Cato Policy Analysis No. 208 May 17, 1994

Policy Analysis

Prison Blues:
How America's Foolish Sentencing Policies
Endanger Public Safety

by David B. Kopel

David B. Kopel is research director of the Independence Institute in Golden, Colorado, and an associate policy analyst of the Cato Institute. A former assistant attorney general for the state of Colorado, he is the author, most recently, of The Samurai, the Mountie, and the Cowboy (Cato/Prometheus, 1992), which was named Book of the Year by the American Society of Criminology's Division of International Criminology.

Executive Summary

The amount of money that American taxpayers spend on prisons has never been greater, and the fraction of the American population held in prison has tripled during the last 15 years, as has national prison capacity. Yet the expected punishment of violent criminals has declined, and violent crime flourishes at intolerably high levels. The seeming paradox of more prisons and less punishment for violent criminals, which means less public safety, is explained by the war on drugs. That war has gravely undermined the ability of America's penal institutions to protect the public. As prisons are filled beyond capacity with nonviolent "drug criminals" (many of them first offenders), violent repeat offenders are pushed out the prison doors early, or never imprisoned in the first place.

As prison crowding worsens, many public officials are embracing alternatives to incarceration, such as electronic home monitoring, boot camps, and intensive supervised probation. Although those alternatives have their place, their benefits have frequently been overstated.

The most effective reform would be to return prisons to their primary mission of incapacitating violent criminals. Revision or repeal of mandatory minimum sentences for consensual offenses, tighter parole standards, and tougher laws aimed at repeat violent offenders can help the state and federal criminal justice systems get back to their basic duty: protecting innocent people from force and fraud.

Background: The Imprisonment Surge

For approximately the last 15 years, the United States has been engaged in the largest imprisonment program ever attempted by a democratic society. The number of persons incarcerated has soared to levels unknown in American history. Consider the following statistics.

In 1981 the total number of state and federal prisoners was 369,930. By 1992 the number had risen to 883,593.(1)

The number of sentenced prisoners per 100,000 population in 1980 was 138, a figure that represented a historic high. By 1993 the figure has risen to 344.(2)

In addition to the state and federal prison population, there were 444,584 persons held in city and county jails as of June 30, 1992, a figure that has more than doubled over the last decade.(3)

In addition to the nearly 1 million persons in state and federal prisons, and the over 400,000 in city and county jails, there are over 2 million persons on probation and over half a million on parole.(4)

The federal prison population, which stood at about 24,000 in 1980, had soared to over 90,000 by December 1993 and is expected to rise to 130,000 by the turn of the century.(5)

The drastic increase of the combined state and federal prison population between 1974 and 1990 is mainly the result, not of demographics, but of policy changes. Population growth accounted for 7.7 percent of the growth of the prison population, increased crime for 19 percent, and more arrests for 5.3 percent. The great bulk of the increase-- 60.9 percent--was the result of decisions to send to prison offenders who otherwise would have been given an alternative sentence. And 7.1 percent of the growth resulted from an increase in time served.(6) The relative impact of the increase in time served may grow larger in coming years, as the 10- and 20-year mandatory sentences enacted in the 1980s have their full impact.

There are no signs of the prison surge's abating. State prison populations are up 59 percent in just the last four years.(7) At the rate prisoners are currently being added to the state and federal systems, the United States needs 1,143 new prison beds each week.(8) A survey of 49 state prison systems plus the federal system found that they expected an average growth of 21.4 percent from January 1, 1993, to January 1, 1995.(9)

In state after state, prison capacity is at record highs, but the prison population is even higher. The average American prison system now operates at 15.4 percent over capacity.(10) For example:

Maryland's prisons, built to hold fewer than 12,180 inmates, now hold 19,799.(11)

Florida prisons held 20,000 inmates in fiscal year 1980-81 and now hold over 48,000.(12)

The Texas prison system, which had 25,000 beds in 1979, had 109,000 by 1993.(13)

Forty states, two territories, and the District of Columbia are currently under court orders as a result of prison overcrowding.(14)

As of January 1, 1993, the federal prison system was 38 percent over capacity.(15)

Prison violence has also increased. Inmate assaults on guards have risen 10-fold in the last five years.(16) Over-crowding also contributes to inmate assaults on other inmates, as two persons with histories of violent assault are placed in a room designed for one.

Prison critics used to note that the United States incarcerated a larger percentage of its population than any nation except the USSR and South Africa. That statistic is no longer true. The number of state and federal prisoners per 100,000 population tripled in the last two decades, and the United States now leads the world in the percentage of its population it keeps behind bars, with an incarceration rate of 343 adults per 100,000 population (Figure 1).(17) In contrast, the Australian imprisonment rate is 90.5.(18)

Paralleling the explosion in the population of prisons has been a surge in the jail population (Figure 2). (Prisons are state or federal facilities that hold convicted criminals. Jails are city and county facilities that hold persons sentenced to shorter terms, typically under a year. Persons arrested and not released on bail pending trial are also held in jail. Thus, at any given time, about half of a jail's population will be people who have not been convicted of a crime.)(19) The number of jail inmates per capita in the

Figure 1
Federal and State Prisoners per 100,000 Poplation
{Bar Graph Omitted}

Source: U.S. Department of Justice, Bureau of Justice Statistics, Sourcebook of Criminal Justice Statistics, 1992, NCJ-143496 (Washington: Government Printing Office, 1993), p. 608; and "Prison Population Jumps 5%"

Figure 2
Jail Inmates per 100,000 Population
{Bar Graph Omitted}
Source: Beck, et al, Jail Inmates 1992, p.2, Table 2.

United States has more than doubled since 1978 and now exceeds the total incarceration rate (jail plus prison) of most other democracies.(20)

Total operating costs of state and federal prisons are approximately $13 billion a year. Adding in the costs of prison construction, and the costs of city and county jails, the national incarceration budget is about $24.9 billion.(21)

In state after state, prisons are eating a greater and greater share of budget revenues. In fiscal year 1994 state spending on corrections will increase 10.8 percent, a rate of increase higher than even the runaway Medicaid budget, which will increase 7.5 percent.(22)

Teetering near bankruptcy, California spends $2.1 billion a year operating its prison system and has $4 bilion in prison construction scheduled.(23) Pennsylvania saw its corrections budget soar from $126 million to $453 million in 10 years.(24) Even in frugal New Hampshire spending on prisons has risen three times faster than other state spending.(25)

In the federal system spending per prisoner per year is $20,072. In New York City the figure is $58,000.(26) And there are enormous nonmonetary costs as well. Half of all imprisoned mothers never see their minor children. One major reason for the low visitation rate is that women are imprisoned at less than 1/15th the rate of men. As a result, there are many fewer women's prisons, and women prisoners are more likely to be incarcerated far from their families. Thirty-nine percent of the mothers in prison are there for violations of drug laws.(27)

The U.S. prison population is larger than the population of some of the smaller American states. The annual total prison budget is more than twice the size of the budget of a medium-sized state such as Missouri.(28) Thus, American taxpayers are in essence paying for the housing and confinement of enough people to populate one small state at a cost appropriate for the entire operation of two medium-sized states.

Total spending on prisons is about as high as spending on the much-maligned Aid to Families with Dependent Children program. Taxpayers have justifiably begun to ask whether the $22.5 billion spent on AFDC helps cure poverty or, in fact, causes poverty through its perverse incentives. It is time to begin asking whether the $25 billion spent on prisons helps promote public safety or actually makes America more dangerous through its own set of perverse incentives.

Expensive as prisons can be, the incarceration of violent criminals is a tremendous bargain. Violent criminals at large can cause hundreds of thousands of dollars in damage each year.(29) When harder to measure costs, such as the pain and suffering of victims, are considered along with the more quantifiable costs (medical care of victims, funerals, and property destruction), it becomes clear that keeping a violent criminal in prison is an extremely efficient use of tax dollars, whether the cost of incarceration is $21,000 or $58,000.

But today state and federal prison systems are less and less likely to house repeat violent offenders. And it is far from clear that incarcerating record numbers of "criminals" (such as prostitutes and drug suppliers) whose only offense is to facilitate a voluntary transaction between adults is a worthwhile expenditure of limited tax dollars.

More important than the issue of fiscal efficiency is the fundamental issue of public safety. As the state and federal governments have taken record numbers of prisoners in the war on drugs, violent criminals have found to their pleasant surprise that there is less and less room for them in prison.

Impacts of the Drug War

The number of adults imprisoned for drug offenses more than tripled from 1986 to 1991. Violent crime rose 41 percent in the same period.(30) Between 1988 and 1993 the average drug sentence more than tripled, from two years to seven years.(31) Forty-four percent of the increase in the state prison population from 1986 to 1991 was attributable to drug crimes.(32) In virtually every state there has been a massive emphasis on imprisoning drug offenders. In Washington State the number of drug prisoners has risen 966 percent since 1980, and those prisoners now make up half of the state's nonviolent prisoners.(33) In New York State 45 percent of all new prison commitments are for drug convictions.(34) Illinois prisons now hold five times as many drug prisoners as they did five years ago.(35) The director of Florida's Department of Corrections described the drug war as "the primary engine fueling the enormous growth experienced by Florida's correctional system."(36) Richard Lanham, the commissioner of Maryland's Department of Corrections, estimates that "at least 40 percent of those coming into the Maryland prison system are there because of minor drug activity."(37)

In Texas the number of drug offenders in prison rose 350 percent from 1989 to 1992.(38) The states with the largest prison populations tend to be the states with the highest percentages of drug prisoners: California (33 percent), New York (34 percent), Florida (34 percent), Ohio (25 percent), Illinois (28 percent), and Georgia (27 percent).(39) Nevada, the state with the highest percentage of drug prisoners (36 percent), also had the second largest increase in total prison population: 804 percent from 1970 to 1990.(40)

Figure 3
Substance Prisoners as Percentage of Federal Prisoners
{Bar Graph Omitted}
Source: Figures for 1984 and before are from Margaret Werner Cahalan, "Historical Corrections Statistics in the United States", 1850-1984, NCJ-102529 (Washington: U.S. Dept. of Justice, 1986), pp. 152-55. Figures for 1990 are from U.S. Dept. of Justice, Bureau of Justice Statistics, "Drugs, Crime and the Justice System" (Washington: U.S. Dept. of Justice, 1992), p. 195

Figure 4
State Prison Admissions by Type of Crime
{Pie Graph Omitted}

About 60 percent of federal prisoners are drug offenders (Figure 3), and that figure is expected to hit 70 percent by 1995, according to the Bureau of Prisons. In 1981 only 22 percent of federal prisoners were drug prisoners.(41) The current and projected percentages of federal prisoners incarcerated for drugs are comparable to the combined figures for drug and alcohol offenses during Prohibition. (Note that many alcohol prisoners remained incarcerated for years after the repeal of Prohibition.) And it is quite clear that drug prohibition has no more saved America from drugs than alcohol prohibition saved America from alcohol. The increasing tendency to imprison drug offenders for lengthy periods means that America's state and federal prisons deal more with drug offenders than violent criminals (Figure 4). In 1990 the number of persons sent to prison for drug crimes (103,800) exceeded the number sent for violent crimes (87,200) or for property crimes (102,400).(42) About a third of all new commitments to state prisons were for drug crimes.(43) As recently as 1960, only 1 in 25 state prisoners was a drug prisoner.(44) If current trends continue, by the year 2000 half of all prison inmates will be drug war prisoners.(45)

Despite those statistics, some defenders of the present criminal justice system argue that there is no significant problem of overincarcerating nonviolent criminals. They argue that "only 6 percent of prisoners are nonviolent first offenders." And for good measure, they point out that the 6 percent include first-time burglars, a group whose imprisonment elicits little objection from most of the public. The statistic is, however, not as persuasive on closer examination as it might first appear.

The figure is taken from a massive Bureau of Justice Statistics study, Survey of State Prison Inmates, 1991 (or from similar, earlier surveys).(46) Accordingly, the figure does not reflect the federal prison system; despite the drug war, no state prison system devotes more than 34 percent of its space to drug criminals, whereas (in 1990) the federal government devoted 54 percent of its space to such offend ers, and the figure is expected to hit 70 percent within a few years.

Of the 6 percent nonviolent first-timers in the state systems, 61 percent are drug offenders, and only 12 percent are burglars. And the 6 percent figure is not the upper boundary of persons who, arguably, should not be in prison. The persons categorized as recidivists include not only persons who have served time before for felony convictions but also persons who have been convicted only of misdemeanors or who have been sentenced to parole or probation for any offense, including simple drug possession. Nonviolent offenders with at least one prior conviction accounted for 32 percent of all state inmates in 1991. Of that nonviolent 32 percent, 38 percent were drug offenders.(47) A good case could be made that imprisonment would be appropriate for some or many of those drug offenders with one or more previous convictions. But it is far from clear that every person who has been convicted twice of anything (including drug possession) should be in state prison. Because even some persons who have two convictions (such as some of the people who have been convicted of two nonviolent misdemeanors) should not be in prison, the use of the 6 percent figure as a conclusive defense of the status quo is not entirely persuasive.

It bears emphasizing that most drug criminals are not violent criminals. Only 21 percent of drug prisoners sentenced in state systems in 1991 had even a single incident of criminal violence in their background.(48) Seventy percent of drug prisoners in the federal system have no record of violence, while 10 percent have a record of minor violence.(49) And half of all prisoners entering the federal system for drug crime are first-time offenders.(50)

Two periods in American history have seen explosive growth in the federal prison population and federal funding of prisons. One period is the present. The other such period was the era of alcohol prohibition, during which the federal prison population more than quadrupled; by 1930-31, 68 percent of incoming prisoners were alcohol law violators (and another 5 percent were drug law violators).(51)

Mandatory Minimums and the Sentencing Guidelines

As legislative bodies in the 1980s grew increasingly determined to prove that they were "doing something" in the war on drugs, mandatory minimum sentences for drug offenses became common. The expectation was that mandatory minimums would reduce the availability of drugs by reducing the number of suppliers, but the expectation has not come true. What has come true is a living nightmare of barbaric punishment for small-time offenders, to the detriment of public safety.

Sentencing Guidelines

Although mandatory minimums are sometimes confused with the federal sentencing guidelines, it is important to understand the distinction. As of 1983 federal judges enjoyed broad sentencing discretion. That discretion allowed judges to tailor the sentence to the facts of the individual case, but it also resulted in large disparities in sentencing.

