Chapter 4 - The Materialisation of Expertise in European Infrastructure Advisory

4.1 How European advisory practice constitutes itself

Expertise in European infrastructure advisory does not arrive pre-formed 💡 . In this chapter we describe how it is materialised, through what we’ll call the integrated three-scale system: an architecture in which the individual consultant, the firm as knowledge-managing collective, and the domain as public-interest tradition are mutually constitutive. The three scales produce one another rather than standing as independent layers: the individual is individuated by the firm and by the domain, the firm is individuated by the individuals it forms and by the domain it operates within, and the domain is individuated by the firms that work within it 💡 .

This chapter is descriptive. Chapters 5 and 7 take up what AI mediation is doing to the system; here I set out the system as it has historically operated, so that the analytical chapters have ground to stand on. 💡

We will draw on three sources. First, the philosophical framework we assembled in Chapter 3: Stiegler on individuation across scales 1 2 , Polanyi and Collins on tacit and contributory expertise 3 4 , MacIntyre on practices and internal goods 5 . Second, the institutional architecture Chapter 2 introduced: the ChMC framework and European infrastructure advisory as a distinctive domain. Third, the lived material from my own practice and from observation of colleagues across firms and sectors, marked clearly where it operates. I add the integration itself: the system is one formation whose character depends on the three scales constituting one another.

4.2 The individual scale: how a consultant comes into being

The individual scale is the level of the consultant (the professional formation and identity that Chapters 5 and 7 will examine directly). The consultant is not a static professional persona but an individuation: an ongoing process of becoming a particular kind of practitioner, carried forward through formative work over the arc of a career. Ibarra’s study (Chapter 3) describes that becoming at close range: junior professionals observe role models, experiment with provisional versions of themselves, and judge the results against their own standards and the feedback of others 6 .

Individual formation in professional advisory work runs through the same components in every context. The pattern 💡 is the general mechanism through which advisers in any domain develop contributory expertise. The content varies: the regulatory vocabulary, the institutional settings, the form of public-interest obligation, and the formative environments specific to the domain.

In European infrastructure advisory, the consultant’s individuation has three components.

The first is formal training. Consultants enter the profession with formal training 💡 and continue it across the career through certifications, professional development, and the institutional architectures of chartering. This training supplies the analytical apparatus and disciplinary vocabulary the consultant brings to the engagement. It is necessary but not sufficient: it sets the conditions under which contributory expertise can develop; the expertise itself develops through engagement with consequence. 💡

The second is therefore engagement with consequential work. Consultants develop contributory expertise through the work itself: building tariff models from blank documents and learning from the corrections, drafting memos that fail and producing revised drafts that land, attending regulator meetings and learning what to listen for, presenting to clients and learning when to hold a position and when to retreat. Contributory expertise (in Collins’s sense 4 , the ability to contribute to the practice through embodied engagement) is built through work of this kind, not through formal training.

Stockholm, 2018. A young analyst runs a stakeholder mapping workshop for an electricity transmission regulatory engagement. She has prepared the materials, rehearsed the framing, structured the agenda carefully. The workshop produces a clean map: actors sorted by interest and influence, colour-coded, annotated. The partner reviews it afterwards and asks one question: where is the grid operator? The national transmission operator is not in the engagement scope (it is not party to the pricing review), but its investment programme constrains every strategic option the client is considering, and the households and firms at the end of the network will pay for whatever is decided; any recommendation that ignores the investment pipeline is advisory in name only. He does not redraw the map for her. The analyst had mapped the regulatory space. She had not yet learned that the regulatory space sits inside a physical network with its own logic, and that learning to see both at once is what distinguishes an infrastructure regulatory analyst from a regulatory analyst who happens to work on infrastructure. She spends that evening redrafting the map herself. She has never since produced a stakeholder analysis without asking what the physical network requires, and she now asks her own juniors the same question, in the same words.

