Personality has been depreciated as a contributor to addiction. Yet, it is obvious that some people's personalities are part and parcel of their addictive habits. Stanton reviews critiques of personality theories, evidence of personality's role in addiction, and the interaction among personality and setting and life-history factors, integrating all together into a complete model of addiction.
In D.A. Ward (Ed.), Alcoholism: Introduction to theory and treatment (3rd ed.), Dubuque, IA: Kendall-Hunt, 1990, pp. 147-156
© Copyright 1990 Stanton Peele. All rights reserved.
Personality and Alcoholism: Establishing the Link
The belief that personality plays a role in alcoholism that it may even be the major or sole cause has a long tradition. Nineteenth-century theorizing both popular and medical on alcoholism and narcotic addiction frequently focused on the "degeneracy" of the addict or alcoholic. Degeneracy was a global concept: thought to be inherited, it encompassed criminality, feeblemindedness or retardation, sexual promiscuity, et al., along with drug and drinking excesses. By the twentieth century, explanations for alcoholism and addiction were regularly put forward by psychiatry and psychoanalysis. Probably the most popular psychoanalytic view twenty to forty years ago was that alcoholics have a dependent personality, probably created by an unusually great dependence on their parents in childhood (see Vaillant, 1983). The psychoanalytic concept of orality, based on the idea that some people are preoccupied with oral consumption, was also often at the heart of early psychiatric theories of alcoholism.
But even brief reflection on such personality factors illustrates the problems personality theories must confront. Are all people who have oral fixations alcoholics? What about those who bite their nails and are also said to be stalled in the oral stage of development why do they chew their nails instead of drinking excessively? In the same vein, don't many overly dependent people simply become dependent on a lover or some ritual, rather than drinking too much? Theories associated with psychoanalysis have generally had difficulty answering such questions, and few people today take seriously the idea of a "dependent" or "oral" personality that characterizes all alcoholics.
These issues illustrate the methodological problems of necessary and sufficient cause: do all alcoholics have similar personalities, and does everyone with a certain "alcoholic personality" become an alcoholic? The answer to these two questions is almost certainly "no," and the recognition that this is the case has brought about a more sophisticated exploration of the topic of personality in alcoholism. At the same time, many have become skeptical that personality is an important factor in alcoholism. Unfortunately, skepticism about the existence of an alcoholic personality may have caused many to throw out the baby with the bath water, and to fail to see clear links that have been established between personality traits and alcoholism.
What is personality?
Personality is based on the individual's distinct and consistent outlooks and actions or overall style of behavior. Inherited or biological traits are not personality traits except inasmuch as they influence behavior. For example, if level of energy is an inherited trait, a person could be predisposed to having a highly energized and active personality or a passive and "laid back" one. Even so, many aspects of the person's personality would remain to be filled out by the person's life experiences, particularly within the family. Obviously, many other biological traits such as height are not part of personality, although they may influence it. Some other examples and counter-examples of personality traits: achievement orientation is a personality trait while intelligence is not; a bright or optimistic outlook (or a negative, pessimistic one) is a personality characteristic while a set of political beliefs is not.
Is Behavior Consistent?
In 1968, distinguished personality theorist Walter Mischel pointed out that correlations between a person's behavior in one setting and another are often surprisingly low rarely exceeding .30 (a perfect correlation is 1.0). For example, people we label very talkative may hardly talk at all at a formal gathering. If personality traits are so inconsistent, Mischel reasoned, then they may be very poor ways to characterize individuals and to explain human behavior. Other theorists, notably Richard Nisbett (1980), took this argument farther and argued that personality traits are merely cognitive delusions people create regarding their friends and family in order to give their worlds and their relationships an artificial aura of predictability.
Yet, people we know seem to have personalities. We feel that they act in consistent ways. Psychologists who believe people are characterized by distinct personalities have sought explanations for why it is hard to get similar measures of personality traits for individuals in different situations. David Funder, for example, has shown that people may vary on some traits from situation to situation, but they can be characterized in different situations by central traits that are important to the individual's identity. When asked about traits that are very true of a particular individual, Funder found, there is often strong agreement among friends, family and the individual that the individual is characterized by this trait, even though these different people see the person in very different situations (Funder & Dobroth, 1987).
