Ron Roizen, Contemporary Sociology 25(1):126-127, 1996.
Drinking Careers: A Twenty-five year Study of
Populations, by Stephen J. Kunitz and Jerrold E. Levy, with Tracy
Chena DuPuy, K. Ruben Gabriel, and Scott Russell. New Haven, CT. Yale
Press, 1994. 280 pp. $28.50 cloth.
How difficult can it be to carry out an adequate empirical assessment of two competing theoretical perspectives on Indian problem drinking? In their celebrated 1974 monograph (Indian Drinking: Navajo Practices and Anglo-American Theories, New York: John Wiley & Sons), Jerald E. Levy and Stephen J. Kunitz answered this question by taking readers on a memorable intellectual expedition. They asked: Does Indian problem drinking derive primarily (1) from the destructive consequences of cultural contact with Anglo-American society or, instead, (2) from relatively enduring features of aboriginal culture? They preferred hypothesis (2) to (1), and the monograph offered the results of their attempt to retest that preference.
Comparisons across two or more Indian groups had to be made. Hypothesis (1) required comparing groups ordered in terms of level of contact with Anglo-American culture; hypothesis (2) required comparing groups ordered in terms of inherent cultural potential for problematic drinking (according to Peter B. Field's sociocultural integrationist thesis). Data bearing on the "dependent variable" in each comparison group -- i.e., appropriate problem-drinking indicators or rates -- also had to be gathered. The authors assembled and analyzed all the available historical and epidemiological evidence relating to this comparative assessment, and supplemented such evidence with newly collected survey-interview data as well.
Countless problems of conceptualization, operationalization, measurement, and analysis dogged their enterprise. Candidate Indian groups had to be identified and evaluated in terms of the needs of their comparative analytical framework. Histories had to be examined to assess levels of Anglo-American cultural contact. Cirrhosis mortality, homicide, and suicide statistics were collected as indirect indicators of the prevalences of problem drinking. Samples of Indian respondents were interviewed using Harold Mulford's pioneering drinking-problem scales to gain more direct empirical evidence of problem-drinking rates. At every turn -- and despite the authors' best efforts -- their data were less than perfect, often woefully so. The historical evidence was sparse and ambiguous with regard to ordering Indian groups in terms of the cultural-contact variable. Mortality statistics could be assembled, but the population aggregates to which they referred did not correspond exactly to the Indian groups under examination. ne survey data were collected from nonprobability samples.
Ironically, the technical weaknesses of Levy and Kunitz's (1974) omnivorous investigative enterprise lent their monograph much of its scholarly strength and charm. Faced with incomplete or dubious data, the authors bravely called upon, and made splendid use of, their full knowledge of the Indian groups they studied. Like a couple of salty old miners examining a lump of stone for the merest traces of ore, they subjected even favorable evidence to searching scrutiny and richly contextualized evaluation. Weaknesses in their data became narrative occasions for displays of erudition and interpretive surehandedness. In the end -- and despite many points in their argument where thoughtful readers might quibble -- Indian Drinking offered readers not only the fruits of their research but also, in effect, a sustained allegory on the theme of how truly difficult it is to conduct a meaningful assessment of cultural-level explanations of behavioral phenomena such as problem drinking.
In their recently published follow-up study, however, the sune authors -- with Tracy Andrews, Chena DuPuy, K. Ruben Gabriel, and Scott Russell -- offer a considerably less compelling narrative. Whereas Indian Drinking (1974) focused chiefly on explaining cross-group differences in drinking-problem rates, this new book focuses primarily on an individual-level question -- namely, What is the character of progress or change in the drinking behavior of individuals over time? The new focus contrasts (1) an "alcoholism" model of lock-step progression in drinking behavior and symptomatology with (2) a looser "drinking career" perspective that leaves open the prospect of considerable variability in drinking behavior outcomes. Once again, the authors preferred hypothesis (2) to (1).
This shift of the spotlight from a group-level to an individual-lcvcl focus on problematic drinking phenomena harbored new kinds of data-collection and analytical pitfalls for the authors. Unfortunately, these were not as successfully dealt with this time around. Unlike the earlier volume, the 1994 monograph found the authors perplexed and hobbled by their sensitivity to methodological shortcomings. The move from a group-level to an individual-level analysis in effect marginalized much of the new monograph's reexamination and update of the sorts of historical and epidemiological data presented In the 1974 monograph. The follow-up monograph fell prey to a topic-by-topic discursiveness, often freighted with overnumerous qualifications and doubts. Oddly enough, even the authors' remembrance of the 1974 book's research agenda and conclusions seems to have become distorted by the rear-view mirror afforded [by] the 1994 volume's preoccupations. (The 1994 text, for instance, incorrectly portrays the 1974 volume as strongly committed to the venerable thesis that alcohol's problematic character in Indian groups largely derives from the rowdy drunkenness of the Anglo-American frontiersmen from whom Indians first learned about alcohol -- in fact, this hypothesis held only a minor place in the earlier volume's argument.) There were, to be sure, the occasional bright spots. I found it intriguing -- even entertaining--to be treated to the new monograph's account of how the spread of "New Age" rhetoric in the wider American society has offered traditional Navajo religion new rhetorical touchstones. On balance however, the follow-up monograph was more distracted than enhanced by its new individualistic focus.