As Howard Becker pointed out almost two decades ago, drawing a random sample of marijuana users is an impossibility. No list of all users, or even a large number of users, exists. There are several organizations concerned with marijuana. LEMAR, for instance, as its name implies, has as its goal the legalization of marijuana. The "Jade Companions" offer legal assistance to those arrested for the possession of psychedelic drugs. It would seem that these organizations provide a starting point for the collection of an informal sample of marijuana smokers. One problem with approaching an organization of this kind is that each one, for good reason, fears publicity, police surveillance, and harassment. LEMAR, for in stance, would certainly attempt to keep a listing of their member ship from falling into the hands of anyone outside its organization. Since the existence of the organization is a matter of public record, they are, even without notoriety, open to the possibility of harassment. If it were known that a sociologist had interviewed their members, the further likelihood of their attracting incriminating attention would be multiplied several times. In fact, at the first organizational meeting of the Stony Brook campus chapter of LEMAR, potential members were dissuaded from joining if they were presently users of the drug; "If you smoke, don't join," they were urged. This advice protects both the individual in that his member ship, if known, would automatically cast suspicion on the legality of much of his behavior, as well as the organization, since a large number of members who are vulnerable to arrest threatens its stability and existence. In any case, none of the individuals associated in a leadership capacity with the drug-related organizations whom I contacted was willing to cooperate with the study. Not wanting to threaten their already dubious relationship to the law and law enforcement agencies, I respected their unwillingness. It was apparent that a less formal means of sample recruitment had to be found.
One of the main channels of access that I used to collect respondents was through acquaintance with individuals who occupied positions in organizations which, although in no way formally drug-related, included marijuana users. This segment of the sample was generally gathered by going with the "gate keeper" individual to the place of employment and getting the names of users willing to be interviewed. Many interviews were conducted on the job, either during the lunch hour or a lull in work; others were carried out after working hours, usually at the residence of the interviewee. These organizations included two large New York universities, a large publishing house, and a market research firm.
The second source of my sample was through friends and their acquaintances. A kind of "snowball" method of gathering names was adopted, whereby each interviewee would supply me with one or two names of people who also used marijuana. Often the original person would contact his acquaintance and ask if he would consent to an interview; in this case, he received the refusal or the acceptance, not I. Frequently, I initiated the contact. Considering the illegal nature of the activities I questioned them about, the number of refusals was negligible. In fact, when I contacted the individual directly, only four refused. As to the rate of refusal through the indirect route, I cannot estimate.
The main concern was that I might be a policeman. It is puzzling to me as to why this was so, but their initial fears were fairly easily quelled; after all, were I actually a policeman, I would reassure them that I was not in the same soothing tones. Perhaps, like everyone else, marijuana smokers often react to stereotypes, and I certainly do not look like a policeman. The second worry was that their names would be kept and used, that it would be known publicly that they were lawbreakers. I was careful to assure them of anonymity, and to explain my procedure as to the use of names which invariably eased their doubts. It seems strange, but these two worries, that I was a policeman and that their names would be taken down, were articulated in a minority of cases; in about two-thirds of the cases, this was not necessary. It seemed that my original contact had vouched for my veracity, and that was apparently sufficient.
It should be emphasized that this is not a representative sample, and is in no way a cross-section of all marijuana users. At this point, the collection of such a sample is impossible. Therefore, to use the composition of this sample as an accurate description of marijuana smokers in general would be completely fallacious and misleading. To reason, for instance, that since 47 percent of my respondents were female, the same percentage of all users are female, would be to distort the meaning of this study. I am not presenting a profile of marijuana smokers, but an analysis of the social structure of marijuana use. This unstructured manner of collecting interviewees for a study of a deviant and illegal activity has both advantages and drawbacks. The potential interviewee in a complete stranger situation will normally fear detection by law-enforcement agencies, and will be unwilling to be interviewed in the first place, or, if willing, would be evasive and even dishonest. Cooperation, then, would, have been problematic had more formal techniques, such as neighborhood sampling, or drawing from a complete listing of individuals working in an organization known to include high proportions of individuals who smoke marijuana. What was necessary for me to be able to approach my interviewees was that someone in the network of social relations be able to vouch for my veracity and basic harmlessness. Only in this way was the cooperation of my interviewees assured. Moreover, this method avoids the oft-trod route of studying individuals who have come to some sort of official noticeincarcerated criminals, for instance, or those who have made some sort of court appearance. As we now know, deviants who have attracted some form of official notice present a systematically biased view of any group under study, unless, of course, that group under study is individuals who have attracted public notice.
