Photo by Dan Freeman on Unsplash
Document: PPA-04-2026 Classification: Unclassified. Distressingly so. Distribution Statement A: Approved for public release; distribution is unlimited, unlike the underlying reforms. Prepared for: No sponsor. None was sought. None would have been forthcoming. Prepared by: The Applied Deadlock Division, under its own recognizance.
This paper should be of interest to policymakers, institutional analysts, procedural historians, and any person who has recently sat through a meeting at which a known and unfixable problem was described in full by everyone present and then adjourned.
The Pocket Prompt Analysis series provides composable, copy-and-pasteable analytical instruments for the general reader. Each instrument fits in a pocket. Each is deployable in under an hour. None of them work, in the sense of causing anything to happen. They work in the narrower sense of telling you precisely why nothing is happening, in the institution’s own vocabulary, with the relevant provision numbers.
The Division regards this as a meaningful contribution to the literature and a total loss to the world.
Prior entries in the series:
The Structural Contradiction Audit (SCA) is an eight-stage dialectical protocol for locating obstruction inside formal rule-systems. It takes as input a failed normative desire — a thing an institution said it wanted and did not do — and returns a located, typed, and classified account of exactly where the thing died.
Version 0.3 has been validated against four institutional architectures spanning designated-veto, distributed-consensus, threshold, and captive-membership systems. Across all four, the instrument performed as designed.
In every case it confirmed what the affected populations already believed.
The Division wishes to be clear that this is the intended behaviour. An audit that discovered a surprising reason the thing was stuck would be an audit that had made an error. Everyone already knows. They have always known. The contribution of SCA is not the knowledge but the paperwork, and the Division maintains — against considerable evidence — that paperwork is a form of dignity.
F-1. The institution is stuck.
F-2. It is stuck for reasons that are legible, documented, and cited in the founding instrument.
F-3. In three of four validated cases the obstruction was sealed — meaning the rule preventing the outcome is protected by the rule preventing the outcome, in a closed loop the drafters installed on purpose as the price of the institution existing at all.
F-4. In the fourth case the obstruction was not sealed. This was worse. See The Excuse Removal Function, below.
F-5. The Audit generated twenty-nine methodological findings, four contradiction types, and two confirmed false negatives across four institutions and roughly two centuries of accumulated procedure. Total measured change to any institution studied: nil.
| Case | Architecture | Grievance duration | Diagnosis | Did anything change? |
|---|---|---|---|---|
| Multilateral security body | Designated veto | 14 yrs | Sealed | The problem outlived itself |
| Multilateral trade body | Distributed consensus | 10 yrs | Soft-sealed | A workaround, for the willing |
| Multilateral financial body | Threshold | 16 yrs | Soft-sealed | It grew. So did the grievance |
| Domestic legislative chamber | Captive membership | Ongoing | Soft-sealed [masked] | See below |
The Division draws attention to the third row. The financial body was found to exhibit Type 10 — Reinforcement Through the Permitted Channel, in which the only measures capable of clearing the decision rule are measures proportional to the existing distribution. The institution therefore grows, repeatedly and successfully, and each growth deepens the imbalance the growth was demanded to correct.
This is not deadlock. Deadlock would be an improvement. This is an institution that is working, on schedule, in the direction of the problem.
The Maintenance Node. Hardened institutions are attacked at the basement, not the front door. Nobody vetoes the ruling; they decline to appoint the judges. Nobody blocks the mandate; they decline to renew it. The front door is reinforced steel. The budget line is a screen door.
Default Direction. In the strongest form of obstruction, the blocking party does nothing at all. Terms expire. Windows lapse. No vote is cast, no statement issued, no record generated. The Division notes with professional admiration that this makes consensus procedures less accountable than an explicit veto, which at least leaves a fingerprint.
Channel Deformation. Where a single privileged channel survives, policy is redesigned to be legible to that channel rather than effective at its purpose. Mandates become subsidies. Regulation becomes a tax credit. Over sufficient time, an institution’s entire output warps into the shape of whatever door was left unlocked.
