The Odd Politics of the Marijuana Tax Act

The passage of the Marijuana Tax Act of 1937 requires an extraordinary amount of explanation. Although the prohibitionist sentiment has been a constant element of our culture, major legislative initiatives to regulate or prohibit intoxicants have been episodic. Two circumstances have typically accompanied successful legislative initiatives to regulate intoxicants. One has been heightened public perception of excessive vice and its related problems — periods of relative strength for temperance movements. The second has been the availability of political coalition partners whose interests overlap those of the prohibitionists.

Regulating the use of intoxicants has not necessarily been a concern of all members of such coalitions. The period from 1790 to 1840, for example, saw a massive transformation of everyday life in the United States, including a strong church-based temperance movement and great reductions in the consumption of alcohol and the public acceptability of drunkenness. (Larkin 1989) Although the imposition of the excise tax on whiskey in 1791 may have been an early event in that temperance movement, other interests dominated the coalition. The political dispute over passage of the Whiskey Excise of 1791 expressed sectional rivalry between the western frontier and the states of the eastern seaboard; tested the newly claimed taxing power of the federal government; and instantiated the debate between the competing federalist and anti-federalist philosophies. (Slaughter 1986) That is, for many supporters of the excise, the crucial attributes of whiskey were its dominant place in trade in the frontier regions — its attractiveness as a revenue source — not its use as an intoxicant.

Commercial and professional interests have also typically been significant. The provisions of the Whiskey Excise of 1791 (Slaughter 1986) and the Pure Food and Drug Act (Musto 1987) both favored larger well-capitalized companies over smaller ones and both resulted, as anticipated, in the concentration of the relevant industries in large companies. Early versions of the Harrison Act were rejected until the coalition promoting it as a temperance measure and as a weapon against the "yellow peril" were able to secure the support of the medical profession and pharmaceutical industry. (Brecher 1972; Musto 1987; McWilliams 1992)

The extent to which the interests of coalition partners can diverge was demonstrated in the implementation of the Harrison Act. Prohibitionists in the federal bureaucracy advanced an interpretation of the act that removed the physicians' right to prescribe opiates in most cases and virtually eliminated the profitable commerce in opiates. Many physicians were arrested by their erstwhile coalition partners. (Brecher 1972; Musto 1987; McWilliams 1992)

None of these typical circumstances preceded passage of the Marijuana Tax Act of 1937. At that time, the use of marijuana as an intoxicant was not widespread, nor was there significant public perception of social or medical problems associated with abuse. In fact, attempts to arouse public concern about marijuana use through a media campaign failed. Prohibition had recently been discredited as a remedy and its repeal was extremely popular. The lurid rhetoric of the media campaign proposed to eradicate commerce and use of marijuana, but some political discussions within the Roosevelt administration and congressional testimony in support of the Marijuana Tax Act downplayed or denied the prohibitive intent and anticipated effects. There was no professional or commercial interest in the regulation of marijuana. The continued commercial use of hemp fiber, seeds, and oil was protected and there was no sign of competition among segments of the relevant industries attempting to achieve competitive advantage by promoting favorable provisions in the statute.

Medical use of marijuana was negligible. Although marijuana had historically been used medicinally, it had been all but replaced by superior alternatives. By the 1930s, marijuana had been dropped from the American Medical Association's list of useful drugs whose only concern in this regard was that research might reveal new uses. The representative of the American Pharmaceutical Association reported that marijuana was very rarely prescribed and expressed only the fear that cumbersome and expensive reporting procedures might lead pharmacists to refuse to fill prescriptions in the few cases in which they were warranted. ("Taxation of Marijuana" 1937) In short, passage of the Marijuana Tax Act was achieved without the initiative of any interests directly concerned with the commerce in marijuana. Put another way, there is no clear reason for the political success of the Marijuana Tax Act — the legislation proceeded without the apparent support of powerful interests to address an insignificant problem.

Explanations for the Passage of the Tax Act

When interest in marijuana reappeared during the 1960s, two sociological accounts (Becker 1963; Dickson 1968) were offered for the passage of the Marijuana Tax Act. Both began with the premise that there had been no problem of marijuana use prior to its passage — no reason to pass the act. And both postulated an extraordinary zealous crusade led by Harry Anslinger from his position in the Federal Narcotics Bureau to account for the passage of the legislation. In both explanations, the zealous crusade, a species of charismatic leadership, overrode tradition by creating new moral restrictions and extending bureaucratic action beyond the existing routine. Methodologically, both Becker and Dickson characterize Anslinger's motivation as a typification or ideal type of expected conduct in a class of circumstances, rather than as a historically particular occurrence.

Substantively, the two accounts differ only with respect to Anslinger's motivation in the crusade. Becker (1963) cast Anslinger as a moral entrepreneur, acting to establish new moral rules in service of his commitment to prohibition as a goal. Dickson (1968) conceded the importance of moral zeal, but argued that Anslinger's primary concern was bureaucratic — to protect his bureau's budget by expanding the areas of its authority.

