The origins of attachment theory

 It is in the monumental trilogy “Attachment and loss” (Bowlby, 1969, 1973, 1980) that we find the point of arrival of a long process concerning the effects of the adequacy of maternal care on the development of the individual. At the origin of attachment theory in fact, there are Bowlby’s earliest formulations based on the linear model of the construction of development, but over the years numerous variations and enrichments have come from the many and varied areas of interest on the question of attachment and relational development. These contributions have a clear intention: to enrich and to give a better understanding of the shared assumptions, to answer the unanswered questions and to investigate the unexplored areas related to the vast, complex question of attachment.
Alongside the classic studies on the dyadic caregiver-child relationship, we therefore find studies on the assessment of attachment styles and on their intergenerational transmission, studies on the the attachment bond in the couple and more recently in the whole family group, and also reflections on the epistemological relation between the intersubjective perspective and the theory of attachment. Another important area of study is found in the question of the versions of this theory in psychotherapeutic treatment, as well as in the construction of mental representations and narratives about attachment experiences.
Working backwards, we can identify in the work of Bowlby (1973) the definition of the term attachment, a concept used by the author to refer to the capacity of the parent figure to support the child’s feelings of security. By extension it consists of the condition in which an individual is emotively tied to another person, generally perceived as older, stronger and wiser; in other words it is the present state of an individual’s attachments, which can be divided into secure and insecure (Holmes, 1993).
Proof of this bond, as Holmes (1993) underlines, is constituted by the pursuit of closeness which for Bowlby coincides with the primary social drive, by the phenomenon of the secure base, expressed as a set of dyadic interactions between the child and the caregiver (corresponding to the background of security described by Sandler in the psychoanalytic model), and by the protest against separation. In the psychological growth model that Bowlby describes, there are numerous possible paths that an individual can take during infancy depending on the interactions with the caregiver and it is in the interactions repeated in infancy with the attachment figure that, from adolescence onwards and following positive attachment experiences, the representation of the secure base is formed.
Unlike the Freudian model, founded on the homuncular theory of development, that is, the theory that development follows fixed, pre-established phases, and that follows the aetiological model,  Bowlby prefers the epigenetic theory (Waddington, 1977), which holds that there are different paths of possible development depending on the environmental conditions. This approach allows for a more refined explanation of the interactions between an individual and his environment and also makes it possible to conceive development as work in progress, free from fixed patterns but tied to an epigentic compromise between attachment needs and the figure capable of sustaining these needs. Although the child has a natural propensity to form attachments, a propensity based on the neotenic nature of the human species which means that thought originates in the relational nourishment received and in the family relational patterns affecting the individual, the nature and the dynamics of the attachments will depend on the interaction between the organism and his environment.
Unlike the psychoanalytical development model, moreover, Bowlby states that attachment patterns last a lifetime, are flexible and if they are insecure they can be changed by good experiences.
The separation or threat of separation from the attachment figure triggers ‘attachment behavior’, which is the behavior that appears in a subject that seeks closeness with another different or favored individual.
At the basis of attachment and of attachment behavior there is therefore the ‘behavioral attachment system’, which includes the resiprocal behavior pattern shown by those who seek and by those who give care and of which both the subjects involved are aware. It is therefore made up of a model of the world containing the representation of the Self, significant others and their interrelations. Closely connected to this is also the ‘explorative behavioral system’, which Ainsworth (1967) identifies in the attachment figure’s willingness to provide a secure base for exploration.
Central to the theory is also the concept of Internal Working Models (IWM). Based on Beck’s cognitive theory, in fact, Bowlby argues that individuals need a map of the world so as to control and manipulate the environment, and this occurs through two different models: an ‘environmental’ model, which gives information on the things of the world, and an ‘organismic’ one, which gives information on oneself in relation to the world. Repeating Craik’s argument that individuals interiorize a relationship with themselves, others, and the relation between these two poles, and take into their mind a small scale model of the external reality (Craik, 1943), Bowlby (1973) states that the IWM are mental models that guide and regulate human functioning, that is, the reference structures that retain the temporal and spatial configuration of the real world.
The child’s repeated interactions with the external world, therefore, lead to the structuring of Working Models of the predictable behavior of significant figures, of oneself and of the interaction between these different behaviors. (Bowlby, 1973). It is therefore the presence/absence of the attachment figures that generates confidence or lack of confidence in his availability and it is on this ontogenesis that the key characteristics of the IWM are based, permitting the child to orient himself amidst relationships, knowing who the attachment figures are, where they can be found and how they will react to requests for care and comfort.
Similarly, Horowitz (1987) talks about ‘role-relationship models’ and ‘Self-Other patterns’, or representational models used by the child to predict the world and place himself in relation to it, while Stern (1985) outlines the concept of ‘generalized event representations’ to explain the nuclear intersubjectivity that substantiates human nature right from birth. While in neuroscience the discovery of mirror neurones is at the basis of intersubjectivity, Stern affirms the precocious existence of internal affective states that develop within the intersubjective matrix in which the subject grows, states that then develop thanks to experiences of interaction, and recognizes, from the second month of life, the growth of affective consonance, which is the cross-modal equivalent of interpersonal affective experiences on which the concept of empathy in nuce is based (Gallese, 2003).
Intersubjectivity and attachment therefore possess the same ontogenesis and as primary motivational systems they support each other, the first by creating the conditions necessary for attachment and the latter by favoring the development of intersubjectivity through closeness to significant people (Lavelli, 2007).
Alongside a cognitive aspect of the IWM, there is their affective nature, according to which the map is constructed starting from the generalization of relational experiences between the subject and the agents of maternal care (caregivers),and which feels the need to defend itself from loss and from painful needs. Therefore while a secure child stores an Internal Working Model of a sensitive, loving, reliable caregiver and of a Self worthy of love and attention and lets such assumptions influence his own relationships, this includes the risk of also internalizing a distorted and incoherent map of the world characterizing the insecure attachment style, in which the world is seen as a dangerous place, in which others must be treated cautiously and the Self does not deserve love. Such assumptions are, moreover, persistent and difficult to change based on later experiences.
It is therefore through the IWM that childhood attachment patterns are transposed into adult life and are transmitted to the following generations.
There have been numerous advances in the last decades compared to Bowlby’s original concepts about  attachment and the construction of the IWM, theoretical advances that Fonagy (2001), using the concept of ‘reflective function’, sums up in a sequence of representational systems based on expectations about the interactive characteristics of the first caregivers (created during the first year of life and later elaborated), about the representations of events and the autobiographical memories to which the subject connects his attachment experiences, and about the capacity to understand other people’s psychological characteristics and distinguish them from his own (Trevarthen, in 1993, connects this capacity to a healthy, effective emotive communication in which the subject lets the other person’s state of mind influence his own).
According to Bretherton (1991), IWM are the products of a series of operations of storing, retrieval and construction of information, in that the concept of Working Model on the one hand consists of the models organized in the long term memory, on the other of the operative memory, since such models are responsible for understanding new situations confronting the individual. The models of operative memory are therefore revised on the basis of the representations of people and things and of the relations contained in basic long-term individual knowledge. This capacity has an adaptational sense and the degree to which the models adequately simulate the relations with the outside world and create IWM of themselves that include their own and others’ representational processes, underlies the main developmental characteristic of the human species. Bretherton also explains the working of the long-term memory in the construction of IWM with the activation of scripts, first described by Schank and Abelson in 1977, and seen as structures of representations organized in sequences that include patterns of actions, relations and emotions, including the contexts in which such processes take place; this is similar to what was stated by Byng-Hall (1995), who says the scripts are representations of multipersonal interactions and they are triggered within the family. It is in fact possible, according to the author, to conceptualize the existence of generalized schemas concerning attachment experiences, so in this perspective the IWM become mental representations composed of systems of schemas, including procedural schemas, organized in a hierarchy. Negative attachment experiences will therefore determine internal relational models that are not very consistent, being victims of defensive processes that distort information, and therefore leading to the construction of insecure attachment patterns, that are particularly hard to change.

