The term "attachment" describes "an infant's tendency to seek closeness to particular people and to feel more secure in their presence" (Atkinson et al, 2000, p90). This essay will attempt to provide a brief and up to date summary of attachment theory and research, show how it is linked to Child Abuse, the Family, and Children and Divorce, critically evaluating attachment's predictive value.
One of the most influential theories in the history of attachment has been that of John Bowlby developed during a study of the mental health of homeless children for the World Health Organisation in 1951. This proposed a multidisciplinary stance in which psychoanalysis appears to be integrated with paradigms such as ethology's "imprinting" phenomenon and "critical period" (Lorenz (1935) cited in Durkin (2000) p83), cybernetic theory of control systems (Bowlby (1988) p3), social, (Hodges & Tizard (1989)), and cultural psychology (Gnaulati & Heine (2001)).
Whereas it seems that Freud and virtually all subsequent analysts had worked from an end-product backwards, whilst agreeing with the importance of the relationship with the mother, Bowlby took the reverse position to understand the origin, function and development of the child's early socio-emotional relations. His early research concluded that the development of a "warm, intimate, and continuous relationship with his mother" Bowlby, (1953) cited in Gross (2000) p550, was crucial for the infant to achieve successful close and personal relationships as an adult. Highlighting the importance of "monotropy" he posited that, without a secure attachment, specifically to the mother, through absence or severe disruption, behavioural and emotional problems would follow in later life and that "individuals suffering from any type of psychiatric disorder always show an impairment of the capacity for affectional bonding" frequently due to interference of bonding in childhood, Bowlby (1969) cited in Rutter (1991) p15. However, the political implications of the era in which it was developed cannot be ignored with men returning from war and needing jobs back! In contrast, Freud & Dann (1951) found that the negative outcome for the child of deprivation (lost or separated from mother), or privation (no particular person to attach to), in concentration camps, was alleviated by the company of other children..
Bowlby's somewhat pessimistic view of the consequences of maternal separation/deprivation has been actively debated and was the catalyst for much contemporary work. Rutter (1971) in Rutter (1999) p65, challenged Bowlby's (1960) findings that linked significant separations in early childhood with stealing tendencies/delinquency in boys. He found that emotional disturbances related to the reasons for separation from the primary caregiver, i.e. serious tensions at home, parents with mental illness and acrimonious divorces were responsible for later delinquency, rather than the separation itself. Tizard (1989) also challenged monotropy finding adopted children did better than those who remained institutionalised or were reunited with their mothers. Kagan (1978) found that what mattered was the quality of the substitute care and Calkin & Fox (1992) gave equality to the parenting ability and the infant temperament. Schaffer and Emerson (1964) cited in Durkin (2000) p.80, found distinct stages in primary attachment development and transferability of attachment to others such as siblings and fathers whilst Verschueren and Marcoen (1999) found anxious/withdrawn attachments were better predicted by the quality of the child/father attachments. Latterly Bowlby's monotropic stance was amended to that of "primary caregiver" which need not necessarily be the mother.
Bowlby's model is primarily a homeostatic mechanism designed to account for the infant's tendency to maintain a degree of proximity to adults. It is a closed mechanism in that, although set goals may be modified, the mechanism remains unchanged (Bartlaffny (1969) cited in Harris (1995). According to contemporary evolutionary thinking, structures and behavioural systems are innate behaviours present in the population because they contributed to the continuation of the species in the environment of evolutionary adaptedness. If the evolution of the attachment behavioural system merely achieves the advantages of closeness to adults however, this goal could as well be served by a tendency to cling unremittingly, such as in baby monkeys, rather than a complex system to organise the balance between attachment and exploratory behaviour resulting in repetitions of separation-reunion experiences that can result in a pessimistic prognosis for adult relationships thus shaping future parental behaviour.
Theory appears to leave us with the somewhat circular argument, does the external environment/social circumstances shape the behaviour of a parent thus primarily arbitrating the attachment type, or is it shaped by the parent's internalised early attachment experience and which allows a predictive ability? Current theory strongly suggests the "type" of attachment formed is very relevant in the quality of survival..
In 1971, Ainsworth et al., cited in Harris (1995), devised the 'Strange Situation'. Three attachment categories were derived from these studies: Type A, Anxious Avoidant; Type B, Securely Attached; and Type C, Anxious Resistant. These formed a basis for a series of studies of attachment worldwide with only one addition since, Type D, Disorganised and disoriented Main & Soloman (1986) (cited in Durkin, 2000, p.94). These classifications are generally held to remain relatively stable throughout life suggesting that adult attachments mirror the attachment formed with the primary caregiver in early life (Fonagy et al (1994)) thus inferring less comfortable outcomes in life for types A, C and D with type D generally being held to predict child abuse.
