Group-as-a-Whole and Complexity Theories: Areas of Convergence. Part II: Application to Group Relations,Group Analysis,and Systems Centered Therapy

Introduction

Part I of this discussion proposed that complexity theory could help re-configure aspects of group theory and practice to take into account modern understanding of phenomena characterized by a concatenation of order (predictability, determinism) and disorder (chaos, uncertainty, turbulence), the latter of which are inadequately considered within group theories as such. It was suggested, moreover, that the notions of emergence and self-organization within complexity theory provide a scientific accounting for group systems that are greater than and different from the individuals who co-create and participate in them. These reflections were based on a review of extant literature on complexity and group therapy by Rubenfeld (2001), Stacey (2005), Grossmark (2007), and Brabender and Fallon (2009).

A vital next step is to consider the potential impact of complexity theory on specific theories and approaches used in the training of group psychotherapists, considering ways in which these frameworks could benefit from the insights of complexity theory. The author has chosen three theoretical and training models used by group therapists: 1) Bion’s basic assumption theory and the Tavistock and A.K. Rice group relations training approaches that incorporate it; 2) Foulkes’ group analysis; and 3) Agazarian’s Systems-Centered Therapy. These are among the currently prevalent and most used approaches to training in dynamic group therapy emphasizing group-as-a-whole and group systems understanding. Each offers its own comprehensive frame of reference. And they themselves are all ‘open systems’ that allow for change and modification based on new experience, research, and ideas. In what follows, each perspective shall be discussed with respect to the ways in which complexity theory might add a dimension and/or serve as an important corrective.

Bion’s Basic Assumptions, Group Relations, and Complexity

Bion (1959) took the then bold step of conceptualizing the group-as-a-whole mentality as a dynamic different from that of an aggregate of individuals, and although he himself did not remain long in the group field, turning instead to psychoanalysis, his work fostered the development of the Tavistock and A.K. Rice Institutes, both of which, following Bion, emphasize unconscious processes manifest at the group level.

Bion was educated in Enlightenment thinking, notably that of the philosopher Kant (Bleandonu, 1994: 191–2). His writing on groups retains a sense of group cultures as ‘things-in-themselves’, i.e. realities that are an inevitable aspect of ‘grouping’. For him, the basic assumption states are present in all groups, structured around universal themes of dependency, fight-flight, and pairing. In addition, the ‘work group’ as he depicts it echoes the crack military units he was asked to help his soldier patients re-join. It involves self-examination and adherence to logic and group protocol, while the basic assumption cultures represent emotional regression to primitive phantasy and defence. Bion viewed group dynamics as revolving around a limited number of well-defined climates, each with specific themes and structures.

For Bion, a group alternated between a rational work mode and irrational ‘psychotic-like’ basic assumption cultures. He only peripherally considered transitional, transformative events at the ‘edge of chaos’ where the group could grow, change, and develop. Later, after he shifted his focus to psychoanalysis, he referred to a transformative force in groups in the form of ‘messiah thoughts’, thoughts and their thinkers which challenged the ‘Establishment’ (Bion, 1970: 115–6), the conservative element that maintained and enforced group norms. In his use of the terms ‘messiah’ and ‘Establishment’ there were obvious allusions to the New Testament and the British aristocracy. Within complexity theory, messiah thoughts could be understood as catalytic agents and possibly even mutations that, although they may be irrational, lead to new adaptations and structures. Bion clearly felt that such changes were of high risk in groups, just as catalytic activity and mutations are subject to the crude indifference of natural selection.

Non-Linearity, Complexity and Foulkes’ Group Matrix

S.H. Foulkes was a British contemporary of Bion and like him a psychoanalyst. Both stimulated the intense, enduring interest in group psychotherapy and group dynamics that began in the 1940s. Foulkes, unlike Bion, became a long-time advocate of group therapy and founded the Group Analytic Institute and Society, mentoring students and colleagues until his sudden death while teaching a seminar (de Maré, 1983: 218). (Complexity theory might have a good deal to say on how this unexpected event everlastingly impacted his students and the Institute.)

Importantly, Foulkes provides a resilient theoretical framework with a clear set of agendas for teaching, research, and practice. His conceptual frame allows for integration of diverse intellectual and scientific currents, which thus far have included developments as far ranging as object relations theory (Gordon, 1994), Bion’s theory of thinking (Hinshelwood, 1996), and the philosophical thought of Bahktin on language and dialogue (Pines, 1998: 132). Complexity theory, as Stacey (2005) first suggested, offers a potent addition to them. In what follows, I will focus on two of Foulkes’ key concepts: the group matrix and the social self and suggest some ways that complexity theory can complement and modify Foulkes’ understanding.