The intersubjective basis of the attachment bond

As has already been said, intersubjectivity and attachment possess the same ontogenesis and as primary motivational systems they support each other, the first by creating the conditions needed for attachment and the second by favoring the development of intersubjectivity through the closeness to significant persons (Lavelli, 2007).
This fact is confirmed in the thought of Siegel (2001), who believes that at the level of the mind, attachment relations help the child’s immature brain by exchanging emotions with the parent. In order for a secure attachment relationship to be formed, these exchanges must be characterized by the adult’s capacity to react promptly and adequately to the signals transmitted by the child, providing responses that favor the production of positive emotional states and that facilitate the control of negative states.
Similarly, Gallese (2003) argues that the study of interpersonal relations, which are based on the attachment experience, shows the social nature of the human mind. At the basis of intersubjectivity he places the construction of a primitive Self-Other space (we-centric space) marked by a paradoxical form of intersubjectivity with no subject. More specifically, the author states that at the beginning of life we live in and share with others a multidimensional interpersonal space, which also occupies in adulthood a consistent part of the social semantic space and that he links to the modelling of objects and events (based firstly on the action of mirror neurones) with which the subject interacts and from which he draws his knowledge of the world.
As well as the organization of experiences, attachment relations, according to Siegel, influence the development of neuronal circuits and have a direct effect on the cerebral activities that mediate the fundamental mental processes (memory, autobiographical narrative, emotions, representations and states of mind), so the presence of insecure attachments will lead to a significant risk of a later manifestation of psychopathological conditions. It is from the exchange and co-regulation of cerebral signals that it is possible, in the author’s view, to observe the development of the affective syntonization described by Stern in 1985. Siegel also points out that the development of the mental model of attachment depends on the activity of the implicit memory (procedural memory that reproduces the already known aspect of the subject’s experience), which allows the subject to foresee the future characteristics of the attachment relationship. The importance of his contribution lies in the observation that attachment experiences participate in the increased capacity for autonoetic knowledge from the earliest years of the subject’s childhood.
The states of mind experienced by the child enter the emotional memory from the first year of age, and are therefore learnt at an implicit level, becoming a stable part of the subject’s brain. The central distinction is between implicit memory (procedural, not associated to subjective inner experience but involved in the creation of mental models) and explicit memory (associated to subjective inner experience and in the case of autobiographical memory to a sense of self through time) which in turn includes episodic or autobiographical memory and semantic memory. The latter is to be seen as a set of prepositional representations that do not evoke the mental journey in time and that are therefore to be linked to noetic knowledge; the implicit memory, on the other hand, is seen as being grounded on autonoetic processes (i.e. knowledge of oneself, a concept that Napolitani linked to the lived experience or erlebnis), mediated by the frontal lobes, which explains the capacity to have a representation of oneself in the present reality and in an imaginary future and which seems to depend on dialogue and on interpersonal relations between child and caregiver, in confirmation of the context-dependent nature of the explicit memory, whose activity is mediated by the hippocampus, the zone responsible for constructing cognitive maps of experience.
Moreover, as Seigel tells us, Wheeler (1997) puts forward a distinction between episodic and autobiographical memory, which seem to be subject to two separate processes, the first implying a sense of self through time, the second lacking this sense. Models of implicit memory, therefore, enable the elements of the explicit memory to be perceived and filtered and it is in turn possible to recognize the influences exerted by implicit memories, also at a non-verbal level, on behavior and on narratives.
The narrative capacity, in fact, plays the fundamental role of leading to the creation of stories that can be shared which in turn leads to particular forms of behavior that can influence the inner world and have effects on the modulation of emotions and on the organization of the Self (Siegel, 1999). This then reflects the social interaction underlying human experience and the structure itself of the individual narrative processes associated to the early experiences of attachment, which in turn are associated to the development of specific modes of autobiographical narration.
Bretherton (1987) also believes that the construction of IWM can be found in the linguistic-narrative approach, similarly to Simonelli (2006) who, in the communicative and narrative styles of interactional experience, underlines the adult’s capacity for narrating his own attachment story. It is here then that one finds the link between attachment behavior and its representation, in that it is precisely in the representational correlates of attachment in the adult that the individual’s inner world takes shape with respect to relationships, by means of narrative communication and of interaction between the two systems of implicit and explicit memory.
The IWM of attachment, therefore, will be established initially as a synthesis of the implicit memories of generalized attachment interactions, and they will later become semantic memories that can be and are formulated verbally.