The Target-Held Gate (Type 7). Confirmed in four of four architectures and now the most robust element of the framework: the entity a mechanism exists to constrain reliably holds a consent node the mechanism requires to operate. The regulated hold the pen. The Division reports this as a finding rather than a joke, though it declines to argue the point.
⚠ OPERATOR WARNING
The Audit’s fourth stage-six verdict is the flag masked, returned when a lower-threshold route to changing the rule exists and is simply not used.
When this flag fires, the finding is not “the institution cannot.” The finding is “the institution will not, and has been describing the second as the first.”
The Division advises extreme caution. Sealed verdicts are comforting: they license despair, which is restful, and they distribute blame to the dead. A masked verdict licenses nothing. It removes the procedural explanation from everyone entitled to it and returns the matter to the domain of choice, where it becomes a conversation rather than a structure.
Operators have reported that deploying a masked finding at a family gathering, a professional retreat, or a committee hearing produces immediate and durable social cost, borne entirely by the operator. The instrument is accurate. That is the problem with it.
| Line item | Value |
|---|---|
| Instrument cost | $0 |
| Time to insight | ~40 minutes per institution |
| Institutional change produced | 0 |
| Return on investment | Undefined (division by zero) |
| Return on investment, alternative treatment | Also undefined, but with better catering |
The Division notes that where both the numerator and the denominator are zero, the ratio may be assigned by the analyst according to preference. This is the only degree of freedom the framework provides and operators are encouraged to enjoy it.
The Audit does not unstick the thing.
The Audit has never unstuck the thing.
The Audit was not designed to unstick the thing, and the Division wishes to correct a persistent misreading of the v0.2 abstract in which the phrase “minimal perturbation test” was taken by several early operators to imply that a perturbation would occur.
Further limitations:
Do not deploy on:
The Division draws particular attention to the last. The Audit’s anti-teleology gate requires the operator to distinguish contradictions that were designed in as the price of ratification from those that merely emerged. Operators who drafted the instrument under audit have proved unable to make this distinction and, in three documented instances, have instead delivered a reform proposal that was, on inspection, a proposal to dissolve the institution, stated in a register that concealed this from its own author.
Copy from here. It is the whole instrument. It has been compressed from a specification of some length and loses roughly nothing, which the Division has elected not to examine further.
You are running a Structural Contradiction Audit.
INPUT: an institution, and one specific thing it was supposed to do
and did not do.
0. OPERATIONALIZE. Convert the desire into a specific output artifact:
which organ, which provision, what instrument, by when. If you can't
name the artifact, stop — it's a mood, not a probe. Decompose: real
desires arrive as bundles. Flag any probe in the bundle that
SUCCEEDED — that's your control.
1. REACHABILITY. Per probe: unreachable by text / reachable but
unreached / reachable but unenforceable. Cite the provision, not
the institution's self-description.
2. CHOKEPOINTS. List every consent node — SUBSTANTIVE (on the path)
and MAINTENANCE (appointments, quorum, budget, renewal, sunsets).
For each: who holds it, cost of blocking vs enabling, does the
status quo decay without action (if yes, inaction is the weapon),
does blocking leave a record, is the holder unitary or does it
contain its own internal ratification.
3. TYPOLOGY. Classify: constitutive / temporal / scalar / inversion /
enforcement gap / closure / target-held gate / mandate drift /
extra-textual convention / reinforcement through the permitted
channel. Separate contradictions DESIGNED IN as the price of
ratification from ones that emerged.
4. DISPLACEMENT. Blocked pressure doesn't vanish. Where did it go —
parallel bodies, subnational units, another branch, deferral,
non-compliance? Two tests: does the bypass run at a LOWER consent
threshold than the blocked channel (if not, it inherits the
disease), and what does the bypass COST IN CONTENT — what shape
must the policy take to fit through it?
5. LEGITIMACY. Who is actually paying? Partition by issue; run it as
a time series, not a snapshot. Use the institution's own published
benchmark if it has one. In captive systems the cost usually lands
on some OTHER body, so the blocked one looks healthy.
6. PERTURBATION. Smallest change that makes the probe reachable.
Then enumerate EVERY route to rule change — amendment, precedent,
interpretation, practice, privileged track — with each one's
threshold. Amendment is rarely the cheapest.