Foulkes conceptualized the group as a ‘matrix’, a network of individuals linked together by their interactions and communications. The members are considered ‘nodal points’, very similar to the notion of agents within complexity theory. The group-as-a-whole consists of the patterns of relating that emerge from the interactions, again consistent with complexity theory. One problem with the group matrix concept is that, at present, it remains locked into the idea of orderly interpersonal connections between the members. The reality is that group participants are carrying out multiple complex functions simultaneously, and a significant amount of the information exchange is non-conscious and undisclosed. The observed group matrix is really a narrow ‘tunnel’ (Metzinger, 2009) within a large informational network operating at high speed, its surface stability emerging from complex underlying systems that feed into it. The group matrix contains fractals, ‘black holes’, and discontinuities that can only be understood through complexity theory.

Foulkes viewed the social self as a relational structure and process that results from and in turn stimulates social interactions. His understanding of the social self is remarkably similar to John Holland’s depiction of the agents of complex adaptive systems (Holland, 1975; Waldrop, 1992: 144–97). A synopsis of Holland’s criteria for complex adaptive systems and their agents will suggest how Foulkes’ thinking about the group matrix and the social self implicitly encompasses them in varying and creative ways.

Holland defines the complex adaptive system as a network of many agents acting in parallel. That is, the components of a system are an interconnected network. They all act in ‘parallel’, simultaneously manifesting their individual structures. The group matrix is just such a network, with social selves as the ‘agents’. Foulkes’ concept of mirroring, in which the members identify with aspects of themselves in the other members, is close to what Holland means by parallel. The components of an adaptive system have parallel features, which operate simultaneously, and it is the similarities which allow them to interact with one another. For example, humans communicate with one another because they recognize their similarities. When they do not acknowledge each others’ humanness, as in some situations with strangers or members of a different culture or ethnic group, a condition called ‘pseudospeciation’ (Erikson and Erikson, 1997: 69) may occur, in which individuals regard others as less than human. Chaos, in the form of scapegoating, war, or ethnic cleansing can result. In an adaptive system, the agents must recognize their common features, or they will ignore or eliminate the other, almost like an autoimmune reaction.

Holland also emphasizes dispersal of control in adaptive systems. There is no central locus of control, no principle or agent that governs the whole system. The features of the system result from actions and decisions of all the agents acting in concert. Similarly, Foulkes advocated for therapy of, by, and for the group, with the leader serving as a facilitating ‘conductor’ who helps the members realize their own goals in the group.

Complex adaptive systems manifest many levels of organization, with agents at one level serving as building blocks for agents at a higher level. In like manner, Foulkes (1964: 15 ff.) developed a schema for group levels of development, from the most primitive ‘foundation matrix’ of archetypes and phantasies to the highest levels of achievement and self-actualization. These levels result from the contributions of the entire membership, with the leader serving as a conductor or facilitator.

For Holland, the agents revise and rearrange their building blocks as they gain experience. This is just what Foulkes means by group therapy. He holds, for example that ‘the group constitutes the norm from which the members deviate individually’ (Foulkes, 1964: 63). The deviations include individual neuroses, and the norms established in the group paradoxically manifest healthy expectations which the members can internalize. The collective norms facilitate growth and change.

Finally, adaptive systems have many niches, thus constituting an ecology held in balance by diverse species and locations that cooperate and compete to preserve a degree of survival and stability over time. Foulkes and his followers were aware of the diverse environments and opportunities for adaptation and growth that emerge in small, median, and large groups, as well as the significant differences in groups of diverse ‘populations’. Moreover, Foulkes’ group analysis pays meticulous attention to individual dynamics and personalities, much more so than Bion’s approach.

The similarities between Foulkes’ understanding of the group matrix and Holland’s adaptive systems suggest that some aspects of complexity theory are already implicit within group analysis and could be further exploited. For example, the processes of member self-selection (via extrusion, acceptance, and ability to accept group norms and rules), creativity, and cooperation versus competition (dependency and power relations) have been neglected in Foulkesian studies and could readily be explored within the concepts of mirroring and social self already elaborated in Foulkes’ schema. The dynamics of survival and evolution implied in Holland’s work deserve more attention from group analysts.

Foulkes’ point of view incorporating Wertheimer’s (Ellis, 1938) and Goldstein’s (1940) principles of gestalt psychology emphasizing neuropsychological wholeness and organization, is strongly influenced by assumptions of holistic organization, an important and enduring counterposition to behaviouristic stimulus-response reductionism. However, it is a one-sided view that emphasizes orderly patterns and wholeness at the expense of the chaotic, entropic, and destructive aspects of living systems. This bias is made explicit and partly corrected for in Nitsun’s (1996) notion of the ‘anti-group’, the destructive forces contained in the group matrix. Nitsun, based on biographical information about Foulkes, suggests that the latter had a personal aversion to aggression and hence neglected its role in groups. Nitsun’s anti-group, by contrast, represents the destructive forces of the group matrix, including hatred, deception, sociopathy, and denial of reality.