Attachment and family relations: the encounter with group-analysis

To sum up what has been said so far, we can conceive attachment theory as going beyond the dyadic mother-infant or caregiver-infant relationship, and find in it a complex space in which the various figures making up the family and other significant figures have a specific place in the subject’s development and configuration of attachment and of his relational skills.
In considering the advances in attachment theory, linked to the attempt at validation of Bowlby’s early formulations starting from Strange Situation Procedure (SSP), Main (1999) underlines the need to do research into the relation between the biological, cultural, ecological and family context and attachment theory, so as to go beyond Bowlby’s monotropism. The author in fact points out the presence of non identical reactions, and therefore of non identical attachment styles, between the infant and the two parents in the Strange Situation, concluding that the different quality of the two attachment relationships is the outcome of two separate experiences occurring between infant and mother and between infant and father.
It is therefore in the concept of monotropism that the aetiological value of the exclusive attachment of the infant with his main carer, usually the mother, lies. Impressed by Lorenz’s studies with geese, which led the ethologist to define the concept of imprinting, Bowlby similarly hypothesizes that there is a parallel process in the infant’s development.
The first objections to such an assumption come from the observation that attachments develop gradually and over a long time span, far longer than the ‘sensitive period’ that Bowlby indicates as the first year of life, and that human attachment may be considered hierarchical, insofar as it depends on a list of carers providing maternal care that extends beyond the parental figures alone, to include grandparents, siblings, uncles and aunts, teachers and other significant figures. This data undermines the idea of the existence of a single attachment bond, leading to the study of the role of ‘multiple attachments’, and therefore to the capacity to create attachment bonds with various affective figures.
Underlying this propensity to interact with two or more people is presumably the basic social motivation that Fivaz-Depeursinge and Corboz-Warney (1999) identify in ‘triangular competence’. This at least triadic relational matrix is already present at three months of age and seems to be at the basis of the development of family alliances.
Although supported by data on the developmental and genetic bases of monotropism (Attili, 2007), one cannot rule out the presence of the child’s neotenic propensity to interact with other significant figures, who will influence the formation of IWM concerning attachment and will contribute to the subject’s adaptation to the physical and social environment.
This shift in viewpoint then becomes possible and finds its foundation in the model provided by Hinde and Stevenson-Hinde (1991). These authors draw attention to the fact that individuals’ characteristics influence the quality of their relations. What seems important is however the role attributed to the complex relational matrix which at the same time contains the relations between groups of subjects in interaction; this is in line, from our point of view, with what is said in group-analysis about the construction of interactive and communicational networks based on the transpersonal nature of the individual mind.
As Marvin and Stewart (1999) argue, for an understanding of attachment in the network of family relations, it is necessary to depict, within the family, the different representations of care and attachment, a process that can be summed up by the authors in the concept of ‘shared Internal Working Models’.
The interest in the influence of attachment within the family constellation therefore seems to lead beyond the theoretical dualism between the real family and the represented family, i.e. the product of mental representations. This is in line with the early attempts, first made by Wynne in 1984, to bring attachment into the epigenesis of relational systems.
Byng-Hall (1995) attributes the delay in exploring attachment theory within family relations to Bowlby’s early decision to concentrate exclusively on the study of the dyadic aspects of the child-caregiver relationship, because of the same complexity that Bowlby felt lay in the family group, a complexity that seemed to make the family an inexplorable place for scientific research.
It was with the studies of Stevenson-Hinde (1990), Byng-Hall & Stevenson-Hinde (1991), Donley (1993) and Doane & Diamond (1994), that family therapists shifted attachment theory from the dyad to the whole family group, an approach explained by Minuchin (1995) in the light of the bi-directional relations between the various members of the family context. Donley (1993) considers attachment a process going beyond the mother-infant dyad, a dyad that the author describes within a broader emotional unit: the family, a dynamic and emotional setting and place for relating.
The definition of these relations, which are similar to the systemic theory, in our opinion is to be found equally in the group-analytic approach, which states that the supremacy of relations holds the key to understanding the existence of an inner relational world determined by the intentional logic of a network that goes beyond the single individual and that offers an image of an individual composed of a mosaic of puzzles arising from the context.
This image sums up the interaction between the subject, the subjectivity and the family and social context of which the subject himself is part (Foulkes, 1973) and which is the foundation of the transpersonal nature of the human species. The natural centrality in the subject of the tendency to regard his own relation with the Other person as fundamental, in fact, lays the basis, in group-analytic discourse, for an understanding of the psychic identity that Napolitani (1987) positions in the ‘idem’, a process in which the human being makes himself identical and identifies with others’ intentional qualities, sinking his own subjective component into an intentionally affective fabric from which to distinguish himself through his reflexive and symbolic capacities, always within the psychic implant grounded by and in the culture.
It is to Byng-Hall & Stevenson-Hinde (1991) that we owe the definition of ‘family script’ as a set of models of work shared by the family members, while in quite a similar way, Stern (1995) talks about multiple family interactions represented both in the scripts (or ‘patterns of being with’ shared within the family context),and in family myths, legends, stories and romances.
In order to illustrate the expectations shared by the family about respecting family roles, Byng-Hall (1995) therefore proposes the term ‘family script’, which the author bases on the acceptance of a fundamental, high priority rule: care. He believes in fact that a script of family attachment includes the interaction between all the attachment relations, entailing on the one hand the distribution of roles among caregivers and those who seek care, and on the other including the mental representation that each member has of family functioning. This therefore represents the tendency to link manifest family interactions to configurations of internal relations, which can be found in the representations of the family group; it also leads to an understanding of the bond existing between narrative style, behavioral patterns and dysfunctions present in the family, according to the typically systemic assumption that every relationship has an influence on all the others.