In the alcoholism field, there has been a general rejection of broad personality theories in alcoholism by researchers. This is because different studies have not consistently found the same personality traits to characterize different groups of alcoholics. Indeed, sometimes different studies have found alcoholics have both more and less of some traits (Miller, 1976)! The principal argument against the idea that people's personalities predispose them to become alcoholic has been that personality traits observed in alcoholics are the result and not the cause of their alcoholism. That is, if alcoholics frequently appear to lack motivation or to be aggressive and antisocial, this is due to their years of drinking and does not represent their original personality.
Based on these issues, we must keep the following questions about personality and alcoholism in mind:
- Do alcoholics as a group of people fit into one or more distinct personality patterns?
- Do people with personalities associated with alcoholism inevitably, frequently, or infrequently become alcoholics, or do they often express this personal disposition in some other way?
- Can we systematically establish that many different groups of alcoholics have similar personality traits, and that groups of nonalcoholics do not display these traits?
- Can we establish the existence of particular personality traits that characterize alcoholics prior to their heavy drinking, so as to establish personality traits as a cause of alcoholism?
- Finally, can we fit the different personality factors connected with alcoholism into an overall picture of the development and persistence of alcoholism, so as both to make sense and to be useful in dealing with alcoholism?
David McClelland and his colleagues' work with the relationship of the need for power and alcoholism illustrates the use of personality measurement techniques in alcohol studies. McClelland et al. (1972) investigated heavy college drinkers and found they were likely to have a high need for power, as measured by the stories they told about a series of illustrations. These stories were then coded according to a set of protocols that assess the story tellers' need for achievement, affiliation, or power. This process, and the picture-story technique, are examples of a "projective" test. The projective test the McClelland group used is the Thematic Apperception Test, or TAT. McClelland and his colleagues found that heavy college drinkers were characterized by a high need for power, but that they did not exercise power in constructive ways, such as by becoming an officer in a fraternity. The McClelland group also measured heavy drinkers after drinking, and found they had more extravagant power fantasies after imbibing alcohol.
While McClelland's work was some of the first to apply modern personality techniques to the study of drinking, alcoholism research has moved beyond it. This is due to a combination of factors. First, projective tests have lost favor in psychology there is often too much ambiguity in coding different responses to a stimulus picture. Most researchers today prefer paper-and-pencil or objective tests based on any of a variety of validated scales, such as the Minnesota Multiphasic Personality Inventory (MMPI). Second, McClelland's effort rarely dealt with alcoholics per se, in favor of using heavy drinking college students or men in working-class bars as subjects. Third, later studies have not always found support for McClelland's power thesis. In many cases, other, related personality traits such as impulsiveness or antisocial personality seem to present a better description of alcoholic or problem drinkers.
The Mac Scale
Since the MMPI is often used to test hospitalized psychiatric patients, a number of efforts have been made to relate various subscales of this instrument to alcoholism. The best developed and most thoroughly explored has been an MMPI subscale developed by Craig MacAndrew, called the MAC Scale. Administering the scale to both alcoholic and non-alcoholic patients in a psychiatric facility, MacAndrew and other researchers have found alcoholic patients have much higher MAC scores. About 75%-85% of all patients can be correctly placed in the psychiatric or alcoholic category by relying on patients' MAC scores. While this is not a perfect categorization, and some patients can be incorrectly identified as being either alcoholic or not based solely on MAC scores, the scale clearly has major diagnostic value.
What exactly does the scale measure? According to one group of researchers that has used the scale:
High scorers on the MacAndrew Scale seemed to be bold, uninhibited, self-confident, sociable people who mix well with others. They show rebellious urges and resentment of authorities. They tell of carousing, gambling, playing hooky, and, generally "cutting up." Yet their answers show they are drawn to religion. (Finney et al., p. 1058)
MacAndrew himself, noting the highest factor loading for the MAC Scale is "boldness," describes the high scorer as "an assertive, aggressive, pleasure-seeking character" which makes alcoholics "resemble criminals and delinquents" (MacAndrew, 1981, p. 617). While MacAndrew's early work was conducted with male alcoholics, later research with the scale with female subjects gives a similar picture of the alcoholic personality in women (MacAndrew, 1986a).
It is worth mentioning that prior to the development of the MAC Scale, other, existing subscales of the MMPI had shown success in identifying alcoholics. The principal of these scales has been the PD or Psychopathic Deviate Scale. The PD Scale, also known as Scale 4 of the MMPI, has often given a similar picture to results of the MAC Scale. In addition, Scale 4 describes an emotional shallowness and inability to create intimate relationships that is important to contrast with the sociability reported by Finney and his colleagues above. Zucker and Gomberg (1986) summarized a range of studies which found that boys who subsequently become alcoholic are characterized by their superficial relationships throughout life.
