Assume a modern society not troubled by Mill's libertarian scruples, but also not tom by battles for cultural dominance or paralyzed by historically fixed definitions. What would be the most sensible ways for it to deal with the use of drugs? Obviously, the range of solutions is wide, and any society would have to use a mixture, but they can be assigned generally to two categories: informal prescriptions and sanctions and formal (mostly legal) ones. The distinction between informal and legal sanctions is not equivalent to the distinction between consensus and coercion. Even Mill included among threats to liberty the informal tyranny of majority opinion as well as formal acts of government. The law differs from social pressures not in coerciveness but in its use of explicit rules and penalties. Societies have, differed vastly in the extent to which they manage drug use by legal compulsion as opposed to education, persuasion, and social pressure to act in accordance with accepted customs and values. A society like ours, complicated in organization and culturally diverse, with many strangers encountering one another or affecting one another's lives, is more likely to want and need formal rules (Turk 1972).

Where alcohol is concerned, there is still some serious argument about the virtues of legal versus informal controls, expressed in a debate between advocates of the integration or sociocultural model of alcohol control and those who favor the distribution of consumption model. The integration model proposes that the seriousness of a society's alcohol problem is independent of the total amount of alcohol consumed. Instead it depends on the informal arrangements and sanctions surrounding alcohol use: the occasions on which people use it, how children are taught about it and so on. Alcohol should be properly integrated into daily life, used at family gatherings and at meals. Drinking should not become a center of attention; it should usually be subordinated to another activity that goes on at the
same time. Both prescriptions and proscriptions for drinking should be clear. Drinking alone and drunken sprees should be discouraged by custom. Abstinence should be socially acceptable, but drinking should not be material for moral enterprise. Feelings about alcohol should not oscillate between puritanism and license. Drinking lore should be transmitted from generation to generation; teachers and learners should not be the same age. If these intermediate social controls can be established, we will not need either state intervention or desperate appeals for individual restraint. Certain ethnic groups — Jews, Chinese, Greeks — are held up as model integrated drinkers with few alcohol problems.

Integration theorists have recommended such policies as advertising that shows families drinking together and education for responsible drinking. They have also proposed mild legal measures to help shape custom: taxes to make wine and beer relatively less expensive than liquor, licensing laws that discourage the sale of alcohol where no food is sold, suspending bartenders' licenses for serving drunks, and even a lowered drinking age, supposedly to prevent furtive and uncontrolled adolescent drinking (Wilkinson 1970). A few daring writers have also suggested integrated use as an approach to the control of illicit drugs (Harding and Zinberg 1977; Zinberg and Harding 1982).

Some theory and evidence support the integration model. For example, alcohol tends to narrow the range of attention and to impoverish conceptual thinking, so that the drinker notices a smaller variety of cues aid may be more likely to respond immoderately or violently. A well-established drinking etiquette means fewer indeterminate cues, a secure cognitive framework for central environmental cues, and enough external cues to prevent chaotic impulses from dominating action (Pernaen 1976). Certain family attitudes toward alcohol seem to be related to later drinking problems among the children — especially situations in which one parent, usually the mother, is an abstainer and hostile to alcohol, and the other parent drinks too much. A smaller proportion of alcoholics than control subjects have parents who do not drink at all, but a larger proportion have parents who disagree about drinking (Wilkinson 1970).

But there are serious objections to the integration model. Certainly the example of France puts the virtues of integrated drinking in doubt. Drinking is highly integrated there; wine is drunk at meals, in most festival and ritual social situations, and as an accompaniment to many daily activities. Children learn to use it from their parents. Yet France has what is possibly the highest rate of chronic alcohol problems in the world and a high rate of acute alcohol abuse. In the case of parents who disagree about drinking and whose children grow up to be alcoholics, it is possible that the parents are in conflict mainly because one of them (usually the father) is an alcohol abuser, who passes on the tendency to his children either by heredity or by example. Even if these empirical difficulties can be evaded, there is likely to be a conceptual one. Unless you are very precise about what integrated drinking means, it is too easy to define it in such a way that it becomes incompatible with alcohol abuse. If integrated drinking means no more than moderate and disciplined drinking, that kind of alcohol use causes few problems by definition.

But the main objection to the integration or sociocultural model is that it suggests no policies that seem likely to be effective. The model has a certain common sense appeal; some ethnic and social groups or cultures manage to discipline their drinking and drug use without total abstinence. But this moderation is almost always the product of a long history, not current injunctions and instructions. It is doubtful how far drinking and drug-taking practices can be manipulated as a conscious policy. There is no simple way to identify how some group does it and to get others to emulate them. The usual suggestions for advertising, drug education in schools, encouraging the use of beer and wine instead of liquor, and so on appear to be very light ammunition against established customs that are deeply embedded in the life of a society. And in fact, they usually prove ineffective. The nineteenth-century temperance movement had to give up believing that beer is better than whiskey. Studies of drug and alcohol education generally fail to show that it changes anyone's behavior very much (Abrams et al. 1973). It is not misleading to say that most people who could profit from a formal drug education course do not need it, because they would learn how to control their drug use anyway.

Current education about illicit drugs is limited in effectiveness because whatever facts are taught, the message ultimately has to be never to use the drug at all. In some cases, the facts and the warning may be hard to reconcile, so that responsible, integrated drug use seems plausible by comparison. But deliberate education for responsible alcohol use has not been a great success, and it is doubtful whether the proposed reform of drug education (even if it were politically conceivable) would help much either. One kind of policy that sometimes works is a religious injunction; Mormons and Moslems have few alcohol problems. Maybe the integration model too would work best in a theocracy, but that is not what its advocates have in mind.

The greatest service of studies on the integration model has been to make it clear how illicit drug users establish "sanctions and rituals" (as Norman Zinberg calls them) that keep ill effects to a minimum. If advocates of the integration model for alcohol were consistent, they would have to recommend it for other drugs too. Except that they are established outside the law, informal social controls on illicit drug use are in principle no different from similar controls on drinking. They vary according to the properties of the drug; they include instructions for using the drug, sumptuary rules, social relations that encourage certain ways of using it, and informal sanctions enforcing the rules (Maloff et al. 1982). In a way, this should be obvious, but the reminder is necessary, since so many people have an image of illicit drug use as an uncontrolled epidemic of disease or criminality. The integration model can teach us to make necessary distinctions and not respond hysterically to all illicit drug use, but it does not tell us much about what to do when drugs and alcohol are abused.

Supporters of the distribution of consumption model believe that alcohol problems vary with the total amount of alcohol consumed in a society, and suggest that the best way to reduce alcohol abuse is to restrict the supply legally through pricing, taxing, and licensing (Popham et al. 1976). In particular, they say that the distribution of alcohol consumption in any population is unimodal (has a single peak) and, with proper mathematical adjustments, can be made to approximate a normal curve. Height, IQ, and many other features of a large population vary in a normal distribution. The shape of a normal curve is symmetrical; it depends only on the mean, which gives its height, and the standard deviation, which defines its spread. In general, whenever values of a variable differ from their mean because of many independently acting influences that sum to produce their effects, the resulting probability distribution is close to a normal curve.

The curve describing the consumption of alcohol by individuals in a population is actually not symmetrical but highly skewed to the right. This means that most people drink only a little alcohol and a few elevate the mean by drinking a great deal. In the United States, for example, 10 percent of the population uses 57 percent of the alcohol, and 20 percent uses 78 percent of the alcohol (Gerstein 1981). This kind of curve often appears when independent random variables combine as a product rather ' than as a sum to produce an effect. It is then known as a lognormal distribution. This means that the curve, which is highly skewed when plotted using the amount of alcohol consumed by each individual in the population, will be close to a normal curve when plotted using the logarithms of those amounts.

According to this model, the range of individual variation is more or less fixed (zero on one side, the limit of physical toleration on the other), and the shape of the curve depends only on the mean in a given population. If the average person drinks somewhat less, the heaviest users will drink a great deal less. That is true of any normal distribution; if the mean height in two populations is slightly different, the proportions of people more than six feet tall will differ enormously. The conclusion is that policies designed to cut total alcohol consumption will greatly reduce alcohol problems. There is no need to be selective or try to educate the populace. The most effective methods are restricted hours of sale and, above all, increased taxes.

