In a large twin study Minnis et al (Reference Minnis, Reekie and Young2007) have demonstrated that attachment disorder behaviours can be differentiated from other common childhood emotional and behavioural disorders and appear to be strongly genetically influenced, particularly in boys. The authors also point out that, even in a population of children that was probably healthier than the general population, behaviours suggestive of attachment disorder were identified. Conventional aetiological factors are addressed but the paper would have benefited from the inclusion of an evolutionary perspective. Evolutionary or Darwinian psychiatry examines, among other things, the potential for adaptive benefits to pre-programmed psychobiological mechanisms (e.g. depressive symptoms or attachment disorders) that are sometimes incorrectly viewed as being simply abnormal or pathological (Reference AbedAbed, 2000).

It was surprising that Minnis et al made no reference to Bowlby's seminal work (Reference BowlbyBowlby, 1958) in the area of attachment. Bowlby's perspective on attachment was an evolutionary one, in that he viewed the associated behaviours as representing evolved and adaptive psychobiological mechanisms, protecting the child from predators and the many other dangers prevalent in our ancestral environment. This ‘adaptionist’ perspective could have been explored by Minnis et al when considering why attachment disorder behaviours occurred at all in this healthy non-clinical sample.

Chisholm (Reference Chisholm1996) and Belsky (Reference Belsky1997) proposed in more recent years an integration of life history theory (Reference LevinsLevins, 1968) and attachment theory. Chisholm (Reference Chisholm1996) argued that, in life history theory, life cycles constitute evolved adaptive strategies. Furthermore, individuals must prioritise the allocation of their time and resources to different components of reproductive fitness (e.g. growth, mating or parenting). Therefore, the sexual strategy employed by parents (e.g. low investment in large numbers of offspring or vice versa) is an integral component of the child's early environment. Belsky (Reference Belsky1997) argued that secure attachment in children functioned to promote a strategy of high-investment parenting, and avoidant attachment (child showing indifference to parent) as representing an adaptation to parental unwillingness to invest (e.g. when the parent invests instead in a short-term mating strategy with relatively little investment in individual offspring).

The anxious/ambivalent style of attachment evolved in response to parental inability (e.g. through illness) to invest, and fostered a ‘helpers at the nest style’ in the children, whereby children would cooperate in rearing siblings. For example, Turke (Reference Turke, Betzig, Borgerhoff-Mulder and WTurke1988) demonstrated (independent of attachment disorders) that women from the Micronesian atoll of Ifaluk were likely to have significantly larger families when their first-born was female: an anxious/ambivalent attachment style may further accentuate such behaviour in female children, perhaps explaining in part the gender differences in attachment disorders raised by Minnis et al.

These are merely a few examples of the insights that evolutionary psychiatry can provide. In the total absence of such an evolutionary perspective, one is reminded of Abed's (Reference Abed2000) cautionary comments: ‘In recent years psychiatry has attempted to circumvent such problems by engaging in an atheoretical research enterprise involving gathering masses of data and calculating sophisticated statistical associations. However, such an endeavour of itself cannot generate a scientific discipline, for science is a method of discovering the world and not simply a body of facts’.