The severe and restricting drawback to this casual and informal technique of gathering respondents is, of course, that the interviewees were certainly not representative of marijuana smokers in general; moreover, just how unrepresentative they were, is unknown. The specific individuals known to me reflect my personal associations; another researcher with a different set of acquaintances would have drawn a somewhat different set of respondents. And the organizations to which I had informal access do not necessarily house a cross-section of the marijuana-using population. Moreover, the persons whom the original contact designates as a potential interviewee will be distinctive in crucial ways. For one thing, he must be willing. For another, those who are so designated are likely to be obvious and conspicuous enough users (to their friends, at least) for the designator to think of him off-hand specifically as a marijuana smoker. (In spite of the fact that I requested users of every level of use.) In all, the sources of bias were strong and have many ramifications. For this reason, we must consider this study exploratory, and its findings tentative. We hope that the guesses and hypotheses suggested by our data will be tested subsequently by more careful instruments.
Since our sample is unrepresentative, it is important to describe its composition. The respondents were slightly more than half (53 percent) male, relatively young (the median age was twenty-two, and three-quarters of the respondents were in their 20s), and overwhelmingly white (8 percent were black, and five respondents, or 2.5 percent, were Puerto Rican.) Slightly over a quarter (27 percent) had parents with a Protestant background, 44 percent were Jewish, and about a seventh had Catholic parents (14 percent) or had parents with different religions (15 percent). Not quite four-fifths (78 percent) were single, and a tenth were divorced. A high proportion were students; 4 percent were high school or grade school students, a quarter (27 percent) were college students, and about a tenth (11 percent) were graduate students. The occupations of the remaining respondents were professional, technical, or kindred, 26 percent; managerial, official or proprietor, 4 percent; sales or clerical, 16 percent; manual laborer, 5 percent; unemployed, 5 percent; housewife, 3 respondents. A third of the fathers of the respondents (34 percent) were professional, between a third and a quarter (29 percent) were managerial, officials or proprietors, and a quarter (24 percent) were manual laborers; the remainder (14 percent) was made up of salesmen or clerical workers
Of the respondents who were not at the time of the interview in school (or, if the interview was conducted in the summer, who did not plan to attend school in the fall), not quite half had dropped out either of college (4/5 of this group) or of high school. In fact, about 25 percent of the total sample was a college drop-out. (It is, of course, impossible to estimate the likelihood of these respondents returning to college. It should be kept in mind that only about half of all those in general who enter college actually receive a bachelor's degree.) About 10 percent of those not presently attending school had a graduate degree, and about twice this number either attended some graduate school without receiving any degree, or had received a B.A. without attending graduate school. All of the respondents were residing in New York or its suburbs at the time of the interview (although a few were in transit); our findings, then, will apply most directly to the New York subset of marijuana smokers, and only by inference to users elsewhere in the country. The data are probably without application outside the United States.
An estimate as to the degree to which my sample varies from the large and unknown universe of all users would be sheer speculation, of course. I suspect, however, that the following differences would be observed between a random sample and mine:
- A random sample of all marijuana users would be overwhelmingly maleprobably about three-quarters.
- It would be more heavily black.
- It might be slightly younger, possibly at a median age of about nineteen.
- It would contain a lower proportion of individuals with any contact with college.
- It might include a lower proportion with a middle-class background.
- Fewer would be Jewish, more would be Catholic and Protestant, and very few would have a mixed religious background.
The interviews took place between February and September 1967. Rapport with the interviewees was, on the whole, excellent. Not one interview was terminated by the interviewee. (I terminated two; one received too many telephone calls for me to finish the interview, and later attempts at scheduling proved fruitless, and the second was a psychotic whose answers bore no relation to the questions.) Many interviewees reported that the interview was interesting; my rapid pace kept their interest from flagging. More important, of course, is that I believe that I received honest answers, although more than a casual check is impossible. I was careful to point out inconsistencies when they did occur, and I rarely allowed vague answers to pass unclarified. Most of the individuals interviewed were supplying information about felonious acts. Although relatively few reported serious crimes beyond drug use, the few that did appeared to be candid about it, although wary. When I asked one chronic user of amphetamine how he was able to pay for such heavy drug use he thought for a moment, turned to the tape recorder, which was running, and said, "Could you turn that thing off? He then proceeded to divulge the nature and frequency of the crimes which he did commit. When I asked a former heroin addict the same question, she responded shyly, "I was a prostitute." Undoubtedly, there were some evasive replies, some probably lied. But I believe that, given the nature of the enterprise which they were describing, this was minimal and exceptional, and certainly not characteristic.
The author conducted all of the interviews (except two). About half were conducted at the interviewee's place of domicile (or, rarely, at that of a friend), a quarter were conducted at his place of employment, and perhaps another quarter was done at the author's residence. A scattered few were done in public placesa coffee house, a restaurant, a bar. The first fifty were tape-recorded, and the remaining one hundred-fifty were transcribed almost verbatim.
