Chapter 4

Transitional Phenomena and Perturbation of Affect Attunement

What kind of processes may be involved in the infant's use of a bit of cloth, or a tuft of hair or wool, or a thumb, or a teddy bear, or some other 'transitional object' as it is termed in Object Relations theory, and which begins to show in Western cultural contexts at about four to twelve months of age? In this chapter I shall attempt to indicate how the infant, by virtue of its self-creative dialogic organization, is able to accomplish the tremendous feat of transforming an "object" (as seen by the adult observer) into an actualized other in the companion space of the infant's virtual other. Even when the original medium for this self-creative transformation may come to be worn out, for example with only dirty traces left of the original piece of wool, of the teddy-bear, or of whatever used as a medium, the transitional and transformed "object" may come to acquire properties as a more or less enduring companion as an actualized other in the inner companion space of the child's virtual other. This act of creation will enable the infant, when left alone, in need of comfort, or going to sleep, to re-enact dyadic togetherness with a actualized companion by virtue of the dyadic organization of the infant's mind.
In the previous chapter an explanation was offered of the infant's engagement in
protoconversation in an interpersonal lifeworld prior to its being structured by language. The theorem of dialogic closure allows for specification, as we have seen, of the dialogic recursive operations of a self-organizing dyad, recreating itself in the infant with a virtual other, and inviting the actual other to replace the virtual other in dyadic closure. If this closure is perturbed, the thesis entails that the infant will renew his invitation to the engage the actual other, and if failing, will resort to dialoging with his virtual other. In this chapter some perturbation studies will be considered in this light and which support the implications of the thesis. I shall also consider in this and the next chapter transitions to modes of mediate understanding, that is, understanding mediated by acquired means of evoking and expressing re-presented and generalized experience. I shall approach the issue of how "objects" come into play, both as a topic of shared attention by the infant and adult in mediational understanding, elaborated in the next chapter, and as means for the infant's self-creative transition in what Donald Winnicott terms "transitional phenomena".1

Transitional phenomena2

The medium for the baby may be a bit of cloth that once belonged to the cot cover or a blanket, or perhaps a bundle of wool, or a word or a tune or a mannerism, which are 'not-me' objects, and yet are weaved by the infant into a personal pattern through the infant's own creation of an intermediate domain between the infant and absenting others in a world of objects. It becomes vitally important to the infant for use at the time of going to sleep. If the familiar smell or taste is washed away, it will no longer be accepted as a medium. The medium may be thumb-sucking, perhaps accompanied by teddy bear stroking, or any other personal pattern of creation, which begins to show at about four to twelve months of age. It is created by the child as a quiet resting place, as an intermediate area of experience which may allow for the differentiation between self and others, and yet keep them inter-related. It is a first not-me creation, co-created by the child with its virtual other within the boundary of the personal. It is based on the confidence in the union with the actual other, and yet created as an intermediate, or in terms of the thesis, as something that may presently fill the virtual other in the absence of the actual other.
As we saw in the previous chapters, when certain requirements concerning the situational context and the initial state of the participants (and the observer) are satisfied, a parent-infant dyad may come to exhibit the form of the (proto)dialogical through the closed operation of one dyadic organization. It recreates itself through the actual other of each replacing the virtual other of each in a reciprocal mode of felt immediacy. As pointed out in the previous chapter, for the mother, who during pregnancy has experienced the child as her actual companion, literally coming to fill her bodily participant space, the moments of loving care after birth offer a qualitatively new realization of being fulfilled, but now in a symmetric mutually fulfilling way through which both, not only the infant, fill each other's companion space. It is entailed by the thesis about the inherent dialogic nature of the human mind permitting a mutual mode of felt immediacy through dialogic closure. There is mutuality and reciprocity in the loving care bestowed upon the other, each fulfilling the other's companion space in a mode of felt immediacy.
In line with the static inscription offered in Fig.3.1, the way in which the infant and adult comes to engage in such dialogic closure is indicated in the two uppermost domains in Fig. 4.1. Domain (i) gives the initial state of the baby B inviting the adult other A to step into the dialogic helix, which recreates itself in domain (ii) when the adult takes the place of the infant's virtual other. This has been succinctly symbolized in this way:

(T) for baby B: B,*A := B,A

We may now express the reverse transform, relevant to perturbation experiments. Upon dissolution or perturbation of the dyadic closure of the infant-adult dyad (domain ii), the infant will, if it fails to re-engage the adult other in reciprocal interplay, disengage and continue in a self-organizing manner, returning to its initial operations with the virtual other (in domain i):

(U) for baby B: B,A := B,*A

Now, in this process the baby may resort to some object, O, as a medium re-creating the infant-adult interplay. With a safe, or not so safe, anchorage in the dyadic closure co-generated with the parent or careperson, the infant may turn to attending 'objects' that are included in the domain of the dyad. A gradual or abrupt expansion of the domain may occur even in the absence of the actual other, whereby the infant with its virtual other may come to create an actualized other on the basis of such 'objects'. That is, objects as identified by some observer, are made into a subject (and later in life, into a subject matter) filling the companion space of the child in a format the resembles the format (T) of the infant-adult interplay, but where the actualized participant companion characteristics of the "object" are endowed through the internal operation of the infant's virtual other. A static inscription of these transitions are given in Fig.4.1 (ii), (iii), (iv). The transitional phenomenon indicated in domain (iv) may be succinctly symbolized in this way:
(W) for baby B: B,*A := B,O(*A)

(W) is to be read: The companion space of B's virtual other *A is filled by some object O actualized as a companion in B's internal dialogue by virtue of the participation of B's virtual other. Hence, while O is an object as discerned by an observer, it is endowed with life as an actualized other in interplay.

In the way specified in fig. 4.1 (iv), that which is identified as an "object" by the observer, may be seen to be brought into the infant's companion space and endowed with "life" as an actualized actual other by virtue of the infant's virtual other. That is, the dyadic organization recreates itself in the child (with the child's virtual other) when the actual other has left, or is excluded from the infant's companion space, for example, when in the process of going to sleep. Using the familiar smell or taste or sense of some medium - a thumb, a tongue, a lock or a doll (that is on the basis of some object, O, as discerned by an observer), the child will create an actualized other, which permits a kind of dialogic closure to maintain itself in the absence of, or to the exclusion of, some actual other. It may relate to the observations of infants' meaningful acts of care accorded for example to the doll (as seen by the observer), imbuing it with a life of its own.3 Winnicott(1986) terms such media transitional objects, serving as material for the infant's creative life in the space-time domain between the infant and the parent.
Now, such acts of creating an actualized companion in the infant's dialoging with himself may come to be a recurrent pattern in the infant's self-organization if the infant frequently fails to engage the adult other in dyadic closure, or if the interplay with actual others is frequently out of tune, preventing dyadic closure in felt immediacy. Dyadic infant-adult closure, as defined in the previous chapter is precarious in the sense that it may easily be disturbed or perturbed by the states of the participants or their environment. Lived moments of affect attunement, which cannot be discerned by a third party, acting as observer unless it is perturbed, involve the infant-adult dyad in immediate reciprocity. When the parent is distracted, for example, by a phone ringing or someone knocking at the door, and brought out of tune with the baby, the baby may exhibit a behavior that indicates that something is amiss. If the infant fails to bring the parent in again in affect attunement, it may resort instead to dialoging with itself in a self-organizing manner. The present thesis entails that this comes about by virtue of inherent the dyadic organization of the infant's mind, not through the copying or imitating some adult actual other. It is further implied that the infant's self-organizing dialogue with itself is a natural and frequent mode of operation on the part of the infant, not a pathological or artificial mode imposed by others. External conditions, however, may perturb or prevent dyadic closure and affect attunement with actual others so frequently that the self-organizing mode of dialoging with the virtual other becomes an enforced resort, instead of a mode in natural alternation with the mode of engaging actual others in dyadic closure.

