buy off foreign and domestic officials and to secure protection for the most vulnerable stages of the drug-cultivation, -manufacturing, and -distribution process. These profits also hire various specialists, from assassins to money launderers to lawyers, needed to cope with interlopers; they pay for technological devices ranging from cellular phones to jet planes; and they ensure that should a trafficker die or land in jail, there will be no shortage of replacements.

It is hardly surprising that these stubborn economic realities, together with the drug war’s uneven and often disappointing results, have led several commentators to question the wisdom of what they call the prohibition policy. What is unprecedented is that these disenchanted critics include mayors, prominent lawyers, federal judges, nationally syndicated columnists, a congressman, a Princeton professor, and a Nobel laureate in economics. They espouse variations of a position that is often called controlled legalization, meaning that the sale of narcotics should be permitted under conditions that restrict and limit consumption, such as no sales to minors, no advertising, and substantial taxation. They cite the numerous advantages of this approach: several billion dollars per year would be realized from tax revenues and savings on law enforcement; crime would diminish because addicts would not have to hustle to keep themselves supplied with drugs; the murders associated with big-city drug trafficking would abate as lower-cost, legal drugs drive the traffickers out of business. Because these drugs would be of known quality and potency, and because they would not have to be injected with shared needles, the risk of overdose and infection would drop. The issue of foreign complicity in the drug traffic, which has complicated American diplomatic relations with many countries, would disappear. Under a policy of controlled legalization, it would be no more criminal or controversial to import coca from Colombia than to import coffee.

The more candid of the legalization proponents concede that these advantages would be purchased at the cost of increased drug abuse. Widespread availability, lower prices, and the elimination of the criminal sanction would result in more users, some of whom would inevitably become addicts. But how many more? Herbert Kleber, a treatment specialist and former deputy director of the Office of National Drug Control Policy, has argued that there would be between twelve and fifty-five million addicted users if cocaine and heroin were legally available. While it is impossible to anticipate the exact magnitude of the increase, history does support Kleber’s argument. In countries like Iran or Thailand, where narcotics have long been cheap, potent, and readily available, the prevalence of addiction has been and continues to be quite high. Large quantities of opium sold by British and American merchants created a social disaster in nineteenth-century China; that Chinese sailors and immigrants subsequently introduced opium smoking to Britain and America is a kind of ironic justice. Doctors, who constantly work with

and around narcotics, have historically had a very serious addiction problem: estimates of the extent of morphine addiction among American physicians at the turn of the century ran from 6 percent to an astonishing 23 percent. In a word, exposure matters.

Kleber has also attacked the crime-reduction rationale by pointing out that addicts will generally use much more of an illicit substance if the cost is low. They would spend most of their time using drugs and little of it working, thus continuing to resort to crime to acquire money. If the total number of addicts rose sharply as availability increased, total crime would also increase. There would be less crime committed by any single addict but more crime in the aggregate.

The debate over decriminalization is, in essence, an argument about a high-stakes gamble, and so far the opponents represent the majority view. At the close of the 1980s, four out of every five Americans were against the legalization of marijuana, let alone cocaine. But if the drug war produces another decade of indifferent results, growing disillusionment could conceivably prompt experiments in controlled legalization.

The controlled-legalization argument rests on the assumption that legal sales would largely eliminate the illicit traffic and its attendant evils. The history of drug use, regulation, and taxation in the United States suggests otherwise. The very phrase controlled legalization implies denying certain groups access to drugs. Minors are the most obvious example. No one advocates supplying narcotics to children, so presumably selling drugs to anyone under twenty-one would remain a criminal offense, since that is the cutoff point for sales of beverage alcohol. Unfortunately, illicit drug abuse in this century has become concentrated among the young—that is, among the very ones most likely to be made exceptions to the rule of legal sales.

Until about 1900 the most common pattern of drug dependence in the United States was opium or morphine addiction, brought about by the treatment of chronic diseases and painful symptoms. Addicts were mainly female, middle-class, and middle-aged or older; Eugene O’Neill’s mother, fictionalized as Mary Tyrone in Long Day’s Journey into Night , was one. Habitual users of morphine, laudanum, and other medicinal opiates in their adolescence were extremely rare, even in big cities like Chicago.

Another pattern of drug use was nonmedical and had its roots in marginal, deviant, and criminal subcultures. The “pleasure users,” as they were sometimes called, smoked opium, sniffed cocaine, injected morphine and cocaine in combination, or, after 1910, sniffed or injected heroin. Non-medical addicts began much younger than their medical counterparts. The average age of addiction (not first use, which would have been lower still) for urban heroin addicts studied in the 1910s was only nineteen or twenty years. They were also more likely to be male than those whose addiction was of medical origin, and more likely to have been involved in crime.