There is, of course, the matter of variables which influence interview rapport. It is possible that the rapport is greatest when the characteristics of the interviewer are the same as those of the interviewee, although there are important exceptions to this, especially for certain kinds of information. During the period of the interviewing, I was twenty-eight years old, while the median age of the respondents was about twenty-two. I was in fact told frankly by a half-dozen respondents that had I been noticeably older, they would not have been willing to be interviewed. Another important factor was attire. To have done the interviews in a suit, white shirt and a tie would have threatened rapport; at the very least, it would have placed a chasm of distance between myself and the respondents. My dress was always informal, usually no different from that of the interviewees. In general, dress may be considered a part of the "hip-straight" continuum. (The word "hip" is both an adjective and a noun. It is permissible to speak of "having hip." "Hip" is also used as a verb: "I hipped him to the scene.") To the users who did think of themselves as "hip," hair style played an important role in their identification, as well as in the identification of others. During the period of the interviewing, I had very long and shaggy hair. Although I did not grow my hair long for the study, it had a peculiar relevancy for many respondents which I had not anticipated. Since the "hip" style is itself so variable, or, at least, there are degrees of "hip," many participants in this subculture might have thought that my innocuous style was exceedingly "square." As I was walking into an East Village artist's loft, two members of a motorcycle cult walked out, claiming that they felt "bad vibrations." It should be noted that among many "hippies," the mere fact of wanting to conduct an interview is "square." It is possible that this does not indicate the success of my unintended disguise, but I was approached several times on the street in the "East Village": "You want to cop some grass, man?" (Sometimes hashish as well.) It is possible, however, that anyone who appeared to have any money would have been approached. Other indicators of my ability to blend into the marijuana scenery were several tribal greetings which I received from people unknown to me on the East Village streets, as I went to and from interviews: the extension of an open palm (which calls for slap into the palm), spoken tokens of phatic communion ("Like, what's happening, man?"), and similar examples of communication Once, while sitting in a coffee shop, three young girls in their middle 'teens, in "hippie" garb sat across the counter. (I quote from my field notes):They smile and wave. To me? I look around. It looks like me. One says, "How's your trip?" Stupidly, I ask: "What trip?" "You know," one says, giggling, with her hand over her mouth, "LSD." Slowly realizing what's going on, I ask "How did you know?" "Oh, we can tellit's the glint in your eyes." Oh, yeah. Anyway, here's the scene: they took me for a hippie. Why? What are the status cues?
Another indicator by which respondents, actual or potential, may have sensed my lack of connection with law-enforcement agencies and were, therefore, willing to confide in me, was in my understanding and occasional use of slang terms. Although some are in current use everywhere, such as "pot" and "grass," others are somewhat more esoteric: "spaced," "zonked," "blind," "wiped," to mention only a few for the notion of being "high." It is interesting that many of the respondents used these terms freelynone ever gratuitously explained them to meassuming that I understood them.
It should be noted that any of these indicators of one's "purity" i.e., lack of affiliation with the policecould be faked. Undercover agents (such as the ones planted on Stony Brook's campus) learn the argot, the manner of dress, the style of life, tonsorial cut (or lack thereof), and so on. In fact, I interviewed an undercover agent, a policeman who possessed all of the necessary appurtenances. The user's naive faith in style leads him to believe that he is able to sense a threat to him, when in fact, I doubt very seriously if this is the case. When asked how he knows something, a "hip" marijuana-using resident of the East Village will often reply, "Vibrations, man." It must be remembered that at least half of my respondents thought of themselves as in no way involved in the "hip" way of life. I interviewed Wall Street lawyers and corporation executives as well as "hippies," dealers, and unemployed wanderers who remained high most of their waking hours. For many, marijuana did not change their basic style of life. Had the lawyers and executives who smoked wandered into the habitat of the "hippies," they would have been thought of as "freaks." (For some strange reason, the term "freak" has a dual usage. It refers to a "square" who would be egregiously out of place in a "hip" environment, and a "hippie" who would be equally unfit for a place in "square" society. Both would be freaks' in the opposite setting. When asked why he wore a silver lame jacket on the stage when it was so hot, a rock performer replied "Because I'm so freaky." It was, of course, a boast. Usually, however, the term "freak" has negative connotations.)
Therefore, these comments on "hip" only apply to a segment, certainly less than half of the sample in full degree, since "hip" is a matter of degree. But as a qualifier to this qualifier, it should be noted that the more one was involved in the drug scene, and with drug use, the more that one was likely to display a "hip" style of life.
Perhaps in one respect, my rapport with the respondents may have reduced the amount of information which I received, at least before I began to probe and ask for elaboration. It was often assumed that I knew what the respondent was describing. I received the response, in the middle of a description: "Oh, I don't have to go onyou know what I mean!" A pose of complete innocence was not possible in many cases, although it is possible that this approach would have yielded more information.
