Of all the theories that grew out of psychoanalysis, attachment theory has most shaped how love is understood today. Its central idea is that the bond between a baby and its caregiver is not a side effect of feeding or of sexual drives. It is a biologically rooted system whose purpose is protection. Later researchers argued that the system keeps working in adulthood, and that much of romantic love, longing, jealousy and grief is its adult form.
This article is part of the Psychoanalysis of Love series. The basic concepts are introduced in the site's articles on attachment theory, adult attachment and the Strange Situation. The focus here is on how John Bowlby's work and its successors came to describe love itself as an attachment process.
A psychoanalyst who disagreed with psychoanalysis
John Bowlby (1907–1990) trained at the British Psychoanalytical Society in the 1930s. He was analysed by Joan Riviere and supervised by Melanie Klein. Early on, however, he questioned a central assumption of the Kleinian circle described in the article on Melanie Klein: that a child's suffering is best understood through unconscious phantasy, the child's inner imaginings.
Working in child guidance, he became convinced that real events, above all long separation from the mother or her loss, had a decisive and under-recognised effect on development. He later recalled being discouraged, in one supervised case, from taking the mother's circumstances into account, an episode often cited as a symbol of the disagreement.
His early study Forty-Four Juvenile Thieves (1944) linked long early separations to what he called an "affectionless" character in adolescence. His WHO report Maternal Care and Mental Health (1951) brought maternal deprivation to world attention.
With James Robertson, whose films showed young children separated from their parents in hospital, Bowlby described a typical sequence of reactions to separation: protest, despair and detachment. In detachment, the child seems indifferent to the returning parent, a defence Bowlby believed had lasting consequences.
To explain these observations, Bowlby turned to ethology (the study of animal behaviour), control systems theory and evolutionary biology rather than Freud's drive theory. His paper "The nature of the child's tie to his mother" (1958) and his talks on infant grief around 1960 were strongly criticised by analysts, including Anna Freud, who felt he had replaced the unconscious with observable behaviour. For decades attachment theory developed outside mainstream psychoanalysis, although Bowlby saw it as a revision of psychoanalysis, not a rejection.
Safe haven and secure base
Bowlby set out his mature theory in the three volumes of Attachment and Loss: Attachment (1969), Separation: Anxiety and Anger (1973) and Loss: Sadness and Depression (1980).
At its core is the attachment behavioural system: an inborn system, shaped by evolution, that keeps a vulnerable person close to someone who can protect them. Behaviours such as crying, clinging and following are switched on by threat or separation, and calm down when closeness and comfort return.
Two roles of the attachment figure became central to later accounts of love:
- Safe haven: the person we turn to for comfort and protection when we are upset.
- Secure base: the person whose reliable presence makes it possible to explore the world, take risks and become independent.
Knowing a partner will be there in the evening can make a difficult day easier to face alone. For Bowlby, confidence in an attachment figure's availability is what makes independence possible. This is close to Winnicott's view (see Winnicott) that the capacity to be alone develops in the presence of a reliable other.
Anger and grief as ways of holding on
The second and third volumes applied the theory to anger and mourning. Bowlby saw the angry protest of a separated child, and of an adult whose partner has withdrawn, as having a purpose: to restore the bond and discourage future separations.
In Loss (1980), he described adult mourning as a long attachment response, in which searching and yearning for the lost person continue long after the loss is known (see also basic concepts of grief). Someone may still reach for the phone to call a former partner months after a breakup.
Bowlby insisted that attachment behaviour shapes human life "from the cradle to the grave" (The Making and Breaking of Affectional Bonds, 1979), which opened the way to applying the theory to adult love.
Internal working models: what we expect from love
Bowlby's main link with the psychoanalytic interest in inner life was the internal working model: a set of inner expectations about relationships. Through repeated experience, a child learns whether the people they depend on can be relied upon, and whether they themselves deserve care.
These models work mostly outside awareness and tend to confirm themselves. A person who expects rejection may become guarded or quick to take offence, and so make rejection more likely.
