Over the past year I have been reaching, in several posts and in the course of my industrial doctorate work, towards a set of ideas or perceptions about design and the design process that, it turns out I did not yet have the right vocabulary for. In Beyond Technomagic I argued that visibility alone does not produce organisational change; in Who Whom? I traced how a programme sustained incompatible realities by keeping them in separate registers; in The Service Designer as Abstract Signifier I explored how designers become screens for organisational projections. Each time I was circling around a crude reading of a particular intellectual tradition. These observations sat uneasily alongside the wider field's growing confidence in design's capacity to transform social systems - a confidence that, as Vink, Wetter-Edman and Ulloa (2017) observe, outpaced the discipline's understanding of what design could actually offer to such change, let alone how institutions absorb or neutralise the interventions directed at them. This post is an attempt to set that tradition down in one place: systems psychodynamics, as developed from the Tavistock Institute from the late 1940s onwards and articulated most recently by Lawlor and Sher (2021, 2023).
Why this tradition matters for design
Design researchers and practitioners who work in complex institutional settings - healthcare, public administration, defence - routinely encounter phenomena that the dominant framings of design struggle to account for. The Scandinavian and participatory design traditions, widely adopted as normatively emancipatory, remain largely under-theorised in their capacity to handle conflict, power asymmetry, or the performative dynamics of participation itself; as Von Busch and Palmås (2023) argue, the very territory where participation and political realism meet is one that co-design's inclusive ethos has been reluctant to occupy. Stakeholders who asked for change resist it when it arrives. Artefacts that make problems visible are absorbed without producing corrective action. Workshops generate apparent consensus that evaporates by time people have got back to their desks, or surface conflicts that stagnate or are avoidantly ignored. Even on "design projects" or in organisations, actively embracing or engaging with design methods, designers find themselves idealised or imbued with messianic status, or conversely crucified, scapegoated or quietly marginalised. These are some of the patterns I have been exploring in the posts linked above, and which are under resolved in the design literature. They point towards a proposition that the systems psychodynamics tradition has been articulating for seventy years: that organisations are not merely rational systems pursuing their stated tasks as much of the design literature would have it, but are also - and often primarily - collective arrangements for managing anxiety, or operate dysfunctionally. The structures, procedures, cultures, and role boundaries of organisations are not just instruments for getting work done. On this account, they are also socially constructed defences against the anxieties that the work evokes.
Origins
Systems psychodynamics emerged from the Tavistock Institute of Human Relations, founded in London in 1947 by a group of psychiatrists and social scientists - including Wilfred Bion, Elliott Jaques, Harold Bridger, and Eric Trist - who had worked together during the Second World War on large-scale army psychiatry and social reconstruction projects (Lawlor & Sher, 2021). The wartime experience was formative: these were people who had seen what happened to human beings under extreme anxiety and organisational pressure, and who had developed practical methods - group therapy, leaderless group selection, therapeutic communities - for working with the irrational dimensions of collective life under those conditions.
After the war, they turned these methods towards civilian organisations: factories, hospitals, schools, communities. The Tavistock Institute became the institutional home for a distinctive approach that fused two bodies of theory - open systems thinking and psychoanalysis - into a paradigm for understanding organisations that neither tradition could supply alone.
It is not a single unified theory but a living tradition, sustained through practitioner communities, conferences, and institutions - from the Tavistock to ISPSO, OPUS, OFEK, and NIODA - that continues to develop. It integrates sociotechnical systems design, group relations theory, psychoanalytic object relations, social defence theory, and more recently complexity and socio-ecological thinking (Lawlor & Sher, 2023). The term "systems psychodynamics" itself signals the fusion: "systems" points to organisations as open systems existing in dynamic relation with their environments; "psychodynamics" points to the forces - largely unconscious - that animate and distort behaviour within them.
Open systems theory
The structural backbone is open systems theory, developed by biologist Ludwig von Bertalanffy and applied to organisations by A.K. Rice, Eric Miller, and Fred Emery at the Tavistock from the 1950s onwards.
The core proposition is that organisations are open systems that must transact with their environments in order to survive. They import energy and materials, transform them, and export outputs. They are irreducibly relational - their boundary with the environment is always active and must be managed.
Several concepts from this tradition are immediately relevant to design researchers:
Primary task
The work an organisation must carry out to survive. Rice (1958) distinguished three forms: the normative primary task (what the organisation ought to do), the existential primary task (what its members believe they are doing), and the phenomenological primary task (what can be hypothesised they are actually doing). The gap between these three - often large - is itself diagnostic.