In 1984 Congress enacted the Sentencing Reform Act, perhaps the most significant change in sentencing policy in American history. To begin with, the act abolished parole in the federal prison system. The only possibility for early release is for a prisoner to earn 54 days off for good behavior during each year served. Thus, every federal prisoner will serve at least 85 percent of his sentence. So a person sentenced to a 10-year term has no possibility of being released in less than 8.5 years. Although there was no change in the nominal length of sentences, the abolition of parole resulted in a major change in the practical length of sentences.

Besides abolishing parole, the Sentencing Reform Act delegated to the U.S. Sentencing Commission broad discretion to create a body of sentencing guidelines that the federal courts would be required to follow. The guidelines have been fully operative in the federal courts for several years.

The guidelines begin by assigning a base-level sentence to each type of crime. That sentencing level is then enhanced by aggravating factors (such as perpetrating the crime in an unusually brutal manner) and reduced by mitigating factors (such as acceptance of responsibility for the crime). The computation then provides the sentencing judge with a particular range within which he may sentence the defendant. For example, the guidelines might specify that the defendant must be sentenced to a term somewhere between 10 and 12 years; the court would then impose a determinate sentence within the acceptable range (e.g., 11 years). Sentenced to a definite term of 11 years, the prisoner would serve the full 11-year term (with no possibility of parole), less the 15 percent reduction available for good behavior.

The sentencing guidelines have been criticized for their rigidity and severity. The defendant's personal characteristics are "not ordinarily relevant." That a person has been employed for the last 30 years, raised three children, and contributed thousands of volunteer hours to charity does not entitle her to any sentence reduction compared a person who has never held an honest job or done anything for the community other than contribute DNA to a string of illegitimate children.

To some critics, the guidelines reflect a preoccupation with regulatory offenses and a disregard for the seriousness of violent crime. As Justice Scalia explained in a dissenting opinion that would have held the guidelines unconstitutional:

Under the guidelines, the judge could give the same sentence for abusive sexual contact that puts the child in fear as for unlawfully entering or remaining in the United States. Similarly, the guidelines permit equivalent sentences for the following pairs of offenses: . . . arson with a destructive device and failure to surrender a canceled naturalization certificate; operation of a common carrier under the influence of drugs that causes injury and alteration of one motor vehicle identification number . . . aggravated assault and smuggling $11,000 worth of fish.(52)

Likewise, "transmission of wagering information" draws a longer sentence than negligent manslaughter.

And the 700 pages of federal sentencing guidelines are, in the words of Federal District Judge Jose Cabranes, "nearly unintelligible to victims, defendants, and observers, not to mention the lawyers and judges involved. Disparity is rife, though much of it is now hidden within the guidelines themselves and in the silent exercise of discretion by police officers and federal agents, prosecutors, probation officers and (even now) federal judges."(53)

Mandatory Minimums

While many federal prisoners are sentenced according to the guidelines created by the U.S. Sentencing Commission, there are also statutory mandatory minimums for certain crimes. When statutory mandatory minimums, enacted by Congress, conflict with the sentencing guidelines, the mandatory minimums prevail. Thus, if the sentencing guide lines specify a sentence in the 5- to 6-year range, but the statutory mandatory minimum requires a 10-year minimum sentence, the 10-year sentence (or longer) is imposed.

Today there are over 100 federal laws specifying mandatory minimum sentences.(54) Although a few mandatory minimums had existed since 1790 (for piracy and murder), such sentences did not become widespread in the federal system until 1956, when they were enacted as part of federal efforts to control narcotics. The federal mandatory minimums were repealed in 1970, as Republican and Democratic members of Congress recognized the flaws in the mandatory minimum approach. As then-congressman George Bush explained in his support of the repeal bill:

Contrary to what one might imagine, however, this bill will result in better justice and more appropriate sentences. . . . Federal judges are almost unanimously opposed to mandatory minimums, because they remove a great deal of the court's discretion. . . . As a result [of repealing mandatory minimums], we will undoubtedly have more equitable action by the courts, with actually more convictions where they are called for, and fewer disproportionate sentences.(55)

In 1984 Congress began a process that continues today, adding vast new numbers of mandatory minimums, particularly for crimes involving drug or firearms offenses. Since 1984 nearly 60,000 persons have been sentenced under the mandatory minimums.(56) Despite the large number of federal laws with mandatory minimums, 94 percent of federal mandatory minimum cases involve four laws covering drugs or weapons.(57) Beginning with New York's "Rockefeller Law" in 1973, almost every state has enacted its own mandatory minimums.

The mandatory minimums are extremely tough on drug offenses and make drug weight almost the sole factor in setting a drug crime sentence. For example, merely possessing more than five grams of crack cocaine requires five years in federal prison. (An individual packet of sugar in a restaurant weighs about one gram.) The only factors other than drug weight that may be considered in the sentencing are (1) if the defendant has prior convictions (in which case the mandatory minimum is raised)(58) and (2) if the U.S. attorney makes a motion stating that the defendant has provided "substantial assistance" in obtaining the conviction of another drug criminal (in which case the court has discretion to sentence the defendant to less than the mandatory minimum).

Although the federal mandatory minimums are closely tied to drug quantities, they are not limited to persons who possess the particular drug quantities for sale. In 1988 Congress added conspiracy to commit a drug offense to the list of crimes with mandatory minimums. Thus, if a woman tells an undercover federal agent where to buy some LSD, and the agent then buys some LSD from a person who possessed five grams of LSD, the woman, as a "conspirator," is subject to the same mandatory minimum as is the person who actually possessed the LSD. As a result of the sentencing guidelines and the mandatory minimums, one-quarter of all federal inmates are serving sentences of 15 years or more; half are serving sentences of over 7 years.(59)

Even more mandatory minimums are currently being proposed. One proposed bill would impose a five-year mandatory minimum for possession of a gun in violation of the Gun Control Act of 1968. The new proposal would have no effect on people with criminal records who illegally buy guns. Those persons are already subject to mandatory minimums pursuant to laws enacted in 1986. The Gun Control Act makes it illegal for the "user" of any controlled substance, including marijuana, to purchase a gun. If someone smoked marijuana occasionally and then bought .22 rifle for use at a target range, that person would be required to serve a five-year minimum term.

Another proposal would require a 5- or 10-year mandatory minimum for delivery of a controlled substance to a minor. If a college student gave her high school brother some marijuana cigarettes, she would be required to serve 10 years in federal prison, with no parole.

In practice, the number of conflicts between the mandatory minimums and the sentencing guidelines is relatively small. That is because the U.S. Sentencing Commission (which meets several times a year to fine-tune the sentencing guidelines) takes the statutory mandatory minimums into account, by using the mandatory minimums as "anchors" to set the sentencing range.

A General Accounting Office study of 573 drug sentences in federal courts found that in only 5 percent of cases were the mandatory minimum sentences higher than the most severe end of the applicable sentencing guidelines. (For example, the guidelines might specify a 2- to 4-year sentencing range, but the mandatory minimum would require 10 years.)(60) Thus, in 95 percent of federal cases, the mandatory minimum does not have a significant impact. Either the mandatory minimum requires a sentence that would have been approximately the same as what the sentencing guidelines would require anyway (the 70 percent of cases in which the mandatory minimum falls within the guideline range), or the mandatory minimum is below what the guidelines would specify (the 20 percent of cases in which the lower end of the guideline exceeds the mandatory minimum, because of aggravating factors). So the only cases where the mandatory minimums have a major impact on the sentence are the 5 percent of drug cases for which the highest range in the guidelines is below the floor of the mandatory minimum. Those 5 percent are the cases for which the sentencing guidelines suggest that simply looking at the total weight of the drugs involved is inappropriate--such as when a first-time offender carries an ounce of heroin as a low-level operative in a large conspiracy to transport 100 pounds of heroin.

Stated another way, the only place the federal mandatory minimums really make a difference is in the 5 percent of cases in which they inflict disproportionate and unfair sentences. In the other 95 percent of cases, the sentencing guidelines already require sentences approximately equal to or greater than the mandatory minimums. (It should be noted, though, that the sentencing guidelines use the mandatory minimums as a starting point; accordingly, if the mandatory minimums were repealed, the U.S. Sentencing Commission could choose to lower the overall severity of punishment for some drug offenses.)

At both the federal and state levels, mandatory mini- mums, when they matter, require grotesquely disproportionate sentences. Consider the following.

Brenda Valencia, a 19-year-old with no prior convictions, or even any evidence of involvement in drug sales, drove her aunt from Miami to a drug dealer's home in Palm Beach County. For that she was sentenced to 12.5 years in prison, which the sentencing judge, Federal District Judge Jose Gonzalez, Jr., termed "an outrage."(61)

Michael Irish was a 44-year-old carpenter in Portland, Oregon. Irish had no criminal record, but he did have a wife, two children, and no money, because his savings had been wiped out to pay for his wife's cancer treatment. One afternoon he was asked to help unload a cargo boat of hashish onto a truck; he did so and ended up sentenced to 12 years in prison without parole.(62)

Gary Fannon, a 19-year-old with no prior convictions, worked to set up, but then became too fearful to complete, a sale of a large quantity of cocaine to an undercover Michigan police officer. Fannon was convicted of conspiracy and sentenced to mandatory life in prison without parole.(63)

Christian Martensen, a young fan of the Grateful Dead, followed the band on tour. When his van broke down, he needed money to fix it, and another fan offered Martensen $400 if Martensen would find him someone who would sell some LSD. Martensen accepted the fan's offer; the fan turned out to be an undercover federal agent, and Martensen is now serving a 10-year mandatory minimum. He has no prior record.(64)

The unfair impact of mandatory minimums is not limited to drug cases. In Massachusetts the Bartley-Fox law requires one year in prison for carrying a gun without a permit. An early test case under Bartley-Fox was the successful prosecution of a young man who had inadvertently allowed his gun license to expire. To raise money to buy his high school class ring, he was driving to a pawn shop to sell his gun. Stopping the man for a traffic violation, a policeman noticed the gun. The teenager spent the mandatory year in prison with no parole.(65)

Another Massachusetts case involved a man who had started carrying a gun after a coworker began threatening to murder him. Wrote the court:

The threat of physical harm was founded on an earlier assault by Michel with a knife and became a real and direct matter once again when Michel attacked the defendant with a knife at the MBTA station. . . . [D]efendant is a hardworking, family-man, without a criminal record, who was respected by his fellow employees (Michel excepted). Michel, on the other hand, appears to have lacked the same redeeming qualities. He was a convicted felon with serious charges pending against him. . . . It is possible that defendant is alive today only because he carried the gun that day for protection. Before the days of a one-year mandatory sentence, the special circumstances involving the accused could be reflected reasonably in the sentencing or dispositional aspects of the proceeding. That option is no longer available in the judicial branch of government in a case of this sort.(66)

The Civil Liberties Union of Massachusetts--normally no friend of gun rights--lobbied against Bartley-Fox precisely because the union recognized the likelihood that people would be inappropriately sent to prison.(67)

President Clinton defends the war on drugs by pointing to his brother Roger. The president credits the arrest of Roger Clinton for selling cocaine with saving Roger's life. Upon conviction, Roger Clinton served 15 months in prison and has since stayed out of trouble with the law. Roger Clinton was arrested for selling cocaine in 1984. If he had been arrested a few years later, when the federal mandatory minimums and sentencing guidelines had gone into effect, Roger Clinton would have served a five-year mandatory minimum. If he had sold crack cocaine instead of powder cocaine, the sentence would have been 10 years. Would public safety, Roger Clinton's life, or any other value have been enhanced by requiring Roger Clinton to spend all those extra years in prison?

Being committed to the war on drugs is not the same as being committed to filling the prisons with small-time drug criminals. Ray Enright, a former assistant to the director of the Drug Enforcement Agency and currently the chair of the Colorado Parole Board, observed: "It's been my experience that we're seeing too many people, too many low-level traffickers and abusers whose sentences, in my opinion, are not commensurate with their crime. I think we've gone in for some overkill."(68)

Mandatory minimums are inappropriate even when tied to a particular aggravating factor. The conduct described by a mandatory minimum trigger often is that of an especially serious offender. But not always. Consider the Uniform Controlled Substances Act, a "model" state law being pushed in state legislatures nationwide. The act includes numerous factors requiring mandatory prison sentences. One factor is sale of drugs to a minor. Does a school janitor who pushes drugs to elementary school children deserve a severe, lengthy sentence? Does an 18-year-old high school student who asks his 17-year-old girlfriend to temporarily store a gram of cocaine in her jacket pocket until his older brother can buy it deserve an equally severe, lengthy sentence?(69) Under the UCSA, the same mandatory sentence applies to both persons since they involved a minor in their drug crime. Similarly, does a Stanford college student who brings four grams of psilocybin mushrooms home to Idaho for personal use on Christmas vacation deserve the same type of very severe sentence meted out to someone who smuggles three pounds of heroin into the state?(70) "Yes," is the answer supplied by the UCSA, since both imported drugs into the state.

The aggravating factors described in the UCSA and other sentencing legislation would be better treated as just that: aggravating factors, rather than mandatory sentence triggers. The presence of one of the aggravating factors could allow but not require the court to impose a sentence in excess of the presumptive sentence.

Cliffs and Mules

Mandatory minimums are frequently enacted with the objective of cracking down on serious drug dealers. Ironically, the most significant impact of mandatory minimums is often felt by the smallest players in the drug trade. As the U.S. Sentencing Commission's 1991 report on mandatory minimums observed, normal sentencing policies increase the penalty as criminal culpability increases. But "mandatory minimums result in 'cliffs,'" which tend to "compromise proportionality, a fundamental premise of just punishment."(71)

The existence of the cliffs has led to a new form of governmental abuse, known as sentencing entrapment. Under a judge-directed sentencing system, it would make little difference whether a defendant had sold 45 grams of cocaine or 51 grams. But statutory mandatory minimums and the sentencing guidelines impose rigid rules based on precise quantities of drugs. Thus, in a Minnesota case, the defendant, according to the court's finding, was entrapped by undercover agents into selling them 50.4 grams of crack cocaine, because the agents knew that once the defendant had sold an aggregate of 50 grams, he would be subject to a 10- year mandatory minimum. The court refused to sentence the defendant on the basis of the 50-gram statute.(72) In a previous case, the same court had described sentencing entrapment as "'outrageous official conduct [that] overcomes the will of an individual predisposed only to selling in small quantities' for the purpose of increasing the amount of drugs . . . and the resulting sentence of the entrapped defendant."(73) Unfortunately, not every court has been willing to recognize the existence of sentencing entrapment.

