Schaffer and Emerson (1964, cited in Cardwell et al., 2003) found that the first attachment is generally made at between 6-8 months of age, when the infant demonstrates separation anxiety, with stranger anxiety being displayed approximately one month later. This links in with Piaget’s cognitive developmental theory and notion of ‘object permanence’, as it appears that when the infant becomes aware that they are separate to their caregiver, they begin to display anxiety when that person goes out of sight. However, Schaffer and Emerson (ibid.) argue against the notion of ‘monotropy’ (Bowlby, 1969, cited in Woodhead et al.,1991) stating that rather than only one main attachment being formed initially, followed by several other attachments, infants are able to attach to a number of individuals equally and that they all form qualitatively different yet similarly meaningful relationships.
They also considered the strength of attachments as being important, stating that highly responsive mothers tend to produce intensely attached infants, whereas lack of interaction leads to a much weaker attachment. It wasn’t always the case that the main attachment was formed to the mother however, (ie. the one who bathed, fed and changed the infant), the one who played and interacted socially with the child being almost as likely to become the main attachment figure. Therefore there is a need to be more than just present to form a secure, beneficial attachment. This has major implications for mothers suffering with post-natal depression, as although they may carry out basic care needs, if they are depressed, show flat effect and don’t play, smile or interact socially with the infant then they may not become the primary attachment figure. Although this is not necessarily detrimental to the child in itself, if there were to be no other caregivers around then this could have potentially damaging consequences for the infant. In this sense, it could be argued that mother and baby units for mothers suffering from post-puerperal psychosis are not necessarily the best option. Perhaps then the mother should be given the time to recover whilst the baby is cared for by other significant people who ensure a climate which is conducive to bonding during periods of contact between mother and child. In the case of a lack of any other family or close friends, this could be a foster carer. Indeed, Robertson (1970, cited in Bowlby, 1988) felt that separation need not lead to deprivation if good substitute care is provided and if the separation is minimised. Clarke and Clarke (1976) also argue that early experiences can be both forgotten and overlaid and therefore the early years are not as crucial as Bowlby and others maintain. However, it is also important to consider this from the point of view of the mother, as although the infant may not specifically remember this period of time, the mother will remember and therefore it is crucial that her needs be given due consideration and every opportunity be made available for contact with the infant in order that she does not begin to, or have to, ‘detach’ herself.
Contrary to the opinion of Clarke and Clarke (ibid.), Walsh et al (2009:online) examined children in middle childhood whose mothers had been hospitalised due to mental health issues, finding that they were ‘…less emotionally open and secure, and generated fewer adaptive coping strategies…’, linking maternal depression to children’s emotional and cognitive difficulties. A further complication, as noted by Sheppard (2002), is that depressed mothers also have difficulty in working in partnership with services and therefore this poses a problem for social workers and a dilemma in terms of balancing risk with supportive and useful intervention.
If depressed and alcohol / drug dependent mothers have difficulty in prioritising their child’s needs above their own, this can not only have a negative impact on attachment (Woodcock & Sheppard, 2002, cited in Taylor and Kroll, 2004) but raises safeguarding issues. Mental health practitioners who focus on the mother’s mental health needs and children’s social workers whose focus is child centred may tend not to take sufficient account of the other’s perspective in order to adequately balance the needs of their respective clients and thus produce the best outcome for the mother and child. This is therefore a multi-agency practice issue that would benefit from further investigation. In fact, Sheppard (1994), who found a clear relationship between maternal depression and child care difficulties in terms of both the mother’s ability to care for the infant and her perception of the child’s behaviour, argues for the connection of mental health and child care services, given the likely high volume of mothers who come to the attention of services and who are likely to suffer from depression.
Cox et al (1987), on the other hand, conducted their own studies into postnatal depression and found that not all depressed mothers fail to engage sensitively with their child. The study found postnatal depression was just one factor in the mother-infant relationship, and that mothers who had poor relationships with their children and subsequently children with disturbing behaviour, were more likely to have problems in their marriage, be younger when they had their children and may not have had particularly responsive parents themselves. Hence, depression in direct relation to childhood or later behavioural problems is not clear cut and involves many other influential factors which must be considered.