Anyone who has worked in a public sector organisation will recognise this gap. I encountered it at SCÖ, where the normative task (develop a federated learning tool for occupational rehabilitation) had become decoupled from the phenomenological task (sustain the consortium, protect the funding narrative, manage reputational risk). The primary task is a heuristic that makes visible the conflicts between competing definitions of purpose.
Boundaries
All systems have boundaries that regulate the exchange between inside and outside. Boundary management is a central leadership task. Disruption to boundaries - excessive permeability, defensive rigidity, unclear demarcation - generates anxiety and dysfunction. Miller and Rice (1967) distinguish boundaries of task, territory, time, transactions, and teamwork. For design, this matters because design interventions typically cross boundaries - between disciplines, between organisational levels, between the espoused and the actual - and boundary-crossing is inherently anxiety-provoking.
Task system and sentient system
Every organisation contains a task system (structured around the work to be done) and a sentient system (structured around feelings of loyalty, belonging, and shared identity). These two systems need to be aligned - "jointly optimised" - but are frequently in tension. The sentient system may seek to preserve itself against task demands; the task system may undermine the relational bonds needed to sustain it. This tension is at the heart of many design implementation failures: a technically sound intervention that disrupts the sentient system will be resisted, not because it is wrong, but because it threatens belonging.
Psychoanalytic theory
The psychodynamic dimension draws principally on Melanie Klein's object relations theory and its development by Bion, but also on Freud, Winnicott, and Bowlby. It offers a theory of unconscious life - in individuals, groups, and organisations.
The core proposition is uncomfortable for anyone trained in the rationalist traditions that dominate design: on this account, much of what happens in organisations is driven not by rational task engagement but by anxiety and the defences erected against it. Unconscious processes - projection, splitting, idealisation, denial - are understood to shape roles, relationships, culture, and structure in ways that are invisible to those inside them.
The paranoid-schizoid and depressive positions
Klein (1946) described two fundamental modes of relating. In the paranoid-schizoid position, anxiety is managed by splitting: the object is either wholly good or wholly bad, and the bad is expelled or attacked. In the depressive position, ambivalence becomes tolerable - the same object can be both good and bad, and concern for the other becomes possible. These are not developmental stages left behind in childhood but ongoing modes of relatedness that any individual, group, or organisation can inhabit. Much organisational dysfunction represents a collective retreat to the paranoid-schizoid position: scapegoating, idealisation of leadership, rigid us-and-them boundaries.
I described exactly this pattern at SCÖ without having the Kleinian vocabulary for it at the time: the programme splitting between an idealised official narrative and a vilified material reality, keeping them in separate registers so they never had to be confronted together. A systems psychodynamic reading would describe that as paranoid-schizoid functioning at an organisational level.
Projective identification
Klein's central concept, and the one most consequential for design practice. Projective identification is the mechanism by which unwanted internal states are not merely attributed to another (projection) but unconsciously lodged there, such that the recipient begins to feel and act as though they own the projected material. In organisational life, this happens constantly: the team that is assigned the role of the "difficult" department begins to behave difficulty; the designer who is positioned as the saviour begins to feel omnipotent; the one who surfaces uncomfortable truths begins to feel - and be treated as - the source of the problem. I suspect that much of what I described in The Service Designer as Abstract Signifier - the way designers become screens for organisational projections - can be productively read as projective identification at work. The miracle worker metaphor, the mechanic metaphor, the teacher metaphor: each represents a different set of organisational needs being projected into the designer's role.
Containment
Bion (1962) developed the concept of the container-contained relationship, drawing on Winnicott's holding environment. The mother's capacity to receive the infant's raw, unthinkable emotional experience and transform it into something bearable - Bion called this "alpha function" - is the prototype for what leaders, organisational structures, and - I would argue - design processes can do: create conditions in which anxiety is held rather than discharged, thought rather than enacted. When containment breaks down, anxiety floods the system and drives defensive and regressive behaviour. The question "who or what is containing the anxiety in this system?" turns out to be one of the most useful diagnostic questions a designer working in a complex institutional setting can ask.
Social defences against anxiety
This is the contribution most directly relevant to design, and the one I have been drawing on most heavily in my own work.
The theory was developed principally by Elliott Jaques (1955) and Isabel Menzies Lyth (1960). The core proposition is that the structures, procedures, cultures, and role boundaries of organisations are not merely rational instruments for achieving the primary task. The tradition argues that they are also - and often primarily - socially constructed defences against the anxieties that the primary task evokes. These defences become built into the fabric of organisational life through collective, unconscious agreement over time.