The study of attachment relations in the family therefore gives fundamental importance to the analysis of the modalities of storing the procedural aspects of family relations, which Emde (1991) finds in complex habits, daily rules, turn-taking, and everyday interpersonal exchanges that take place in the complex family context.
This seems closely connected to the fundamental concept of ‘secure base of the family’, underlying the shared awareness that attachment needs must be protected in the network of family relations. The secure family base is in fact seen as the family that provides adequate attachment figures that make all the members of the family sufficiently secure and also implies the idea of a shared family responsibility that assures help, and therefore security, to every family member.
Among the factors that affect security, Byng-Hall (1995) underlines the loss or the threatened loss of an attachment figure (in line with Bowlby’s first formulations), the competition to take care of another member of the family, the turning to inadequate attachment figures, the repetition of losses experienced in the previous generations and enacted, through excessive precautions, by the present parents, the presence of abuse in the family and the existence of conflicts, like power struggles or distance conflicts, which interrupt the experience of care-giving and the explorative capacity. This is expressed, according to Byng-Hall (1999), in the search for Co-working Models of collaboration between the family members when providing care, in the priority of the need to offer care  and in belonging to the family.
It seems that the very concept of belonging to the family can be linked to the construction of social identity relations, the guarantee for individuals of the capacity to predict the consequences of others’ behavior (Gallese, 2003) by attributing an identity status that reduces the differences in a shared belonging that assures continuity, also through the creation of what Reiss (1989) calls group memory. This is the memory that each member of the family has and that is created in the concrete encounter between all the family members, who collaborate in the construction of a shared memory.
The family is therefore to be regarded, in line with group-analytic thought, as a place where multidimensional models are learnt, and family thinking in such a context is seen as the mediator between the passing of generations and between the group nature of the family and the complex reality of the world.  The family can therefore be defined as the family matrix that provides the child with the crucial meanings for the constitution of a stable image of the Self, and its mobility makes the family member capable of transforming these meanings into new interpretations of the world (Lo Verso, 1994). As Foulkes (1973) says, the child is determined by the relational context, that is, the intrapsychic relational equilibrium is always tied to that of interpersonal relations.
 In the field of this broad and multifaceted relational panorama there are other studies on the development cycle of the attachment bond, based on the assumption that the latter undergoes constant processes of reorganization and transformation throughout the subject’s life. The development cycle of attachment, according to Carli (1999), consists of a succession of dyadic relationships which, from asymmetrical in childhood, become symmetrical in adulthood, in the relationship with the partner, and go back to being asymmetrical in maturity due to the inversion of the roles in the parent-child relationship typical of late adulthood.
Considering the family as a single relational entity constantly relating with the outside world, one can also apply the double classification of family organization provided by Minuchin (1974) connected to the two types of insecure infantile attachment described by Ainsworth: enmeshed families, with a prevailing anxious-ambivalent pattern, and disengaged families, in which an avoiding attachment style dominates.
In the same way Stevenson-Hinde (1990) distinguishes between adaptive, supportive and autonomous family functioning, corresponding to the secure attachment style, and forms of functioning based respectively on avoiding, ambivalent and disorganized styles. The first is characterized by a distancing and insensitive functioning that rejects family members, the second has an enmeshed and over involved family functioning while the third, which can be called chaotic, reveals a family functioning characterized by the lack of structure and by abuse.
These family modalities based on the development of insecure attachment relations are therefore in contrast with family groups featuring supportiveness and trust, which enable each person’s potentialities to be developed; this assumption, though risking over-classificationism, is not far removed from the conception of Foulkes’ concept of the unsaturated and saturated matrix. The unsaturated matrix actually includes the possibility of creativity and an opening towards the new as well as a revisitation of the mythological dimension of the family, aspects echoing both the function played by narration as a system of transmission of relations and the concept of reflective function; in contrast, the saturated matrix implies the impossibility of creativity and a block on the emergence of individual thought, which seems to recall the impossibility of generating emotions and affects concerning relational experiences that is found in distancing attachment and in the enmeshing family which saturates the spaces of security and which is found in the anxious style.
The aim of group-analysis, as we have already said, is therefore to place psychic functioning in a social, developmental context, affirming the interpersonal conception of the human being in relation to others (Marrone, 1999), very close to what Bowlby states regarding the child’s inner representations. The founder of attachment theory in fact wrote in 1988 that to understand infantile mental representations and the relationship between the child and his parents in the actual reality, it is necessary to unite the inner and outer worlds, in line with what Foulkes wrote in 1975, saying that object relational dynamics are processes that originate in a multiperson network of interactions, represented first of all by the primary family.
We can therefore say, in agreement with the thought of Marrone (1999), that group-analysis and attachment theory have essential principles in common and that the fundamental link between these two models of studying and understanding the social nature of the subject and of the relational structure of identity lies in the concepts of the social matrix of psychic life and of De Marè’s Koinonia (which is dialogue in communion, sharing and fusion).

