Introducing a Thesis:

Born with a Virtual Other in Mind

"Some people,' said Humpty, looking away from her as usual, "have no more sense than a baby!"
('THROUGH THE LOOKING GLASS and what Alice found there' by Lewis Caroll, 1872)

Current research on infancy and child development invites a revision of our picture of the human mind, of the roots of social understanding and concern, and of the relation between thought, feeling, and communication. Some classical conceptual distinctions are found wanting. For example, in need of complements are Darwin's categories of emotions, Freud's attribution of a superego as the mind's agent of morality, and Piaget's concept of decentration required for the child's partaking in an interpersonal lifeworld.
We have once taken part in the lifeworld of infants, and may again, as parent or caretaker, come to share in some of the pains and delights involved in infancy. The study of human nature in infancy and beyond can be a fascinating mixture of story-telling and conversations, video-recording and participation, and our trying to turn the looking-glass towards ourselves in wonder.
One's particular perspective limits the focus and determines the levels of resolution and details of that which is captured and constructed. When writing about infancy one attempts in a way to offer a looking-glass to the reader. It may appear as an objective undertaking. It is not. Feelings, even dreams, play a part, as they do in the genesis of the thesis behind the explanations offered in this book. And yet, the attempt to explain is not a subjective undertaking. To invite others to share a given view, even tentatively and as a part of dialogue between complementary views, is an intersubjective undertaking. What the other may see through the looking-glass, may fit with preconceived notions, or make no sense, or perhaps sometimes turn out like the looking-glass offered to Alice, turning things on their head.
The view of the nature of the mind in infancy, childhood, and adulthood submitted in the following appears to turn things on their head. For example, when looking at the infant and the adult in interplay the reader is invited to see them as one self-organizing system, not two. When looking at the infant left to himself, the reader is invited to see not one, but two participants
making up a self-organizing dyad. The same form of self-organizing processes that unfolds itself in the infant's protodialogue with an actual other is seen to unfold itself in the infant when by himself, in dialogue with what I term the virtual other.
The virtual other1 can be preliminary defined as a companion process, and associated space, in the mind's dialoging with itself and which invites replacement by actual others. By virtue of its inner companion space for communion with a virtual other, the infant is expected to invite and feel the perspective of the actual other who replaces the virtual other, and to be able to continue by himself when the actual other leaves or is left out. Examples are the baby's use of what Winnicott calls "transitional objects", such as a blanket piece or a bundle of wool, or, later in life, the preschooler's "egocentric speech", found by Vygotsky to be exhibited even by the problem-solving school child when in trouble. Such occurrences will be specified in the chapters to follow as instances of the child's dialoging with his virtual other.
When during the first months of life the parent or careperson as actual others replace the infant's virtual other, the self-organizing dyad that thereby recreates itself exhibits patterns that appear impressive or puzzling to the outside observer. Examples are perfectly timed turn-taking and sometimes minute inter-bodily coordination and apparent imitation of gestures.
As a scientist or an observer one is outside children's lifeworlds and yet attempt to make sense of them. Perhaps understanding may discoveries are facilitated by being in a boundary position, able to shift between being an inside participant in adult-infant dyads and an observer applying the looking-glass. Mary Catherine Bateson's finding is a case in point:
In 1970 she carries out studies in a tiny film projection room at M.I.T. Between breastfeedings of her own newborn daughter, she spends hours studying and analyzing films and tapes of mothers and their children in their first months of life. She picks out series of brief passages in the films and analyzes them frame by frame. She does acoustic analyzes of the infants' and the mothers' vocalizations. She finds that by three months there is a kind of flowing pattern of alternating vocalizing turns and accompanying gestures between infants and mothers, sometimes exhibited already within a scant month after birth. This pattern she later terms "protoconversation". It allows for a description in the metaphor of a dance, and as conforming to the turn-taking pattern of adults in conversation. Colwyn Trevarthen and his coworkers who have carried out extensive laboratory studies of mother-baby interplay and protoconversations, find that they are evidences of what he terms 'primary intersubjectivity' on the part of the infant.
These are instances of infancy research findings during the past decades which call for a revision of the way in which infancy has been understood. Infants turn out to be more sociable than previously assumed, ready to engage in mutual interplay and reciprocal interaction, capable of being in affect attunement with the other, and even showing prosocial behavior, that is, caring for and helping the other in need or distress.
For example, as I am writing this introduction in my summer cabin, I am interrupted by my daughter's telling me about how her youngest child, Katharina (26 weeks) comforted her elder sister, Kine (4 year and 3 months). They were in bed with their mother. Kine kept begging for her dummy and juice bottle and then starts sobbing. Katharina, sitting on her mother's stomach, turns over, extends her arms towards Kine, and then puts her face against Kine's face. This case of Katharina (five months old) showing concern for her sister in distress is but one of numerous examples of the sociability, care and understanding that infants and young children may exhibit. The case is consistent with what has been revealed in recent infant research, and perhaps with what parents have known in their heart for ages. When parents meet with their babies, for example, it is hard for them not to notice how the children will pay each other acute attention. As will be exemplified later, even an infant born three months before term, when carried in "kangaroo"-manner by the parent, can engage in reciprocal visual and auditory contact with the adult.
These examples do not fit the view of the infant as asocial, egocentric, or lacking contact abilities. Hence, they pose a challenge to theories of child development that traditionally have entertained such viewpoints. Revealed phenomena of infant "imitation" and affect attunement, of protoconversation and prosocial behavior challenge both individualistic and collectivistic perspectives. They also partly defy explanations from an information processing perspective of mediating symbolic representations. A complementary perspective is proposed in this book. It will be seen to permit the modeling of processes within and between infants and adults in terms of the same dyadic format: the self-organizational form of the dialogical.
The core of the thesis that will be employed in explanations offered of the above and other findings and phenomena is this:

(q) The infant is born with a virtual other in mind that invites replacement by some actual other in felt immediacy. Hence, the immediate form of understanding and feeling is assumed to have the primary form of the dialogical, irrespective of whether it recreates itself (i) in the mind (in dialogue with the virtual other), or (ii) between minds (in dialogue with an actual other).

When the adult actual other takes the place of the baby's virtual other, the self-organizing dyad may be seen to recreate itself in the same operational form, now realized by two organisms instead of one. This will be seen to permit an explanation of how a prematurely born girl baby, long before term is able to engage in a reciprocal duet with her father. When contact with the caring other is aborted or perturbed, and the baby is left to her herself she will, before going to sleep, come to engage in a self-organizing manner in a continued contact with herself, that is, with her virtual other which endows "life" to some transitional object, such as her thumb, tongue, or a dummy.
No qualitative jump need be involved when the baby shifts between (i) being engaged in such contact with herself and (ii) being engaged again in contact with the actual other. This is indicated in Fig.0.1.

The self-organizing dyad of the baby B recreates itself in the same form of felt immediacy when the baby's virtual alter *A is replaced by adult actual alter A. By virtue of feeling its own lifeform through its virtual alter the baby will feel the lifeform of the actual alter replacing the baby's virtual alter. Noticing how natural and effortlessly the baby participates in protodialogue, Newson refers to those "mysterious processes" which underlie the baby's capacity to share states of feeling with the adult. In his studies of emotions in man and the animal Darwin suggested an instinctual emotional basis for infant's ability to sense feelings in others. Stern identifies what he terms vitality affects as a complement to Darwin's categorical emotions, and points to infants' apparent innate capacity for cross-modal and a-modal perception of the other's emotional expressions. As we shall see, when offering an account of protodialogue, the puzzling aspects of shared feelings and affect attunement, as well as their perturbation, may be considered in similar terms.
Based on infant studies in his laboratory at the Edinburgh Centre for Research in Child Development, Trevarthen was the first to come up with experimental evidences and corroborating statements about processes corresponding to a virtual other in the infant's mind. In its emotional and communicative organization there appears to be something that expects and welcomes the mother as the actual other. The model2 formulated in line with Fig. 0.1 (specified in chapter 3) was found to explain protoconversation.3C. Trevarthen (1989): Origins and Directions for the Concept of Infant Intersubjectivity. Newsletter for the Society for Research in Child Development, Autumn 1989.pp.1-4.
When infant-adult communication is aborted or perturbed, the model predicts that the infant will resort to modes of self-organization in dialogue with himself, that is, with his virtual other, rather than randomized or chaotic behavior. Confirmation in this respect has been provided by Lynne Murray. At the Cambridge University Winnicott Research Unit she has recently shown how data from experimental perturbation of mother-baby communication and from studies of the impact of postnatal depression are consistent with Object Relations theory and with the model implications.4 "Until recently, accounts of infant emotional and cognitive development in terms of British Object Relations theory and of developmental psychology have been largely provided in separate intellectual domains with little possibility of integration.
The development of the model infant intersubjectivity and the role of the virtual other has provided a bridge between these two disciplines."
This will be turned to in chapter 4. The organization of the book reflects the way in which the self-organizing dialogue of the mind is seen to unfold itself in primordial modes of immediate understanding during the first months of life, and come to be complemented by the cultural and constructed modes of mediate understanding that arise from interactional experiences.