Initially the pleasure users were the smaller group, but

during the first two decades of this century—the same period when the police approach to national drug control was formulated—the number of older, docile medical addicts steadily diminished. There were several reasons: doctors became better educated and more conservative in their use of narcotics; the population grew healthier; patent-medicine manufacturers were forced to reveal the contents of their products; and the numerous morphine addicts who had been created in the nineteenth century began to age and die off. Drug use and addiction became increasingly concentrated among young men in their teens and twenties, a pattern that continues to this day.

In 1980, 44 percent of drug arrests nationwide were of persons under the age of twenty-one. There were more arrests among teen-agers than among the entire population over the age of twenty-five; eighteen-year-olds had the highest arrest rate of any age group. By 1987 the proportion of those arrested under twenty-one had declined to 25 percent. This was partly due to the aging of the population and to the effects of drug education on students. But when large numbers of “echo boomers”—the children of the baby boomers—become adolescents during the 1990s, the percentage of under-twenty-one drug arrests will likely increase.

So, depending on timing and demographic circumstances, at least a quarter and perhaps more than a third of all drug buyers would be underage, and there would be a great deal of money to be made by selling to them. The primary source of supply would likely be diversion—adults legally purchasing drugs and selling them to customers below the legal age. The sellers (or middlemen who collected and then resold the legal purchases) would make a profit through marking up or adulterating the drugs, and there might well be turf disputes and hence violence. Some of the dealers and their underage purchasers would be caught, prosecuted, and jailed, and the criminal-justice system would still be burdened with drug arrests. The black market would be altered and diminished, but it would scarcely disappear.

Potential for illegal sales and use extends far beyond minors. Pilots, police officers, fire fighters, drivers of buses, trains, taxis, and ambulances, surgeons, active-duty military personnel, and others whose drug use would jeopardize public safety would be denied access to at least some drugs, and those of them who did take narcotics would be liable to criminal prosecution, as would their suppliers. Pregnant women would also pose a problem. Drugs transmitted to fetuses can cause irreversible and enormously costly harm. Federal and local governments may soon be spending billions of dollars a year just to prepare the impaired children of addicts for kindergarten. Society has the right and the obligation to stop this neurological carnage, both because it cruelly handicaps innocents and because it harms everyone else through higher taxes and health-insurance premiums. Paradoxically, the arguments for controlled legalization might lead to denying alcohol and tobacco to pregnant women along with narcotics. Alcohol and tobacco can also harm fetal development,

and several legalization proponents have observed that it is both inconsistent and unwise to treat them as if they were not dangerous because they are legal. If cocaine is denied to pregnant women, why not alcohol too? The point here is simply that every time one makes an exception for good and compelling reasons—every time one accents the “controlled” as opposed to the “legalization” —one creates the likelihood of continued illicit sales and use.

The supposition that this illegal market would be fueled by diversion is well founded historically. There has always been an undercurrent of diversion, especially in the late 1910s and 1920s, when black-market operators like Legs Diamond got their supplies not so much by smuggling as by purchases from legitimate drug companies. One possible solution is to require of all legal purchasers that which is required of newly enrolled methadone patients: consumption of the drug on the premises. Unfortunately, unlike methadone, heroin and cocaine are short-acting, and compulsive users must administer them every few hours or less. The dayrooms of drug-treatment clinics set up in Britain after 1968 to provide heroin maintenance were often clogged with whining addicts. Frustrated and angry, the clinic staffs largely abandoned heroin during the 1970s, switching instead to methadone, which, having the advantages of oral administration and twenty-four-hour duration, is far more suitable for clinic-based distribution. Confining the use of heroin or cocaine or other street drugs to clinics would be a logistical nightmare. But the alternative, take-home supplies, invites illegal sales to excluded groups.

Another historical pattern of black-market activity has been the smuggling of drugs to prisoners. Contraband was one of the reasons the government built specialized narcotic hospitals in Lexington, Kentucky, and Fort Worth, Texas, in the 1930s. Federal wardens wanted to get addicts out of their prisons because they were constantly conniving to obtain smuggled drugs. But when drug-related arrests multiplied after 1965 and the Lexington and Fort Worth facilities were closed, the prisons again filled with inmates eager to obtain drugs. Birch Bayh, chairing a Senate investigation of the matter in 1975, observed that in some institutions young offenders had a more plentiful supply of drugs than they did on the outside.