Lamb et al (1985) cited in Durkin (2000) p.107, would argue cultural limitations on the Strange Situation but Van Ijzendoorn (1990) suggests this simply underlines its universality. Interestingly, deMause (1994) outlines cultures in New Guinea (the Bimin-Kuskusmin), India and China that habitually perform what we would consider as child abuse including incest, seen as totally acceptable within that culture. It must surely be considered doubtful that these children would demonstrate Type D attachment behaviour as a result of, what is to them, normal acceptable and unthreatening behaviour. This incites a strong argument for the external socio-cultural pressures predicting behaviour rather than attachments formed.
Carver (1980) pp.52-67 outlines many indicators of child abuse including physical signs and behavioural aspects like "frozen watchfulness" which fit well with the Disorganised, Type D attachment behaviour. She also remarks that "the quality of the parent/child relationship at home" can give clues as to whether a child is being abused, further supporting attachment theory but, included in a list of causative factors in abusive families are, "one parent family, unusual family structure, unemployment, low income, low educational attainment and occupational status of parents, marital problems, mental illness, isolation and psychosocial malfunctioning". This telling list could also suggest that abusive behaviour is not necessarily directly related to the abuser's own childhood attachment formation, but perhaps to a change in environmental causes such as lack of social support due to scattered family or merely socio-economic difficulties perhaps due to divorce, redundancy, death of a partner or even due to congenital retardation.
The implication that the underlying community dynamic of child abuse is considered to be the breakdown of the family is supported in a seminal British study by Whelan (1993), who confirms that a child is safest from abuse when its biological parents are married and least safe when the mother is cohabiting with a man other than her husband. Fagan et al (1997) reiterates these findings in his American study and finds that research on crime and delinquency in both the U.S. and Britain illustrate similar social trends and relationships between family breakdown and social problems. They also cite findings remarkably similar to Carver's intimations in that the subculture of abuse is more prevalent in communities where "marriage is less common, individual families are more isolated, alcohol abuse is widespread and drug trafficking is high". Further, they state there is an acceptance among men in high abuse communities that abusing women is normal and acceptable and that low income is a common factor. It cannot, therefore, possibly be considered feasible to suggest that abuse and neglect of children stems solely from parents without secure early attachments, rather than the social influences of the apparently dysfunctional community around it.
Arguably the "nuclear family" of the 60's has undergone such radical changes as to be virtually unrecognisable. Taking stable attachment to its logical conclusion, unless you are a "Type B", securely attached, you are unlikely to form successful relationships and must therefore be responsible for the breakdown of the "family unit", whatever that is. Could it not merely be that we are not psychologically adapted for monogamous relationships lasting throughout our lifespans which, due to "progress", have extended considerably. Marriages lasting 25 years can now last twice as long with parties changing considerably throughout. With rising divorce rates, social acceptance of sexual relations outside marriage, single sex partnerships, commonness of cohabitation and decrease in localisation of supportive family groups all contributing to variations in the construct of what a "family" is, there has been a huge impact on the way in which children are raised and by whom. Yet again a circularity of argument arises - can attachment type realistically be said to be to blame for divorce, or is divorce responsible for the attachment type?
That parental separation as in divorce, causes distress in children in nearly all cases, varying with their age and level of understanding, has been well researched Bernstein and Brannen (1996) p.189. Growing awareness and understanding through pre-school to adolescence alters whom children blame and how they cope. Grych and Fincham (1990) examined effects of intense marital conflict and violence in the home establishing the damaging effects it had on the children. It is arguable which is more damaging, the effects on the parents themselves, impairing their health and parenting abilities or the hearing/observing of the conflict/violence. Amato (1993) cited in Bernstein and Brannen (1996) holds that the parenting style and relationship with the child is of more importance than whom the relationship is with. Although post-divorce children do not necessarily suffer from being brought up by one parent, financial stress does impact on a child Oppenheim (1993) and female-headed households tend to be less well off.
Resultant of divorce, change in home and or school environment causes stress for children and the re-marriage of the primary caregiver may be causal in additional difficulties depending whether it brings conflict or additional support! In summary it is difficult to decide whether a secure attachment would enable a child to survive divorce or whether environmental pressures disturb it suggesting that attachment is subject to change and cannot be stable over time. If this is the case it means we cannot necessarily assume the parental attachment is mirrored in their current attachment with their child. Additionally relationships with siblings may well prove to be the critical factor in surviving divorce linking with Freud and Dann's (1951) study.
