The Dominant-Player Constraint

Governance prevents the strongest player from picking the rules, evidence, forum, timing, closure condition, and enforcement substrate all at once.

Elias Kunnas

The corpus’s existing primitives — Procedural Object, Record Gate, Measurement Anchor, Refusal to Compute, Feedback Authority, Implementation Ledger, Structural Residue, Corrective Closure Ownership, The Reproduction Test — read as a disconnected toolbox of sender-side, receiver-side, and ledger-side mechanisms. They form a single architecture: each blocks a specific dominant-player capture lever. Naming the lever set makes the existing primitives easier to use and shows which levers no primitive yet blocks.

I. What a dominant-player constraint is

A dominant player, in this essay, is the actor with unilateral leverage over one or more variables of a specific contested decision. It is a structural, role-relative position, not a class identity. The same actor can be dominant in one contest and constrained in the next; class position often predicts dominance but does not exhaust it. The IRS is the dominant player when collecting from a small-business taxpayer; that same taxpayer is the dominant player when its supplier sues for late payment in a court the taxpayer has lobbied for years.

When such an actor faces a contest, six recurring levers can tilt the outcome in advance.

Rules. What the procedure says, what counts as relevant, how categories are drawn. Whoever sets the rules wins most contests before they begin.
Evidence. What gets measured, what gets recorded, what counts as admissible proof. Whoever curates evidence renders inconvenient facts inadmissible without ever denying them.
Forum. Where the dispute is heard, by whom, under whose procedural rules. Whoever picks the venue gets a sympathetic referee.
Timing. When the decision closes, when it reopens, whose clock runs. Whoever controls timing can run out the clock or strike when opponents are unready.
Closure condition. What counts as resolved, what residue is acceptable, what reopening looks like. Whoever owns the closure condition can declare victory regardless of whether the underlying problem is fixed.
Enforcement substrate. The carriers — judges, auditors, journalists, civil servants, technical specialists — that make the other five constraints real in practice. Whoever can let the substrate decay can render the formal constraints decorative even where they remain formally in force.

Strong governance means no single actor controls all six levers in a contest that affects others. The threshold matters: partial capture is more common than total capture. An actor that holds rules + timing + closure has often already won the contest without ever appearing to control the forum or the substrate. The diagnostic question is therefore not “does anyone have all six?” but “how many does this actor hold in this contest, and which combinations are independently fatal?”

The frame applies to capture by a unilateral dominant actor. It is silent on governance failures driven by coordination breakdown, fragmented vetoes, or commons collapse — those are a different family of failure mode the corpus addresses under different primitives. Cancer Failures names the binding-mass overload case; the dominant-player frame names the unilateral-capture case.

The six are procedural levers, observable inside a contest as it runs. §IV identifies further levers (frame, agenda, standing, burden, scope, capacity-hollowing) that operate upstream of or across any single contest. They are not collapsed into the six because no current corpus primitive blocks them and the analytical work they require is distinct. The six are best read as the recurring levers the existing primitives already address; the upstream set is the recurring levers the corpus still owes primitives for.

II. The corpus primitives as constraints on the six levers

Each governance-arc primitive constrains one or more of the six levers. The mapping is many-to-many — the same primitive often pushes back on more than one lever — but the primary effect of each primitive is identifiable. Below, the primary lever it constrains and the move it makes structurally costlier.

