Norm Production: An Operational Map of Transformation, Part II

Document version: 0.02-alpha

Series: Norm production. Part II of II. Part I (infrastructure) published 26 March 2026.

Whoever built the infrastructure had better have something to put inside it.

This text describes the mechanisms of symbolic transformation symmetrically, without indicating in which direction to apply them. Who applies them, and to what end, is the reader’s choice. The time horizons in the Part IV examples are drawn from specific cases and are not normative.

Every example here refers to the American context, because the American system for producing cultural norms is at present the most developed and the best documented. Other systems partly reproduce its structure and move in the same direction with a lag. The examples are local; the mechanisms are general.

The framework is built for practical use, not for academic completeness. Many mechanisms are described in simplified form. That is the unavoidable price of operationality. Like any framework, this one reduces reality to a set of levers. The instrument describes the mechanics. The content of the code and the intended outcome lie outside its competence.

Part I. The CulturalBI framework v0.01-alpha (Module II)

Foundation

The first module of the framework describes the infrastructure of norm production. It answers the question "how": how to enter the system, how to accumulate position, how to fix an operational criterion through directive, reflex and a self-sustaining loop of legitimacy. The first module closes where the next question opens. An operational criterion sets the personnel pipeline in motion. But why is one criterion read as legitimate beyond the community that produced it, while another remains foreign? That is a question not about infrastructure but about symbolic codes.

The second module answers it. It translates the academic conclusions of Jeffrey Alexander and adjacent authors in cultural sociology from descriptive into operational language [36]. The Gramscian module applied an engineering optic to infrastructure. The Alexanderian module applies the same optic to symbolic structures: what to assemble, in what order, which parts break first, and which design decisions buy durability. The ideological content has been removed; the design optic has been added.

Eight institutional reports in the Alexanderian series (Disney, Netflix, AMPAS, the Ford Foundation, the NEA, the Iowa Writers’ Workshop, the Recording Academy, the Federalist Society) describe particular codes in particular circumstances. This document does not add a ninth case. It distils the mechanisms of symbolic work that recur regardless of industry or political vector. The eight cases span three industries, two political vectors and horizons ranging from a few years in the newest reforms to eight decades at Iowa. Each mechanism is traceable across several independent cases.

The symbolic level of the norm-production system

The first module described infrastructure on three levels: legitimation, the personnel pipeline, distribution. The symbolic level is not a fourth storey of that building. It is another dimension of the same building. Infrastructure answers the question of who controls production. The symbolic level answers the question of why what is produced convinces.

The central dynamic of the symbolic level is re-fusion and de-fusion. An institution performs its code in a ritual. If the audience believes the performer believes what is being performed, re-fusion occurs: the performance convinces. If the audience doubts, de-fusion occurs: the seam between the performer’s intention and the audience’s perception becomes visible.

Alexander decomposes performance into six elements: script, actor, audience, means of symbolic production, mise-en-scène, social power [36]. Re-fusion occurs when all six coincide in a single event. De-fusion occurs when even one diverges. The contrast between two AMPAS ceremonies shows the mechanism in pure form. At the 86th ceremony (2 March 2014, 43.7 million viewers [1]) the host broke the script with an everyday gesture. The selfie became the most retweeted image of its time [2], the boundary between the auditorium and the television viewer dissolved, and all six elements converged. That is re-fusion in its maximal form. At the 93rd ceremony (25 April 2021, 10.4 million viewers [3]) the venue moved to a railway station, there was no host, the room stood half empty, the camera in an empty space exposed the construction, and the elements of the performance came apart. The ritual was executed, re-fusion did not follow, and the ceremony went into de-fusion. [AMPAS]

Re-fusion cannot be engineered directly. It can only be prepared for, by creating the conditions. The five mechanisms below describe precisely those conditions. Each of them, when present, raises the probability of durable re-fusion. Each of them, when absent, opens a specific line of de-fusion.

The eight cases and how they distribute

The eight institutions of the series distribute across six ritual types and three audience types. The ritual type sets the main channel through which the code becomes visible. The audience type sets who verifies the performance of the code from outside.

The distribution covers four broad vectors. Commercial institutions with a mass audience (Disney, Netflix). Ceremonial arbiters with a dual audience (AMPAS, the Recording Academy). Grantmaking institutions with a professional audience (Ford, the NEA). Institutions with a closed professional audience and a long transmission cycle (Iowa, the Federalist Society).

The sample is not representative of every type of institution. Religious traditions, academic schools, genre networks in media, and military, medical and legal corporations operate in other registers. The CulturalBI series was selected by the project’s thematic criterion — the production of norms in the American public cultural sphere. The mechanisms described below have been tested on this sample. Transferring them to other types of institution requires separate testing.

The binary codes of the eight institutions are set out in the table below. The fourth column records the operationality of the sacred pole, the variable to which Mechanism 3 and Part III return.

Precondition 2. Structural incompleteness of the code

The first module begins with the precondition of a normative vacuum: an organisation recognised as an arbiter of quality leans on an abstract principle where an operational criterion is required. That is the infrastructural formulation. At the symbolic level the vacuum has an exact counterpart — structural incompleteness of the code.

