The early literature on women with alcoholic husbands outlined a predominantly negative view of these women. Such women were seen as neurotic, poor copers who were obsessed with controlling their husbands' drinking. They were seen to have partnered alcoholic men in order to satisfy their own pathological needs (Kalashian 1959; Whalen 1953). It was further asserted that some women would sabotage the drinker's attempts to abstain in order to continue meeting these needs (Futterman 1953). Edwards, Harvey & Whitehead (1973) termed such notions the disturbed personality theory. They criticised the lack of empirical support for such notions and cite a number of research findings which support the alternative view: that it is the stress created by the drinking which affects the partner's psychological functioning. Subsequent reviews cite further research supporting the stress model and refuting the disturbed personality model (Finney, Moos, Cronkite & Gamble 1983; Gomberg 1989; Watts, Bush & Wilson 1994; O'Farrell, Harrison & Cutter 1981).

When the term chemical dependent emerged as the new label for both alcoholics and drug addicts, the term codependent was coined to describe their partners (Beattie 1989; Bradshaw 1988; Cermak 1986; Mendenhall 1989; Rothberg 1986; Schaef 1986). Rothberg (1986) articulates the notion that problem drinkers and their partners develop complementary relationships in which each reinforces the pathological needs of the other. Such notions were developed from a crude and simplistic adaptation of systems theory incorporating aspects of the disturbed personality model which Edwards et al. (1973) had discredited more than ten years earlier.

Adults raised in families affected by parental problem drinking were also labelled codependents. It was argued that living in such a family results in the person learning the dysfunctional coping responses seen in the partners of alcoholics and developing a similar personality profile (Cermak 1986). The central theme of the vast literature on codependency is that all members of any family in which one member has a drinking problem are psychologically disturbed and in need of treatment. There is no doubt expressed by any of the proponents of the codependency model that there exists a distinct syndrome of maladjustive coping behaviours and that this can be observed within every family in which a drinking problem exists. Some writers are explicit in describing codependency as a personality disorder (for example, Cermak 1986) and others go so far as to describe it as a disease (for example, Schaef 1986; Young 1987).

The term has been further generalised to also refer to the partners of anyone with any form of major behaviour problem (excessive gamblers, violent and abusive men, workaholics, psychiatrically disturbed individuals, etc.) and to anyone who had grown up in a family affected by any major disturbance (Bradshaw 1988; Schaef 1986; Subby & Friel 1984).

Treatment programs for codependents have been developed and hundreds of self-help books on codependency have been published. Typically these books comprise discussions of the characteristics of the codependent person, disclosures of personal experience, case histories of codependents, explanations of why long-term therapy is seen to be required, and advice on self-change strategies. Such books sell extremely well and an entire industry has developed around them. Leading writers tour the world conducting workshops and seminars, therapists advertise that they provide treatment for codependency, and numerous support groups and family counselling services have developed programs based on the codependency model.

Such developments have all taken place in the absence of any research support for the model and the lack of an accepted formal definition for the proposed syndrome. A number of other criticisms of the model have also been raised:

*that the model is incorrect in asserting that there is a distinct coping pattern found among the partners of problem drinkers (Gierymski & Williams 1986; Gomberg 1989; Haaken 1990; Hands & Dear 1994).
* that the model is at odds with the research on family coping in that it promotes the notion that most family members adopt ineffective and pathological coping responses (Gomberg 1989; Haaken 1990; Hands & Dear 1994; Raven 1994; Watts et al. 1994).
* that the model is demeaning to women in that it describes socially sanctioned feminine role behaviours as evidence of personal inadequacy and dysfunction (Appel 1991; Haaken 1990; Hagan 1989; Hands & Dear 1994; Krestan & Bepko 1990).

Despite such stringent criticism, and the complete lack of any research support, the model continues to be widely used in the alcohol and other drug field. It is also becoming more common in other areas of the health and welfare arena, and continues to be a prominent concept in the personal growth industry.

Application to Domestic Violence

One area where the codependency model has recently gained some degree of acceptance is in the development of counselling services for women who have been physically abused by a partner or other family member. Domestic violence is specifically listed by a number of the leading writers on codependency as a relevant clinical area for applying the concept (for example, Bradshaw 1988; Cermak 1986). Cermak (1986, p.33) states that `One of the most reliable symptoms of codependence is the inability to leave a chronically abusive relationship behind'.