Primitive	Primary lever	What the primitive installs
Procedural Object	Rules — admissibility of artifacts	Admissibility becomes a property of the artifact, removing the receiver’s unilateral discretion to refuse inconvenient claims as “non-procedural.”
The Record Gate	Rules — admissibility of cognition	The threshold at which institutional thinking becomes binding is architecturally specified, blocking the strategic use of informality to keep load-bearing cognition outside the record.
Measurement Anchor	Evidence	Four conditions (causal audit path, validity envelope, bounded proxy status, independent verification) keep the consequential proxy coupled to its territory and block Goodhart drift.
Refusal to Compute	Evidence — via the rule that triggers obligation	Institutional refusal to bound a consequential relation becomes visible as a procedural failure, removing “we cannot estimate this” as a free option for avoiding obligation.
Feedback Authority	Closure	The grade of authority the architecture actually carries is named and graded, preventing the labeling of advisory channels as consequential.
Implementation Ledger	Timing — with closure as secondary	Conditional closure with preauthorized re-entry constrains how long acceptance can stand in for execution; “accepted” stops substituting for done.
Structural Residue	Closure	The causally live remainder after correct refusal is named as a live load that must be routed somewhere owned, preventing the receiver from treating residue as someone else’s problem.
Corrective Closure Ownership	Closure — re-entry	“Who must notice and repair when this decision fails?” gets a named, funded, independent, triggered, authorized answer, replacing entropy-based correction.
The Reproduction Test	Enforcement substrate	Substrate health is itself audited at or above attrition rate, blocking reliance on generator-chain decay (fewer judges, fewer auditors, fewer journalists, fewer civic-knowledge carriers) to make formal constraints decorative.

The table makes the architecture visible. Two primitives constrain Rules (Procedural Object, Record Gate). Two constrain Evidence (Measurement Anchor, Refusal to Compute). One constrains Timing (Implementation Ledger; Closure is its secondary effect). Three constrain Closure or its conditions (Feedback Authority, Structural Residue, Corrective Closure Ownership). One constrains the Enforcement substrate (The Reproduction Test). No existing primitive constrains Forum. That is the most visible gap.

III. The pattern

The collection is not arbitrary. Each primitive isolates a specific dominant-player lever and installs a constraint that makes the corresponding move costlier or impossible. The collection extends the corpus’s master line — a system fails where reality enters through a channel the institution does not own — by specifying which channels need owning under which capture threats.

The pattern is best read as an organizing hypothesis, not a closed typology. Three caveats are visible from the table itself:

Many-to-many mapping. Most primitives push back on more than one lever. The matrix is dense, not diagonal.
Uneven coverage. Closure has three primitives (with Implementation Ledger’s secondary touch as a fourth). Forum has none. Several upstream levers (frame-and-categories, agenda-setting, participation-and-standing, burden-of-proof, scope-and-unit-of-analysis) are not constrained by any current primitive and probably warrant their own.
Substrate vs procedure. The Reproduction Test sits at a different layer than the others — it preserves the carriers that make all five procedural constraints real, rather than constraining a procedural move directly.

These are properties of the architecture, not defects in it. They show where the typology has been built out and where the next work belongs.

IV. Generative gaps

Apply the question — which dominant-player lever does this primitive block, and would the move otherwise be available? — to the lever set, and the corpus’s missing primitives become visible.

Forum-Selection / Venue-Architecture. Health-law cases routed to administrative committees, tax disputes to revenue-friendly courts, environmental claims to permit boards instead of citizen suits, platform disputes to forced arbitration. The corpus has Procedural Object (sender-side) and Feedback Authority (receiver-side response duty) but no primitive that names the venue-selection move itself.
Standing / Participation. Forum-Selection is where; this is who appears in the forum, with what discovery rights, and what standing to appeal. Capture often runs through standing restrictions before it reaches the procedural rules.
Frame / Category. Foucault- and Lakoff-style framing power precedes rule selection: it defines what the dispute is about before procedure starts. Adjacent corpus work (the framing literature, Calculemus, Non-Compilation) bears on it; no primitive names the move.
Agenda / Contestability. The Bachrach-Baratz second face of power: preventing the issue from becoming contestable at all. Procedural Object assumes the contest already exists; agenda-control happens earlier.
Burden-of-Proof. A meta-rule over evidence: who must prove, to what standard, after what presumption. Often the decisive move in regulatory and welfare contests.
Capacity-Hollowing. Defunding or de-staffing the agency that runs an otherwise sound procedure. The Reproduction Test covers substrate decay at the carrier-chain level; the corpus’s earlier work on hollowed-out analytical capacity inside government (Powerless Intelligence) names the pattern; neither functions as a per-decision constraint.