Clinical relapses: attachment theory in group-analytic psychotherapy

In Holmes’s view, the construction of the attachment system is a creation of the psychological immune system in which the subject’s defences are weakened, also towards the perception of the failure or absence of an internalized secure base, and the aim of psychotherapy becomes that of increasing the awareness of the individual’s mental life, of immunizing him by strengthening the narrative powers which are the psychological equivalent of the immune component (Holmes, 2001).
Probably because of the changes that took place in the transition from traditional cultures, in which the secure base is provided by the family and the tribal group, to the modern individualistic culture, dominated by a nuclear and subnuclear family model, meaning that the secure base has to be sought inside oneself, psychotherapy seems to be the only chance of developing a ‘secure internal base’, or Internal Working Model of security (Holmes, 2001).
Seen in this way, however, the function of psychotherapy seems to rest on empowering the individual and his resources with the risk of isolating him from the transgenerational and transpersonal context underlying psychic life, forgetting that in traditional societies the secure base already belonged to a group context and to group relations that generated security.
It therefore seems that a recuperation of the relational system of group life, typical of traditional cultures, might help us to understand the social nature of the human species, one of the first manifestations of which is attachment and intersubjectivity; it might also bring home the value that group psychotherapy, with its therapeutic and support features, attaches to the construction through therapy of a secure base shared and supported by the group itself, in which the figure of the leader becomes central but above all, in which the members of the group will often be seen as brothers and sisters.
Proceeding with the analogy with traditional cultures, we can in fact find in the figure of the leader, the group’s first patient according to Foulkes (1975), the first attachment figure in a hierarchical system of care-giving, in the same way that in tribes a top role is always given to the mother, the main carer, but without making the mistake of considering only the dyadic therapist-patient relationship. What we are saying is that in the networkof relations in the group setting, the relationships are one to many and many to many and that each person’s attachment bond will therefore determine his way of relating to the group, including the leader.  At the same time, we believe that the configurations of group attachment are created as a network (due to the group’s creative matrix, its affiliative behavior and its cohesion) and these bonds will be affected by the life of the group itself and will be transformed, once a group culture has been established, into secure forms of attachment. This process, which Ondarza Linares (2004) calls “consensual convalidation of Internal Working Models”, is therefore based on the interplay of the IWM of the subjects in the group network and on the construction of transpersonal relational models.
Seen in this way, group-analytic psychotherapy presents itself as an intermediate space between the individual Self and the group Self, a consideration that we think brings this model close to the relationship-centred view of attachment theory. In historical terms, in fact, it is in the evolution achieved by the psychoanalytical apparatus in leaving the mechanistic drive oriented monad and shifting towards the relationship that the closeness between attachment theory and group-analysis is based, the first being positioned on an interpersonal level, the second on a transpersonal level.
In psychotherapy it therefore seems useful, due to the complex nature of the individual, to consider all the relational dynamics and the defensive strategies enacted by the subject and brought to light by a clinical application of attachment theory, since the motivational system of attachment is one of the systems involved in the construction of a person’s mind and his relational capacities (Muscetta, 2006).
Bowlby (1988) himself believes that the three main forms of analytically oriented psychotherapy (individual, family and group) are influenced by the theory of personality development and psychopathology built on attachment theory. In particular he observes that the therapist’s aim is to help the patient explore the representational models of himself and of his attachment figures and to do this by providing a secure base, encouraging his explorations of the way he forms relations with significant people and of his own feelings and perceptions of attachment experiences, analyzing transference relations and leading him towards awareness of the models of self and others deriving from childhood.
With the application of the Adult Attachment Interview (George, Main & Kaplan, 1994) in therapy it is observed, however, that the mere fact that the therapist offers the patient a secure base does not explain the influence attachment theory exerts over therapeutic practice.
In the same way, the evidence that the mirroring of mutually regulated interactions is not enough to explain the success of psychotherapy (Dozier; Cue & Barnett, 1994) leads us to consider the importance of the therapist’s role in terms of his capacity for empathy and responsiveness (which is the capacity of a reflective function), both as regards attachment and separation, as well as the value of syntonization in setting up a creative encounter that can extend the endopsychic field  (Rycroft, 1985) and also substantiate the creation of a good therapeutic alliance (Speranza, 2006). According to Holmes (2001), this will lead to the revival of the subject’s narrative capacity, which is awareness of self, of significant others and of the relationships with them, and therefore to a new integration of the Self, thanks to the therapist’s capacity, and also that of the group in the group setting, to tolerate and support the crisis. 
In the group setting this comes about both through free flowing communication (Foulkes, 1973) and through the creation of support factors and therapeutic factors proper. Specifically, support factors, identified mainly in cohesion, belonging, sharing, regressive modification of defence mechanisms, and empathy, and the analytical therapeutic factors, which include mirroring, resonance, corrective emotional experience, training the Ego in action, identification of the Self, and socialization (Pisani, 2000), seem to enable the creation of the secure base, indispensable for the structuring of the narrative capacity and for fortifying the group matrix containing it.