A universal lognormal distribution would refute the integration model, which requires differences in the shape of the consumption curve in different societies. The number of heavy alcohol users or abusers should not necessarily be determined by the same forces that determine how much the average person drinks. Appropriate customs are supposed to ensure that almost everyone drinks moderately and almost no one drinks too much or not at all. It should be possible to find two societies in which the one with the greater average alcohol consumption has less serious drinking problems. The empirical and statistical assumptions of the distribution of consumption model have therefore been strongly challenged, and the challenge has engendered a complicated dispute about consumption measures, sampling, and \statistical theory (Room 1978; Pittman 1980; Beauchamp 1980, pp. 102-11; Hunt 1982).

Whatever the precise shape of the consumption curves, there is strong evidence that raising the relative price of a drug or restricting access to it cuts down its use. Heroin addicts apparently adjust the size of their habit 0 the price of heroin; it is not true that they must have a certain daily dose at all costs (Goldman 1981). And one of the main reasons they stop taking heroin is that the drug becomes too hard to get (Vaillant 1970). England imposed a tax and sales control on gin in 1751 and reduced consumption from eleven million to two million gallons in a few years (Coffey 1966). When Finland liberalized its laws on beer in 1968-9, lowering the drinking age and allowing more outlets, alcohol use increased 46 percent per capita in the first year, and arrests for drunkenness rose substantially (Beauchamp 1980). As we have mentioned, alcohol consumption in 1918-23, at the start of Prohibition, is estimated as one-third of the 1913 average, and in 1927-30 as two-thirds of the 1913 average. Today the level of state taxes on alcohol is correlated with total consumption (Cook 1981). The price of alcohol is relatively low in France, where consumption is very high, and relatively high in Ireland, where (despite myths about Irish drinking) consumption is low. In the United States, the real cost of alcohol dropped substantially from 1960 to 1980 (by 48 percent for liquor, 27 percent for beer, and 18 percent for wine), and consumption rose greatly (Moore and Gerstein 1981, p. 68).

It also seems to be true that where the population as a whole uses less alcohol, there are fewer serious alcohol problems. The rate of cirrhosis of the liver in the United States was 21.5 per million in 1911 and 10.7 per million in 1929 (Aaron and Musto 1981). Between 1969 and 1979, consumption of wine and spirits doubled in Great Britain; so did the rate of alcoholism (Royal College 1979, p. 134). In American states where alcohol taxes are high, there are fewer drunken driving accidents as well as less cirrhosis (Cook 1981). During wartime, when alcohol is rationed, cirrhosis rates drop — for example, in occupied Paris during World War II (Kalant 1981). Countries and peoples held up as models of integrated drinking do not have a low level of alcohol problems unless the drinking level is also relatively low. Italy is often contrasted favorably to France for its sensible drinking customs, and it is true that conspicuous drunkenness and the homeless alcoholic are less common there; but Italy also has a high average consumption of alcohol, close to that of France, and a high level of serious chronic health problems, including cirrhosis. Israel, on the other hand, has few alcohol problems, not just because of integrated drinking but because its average alcohol consumption is the lowest in the industrialized West. Scandinavians, despite their reputation for lone drinking and bingeing, which integration theorists abhor, use little alcohol compared with most other Europeans and have relatively mild acute and chronic alcohol problems. Even if the relationship between per capita consumption and alcohol problems does not follow exactly the pattern suggested by the distribution of consumption model, there is much evidence that the relationship exists.

Does cutting down the supply of a drug simply cause its users to replace it with another drug? If marihuana smokers turn to alcohol or heroin when the price becomes higher, it might be better not to .allow that to happen. The evidence is not entirely clear, but very commonly, at any rate, one drug is added to another rather than substituted for it, at least up to the limit of physiological toleration. In one study, patients in a drug treatment program tended to use more alcohol when they were using illicit drugs less often, but the difference was not statistically significant (Simpson and Lloyd 1981). In an experiment in which subjects were supplied with marihuana alone, alcohol alone, or marihuana and alcohol, they used more of both when both were available (Mello and Mendelson 1978). A study of two villages in Pakistan, one in which half of the population used opium and one in which only 5 percent used it, showed that alcohol and marihuana use was about the same in both (McGlothlin 1975).

Another reason for taking legal controls on alcohol seriously is that if raising prices and restricting access do not lower the average consumption of a drug, or if lowering the average consumption does not reduce its ill effects, then the twentieth-century regulatory system for most recreational drugs makes no sense at all. This system does not distinguish between moderate and excessive use or integrated and unintegrated customs; it simply aims to keep everyone from using the drugs, and for that we are willing to pay a substantial price. Serious defenders of current heroin policy admit that it actually makes life worse for addicts, but they contend that heroin must be as expensive and inaccessible as possible to keep down the number of new users and therefore the number of addicts (Moore 1977). Even in the rare cases in which a respectable public body proposes legalization (as opposed to decriminalizing possession) of an illicit drug, it assumes that legalization will lower the price, increase the availability, and therefore increase the total consumption and the problems created by abuse of the drug; the only question is whether the price paid for the current laws is too high (National Research Council 1982).

The evidence for legal restraints has nevertheless been challenged. The gin law passed in England in 1751 reduced alcohol problems, but a harsher law passed in 1736 proved ineffective and led to riots. Was the earlier law too harsh? Was it the spread of coffee, tea, and Methodism that made the English poor more sober? During our Prohibition years, alcohol consumption dropped not only in the United States but throughout the industrial world. As so often happens in social science, the problem is how to distinguish cause from effect when there is no accepted theoretical framework. Legal restraints on alcohol may be just a symptom of a popular trend toçvard disapproval of alcohol use; liberalization of laws or lowering of prices may be a byproduct of a change in attitudes that makes people tolerate more alcohol use: stateways following folkways. Apparently there are no prospective studies of a geographical region before and after a new alcohol restraint or pricing system is introduced, with a control region for comparison (Smart 1976). And no doubt even if there were, reasons could be given to explain why the control region was not precisely comparable to the experimental region.

There are other related problems. If use of a drug is well enough entrenched in a society, the people will simply not tolerate prices that are too high. High legal prices and restricted legal availability may defeat their purpose by creating a vast black market for that drug or increasing the interest in another one. Some people think that we reached that point with alcohol during Prohibition and are reaching it with marihuana now. Besides, a drug may create such a strong dependency that demand becomes inelastic and raising the price has only a limited effect. This is often said about heroin, but it is more clearly true of tobacco. When James I of England put a confiscatory customs duty of 4,000 percent on tobacco to keep it out of the country, smuggling became so common and lucrative that a few years later he had to reduce the duty and use it for revenue s instead of prohibition.

Laws work best in the short term, and especially when the drug problem is concentrated in a small sector of the society. New relaxations or tightenings of alcohol laws often have their greatest effect in the first few months. Japan in the 1950s and Sweden in the 1960s dealt with serious but limited amphetamine problems partly by cutting off the legal sources of supply. India dealt similarly with opium addiction in the 1950s and 1960s. Great Britain kept a small heroin problem from growing by instituting a clinic system in 1968 to replace prescriptions by individual doctors. In all of these cases, the drug users were a small, socially marginal group, often seriously disturbed or criminal; the legal change had no effect on the habits of the majority or even a substantial minority. And because there was no established illicit mass distribution system, the problem could be eliminated fairly quickly, without a long-drawn-out struggle.

A natural suggestion is to combine laws that reduce the total supply with efforts to promote integrated drinking or drug use. Some studies of preliterate societies suggest that the total consumption of alcohol is correlated highly but not perfectly with drinking problems; the rules for alcohol use also make a difference (Frankel and Whitehead 1979). The law can be used to reinforce socially accepted values and make marginal people conform, as in the Swedish and Japanese treatment of the amphetamine problem. It helps put temptation out of the way. Restrictions on a drug's availability may help to promote safer and healthier ways of using it. Illicit drug users on the verge of serious abuse sometimes approve ot the law, even when they intend to go on disobeying it. In the same way, a posted speed limit may keep drivers moving more slowly even when they do not obey it.

One interesting combination of sociocultural and legal controls is the local option system. Each small, culturally homogeneous geographical area makes its own laws on alcohol sales, so that legal restrictions are reinforced by social attitudes. In the political terms discussed in Chapter 1, an illicit drug user who wants his or her favorite drug to remain illegal is in the position of Ulysses ordering his men to protect him from his own attraction to the sirens. A small town that bans alcohol sales is close to the position of workers who have collectively agreed to a contract that forbids any one of them to accept longer hours or lower wages than those agreed on by the union (as opposed to workers obeying a minimum wage and maximum hours law for which they have not contracted).