Structural incompleteness of the code is the condition in which the narrative an institution carries is under-articulated on at least one of the seven anatomical layers. The seven layers are examined in detail in Part II. Here it is enough to know that a code may be complete or incomplete by the number of layers worked through, and that the layer of the operationality of the sacred is the decisive one among them. The seven layers are neither equivalent nor simultaneously necessary. But incompleteness accumulating across several layers leaves a code vulnerable to pressure that a complete structure withstands.

The link between the two preconditions is direct. The normative vacuum of the first module is a special case of structural incompleteness. The absence of an operational criterion means incompleteness on the layer of the operationality of the sacred. The Recording Academy held an empty sacred signified in its rule for nearly seven decades, from its founding in 1957 [10]. The sacred was defined by negation alone — "without regard to album sales" [11]. This is at once a normative vacuum in the terms of the first module and an extreme form of structural incompleteness in the terms of the second. [Recording Academy]

Structural incompleteness, like a vacuum, can be filled without a durable result. Pressure, scandal or a change of composition can displace an old code and put a declaration in its place. But replacing one incomplete code with another reproduces the vulnerability in a new form. The Recording Academy replaced the empty centre of the peer habitus with a demographic proxy metric [16]. One unspoken principle gave way to another unspoken principle. The centre remained empty. [Recording Academy] Within the sample, structural completeness is associated not with transformation itself but with its durability. The strength of that association is limited by the single complete case and is examined under Mechanism 1.

Three conditions for symbolic transformation

Symbolic transformation is triggered when three conditions hold at once. They run parallel to the three conditions of the first module and are tested on the same cases.

• •The existing code is structurally incomplete or has undergone de-fusion. The performance has stopped convincing a sufficient number of participants.
• •Pressure — external, internal or both — that makes the incompleteness visible.
• •A change agent carrying a code that reads as an answer to that pressure.

The third condition needs qualifying. To trigger transformation, the agent’s code need not be more complete or more operational than the one it displaces. The sample shows this directly. The Recording Academy and AMPAS both changed code under pressure, and in both cases the incoming code replaced an empty centre with another emptiness. Completeness and operationality determine not the launch but the hold. They are examined under Mechanism 1 and in Part III. [Recording Academy], [AMPAS]

If even one of the three conditions is missing, the process does not start. Structural incompleteness can persist for decades. The Recording Academy lived with an empty centre for roughly six decades — from the first ceremony in 1959 [10] to the 2018 press conference that made the emptiness visible [14]. [Recording Academy] AMPAS held a latent code of professional excellence for more than eight decades, from its founding in 1927, before #OscarsSoWhite supplied the trigger [6]. [AMPAS]

States of the sacred

From the standpoint of symbolic work, the sacred pole exists in three states. The state determines which test the code withstands and which it does not.

Operational sacred admits independent verification. The procedure is open to a third party who can apply it and assess the result. This is a property of the procedure, not a guarantee of a single answer. Carriers may diverge, and the degree of their convergence can itself be disputed. The originalism of the Federalist Society works this way. Original public meaning is reconstructed from period dictionaries, ratification debates and legal texts of the era [26], and although originalists diverge on particular cases, the criterion itself remains open to third-party verification. Confirmation hearings test whether a candidate applies that procedure, and such a test is available to examiners of differing political positions [27]. [Federalist Society]

Iconic sacred admits no procedural verification but has content. It rests on recognition rather than on a test. The content lives in the habitus of a homogeneous community of carriers and is written down nowhere as a rule. Iowa has neither manifesto nor criteria, yet the criterion by which prose is selected can be reconstructed from outside with considerable precision: the concrete above the abstract, private experience above polemic, character above idea [34]. The code exists as a group habitus and is reproduced through the workshop format [33]. [Iowa Writers’ Workshop]

Empty sacred has no content. It is set by negation or by a proxy metric and rests on the presumption that the carriers "know". Until 2018 the Recording Academy looked like an iconic case: colleagues vote, and colleagues know. Yet the rule had been empty since 1957, the sacred defined in it by negation alone — "without regard to album sales" [11]. In the early decades the content was supplied by the habitus of a small homogeneous electorate: the first ceremony of 1959 was held simultaneously in Beverly Hills and New York, around two founding chapters whose biographies overlapped [10]. This was an iconic sacred laid over an empty rule. As the electorate grew and differentiated by genre, the habitus stopped producing a result the industry recognised as professional. The secret nominating committees of fifteen to thirty people, introduced in 1989 for genre categories and in 1995 for the Big Four [12], restored a small homogeneous group where the large one had lost it. This is a prosthesis for the iconic sacred, and its appearance dates the erosion of content. By 2018, when pressure demanded that the criterion be named, there was nothing to name; only the rule remained. The difference between Iowa and the Recording Academy runs exactly here. One unwritten code is reproduced by a format and holds for eight decades. The other rested on homogeneity, lost it, and failed to notice the loss. [Recording Academy]

The three states protect a code differently. The operational is robust against substantive criticism and can be defended by argument, but it is vulnerable to attack precisely because it is documented. The iconic is robust against direct attack — there is nothing to attack — but vulnerable to erosion from within: if the community of carriers disperses or loses its homogeneity, the content vanishes with it, quietly and without announcement. The empty is protected neither by argument nor by habitus. At the first demand to name the criterion, it fills with whatever is at hand.