The construction of narratives, which underlies the whole psychotherapy apparatus, enables events and stories to be restructured, permitting relations and bonds to be established. This is linked to the transformation and the passage, described by Napolitani (1987), from the happening, in the sense of any phenomenon that has not yet found space in the subject’s mental field and that flows before the consciousness without finding space in signification, in invento (from the Latin in-venire, find something inside), which is a completed process of genuine subjective discovery on the part of the subject. It is through the process of symbolopoiesis (inner mental symbolization that gives everyday happenings a subjective meaning and existence) that the happening can be transformed into invento, when the mental field of thought is organized so as to enable new happenings to be received and elaborated (Di Maria & Lo Verso, 2002), and then narrated and used as a source of relations and bonds.
The narrative plots outlined are connected to the unfolding of changes and transformations inherent to the subjects and their identity and envisage the psychological implication that the knowledge, observation and reconstruction of the question of truth is the construction and reconstruction of the patient’s story, which becomes a text constructed within the relationship itself (Montesarchio, 2002). In narrative theory the Ego is therefore an autobiographical, polyphonic Self, and that is a tale of the multiple modes and multiple voices that interact to create a self-narration. The therapeutic encounter becomes the attempt at a new realization of the process of “digesting” emotive experiences, linked first of all to attachment experiences, insofar as what is narrated constitutes a shared lexicon of feelings that, if considered closely, correspond to the reflective function.
As Ortu et al. (1992) observe, while secure subjects present a fluent integrated narrative of their childhood experiences and know how to consciously assess the influences on their mental states, distanced subjects undervalue the attachment experience, presenting a narration with few memories and the worrying experiences seem to be involved in early attachments, an element that is reflected in a confused, untruthful narrative, overpowered by the past.
Attachment strategies therefore influence not only the interactions between partners who have an  affective bond, but also condition the narrative quality regarding attachment experiences and therefore metacognitive capacity, which is both recognition of representational capacities and representational changes. The aim of the Adult Attachment Interview (George, Main & Kaplan, 1994) is therefore that of assessing the subjective aspects of the subject’s autobiographical account, that is, the state of mind towards attachment, by placing the subject interviewed in a strange, unusual situation that takes the unconscious by surprise.
Based on the different narrative capacities of subjects with different attachment configurations, Main (1995) indicates the existence of three prototypical pathologies of narrative capacity: the clinging to rigid stories, the non historicization of experience and the narration’s inability to limit suffering, pathologies that are evident in the analysis of the patient’s narratives.
Moreover, by paying attention to these specific pathologies of narration and of the subjects’ attachment configurations, there will clearly emerge a way of considering psychological structure in terms of a continuum between structure and affective regulation (Slade, 2002).
On the basis of the relation outlined here between attachment theory, intersubjectivity and group-analytic theory, we will conclude by recalling what Foulkes said in his writings about the role of the group mechanism. Thanks to the setting and to the group function, in fact, according to the author, there emerge configurations prompted by the interactions between individuals and by the group context in the background. What follows from this is a network of communications organized vertically and horizontally, in which the communication becomes a spiral passing through the group and in which the group offers itself as a natural context for the processing of the historicity of the events both at the level of experience and at that of reflection. Within the group there can in fact coexist, thanks to the role of the therapeutic factors mentioned, the implicit and the explicit forms of memory (Ondarza Linares, 2004). This is consistent with what has been said on attachment systems.
The individual story is thus deconstructed and with the involvement of the group network takes on a new sense, in virtue of the symbolopoietic process, which therefore corresponds to the Working Model contained in the narration and shared by the group. The Self’s reflective function which in one-to-one psychotherapy normally occurs between patient and analyst, in group-analysis occurs instead in the dimension of the group Self,  in its double aspect: the internalized models of the network, or internal groupality, and the internalized models of the external group as the more or less structured reality (Ondarza Linares, 2004).
Proof of the validity of these arguments and the possibility of synthesizing what has been said so far is to be found in Marrone (1999), who identifies seven tasks that the leader must perform in a group-analytic process oriented to attachment theory.
-  increasing the cohesion and affiliation so that the group becomes a secure base;
- exploration of the present relational situation;
- reflection on the responses expected in relating to others;
- reinforcing episodic and semantic memories, in order to consider the IWM of the past and their influence on the present;
- identification of the crises of empathy among members, in order to reflect on reactivation, typical of insecure attachments, fear and non-empathy;
- identification of the dysfunctional strategies of a group member so as to regulate the closeness to others and self-esteem, and to reject vulnerability;
- promotion of interpersonal knowledge, stimulating a culture of reflection.
According to Marrone this will happen in the light of four main characteristics activated in the group setting, namely learning that observed behavior is determined by interpersonal and intrapsychic causes, understanding from experience that people express different emotions from those they feel and that they can even unconsciously try to deceive others, and learning that people are often not aware of their own motivations and energies, which are central aspects of the dynamics present in group-analytic work, the purpose of which is always that of achieving a genuine, lasting change in the patient’s  mental organization. In the group it is therefore possible, thanks to the interaction with others, to develop the metacognitive capacities underlying the reflective function (Marrone, 1999).
To conclude, we can therefore say that in the group-analytic setting, at the group level and in virtue of the group nature of the mind, there is an increase in the subject’s reflective capacity and in the transformational possibilities of thought, which is both the purpose of the psychotherapy intervention and the confirmation of the value attributed, in attachment theory just as in group-analytic theory, to an interpersonal conception of the human being, who exists only when relating with others.