Mandatory minimums have the stated goal of ensuring that similar offenders receive similar sentences, but in practice, perverse, unequal consequences often result. Mandatory minimums tend to fall hardest on people who are not habitual criminals (who are already covered by repeat offender laws). For example, in 1989 Delaware enacted a three-year mandatory minimum, with no parole, for possession of 5 to 15 grams of an illegal substance. Seventy-two percent of the persons convicted under the new law had never before been imprisoned for any crime.(74)

In New York most of the people sentenced to mandatory minimums are drug addicts carrying relatively small quantities. According to prison worker Sister Marion DeFeis and correctional specialist Tracy L. Huling, the Rockefeller laws (named for the governor who originated them) sweep in many foreign women who are duped into unknowingly carrying drugs. One woman from Ghana, a nurse's aide, told a story of a man who proposed marriage. On an international flight, she carried a pair of sneakers for him, sneakers that contained four ounces of cocaine.(75)

Mules are often foreigners, recruited for a single trip. When the mandatory minimums are applied to those low- level couriers, American taxpayers end up spending large amounts of money to imprison foreigners for years on end. A Sentencing Commission study of 1,100 cases found that one- fourth of all persons sentenced under the mandatory minimums were not American citizens.(76) As more and more persons reject the idea of paying welfare to illegal aliens, it may be asked whether it is appropriate to spend so many tax dollars paying room and board for aliens within America's ultimate welfare state, the prison system.

Even Chief Justice Rehnquist, not known for his lenient approach to crime, observed:

These mandatory minimum sentences are perhaps a good example of the law of unintended consequences. There is a respectable body of opinion which believes that these mandatory minimums impose unduly harsh punishment for first-time offenders-- particularly for "mules" who played only a minor role in a drug distribution scheme. Be that as it may, the mandatory minimums have also led to an inordinate increase in the federal prison population and will require huge expenditures to build new prison space. . . .

Mandatory minimums . . . are frequently the result of floor amendments to demonstrate emphatically that legislators want to "get tough on crime." Just as frequently they do not involve any careful consideration of the effect they might have on the sentencing guidelines as a whole. Indeed, it seems to me that one of the best arguments against any more mandatory minimums, and perhaps against some of those that we already have, is that they frustrate the careful calibration of sentences, from one end of the spectrum to the other, which the sentencing guidelines were intended to accomplish.(77)

Justice Anthony Kennedy, also known for his stern views on crime, has stated, "I think I am in agreement with most judges in the Federal system that mandatory minimums are an imprudent, unwise and often unjust mechanism for sentencing."(78)

Chief Justice Rehnquist's observation that proposals for mandatory minimums "do not involve any careful consideration" of their practical effect is understated. In 1990 U.S. senators were showing their dedication to fighting crime by voting for a federal sentencing bill that required that certain defendants not be eligible for parole--even though Congress abolished parole back in 1984.

Under the federal mandatory minimums, a manual laborer who unloads a boatload of marijuana faces the same 30-year mandatory minimum as the man who masterminded the smuggling. In contrast, the federal sentencing guidelines consider not only the weight of the drug involved but also the role of the defendant in the drug conspiracy. Under the sentencing guidelines, the head of a drug-trafficking conspiracy will receive a sentence 50 percent greater than an underling receives and twice as great as the sentence meted out to a low-level defendant with a minimal role, such as a courier who carries only a small quantity of drugs.

It is the first offender and the person lower down in the chain of criminality who is often likely to receive the mandatory minimum because, unlike criminals higher up the chain of command, the small-time criminal often cannot turn in anyone else and thereby render "substantial assistance" that merits a downward departure from the guidelines.

The Judicial Reaction

As long as there have been mandatory minimums, there have been judges who have found the resulting sentences repugnant to principles of decency. In New Mexico in 1981 one judge resigned after being forced to send to prison a man with a clean record who had brandished a gun during a traffic dispute.(79)

By May 1993, 50 senior federal judges, such as Jack B. Weinstein and Whitman Knapp of New York, had exercised their prerogative to refuse to hear drug cases.(80) (Senior judges are allowed much more control over their dockets than are ordinary district judges.) Many conservative, Reagan-appointed federal judges have denounced the 5- and 20-year mandatory minimums as draconian miscarriages of justice. Federal District Judge Stanley Harris remarked, "I've always been considered a fairly harsh sentencer, but it's killing me that I'm sending so many low-level offenders away for all this time."(81)

A Gallup survey of 350 state and 49 federal judges who belong to the American Bar Association found 8 percent in favor of and 90 percent opposed to the federal mandatory minimums for drug offenses.(82) The sentencing guidelines did somewhat better; 27 percent of the judges thought they had worked well, while 59 percent thought they had worked poorly or not at all. Fifty-nine percent of the judges thought the federal sentencing guidelines should be scrapped, while 30 percent did not.(83)

Although many judges are dissatisfied with the sentencing guidelines, official judicial bodies have not formally opposed them. In contrast, the judges of every feder- al circuit in the United States have enacted resolutions calling for repeal of the federal mandatory minimums. The Judicial Conference of the United States, which represents federal judges, has endorsed repeal of mandatory minimums, as have the American Bar Association and the Federal Court Study Committee (created by Congress).(84)

Although lobbyists who represent prosecutors are usually strongly in favor of mandatory minimums, not all frontline prosecutors agree. A U.S. Sentencing Commission survey of assistant U.S. attorneys (federal prosecutors) found that only 53 percent thought that the mandatory minimums had any positive effects.(85)

Some courts are finding constitutional flaws in the mandatory minimums or in portions of the sentencing guidelines. Judge Harold Greene, of the Federal District Court of the District of Columbia, ruled the federal sentencing guidelines unconstitutional as violative of the due process clause and the cruel and unusual punishment clause of the Constitution. The decision arose in a case involving a sentencing guideline for a 30-year term for a person con victed of possessing with intent to sell eight grams of heroin, after two previous convictions for small-scale heroin sales.(86)

In Minnesota the state supreme court found unconstitutional a statute that permitted only the prosecutor, and not the judge, to allow the defendant to be sentenced to a term below the mandatory minimum. The court ruled that the sentencing scheme violated the separation of powers.(87) Courts in other jurisdictions have also found certain aspects of mandatory minimum laws to violate the separation of powers.(88)

The Defense of Mandatory Minimums

Some prosecutors assure critics of mandatory minimums that prosecutorial discretion will prevent the mandatory minimums from being imposed unfairly. Yet the prosecutors' ability to exercise discretion simply reflects the fact that mandatory minimums and sentencing guidelines transfer sentencing discretion from the place where it belongs--the court--to the prosecutor. That is why some judges have found the mandatory sentencing scheme to violate the separation of powers.

One argument in favor of mandatory minimums is that many (although far from all) persons sentenced under mandatory minimums are repeat offenders. But that fact hardly legitimizes in all cases the application of such cruel and severe sentences. In Houston, for example, a 37-year-old stevedore was prosecuted and convicted of possessing 1/1,000th of a gram of crack (the residue on a crack pipe). Because the defendant had two previous drug convictions, he was required to be sentenced to a 25-year minimum term as a "habitual offender."(89)

In any case, the combination of mandatory minimums and the sentencing guidelines results in severe sentences for first-time drug offenders. In 1990, 88.9 percent of all drug offenders in federal court who had no prior conviction for any offense were sentenced to prison. The percentage of first-time drug offenders sent to prison was higher than the percentage of first-time violent criminals sent to prison (79.4 percent) and over twice as high as the percentage of first-time property offenders sent to prison (37.7 percent for fraud, 22.9 percent for other property crimes).(90) Drug offenders with no prior record were sentenced to an average prison term of 68.4 months, compared to 56.2 months for violent criminals with no prior record.(91)

Rep. Charles Schumer (D-N.Y.) argues that the number of persons unfairly sentenced under mandatory minimums is small. According to a report prepared at Schumer's request by the U.S. Sentencing Commission: "Of the 38,000 persons sentenced under the federal sentencing guidelines in 1992, only 3,189 non-violent, first-time offenders with no aggravating role in the offense were sentenced to a mandatory minimum. That is less than 10 percent of the persons sentenced under the guidelines." Schumer suggests that the numbers prove that repeal of any mandatory minimum is unwarranted.(92) Close analysis suggests that Schumer's numbers are not as convincing as he thinks.

First of all, anyone sentenced federally in 1992 would have been sentenced under the federal sentencing guidelines. The number of persons sentenced under federal mandatory minimums would have been much smaller, because the mandatory minimums would have come into play only when there was a statutory mandatory minimum that exceeded the sentencing guidelines for a particular offense.

Accordingly, Schumer's figure of less than 10 percent is based on a comparison of apples and oranges. The more appropriate question would be, what percentage of persons sentenced to mandatory minimums (not what percentage of all persons sentenced federally) were nonviolent first offenders?

In addition, Schumer defines a "nonviolent" offender as one "who was not subject to a penalty enhancement for use of a firearm." That definition, however, excludes many nonviolent offenders. The Supreme Court has ruled that penalty enhancements for "use" of a firearm can be applied when the firearm was "used" by being traded for drugs. Likewise, the severe federal laws that prohibit any person with a felony conviction from ever possessing any firearm can sweep within their net a nonviolent ex-offender who was simply unaware that his possession of .22 squirrel rifle was unlawful.

In response to Schumer's claim that there are only a small number of "horror stories" associated with the federal mandatory minimums, Judge Vincent Broderick, chair of the Criminal Law Committee of the Judicial Conference of the United States, said: "I respectfully submit that the mandatory minimum system in place is itself the 'horror' story. . . . There is no single issue affecting the work of the federal courts with respect to which there is such unanimity. . . . Most federal judges . . . believe, and this is predicated upon their experience, that mandatory minimums are the major obstacle to the development of a fair, rational, honest, and proportional federal criminal justice sentencing system."(93)

As Eric Sterling, director of the Criminal Justice Policy Foundation, observes, the mandatory minimums are inappropriate not just in the case of a first-time drug courier who happened to carry a large quantity of drugs. Even for offenders who fall more securely within the intended scope of the mandatory minimum, the sentences are often sadistic. Why should any quantity of marijuana result in a 30-year prison term, or a death sentence, as some versions of the 1994 congressional crime bill require?

A recent report from the Department of Justice, which contains the most detailed analysis ever conducted of the frequency of the imposition of severe sentences on minor offenders, revealed the following: 21.2 percent of the total federal prison population, and 36.1 percent of all federal drug prisoners, are "low-level" drug offenders with no record of violence, whose offense did not involve sophisticated criminal activity, and who had no record of serious prior offenses. Those 16,316 federal prisoners are serving an average of 5.75 years apiece, about 150 percent more than what they would have served under laws in effect before 1986. Of the low-level offenders, 42.3 percent were couriers or had other peripheral roles in the drug offense. The low-level offenders were much less likely to recidivate than were higher level offenders, and the length of incarceration had no influence on likelihood of recidivism.(94)

It is no defense of mandatory minimums to point out that any particular percentage of persons sentenced under them are repeat offenders or violent criminals or major drug dealers. The federal judiciary is dominated by Reagan-Bush appointees, not known for their softhearted attitude toward crime. Courts would gladly impose stiff sentences on genuinely serious offenders anyway, without need of a congressionally imposed mandatory minimum. The only cases in which mandatory minimums make a major difference are, by definition, cases in which the court would choose to impose a lesser sentence if not for the mandatory minimums.

William E. Wilkins, chairman of the U.S. Sentencing Commission, offers a proposal somewhat similar to Schumer's amelioration option: retain the mandatory minimums but make them starting points rather than inflexible barriers.(95) Thus, the severity of the mandatory minimums could be retained in most cases, but courts would be allowed to impose lower sentences in the minority of federaldrug cases in which consideration of individual factors would result in a sentence below the mandatory minimum.

Significant Reform Proposals

Schumer's "amelioration" and Wilkins' "starting point" proposals are worthwhile reforms, but they do not go nearly far enough. A more satisfactory approach to mandatory minimums would be to require a sunset provision for all of them. Many states have laws that require that all regulatory agencies undergo sunset review; that is, an agency's existence automatically terminates after a set period of time (such as seven years) unless the legislature affirmatively reauthorizes its existence. The sunset review ensures that obsolete agencies do not continue to exist for decades simply because of inertia.(96) In some states, newly enacted substantive criminal laws are also subject to sunset review, as are administrative regulations of all types.

It makes sense to extend sunset review to mandatory minimums. Mandatory minimum prison sentences would expire on the same sunset schedule as other state programs (e.g., every seven years). Unless the legislature specifically reenacted the mandatory minimum, ordinary sentencing rules would go back into effect.

In Colorado and many other states, every proposed statute that would have a fiscal impact on the state requires a "fiscal note" setting forth the anticipated budgetary impact. The fiscal note helps to ensure that legisla- tors not only consider the content of proposed legislation but also make sure that it will be paid for; all bills with fiscal notes must be sent, after approval by the committee dealing with the substance of the bill, to the finance or appropriations committee. A constructive step toward fiscal responsibility in sentencing policy would be to require that all criminal legislation include a "prison impact assessment." The assessment would attempt to quantify the increased prison population, and hence the increased expense, that might result from new sentencing policies. The Gallup/American Bar Association survey of state and federal judges found 80 percent support for requiring impact statements.(97) If prison impact assessments had been required in the mid-1980s, and if the balanced-budget amendment had been enacted, Congress might well have been precluded from enacting the frivolous and poorly considered increases in mandatory minimum sentences that characterized the drug hysteria of that period.(98)

Capping sentences for small-scale drug offenses at 13 years (down from the current 30-year maximum) has been proposed by a panel of federal judges.(99) Thirteen years is a long time to spend in a cage and is more than sufficient punishment for almost any drug crime, let alone a small-quantity offense. Similarly, it would be appropriate to cut by at least 25 to 50 percent the mandatory sentences that have already been imposed on small-time drug offenders.

Significantly, the state of Florida, which helped lead the United States into the drug war, has changed its mind on mandatory minimums. Gov. Lawton Chiles, an advocate of mandatory minimums in the 1980s, in May 1993 signed a law repealing 23 mandatory sentences.(100) The repeal actually allowed the state to increase sentences for violent crimes. Among the changes was removal of persons repeatedly arrested for small-scale drug offenses from the "habitual offender" classification.(101) Tennessee took a similar step in 1992, halving the sentence for low-level drug dealers and using the space saved to increase sentences for sex offenders.

