THE NEW DRAFT LOTTERY:
SOME RESEARCH IMPLICATIONS*
Social scientists long have accepted the inherent discrepancy between ideal experimental design and the exigencies that demand a compromise of the design in non-laboratory settings. More often than not, social scientists lack the ability to specify and control significant experimental variables. Additionally, they display an ethical hypersensitivity (Baumrind, 1964; Daedalus, 1969) that further tends to restrict the employment of experimental research of major significance either in or out of the laboratory. While others, notably public officials, casually effect alterations in the social system, social scientists are reluctant to adversely affect subjects' "real life chances," coerce their participation, or misrepresent circumstances to them.
Although unable to initiate the experimental conditions desired, some social scientists, following what Campbell and Stanley (1969:34) refer to as "quasi-experiments," have responded to external events by constructing ex post facto designs. Such opportunities take advantage of variables linked to natural calamities or man-made crises. Unfortunately, studies of this variety, perhaps best exemplified in Barton's Communities In Disaster (1969), are both few in number and restricted in range. Illustrative is a study by George and Pat Nash (1965) who took to the streets of New York within hours of the beginning of the city-wide blackout and interviewed respondents using cluster samples from four distinct sections of the city. Yet such research is largely fortuitous; most social scientists lack the alertness, responsive capacity, resources, and plain luck that it requires.
It is against this background and the growing interest in "field experiments" (Skipper and Leonard, 1968) that I would like to review certain research opportunities created by the recently introduced national draft lottery. This event is unusual within the context of fortuitous research in that it has been both preplanned and foretold. Furthermore, it closely approximates true experimental conditions. Before examining the research potential of the new Selective Service program, let me briefly review its substance.
The draft lottery has taken the total male population of the United States between the ages of nineteen and twenty-six and randomly assigned to each member a number between 1 and 366. This number, determined by date of birth, represents the sequence in which members will become eligible for military induction. Various mitigating factors, such as temporary deferments and physical and mental disabilities, continue to be recognized. Bu what were previously arbitrary and idiosyncratic (and often inscrutable) actions by individual draft boards (Dolbeare and Davis, 1968) are now assumed to have beer standardized, although this, in fact, is not the case (Kendall and Ross, 1970).
Apparently extrapolating from recent trends, the mass media—and very likely the average person—immediately grouped the 366 numbers into thirds. Typical of this popular reduction is the following quotation from Time magazine of December 12, 1969:
Those whose birth dates were among the first third drawn are virtually certain to be called. Those in the middle third have a fifty-fifty chance of receiving induction greetings. Barring a national emergency, those in the last third are home free.
It may be relevant to note that other estimates based or projected draft calls denoted the "inevitables" as only the first fifth, while later evaluations suggested that the assumption of any increase in predictive power is misleading, since local Selective Service boards will not all have the same distribution of eligible inductees (Fienberg 1971:259; Kendall and Ross, 1970).
Whether we take thirds or fifths—and W. I. Thomas's famous theorem commends the former—previously ambiguous prospects have been more clearly defined. Borrowing from Alfred Schutz (1967:229-234), "the constitutive accent [has been] endowed upon a specific province of meaning."
Viewed in terms of an experimental model, it is striking that except for draft eligibility sequence the three thirds represent fully comparable groups randomly constituted from the same universe. Even furthering the approximation to strict experimental design, it seems legitimate to suppose that nineteen- to twenty-six-year-old males with known "ineligibilizing" factors—especially non-debilitating non-stigmatizing, and nonpublic disqualifying personal features such as epitomized by IV-F classification—would amply serve as a pre-stimulus control group. This group could be retrieved and constructed through examination of Selective Service records, inquiry into date of physical or other disqualifying exam, or subject's prior knowledge of disqualifying features. It should be noted that data are now available (Fienberg, 1971) that seriously question the randomization of the 1970 drawing. However, with the aid of professional statisticians, the 1971 drawing was substantially improved (Rosenblatt and Filliben, 1971).
One result of the draft lottery is that it is now possible to investigate, in a way importantly different from any previous way, some of the effects of our Selective Service procedures. Draft eligibility position, available through knowledge of birth date alone, may be directly related to various behaviors within each third. To the extent that it is possible to demonstrate reasonable connections between draft eligibility position (particularly the significance of the position to the holder) and specific behaviors such as marrying, choosing a career, or determining an educational program, we can compare thirds for the occurrence of significant and meaningful differences.
It is this somewhat tardy recognition that precipitates the hope that some social scientists were not merely responsive to the draft lottery but anticipated it with sound research design, even to setting up and measuring experimental and control groups prior to the experimental event. Whether or not this was done the literature may eventually reveal, but it is well to realize that there remains a fruitful field of research here.
For example, out of the sociologist's conceptual stockpile, we can apply our concern with the "unanticipated consequences of purposive social action" (Merton, 1936; 1957:19-84). We can predict not only that the new draft lottery will have consequences beyond its explicit intention, we can postulate what the consequences may be and, more important, we can scientifically investigate and determine some of these consequences.
