Attachment Theory : how children are able to make relationships

Date Posted: Friday, 27 February 2009

I wrote this article in 1989 for adult students at an evening class I was teaching at a local College of Further Education. The student group was made up of mothers who wanted to pursue their interest in the development of their young children and it also consisted of child care and youth workers who felt the course would might help them understand their work better. The article is perhaps not as I would have written it today – for instance I think I would now use the words ‘relationships with others’ rather than socialisation – and I would have dwelt less on Sigmund Freud’s theories, important though I still think they are.  In 2007 I added some additional material relating to the work of Jeremy Holmes and Peter Fonagy.

The nature of the article is introductory and it is one of a series about different psychological theories which have informed and influenced those who have  had, in one way or another, a responsibility to look after children and young people. In my view the theories are not prescriptive but they can help inform thoughtful and reflective child care.

Attachment Theory : children are able to make relationships 

Human beings seem to have a strong tendency towards sociability. From birth the infant appears to have the facility to respond to other human beings, and a number of studies have shown how very young babies respond in particular to the stimulus of the human face(see for instance, Diamond & Carey, 1977 and more recently De Haan, Pascalis & Johnson, 2002).The process of socialisation is continuous from birth but it is now generally agreed that an attachment bond to a parenting figure marks the child’s first step towards socialisation. In this article I give attachment theory my own narrative which starts with Sigmund Freud and reaches the present day with the ideas of Jeremy Holmes and Peter Fonagy

Sigmund Freud

Freud, though he never took a theoretical stance which was specifically concerned with the idea of attachment, nevertheless believed that a human being’s capacity to make attachments to other people was the consequence of a process which occurs during the first five years of life. The process was related to the different psychosexual stages which he proposed a child goes through. According to Freud the important stages were the oral stage, the anal stage, and the phallic stage, (the latter at about 4 to 5 years of age). Freud’s emphasis reflects a culture and society which was a male dominated one and it has been argued his theoretical positions relate more specifically to boys rather than girls. His principal observations on the attachment of a child to an adult concern his notion that young boys develop an unconscious desire for a sexual relationship with their mothers, and begin to see their fathers as rivals for their mothers’ affections. Freud’s view is that a boy eventually would, “Identify with the aggressor”, (his father) and join sides with him for fear that his father would ‘castrate’ him. This theory of the ‘Oedipus complex’is based on the story of King Oedipus. Freud suggests that it is during the phallic stage, that the boy resolves the Oedipus complex by learning to identify with his father. Freud and later psychoanalytic theorists applied a similar theory relating to girls. This theoretical stance suggests that girls unconsciously develop what was termed as “penis envy”, believing themselves to be already castrated. This leads a girl to seek a strong love relationship with her father. The notion is that she will replace the lost penis by having her father’s child, but when it is apparent that this is not an acceptable wish, the girl finally identifies with her mother, to become like her, and so resolve what is termed as the ‘Electra complex’.

According to Freud the personality developed in childhood through the child’s interaction with parents. For Freud the personality consisted of three elements : the id, the ego and the superego. A balance between these elements means that the personality as a whole is balanced, and the person will experience no serious emotional problems. The superego is said to operate as a kind of conscience and takes the role of an authoritarian father. Since the adult personality would depend on how well the child and parent formed an attachment, if the superego (the authoritarian father) is too well developed, then the person will feel oppressed.

Klein, Winnicott, Erikson and Bion

Subsequent psychoanalytic theorists, such as Klein, Winnicott, Erikson and Bion, suggested that a well integrated child is one for whom the attachment between the infant and a parenting figure – usually the mother – is engendered within a holding or containing environment which allows the infant time to establish a sense of being an individual who is separate from the primary caring figure. Klein (1952) emphasised the important role the parenting figure has in holding the infant’s primitive fears and anxieties. Winnicott and Bion with their respective notions of ‘the facilitating environment’ (Winnicott,1965) and ‘maternal reverie'(Bion,1962) place a great deal of emphasis on the significance of mutuality in the primary attachment relationship. For them this relationship has a different but equally as intense significance for the maternal figure as it does for the infant.Erikson (1950) believed a healthy consistent attachment relationshiplead to the child being able to develop trusting relationships.