This kind of work is formative in the Stieglerian sense: it is how the consultant individuates as a particular kind of practitioner. It is shaped by the firm (which decides what engagements are available and how supervision operates), by the domain (which supplies the substance the consultant works on), and by the wider professional tradition the consultant is being formed into.

The third is senior judgment witnessed and absorbed. Individuation also runs through witnessing senior practitioners exercising professional judgment: in client meetings, in internal partner discussions, in the moments when senior practice is visible to the more junior practitioner who is being formed. Senior judgment is partly a matter of what Polanyi calls tacit knowledge 3 : picked up by working alongside rather than taught explicitly, and not always articulable in propositional form.

Recent work in cognitive science backs the durability of this component of formation. Oakley et al. 7 argue that consolidating declarative and procedural memory (the systems on which expert judgment depends) requires sustained engagement with the content over time, and that the schemata absorbed through this consolidation are what enables the practitioner to evaluate, refine, and guide externally available content. On this reading, the senior judgment that infrastructure advisory practice depends on consists of long-consolidated memory systems rather than a static stock of information, and interactional fluency with external systems cannot shortcut their production.

Brussels, 2020. An advisory team is preparing a formal regulatory challenge on behalf of a Belgian energy client. The commercial case is sound: the numbers support it, the procedural grounds are defensible, the regulatory determination has a genuine weakness the challenge would expose. The senior partner (who has worked with the federal energy regulator since it was set up) says: not now. He cannot fully articulate why. Something about the change of leadership at the regulator, about a relationship the client cannot afford to damage before the next regulatory period opens, about a signal in the new chair’s recent speeches that the team has not registered. The team thinks he is overcautious. Eighteen months later the regulator opens a sector-wide market investigation that directly affects the client’s revenue stream. The partner had read something in the institutional temperature that existed only in the accumulated residue of two decades of working with that one institution. It was in no document, could not have been transferred through a briefing, and saved the client a great deal. The junior who took the minutes that day still cannot say what he saw; she knows only that she wants to be able to see it. 💡

The three components together produce the consultant as a particular kind of practitioner: none of them alone, but their integration across a career, mediated by the firm and the domain within which that career runs.

The ChMC three-stage architecture (Foundation, Applied, Chartered) is the institutional framework within which this individuation is structured 8 . The Foundation stage is where the formative engagements are most concentrated and the consultant develops the basic tacit-knowledge ground that contributory expertise builds on. At the Applied stage the consultant develops autonomy: the formative engagements move from highly supervised to more independent, and the consultant begins to exercise the professional judgment that Chartered practice will then operate on. At the Chartered stage that judgment becomes the consultant’s primary contribution, exercised in the contexts the later chapters take up.

The formative engagements contribute more than content: they contribute disorder. The model built from a blank document and the memo that fails, the engagements of the second component above, are formative because they go wrong first: the wrong structure, the assumption that does not hold, the draft that does not land. The profession has recently found a name for this. The HEC Alumni white paper (Chapter 2) observes that juniors who receive polished, ready-to-use answers from a copilot “ne traversent plus le ‘chaos structurant’ qui formait autrefois leur jugement” 9 , and that critical sense, which used to be exercised “tout au long de la construction” of a proposal or deliverable, now meets a draft that is largely correct and “arrive d’un seul bloc” 9 . Structuring chaos is a precise term for the temporal content of formation: the duration of doing the work badly, under correction, until the corrections become one’s own. It is what integrates the three components above, and it is what an adequate first draft removes. The term is the profession’s, and I adopt it. 💡

4.3 The organisational scale: the firm as a knowledge-managing collective

The organisational scale of formation (the firm as a knowledge-managing collective) looks much the same across the advisory landscape. Large firms in every sector maintain methodological architectures, case databases, and apprenticeship systems through which a distinctive way of seeing passes from one generation of practitioners to the next. Boutiques concentrate that distinctiveness in the accumulated judgment of a small number of senior practitioners whose psychic individuation the firm depends on. The mechanism is general; an infrastructure advisory boutique and the strategy arm of a large firm differ in the identity at stake, not in the mechanism that reproduces it and now puts it at risk. The analytical chapters engage this identity directly.

