Preface

This book is aimed at those who are seriously concerned with the narcotics problem. Such an audience includes people with a wide variety of professional backgrounds—lawmakers; behavioral scientists; police officers; social case workers; psychiatrists; the personnel of correctional institutions, probation departments, and voluntary social agencies; representatives of local, state, and federal government directly charged with doing something about the problem; and the informed and alert citizenry with no direct contact with the problem but without whose participation in public affairs rational efforts to solve many of our society's ills would be hopeless.

It may be assumed that the majority of such a hoped-for audience is uninformed on technical aspects of research design and statistical data. Despite the fact that most of Part One is based on research, we have been advised to play down these aspects of our work. The typical reader, we have been assured, would be willing to take our technical competence on faith and would want to know what we have found out, without being bothered with issues of method or with the details on which our conclusions are based. The typical reader, it is said, is carried along by the dramatic quality of the presentation and is only alienated by the intrusion of evidence expressed in numbers and considerations of logic and of the nature, relevance, and limits of evidence.

Perhaps so. Yet, even if we had the competence to write these sections of Part One in a way that would hold the "typical reader" spellbound, we would not be inclined to do so. We have had a number of opportunities to see what happens to our materials when they are worked over by skilled publicists with such "typical readers" in mind The results, as they have appeared in newspapers and popular magazines, may have made interesting reading, but they have somehow lost their authority—not the pseudoauthority of ascription to eminent professors and years of university-based research nor the pseudoauthority of polysyllabic technical jargon, but the true authority of respect for the reader's intelligence. That sort of authoritative writing presents him with the essential materials in as unbiased a fashion as possible and with the thinking that the investigators have contributed. It gives the reader a fair basis on which to accept or reject the conclusions reached or to form his own.

The worth of research cannot be measured in years of effort or in the reputation of those who carry it out; even professors have been known to bumble along for unproductive years or to wander into unavoidable blind alleys, and the time spent in writing a book may reflect the pressures of other commitments, indecisiveness of thought, and the sheer lack of writing skill or inspiration. The popular articles we have referred to appealed to these false bases of authority and selected from our materials the kinds of statement and fact that would hold the interest of "typical readers," but they were, in fact, no more authoritative than countless other journalistic efforts. If we were to try to emulate the publicists, we could only do worse—we lack the necessary writing skills—and we would only be adding to the reams of paper already devoted to the topic and as quickly forgotten.

Our convictions about the kind of book we should write were strongly reinforced by certain aspects of the topic with which we are dealing and by a related experience that occurred early in the course of our research. The question of what to do about drug use is highly controversial. Certain issues of fact are germane to this controversy. In the past few years, there have been many hearings by various bodies—a committee of the United States Senate, special legislative commissions in various states, and local government agencies. At these hearings, there are always many witnesses to testify on the facts. The experience we have already referred to gave us some perspective on the quality of these facts.

It happened that, toward the end of the first year of our investigations, a major hearing was to focus on the narcotics problem in New York City. It was already widely known that we had been collecting data on drug use in the juvenile population of the city. A few days before the hearing, we began to receive visits from representatives of certain public and private agencies. Each of our visitors had the same request, worded somewhat differently, but, in essence: "I have to testify at the hearing, and I will surely be asked how many addicts there are in the city. I hope you can tell me." We explained the nature of our statistics; they referred to individuals involved with narcotics—most of them known to have taken narcotics, and the bulk of the remainder could be safely assumed to have done likewise. At that time, we took it for granted that a user, if not already addicted, would almost certainly end up an addict; but, even then, we were careful to distinguish between "users" and "addicts" (a distinction which will be discussed at some length in Chapter II).

We are not here concerned, however, with the failure of our visitors to keep the distinction in mind, but with a simpler issue. We pointed out that we were dealing with only three of the five boroughs; that our figures referred to known cases; and that, to arrive at the true figures, one would have to multiply our statistics by an unknown factor—perhaps two, or three and one-half, or seven. In the next few days, we were both amused and dismayed by newspaper reports (subsequently verified in the published record) of the testimony given by each of our visitors. The figures they cited differed wildly (we assumed that they arbitrarily chose different multipliers and added a few odd cases to eliminate the imprecision that might be suggested by too rounded numbers); there was no reference to the common source and no indication of the operations they had performed on the figures we had given them. And—this is the crux of the matter—apparently no one thought of asking them how they arrived at the figures.

The facts and figures of one "expert" are seemingly as good as those of another. How the expert arrives at his facts is a matter of technical detail that the ordinary person is not supposed to look into, to say nothing of the question of what the facts are supposed to mean. If the experts disagree, one takes his pick according to his predilections; according to which expert is most articulate; according to what fits best with one's preconceived solution of the problem; or he concludes as we have heard at a recent hearing, that, "in the field of drug addiction, there are no experts." This is what comes from not looking at how the "facts" are obtained.

We simply do not want our facts and figures to be lost in a mass of undocumented assertions. We do not mean to suggest that we are infallible. We do mean that, if someone disagrees with us, then the quality of what he disagrees with ought to be such as to place on him the burden of demonstrating the flaws in our procedures or argument and, perhaps, of producing something better.

