The Hidden History of Relational Public Services: From the Coal Mine to the Front Door
Physics, Ethics, Economics and Prototypes
We often talk about "relational public services" as a modern reaction to the cold efficiency of New Public Management. But the lineage of this idea is much deeper.
It is not a single invention, but a recurring discovery made by different people in different decades (miners, social workers, nurses, economists, and researchers) who all stumbled upon the same truth: when you treat human beings like industrial components, the system fails.
We have been reading and working in this field for about 20 years. What follows is not comprehensive; it is our current understanding of the lineage that matters most. We have identified an underlying logic and nine stages that have shaped how we think about relational services and teams. Others would choose differently, (and do!). This is the timeline that makes sense to us.
The underlying logic: Deming and Argyris
Before we trace the nine stages, we need to name the thread that runs through all of them. Two thinkers, working in parallel with many of the figures below, provided the conceptual grammar that makes sense of everything else.
W. Edwards Deming is often filed away as a "quality control" guru, but his real intervention was moral and managerial: stop blaming people for outcomes created by the system they work in. Deming estimated that the vast majority of trouble and improvement potential "belongs to the system," meaning leadership has to change conditions, not merely demand more effort.
This way of thinking became a major influence on John Seddon's work in services. If failure is largely systemic, then "holding individuals accountable" for predictable consequences of the design is not only unfair; it is a basic misunderstanding of variation and learning.
Deming's first job for leaders was famously blunt: "Drive out fear." A fearful frontline cannot surface reality, learn, or exercise judgement; it can only comply. That thread runs through what follows. Wherever you see high-discretion work done well, such as the coal miners, Buurtzorg, or the Nordic welfare states, you find the same underlying architecture: trust, competence, and feedback close to the work.
Chris Argyris looked at how we think, and why organisations resist learning. He distinguished between "Single-Loop Learning" (fixing a thermostat when the room is too cold) and "Double-Loop Learning" (asking why the room is too cold in the first place, or questioning whether we even need it that warm).
Most public service organisations are trapped in what he called "Model I" behaviour: defensive routines designed to maintain control, save face, and avoid embarrassment. In Model I, admitting you don't know the answer feels like failure. So, we pretend we do know. We follow the protocol. We blame the citizen for not fitting the service.
Relational work is essentially an attempt to move public services into "Model II": where we can be vulnerable enough to acknowledge uncertainty, test assumptions, and learn from what actually happens rather than what we thought would happen.
This matters because you cannot do relational work in a Model I culture or system. If workers are afraid to admit they don't have the answer, they will reach for the referral form or the standardised pathway. If they are afraid of making mistakes, they will stick rigidly to process even when it is clearly not helping.
Model II requires what Argyris called "valid information, free and informed choice, and internal commitment." In plain English: tell the truth about what you know and don't know, make decisions together based on what's actually happening, and commit to learning rather than defending.
This is the logic that connects everyone who came after. The coal miners practised it. The Nordics built systems around it. Buurtzorg proved it scales. Human Learning Systems finally gave it a management framework. We reflect it in our Relational Service Design. Keep Deming and Argyris in mind as you read what follows; they are the bass line beneath the melody.
The wider currents: Psychology and Democracy
Before we begin the timeline, two deeper traditions are worth naming. They run alongside everything that follows.
The psychology of connection: Carl Rogers showed that the therapeutic relationship itself is what heals. His "Person-Centred" approach (that the expert in the room is usually the citizen, not the professional) shaped Biestek and echoes through everything from Participle to Human Learning Systems.
More recently, Iain McGilchrist offers a (contested) neuroscientific framework. His work on the brain's hemispheres suggests modern bureaucracy is a product of the "left brain" (abstract, categorising, controlling) which has usurped the "right brain's" ability to sustain context, empathy, and connection.
The democratic tradition: Paulo Freire in Brazil taught us that liberation comes through dialogue, not deposit-making, which treats students like empty bank accounts to be filled. Asset-Based Community Development (ABCD) flipped the script from "needs" to "strengths," asking what a community can do for itself before the state interferes.
In Aotearoa New Zealand, Whānau Ora demonstrates that the unit of wellbeing is not the isolated individual but the extended family (whānau). It is perhaps the most advanced example of a state system that funds relationships rather than transactions. These currents surface throughout the timeline. Watch for them.