A sacred that is publicly named but not operationalised appears twice in the sample. Ford and the NEA declare content — "structural justice" [43], "inclusive equity" [41] — but have no verification procedure. At Ford the content is lodged in the habitus of program officers, and its sacred is closer to the iconic. At the NEA the content rests on a document alone and disappears when the document changes, which brings it closer to the empty [42]. [Ford Foundation], [NEA]

Note. In practice the states of the sacred are almost never pure. A single institution may combine operational, iconic and empty elements and shift the balance over time. The division is a convention, introduced for analytical clarity.

Part II. The mechanisms of symbolic work

Anatomy of a code: seven layers

Before describing the mechanisms, a grid is needed. Any institutional code decomposes into seven layers. The layers provide a language for describing structural completeness and a point at which incompleteness can be diagnosed.

Layer 1. Narrative structure. Seven components: initial state, crisis, hero, antagonist, method, sacrifice, restoration. The Federalist Society deployed all seven: the Constitution of 1787, the Warren Court, the methodologically disciplined jurist, judicial activism, originalism, the Bork hearings of 1987 [27], a line of decisions from Heller [29] to Dobbs [30]. The Recording Academy is folded up: the initial state is vague, the crisis undated, hero and antagonist unarticulated. [Federalist Society], [Recording Academy]

Layer 2. Sacred categories. What is declared worthy. The Federalist Society defines the sacred positively and hierarchically. The Recording Academy defines the sacred by negation alone [11]. [Federalist Society], [Recording Academy]

Layer 3. Profane categories. What is declared unworthy. At the Federalist Society every sacred category has a profane counterpart. The original is opposed by activism, neutral principles by result-oriented adjudication, judicial restraint by the imperial judiciary. The symmetry makes the narrative applicable to any case. [Federalist Society]

Layer 4. Hierarchy of categories. The division of labour among categories. Original public meaning as the fundamental criterion, neutral principles as the test of honest application [26], judicial restraint as the meta-rule about limits, constitutional fidelity as a virtue category with moral charge. At most institutions in the series the hierarchy is undeveloped and the categories work in parallel. [Federalist Society]

Layer 5. Trauma claim. The appropriation of collective trauma as a source of legitimacy. Alexander and Eyerman describe this mechanism [37]. A trauma is declared "ours", and the institution positions itself as its keeper and healer. The claim has two layers — a supra-group framing and a specific group — and in the sample both layers are present at every institution. What differs is which layer is pushed to the foreground and carries the load. The Federalist Society foregrounds the trauma of the constitutional order, supra-group in form, and ties it to specific groups in the background. Disney, Netflix and AMPAS foreground a group claim — "historic underrepresentation" — with a universalist frame of justice in the background. [Federalist Society], [Disney], [Netflix]

Layer 6. Boundary. Who is inside and who is outside. Lamont distinguishes three types of boundary: moral (worthy or unworthy), cultural (educated or uneducated), socioeconomic (successful or marginal) [39]. A boundary with an operational marker yields a stable distinction across wide demographic variation. A demographic boundary is vulnerable to counter-claims from other groups. [Federalist Society], [Netflix]

Layer 7. Rules of self-utterance. How the narrative speaks about itself: declarable, semi-declarable or constitutively undeclarable. The Federalist Society unfolds the narrative inside the network and compresses it at hearings — a two-register utterance [28]. The Recording Academy is constitutively undeclarable: uttering the commercial genealogy of 1957 would destroy the promotional function [10]. [Federalist Society], [Recording Academy]

Running through all seven layers is a transverse property — the operationality of the sacred (Part I). It is not a separate layer but a characteristic of layers 2 and 4. Sacred categories either admit independent verification or they do not.

Mechanism 1. Structural completeness as a resource for holding

Structural completeness is the articulation of a code across all seven layers at once. A complete architecture supplies internal resources for reproduction and adaptation that incomplete structures do not possess.

Completeness is measured on two scales, and they must not be conflated. The first is the seven anatomical layers. The second is the seven components of narrative structure inside the first layer. An institution may hold all seven narrative components and still be incomplete on the layer of hierarchy or of utterance.

In the series sample, one institution holds a complete structure across all seven layers: the Federalist Society. The narrative survives across a forty-year horizon [28]. The other seven institutions present incompleteness of varying degree. The Recording Academy is folded up on nearly every layer. Before 2016 AMPAS held three of the seven narrative components; after Academy Aperture 2020 [8] it contains two partly incompatible structures at once. Disney and Netflix carry DEI components in corporate rather than dramatic form, without sacrifice or a final point of restoration. Iowa is iconic: the hero is supplied by the canon of alumni, while the antagonist — Soviet collectivism — was set by Engle in the 1950s and has not been articulated publicly since 1967 [33]. There is neither sacrifice nor restoration. [Federalist Society], [Recording Academy], [AMPAS], [Disney], [Netflix], [Iowa Writers’ Workshop]

Activation condition

• •The code is articulated on most of the seven layers.
• •The layers support one another rather than working in isolation.

Completeness as a resource, not as a cause

Structural completeness on its own does not explain the survival of a narrative over a long horizon. Otherwise the reasoning would be circular. Completeness is defined through the single complete case, and causal force cannot be ascribed to it on the basis of that same case. The sample contains one complete code, so the independent contribution of completeness cannot be separated from the other features of that case by the sample’s own means. Completeness is offered below as a resource consistent with the single complete case, not as a demonstrated necessary condition. Explaining the hold requires three distinct causal layers at once.