The Pre-Alcoholic Personality
MacAndrew noted that younger (16-21 year-old) drinkers who were treated for alcohol problems also showed elevated MAC Scale scores. He modified the MAC Scale for this young group to create a new scale (the SAP Substance Abuse Proclivity Scale) which has shown a good identification rate for young problem drinkers and drug abusers (MacAndrew, 1986b). This strongly suggests that this personality type precedes alcoholism, rather than resulting from it, since few if any of these youngsters could be clinically classified as fully developed alcoholics.
Another tack for seeing whether a personality type predisposes the individual to become alcoholic is to take measurements of young people and to see whether a scale predicts which will become alcoholics and which will not. Exactly this strategy was employed by a group of Minnesota researchers. Hoffman et al. (1974) and Loper et al. (1973) compared the MMPI scores of college students who later became alcoholics with those who did not. The MMPI scores between the two groups showed differences very similar to those noted according to the MAC Scale in the direction of higher sociopathy, defiance of authority, and impulsiveness among those who were to later become alcoholics. Clearly, repeated findings have shown this kind of anti-social impulsiveness to characterize alcoholic males and females both before and after they are clinically diagnosed as alcoholics.
Drinking in Response to Depression, Anxiety, or Other Emotional Conflict
MacAndrew found that about 15% of alcoholics were not correctly identified by the MAC Scale. MacAndrew (1986a) identifies the alcoholics who do show high MAC scores as "primary" alcoholics and the group not identified by the scale as "secondary" alcoholics. This distinction between primary and secondary alcoholics is often made in the alcoholism field by those who espouse the disease theory. In the disease view, "real" alcoholics drink because they have an inbred disease, while others imitate alcoholic drinking because they have life problems that drinking palliates. Other research discounts this distinction, however. Epidemiological surveys (surveys about the incidence and population distribution of drinking problems) find that all those who abuse alcohol show variations in their alcohol abuse according to problems and other environmental pressures.
Secondary alcoholics in MacAndrew's typology fit the description of "reactive" alcoholics. That is, they drink alcoholically in reaction to tension and depression. This approach suggests that while sociopathic deviance is more likely to lead to problem drinking and/or alcoholism, depression or anxiety sometimes but less reliably leads to alcoholic and problem drinking for palliation of these feelings. Other personality typologies find a similar distinction to MacAndrew's between drinking and personality types. For example, Skinner (1982) found a sociopathic variety of alcoholic while another group is distressed and neurotic. The first group, like MacAndrew's primary alcoholics, are impulsive and act out because they do not have sufficient concern for societal standards, while the neurotic group show unusual emotional distress that drinking alleviates.
Anxiety is a separate category for drinking. Like distressed, neurotic, or depressed drinkers, some anxious people may find alcohol to effectively relieve their anxiety, at least for a time. One typology of alcoholics related to anxiety has been offered by Petrie (1967), who claimed alcoholics were driven to drink in order to modulate the intensity of stimuli. Alcoholics require this modulation, Petrie's research shows, because their natural tendency is to augment external stimuli which then strike them as too intense. Without alcohol, these stimuli would be too painful or anxiety-provoking for them to tolerate. An argument against this model is that drinking does not relieve anxiety or reduce tension in all cases. Sher and Levenson (1982) indicate (just as does Petrie), rather that it only does so for special groups of drinkers, those who may have elevated or special kinds of anxiety.
One paradox with drinking to relieve anxiety, depression, or other emotional distress is that alcoholics continue to show these problems, often in quite severe form. That is, alcoholics who drink in response to depression and anxiety actually show greater anxiety and depression after drinking. Here we see that drinking for emotional purposes, while it provides immediate rewards for the drinker, can also hold out the possibility of developing more severe emotional disabilities. In other words, drinking creates a vicious cycle for these alcoholics which reinforces and potentially exacerbates their drinking and emotional problems.