Although mandatory minimums are sometimes promoted by politicians who believe that, regardless of the fiscal and human cost, the minimums are politically popular, public opinion polling suggests otherwise. Polling conducted in Alabama in 1989 and Delaware in 1991 found strong support for imprisoning nonviolent offenders. But support for incarceration of nonviolent offenders fell substantially if respondents were told about the availability of alternatives to incarceration, such as strict probation, boot camps, or electronic home detention.(102) Polling in Pennsylvania in 1992 yielded similar results.(103)

The Devastating Impact on Public Safety

Violent Criminals Set Free

Multiplying the risk of apprehension by the average sentence served upon conviction, Texas A&M economist Morgan Reynolds finds that the average 1990 murderer could expect to serve 1.8 years in prison. The expected punishment for murder fell by 20 percent from 1988 to 1990. Similarly, the expected punishment for rape (60 days in prison) fell by 25 percent in just two years, and the expected punishment for robbery fell by 50 percent (to a mere 23 days). Expected punishment for motor vehicle theft is only 1.5 days in prison. From 1950 to 1990 the expected punishment for all serious crimes, taken as a whole, fell by 65 percent.(104) Simply put, crime pays.

Interestingly, Sen. Phil Gramm (R-Tex.) uses the Reynolds expected-punishment calculations as proof of the need for mandatory minimums. After setting forth the Reynolds data, Gramm writes: "Mandatory minimum sentences deal with this problem directly. When a potential criminal knows that if he is convicted he is certain to be sentenced, and his sentence is certain to be stiff, his cost-benefit analysis changes dramatically and his willingness to engage in criminal activity takes a nose dive."(105)

But 10 years after Congress, often led by Senator Gramm, imposed severe mandatory minimums for drug sales, it is apparent that criminal willingness to engage in illegal drug transactions has not taken "a nose dive." To the contrary, cocaine is now cheaper, purer, and more readily available than it has been in decades. And the mandatory minimums have the perverse effect of increasing the profitability of the drug trade.

The mandatory drug minimums have led to reduced punishment for violent crime. Consider Senator Gramm's home state of Texas. "The most extraordinarily violent criminal ever to set foot in Falls County, Texas," was how the county district attorney described murderer Kenneth McDuff. McDuff had murdered two teenage boys, then raped a girl, and then snapped her neck with a broomstick. Every law enforcement official who had encountered McDuff stated that McDuff would kill again if he ever had the opportunity. In 1989 the war on drugs gave McDuff the opportunity.(106)

Although Texas doubled its prison capacity in the 1980s, the state also quadrupled its incarceration of drug offenders.(107) To cope with prison-space limitations, the Texas Parole Board lowered its standards for eligibility for parole and set McDuff free. Three days later, the naked, strangled body of his first new victim was found. McDuff was apprehended a year later, charged with three murders, and put under investigation for six more.(108)

Figure 5
Federal Sentences, 1980-90
{Bar Graph Omitted}
Source: U.S. Department of Justice Statistics, Federal Criminal Case Processing, 1989-90, NCJ-136945 (Washington: U.S. Department of Justice, 1992)

The McDuff story was repeated, on a less sensational scale, throughout the last several years in Texas. According to Reynolds, the expected punishment for serious nondrug crime in Texas fell 43 percent in the 1980s. The average time served by violent offenders in Texas dropped from 28.4 months in 1985 to 24.2 months in 1991.(109) Not surprisingly, the Texas crime rate for those offenses rose 29 percent.(110) An Illinois study linked the huge increase in drug law-enforcement in the state to a sharp increase in violent crime; one reason was that greater numbers of violent criminals were released from prison early to make room for the surge of drug offenders.(111) In the federal prison system, between 1980 and 1990, the average sentence imposed for rape, robbery, and kidnapping fell, while the average drug sentence increased sharply (Figure 5).(112)

Sociologist Robert Figlio points out that most violent crimes are perpetrated by a fairly small number of sociopaths. Those criminals start committing violent crimes early in life and continue doing so long after age 30, the age by which most other criminals have settled down. Figlio suggests that failing to incarcerate those repeat, violent offenders for lengthy periods greatly endangers public safety, and the reason those offenders are not incarcerated is that so much prison space is consumed by drug offenders.(113)

Though inner-city delinquent teenagers may not have calculated the mathematical risks of arrest, they are well aware of how minimal punishments are for even the most serious armed, violent offenses. They cannot help but infer that society does not really take violent crime seriously. And are they not right in their perceptions of today's criminal justice system, where first-offense burglary has practically been decriminalized? If a society is so intent on sending first-time drug vendors to prison that first-time muggers often do not go to prison, should it be surprising that burglary and mugging increase?

What kind of society treats mere possession of a drug more severely than drunk driving, even though drunk driving is, by definition, always dangerous to other persons and is known to cause the deaths of thousands of sober people every year? In contrast, mere possession of drugs is generally not dangerous to anyone except the possessor.

There is one other way in which mandatory minimums may, occasionally, encourage violent crime. It is possible that the prospect of mandatory minimums may increase the willingness of some offenders to perpetrate murders in order to avoid detection. If the mandatory minimum sentence for a drug or gun possession crime is not much less than the sentence for homicide, some criminals will find perpetrating the murder a risk-aversive choice. As one prisoner put it:

If I got out today, I'd have a gun today. If a Government agent showed up at my doorstep, I'd kill her. I'm not going down again. This law is creating a monster--it's creating a lot of people who will kill agents because you've made them hardened criminals--what does he have to lose? A guy facing mandatory sentences will resort to violence.(114)

Cesare Beccaria, the father of modern criminology, argued that penalties should be graduated, based on the seriousness of the offense; otherwise, criminals would as soon commit heinous crimes as minor ones. Or, as Jeremy Bentham observed, penalties should be proportioned "to induce a man to choose always the lesser of two offenses." Would Bentham approve of the trend in the federal criminal justice system toward treating rape less seriously than drug sales?

In defense of mandatory minimums for drug offenders, some elected officials, such as Rep. Romano Mazzoli (D-Ky.) claim that drug sales or use is an "inherently violent" crime.(115) Similarly, Senator Gramm labels drug use "violent" and claims that "the violence drug traffickers inflict on society is massive."(116) To the contrary, the violence related to drugs is due mostly to the drug laws. Studies by Professor Paul Goldstein and others of 218 homicides in New York that had been classified as drug related found that 21 were caused by the pharmacological effect of alcohol and 5 by the effect of crack; the rest resulted from the turf wars, robberies, and other violence engendered by drug prohibition, just as alcohol prohibition caused violence in a previous era.(117)

Whatever benefits the war on drugs has brought America, there has been a terrible price exacted by the violent criminals set free to make room for drug war prisoners. "Drug dealers and users have consumed the single most valuable resource in the criminal justice system: jail cells," argues San Francisco sheriff Michael Hennessey. "We desperately need the limited space in our nation's jails and prisons to house violent offenders, not minor league dope addicts and dealers."(118)

hy Imprisonment Works for Violent Crime Control but Not for Drug Control

The fundamental flaw of the entire strategy of trying to control drug use through imprisonment is that it cannot work, as free-market analysis demonstrates. In contrast, imprisoning violent criminals does work, and the removal of violent criminals from prison cells to make room for drug criminals replaces a policy that does work with one that does not.

Although there has long been scholarly debate about the extent, if any, to which prisons deter crime or rehabilitate criminals, there is no denying that prisons incapacitate criminals. As long as a violent criminal is in prison, he will not endanger anyone except other prisoners and the prison guards. Imprison one armed robber, and there is one fewer armed robber on the street. Imprison half the armed robbers, and the armed robbery rate will decline about 50 percent. The same analysis applies to child abusers, burglars, and most other criminals who repeatedly attack innocent people. The more of them are imprisoned, the less crime will occur while they are imprisoned. As New York assemblyman Daniel Feldman observes, "Incarcerating a sex offender does not create a job vacancy to be filled by another sex offender."(119)

In contrast, the imprisonment of one drug dealer (or the destruction of one drug distribution network) does not diminish the availability of drugs for long. Nearly as soon as one supplier is removed, another supplier will move in to take his place. The law of supply and demand states that as long as there is a demand for a product, a market will make that product available at some price. Thus, removing one cocaine-addicted junkie who sells drugs on a street corner offers another junkie the opportunity to sell drugs on that corner. Likewise, removing one network of drug suppliers simply opens up a market for other suppliers. Allocating vast amounts of prison space to such easily replaced offenders is a dangerous waste of public resources.

The very illegality of drugs means that drug dealers can charge a "risk premium" as compensation for the risk of going to prison. The more severe the mandatory minimum punishment for drug dealers, the greater the risk premium that will be charged. Thus, the drug supply can only be reduced by imprisoning almost every drug addict (the group that forms the bulk of street-level dealers) and also imprisoning everyone who is willing to break the law for enormous financial rewards (the higher level operatives in the drug network). Having already been tripled, national prison capacity could be tripled again without achieving that goal.

Alternatives to Incarceration

As prison crowding increases, alternatives to incarceration are becoming more popular. Among the recent innovations in alternatives are electronic home detention, intensive supervised probation, and boot camps. Although all of those alternatives have merit in certain cases, they have not, thus far, significantly reduced prison crowding.

Instead, the large majority of convicts diverted into alternative programs tends to be persons who would not have been imprisoned in the first place. Thus, a person who might have been sentenced to loosely supervised probation would instead be sentenced to home detention with electronic monitoring. For any of the programs described below to have any effect in reducing the prison capacity crisis (thereby enabling prisons to hold repeat violent offenders for longer periods), the alternatives must be used as alternatives to imprisonment, rather than as alternatives to standard probation. One important step toward better use of alternatives would be to establish the presumption that alternatives, rather than imprisonment, should be used in the large majority of first-time drug offenses.

Electronic Monitoring and Home Detention

No form of alternative sanction grew more rapidly in the late 1980s than did electronic monitoring. Now, between 50,000 and 70,000 persons are under electronic monitoring every day.(120)

In a "passive" monitoring system, a computer randomly calls the prisoner's home, and the prisoner must respond by verifying his presence. In some jurisdictions, the computer analyzes the voice at the other end of the telephone and checks it against the prisoner's known vocal pattern. In an "active" system, the prisoner has a miniature radio transmitter attached to his ankle. The transmitter may send signals back to a receiver in the prisoner's home, or it may transmit a signal to a central receiver. A passive system costs about $2,500 per participant per year; an active system costs about twice as much.(121)

An active, ankle-transmitter system can be especially effective in helping women who are threatened by stalkers, ex-boyfriends, and the like. After the electronic cuff is put on the offender's ankle, a receiver is set up in the woman's home. If the offender comes within a few hundred yards of her home, the receiver alerts the woman, notifies the police, and even sends a fax to the offender's probation officer. The signal activates a microphone in the receiver at the woman's home, so that officers on the way to the scene can learn what is going on. The receiver has a "panic button" for the woman to use, and she is also given a portable "panic button."

Electronic monitoring functions effectively in cases in which some form of punishment is appropriate but imprisonment is considered too severe. Electronic monitoring is, however, not a defense against a prisoner who decides that he is willing to commit the crime and does not care about the implications of getting caught. In addition, some ankle cuffs can be removed without activating an alarm.

Within the next 15 years, electronic monitoring will become much more sophisticated. A monitor might be able to detect physiological changes in the prisoner and to respond by releasing doses of incapacitants. For example, a monitor placed on a sex offender might be configured to release chemicals that would make the offender go to sleep whenever the offender left his home and had an erection. Other monitors might be able to report on a prisoner's activities within his home and the activities of other persons in the home.

Increasingly sophisticated electronic monitoring programs raise significant civil liberties questions. On the one hand, they offer a future in which imprisonment may become obsolete, as all but the most dangerous criminals could be controlled with neurochemical implants. Yet as governments become ever more able to regulate the movement and the thoughts of the governed, history provides ample reason to fear that thought control will become commonplace, as temptations to abuse power are often not resisted.(122) Civil libertarians might do well to consider how to control the emerging monitoring technologies now, rather than wait until they have become commonplace.

Boot Camp

No boot camps for prisoners existed in 1983; in 1986 three states had set up boot camps, and by 1990, 26 states had boot camps, as did the federal prison system.(123) The spread of boot camps is attributable in large part to the intuitive appeal of the idea that some criminals, if subjected to military-style discipline for several months, might get their lives in some kind of order. Boot camps usually begin with a camp phase that lasts from two to nine months. The camp phase is often followed by a period of probation, or intensive supervised probation or time in a halfway house.

A Louisiana study found that inmates who entered boot camps had more "pro-social" attitudes than did inmates who chose not to enter, or who were not selected for the program. Boot camp inmates who completed their stay in the camp were found to have become significantly more "prosocial."(124) Nevertheless, a national study found the recidivism rate of boot camp graduates to be comparable to that of similar persons not sent to boot camp.(125)

Sending convicts to boot camps tends to be cheaper than imprisoning them, but only because boot camp sentences (usually 90 to 120 days and rarely more than 240 days) are so much shorter than the prison sentences that would have been imposed. On a day-to-day basis, boot camps are more expensive than prisons, because staffing levels must be higher.(126) As is the case with most of the alternatives discussed in this section, the persons sentenced to boot camp tend to be persons who would not have been sent to prison anyway. Only if about 80 percent of boot camp inmates are persons who otherwise would be in prison does a boot camp "break even" in reducing the net prison population. That is because persons who break camp rules and are expelled are typically sent to prison. If the inmate who moved from boot camp to prison would have (in a campless world) simply been sentenced to probation, the boot camp program increases the prison population.(127)


Unlike boot camps and electronic monitoring, probation is a long-standing alternative to incarceration. Instead of being sent to prison, a defendant sentenced to probation is placed under the supervision of a probation officer, to whom the defendant is required to report at regular intervals. Conditions may be placed on probation, such as a requirement that the defendant get a job or submit to random drug testing. If the probationer fails to comply with the terms of probation, the probation officer may recommend that the probationer be sent to prison.

Probation has been criticized, often with some merit, for failing to impose sufficient punishment or control on probationers. Probation is also criticized because of the supposedly high rate of recidivism of probationers. The latter criticism merits careful scrutiny.