For the organisational scale, we will draw on Nonaka and Takeuchi’s SECI model 10 11 as the device for describing how consulting firms manage knowledge. It identifies four modes through which knowledge moves between tacit and explicit forms across the organisation:

1. Socialisation (tacit to tacit, through shared experience and apprenticeship)
2. Externalisation (tacit to explicit, through articulation into transmissible forms)
3. Combination (explicit to explicit, through recombination of articulated forms)
4. Internalisation (explicit to tacit, through absorption into individual practice).

The four modes operate as a continuous loop 💡 through which the firm’s knowledge is reproduced across time.

A mid-sized London advisory firm specialising in regulated infrastructure has a concept it calls “the regulatory clock”: an informal framework for reading where any given regulatory process is in its cycle and what that implies for what can and cannot be moved. It is not in any playbook. It is not on any slide. It lives in the onboarding conversations senior consultants have with junior ones, in the casual annotations partners add to client briefing notes, in the way strategy meetings are opened: where are we in the clock? Clients hear it as the firm’s way of thinking, and some hire the firm for it. An analyst joins from a large management consultancy with sophisticated knowledge management systems: databases, playbooks, frameworks for everything. She spends three months confused by the clock’s absence from any document. By month six she is using it instinctively, though she cannot say precisely where she learned it. By year three she is teaching it to someone new, in exactly the kind of onboarding conversation she cannot remember having been in. 💡 💡

To the standard SECI account we can add a reading of organisational knowledge management as identity-producing practice 💡 , not merely as productivity-supporting infrastructure. Through the SECI cycle the firm reproduces not just knowledge that supports productive work but whatever makes it a particular kind of firm: its framings, its sectoral instincts, its methodological reflexes, the vocabulary in which it recognises a situation. These make one infrastructure advisory practice distinguishable from another even when they sell similar services.

This identity-producing function has been underrecognised in the existing literature on consulting under AI mediation. Information-systems research on AI in consulting, including Tuczek et al. 12 , engages organisational knowledge management at the level of productivity and efficiency. We can engage it at the level of identity: what the firm is as a knowledge-managing collective, and what is at stake when that identity comes under contemporary AI mediation.

Market-facing analysis of the same mechanism has begun to recognise the learning loop at the operational level without arriving at the identity question. Fuller’s prescription for AI-mediated consulting practice describes, from the side of an AI tooling vendor, what the externalisation phase of the SECI model produces when it runs through a firm-controlled loop: every engagement recorded by agents as a byproduct, anonymised, and recirculated to future projects, so that the firm’s accumulated knowledge compounds through its own models, including post-trained open-source ones, rather than locking the firm into a single lab’s 13 . 💡 The prescription is commercially coherent and, for large well-capitalised firms with the infrastructure to build and govern proprietary AI capability, is a real operational response to the extraction dynamic. But it is available only to a minority of the consulting ecosystem, and it engages the SECI cycle at the productivity level: it describes how knowledge is retained and recirculated, not what kind of firm that knowledge constitutes. The identity-producing dimension 💡 requires the reading we are adding here.

Firm-distinctive identity lives in four places.

• Internal vocabulary and cultural identity. The internal vocabulary, the shared reference points, the house ways of framing a problem: the firm’s collective way of talking is part of what makes its analytical work its own.
• Senior practice and apprenticeship. Foundation-stage consultants are formed not only through formal training but by working under senior practitioners whose analytical voice they absorb. Ibarra’s juniors described that absorption as observation: watching seniors, they built “a store of tacit knowledge, attitudes, routines, impression management techniques” for the role 6 . Apprenticeship (partner mentorship, supervisory practices, the rituals of senior-to-junior engagement) is how the firm’s identity passes across generations.
• Case databases and institutional memory. Firms keep a memory of past engagements: case studies, internal documentation, a working knowledge of which approaches have worked where. It is more than an archive: it is how the firm’s accumulated experience reaches current practitioners, and how Foundation-stage and Applied-stage consultants are formed in the firm’s way of seeing.
• Methodologies and frameworks. Consulting firms develop methodological apparatus (analytical frameworks, sectoral playbooks, structured approaches to recurring engagement types) that constitutes the firm’s own way of approaching the work. The methodologies are quite specific: they bear the marks of the firm that produced them and distinguish one firm’s analytical voice from another’s.

