One of the theories criticized by Bard, Keller, and Leavens for being WEIRD, i.e., for relying mostly on research about people from Western, Educated, Industrialized, Rich, and Democratic countries (Henrich, Heine, & Norenzayan, Reference Henrich, Heine and Norenzayan2010), is John Bowlby’s and Mary Ainsworth’s Attachment Theory (AT). Building upon an extensive body of work showing the diversity of childcare practices in different cultures, the authors challenge the universal and normative character of the attachment system that Bowlby postulated. Despite agreeing with their goals and proposals, I fear that their thorough demonstration of the WEIRD character of AT will not convince its supporters to give it up.

In my view, discussions about AT are at an impasse that cannot be resolved by providing more empirical evidence about cultural diversity in children’s social relations. Many attachment theorists have recognized the contextual dimensions affecting an infant’s attachment to his or her caregivers. Nevertheless, they continue to defend the cross-cultural validity of AT (van IJzendoorn & Sagi-Schwartz, Reference Van IJzendoorn, Sagi-Schwartz, Cassidy and Shaver2008; Cassidy & Shaver, Reference Cassidy and Shaver2018; Duschinsky, Reference Duschinsky2020). Following the founders of AT, they believe that, underlying all the cultural diversity, there is a common biological system with prescriptive implications.

For Bowlby and Ainsworth, the universality and normativity of the attachment system derived from its biological basis. They presented the attachment between infant and mother as an adaptation forged in what Bowlby called “man’s environment of evolutionary adaptedness” (EEA), defined as the ancestral environment human beings evolved in and are adapted to. Bowlby saw that system as universal and normative, claiming that “the more the social environment in which a human child is reared deviates from the environment of evolutionary adaptedness … the greater will be the risk of his developing maladaptive patterns of social behaviour” (Bowlby, Reference Bowlby1982, 166). Ainsworth and colleagues agreed: “To the extent that the environment of rearing departs from the environment to which his [an infant’s] behaviors are adapted, developmental anomalies may occur” (Ainsworth et al. Reference Ainsworth, Blehar, Waters and Wall1978, 9).

Though most attachment researchers today still uphold the view of secure attachment as a biological adaptation, several scholars since the 1980s have pointed out serious issues in their use of biological concepts (Hinde, Reference Hinde, Murray Parkes and Stevenson-Hinde1982; Hinde & Stevenson-Hinde, Reference Hinde and Stevenson-Hinde1990). Michael Lamb and his colleagues noted that the attachment literature often conflates the notion of psychological adaptation with the biological notion of adaptation (Lamb et al., Reference Lamb, Thompson, Gardner, Charnov and Estes1984). Vicedo (Reference Vicedo, Keller and Bard2017a) argued that recent attempts to provide evolutionary explanations of attachment remain as speculative as Bowlby’s original proposal.

Bowlby’s idea that the secure attachment system developed as an adaptation in the EEA was at best a hypothesis, and one that has not been substantiated to date. It is very difficult to figure out the fitness values of behaviors from a couple of million years ago. AT says that babies who cried increased their chances of survival by getting their mothers’ attention. But crying babies also alerted predators to their location. Maybe being quiet rather than crying increased an infant’s fitness in the EEA. The same goes for sensitive mothering. Focusing too much on your baby’s signals may distract you from paying attention to sources of danger. So, maybe being an insensitive mother rather than a sensitive one increased a woman’s fitness in the EEA. Population biologist Richard Lewontin and paleontologist Stephen J. Gould famously showed how easy it is to construct “just so stories” to “explain” how things may have happened during evolution (Gould & Lewontin, Reference Gould and Lewontin1979). But it is not easy to prove that an organism’s trait is an adaptation, and it is especially difficult in the case of behavior (Lewontin, Reference Lewontin1978).

The view of the secure attachment system as adaptive in the Pleistocene is an instance of what American paleoanthropologists Marlene Zuk and Leslie Aiello have called “paleofantasies.” For Zuk, a paleofantasy assumes that “humans, or at least our protohuman forebears, were at some point perfectly adapted to our environments” (Zuk, Reference Zuk2013). But the idea that evolution leads to an optimal fit of an organism with its surroundings is incorrect (Lewontin, Reference Lewontin1983; Zuk, Reference Zuk2013). In addition, it underestimates the extent to which organisms and environments are in constant flux. Human evolution did not stop in the Pleistocene. Thus, even if a secure attachment system was adaptive way back then, it does not follow that it is adaptive today, either biologically or psychologically.

As I see it, the entrenched commitment to the static and speculative evolutionary picture postulated by Bowlby prevents attachment researchers from fully understanding the import of the diversity encountered in human communities. For example, Mesman et al. (Reference Mesman, Minter, Angnged, Cissé, Salali and Migliano2018) maintain that sensitive responsiveness as defined by Ainsworth is universal, though it can be expressed in varied ways in different cultures. Duschinsky (Reference Duschinsky2020, 104), summarizing the current position in the attachment literature, says: “In general it is agreed that experience, circumstance, and culture can all shape the conditions under which the attachment system is activated, the forms of behaviour recruited by the attachment system, and how these are expressed.” Thus, in this view, the environment (culture, ecological context, experience) is only “shaping,” “activating,” or modulating the expression of a system that has already been set.

However, students of evolution and development have shown that organisms are not mechanisms with fixed designs, but agents that create and change their environments (Jablonka & Lamb, Reference Jablonka and Lamb2005; Walsh, Reference Walsh2015; Sultan, Reference Sultan2015). Furthermore, ontogeny is not a simple process of “expressing” an innate design “activated” by the environment. In development, biological, experiential, and environmental factors all interact in complex ways to give rise to different phenotypes in different ecological contexts. Bowlby was wrong in claiming that evolution leads to “environmentally stable behavioural systems” that are necessary for a species’ survival (Bowlby, Reference Bowlby1982).

As Bard, Keller, and Leavens propose, it is time to recognize that neither our current understanding of evolution nor our knowledge about child development support AT. Only then will we be able to appreciate the true extent and value of human diversity.