We are tired—and we think that the interested public ought to be tired—of conclusions presented with no indication of the grounds on which they were reached. We are also tired of appearing as witnesses at public hearings and as participants in conferences where the sheer number of voices clamoring to be heard precludes any extensive critical examination of what is being said and where those responsible for formulating recommendations all too often show signs of having made up their minds in advance or of being concerned with little more than feeling the pulse of that portion of the public which is sufficiently stirred by the problem to want to be heard.

We do not mean to suggest that most of those who give testimony are uninformed or inexperienced, but only that what they have to say cannot be weighed in any intelligent fashion without going beyond the conclusions to the premises, the data, and the reasoning on which they are based and without going beyond the statistics to the circumstances and procedures under which they were collected and processed and the statistical logic from which their cogency must be inferred.
Nor do we mean to suggest that it is senseless to take action until every loophole in the evidence has been closed, every premise exposed, every argument formulated with unassailable mathematical rigor, and every possible conclusion but one disproved. Far from it. The requirement of absolute certainty, even if it were possible in principle with regard to any issue of action, could only result in paralysis; and there are situations in which doing something is so urgent that one may be quite justified in acting blindly, even at the risk of worsening the situation.

Our point is that one should never proceed more blindly than one must. If, in the potential audience that we have envisioned, there are those who, for one reason or another, prefer to do so, this is, of course, their choice. If they open the book at all, they will skip the difficult sections and the discussions of methodological issues in Part One and look for the summary sections; there may even be some entire chapters that will hold their interest. We have, however, felt obliged to tell our story as fully as we could. If there are flaws in the materials of which we are cognizant, we have tried to call attention to them. If there are points at which we have felt that a particular kind of expert would be inclined to lift an eyebrow, we have attempted to acknowledge them. We have not wittingly attempted to conceal weakness for the sake of making a more convincing case for what we advocate.

There may be technical matters that are simply beyond the reach of individuals without relevant specialized training. We have, however, written this book with the conviction that such is not the case with respect to the kinds of material that we have to present and with respect to our potential audience. We are quite willing to give up the "typical readers" in this audience, but we assume that there are enough left who will put in the extra effort required. There must, of course, be many among these who will never have heard of a variance, of a multiple correlation, or perhaps even of a simple correlation coefficient; but we believe that these and other statistical concepts can be made reasonably intelligible as the need for them arises and in the applications that we have made.

We have attempted to make the text self-sufficient, with a minimum of explanation of such concepts, but we have also included somewhat fuller explanations, in terms comprehensible to the statistical laity, in several appendixes and a number of footnotes. Specifically, Appendix C deals at some length with the meaning of correlation coefficients; Appendix E deals similarly with cluster analysis; and Appendix J deals briefly with the logic of statistical inference. In the course of these appendixes, we also deal with a number of other statistical and psychometric or sociometric concepts (e.g., the concept of a standard score) and some special versions of the correlation coefficient (e.g., the tetrachoric correlation).

Throughout, we have designated footnotes concerned with technical statistical points by superscript letters. Thus, we have dealt with some of the basic notions of multiple correlation in such footnotes. These footnotes are intended for the critical reader who may otherwise be puzzIed by some point in the text and who is entitled to be informed of special statistical issues as they arise. A small number of such footnotes and Appendix F are also addressed to readers with special statistical training. Such items are concerned, in the main, with points that might otherwise be bothersome to statistically trained readers, but which are not relevant enough to the text to justify detailed exposition.

Ordinary footnotes, containing additional remarks or information relevant to the text, but not incorporated into it, have been designated by superscript numbers. All the appendixes not already mentioned are also of this character. Footnotes to tables have been uniformly designated by asterisks, daggers, etc., although most belong in the present category.

Several chapters of Part One do not deal with statistical data at all, and even those that do devote more space to the clarification of the research issues than to statistics.

In the years since we began our work on drug addiction, the number of agencies and individuals to whom we are indebted has become unmanageably large. We have received active assistance from the various courts of New York City, the probation and parole departments attached to them, the Youth Counsel Bureau, Riverside Hospital, the New York State Training School at Coxsackie, the public school system of New York City, the school system of the Archdiocese of New York, the New York City Youth Board, the New York City Planning Commission, the New York School of Social Work, and the School of Social Work at New York University.

The studies carried out at the Research Center for Human Relations of New York University under the general direction of Isidor Chein, the reports on which constitute the bulk of this book, and the preparation of the first draft of the book were made possible by a series of special grants from the National Institute of Mental Health of the United States Public Health Service.

To attempt to list the many individuals to whom we are indebted would detract from the credit due each of them. We hope that they will not consider us unappreciative of their help.

Two exceptions, however, must be made. One is John A. Clausen, formerly chief of the Laboratory of Socio-Environmental Studies, National Institute of Mental Health, without whose help, advice, and encouragement our work could not have proceeded. The other is Daniel M. Wilner, now at the School of Public Health of the University of California at Los Angeles, who was an active partner in much of our research. Although he has had no part in the actual writing of this book, we have considered it entirely appropriate to list him on the title page as a collaborator.

It should be entirely unnecessary to add that none of the agencies or individuals that have helped us should be considered in any way responsible for the opinions we express and that there is no basis whatever for any assumption that they necessarily share these opinions to any degree.

ISIDOR CHEIN

DONALD L. GERARD

ROBERT S. LEE

EVA ROSENFELD

New York City
November 1963