Pricing policies might be regarded as a way of making people buy insurance for the risk to themselves and others in their misuse of alcohol or drugs. With taxes on alcohol, we pay to solve the problems created by alcohol abuse. But this idea has serious difficulties. It goes without saying that existing taxes on alcohol are not nearly high enough to pay for the costs of alcohol abuse. But more important in principle is the fact that the price of this insurance is not paid by the right people. Alcohol abusers pay more in taxes than moderate drinkers, but not nearly enough to cover the extra expense they create. Mill complained that this kind of restraint makes everyone suffer for the intemperance of a few. In particular, the cost, if it is a cost, falls on people who have less income. Any law that makes a commodity relatively more expensive obviously reduces the amount used by the poor more than the amount used by the rich.

If alcohol is an evil, this does the poor a favor; so prohibitionists sometimes argued. But as we saw in our historical survey, the motives of paternalism are always open to question. In the seventeenth-century debate on tobacco or the eighteenth-century English debate on gin, it was quite acceptable to talk about keeping those drugs out of the hands of the dangerous classes — dangerous to themselves and others. Since then we have become more democratic; an open appeal to class fears is out of tfie question, and even paternalistic attitudes toward the poor are not in favor. But the effect of high prices remains the same.

The idea of reducing alcohol problems by reducing per capita consumption is less plausible if the pathological condition of alcohol abusers makes them immune to legal measures that might change the habits of the average drinkei. So, some advocates of the disease theory of alcoholism have been especially hostile to the distribution of consumption model, with its implication that there will be fewer abusive drinkers if the average drinker uses less alcohol. It is significant that the disease theory in its modern form was first successfully promulgated after Prohibition had failed and the effort to regulate drinking by law was in bad repute. This is not to say that the distribution of consumption model is logically incompatible with the disease theory. Whatever the definition of alcoholism, not all alcohol problems are alcoholism problems. In any case, lack of easy availability might prevent some susceptible people from becoming alcoholics, just as heroin law enforcement is supposed to prevent susceptible people from becoming addicts. But obviously, to vary Mill's language, it would be better not to make everyone suffer because of the illness of a few; and that is a more serious danger if diseased and healthy alcohol consumers represent two different populations rather than different points on a single unimodal distribution curve. Those who insist on total abstinence as the only cure for alcoholics are most likely to reject the distribution of consumption model, because it suggests that even abusive drinkers can improve merely by drinking less. Conceiving of alcoholism as a single disease fits the distribution of consumption model less well than conceiving of it as a mixture of alcohol abuse problems that have numerous sources and come in numerous combinations.

Pricing, licensing, and restrictions on hours of sale are accepted as ways to keep alcohol consumption down. But the gulf between alcohol and other abusable drugs is enormous when it comes to a special kind of legal restraint, the criminal law. Only a few drinking practices are criminal — mainly drunk driving and public intoxication. But decriminalizing marihuana or other drugs means only making the laws slightly less lenient than the Volstead Act. Criminalizing sale or possession is no longer taken seriously as a way to control alcohol, but it is the first line of defense against other forms of drug abuse. Social pressure is directed mainly toward getting people to obey the law, that is, not to use the drugs at all. Despite the interesting work on informal social controls over illicit drug use, we are not likely to see a campaign for "responsible" use of marihuana, cocaine, or LSD resembling the publicity for "responsible drinking." The argument for using the criminal law is very simple: If less drastic means of raising prices and restricting availability work, this one should work even better; and even if less drastic means will ndwork, this one may. As one authority puts it, harassment by the criminal law is worthwhile if it can increase the time needed to obtain an intoxicating dose of heroin from five minutes to two hours (Moore 1977).

Criminal laws differ qualitatively from other legal restraints. To discuss what acts should be criminal raises in a new form the issues of fegal moralism and cultural symbolism. Especially when the user is made a criminal, morality has often taken precedence over health and consumer protection. The strange argument that removing criminal penalties on marihuana would amount to public endorsement of the drug reflects the symbolic weight given to the criminal law as a mark of social disapproval. For our purposes the symbolism is important only insofar as it provides a deterrent in itself, apart from the formal penalties. But where drugs are concerned, the value of criminal law symbolism as a deterrent is much smaller than its value as a mark of cultural dominance; that is one reason the public will often refuse to give it up even when the deterrent effect seems close to nil.

Against criminalization there is a long list of objections: the enormous burden on police, courts, and prisons; the preemption of resources that could be devoted to other crimes; the forgoing of tax revenues on a legitimate commodity; the unavoidability of unequal and selective law enforcement; promotion of disrespect for the law when large numbers of people do not accept it and disobey it shamelessly; ruined lives or careers of convicted drug users who have done no harm to others; economic niches created for organized crime; occasions for police corruption and violation of citizens' rights; increased dangers of drug misrepresentation and adulteration; addict crime made necessary by high black market prices.

There is a large critical literature with plenty of evidence on these disadvantages and unpleasant side effects of the criminal law approach to drug control. But most of it does not suggest general rules for judging when criminal penalities for the possession or sale of a particular drug in particular circumstances are unwarranted. The proper limits of the criminal sanction are an old topic of debate. We mentioned the paradox that people who could profit from formal drug education probably do not need it. The legal scholar E. A. Sutherland stated an analogous paradox: Where customary restraints are adequate, laws are unnecessary; where customs are inadequate, laws are useless. If this saying were wholly true, an anarchist utopia of positive liberty would be the only solution. We want to know when it applies in the existing imperfect world.

In the following general situations, criminal sanctions are likely to be wrong:

1. The act under consideration does more good than harm, or only slight harm, or the facts are just not known well enough.

2. There are ways of dealing with the problem that impose less serious restrictions on liberty — education, civil penalties, the whole range of solutions suggested for alcohol problems.

3. The law is not backed by a popular consensus and therefore has little symbolic value as a mark of disapproval.

4. The cost of enforcement, direct and indirect, in money, time, resources, and side effects, is a greater burden on society than the consequences of the crime. In the extreme, the law is unenforceable. As Jeremy Bentham pointed out long ago, sometimes the danger of detection is so small that enforcement would have to be disproportionately harsh to make the law a true deterrent:

With what chance of success, for example, would a legislator go about to extirpate drunkenness and fornication by dint of legal punishment? Not all the tortures which ingenuity could invent would encompass it. . . The great difficulty would be in procuring evidence; an object which could not be attempted . . . without spreading disarray through every family, tearing the bonds of sympathy asunder, and rooting out the influence of all social motives. (Sinclair 1962, p. 178)

5. It is too hard to formulate the law in a way that prevents inequities in punishment. As the Canadian Commission of Inquiry into the Non-Medical Use of Drugs reported (Final Report 1973, p. 98), any criminal classification of drugs will be subject to "incoherences and anomalies," and drug laws "do not distinguish the extent of harm" very well.

If one of these situations exists, others are likely to exist as well. If a law lacks popular support, the cost of enforcement will be burdensome; a law will become unpopular if the means necessary to enforce it outrage many people; inequities in punishment are more likely if the act usually does little harm; and so on. (For more discussion, see Packer 1968; Kittle 1971; Turk 1972; Irwin 1973; Greenawalt 1974).

If drug enforcement is taken seriously, enormous resources have to be devoted to it. We mentioned that under Prohibition, a third of the inmates in federal prisons were there for alcohol trafficking. Today the Drug Enforcement Administration (DEA) has an annual budget close to $1 billion. About three-quarters of all search warrants are issued for drugs or evidence of drug offenses; electronic surveillance and wiretapping are reserved mostly for drug enforcement under the official designation of "organized crime." Here is what the public money pays for: "DEA's second command post in El Paso, Texas looks like the War Room in Dr. Strangelove. Some men sit hunched over video display terminals while others monitor drug flow charts and airspace indicators. The agency also employs surveillance satellites, fixed radar and microwave installaiions, AWACS, radar planes, and one of the largest computerized intelligence files in the government" (Dubro 1981, p. 13).

