Journal of Substance Abuse 4:91-106, 1992. Also published in Swedish in Nordisk Alkohol Tidskrift, 1991


Robin Room

Addiction Research Foundation

33 Russell Street

Toronto, Ontario M5S 2S1



Two main prescriptions are offered for reducing rates of alcohol problems in English-speaking and Nordic cultures: a "dry" solution of reducing the physical and cultural availability of alcohol, and a "wet" solution of reducing problems of intoxication by better integrating drinking into the culture. Empirical evaluations of change in particular cultures have tended to support the "dry" and not the "wet" solution. But such studies focus on relatively short-term effects, and there is evidence that long-run effects may be weaker or may even be reversed from short-term effects. Some particular societies which have been put forward as examples of the long-term success of "wet" strategies are considered; the most likely success from this perspective is the Netherlands case, which is also an exceptional society in terms of the "wetness" of its drug policies. In the light of these cases, consideration is given to some issues concerning the criteria for evaluating success of one or the other solution. It is argued that alcohol policy discussions need to recognize that intoxication has a particular social position in our societies, in terms of characteristics of the drinker and of the occasion, in terms of intoxication's cultural significance, and in terms of its entrenchment in social worlds of heavy drinking. Some implications for policy are noted.

Are we satisfied with the cultural position of alcohol and with the rates of drinking problems in our own society? Though these are questions which would not occur to social commentators in many societies, they are questions which regularly come up in "temperance cultures" (Levine, 1990) -- societies, such as the English-speaking and Nordic societies, which have a substantial historical tradition of singling out drinking behavior for examination and identifying it as a potential source of problems. Once the questions are asked at all, the answer is very likely to be "no". In the contemporary zeitgeist, it would be the exception for a commentator in such a society to conclude explicitly that the present level of problems associated with drinking are an acceptable cost and that there is thus no justification for seeking change.


If change in drinking behavior is sought, there are two broad directions to look for it. One direction is to cut down the amount of drinking in the society, on the premise that reducing the level of drinking will reduce the level of alcohol-related problems. The second direction is to focus not on changing the amount of drinking per se, but rather on changing what is problematic about the drinking. Changes in these two directions are not mutually exclusive, although it is certainly possible for a change in one to interfere with a change in the other (Room, 1975b). But at the level of policy advocacy, there is a strong tendency for arguments to be pitched toward one goal or toward the other. In fact, it is commonplace for the directions of change to be presented as policy alternatives. What we may call the "dryer" side argues for measures to cut down the amount of drinking, in order to reduce the rates of alcohol-related problems. Against this, the "wetter" side argues that the problem is not with the level of "normal" or "social drinking", but rather with particular problematic patterns or aspects of drinking.

In line with an old tradition of social criticism, those arguing in behalf of changing the problematic aspects of drinking commonly offer a vision of an alternative social order, in which drinking has taken on its rightful character -- what Börje Olsson (1990), in his recent study of the policy debate on alcohol controls in the Swedish press, terms a "dream of a better order". In North American discussions, various particular cultures have often been singled out as offering an approximation of this dream: at one time or another, the choices have been Italian, Jewish, Chinese or Japanese drinking patterns. In Sweden, Olsson notes, the objective correlative of the dream is often a generic "continental drinking culture". As this alternative Swedish social order is envisioned,

the central role of alcohol still remains, but ... it is less dramatic and ... the negative consequences of alcohol are believed to be minimized. The continental drinking culture is the theme of this dream, nourished by the shame felt about what is felt to be the dominating drinking culture in our own society, the Scandinavian way of drinking, which is characterized by heavy drinking, drunkenness, and violence. (Olsson, 1990)

What are the changes in drinking which are being sought in this "dream of a better order"? Most discussions contain one or more of the following components:

* a change away from intoxication in the drinking occasion, by limiting the number of drinks consumed on an occasion and/or limiting the repertoire of drinking occasions -- e.g., drinking only with meals.

* a change in behavior while drinking, away from aggressive and other socially disruptive behavior.

* a change in the patterning of heavy drinking over time in the drinker's life, away from alcoholism, alcohol dependence or habitual heavy drinking.

The literature centering around this "dream of a better order" tends not to offer much specific detail on how the existing concrete society is to be changed into this envisioned alternative society. For some writers, it is enough to offer evidence, by naming specific cultures, that the kind of drinking culture offered as an ideal is a practical possibility. Such evidence, of course, invites ripostes like Haggard's, when Keller proposed a study of why Jews do not become alcoholics:

All right, Mr. Keller -- and suppose we discover why the Jews don't become alcoholics. How are we going to convert 150 million Americans to Judaism? (Keller, 1970)

Other writers do offer specific suggestions on how to get there from here. Usually the answer is through education; in particular, through teaching youths how to drink in an unproblematic fashion. As a recent discussion put it,

teaching young people honestly about the nature and effects of alcohol, and instilling them with norms about the limits of appropriate and inappropriate behavior, serves as a kind of `immunization' against alcoholism and drinking problems. (Heath, 1990, p. 78)

Other common suggestions have included making lighter-strength alcoholic beverages more available, and encouraging their use. The recent release of beer onto the home market in Iceland was supported with the argument that spirits-drinking might be reduced by making beer more readily available. A U.S. report from the late 1960s on the prevention of drinking problems applied the same logic in proposing the introduction and official promotion of what we would now call spirits-based coolers:

The giant distillers should seriously consider marketing a new, low-proof alcoholic fruit drink, having some of beer's versatility, to be served with meals, snacks, barbecues, etc., as well as alone. When marketed, it would be included in government material promoting food with low-proof alcoholic drinks. (Wilkinson, 1970, pp. 122-123)

Such proposals have fit quite well into general ethical themes in industrial societies in the postwar world. Looking back a few years later on the proposals made in the earlier report and by its sponsor, the Cooperative Commission on the Study of Alcoholism, the report's author acknowledged that

a particular model was preferred partly because it suited our views of human nature, of society in general, and certain societies in particular.... We rejected a semi-dry approach that would concentrate on lowering alcohol consumption per capita ... rather than focus on how people drank.... Some of us ... felt that a great focus on restricting consumption accorded ill with our own basic like of drinking(!) and an image of ideal man as both unrepressive and relatively unrepressed. (Wilkinson, 1975, pp. 374-375, emphases in original)

It is not surprising, then, to find that such suggestions have repeatedly been put to the practical test in recent decades, in some cases with formal evaluations of the results of the tests.

In general, these models for the prevention of alcohol problems have not fared well in the evaluations.

* Programs to educate or persuade young people on drinking issues have difficulty showing any specific effects on drinking behavior, no matter what the goals or method of education are (Moskowitz, 1989). It should be noted that the programs evaluated have usually had the aim of postponing or avoiding drinking, since a formal program to teach teenagers how to drink is politically unacceptable in societies doing such evaluations.

* Efforts to promote particular forms of alcoholic beverages have often (not always) been successful. But the usual result has been that consumption of the new beverage has been added to, rather than substituted for, the existing consumption patterns. This "additive effect" has been documented for a variety of cultures and circumstances.

Perhaps the best-documented case of the application of a "wet" policy with preventive intentions is the alcohol policy changes of 1968 in Finland (see Mäkelä, Österberg and Sulkunen, 1981). The conclusions of Finnish researchers were that the 1968 reforms failed in terms of all three of the criteria noted above: not only did the per-capita consumption of ethanol rise, but also the prevalence of occasions of drunkenness and of alcohol-related casualty and social problems rose. Even in terms of problems-per-litre of ethanol, there was no clear downward trend -- in other words, there was no substantial evidence that the "new" consumption successfully encouraged was less harmful than the "old" consumption. Habitual heavy drinking, as measured by cirrhosis mortality, also tended to rise. Bruun, among others, has discussed how these findings were a shock to Finnish alcohol researchers, putting into question their prevailing assumptions.

Against this rather dismal success record of "wet" policies must be set the partial success record, at least, of "dry" policies. In two areas the results have been well evaluated and are unequivocal (Room, 1984a; Smith, 1988):

* raising the taxes and thus the price of alcohol, at least within broad limits, tends to lower the consumption of alcohol, with at least as much effect on heavier drinkers as on lighter drinkers;

* at least in North America, raising the minimum drinking age has reduced the rate of teenage drinking-related casualties.

Results for other alcohol control measures have been less studied and are more mixed. It seems likely, however, that alcohol rationing and other measures which substantially impede or raise the price of heavy drinking reduce the rate of heavy drinking and of health and some social problems associated with drinking. In general, alcohol control measures seem to have at least as strong an effect on intoxication occasions as on other drinking occasions, and on habitual heavy drinkers as on other drinkers.

At its face value, then, the evaluative literature on the effects of alcohol prevention and policy measures seems to be reaching a clear conclusion: in terms of the aims enunciated by proponents of "wet" policies, "dry" policies work, at least sometimes, and "wet" policies do not. This imbalance presumably partly underlies the gross imbalance Olsson (1990) found in the 1989 Swedish alcohol debate: that the "dry" side of the debate was conducted in rational terms, mostly put forward by experts, while the "wet" side tended to be stated in phenomenological terms, and more by private persons than by experts. For those of us living in cultures which worry about drinking, the "dream of a better order", the evaluation literature tells us, is an unattainable mirage. Within unspecified limits, the most effective way to attain the aims of the "wet" protagonists is to dry up the society as much as possible.


In my view, this is too absolute a conclusion, if we adopt a larger frame of reference than the formal evaluation literature. In the remainder of this paper I will argue why this is so and consider the state of the evidence in this broader frame.

The evaluation literature has some clear limits (Room, 1990). Some of these limits do not matter much for the present purposes. For instance, that most of the studies have been carried out in societies which worry about drinking is an advantage rather than deficiency in the present context. That most have been carried out in a period when government policy changes have mostly been moving in a "wet" direction also does not impede the interpretation of their results. But a third limitation is more serious: evaluation studies are oriented toward measuring short-term effects rather than long-term effects. In part this reflects funding cycles and the impatience of funding agencies and of researchers; usually, the baseline measurement, the preventive intervention, and the follow-up measurements all have to be fitted in a 3-year or at most a 5-year funding period. But more fundamentally, evaluation technology is inherently better at measuring short-term results, and at measuring the results of a discontinuous change in policy or program, than at measuring results which are longer-term and incremental. In the longer run, history keeps happening, and individuals move into different life-stages -- and keeping track of an effect of a specific intervention becomes increasingly difficult.

