Heidi Keller
Heidi Keller is a cultural psychologist at the University of Osnabrück and the most serious critic attachment theory currently has. Her argument is that the theory encodes a Western middle-class model of childhood, roughly five per cent of the world's population, and then claims universality for it.
Who they are
Keller's 2018 paper in PNAS sets out four assumptions she says the field treats as universal: that all children form attachments, that security is the normative and desirable outcome, that sensitivity produces it, and that security produces later competence. Her challenges are empirical before they are philosophical.
The Strange Situation assumes a child who is wary of strangers. Among the Beng of Côte d'Ivoire and the Nso of Cameroon, being picked up by an unfamiliar adult is an ordinary event, and infant stress hormones fall rather than rise as it happens. The procedure also assumes a single principal adult who can be removed from the room. In parts of South Madagascar an infant's main social partners are other children under five.
Her hardest point is about sensitivity itself. The construct presupposes a child who leads and an adult who follows, which inverts what counts as good parenting in subsistence farming communities where a child is taught to fit into an existing order.
The second half of her paper is ethical rather than empirical. She opens on a documented Norwegian child-welfare removal, and argues that exporting the framework into child-protection decisions and parenting programmes risks reading culturally competent parenting as neglect.
The other side of the argument is held by Judi Mesman, Marinus van IJzendoorn and Abraham Sagi-Schwartz, whose evidence from the Dogon of Mali and Efe foragers finds security still the most common pattern and sensitivity still predictive. Both camps are right about their own data, and the dispute is really about the instrument rather than about whether babies bond.
Key papers
Key papers
Keller, H. (2018). Proceedings of the National Academy of Sciences, 115(45), 11414-11419. Sole author. The paper that sets out the four assumptions she says attachment research treats as universal, and tests them against fieldwork: stranger wariness among the Beng of Côte d'Ivoire and the Nso of Cameroon, where infant stress hormones fall rather than rise when an unfamiliar adult picks a child up, and the single removable caregiver against parts of South Madagascar where an infant's main social partners are other children under five. The second half argues the ethical case about child-protection decisions.
Mesman, J., van IJzendoorn, M. H., & Sagi-Schwartz, A. (2016). In Handbook of Attachment, 3rd edition. The reply, and it should be read alongside rather than after. Drawing on evidence from the Dogon of Mali and from Efe foragers, these authors report security as still the most common pattern and sensitivity as still predictive.
van IJzendoorn, M. H., & Kroonenberg, P. M. (1988). Child Development, 59(1), 147-156. The cross-national pooling of thirty-two Strange Situation samples that both sides of this argument use, and which found variation between two samples from the same country larger than variation between countries.
Key ideas
The instrument carries assumptions. Stranger wariness and a single removable adult are design requirements, not human universals.
Sensitivity is a culturally specific ideal, not a neutral description of good care.
The applied stakes are real. This is not only an academic dispute when the framework informs who keeps their children.
Concepts on this site
Sources
- Keller, H. (2018). PNAS, 115(45), 11414-11419. DOI 10.1073/pnas.1720325115.
- Mesman, J., van IJzendoorn, M. H., & Sagi-Schwartz, A. (2016). In Handbook of Attachment, 3rd edition.
- van IJzendoorn, M. H., & Kroonenberg, P. M. (1988). Child Development, 59(1), 147-156.