John Bowlby

Since the 1950s John Bowlby’s work on attachment has been influential. His theoretical position was influenced by Harlow’s experiments with rhesus monkeys which concluded that there was a critical period for attachment. Harlow in his research with rhesus monkeys showed that infant monkeys needed comfort and security. Harlow concluded if infants were separated from their parents at birth, they missed a crucial period of attachment formation, which had negative implications for their socialisation in later life. Although drawing similarities between monkeys and children should be treated with caution, and however unethical we may find Harlow’s studies today, his findings support theories of attachment and socialisation. Bowlby suggested that unless firm attachment was formed between the child and his mother within the first five years of life, the child would develop an affectionless psychopathy that is being unable to feel any warmth for anyone else, or to show any concern for their welfare (Bowlby,1988). According to Bowlby separation from the mother, “maternal deprivation” could result in physical and psychological problems, and delinquency in adolescence. After studying 44 delinquent adolescents, 17 of whom had experienced separation from their mothers before the age of five years, Bowlby concluded that maternal deprivation was a cause of delinquency during adolescence

Bowlby’s position is called monotropy theory, in that the child relates only to its mother. Michael Rutter (1981) disagreed with Bowlby. He studied a group of adolescent boys to see if there was a relationship between delinquent and anti-social behaviour and early separation due to hospitalisation and also due to family problems. He found that when such children returned to a stable environment, they would settle down and become less inclined to anti-social behaviour. While Rutter concluded that family arguments and unsettled family circumstances were the causes of anti-social behaviour, he argued that Bowlby was wrong to relate his work to Harlow’s studies with monkeys because firstly, Harlow had conducted privation studies, (lack of mother) while Bowlby had conducted deprivation studies (losing of mother), and also that Harlow used an experimental method, while Bowlby used an observational method.

Anna Freud

Anna Freud, conducted a series of case studies of a group of 6 children from a war time concentration camp who had been orphaned. She discovered that though the children experienced difficulties in their attachment to adults, they were firmly committed to each other. They regarded their peers as the central figures of attachment, rather than their parents. This work like Rutter’s suggests that bonding with the mother is not always necessary for successful attachment and socialisation (Freud,A. and Burlingham, 1942 and 1944).

Social LearningTheory

Social Learning Theory which is associated with the work of Bandura, suggests that children learn behaviour from observing adults and copying them. According to Social Learning Theory attachments are formed by children taking on the behaviours of adults with whom they interact. Research studies have concluded that attachment bonds develop more quickly if parents and infants interact. Interlocution , “body talk”,and crying are powerful signals, and eye contact is generally considered a critical form of communication.

Mary Ainsworth

Mary Ainsworth who for a number of years worked closely with John Bowlby is a significant attachment theorist. Her concept of a‘fear of strangers’ is one she argues is developed in the first few months of an infant’s life. It is a process by which infants seem to discriminate between the preferred adult and others. For Ainsworth, the preferred adult is normally the person who interacts in accordance with the child’s liking and is usually the mother, though not necessarily so, and neither need it necessarily be the person who looks after the child. In her studies Ainsworth set up experimental situations which she called ‘strange situations’. These were procedures of 8 fixed actions which involved parents and a stranger entering a room where the infant was. She categorised 3 types of attachment based on observation of how the infant reacted to the situation. Firstly, in ‘secure attachment’ the infant is reasonably confident, well adjusted and self aware. Secondly, in ‘insecure attachment’ the child tends to cling to the parents and is reluctant to venture out on its own. Finally there is the ‘insecure unattached’ child who seems indifferent to whether the parent is there or not.

Factors which Ainsworth sees as contributing to which kind of attachment develops are firstly the temperament of the child, whether it is friendly and good natured, or bad tempered. Secondly the behaviour of the parents towards the child, whether they are sympathetic and respond to the child’s needs. According to Ainsworth the more parents accept the child on the child’s terms, the more securely attached the child is (Ainsworth and Bowlby, 1965).


Whichever theory of attachment is considered, certain common conclusions can be drawn.

Children need at least one firm, caring relationship from an early age; this relationship would be with the main caregiver but not always the preferred adult; this relationship needs to last throughout the first years of the child’s life.

Children need a secure environment in which they feel they have a clear identity and a role to play.

They need good role models whose behaviour they will imitate.

They need praise and warmth.

They need positive affirmation that they are loved and valued.

Caregivers need to be consistent in their attitudes towards, and their dealings with children and avoid displaying extreme mood swings or “blowing hot and cold”.

Caregivers should cater for the child’s wants and needs on a sensible basis.

The caring environment should be rich in stimuli and resources to encourage exploration, discovery and creativity.