Table 0.1 Primordial, constructed, and distorted domains of self-and other-relations, cutting across intra- and interpersonal levels

(i) INTRAPERSONAL (ii) INTERPERSONAL (ego and virtual alter) (ego and actual alter) (Do) (D1) PRIMORDIAL the infant in the infant in DOMAINS self-organizing self-organizing OF dialogue with protodialogue IMMEDIATE with its with the adult RELATIONS virtual other actual other (D2) (D3) CONSTRUCTED the experienced the experienced DOMAINS infant in self- child in OF conversation with conversation with MEDIATE transitional objects actual others RELATIONS and creative media through symbolic use later in life representations COLLAPSED collapse of collapse of DOMAINS intrasubjective intersubjective THROUGH feelings feelings DISTORTIVE (divorce from (exclusion of RELATIONS the virtual other) actual others) The way in which the same operational format will be applied in the modeling of processes across the intra- and inter-personal level has been indicated in Fig.0.1. They are seen to involve the self-organizing mind as it recreates itself in dialogue (i) with the virtual other in domain (Do), and with the actual other in domain (D1). They are posited as primordial modes involving intra- and intersubjective feelings.5 They may, however, come to be converted or be brought to periodic or more permanent collapse by extremely distortive experiences, bringing about a divorce from the virtual other and exclusion of actual others from felt immediacy.
The co-constructed domains (D2) and (D3) apply only to the experienced infant and to processes later in life, for example, how the baby, having resorted to transitional objects, later in life may find a similar form in the creative use of cultural media. This does not entail, however, that the primordial domains (Do) and (D1) apply exclusively to the beginning of life. Unless permanently perturbed and converted through previous interaction experiences, they will be seen to remain domains for potential processes of felt immediacy throughout life, complementing those of mediate understanding. They pertain to the complementary process of children taking the perspective of the actual other. In the primordial domain (D1) of immediate understanding the actual other's perspective is felt by virtue of the child's virtual other. In the constructional domain (D3) of mediate understanding it is taken by virtue of the experienced child's re-presented or generalized other. The infant's inherent ability to relate to the actual other in what Whitehead terms "presentational immediacy" will be seen to provide the pre-understanding required for acquiring through interaction the symbolic means for understanding also in re-presentational mediacy. For example, as we shall turn to in chapter 8, while infants appear to grasp immediately the other's emotion, Harris finds that young children, when asked about it, are unable to acknowledge expressions of mixed feelings in themselves or in others. While the former pertain to the domain of immediate understanding, the latter concerns the domain of mediate understanding in which the child becomes capable of imagining himself in the other's shoes through re-presentational means for simulating the other.
Findings and phenomena in infancy and child development that will be considered in the light of the thesis are these:

. the ability of the infant to engage in protoconversation, affective communion and attunement with the parent or careperson; and infants' apparent ability to "imitate" expressions of affect (chapters 1 - 4);

. phenomena of transition involving the infant's use of some "transitional object", and the creative use of other media later in life for conversations "with oneself", including so-called "egocentric speech" exhibited by the preschool child (chapters 4 - 6, and 12);
. infants' prosocial capacity to take the others perspective and to show feelings of care and exhibit concern for the other in need or distress, children's mixed feelings when engaged in hurting the other, and the complementarity between sentiments and judgments in moral development (chapters 7 - 9).