People who argue that probation should be abolished because persons on probation are likely to commit new crimes usually begin by citing a RAND corporation study of probationers in Los Angeles and Alameda Counties, California. The RAND study found that 65 percent of felony probationers were rearrested within three years, and 51 percent were convicted of a new offense.(128) Far less noticed, however, have been more recent, broader studies that found a much lower recidivism rate. A Kentucky study found only 22 percent of felony probationers rearrested and 18 percent convicted of another crime within three years.(129) A Missouri study found 22 percent rearrested and 12 percent reconvicted in a 40-month period.(130) A study of 10,000 cases in 16 jurisdictions found that 34 percent of probationers were rearrested within 33 months. The study also observed that drug offenders were less likely to recidivate than were other offenders.(131)

One of the most thorough probation studies was a New Jersey analysis involving 2,000 probationers. That study found that 47 percent of felons on probation for robbery or burglary were convicted of another felony within four years, whereas 30 percent of felony drug probationers were reconvicted, as were 40 percent of total New Jersey felony probationers. Notably, 31 percent of the robbers and 32 percent of the burglars, but only 14 percent of the drug offenders, were charged with three or more offenses. Not surprisingly, the probation-period crimes perpetrated by the robbers and burglars on probation were much more likely to be burglaries or assaults than were the crimes committed by drug offenders on probation.(132)

The research on probation highlights the irrationality of mandatory minimums for drug offenders. Mandatory sentences deny probation to the group of offenders most likely to comply with probation. And to keep prison space available for the no-probation mandatory drug offenders, judges too often must sentence to probation increasingly large numbers of burglars and robbers--the very groups who are most likely to violate probation and commit interpersonal offenses.

Intensive Supervised Probation or Parole

If ordinary probation is sometimes insufficiently onerous, a new form of probation--intensive supervised probation--is carefully designed to be extremely onerous. About a decade old, ISP programs place offenders under much tighter control than does standard probation. Intensive supervision principles are also increasingly used on select parolees (persons under supervision after release from prison). Boot camp programs often put their graduates under intensive supervised probation or parole after completion of the camp phase.

Caseloads of standard probation officers are immense: 150 cases in Alabama, 200 in New York, 300 in Los Angeles.(133) But under ISP, a the probation officer has only about 25 cases, which allows him to monitor his charges much more thoroughly. Georgia's ISP program, begun in 1983, requires more than 100 hours of community service, frequent urine testing for drugs and alcohol, electronic home detention in many cases, contact with the probation officer almost every day in initial stages, and unannounced home visits. Revocation of probation for failure to comply with program rules is common.

Whereas ordinary probation officers usually depend on probationers to get to scheduled meetings, ISP officers leave the office to check on probationers at night and on weekends. Partly because ISP participants are watched so much more carefully than ordinary probationers or parolees, detection of violations of conditions of probation or parole is common. Thus, the return-to-prison rate for persons on ISP is about 50 percent; most of the returns are for technical violations rather than for new crimes.(134)

A Philadelphia study of juveniles placed in an "Intensive Aftercare Probation" program found that, while the percentage of juveniles who committed a crime during probation was similar to the percentage in standard juvenile probation programs, the total number of offenses was lower. The probation officers believed that because they had significantly lower caseloads than did probation officers in standard programs, the IAP officers could stay on top of juveniles who might otherwise have slipped through the cracks of the system. In addition, the IAP officers, because of their familiarity with their smaller caseload, were able to recommend swift revocation of parole in appropriate cases.(135)

A major nine-state study of intensive supervised probation or parole programs found the following results: ISP programs produce much better surveillance of probationers and parolees than do standard programs. Nevertheless, recidivism rates were no better than in standard probation programs. ISP programs, because of the high level of surveillance, did function effectively as intermediate-level punishments, designed for offenders for whom prison might be too severe but for whom a mere warning would be insufficient punishment. Indeed, ISP's restrictions on freedom are so stringent that a significant number of prisoners (25 percent in Oregon) chose a shorter prison term over a longer period of ISP. ISP programs do not save total prison costs, because ISP participants (being more closely monitored) are much more likely to have their probation revoked and be sent to prison than are standard probationers or parolees.(136)

Intensive probation is significantly more expensive than traditional probation or parole. In Georgia, for example, the intensive program costs $1,600 per participant per year, compared to only $300 for standard probation.(137) The nine-state national survey estimated the annual cost of ISP per offender at $4,000, compared to $12,000 for imprisonment.(138)

One important way to reduce the cost of parole and probation, including the intensive version of each, would be to reduce the imprisonment rate for technical violations (such as not keeping a job or using alcohol). Imprisonment is still, of course, the appropriate response to the perpetration of a new crime while on parole or probation. Washington State has recently mandated that imprisonment not be used as a response to technical parole or probation violations; a 60-day jail term should be the maximum punishment. In addition, Washington now requires that conditions of parole or probation be based on the offender's particular offense and past behavior. For example, frequent drug testing would be appropriate for a drug addict who was convicted of burglary but not for an embezzler with no record of substance abuse. Because, nationally, about 30 percent of persons admitted to prison have violated parole or probation,(139) a Washington-style approach to technical parole and probation terms may reduce the overuse of imprisonment for technical violations, thereby allowing prisons to keep their existing inmates in prison longer.(140)

Parole and the Abolition of Parole

Parole is similar to probation, except that parole is a program for people who have served time in prison and are under supervision after release. In many state prison systems, parole can replace part of a sentence. For example, a person sentenced to a six-year prison term might, after three years in prison, apply for parole and, if the parole board approved, spend the final three years of the sentence on parole rather than in prison.

Because of the early release available for parole, and the crowded prison conditions resulting from the war on drugs, many violent criminals serve fairly short sentences. James S. Gilmore III, attorney general of Virginia, pointed out that a first-time murderer in Virginia will actually serve about five years in prison, which will then be followed by release on parole. A first-time rapist will serve four years, a robber three, and a person convicted of aggravated assault two.(141)

Persons on parole have a high recidivism rate. A 1983 study of 108,000 parolees found that 62.5 percent were rearrested within three years. Forty percent of those persons were rearrested within six months of release.(142) A study of prisoners entering state prisons in 1989 found that 28 percent of the incoming male prisoners would still have been in prison had they not been released on parole.(143) Similarly, a 1989 Orlando Sentinel study of 4,000 prisoners released early from Florida prisons because of prison over-crowding (caused by the drug war) found that 31 percent of the persons released perpetrated a crime during time they would have been in prison, had they served their full sentences. A study of parolees in 11 large states found that nearly half were convicted of another crime within three years of release. Persons paroled for violent crimes were, unsurprisingly, much more likely to commit violent crimes than were other parolees.(144)

A Washington Post investigation of seven murders perpetrated in the District of Columbia in a single week puts a human face on those statistics. Two of the seven murderers walked out of halfway houses. Another murderer perpetrated his crime while free on bail for another crime. Another murderer was on parole from Virginia and was also free on bond in Prince George's County on robbery charges. A fifth murderer (who was the target of a playground shooting that killed a four-year-old) was on parole in the district. He had been arrested several months before for attempted murder.(145)

With similar stories being reported across the country, it is understandable that there is a strong citizen movement to abolish parole for violent crimes. George Allen was elected governor of Virginia in a landslide in 1993 in part on his pledge to abolish parole. And proposals abound in Washington to use federal grants as a carrot to make states abolish parole and adopt the federal system, under which prisoners serve at least 85 percent of their sentences--a plan known as "truth in sentencing."

Such proposals have some merit, if properly implemented. First of all, it must be noted that if there is a problem with parole boards' releasing dangerous violent criminals, the problem is not caused mainly by the parole boards but by the legislatures that have filled the prisons with drug offenders serving mandatory minimums--thereby creating a prison-space crisis that forces parole boards to release undesirable applicants for parole.

In Virginia, simply implementing the Allen proposal through additional prison building would cost $4 billion in construction, as well as large sums in continued prison operating costs.(146) Virginians concerned about controlling the cost of government may understandably balk at such heavy expenditures. Yet Virginia has available a low-cost method to implement the Allen plan. Forty-five percent of Virginia's inmates are currently imprisoned for nonviolent offenses. Virginia judges currently lack authority to sentence defendants to electronic home monitoring.(147) Increased use of electronic home monitoring of some nonviolent offenders could free enough prison space to implement much of the Allen plan. And certainly replacing nonviolent criminals with violent ones in Virginia prisons would be a progressive step for public safety.

While abolishing or restricting parole for violent criminals can have a positive effect, it is important to understand that the benefits may be smaller than the public may expect. A New Orleans study found that only 8 percent of burglary and robbery arrests of adults involved probationers, and only 1 to 2 percent involved persons on parole.(148) Thus, while the abolition of parole would certainly prevent some crimes, it might not be realistic to expect a major drop in violent crime as a result.

Sentencing Reforms

Previous sections have suggested reforms of mandatory minimum sentencing for drug crimes and of policies regarding parole and probation. Those reforms are designed to increase the number of violent criminals in prison while reducing the number of nonviolent criminals, especially drug offenders. This section offers additional reforms intended to restore justice to current American sentencing policy.

Abolish "Real-Offense" Sentencing

Most people believe that, in the American criminal justice system, a person may be sentenced for a crime only if he pleads guilty to the crime or is convicted of the crime after a trial. Although conviction-based sentencing was the practice in America for most of its history, it is now being replaced with "real-offense" sentencing. Under real-offense sentencing, a person who is convicted of any crime may have his sentence increased on the basis of any other offense that the prosecutor alleges was committed-- even though the supposed real offense was never proven in a court of law.

Incorporated in the federal sentencing guidelines, real-offense sentencing is also used in New York and other states.(149) Allegations about the supposed real offense may be based on hearsay, reputational evidence, and other "evidence" that would not be admissible at trial.(150)

The following are some actual results of real-offense sentencing.

A defendant was acquitted of possessing a certain quantity of drugs and convicted of possessing a smaller quantity. The court sentenced him on the basis of the higher amount, even though he was acquitted.(151)

Police illegally broke into someone's apartment and seized more than a kilogram of cocaine. The illegally seized evidence could not be used in court. Later, the suspect was arrested for selling a gram of crack co- caine and convicted of the small sale. The conviction for the sale of a gram carried a 33-month sentence, but the defendant was sentenced to 19.5 years on the basis of the illegally seized evidence.(152)

A defendant said he would sell an undercover narcotics agent 2,000 pounds of marijuana, but the defendant only produced 500 pounds for sale. The defendant was sentenced for the sale of 2,000 pounds.(153)

As police and prosecutors have become aware of the possibilities of real-offense sentencing, it has become increasingly common to manipulate the charges against the defendant to gain the maximum sentence from the minimum proof. A prosecutor may indict a defendant for only a low- level, easy-to-prove offense. Then at sentencing, the prosecutor will ask that the defendant be sentenced for the alleged commission of other, unproven crimes--crimes that the prosecutor knew could not be proven beyond a reasonable doubt in court.(154)

Simply put, real-offense sentencing destroys the requirement of proof beyond a reasonable doubt, the meaningfulness of a jury's acquittal, and the accused's right to confront the accuser. Real-offense sentencing ought to be abolished, but the U.S. Sentencing Commission has resisted even minor reform. In early 1993 the commission rejected proposals to allow judges to exercise discretion in sentencing defendants for real offenses of which the defendants had been acquitted.(155)

Stop Dual Prosecution Abuses

Federal prosecution and imprisonment of a defendant who has already been prosecuted in state court is referred to as "dual prosecution." During the Reagan and Bush years, administrative guidelines on dual prosecutions were greatly relaxed. The double-jeopardy clause of the Bill of Rights might be thought of as a protection against dual prosecution, but that clause, like the Fourth Amendment, is not nearly as powerful as it used to be, as interpreted by courts who confuse being tough on crime with being tough on the Constitution.

Thus, there are more and more cases like those of Rufina Canedo. She pleaded guilty to possession of 50 kilograms of cocaine and was serving a six-year state sentence. Federal prosecutors came and demanded that she testify against her husband, which she refused to do. Her guilty plea in state court was usable evidence in federal court. And so she was sentenced to a federal 20-year manda tory minimum. Her state prison time is not credited against her federal sentence.(156)

California actor Joe Renteria served 11 months in state prison for conspiring to buy marijuana and cocaine. After returning home, he resumed his career and began writing a script. He was federally prosecuted and sentenced to a five-year mandatory minimum. At the sentencing hearing, Judge David Kenyon stated:

The court is very bothered that the government would let this man or anybody go through an entire sentencing in state court on the exact same facts, wait until he's out of prison, he's starting a new life, he's married, he's working, and then announce, "Now we're going to prosecute you on the federal side." There's something wrong about that. No matter what the person does wrong, that too is wrong.(157)

Dual prosecution is wrong, and the attorney general of the United States could stop it with the stroke of a pen.(158)

Three Strikes, You're Out

Perhaps the most popular proposal for sentencing reform is known as "three strikes, you're out," which would impose a mandatory life sentence (or 20 years to life) on persons convicted of three violent crimes.

In Washington State, the proposal was dubbed the "Persistent Offenders Accountability Act." After being bottled up in the legislature and failing to gain enough signatures to make the ballot in 1992, the measure appeared on the ballot in November 1993, thanks to a donation of $60,000 from the National Rifle Association to pay for signature gathering.(159) The initiative received the largest positive vote of any initiative in the state's history.

A similar initiative may appear on the California ballot in 1994, again with heavy NRA support. Sen. Trent Lott of Mississippi has proposed a similar federal law, and President Clinton has followed suit.

The Washington State law is expected to affect about 40 to 70 criminals a year.(160) In the most extreme scenario, the initiative could, in 20 years, add as many as 746 people to Washington's 10,200 inmate population.(161)

In many cases, the three-strikes law may give rise to an appropriate increase in punishment for violent criminals. The Washington initiative replaces existing state sentencing guidelines, under which a rapist with two previous sex offenses could be sentenced to a 14-year term, or a child molester with two previous offenses could receive 9.5 years.(162) Likewise, a person with two violent felony convictions who perpetrated a robbery in which he severely beat the victim would be sentenced to only five years and serve three and a half.(163)

But the Washington law also counts too many nonviolent, or lower level violent, offenses as a "strike." Among the covered offenses are possession of an incendiary device,(164) unarmed burglary (which includes knocking someone down and stealing a jacket), second-degree assault (including bar fights), reckless driving causing nonfatal injury, and even stealing cattle while carrying a gun.(165)

One of the most serious problems with the Washington initiative is that delivery of cocaine or possession with intent to deliver is counted as a strike.(166) As mentioned earlier, New York State has experimented with similar mandatory sentencing laws for cocaine sales and found that the number of drug addicts who commit repeat, small-scale offenses (such as selling small quantities of cocaine to other users to support their habit) is so large that mandatory sentencing, even when limited to repeat offenders, will flood the prison system.