Yet the debate about alcohol policy directions is about changes in the long as well as the short run. To regard the formal evaluation literature as settling the debate involves assuming that long-term change is simply an extrapolation of short-term change. This assumption needs to be tested on a wider canvas, taking into account the historical record on alcohol, and for that matter on other psychoactive drugs, in diverse societies. Let me propose some conclusions from this wider frame:

* There is ample evidence, some of it from evaluation studies, that restrictive and punitive policies often are more effective in the short run that in the long run.

Market restrictions. Prohibition of sales or very restrictive control policies create an opportunity for illicit production and/or distribution. Yet it takes time for the illicit market to get organized, so that the effects of such policies will be at their most dramatic immediately after adoption, and will tend to decay with time. Thus, for instance, it is estimated that in the initial stages of Prohibition in the U.S., consumption of alcoholic beverages fell to one-third of what it had previously been, but that by the end of Prohibition it had risen to two-thirds of the pre-Prohibition level (Warburton, 1932). Similarly, the present prohibition of alcohol sales in Truk was more effective when first adopted than after a couple of years (Marshall & Marshall, 1990).

Analogously, the evidence from the adoption of rationing in Greenland (Schechter, 1986), and from the 1985 reduction in licit alcohol production in the Soviet Union (Anonymous, 1990), is that consumption fell sharply when the measures went into effect, but then had risen considerably by a couple of years later.

These cases involve a relatively isolated market where the illicit market depended on illicit production or or where smuggling was a substantial logistical task. The evidence from the Scandinavian strike studies is that, where illicit importation requires less organization and effort, an alternative unofficial market may already be starting operations in a matter of mere weeks (e.g., Mäkelä, 1980).

Deterrence and punishment. It has been a very common experience that punitive drinking-driving countermeasures have a much stronger effect in the short term than in the longer term (Ross, 1984; Hingson et al., 1987). A leading theory of why this is so suggests that the initial effect of the crackdown is due in considerable part to the perceived increase in the threat of punishment. But drinking drivers learn within a matter of months that the realistic odds on being caught and punished are far less than the publicity surrounding the new measures suggested. Furthermore, there is a tendency for the police resources devoted to the crackdown to decline as new policing problems compete for attention. This line of thinking has been supported by Australian findings that the effects of countermeasures (in this case, intensive random breath-testing) do not decay if the intensive enforcement is kept up (Homel et al., 1988).

* There are clear historical instances where a policy with one effect in the short term has opposite effects in the longer term.

Restrictive policies on alcohol or other drug use offer an inviting symbolic target for generational revolts. In U.S. 20th-century history, there have been at least two such revolts, involving alcohol for the college generation of the 1920s and 1930s (Room, 1984b) and marijuana, LSD and other drugs for the college generation of the late 1960s and 1970s. It can be argued that cigarette smoking, too, particularly for women, became entrenched as part of a status rebellion.

In the case of alcohol prohibition, while there were immediate positive effects from a public-health perspective, the generational reaction against prohibition in the long run carried rates of alcohol consumption and of alcohol-related problems to considerably higher levels than the rates in around 1910.

Other examples of the same process of generational revolt against restrictive policies resulting in the entrenchment of drug use as a high-status behavior can be found in Austin's historical compilation (1979) -- for instance, for tobacco in Turkey, Italy and Russia.

This process of generational revolt against restrictive policies can be regarded as one among many elements in the long waves of societal response to drinking which, it can be hypothesized, form part of the dynamic of the "long waves of alcohol consumption" which have been noticed in many industrial societies (Mäkelä et al., 1981; Room and Roizen, 1990).

* The adoption of restrictive or punitive policies may contribute to the creation of an embattled contraculture of drug use, increasing rates of conflict and disruption in the society.

A clear example of this is heroin policy in the U.S. 1900-1960. The Harrison Narcotics Act and the other measures which followed it almost certainly did curb the rate of opiate consumption in American society, both by the persuasive effect of the law and by driving up the price of opiates outside the medical system. But the population of users was radically changed, not only in social composition but also in its relation to the society. The need to connect to illicit supplies bound users together into tightly-defined subcultures. Such subcultures necessarily became embattled contracultures, with a great deal of violence and other crime associated with distributing and paying for the drug supplies. These subcultural processes emerged over a number of years, rather than immediately after the legislation went into effect.

On a cross-sectional basis, the same interrelations can be seen operating for alcohol, and where the drinking is disfavored as well as where it is actually prohibited (Room, 1975a). In traditionally "dryer" social environments, alcohol consumption tends to be more marginalized into definable subcultures, and more socially disruptive. But it is harder to be sure about causal directions for this quite pervasive pattern. It can be plausibly argued that the higher level of social disruptiveness surrounding drinking in such environments is long-standing, and that the marginalization of drinking has been more a response to than a cause of the disruption. Furthermore, the marginalization may reflect broad and informal patterns of societal response, with the restrictive policies more a reflection than a cause of the underlying dynamic. I can think of no existing case study, following a society through time, which provides clear answers on these issues. Case studies oriented to these questions would be of great interest.