References

Ainsworth, M.D.S. (1967). Suicidal behavior and attachment: A developmental model. In B.M. Sperling, & H.W. Berman (Eds.), Attachment in adults: Clinical and developmental perspectives (pp. 275-298). New York: Guilford Press.

Attili, G. (2007). Attaccamento e costruzione evoluzionistica della mente [Attachment and evolutionistic building of the mind]. Milano: Raffaello Cortina.

Bowlby, J. (1969). Attachment and Loss (Vol. 1). New York: Basic Books.

Bowlby, J. (1973). Attachment and Loss (Vol. 2). New York: Basic Books.

Bowlby, J. (1980). Attachment and Loss (Vol. 3). New York: Basic Books.

Bowlby, J. (1988). A Secure Base: London: Routledge.

Bretherton, I. (1987). New perspectives on attachment relations: security, communication, and internal working models. In J. Osofsky (Ed.). Handbook of infant development (pp. 1061-1100). New York: John Wiley and sons.

Bretherton, I. (1991). The origins of attachment theory: John Bowlby and Mary Ainsworth. In In C. M. Parkes, J Stevenson-Hinde & P. Marris (Eds.), Attachment Across the Life Cycle. London: Routledge.

Byng-Hall, J. (1995). Rewriting family scripts. New York: The Guilford Press.

Byng-Hall, J. (1999). Family and Couple Therapy: Toward Greater Security. In J. Cassidy, & P. R. Shaver, Handbook of Attachment: Theory, Research, and Clinical Applications (pp. 707-731). New York: Guilford Press.

Byng-Hall, J., & Stevenson-Hinde, J. (1991). Attachment relationships within family system. Infant Mental Health Journal, 12, 187-200.

Carli, L. (Ed.) (1999). Dalla diade alla famiglia. I legami di attaccamento nella rete familiare [From the dyad to the family. The attachment bonds in the family net]. Milano: Raffaello Cortina.

Craik, K. (1943). The nature of explanation, Cambridge: Cambridge University Press.

Di Maria, F., Lo Verso, G. (Eds.). (2002). Gruppi [Groups]. Milano: Raffaello Cortina.

Doane, J. A., & Diamond, D. (1994). Affect and Attachment in the Family: A Family Based Treatment of Major Psychiatric Disorder. New York: Basic Books.

Donley, M. G. (1993). L’unità emotiva dell’attaccamento [The emotional unit of the attachment]. In L. Carli (Ed.), Dalla diade alla famiglia (pp.65-92). Milano: Raffaello Cortina.
 
Dozier, M., Cue, K., & Barnett, L. (1994). Clinicians as caregivers: role of attachment organization in treatment. Journal of Counsulting and Clinical Psychology, 62, 793-800.

Emde, R. N. (1991). The Wonder of Our Complex Enterprise: Steps Enabled by Attachment and the effects of relationships on relationships. Infant Mental Health Journal, 12 (3), 164-173.

Fivaz-Depeursinge, E., & Corboz-Warney A. (1999). The primary triangle. New York: Basic Books.

Fonagy, P. (2001). Attaccamento e funzione riflessiva [Attachment and reflective function]. Milano: Raffaello Cortina.

Foulkes, S.H. (1973). The Group as Matrix of the Individual's Mental Life. In L.R. Wolberg & E.K. Schwartz (Eds.), Group Therapy (pp. 211-220). New York: Intercontinental Medical Book Corporation.

Foulkes, S.H. (1975), Group-Analytic Psychotherapy: Method and Principles. London: Gordon & Breach.

Gallese, V. (2003). La molteplice natura delle relazioni interpersonali: la ricerca di un comune meccanismo neurofisiologico [The multi-facets of the interpersonal relations nature: the research of a common neurophysiologic mechanism]. Networks, 1, 24-47.

Hinde, R.A., & Stevenson-Hinde, J. (1991). Perspective on attachment. In C. M. Parkes, J. Stevenson-Hinde, & P. Marris (Eds.), Attachment across the life cycle. London: Routledge.

Holmes, J. (1993). John Bowlby and Attachment Theory. London: Routledge.