Perturbation of dyadic closure

The present thesis permits us to regard occurrences in the domain of infant-adult dyads as being generated by the same kind of organization that is enfolded and realized in the individuals when by themselves. Activities by the individual in solitude, for example, when engaged in playing with a toy, or talking to himself, are seen to come about through operations of the same kind of reciprocal organization that is realized when the individual engage in mutual play or protoconversation with an actual other, in line with (T). When the dyadic closure of the infant-adult interplay is perturbed or dissolved, and the infant fails to re-engage the actual other, the infant will continue, in line with (U), in a self-organizing manner the dialogue with his virtual other, *A, replacing the previous actual other, A.
The thesis entails, in line with (U), that when infant-adult closure is aborted or perturbed the infant is expected to resort to modes of self-organization in dialogue with himself, that is, with his virtual other. When the adult frequently is unable to comply with the infant's natural invitation to enter into dyadic closure, the infant's resort to dialoging with the virtual other may become defensive. Rather than being an experimental ground for co-creative genesis of mediate self-other distinctions from the safe anchorage of dyadic closure, the companion space of the infant's virtual other may become restricted to a domain resorted to upon contact failures.
These expectations may also be expressed in Winnicott's theory, being as they are, implied by his theory and practice.
Key terms are "holding" and "let down". He identifies holding as a form of loving offered to the infant:

"Holding includes especially the physical holding of the infant, which is a form of loving. It is perhaps the only way in which a mother can show the infant her love of it."4

Such holding is contrasted with experiences of environmental let down.5 In terms of the present thesis it means that the infant's invitation to engage in dyadic closure is not complied with. Instead of holding which involves loving fulfillment by the actual other in the companion space of the child, providing a safe anchorage also for the child's self-creative dialoging with itself, the space-time domain between the child and the parent is empty. It may enforce the child to resort to itself as a means of overcoming the suffering of a let-down, that is, to a continuous self-organizing engagement with only the virtual other, without the fulfilling recreation of experiencing the actual other's replacing the virtual other in dyadic closure.

Corroborating studies on the impact of perturbation

Perturbation of dyadic closure and the infant's reaction have been studied by Colwyn Trevarthen and Lynne Murray. The experimental set-up (shown in Fig. 4.2) involves a mother in direct audio- and television link with her 8-week-old baby.

Audio-visual contact is channeled through monitors reflected by partial mirrors in front of the baby and the mother (upper left part of Fig. 4.2). Even though they cannot touch each other they are shown to be happily engaged in communication (Cf. (i.a) and (i.b)). In one case the mother is asked to maintain periodic voice emissions coordinated with head and face movements. The baby may be seen to show an increased frequency of smiling and playfulness as they proceed. When, however, a two minutes record of the mother's engagement in such happy protoconversation is replayed to the infant half a minute after it was recorded, the infant is seen to disengage himself, even though faced with the same sounds and moving images of his mother which earlier had elicited his happy engagement. This is illustrated in Fig. 3.2.
The 8-week-old infant who in television communication with the mother exhibits expressive smiles and gestures (cf. (i.b)), gives signs of distress and exhibits "gaze avoidance" upon a video-replay of the mother (Cf. ii.b). (Murray and Trevarthen, 19856; Trevarthen, 1986:229-231).
Two important features are demonstrated by these experiments and other investigations in which the baby's responses are played back to the mother. First, they demonstrate
the reciprocal nature of natural protodialoging: Both participants are vital to the process as lived moments of reciprocal participation. Second, they show that when the protodialogic closure is perturbed by the live mother being replaced by a replay-mother, the infant does not resort to chaotic or break-down behavior. When the timing of the mother's respons to the infant is disturbed by means of the closed circuit television system, the infant exhibits reactions resembling puzzlement and confusion, followed by gaze avoidance of the mother and self-organized behavior. Thus, the disappointed infant, after failed attempts to establish reciprocal contact, will not collapse in a random or chaotic manner, but will fold back on himself in a self-organizing manner.
This is demonstrated also in other perturbation experiments. Murray reports results (Murray, 1980; Murray and Trevarthen, 1985) that point in the same direction as some previous studies of maternal perturbation during mother-infant interplay (Brazelton et al.,1975; Papousek and Papousek, 1975; Tronick et al., 1978).7 In addition to the out-of-phase conditions, reported above, two other forms of perturbation of the regular mother-infant interplay have been employed8:
First, in an "interruption condition", the mother, whilst talking to her child, is interrupted by the experimenter, diverting her attention from the child. When this occurs the infant quietens, reduce his smiling and dampens his communicative gestures, while watching both mother and experimenter attentively.
Second, in an "blank face condition", the mother complies with the experimental instruction to adopt a blank or still face while continuing to look at the infant. When this occurs the infant first tries to engage more intensely with the mother, frowning at her and thrashing the arm in an agitated fashion. When this fails to elicit respons, the infant appears to withdraw from engagement in the other, becomes self-absorbed, gazing at his hands, or looking blankly into space whilst fingering his clothes or touching his face. This is consistent with the above expectations. As Murray points out, when the recordings are analyzed, the infant's behavior in each sequence is found to be coherent and highly organised. Changes in the direction of gaze were systematically linked to changes in facial expression and communication. The patterning of gaze itself was sensitive to the quality of maternal response. In the addition to the self-organizing nature of infant responses to breakdown in communication, infant distress and avoidance provoked agitation and concern in the mother that led to a heightened solicitousness when normal communication was resumed.

