Social and Cultural Aspects of Drinking
Culture, chemistry and consequences
"The cross-cultural study of alcohol represents a classic natural experiment: a single species (Homo Sapiens), a single drug substance (ethanol) and a great diversity of behavioural outcomes." Marshall 1979
It would be inappropriate to present a detailed account of the biochemical and physiological effects of alcohol in this report, or to dwell on health issues, as these are covered in separate reports. We have, however, criticised some of our colleagues for ignoring ‘non-cultural’ aspects of alcohol, and in the context of considering the roles of culture and chemistry in the behavioural outcomes of drinking it is important to note the relevant biochemical effects.
Independent of cultural context, it is clear that ethanol produces dose-related changes in central nervous system (CNS) functioning which in turn affect basic physiological processes. In layman’s terms, after ingestion of moderate to high doses of alcohol: reaction times are generally slower; short-term and intermediate memory may be affected; performance on problem solving tasks decreases and muscle control, dexterity and eye-hand co-ordination may be impaired.
While these physiological and psychological correlates of alcohol consumption are not disputed, the effect of alcohol on affective or emotional processes is far more variable and complex, and results of experiments have proved inconclusive and unreliable. Effects on emotional states and specific forms of behaviour are clearly extremely hard to demonstrate, as to prove that alcohol produces, by way of action on the CNS, psychological changes which lead to particular behaviours, one must control for the influence of social and cultural factors and individual expectations regarding the effects of alcohol.
The ‘natural experiment’
This necessitates the use of cross-cultural methods which allow measurement of the extent of variation attributable to sociocultural influences. For alcohol to be seen as a causal variable, in any meaningful sense, with regard to emotional states and behaviour, we would expect it to produce an invariant pattern of responses across all cultures. If , however, alcohol evokes only diffuse psychological states which have no direct prescriptions for behaviour, then we would expect to see responses varying according to different social and cultural mores. In other words, beyond consistency in the straightforward physiological, psycho-motor and cognitive effects outlined above, we would expect high levels of emotional and behavioural variance around the world following drinking. Such variance would not mean that alcohol plays no role at all in the generation of, say, promiscuous or aggressive behaviour; rather, it would indicate a strong alcohol-culture interaction, with cultural variables modifying, directing or even overriding the physiological and psychological effects of alcohol.
The ‘natural experiment’ of cross-cultural study finds levels of variance which rule out any direct causal effects of alcohol on behaviour. We have already noted that in some societies drunken aggression and belligerence are commonplace, while in others the same doses of ethanol result in quite opposite behaviour, characterised by calmness, passivity and good humour. Even within a single culture, the diversity of behavioural effects is often striking: the Aztec name for their alcoholic beverage pulque was centzonttotochtli or ‘four hundred rabbits’, in recognition of its almost infinite variety of effects on those who drank it (Marshall, 1983).
One of the earliest attempts to make use of Marshall’s ‘natural experiment’ to test commonly accepted hypotheses regarding the effects of alcohol, MacAndrew and Edgerton’s Drunken Comportment (1969), is still the most widely quoted in the anthropological literature. The authors saw their task as:
"A reconsideration of the conventional understanding of one aspect of man’s relationship to alcohol - the proposition that alcohol, by virtue of its toxic assault upon the central nervous system, causes the drinker to lose control of himself and do things he would not otherwise do."
MacAndrew and Edgerton note the general reluctance of scientists to admit empirical evidence that is at odds with accepted theory, pointing out that in the history of every scientific discipline, the discrepancy between conventional explanations and observable fact has frequently had to become "downright scandalous" before such explanations are abandoned. In the case of the ‘alcohol as disinhibitor’ theory, they argue that:
"The disjunction between the conventionally accepted formulation of alcohol’s effects upon man’s comportment, and presently available fact concerning what people actually do when they are drunk is even now so scandalous as to exceed the limits of reasonable toleration."
The empirical evidence presented in Drunken Comportment includes detailed accounts of:
- societies in which drunkenness does not result in any ‘disinhibited’ behaviour at all
- societies in which the type of behaviour associated with drunkenness has undergone radical changes over time
- societies in which drunken behaviour varies dramatically according to the circumstances in which alcohol is consumed
- societies in which apparently ‘disinhibited’ drunken behaviour remains within well-defined, culturally sanctioned limits.
Almost 30 years on, despite the ever-increasing weight of conflicting evidence, the ‘conventional formulation’ still has its supporters, and the ‘reconsideration’ is still in progress.
Drunken Comportment remains, however, the only global-scale, systematic analysis of empirical evidence relating specifically to drunken behaviour. Although individual studies and ethnographies have contributed further examples of variation in alcohol’s effects on human behaviour, descriptions of drunken behaviour are often incidental to the main focus or argument of the work, and therefore may not provide a reliable indication of the predominant behaviours associated with drunkenness in a particular culture.