Holmes, J. (2001). The Search for the Secure Base: Attachment Theory and Psychotherapy. London: Routledge.

Horowitz, M. (1987). States of mind. New York: Plenum Press.

Lavelli, L. (2007). Intersoggettività. Milano: Raffaello Cortina.

Lo Verso, G. (1994). Le relazioni soggettuali [The subject relathionships]. Torino: Bollati Boringhieri.

Main, M. (1995). Attachment Overview with implications for clinical works. In S. Goldberg, R. Muir, & I. Kerr (Eds.), Attachment theory: social development and clinical perspective (pp. 407-474). Hillsdale: Analytic Press.

Main, M. (1999). Una visione di insieme sulla teoria dell’attaccamento [An overall view on the theory of the attachment]. In L. Carli (Ed.), Dalla diade alla famiglia. I legami di attaccamento nella rete familiare (pp.1-13). Milano: Raffaello Cortina.

Marrone, M. (1999). Attaccamento e interazione [Attachment of interaction]. Roma: Borla.

Marvin, R. S., & Stewart, R. B. (1999). Una cornice sistemico-familiare per lo studio dell’attaccamento [A systemic-familiar frame for the attachment study]. In L. Carli (Ed.), Dalla diade alla famiglia. I legami di attaccamento nella rete familiare (pp. 19-64). Milano: Raffaello Cortina.

Minuchin, S. (1974). Families and family therapy. Cambridge (Mass.): Harvard University Press.

Minuchin, S. (1995). Families and individual development: Provocations from the field of family therapy. Child development, 56, 289-302.

Montesarchio, G. (Ed.). (2002). Colloquio in corso [Colloquy in progress]. Roma: FrancoAngeli.

Muscetta, S. (2006). Attachment e psicoterapia: teoria e ricerca empirica [Attachment and psychoterapy: theory and empiric research]. In N. Dazzi, V. Lingiardi, & A. Colli (Eds.), La ricerca in psicoterapia: Modelli e strumenti (pp. 567-590). Milano: Raffaello Cortina.

Napolitani, D. (1987). Individualità e gruppalità [Individualità and groupality]. Torino: Bollati Boringhieri.

Ondarza Linares, J. (2004). Gruppoanalisi e teorie dell’attaccamento: Gli aspetti metodologici e tecnici [Group analysis and attachment theory: methodological and technical aspects]. Attualità in Psicologia, 19 (1-2), 51-60.

Ortu, F., Dazzi, N., De Coro, A., Pola, M., & Speranza, A.M. (1992). Un contributo di ricerca sugli stili di attaccamento in preadolescenza: La coerenza della verbalizzazione nell’intervista [A research contribution on attachment style in preadolescente. The coherence of the interview recording]. Adolescenza, 3, 269-287.

Pisani, R. (2000). Elementi di gruppoanalisi [Elements of group analysis]. Roma: Edizioni Universitarie Romane.

Reiss, D. (1989). The represented and practising family: Contrasting visions of family continuity. In A.J. Sameroff & R.N. Emde (Eds.), Relationship disturbances in early childhood: A developmental approach (pp. 30-51). New York: Basic Books.

Rycroft, C. (1985). Psychoanalysis and Beyond. London: Chatto.

Schank, R.C., & Abelson, R.P. (1977). Scripts, plans, goals and understanding. Hillsdale: Erlbaum.

Siegel, D.J. (1999). The Developing Mind. Toward a Neuro-biology of Interpersonal Experience. New York: Guilford.

Simonelli, A. (2006). La prospettiva rappresentativo-narrativa dell’attaccamento [The representation-narrative view of the attachment]. In O. Codispoti & A. Simonelli (Eds.), Narrazione e attaccamento nelle patologie alimentari (pp.33-61). Milano: Raffaello Cortina.

Simonelli, A., & Calvo, V. (2002). L’attaccamento: Teoria e metodi di valutazione [The attachment: theory and methods of assessment]. Milano: Carocci.

Slade, A. (2002). Attachment Theory and Research. In J. Cassidy, & P. R. Shaver (Eds.), Handbook of Attachment: Theory, Research, and Clinical Applications (pp. 652-673). New York: Guilford Press.

Speranza, A.M. (2006). La valutazione dell’attaccamento [The assessment of the attachment]. In N. Dazzi, V. Lingiardi, & A. Colli (Eds.), La ricerca in psicoterapia. Modelli e strumenti (pp. 539-566). Milano: Raffaello Cortina.

Stern, D.N. (1985). The interpersonal world of the infant. New York: Basic Books.

Stern, D.N. (1995). The motherhood Constellation. A Unified View of Parent -Infant Psychotherapy. New York: Basic Books.

Stevenson-Hinde, J. (1990). Attachment within the family system: An overview. Infant Mental Health Journal, 11, 218-227.

Trevarthen, C. (1993). The function of emotion in early infant communication and development. In J. Nadel, & L. Camaioni (Eds.), New perspectives in early communicative development (pp. 48-81). London: Routledge.

Waddington, C. (1977). Tools for Thought. London: Cape.

Wheeler, M.A., Stuss, D.T., & Tulving, E. (1997). Toward a theory of episodic memory: The frontal lobes and autonoetic consciousness. Psychological Bulletin, 121, 331-354.

Wynne, L. (1984). The epigenesis of relational systems: A model for understanding family development. Family Process, 23, 297-318.

Notes

* Department of Psychology, University of Palermo. Top