Becker's and Dickson's accounts are both consistent with some aspects of the historical record. However, a more detailed examination undermines the substance of both accounts. There was indeed virtually no media interest in marijuana until Anslinger began to promote marijuana use as a serious concern and to provide information to the press, including lurid case studies. This resulted in a flurry of magazine articles and some newspaper editorials — an orchestrated media campaign. However, the media campaign was ineffective in arousing the public. Contemporary law enforcement officials cited the continuing apathy of the public as an obstacle to "the eradication of America's foremost drug." (Wolf 1936) Clinton Hester, assistant general counsel for the Department of the Treasury, testified that marijuana was especially insidious because the public failed to understand its fatal qualities. ("Taxation of Marijuana" 1937) That is, supporters of the Marijuana Tax Act claimed that action was needed because of public apathy and unawareness of a serious problem, not a public outcry.

Neither were members of Congress actively concerned with marijuana use. Questions by members of the Ways and Means Committee ("Taxation of Marijuana" 1937) indicated that many of them were not aware of marijuana as a problem until the tax act was introduced; were unaware of the content of the recent flurry of magazine and newspaper articles; were not previously familiar with the clinical research brought to their attention in support of the act; and were not familiar with the relationships among marijuana, other drug use, and crime, as represented in the literature of the day. Rather, as Musto (1970:106) summarized the situation, "the government's witnesses could also be fairly confident that the congressmen had no preconceived favorable or even informed opinions."

Even more problematically for Becker's and Dickson's explanations, Anslinger was not a rader in the promotion of the Marijuana Tax Act. Musto (1970) documented that as late as January of 1937, Anslinger regarded adoption of the Uniform Narcotic Act by the states as the appropriate means to control marijuana, although the administration had apparently already made the decision to pass national legislation. In a conversation with Musto (1970) Anslinger recalled doubting the constitutionality of the law and its ability to be passed by Congress. He recalled that the bill was political in origins, beginning with pressure from law enforcement agents and politicians in the Southwest due to problems with Mexicans and transmitted through the Secretary of the Treasury. The idea of a transfer tax, the core of the federal regulation, was not his, but is attributed to the Treasury Department's General Counsel's Office. Finally, even the idea for the lurid media campaign was not Anslinger's. At a conference among administration officers, Anslinger included lurid case studies in response to a request for proof of the dangers of marijuana. It was S.G. Tipton of the General Counsel's Office who recommended gathering a lot of such cases, and specified that horror stories were wanted. ("Conference on Cannabis Sativa L." 1937).

Musto (1970) correctly argued, contrary to Becker and Dickson, that the passage of the Marijuana Tax Act could not be attributed to activity at the Federal Bureau of Narcotics, nor to Anslinger's "determined will." He argued that passage of the Marijuana Tax Act must be placed in cultural and institutional context and explained as political action within the political system. Following Anslinger's recollections, Musto (1970) identifies the appropriate political context as a variety of concerns about Mexican immigrants, among whom marijuana use was relatively widespread. Among those concerns, was crime and deviant behavior by Mexicans associated with marijuana intoxication. The unrestricted Mexican immigration of the period was identified as the source of the growth of the marijuana problem.

Musto's (1970) replacement of typifications by historical particulars and his identification of an agenda and some interests served by passage of the Marijuana Tax Act were both steps in the right direction. But, inexplicably, Musto (1970) relies un-skeptically on the accuracy of Mr. Anslinger's recall and limits his review of evidence of the period to documenting the concern about Mexican immigrants recalled by Anslinger. As a result, evidence of the attractiveness of anti-marijuana legislation as a weapon against other identifiable groups was not considered. In addition, Musto limits his discussion of the impact of the new law to its manifest content — the control of marijuana. Its potential impact in terms of other agendas, such as the control of the labor market, is not considered. We are left with Mr. Anslinger's improbable explanation that a fear of marijuana, grounded in nativist sentiment against Mexican 'Americans, had sufficient strength to motivate legislation without widespread public or congressional interest and with no support from commercial or professional interests.

Mr. Anslinger's recall, however, is problematic. In his account of events, the origin of the movement to regulate marijuana was the concern of law enforcement officials in the Southwest, who were alarmed by its dangerous effects in the Mexican communities. But Morgan (1990) provides evidence that there was no widespread call for action from the Southwest. Instead, the alarm of law enforcement officials in the Southwest was exaggerated or fabricated to justify national marijuana legislation.

In addition, evidence from the period indicates that in addition to Mexicans, at least two other groups — African-Americans and Bohemian/intellectual elements — were also identified as marijuana users and targeted as dangerous. The typed script of a speech delivered by Mr. Anslinger (1938) to the National Catholic Welfare Conference mentions that the principle clientele of marijuana pads in New York was "colored men of the racketeer type and white prostitutes" listening to swing music. While the expanded target list may still not be complete, it suggests a broader and more credible base of interests for successful political action.