First layer. Conditions of initial formation. A specific historical moment. For the Federalist Society this was the CLS vacuum of 1971–1980, a ready articulating text (Bork, "Neutral Principles", 1971 [26]), institutional infrastructure (Yale, Chicago), financial resources and a political moment. These conditions are non-repeatable. [Federalist Society]

Second layer. Structural completeness as a resource for holding. A complete architecture supplies three specific resources an incomplete one lacks. A symmetrical profane pole allows new situations to be classified without recourse to the code’s author: any new case can be sorted into activism or restraint by a rank-and-file carrier. A hierarchy of categories turns a set of slogans into a procedure, and the carrier knows which category settles which type of question. The components of sacrifice and restoration supply a mobilising memory: the Bork hearings of 1987 work as a date around which the network reassembles itself each generation [27]. In the sample these resources are observed in one complete case and function as a condition of holding, not as an independently demonstrated cause. [Federalist Society]

Third layer. Continuing structural and environmental conditions. The environment, not the narrative. For the Federalist Society this is Article III, which grants a horizon of thirty to forty years per appointee [44]. The hold also depends on administrations able to recruit judges through the network, and on the ability to recruit new generations. [Federalist Society]

Weakness in any of the three layers weakens the hold. This rescues the argument from circularity. Each factor has a bounded explanatory role and predicts the change under which it will stop working.

A counterfactual test through mirror institutions confirms the environmental layer of the model in negative form. Networks of conservative legal thought in other systems do not reproduce the result of the Federalist Society. In Canada (the Runnymede Society) [31] and Australia (the Samuel Griffith Society) [32], mandatory judicial retirement at seventy-five [45] and seventy [46] shortens the horizon of the result, and the external support is weakened. In the United Kingdom, parliamentary sovereignty and the absence of a written constitution make the model non-reproducible. These cases vary the environment, not the structural completeness of the code, which has not been measured separately for them. They therefore show the dependence of the hold on the environment but do not isolate the contribution of completeness. The hold is explained by completeness and environment acting together, and the available data do not disentangle the two. [Federalist Society]

[Federalist Society / complete structure], [Recording Academy / folded structure], [AMPAS / partly incompatible structures]

Mechanism 2. Carrier groups: the living bearers of a code

Carrier groups are the specific social groups that bear and transmit a code inside an institution. Not an abstract "company" or "organisation", but specific bearers. The concept belongs to Weber. Alexander borrows it in the theory of cultural trauma to describe the groups that formulate a traumatic narrative and win recognition for it [37].

The distinction between two modes of transmission — directive and reflex — was described in the first module. Here it acquires symbolic content. A directive is dismantled by order. A reflex is not. Carrier groups are where the reflex lives.

Activation condition

• •The code is transmitted through groups possessing institutional resources or professional autonomy.
• •The bearers have adopted the code as their own norm rather than performing it for compensation.

Carrier groups cannot be recalled by order

The variable that explains why one institution holds its code under external pressure while another does not is neither the pressure nor the political camp of the leadership. It is carrier groups. The comparison of the Ford Foundation and HHMI shows the mechanism. In February 2025 HHMI closed the Inclusive Excellence programme without explanation — 60 million dollars, 104 institutions [17]. In those same months Ford stated publicly that the pressure was not stopping it [18]. At HHMI the bearers of the DEI code held no institutional power; the programme was run from above. At Ford the bearers of the code are the institution: program officers carry the code as professional habitus. An order can close a programme. An order cannot change the habitus of staff. [Ford Foundation]

The same explains why dismantling Disney’s formal infrastructure — renaming the BERGs "Belonging" and taking down the Reimagine Tomorrow page in February 2025 [23], the departure of the chief diversity officer in June 2023 [24] — is not equivalent to dismantling the code. The bearers stayed. [Disney]

The carrier-group paradox

Bearers carry a code consistently even when doing so runs against management’s interests. This gives durability and creates vulnerability at the same time. Carrier groups embedded in self-organising structures without a recall mechanism can turn against the institution. In October 2021 the Trans ERG organised a public walkout at Netflix: more than a hundred employees gathered outside the office without leadership’s sanction [19]. To coordinate, the group used internal infrastructure created under the inclusion policy. The bearers applied the code against management’s decision to keep a contested show. [Netflix] The same episode records the mechanics of de-fusion through performance. The co-CEO first defended the decision without acknowledging employees’ pain — the seam between the declared code and the behaviour became visible — and the public reversal that followed did not close the seam but fixed it in place [20]. [Netflix]*

Iconic consciousness cannot be assigned

Carrier groups transmit a code, but icons are made by the audience. Iconic consciousness arises when form and meaning fuse in perception, not when an institution announces the fusion. Disney tried to create icons by administrative mandate: every release was positioned as a historic milestone. So far no character from the 2016–2022 period has reached the iconic status of earlier periods. The qualification matters: iconic status forms over decades, so at short remove this is a provisional observation, not a closed verdict. At the same time the new code attacked existing iconic consciousness: replacing the voice of an iconic character produced a reaction out of all apparent proportion to a technical decision. [Disney]