There is substantial reason, then, to focus on the longer term as well as the shorter term in assessing the effects of alcohol policy initiatives. It is clear that in terms of short-term effects the evidence favors the "dry" side of the argument. But what about the longer term? If the effects of restrictive legislation tend to decay over time, one possible response is a policy of frequently adopting new legislation (as with U.S. anti-drug laws), and another is to step up and maintain the level of enforcement (as with random breath-testing in Australia). The fact that a piece of legislation may create an outlaw contraculture has rarely been an effective argument against its passage: this is often seen as an unpleasant but acceptable cost, outweighed by the intended benefits. It can even be argued that such an outlaw contraculture is functional for the modern state, providing a "suitable enemy" (Christie and Bruun, 1985) to be combatted with social "wars". As for perverse long-term effects from generational rebellions, not every restrictive policy provokes such a rebellion, and such a response is often long delayed. Again, it would be interesting to make a comparative historical study of the circumstances in which such generational rebellions, oriented around symbolic drug-using behaviors, emerge.

A crucial piece of the argument, it seems to me, is whether examples can be found which fit the "dream of a better society" as a process rather than as a state; that is, examples of societies in which, as they moved away from being relatively restrictive and "dry" about drinking, the prevalence of drunkenness occasions, of aggressive and disruptive behavior while drinking, and of habitual heavy drinking, declined in the medium term -- say over a 20-year period.

The societies usually cited as exemplars of what a "better society" would look like may not be very promising for detailed case studies of processes of change. For two of the cases -- the Chinese and the Jews -- there seems indeed to have been a big cultural shift in the position of alcohol. Keller (1970) dates the change for the Jews from "long and hard drinking" to the disappearance of drunkenness as a practical concern to the period of two hundred years after 537 BCE -- a period remarkable, as he notes, for "the paucity of historical record". In the case of China, there were periods in which drunkenness was very prevalent, at least among elites (Lee, 1987), but there does not seem to be an adequate modern history of the disappearance of drunkenness in Chinese culture. For the Mediterranean wine cultures, there are certainly changes in the amount of drinking (consumption rose threefold in France in the course of the 19th Century), but there is no adequate study and probably little historical record of how they came to be "wine cultures" in the first place.

In any event, case studies of societies in the modern period are more likely to be determinative, because of the greater availability of social and health statistics, and more relevant, because of the changed social conditions of modern life. A number of likely candidates can be put forth. Each is briefly discussed in turn.

Denmark. Danish drinking has played two main roles in the alcohol literature. On the one hand, in comparisons with the rest of Scandinavia since the 1950s, Danish drinking consistently shows a more "continental" pattern, with fewer drunkenness episodes and fewer acute consequences of drinking. On the other hand, there is the well-known history of the 34-fold increase in Danish spirits taxes in 1917, which overnight turned a spirits-drinking country (with the highest recorded spirits consumption among western European countries) into a beer-drinking country, while drunkenness arrests and delirium tremens admissions fell by three-quarters (Wilkinson, 1970, p. 38). To what extent is Denmark an example where, in the crisis situation of a neutral country in wartime, the cultural position of alcohol was successfully shifted? If this is true, the Danish case is an example for the "dry" side of the argument, for a long-term change resulted from control more directly than from persuasional measures. It should be added that, like many other industrialized countries (Mäkelä et al., 1981), there was a substantial increase in per-capita consumption in Denmark in the 1960s and 1970s, and accompanying rises not only in liver cirrhosis but also in traffic casualties and alcohol-related psychiatric treatment and deaths, in a period of a professedly liberal alcohol policy (Thorsen, 1990). In this period, the Danish experience weighs against the "wet" side of the debate.

Netherlands. In the 19th Century, alcohol consumption had been very high, ranging from 5 to 6.5 litres of ethanol per capita. It fell sharply in the early 20th century, reaching a low of 1.7 litres in the 1940s, and then rose fourfold in the postwar period, reaching a level in the 1970s (7.2 litres) above 19th century levels (de Lint, 1981, p. 89). De Lint notes that "dry sentiment was widespread during the first half of the 20th century but only a few alcohol control measures were ever implemented. It follows that the low rates of alcohol consumption in the 1930s, 1940s, and 1950s may have been the result of the massive anti-drink propaganda during the early 20th century but they definitely were not brought about by restrictive legislation." The strength of temperance sentiment in the early 20th century, and its evaporation in the 1960s and 1970s, "leaving a handful of aged abstainers wondering what happened", are attributed by de Lint (1981, pp. 94-95) to the close identification of a strong moral-persuasion temperance movement with social reform and working-class movements.

The Netherlands in the early 20th century thus seems to be a case which escapes the current bounds of the debate on wet and dry approaches: a case where the consumption level, and apparently also the rates of alcohol problems, were reduced neither by integrating drinking into the culture, nor by alcohol control measures, but by a strong moral-persuasion social movement.

The very large increase in consumption in the 1960s and 1970s appears to have been accompanied by less-than-proportional increases in health and casualty consequences of drinking, and by an actual decrease in public drunkenness arrest rates (not compensated for by a smaller rise in other intoxicated offenders) (van Ginneken et al., 1983). But rates of alcohol problems in the Netherlands, both before and after the rise in consumption, tended to be lower than elsewhere in Europe. The postwar experience in the Netherlands thus appears to be a case which might fit the "wet" prescription, where alcohol consumption became less problematic, at least in the limited sense that the problems per litre of alcohol declined. Nevertheless, the rise in absolute terms in alcohol-related health problems has concerned public health authorities and led to a renewed debate on alcohol policy (e.g., van Iwaarden, 1989).