Approved as a ballot initiative, the Washington three- strikes law went into effect on December 1, 1993, and is immune from legislative modification or repeal for two years. At the end of the two-year period, it would be sensible for the legislature to fine-tune the law by removing from its scope the offenses mentioned. The paring would leave intact the core of the legislation.

The Washington law also has some carefully thought out provisions that should serve as models for other, similar laws. First, the statute includes a "wash-out" provision, so that, if an offender stays free of crime for 5 or 10 years after conviction of certain felonies, the felony no longer counts as one of the three strikes. The most serious felonies, however, never wash out. In addition, the three- strikes counting is based on distinct felony convictions arising at different times. Thus, if a person with no previous convictions went on a crime spree and committed three serious felonies, the three-strikes provision would not come into play. The provision applies only to persons with two prior convictions who are then convicted of a third, separate offense.

In contrast, federal law dealing with possession of firearms by repeat offenders has been interpreted to apply to a person who has been prosecuted only a single time.(167) The federal statute should be modified to comport with the Washington State model.

Backers of three strikes appear to endorse the theory of selective incapacitation. For example, John Carlson, father of the Washington initiative, points out that "the clear majority of violent crime is committed by a small fraction--about 6 percent--of the criminal population."(168) Indeed, one study showed that while the average prison inmate might commit about five burglaries a year, the top 10 percent of burglars committed about 232; while the average inmate robbed 5 persons a year, the top 10 percent robbed 87.(169)

Thus, Carlson's point about the small fraction of the criminal population is clearly true, as applied to high-rate predatory crimes such as robbery or burglary. Other research suggests that incapacitating large numbers of robbers and burglars can have a major impact on the robbery and burglary rate.(170) A famous study of Philadelphia arrest records by criminologist Marvin Wolfgang found that 7 percent of all males perpetrated two-thirds of all violent crime, three-quarters of rapes and robberies, and almost all murders.(171)

It must be recognized that three strikes is an extremely blunt instrument for targeting the high-rate violent criminal. Past involvement in crime is, compared to any other single variable, the best predictor of future criminality. But criminal convictions, or even arrest records, do not necessarily correlate closely with a person's actual criminal past.(172) Even models that take into account many factors besides felony convictions (such as job stability, marital status, juvenile criminal record, and substance abuse) have only moderate success in categorizing the criminals who are the uniquely active high-rate offenders. Analysis of a person's criminal record only is not a particularly accurate way to determine if he is a high-volume offender.(173)

Thus, while selective incapacitation is sensible theoretically, the three-strikes proposals are, at best, imperfect instruments for achieving the goal. The goal is better served, ironically, by allowing judges wide discretion in sentencing, so that a court, looking at all the relevant facts, can decide whether the criminal is a high-rate offender who should serve an especially long sentence.

It should also be remembered that few violent criminal careers, even those of repeat offenders, persist far into middle age, and virtually none persist into old age. Thus, the continued incarceration of a 55-year-old who may have perpetrated armed robberies in his teens and twenties may do little to benefit public safety. Indeed, prison cells that are used to hold geriatric prisoners who are very unlikely to commit violent crime are unavailable to hold younger, active violent criminals. Whatever value there is in incarcerating the 55-year-old until he dies in prison 20 years later is derived from the social interest in retribution, rather than from a public safety interest in incapacitating an active criminal, as well as the difficult-to-quantify deterrent effect that a three-strikes law might have on criminals with one or two convictions today. Balanced against the possible deterrent effect is the fact that, in the absence of any realistic potential for imposition of the death penalty, a logical criminal with two strikes against him would have no incentive not to kill witnesses and victims; the punishment for the third felony (life in prison) would be no less than punishment for a homicide (life in prison).

To the extent that three-strikes laws are carefully drafted to include only serious violent felonies, the laws can be a helpful step forward. But to the extent that three-strikes laws include consensual offenses or make it impossible for judges to treat relatively less dangerous offenders differently than they do violent predators, such laws simply continue, rather than reform, the failed sentencing policies that endanger public safety.


In the last 15 years, American elected officials have required prisons to engage in a bold social experiment. The historical prison policy--incarceration of violent criminals--has been replaced with a policy of using prisons mainly to punish drug offenders with increasingly severe, mandatory terms in increasingly overcrowded prisons.

The social experiment has been a failure. Some advocates of the drug war point to the enormous imprisonment figures as proof of their success, and the imprisonment figures are touted as proof that "we are winning the war." But to a growing number of Americans, it is becoming increasingly clear that no matter how many of "the enemy" (Americans who use or sell illegal substances) the American prison system incarcerates, the nation is not even coming close to winning the drug war. It is simply beyond the capability of the American military-industrial-law enforcement complex to subdue and control millions of determined people who do not want to be subdued, even if the control is "for their own good."(174)

Mandatory minimums and other tactics to increase drastically punishment for nonviolent drug offenders may appear to cause no problem for people who do not use drugs. But, in truth, mandatory minimums threaten the safety of every American. While taxes and debt rose in the last decade, in part to help pay for more than doubling national prison capacity, most of the prison space went to incarcerate people for nonviolent offenses. Today, prisons are bursting at the seams, and there is insufficient room for hard-core violent criminals because the space is already taken by nonviolent criminals serving mandatory minimums. Mandatory minimums are the best thing that ever happened to violent criminals, because mandatory minimums prevent today's judges from doing what they want--putting violent thugs away for a long time--and force the prison system to waste precious space on nonviolent offenders. The violent criminals out on parole are given their opportunity to commit more crimes by a criminal justice system fixated on drugs.

Professor Dan Polsby of the Northwestern University School of Law suggests that the mania for locking up drug offenders has been "one of the great bipartisan public policy fiascoes of recent times."(175) It is time for the fiasco to end. It is time to return to America's historic sentencing policy. It is time for America's prisons to recognize that their primary mission is not the punishment of voluntary capitalist acts between consenting adults, even if the acts involve substances disapproved by the majority. It is time for America's prisons to be redevoted, as they were in previous, safer generations, to the incapacitation of violent criminals.


(1) U.S. Department of Justice, Bureau of Justice Statis tics, Prisoners in 1992, NCJ-141874 (Washington: U.S. De partment of Justice, May 9, 1993), p. 1. Cited hereafter as Prisoners in 1992.

(2) U.S. Department of Justice, Bureau of Justice Statis tics, Sourcebook of Criminal Justice Statistics, 1990, ed. Kathleen Maguire and Timothy J. Flanagan (Washington: U.S. Department of Justice, 1991), p. 604; and "Prison Population Jumps 5%, Passes 900,000 Mark," Criminal Justice Newsletter 24, no. 16 (September 16, 1993): 7.

(3) Allen J. Beck, Thomas P. Bonczar, and Darrell K. Gil liard, Jail Inmates 1992, NCJ-143284 (Washington: U.S. De partment of Justice, August 1993), p. 1.

(4) As of January 1, 1993, there were 2,079,881 adults on probation in the state and federal prison systems and 573,844 on parole. George M. Camp and Camille Graham Camp, The Corrections Yearbook 1993: Probation and Parole (South Salem, N.Y.: Criminal Justice Institute, 1993), pp. 18, 46.

(5) U.S. Department of Justice, An Analysis of Non-Violent Drug Offenders with Minimal Criminal Histories (Washington: U.S. Department of Justice, February 4, 1994), p. 13.

(6) Advisory Commission on Intergovernmental Relations, The Role of General Government Elected Officials in Criminal Justice (Washington: ACIR, 1993), p. 11.

(7) John Hanchette and Paul Barton, "Criminals Freed Early, Even as More Cells Being Built," (Cherry Hill, N.J.) Couri er-Post, June 6, 1993, p. 6B.

(8) Prisoners in 1992, p. 1.

(9) Camp and Camp, p. 9.

(10) Ibid., p. 35.

(11) Paul W. Valentine, "'You Can't Build Your Way Out,' Maryland Prison Official Says," Washington Post, May 24, 1993; and William F. Zorzi, Jr., "Home Detention Is Urged," Baltimore Sun, June 22, 1993.

(12) Jack Eckerd, "Released Prisoners Lack Skill to Be Productive in Society," (Fort Myers, Fla.) News Press, June 6, 1993.

(13) Molly Ivins, "'Lock 'em Up' Doesn't Work Anymore," Detroit Free Press, April 16, 1993, p. 11A.

(14) American Civil Liberties Union, National Prisons Proj ect, Status Report: State Prisons and the Courts--January 1, 1993, discussed in "Forty States Currently under Court Orders to Improve Prison Conditions, Alleviate Crowding, ACLU Reports," Justice Research, January-February 1993, p. 1. Alabama's prison system, built for 14,500 inmates, now holds 18,000. Allen L. Tapley, "Hope in Tough Times," Overcrowded Times, April 1993, p. 18.

(15) Camp and Camp, p. 36.

(16) Hanchette and Barton.

(17) Camp and Camp, p. 8.

(18) Australian Institute of Criminology, Prisoners in Asia and the Pacific 1993 (Canberra: Australian Institute of Criminology, November 24, 1993). It must be noted, however, that part of the higher U.S. incarceration rate stems from higher American crime rates. For example, Carnegie Mellon's Alfred Blumstein's analysis demonstrates that the United States does not have more prisoners per murder or prisoners per robbery than do other nations. Blumstein, "Prison Populations: A System Out of Control?" Crime and Justice 10 (1988): 233-37. The Blumstein demonstration is limited, however, because it looks only at the total number of per sons imprisoned, not at what they are imprisoned for. Thus, if one nation (New Zealand, for example) imprisoned most persons convicted of robbery but relatively few drug offend ers, while another nation (the United States, for example) imprisoned a smaller percentage of robbers but a great many drug offenders, the two nations would have an equal ratio of "prisoners per robbery"--even though the two nations had very different policies regarding imprisonment for robbery.

(19) In 1992 convicted inmates made up 49 percent of the jail population. Beck et al., Jail Inmates 1992, p. 2.

(20) Beck et al., Jail Inmates 1992; and Stan C. Proband, "Jail Populations Up, Racial Disproportions Worse," Over crowded Times, August 1993, p. 4.

(21) Hanchette and Barton.

(22) "State Budget and Tax Actions 1993," City and State, August 16, 1993.

(23) Stephanie Mencimer, "Righting Sentences," Washington Monthly, April 1993, p. 27.

(24) Pennsylvania Economy League study quoted in Kay Monaco, "Pennsylvania Coordinator's Report," Overcrowded Times, April 1993, p. 17.

(25) Francis X. Clines, "Prisons Run Out of Cells, Money, and Choices," New York Times, May 28, 1993.

(26) Hanchette and Barton.

(27) National Council on Crime and Delinquency, Why Punish the Children? A Reappraisal of the Children of Incarcerated Mothers in America (San Francisco: NCCD, 1993).

(28) Lawrence J. Fleming, "The War on Crime--A Domestic Vietnam," Champion, September-October 1993, p. 34; and U.S. Department of Justice, Bureau of Justice Statistics, Source book of Criminal Justice Statistics, 1990, p. 2, Table 1.1.

(29) One analysis suggested that the financial benefits, including indirect benefits, of incarcerating a single inmate were between $172,000 and $2,364,000. (The study did not look into the costs of homicide, except when it accompa nied another felony, and did not look into potential costs of drug crimes.) David P. Cavanaugh and Mark A. R. Kleiman, A Cost Benefit Analysis of Prison Cell Construction and Alternative Sanctions, prepared for the National Institute of Justice, May 1990; cited in U.S. Department of Justice, Office of Policy Development, The Case for More Incarcera tion, NCJ-139583 (Washington: U.S. Department of Justice, October 28, 1992), p. 18.

(30) "Drug Cases Swamping CJ System," Law Enforcement News, February 28, 1993, p. 3.

(31) "Reno Orders Review of Sentencing Rules for Low-Level Drug Offenders," Justice Bulletin, May 1993, p. 2.

(32) Allen Beck et al., Survey of State Prison Inmates, 1991, NCJ-136949 (Washington: U.S. Department of Justice, March 1993), p. 4.

(33) David Foster, "Prisons Feel Squeeze on Drugs, Crime," Seattle Times, April 12, 1993.

(34) Francis X. Clines, "Tough Sentences for Low-Level Drug Felons," New York Times, March 23, 1993. Ninety percent of New York drug offenders are sentenced under mandatory mini mum laws. Correctional Association of New York, Mandatory Sentencing Laws and Drug Offenders in New York State, Febru ary 1993; cited in Campaign for an Effective Crime Policy, Evaluating Mandatory Minimums Sentences (Washington: CECP, October 1993), p. 5.

(35) Stephen Chapman, "Many a Draconian Sentence Just Doesn't Fit the Crime," (Colorado Springs) Gazette-Tele graph, March 23, 1993, p. B5 (syndicated from Chicago Trib une).

(36) Harry K. Singletary, "State Corrections Approach to Treatment," in U.S. Sentencing Commission, Drugs and Vio lence in America (Washington: U.S. Sentencing Commission, 1993), p. 226.

(37) Michael James, "Prison Boss Prefers Alternatives," Baltimore Sun, August 3, 1993, p. 2B.

(38) Ivins.

(39) David L. Feldman, "Behind Bars," New Democrat, October 1992, p. 24.

(40) Ibid. Hemp possession is a felony in Nevada.

(41) Hanchette and Barton.

(42) Prisoners in 1992, p. 10, appendix Table 2.

(43) Prisoners in 1992, p. 10, appendix Table 1. The graphs in Figure 4 are for new court commitments to state prisons and thus do not include commitments for violations of parole or probation.

(44) "Kicking the Prison Habit," Newsweek, June 14, 1993.

(45) Mencimer, p. 28.

(46) Beck et al., Survey of State Prison Inmates, 1991, p. 11.

(47) Ibid. It is probably true that some of the drug of fenders also committed violent crimes for which they were never charged or convicted. A free society can only impris on persons for crimes of which they have been convicted as individuals, not for crimes that some persons who have similar traits might have committed.

(48) Stuart Taylor, Jr., "How a Racist Drug War Swells Violent Crime," American Lawyer, April 1993, p. 31. The percentage with "violent" offenses would be higher if bur glary were counted as violent crime. It should be noted, though, that the burglaries themselves may be the result of drug laws. The reason that cocaine and heroin addicts commit burglaries to feed their habits, while nicotine and alcohol addicts do not, is that the price of illegal drugs is artificially inflated by their illegality and the conse quent risk premium charged by sellers.

