I've just returned from a period of leave, and I'm trying to make sense of where I am. Six months into the industrial doctorate at SCÖ, and something feels off - not in a way I can easily articulate, but persistently, like a note that doesn't quite resolve.
The formal project documents describe one reality: a collaboration between universities and public agencies to explore machine learning for vocational rehabilitation. Clear objectives, defined work packages, measurable outcomes. The lived experience is something else. Meetings that circle without landing. Technical requirements that assume infrastructure that doesn't exist. A growing gap between what the project claims to be doing and what is materially possible.
But what troubles me most isn't the gap itself - gaps between aspiration and reality are normal in complex projects. It's the expectations placed on me, and on "design", to somehow bridge it. There's an unspoken belief that I can make things happen that the structural conditions don't support. That a good workshop, a compelling visualisation, the right facilitation - that these will somehow transform the situation.
I've started calling this, privately, the "Jesus metaphor". The expectation that the designer will work miracles. Turn water into wine. Make the impossible possible. Heal what is fundamentally broken.
It's flattering and terrifying in equal measure. What I'm beginning to recognise, drawing on Matthews' work on the sacred in service design, is that this expectation may not be incidental. The projection of mythical or emancipatory capability onto the designer - the belief that design can transcend structural constraints through something approaching faith - has the character of a sacred attribution rather than a rational expectation. The organisation doesn't need the miracle to happen; it needs someone to occupy the role of miracle worker.
Finding Morgan
Trying to understand this dynamic, I've been reading Gareth Morgan's work on organisational metaphors. His book Images of Organization (1986, updated 2006) argues that we understand organisations through metaphors - and that different metaphors generate radically different possibilities for thought and action.
"Taking the lens of an image always determines not only a way of seeing but also a way of not seeing" (Morgan, 2006, cited in Örtenblad et al., 2016).
This resonates with my earlier post on generative metaphor. Schön showed how metaphors set problems in social policy; Morgan extends this to how we imagine institutions themselves. The way we see an organisation shapes what we think is possible within it - and what we think is wrong when things aren't working.
Morgan identifies eight root metaphors for understanding organisations. Each illuminates certain aspects while obscuring others. Each generates particular diagnoses and prescriptions. And - this is the crucial point - different stakeholders may be operating with different metaphors without realising it.
Eight Images
Let me briefly sketch all eight before focusing on those most relevant to my situation.
The Machine. The dominant metaphor of classical management. Organisations as assemblages of interlocking parts, each with clearly defined roles. Efficiency through standardisation. The right person for the right job. Management as engineering - designing the machine, maintaining it, optimising its operation.
The Organism. Organisations as living systems that must adapt to their environments. Different "species" of organisation suited to different ecological niches. Focus on needs, growth, survival, evolution. Health as adaptation; pathology as failure to fit the environment.
The Brain. Organisations as information-processing systems. Focus on learning, intelligence, knowledge management. The "learning organisation" that can sense its environment, process information, and adapt its behaviour. Distributed intelligence rather than centralised control.
The Culture. Organisations as systems of shared meaning. Focus on values, beliefs, rituals, symbols, stories. "The way we do things around here". Reality as socially constructed; change as cultural transformation.
The Political System. Organisations as arenas of competing interests. Focus on power, conflict, coalition-building, negotiation. Decisions as outcomes of political processes rather than rational analysis. Who wins and who loses.
The Psychic Prison. Organisations as contexts where people become trapped by their own thoughts, ideas, and unconscious processes. Focus on the hidden constraints we impose on ourselves. The ways organisations enact and reinforce particular constructions of reality that limit what can be imagined.
Flux and Transformation. Organisations as constantly changing systems. Focus on paradox, complexity, self-organisation, emergence. Nothing is stable; apparent stability is just slow change. Patterns rather than structures.
Instrument of Domination. Organisations as systems of exploitation. Focus on whose interests are served, at whose expense. The ways organisations extract value, impose discipline, and maintain inequalities. Power as structural, not just interpersonal.
Each metaphor is partial. None captures the full complexity of organisational life. Morgan's point is not that one is correct but that "new meaning is essentially created by the interplay of various and partially contradicting perspectives" (Örtenblad et al., 2016). Understanding an organisation requires holding multiple images simultaneously - and recognising which images different actors are using.
The Metaphors at SCÖ
Four of Morgan's images seem particularly relevant to understanding my situation.
Machine vs Organism
The ESF funding structure treats the project as a machine. There are defined inputs (funding, personnel, partner contributions), expected outputs (deliverables, reports, training sessions), and measurable KPIs (participants reached, assessments completed, outcomes achieved). The logic is mechanical: if we assemble the right parts and follow the right procedures, the machine will produce the expected results.