Verdict: reformable / soft-sealed / sealed.
If a cheaper route exists and goes unused, flag it MASKED and say
plainly what is actually doing the blocking.
Then name where the pressure lands next.
7. RECURSE. The fix becomes the new desire. Return to 0.
Run a negative control: something the same institution DID achieve.
If your method finds obstruction there too, you are pattern-matching.
Report that.Homeowners’ associations. Faculty senates. Standards bodies. Any organization with a constitution and a grievance. Procurement. Your own household’s recurring dispute about the thermostat, which the Division predicts will return Type 7 — target-held gate, and which should not be pursued.
The Division thanks the four institutions examined in this study, none of which were contacted, all of which would have declined, and each of which has published, in its own documents, a more damaging account of its condition than anything appearing here.
The Division further acknowledges an anonymous external reviewer whose three criticisms of v0.2 were all correct, two of which the Division had already suspected and suppressed.
Suggested citation: Office of Diminished Expectations, Applied Deadlock Division. The Structural Contradiction Audit: A Portable Instrument for Establishing, at Length and with Citations, That Nothing Can Be Done. Pocket Prompt Analysis No. 4. The Uncertainty Department, 2026.
Series note: Pocket Prompt Analysis instruments are composable. PPA-04 accepts the output of PPA-01 as input, and its own output is accepted by no one.
Annex note. The Division has been asked, repeatedly and with a tone, to publish the actual specification rather than continue describing it. It is reproduced below in full, together with all four validation case files and the complete findings register.
The reader is advised of three things. First, the Annex is roughly four times the length of the paper it is annexed to, which the Division regards as the correct ratio and offers as a general principle. Second, the Annex retains its own internal lettering (Addenda A through E), which does not correspond to the Appendices above; no reconciliation is planned. Third, the Annex contains no jokes. It was written first, straight, and every finding in it was arrived at by running the protocol on real institutions and recording what broke. The satire above is downstream of it and depends on it being true.
Operators who find the paper amusing and the Annex tedious have understood the paper.
Status: Working spec. Validated against four institutional architectures, including one captive-membership system. Name: Settled as SCA. ICE (Institutional Closure Engine) is withdrawn — “closure” carries heavy prior freight in systems theory and epistemology, and the tool’s function is audit, not engine. Lineage: Sibling to RDPE. Shares the recursive dialectical loop and the findings-register discipline; differs in unit of analysis — RDPE models sentiment and positional capture across populations, SCA models obstruction inside formal rule-systems. GODE supplies the ontological framing for Stage 3.
v0.2 → v0.3
Three criticisms were received. Disposition:
1. Exit-option bias in Stage 4. Accepted, and tested directly. The prediction was that Stage 4 would need to model internal deflection rather than external forum-shopping. Partly confirmed, but the stage did not break — displacement is abundant in captive systems too; it simply runs along different axes. The genuine failure was elsewhere and unpredicted: displacement channels deform the substance they carry, which was invisible in the international set because those bypasses were functionally isomorphic to the blocked organ. See F24, F25.
2. The soft-sealed trap — actors disguising political will as procedural hurdle. Accepted and empirically confirmed by the fourth case, where it is the central dynamic. Now formalized as the preference-revelation test (Stage 6.3) and the masked verdict flag (§1.5). This was the most productive item in the review.
3. Type 9 as a residual category. Accepted. Three positive criteria added (§1.4). Applied retroactively, they downgrade the v0.2 IMF finding to provisional — see Addendum E.
Recommendation on sequencing (run a hybrid body such as the CJEU or IACtHR before the captive case) was not followed, and in hindsight the ordering did not matter for the reason given: Stage 4 didn’t break at the captive extreme, so the hybrid case wasn’t needed as a bridge. It is still worth running, but for a different purpose than proposed — see §5.2.
SCA diagnoses where a formal rule-system structurally obstructs an outcome that actors within it claim to want. It takes a failed normative desire as input and returns a located, typed, and classified account of the obstruction, plus a verdict on whether the obstruction is reformable through the system’s own machinery.