Bowlby also advocated the concept of a ‘critical period’, stating that if an attachment had not developed before 2 ½ years old then it never would. If this belief is upheld then this could lead to a feeling of uselessness in terms of social work intervention and therefore possible inaction on the part of the authorities. Again, this concept has been highly criticised by various researchers. Tizard et al (1978, cited in Woods, 1997) showed that children adopted from an institution were able to form attachments with their adopters beyond the said critical period. These children also showed greater intellectual and social development than those who were returned to their own families, suggesting that the birth family may not always be the best place to raise children. This has important implications for practice and could be seen to conflict with the ethos of the Children Act 1989, which states that local authorities have a duty to promote the upbringing of children within their families. Similarly, Tizard (1999, cited in Clarke and Clarke,2000) also states that a major error of social work practice is to delay permanence in favour of supporting parents until such a point that the child becomes difficult to place. This seems to support Bowlby’s theory, particularly the concept of a ‘critical period’ and the impact that the lack of sensitive caregiving can have on future behavioural difficulties.
Bowlby (1946; 1951, cited in Gross, 1992) also believed that prolonged separation of the child from the mother (‘maternal deprivation’) was a major cause of ‘delinquent’ behaviour and mental health issues. This hypothesis is separate to, and came some twenty years prior to, his attachment theory. Emotional ‘deprivation’, it was felt, may harm social and emotional development, whereas ‘privation’ (the lack of the formation of any attachment and a chronic lack of basic care and stimulation) would cause harm to the emotional and social development and in this sense would be irreversible. After the age of five, it was suggested that children are better able to cope with separation, but until then, they would need a single, continuous, unbroken attachment. However, Bowlby did not say that this had to be the mother (Crawford & Walker 2007). Also, attributing problems in behaviour, and in later life, to maternal deprivation, Bowlby has been said to deny the impact of other factors, especially the impact of privation. Rutter et al (1998, cited in Clarke and Clarke, 2000) studied Romanian orphans adopted in the UK and found that a later adoption age led to slower progress, thus concluding that a longer experience of privation can result in a longer recovery period.
Rutter (1981) concluded that delinquency and disturbance were most commonly found in children who had been reared in unhappy homes, regardless of whether the home was ‘broken’ or whether the children had suffered from maternal deprivation. He argued that the juveniles that Bowlby had used in his study had never been looked after by their families and that nothing had been taken away from them, therefore they could not be seen as ‘deprived’. Rutter (ibid) also pointed out that Bowlby’s theory denies the opportunity to reverse the effect of negative early experiences, equally feeling that positive experiences in early life do not necessarily protect a child from later emotional damage (cited in Crawford & Walker, 2007).
Indeed, Bowlby’s theories of attachment and maternal deprivation have been criticised on a number of levels. His methodology is questionable in terms of unrepresentative samples and the use of a retrospective case study method. The concept of ‘monotropy’ has also been questioned as to its cultural validity. Does attachment theory apply universally because it is an innate concept or is it culturally bound? Generalisations about ‘ideal’ attachment behaviour or patterns tend to be made in a Eurocentric context. Studies conducted on infants in Northern Germany, Japan and those in Kibbutz care in Israel found that attachment theory could not be reliably applied cross culturally (Goldberg, 2000). Grossman et al (1985 cited in Goldberg, 2000) found that German mothers value early independence and hence see avoidant attachment as the ‘ideal’ rather than the universally accepted preference for secure attachment patterns. In theory German parents like to keep a greater parental distance than American parents, preferring a child who obeys and does not question. Research has found that there is no evidence that the larger number of anxious/avoidant German children is because German mothers are unresponsive, indifferent or insensitive but rather they merely value and thus nurture, independent children. In this respect, perhaps to a German parent a secure American child would be seen to be ‘clingy’ and ‘spoilt’ and perhaps therefore overly attached to a detrimental level.
