NUMBER 1035 • 30 AUGUST • insecure attachment

     The greatest difference between the group of adolescents in residential care and the other groups of adolescents studied until now is the high proportion in the CC* attachment category, where contradictory patterns in the AAI are characteristic. In non-clinical samples this category is distinctly unusual. It has been found to such a high degree only among adults in connection with psychiatric conditions, violence and criminality (Hesse, 1999). The significance of this attachment category at an adolescent age has not yet been explained. For the adolescents studied here it may well be assumed that the precarious life experiences they had in their original families, marked by neglect, maltreatment and sexual abuse, sharp conflicts between their attachment figures, and the frequent change of such figures, did not permit integration along the lines of a consistent attachment strategy. As a consequence it was not possible to construct a coherent attachment organization. Multiple attachment disruptions which may be seen as a marker or summary variable for cumulative risk factors (Kobak et al., 2001) were the normal state of things in their life. For that reason such traumatic experiences have quite often caused a global breakdown in attachment strategies.

Perhaps, however, a more optimistic view is also possible, according to which the high prevalence of CC attachment patterns can be explained by the fact that these adolescents have not yet succeeded in integrating new and better attachment experiences into their internal working models. In that case the CC classification would rather be a sign of hope. It suggests itself, however, that the specific residential context with its large number of relationships associated with shift work - which relationships are therefore almost never exclusive - makes the task of such a development even harder. The present findings provide, finally, no indications that could suggest that the youth welfare measure of residential care could have improved the degree of organization in the youth’s attachment representation, in the sense of a corrective attachment experience (e.g. Schleiffer, 2003). The statistically demonstrable connection between the term of the stay in the residential home and an especially insecure attachment representation does not suggest that such an intervention is efficient.

Even if the frightening similarity between the attachment distribution of the adolescents and that of criminals with psychiatric problems may by no means lead to the conclusion that the adolescents now in residential care will probably become criminals with psychiatric problems themselves, this result must nevertheless suggest that further research is necessary. According to the probabilistic developmental psychopathological approach the factors have to be studied that determine who becomes psychically disturbed and/or criminal.

For the time being one may only be sceptical about the future development of the adolescents, especially considering that the Adult Attachment Interview definitely has a predictive validity for psychopathological conditions (Van IJzendoorn, 1995). This can he no reason for surprise since we know of the relationships between the current attachment representation of adolescents and a series of adjustment variables in this age-group, for example ways of coping with stress, the development of self-esteem or the form of relationships not only to parents but especially to peers - people of the same age – and therefore to future partners (Allen & Land, 1999). Consequently an insecure, and especially a highly insecure attachment representation even in adolescence must be viewed as a by no means negligible potential risk factor for the person’s future life.

[* NOTE: Thirty-three of the adolescents studied could not be classified as secure, dismissing or preoccupied because in their interviews there was a mixture of contradictory mental states such that no predominant strategy was to be identified. In 11 of these 33 cannot classify interviews a best-fitting dismissing subcategory was assigned, whereas in 22 cases two or three dismissing and preoccupied sub-categories, which normally will be contradictory, seemed applicable. High scores for idealization and for insistence upon lack of memory correspond with a very low score for coherence of mind.]



Schleiffer, R., & Müller, S. (2004) Attachment representations of adolescents in institutional care, International Journal of  Child & Family Welfare, Vol. 7(1), p. 70



































Allen, J.P.,  & Land, D. (1999). Attachment in adolescence. In J. Cassidy, & P.R. Shaver (Eds.), Handbook of attachment, (pp. 319-335). New York: Guilford Press

Hesse, E. (1999). The adult attachment interview: historical and current perspectives. In J. Cassidy, & P.R. Shaver (Eds.), Handbook of attachment, (pp. 395-433). New York: Guilford Press

Schleiffer, R. (2003). Bindungsbeziehungen im Heim (2nd edit.). Münster: Votum

Van IJzendoorn, M.H. (1995). Adult attachment representations, parental responsiveness and infant attachment: A meta-analysis on the predictive validity of the Adult Attachment Interview. Psychological Bulletin, 117, 387-403