"This evidence, together with the results of other studies that have employed perturbations to maternal communication, supports the view that, by at least six weeks, the infant seeks interpersonal engagements in which the characteristics of the partner's response (eg. form, affect and timing) are finely specified; and that if the appropriative response is not forthcoming the infant will avoid engagement with the environment and will instead fall back on experiences that are generated and controlled by himself. These conclusions are wholly consistent with Object Relations theory and the model of intersubjectivity developed by Braten." (Murray, 1991:8)9

The results of the blank-face conditions may throw some light on the outcome of a sad-face and tone condition imposed upon mothers when engaging their ten-week old babies in dialogue. In their study of the infants' responses to three emotional expressions, happiness, anger, and sadness, Haviland and Lelwica (1987) ask each mother, facing her baby, to adopt a facial expression and tone of voice that repeatedly display a given emotion, with rest periods in between. In the happy face and tone condition the infants appear pleased and happy. In the angry face and tone condition they looked angry or remained still - perhaps in puzzlement since the mother, after all, is faking anger, and hence need not give a consistent performance. This may also apply to the conditions in which the mothers adopt a sad face and sad tone of voice. In this condition, however, the infants would often engage themselves in mouthing, chewing and sucking.10 Harris (1989) points out that the results demonstrate that even the ten-week-old babies clearly distinguish between (the mothers' expressions of) the different emotions in their reaction, particularly when faced with it for the first time in these perturbations.11 But there may be more to these results. In the sad face and tone condition, the baby's mouthing, perhaps even the chewing, could be interpreted as communicative expressions in the ten-week old baby's attempt to re-engage the mother in authentic protodialogic closure, perturbed by the mother repeatedly adopting the sadness expressions. Chewing and sucking may be interpreted as reactions upon the adult other's failure to comply with the infant's invitation to engage in dyadic closure. As in the blank-face condition the baby may be seen to disengage himself in a self-organizing mode, involving his virtual other in the place of the uncomplying actual other.