The authenticity of the bearer

A reflex holds when the code reads as the bearer’s own, because it answers a real demand. A bearer’s biography can work as a performance of authenticity. Walker appropriated the collective trauma of racial inequality, and his biography strengthened the performance: a man who came through the system leads the institution meant to change it [43]. A biographical fit with the content of the code makes the performance read as authentic, irrespective of what is happening in the bearer’s inner life. [Ford Foundation] The reverse case: a corporation capitalised at over two hundred billion dollars declares proximity to the suffering of marginalised groups. The gap between the performer’s economic power and the declared vulnerability does not close, and the audience reads the performance as strategic positioning [22]. In both cases the assessment turns on how the performance reads from outside, not on presumed sincerity. [Disney]

[Ford Foundation / officers’ habitus], [Netflix / walkout], [Disney / icons by mandate]

Mechanism 3. The external point of verification

The external point of verification is an indicator independent of the institution through which re-fusion or de-fusion becomes publicly visible: box office, ratings, membership attrition. The point of verification works in both directions. It exposes de-fusion where the code fails to match the result. It reinforces the code where the code withstands the test.

Activation condition

• •An indicator exists that the institution does not control.
• •The indicator is publicly observable and reads as a judgement on the code.

De-fusion as a rupture along an axis

Griswold describes culture through four poles: creator, cultural object, receiver, social world [40]. Legitimacy requires all four to align. De-fusion is a rupture along a specific axis, and the external point of verification makes the rupture visible. The axis is located by asking which pole the audience addresses its refusal to. Disney positioned every release as a statement by the studio, and refusal read as a reply to the creator. The AMPAS audience walked away from the broadcast as an object without contesting the Academy as creator.

A rupture along the creator–receiver axis is visible through box office. Disney did not control the box office, and the mismatch between the code of representation and the market result became publicly visible [25]. Box office, however, is multi-causal. Franchise fatigue, the shift from cinema to streaming and the quality of individual films all bear on it, and these factors do not separate cleanly from the code. The indicator does not track the code alone, but it is on this indicator that the mismatch is presented to the audience, and it discredited not only the directive but the principle behind it. [Disney]

A rupture along the object–receiver axis is visible through ratings. AMPAS lost its ceremony audience from 43.7 million (2014) [1] to 17.9 million (2026) [4] — a fall of 59 per cent over twelve years. That long figure mixes a code-specific decline with a format-wide one, and the two are only partly disentangled, as shown below. [AMPAS] The Recording Academy lost television audience from 26.1 million (2017) to 19.8 million (2018) — a fall of 24.1 per cent — and it happened before any reform, at and immediately after the press conference [15]. What is firmly established is the order of events: the audience left before the new code, which did not yet exist in 2018. What actually drove it away — the scandal, that particular evening’s programme, or the general decline of the ceremonial format — cannot be disentangled from a single figure. The scandal is the most probable factor here, but not the only one. [Recording Academy]

In diagnosis it matters to distinguish de-fusion through the collapse of a performance from de-fusion through a change of code. The AMPAS de-fusion of 2021 is explained by the collapse of the performance — pandemic, empty room, no host [3] — not by a change of code: mandatory compliance with the RAISE standards took effect only in 2024 [9]. Multiple causation and a shared format factor are confirmed by the fact that in 2026 every ceremonial broadcast declined at once [4]. [AMPAS]

An operational sacred withstands the test

If the sacred is operational, external verification strengthens it. The originalism of the Federalist Society worked this way. Confirmation hearings tested the application of the procedure, and that test was available to examiners of differing political positions even when they diverged in their conclusions [27]. [Federalist Society] If the sacred is not operational, the institution has no resource for passing an external test. After 2021 the sacred of the Recording Academy is defined entirely by an external proxy metric [13][16], and should the political context in which that metric operates change, the institution has no reserve sacred code to fall back on. [Recording Academy]

The latent absence of a point of verification

The configuration most resistant to de-fusion is the absence of an external point of verification inside a closed loop. The first module described this as self-sustaining legitimacy. At the symbolic level it means there is no one to confront the institution with the mismatch between code and result. Grantmaking institutions with a professional audience in the low thousands (Ford, the NEA) operate without an external box-office test. The audience is hidden and the result does not convert into a publicly observable indicator. The loop stays closed until an external parallel system of sufficient institutional weight contests the consensus. [Ford Foundation], [NEA]

[Disney / box office], [AMPAS / ratings], [Recording Academy / 2018 ratings], [Federalist Society / hearings], [Ford Foundation / closed loop]

Mechanism 4. Settled and unsettled modes

Swidler distinguishes two modes of culture [38]. In settled periods culture works as an invisible habitus: the code is not uttered, because everyone shares it. In unsettled periods the institution actively constructs ideology: it is visible, spoken explicitly, imposed. Explicit ideology is a signal of instability.

Activation condition

• •The code has undergone de-fusion or has met a competing code.
• •The institution is forced to utter what was previously implied.