Altogether, probably more than $10 billion a year is spent to suppress drugs. Much of it is used to stop smuggling, but at most 10 percent of imported illicit drugs are seized. A relaxation of the marihuana law g in California saved at least $7.4 million in court costs in the first six months (Report 1977, p. 28). According to a study conducted in Los Angeles County in the early 1970s, more than a third of all felony arrests were . on drug charges, most for possession; 40 percent of these were rejected in court, mainly because of insufficient evidence (Silberman 1978, p. 357). The attorney general of the United States reported in 1977 that 55 percent of the cases in federal courts in Tucson, Arizona, involved drugs. When the harsh Rockefeller laws were passed in New York and plea bargaining in drug cases became more difficult, court dockets were immediately flooded; there was no evidence of a decline in heroin use (Pekkanen 1980).

The zeal of police against violators can have very unpleasant effects:

In separate actions [on April 23, 1973, in Collinsville, Illinois] these agents [of a Justice Department drug abuse task force] conducted raids at the homes of Mr. and Mrs. Herbert Giglotto and Mr. and Mrs. Donald Askew. The shabbily-clad unshaven raiders barged in on the two houses without. . . search warrants. . . they kicked in the doors without warning while shouting obscenities and threatening the inhabitants with drawn weapons. The Giglottos and Askews . . . were forced to stand idly by while their homes were ransacked and their lives threatened. At no time . . . did the agents satisfactorily identify themselves. . . . When they discovered that they had raided the wrong premises, they simply left — no apologies, no explanations, no offers to fully compensate for the damage they had done.

The agony of the Giglotto and Askew families ended only temporarily when the agents departed . . . both families suffered intermittent harassment . . . late-night threatening telephone calls. . . . But perhaps more demoralizing . . . has been the social ostracism both families have suffered. . . . By July, the intimidation and social pressures had become so intense that the Giglottos were forced to permanently flee Collinsville. (Percy 1974, p. 5)

Another example: "A New Jersey man left his jacket lying in a neighbor's back yard, where it was found and taken in to the police. The man was arrested because there were eight Libriums in the pockets. Even though he had a prescription. . . he was arrested for carrying them in an unmarked bottle. That, according to New Jersey law, is an offense that carries the maximum of three years imprisonment and a $1000 fine" (Dubro 1981). Presumably that man was not jailed, but he may have a criminal record, like the hundreds of thousands arrested each year for possession of marihuana.

In a fascinating pamphlet (Kirkpatrick 1975), prosecuting attorneys charged with enforcing the drug laws speak anonymously. On strict laws and selective enforcement: "Borrowing some pills from a friend — every time someone borrows even penicillin from you. . . they are in technical violation of the drug laws and could get a severe sentence, but wouldn't, of course" (p. 9). On community response to the drug menace: "they knew that drugs are supposed to be a big issue. Well, they don't have any drug problem yet, but they feel like they ought to. . . . But you can't say, 'Well, you people are out of your minds,' or you're earmarked as somebody who's anti-law enforcement or soft on drugs or soft on crime" (p. 26). On questionable enforcement methods: "undercover agents were placed in a school for the deaf and they posed as deaf workers for six months . . . in order to catch people enrolled at the college for the deaf who were using marihuana. And when the police came in here with a case like that, that they're so proud of, and you appear maybe a little shocked or reluctant to get involved . . . then it's bad for your relationship" (pp. 29-30).

On drug searches and seizures: "When you ask for a warrant knowing that it's going to be quashed, when you know it's bad to begin with, you can still get one issued in most counties, make the search, and have it thrown out later. You've busted in the guy's house and ransacked the joint, or you've impounded his car and torn the insides out of it. So the warrant is quashed, so what? . . . And some people are quite willing to abuse the law in that way, to make life unpleasant for people they have in their minds identified as criminal anyway" (p. 39). All of these statements bring to mind Bentham's comment about procuring evidence to extirpate drunkenness and fornication by legal punishment; it is "an object which could not be attempted . . . without spreading disarray through every family, tearing the bonds of sympathy asunder, and rooting out the influence of all social motives." (For further examples and discussion, see Manning 1981.)

The police have several strategies for controlling drug traffic, all adjusted I to the absence of complainants and the problems of obtaining evidence. (The test for possession of a drug is formally quite demanding — either `1 immediate or exclusive access to the drug or dominion and control.) General street patrol is ineffective as long as search and seizure are not permitted without probable cause. But people who are stopped for some other reason can be searched and then charged if drugs are found. Most , drug arrests are made by ordinary patrol officers on possession charges, but they yield relatively few convictions; in one study of six cities, only 14 percent of those arrested served time in jail, and most of them were sellers (Johnson and Bogomolny 1973). Most crimes are solved by retrospective investigation, but that is rather ineffective in drug cases except on rare occasions when an informer has been planted at a high level in a distribution network. Long-term covert operations are difficult and un- I common. A better technique is planting an informant to get a search warrant; often the key to drug law enforcement is the use of informers who cooperate to reduce their own sentences (Moore 1977). This practice obviously creates enormous potential for gross error and injustice. The most important police strategy for arresting the larger dealers is "buy and bust." After buying drugs several times to establish a reputation and a. business relationship, the undercover police officer arranges a new sale and makes an arrest when the money and drugs have changed hands. The procedure creates many opportunities for corruption, since the police collect so much untraceable cash.

The institutionalized corruption, betrayal, chaos, and terror promoted I by drug law enforcement are not only unfortunate byproducts of a nasty but necessary business, they are essential means of attaining the law's ends. The most serious defenders of the criminal law's regulatory system recognize this explicitly (Moore 1977). Only a small proportion of drug users or even dealers are actually arrested and convicted. But they have nowhere to turn when they are robbed, cheated, or assaulted by other users and dealers. By making the business as nasty and risky as possible, law enforcement is supposed to limit supplies and keep retail prices high. So those who support the system are in an uneasy position when they profess outrage at contemptuous, cruel, and lawless behavior by narcotics agents. It is certainly hypocritical to pretend to indignation and alarm about the murderousness of drug trafficking itself, since we have in effect deliberately tried to shape it in a way that makes it attractive to the most reckless and callous people and as nerve-racking as possible for everyone involved. The main reason marihuana and heroin dealers leave the business is apparently not the threat of prison but "interpersonal strain" — an indirect but no less powerful effect of the criminal control system (Fisher and Bruhnsen 1982). The unpleasantness is sometimes even blamed on the drugs themselves, as though it were a pharmacological effect, but that is largely myth making and only confuses the issue. To be honest, we must admit that guaranteeing the existence of an underworld of treachery and terror is simply one of the methods society has reluctantly chosen to prevent drug abuse, or at least to quarantine the carriers of the drug traffic.

Drug laws notoriously ask more of the criminal justice system than it can deliver. Political principles and social realities that are ignored when the laws are passed must be considered when it comes to enforcement. We make laws that cannot be enforced without paying a price, in money or in liberty, that we are unwilling to pay. The cost of the enforcement machinery is only part of the story. Vested interests of a capitalist economy become involved (we will say more of this later), and so do vested interests of a liberal political system. Mill anticipated some present dilemmas of drug law enforcement when he wrote in On Liberty,

It is only because the institutions of this country are a mass of inconsistencies, that things found admittance into our practice which belong to the system of despotic, or what is called paternal government, while the general freedom of our institutions precludes the exercise of the amount of control necessary to render the restraint of any real efficacy as a moral education. (Mill 1859, chap. 5)

It is useless to ignore questions of liberty in a cost—benefit analysis of the effect of drug laws. The tactical problems cannot even be defined without some acknowledgment of the freedoms we are committed to protecting, whether commercial enterprise or individual rights. The "general freedom of our institutions" is what precludes full exercise of the paternal control written into the drug laws. The Wickersham Commission admitted that alcohol prohibition could never be fully enforced without the use of methods intolerable in a democracy; the same is clearly true of other drug laws, however necessary they may be. A study in New York City in 1969-70 found that 70 percent of the misdemeanor charges and 33 percent of the felony charges involving drugs were dismissed in court, more than half of them because of faulty police procedures, especially insufficient evidence or illegal search and seizure (Cooper 1973).

Of course, costs and individual rights have to be weighed in judging any criminal law, but the drug laws create such a peculiar situation quantitatively that they seem qualitatively unusual. The issues we neglect in making substantive drug legislation return to haunt us in the form of problems that are described as procedural. In a standard textbook on criminal law (Vorenberg 1975), drug and alcohol offenses compose about half of the cases reprinted in four full chapters on admissibility of evidence, stops and arrests, search and seizure, and entrapment. Police and advocates of law and order tend to regard procedural scruples as merely an obstacle and a nuisance, an elaborate game in which the authorities are prevented from doing their work. But just as, according to Sir Henry Maine's famous saying, the substantive common law of England seems to have been secreted in the interstices of procedural decisions, so the substance of our drug laws, which are formally so harsh, is often created in procedural guise.