Cognitive Orientation Dependence On External Standards and Structures
Several scales have been developed by psychologists to indicate that some people search for external forces to explain their behavior (such as "fate"), or else that they are very dependent on external means for making decisions or orienting their thinking. The most prominent example of the former is Rotter's internal-external locus-of-control research. The purpose of Rotter's scale is to determine whether the individual believes that one can attain goals through one's own efforts, or whether the person feels that fate or chance determines what happens to people. Rohsenow (1982) has surveyed a number of studies that show that it is people with an external orientation who are most likely to become heavy or problem drinkers and also alcoholics. At the same time, not all research has found this to be the case.
The second type of measure which incorporates the concept of external reliance is field dependence, developed by Witkin and his colleagues. Field-dependent people are more suggestive in their interpretation of what they perceive and think, relying on external indicators to suggest the direction for their perceptions. Barnes (1979) has surveyed a number of such studies to find "field dependence appears to be an important characteristic in the alcoholic personality. Studies conducted to date have consistently shown that alcoholics are more field dependent than controls (that is, nonalcoholic comparison groups)" (p. 599). Based on these two groups of research, then, alcoholics seem more passive and ready to depend on alcohol to determine their mood or to solve their problems.
Do the Same Personalities Predispose People to Drug Abuse and Other Behavior Problems as Well as Alcoholism?
Recall that MacAndrew, the developer of the MAC Scale, also devised the SAP or Substance Abuse Proclivity Scale which successfully predicted which men were more likely to abuse drugs and alcohol. Consider that earlier research by MacAndrew and by Hoffman and his colleagues came at a time when relatively fewer people were likely to abuse drugs. If we examined young people with an alcoholic personality today, we might wonder whether many of them might not become drug abusers or addicts rather than alcoholics. At issue is whether the alcoholic personality type is really specific to alcoholism, or whether it is a more general predisposition that opens a person to a range of problem behaviors.
In support of this idea, the same people often abuse a range of substances, while they often misbehave in other ways as well. To take one example, the very same people who are more often arrested for drunk driving are more often stopped for reckless driving even when they haven't been drinking. The most comprehensive view of a similar personality or temperament that can lead either to alcoholism or to other behaviors such as drug abuse, overeating, and anti-social criminality is Tarter and his colleagues' (1985) model of an impulsive personality. For Tarter et al., a person born with such a temperament may engage in any of a range of excessive behaviors either as an expression of this impulsiveness, or else as a way of trying to modulate or control their impulsive temperaments.
Problems with this point of view are that impulsiveness and anti-social acting out are also related to social status. That is, when they look at large population groups, Cahalan and his colleagues have found that young men in lower socioeconomic groups are far more frequently likely to drink and act out aggressively by breaking laws, fighting, or other anti-social actions. Indeed, Cahalan and Room (1974) found, for groups of young, working-class men in certain ethnic categories, drunken misbehavior is typical, a seeming rite of passage. In this sociological view, drinking and other similar kinds of behavioral problems stem from social outlooks and values, rather than representing individual personality traits.
Peele (1985) argues instead that addictions to a range of activities including gambling, overeating, or even compulsive exercise or destructive love relationships are equivalent. While they may stem from from similar personal impulses, the form addiction takes for a specific individual is determined by the person's background and outlook and by environmental forces. Nonetheless, a person more likely to be addicted to one thing is probably more likely to be addicted to another. Addiction in this sense is an underlying disposition that takes a particular form in a given set of circumstances.
How Does Personality Translate Into Alcoholism, or Anything Else?
In order to construct a scientific theory of personality and alcoholism, or anything else, we need to draw a picture a set of logical connections that takes us from the hypothesized determining factor to the projected outcome. How does a particular personality type lead to alcoholic drinking? In the case of Tarter and his colleagues' view that impulsiveness is the key to alcoholism, although it also predisposes people to other misbehaviors or behavioral problems, can we not also imagine that some extremely impulsive people get involved in business dealings or sports, where they can harness their impulsiveness to more constructive ends? Indeed, when we consider that probably roughly similar proportions of different social groups may be born with similar temperaments, we are left to wonder why it is that so many born to some groups become alcoholics while those in other groups do not.
George Vaillant (1983), for example, discovered that 7 times as many Irish Americans as Italian Americans and others in his Boston sample became alcoholic. When comparing a group of Harvard students with an inner-city group of white ethnics, he also found more than three times the alcoholism rate among the inner-city group (all these men's drinking histories were followed for over 40 years). Why do those in some groups, like the Irish, with an impulsive personality drink too much and those in other groups do not? Many have noted the special allure and yet the danger the Irish attribute to booze. For those from this background, there is a heightened tendency to see alcohol both as a salvation and as an evil force that can take control of the individual (remember the research on locus of control).