Similarly, Takahashi (1990, cited in Robinson, 2007) did not find consistency across cultures. Studying Japanese children, he found infants to demonstrate extreme distress on separation. As Japanese children tend never to be left alone (they sleep and bathe with caregivers), the ‘Strange Situation’ induced severe distress (ethically questionable in itself), thereby leading to a larger number of Japanese children being rated as insecurely attached. Japanese children are also taught that it is impolite to show avoidance, even to strangers, and therefore their lack of ‘stranger anxiety’ in this case could be due to socialisation rather than a faulty attachment.
Fox (1977, cited in Cardwell et al, 2003) also studied the Israeli kibbutz where multiple attachments to the mother and nurse (‘metapelet’) were found. The infants protested equally on being separated from the mother or the metapelet, but were more comforted by their mother on reunion. These studies found a very different distribution of attachment patterns compared to those found in middle class North American samples. This is not to say that the concept of attachment is not important because attachment relationships occur and impact upon humans universally. However, measuring attachment using the ‘Strange Situation’ during infancy in these cultures raises interesting questions, including the validity of the assessment in regard to cultural influences and practices. Further criticisms include the fact that the ‘Strange Situation’ does not measure the child but measures specific relationships and as only the relationship with the mother is usually examined (Lamb 1984, cited in Woods, 1995), other equally important attachments are excluded.
However, the ‘Strange Situation’ has also rendered some similar findings across different cultures in terms of ‘monotropy’. Ainsworth (1967, cited in Bowlby, 1988) studied the pattern of care of Ugandan infants who have multiple carers, finding that these infants still formed one main attachment. Similarly, Tronick et al (1992, cited in Robinson, 2007) studied infants from an African tribe where the norm is for extended family groups to care for infants and for infants to be breastfed by different women but sleep with their mothers. Here, the main attachment again was found to be to the mother, despite the numerous people involved in the care of the infant. Although some consistency demonstrates that there may be universal characteristics that underpin infant and caregiver interactions, significant variations demonstrate that universality is somewhat limited. However, as Fahlberg (1991; 1994, cited in Wilson et al, 2008) has demonstrated, whilst working with problems of attachment, the process of attachment appears to be culturally similar but with varieties in how attachment happens. Here, the cycle of arousal and relaxation which takes place when a child has a need which needs responding to and the carer is responsive to that need, has, at its centre, concepts of security and trust. It is argued that the security which results from trust not being broken crosses many cultural boundaries.
The parental role therefore seems to be a central factor in the resultant behaviour of the child, but this must not be considered in isolation to many other varying factors. A supportive and responsive relationship between parent and child appears to encourage normative development, whilst a negative, non-responsive relationship could be considered to be a key factor in the development of disturbing, maladaptive behaviour. However, although this assumption is supported by many studies, this explanation is perhaps too simplistic and ignores other factors. A key ‘other factor’ is the child themselves. In this respect, attachment theory has been criticised for its claims of universality from an individualist perspective. Dunn (1993) believes that individual differences in children affect the relationship they have with their parents, thereby suggesting a child’s temperament and personality can impact on attachment style ‘tests’:
To any parent with more than one child, the idea that differences in their relationships with their various children are attributable solely to parent differences is hard to reconcile with their own experience of the impact of their children’s individual differences on other family members.
(Dunn, 1993:36).
It is not useful therefore to associate set patterns with secure attachment, classifying them as either ‘normal’ or ‘abnormal’. ‘Instead, think of them as children adapting to the relationship with their attachment figure and the circumstances they are in’. (Wilson et al, 2008:42). An example of this is how attachment may be measured in children who are unable to interact, who perhaps have a physical or learning disability.
Contrary to the findings of researchers such as Ainsworth et al. (1978, cited in Howe, 2005) who feel that specific attachment types are caused by the mother’s behaviour and her sensitivity, Kagan’s (1984) ‘temperament hypothesis’ advocates that some infants are born to be more sociable and this in turn can lead to more secure attachments. An innate predisposition to stress, for instance, could cause ‘stranger anxiety’, whereas a high level of sociability in an infant could lead to a complete lack of distress on separation from their caregiver. Here therefore, the sensitivity of the caregiver is not necessarily of any central concern. Rather the temperament and personality of the child are deemed to be equally as crucial. In a relationship between two people, both drive the nature of the relationship and have an affect on it. The way that the infant behaves and interacts, her ‘temperament’, can therefore affect the way that she is treated by her caregivers.