Affect attunement noticed only when broken

In the above experiments the artificially imposed condition, be it the imposed out-of-phase condition, the disruption condition, or the blank or sad face condition, will inevitable break or prevent whatever affect attunement that otherwise might have arisen in the dyad under natural conditions. Affect attunement is an aspect of infant-adult interplay that easily escapes the notice of the observer due to the fact that only when it is broken is there any noticeable change in the infant's behavior.
During the first month or so after birth, there is an absorbed inward-oriented focus in the infant-adult dyad upon the baby's states and attending bodily needs. But soon a rhythm may be come to be established. There is mutual attention and protoconversation may be exhibited in a mode of felt immediacy. Up to the fourth month in the baby's life the outside world of the dyad may be excluded from the joint attention of the dyad. There is interaffectivity, or what Trevarthen terms "primary intersubjectivity", in a manner that does not yet involves much attention to outside "objects". But gradually, other things are brought into focus of attention. For example, a mother and her baby girl (4 1/2 months) has established a pattern whereby noises she buries her head in the baby's body, accompanied by loud exclamations, and the baby laughs, whenever this is repeated.
After a while, another "object" is brought into the play. The mother places the baby in her lap with the girl's face outward. In front of the baby her mother holds a play-bear (her older brother's), which is brought to close upon and burying itself in the baby's body, accompanied by the mother's voice. The baby laughs. The mother repeats the process, withdrawing the play-bear and bringing it close again, and the baby laughs. An object has been introduced, whereby the self-enclosed boundary is extended to comprise a third element. The baby attends, with the mother, the approaching object in her mother's hands. There is not focus upon the play-bear as such, only participation in the event involving this new element.
But the baby is also beginning to pay attention to objects and events outside of the dyad. For example, during video-recording of the mother-baby interplay at the nursing table, the baby may shift her gaze from her mother to the camera-man, "studying", as it were, for a prolonged period this strange apparition, continuing to look even when the mother calls upon the baby to re-direct her attention to the mother. While left alone, in a semi-sitting position, the baby in the fourth or fifth months of her life, may be seen to gaze upon the panorama in front of her, even upon the changing patterns on the television screen, and will lighten up, with smiles and gestures, when some actual other approaches her with inviting noises. Conversely, while engaged in intimate contact with an actual other, the baby may break off, as it were, the contact, and gaze upon the other in a different manner, "scrutinizing" the face of the other almost as an object of attention in an impersonal, detached manner. For example, as I am sitting and rocking my grandchild, holding the baby (4 months old) in my lap, while she is making content noises in a kind of humming way, she suddenly stops and lifts her gaze, studying my face for a prolonged period. It may have been brought about by my starting to think about this book I am writing, perhaps bring my rocking out of tune with her humming. Her gazing at me gives me the feeling of being studied in an almost detached, yet friendly manner. She makes me aware of smelling and looking differently than her mother. At this moment we are clearly out of tune with each other. In this mode of being gazed at there is distance, felt radically different from the feeling when both of us are involved in the rocking and humming, while the baby lets her gaze wander about in the room. or, at other instances, when we are engaged in protoconversation. An event of protoconversation may also be broken off, for example by me, required to do something elsewhere, and offering instead to the child something else to focus upon. For example, in one instance I offer the child a brightly coloured leaflet. The baby tries to grasp it with her hands and bring it to her mouth in intense concentration. When losing the leaflet, I retrieve it for her, and she repeats the process.
There is, then, a phenomenological ground for attributing to the infant the emergent ability to sense and distinguish things and even sometimes persons as "objects" of attention, study and manipulation. The term "object" used in Object Relations theory and in the Piagetian tradition makes sense in this respect of applying to some topic for which the source distinguished from the other, actual or virtual, qua participant in the self-creative dialogic organization of the child. When I am scrutinized by the child, her scrutinizing makes me into an object of reflection. But the term "object" is not quite suitable as a label when I am the actual other participating in protoconversation with the child. The child and the adult may be "objects" of observation to the observer, or the adult may become an "object" of symbolic representation to the child who has acquired mediating means for reflection, or the adult may be remembered as an "object" related to in childhood. But in the immediate present of an actual other in protoconversation, the actual other is no object, although the very same person in another situation and in other circumstances may come to acquire stable and object-like characteristics.
The initial interaffectivity of the dyad may be retained as the dyad transforms itself into a system involving objects. Even as the baby, about nine month old or so, may reach for or be engaged in some activity involving an "object", such as a toy, both mother and child behave in a way that express the quality of feeling a shared affect state. There is affect attunement, as considered in the previous chapters, an attunement which takes the shared felt resonance and recast that Erlebnis into another form of expression. Stern lists some examples:

"A nine-month-old girl becomes very excited about a toy and reaches for it. As she grabs it, she lets out an exuberant "aaaah!" and looks at her mother. Her mother looks back, scrunches up her shoulders, and performs a terrific shimmy with her upper body, like a go-go dancer. The shimmy lasts only about as long as her daughter's "aaaah!" but is equally excited, joyful, and intense.

A nine-month-old boy reaches for a toy just beyond reach. Silently he stretches toward it, leaning and extending arms and fingers out fully. Still short of the toy, he tenses his body to squeeze out the extra inch he needs to reach it. At that moment, his mother says, "uuuuuh...uuuuuh!" with a crescendo of vocal effort, the expirations of air pushing agist her tensed torso. the mother's accelerating vocal-respiratory effort matches the infant's accelerating physical effort...........................