Caregivers and children need to spend time in play and exploration.

Caregivers need to encourage play and exploration with the child, while keeping an eye on appropriate behaviour.

2007 : some further observations : early attachment experience and parenting

The way children and young people feel about their world is influenced by the nature of their attachment relationships. Children who have been dominated by their parenting figures may withdraw while those who have been over-protected may feel threatened by new experiences. Children who have been left to their own devices may be indifferent to experience and affectionless towards other human beings. Children who have been projected into the role of carers for their parentinig figures may miss out on necessary experiences in their childhood and adolescence. John Bowlby, supported in large measure by the research of Mary Ainsworth has demonstrated how children who have had unsatisfactory attachment relationships are affected by this for the rest of their lives (Ainsworth and Bowlby, 1965). Using the findings of more recent research,Peter Fonagy (2001) has been able to illustrate how our early attachment experiences influence our own capacity as parents to attach satisfactorily to our own children.The quality of attachment relationships made in their early childhood from secure to insecure largely defines the developmental pathways and the life experiences of all young people. The securely attached young person has both an internal and external secure base to which he can return when under threat. This in turn provides him with robust inner and outer worlds which he can use to navigate safely through the difficulties he may face as he grows into adulthood.

Attachment and the adolescent

Jeremy Holmes (2001) has established the connection between secure attachment in childhood and the development of the capacity in the adolescent to make sense of his life. The young person who throughout his childhood has experienced consistent parental care along with the stability which this brings, has the capacity to develop trusting relationships in which he is a healthily critical, yet cooperative partner. Such a youngster is able to make sense of his life history, and is fluent, coherent and insightful when describing his childhood attachments. The youngster who has not had the continuity of care which a secure attachment brings and so has not experienced the mutual stability afforded by a good attachment relationship has no ‘secure base’ to return to, and is encumbered by unresolved attachment issues. Such a young person has little or no inner or outer resources to help him deal with any threat in a socially helpful way. His disorganised life patterns brought on by severe deprivation or trauma creates a propensity to be resistant to support, and an inability to trust when help is offered.


Ainsworth, M. and Bowlby,J. (1965) Child Care and the growth of Love London : Penguin Books

Bion, W. (1962) Learning from Experience London : Routledge

Bowlby, J.(1988) A Secure Base : Clinical Applications of Attachment Theory London : Routledge

De Haan, M., Pascalis, O., & Johnson, M. (2002) ‘Specializations of Neural Mechanisms Underlying Face Recognition in Human Infants’ in Journal of Cognitive Neuroscience 14

Diamond,R. Carey S.(1977) ‘Developmental Changes in the representation of faces’ in The Journal of Experimental Psychology 23 pp1-22

Erikson, E.(1950) Childhood and Society Harmondsworth : Pelican 1965

Fonagy,P.(2001) Attachment Theory and Psychoanalysis New York : Other Press

Holmes, J. (2001) The Search for the Secure Base : Attachment Theory and Psychotherapyspan> Hove : Brunner-Routledge pp80-94

Freud, A. and Burlingham, D. (1942) War and Children New York : International Universities Press

Freud, A. and Burlingham, D. (1944) Infants without Families New York : International Universities Press

Freud, S. (1920) ‘Infantile Sexuality’ in On Metapsychology : The Penguin Freud Library Vol 7 London : Penguin Books 1977

Freud, S. (1923) ‘The Ego and the Id’  in On Sexuality : The Penguin Freud Library Vol 11 London : Penguin Books 1984 pp339-408

Harlow, H.(1964) ‘Early social deprivation and later behavior in the monkey’ in Unfinished Tasks in the Behavioral Sciences (A.Abrams, H.H. Gurner & J.E.P. Tomal, eds.) Baltimore : Williams & Wilkins pp154-173

Klein, M. (1952) ‘On Observing the Behaviour of Young Infants’ in Envy and Gratitude and Other Works 1946-1963 London : : Vintage 1988 pp94-121

Rutter, M. (1981) Maternal Deprivation Reassessed (Second edition) Harmondsworth : Penguin Books

Winnicott, D. (1965) The Maturational Processes and the Facilitating Environment London : Hogarth Press

© Charles Sharpe 1989, (with some further observations, 2007)
© and Charles Sharpe 2008

Editor’s note : an essay by John Fallowfield “Attachment theory and social work with ‘looked after’ children and their families” which deals with many of the issues raised in this article can be found at