The visibility of the code as a diagnostic sign

A settled code rests on the homogeneity of its bearers, not on an infrastructure of declarations. The Recording Academy electorate of the early decades consisted of people whose professional biographies intersected in three music centres — Los Angeles, New York, Nashville [10]. The criteria were shared because the formative experience was shared. Settled culture worked in the strict sense: the habitus stayed invisible while it was common to everyone in the room. When the electorate grew and shared experience ceased to be shared, the committees of 1989 and 1995 narrowed the room to fifteen or thirty people and extended the mode by another three decades [12]. [Recording Academy] AMPAS held a settled code of professional excellence for more than eight decades, and a mission that explicitly excluded economics and politics became an invisible habitus. [AMPAS]

When a code becomes visible, that marks the transition into an unsettled mode. AMPAS in 2015–2016: the settled code becomes visible through #OscarsSoWhite [6]. [AMPAS] Iowa shows the opposite possibility: a settled code holds for eighty years because it is transmitted through a workshop format that does not depend on external verification [33][35]. [Iowa Writers’ Workshop]

Unsettled as a permanent mode

Some institutions in the series live in the unsettled mode as a permanent condition. Ford after 2013 [43], the NEA after 2016 [41], the Recording Academy after 2018 [14] utter their code continuously. Continuous utterance is not strength but a sign that the code has not settled into habitus and requires constant maintenance. [Ford Foundation], [NEA], [Recording Academy]

The sign needs qualifying, or it yields a false reading on the Federalist Society. That institution has articulated its code continuously since 1982 [28] and nonetheless holds it as settled. The distinction runs through the addressee of the utterance. Articulation directed outward — at recruitment and at argument with a competing code on the field’s boundary — is compatible with a habitus settled inside the community of bearers. Within the network, originalism is applied as a method rather than debated as a position. Articulation directed inward — at holding one’s own bearers, or at justifying oneself before one’s own audience — marks an unsettled mode. Ford tells its field that pressure is not stopping it [18], the Recording Academy explains its criterion to its voters and viewers, Disney explains its values to its audience. The diagnostic sign of an unsettled mode is not the explicitness of the code but the need to prove its legitimacy to those who are already supposed to carry it. The settled mode is assessed relative to the community of bearers, not to the public sphere. [Federalist Society]

Re-fusion demands an authenticity infrastructure cannot produce

A new code can be installed institutionally: declarations, infrastructure, external ratings, financial incentives. But re-fusion demands perceived authenticity, which infrastructure cannot produce. The Disney code of 2016–2022 was installed institutionally and proved vulnerable on both sides. The audience of the old code read the new one as a replacement; the audience of the new code read its execution as insufficiently authentic. In March 2022 Pixar employees published an open letter about the systematic blocking of same-sex scenes [22] — the voice of the new-code audience on inadequate execution. The code was installed without a cultural foundation and turned out to be vulnerable along several axes at once. An order can create infrastructure but not the belief of an audience. [Disney]

[Recording Academy / settled until 2018], [AMPAS / the 2016 transition], [Iowa Writers’ Workshop / settled through format], [Disney / a code without foundation]

Mechanism 5. Declarability: the mode of utterance

Declarability is the mode in which an institution speaks about itself. The parameter was derived from the Recording Academy case and tested on the rest. It has two objects, and they must be distinguished. The first is the mandate, the publicly stated function of the institution. The second is the code, the substantive criterion of the sacred. They can diverge. Iowa declares its mandate and does not declare its code: the institution calls itself pedagogical in public, while the criterion by which prose is selected has never been written down and lives as the habitus of the workshop [33][34]. That divergence is precisely the source of durability. A declared mandate supplies a language of defence; an undeclared code deprives criticism of a target. At the Recording Academy neither object is declared, and there is nothing to defend [11]. At the Federalist Society both are declared, in two registers [28]. What follows concerns above all the declarability of the mandate. The declarability of the code is recorded in layer 7 of the anatomy. [Iowa Writers’ Workshop], [Recording Academy], [Federalist Society]

Activation condition

• •The institution meets pressure demanding a public answer about its own criterion.
• •The institution either has or lacks a language of defence that does not destroy its own foundations.

Declarable, semi-declarable, constitutively undeclarable

A declarable narrative mobilises through identification and at the same time creates a target for attack. A constitutively undeclarable one is shielded from direct criticism but stripped of the resources of mobilisation. This is the double effect of visibility, not a paradox.

Iowa works in a declarable pedagogical register. The institution states publicly that it is pedagogical, and criticism is received as a discussion of pedagogy rather than a scandal about genealogy [33]. When pressure for diversification arrives, Iowa answers in the pedagogical register without abandoning the role. This is the most durable configuration for utterance in the series. [Iowa Writers’ Workshop]

The Recording Academy works in a constitutively undeclarable register. The pedagogical function exists in fact — the formation of mass taste through the instruments of taste formation — but it has never been publicly articulated. The undeclarability is not accidental; it is constitutive. Public articulation would require opening up the commercial genealogy of 1957 [10], which would undermine the legitimating structure. This is the most vulnerable configuration. The institution cannot defend what it does not publicly declare, and it cannot begin declaring without destroying its foundations. [Recording Academy]

The Federalist Society works in a two-register mode of utterance. The narrative is unfolded inside the network and compressed at confirmation hearings [27][28]. This combines mobilisation with protection against direct attack. [Federalist Society]

Declarability as a predictor of vulnerability

The link between declarability and the depth of de-fusion is structurally predictive, though not logically necessary. An institution with a constitutively undeclarable mandate, on meeting pressure, has no language of defence that does not destroy its foundations. Its only available move is a language that requires no articulation of function. That move leads to substituting a proxy metric for the missing sacred [16]. This is the deepest form of de-fusion. Under constitutive undeclarability, deep de-fusion is not inevitable, but it is probable. [Recording Academy]