An interesting symptom is the repeated efforts to impose mandatory minimum sentences for drug offenses. Before 1970, narcotics crimes were the only ones for which the federal criminal code both specified a minimum sentence and denied the possibility of probation or parole. The Drugs Abuse Prevention and Control Act eliminated most mandatory sentencing, but state laws since then have repeatedly imposed it. The implication here is that in its day-to-day workings, the criminal justice system cannot be trusted to be harsh enough; the substance of the law will be undermined by the procedures for applying it in drug cases. The most common effect of such laws is to give prosecutors sentencing power by allowing them great discretion in deciding what crimes they will impute to a drug dealer for purposes of plea bargaining. Sometimes the result is that the defendant pleads guilty to a lesser crime that could not have occurred in order to avoid conviction and a mandatory jail term for a more serious crime. These distortions arise because of our unwillingness or inability, for technical and principled reasons, to follow through on the laws we make.

Drunk driving is one kind of behavior related to intoxicant use for which criminal sanctions apparently make sense. Here the harm is obvious, the danger is immediate, and cause and effect seem clear. Even Mill would probably have admitted penalties for drunk driving, just as he was prepared to penalize soldiers for being drunk on duty. But even in this 'case, many studies suggest that the law has little effect. First, only a 'small minority of drunk drivers are caught; we could not catch many more of them without devoting more resources to the job than we are willing to make available. When roadblocks and breathalyzer tests are introduced, drunk driving may become less common for a while — until people realize that it is too costly to use roadblocks and breathalyzers long enough on a large enough scale to be effective. Second, when the laws are made harsh, judges, juries, and prosecutors are unwilling to apply them (Waller 1978), and not only because most of them can easily conceive of themselves as getting into the same kind of trouble. After all, only a small proportion of drunk drivers become involved in serious accidents; even fewer such accidents cause deaths, and if anyone dies, it is much more likely to be the drunk driver than an innocent stranger. Any punishment harsh enough to deter might not be defensible as a criminal sanction. License suspension would be a good alternative if only enough drunk drivers could be caught. Scandinavian countries are famous for their strict drinking—driving laws, but there is some evidence that habits and moral education are responsible for their impressive record of highway safety rather than the laws, which are more a symptom than a casse of the regard for prudent driving (Ross 1975, 1982).

Removing penalties for possession of drugs is a widely recommended solution for some of the problems created by the criminal law. Possession of narcotics did not become a federal crime until 1914, but by 1960 it was actually a felony under federal and most state laws. Public opinion thenviewed illicit drug use as a vice and the illicit drug user as immoral. Although federal possession penalties were added for amphetamines, barbiturates, and other psychoactive prescription drugs in 1968, attitudes had already begun to change; the illicit drug user had begun to seem more a victim than an evildoer. Federal possession penalties are lower under the 1970 law, and more than thirty states as well as the federal government now allow conditional discharge of first offenders after trial without an adjudication of guilt. The Canadian Commission of Inquiry into the Non-Medical Use of Drugs and a recent National Academy of Sciences report have both suggested decriminalization of marihuana possession, and several states have reduced the penalty to something like a traffic ticket. In the Netherlands, marihuana and hashish can even be sold openly; the police have an announced policy of not interfering with small-scale distribution. In Alaska, cultivation of marihuana at home for personal use is legal. Although arrests for drug possession, especially marihuana possession, are still very common, few of those arrested now go to jail if that is the only charge.

Putting the burden of criminal sanctions entirely on the grower, manufacturer, and distributor rather than the user, sometimes called the "vice" model of drug control, has many advantages, both tactical and moral. By cutting off the drug supply where it is concentrated in a few hands, it seems possible to achieve the maximum effect at a minimum cost. The problem of enforcement is reduced to more manageable proportions, since we are no longer making criminals of a large part of the population, with all of the complications and injustices entailed. And whatever the effect of criminal sanctions on suppliers, decriminalization of possession alone does not necessarily increase the use of a drug, if surveys on marihuana in states where the law has changed are correct.

Punishing only the seller also eases some of our moral and political doubts about paternalism. Even Mill was willing to consider punishing people who profit by persuading others to do what is bad for them; brothel keepers and gambling house proprietors were his examples. The evil reputation of the large-scale drug trafficker today is even greater than that of the brothel owner in Victorian times, and he is a suitable object of outrage. The symbolic value of the criminal law is greater if it is directed at the seller rather than at the user. Even people who use the drug may despise the dealer, just as slaveholders in the ancient world and in the American South often felt contempt for slave traders. The consumer or medical model of drug control also dictates that users be left alone, since they are at most victims; we do not fine or jail the patients of someone who practices medicine without a license or the customers of a manufacturer who disobeys consumer protection laws. Finally, the incohérences and anomalies in the distribution of punishment that are hard to avoid in any criminal classification of drugs seem less troubling when only the person who profits from other people's drug use goes to prison.

But not too much should be expected from this kind of change in the law. Alcohol use was not criminalized during Prohibition, which still produced most of the same evils as current drug laws. The idea that the conviction of a few big drug traffickers would break up the whole commerce is seductive but false. The problem ultimately lies with the user, not the dealer. As long as there is a market, any so-called kingpin is replaceable. Besides, it has been argued that the possession penalty is useful to get convictions for sale and distribution; for example, persons arrested for possession can serve as informers. One student of heroin policy notes that small-time dealers, especially those who are also addicts, are least cautious and therefore tend to bear the brunt of enforcement; but he regards that as acceptable, since the least cautious dealers are the ones who accommodate new heroin users (Moore 1977). Another reason suggested for preserving the possession penalty is to induce drug abusers to enter treatment. That is a neat solution for anyone who adopts the disease—crime model of drug abuse discussed in Chapter 2. Faced with a crime whose perpetrators need treatment more than punishment, which is also a disease for which the sufferers refuse to take a cure, we might do best with this two-stage procedure: Arrest them as criminals and then impose treatment on them as sick people.

Principle and practice, substantive and procedural problems, abstract rights and concrete actualities confront one another most directly in cases involving constitutional issues. We have already discussed the courts' reluctance to invalidate civil commitment of addicts on constitutional grounds. They have also rejected almost all constitutional challenges to the substance of the drug laws. But drug laws continue to provide the most common occasions for raising many procedural issues.

The main possibility for a constitutional challenge to drug laws lies in the right of privacy, which developed slowly out of interpretations of the First Amendment (freedom of expression, the right to read), the Fourth Amendment (against unreasonable searches and seizures), the Fifth Amendment (against forced self-incrimination and deprivation of life, liberty, or property without due process of law), and the Ninth Amendment (the people may retain rights not specifically enumerated in the Constitution). The Fourteenth Amendment charges states to abide by the due process clause of the Fifth Amendment; in 1973 the Supreme Court ruled that this required states as well as the federal government to respect the right
. of privacy, which had become incorporated into the Court's interpretation of the Bill of Rights. The idea is that in some areas the control of individuals over their own thoughts and actions should be nearly absolute, and government interference is warranted only if there is an unusually strong reason for it. When an action involves a right "fundamental to ordered liberty," the Fourteenth Amendment prohibits interference with it unless there is a "compelling state interest." The protected areas of intimacy are the mind, the body, and the home.

The right of privacy is an obstacle to legal moralism and therefore provides some of the same protections as Mill's principle of liberty. In a sense, the distinction between the moral and the legal is better stated as a distinction between the private and the public (Shklar 1964). Some people may feel distaste for what goes on in other people's homes or minds, or what they do with their bodies, but the law, in effect following Mill's recommendation, will not treat this distaste as injury to the public. The affront to sensibilities in a breach of public order or decency is distinguished from feelings of disgust or annoyance about what is happening somewhere behind the scenes. Any act may eventually have public consequences, but that is not considered a good enough reason to invoke , the law.

Abortion, birth control, and much pornography now come under this rule, but drug use does not. The Supreme Court has upheld the legality of abortion and birth control and the right to have pornography in the home. But drug laws are not considered to be invasions of the private realm. Federal courts have rejected the claim that a right to use any drug, in the home or anywhere else, is fundamental to ordered liberty. By refusing to hear appeals in which people demand the right to use the cancer drug laetrile, the Supreme Court has indicated that it intends to leave this issue to the legislatures. Only the Supreme Court of Alaska has ruled that cultivation of marihuana for personal use is legally protected by an explicit guarantee of the right to privacy in the state (not federal) constitution. And no court will ever rule that the right of privacy protects any commerce in drugs.