Menzies Lyth's study of nursing in a London teaching hospital remains the foundational case. She demonstrated that the nursing service - its rituals of task fragmentation, depersonalisation of patients, routinisation of care, avoidance of decision-making - could not be understood as rational task management. It was a collectively maintained defence against the intolerable anxieties of nursing work: proximity to death, responsibility for life, the demands of suffering bodies. The structural arrangements relieved anxiety in the short term but prevented nurses from developing more mature ways of relating to their work and patients, and drove turnover and demoralisation.
The anxiety-culture-defence model (Hinshelwood & Skogstad, 2000) generalises this: anxieties evoked by the primary task generate a collective defensive response, which becomes culturally and structurally embedded, into which new members are inducted, and which the system then self-maintains.
A key diagnostic question is always: what anxiety is this structure or procedure defending against, and at what cost to the primary task?
This is the question I was reaching towards in Beyond Technomagic - where I argued that technomagic beliefs persist not because organisations lack information but because acknowledging the real constraints would require confronting institutional interests, power relations, and the anxiety of not knowing. It is also the question behind Owning the Problem Space - where I argued that design, done honestly, threatens, because owning the problem space means being transparent about what is not known, and organisations defend against that transparency.
Importantly, not all anxiety is to be eliminated. Lawlor and Sher (2021) are clear that a disproportionate amount of anxiety produces defensive and rigid organisations, but too little anxiety - complacency - is equally dangerous. The goal is not to eliminate the discomfort that design interventions produce, but to ensure that the system has sufficient containment to work with that discomfort productively rather than defending against it. This reframes the designer's task: the question is not "how do I avoid making people uncomfortable?" but "is there enough containment here for the discomfort that honest design work will inevitably surface?"
Group relations
Group relations theory, developed principally by Bion (1961), provides a framework for understanding the unconscious life of groups - and therefore for understanding what happens in every design workshop, co-creation session, and multidisciplinary team meeting.
The Work Group
Bion identified two co-existing modes of group functioning. The work group is genuinely engaged with its task, operating realistically, able to tolerate frustration and uncertainty, capable of learning from experience.
The Basic Assumption Group
The basic assumption group behaves as if it were united around an unconscious assumption that has nothing to do with its stated task, but which manages collective anxiety. Bion identified three basic assumptions:
Dependency
In dependency (baD), the group acts as though its purpose is to be sustained by an omnipotent leader, who is expected to solve all problems and spare the group the pain of independent thought. Anyone who has facilitated a design workshop and found the room looking at them expectantly, waiting for the "answer", has encountered dependency.
Fight-Flight
In fight-flight (baF), the group acts as though survival depends on either attacking an enemy or fleeing from danger. A leader who will lead the fight or organise the flight is sought; anyone who points to internal difficulties is resisted. The scapegoating of designers who surface uncomfortable truths - the pattern I described at SCÖ - is a fight-flight dynamic.
Pairing
In pairing (baP), the group invests hope in a future product of two of its members - or in a future event, technology, or reorganisation - and is thereby released from the anxiety of the present task. The hope that the next technology platform, the next reorganisation, the next "transformation programme" will solve everything - without having to do the difficult work now - is pairing in its organisational form.
These are not pathologies but universal features of group life. The work group and basic assumption group always co-exist. The skill is to recognise when basic assumption behaviour is predominating and to work with it, rather than being unconsciously enacted by it.
BART and what it means for the designer
BART - Boundaries, Authority, Role, Task - provides the practical conceptual frame for systems psychodynamic work. In the tradition, it is typically applied to the organisational consultant. I want to suggest that it applies equally - and perhaps more urgently - to the designer working in complex institutional settings, who faces many of the same dynamics but rarely has the conceptual vocabulary to make sense of them.
Role
Every individual in an organisation occupies a role defined by its relationship to the primary task. How people take up their roles - whether they act in role (from the formal and sanctioned position) or out of role (driven by personal need or unconscious group pressure) - is a central diagnostic lens.
Designers in organisations are constantly navigating role confusion: are you the person who makes the wireframes, the person who represents the user, the person who challenges the brief, the person who makes the programme look innovative? The answer is usually "all of these, and the organisation switches between them depending on its needs at any given moment". BART gives this a name and makes it analysable.
Authority
The right to use one's own resources, knowledge, and initiative in service of the task. Authority can be granted from above (formal authority), given from below (by those one serves), and claimed from within (through one's own sense of purpose and competence).