If, indeed, attachments between primary caregiver and child are a result of the parent's internalised childhood experiences, then it would seem logical that children of that parent should develop identical attachments which is clearly not the case. Harris (1995) posits that within-family environmental differences such as child-driven effects, parent-driven effects and family context such as birth-order effects can explain the variance. A first-born child, for instance, will have a different experience to its siblings. It will have been the sole recipient of parental attention for at least a year, for which it must then enter into competition (and for what it may consider as "it's" possessions) with the newcomer. The second-born has never experienced life without an older and more experienced sibling and the parents have "cut their teeth" on the first born and may be more relaxed and adept at handling a child by the time the second born has arrived. Teti and Ablard (1989) cited in Durkin 2000, p.126. found that Type B infants "developed more positive relationships with their siblings" firmly supporting attachment's predictive ability although Dunn cited above would not discount "child-driven effects". Dunn and Kendrick (1982) also cited in Durkin 2000, p.127 highlight the importance of the attachment with the older child. This suggests that in addition to the question of what influences later relationships; primary caregiver or socio-cultural/economic environment, the attachment to an older sibling must be considered as a possible model for later relationships.
Whilst attachment types have supplied a predictive formula for characteristics in later life and indeed, claimed some success (Fonagy, Steele and Steele (1991) Hazen and Shaver (1987)), there are many variables involved in individual development apart from the relationship with the primary caregiver. For instance: the biological trajectory, which "may wax and wane at different periods and may differ for different processes" examples given are the adolescent growth spurt and "growing heritability for IQ in adolescence" Wilson (1985); the psychosocial trajectory and "interactional/transactional processes", Scarr and McCartney (1983) and chance encounters or events (Bandura (1982) both cited in Bernstein and Brannen (1996). Reflection on this complexity of variables is indicative of the problems encountered with predictions in individual development and Dunn, (1993) p114, would argue that the use of the attachment framework for predicting relationships in later life is "both limited and limiting". Indeed, it is doubtful whether attachment categorisation would hold true for children with congenital abnormalities and special needs such as autism, which classification alone offers predictive potential in its own right. In retrospect, that attachment categorisation alone can accurately predict a child's circumstances or future appears to be a somewhat of a tenuous claim and raises the question as to whether an attachment construct has any value at all from logical and scientific point of view (Weinraub et al. (1977).
Ainsworth, M.D.S.(1989). Attachments Beyond Infancy. American Psychologist, 44:4 709-716.
Atkinson, R., Atkinson, C., Smith, E., Bem, D., Nolen-Hoeksema, S., (2000) Hilgard's Introduction to Psychology, Harcourt Brace College Publishers: Orlando (13th edition)
Bernstein, B. and Brannen J., (Eds). (1996). Children, Research and Policy. Taylor and Francis: London.
Bowlby, J. (1998). A Secure Base: Clinical Applications of Attachment Theory. Routledge: London
Carver, V. (1980). Child Abuse, A Study Text. The Open University Press: Milton Keynes
DeMause, L. (1994). A History of Childhood As the History of Child Abuse. The Journal of the International Primal Association, 11: 48-62 .
Durkin, K. (2000). Developmental Social Psychology: From Infancy to Old Age. Blackwell Publishers: Oxford
Dunn, J. (1993). Young Children's Close Relationships. Sage Publications: London
Fagan, P.F., FitzGerald, W.H.G. Snr., and Hanks, D.B. (1997). The Child Abuse Crisis: The Disintegration of Marriage, Family and the American Community. The Heritage Foundation Roe Backgrounder, 1115.
Fonagy, P., Steele, M., Steele, H. (1991). Maternal representations oa attachment predict the organization of infant-mother attachment at one year of age. Child Development, 62: 891-905.
Fonagy, P., Steele, M., Steele, H., Higgitt, A., Target, M. (1994). The theory and practice of resilience. Journal of Child Psychology and Psychiatry, 35: 231-257
Gross, R., (1999). Psychology, The Science of Mind and Behaviour Hodder and Stoughton: London
Grych, J.H., Fincham, F.D. (1990). Marital conflict and children's adjustment: A cognitive-contextual framework. Psychological Bulletin, 108: 267-290.
Harris, J.R. (1995). Where Is the Child's Environment? A Group Socialisation Theory of Development. Psychological Review, 102:3, 458-489.
Hazen, C., Shaver, P. (1987). Romantic Love Conceptualized as an Attachment Process. Journal of Personality and Social Psychology, 52:3, 511-524.
Rutter, M. (1991). Maternal Deprivation Reassessed Penguin Books: London (Second Edition).
Van Ijzendoorn, M. H. (1990). Developments in cross-cultural research on attachment: some methodological notes. Human Development, 33:3 3-9.
Verschueren, K. and Marcoen, A. (1999). The Developmental Patterns in Security Attachment to Mother and Father in Late Childhood. Child Development, 67:5 2493-2511.
Weinraub, M., J. Brooks, & M. Lewis. (1977). The social network: a reconsideration of the concept of attachment. Human Development, 20: 31-47.
Whelan, R. (1993). Broken Homes and Battered Children: A study of the relationship between child abuse and family type. London: Family Education Trust
Wilson, R. S. (1985) The Louisville Twin Study: developmental synchronies in behaviour. Child Development, 54: 298-316.