Another crucial factor in terms of individual differences is that of resilience, which is felt to help to protect against mental health problems in later life. Better resilience to faulty attachments is felt to stem from both personal characteristics and environmental influences, with some positive interactions, experiences and relationships (ie. with a teacher or other adult) arguably being enough to protect against long term detrimental effects (Clarke and Clarke, 2000). Most of the children in Bowlby’s studies not only lacked maternal care but lacked care in general and so it is dangerous to make assumptions about the impact solely of maternal deprivation.
There has been an extensive feminist critique of Bowlby’s post-war work into separation and loss, his pre-occupation with the mother as primary caregiver and his concept of ‘maternal deprivation’. As Burman (2008:132) states:
Women were portrayed as devoted to, attending to and attuning to
their children; they bask in her presence while she orients to their
unspoken needs.
Here then the mother’s role is perfectly laid out and any deviation for this social norm leads to blame, without any examination of the part that society has played in the role that has been created for women. In fact, the social construction of childhood and motherhood, which is still very much currently portrayed via the media, places great emphasis on the mother-child bond, with concepts such as ‘breast is best’, the importance of skin to skin contact following birth for bonding purposes and the concept of the ‘maternal instinct’ that will come naturally to any woman who gives birth. This all then results in an immense pressure on the woman to be the ‘perfect mother’, despite the sleepless nights, the endless nappies, the mastitis and the episiotomy! Despite attempts by professionals to relieve any guilt felt by the mother through not aspiring to these expectations, society still does not look kindly on the mother who lacks any maternal drive or instinct, who puts a career before the care of her children, or who takes ‘time out’ for herself. As Burman (1997) states ‘…instead of poverty, unemployment and frustration we have evil children, bad mothers, absent fathers and broken homes’. There is a failure then to recognize the impact of poverty and social conditions as a mitigating factor in family difficulties which may impact firstly on the infant’s attachment and secondly on the family as a whole.
As Bowlby stated that ‘…a bad home was better than a good institution’ (Burman, 2008:131), we have to now question whether things have in fact moved on from this perspective, despite Burman’s ascertion that ‘…the wisdom [of the statement] has increasingly been challenged’ (p.131). Often children deemed to be ‘in need’ or ‘at risk’ are found places within nurseries or with child minders in order to safeguard and promote their well-being, whilst keeping them within their families, rather than removing them into foster care or residential care. Does this therefore reinforce Bowlby’s concept and, as such, present a counter criticism to feminism, particularly considering the many number of children’s homes that have now been closed down?
In defence of Bowlby’s work however, although he does not retract from his original thoughts about the necessary pre-requisites of securely attached children, his later research does begin to include the role of the father and other caregivers (Bowlby, 1988). He still advocates for sensitive and attentive care-giving in a secure family home but also stresses the importance of supporting those parents who struggle with this, either because of their own upbringing or because of other adversarial conditions. In this sense then he has ‘moved with the times’ somewhat and has acknowledged further research that has been undertaken in relation to attachment theory whilst also protecting his original ideas. The fact that his theory has generated such a vast amount of further research is evidence in itself of its importance. Although Bowlby did not originally consider other kinship arrangements, as would be advocated today, his defence here may be in considering the reasons as to why he should be expected to be ahead of his time. His theory was based around the current societal happenings and expectations, the nuclear family and the centrality of the mother being exactly those. We must remember that Bowlby did not create those conditions but merely conducted research within them.
Despite the various criticisms and counter-criticisms outlined above, the assessment of attachment plays a key role in social work with children and families and thus has a significant impact on professional practice. Brandon et al, (1998:40) state that:
attachment is a central concept in social work practice with
children because it is most often in the area of children’s close family relationships that we see the origins of emotional and behavioural problems.