A general process in representative institutions

The declarability parameter works beyond the series. Representative institutions in different fields — academic peer review, editorial control in journalism, medical expertise, curatorial selection, literary prizes, grant panels — occupy structurally the same position. A group of co-opted experts certifies, on behalf of the profession, that a given work deserves recognition. From roughly 2010 onwards, pressure has been mounting in every field at once. Digital platforms let producers bypass the representative filter, algorithmic aggregation offers an alternative criterion, demographic pressure demands a redefinition of composition, and polarisation erodes the presumption of a single consensus. The degree of declarability of the mandate can be used as an indicator of an institution’s probable durability under such pressure. This observation requires confirmation by comparative research beyond the scope of the series. [Recording Academy]

[Iowa Writers’ Workshop / declarable], [Recording Academy / constitutively undeclarable], [Federalist Society / two-register]

Five mechanisms across eight cases

The table maps the five mechanisms across the eight cases. The cells summarise the reports in the series. Where a report did not examine the corresponding layer, the cell records that.

The table shows that no single mechanism distinguishes the cases on its own. Iowa and Ford both lack an external point of verification and both hold their code through the habitus of bearers, yet Iowa is settled and Ford unsettled. The difference comes from the format of transmission. A workshop reproduces the code without uttering it; a program office reproduces it through continuous declaration. Disney and Netflix coincide on four mechanisms out of five and diverge on the external point. Netflix is shielded by subscription, so its de-fusion was recorded by the behaviour of people rather than by a number [19]. [Iowa Writers’ Workshop], [Ford Foundation], [Disney], [Netflix]

When the mechanisms fail to hold

Structural completeness, living bearers, passing an external test, a settled habitus and a working mode of utterance are conditions that raise the probability of durable re-fusion. They do not guarantee it. The mechanisms may fail for many reasons, mutually incompatible and dependent on the specific context. The framework describes the conditions necessary for holding. It does not describe sufficient ones.

Part III. On the sacred

On the nature of the operational sacred

The second module poses the question at which the first stops: what exactly makes a code durable beyond the community that produced it. The answer is assembled from the five mechanisms and reduces to a single linkage — the operational sacred.

The first module showed that an operational criterion is a necessary condition for starting the personnel pipeline. The second module shows what an operational criterion becomes at the symbolic level: a sacred pole open to independent verification. This is the point where the two modules meet. Originalism is at once the verifiable criterion of the first module and the operational sacred of the second [26]. Operationality is a characteristic of the procedure, not a promise of a single correct decision. Originalists diverge on particular cases, and the degree of their convergence is itself a matter of dispute. What is structurally significant is not that the procedure always yields one conclusion, but that it is open to third-party verification. That same openness makes the sacred defensible in court, in the academy and in public debate [29][30]. [Federalist Society]

This explains why one sacred is structurally open to resonance beyond its own community while another is structurally tied to its group. An appeal to a verifiable marker rather than to group membership is in principle open to people arriving from different starting positions. In practice the Federalist Society recruited bearers beyond a single political niche, though critics contest the actual reach of that resonance as partisan [28]. A sacred defined through a demographic proxy metric leans on a universalist frame of justice, yet its working boundary runs along group membership [39]. It therefore mobilises one group and is structurally limited in reaching another, even when it appeals to everyone. Both sides receive here a structural property and an empirical qualification. [Federalist Society], [Recording Academy]

The framework does not substitute for the work of creating an operational sacred. A durable sacred cannot be constructed quickly to order. It must answer the question of what quality means in the given field, withstand intellectual challenge, and be open to third-party verification. The form of a sacred can be copied, along with its vocabulary and its ritual. The property of withstanding independent verification cannot. It is either built into the criterion itself or absent, and it is tested before the outcome, not retrospectively by how long the code lasted. A criterion that is open to verification and withstands challenge is distinguishable from an imitation at the moment of construction, regardless of what later becomes of it. That distinction is structural, not evaluative. Creating an operational sacred means building a verifiable foundation, not performing an act of positioning. [Federalist Society]

Discursive position

The symbolic frame removes a number of standard critical positions from the agenda and replaces them with a deeper question. The Recording Academy example shows the mechanics.

Public discussion of the reform runs in two camps. The progressive position holds that before the reform the institution had a working aesthetic standard applied with a demographic skew, and that the reform corrected the skew. The right-wing position holds that the institution had a standard developed over seventy years, which the reform destroyed by substituting a quota for expertise. Both positions share a hidden assumption: that a substantive aesthetic code existed, and was either applied wrongly or destroyed. [Recording Academy]

Analysis through the seven layers shows the assumption to be false in both versions. The sacred was never formulated substantively. In the rule it is set by negation alone — "without regard to album sales" [11]. Whether this was a deliberate decision by the founders to keep the centre empty, or an abstraction never operationalised, does not follow unambiguously from the available sources. What is structurally significant is the result, not the intention. A substantive standard existed only as the habitus of a homogeneous electorate and dissolved along with that homogeneity from the late 1980s, when secret committees became necessary to restore the result [12]. One cannot correct the application of a criterion that is not in the rule. Nor can a quota destroy a standard that had disintegrated from within three decades before the quota and never existed in a form that could be applied with a skew. The emptiness of the centre made it possible to substitute a proxy metric at the moment when pressure demanded a public answer about the criterion [16]. The question is not whether the code was applied correctly, but how an institution with an empty centre works, and what happens when the emptiness is forced to be filled. [Recording Academy]

The move generalises. Before positing a change of code, it is worth checking whether the observed changes can be explained by a change in the inputs to an unchanged mechanism. Iowa shows the pure case. The outward signs of change are maximal — demographics, subject matter, public language — while the internal procedure of the workshop is unchanged across an eighty-year horizon [33]. To posit a new code in the 2010s would be an attribution error. The rule applies where an institution has a separable layer of reproduction through which new bearers absorb the code irrespective of where the code is formulated. [Iowa Writers’ Workshop]

Part IV. Application

The parenthetical examples show the type of analysis required, drawn from cases in the series. A full answer to each question requires a separate analytical document.