This use of the codependency model in the area of domestic violence is of considerable concern. The notion that all women who have difficulty leaving violent and abusive men have some form of personality disturbance is dangerous because it blames the victim for not being able to prevent, avoid or cope with the violence (McIntyre 1984; Queensland Domestic Violence Task Force 1988; Roxburgh 1991). Moreover, blaming the victim further undermines her ability to take action against the violence (Dobash & Dobash 1987; Roxburgh 1991). As Roxburgh (1991, p.143) explains, blaming the victim:

reinforces the abused woman's low self-esteem . . . ; can contradict her interpretation of the violent situation and distort her version of what is happening . . . ; can weaken her resolve to act because she feels responsible for and therefore deserving of the violence; makes her feel undeserving of other assistance; diminishes the capacity of the service provider to offer assistance which will be of real benefit to the woman; and is untrue.

Orr (1991, p. 120) concludes her review of the various theories put forward to explain family violence by stating that an `understanding of the differences in the gendered identity of men and women is crucial to elucidating why family violence occurs, and to replacing the common myths about the causes of family violence with a stronger knowledge of who benefits from its continual perpetration'. The Queensland Domestic Violence Task Force (1988) also emphasised the importance of such an approach to understanding family violence. The codependency literature, however, comprehensively fails to examine sociocultural processes and gender related power issues and hence leads to an incomplete understanding of the dynamics of family violence.

Norwood (1985), for example, writes of the women `who love too much'. She avoids examining the cultural processes which obstruct domestic violence victims from obtaining a position of safety and empowerment. Rather she analyses intrapersonal processes in order to explain their lack of power. Hagan (1989) has strongly criticised this approach. She argues that the concept involves `a classic reversal: women are at fault again, this time for loving- what we've been reared to do- too much' (p. 9). She is highly critical of the lack of social analysis which only serves to maintain the processes that enable domestic violence to thrive.

As Roxburgh (1991, p. 130) explains, family violence `isolates the victim from assistance, a consequence the perpetrator frequently seeks to maintain'. Self-help books which promote concepts of personal inadequacy and disorder could be expected to instil a sense of personal responsibility for preventing the violence and hence further isolate the victim from those services which may provide a more realistic solution.

Victims of domestic violence need to have their feelings of fear and trauma legitimised (Queensland Domestic Violence Task Force 1988). They need clear messages which counter the myth that they are in any way responsible for being abused. They need to be able to explore their fears and anxieties and discuss the difficulty they experience in removing or protecting themselves without feeling that this indicates there is anything wrong with them. It is questionable whether a model which employs notions of personal inadequacy can be made consistent with such aims.

The codependency model does not provide any meaningful contribution to the understanding of domestic violence. Given this, and the extensive problems inherent in the model, there is no justification for using it in family violenceprograms. To do so is in fact unnecessary, given that there are more established models of stress and coping which can be used as the basis for developing positive counselling programs for families (for example, Lazarus & Folkman 1984, Orford 1987, Roth & Newman 1991). Such counselling programs need to be coordinated with other supportive and refuge services, and they need to be philosophically consistent with these other services (Dobash & Dobash 1987; Roxburgh 1991).

Counselling programs for survivors of family violence need to help participants understand that they are coping as best they can under difficult circumstances and that with appropriate support, and an opportunity to learn more effective coping strategies, they can minimise the trauma they experience and improve the quality of their future life. It is also important for these programs to provide participants with an opportunity to examine how gender- based power issues have impacted and continue to impact on their lives. The aim is to empower participants to develop more self-protective and self-fulfilling social roles. While this aim is also the declared aim of the codependency movement, the manner in which this objective is addressed within the codependency model is likely to be counter-productive.

Apart from the conceptual arguments against applying the codependency model within the domestic violence field, the lack of research support for the model dictates against such a move.

A Lesson for Victimologists

The immense popularity that the codependency model has gained over the past decade carries an important lesson for victimologists, as it does for all applied behavioural scientists.

Victimologists need to be vigilant in relation to the various conceptual models which emerge in the popular literature. In concentrating on our own theoretical models it is easy to lack an awareness of the degree of influence which popular models can exert despite their lack of rigour and empiricism. We need to remember that a crucial aspect of the academic's role is to monitor and investigate those models that emerge from the lay movement. Such endeavours are needed not only to safeguard the public from suspect theories but also to pursue those ideas which may prove to be useful or which highlight the limitations of our own perspectives.

It is important not only to investigate emerging conceptual frameworks from an empirical perspective but also to reliably determine whether a given model will appropriately serve the victim or will generate responses which could be regarded as a disservice. It is crucial to ensure that we do not revictimise our clients. The codependency model (with its inherent victim blaming attributes) is therefore not appropriate for use within domestic violence services as it is virtually guaranteed to revictimise clients.