Alternately, those in the Harvard versus the inner-city group may have had their energies channeled into more constructive pursuits. We see here that values play an important role in determining whether a given personality type expresses itself in alcohol or drug abuse or other misconduct. Jessor and Jessor (1977) have developed a complex model which takes into account personality, the person's immediate environment, and the larger social groups and values that the individual pursues. According to Jessor and Jessor, young people who are oriented toward achievement or other prosocial activity are unlikely to abuse drugs or alcohol whatever their personality type.
Alcoholism Through The Life Cycle
One other argument against the idea that alcoholics are "born" with a certain personality is offered by the life study of drinking. Vaillant's work assessed what he calls The Natural History of Alcoholism (the title of his book). He found that most alcoholics and problem drinkers outgrow this problem somewhere in their lives, more often than not without undergoing treatment or joining AA. At some point, they either quit drinking or, Vaillant maintains, in some cases they control their drinking. Indeed, Vaillant cites research by Fillmore (1975) which found that only a small minority of college students who drank regularly until they blacked out were blackout drinkers 20 years later, when they were middle-aged, even though they still continued to drink. Many of us, after all, have encountered people who remedied their youthful heavy or problem drinking as they matured.
If, then, a certain personality predisposes the person toward alcoholism, how is the same individual whose personality is presumably fairly constant able to forego problem drinking and even alcoholism (Ludwig, 1986)? We can think of our answer in the same terms as we did the values issues in the previous section. As drinkers mature, both their values and their life alternatives broaden. Rather than being concerned primarily with self-expression or hedonistic enjoyment, they become more focused on family, career, future, and health all values that work against uncontrolled expression of personality urges. At the same time, they become engaged in work and family in a way that channels their energies into constructive activities without which they could not lead an adult life. In this sense, then, the "alcoholic personality" is extremely dependent on the age of the person.
A Model of Personolity and Alcoholism
We promised earlier to describe how personality can lead to alcoholism. Given the previous sections on personality traits, values, and life developments, the model goes like this: Some people find that drinking modifies their experience in a way which they enjoy or find beneficial. For example, it makes them feel powerful or socially at ease or lessens their anxieties. If they belong to a group which regularly relies on drinking as an entertainment and which also tolerates or even encourages anti-social acting out, the person may develop a pattern of drinking to excess and misbehaving when drunk that has negative consequences. If this drinker does not become engaged in healthier forms of expression or does not develop more mature values, such problematic drinking can develop and expand. Compounding the problem, some people decide they are incapable of functioning in certain key situations without the assistance of alcohol for example, at a party or in dealing with members of the opposite sex. When they become convinced that their functioning requires alcohol, they may be said to depend on drinking.
All of this is nonetheless a long way from alcoholism. That is, someone can drink regularly in order to relax at parties for years, and decades, without becoming an alcoholic. But when drinking rules out all other sorts of activities and feelings, so that it becomes the core of the person's lifestyle and, eventually, their self-concept, a dangerous point has been reached. Here alcohol has the possibility of affecting the person's personality, the way he or she comes across to others, conceives of himself or herself, and acts so as to determine his or her environment. When people allow alcohol to determine their personalities, drinking can become so inextricably mixed with identity that separating the two becomes a difficult or painful task. And drinking's dangerous consequences can become the major problem in which the person is engaged, a preoccupying force that has taken an increasing role in all the person does.
Even at this point the role of alcohol in a life can be disowned and drinking eliminated or tightly controlled. No personality type has been found to predispose the person for all time to rely on drinking or any other negative addiction. Think simply of people you may know who suddenly quit smoking after literally decades of addiction. They required cigarettes to allay their anxieties, give them something to do, and create their self-images for years and years, yet these individuals found they had changed their values, views, and dispositions enough to leave cigarettes behind. Such, too, is the case with problem drinking and alcoholism.
In this sense, then, the idea of an indelible alcoholic personality the belief that a leopard can never change its spots can actually contribute to or create alcoholism. That is, believing that one has a personality that means he or she can never hope to control his or her drinking can be a self-fulfilling prophecy. Trying to persuade people that they have such a personality can literally do more harm than good. After all, some people who appear completely out of control of their actions at one point significantly change their outlooks and ability to regulate their behavior later in life.