The organisational scale is not homogeneous. The distinction between large advisory firms and SME and individual boutique practices governs how the proletarianisation risk distributes across the sector.

Larger firms

In large advisory firms 💡 , savoir-faire has historically been organised through the four structures above. The firm’s competitive asset sits substantially in that knowledge architecture: methodologies and frameworks, case databases, training curricula, the SECI cycle through which its way of seeing is reproduced. Individual senior practitioners carry important personal judgment, but the firm’s identity as a knowledge-managing collective depends on the structures that reproduce its character from one cohort to the next, not on any one of them. For large firms, the Stieglerian mechanism identified in Chapter 3 operates at this institutional level: the firm’s externalised knowledge and its formal formation pipeline stand exposed when AI mediation substitutes for the formative engagements through which junior practitioners are made into senior ones.

The financial logic that has sustained this pipeline is also under pressure. The traditional leverage model (large junior cohorts whose billing subsidises the cost of formation, with attrition leaving the few who progress to partner) depends on the economic case for junior work. When the entry-level work is automated, that case weakens, and the institutional vehicle for the formation pipeline is itself at risk. Tarki and Raczynski (2025) report the same pressure in HBR: many professional services firms are cutting entry-level hiring in the belief that AI can automate much of the junior work, without a real plan for what the cut will do to their partnership pipeline 14 . 💡

The pyramid’s replacement is already being drawn, and the drawing is instructive. The HEC Alumni white paper replaces the pyramid with an “obélisque”, a leaner model “avec moins de niveaux hiérarchiques et des équipes plus restreintes” (with fewer hierarchical levels and smaller teams), and with new roles such as an AI strategy director or a chief AI ethics officer 9 . Read as a shape, it keeps the summit and narrows the base. The image answers the question it was drawn for, which is a question about margin: once a model does the junior’s work, the base is expense without leverage. Blangeois and Lebrument’s study of the earnings calls of seven large digital-services firms names the bind a “self-cannibalization paradox”: firms “must adopt a technology that erodes their effort-based business model to remain competitive” 15 . The effort-based model being eroded is the one that paid for the base. 💡

A drawing made to answer the margin question cannot answer the one this chapter asks. The pyramid did three things nobody designed it to do. The first, financing early learning and apprenticeships out of billable hours, is the leverage model just described. It also absorbed junior error at a price the engagement could carry, and it supplied the escalation boundaries, the problems one cannot yet solve but learns to recognise, through which knowledge hierarchies are crossed 16 . An obelisk has seniors and no account of where they come from, so the missing base in the profession’s own successor image is the formation question in visual form rather than a flaw in the drawing. 💡

Smaller organisations

In SME and individual boutique practices, the knowledge architecture is organised differently. The boutique’s value typically resides in the accumulated personal experience and judgment of one or a small number of senior practitioners: decades of knowing a specific regulatory environment, a specific type of project, a specific stakeholder field. Externalisation is minimal and often actively undesirable: the boutique’s competitive asset is the individual’s personal judgment, not a transferable framework a competitor could reproduce. 💡 The SECI cycle is present in attenuated form; what cannot be attenuated is the senior practitioner’s individuation: the distinctive analytical voice, the relational intelligence, the accumulated instinct that makes them who they are.