A ten-month-old girl finally gets a piece in a jig saw puzzle. She looks towards her mother, throws her head up in the air, and with a forceful arm flap raises herself partly off the ground in a flurry of exuberance. The mother says "YES, thatta girl." The "YES" is intoned with much stress. It has the explosive rise that echoes the girl's fling of gesture and posture." " (Stern, 1985:140-141)

There occurs what deserves the label of being a communion involving affect attunement as the infant reaches for or relate to the some object with the adult other participating in the event by way of expressions attuned to the occasion. As long as there is attunement, the baby may continue the activity. But at the slightest "false note" or indication that the parent is out of tune, the baby may stop his activity and take notice. Stern describes observed cases where the mother is instructed to bring her accompanying voicing out of tune with the child, either overplay or underplay expression of the actual state of feeling she is sharing with the child, engaged in playing. While some mothers found this difficult to do at all, some succeeded in perturbing the affect attunement. In these artificial conditions, in contrast with natural conditions of attunement, the child aborted the activity, for example, stopped playing and looked around at her (Stern, 1985:150).
Stern gives an example of a natural condition. In the videotape of a play period, a nine-month-old infant is crawling away from the mother towards a new toy. He grabs it and begins to bang and flail with it in a happy manner. Mother approaches him from behind, out of sight, puts her hand on his bottom while giving an animated jiggle side to side. Her jiggle appears to have no effect on the infant, who continues his play, not missing a beat. When again the infant wanders from her and becomes involved in other toys, she again leans over and jiggles his bottom, or his leg or foot. As in the former episode, the infant continues as if nothing has occurred (Stern, 1985:150). Now to the uninformed observer, in these episodes the infant may appear to be unaware of the mother, solely focused on the play. Not so. They are in affect attunement, indicated by the way in which the speed and intensity of her jiggle matches the intensity and rate of his movements, breathing, and vocalization.
This is demonstrated by the baby's reaction to the mother's perturbed jiggling in the artificial condition, when instructed to underplay or overdo in intensity her accompanying the baby's level of joyful animation. When the mother jiggled somewhat more slowly and less intensely than she felt, the baby stopped playing and looked around at her. When she again repeated her efforts to be out of tune, the result was the same. When trying to overdo her jiggling, pretending that the baby was at a higher level of animation than she felt he was, again the result was the same: The infant responded by stopping playing. When going back to jiggling appropriately in the natural condition, in accord with her feeling with the baby, the infant continues playing without any apparent notice, as in the initial period.
In another case of natural attunement, an eleven-month old infant is accompanied by his mother's comment, "Yeah, ya like that", when the infant brings it to the mouth with much excitement. As in the above natural episode the infant does not show any response to her commentary. When the mother complies with the experimenter's wish to over-shoot or under-shoot her commentary compared to her feeling of match with the infant, the infant takes notice and looks at her, as if for further clarification.

Impact of parent depression

The experience in infancy of affect attunement in a positive sense with the parent or careperson is important to the way in which the child will come to be able to recognize and share feelings with others. As Stern points out:

"It is clear that interpersonal communion, as created by attunement, will play an important role in the infant's coming to recognize that internal feeling states are forms of human experience that are sharable with other humans. The converse is also true: feeling states that are never attuned to will be experienced only alone, isolated from the interpersonal context of shareable experience. What is at stake here is notion less than the shape of and extent of the shareable inner universe."(Stern, 1985: 151-152).