Since then more jails have been crammed with more prisoners, and these prisoners are more likely than ever to have had a history of drug use. In 1989, 60 to 80 percent of male arrestees in twelve large American cities tested positive for drugs. It is hard to imagine a controlled-legalization system that would permit sales to prisoners. Alcohol, although a legal drug, is not sold licitly in prisons, and for good reason, as more than 40 percent of prisoners were under its influence when they committed their crimes. If drugs are similarly denied to inmates, then the contraband problem will persist. If, moreover, we insist that our nearly three million parolees and probationers remain clean on the theory that drug use aggravates recidivism, the market for illegal sales would be

so much the larger. By now the problem should be clear. If drugs are legalized, but not for those under twenty-one, or for public-safety officers, or transport workers, or military personnel, or pregnant women, or prisoners, or probationers, or parolees, or psychotics, or any of several other special groups one could plausibly name, then just exactly who is going to buy them? Noncriminal adults, whose drug use is comparatively low to begin with? Controlled legalization entails a dilemma. To the extent that its controls are enforced, some form of black-market activity will persist. If, on the other hand, its controls are not enforced and drugs are easily diverted to those who are underage or otherwise ineligible, then it is a disguised form of wholesale legalization and as such morally, politically, and economically unacceptable.

One of the selling points of controlled legalization was also one of the decisive arguments for the repeal of Prohibition: taxation. Instead of spending billions to suppress the illicit traffic, the government would reap billions by imposing duties on legitimate imports and taxes on domestically manufactured drugs. Not only could these revenues be earmarked for drug treatment and education programs, but they would also increase the prices paid by the consumer, thus discouraging consumption, especially among adolescents.

The United States government has had extensive historical experience with the taxation of legal narcotics. In the nineteenth and early twentieth centuries, opium was imported and subject to customs duties. The imports were assigned to one of three categories. The first was crude opium, used mainly for medicinal purposes and for the domestic manufacture of morphine. Foreign-manufactured morphine, codeine, and heroin made up the second class of imports, while the third was smoking opium, most of it prepared in Hong Kong and shipped to San Francisco.

The imposts on these imported drugs fluctuated over the years, but they were generally quite stiff. From 1866 to 1914 the average ad valorem duty on crude opium was 33 percent; for morphine or its salts, 48 percent. From 1866 to 1908 the average duty on smoking opium was an extraordinarily high 97 percent. This last was in the nature of a sin tax; congressmen identified opium smoking with Chinese coolies, gamblers, pimps, and prostitutes and wished to discourage its importation and use.

These customs duties produced revenue; they also produced widespread smuggling, much of it organized by violent criminal societies like the Chinese tongs. The smugglers were as ingenious as their latter-day Mafia counterparts. They hid their shipments in everything from hollowed-out lumber to snake cages. Avoiding the customs collectors, they saved as much as three dollars a pound on crude opium, three dollars an ounce on morphine, and twelve dollars a pound on smoking opium. Twelve dollars seems a trifling sum by modern standards, hardly worth the risk of arrest, but in the nineteenth century

it was more than most workers earned in a week. Someone who smuggled in fifty pounds of smoking opium in 1895 had gained the equivalent of a year’s wages. One knowledgeable authority estimated that when the duty on smoking opium was near its peak, the amount smuggled into the United States was nearly twice that legally imported and taxed. Something similar happened with eighteenth-century tobacco imports to the British Isles. More than a third of the tobacco consumed in England and Scotland circa 1750 had been clandestinely imported in order to avoid a duty of more than five pence per pound. The principle is the same for domestically produced drugs: If taxes are sufficiently onerous, an illegal supply system will spring up. Moonshining existed before and after, as well as during, Prohibition.

The obvious solution is to set taxes at a sufficiently low level to discourage smuggling and illegal manufacturing. But again there is a dilemma. The most important illicit drugs are processed agricultural products that can be grown in several parts of the world by peasant labor. They are not, in other words, intrinsically expensive. Unless they are heavily taxed, legal consumers will be able to acquire them at little cost, less than ten dollars for a gram of cocaine. If drugs are that cheap, to say nothing of being 100 percent pure, the likelihood of a postlegalization epidemic of addiction will be substantially increased. But if taxes are given a stiff boost to enhance revenues and limit consumption, black marketeers will reenter the picture in numbers proportionate to the severity of the tax.

Tax revenues, like drugs themselves, can be addictive. In the twelve years after the repeal of Prohibition, federal liquor tax revenues ballooned from 259 million to 2.3 billion dollars. The government’s dependence on this money was one important reason anti-liquor forces made so little progress in their attempts to restrict alcohol consumption during World War II. Controlled drug legalization would also bring about a windfall in tax dollars, which in an era of chronic deficits would surely be welcomed and quickly spent. Should addiction rates become too high, a conflict between public health and revenue concerns would inevitably ensue.