1. The ethical and realistic foundation (1957–1980)
The movement began by defining what good work looks like, and who actually holds the power.
The ethics (1957): Felix Biestek wrote what became the Old Testament of social work: The Casework Relationship. He defined the seven principles of care, including acceptance and non-judgmental attitude. His insight was simple but radical: the relationship is the tool. Everything else follows from that.
Biestek was articulating in social work what Carl Rogers was developing in therapy, specifically the insight that the expert in the room is usually the person being helped, not the professional. This was not soft thinking. It was foundational: you cannot help people if you don't see them as whole human beings first.
The reality (1980): Decades later, Michael Lipsky revealed a crucial truth in Street-Level Bureaucracy: frontline workers such as teachers, police officers, and social workers are the actual policymakers. They exercise discretion every day, making hundreds of small decisions that shape people's lives. Bureaucracy tries to crush this discretion through rules and targets; relational services try to liberate it.
2. The laws of complexity (1949–1960s): The coal miners and the cyberneticians
While Biestek wrote about individuals, Trist and Bamforth looked at teams.
In the post-war coal mines of northern England, they saw that new mining technology was backfiring whenever work stayed organised in rigid, Taylorist ways. Where miners re-organised themselves into autonomous, multi-skilled groups, the same machines delivered higher productivity and better morale. Their conclusion was clear: you cannot optimise the technical system without redesigning the social system around it. This became socio-technical systems (STS): the idea of jointly designing technology and work organisation, and the intellectual ancestor of today's self-managing teams.
Fred Emery, an Australian psychologist, joined Trist at the Tavistock Institute in the late 1950s and became STS theory's most important developer. Their 1965 paper, "The Causal Texture of Organizational Environments," became foundational to the field. But Emery's real contribution came after he returned to Australia in 1969 when he created practical methods that turned STS from theory into practice. His Participative Design Workshop method gave organisations a concrete way to redesign work around self-managing teams.
Emery proved that the coal miners' discovery wasn't a one-off. It was a blueprint that could work anywhere, from factories to hospitals to social services.
At the same time, the cybernetician W. Ross Ashby was giving formal language to what the miners were learning in practice. In An Introduction to Cybernetics (1956), he set out the Law of Requisite Variety: for any system to stay in control, it needs enough flexibility to respond to the range of situations it faces. Often paraphrased as "only variety can absorb variety".
Applied to work like mining, social care, or crisis response, the implication is straightforward. When the environment is complex and unpredictable, frontline teams need discretion, meaning room to judge, adapt, and coordinate locally. Strip that discretion away, try to manage everything through rigid rules and centralised oversight, and you reduce the system's ability to cope with real-world variation. Failures aren't guaranteed in every case, but the risk becomes structural: the organisation cannot reliably meet complexity with complexity. Errors, workarounds, and breakdowns multiply.
The miners hadn't discovered a "nicer" management style. They'd bumped into a deep principle about how control works in complex systems. This was Deming's principle in action: the workers know best how to organise the work. Management's job is to create the conditions, not control the process. Decades later, Agile would repackage the same lesson as a delivery discipline: self-organising teams, fast feedback, and learning by doing.
3. The economic proof (1990–1999): The licence to trust
While the philosophers wrote, practitioEconomists then provided the proof that top-down control wasn't the only way. They gave the movement its economic licence to exist.
The commons (1990): Elinor Ostrom destroyed the myth that we need top-down control to manage resources. Ostrom's empirical research and institutional analysis proved that local communities, given the right conditions, can self-govern complex systems without either top-down direction or a market. She won the Nobel Prize for showing what the coal miners had already discovered: people closest to the work know how to organise it. Her research gave policymakers permission to trust.
The capability approach (1999): Amartya Sen revolutionised how we think about poverty. He proved that "welfare" isn't about delivering goods; it's about expanding capabilities and the freedom to achieve the life you value. This shifted the question from "What does this person need?" to "What does this person need to be able to do?". It became an economic foundation for Hilary Cottam's Radical Help.
Both Ostrom and Sen were arguing against the assumption that the state knows best. They were providing the economic backbone for what would become relational practice.