This discussion is guided by Cohen, March and Olsen's (1972) "garbage can" model of decision processes in organized anarchies. They argue that all organizations sometimes present decision situations with the characteristics of organized anarchies. In such environments problems that need definition as well as solution coexist with personal, perhaps secret agendas — prior, inflexible commitment to programs or means, whose adoption is recommended as a solution to any problem with which they can be associated. Appropriately defined and promoted, an agenda or solution can be successfully attached to a wide variety of problems and a problem can be addressed by a variety of means. That is, the conditions of organized anarchy, the organization acts as a sort of garbage can for ideas — problems and solutions — which can stick to one another in unlikely combinations. Thus, to understand organizational decisions — in this case the passage of the Marijuana Tax Act — we must identify the problem it addressed without assuming that the problem is related in a straightforward way to the manifest content of the solution. In one respect, this was recognized at the time. The act was explicitly intended to regulate indirectly through taxation, in an area in which direct regulation was reserved for the states.

In identifying the problem to which passage of the Marijuana Tax Act was the solution, the derivative publications that resulted from Anslinger's campaign are discounted. Policy-makers may have been concerned, however, by then current theories (Fossier 1931; Walton 1938) that marijuana effected different racial and cultural groups differently and was an important factor in the relative status and dominance of civilizations.

Other concerns had little to do with marijuana, but rather with other characteristics of the groups in which marijuana use was believed to be widespread. As a result of internal migrations the African-American communities in urban areas, especially in the North and West, grew rapidly and underwent fundamental changes in its class structure. Race relations in these areas deteriorated. The Mexican population in the United Sates quadrupled over a 20-year period, and Mexicans also participated in internal migrations. I argue that the act was passed primarily to advance the agendas and interests served by controlling those groups and a loosely defined group of intellectual elites who advocated unusual politics and Bohemian, unconventional life-styles. The law established a basis for control of marijuana users, which was desirable because of their other characteristics.

Rather than a puzzle, the limited scale and distribution of marijuana smoking and political indifference to it were strengths of this legislation. Public, political, and Business interests were mostly indifferent to this legislation because they were not significantly affected by its provisions. The law in its majesty may equally forbid the rich and the poor from stealing bread and sleeping under bridges, but it does not inconvenience them equally. This explanation characterizes the passage of this law as perhaps extraordinarily cynical, but certainly as part of the normal political process.

Mr. Anslinger's contribution appears to have been to provide sufficient rhetorical cover so that the supporters of the tax act need not express their own agendas publicly. The evidence for this interpretation is strong but partly circumstantial, requiring corroboration from further detailed investigation of how the coalition that supported the law was assembled.

The Dangers of Marijuana: Versions from the 1930s

The flurry of publications influenced by Anslinger's campaign complicate the task of assessing the knowledge and goals of policy-makers of the period. For these publications, although contemporaneous, did not enter policy decisions. Rather, their publication was encouraged and their contents influenced in service of policy decisions already taken. Since evidence indicates that they failed to arouse public and congressional concern, they did not even have the effect of recruiting additional support to the policy originators. And since part of the information distributed was intentionally selected lurid horror stories, these contemporaneous articles can not be assumed to accurately reflect professional expert opinion of the period. That is, in reflecting on the policy decisions of the period, the inflammatory, derivative publications must be discounted.

Six publications stand out as important primary sources of the period — Stanley (1931), Simon (1921), Siler et al. (1933), Bromberg (1934), Fossier (1931) and Walton (1938). None of these sources was influenced by Anslinger's media campaign — all but one being published before it began in earnest. To the contrary, all of these publications except Walton (1938) informed Anslinger's campaign. Stanley's (1931) article was included in the record of the congressional hearings concerning the tax act. ("Taxation of Marijuana" 1937) The other publications of the period are derivative of these few, either directly or through Anslinger's press release summaries of them.

Derivative Consensus on the Threat of Marijuana Use

In the derivative literature of the period, the threat of marijuana use was both crime and deviant behavior caused by and during marijuana intoxication. The effects of marijuana smoking, as described in these sources begin with feelings of hilarity. Perceptions of time and space are distorted. Mental excitation reduces the ability to focus attention, to sustain or direct a train of thought. Thinking becomes fragmented and disjointed. Initially, a stuporous state is induced which includes vivid dreams and hallucinations. With extended use, the stupor gives way to a state of high energy and delusional behavior. (Some sources, omitted the stuporous state and the distinction between initial effects and long term use.) The delusions and conduct are often sexual. There is a general loss of inhibition and emotional control, which leads to acting out impulses and conflicts that are ordinarily suppressed. This aspect of the effects of marijuana varies with the personality of the user. A destructive and violent rage occurs, which leads the intoxicated person to strike out at any convenient targets, without warning or reason. (Some sources attribute the destructive rage to the delusions or to specific delusional psychological processes as "persecution complex.") When the effects of the marijuana wear off, the person returned to normal, becoming indistinguishable from non-users. Some sources indicated that there was no memory of the intoxicated period. Sources differed concerning whether marijuana was addictive. In sum, while under the influence of marijuana, individuals are prone to criminal or delusional psychotic behavior, often sexual and violent.

The Threat of Marijuana in the Primary Literature

While the derivative literature was consistent with the views of law enforcement officials (Simon 1921; Stanley 1931) in the primary literature, it contradicted the primary clinical and statistical studies of the period (Bromberg 1934; Siler et al. 1933) in important respects. In addition, a major theme of the primary literature — the importance of racial and cultural differences (Fossier 1931; Walton 1938) — was not discussed in the derivative publications. Part of Anslinger's manipulation of the media was to omit that theme from his press releases and campaign, even while using other information from the sources in which it was discussed.