Part Two is devoted to the roots and collapse of concern. Explanations will be offered, for example of how it was that Katharina, five months old, could be able to exhibit caring concern for her elder sister in distress. Reports by Anna Freud, Zahn-Waxler, Radke-Yarrow and others on children showing care and concern for the other in need or distress will be seen in the light of implications of the thesis. Children not only engage in helping and comforting each other, but also in hurting each other, as most parents will have observed. I shall discuss reports on children's "mixed feelings", as studied by Judy Dunn, Paul Harris, and others, and on how neglected or abused toddlers may sometimes engage in abusing others.
This brings us to the "dark" side of human nature, the converse of the inherent capacities entailed by thesis, or rather, the way in which the dialogic and dyadic self-organization attributed to the mind in term of (q) may convert into distorted forms:
(p) Distortions may bring about a periodic or more permanent collapse of the self-organizing dialogue of the mind, ruling out any complementary perspective, or even excluding the other, virtual and actual, from felt immediacy.

In the extreme case the capacity to feel with the other, actual or virtual, can be distorted in a way that generates a separate self in divorce from the virtual other.
In chapter 10 I turn to the way in which some subjects in Milgram's electric chair experiments appear to suffer periodic collapse during the experiment. First, they submit to the exclusive perspective of the experimenter, ruling out their own in the process. Second, they exclude the victim from felt immediacy, including instead the experimenter in their immediate companion space. When, however, the victim is brought into closer immediacy, or the immediate closeness to the experimenter is reduced, the proportion of subjects administering shock to the limit is drastically reduced.
This chapter will otherwise be devoted to the shattering studies by Christie and Lifton of Nazi war criminals. Lifton describes doctors who discard the Hippocratic oath and commit genocide. Such atrocities committed for the sake of the cause can hardly be considered merely in terms of aggression instincts, as Lorenz' points about aggression in man and animal lend itself to be interpreted.6 Mankind repeatedly shows through its history how a monolithic perspective - religious, moral or economic-political - co-constructed in the culture and internalized in the individual born to the culture, can mobilize and give coherence to destruction and atrocities, even genocide. This relates to a point made by the brother of one of those responsible for the Red Khmer mass murders. In his childhood and youth this man was a kind boy, recounts the brother, "he could not even bring himself to kill a hen for their dinner meal." In his analysis of totalitarian logic, Herbst (chapter 9) has shown how a dictum, such as 'Let there be a world free of evil!', can be lethal when embedded in an Aristotelian structure of two-valued logic that defines the enemy or the oppressors as evil and beyond salvation. Submitting to a monolithic perspective, the persons committing such crimes transform themselves into instruments of destruction, collapsing in the process the inherent qualities that makes them human and alive. Distorting their inherent capacity to feel the lifeform of the actual other the Nazi doctors developed what Lifton calls the Auschwitz self. Defying comprehension, genocide compels consideration in the hope, perhaps futile, that awareness of some of the possible mechanisms behind it, unique to the species of man, may reduce its probability of recurrence.
This dark chapter ends Part Two.
Part Three is devoted to some clinical and cognitive science implications of the thesis (q) and its converse (p).
In chapter 11 will be considered the ways in which the childhood victim of abuse may try to cope with the injuries suffered through developing multiple personalities. In terms of (p) this involves a divorce from the virtual other, which then runs its own course and give rise one or several separate selves. With reference to cases I shall venture some reflections on possible paths of recovery. Some clinicians commenting their own practice will be cited will be cited in the light of the thesis.
Its conception was prompted by a problem that have arisen in connection with ideas on self-organizing systems.7 The question is this: How can two self-organizing systems, A and B, be seen to constitute a self-organizing dyad AB without loss of their respective identifying organization? I had been grappling with this problem for a long time until a winter morning (in February 1986) when I woke up with the solution in mind:8 The scenery is a harbour with a large warehouse. I am standing on the dock. Two container ships come front-end to the dock. The container wagons of the ships rumble along on parallel rails that run from the ships over the wharf and into the warehouse. Standing outside the warehouse, I am trying to peer in through the closed side-door of the warehouse at the container-wagons inside. It is relatively dark inside, but I manage to see the wagons. They stand at rest in parallel, on their respective rails, each with its own load enclosed in the container. While standing there outside and peering in, I felt that if some of the bulk in one of the containers somehow could be shifted and replaced in the other container, then the problem would be solved.
Given that the respective organization of both A and B is already dyadic, involving a virtual other, then this very same organization can recreate itself without any qualitative jump, when each, as an actual other, replaces the virtual other. The dialogue continues, irrespective of whether it recreates itself in the mind's dialogue with itself, that is with the virtual other, or in dialogue with the actual other.
A Gordon Research Conference turned out to be the ideal forum for the first introduction of the thesis.9
In addition to researchers on child development and Piagetian constructivism, the participants10Heinz von Foerster, Ernst von Glasersfeld, Lynn Hoffmann, Oliver Selfridge, and Hermina Sinclair. I am grateful to all of them, on that or on later occasions, for stimulating comments and valuable references, and also to many of the other participants. comprised people, some of them pioneers, in such diverse fields as cybernetics of self-organizing systems, neurocomputational simulation of perception, and family therapy. These fields have generated questions that invite consideration in the light of the thesis. Examples are these:

. In view of their different frames and the time delays in each of them, how is it that two participants in intimate conversation sometimes can exhibit a perfect, or near-perfect, coordination of gestures and movements? (Chapter 14, connecting with chapter 3).

. What is the nature of the process of transference that may occur during therapeutic conversation, and how is it that a user in conversation with a computer program sometimes may feel himself being understood by the program? (Chapter 11 - 12, connecting with chapter 4).

. What may be involved in perceptual reversals upon exposure to ambiguous figures such as Rubin's vase or the Necker cube, and how is that in neodissociation experiments a "hidden observer" appears to be disclosed in some, but not all, subjects submitted to hypnotic suggestions? (Chapter 13).

Thus, the remaining chapters concern cognition, perception, and interbodily coordination. I shall consider the ways in which understanding is approached in cognitive science, and offer an explanation of the so-called Eliza-effect found by Weizenbaum in some computational contexts. Some "neurocomputational" simulation explorations on visual ambiguity in terms of the dyadic conception of the mind's operation will be reported from. The issue of divided or unified consciousness will be considered with reference to Sperry's split brain studies, Neisser's studies of attention, and Hilgard's neodissocation experiments.
In the final chapter interpersonal coordination will be accounted for in terms of the thesis. Inviting disconfirmation I shall venture this qualitative prediction: If persons can be found to be synchronous with themselves, then one should be able to register at least some occurrences of interpersonal synchrony. This brings me out on thin ice in a double respect. First, original findings of so-called "interactional synchrony" have been difficult to replicate. Second, even though its domain pertains to observers, not participants, relativity denies that two persons can be simultaneous due to their different frames. Organizing this book I have arrested my inclination to begin in the abstract with the history of ideas and then descend to the concrete level of phenomena purported to be accounted for. At the same time I have tried to avoid the fallacy of what Whitehead terms "misplaced concreteness". In this I may have failed. What we capture by way of scientific data reflect our own co-constructions and mediations in the language at our disposal.
The lived moments of that for which there is no single English word 11 and which afterwards, through mediation, is turned into experience, elude scrutiny and elicit wonder in the sensitive adult. Our immediate feelings and fleeting thoughts are transformed as they are turned into objects of reflection and become experience. For example, only when there is interruption of a lived moment of what Stern terms 'affect attunement' does something emerge as an object for reflection. That which is experienced is the loss and absence of the immediate and shared quality involved in affect attunement. The attempts in this book to indicate such phenomena of felt immediacy in the mediational language of science involve a paradox.