Perhaps it is worth noting that, in the context of policies on illicit drugs, the Netherlands has been an exceptional case in its willingness to tolerate youthful drug use and to integrate rather than marginalize drug users from the society. It may be that the Dutch experience with handling the social position of alcohol will prove as hard to apply in other societies as its drug policy experience.

Iowa. In articles comparing survey data on Iowa from 1961 and 1979, Mulford and Fitzgerald (1983a) found that, although recorded per-adult consumption had risen by 67%, the rates of self-reported drinking problems had shown no significant change. The authors argued that, in the context of their findings of "relaxed attitudes regarding social drinking" but continued disapproval of "excessive" drinking in the later survey, the patterns "make up a fairly consistent and coherent pattern of changes" fitting "the alcohol-integration model" (Mulford and Ftizgerald, 1983b, p. 685). The study was accordingly quite widely cited in beverage-industry arguments for "wet" policies.

However, in an extension of the survey series to 1985, the authors found another picture (Mulford and Fitzgerald, 1988). Although recorded per-adult consumption had declined by 12% between 1979 and 1985, the rates of self-reported drinking problems had mostly increased. Other survey-derived indicators, not shown in the earlier analysis, showed several different patterns: self-reported drinking and driving had significantly risen between 1961 and 1979, reported troubles in the last year with drinking in the respondent's household showed a peak in 1979, although the differences were not significant, and the proportions reporting relatively heavy drinking declined from 1961 on (differences in method may have affected this). The authors interpret the results as problematic for the distribution-of-consumption model, but they must be seen as even more problematic for the alcohol-integration model.

Australia. Historically, drunkenness, and indeed, any alcohol consumption at all, became a major cultural marker of the boundary between the rough and the respectable, the frontier and the suburb, the male drinking group and the more feminine domain of the home (Room, 1988). Until the 1950s, the "hotel" (tavern) was an all-male environment, featuring the rushed heavy drinking of the "six o'clock swill" in the hour between the end of work and the closing of the bar. In the ensuing decades, the social position of drinking changed remarkably, but without much drama. "New Australians" many immigrating from southern Europe, enriched the repertoire of Australian eating and drinking habits. On the one hand, women entered the public drinking-place, and restaurants became a major location of public drinking; on the other hand, drinking entered the home, and became a part of the companionate marriage or relationship.

Between the nadir of recorded consumption in Australia in the early 1930s and the peak in the early 1980s, per-capita alcohol consumption almost quadrupled. Liver cirrhosis mortality rates a little more than doubled. And public drunkenness arrests rose to 2 times, and then settled back to 1 times prewar rates (Room, 1988). The Australian postwar experience could therefore be regarded as supporting a more restricted version of the "wet" model, in the sense that the relaxation of controls on drinking and the integration of drinking into everyday life resulted in fewer problems per litre of alcohol consumed. On the other hand, the rates of alcohol-related problems did generally rise as alcohol consumption rose.

ISACE. The Australian experience is a quite strong example of general trends that were found in the International Study of Alcohol Control Experiences. As the study's authors summarized the postwar experience in seven industrial societies,

even in societies where conflicts related to drinking have increased in absolute figures, their rate of increase has fallen below the increase in aggregate consumption. This can be seen as an indication of less conflict-prone patterns of drinking behavior.... The role of drinking as a causal factor in social disturbance began to be seen as less prominent. Whereas at the beginning of the period, in many of the societies studied, it was considered a sufficient explanation for a wide array of disturbance ranging from wife-beating to sports hooliganism, by the 1970s mass media and public discussions have downplayed the role of drinking. Poland and the U.S. are perhaps exceptions to this trend: the public and the media continue to associate alcohol with a wide range of social evils. (Mäkelä et al., 1981, pp. 61-62.)

Consideration of these examples suggests that whether one can find an exemplar of the "dream of a better society" as a process depends considerably on how the criteria for such an exemplification are defined. One issue is which alcohol-related problems to focus on. The "dry" side of the debate, it has been noted (Room, 1978), tends to focus on cirrhosis and other long-term health consequences of drinking, while the "wet" side is more concerned with social disruption around drunkenness. Long-term health consequences are presumably more directly related to rates of sustained heavy drinking, while social disruption tends to be more related to rates of intoxication episodes. It is certainly possible to imagine a process of change in drinking habits which would result in lower rates of social disruption but higher rates of cirrhosis. Empirically, however, such cases seem to be few and far between.

A second issue is what we mean by drinking becoming less problematic: do we mean an absolute reduction in problems rates, or do we merely mean that there are less problems per litre consumed? There seems to be no shortage of examples in recent decades of societies in which rates of at least some alcohol-related problems appear to have risen more slowly than a rise in per-capita consumption, so that the problems per litre fell. But it is difficult to find an example of a society in which, without a fall in per-capita alcohol consumption, rates of all alcohol-related problems fell. Where a particular alcohol-problem rate has fallen while consumption rose, it tends to have been a rate, such as public drunkenness arrests, in which societal reaction to the drinking plays an intrinsic role. In one sense, of course, this means that the rate of the problem has indeed fallen, if only by the society deciding to reclassify behavior as unproblematic. But we are left uncertain whether the prevalence of the behavior being reacted to -- in this case, drunkenness in a public place -- has changed, and in which direction.