These are the most visible missing primitives. Naming the lever set is what makes them visible.

V. Why the typology matters

The typology converts the corpus’s accumulated technical work into a single recognizable claim: governance is the architecture that prevents the strongest player from picking the whole game. That vocabulary — corruption, capture, oligarchy, regulatory failure, elite immunity — is the one readers already use. The technical primitives are the joints of that architecture, not a separate engineering project.

The plumbing is the political mechanism. Separating the six levers is what makes political contest real; without separation, one actor holds all of them and the contest is decorative. Whether any given pipe is itself captured is the standard next question, with standard answers: independence, contestability, auditability, adversarial access, maintenance funding, re-entry, and sanctions. The separation is the precondition for those questions being meaningful at all.

VI. Compression

Governance is the architecture that prevents any single actor from picking the rules, the evidence, the forum, the timing, the closure condition, and the enforcement substrate of a contest that affects others. Each governance-arc primitive constrains one or more of these six levers. The matrix is dense, not diagonal: most primitives push back on multiple levers; some levers have several primitives; one (Forum) has none. The gaps identify where the next primitives belong.

Related corpus essays, by lever:

Rules: The Procedural Object, The Record Gate
Evidence: Measurement Anchor, Refusal to Compute
Timing: Implementation Ledger
Closure: Feedback Authority, Structural Residue, Corrective Closure Ownership
Enforcement substrate: The Reproduction Test
Cross-layer (binding-mass overload, the other failure family): Cancer Failures

Sources and Notes

Anti-domination as an explicit political target.

James Madison, Federalist No. 51 (1788) — “ambition must be made to counteract ambition.” The earliest explicit articulation of the anti-domination architecture at the constitutional layer; the six-lever frame is a mechanism-level translation of the same move.
Philip Pettit, Republicanism: A Theory of Freedom and Government (Oxford University Press, 1997) — non-domination as the explicit political target, distinguished from non-interference. The lever-set is one operational reading of what non-domination requires in practice.

Comparative-political-economy frames.

Daron Acemoglu and James A. Robinson, Why Nations Fail (Crown, 2012) — extractive vs inclusive institutions. The dominant-player frame is roughly extractive-vs-inclusive at the mechanism level; the six levers identify what extractive institutions concentrate and inclusive ones separate.
Douglass C. North, John Joseph Wallis, and Barry R. Weingast, Violence and Social Orders (Cambridge University Press, 2009) — open-access vs natural states. Open-access transitions are the historical move from one actor holding most levers to multiple actors each holding some.

Upstream-power frames the typology does not yet constrain.

Peter Bachrach and Morton S. Baratz, ‘Two Faces of Power,’ American Political Science Review 56:4 (1962); Steven Lukes, Power: A Radical View (Macmillan, 1974; 2nd ed. 2005) — agenda-control and frame-control as forms of power prior to procedural contest. The current typology covers the procedural layer; the agenda and frame layers are the most visible missing primitives.
Michel Foucault, Discipline and Punish (Gallimard, 1975); George Lakoff, Don’t Think of an Elephant! (Chelsea Green, 2004) — on category-construction and frame-setting as moves that precede formal procedure. A Frame-Gate or Category-Gate primitive is one of the visible gaps in §IV.

Note on completeness. The six-lever set is not asserted as exhaustive. It is the set the existing corpus primitives already address; the upstream-lever list in §IV is the set the corpus still owes primitives for. As the corpus matures, the lever count may expand, and individual primitives may be reclassified across levers without disturbing the underlying claim — that governance is the architecture that prevents any single actor from holding the contest-defining levers all at once.