, or injury; who are not considered major players in the drug trade; and who cooperated with the authorities. In 1997, this provision led to significantly lower sentences - for example, three years off a 10-year term - for about 4,000 people, 21 percent of the drug offenders sentenced by federal courts that year.

On the other hand, Congress has blocked the commission's attempt to do away with especially harsh sentences for crack cocaine, which is treated as if it were 100 times worse than cocaine powder, even though the active ingredient in the two substances is identical. A first-time offender with five grams of crack gets five years in federal prison; he would need 500 grams of cocaine powder to trigger the same penalty. Since nine out of 10 federal crack defendants are black, this policy has a strikingly disproportionate racial impact. Rep. Charles Rangel (D-N.Y.), ordinarily a hard-line drug warrior, has introduced a bill that treats crack and cocaine powder the same for sentencing purposes. The other members of the Congressional Black Caucus also support this reform, while the Clinton administration favors reducing the disparity rather than eliminating it.

At the state level, the most dramatic change has occurred in Arizona, where a 1996 ballot initiative required that nonviolent people convicted of drug possession receive probation. The state legislature tried to override the initiative, but voters approved it again in November 1998. Last year in Connecticut, the state legislature unanimously passed a bill that allows parole for nonviolent drug offenders who have served at least half of a two-to-four-year term; previously, such prisoners were eligible for parole after serving at least two-thirds of their sentence. Also last year, Michigan changed its 650 lifer law, making the minimum sentence 20 years and allowing parole after 15. The parole provision was retroactive, so the 220 people already serving life sentences became eligible for future release.

In New York, Pataki has backed away from his initial criticism of the Rockefeller drug laws and suggested only marginal changes. Last May, he proposed allowing appellate judges to reduce the 15-year minimum to 10 years for first-time offenders convicted of possessing small drug amounts. Even that minor reform was tied to his demand for eliminating parole. Meanwhile, the Democratic legislators who were once reliable critics of the Rockefeller drug laws are avoiding the issue, worried about seeming soft on crime.

In other states, there is still support for cracking down even harder on drug offenders. Last year, a group of 38 Republican legislators in Kansas pushed a bill that would have required a life sentence without parole for anyone convicted of growing 100 or more marijuana plants. They were thus proposing to treat marijuana growers more harshly than first-degree murderers, who in Kansas are eligible for parole after serving 25 years.

Nevertheless, there has been a notable change in the climate of opinion among supporters of the war on drugs. Joseph Califano, president of the prohibitionist Center on Addiction and Substance Abuse, declared last year that "mandatory minimum sentences are a round-trip ticket back to jail and into a life of crime." Even Ed Meese, attorney general when Ronald Reagan signed the legislation establishing most of the current mandatory minimums, is having second thoughts. "I think mandatory minimum sentences for drug offenders ought to be reviewed," he told The New York Times in May. "We have to see who has been incarcerated and what has come from it."

Toward the end of his National Review article, DiIulio suggested that opponents of prohibition make it harder to win repeal of mandatory minimums because they "too often characterize all persons incarcerated for drug crimes as casualties of the War on Drugs." Libertarians, of course, would insist that thugs should be locked up for violating other people's rights, not for engaging in consensual activities such as drug dealing. And it's important to keep in mind that prohibition encourages property crime by inflating drug prices and fosters violence by creating a black market. More than one-fifth of the male prisoners in DiIulio et al.'s Manhattan Institute study said they got involved in crime to raise money for drugs, and some drug offenders may have a history of violence because they tried to protect their own lives and property within the context of a black market. We have no way of knowing whether these same individuals would be getting arrested in the absence of prohibition, but it seems likely that at least some of them would not.

Still, DiIulio has a point: It is easy for critics of the war on drugs to imagine that everyone locked up for drug offenses is a medical marijuana user or a hapless mule who swallowed condoms full of heroin out of financial desperation, rather than a vicious, ruthless criminal. But if opponents of prohibition have too rosy a view of drug offenders, policy wonks like DiIulio have been guilty of the opposite error, implying that virtually everyone incarcerated for a drug crime is a predatory criminal. The reality, it seems, is quite different, and DiIulio has been courageous enough to admit that. "You've got to give him credit for being willing to stick his neck out and change," says Stewart, the president of FAMM. "I wish politicians would do that too."

Senior Editor Jacob Sullum ([e-mail]) is the author of For Your Own Good: The Anti-Smoking Crusade and the Tyranny of Public Health (The Free Press).