The Department of Health (2000) states that the ‘nature and quality of early attachments’ and ‘a stable and affectionate relationship with parents and caregivers’ are key dimensions of a child’s developmental needs. Where there are concerns about a parent’s capacity to meet these needs, although there is value in supporting parents to make positive changes, the focus must always be the needs of the child. As Atwool (2005:230) points out:
behavioural management strategies and individual counselling
are largely ineffective when unmet emotional needs and conflicted
relationships continue to characterise the child’s environment.
This study has demonstrated the importance that should be placed on viewing child developmental theory, and indeed the concept of childhood per se, within its historical context and understanding that the interpretation of theory as ‘fact’ can serve to reinforce the political, economic and social values of the time. Bowlby’s post war research and resultant maternal deprivation theory being a prime example of how the economic climate at the time acted as a catalyst for the theory to be widely accepted by society, resulting in mothers returning from the workplace (thereby freeing the jobs for the men returning from war) and back into the home.
In summing up the relevance of attachment theory today, it would appear that Bowlby may have been mistaken when he proposed a ‘critical period’ for human infants to form attachments. Also, Bowlby was not able to show whether all those who suffer maternal deprivation later display maladaptive behaviours. In the studies that link maternal deprivation and delinquency, maternal deprivation is only one factor amongst many others. Childhood experiences that span Bowlbys’ critical years do not necessarily produce irreversible detrimental effects. However, the miscellaneous clinical experiences of psychiatrists suggest that the inability to form an attachment may socially and emotionally disadvantage an individual throughout their life. In terms of relevance to current practice, Aldgate and Jones (2006:85) state that:
Attachment today is just as important for children as it was
when Bowlby and Ainsworth carried out their pioneering
work. For this reason, practitioners in different disciplines
need to be able to assess attachment in order to safeguard
the welfare of the developing child.
It is clear that developmental psychology must not underestimate the enormous contribution that attachment theory has made to understanding socio-emotional behaviours or to the importance of the attachment bond during infancy. Recognising that secure attachment relationships are a huge contributor enabling people to explore the world during all stages of life, but also appreciating the adaptability and flexibility of the human being to adjust to different circumstances is necessary. The evidence reviewed here has found mixed results despite seeming to suggest that a secure attachment during infancy is more likely to produce positive later outcomes. However, consistently the research questions whether later outcomes are actually the result of a secure infant attachment relationship, or whether they are related to other variables such as infant temperament and stable family and environmental circumstances, including the emotional availability and mental health of the mother. Overall, the evidence seems to be that whilst the attachment bond during infancy is important, it is not realistic to view it as ‘crucial’ for later outcomes. Clarke and Clarke (2000:103), state that:
one must take account of the ever greater number of factors known
to be relevant over and above early experience itself. Subsequent
influences can modulate the earlier effects; we know that early
experience does not have a dominant role on its own and that reversal
may be feasible.
They summarise (2000:105) succinctly by stating that ‘theories ascribing overwhelming, disproportionate and pre-deterministic importance to the early years are clearly erroneous’. Similarly, Bronfenbrenner (1977; 1983, cited in Clarke and Clarke, 2000) advocates looking at the wider context and inter-relationships within society in general. We must conclude then that it is not just the early years, but lifelong experiences and interactions of these experiences, including major life transitions and how these are dealt with that contribute towards the development of the whole person. Again we must reflect on Runyan’s (1978, cited in Sugarman, 1986:2) statement on the importance of the whole life span, along with giving due consideration to both individual and environmental factors which influence, predict and determine the outcome of the person, be that during infancy, adolescence, adulthood or old age.
REFERENCES
Aldgate, J., and Jones, D. (eds.) (2006) The Developing World of the Child. London: Jessica Kingsley
Atwool, N. (2005) ‘Working with Adults who are Parenting’ in Nash, M., Munford, R. and O’Donoghue, K. (eds.) Social Work Theories in Action. London: Jessica Kingsley Publishers, pp229-231
Berlin, L., Cassidy, J. & Belsky, J. (1995). Loneliness in Young Children and Infant-Mother Attachment: A Longitudinal Study. Merrill-Palmer Quarterly, Vol. 41, no. 1, pp91-103
Bowlby, (1988). A Secure Base: Clinical Applications of Attachment Theory. Abingdon: Routledge
Brandon, M., Schofield, G., Trinder, L and Stone, N. (1998). Social Work with Children. Basingstoke: Macmillan
Burman, E (2008). Deconstructing Developmental Psychology 2nd ed., London: Routledge
Cardwell, M., Clark, L. and Meldrum, C. (2003). Psychology for AS Level. Hammersmith: Harper Collins Publishers Limited.