For boutique practices, the Stieglerian mechanism operates differently. The primary risk is not to a formal formation pipeline (boutiques rarely form practitioners at scale as large firms do) but to the practitioner’s individuation under generic mediation: whether the boutique advisor, using AI tools whose default outputs are common across the market, can keep a personal analytical voice distinct, or whether the cumulative weight of that mediation flattens it. The dis-individuation pressure lands on the person, not the institution, and the defences are correspondingly personal: the individual’s deliberate cultivation of the voice and relational depth that generic mediation tends to smooth away.

So what

The distinction carries into the analytical chapters. Chapter 5 shows generic dis-individuation operating on different substrates (firm-distinctive institutional knowledge in large firms, individual judgment in boutiques), each with its own vulnerabilities and its own forms of resistance. Chapter 7’s formation-pipeline argument applies primarily to large firms; boutiques face the disruption at the individual level, where the question is whether the particular senior practitioner’s judgment survives generic mediation intact.

Neither form keeps its adaptations to itself. An adjustment in one firm’s routines becomes, in time, the condition of the whole field, and that passage is now being observed directly. Killewald and Haskamp’s multiple-case study of two European management consultancies, a large strategy firm and a sector boutique (the two forms we just distinguished), built on thirty-one interviews, derives four mechanisms by which generative AI reconfigures consulting work and integrates them into a framework of “how local adaptations of routines accumulate into broader field-level change” 17 .

1. Negotiated legitimacy shifts what consultants believe is acceptable to use, under the paradox of a field that prizes both efficiency and prudence;
2. provisionality of reliance describes the model called upon under time pressure or a knowledge gap, producing results “formally convincing, yet oftentimes remain[ing] fragile”, with review and validation routines “shortened, bypassed, or readjusted” 17 ;
3. skill polarity is the duality in which the tool lets juniors tackle complex tasks earlier while “practice loops are shortened or skipped, weakening basic training of competencies, such as analytical thinking, critical assessment, and clear argumentation” 17 ;
4. and the contestation of value aggregates these adjustments into a field-level shift in which value moves toward “what cannot be commoditized: using own collected experience to tailor solutions to clients’ situations” 17 .

Two of the four bear directly on this chapter. The third mechanism is the formation argument observed in the field: in “a domain that traditionally relies on a gradual apprenticeship”, the authors write, fundamental skills are built “through repeated practice in close collaboration with and supervision of a few direct colleagues and mentors”, and a manager they interview finds the resulting gaps are “now virtually my job to fix” 17 . The fourth is the propagation the three-scale account requires: what one firm normalises in its delivery routines becomes, through clients, staff mobility and competitive imitation, what the domain expects, and what the domain expects is what the next cohort is formed into.

Read together, the third and fourth mechanisms say something the study does not state outright: the value the field now relocates to, collected experience, is what skill polarity stops producing. In either form, large firm or boutique, the firm is a knowledge-managing collective: the organisational scale of the system. 💡

4.4 The domain scale: what infrastructure advisory actually requires

The domain scale is where this work’s primary research site comes into clearest focus. The individual and organisational scales look alike across advisory contexts; the domain scale is specific. European infrastructure advisory is the domain whose public-interest orientation, regulatory architecture, contested knowledge, and long time horizons the analytical chapters work on, and it is where the specific-level argument is most legible, for the reasons Chapters 1 and 2 gave.

Chapter 2 (§2.3) introduced the institutional features of this domain as a public-interest tradition (its public-purpose engineering tradition, regulatory architecture, and EU governance frame); this section asks what those features mean for the character of practitioner knowledge and what they require of the formation pathway. The domain supplies the substance: the problems the individual consultant develops contributory expertise in, and the material the firm reproduces through its knowledge management.

Four components constitute the domain.