In addition to the responses to the formal interview, I observed a great deal of drug-related behavior casually; I was a "participant observer." In addition to the interview situation, I interacted informally with many of my interviewees. Knowing that I was doing the study, as well as for more personal reasons, they often invited me to observe and take part in various drug-related social events, such as parties, the "turning on" of a curious potential marijuana smoker, the baking and eating of various foods in which marijuana was cooked, feasts and dinners eaten while high, LSD trips, "be-ins," "smoke ins," concerts (listened to while high), drug sales and transactions, and so on; I was called twice to calm frightened LSD "trippers," so that I spent two evenings doing just that (which taught me both that LSD is not the harmless drug it is sometimes portrayed by drug users to be, and the difference between temporary panic, and hospitalization is often an understanding guide). I would estimate that I observed about two or three thousand man-hours of marijuana use in the eight months of the field research. That is, the number of people times the number of hours I spent observing; I spent about five hundred of my own hours in the company of someone who was high.
One problem which any sociologist of deviant behavior faces is that he has access to information concerning illegal activities of his respondents which, if publicity and conviction ensued, could result in long prison sentences for those so generously supplying the information. In order to secure this information in the first place, the researcher must assure his subjects of complete confidentiality, that their names will be revealed to no one. A betrayal of trust would be suicidal. This is both a matter of professional ethics, as well as a question of sheer practicality: if it became known that sociologists were unable to keep their word concerning confidentiality, that, by revealing compromising information about oneself, one thereby was placed in serious trouble, the student of criminal behavior would not long be in business. Adhering to the rule of confidentiality is absolute, a rule that must not be broken.
No serious researcher of crime questions the maxim. But alas, it is not so clear-cut. The sociologist is often called upon not only to find out about illegal activities, but, as I said, to observe them as well, and occasionally to participate in them. The journalist-writer of a recent book, The Seekers, Jess Stearn, refused even to be present when drugs were present in the same room: "It was against the law to knowingly stay where marijuana was smoked." As the book testifies, Stearn learned nothing about drug use. There is no handy rule of thumb here. My colleague, Ned Polsky, admits that he has been unwilling to witness a number of illegal acts which were morally repugnant to him (beatings, for instance), and, in so doing, slightly compromised his role as a sociologist.
In my case, then, I witnessed hundreds of cases of drug use, possession, and/or sale. As I said about one hundred of the interviews were conducted in the place of residence of the respondent, or a friend of his. In about half of these cases, or about fifty interviews, the respondent used marijuana during the interview. The likelihood of that many respondents smoking during the interview was far greater than would have been expected randomly, judging from the fact that the overwhelming bulk of the sample claimed to smoke less than daily, and even the daily smokers did not typically remain high during the whole day. My chances of hitting the respondent while he was smoking, then, was far smaller than the 50 percent of the at-home interviews I did, or even the 25 percent of the total. This was a most curious tendency. Possibly talking about the drug stimulated the respondents to smoke. When I asked one respondent to describe the marijuana high, he said, "I'll have to get high first," and lit up a pipeful of hashish.
There is, of course, the issue of criminality of the author's behavior. It was witnessing crimes "taking place," i.e., I knew that the respondent possessed and used marijuana. In addition, I was present during several purchases, which is also a criminal offense. By law, one must report felonies of which one has knowledge. And in this sense, in not reporting criminal acts, my behavior was criminal. Obviously, I share this trait with every other criminologist or researcher of deviant behavior who does his research "in the field," i.e., in the open air outside the jail cell or correction house. (And any criminologist who does all of his work within the confines of an institution of incarceration itself cannot be taken very seriously as a criminologist.)
I felt it to be of great importance to protect my respondents' identity in any way that I could. I followed a number of procedures to assure them of a relative degree of protection. When a subject was contacted, I wrote his name on a small slip of paper. After the interview was complete, I destroyed the slip, and I no longer had either the name or any way of contacting the individual. At no time did I have a list of any more than twenty individuals (who were potential interviewees); at no time did I have a list of any of the respondents who had already been interviewed. Moreover, since I have gradually allowed myself to forget the names of all the respondents whom I interviewed, I am not at this time able to get in touch with any but a tiny handful of the subjects of my study. Truman Capote claims to have cultivated the ability of almost total recall after an interview, an ability which he employed in the writing of In Cold Blood. With regard to names at any rate, I cultivated precisely the opposite skill: that of forgetfulness.
This precaution was a way of making sure that the fact that I had written down names and telephone numbers of respondents and potential interviewees did not place any of them in jeopardy. They were assured anonymity, and I felt that it was necessary to do whatever I could to protect that. If, for some reason or in some way, the research became known to law-enforcement officers who saw me as a route to the names of users and, possibly, suppliers, I would be able to at any given time to be of as little use to them as possible, even in the case that they confiscated the names. After the interviews were completed, of course, I became of no use to them whatsoever.