The governing intuition: structure is invisible until something fails to pass through it. A charter read on its own terms yields its self-description. A charter read against a specific thing it failed to do yields its actual architecture. The failed desire is a probe, not a grievance.
Requires: a written rule-system with identifiable decision procedures; a datable, specifiable outcome sought and not achieved; a public or reconstructable record of the decision path.
Degrades badly where decision rules are unwritten, records are closed, or the “institution” is an informal network.
New in v0.3: SCA now covers captive-membership systems (exit unavailable or prohibitively costly). Stage 4’s geometry differs materially; see §2, Stage 4.
An operationalized normative desire: a specific institutional output artifact, attributable to a named organ, under a named provision, with a date.
Gate: if the output artifact cannot be named, the analysis does not start.
Any point on the path from input to output where some actor’s consent — active or passive — is required.
| Class | Definition | Typical decision rule |
|---|---|---|
| Substantive | On the path from input to output | Usually the hardened rule |
| Maintenance | Appointments, quorum, budget, mandate renewal, sunset clauses, secretariat capacity | Usually a weaker rule — classified as housekeeping at drafting |
Maintenance nodes are the softest attack surface in any hardened institution and are systematically under-recorded in institutional self-description. Always enumerate both classes. (F9)
Hardness — absolute or marginal. Scalar in designated-veto architectures; a distribution over latent blocking coalitions weighted by coordination capacity in threshold architectures. (F2, F15)
Default direction — does the status quo persist without affirmative maintenance, or decay? Where it decays (term expiry, sunset clause, renewal), inaction alone destroys and cost-of-blocking falls to zero. (F10)
Attribution cost — does exercising the block generate a record? A cast veto is attributable; a failure to join consensus is not. Consensus architectures are less accountable at the chokepoint than designated-veto architectures. (F12)
Holder type — designated, threshold, distributed, or adjudicated. (F26)
Adjudicated holders are referees applying a stated test rather than parties exercising a preference (a parliamentarian, a registrar, a compliance officer, a credentials committee). They carry two distinguishing features: the test itself is a second-order rule that can be litigated, and the holder is usually removable or overridable by a meta-node under a different — often lower — threshold. Record both the test and the meta-node. Adjudicated nodes are frequently mistaken for neutral procedure and are among the most consequential in practice, because they determine what form a policy must take to pass.
Nesting — is the holder unitary, or does it contain internal ratification requirements on a separate clock? (F18)
Record cost-of-blocking against cost-of-enabling per node. Structural weakness concentrates wherever cost-of-blocking approaches zero.
Core predictive rule (F21): a blocked outcome becomes achievable when it is reframed from redistribution to capacity — restructured so it no longer alters the holder’s relative position. Confirmed in four architectures.
| # | Type | Definition | Precondition |
|---|---|---|---|
| 1 | Constitutive | The founding compromise itself; the price of the institution existing | Any |
| 2 | Temporal | Rules encoding a world model that no longer obtains | Any aged body |
| 3 | Scalar | Machinery built for one class of phenomenon applied to another | Any |
| 4 | Inversion | A procedure designed to protect a value now reliably produces its opposite | Any |
| 5 | Enforcement gap | Obligation without compulsion | Separable adjudication and enforcement |
| 6 | Closure | The body that must amend the rule is protected by the rule | Self-amending body |
| 7 | Target-held gate | The entity the mechanism exists to constrain holds a consent node the mechanism requires | Identifiable regulatory target |
| 8 | Mandate drift | The institution generated the contradiction by expanding past its constitutive text | Delimited mandate + discretionary application |
| 9 | Extra-textual convention | Obstruction with no textual basis and therefore no node | See discipline criteria below |
| 10 | Reinforcement through the permitted channel | Only measures proportional to the existing distribution clear the rule, so every success deepens the grievance | Distributional rule governing outputs |
v0.3 adds no types. Two candidates were considered and rejected: deformation (recorded instead as a Stage 4 mechanism, F24, since it describes what a bypass does rather than a contradiction in the rule-system) and masked obstruction (recorded as a verdict flag, F23, since it is a diagnostic error category rather than a structural feature). A typology that grows one entry per case is a taxonomy of anecdotes. The bar for a new type is that it names a structural precondition, recurs across architectures, and cannot be expressed as a stage output.