For the "strong" version of the "dream of a better society", then, where alcohol becomes more integrated into daily life and problem rates decline, it is difficult to find a convincing example. For a weaker version, where at least some problems rise more slowly than consumption, there is considerable empirical support. But recent experience suggests that in the long run societies with a tradition of temperance concerns will not conclude that this results in a better society -- that the dream has now been realized. The continued rise in the absolute levels of alcohol-related problems eventually creates a more restrictive counter-reaction, both at the level of popular concerns and (except in Holland?) of legislation. Already, the ISACE report noted, the 1970s "saw the resurrection of alcohol as an explanation for problems in many of the study societies", and alcohol consumption levels began to fall, in another turn in the "long waves of alcohol consumption" (Mäkelä et al., 1981, p. 62, 7).


It would certainly be useful, for each of the societies noted above, and for that matter for others, to examine the available data on changes in drinking in greater detail. But perhaps it is worth discussing some impediments to the "dream of the better society" in temperance cultures. To a considerable degree, the "dream of a better society" is a sober drinker's dream. Olsson (1990) notes, in fact, that the Swedish newspaper dreamers all discuss the effects of alcohol controls in terms of their putative effects on others rather than on themselves. Similarly, alcohol expectancies in the U.S. -- people's expectations about the effects of drinking -- are very different, and much more benign, for their own drinking than for others' drinking (Leigh, 1987). The "dream of a better society" is always a dream about how others are going to behave -- usually, just like how one thinks of oneself as behaving. And, of course, the responses to the newspaper reporter or to the survey interviewer are usually given when sober, and may not reflect how one would think or feel when intoxicated (it would be interesting to compare responses on expectancies when sober and intoxicated). Thus the "dream of a better society" involves assumptions about the functions of drinking that may not match the reality even for the dreamer after a few drinks, and certainly may not match the reality for those who do get drunk and sometimes get obstreperous in the society.

By and large, in temperance cultures, at least, intoxication is a young person's game, and particularly a young man's game. Often, indeed, when middle-aged men get drunk, it is on occasions of nostalgia for the days of their youth (reunions of students or military veterans). Occasions of intoxication, like occasions of drinking, are primarily sociable in nature, and occur in particular times and circumstances: on weekends or at least outside working hours, and usually outside the home. Intoxication occasions are thus to a considerable extent beyond the direct reach of social controls operating through the workplace or school or the home. The call for replacing intoxication with frequent light drinking is essentially a call for youths to behave like the middle-aged -- something which will eventually happen in the individual life-course, but which is usually hard to sell to the young when they are young.

In our societies, intoxication has functions for the drinkers (and sometimes for the society) that are not the same as the functions of having one or two drinks. For youths, getting drunk can be a voyage of discovery and of self-discovery -- a "trip", in the psychedelic jargon of the 1960s -- in a lifestage in which experimentation has its social functions. For one example, societies that have not forsworn military self-defense have a need for some youthful risk-taking; in this connection, it is unlikely that the traditional association of intoxication and military service will be easily broken. Since intoxication is viewed suspiciously by "settled folks" in the society, and is the reason alcoholic beverages are denied to children, getting drunk is frequently for youth at once a symbol of emancipation and of rebellion. Since intoxication is culturally regarded as causing obstreperous or evil behavior, getting drunk indeed has these effects, and to an extent legitimates them; a desire to be obstreperous may thus motivate a drunkenness episode. It is doubtful whether teaching children or youths "responsible drinking" patterns will do much to undo these deep cultural associations and values around intoxication.

Most intoxication occurs in particular subcultures or "social worlds" of heavy drinking. The tavern is a "home territory" for such social worlds of heavy drinking, but it is not the only one (Clark, 1988). Social worlds of heavy drinking tend to overlap considerably with other deviant subcultures -- for instance of illicit drug use and of criminality. But in a society in which intoxication is only mildly devalued, these links are not strong, and the boundaries around heavy drinking worlds are so permeable as to be not easily discernible. Policies which directly attack the world of heavy drinking will necessarily have the result of creating a much sharper boundary around the world -- particularly policies like alcohol prohibition which make behaviors intrinsic to intoxication illegal. The world will become much more sharply defined into a contraculture, with new behavioral patterns of "secondary deviance", many focused around access to the means of maintaining the behaviors.

A realistic discussion of how to attain the "dream of a better society" needs to address these matters of the social locations and meanings of intoxication. Though there is by now a copious literature on alcohol expectancies, there is as yet little on how expectancies might be changed -- at the individual, let alone at the societal level. To the extent subcultures which are deviant or on the edge of deviance figure in the policy literature, the discussion is usually in terms simply of often-futile attempts at suppression. There are relatively few discussions of strategies of cooptation (though peer-counseling strategies may sometimes actually be such a strategy).

In terms of policy, these matters of the social location and meaning of intoxication hold several implications.

(1) A policy which increases the marginalization of intoxication holds long-run dangers. It may help "harden" a large but diffuse social world of heavy drinkers into a smaller but more problematic contraculture. It may increase the attractiveness of intoxication as a symbol of youthful rebellion -- both at individual and at generational levels.