A parent or careperson may sometimes, when feeling depressed, fail to establish a protodialogic "dance" in dyadic closure with the infant. When in a depressed state the adult may also find it hard, sometimes impossible, to sing a nursery rhyme to the child. The depressed state may make the adult self-absorbed and disturb her otherwise natural capacities for dyadic closure and attunement with the child.
Studies have been carried out by Lynne Murray and co-workers at Cambridge University Winnicott Research Unit of how perturbations of infant-adult contact from two months onwards may affect the infant's self-other differentiation and attachment to the mother at eighteen months.12 The quality of the mother's relationship to the baby was directly assessed from videotaped interactions throughout an eighteen months period. Depressed mothers, being preoccupied with their own concerns, showed a lack of identification with the infant. The impact was measured through infant performance of object concept task, and assessment of infant attachment to the mother at eighteen months. Murray (1989,13
1991,14) reports these results of the preceding infant-adult contacts, some of which are perturbed by postnatal parent depression:15 First, the infant's capacity to distinguish between self and other was found to be impaired and to be related to the quality of early parent-infant engagements at two months. Second, the infants whose mothers had experienced depression appeared to be less able to integrate the distress of separation from the mother and instead responded with avoidance and disorientation or resistance.(Murray, 1991: 13-14).16 Both of these results are considered to be supportive of ideas from Object Relations theory and of the present thesis.17
Now, the above findings may evoke worry or guilt feelings on the part of parents who find themselves depressed from time to time, and who in those periods feel that they cannot comply with their child's invitation to "dance" in protodialogue or vocalize in unison, or feel that they are unable to tune in, for the time being, with their baby. There need be no great cause for concern as long at the parent is aware of the baby's sensitivity to adult feelings and is capable of holding the baby in love, not hostility.
First, as long as not parent hostility evoke in the child traumatic feelings of being let-down, the impact of temporarily parent depression may wear off as there are renewed opportunities for reciprocal contact and affect attunement in a positive sense. There cannot be a continuous attunement all the time, as long as there is holding in the physical and loving sense that Winnicott emphasizes in contrast to environmental let down and hostility on the part of the parent.
Second, when feeling depressed the parent or careperson if turning to the child for comfort, instead of showing disdain or aversion through own self-absorbed attention, would find, perhaps to her or his surprise, that the infant already in the first year of life is capable of feeling with the other in distress in a way that may elicit comforting gestures. Compare the story in the introduction about Katharina (26 weeks) who, albeit from the lap of her mother, comforts her elder sister. Other cases pointing in the same direction will be taken up in chapters 7 and 8.
Third, while supported by the above findings, Winnicott's theory and the present thesis implies that no adult plays a direct part in the self-creative moments of the infant's transitional phenomena. Even when parents or careperson are always present and ready to engage in reciprocal contact, they will find that the child, from time to time, prefers to enter into dialogue with itself, excluding the adults from the process. The transitional period before going to sleep is but one such instance, albeit the quality of turning to the dialogue with a transition object will depend on the previous felt quality of dyadic closure with the adult other. The self-organizing mode of transitional events involve precious moments of self-creative achievements which exclude the participation of present actual others, whether they have been in previous attunement or not. It prepares the way for the creative achievements that is yet to come later in life.

Transitional phenomena as the first acts of mediate creation

There is nothing mysterious about transitional phenomena. Infants do it most of the time, and even adults frequently resort to similar processes, for example, in the way in which some object of magical or religious significance are used in spiritual communion, filling, as it were, the companion space of the believer's virtual other. They may be seen to conform to the same pattern if one applies the perspective of dyadic closure not only to the infant in transition, but also to the way in which what Winnicott has identified as transitional phenomena as the first lesson in how contact may be re-established and re-enacted through a self-organizing creative process of mediation throughout life:

"because in the space-time area between the child and the mother, the child (and so the adult) lives creatively, making use of the materials that are available - a piece of wood or a late Beethoven quartet."(Winnicott 1986:36-37)18

Hence, separation later in life, may be creatively overcome, as it were, in a modus that corresponds to perhaps the first impressive creative act of mediation that the infant resorts to. The first act of creation will enable the infant, when left alone, in need of comfort, or going to sleep, to re-enact dyadic togetherness with a actualized companion that will be available to the child during preschool years, and, for some, throughout life.
Involving the use of media such processes will come to involve also the use of language, serving as a transitional medium. For example, Katherine Nelson followed with a group of infant researchers, how a girl, after her second birthday, engaged in a protodialogue with her father, and then continued dialoging with herself after he had left.19 Nelson had recorded her "crib talk" before and after her second birthday. The group met monthly to observe the events. Stern notes how she in her "monologue" after the father left could be seen to struggle with finding the right linguistic forms, and with successive trials, to move closer to more satisfying verbal expressions. But while she thus appeared to be engaged in the practice of languaging, she also appeared to be engaged in a "transitional phenomenon" in Winnicott's sense:
"it was like watching "internalization" happen right before our eyes and ears. After father left, she appeared to be constantly under the threat of feeling alone and distressed. (A younger brother had been born about this time.) ...she repeated in her soliloquy topics that had been part of the dialogue with father. Sometimes she seemed to intone in his voice or recreate something like the previous dialogue with him, in order to reactivate his presence and carry it with her towards the abyss of sleep." (Stern, 1985:173).