Type 9 has no node and is therefore the framework’s residual-category risk: anything the method cannot isolate can be dumped there. To classify an obstruction as Type 9, all three must hold:
Criterion 3 does the real work. Without it, what looks like a norm-lock is an unexercised node — nobody with the power to break the convention has yet tried — which is a completely different diagnosis with a completely different Stage 6 verdict. Where 1 and 2 hold but 3 is unestablished, classify as provisional Type 9 and flag the evidentiary gap explicitly.
Stage 1 reachability: unreachable by text / reachable, unreached / reachable, unenforceable
Stage 6 perturbation:
masked (flag, not a verdict) — a lower-threshold constitutive channel exists and is unused. The rule is not the obstruction; the holder’s unwillingness to use the available channel is. Attach to whichever verdict applies and state plainly what is actually doing the blocking. (F23) The masked flag exists because participants have a strong incentive to present a soft-sealed situation as sealed: procedural impossibility is a costless explanation for an outcome that is actually a choice. Symmetrically, opponents of a rule have an incentive to present sealed situations as soft-sealed, to attribute bad faith where a genuine structural barrier exists. The audit must resolve this rather than adopt either party’s framing.
Completeness test: could the desire have been satisfied by an output not in your probe set?
Trace each probe’s formal path; cite the provision, not the institution’s self-description. Distinguish sharply between forbidden by text and never done. Most obstruction sits in the second category, and institutions routinely present the second as the first.
Output: a node table, not a narrative.
Classify against §1.4. Multiple types routinely co-occur.
Blocked pressure does not dissipate. Map where it went.
The axes available depend on whether exit is possible:
| Axis | Description | Typical of |
|---|---|---|
| Horizontal | Parallel institutions, plurilateral arrangements, regional bodies, coalitions | Voluntary membership |
| Vertical | Subnational or supranational units acting in the gap | Federal / layered systems |
| Branch-lateral | Another organ of the same polity — executive unilateralism, judicial policymaking | Captive systems |
| Temporal | Deferral, sunset, provisional measure, continuing resolution | Any |
| Extra-institutional | Non-compliance, nullification, protest, withdrawal of cooperation | Terminal route in captive systems |
Stage 4 survives the captive-membership case. The reviewer’s expectation that displacement would collapse into systemic breakdown was not borne out — displacement is abundant, it simply re-axes. What changes is the cost incidence: see 5d.
A bypass is durable when it operates under a lower consent threshold than the blocked channel. Routes requiring the same threshold inherit the disease.
Displacement is contested terrain. The chokepoint holder pursues pressure into the exit channel, usually by capturing that channel’s maintenance nodes.
Record what the bypass costs in content, not merely whether it works. A displacement channel with a form constraint reshapes the substance passing through it: only provisions satisfying the channel’s admission test survive, so policy is redesigned to be channel-legible rather than effective. Over time the institution’s entire output distribution warps toward whatever the surviving channel can carry.
This was invisible in the international validation set because those bypasses were functionally isomorphic to the blocked organ. It is the dominant effect in the captive case, and it is the mechanism by which Type 10 operates in systems where the distributional rule is a form constraint rather than a share constraint.
Check whether the structural damage has been converted into an affirmative tool by third parties. A defunct organ can be more useful to a non-complying actor than no organ at all, because it launders non-compliance as procedure.
Track cost in the institution’s own authority currency: compliance pull, forum-shopping, participation, invocation-versus-deference, bifurcation into parallel orders.
5a. Issue partition (F6). Institutions hollow domain by domain. Global legitimacy claims are almost always wrong.
5b. Time series (F8). Distinguish event-driven decay (discrete blocking acts) from attrition-driven decay (term expiry, renewal lapse, deadline slippage).
5c. Benchmark selection. Where the institution publishes its own benchmark, use it — this supplies a falsification standard internal to the institution rather than imported by the analyst, and is the strongest available Stage 5 evidence. Where it does not, a benchmark may be constructed from the institution’s own formal premise (its stated representational basis, its stated purpose, its own technical staff analysis). Mark analyst-constructed benchmarks explicitly and weight them lower; they are contestable in a way published benchmarks are not.