Marijuana and Military Discipline. Siler's (1933) report was the culmination of a decade of debate about the impact of marijuana smoking on military discipline and includes discussion of three separate studies over that period. During the 1920s military commanders became concerned about disciplinary problems caused by marijuana smoking in the Panama Canal Zone. Possession of marijuana was prohibited in the zone in 1923. In 1925, a committee studied the matter and found that marijuana was not habit forming and had no apparent deleterious effects on users. Disciplinary records were studied and only a small percentage of cases involving insubordination or violence could be attributed to marijuana. As a result, the prohibition of marijuana was ended in the zone in 1926.

But the officers in direct command of troops did not concur with the findings and a second study was commissioned. Accurate disciplinary and clinical records were generated and preserved during 1928 and the conclusion was reaffirmed that marijuana posed no disciplinary problem and that penalties should not be reinstituted. In 1930, however, a new commander declared marijuana to impair efficiency and reinstituted the prohibition. In 1931, a new committee was established to investigate the extent and effects of marijuana smoking. The percentage of marijuana smokers in the eight forts in the canal zone was estimated. In one, 20 percent of the men were estimated to use marijuana, while in the other seven forts, the percentage ranged from 5.4 percent to 0.6 percent.

A sample of known marijuana smokers were studied on a volunteer basis. Eighty-five percent were characterized as mentally abnormal — 62 percent constitutional psychopaths and 23 percent morons. Only 15 percent reported missing marijuana when deprived of it and, in fact, 71 percent of the subjects preferred tobacco, choosing it when both were available. The subjects reported using marijuana as a pleasant pastime when they were off duty and had nothing to do. Their impression was that almost all recruits tried marijuana and those that liked it continued its use.

The subjects were allowed to smoke marijuana freely during the study. While intoxicated, they became less reserved and more animated and talked foolishly. Neurological and mental tests were performed, and no ill effects were observed even when the soldiers were allowed to smoke freely for several consecutive days. There was no evidence of combativeness and destructiveness. Disciplinary records indicated that only 1.17 percent of soldiers brought to military trial had involvement with marijuana. This was less than the estimated percentage of marijuana smokers. Only three of 94 discharges for undesirable habits or traits of character were attributed to marijuana. The committee concluded that "commanders ... have unduly emphasized the effects of marijuana, disregarding the fact that a large proportion of the delinquents are morons or psychopaths, which conditions themselves would serve to account for the delinquency." (Siler et al. 1933)

Marijuana, Personality and Crime. Bromberg's (1934) study included psychiatric investigations of practically all (n=2,216) convicted Negro and white serious offenders in New York City during 1932- 1933. No confirmed marijuana addiction was noted. None of the assaults was committed under the drug's influence. None of the crimes was committed during or immediately after intoxication. None of the sexual crimes was due to marijuana intoxication. That is, no serious crime was associated with, let alone caused by, marijuana intoxication.

But Bromberg found a different causal role for marijuana in crime. Bromberg agreed that the drug was more widely used by psychopathic, unstable and socially inadequate types. While Siler (1933) argued that the underlying condition alone accounted for criminal actions, Bromberg (1934) argued that marijuana releases these tendencies in the individual. That is, marijuana uncovers and releases anti-social, aggressive, and sadistic elements in the personality if they are present. Once released, apparently, these tendencies need not be reactivated by intoxication.

Both clinical interpretations are consistent with both the extremely weak correlation found between marijuana intoxication and crime and the variable effects of the drug. While Siler's (1933) interpretation is more parsimonious, Bromberg's (1934) interpretation is more consistent with the subjective reports of marijuana users and the impressions of law enforcement officials and the military base commanders. Neither study is adequately designed to test this difference of interpretation. The interpretation of marijuana as a releaser of preexisting character defects may have reflected an occupational preference or interest. Some sources (e.g., McCracken 1937; Reznick 1937) drew the obvious implication that since the effect of the drug was to actualize anti-social potentials inherent in the character of the smoker, intervention must include not only the imposition of abstinence, but also extended psychotherapy.

Despite the unclarity of the role of marijuana as a trigger or releaser of underlying personality tendencies, the clinical studies are clear, and clearly contradict the information in the derivative literature in several key respects. The clinical studies indicated that marijuana was not addictive, nor even preferred. No crime could be attributed to marijuana intoxication. Observation and testing indicated no tendency to violence or aggressiveness while intoxicated.

The Racial Factor. Cannabis Sativa was native to Asia and was cultivated for fiber, oil and feed. It was transported westward through the Arabian regions, Europe, and Africa and introduced in the English Colonies as a planta-17th century. Although its
potential for use as a medicine and intoxicant was widely known, the extent of its actual use varied greatly among places where it was available.