Systems psychodynamic thinking is alert to how authority is surrendered, usurped, refused, or distorted under anxiety.
Designers in public sector settings will recognise the pattern: formal authority to "do design" is granted, but the actual authority to act on what design reveals is withheld. The designer is authorised to look or perform some rituals or symbolics process, but not to change the current organisational structure, or even products of the organisation based what they see or these group activities reveal.
Boundaries
As discussed above - the regulation of inside and outside, belonging and not-belonging, what can enter and leave.
Task
The distinction between normative, existential, and phenomenological primary task, as above.
The systems psychodynamic tradition describes a distinctive practitioner stance that I think transfers directly to design practice in complex systems:
Self as instrument
The practitioner's own emotional and somatic responses are primary data. What you find yourself feeling in the presence of a system is not merely personal reaction but information about the unconscious dynamics of that system. This is countertransference used diagnostically - and it is something designers already do, intuitively, when they pay attention to the atmosphere in a room, the tension in a meeting, the feeling of walking onto a ward or into a programme board. The difference is that the psychodynamic tradition provides a framework for taking these responses seriously as data, rather than treating them as noise to be managed or personal weakness to be suppressed. At SCÖ, I spent months feeling confused, inadequate, and increasingly invisible. I treated this as a personal failing. A systems psychodynamic reading would treat it as data: the system was communicating something through what it induced in me - perhaps the confusion and helplessness that the programme itself could not bear to acknowledge.
Hypothesis generation
Rather than arriving with diagnostic frameworks to impose, the practitioner develops working hypotheses - provisional interpretations of what is happening - which are offered tentatively and tested against the system's response. This is not so different from the reflective practice tradition in design (Schön, 1983), or assumption-based agile development, but it extends the domain of reflection from the conscious and cognitive to the unconscious and emotional.
Working at the boundary
Entry into a system exposes the designer immediately to projective identification. Holding one's role - neither enacting the projections nor defensively refusing them - is the central discipline. When the organisation needs someone to blame, or to idealise, or to carry its confusion, the designer is often the nearest available container. Recognising this as a systemic dynamic rather than a personal predicament changes how designers can understand their own practice - developing practices that manage this projective identification more fluently.
What this tradition offers design
The concepts outlined here - primary task, social defences, projective identification, containment, basic assumptions, BART - are not a replacement for the theories and methods that design research already has. They offer a dimension that design research has not typically foregrounded: an account of what happens beneath the surface when designers enter complex institutional settings and attempt to work with the social and organisational material they find there. Design has sophisticated frameworks for participation, for co-creation, for making things visible; it has tended to address in different terms what happens when the system does not want things to be visible, when participation is unconsciously co-opted, when the designer's own emotional responses are data about the system rather than noise to be suppressed.
References
Armstrong, D. (2005). Organisation in the Mind. London: Tavistock Clinic Series.
Armstrong, D. & Rustin, M. (2019). Social Defences Against Anxiety: Explorations in a Paradigm. London: Karnac.
Bion, W.R. (1961). Experiences in Groups. London: Tavistock.
Bion, W.R. (1962). Learning from Experience. London: Heinemann.
Hinshelwood, R.D. & Skogstad, W. (2000). Observing Organisations: Anxiety, Defence and Culture in Health Care. London: Routledge.
Jaques, E. (1955). Social Systems as a Defence Against Persecutory and Depressive Anxiety. In M. Klein, P. Heimann & R. Money-Kyrle (Eds.), New Directions in Psycho-Analysis. London: Tavistock.
Klein, M. (1946). Notes on Some Schizoid Mechanisms. International Journal of Psycho-Analysis, 27, 99–110.
Lawlor, D. & Sher, M. (2021). An Introduction to Systems Psychodynamics. London: Routledge.
Lawlor, D. & Sher, M. (2023). Systems Psychodynamics: Innovative Approaches to Change, Whole Systems and Complexity. London: Routledge.
Menzies Lyth, I. (1960). A Case Study in the Functioning of Social Systems as a Defence Against Anxiety. Human Relations, 13(2), 95–121.
Miller, E.J. & Rice, A.K. (1967). Systems of Organization. London: Tavistock.
Obholzer, A. & Roberts, V.Z. (1994). The Unconscious at Work. London: Routledge.
Rice, A.K. (1958). Productivity and Social Organisation: The Ahmedabad Experiment. London: Tavistock.
Winnicott, D.W. (1971). Playing and Reality. London: Tavistock.