Even if a right is not judged fundamental to ordered liberty, it may be guaranteed by the due process clause, or by the clause in the Fourteenth Amendment that requires states to provide equal protection of the law to all people within their jurisdiction. But the courts have interpreted legislative restraints as constitutional under these clauses so long as they have some rational relationship to a permissible state end; no compelling state interest is necessary. In particular, alleged inequities and anomalies in the distribution of penalties are not generally judged to be in violation of equal protection or due process. Only in the case of a fundamental right like freedom of speech will the Constitution be said to forbid serious disparities. In obscenity cases, the courts always look for a consistent rule that justifies banning some books or pictures and not others. But no principle is required to establish the constitutionality of a legislative distinction between two drugs.

The courts will not provide a judicial remedy for every dubious law, and they will rarely invalidate the legislature's classification of crimes, They would not overrule a state law that made picking pockets a more serious crime than shoplifting, or a law that required safety devices on chainsaws but not on power mowers. For similar reasons, they have not been willing to overrule penalties for marihuana possession and sale on the ground that they are too severe in relation to the penalties for other drugs or the dangers of marihuana itself. They have upheld state laws imposing a considerably longer maximum sentence for the sale of cocaine than for the sale of amphetamines, which are pharmacologically similar. And the special treatment of alcohol is not a constitutional problem because the Twenty-First Amendment, which repealed Prohibition, has been interpreted so that it supersedes other applicable constitutional provisions and gives the states full power over alcohol regulation. The problem of distinguishing the extent of harm, as we mentioned earlier, may reduce the effectiveness of criminal laws on drug use and sale, but it has not been made into a constitutional obstacle.

Despite its rejection by the courts as a substantive right in drug cases, privacy has had an important indirect influence on the enforcement of drug laws. The Fourth Amendment's guarantee against unreasonable search and seizure, extended to the states through the due process clause of the Fourteenth Amendment, has served as a constitutional protection for drug offenders. The relationship to privacy is obvious in the Fourth Amendment's reference to protecting "persons, houses, papers, and effects" against searches and seizures. Because the provision was hard to enforce in any other way, the Supreme Court ruled in 1961 that evidence obtained by illegal searches was not admissible in state criminal trials. The ruling has been subjected to much criticism, and the Court is reconsidering it. Important exceptions to the exclusionary rule may eventually be established.

The Fourth Amendment is now interpreted as allowing two kinds of legal search. First, police may obtain a warrant "upon probable cause, supported by vow or affirmation," with a suitably precise description of the place to be searched and the things to be seized. Probable cause is a difficult requirement. Exploratory searches based on mere suspicion do not come up to this standard. The hardest cases involve anonymous informants, and here the courts have not known where to draw the line. Questions arise about when informants have a good enough reputation for reliability and when their names must be disclosed. A warrant is required for telephone tapping unless the police have the consent of one of the parties to the conversation. The courts have also had to decide whether drug courier profiles providing a formal list of suspicious characteristics give probable cause for a search.

Searches incident to arrest are much more common than those conducted with warrants. They are permissible in order to find weapons or to prevent escape or the destruction of evidence; normally, anything in plain view can also be seized. Since the decision excluding illegally obtained evidence from admission in court, the police have taken to lying about this issue. They tend to regard the exclusionary rule as a meaningless obstacle to law enforcement, and they are unwilling to allow it to interfere with the conviction of people who, as they see it, are clearly guilty. Thus they often feel no compunctions about perjuring themselves. There has been an epidemic of what is called "dropsy" testimony; a police officer states that the defendant "dropped" a package of drugs on the floor in plain view while being apprehended. Police also refer to "furtive gestures" that made them think that the defendant was reaching for a gun or trying to get rid of evidence. To justify legally a successful search undertaken after a tip by an informer, the police will sometimes fabricate a history that makes the informant seem reliable (Heller 1973).

The Eighth Amendment's prohibition against cruel and unusual punishments has also been extended to the states through the Fourteenth Amendment, and under this principle both state and federal courts have invalidated a few particularly savage sentences in drug cases. For example, in People v. Lorenzen, decided by the Supreme Court of Michigan in 1972, a twenty-year prison term for the sale of a small amount of marihuana was invalidated under a Michigan state constitutional clause prohibiting "excessive" punishment, interpreted by the court as cruel or unusual. Noting that the maximum penalty for assault with intent to maim was ten years, for cruelty to children four years, for selling a machine gun five years, and for adulterating food ninety days, the court ruled that the marihuana sentence shocked the conscience. Criminal penalties for drug transactions are generally severe, but as we have said, the courts have never invalidated any of them under the equal protection or due process clause. Only occasionally, in extreme cases, will courts challenge the , legislature's distribution of penalties by invoking the Eighth Amendment and its equivalents.

Another issue that occasionally arises in drug enforcement is the Firth Amendment's protection against self-incrimination. In 1952 the Supreme Court decided the case of Rochin v. California, in which the defendant was convicted of possessing morphine. He had swallowed the morphine capsules as Los Angeles police officers broke into his house and burst into his room. The police took him to a hospital and ordered a doctor to give him an emetic through a tube inserted into his stomach. The morphine was recovered, and the defendant was sentenced to sixty days in jail. The Supreme Court's majority said that the breaking and entering, assault, and torture sufficiently shocked the conscience so that the due process clause was violated. Justice Black considered this statement too vague; he said that the due process clause of the Fourteenth Amendment extended to defendants in state courts the protection of the self-incrimination clause of the Fifth Amendment, which was violated here.

Drug cases raise so many procedural issues so often because of the peculiar nature of the laws. Since drug use and sale are common activities. that have been made serious crimes, the number of cases is large. The consensual nature of the crime makes evidence difficult to get. Penalties sometimes seem excessive or irrational, but courts have been unwilling to say that drug laws violate any substantive rights. As a result, the procedural system has to bear the burden of our doubts and second thoughts. In establishing limitations that are formally procedural, judges in a given case are often influenced by substantive questions — what drug is involved, how much, whether there is a sale or an organized criminal enterprise. Procedural adjustments have to accommodate distinctions in degree of harm not effectively made by the statutes, enforcement problems not anticipated by legislatures, and potential rights violations otherwise ignored.

Nevertheless, if the obstacles are overcome and a drug conviction is obtained, it is rarely invalidated for constitutional reasons. Big drug dealers especially fall under what has been called the "bad guy exception" to the Bill of Rights. An appeals court may declare a procedural error committed during the trial to be harmless even when it was in fact substantial. The court may also reject a drug dealer's appeal because his lawyer did not fill out the legal forms in exactly the right way. When it can find no constitutional justification at all for the trial procedure, the higher court may deny the appeal without giving a reason — a recourse that is supposed to be used only for appeals that are entirely frivolous. So, especially in drug cases, "The Constitution is rhetorically overvalued by the courts and then systematically devalued in legal practice" (Hughes1982, p. 27).

Taxation, restrictions on the time or place of sale, and criminal sanctions on users or sellers are the most common legal methods of reducing the supply of a drug. But the problem can also be approached more specifically by making the drug legally available only to designated users or for designated purposes. The most nearly universal designated user group, of course, is adults. Even Mill made the obvious distinction between adults and children. Other forms of discrimination among classes of users are less acceptable today. We mentioned the early American attempts to prohibit alcohol to sailors, servants, and Indians Plato, in the Laws, proposed that slaves, magistrates, and soldiers should not be allowed to use alcohol. In the Soviet Army, only officers, not enlisted men, are permitted to drink. These forms of discrimination now seem inequitable, to say the least. The passbook system that operated in Sweden for many years was meant to limit the amount of alcohol any person could consume, thus restricting it to moderate drinkers; that system failed. It is true that even the methods now in use often have the indirect effect of designating classes of consumers. High prices exclude the poor; criminal sanctions may exclude people not adventurous or reckless enough to ignore them. But except for children, we now rarely consider using the law explicitly to keep drugs away from a specific class of people who are thought to be especially subject to their ill effects. The reasons are partly empirical — we simply cannot predict well enough who is at risk — and partly a matter of democratic principle.