Experts of the period developed a racial and cultural explanation for the hit-and-miss character of its adoption as an intoxicant in the various countries and cultures in which it became available. For example, use of marijuana was widespread in India, but not in China, widespread in the Moslem countries and Africa, but only widespread in a relatively small number of ethnic communities in Europe. First, the effects of marijuana were thought to be different in different racial groups (e.g., Bromberg 1934; McCracken 1937), although the sources disagreed on which groups suffered the greatest problems. Second, different races or cultures were thought to have different susceptibility to marijuana as a result of temperament or culture (e.g., Fossier 1931; Walton 1938). Where marijuana use takes hold as a widespread vice, entire cultures or nations could be degraded and destroyed (e.g., Fossier 1931; Walton 1938). Thus the prevalence of marijuana smoking was offered as a key indicator and explanation of the differential status of races and nations.

Fossier (1931, 249) argued that the use of intoxicants was a necessary evil at this stage of civilization, but substituting opiates or hashish for alcohol "could be considered the greatest calamity that can befall a nation." "The dominant race and most enlightened countries are alcoholic, whilst the races and nations addicted to hemp and opium, some of which once attained to heights of culture and civilization have deteriorated both mentally and physically."

Walton (1938) argued that two conditions must be present for the vice of marijuana use to become established. First, there must be contact with a group that already indulges in the vice. Second, the contacted group must be susceptible due to constitutional weakness of temperament or flaws in the culture. He drew a contrast between Kentucky pioneers who cultivated hemp for fiber but had no record of the vice and the idle irresponsible classes in modern America among whom the vice was spreading. He identified the vice as widespread among Negroes and Mexicans. His theory sees a special danger to children whose character is not yet firmly established, and could be extended to any cultural or ethnic subgroup that might have weakness of temperament or culture. For example, he indicates the vice is being established among decadent wealthy circles in London.

Interaction of Clinical and Social Beliefs. The susceptibility of a class, race or nation to the use of marijuana as an intoxicant is related to a constitutional defect of the people or to flaws in the culture. That is, dangerous flaws of character or personality may be widespread in the constitution and culture of a group. At the individual level, traits of character or personality determine which individuals are susceptible to indulgence in marijuana smoking and which will suffer from the release of personality defects. Thus, Bromberg (1934) observes that the smoking of marijuana is more common among Negroes, Mexicans, and Puerto Ricans in New York than among people of American or European stocks. Marijuana users in the latter group are more likely to be psychopaths. Although the lurid case studies of the era focus on individual criminals and their pathologies, the true menace in the primary literature is the spread of the vice through contact between different peoples, resulting in the establishment of the vice among an inferior people which would produce violent, delusional criminals in very large numbers. Bluntly put, in superior peoples and cultures, the extent and consequences of marijuana smoking would be limited by the rarity of psychopaths and morons while in inferior ones the extent and consequences would be unlimited.

Anslinger emphasized the contamination of new groups through social contact in his testimony and placed a letter containing the following passage in the record: "I wish I could show you what a small marijuana cigarette can do to one of our degenerate Spanish-speaking residents. That's why our problem is so great; the greatest percentage of our population is composed of Spanish-speaking persons, most of whom are low mentally, because of social and racial conditions." ("Taxation of Marijuana" 1937)

Prohibition as a Remedy. The secondary literature of the period unambiguously recommends eradication of the use of marijuana — a prohibition. However, there is some evidence that policy-makers faced resistance within the government to a new prohibition. During a discussion of the proposed legislation ("Conference on Cannabis Sativa L." 1937: 32-33) the issue of the application of the prohibitive tax for growing marijuana for personal use was raised explicitly. Mr. Tipton of the General Counsel's Office, replied "No, by paying $25 I think you can grow and smoke all the marihuana you like, yourself." During the congressional hearings ("Taxation of Marijuana" 1937: 81-84) the issue of uncultivated marijuana growing as a weed was introduced. The concern was raised by Mr. Vinson of Kentucky, a state in which marijuana had been grown as a plantation crop, that the law might impose the cost of very difficult and expensive weed eradication on farmers or landowners. He was assured by Mr. Hester, Assistant General Counsel for the Department of the Treasury, that no such cost was imposed. The landowner could pay the non-prohibitive occupational tax and allow the weed to grow. Payment of the tax would provide information to the government about the location of the weed, allowing action at the government's cost.

Were they genuine, these two assurances would amount to a limitation of federal influence to controlling commercial traffic in marijuana, while not prohibiting nor even heavily taxing use, cultivation, or possession. But these assurances were disingenuous, because as other testimony indicated, one effect of the tax act would be to permit the Bureau of Narcotics to lend support to the enforcement of state laws. The disingenuousness of these responses suggest that the marginally informed and relatively disinterested acquiescence to the act would not necessarily extend to involving the federal government in another prohibition.