Nonetheless, our role as scientists obligates us to examine certain consistencies that appear across different individuals and that persist through the life cycle in the case of alcoholism. In this chapter, we have seen that an antisocial outlook (including a lack of achievement orientation), lack of an ability to create intimacy, a tendency to seek feelings of power through drinking, use of alcohol to palliate depression or anxiety, a tendency to feel one does not have control over one's life and to define oneself externally, can combine under certain circumstances especially when not balanced by other values and life involvements into alcoholism, or else into other excessive and self-destructive involvements, from overeating to compulsive gambling.
Barnes, G. E. (1979). The alcoholic personality: A reanalysis of the literature. Journal of Studies on Alcohol, 40, 571-634.
Cahalan, D., & Room, R. (1974). Problem drinking among American men. New Brunswick, NJ: Rutgers Center of Alcohol Studies.
Fillmore, K. M. (1975). Relationships between specific drinking problems in early adulthood and middle age: An exploratory 20-year follow-up study. Quarterly Journal of Studies on Alcohol, 36, 882-907.
Finney, J. C., Smith, D. F., Skeeters, D. F., & Auvenshine, C. D. (1971). MMPI alcoholism scales: Factor structure and content analysis. Quarterly Journal of Studies on Alcohol, 32, 1055-1060.
Funder, D. C., & Dobroth, K. M. (1987). Differences between traits: Properties associated with interjudge agreement. Journal of Personality and Social Psychology, 52, 409-418.
Hoffman, H., Loper, R. G., & Kammeier, M. L. (1974). Identifying future alcoholics with MMPI alcoholism scores. Quarterly Journal of Studies on Alcohol, 35, 490-498.
Jessor, R., & Jessor, S. L. (1977). Problem behavior and psychosocial development: A longitudinal study of youth. New York: Academic Press.
Loper, R. G., Kammeier, M. L., & Hoffman, H. (1973). MMPI characteristics of college freshmen males who later become alcoholics. Journal of Abnormal Psychology, 82, 159-162.
Ludwig, A. M. (1986). Cognitive processes associated with "spontaneous" recovery from alcoholism. Journal of Studies on Alcohol, 47, 53-58.
MacAndrew, C. (1981). What the MAC Scale tells us about men alcoholics. Journal of Studies on Alcohol, 42, 604-625.
MacAndrew, C. (1986a). Similarities in the self-depictions of female alcoholics and psychiatric outpatients: Examination of Eysenck's dimension of emotionality in women. Journal of Studies on Alcohol, 47, 478-484.
MacAndrew, C. (1986b). Toward the psychometric detection of substance misuse in young men: The SAP scale. Journal of Studies on Alcohol, 47, 161-166.
McClelland, D. C., Davis, W. N., Kalin, R., & Wanner, E. (1972). The drinking man. New York: Free Press.
Miller, W. R. (1976). Alcoholism scales and objective assessment methods: A review. Psychological Bulletin, 83, 649-674.
Mischel, W. (1968). Personality and assessment. New York: Wiley.
Nisbett, R. N. (1980). The trait construct in lay and professional psychology. In L. Festinger (Ed.), Retrospections on social psychology (pp. 109-130). New York: Oxford.
Peele, S. (1985). The meaning of addiction: Compulsive experience and its interpretation. Lexington, MA: Lexington Books.
Petrie, A. (1967). Individuality in pain and suffering. Chicago: University of Chicago Press.
Rohsenow, D. J. (1983). Alcoholics' perceptions of control. In W. M. Cox (Ed.), Identifying and measuring alcoholic personality characteristics (pp. 37-48). San Francisco: Jossey-Bass.
Sher, K. J., & Levinson, R. W. (1982). Risk for alcoholism and individual differences in the stress response dampening effect of alcohol. Journal of Abnormal Psychology, 91, 350-367.
Skinner, H. A. (1982). Statistical approaches to the classification of alcohol and drug addiction. British Journal of Addiction, 77, 259-273.
Tarter, R., Alterman, A., & Edwards, K. (1985). Vulnerability to alcoholism in men: A behavior-genetic perspective. Journal of Studies on Alcohol, 46, 329-356.
Vaillant, G. E. (1983). The natural history of alcoholism. Cambridge, MA: Harvard University Press.
Zucker, R. A., & Gomberg, E. S. L. (1986). Etiology of alcoholism reconsidered: The case for a biopsychosocial process. American Psychologist, 41, 783-793.