(2) Alcohol control measures which do not have a high symbolic profile will probably usually reduce the level of problems from intoxication. To raise alcohol prices, for instance, does not single individual drinkers out for labeling and does not much affect the symbolic significance of taking a drink -- at least up to the point where a substantial illicit market would appear. Alcohol rationing (again within the limits set by the emergence of a substantial illicit market) seems to be particularly effective in reducing the rates of problems from intoxication, and a nondiscriminatory rationing system might not attract a great deal of symbolism. Modern experience with rationing is, however, mostly short-term -- because it is unpopular in modern consumption-oriented societies.

(3) There are other prevention strategies which hold the potential of ameliorating alcohol-related problems without increasing the symbolic power of drinking -- for instance, strategies which seek to insulate intoxication from its potential consequences, whether physically, temporally or culturally (Room, 1975b). Such strategies outflank the wet/dry discussion, since the focus of change is elsewhere than on the drinking behavior itself. Airbags in automobiles, separating tavern-going from driving, and providing safe harbors for the intoxicated are all measures which have little effect on the symbolic value of intoxication (though it is possible that by reducing deterrents to intoxication they may increase its prevalence).

(4) All in all, taming intoxication in a society with a strong temperance history must be seen as a formidably difficult aim. Perhaps the way that Mediterranean wine cultures emerged, with what might be seen as a disbelief in the intoxicating power of alcohol, or at least a strong norm against showing any behavioral effects of heavy drinking, involved centuries-long processes. There may have been enormous casualties along the way -- and, of course, in terms of cirrhosis, traffic injuries, and other alcohol-related problems which are not intoxication-specific, the casualties continue. It is doubtful that temperance cultures have the patience for such a long term, or for the level of alcohol-related chronic health problems in "wet" societies. Certainly, the history of the last century and more, with the "long waves of consumption" interacting with long waves of societal response to drinking (Room and Roizen, 1990), suggests that the increased absolute levels of alcohol-related problems which tend to come with a period of relaxation and "wettening" touch off a societal reaction, moving the society into a new temperance phase. The challenge of alcohol policymaking for the long-term is thus the formidable challenge of escaping the social dynamic which underlies the "long waves".


Anonymous, "Alcohol Policies: Perspectives from the USSR and Some Other Countries", Drinking and Drug Practices Surveyor 23:2, 59-62, 1990.

Gregory Austin, Perspectives on the History of Psychoactive Substance Use, NIDA Research Issues 24, Rockville, MD.: National Institute on Drug Abuse, DHEW Publication No. (ADM) 79-810, 1979.

Nils Christie and Kettil Bruun, Den Goda Fienden: Narkotikapolitik i Norden (Suitable enemies: Narcotic policy in the Nordic countries), Kristiansand: Boktrykeri AB, 1985.

Walter Clark, "Places of Drinking: A Comparative Analysis", Contemporary Drug Problems 15:399-446, 1988.

Jan de Lint, "Anti-Drink Propaganda and Alcohol Control Measures: A Report on the Dutch Experience", pp. 87-102 in: Eric Single, Patricia Morgan and Jan de Lint, Alcohol, Society and the State: 2. The Social History of Control Policy in Seven Countries, Toronto: Addiction Research Foundation, 1981.

Dwight Heath, "Flawed Policies from Flawed Premises: Pseudo-Science about Alcohol and Drugs", pp. 76-83 in Ruth Engs, ed., Controversies in the Addictions Field: Volume One, Dubuque, Iowa: Kendall/Hunt, 1990.

Ralph Hingson, Timothy Heeren, David Kovenock, Thomas Mangione, Allan Meyers, Suzette Morelock, Ruth Lederman and Norman Scotch, "Effects of Maine's 1981 and Massachusetts' 1982 Driving-Under-the-Influence Legislation", American Journal of Public Health, 77:593-597, 1987.

Ross Homel, Don Carseldine and Ian Kearns, "Drink-Driving Countermeasures in Australia", Alcohol, Drugs and Driving 4:113-144, 1988.

Mark Keller, "The Great Jewish Drink Mystery", British Journal of Addiction 64:287-296, 1970.

Julia Lee, "Chinese, Alcohol and Flushing: Sociohistorical and Biobehavioral Considerations", Journal of Psychoactive Drugs 19:319-327, 1987.

Barbara Critchlow Leigh, "Beliefs about the Effects of Alcohol on Self and Others", Journal of Studies on Alcohol 48:467-475, 1987.

Harry Gene Levine, "Temperance Cultures: Alcohol as a Symbol of Danger in Anglo and Nordic Societies", presented at a Society for the Study of Addiction seminar on "Substance Misuse -- What Makes Problems", Windsor Great Park, England, 25-27 April, 1990.

Klaus Mäkelä, "Differential effects of Restricting the Supply of Alcohol: Studies of a Strike in Finnish Liquor Stores", Journal of Drug Issues 10:131-144, 1980.

Klaus Mäkelä, Esa Österberg and Pekka Sulkunen, "Drink in Finland: Increasing Alcohol Availability in a Monopoly State", pp. 31-60 in: Eric Single, Patricia Morgan and Jan de Lint, eds., Alcohol, Society, and the State: 2. The Social History of Control Policy in Seven Countries, Toronto: Addiction Research Foundation, 1981.

Klaus Mäkelä, Robin Room, Eric Single, Pekka Sulkunen and Brendan Walsh, with 13 others, Alcohol, Society, and the State: 1. A Comparative Study of Alcohol Control, Toronto: Addiction Research Foundation, 1981.