References

Appel, C. 1991, `Co-dependency: a critical appraisal of social and cultural aspects from a feminist perspective', Contemporary Drug Problems , vol. 18, no. 4, pp. 673-86.
Beattie, M. 1989, Codependent No More , Collins Dove, Victoria.
Bradshaw, J. 1988, Bradshaw on the family, Health Communications, Pompano Beach, Florida.
Cermak, T. 1986, Diagnosing and Treating Co-Dependency , Johnson Institute Books, Minnesota.
Dobash, R.E. & Dobash, R.P. 1987, `Violence towards wives' in Coping with Disorder in the Family , ed. J. Orford, Croom Helm, Kent, pp. 169-93.
Edwards, P., Harvey, C. & Whitehead, P.C. 1973, `Wives of alcoholics: A critical review and analysis', Quarterly Journal of Studies on Alcohol , vol. 34, pp. 112-32.
Finney, J., Moos, R., Cronkite, R. & Gamble, W. 1983, `A conceptual model of the functioning of married persons with impaired partners: Spouses of alcoholic patients', Journal of Marriage and the Family , vol. 45, pp. 23-34.
Futterman, S. 1953, `Personality trends in wives of alcoholics', Journal of Psychiatric Social Work , vol. 23, pp. 37-41.
Gierymski, T. & Williams, T. 1986, `Codependency', Journal of Psychoactive Drugs , vol. 18, no. 1, pp. 7-13.
Gomberg, E. 1989, `On terms used and abused: The concept of codependency', Drugs and Society , vol. 3, pp. 113-32.
Haaken, J. 1990, `A critical analysis of the co-dependent construct', Psychiatry , vol. 53, pp. 396-406.
Hagan, K. 1989, Codependency and the myth of recovery: A feminist scrutiny , Escapadia Press, Atlanta.
Hands, M. & Dear, G. 1994, `Codependency: a critical review', Drug and Alcohol Review , vol. 13, no. 4, pp. 437-45.
Kalashian, M.M. 1959, `Working with the wives of alcoholics in an outpatient clinic setting', Marriage and Family , vol. 21, pp. 130-3.
Krestan, J. & Bepko, C. 1990, `Codependency: the social reconstruction of the female experience', Smith College Studies in Social Work , vol. 60, no. 3, pp. 216-32.
Lazarus, R.S. & Folkman, S. 1984, Stress, Appraisal and Coping , Springer, New York. McIntyre, D. 1984, `Domestic violence: a case of the disappearing victim', Australian Journal of Family Therapy , vol. 5, no. 4, pp. 249-58.
Mendenhall, W. 1989, `Co-dependency definitions and dynamics', Alcoholism Treatment Quarterly , vol. 6, no. 1, pp. 3-17.
Norwood, R. 1985, Women who Love Too Much , Pocket Books, London.
O'Farrell, T.J., Harrison, R.H. & Cutter, H.S.G. 1981, `Marital stability among wives of alcoholics: an evaluation of three explanations', British Journal of Addiction , vol. 76, pp. 175-89.
Orford, J. 1987, `Integration: A general account of families coping with disorder' in Coping with Disorder in the Family , ed. J. Orford, Croom Helm, Kent, pp. 266-93.
Orr, L. 1991, `Explanations of family violence' in Family Violence: everybody's business, somebody's life , The Federation Press, Annandale, pp. 94-126.
Queensland Domestic Violence Task Force 1988, Beyond These Walls , Queensland Government Printers, Brisbane.
Raven, M. 1994, `Codependency: Fact or artefact?' in Proceedings of the 1994 APSAD National Conference: Alcohol, Drugs and the family , ed. J.W. Toumbourou, Australian Drug Foundation, Melbourne, pp. 191-6.
Roth, S. & Newman, E. 1991, `The process of coping with sexual trauma', Journal of Traumatic Stress , vol. 4, no. 4, pp. 279-97.
Rothberg, N.M. 1986, `The alcoholic spouse and the dynamics of co-dependency', Alcoholism Treatment Quarterly , vol. 3, no. 1, pp. 73-86.
Roxburgh, T. 1991, `Support services', in Family Violence: everybody's business , somebody's life , The Federation Press, Annandale, pp. 130-68.
Schaef, A.W. 1986, Co-Dependence: Missunderstood and Misstreated , Harper & Row, San Francisco.
Subby, R. & Friel, J. 1984, `Co-Dependency: A paradoxical dependency', in Co- Dependency: An Emerging Issue, Health Communications, Florida, pp. 31-44.
Watts, S., Bush, R. & Wilson, P. 1994, `Partners of problem drinkers: moving into the 1990's', Drug and Alcohol Review , vol. 13, no. 4, pp. 401-7.
Whalen, T. 1953, `Wives of alcoholics: four types observed in a family service agency', Quarterly Journal of Studies on Alcohol , vol. 14, pp. 632-41.
Young, E. 1987, `Co-Alcoholism as a disease: implications for psychotherapy', Journal of Psychoactive Drugs , vol. 19, no. 3, pp. 257-68.