But vocational rehabilitation - the domain the project addresses - is fundamentally organic. Each person's situation is different. Progress is non-linear. What works depends on relationships, trust, timing, context. The "outputs" that matter (sustainable employment, improved quality of life, reduced marginalisation) emerge unpredictably from complex interactions, not mechanically from defined processes. As I explored in an earlier post on networks in rehabilitation, the evidence on kin networks, social capital, and informal support systems suggests that recovery is embedded in relational infrastructure that simply does not register in the machine metaphor's input/output logic. The ICF framework recognises this through its participation and environmental factors domains, but the project's governance treats these as background rather than as the material conditions on which any technical intervention depends.
These metaphors generate different expectations. The machine metaphor expects predictability: we defined the project, we funded the project, therefore the project will deliver what was defined. The organism metaphor expects adaptation: we'll try things, see what works, adjust based on feedback, accept that outcomes can't be fully specified in advance.
The tension is palpable in meetings. Projects are governed deliverables and timelines (machine logic). Practitioners talk about relationships and trust-building (organism logic). Both are right within their own frame. Neither can fully hear the other.
Brain (Aspiration) vs Psychic Prison (Reality)
The project aspires to be a "learning organisation". The stated goal is to use AI and data science to learn from rehabilitation data - to identify patterns, improve predictions, enhance decision-making. The brain metaphor promises intelligence emerging from information.
But I increasingly suspect the project is also a psychic prison - trapped by assumptions it cannot examine. The belief that machine learning can be applied to vocational rehabilitation assumes that relevant data exists, is accessible, is structured appropriately, and captures what matters. None of these assumptions holds.
There is no unified database. The data that exists is fragmented across agencies with different systems, different definitions, different legal constraints on sharing. The variables that determine rehabilitation success - motivation, social support, labour market conditions, the quality of the caseworker-client relationship - are not captured in administrative data. The technical infrastructure for "federated learning" doesn't exist and would take years to build.
Yet the project proceeds as though these obstacles are surmountable through sufficient effort and ingenuity. The machine learning imaginary functions as what Morgan might call a psychic prison: a construction of reality that constrains what can be thought. To question whether machine learning is possible here risks appearing obstructive, pessimistic, not a team player.
I find myself wondering: am I the one seeing clearly, or am I trapped in my own prison - unable to imagine possibilities that others can see?
Political System
Beneath the harmonious language of "collaboration" and "partnership", there are real conflicts of interest. Different agencies have different priorities. Who controls the data? Who defines success? Whose voice counts in decisions about what the "AI" should optimise for?
The municipalities want tools that reduce their caseloads. The employment service wants to meet its placement targets. The social insurance agency wants to manage costs. The university researchers want publications and demonstrable innovation. The funding body wants evidence of "impact". These interests don't necessarily conflict, but they don't automatically align either.
Design work - my work - is caught in these political dynamics. When I produce a concept map or facilitate a workshop, I'm not operating on neutral ground. I'm making choices about whose perspective to centre, what to make visible, how to frame trade-offs. The appearance of neutrality ("we're just helping you think together") masks the political nature of the intervention.
The Designer's Role - What Metaphor?
This brings me to what's been troubling me most. Different stakeholders seem to hold different metaphors not just for the organisation and the project, but for what I, as a designer, am there to do.
Mechanic. Fix the broken parts. Diagnose what's not working, apply the right tools, make the machine run better. The designer as technician - skilled, useful, but fundamentally in service of a system designed by others.
Facilitator/Coach. Help people learn and develop. Build capability. Run workshops that enable stakeholders to think better together. The designer as process expert - neutral, supportive, not invested in particular outcomes.
Visualiser. Make things visible. Translate complexity into comprehensible diagrams. The designer as sense-maker - taking the mess and giving it shape that others can engage with.
Magician/Miracle Worker. Transform the situation through mysterious powers. Make the impossible possible. The designer as shaman - possessing special abilities that can overcome obstacles impervious to ordinary effort.
These aren't just different job descriptions. They're different imaginaries about what design is and can do. And they generate different, often incompatible, expectations.
The "Jesus metaphor" - the miracle worker expectation - is the most troubling. It combines magical thinking with displacement of responsibility. If the designer is expected to work miracles, then:
- Structural problems become performance problems (the designer didn't try hard enough)
- Political conflicts become facilitation failures (the designer didn't run the workshop well enough)
- Material impossibilities become imagination failures (the designer didn't envision creative enough solutions)
The miracle worker metaphor protects the organisation from confronting its own contradictions. If change doesn't happen, it's not because change was structurally impossible - it's because the magician's spell didn't work.