An ethnographer’s account of a drunken event at which, for example, aggressive or sexually promiscuous behaviour was exhibited does not indicate that these behaviours are invariably, or even typically, associated with drinking in the society concerned. Nor can we draw any such general conclusions from ‘isolated’ accounts of drunken occasions at which no aggressive or promiscuous behaviours were evident. To bring MacAndrew and Edgerton’s work up to date - with any degree of scientific rigour - would therefore require more than a review of the more recent anthropological literature.
Despite this caveat, the cross-cultural evidence of wide variation in drunken behaviour is extensive, and must cast doubt on any purely biochemical explanations. The problem-oriented approach of recent years has led to a disproportionate focus on just two of the many possible behavioural outcomes of drinking, aggression and sexual promiscuity, at the expense of what the Aztecs might call ‘the other 398 rabbits’.
Some people, in some societies, may indeed behave in an aggressive or promiscuous manner when drunk, but the range of behavioural outcomes also includes calmness, joviality, passivity, indolence, affability, tolerance, sociability, generosity, volubility, confidence, loquaciousness, sentimentality, gaiety, euphoria, animation, tenderness, tranquillity, boastfulness, jocularity, silliness, laziness, effusiveness, vivacity, cheerfulness, relaxation, drowsiness, peacefulness, etc. In global terms, the most frequently emphasised outcomes are relaxation and sociability.
Learned effects of alcohol
It is not enough, of course, simply to demonstrate that the biochemical disinhibitor theory cannot explain the observable facts of drunken behaviour. To use Marhsall’s ‘natural experiment’ merely to test and reject an untenable hypothesis, without attempting to provide a more satisfactory explanation of the variance in behavioural outcomes of drinking, would be, at the very least, a waste of good ethnography.
The wide variations in responses to alcohol across cultures and within cultures led MacAndrew and Edgerton not only to reject simplistic pharmacological disinhibition models, but to consider the specific aspects of culture which lead to the learning of alcohol-related behaviour. The notion that the behavioural outcomes of drinking are determined by cultural norms had already been proposed in a seminal paper by Mandelbaum (1965):
"When a man lifts a cup, it is not only the kind of drink that is in it, the amount he is likely to take and the circumstances under which he will do the drinking that are specified in advance for him, but also whether the contents of the cup will cheer or stupefy; whether they will induce affection or aggression, quiet or unalloyed pleasure. These and many other cultural definitions attach to the drink even before it reaches his lips"
MacAndrew and Edgerton, and subsequent researchers (Marshall, 1976; Douglas, 1987; McDonald, 1994; etc.), have provided overwhelming evidence to support Mandelbaum’s statement, and to illustrate the learning process summarised in Drunken Comportment:
"Over the course of socialisation, people learn about drunkenness what their society ‘knows’ about drunkenness; and, accepting and acting upon the understandings thus imparted to them, they become the living confirmation of their society’s teachings."
The anthropological literature shows how central aspects of culture can radically shape the ways in which people learn to drink and the patterns of behaviour which are associated with alcohol consumption. It is also clear that the process of ‘acculturation’, whether induced by colonial domination, tourism or economic and cultural ‘convergence’ such as that currently occurring in parts of Europe, can introduce styles of drinking with which previously existing cultural frameworks are unable to cope (Cottino, 1995; Gamella, 1995; Heath, 1995; McDonald, 1994).
To understand why, in some societies, drinking leads to problematic behaviours, we therefore need to be concerned with aspects of culture and cultural change, rather than with the pharmacology of alcohol. The different patterns of learning fostered by different cultures, and the novel modes of learning that acculturation can present, do not only provide models of appropriate and inappropriate drinking habits, they also create sets of expectancies regarding the behavioural effects of alcohol. Reviews of both ethnographic and psychological literature show that the behavioural consequences of drinking are always in accord with what people in a given culture expect to happen, and that individuals internalise such expectations during the learning process of socialisation.
Critchlow (1986) convincingly argues that if we are to change problematic drinking behaviours we must tackle beliefs about the effects of alcohol. Recent psychological studies confirm her view by showing that the responses of experimental subjects vary widely in line with their previous expectations of the effects of alcohol. Similarly, manipulation of expectancies in the experimental setting, while maintaining dosage of alcohol at the same level, also produces significant changes in responses (Gustafson, 1987; Christiansen, 1982; Rohsenow, 1984; Vogel-Sprott, 1992; Neff, 1991; Milgram, 1993, etc). Heath (1998) provides the following clear summary of the ethnographic and psychological findings:
"There is overwhelming historical and cross-cultural evidence that people learn not only how to drink but how to be affected by drink through a process of socialisation…Numerous experiments conducted under strictly controlled conditions (double-blind, with placebos) on a wide range of subjects and in different cultures have demonstrated that both mood and actions are affected far more by what people think they have drunk than by what they have actually drunk…In simple terms, this means that people who expect drinking to result in violence become aggressive; those who expect it to make them feel sexy become amorous; those who view it as disinhibiting are demonstrative. If behaviour reflects expectations, then a society gets the drunks it deserves."