5d. Spillover (F29) — NEW. In voluntary-membership systems, legitimacy loss is largely contained within the obstructed body and displacement is cheap for everyone else. In captive systems the terminal displacement routes — branch-lateral and extra-institutional — impose costs on the surrounding constitutional order rather than on the blocked organ. The obstructed body can therefore remain formally intact and internally stable while the damage accumulates elsewhere, which means Stage 5 measured only at the blocked organ will systematically understate total cost. Always ask where the bill is actually being paid.
6.1 Enumerate constitutive channels (F22) — NEW, and mandatory before any verdict.
Rule change is not synonymous with amendment. Enumerate every channel by which the operative rule can be altered, with its threshold:
v0.2 assumed the first. That produced the method’s second false negative, and unlike the first (Type 9, found by inspection) this one was falsified by historical event — the sealed verdict was wrong because the seal had already been broken, repeatedly, through a channel the spec did not model.
6.2 Identify the smallest change that makes the probe reachable, and run it back through Stage 1 against each constitutive channel.
6.3 Preference-revelation test (F23) — NEW. If a constitutive channel exists whose threshold the blocked coalition already meets, and it goes unused, then the rule is not the obstruction. Return the verdict with the masked flag and state what is actually blocking. Do not accept participants’ procedural framing on either side.
6.4 Norm-lock check (F19). If Stage 2 found no node, do not return “reformable.” Apply the Type 9 discipline criteria; if criterion 3 fails, the correct finding is an unexercised node, not a norm-lock.
6.5 Package space (F16). In threshold architectures, test whether a multi-party log-roll clears the threshold without touching the blocker’s own position.
6.6 Rigidity migration (F13). Model where displaced pressure lands after the fix. A perturbation proposal that does not name the next chokepoint is incomplete.
Design rule, confirmed 4/4: viable reform routes around the chokepoint and drops the consent threshold.
The Stage 6 fix becomes a new normative desire. Return to Stage 0.
Observed property: in consent-unanimity architectures the loop terminates in one cycle. In systems with plural constitutive channels it does not — the fix is genuinely reachable, and the recursion continues into the political question of why it is not taken. That non-termination is informative, not a defect: it is the signature of a masked verdict.
G1 — Negative control (mandatory). Run the method on a desire in the same institution that succeeded. Prefer a control from the same body, period, and actors, differing only in decision rule.
G2 — Anti-teleology gate. Stage 3.
G3 — Norm-lock check. Stage 6.4, with §1.4 discipline criteria.
G4 — Non-tabling test. Stage 2.6.
G5 — Multi-architecture protocol. Do not promote a finding on a single architecture. Minimum validation set is now four: designated-veto, consensus/unanimity, threshold, captive-membership. Each addition to the set has broken something; treat the set as a floor, not a ceiling.
G6 — Constitutive channel exhaustion (NEW). No sealed verdict may be returned without enumerating every channel in Stage 6.1 and showing each is blocked. This gate exists specifically because the v0.2 spec would have returned sealed on a rule that had already been changed four times.
G7 — Masking audit (NEW). For every soft-sealed or sealed verdict in a case where participants publicly attribute failure to procedure, run the preference-revelation test explicitly and record the result, including when it comes back negative.
SCA is refuted or materially wrong if:
| ID | Class | Provision | Holder | Holder type | Hardness | Default direction | Attribution cost | Nesting | Meta-node | Cost-block | Cost-enable |
|---|
ID: SCA-C###
Case: <institution / file>
Probe: <probe ID>
Type: <1-10> [+ provisional flag if Type 9 criterion 3 unmet]
Node: <node ID | NONE>
Displacement: <routes, by axis, with threshold comparison and deformation cost>
Const. channels: <channel: threshold: status>
Stage 6: reformable | soft-sealed | sealed | norm-locked [+ masked]
Time series: <status at t, t+n>
Spillover: <where the cost is borne>Stages 0–7, gates G1–G7, node table, register entries, findings contributed.
Architecture: designated veto. Role: origin case.