There may be a deeper dynamic at work here, one I am exploring through Giesen's framework of sacred and profane performance. The miracle worker expectation might be more than displacement of responsibility. It might be a sacred role - the designer as ritual specialist whose function is not to solve problems instrumentally but to perform the possibility of transformation. The miracle doesn't need to happen; the performance of belief that it could happen is what sustains the collective. This would explain why failure doesn't discredit the expectation: the sacred is sustained by faith, not evidence.
A Pattern Recognised
This isn't the first time I've encountered this dynamic. Looking back to my earlier work at Experio Lab and in Swedish healthcare service design more broadly, I can see the same pattern. The expectation of "heavy lifting" that design can't actually do. The belief that a good workshop or a compelling visualisation can somehow overcome institutional barriers.
I remember projects where service design was brought in to "transform" systems that had no appetite for transformation. Where the methods - journey mapping, prototyping, co-design - were performed with enthusiasm but nothing changed. Where the design work provided cover: "We did service design, therefore we are innovative".
At the time, I attributed these failures to contingent factors - the wrong stakeholders, insufficient mandate, bad timing. Now I wonder if the problem was more fundamental. If the metaphor for design itself was the issue. If stakeholders were operating with a miracle-worker imaginary that no amount of method could satisfy.
Morgan's framework helps me see this as a structural issue, not a personal one. Different metaphors generate different expectations. When the metaphors conflict - when I see myself as a facilitator and others see me as a magician - frustration and disappointment are built into the situation. Not because anyone is acting in bad faith, but because we're operating in different imaginaries.
What Morgan Teaches
Several things follow from this analysis.
Multiple metaphors are always in play. No organisation is only a machine or only an organism. Different aspects are illuminated by different images. The task isn't to find the "right" metaphor but to recognise which metaphors are operating and what each reveals and conceals.
Metaphor conflicts are often invisible. People don't announce which metaphor they're using. It's tacit, embedded in how they talk about problems and possibilities. Surfacing the metaphors requires interpretive work - attending to language, asking what's being assumed, noticing what isn't being said.
Surfacing metaphors is itself political. Making organisational imaginaries visible isn't a neutral act. It can expose contradictions that stakeholders prefer to keep hidden. There may be good reasons - political, emotional, institutional - why certain metaphors remain tacit. The designer who surfaces them takes a risk.
The designer can't work miracles - and shouldn't accept that expectation. This is perhaps the most practically important implication. Recognising the miracle-worker metaphor for what it is - a displacement of responsibility, a way of avoiding structural confrontation - creates space for a different kind of conversation. Not "how can design transform this?" but "what can design actually do here, given the constraints that exist?"
What Comes Next
I don't have neat conclusions. The situation at SCÖ remains what it is - a project with structural contradictions that design cannot resolve. But Morgan's framework helps me understand why I feel caught, and that understanding is itself valuable.
If organisations are understood through competing metaphors, and technology (like AI) is understood through metaphors, then what about "service design" itself? What metaphors shape how people understand the practice and the role? That's where I want to go next - to examine the metaphors that structure expectations of design, and what happens when those metaphors conflict.
The miracle-worker imaginary may be a persistent feature of how design is understood in organisations. If so, naming it might be the first step toward more realistic - and more honest - conversations about what design can and cannot do.
References
Matthews, T. (2021). Exploring Sacred Service Design. PhD thesis, The Oslo School of Architecture and Design.
Morgan, G. (2006). Images of Organization (Updated ed.). Sage Publications.
Örtenblad, A., Putnam, L.L., & Trehan, K. (2016). Exploring Morgan's Metaphors: Theory, Research, and Practice in Organizational Studies. Sage Publications.
Schön, D.A. (1993). Generative metaphor: A perspective on problem-setting in social policy. In A. Ortony (Ed.), Metaphor and Thought (2nd ed., pp. 137-163). Cambridge University Press.
Revision Note: This post was written in late October 2022, after returning from leave and trying to make sense of the first six months at SCÖ. The "Jesus metaphor" framing emerged from conversations with my supervisors about the expectations being placed on the design work - and from my own discomfort with those expectations.
Looking back now, I can see that I was beginning to articulate what would become a central theme of the doctoral research: the limits of design's "making visible" orthodoxy. Morgan's framework helped me understand the dynamics as structural rather than personal, which was important for maintaining the critical distance needed to continue the work.
The connection to Experio Worklab is significant. The pattern I describe - design as cover for transformation that doesn't happen - would prove relevant not just to SCÖ but to the broader question of why service design seems to succeed at the level of method or myth while failing at the level of change. The miracle-worker metaphor is part of that story.
The question posed at the end - what metaphors structure expectations of design itself - would be taken up in the NORDES paper with Ana Kustrak, where we developed a typology of generative metaphors for the service designer role in Swedish public sector contexts.