Unlike Klein's internal objects (the inner images of others we carry within us), working models are thought to reflect real interactions fairly accurately; like them, they are unconscious, emotionally charged and hard to change. A person may also hold contradictory models of the same figure, such as a conscious, idealised image of a parent alongside an unconscious expectation of rejection, an idea close to Fairbairn's account of split internal objects in the article on Fairbairn.
Patterns of attachment in children
After observational studies in Uganda and Baltimore, Mary Ainsworth gave the theory a research method. Her Strange Situation procedure (Ainsworth et al., 1978) observed how infants respond to a brief separation from their caregiver and to the reunion. She identified three organised patterns, secure, insecure-avoidant and insecure-ambivalent (resistant), and linked them to how sensitive the caregiving was at home.
Mary Main and Judith Solomon (1990) later described a fourth category, disorganised/disoriented. These infants show contradictory or frightened behaviour towards the caregiver, as if the source of safety were also a source of alarm, a pattern since associated with unresolved loss or trauma in caregivers.
For understanding love, the strategies behind the labels matter most. The avoidant infant plays down its need to keep a rejecting caregiver close; the ambivalent infant turns up its distress to hold the attention of an inconsistent one. Both adapt to a particular relationship, and both reappear, in changed form, in adult couples.
Romantic love as attachment
In "Romantic love conceptualized as an attachment process" (1987), Cindy Hazan and Phillip Shaver used a newspaper questionnaire and a student sample to ask adults which of three descriptions best matched their feelings in close relationships:
- Secure: relationships marked by happiness, friendship and trust.
- Anxious-ambivalent: obsessive preoccupation, emotional highs and lows, jealousy and a strong wish to merge with the partner.
- Avoidant: fear of intimacy and emotional distance.
Memories of childhood relationships with parents were related to adult style, though the authors avoided claiming cause and effect.
Its lasting contribution was the insight that many features of infant attachment reappear in adult romance: the wish to be close, distress at separation, seeking comfort from the partner, and looking into each other's eyes. Romantic love thus shows clearly how early patterns continue into adult life.
Attachment, care and sex
Shaver, Hazan and Bradshaw (1988) proposed that adult romantic love brings together three systems:
- attachment: seeking comfort and closeness
- caregiving: offering comfort and protection
- sexuality: desire for the partner
Unlike the one-way infant bond, in adult couples each partner is ideally both attachment figure and caregiver for the other, and desire is added. Many difficulties in love are imbalances between the three: one partner only giving care, sex used mainly to calm anxiety about the bond, or passion that never becomes a secure base. The view is close to Kernberg's account of uniting tenderness and erotic desire (see Otto Kernberg).
From four styles to two dimensions
Kim Bartholomew and Leonard Horowitz (1991) built a four-category model from Bowlby's two working models, of the self and of others.
- Secure people see both themselves and others positively.
- Preoccupied people see others positively but doubt their own worth, and so seek reassurance.
- Dismissing people protect a positive self-image by playing down the importance of closeness.
- Fearful people see both negatively, wishing for intimacy but expecting to be hurt; this category is often linked with trauma.
Later research, notably by Brennan, Clark and Shaver (1998), found that adult attachment is better measured on two continuous scales than in separate types: attachment anxiety (worry about being abandoned or unloved) and attachment avoidance (discomfort with closeness and dependence). Mikulincer and Shaver (2007) called the matching strategies hyperactivating (urgent bids for closeness) and deactivating (pushing need aside and stressing self-reliance).
When an anxious and an avoidant partner meet
A well-documented pattern pairs a partner high in attachment anxiety with one high in avoidance. Each strategy sets off the other: one partner pursues, the other feels crowded and withdraws, which confirms the fear of abandonment and intensifies the pursuit.
One partner may send several messages when a reply is late, while the other, feeling pressured, replies even more slowly. From an attachment point of view, both are managing the same fear of losing the bond, by opposite means. In Bowlby's terms, the anger is protest: arguments about small matters are often really about whether the other person can be reached.