4. The philosophical awakening (1998–2008): The theory
By the late 90s, the industrial model had taken over public services. Three thinkers explained why this was doomed to fail.
The blind spot (1998): James C. Scott's Seeing Like a State revealed why large organisations keep making the same mistakes. Bureaucracy needs to make complex lives "legible" or to simplify messy reality into neat categories it can process. The cost? It destroys what Scott calls mêtis: the local, practical knowledge that actually makes things work. What gets measured gets managed. What can't be measured gets ignored.
The ethical alternative (2001): Andries Baart's Theory of Presence provided the philosophical backbone. He argued that care isn't about "fixing" a problem and moving on, which he called intervention. It is about "staying with" someone through complexity, or what he called presence: attentive nearness over distance and protocol. This became the vocabulary the movement needed to fight for relational work.
The logic of care (2008): Annemarie Mol challenged how we think about healthcare itself. In The Logic of Care, she argued that we treat patients like consumers making choices, which is the logic of choice. But good care isn't about choosing from a menu. It is about tinkering, adapting, and learning what works for this person in this situation: the logic of care. Real care is iterative, not transactional.
5. The diagnosis (1990s–2000s): What's actually going wrong
FWhile the philosophers wrote, two practitioners looked at how services actually worked, and found the same disease.
The efficiency sceptic: John Seddon In the 2000s, Seddon looked at call centres and saw something everyone else had missed. "Efficiency targets," like call handling times, weren't making things more efficient. They were creating what he called "failure demand": people calling back because their problem wasn't solved the first time. The system was optimised for speed, not for solving problems.
His work proved that chasing efficiency without understanding demand makes everything more expensive, not less. This was Deming's 94% made visible: the system creating waste, then blaming workers for not processing it fast enough.
The social worker: David Thorpe Starting in the 90s, Thorpe looked at the "front door" of children's social services and saw it was clogged with referral forms. He realised forms forced workers to process data instead of people. The form became a way of managing anxiety (pass it on, cover yourself!) rather than actually helping. He scrapped the forms and replaced them with conversations. The result? Fewer children escalated into the system, better outcomes, and lower costs. Thorpe proved that bureaucracy doesn't manage risk; it creates failure demand. His influence and impact on children's services has been profound and lasting.
Seddon and Thorpe were effectively applying Ashby's law to the service sector: when you standardise the work (reducing variety), you strip the frontline of the ability to solve the problem, forcing people to come back. This was pure Deming: the system creates 94% of the problems. Fix the system, not the people.
6. The operating rhythm: Agile (2001–present)
In the early 2000s, a parallel tradition in software development codified a way of working that turns learning in complexity into a routine: Agile.
The Agile Manifesto valued individuals and interactions, rapid feedback, and responding to change over following a fixed plan. This is not because plans are bad, but because complex environments keep moving. What matters for relational public services isn't "Scrum" or the jargon. It's the underlying mechanism: short learning cycles and self-organising teams.
In Ashby's terms, Agile is a method for increasing a team's capacity to absorb the variety it faces by pushing judgement and adaptation closer to the work and by building fast feedback loops so teams can adjust continuously.
More recently (2017), practitioners have translated this directly into service contexts through the Agile Service Manifesto. It shifts the centre of gravity from expert solutions and business cases toward user needs, working prototypes, and learning through change. Its principles foreground co-design, experimentation, building people's capabilities, and trusting teams with real responsibility, exactly the conditions relational work requires.
7. The prototypes (2006–2007): Building the alternative
In 2006 and 2007, we saw two very different experiments that proved relational services could scale.
The nurse: Jos de Blok and Buurtzorg In 2006, de Blok founded Buurtzorg in the Netherlands, proving that stripping away managers and trusting nurses lowers costs and improves lives. Self-managing teams of district nurses working in small neighbourhoods deliver better care at lower cost. No middle management. No call centres. Just nurses doing what nurses know how to do.
Frederic Laloux later held it up as the defining example of "Teal" organisations: places where hierarchy gives way to trust. But what's often missed is that de Blok was essentially operationalising what Emery had developed decades earlier: Participative Design, finally made to scale. Buurtzorg now has over 15,000 nurses and has been replicated in multiple countries. It proved this wasn't idealism. It was a better business model.