Variation across subdomains. The family of European infrastructure subdomains varies in regulatory weight, time horizon, stakeholder shape, and decision character. Energy and water are the most heavily regulated, with an extensive apparatus of price regulation, service standards, and public-interest obligations. Defence is institutionally distinctive, operating through procurement, security clearances, and strategic-autonomy concerns. Transport is moderately regulated with subdomain variation between rail and road. Environmental, social, and urban subdomains have variable regulatory architectures depending on the specific institutional context. I keep this variation in view throughout the analytical chapters (the argument fits the whole family).

Paris, 2017. An advisory firm is engaged on procurement governance for a national nuclear decommissioning programme: a sixty-year horizon, multiple generations of contractors, oversight by the nuclear safety authority layered over the commercial governance of the operator and the state research body. The first client meeting lasts four hours. The senior consultant from the advisory firm does not speak for the first ninety minutes. She is learning the interlocking governance structure before she says anything useful. A colleague who has advised on airport privatisation and toll roads thinks this is excessive caution. She explains afterwards: in nuclear, a wrong framing embeds itself in a procurement framework that governs decisions thirty years from now. The consequence of getting the framing wrong is not merely a bad report but a governance structure that cannot be corrected by anyone who comes after us, answered for by people who were not in the room and some who are not yet born. The time horizon is a constraint on what kind of advice is responsible, not just a parameter in the model. The junior beside her learns more from the ninety minutes of silence than from her four pages of notes.

The unifying public-interest texture. Despite subdomain variation, the family is unified by the public-interest tradition that Chapter 2 introduced. The work 💡 is shared across the subdomains, and it is what distinguishes the domain from generic management consulting. 💡

The dual character of technical depth and political judgment. European infrastructure advisory sits at the intersection of technical depth (the analytical apparatus required for engineering, financial, regulatory, and environmental analysis) and political judgment (the understanding of the political and institutional contexts within which infrastructure decisions are made). Neither alone constitutes the domain; their integration sets infrastructure advisory apart from generic technical consulting on one side and generic political consulting on the other.

The role of contested knowledge. Infrastructure decisions are contested. The same engagement can be analysed in multiple ways: through cost-benefit, equity, climate-transition, or strategic-autonomy frameworks. Advisory work happens where that contest is played out: public consultations, regulatory hearings, multi-stakeholder processes, intergovernmental negotiations. Contested knowledge is the ground the work stands on, not a problem to be solved.

London, 2019. An advisory firm is commissioned by a statutory body to produce an economic assessment of an airport third runway. The technical case is broadly supportable. The climate compliance question (whether the expansion is compatible with the national carbon budgets) is contested between government and its statutory climate advisers, and no client brief has resolved it. In a public hearing before a parliamentary committee, the senior partner is asked directly whether the firm’s analysis assumes climate compliance or treats it as a separate policy question. He says it is a separate policy question. He is technically correct; the firm was not commissioned to answer it. He is also, in saying so, taking a position in a genuine disagreement about what the advisory role is for: whether infrastructure advisers are responsible for the consequences of their analytical framings for the publics affected or only for the technical correctness of their models. The room knows it, and so does he. So does the analyst who prepared his briefing pack, watching a partner decide in public what the firm answers for. The answer he gives is defensible and incomplete and he has thought about it since.

The domain scale has not been static. Energy and rail deregulation waves, the privatisation of utility companies, and the cumulative weight of EU governance architecture shifting from liberalisation to climate transition have reshaped the knowledge requirements and formation conditions of European infrastructure advisory over three decades. AI mediation arrives into a domain already in transformation, not a settled system, and adds a further pressure to those underway. 💡

These four components together make up European infrastructure advisory as a domain. Chapter 5 will argue that the domain’s distinctive cultural and cognitive character (the shared vocabulary, the sectoral instincts, the contested-knowledge frameworks through which the domain recognises and engages its problems) is most exposed to generic dis-individuation: the symbolic misery that results when generic technical mediation progressively impoverishes the shared cognitive world through which a domain’s distinctive intellectual life is experienced and sustained 18 .