Mac and Leslie Marshall, Silent Voices Speak: Women and Prohibition in Truk, Belmont, CA: Wadsworth, 1990.

J. M. Moskowitz, "The Primary Prevention of Alcohol Problems: A Critical Review of the Research Literature", Journal of Studies on Alcohol 50:54-88, 1989.

H.A. Mulford and J.L. Fitzgerald, "Changes in Alcohol Sales and Drinking Problems in Iowa, 1961-1979", Journal of Studies on Alcohol 44:138-161, 1983a.

H.A. Mulford and J.L. Fitzgerald, "Changes in the Climate of Attitudes toward Drinking in Iowa, 1961-1979", Journal of Studies on Alcohol 44:675-687, 1983b.

H.A. Mulford and J.L. Fitzgerald, "Per Capita Alcohol Sales, Heavy Drinker Prevalence and Alcohol Problems in Iowa for 1958-1985", British Journal of Addiction 83:265-268, 1988.

Börje Olsson, "Alcohol Policy and the Phenomenology of Alcohol: Conceptions Articulated in the Press", presented at the 16th Annual Alcohol Symposium, Kettil Bruun Society for Social and Epidemiological Research on Alcohol, Budapest, Hungary, 3-8 June 1990.

Robin Room, "Normative Perspectives on Alcohol Use and Problems", Journal of Drug Issues 5:358-368, 1975a.

Robin Room, "Minimizing Alcohol Problems", pp. 379-393 in: Morris Chafetz, ed., Proceedings of the Fourth Annual Alcoholism Conference of the National Institute on Alcohol Abuse and Alcoholism: Research, Treatment and Prevention, Washington DC: USGPO, DHEW Publication No. (ADM) 76-284, 1975b.

Robin Room, "Evaluating the Effect of Drinking Laws on Drinking", pp. 267-289 in: John Ewing and Beatrice Rouse, eds., Drinking: Alcohol in American Society -- Issues and Current Research, Chicago: Nelson-Hall, 1978.

Robin Room, "Alcohol Control and Public Health", Annual Review of Public Health 5:293-317, 1984a.

Robin Room, "A `Reverence for Strong Drink': The Lost Generation and the Elevation of Alcohol in American Cultre", Journal of Studies on Alcohol 45:540-546, 1984b.

Robin Room, "The Dialectic of Drinking in Australian Life: From the Rum Corps to the Wine Column", Australian Drug and Alcohol Review 7:413-437, 1988.

Robin Room, "Thinking about Alcohol Controls", pp. 68-75 in: in Ruth Engs, ed., Controversies in the Addictions Field: Volume One, Dubuque, Iowa: Kendall/Hunt, 1990.

Robin Room and Ron Roizen, "Studying Shifts in the Cultural Position of Alcohol in the U.S.", presented at the 16th Annual Alcohol Epidemiology Symposium, Kettil Bruun Society for Social and Epidemiological Research on Alcohol, Budapest, Hungary, June 3-8, 1990.

H. Lawrence Ross, Deterring the Drinking Driver: Legal Policy and Social Control, revised ed., Lexington, Mass.: Lexington Books, 1984.

Elaine J. Schechter, "Alcohol Rationing and Control Systems in Greenland", Contemporary Drug Problems 13:587-620, 1986.

D. Ian Smith, "Effectiveness of Restictions on Availability as a Means of Preventing Alcohol-Related Problems", Contemporary Drug Problems 15:627-684, 1988.

Thorkil Thorsen, "Dansk Alkoholpolitik: Oplysning, Afgifter -- og Hvad Mere?" (Danish alcohol policy: education, taxation -- what else?), Alkoholpolitik 7:1-13, 1990.

Sandra van Ginneken, Harm J. van der Wal and Wil de Zwart, "Alcohol Consumption and Alcohol-Related Problems in the Netherlands, 1950-1980", pp. 71-85 in: Norman Giesbrecht, Monique Cahannes, Jacek Moskalewicz, Esa Österberg and Robin Room, eds., Consequences of Drinking: Trends in Alcohol Problem Statistics in Seven Countries, Toronto: Addiction Research Foundation, 1983.

M.J. van Iwaarden, "The Development of an Alcohol Control Policy in the Netherlands", presented at the Alcohol Policy Section, 34th International Institute on the Prevention and Treatment of Alcoholism, Pontault-Combault, France, 31 May - 9 June, 1989.

Clark Warburton, The Economic Results of Prohibition, New York: Columbia University Press, 1932.

Rupert Wilkinson, The Prevention of Drinking Problems: Alcohol Control and Cultural Influences, New York: Oxford University Press, 1970.

Rupert Wilkinson, "Prevention: A Retrospect", pp, 369-378 in Morris Chafetz, ed., Proceedings of the Fourth Annual Alcoholism Conference of the National Institute on Alcohol Abuse and Alcoholism: Research, Treatment and Prevention, Washington DC: USGPO, DHEW Publication No. (ADM)76-284, 1975.

1. Revised from a paper presented at a Nordic Council for Alcohol and Drug Research conference on Alcohol Policy and Social Change, Hverageri, Iceland, 3-7 September, 1990. Preparation was supported by a National Alcohol Research Center grant from the U.S. National Institute on Alcohol Abuse and Alcoholism to the Alcohol Research Group, Medical Research Institute of San Francisco.