Groups With High Recognized Incidence of Marijuana Use

The sources of the 1930s (e.g., Simon 1921; Walton 1938; Bromberg 1934; Literary Digest, Oct. 1936; Parry 1935; Fossier 1931) were in general agreement that marijuana use was widespread among African-Americans, Puerto Ricans, West Indians, and East Indians, as well as Mexicans. In addition, widespread drug use was attributed to a loosely defined cultural group from the "leisure classes, where unrestricted vice may be maintained secretly in an atmosphere of supposed refinement." (Simon 1921) Among these are artists, writers and slumming millionaires. Parry (1935) reported marijuana use among "the sickly aesthetes," Bohemian intellectuals and residents of Greenwich Village. A similar group was indicated to exist in London (Walton 1938). The use of marijuana by musicians, artists and writers and its prominence in the night life of the period were frequently noted (e.g., Literary Digest 1936; Parry 1935). This loosely defined group is characterized by unconventional values and life-styles and radical political and philosophical views. Communism was among the troublesome political views held in this group. In addition to opposition to Communists within organized labor, sources of the period link rumor-mongering by Communists to the bank failures of the Depression. (Literary Digest 1936) Several sources specifically mention the mingling of African-Americans and whites in various social contexts as part of the involvement of the Bohemian/intellectual types with marijuana.

Trends Between the 1920 and 1930 Censuses and the Location of Marijuana Arrests: The Context of the Marijuana Tax Act

Demographic and Social Trends. Concurrent with the 1930 census, President Hoover established a committee of scientists to review social trends in the United States, including the status of racial and ethnic groups (Woofter 1933) . Two important trends reflected in census data were: 1) a very large increase in Mexican immigration to the United States, attributed to instability in Mexico and prosperity in the United States; and 2) the internal migration of Mexicans and Negroes within the United States. This internal migration was attributed to the impact of immigration quotas established during the 1920s on supplies of labor in northern cities, and, in the case of African-Americans, worsening racial relations in southern states.

As a result of immigration, the number of Mexicans grew from 381,000 in 1910 to 701,000 in 1920 and more than doubled again to 1,423,000 in 1930. During this period, the geographical distribution of the Mexican population within the United States also changed. Between 1920 and 1930, the percentage of Mexicans living away from the border counties of Texas, Arizona and California increased 20 percent from 15 percent to 18 percent of the total Mexican population. Coupled with a quadrupling of the population base over a 20-year span, this resulted in dramatic increases in the Mexican populations of Oklahoma, Colorado, Kansas, Illinois, Michigan and Indiana. New York, New Jersey and Pennsylvania also experienced increases. In addition to the rapid increase in the size of Mexican communities throughout the country, internally migrating Mexicans became visible by beginning to work in industry in addition to migrant, seasonal agricultural work.

During the Depression, opposition to Mexican labor developed from labor unions, which pressed for greater controls on illegal immigration from Mexico and for quotas on Mexican immigration. (cf, McLean 1930) In California, a specific concern was the high expenditures on welfare to support seasonal Mexican workers during their annual period of unemployment. (Morgan 1990; Sowell 1981) During the 1930s, Sowell (1981) reports, there was a systematic campaign to deport Mexicans by a variety of means other than formal deportation hearings. Some Mexican applicants for relief were referred to agencies that deported them instead. Relief was sometimes denied until the person agreed to voluntarily leave the United States. He estimates that about 9,000 Mexicans were deported each year during the 1930s and that the Mexican population in the United States decreased by about one-third. Only a few months after the passage of the Marijuana Tax Act, Walton (1938) reported that arrests under the Marijuana Tax Act were also used to deport Mexicans.

During the 20-year period from 1910 to 1930, the African-American community transformed itself by migration to urban settings. In 1910, 2,685,000 African-Americans lived in cities while, 7,143,000 lived in rural areas. By 1930, 5,194,000 lived in urban settings and 6,697,000 lived in rural areas. This trend had accelerated between 1920 and 1930 and was continuing. Although the African-American population in southern cities grew as well, the northern and western cities received the biggest influx of migrants. Men were overrepresented in the first wave of migrations, during the period between 1910 and 1920, producing gender imbalances in both the rural south and the cities. The surplus of young men in northern cities increased delinquency. Gender balance was reestablished by the increased migration during the 1920s.

Woofter reports two contrasting social trends during this period. Urban African-Americans became established in machine industries and industrial labor, moving beyond the menial unskilled positions to which they had originally been confined. Surveys of the period indicated that their reliability, efficiency and turnover were about the same as those of whites. While still concentrated in unskilled jobs, the percentage of African-Americans working in production jobs, as clerks and as foremen grew. The number of businesses owned by African-Americans increased and efforts were being made to persuade white business owners serving the African-American community to employ African-Americans. Sowell (1981) reports that prior to the internal migrations, residential segregation was low as well.

At the same time, Woofter (1933) reported increasing resistance to the upgrading of opportunities for African-Americans from organized labor and from the white customers of businesses serving both communities. This resistance was seen at the time as similar to that encountered by other migrant groups. Sowell (1981) observed that the internal migrations completely altered the class structure of the urban African-American communities. Originally small, they were stable working class communities with a growing middle class. But the migrants were illiterate, unskilled rural people who were rejected by both the white and black middle classes. Among the problems of the day were inferior schools, disruption of family structures due to the gender imbalance created by uneven migration, and increased disease. As the huge migration changed the character of the African-American urban communities, the white community responded with an increase in discrimination against the entire community. As the number of African-Americans increased in northern cities, the residential neighborhoods became more segregated and public accommodations such as restaurants and theaters became segregated. (Woofter 1933)

A pattern of discrimination against African-Americans in the legal system was also beginning to be documented. The crime rates in the African-American community could only be evaluated when the tendency to arrest African-Americans "on suspicion" was taken into account. (Woofter 1933) Sutherland and Gehlke (1933) reported that African-Americans had higher rates of being both charged and convicted of both misdemeanors and felonies than whites, and that these rates were growing.