4.5 How the three scales constitute one another

The Hague, 2017–present. A Dutch advisory practice founded in 1998 has worked almost exclusively in water and environmental infrastructure: drinking water, wastewater treatment, flood protection, water board governance. Over twenty years it has developed a way of framing regulatory problems that treats the infrastructure asset as a service-delivery system first and a cost-recovery problem second. This is the framing the Dutch water sector’s publieke voorzieningen tradition produces rather than the default framing in regulatory economics: the tradition the firm grew up in, the tradition absorbed through decades of client relationships with the water boards and the national ministry, the tradition that shapes which problems infrastructure clients bring to this firm and which they take elsewhere. A new partner joins in 2019 with a background in energy regulatory economics: rate-of-return frameworks, WACC calculations, efficiency benchmarking. She is capable. Within eighteen months she is framing water regulatory problems in the language of service delivery, not of rate of return. Nobody has instructed her to do so. The clients ask questions in that language. The firm’s reputation is built on answering in that language. The domain, the organisation, and a twenty-year history of individual consultants being formed through that domain have together produced a way of seeing that recruits her without coercion. She brings her energy expertise to that framing but leaves the energy framing behind. A style guide could describe that framing; only the work and the people who do it keep it alive. The three scales have constituted one another so completely that the firm cannot tell you where its approach came from; it can only show you that it works.

Three relationships carry the mutual constitution:

The individual is individuated by the firm and by the domain. The professional identity 💡 of the individual consultant 💡 is produced through the formative engagements the firm makes available, through the supervisory practices the firm operates, and through the problems the domain puts in front of them, never in isolation. The Foundation-stage consultant formed at one firm in one subdomain becomes a different practitioner from one formed at another firm in another subdomain, even where their formal training and personal capacities are equivalent. This individuation is integrated and cannot be abstracted from the firm and the domain within which it operates.

The firm is individuated by the individuals it forms and by the domain it operates within. The identity of the firm 💡 is reproduced across generations through the practitioners it forms and shaped by the domain within which it operates. It is not a static inheritance: a firm that has worked extensively on European water utility regulation develops a repertoire distinct from one that has worked on European energy market design, even where their methodological apparatus and institutional architectures are formally equivalent. The firm’s individuation is inseparable from its domain.

The domain is individuated by the firms that work within it and by the wider European public-interest tradition. The domain is reproduced through the engagements the firms within it produce, through the regulatory and political institutions that frame it, and through the wider European public-interest tradition it belongs to. Nor is this a static institutional reality: the domain at any moment is what its firms have been making of it.

Together the three relationships make the system integrated. Abstract any one scale from the others and you lose what constitutes legitimate practice in the domain. One consequence matters for everything that follows: a change at any one scale propagates to the other two. Change how the individual consultant is formed and you change the firm, whose identity is reproduced through the people it forms, and the domain, whose character depends on those firms. Change how the firm reproduces its way of seeing and you change the individuals formed within it and the domain that depends on firms seeing distinctly. Change the domain and you change the firms working within it and the individuals being formed in those firms.

4.6 The integrated three-scale system

European infrastructure advisory materialises its expertise through three mutually constitutive scales.

• At the individual scale, formation proceeds through formal training, sustained engagement with consequential work, and senior judgment absorbed across the arc of a career.
• At the organisational scale, the SECI cycle turns tacit knowledge into institutional assets (methodologies, case databases, cultural vocabulary) while apprenticeship and senior practice continuously replenish what those assets cannot hold.
• At the domain scale, public-interest texture and contested knowledge keep the firm grounded in something larger than its own institutional identity: the European regulatory tradition and the third box whose conditions of life the work shapes.

We will ask in Chapter 5 what contemporary AI mediation is doing to this integrated system: whether generic mediation operating across all three scales simultaneously threatens the sovereignty of the integrated identity, and whether the extraction dynamic 💡 is the mechanism through which the political economy of AI mediation operates on European infrastructure advisory specifically. Chapter 7 follows the consequences through the three ChMC stages of professional progression.