Expectations and excuses
Expectations not only shape drunken behaviour, they also enable subsequent rationalisation, justification and excuses (MacAndrew and Edgerton, 1969; Gusfield, 1987). In cultures where there is an expectation that alcohol will lead to aggression, for example, people appeal to the fact that they were drunk in order to excuse their belligerent conduct. This is particularly evident in Britain, where defendants in court often plead for mitigation on the basis that they were intoxicated at the time of the offence. Perhaps surprisingly, British courts often accept such pleading, arguing that the behaviour was ‘out of character’ - a standard metaphor for disinhibition. In more informal social contexts, excuses such as "it was the drink talking" are even more likely to be accepted.
In cultures where learned expectations about the effects of alcohol are very different from the British, appeals to drunkenness as an excuse for aggressive behaviour would not only fail to be persuasive, they might actually compound the severity of the offence. Among Italian youth, for example, attempts to excuse violent or anti-social behaviour on the grounds that the person was drunk would meet with incredulity (Marsh and Fox, 1992).
Room (1983, 1984b) argues that negative expectations about the effects of alcohol may derive from current ‘amplification’ or exaggeration of alcohol problems, particularly in the US and UK. This may seem surprising given Room’s suggestion, noted earlier, that some anthropologists have been guilty of ‘problem deflation’ in their studies of alcohol. Nonetheless, he points out that drinking is more likely to serve as an excuse for anti-social behaviour if we increasingly attribute strong powers to alcohol. If we believe that the powers are ‘real’, they become real. Following Room’s argument directly, Critchlow (1986) summarises:
"On a cultural level it seems to be the negative consequences of alcohol that hold most powerful sway over our thinking. Because alcohol is seen as a cause of negative behaviour, alcohol-related norm violations are explained with reference to drinking rather than the individual. Thus, by believing that alcohol makes people act badly, we give it a great deal of power. Drinking becomes a tool that legitimates irrationality and excuses violence without permanently destroying an individual’s moral standing or the society’s system of rules and ethics."
Gusfield (1987) also argues that current ‘problem inflation’ and warnings about the disinhibiting effects of alcohol provide drinkers with a convenient excuse for anti-social behaviour:
"The very derogation of drinking among large segments of American society creates its meaning as quasi-subterranean behaviour…by shifting the burden of explaining embarrassing moments from a reflection of the self to the effects of alcohol, drinking provides an excuse for lapses of responsibility, for unmannerly behaviour; for gaucheries, for immoral or improper actions. ‘I was not myself’ is the plea of the morning after."
Changing people’s expectations about the behavioural effects of alcohol may seem to be a daunting task, especially as dire warnings about the links between drinking and problem behaviours are currently a standard feature of both sensationalist media coverage of alcohol issues and supposedly ‘responsible’ alcohol-education programmes.
Engineering a shift in beliefs might, however, be one of the most effective strategic approaches to reducing alcohol-related problems. The historical and cross-cultural evidence shows that such changes in beliefs regularly occur, suggesting that attempts to promote a shift in expectations would not be ‘going against nature’ in any sense. Presenting a basis for alternative beliefs about the effects of alcohol, stressing social harmony and relaxation rather than aggression, promiscuity or anti-social conduct, should, according to the evidence available, result in corresponding changes in alcohol-related behaviour.
Studies comparing the wider consequences of drinking in different cultures - such as Levine’s (1992) influential paper on ‘Temperance’ and ‘Non-temperance’ cultures - suggest that a shift toward more positive beliefs about alcohol could affect not only immediate behavioural outcomes but also longer-term social and even health consequences of drinking (Harburg et al, 1993). Consistently positive beliefs are associated not just with good behaviour when actually ‘under the influence’, but more healthful, less physically damaging patterns and modes of drinking. As Heath (1998) points out:
" … the impact of two drinks on each of seven days is very different from that of fourteen drinks on a Saturday night … Similarly, five drinks gulped down fast with no food will have an impact markedly different from five drinks spaced throughout a six-hour dinner party."
The episodic, binge-drinking patterns of ‘ambivalent’, ‘Temperance’ cultures are associated with a wider range and higher frequency of alcohol-related problems - including alcoholism, accidents, drink-driving, hangovers and hypertension - than the more harmonious drinking styles of ‘non-Temperance’ cultures, although the latter almost always have significantly higher per-capita consumption (Peele, 1997). Indeed, it is now a ‘cross-cultural commonplace’ to observe that alcohol-related problems (with the exception of liver cirrhosis) often occur in inverse proportion to consumption levels (Heath, 1998). The cumulative evidence on culture, chemistry and consequences seems to indicate that alcohol policies designed to change beliefs and expectations are likely to be more beneficial than attempts to reduce overall consumption.