Foulkes was deeply committed to a view of groups as a hopeful, cooperative aspect of human nature, and, always the aesthetician with a lifelong interest in the arts, he deeply appreciated the orderliness and coherence of groups (which he compared to a symphony orchestra). Therefore, his vision of groups, while refreshingly optimistic, minimized the ‘fractal’ areas of chaos, disorganization, and destructiveness inherent in the matrix.

Agazarian’s Systems-Centered Therapy: General Systems Theory and the Edge of Chaos

Systems-Centered Therapy (SCT) is a comprehensive theory of groups and psychotherapy, developed by Yvonne Agazarian and members of the Systems-Centered Training and Research Institute. Since its fundamental operating principles derive from the theory of living human systems, SCT dovetails in many respects with the systems aspects of complexity theory. SCT incorporates insights from general systems theory, Lewin’s (1951) field theory, and Shannon and Weaver’s (1964) information theory into a broad perspective on groups, leadership, and psychotherapy. The details of SCT are evolving and manifold and have been discussed elsewhere (Agazarian and Gantt, 2000; 2005). Here we will take up a few key concepts with attention to the ways in which complexity theory is consonant with, diverges from, or challenges SCT understanding of groups.

SCT proposes four component systems of all groups: person, member/role, subgroup, and group-as-a-whole which are ‘nested’ within one another, i.e. interdependent and overlapping (Agazarian and Gantt, 2005). The concept of nesting, which originated in the work of the biologist James Grier Miller (1978), takes into account complexity and non-linearity in the group as a living system, an important corrective to both Bion and Foulkes, neither of whom sufficiently considered the diverse ways in which groups can structure and self-organize through the interactions among subsystems. Like Holland’s adaptive self-organizing systems, SCT systems organize their energy and correct their courses based on ongoing feedback.

SCT bases much of its systems work on the principle of isomorphism, in which all systems have similar structure and function (Brabender and Fallon (2009: 45). Thus, for example, the therapist could address conflicts between a) parts of the self, b) role relations, or c) subgroupings in analogous ways, by analysing the forces and informational exchanges involved. Isomorphism is a general systems theory (GST) construct that preceded complexity theory as such. GST recognizes isopmorphism in the ways that cells, organs, whole organisms, and collectivities of organisms (communities) all metabolize energy, process information, and develop structures according to the same biological necessities and laws of chemistry and physics. In these respects, they are isomorphic. Physicians rely on this when they treat an epidemic of, say, smallpox by inoculating individuals. The cells, organs, persons, and communities all heal because they share a common biological base.

However, treating a sick person and healing a whole community require two different disciplines: immunology and epidemiology, because the complexities of these systems differ significantly. Similarly, in group psychotherapy, the members and the group-as-a-whole must be dealt with in very different ways. They are far from being isomorphic. One feature of groups that is different from individuals is that groups disperse temporally and spatially, while individuals cohere as unified organisms in time and space, except when they reproduce. Groups vary in cohesion, while biological individuals are by definition a unit. As Hopper (2003: 91–107) has suggested groups may act as aggregates (‘crustaceans’) or masses (‘amoebae’), and the therapist must respond differently to these two forms of collectivity. A group is not defined by contiguity in time and space, and social distance may vary considerably. ‘Iso-morphic’ literally means ‘same shape’, a parallel, which simply does not exist between the individual and the group.

This difference between SCT and complexity theory has to do with the way in which systems in a hierarchy is seen to develop. In SCT, systems develop cybernetically through feedback loops that seek equilibrium. In complexity theory, systems create higher levels through turbulence, such as the way waves develop on the sea. Waves are not properties of water molecules; rather, they result from wind, gravity, and water coming together. Similarly, group-as-a-whole systems have emergent properties, structures, and functions that individuals and subgroups do not. For example, a group has structures for initiation rites and cultural transmission.

In SCT, ‘human systems function, survive, develop towards increasing complexity by discriminating and integrating differences’ (Brabender and Fallon, 2009: 45.) This view is consistent with complexity theory, which regards change as a function of the interaction among agents that have sufficient similarity to ‘relate’ to one another yet are sufficiently different to produce a ‘reaction’ among them.

When trying to understand how life could develop from inanimate complex molecules, complexity-oriented biologist Kauffman hypothesized that additional agents called autocatalysts could allow such molecules to interconnect more rapidly and increase the probability of a new structure emerging from the ‘primal soup’ (Waldrop, 1992: 117–33). Such autocatalysts are crucial as well in therapy groups, where the leader or an especially influential or creative member, or even a sudden event may propel the group into a productive state change. SCT therapists actively catalyze state changes in their groups that promote growth, in sharp contrast to analytic therapists who often wait for long periods of time for the group become ripe for such changes. SCT utilizes Davanloo’s (2000) model of short-term dynamic psychotherapy, in which the therapist plays an active role in producing change, as a paradigm for therapeutic interventions.