Marijuana Arrests and Seizures under State Laws. Arrests and seizures under state marijuana laws just prior to the passage of the act were not concentrated in the border states ("Taxation of Marijuana" 1937; Walton 1938). Of 18 arrests nationally in 1936, only one arrest was made in Arizona and 17 in Texas. (Arrests in California were excluded because arrests in its border counties could not be distinguished from those in other areas to which there had been internal migration of African-Americans and Mexicans.) This is consistent with Anslinger's alarm at the "spread" of marijuana throughout the country, but the marijuana crime wave in the areas of concentrated Mexican population is contradicted as Morgan (1990) suggested.

But neither are the arrests randomly scattered. Rather tfiey are concentrated in areas to which there had been internal migrations of Mexicans and African-Americans. The newspaper editorials cited in Anslinger's testimony, calling for national legislation were from Washington, D.C., Chicago and St. Louis. The states of Illinois, Missouri, Michigan, Ohio, New York, and New Jersey reported significant increases in marijuana use. If marijuana use was a genuine concern at all, it was only a concern where its users had recently migrated and were unwelcome for many reasons.

Summary

The Marijuana Tax Act of 1937 was cynical legislation. It was acknowledged by its supporters to circumvent the constitutional reservation of rights to the states and people by using taxation to regulate. In our terms, the solution of taxation was utilized to solve the problem of regulation. But regulation of what? I have argued that it was also cynical in the sense that support of the act was motivated by concerns to control the labor markets, Mexicans, African-Americans and intellectual/Bohemian elites, rather than concern about marijuana per se.

Media concern with the issue was largely the product of a campaign undertaken by Mr. Anslinger and the Federal Bureau of Narcotics. It did not succeed in arousing public or congressional alarm. In informed circles, arguments supporting control of marijuana were not addressed to the problem found in the derivative literature — psychotic episodes brought on by marijuana intoxication leading to crimes of violence and sexual excess. Statistical and clinical studies of the period found no evidence of crime caused by marijuana and no tendency to violence aggression or sexual assault. Instead, the expert literature of the period and, to some extent, congressional testimony express concern with the consequences of marijuana use among socially, culturally, and constitutionally inferior races and peoples and the threat they posed to American society. Other social trends of the period suggest that this problem was important in the passage of the act. Mexicans and African-Americans were the target of varied social control efforts as a response to: 1) their internal migration to new areas of the United Sates; 2) changes in the labor market caused by immigration quotas established in the 1920s; and, of course, 3) the Depression of the 1930s.

A strong , if partly circumstantial, case emerges that the main purposes of the act included: 1) a means to attack and control groups living and encouraging alternative Bohemian values and political views; 2) the establishment of de facto immigration restrictions as a labor market intervention and to limit relief expenditures on Mexicans; and 3) the establishment of de factoJim Crow laws, contributing to the growing discrimination in the justice system and to efforts to increase residential and employment racial segregation in the northern cities.

Discussion

The historical record amply documents contemporary concern for the several agendas discussed in this paper. However, the link to the Marijuana Tax Act is offered as a hypothesis. An attempt to gather direct evidence for this hypothesis should focus within the New Deal coalition on the "Dixiecrats" who had the most obvious nativist agenda of the period, the labor unions, and fervent anti-Communists. Since Musto estimates that the decision to seek this legislation was made in 1935, the period to be investigated should probably begin shortly before the congressional elections in 1934.

The passage of the Marijuana Tax Act is instructive concerning decisions in organized anarchies. First, in this instance, the reasons for supporting the act and its anticipated outcomes were quite different from the reasons expressed and publicized. In such situations, rational, principled discussion of the manifest issues will not be effective, nor even germane. This suggests an important limit on the utility of rational response as a means for influencing organizational decisions.

In addition, evaluation and defense of the outcomes becomes complex. A program may fail with respect to its stated intentions, while satisfying the unstated purposes of its supporters. In this instance, for example, the tax act failed to control marijuana use, but did allow the deportation of Mexicans, arrests of African-Americans, and, at least later during the 1960s, the infiltration and disruption of politically troublesome groups. In other words, it was successful in terms of several underlying agendas.

This suggests that program evaluation can only be effective when fixed criteria are utilized. Flexible criteria are subject to the cynical exploitation of mission creep. In program evaluation, coalitions favoring the continuation of programs will invariably include and be informed by occupational groups created by the program. Expert opinion, then, which greatly overlaps these occupational groups, will be strongly interested in extending the life of programs by substituting untested evaluation criteria for criteria in terms of which the program has demonstrably failed. In a cynical coalition, a cycle can be created which extends program life indefinitely by repeatedly redefining the criteria for evaluation and extending program life by the time required to apply the new criteria.