Children Act 1989 (c.41) London: OPSI
Clarke, A. and Clarke, A. (eds) (1976). Early Experience: Myth and Evidence. London: Open Books
Clarke, A. & Clarke, A. (2000). Early Experience and the Life Path. London: Jessica Kingsley Publishers
Cox, A., Puckering, C., Pound, A. and Mills, M. (1987). The Impact of Maternal Depression on Young Children. Journal of Child Psychology and Psychiatry and Allied Disciplines Vol. 28, no. 6, pp917-928
Crawford, K. and Walker, J. (2007) Social Work and Human Development (2nd ed.) Exeter: Learning Matters
Department of Health (2000). Framework for the Assessment of Children in Need and their Families. Cmnd. London: TSO
Dunn, J. (1993). Young Children’s Close Relationships: Beyond Attachment. California: Sage.
Goldberg, S. (2000) Attachment and Development. New York: Arnold
Gross, R. (1992). Psychology: The Science of Mind and Behaviour (2nd ed.) London: Hodder and Stoughton
Howe, D (1995) Attachment Theory for Social Work Practice. London: Macmillan Press
Howe, D., Brandon, M., Hinings, D and Schofield, G. (1999) Attachment Theory, Child Maltreatment and Family Support: A Practice and Assessment Model. London: Palgrave Macmillan
Howe, D. (2005) Child Abuse and Neglect: Attachment, Development and Intervention. Basingstoke: Palgrave Macmillan
Kagan, J. (1984) The Nature of The Child. New York: Basic Books
Lamb, M. (1987). Predictive Implications of Individual Differences in Attachment. Journal of Consulting and Clinical Psychology, Vol. 55, no.6, pp817-824.
Parkes, C., Stevenson-Hinde, J. & Marris, P. (1991). Attachment Across the Life Cycle. London: Routledge.
Robinson, L. (2007). Cross-Cultural Child Development for Social Wokers: An Introduction. Basingstoke: Palgrave Macmillan.
Rutter, M (1981) Maternal Deprivation Reassessed (2nd ed.) Harmondsworth: Penguin,
Sheppard, M. (1994). Maternal Depression, Child Care and the Social Work Role. British Journal of Social Work, Vol. 24, pp33-51
Sheppard, M. (2002). Depressed Mothers’ Experience of Partnership in Child and Family Care. British Journal of Social Work Vol. 32, pp93-1112
Sugarman, L. (1986) Life-Span Development: Concepts, Theories and Interventions. London: Routledge
Taylor, A. and Kroll, B (2004). Working with Parental Substance Misuse: Dilemmas for Practice. British Journal of Social Work Vol. 34, pp1115-1132
Thompson, R. (2000) The Legacy of Early Attachments. Child Development, Vol. 71, no.1, pp145-152
Walsh, J., Schofield, G., Harris, G., Vostanis, P., Oyebode, F. & Coulthard, H. (2009). Attachment and Coping Strategies in Middle Childhood Children Whose Mothers Have a Mental Health Problem: Implications for Social Work Practice. British Journal of Social Work, Vol. 39, no. 1, pp81-98
Wilson, K., Ruch, G., Lymbery, M & Cooper, A. (2008) Social Work: An Introduction to Contemporary Practice. Essex: Pearson Education Ltd
Woods, B. (1995). Basics in Psychology. London: Hodder & Stoughton
Woods, B. (1997). Discovering Psychology. Abingdon: Hodder & Stoughton
Woodhead, M., Carr, R and Light, P (eds.) (1991) Becoming a Person. London: OUP
.