The designers: Hilary Cottam and Charlie Leadbeater One year later in 2007, Hilary Cottam and Charlie Leadbeater, alongside Hugo Manassei, launched Participle. Cottam brought design thinking: the discipline of prototyping and iteration. Leadbeater brought social entrepreneurship: the belief that innovation could come from citizens themselves.
Together, they ran experiments like The Circle (supporting older people through relationships rather than services) and Life (rethinking the jobcentre). They proved that we must build people's capabilities, not just service their needs. Welfare isn't something you deliver. It's something you grow. This echoed Sen's capabilities approach: expand what people can do, don't just give them things. Cottam would later codify these lessons in Radical Help (2018), one of the most important books on public services in a generation.
8. The Nordic countries: Trust as infrastructure
While the UK was busy perfecting New Public Management, the Nordic countries absorbed its influence but continued quietly building something different: a state built on trust as infrastructure.
In 2012, the UK think tank IPPR tried to give this "something different" a name. Their report The Relational State, with lead essays by Geoff Mulgan and Marc Stears, argued that government's job isn't just to deliver services but to foster relationships and build social capital. They were looking, in part, at what the Nordics had been doing for decades.
But the Nordics didn't need a name for it. Their welfare systems were built on something we struggle to even talk about in British policy: trust as infrastructure. High trust between citizens and state. High trust within organisations. The assumption that people will do the right thing if you give them the conditions to do so.
This was not accidental. It was the direct legacy of the Norwegian Industrial Democracy Project in the 1960s. Einar Thorsrud and Fred Emery proved that "industrial democracy" (giving workers control over their own tasks) wasn't just ethical. It was more productive.
Over time, that "working-life democracy" sensibility flowed into the wider Nordic way of governing and managing work across sectors, including the public sector: strong unions, negotiated change, and a presumption that professional discretion and team autonomy are assets, not threats. You can also see its fingerprints in later Scandinavian traditions like participatory design, where public-sector staff (from administrators to nurses) are involved in shaping the tools and technologies they rely on.
And in doing so, they inadvertently built what cybernetician Stafford Beer was calling, in the early 1970s, a "Viable System". Beer argued that a system can only handle complexity if decisions are made at the point of action. By pushing decision-making to the edges (to the nurse, the teacher, or the social worker) the Nordics solved the problem of "variety" that paralysed the UK. They built a system structurally engineered to survive complexity, and Scandinavian participatory design was part of the toolkit.
This was not just soft policy. It had teeth. Sweden's approach to social services emphasised continuity of care and professional autonomy. Danish job centres trusted frontline workers to exercise judgment rather than follow scripts. Finland embedded collaborative decision-making into child protection, with families helping design their own support plans. The Nordic model proved something important: relational approaches aren't incompatible with universal services. They might be the only way universal services actually work.
What this history tells us
Here's what strikes me about this lineage: we didn't need more evidence. We had it by 1990. The coal miners proved the practice. Ashby provided the physics. Ostrom delivered the economics. By the time Buurtzorg launched in 2006, the case was already airtight. And yet we spent those decades doubling down on targets, call-handling times, and referral forms.
So the question isn't whether relational services work. The evidence is overwhelming. The question is why we keep forgetting.
Part of it is structural. New Public Management gave us a language of accountability that felt modern and rigorous. Targets are legible. Relationships aren't. And as Scott showed, states have a deep bias toward legibility: toward things they can count, categorise, and control.
Part of it is cultural. Argyris was right: most organisations are stuck in Model I. Admitting uncertainty feels like failure. So we reach for the certainty of a protocol, a pathway, or a referral form. We pretend we know the answer because the alternative (sitting with someone in complexity) is harder.
But I think there's something else. The shift to relational work takes energy, and people in public services feel they don't have the bandwidth. It asks more of workers, more of managers, more of commissioners. It requires us to trust people we've been trained to monitor. To measure learning instead of outputs. To redesign not just frontlines but whole organisations.
That's uncomfortable. It feels easier to buy some AI that makes the organisation do the work faster, but not fundamentally better.
And yet the history also shows us something else: the effort releases energy. Working more closely with residents, seeing the impact, reconnecting with purpose – these aren't just nice-to-haves. They're what makes the work sustainable. Every case study in this lineage found that relational approaches weren't just more effective, they were more energising.