Earned security: change is possible
The Adult Attachment Interview, developed by Mary Main and colleagues, moved the study of attachment "to the level of representation" (Main, Kaplan & Cassidy, 1985). It assesses not what happened in childhood, but how coherently a person can talk about it.
Some adults who describe harsh early experiences still give coherent, balanced accounts and are classified as secure, which has been called earned security. This suggests that working models can change through later relationships, including a reliable partner or a therapeutic relationship, and that what predicts security in the next generation is less the history itself than the ability to reflect on it.
Bringing attachment back to psychoanalysis
Peter Fonagy has done more than anyone to reconnect attachment theory with psychoanalysis. In a study following families over time, Fonagy, Steele and Steele (1991) found that mothers' Adult Attachment Interviews during pregnancy predicted their infants' Strange Situation patterns at one year.
Fonagy and colleagues proposed that the key linking capacity is mentalisation, or reflective function (Fonagy et al., 2002): the ability to understand one's own and other people's behaviour in terms of what is going on inside: feelings, wishes and intentions.
Fonagy's Attachment Theory and Psychoanalysis (2001) explored where the two traditions meet and differ. Applied to love, mentalisation explains why secure relationships survive conflict: partners who can think about each other's feelings can bear hurt without concluding the bond is destroyed. A silent partner may be tired or worried rather than rejecting. When mentalising breaks down under stress, partners react to behaviour rather than meaning and fall back on older working models.
Criticisms and limitations
Attachment theory has been criticised in several ways:
- From psychoanalysis. Critics argued that Bowlby neglected unconscious phantasy, sexuality and inner conflict, and that attachment research says little about desire, the erotic, or the mixed feelings Freud and Klein placed at the centre of love. The article on Lacan presents a view of love grounded in lack rather than security.
- The mother as the only figure. Bowlby's early emphasis on the mother as a single primary figure was challenged by Michael Rutter's Maternal Deprivation Reassessed (1972) and by feminist critics; later research gives more weight to multiple attachment figures.
- Temperament. Some researchers argue that temperament and genetic factors affect infants' Strange Situation behaviour more than early research assumed.
- Culture. Attachment patterns vary across cultures, and some authors, for example Rothbaum and colleagues (2000), have questioned whether the definition of security reflects Western values of independence.
- Continuity. Long-term studies show only moderate stability from infancy to adulthood, and romantic attachment questionnaires match the Adult Attachment Interview only modestly. Popular accounts of fixed "styles" go beyond the evidence.
Why attachment ideas matter in therapy
In A Secure Base (1988), Bowlby described the therapist as offering a reliable base from which painful working models can be explored and revised. Attachment ideas help people recognise patterns such as:
- pursuing a partner harder whenever they seem distant, or shutting down whenever a partner wants more closeness
- intense jealousy or constant need for reassurance
- doubting one's own worth in relationships, or avoiding dependence altogether
- grief after a breakup that includes protest, despair and searching for the lost partner
These can be understood as ways of managing the fear of losing the bond, or as expected phases of grief, rather than as character flaws or weakness. The therapeutic relationship itself can contribute to earned security.
The best-known application is Emotionally Focused Therapy (EFT), developed by Leslie Greenberg and Susan Johnson in the 1980s and elaborated by Johnson as a couple therapy based explicitly on attachment theory. EFT helps partners identify their negative cycle, such as pursue–withdraw, and treats the cycle, not either partner, as the problem. It then reaches the attachment fears beneath anger or distance, and helps partners have new experiences of reaching for and responding to each other. (See also couple therapy.)