1The term 'the virtual other' is new. The general notion of a virtual entity or process is not a new one. In physics one speaks of virtual particles. In computational contexts we may write a program with virtual properties, for example a software-implemented virtual machine which in certain respects may operate as an actual hardware machine, or a virtual computer memory that may have some of the efficiency of an actual hardware computer memory. Thus, as defined by Charles Sanders Peirce, a virtual X, is something, not X, which has the efficiency (virtus) of X. In terms of an identified operation, this means that an actual X may replace or be replaced by a virtual, *X, without change in the identifying characteristics of the operation. In this respect both differ from a symbolic representation 'X', albeit this distinction is not always adhered to in computer science.
Consistent with the above language use, when I qualify the other (alter) as a virtual, marking with a star, *, it means that the virtual alter *A may replace or be replaced by some actual alter, A, without changing the form of the (proto)dialogical inter-operations that occur in "the communion space" of the virtual alter. Such inter-operations give rise to the subsequent formation in this "space" of re-presented or prototypical altera, 'A1', 'A2',...'An' on the basis of interactional experiences. The term "space" is here used metaphorically, as will be expressions about being "(ful)filled" by the actual other who replaces the virtual other, indicating the quality of feelings in such lived moments.
2S. Bråten (1988): Dialogic Mind: the Infant and the Adult in Protoconversation. In: M. Carvallo (ed.): Nature, Cognition and Systems I, Dordrecht: Kluwer Academic Publ., 187-205.
3C. Trevarthen (1988): Infants trying to talk. How a child invites communication from the human world. In: R. Söderbergh (ed.): Children's creative communication. Lund: Lund University Press, 1988, pp.9-31.
4L. Murray (1991): Intersubjectivity, Object Relations Theory and Empirical Evidence from Mother-Infant Interactions. Infant Mental Health Journal, Autumn 1991. Lynne Murray considers the present thesis to offer a bridge between development psychology and object relations theory:
5I shall later replace the term "communion" with the more technical term "dialogic closure" applying to processes evoked in these primordial domains. It will be defined in chapter 3. Until then the term "communion" may suffice (Cf. the root of and use of commune (Hold intimate intercourse (with person, one's own heart, together) (The Concise Oxford Dictionary, Oxford University Press, 1964, p.244)). Various self-organizing systems notions of operational closure will be referred to. Examples are Piaget's notion of auto-regulation (autoréglage) applied to mental structures, and the principle of autopoiesis for self-production; self-creation) applied by Maturana and Varela to living systems, and by Luhmann to social systems. Dialogic closure will be defined in a related sense, albeit qualified as reciprocal and dyadic, not ego-centric or monadic.
6Konrad Lorenz: Den såkalte ondskap. (Das sogenannte Böse). Cappelen, Oslo 1968.
7Ref to meeting with Pask and discussions with Luhmann and Maturana.....S. Bråten (1982): Simulation and Self-organization of Mind, In: Contemporary Philosophy Vol. II: Philosophy of Science (ed. by G. Flöistad) The Hague: Martinus Nijhoff, pp. 189-218.
8The idea of the inborn companion space of the virtual other actually presented itself by way of this dream:
9Bråten, S. (1986): Consent and Dissent - Crossing Boundaries during Dialogue. Invited talk at the Gordon Research Conference on Cybernetics of Cognition, June 1986, organized by Heinz von Foerster and Ernst von Glasersfeld. A revised and extended version has appeared in M. Campanella (ed.): Between Rationality and Cognition. Torino: Albert Meynier, 1988, pp. 205-236, with contributions also by Hayward Alker Jr., Karl Deutsch, Klaus Krippendorff, Herbert Simon and others.
10Among those present were these:
11The lack of a single term in English for such lived moments presents a continuous problem to the English text of this book. Lived moments in an immediate sense is termed "Erlebnis" in German, and "opplevelse" in Norwegian, in contradistinction to experience (termed Erfahrung in German and erfaring in Norwegian). In contrast to experience, entailing something brought 'out' (ex-) or 'forth' or before the senses as an object of perception or reflection, Daniel Stern (personal communication) has suggested the term "lived moment". In the text I shall mostly adhere to the complementary use of the qualifying terms "immediate" and "mediate", or "immediacy" and "mediacy", to differentiate between domains of directly lived moments and domains of reflective experience.