Legal limitation according to the purpose of drug use is much more common and effective. One form of limitation by purpose is confining a drug to religious use; for example, during Prohibition, sacramental use of alcohol was licensed. The only legitimate religious use for a banndrug today is in the Native American Church ritual, discussed in Chapte 1. In an important case, People v. Woody (1964), the California Supreme Court ruled that no compelling state interest justified a ban on peyote use in the Native American Church. The "burden" on religious practice created by a ban on peyote would amount to an infringement on the free exercise of religion. The case was different from that of polygamy in the Mormon Church because here the practice being challenged was "central"; It could not be eliminated without destroying the church, and it coub not be replaced. In Leary v. United States (1967), a federal appeals co refused to extend this right, saying that a compelling state interest i preventing drug abuse outweighed the petitioner's claim to the ruse o marihuana in his religious practice as a Hindu.

It is hard to find any consistency in the principles used in cases deck.' under this clause. Courts have not questioned the good faith of claims a religious exemption; that issue has always been left to juries. And the have not tried to define religion, out of fear that any definition mil violate the First Amendment clause prohibiting the establishment of reli:, ion. But the court stated in the Leary case that conduct otherwise crimin would not be made legal by a religious justification — a comment th: if taken literally, would not only contradict the Woody decision but al make the free exercise clause meaningless (On these cases, see Philli• 1968 and Hallucinogens 1968.)

Possibly one of the unstated principles is a rejection of patpaternalismr example, people have been allowed to refuse blood transfusions f• religious reasons. Quoting a remark in the state attorney general's brithat peyote "obstructs enlightenment and shackles the Indian to primiti conditions," the court in the Woody case declared that making this • ternalistic judgment was not the attorney general's business. But judges also had something more in mind, as suggested by their referen. to "subcultures that flow into the mainstream of our national life . give it depth and beauty" (Uelman and Haddox 1974, p. 239). It w. important that in this case, as opposed to the Leary case, the religio practice created a cultural community. The integration model of dru control emphasizes informal social sanctions and rituals as a protection against wrong use of drugs. By making impulsive and careless drug use unlikely, the seriousness of purpose implicit in a religious ritual adds to the protection. Religious sanctions are always among the most effective; as we have suggested, the integration model might work best in a theocracy. This legal limitation on purpose is supposed to ensure a community in which the drug is used in a decorous and disciplined way.

But religious purpose does not guarantee that drug use will be beneficial. It is not only that some religious drug use — think of the Dionysian rites — is not decorous and disciplined. Religion is too equivocal a notion. In two novels published thirty years apart, Aldous Huxley portrayed two contrasting types of religious drug use. The "soma" of Brave New World is a euphoriant or hallucinogen used by the people of a dystopian future society to cement their communal solidarity. They take the drug as they sit and chant ritually about melting into the Greater Being, which they identify with the Social River. Soma is a tool for the quiet suffocation of individuality and independence in a world where even human beings are cultured in laboratories according to a formula (Huxley 1932).

The religious drug use in Huxley's last book, Island, is very different (Huxley 1962). Here he presents a true utopia in which a decentralized political system is guided by a Western technology and science stripped of their excesses and a religious—philosophical tradition derived largely from Buddhism. A psychedelic or hallucinogenic drug called "moksha medicine," after the Sanskrit word for spiritual liberation, is used on carefully defined social and personal occasions. The emphasis on community, tradition, and discipline prevents the drug use from being merely an experience to be enjoyed or endured, and the guidance of religion and science ensures that it will encourage rather than stifle individuality, variety, and freedom.

But Huxley is pessimistic about the prospects of any such utopia. The island society is stable and successful only because it has been isolated for more than a century from the forces of modernity by the ocean and a benevolent government. In the horrifying last chapter, it is taken over and its culture is destroyed by a nearby tyrant who represents a combination of all the worst features of capitalism, communism, and Third World nationalism. Huxley almost admits that in the modern world his dream is a hopeless one.

The courts have in effect given the Native American Church an exemption from modernity resembling the one that Huxley granted his island. The rules that put drugs and religion in two different categories, protecting religion from contamination, are modified only for a special kind of culture. When official policy toward Indians was assimilation, peyote was regarded as a drug abuse problem. Now that Indians are being allowed to reclaim their cultural identity, peyote has been made legal. But if any other group were involved — say, hippies — the courts would not be so indulgent.

Even apart from the law, it is hard to build modern religious institutions based on drug use. From older forms of religious drug use, we might learn how to shape a drug experience through ritual and how to balance democracy and authority in allowing access to it. But these lessons are hard to learn because we lack the traditions and consensus that would make consciousness changes produced by drugs religious in the old sense. What serves some as moksha medicine may look like soma to others.

The other major limitation by purpose is medicine. Although it emerged from a religious background, medicine is obviously much more important than religion today as a basis for drug control. Religious and medical control both now require government endorsement; aside from that, they are formally similar in certain ways. Both methods have a customary rather than a legal origin, and the medical gatekeeper with Aesculapian authority plays an almost priestly role. Modern medicine has een been sarcastically likened (by Thomas Szasz, for example) to a state religion, with imposing organizational strength, an intellectually powerful ideology, and the support of civil authority. Medicine has its divine powers (medical miracles and wonder drugs), and can also call forth and exorcise demons (drugs of abuse). It even borrows images of purity from religion: and it regards unorthodox healing practices as heresy or superstition to be eliminated by a mixture of official coercion and missionary activity.

The priestly pretensions of doctors may sometimes seem arrogant, and they were challenged by the open invitation to experiment with drugs, issued during the 1960s, which many people continue to accept. But it could be argued that in modern industrial societies the medical profession has to supply the ritual context that makes drug technology relatively safe. Where the ritual of prescription is unavailable, as with alcohol, the results can be disastrous. Some control is needed, and nineteenth-century individualism or openly religious restrictions are impossible in present social conditions.

As we mentioned in discussing the issues of risk and liberty, a medical definition provides a stay against confusion. That is one reason it has become important in public thinking. In a poll taken in the early 1970s, people were asked to choose the best of several definitions of drug abuse; 27 percent chose "excessive use," but 30 percent chose "use for nonintended purpose, nonmedical use" (National Commission 1973, p. 12). Despite' occasional scandals confined to a small minority of doctors and patients, people in general tend to be conservative about prescription drugs, including psychoactive ones; they rarely take more than the doctor recommends and often take less (Manheimer et al. 1973; Boffey 1981). Medical control, if it could be imposed without explicit penalties, merely by respect for the authority granted to doctors, would be a kind of integration model; for example, it supplies clear prescriptions and proscriptions, and it provides for transmission of drug lore by knowing elders. The rituals and social sanctions are somewhat more restrictive and formal than those contemplated by the integration model, but the principle is not necessarily different. In preliterate cultures in which medical and religious uses are not always clearly separated, the similarity is obvious; the religious- medical use of hallucinogens in some preliterate cultures (and on Huxley 's utopian island) is a model of integrated drug use.

In all drug controls that are legally imposed according to purpose, the main problem is to make sure that the purpose is defined narrowly enough. Because courts have not dared to define religion for fear of establishing a church or churches, they have had to use roundabout methods of narrowing the qualifications for a religious exemption from the drug laws. Legal limits on the definition of medicine are both direct and indirect. The government, through the FDA, now decides what drugs will be made available to doctors. Pharmacists must also keep records of prescriptions for controlled substances, and doctors who seem to be prescribing too freely may be called on by the government to justify themselves. The famous Supreme Court decisions against opiate maintenance imposed a more explicit legal definition of medical purpose in drug use.

Despite these legal sanctions at the margins, the decision on what is a medical purpose has been left largely to the medical profession itself. Any doctor with the proper license may still legally prescribe a psychoactive drug for almost any purpose, subject to the threat of malpractice suits and professional discipline. We discussed the historical development of the idea of a medical purpose in Chapter 3. It has left us with the need to distinguish medicine from pleasure, or more often from the kind of casual therapy used to ease problems of daily life. Medicine must also be distinguished from the enhancement of performance in some task or enterprise. Obviously there are many borderline cases. People attending funerals have long used wine and whiskey as medicine for grief, and today a doctor or friend may supply diazepam (Valium) to a mourner. Many people also use Valium for everyday stress. A writer describing her use of the drug says that it gives her a feeling of detachment and makes it easier for her to feel both joy and anger by dispelling anxiety (Carpenter 1980). Whether this use of the drug is nonmedical (and therefore wrong?) is a hard question. When amphetamines are given to children with attention deficit disorder (hyperactive children), some regard it as a' highly questionable pedagogical technique, others as simply a treatment for a disease.