Agazarian further made an important discovery that subgroups serve as lynchpins for change. The key interventions of SCT therapists are aimed at subgroups rather than individuals or the group as a whole. Through interchanges of information and energy, subgroups, which initially form around motives of anxiety and avoidance, can become more functional and productive, and will self-organize into therapeutic subcultures. This is analogous to the way complex living systems such as embryos self-organize in the process of cell division and multiplication. However, complexity theory would point out that each subgroup consists of diverse individuals who propel it from within. In addition, complexity theory would want to articulate how the information contained in a subgroup got there in the first place. Evolution and culture may give human groups a set of inherent structures, templates or archetypes that contain ‘instructions’ to help them self-organize via rituals, role differentiation, and so on. Ettin (2005), without acknowledging complexity theory as such, developed a schema of such rudimentary building blocks of groups.

The difference between SCT and complexity theory is highlighted by the former’s application of Shannon and Weaver’s (1964) information theory to understand differentiation and integration, especially between subgroups, but also at other levels of organization. Agazarian (Agazarian and Gantt, 2000: 134) states that ‘Ambiguity, contradictions and redundancy are at the source of entropic noise in the communication, according to Shannon and Weaver’s information theory. This, then, is the criteria variable I use when listening to group communication’. She advocates minimization of ‘noise’ to facilitate information exchange, differentiation, and integration.

Shannon and Weaver’s theory is based on telephonic and radio communication, in which information occurs in bits or bytes that either are or are not registered by the receiver. This is a linear model. Human emotional and group communication differ from telephonic communication in that emotion-laden group interactions involve multiple dimensions that are transmitted all at once and contain a high degree of ambiguity and turbulence which is yet informative. Thus, one never quite knows everything about another’s emotional state even when it is effectively expressed. The information is complex, like the ambiguous facial expression of da Vinci’s Mona Lisa.

There is an adaptive basis for ambiguity, contradiction, and redundancy in human communication, namely that, as aspects of Winnicot’s ‘potential space’ (Ogden, 1985), they generate possibilities for groups. That is, they constitute the edge of chaos that produces emergent phenomena. In complexity theory, ambiguity, contradiction, and redundancy are not just entropic noise that must be filtered out by therapeutic clarification, confrontation, and interpretation. They also constitute features of the ‘edge of chaos’, which can promote change. A good example is the group’s use of metaphors. Emotions are best communicated by metaphors (cf. Lakoff and Johnson, 1980) because they allow for the ambiguity (paradox), contradiction, and redundancy that are an inherent part of meaning-making in groups.

Summary and Conclusions

This exploration of complexity theory’s potential impact on three theories of the group-as-a-whole is intended to stimulate thinking and call into question several ‘shibboleths’ of these theories that often go unnoticed. Such taken-for-granted beliefs include the notion of a group mind (Bion), the progressive development and goal-directedness of groups (Foulkes; Agazarian), the isomorphism of hierarchical living systems (Agazarian), and their linear information-cybernetic exchange (Agazarian). Such formulations are useful until one encounters complex phenomena that contradict them. Chaos theory suggests that nature has segments that are orderly and others that are in disarray. Complexity theory studies the interface between order and disorder and suggests that it is at the ‘edge of chaos’ that self-organization and hierarchical systems such as groups can emerge. Since adaptive self-organized change is what group therapy is all about, our theories need to be reconfigured to reflect that complex ‘edge’.

A promising bridge between group theory and complexity theory comes from contemporary neuroscience. Neuroscientists are paying increasing attention to how the brain processes social information (Damasio, 1999; Cozolino, 2006), thus providing a linkup with group psychology. They also are incorporating insights from complexity theory. In particular, Edelman’s (1987) theory of ‘neural Darwinism’ emphasizes probability, selection of competing responses, and how neural networks change as a result of experience, all of which are of interest to complexity theorists. Increasing attention is being paid by group theorists to modern neuroscience, which is partly embedded in tenets of complexity theory. An integration of group psychology, neuroscience, and complexity theory will very likely prove to be an important next step in the evolution of group-as-a-whole theory.

Complexity theory is interested in how order and disorder, predictable causality and turbulence, come together to produce hierarchical living systems. As such, it poses a challenge to group-as-a-whole theories that ‘grew up’ within the emphasis on reason, order, design, and predictability in the natural and social worlds. Because complexity theory does not reject such orderliness, but incorporates it in a wider perspective, it may lead us to reconsider some of our assumptions, question long-held concepts, and seek explanations for group phenomena that would otherwise escape our perceptual and cognitive nets.