The Marijuana Tax Act of 1937 is almost certainly unconstitutional. It has had not only had disparate impact on different ethnic groups as a matter of fact, but was designed to do so. •

Warren Handel is a professor in the Department of Sociology and Social Work at Southern Illinois University, Peck Building, Edwardsville, IL 62026-1455.

REFERENCES

Literary Digest 108. "Whispering reds as bank wreckers." January 1931: 10.

Scientific American. "Marihuana Menaces Youth." March 1936: 150-151.

Literary Digest. "Facts and Fancies About Marijuana." October 24, 1936: 7-8.

1937. Newsweek. "Marihuana: New Federal Tax Hits Dealings in Potent Weed." August 14, 1937: 22-23.

"Conference on Cannabis Sativa L." 1937. Box 9, File 22. Harry J. Anslinger Collection. Pattee Library. Pennsylvania State University.

"Taxation of Marijuana." 1937. Hearings Before the Committee on Ways and Means. House of Representatives. Seventy-Fifth Congress, First Session on H.R. 6385.

Scientific American. "Marijuana More Dangerous than Heroin or Cocaine." May 1938: 293.

Becker, Howard. 1963. Outsiders: Studies in the Sociology of Deviance. New York: Free Press.

Brecher, Edward. 1972. Licit and Illicit Drugs. The Consumers Union Report. Mount Vernon: Consumers Union.

Bromberg, Walter. 1934. "Marijuana Intoxication: A clinical Study of Cannabis Sativa Intoxication." American Journal of Psychiatry 91. September: 303- 330.

Brown, Clair. 1938. "Marihuana." Nature Magazine. July: 271-272.

Cohen, Michael, James March and Johan Olsen. 1972. "A Garbage Can Model of Organizational Choice." Administrative Science Quarterly 17, March: 1-25.

Curtis, Howard. "Psychosis Following the Use of Marijuana with Report of Cases." Journal of the Kansas Medical Society 40: 515-517, 526-528.

Dickson, Donald. 1968. "Bureaucracy and Morality: An Organizational Perspective on a Moral Crusade." Social Problems 16, Fall: 142-156.

Fossier, A.E. 1931. 'The Marijuana Menace." New Orleans Medical and Surgical Journal 84, May: 247- 252.

Larkin, Jack. 1988. The Reshaping of Everyday Life 1790-1840. New York: Harper.

Lewitus, Victor. 1936. "Marihuana." American Journal of Nursing Volume XXXVI, No. 7: 677-678.

McCraken, W.H. 1937. "Cannabis Sativa." Journal of the Michigan State Medical Society 36: 848-849.

McCormack, George. 1937. "Marihuana." Hygeia. October: 898-899.

McLean, Robert. 1930. 'Tightening the Mexican Border." The Survey 1, April: 28-29, 54-56.

McWilliams, John. 1992. 'Through The Past Darkly: The Politics and Policies of America's Drug War." Pages 542 in William O. Walker III (ed). Drug Control Policy. University Park: Pennsylvania State University Press.

Morgan, Patricia. 1990. "The Making of a Public Problem: Mexican Labor and the Marijuana Law of 1937." Pages 233-252 in Ronald Glick and Joan Moore (eds), Drugs in Hispanic Communities. New Brunswick: Rutgers University Press.

Musto, D.E. 1987. "The History of Legislative Control over Opium, Cocaine and Their Derivatives." Pages 37-72 in Ronald Hamowy (ed.), Dealing With Drugs. Lexington: Lexington Books.

Musto, David. " 1970. The Marihuana Tax Act of 1937." Archives of General Psychiatry 26, February: 101-108.

Parry, Albert. 1935. "The Menace of Marijuana." American Mercury. September - December: 487490.

Reznick, Sara. 1937. "Marihuana Addiction." Social Work Technique 2, July: 173-177.

Siler, J.F., et al. 1933. "Marijuana Smoking in Panama." The Military Surgeon 73, November: 269- 280.

Simon, Carleton. 1921. "From Opium to Has Eesh." Scientific American. November 1921. 14-15.

Slaughter, Thomas P. 1986. The Whiskey Rebellion: Frontier Epilogue to the American Revolution. New York: Oxford University Press.

Sowell, Thomas. 1981. Ethnic America. New York: Basic Books.

Spencer, R. 1936. "Marijuana." Health Officerl: 299-305.

Stanley, Eugene. 1931. "Marihuana as a Developer of Criminals." American Journal of Police Science 2: 252- 261.

Sutherland, Edwin and C.E. Gehlke. 1933. "Crime and Punishment." Pages 1114-1167 in Recent Social Trends in the United States II. Report of the President's Research Committee on Social Trends. New York: McGraw Hill.

Wolf, William. 1936. "Uncle Sam Fights a New Drug Menace ... Marihuana." Popular Science Monthly. May: 14-15, 119-120.

Woofter, TJ., Jr. 1933. 'The Status Of Racial and Ethnic Minorities." Pages 553-601 in Recent Social Trends. Volume I. York, Pa.: Research Committee on Social Trends.