When both proponents and opponents of controlled legalization talk about drug taxes, they generally assume a single level of taxation. The assumption is wrong. The nature of the federal system permits state and local governments to levy their own taxes on drugs in addition to the uniform federal customs and excise taxes. This means that total drug taxes, and hence the prices paid by consumers, will vary from place to place. Variation invites interstate smuggling, and if the variation is large enough, the smuggling can be extensive and involve organized crime.

The history of cigarette taxation serves to illustrate this principle. In 1960 state taxes on cigarettes were low, between zero and eight cents per pack, but after 1965 a growing number of states sharply increased cigarette

taxes in response to health concerns and as a politically painless way of increasing revenue. Some states, mainly in the Northeast, were considerably more aggressive than others in raising taxes. By 1975 North Carolina purchasers were paying thirty-six cents per pack while New Yorkers paid fifty-four cents. The price was higher still in New York City because of a local levy that reached eight cents per pack (as much as the entire federal tax) at the beginning of 1976.

Thus was born an opportunity to buy cheap and sell dear. Those who bought in volume at North Carolina prices and sold at New York (or Connecticut, or Massachusetts) prices realized a substantial profit, and by the mid-1970s net revenue losses stood at well over three hundred million dollars a year. Much of this went to organized crime, which at one point was bootlegging 25 percent of the cigarettes sold in New York State and half of those sold in New York City. The pioneer of the illegal traffic, Anthony Granata, established a trucking company with thirty employees operating vehicles on a six-days-a-week basis. Granata’s methods—concealed cargoes, dummy corporations, forged documents, fortress-like warehouses, bribery, hijacking, assault, and homicide—were strikingly similar to those used by illicit drug traffickers and Prohibition bootleggers.

Although high-tax states like Florida or Illinois still lose millions annually to cigarette bootleggers, the 1978 federal Contraband Cigarette Act and stricter law enforcement and accounting procedures have had some success in reducing over-the-road smuggling. But it is relatively easy to detect illegal shipments of cigarettes, which must be smuggled by the truckload to make a substantial amount of money. Cocaine and heroin are more compact, more profitable, and very easy to conceal. Smuggling these drugs to take advantage of state tax differentials would consequently be much more difficult to detect and deter. If, for example, taxed cocaine retailed in Vermont for ten dollars a gram and in New York for twelve dollars a gram, anyone who bought just five kilograms at Vermont prices, transported them, and sold them at New York prices would realize a profit of ten thousand dollars. Five kilograms of cocaine can be concealed in an attaché case.

Of course, if all states legalized drugs and taxed them at the same rate, this sort of illegal activity would not exist, but it is constitutionally and politically unfeasible to ensure uniform rates of state taxation. And federalism poses other challenges. Laws against drug use and trafficking have been enacted at the local, state, and federal levels. It is probable that if Congress repeals or modifies the national drug laws, some states will go along with controlled legalization while others will not. Nevada, long in the legalizing habit, might jettison its drug laws, but conservative Mormon-populated Utah might not. Alternately, governments could experiment with varying degrees of legalization. Congress might decide that anything was better than the current mayhem in the capital and legislate a broad legalization program

for the District of Columbia. At the same time, Virginia and Maryland might experiment with the decriminalization of marijuana, the least risky legalization option, but retain prohibition of the nonmedical use of other drugs. The result would again be smuggling, whether from Nevada to Utah or, save for marijuana, from the District of Columbia to the surrounding states. It is hard to see how any state that chose to retain laws against drugs could possibly stanch the influx of prohibited drugs from adjacent states that did not. New York City’s futile attempts to enforce its strict gun-control laws show how difficult it is to restrict locally that which is elsewhere freely available.

I referred earlier to the legalization debate as an argument about a colossal gamble, whether society should risk an unknown increase in drug abuse and addiction to eliminate the harms of drug prohibition, most of which stem from illicit trafficking. “Take the crime out of it” is the rallying cry of the legalization advocates. After reviewing the larger history of narcotic, alcohol, and tobacco use and regulation, it appears that this debate should be recast. It would be more accurate to ask whether society should risk an unknown but possibly substantial increase in drug abuse and addiction in order to bring about an unknown reduction in illicit trafficking and other costs of drug prohibition. Controlled legalization would take some, but by no means all, of the crime out of it. Just how much and what sort of crime would be eliminated would depend upon which groups were to be denied which drugs, the overall level of taxation, and differences in state tax and legalization policies. If the excluded groups were few and all states legalized all drugs and all governments taxed at uniformly low levels, then the black market would be largely eliminated. But these are precisely the conditions that would be most likely to bring about an unacceptably high level of drug abuse. The same variables that would determine how successful the controlled-legalization policy would be in eliminating the black market would also largely determine how unsuccessful it was in containing drug addiction.