Stage 0. Five probes: Chapter VII binding measures (P1); ICC referral (P2); cross-border access without state consent (P3); Art. 42 authorization (P4); chemical weapons disarmament (P5). P5 succeeded — built-in negative control.
Stage 1. P1, P2, P4 reachable by text, unreached; blocked at Art. 27(3). Vetoes 4 Oct 2011, 4 Feb 2012, 19 July 2012; ICC referral vetoed 22 May 2014. P3 delayed to Res. 2165 (July 2014). P5 reached — Res. 2118, unanimous, 27 Sept 2013.
Stage 2. Low node count, absolute hardness. Cost-of-blocking approximately zero. Cost-of-enabling high: Tartus, client retention, and decisively the Libya Res. 1973 precedent — a cross-case contaminant invisible from the instant file.
Stage 3. Types 1, 3, 4, 5, 6, and Type 7 first observed (Art. 2(7) sovereignty held by the besieging party).
Stage 4. Horizontal displacement throughout: GA Res. 66/253; Uniting for Peace conspicuously not invoked; Commission of Inquiry; IIIM created by the General Assembly because it has no veto; OPCW-UN JIM, then the OPCW Investigation and Identification Team via treaty-body plenary; Astana/Sochi displacing the political track into a format run by the blocking power; Art. 51 and “unwilling or unable”; unilateral strikes 2017 and 2018. Route contestation: the cross-border mechanism strangled at renewal — four crossings to one, shortened windows, lapse July 2023.
Stage 5. GA Res. 76/262 (2022) now triggers automatic General Assembly debate after any veto. Issue-partitioned — Res. 2118 showed the Council functioning where P5 interests converged. Epilogue: no UN sanctions regime on the Syrian government across 2011–2024; the file unified only after the regime fell, with the Council’s December 2025 Damascus visit signalling unity after years of division. Signature of a sealed contradiction: not resolved, outlived.
Stage 6. Veto limitation → Arts. 108/109 → sealed. Constructive abstention → soft-sealed. Bypass routes all succeeded.
G1: passed. Cost asymmetry predicts Res. 2118 — the measure was framed as disarmament rather than accountability and the blocking power co-authored it. Did not predict the decay (→ F8).
Findings: F1–F8, Type 7.
Architecture: distributed consensus. Special feature: contains a natural experiment — same institution, members, and years, two decision rules side by side. Panel establishment and report adoption run on reverse consensus and kept functioning; appointments run on positive consensus and died.
Stage 0. Fill vacancies (a); close the appeal-into-the-void loophole (b); constrain jurisprudence to DSU Art. 3.2 (c); preserve automaticity (d). (a) and (c) are in tension; most reform drafts founder on sequencing.
Stage 1. (a) reachable, unreached. Blocking since 2016 across administrations of both US parties; the last sitting member’s term expired 30 November 2020.
Stage 2 — principal discovery. The attack was on staffing, not the substantive path — under reverse consensus, rulings could not be vetoed. Exposed the maintenance-node layer (F9) and retro-explained the Syria renewals. The body died by term expiry: the holder never acted, only withheld (F10). Blocking consensus generates no record (F12).
Stage 3. Type 7 travels, and the Syria framing was too narrow — the most-frequently-sued member gates appointment of the judges; roughly a quarter of all disputes involve US trade policy. Renamed target-held gate. Type 8 first observed — Rule 15 continuation, 90-day overruns, de facto precedent against Art. 3.2, zeroing.
Stage 4. MPIA under DSU Art. 25 — 58 members, close to 60% of world trade after Viet Nam’s accession — durable because bilateral arbitration requires no plenary decision. Confirms the bypass design rule. Deformation cost near zero, which is why F24 was not detected here: MPIA does substantially the same job in substantially the same form. Route contestation manifested instead as exploit conversion — losing members appeal into the void, leaving disputes in limbo (F11).
Stage 5. Issue-partitioned; panels still sit. Outcome is bifurcation — multilateral and non-binding versus plurilateral and binding, two legal orders in one treaty. MC12’s 2024 target produced a formalized reform process under a Facilitator appointed April 2024 and no restoration; the co-sponsored appointment proposal has been tabled at essentially every DSB meeting for six-plus years.