(49) William Booth, "Drug War Locks Up Prisons," Washington Post, July 7, 1993, p. A14.

(50) "Mandatory Minimums Undermine Sentencing Guidelines, Studies Show," Justice Bulletin, August 1993, p. 5. It may be asked whether the federal government has any legitimate role at all in criminal law enforcement. The Constitution specifically authorizes federal enforcement of only two types of laws, both of which involve uniquely federal con cerns. The first authorized federal criminal law enforce ment is based on the congressional power "to provide for the punishment of counterfeiting the securities and current coin of the United States." The second congressional criminal power is the power "to define and punish piracies and felo nies committed on the high seas, and offenses against the law of nations."

(51) Margaret Werner Cahalan, Historical Corrections Statis tics in the United States, 1850-1984, NCJ-102529 (Washing ton: U.S. Department of Justice, 1986), pp. 151-54, Tables 6-8 and 6-9.

(52) Mistretta v. United States, 488 U.S. 361, 414-15 (1989) (Scalia, J., dissenting, quoting dissenting views of a sentencing commissioner).

(53) Jose A. Cabranes, "Incoherent Sentencing Guidelines," Letter to the editor, Wall Street Journal, August 28, 1992.

(54) Don Edwards, "Congress Swamped the Courts," Washington Post, July 7, 1993, p. A21.

(55) George Bush, speaking for the Comprehensive Drug Abuse Prevention and Control Act of 1970, H.R. 18583, 91st Cong., 2d sess., Congressional Record, September 23, 1970, vol. 116, pt. 25, p. 33314. The repeal bill is Public Law 91- 513, 84 Stat. 1236.

(56) "Mandatory Minimums Undermine Sentencing Guidelines," p. 7. Current federal mandatory minimums for drug offenses require five years in prison without parole for possession of more than 5 grams of crack cocaine and for any role in the distribution of any of the following: 500 grams of powder cocaine, 100 grams of heroin, 1 gram of lysergic acid diethylamide (LSD), 100 kilograms of marijuana, or 100 marijuana plants. In computing the weights, the weights of any adulterants or carriers are added to the weight of the actual drug. A 10-year mandatory minimum (again with no parole) is required for any role in the distribution of 50 grams of crack cocaine, 5 kilograms of powder cocaine, 1 kilogram of heroin, 10 grams of LSD, and 1,000 kilograms (or 1,000 plants) of marijuana. The sentences are doubled if the offender has had any previous drug felony conviction, no matter how minor (such as for marijuana possession) or how remote in time.

(57) U.S. Sentencing Commission, Mandatory Minimum Penalties in the Federal Criminal Justice System (Washington: U.S. Sentencing Commission, August 1991), p. 10; and Henry Scott Wallace, "Mandatory Minimums and the Betrayal of Sentenc ing," Federal Bar News & Journal, March-April 1993, p. 158. The four statutes are 21 U.S.C. sec. 841 (manufacture or distribution of controlled substances), 21 U.S.C. sec. 844 (possession of cocaine base), 21 U.S.C. sec. 960 (import or export of controlled substances), and 18 U.S.C. sec. 924(c) (5- to 20-year sentence enhancements for using or carrying a firearm during a drug crime or a violent crime).

(58) It is irrelevant how remote in time the prior convic tion was. Under 21 U.S.C. sec. 841, the prior conviction must be for a "crime of violence," but a crime of violence can be breaking into an automobile to steal a car stereo or a jacket. 18 U.S.C. sec. 924(c). See also U.S. Sentencing Commission, Mandatory Minimum Penalties, p. 30n. 91.

(59) William Shwartzer, "Judicial Discretion in Sentencing," Address to National Conference on Sentencing Advocacy, April 20, 1991, quoted in Lawrence J. Fleming, "The War on Crime-- A Domestic Vietnam," Champion, September-October 1993, p. 33. One study found that the mandatory minimums had the greatest impact on increasing the severity of sentences for large quantities of marijuana. Barbara S. Meierhoefer, "The Role of Offense and Offender Characteristics in Federal Sentencing," Southern California Law Review 66 (1992): 367, 373-75.

(60) "Mandatory Minimums Undermine Sentencing Guidelines," pp. 8-9.

(61) Stuart Taylor, "A Test of Reno's Courage," Legal Times, August 9, 1993, Supplement, p. 36.

(62) "Mandatory Minimums Undermine Sentencing Guidelines," p. 9.

(63) Mike Sager, "The Case of Gary Fannon," Rolling Stone, September 3, 1992.

(64) Dirk Johnson, "For Drug Offenders, How Tough Is Too Tough?" New York Times, November 8, 1993, p. A8.

(65) David Hardy, "Legal Restrictions on Firearms Ownership as an Answer to Violent Crime: What Was the Question?" Hamline Law Review 6 (July 1983): 407.

(66) Commonwealth v. Lindsey, (Mass. Supreme Judicial Court, March 5, 1986). Eventually, the defendant was pardoned by the governor.

(67) Brief for Civil Liberties Union of Massachusetts as amicus curiae, Commonwealth v. Jackson, pp. 22-26, cited in James Beha, "And Nobody Can Get You Out," Boston University Law Review 57 (1977): 110n. 55. The Massachusetts legis lature's black caucus had also opposed the bill, because of concern about discriminatory licensing and arrests. Ibid., p. 108n. 45.

(68) "Cutting Drug Sentences by Half Proposed," Rocky Moun tain News, December 1, 1992, p. 12.

(69) The 18-year-old student would have "induced . . . a

child . . . to act as his agent in the unlawful possession for the purposes of sale any controlled substance."

(70) Both the student and the smuggler would be guilty of bringing a schedule-one controlled substance into the state.

(71) U.S. Sentencing Commission, Mandatory Minimum Penal ties, p. iii.

(72) United States v. Barth, no. 92-2152, 1993 WL 98039 (8th Cir. 1993).

(73) United States v. Rogers, 982 F.2d 1241, 1245 (8th Cir. 1993). A different type of sentencing entrapment occurs in "reverse sting" operations, in which government undercover agents sell drugs to the defendant. Aware of the sentencing guidelines' heavy focus on drug weight, agents will some times sell large quantities of drugs at extremely low prices in order to entrap the defendant into possession of a larger quantity of drugs. Modifications to the sentencing guide lines now allow a downward departure in sentences involving such reverse stings, to reflect the quantity of drugs that the defendant would have bought under normal conditions. Sentencing Guidelines, Application note 17, sec. 2D1.1, in Roger W. Haines, Jr., Kevin Cole, and Jennifer Woll, Federal Sentencing Guidelines Handbook (Colorado Springs: McGraw- Hill, 1993), pp. 182-83.

(74) Mencimer, p. 28.

(75) Clines, "Tough Sentences for Low-Level Drug Felons."

(76) U.S, Sentencing Commission, Mandatory Minimum Penalties in the Federal Criminal Justice System, p. 82.

(77) William H. Rehnquist, "Luncheon Address," in U.S. Sentencing Commission, Drugs and Violence, pp. 286-87.

(78) Anthony Kennedy, Testimony before U.S. House Appropri ations Subcommittee, quoted in "Mandatory Sentencing Is Criticized by Justice," New York Times, March 10, 1994.

(79) "Judge Hears Conscience, Not 'Gun Law' Sentence," American Bar Association Journal 67 (November 1981): 1433.

(80) Cris Carmody, "Revolt to Sentencing Is Gaining Momen tum," National Law Journal, May 17, 1993, p. 10. Under 28 U.S.C. 294(b), a senior judge may "perform such judicial duties as he is willing and able to undertake . . ." (empha sis added).

(81) Ibid.

(82) "The Verdict Is In," ABA Journal, October 1993, p. 78.

(83) Ibid.

(84) U.S. Sentencing Commission, Mandatory Minimum Pena lties, Appendix G, "Congress Swamped the Courts."

(85) Ibid., p. 107.

(86) United States v. Spencer, 817 F. Supp. 176 (D.D.C. 1993) (the case is on appeal as this paper is written, and the decision is not binding on other courts or outside the District of Columbia); and "Making Sense of Sentences," Washington Post, May 1, 1993. Sentencing Commission guide lines require a 30-year minimum sentence for a "career offender," defined as anyone with two prior felony convic tions for drug crimes or crimes of violence. In the Spencer case, the defendant had previous convictions for small- quantity heroin offenses. Judge Greene sentenced the defen dant to 10 years, which was the statutory mandatory minimum specified by Congress. The court emphasized that had Con gress, rather than the U.S. Sentencing Commission, enacted the 30-year sentence, the court would have found the sen tence constitutional.

(87) Minnesota v. Olson, 325 N.W.2d 13 (1982).

(88) State v. Prentiss, 786 P.2d 932 (Ariz. 1989) (a drunk- driving statute improperly allowed only the prosecutor to consider mitigating circumstances); and State v. Lagares, 601 A.2d 698 (N.J. 1992) (mandatory sentencing provision for repeat drug offenders saved from unconstitutionality by court order that attorney general promulgate guidelines on which defendants should receive lesser or greater sen- tences).

(89) The judge in the case, rebelling against the prospect of the defendant's incarceration's forcing the release of a more dangerous criminal from the Texas prison system, de layed the sentencing hearing, in anticipation of Texas's reforming its penal code. John Makeig, "Judge Draws Line on Crack Crackdown: He Won't Push Convicts Out of Prison by Sending Up Small-Time Drug Users," Houston Chronicle, April 7, 1993, p. 21A.

(90) U.S. Department of Justice, Bureau of Justice Statis tics, Compendium of Federal Justice Statistics, 1990, NCJ- 143499 (Washington: U.S. Department of Justice, September 1993), p. 42, Table 4.4.

(91) Ibid., p. 43, Table 4.5. First-time property fraud offenders were sentenced to an average term of 14.7 months, and first-time offenders for other property crimes to 13.3 months.

(92) Charles E. Schumer, "Finding a Middle Ground on Manda tory Minimums," Legal Times, August 23, 1993, p. 21. Sen. Phil Gramm (R-Tex.), another defender of mandatory minimums, uses U.S. Sentencing Commission's data to show that "only 662--just 7 percent--of the 9,221 individuals sentenced in 1992 to a mandatory minimum term of imprisonment for a drug offense were not armed, had no prior contact with the crimi nal justice system and played only a minor role in the criminal activity." Gramm, "Protective Floor," Washington Times, October 10, 1993, p. B4. Gramm's statistic is sub ject to much of the same criticism as is Schumer's. The mere fact that an individual has had "prior contact with the criminal justice system"--including a low-level misdemeanor arrest--does not prove that any mandatory minimum to which he is sentenced will be just. More fundamentally, Senator Gramm's 7 percent are precisely the persons who are most affected by mandatory minimums, for they are the defendants for whom judges would be most likely to exercise discretion in imposing a lesser sentence.

(93) Vincent Broderick, Testimony of July 28, 1993, before U.S. House of Representatives, Subcommittee on Crime and Criminal Justice; quoted in "House Panel Cool to Appeals for Mandatory Minimum Repeal," Criminal Justice Newsletter 24, no. 15 (August 2, 1993): 5. The Judicial Conference is the policymaking body of the federal courts.

(94) U.S. Department of Justice, "An Analysis of Non-Violent Drug Offenders with Minimal Criminal Histories," February 4, 1994. Persons with zero criminal-history points (under U.S. Sentencing Commission guidelines) accounted for 12,727 low- level drug offenders, constituting 28.2 percent of federal drug prisoners and 16.6 percent of total federal prisoners (pp. 2-3). The figures are all based on the federal prison population as of June 1993.

(95) "Minimums Undermine Sentencing Guidelines," p. 10. Judge Wilkins's proposal would apply to what he describes as "unusual cases with truly compelling circumstances." In those cases, "courts would be permitted to depart below the guideline range according to well-established statutory and case law criteria . . . appropriate adjustment for mitigat ing factors would not be left to unguided discretion, but rather would occur in a certain and predictable fashion." Quoted in ibid. The government would be allowed to appeal in cases in which the court was thought to have abused its discretion.

(96) The author immodestly notes that his father, former Colorado state representative Gerald Kopel (D-Denver) was the chief sponsor of the nation's first sunset law.

(97) "The Verdict Is In."

(98) The Democratic 1993 crime bill included a requirement for prison impact assessments, to be performed through a collaboration of the U.S. Department of Justice, the U.S. Sentencing Commission, and the Administrative Office of the U.S. Courts. A similar system is currently in effect in Minnesota, where the Minnesota Sentencing Guidelines Commis sion is explicitly required by statute to give "substantial consideration" to prison capacity when setting guidelines. (99)Dennis Cauchon, "Judicial Panel Proposes Cap on Drug Sentences: Prison Costs, Crowding Cited in Recommendation," USA Today, March 22, 1993. The panel also recommended treating crack cocaine the same as powder cocaine, abolish ing the practice of sentencing defendants for crimes of which they were acquitted, giving judges discretion to reduce sentences for persons who have informed on other persons (federal law currently allows only the prosecutor to request a reduction for cooperation), and revising LSD sentences to eliminate the effect of carrier weight.

(100)Don J. De Benedictis, "How Long Is Too Long?" ABA Journal, October 1993.

(101)"With Prison Costs Rising, Florida Ends Many Mandatory Sentences," New York Times, May 29, 1993.

(102)Public Agenda Foundation (New York) research discussed in Donna Hunzeker, "Bringing Corrections Policy into the 1990s," State Legislative Report 17, no. 5 (March 1992): 4.

(103)The poll was conducted by Public Agenda for the Edna McConnell Clark Foundation. Respondents were asked about whether to incarcerate particular offenders. The respon dents were then told about various alternatives to prisons, such as boot camps or intensive supervised probation. Support for use of the alternatives was very high. For example, in the pretest, 52 percent had supported imprison ing a drug dealer on the first conviction; in the posttest, 18 percent preferred prison, while 67 percent chose sentenc ing alternatives. Public Agenda Foundation, Punishing Criminals: The People of Pennsylvania Speak Out, 1993, summarized in Steve Farkas, "Pennsylvanians Prefer Alterna tives to Prison," Overcrowded Times, April 1993, pp. 1, 13-15.

