Referencing John Bowlby: Not Your Grandpa's Attachment Theory Any More

Thousands and thousands of articles on attachment theory and research have been published over the last 40 years or so. This topic, with its emphasis on early experiences and social-emotional outcomes, has fascinated many readers and received faddish attention and emphasis both within the field of psychology and outside it. Unfortunately, not everyone who writes about attachment theory is on the same page; there is a mainstream view of attachment, developed through decades of discussion and research analysis, and there is an alternative attachment theory that has given rise to the use of holding therapy and similar non-evidence-based treatments.

How do you tell the difference between these when reading Internet articles? One way is to notice that the alternative attachment theorists quote John Bowlby, the “father of attachment theory”, to the exclusion of any more modern work. This practice is problematic, because some strongly-held tenets of Bowlby’s theory are much at odds with modern empirically-based work.

I am going to comment on some of Bowlby’s ideas that no longer appear credible. Please understand that this critique is not an attack on Bowlby himself or on attachment theory in general. John Bowlby was a man of his time, and his formulation of attachment theory was influenced by other ideas current during his professional lifetime. In addition, his thinking changed and developed over a long professional life, with relevant experiences in both wartime and peace. These statements about changing ideas are true of any influential theorist, in psychology or any other discipline—for example, asking what Freud thought about something really makes sense only if we say whether we are asking about his thoughts in 1900 or in 1935.  

References of alternative attachment theorists to John Bowlby’s work often take their information from Bowlby’s report to the World Health Organization in 1951 on maternal care and mental health (later published by Schocken in 1966 in a volume including responses of both supporters and critics to the original report). Here are some ideas rom that volume that no longer seem credible to mainstream attachment theorists and researchers:

1.    1.   Genetic factors play little or no role in the development of mental illness; early experiences, especially separation from the mother or deprivation of maternal care, are the primary causes of later mental illness and/or criminality. [This stress on experiential rather than biological factors was characteristic of Bowlby’s time and can be seen in the work of researchers working on juvenile delinquency like Glueck and Glueck. It is an optimistic view of mental and behavioral problems, because it implies that new, appropriate experiences may be able to correct the effects of early, bad experiences.]

2.     2.  Infants form attachments to a single person rather than to several caregivers and have a great deal of difficulty in forming new attachments if separated. Because of this tendency, called monotropy, the great preponderance of infant attachments are to the mother. [ During the Victorian period and later, emphasis on the powerful importance of mothers was generally assumed to be correct. By the 1940s, mothers were both praised for their benign influences and blamed  for their potential harmfulness—for example, the attribution of autism to “refrigerator mothers”.]

3.     3.  Attachment and attachment behavior are characteristic of the human species and are parallel to the rapid learning to follow other ducks shown by ducklings (called imprinting). When imprinting “goes wrong” (for instance, when a duckling imprints on a human being rather than another duck, and in adulthood tries to court the human rather than mating with another duck), it is very difficult to correct the problem, so attachment difficulties are also likely to cause serious adult problems that are hard to treat. [This ethological approach to human behavior was an outgrowth of comparative studies of  animal behavior that were a feature of psychological work from the 1930s to about 1980. We see little of this time-consuming observational research today and most people do not think of it as part of psychology, but it was of serious interest in the past, when undergraduate psychology curricula normally included a “comparative psych” course.]

4.      4. When children have been separated from their mothers or have otherwise been deprived of maternal care, encouraging regression by bottle-feeding and other “babying” can be an effective way to treat older children who have behavior problems. [ Bowlby included this idea in his 1951 report because such treatment was being used in Sweden, for example, in the care of children who had suffered severe deprivation and separation during World War II. The effect of the war, the number of orphans to be cared for, the evacuation of British children from cities that were being bombed, and so on, cannot be underestimated as a factor in Bowlby’s thinking.]

5.      5. Separation from their mothers causes psychological trauma in young children. [Bowlby based this view on observations of separated, evacuated British children, European Jewish children sent to England by their parents as Hitler came to power, and children being treated in hospitals where their parents might not be allowed to visit them for weeks.]

Much further thinking about attachment, and empirical research on the topic, have allowed us to reject these early assumptions of Bowlby’s while keeping some important tenets of attachment theory. (A paper by Michael Rutter in 1995 stated this very clearly and can be considered a beginning for a more advanced form of attachment theory, in which I do not include Schore’s so-called “modern attachment theory”.) Here are some reasons for rejecting the tenets listed above:

1.    1.  Genetic factors have been clearly shown to play important roles in the development of both mental illness (such as autism) and criminal behavior. Experiences are also important, but individuals with different genetic make-ups may respond to experience differently.

2.     2.  The research evidence indicates that infants commonly have attachments to several different caregivers, none of whom need be the mother.

3.     3.  Although human beings show many characteristic behaviors related to attachment, both in terms of seeking and of giving social interaction with others, it is deceptive to think of these as parallel with imprinting. From an evolutionary point of view, it would seem maladaptive for human beings, with their long childhoods and the relatively high chance of death of a parent (especially a mother) during that time, to have an unbreakable emotional connection with any individual and to have development badly distorted by separation from a single person.

4.    4.   Treatment of childhood mental illness by regression methods has never been shown to be effective. Parenting methods that demand a reasonable level of maturity are advantageous, so it would be surprising if the opposite treatment for mental illness worked.

5.     5.  Older infants (more than about 6 months old) and young children show great immediate distress in response to separation and continuing distress over weeks or months in cases where they receive little comforting attention and care. However, as John Bowlby’s colleague John Robertson showed in an observational study, plenty of responsive care from a small number of concerned adults reduces this distress a great deal. It would appear that it is the withdrawal of sufficient concerned care from adults that is potentially traumatic, not the separation itself. (It should be noted that the youngest infants, before about 6 months, do not appear to show distress about separation, and this is certainly true of newborns, despite some alternative theorists' claims to the contrary.)

Once again, this post is not an attack on John Bowlby or on the essentials of attachment theory. It is simply a statement that some of Bowlby’s ideas have not stood up to continuing thought and research. When advocates of alternative attachment theories and treatments for children cite  some of Bowlby’s original ideas as received knowledge or “scripture”, readers should see this practice as evidence that these people do not understand evidence-based concepts of attachment, and their other pronouncements should be approached warily.

Incidentally, material that invokes the name of Sir Richard Bowlby should be regarded with suspicion. Richard Bowlby, John Bowlby’s oldest son, has been a medical photographer. He inherited his title from his baronet uncle (John Bowlby was a younger son and not in line for the title.) There was a period of time when Sir Richard was invited to speak at conferences of alternative attachment theory proponents, presumably because of the magic Bowlby name and the magic baronetcy.  However, he did not know very much about attachment. He still has a website, though.