The lineage in this article isn't a history lesson. It's a toolkit. Biestek gave us the ethics. Ashby gave us the physics. Ostrom gave us the economics. De Blok and Cottam showed it scales. Lowe and colleagues gave us the management theory.
We have everything we need. The only question is whether we're willing to use it.
Dennis Vergne (c) Basis ltd, 2025
References
Agile Alliance (2001) Manifesto for Agile Software Development. Available at: agilemanifesto.org
Agile Service Manifesto (2017). Available at: agileservicemanifesto.org
Argyris, C. and Schön, D.A. (1978) Organizational Learning: A Theory of Action Perspective. Reading, MA: Addison-Wesley.
Ashby, W.R. (1956) An Introduction to Cybernetics. London: Chapman & Hall.
Baart, A. (2001) Een theorie van de presentie [A Theory of Presence]. Utrecht: Lemma.
Beer, S. (1972) Brain of the Firm. London: Allen Lane.
Biestek, F.P. (1957) The Casework Relationship. Chicago: Loyola University Press.
Cooke, G. and Muir, R. (eds.) (2012) The Relational State: How Recognising the Importance of Human Relationships Could Revolutionise the Role of the State. London: Institute for Public Policy Research (IPPR).
Cottam, H. (2018) Radical Help: How We Can Remake the Relationships Between Us and Revolutionise the Welfare State. London: Virago.
de Blok, J., Jansen, T., Suichies, H. and Vogelpoel, L. (eds.) (2016) Het kleine alternatief voor de zorg: Humaniteit boven bureaucratie [The Small Alternative for Care: Humanity Over Bureaucracy]. Amsterdam: Stichting Beroepseer.
Deming, W.E. (1986) Out of the Crisis. Cambridge, MA: MIT Press.
Emery, F.E. and Thorsrud, E. (1976) Democracy at Work: The Report of the Norwegian Industrial Democracy Program. Leiden: Martinus Nijhoff.
Emery, F.E. and Trist, E.L. (1965) 'The Causal Texture of Organizational Environments', Human Relations, 18(1), pp. 21–32.
Freire, P. (1970) Pedagogy of the Oppressed. New York: Herder and Herder.
Kretzmann, J.P. and McKnight, J.L. (1993) Building Communities from the Inside Out: A Path Toward Finding and Mobilizing a Community's Assets. Chicago: ACTA Publications.
Laloux, F. (2014) Reinventing Organizations. Brussels: Nelson Parker.
Lipsky, M. (1980) Street-Level Bureaucracy: Dilemmas of the Individual in Public Services. New York: Russell Sage Foundation.
Lowe, T. and Plimmer, D. (2019) Exploring the New World: Practical Insights for Funding, Commissioning and Managing in Complexity. London: Collaborate for Social Change.
Lowe, T. et al. (2021) Human Learning Systems: Public Service for the Real World. London: Collaborate CIC.
McGilchrist, I. (2009) The Master and His Emissary: The Divided Brain and the Making of the Western World. New Haven: Yale University Press.
Mol, A. (2008) The Logic of Care: Health and the Problem of Patient Choice. London: Routledge.
Ostrom, E. (1990) Governing the Commons: The Evolution of Institutions for Collective Action. Cambridge: Cambridge University Press.
Rogers, C. (1961) On Becoming a Person: A Therapist's View of Psychotherapy. Boston: Houghton Mifflin.
Scott, J.C. (1998) Seeing Like a State: How Certain Schemes to Improve the Human Condition Have Failed. New Haven: Yale University Press.
Sen, A. (1999) Development as Freedom. Oxford: Oxford University Press.
Te Puni Kōkiri (2010) Whānau Ora: Report of the Taskforce on Whānau-Centred Initiatives. Wellington: Te Puni Kōkiri.
Thorpe, D. (2007) 'Re-thinking Child Protection', in White, S. (ed.) Child Protection: Practice and Malpractice. Essex: APT Press.
Trist, E.L. and Bamforth, K.W. (1951) 'Some Social and Psychological Consequences of the Longwall Method of Coal-Getting', Human Relations, 4(1), pp. 3–38.