I was so delighted to see that a comment of substance was posted on this blog yesterday. The contributor stated that, contrary to what I had said in a recent post, attachment did not follow a stage pattern of development. This was of great interest to me, especially in the context of the differences between conventional and alternative attachment theories which I had addressed last week in a post entitled "Attachment, ‘silliness', and disagreement".
The contributor said that attachment does not involve stages, whereas Erikson's view of personality development (for example) does have a stage framework. To examine this statement, we need to consider both the meaning of the term "stage" and the development of attachment as described by John Bowlby.
Students of introductory psychology often memorize lists of stages like those suggested by Erikson, and may come away with the idea that a stage is an item taken from such a list. By the time they reach their developmental psychology course, however, they should be exposed to a more general concept of stages of development. Of course, if people have no reason to use this concept, it fades quickly, so let me offer a refresher for those who used to know, and an introduction for those who never studied this important idea.
The term "stage" refers to a period of life during which a particular aspect of the individual appears in a form which is qualitatively different from what we would see at other periods of life. By qualitatively different, I mean that the person actually does things in a different way, rather than simply faster or slower; for example, crawling is qualitatively different from walking. Any stage theory describes a series of life periods, each of which involves a function that is qualitatively different from functions seen at other periods. This series is a predictable sequence, so that in normal development each stage will occur in a consistent time relationship to other stages. As was discussed many years ago (e.g., von Glasersfeld and Kelley, "On the concepts of period, phase, stage, and level", Human Development, 1982), stage theories assume that development involves a progression toward an end state and thus some predictability. However, the length of a particular stage, or its connection with a specific chronological age, are not necessarily predictable. Depending on the aspect of development being considered, the stage progression may be caused by genetic factors, and/or may be driven by transactional engagement with the environment, as described by the late, great Gilbert Gottlieb and others. Depending on the mechanisms of development at work, the progression of stages may or may not involve the building of later stages on earlier ones, but many stage theories of human development do make that assumption.
So, is Bowlby's attachment theory a stage theory, even though we see no lists of stages represented in textbooks? It certainly involves qualitative differences in behaviors toward other people, associated with different periods of life. Babies in the first months show similar social responses to most friendly people, but by the end of the first year have strong preferences for familiar caregivers and show fear of people whom they would have smiled at in the past. Older toddlers and preschoolers show less stranger or separation anxiety, but specialize in the qualitatively different negotiation of separation I described the other day. As the internal working model of social relationships continues to mature, older children develop the capacity for goal-corrected partnerships that help preserve their emotional connections with parents and teachers. Later periods of life were much less carefully described by Bowlby, but in adolescence sexuality becomes part of the internal working model, and in adulthood the model comes to include long-term intimate relationships, the care of children, the care of aging parents, and eventually one's own need to accept care from younger individuals.
Bowlby's attachment theory thus also involves a predictable sequence, so we would be surprised and concerned if an adolescent showed the exaggerated stranger anxiety that is normal for a young toddler. In addition, the progression of attachment is assumed to be toward a complex mature working model of relationships. Most thinking about attachment also assumes that characteristics of early stages of attachment help to establish what happens in later stages, and this issue is one of the reasons for research on the stability of attachment status over years of development.
I think it is probably true that alternative theories of attachment regard the early events of attachment as all-or-nothing processes that establish attitudes and behaviors for an indefinite future period--- in other words, they are not stage theories. However, this is not true of conventional attachment theory derived from Bowlby's formulation, which is a stage theory.