Conclusion
Bowlby, a psychoanalyst who believed real separation and loss had been underestimated, founded a tradition that changed how love is studied. Attachment theory sees romantic love as an adult version of the infant's search for a safe haven and a secure base, shaped by early expectations that can still change. It is less concerned with desire than with safety and reachability, which is both its strength and its limitation. Read alongside Freud, Klein and Winnicott, it adds a developmental and research-based grounding the earlier theories lacked. Where fears of abandonment or difficulties with closeness repeatedly shape relationships, psychodynamic psychotherapy can help to understand and revise these patterns; online sessions are available to book here.
References
- Ainsworth, M. D. S., Blehar, M. C., Waters, E., & Wall, S. (1978). Patterns of attachment: A psychological study of the Strange Situation. Lawrence Erlbaum.
- Bartholomew, K., & Horowitz, L. M. (1991). Attachment styles among young adults: A test of a four-category model. Journal of Personality and Social Psychology, 61(2), 226–244.
- Bowlby, J. (1944). Forty-four juvenile thieves: Their characters and home-life. International Journal of Psycho-Analysis, 25.
- Bowlby, J. (1951). Maternal care and mental health. World Health Organization.
- Bowlby, J. (1958). The nature of the child's tie to his mother. International Journal of Psycho-Analysis, 39, 350–373.
- Bowlby, J. (1969). Attachment and loss: Vol. 1. Attachment. Hogarth Press.
- Bowlby, J. (1973). Attachment and loss: Vol. 2. Separation: Anxiety and anger. Hogarth Press.
- Bowlby, J. (1979). The making and breaking of affectional bonds. Tavistock.
- Bowlby, J. (1980). Attachment and loss: Vol. 3. Loss: Sadness and depression. Hogarth Press.
- Bowlby, J. (1988). A secure base: Parent-child attachment and healthy human development. Basic Books.
- Brennan, K. A., Clark, C. L., & Shaver, P. R. (1998). Self-report measurement of adult attachment: An integrative overview. In J. A. Simpson & W. S. Rholes (Eds.), Attachment theory and close relationships. Guilford Press.
- Fonagy, P. (2001). Attachment theory and psychoanalysis. Other Press.
- Fonagy, P., Gergely, G., Jurist, E. L., & Target, M. (2002). Affect regulation, mentalization, and the development of the self. Other Press.
- Fonagy, P., Steele, H., & Steele, M. (1991). Maternal representations of attachment during pregnancy predict the organization of infant-mother attachment at one year of age. Child Development, 62(5), 891–905.
- Greenberg, L. S., & Johnson, S. M. (1988). Emotionally focused therapy for couples. Guilford Press.
- Hazan, C., & Shaver, P. (1987). Romantic love conceptualized as an attachment process. Journal of Personality and Social Psychology, 52(3), 511–524. doi:10.1037/0022-3514.52.3.511
- Holmes, J. (1993). John Bowlby and attachment theory. Routledge.
- Johnson, S. M. (2004). The practice of emotionally focused couple therapy: Creating connection (2nd ed.). Brunner-Routledge.
- Main, M., Kaplan, N., & Cassidy, J. (1985). Security in infancy, childhood, and adulthood: A move to the level of representation. Monographs of the Society for Research in Child Development, 50(1–2), 66–104.
- Main, M., & Solomon, J. (1990). Procedures for identifying infants as disorganized/disoriented during the Ainsworth Strange Situation. In M. T. Greenberg, D. Cicchetti, & E. M. Cummings (Eds.), Attachment in the preschool years: Theory, research, and intervention. University of Chicago Press.
- Mikulincer, M., & Shaver, P. R. (2007). Attachment in adulthood: Structure, dynamics, and change. Guilford Press.
- Rothbaum, F., Weisz, J., Pott, M., Miyake, K., & Morelli, G. (2000). Attachment and culture: Security in the United States and Japan. American Psychologist, 55(10), 1093–1104.
- Rutter, M. (1972). Maternal deprivation reassessed. Penguin.
- Shaver, P. R., Hazan, C., & Bradshaw, D. (1988). Love as attachment: The integration of three behavioral systems. In R. J. Sternberg & M. L. Barnes (Eds.), The psychology of love. Yale University Press.