One purpose for which most drug use has become unacceptable is sports. It seems almost self-evident to most people today that using drugs in athletic competition is wrong, and many consider it potentially disastrous for sports. But it is curiously difficult and complicated to justify this position. What seems self-evident is not so at all. In 1900, for example, a patent medicine whose active ingredient was cocaine won a prize at an English exhibition with a citation calling it "wine for athletes." This drink, Vin Mariani, was openly and unapologetically used at the time by athletes in many sports. If by now it seems almost unnecessary to give justifications for not allowing drugs in sports, that is largely because since the time of Vin Mariani medicine has annexed more territory. Medical use has become the only acceptable use of most drugs, and medicine is defined more strictly now than it was in 1900. Without this medical background, the arguments against drug use in sports might seem feeble.

Four standard reasons given for banning drugs from sports are that they are dangerous, they are unfair, they produce abnormal states of the body and mind, and they are artificial or unnatural. Taken at face value, these reasons would hardly be good enough. For example, it is just not plausible that drugs are banned because they are so risky. Athletes ici many sports are not only permitted but constantly required to take much more serious risks, both short-term and long-term, than those usually involved in drug use. To cite only the most obvious example, everyone knows that professional boxers often die in the ring or suffer severe brain damage, because the threat of damaging blows to the head is essential to the excitement of the sport. It would be implausible to say that any effect of drugs could be worse. But drugs are under medical control, and boxing is not.

Maybe we are implicitly assuming that the risk is not voluntarily assumed in the case of drugs because drug takers simply do not understand what they are doing. A boxer is supposed to know what he is letting himself in for when he climbs into the ring, but not when he takes a pill to fight better. But this is easy to believe mainly because our system of drug regulation is based on the assumption that most people are not fit to make their own choices about the more powerful drugs. And that is because most drug use is now acceptable only under the control and supervision of physicians.

The charge that drug use is unfair makes little sense when examined closely. The implication is that a player who gets his strength and spirit from a pill is not winning by his own efforts, but the same case could be made against any sports equipment. No one says that fielders using gloves or auto racers using mechanics and pit crews are not winning by their own efforts. It is hard to see why drug use is any different in principle. Drug use happens to be unfair right now because it involves breaking the rules to gain special advantage, but we could eliminate that unfairness simply by changing the rules.

There is also a suspicion that athletes who take drugs to perform better are using them in abnormal amounts and in an abnormal way. That sounds sinister until we try to define exactly what it means. Probably the best interpretation is that drugs are acceptable only for normalizing, not for optimizing. They can be legitimately used to cure disease or bring physical condition up to par, but not to improve performance at the high level of athletic competition, where they would violate some standard of human normality. We can even argue that this is what makes drug use unfair: Taking a drug to become well enough to play is all right, but taking it just to play better is a form of cheating.

Whether this position makes any sense depends on what normality means. If it is only a statistical average, there is no reason not to try to raise it. Giving drugs to pole vaulters to improve their jumping is no different from giving them fiberglass vaulting poles. If normal use of drugs or other equipment means use that is conducive to health, there may be a case for the objection. We sometimes think of sports as a way to cultivate a superior or ideal naturalness and balance — the classical Olympic ideal. But this standard, or even the distinction between normalizing and optimizing, is not applied very carefully when whole classes of drugs are permitted or banned by name. The issue is often settled by a kind of medical definition — a decision that the only normal way to use drugs is the way that doctors use them (or should use them): to relieve pain, treat illness, or compensate for disability. Athletes who feel pain because their bodies are demanding rest may take local anesthetics or antiinflammatory drugs to compete, although this is hardly a healthful practice; but a gold medal winner in the Olympics was disqualified when his blood showed traces of ephedrine, an asthma medicine that is banned because it has stimulant side effects. According to some accounts, amphetamines are used in football mainly by older players who feel a loss of their youthful energy and competitive enthusiasm. The drug brings them up to what could be regarded as a normal level — the level of their past play (Mandell 1978). The question is why is this wrong if antiinflammatory drugs are acceptable, or if wearing eyeglasses to compensate for deteriorating vision is acceptable. The answer seems to be that prescribing antiinflammatory drugs or local anesthetics to relieve pain is something that doctors do, and giving stimulant drugs is something they are not supposed to do. Talk of normalizing, optimizing, and abnormality misses the operative distinction here.

The final complaint against drug use in sports is artificiality or unnaturalness. But this assumes a totally unworkable distinction between natural (good) and artificial (bad) ways to improve performance. It is natural for human beings to use artifices; artifacts and artificiality are pervasive in sports. Bare feet are natural, so shoes must be artificial. To say that using drugs to cure illness or relieve pain is natural, but using them to run faster or throw farther is artificial, would be equivalent to calling the use of shoes to protect the feet natural and their use in track competition artificial. The charge of artificiality is really a variant on the charges of unfairness and abnormality; we have already seen how much our idea of what is unfair or abnormal where drugs are concerned depends on medical definitions, usually implicit rather than explicit.

Limitation by purpose in the legal regulation of controlled substances is so important that we tend to be suspicious of any proposal for using them that leaves the purpose ambiguous or vague. The modern doubt about what qualifies as religion makes it hard to justify drug use for religious purposes. The federal court in the Leary case must have wondered, apart from all questions about good faith, whether the petitioner's purpose in using marihuana was religious in any sense that differentiated it clearly from other purposes. The same difficulty arises with most social arrangements suggested for the use of psychedelic or hallucinogenic drugs. Any mixture of medical with religious and other claims now makes us uneasy. There have been proposals for psychedelic centers where people could take LSD and related drugs in a safe environment surrounded by helpful people. But the proponents do not know whether to conceive of these centers as analogous to resorts, religious retreats, psychiatric clinics, or scientific research institutions, and the ambiguity is enough to create suspicion. Our laws exclude such institutions not just incidentally but by the very categories in which they are framed.

The historical troubles of opiate maintenance illustrate this mistrust of ambiguity. The Supreme Court itself could not decide conclusively whether the addict's problem was a medical one and maintenance a medical solution. To some people, it has always seemed outrageous that we should give free opiates to addicts. Do we provide free whiskey to alcoholics or, if that example is too absurd, do we supply free ski lift tickets to skiing fanatics? These doubts about opiate maintenance are, of course, just one of the social complications created by the disease concept of drug abuse. It has occasionally been suggested that we establish recreational drugstores, analogous to liquor stores, but with the safeguard of requiring a doctor's permission for each customer — a procedure that resembles licensing for airplane pilots. The prospects of this idea are very poor, and not only because of the obvious administrative problems. What troubles so many people about opiate maintenance is precisely that it appears to be a kind of licensed recreational drug use that casts medical professionals in an inappropriate role. The practical problems of opiate maintenance schemes make this anxiety plausible.

Andrew Weil (Weil 1972) recommends that we use drugs for positive reasons, to enhance life rather than just to relieve boredom, pain, and misery; he is rejecting the standard medical notion of appropriate drug use, which is almost exactly the opposite. And for most drugs, medical use is the only kind considered appropriate. Yet by the World Health Organization's definition of health as "total well-being, physical, mental, and social," Weil is promoting health just as much as another doctor who thinks that no one who is already feeling fairly well should ever take drugs. What happens here is that the social definition of health varies according to the context. When we talk about the dangers to health caused by drugs, we tend to use the broadest possible definition of health in order to justify the strongest restrictions. When we establish legitimate purpôses for using drugs, of which health is obviously one, we try to define health narrowly so that again we can justify severe restrictions. Health as positive liberty — total well-being — is a legitimate reason for banning drugs but not for using them.

If any generalization can be drawn from this survey of the types of direct drug control, it is that they are all of limited effectiveness. The integration model of prudent, moderate drug use for everyone may be an excellent goal, but it is hard to see how to turn it into a policy. Legal controls, including criminal sanctions, have some effect, but we pay a substantial social price for them. Legally limiting drug use to certain classes of people is generally no longer acceptable. Drug use can be limited to narrowly defined religious or medical purposes, but this policy creates difficulties of its own and often proves irrelevant in solving the larger drug problem. Maybe we are not looking at things the right way when we devote too much attention to controls on drug use. Perhaps we can get a better view by moving back a little.