Diagnosing a code: seven questions

To assess the durability of a code, the analyst runs through the seven anatomical layers and the five mechanisms as a checklist. The seven questions below set the order of that pass and do not coincide with the seven layers. Each question touches one or more layers and mechanisms.

1. Is the narrative structure unfolded? Check for the seven components: initial state, crisis, hero, antagonist, method, sacrifice, restoration. A folded structure is vulnerable to pressure that a complete one withstands. (The Federalist Society holds all seven components. The Recording Academy is folded.)

2. Is the sacred operational? Check whether the criterion is open to application and assessment by a third party. An operational sacred is defensible by argument and withstands external verification even when its carriers diverge on particular conclusions. An iconic sacred cannot be defended by argument; it rests on the homogeneity of its bearers. An empty one fills with the first available proxy metric. (Originalism is open to verification. The sacred of the Recording Academy is empty.)

3. Who carries the code, and can the bearers be recalled by order? Check whether the code rests on a financial incentive that a decision can remove, or on the habitus of bearers with institutional power, which no order can cancel. (The habitus of Ford’s officers against the top-down HHMI programme.)

4. Is there an external point of verification, and what does it show? Find an indicator the institution does not control: box office, ratings, attrition. Passing the test strengthens the code; failing it exposes de-fusion; the absence of a point closes the loop of self-sustaining legitimacy. (Disney’s box office, the AMPAS ratings, Ford’s closed loop.)

5. In which mode does the code work, settled or unsettled? Check whether the code is uttered explicitly or works as an invisible habitus. Continuous utterance addressed to one’s own bearers and audience signals that the code has not settled and requires constant maintenance. Utterance directed outward, at recruitment and argument, is compatible with the settled mode. (Iowa is settled through format. Ford is permanently unsettled.)

6. What is the mode of utterance of the mandate? Determine whether the mandate is declarable, semi-declarable or constitutively undeclarable. Constitutive undeclarability is a predictor of deep de-fusion under pressure. (Iowa is declarable. The Recording Academy is constitutively undeclarable.)

7. Along which axis of the Cultural Diamond does the rupture run? If de-fusion is already under way, locate the axis: creator–receiver, object–receiver, creator–social world, object–social world [40]. The axis indicates what exactly has stopped aligning. (Disney: a creator–receiver rupture through box office.)

Parts I through III describe the mechanics. Part IV translates them into operational questions and preserves the same diagnostic neutrality. The seven questions above set out a general diagnosis of a code. The same questions take different priority depending on the reader’s position. The four roles below cover the whole field, from defending an existing code to building a new one, and the instrument works from any position without itself choosing a direction. Each role comes with a worked example from a case in the series. The example shows which questions the role asks first and what answer the particular case gives.

Diagnosing your role

If the role is not obvious, three questions will determine it.

• •Are you inside the institution whose code you want to hold or change, or outside it?
• •If outside, do you have a ready alternative code, or only pressure?
• •If you have a code, are you building a new institution or entering an existing one?

The answers point to a role: incumbent arbiter, entering change agent, source of pressure, architect of a new code. Switch to the corresponding example.

What remains outside the frame

The five mechanisms describe the conditions of durable re-fusion. Structural completeness supplies a resource for holding; living bearers make the code irrevocable by order; an external point of verification exposes or strengthens the code; the settled mode conceals a habitus that has settled; the mode of utterance determines whether a language of defence exists. The operational sacred binds the symbolic level to the infrastructural one.

But the framework describes mechanics, not content. Which values the sacred activates is a question of content. Which boundary the code draws is a question of social function. The instrument shows what makes a sacred operational and durable. It does not say what the sacred ought to be. The content of the code and the intended outcome lie outside the instrument’s competence and are the choice of whoever applies it.

The end state of the symbolic cycle is the moment when the code reproduces through a distributed network of bearers without a coordinator and withstands external verification. The change of control over production is the same in either outcome: one code displaced another. The two outcomes differ in the structure of the new code. In the first, the new code is structurally weaker than its predecessor or equal to it, the incompleteness has reproduced in a new form, and the door stands open again for the next change agent. In the sample this is the Recording Academy and AMPAS. In the second, the new code is more complete and more operational than its predecessor, and the next claimant will have a considerably harder time. In the sample this is the Federalist Society across a forty-year horizon.

The map of norm production is closed

The first module describes the infrastructural level: how to enter, accumulate position, fix a criterion. The second module describes the symbolic level: what makes a fixed criterion durable and legitimate beyond the community that produced it.

Neither module suffices without the other. An operational criterion without an infrastructure of transmission remains a text without bearers. Infrastructure without an operational sacred produces formal compliance without an audience’s belief and collapses at the first external test.

The first module is closed. The second module is closed. The map of norm production is closed by both.