(104)Morgan Reynolds, "Why Does Crime Pay?" National Center for Policy Analysis Policy Backgrounder no. 123, Dallas, December 8, 1992.

(105)Phil Gramm, "Don't Let Judges Set Crooks Free," New York Times, July 8, 1993, p. A19.

(106)Mencimer, p. 26.

(107)Carl Reynolds, "Texas Commission Proposes Corrections Overhaul," Overcrowded Times, April 1993, p. 16.

(108)Mencimer, p. 26.

(109)Morgan Reynolds, "Crime in Texas," National Center for Policy Analysis Policy Backgrounder 110, Dallas, 1991; and Reynolds, "Why Does Crime Pay?"


(111)Bruce L. Benson and David W. Rasmussen, "Illinois' War on Drugs: Some Unintended Consequences," Heartland Policy Study no. 48, Chicago, 1992.

(112)U.S. Department of Justice, Bureau of Justice Statis tics, Federal Criminal Case Processing, 1989-90, NCJ-136945 (Washington: U.S. Department of Justice, September 1992), p. 17, Table 17.

(113)Robert M. Figlio, "Self-Reported and Officially Defined Offenses in the 1958 Philadelphia Birth Cohort," in Cross- National Longitudinal Research on Human Development and Criminal Behavior, ed. Elmar G. M. Weitekamp and Hans-Jrgen Kerner (Dordecht, Boston, and London: Kluwer Academic Press, 1994).

(114)U.S. Department of the Treasury, Bureau of Alcohol, Tobacco and Firearms, Protecting America: The Effectiveness of the Federal Armed Career Criminal Statute (Washington: BATF, March 1992), p. 15.

(115)"Mandatory Minimums Undermine Federal Guidelines," p. 9.

(116)Gramm, "Protective Floor," p. B4.

(117)P. J. Goldstein, H. H. Brownstein, and P. J. Ryan, "Drug Related Homicide in New York: 1984 and 1988," Crime and Delinquency 38 (1992): 459-76; and P. J. Goldstein, H. H. Brownstein, P. J. Ryan, and P. A. Bellucci, "Crack and Homicide in New York City, 1988: A Conceptually Based Event Analysis," Contemporary Drug Problems 16 (1989): 651-87.

(118)Michael Hennessey, "Our National Jail Scandal," Ameri can Jails, July-August 1993, p. 15.

(119)Feldman, p. 24.

(120)J. Robert Lilly, "Electronic Monitoring in the U.S.: An Update," Overcrowded Times, October 1993, pp. 4, 15.

(121)American Legislative Exchange Council, "Ensuring Truth in Sentencing" in Equal Justice under Law, ed. Samuel A. Brunelli, Gary Barrett, and Adam DeVries (Washington: Amer ican Legislative Exchange Council, 1993), p. 19. For some preliminary research on the behavior of persons subject to home detention, see Terry L. Baumer, Michael G. Maxfield, and Robert I. Mendelsohn, "A Comparative Analysis of Three Electronically Monitored Home Detention Programs," Justice Quarterly 10, no. 1 (March 1993): 121-42.

(122)Many aspects of the American criminal justice system, such as the prohibition on compelled self-incrimination, are founded on respect for individual autonomy. While society may choose to incarcerate a person, society has not, thus far, claimed the authority to replace a prisoner's own thoughts with thoughts created by the government. Neuro chemical monitoring, by making it impossible for a person to sustain certain thoughts, is a form of mind control. If mind control is taken far enough, it operates as a kind of death sentence--the prisoner's true mind is replaced by a state mind. Even though the prisoner's body may survive for decades, the prisoner's personality may be obliterated.

It might be naive to assume that neurochemical controls will be used only on violent prisoners. As the controls gained greater acceptance and familiarity, pressure would inevitably build for neurochemical controls to be used to suppress the desire for altered states of consciousness, thoughts of rebellion, and feelings of "prejudice" (as "prejudice" might be defined according to prevailing norms of political correctness).

(123)Dale G. Parent, "Boot Camps Failing to Achieve Goals," Overcrowded Times, August 1993, p. 1; and General Accounting Office, Prison Boot Camps: Short-Term Prison Costs Reduced, but Long-Term Impact Uncertain (Washington: General Account ing Office, 1993), p. 3.

(124)Doris Layton MacKenzie and James W. Shaw, "Inmate Adjustment and Change during Shock Incarceration: The Impact of Correctional Boot Camp Programs," Justice Quarterly 7, no. 1 (March 1990): 125-50.

(125)Parent; and General Accounting Office. Recidivism analysis is necessarily tentative, since most camps have been in operation such a short time. Data from Colorado's new boot camp at Buena Vista show that the graduates have a slightly higher recidivism rate than a control group. John Sanko, "Legislator Wants to Cap Prison Beds," Rocky Mountain News, November 10, 1993.

(126)General Accounting Office.

(127)Parent, pp. 1, 12-15.

(128)Joan Petersilia, Susan Turner, James Kahan, and Joyce Peterson, "Executive Summary of Rand's Study, 'Granting Felons Probation: Public Risks and Alternatives,'" Crime and Delinquency 31 (1985): 379-92.

(129)Genneard F. Vito, "Felony Probation and Recidivism: Replication and Response," Federal Probation 50, no. 4 (1986): 17-25.

(130)Johnny McGaha, Michael Ficter, and Peter Hischburg, "Felony Probation: A Re-Examination of Public Risk," American Journal of Criminal Justice 11, no. 1 (1987): 1-9.

(131)M. A. Cuniff, A Sentencing Postscript: Felony Probaioners under Supervision in the Community (Washington: National Association of Criminal Justice Planners, 1986).

(132)John T. Whitehead, "The Effectiveness of Felony Probation: Results from an Eastern State," Justice Quarterly 8, no. 4 (December 1991): 525-43.

(133)Todd R. Clear and Edward J. Latessa, "Probation Officers' Roles in Intensive Supervision: Surveillance versus Treatment," Justice Quarterly 10, no. 3 (September 1993): 443.

(134)"Expanded Parole Plan Might Ease Overcrowding," Denver Post, December 13, 1993, p. 12A.

(135)Henry Sontheimer and Lynne Goodstein, "An Evaluation of Juvenile Intensive Aftercare Probation: Aftercare versus System Response Effects," Justice Quarterly 10, no. 2 (June 1993): 197-227.

(136)Joan Petersilia and Susan Turner, "Evaluating Intensive Supervision Probation/Parole: Results of a Nationwide Experiment," RAND Corporation study for the National Institute of Justice, NCJ-141637, May 1993.

(137)"Ensuring Truth in Sentencing," p. 18.

(138)Petersilia and Turner, p. 7.

(139)Prisoners in 1992, p. 1.

(140)Petersilia and Turner, p. 9, citing Washington State Sentencing Guidelines Commission, Preliminary Evaluation of Washington State's Sentencing Reform Act (Olympia: Washing ton State Sentencing Guidelines Commission, 1983). See also Daniel Glaser and Ronald Watts, "Electronic Monitoring of Drug Offenders on Probation," Judicature, October-November 1992. "Incarceration and revocation of probation as reac tions to any positive results from drug tests for employed offenders seem not to be cost-effective public policies, but they are routinely recommended by some officers, whose advice many judges accept almost automatically" (p. 117).

In a national study of state prisoners whose probation or parole had been revoked, 74 percent of revocations were for arrest for or conviction of a new offense, 5 percent were for failing a drug test, and 4 percent were for failing to report for drug or alcohol treatment. Beck et al., Survey of State Prison Inmates, 1991, p. 14.

(141)James S. Gilmore III, "Keeping Convicts behind Bars," Washington Times, November 1, 1993.

(142)U.S. Department of Justice, Bureau of Justice Statis tics, Recidivism of Prisoners Released in 1983, NCJ-116261 (Washington: U.S. Department of Justice, April 1989), p. 3.

(143)U.S. Department of Justice, Bureau of Justice Statis tics, Examining Recidivism, NCJ-96501 (Washington: U.S. Department of Justice, February 1985), p. 4.

(144)U.S. Department of Justice, Bureau of Justice Statistics, Sourcebook of Criminal Justice Statistics 1989, ed. Timothy J. Flanagan and Kathleen Maguire (Washington: Government Printing Office, 1990).

(145)Keith A. Harrison, "Crimes Cast Suspicion on Justice System," Washington Post, October 6, 1993.

(146)"Parole System Defended," Washington Times, July 6, 1993, citing Virginia state parole board chair.

(147)"Cheaper than Prison," Virginia Pilot & Ledger Star, August 28, 1993, p. A10.

(148)Michael R. Geerken and Hennessey D. Hayes, "Probation and Parole: Public Risk and the Future of Incarceration Alternatives," Criminology 31 (1993): 549-64.

(149)The federal law is contained in Sentencing Guidelines, sec. 1B1.3 (relevant conduct). For more, see Kevin R. Reitz, "Sentencing Facts: Travesties of Real-Offense Sentencing," Stanford Law Review 45 (1993): 523, 531n. 49 (list of states allowing defendant to be sentenced for an offense of which he was acquitted).

(150)Reitz, p. 549. There are carefully prescribed circum stances at trial when certain kinds of hearsay or reputa tional evidence may be used; the use of hearsay and reputa tional evidence at sentencing is limited by none of the safeguards that apply during trial.

(151)United States v. Manor, 936 F.2d 1238 (11th Cir. 1991). The defendant was acquitted of distributing 250 grams of co caine and convicted of distributing 19 grams; the defendant was sentenced for 250 grams. For other cases of defendants' being sentenced for crimes of which they were acquitted, see Elizabeth T. Lear, "Is Conviction Irrelevant?" UCLA Law Review 40 (1993): 1179, 1182n. 7.

(152)Tracy Thompson, "Facts beyond Case at Hand Can Lengthen Prison Term," Washington Post, July 22, 1991, p. A1; and "You Make the Call," Criminal Justice Policy Foundation Newsbriefs, August 1991, p. 5.


(154)Lear, p. 1206. A study of sentencing in the Eighth Circuit found that half of all sentences were increased on the basis of uncharged conduct. George W. Hearney, "The Reality of Guidelines Sentencing: No End to Disparity," American Criminal Law Review 28 (1991): 161-232.

(155)Some courts, however, have suggested that in at least some cases, real-offense sentencing can violate the confron tation clause or the due process clause of the Constitution. United States v. Fortier, 911 F.2d 100 (8th Cir. 1990); and United States v. Roberts, 726 F. Supp. 1359 (D.D.C. 1989). Compare United States v. Kikumura, 918 F.2d 1084 (3d Cir. 1990) (rejecting confrontation clause challenge, but holding that due process standards require proof of uncharged con duct by "clear and convincing" evidence; reversing sentence imposed below; concurrence decries prosecutor's manipulation of defendant's charges in order to avoid having to prove more difficult offenses).

(156)Dennis Cauchon, "Dual Prosecution Can Give One Crime Two Punishments," USA Today, March 29, 1993, p. 10A.


(158)A limited restriction on dual prosecution is proposed in Daniel A. Braun, "Praying to False Sovereigns: The Rule Permitting Successive Prosecutions in an Age of Cooperative Federalism," American Journal of Criminal Law 20, no. 1 (1992) (proposal to bar dual prosecutions when the state and federal governments behave like a single government, as when there is a cooperative state and federal investigation).

(159)"Three Strikes' Backers Push for Ballot," Journal American, May 5, 1993; and "Washington Hits Repeated Crimes," (Broward County, Fla.) Sun-Sentinel, October 10, 1993.

(160)Kathy George, "Tough-on-Crime Measure Appears Headed for Ballot," Seattle Post Intelligence, June 30, 1993, p. A1.

(161)George, citing worst-case analysis by the state's Sentencing Guidelines Commission.

(162)"Washington Hits Repeated Crimes."

(163)"Crime Strikes Out," Wall Street Journal, May 13, 1993.


(165)"Washington Hits Repeated Crimes."

(166)Peter Lewis, "'Three Strikes' Initiative Gains Strength, Takes Aim at Felons," Seattle Times, August 31, 1993.

(167)Deal v. United States, 113 S.Ct. 1993 (1993) interprets the phrase "second or subsequent conviction" in 18 U.S.C. sec. 924(c)(1) to include two or more convictions arising in a single prosecution. The majority rejects the argument that recidivism statutes are intended to apply to persons who fail to reform after their first conviction.

Indeed, the "second or subsequent conviction" 20-year enhancement can even apply to a single crime that gives rise to two separate criminal offenses (such as possession of drugs with intent to distribute and conspiracy to distribute drugs). United States v. Parra, 2 F.3d 1058 (10th Cir.), cert. denied 62 U.S.L.W. 3409 (1993).

(168)John Carlson, "Taking the Initiative on Turnstile Justice," Seattle Times, May 11, 1993.

(169)M. Chaiken and J. Chaiken, "Offender Types and Public Policy," Crime and Delinquency 30 (1985): 195; and A. von Hirsch, Past or Future Crimes: Deservedness and Dangerous ness in the Sentencing of Criminals (New Brunswick, N.J.: Rutgers University Press, 1986).

(170)Alfred Blumstein, Jacqueline Cohen, Jeffrey A. Roth, and Christ A. Visher, eds., Criminal Careers and "Career Criminals" (1986), p. 6; cited in U.S. Department of Jus tice, Office of Policy Development, p. 8.

(171)Marvin Wolfgang, Robert Figlio, and Thorsten Sellin, Delinquency in a Birth Cohort (Chicago: University of Chicago Press, 1972).

(172)Alfred Blumstein, David P. Farrington, and Soumyo Moitra, "Delinquency Careers: Innocents, Desisters, and Persisters," Crime and Justice: An Annual Review of Re- search 6 (1985).

(173)Andrew von Hirsch, "Selective Incapacitation: Some Doubts," in Principled Sentencing, ed. Andrew von Hirsch and Andrew Ashworth (Boston: Northeastern University Press, 1992), pp. 158-64.

(174)A recent survey found that large numbers of Americans believed illegal drugs were easy to obtain and did not view occasional drug use as a big concern. Spencer Rich, "Survey Finds Little Concern about Occasional Drug Use," Washington Post, April 10, 1994, p. A3. See also Joseph B. Treaster, "Drug Use Making Comeback, Study Says," New York Times, July 16, 1993, p. D12; and Joseph B. Treaster, "20 Years of War on Drugs, and No Victory Yet," New York Times, June 14, 1992, sec. 4, p. 7.

(175)Quoted in Booth, p. A1.

1994 The Cato Institute
Please send comments to webmaster