Among the more interesting questions we might attempt from this perspective are the following:
1. Will draft eligibility position affect the types of occupational choices and educational programs young men elect? Will it create dilemmas of decision for young men who leave high school prior to their nineteenth birthday?
2. What effects, if any, will draft eligibility position have upon participation in radical student movements, anti-Vietnam war protests? What of attitudes toward the draft, pacifism, renunciation of American citizenship— might these also be differentially affected?
3. Will draft-eligible males establish draft-group cohorts—will they tend to over associate with those who have similar draft chances? And would measures of fatalism, ambivalence, tolerance for ambiguity, and optimism begin showing up differentially among the three groups?
4. Some effect, of course, can be expected upon men's marriage plans, but what of women's marriage and dating behaviors? Might not one imagine, for example, a positive correlation between degree of female desire for marriage and draft eligibility rank of males most often dated?
5. One might further imagine draft eligibility positions creating structural strains in primary group relations, perhaps epitomized in the bizarre possibility that male twins—one born before, the other after midnight—might fall respectively into the first and last thirds of the draft pool.
6. Finally, we might raise the larger question of what effects these experiences—and unmentioned others—might have upon a national people who by tradition are not given to passive acquiescence before the chance determination of fate.
These are the unintended consequences that come to mind. Without doubt there are others. There is also an. other level of social reality to explore. Here I would urge that the more important implications for sociological analysis lie in the subtle, suasive, and symbolic dimensions of this whole phenomenon. While the questions mentioned above have an obvious, immediate, and practical import, their investigation may further the opportunity to "gel inside" the more generic mechanisms of human inter. action. Let me briefly summarize a few of the more purely sociological ideas that occur to me:
First, the draft lottery has transformed a relatively mundane piece of data into a thing of symbolic persona] significance. A young man's birthday, circumstantially determined some two decades past, suddenly becomes f potential harbinger of death, a license to pursue private goals without worry or interruption, or assignment to 2 close watch of casualty rates, draft calls, enlistment trends, and manpower pools for twelve anxiously tentative months. In that situation, I would expect that many among those young men would experience the humbling fait accompli of ascription in a painful, alien, and forceful way.
Second, for males aged nineteen to twenty-six, birth date—now draft sequence number—moves from the particularistic to the universalistic sphere. The sociological perspective leads me to hypothesize that birth date will in some way, become a salient feature of private identity at least in certain social contexts, and will become incorporated into that elusive silent language of symbolic interaction that sociologists, much like neutrino hunters have such a difficult time trapping.
Finally, if birth date, alias draft eligibility position does achieve symbolic interactive salience, it may serve as a unique vehicle through which to investigate how behavior is constructed around and in response to the subtle qualities of personal identity. This brings forth the question, "Will draft position become a feature of interaction that draft-age males give off in some Goffmanesque way?' If so, what will this help us to learn about how information that is given off (Goffman, 1959:1-16) is transmitted and received by social actors?
We should not stop here, of course. The nearly perfect field experimental situation that the draft lottery provides) should stimulate research beyond what is mentioned here research that relates directly to the draft and explores beyond to the interstices of social reality.
Communities in Disaster. Garden City, N.Y.: Doubleday Anchor.
Barton, Allen H.
Campbell, Donald T., and Julian C. Stanley.
1969 Ethical Aspects of Experimentation With Human Subjects. 98 (Spring) :219-594.
Dolbeare, Kenneth M., and James W. Davis, Jr.
1968 Little Groups of Neighbors: The Selective Service System. Chicago: Markham.
Fienberg, S. E.
1969 "Randomization and social affairs: the 1970 draft lottery." Science 171 (January 22): 255-261.
1959 The Presentation of Self in Everyday Life. Garden City, N.Y.: Doubleday Anchor.
Kendall, D., and L. Ross
1969 "Draft odds." New Republic 162 (January 31) :9-10.
Merton, R. K.
1936 "The unanticipated consequences of purposive social action." American Sociological Review (December):
1957 Social Theory and Social Structure. Rev. and enlarged ed. Glencoe, 111.: Free Press.
Nash, G., and P. Nash
1964 "Attitudes during the black-out." New York: Bureau of Applied Social Research, Columbia University.
Rosenblatt, J. R., and J. J. Filliben
1969 "Randomization and the draft lottery." Science 17 (January 22) :306-308.
1964 Collected Papers, Vol. 1. The Problem of Social Reality. The Hague: Martinus Nihjoff.
Skipper, J. K., Jr., and R. C. Leonard
1968 "Children, stress, and hospitalization: a field experiment." Journal of Health and Social Behavior (December):
1969 "The draft." December 12:26.
*Published in The American Sociologist, Vol. 6, (Supplementary Issue), June 1971