In this case, perhaps no noticeable "transitional object", like a doll, or a blanket, or a thumb, was involved. Instead, language expressions may have served as the medium for her reactivation of the dialogue in the companion space of her virtual other.
In the next chapter will be considered how infant-adult dyads bring objects qua objects or topics of common concern into play and language in a scaffolding manner that paves the way for the transition to mediate modes of understanding.

1D. Winnicott: 'The Theory of the Infant-Parent Relationship' in: P. Buckley (ed.) Essential Papers on Object Relations, New York University Press, New York 1986. I am grateful to James Griffith for directing my attention to this work.
2Parts of this chapter is based on S. Bråten (1989): Toward a dialogue theory of mind: The prosocial infant in protodialogue. Prepared for a book dedicated to the work by Ragnar Rommetveit, edited by A. Heen Wold (ed.): The Dialogic Alternative. Oslo: Universitetsforlaget/Oxford University press (in press).
3H.L. Rheingold and G.N. Emery: 'The Nuturant Acts of Very Young Children' in: D. Olweus et al. (eds.) Development of Antisocial and Prosocial Behaviour, Academic Press, Orlando 1986, pp.75-95 (p.77).
4D. Winnicott: 'The Theory of the Parent-Infant Relationship' in: P. Buckley (ed.) Essential Papers on Ojbect Relations, New York University Press, New York 1986, pp.254-271 (p.245).
5S. Winnicott: Home is where we start from, Penguin., Harmondsworth, 1986, p.31.
6L. Murray and C. Trevarthen (1985): Emotional regulation of interactions between two-months-olds and their mothers. In: T.Field and N. Fox (eds.): Social Perception in Infants. Norwood, N.J.: Ablex.
7The other studies referred to by Murray also involve experimental perturbation of infant-mother communication through brief alteration of form, timing and "comprehensibility" of the maternal behavior .....Brazelton et al, 1975; Papousek and Papousek, 1975; Tronick et al. 1978....
8Lynne Murray: The sensitivities and expressive capacities of young infants in communication with their mothers. Ph.D. thesis University of Edinburgh 1980; L. Murray and C. Trevarthen: Emotional regulation of interactions between two months olds and their mothers. In: T. M. Field and N. Fox (eds.): Social Perception in Infants. Ablex, New Jersey 1985.
9Murray, op.cit., 1991, p.8
10Haviland, J.M., and Lelwica, M. (1987). The induced affect response: 10-week-old infant's responses to three emotional expressions. Developmental Psychology, 23, pp.97-104 (Referred to by P. Harris: Children and Emotion, Basil and Blackwell, Oxford, p.18)
11Paul Harris: Children and Emotion. Oxford: Basil Blackwell, 1989, pp.18.
12another.... the findings of Björg Röed Hansen in a projective study of infant-adult dyads focusing on the mutual gazing and attention appear to be consistent with the model implications.
13Lynne Murray: The effects of postnatal depression on the infant. Baillière's Clinical Obstetrics and Gynecology, Vol.3, no. 4, December 1989, pp.921-933.
14Lynne Murray: Intersubjectivity, Object Relations Theory and Empirical Evidence from Mother-Infant Interaction. Infant Mental Health Journal, 1991 (in print).
15another.... the findings of Björg Röed Hansen in a projective study of infant-adult dyads focusing on the mutual gazing and attention appear to be consistent with the model implications.
16Murray, op.cit., 1991, pp. 13-14.
17Lynne Murray: Intersubjectivity, Object Relations Theory and Empirical Evidence from Mother-Infant Interaction. Infant Mental Health Journal, 1991 (in print).
18D. Winnicott: Home is where we start from, Penguin, Harmondsworth 1986, pp.36-37.
19Reported by Stern (1975:172-173). The group consisted besides Katherine Nelson and Daniel Stern, of Jerome Bruner, John Dore, Carol Feldman and Rita Watson.