theory, “The central importance of social relationships (in psychoanalytic terms, ‘object relations’) in shaping the person’s emotional and cognitive growth is the clinical essence of psychoanalysis” (1977, 122). Or as Greenberg and Mitchell put it, in person-oriented object relations theory “the unit of study of psychoanalysis is not the individual, but the relational matrix constituted by the individual in interaction with significant others” (1983, 220).
Bowlby himself rarely puts into play the conceptual vocabulary of object relations. Why is that, one may ask, and whatever happened, intellectually and emotionally, to Bowlby’s own analyst, Joan Riviere, and to Meianie Klein, one of Bowlby’s supervisors? Bowlby does not neglect to acknowledge his debt to them “for grounding me in the object-relations approach to psychoanalysis, with its emphasis on early relationships and the pathogenic potential of loss” (1969, xvii), yet most of his work departs radically from Klein’s. The necessary inference for those familiar with Bowlby’s methodology and cognitive style is that in rebelling against certain features of contemporary British object relations theory Bowlby bent over backwards to avoid any inferences not based on solid, empirical evidence. Yet his subject matter, as distinguished from his methodology, is entirely object relational. He himself declares that his most “central concepts” are “object relations, separation anxiety, mourning, defence, trauma, [and] sensitive periods in early life” (xv), a group of categories that can be lumped together without any distortion as “object relations.” Bowlby specifies that attachment theory derives from object relations theory and has much in common with the work of Meianie Klein, Fairbairn, Balint, and Winnicott (17).
Positioning attachment theory vis-à-vis object relations theory necessitates supplying what attachment theory leaves out, such as attention to particular individuals, and to internalized representations and their processing, while emphasizing those features of object relations theory, collectively considered, with which attachment theory is correlative and compatible, such as psychological responses to loss. When I say “object relations theory, collectively considered,” I refer as well to ideas deriving from self theory not hitherto part of earlier versions of object relations theory, such as the concept of intersubjectivity and the process Stern calls “affect attunement.” Within an expanded framework of self theory and object relations theory, attachment theory as we find it in Bowlby constitutes a special branch, one that continues to grow through his followers’ contributions.
After relating the anecdote about Fairbairn’s question to the abused child, Guntrip mentions that the story illustrates Fairbairn’s concern about the quality of parent-child relations. As a rule, Bowlby pays little direct attention to the quality of parenting. He speaks instead, along quantitative lines, of the presence, or temporary absence (separation), or permanent absence (loss) of parenting figures. Since his method is prospective rather than retrospective, he does not rely on case histories of adult individuals for illustration. He limits his attention to pathology pretty much to the directly observable consequences of separation and loss—such as those mentioned in experiments with animals, especially Harlow’s. Yet even though Bowlby does not talk much about pathological object relations direcdy, he does do so on occasion, one of them being when he approvingly cites Bateson’s double-bind theory of the origin of schizophrenia (Bowlby 1973, 317-19). Another instance that comes to mind is when Bowlby mentions two cases of matricide: “One, an adolescent who murdered his mother, exclaimed afterwards [presumably without irony], ‘I couldn’t stand to have her leave me.’” In the other case, “a youth who placed a bomb in his mother’s luggage as she boarded an airliner explained, ‘I decided that she would never leave me again’” (1973, 251). The point to be registered is that although Bowlby keeps neurosis and psychosis in the background of his discussion in the Attachment and Loss trilogy (1969, 1973, 1980), and although he does not spend much time focusing on separation as a source of crippling emotional conflict or behavioral maladaption except when discussing experiments with animals, a comprehensive theory of object-relational conflict cannot possibly avoid attending to the themes of attachment, separation, and loss, particularly insofar as the effects of pathological parenting can be regarded as comparable to those of separation and loss. The beginnings of such an expansion of attachment theory have already been initiated by such figures as Ainsworth, Main and Weston, Henderson, Brown, Adam, and Parkes (all in Parkes and Stevenson-Hinde, 1982), and Bowlby’s later work (1979, 1988) addresses the issues of etiology and psychopathology more direcdy than the Attachment and Loss trilogy.
Guntrip’s anecdote concerning Fairbairn’s question to the litde girl implies the presence of a sexual factor when Guntrip remarks (presumably paraphrasing Fairbairn) that the girl’s response reflects “the intensity of the libidinal tie to the bad object” (1975, 146). Is this just another instance of “libidinal” being used loosely as a synonym for “emotional,” or are such ties erotic? Attachment theory assumes they are not erotic, the need for attachment itself being the primary instinct in operation. What, then, may be said concerning the relation of attachment behavior to sexual behavior, especially when Bowlby expressly declares attachment theory to be an alternative to libido theory (1969, 17)? The answer is that while Bowlby jettisons the theory of psychical energy, and while he tends to exclude sexual behavior from the areas of his attention, he does not in fact deny the existence or even the importance of sexual behavior. He treats sexual behavior (1969, 230-34) as a separate system of activity that has “close linkages” to attachment behavior. These otherwise separate systems of behavior may “impinge” upon and “overlap” each other, the examples he gives of sharing behavioral components being adult clinging and kissing. Presumably only King Solomon could separate erotic factors from attachment factors in lovers’ kisses—or in their sexual intercourse, for that matter. For Freud, even thumb sucking is an erotic activity. But Fairbairn believes babies suck their thumbs because there is no breast to suck, so that thumb sucking “represents a technique for dealing with an unsatisfactory object-relationship” (1952, 33). And for Winnicott also, thumb sucking, a transitional phenomenon, clearly pertains as much to other as to self (1971). What matters in this connection is not to locate particular instances of unmixed instinctive behavior but to recognize the high degree of ambiguity often prevailing in human action with respect to the kind, and proportion, of instincts involved. Granting the presence of that ambiguity makes it understandable that what has usually been interpreted as sexual behavior under the aegis of Freud may in fact have been primarily or essentially motivated by attachment needs, a proposition that will be illustrated at length in the reading of Freud’s cases in chapter 3.
A factor to consider in the task of positioning attachment theory in a broader theory of object relations concerns the common practice of using the term “attachment” in a literal and very circumscribed manner, often with a sharp distinction between “attachment” and “attachment behavior”(Bowlby, 1982, 371; 1988, 28). Used in this way, the child’s answer to Fairbairn, “I want my own Mummy,” denotes a fairly literal tie, or emotional bond, to what is by definition the child’s primary attachment figure. Although Bowlby generally limits his discussion of attachment behavior to such instances in early childhood, he recognizes that “attachment behaviour does not disappear with childhood but persists throughout life” (1969, 350). He also clearly links transference activity to attachment behavior (1969, 17; 1973, 206, 271). A particularly good instance of Bowlby’s use of “presence” and “absence” in a non-literal way occurs when he writes, “A mother can be physically present but “emotionally’absent. What this means, of course, is that although present in body, a mother may be unresponsive to her child’s desire for mothering” (1973, 23). The point being led up to is this: if attachment theory is to be part of a broader theory of object relations instead of being confined for the most part to developmental psychology, then the concept of attachment must be deliteralized and broadened in a way that recognizes its endless permutations. Freud remarks that “the finding of an object is in fact a refolding of if (1905b, 222). By the same token, one can say that subsequent attachments to some extent replicate earlier ones. All major attachments in adult life constitute versions, or permutations, of earlier attachments, which is tantamount to saying that adult interpersonal relationships reflect the object-relational history of the individuals concerned.