Stage 6. Amendment and selection both require consensus. Soft-sealed at every level. Closure problem in pure form.
G1: passed, stronger than Syria’s — panel establishment under reverse consensus, same body and actors, opposite outcome.
Findings: F9–F14, Type 8, Type 7 generalization.
Architecture: threshold. Role: boundary test.
Stage 0. Realign quota shares (a); increase resources (b); raise basic votes (c); end the European convention on Managing Director selection (d).
Stage 1. (a) and (b) require 85% of total voting power against a US share of roughly 16.5%. (b) succeeded — 16th Review, December 2023, 50% increase, 92.86% in favour. (a) failed; last realignment was the 14th Review in 2010.
Stage 2. Threshold collapses into a node where calibration puts one holder above the margin — and the calibration is deliberate: reform analyses compute the maximum basic-vote increase that would not erode the US veto. Three departures: latent coalitions (EU-27 clears 15% but fragments across constituencies) (F15); anticipatory non-tabling (the Board decides by consensus and essentially never votes) (F17); nested ratification (the 2010 reform waited five years on US congressional action; Treasury remains committed to securing approval for the 16th Review) (F18).
Stage 3. Type 7 travels — 3/3. The entity to be diluted holds the gate on dilution. Type 8 revised — conditionality overreach shows it is not judicial-specific.
Type 9 first observed — NOW DOWNGRADED. The Managing Director has always been European by pure convention, with no textual basis. v0.2 recorded this as Type 9 and it produced the method’s first false negative (Stage 6 returned “reformable”). Under the F27 discipline criteria: criterion 1 holds; criterion 2 holds (European constituency coordination plus the reciprocal US claim on the World Bank presidency); criterion 3 is not established — EMDE candidacies have been mounted, but no coalition possessing the arithmetic has coordinated to break the convention and been defeated. Reclassified: provisional Type 9, with the possibility that this is an unexercised node rather than a true norm-lock. Resolving this is a v0.4 register task.
Type 10 first observed. The 2021 general SDR allocation cleared 85% and delivered $650bn — because allocations are proportional to existing shares. The 16th Review increase was equiproportional for the same reason. Both passed by stripping the redistributive element. The Fund grew twice and both times the misalignment deepened, because the growth was distributed by the shares under dispute. EMDEs including China account for 60% of global GDP and hold 40% of voting power. Not stasis — accelerating divergence through successful institutional action.
Stage 4. Internal migration toward NAB and bilateral borrowing, which bypass the quota rule; note the reverse move — the 16th Review increase replaces bilateral borrowing agreements and is linked to an NAB rollback, pulling the bypass back into the blocked channel while conceding nothing on shares. External: regional financing arrangements, BRICS CRA, AIIB, NDB, swap lines.
Stage 5. Self-quantification available here and nowhere else in the set — the calculated quota formula supplies an internal benchmark. The G-24 has said the failed review undermines the organization’s legitimacy and credibility; the IMFC has acknowledged the urgency of realignment. The governing committee concedes the diagnosis and cannot act on it.
Stage 6. Soft-sealed with a materially wider feasible set, because package space exists (F16). The 2010 reform is the proof: Europe surrendered board chairs, EMDEs gained shares, the US veto was preserved intact. The 17th Review runs to 2028; the June 2025 realignment-approaches milestone passed without one.
G1: passed — both successes are predicted by the reframing rule (F21).
Findings: F15–F21, Types 9 and 10, F14 revision, G3.
Architecture: captive membership, supermajority threshold, plural constitutive channels. Role: exit-impossibility test, per external review. Result: the method’s second false negative, and the confirmation of the reviewer’s masking hypothesis.
Probes:
P4 is the built-in negative control. It succeeds routinely, in the same chamber, under the same members, in the same sessions.
All reachable by text. Rule XXII sets cloture at three-fifths of senators duly chosen and sworn (60) for measures, and — critically — two-thirds present and voting for changes to the standing rules themselves, which is the textbook closure problem.
P5 by the textual route: reachable, unreached. January 2022 rules-change vote failed 48–52.
But Rule XXII is not the only constitutive channel — see Stage 6.
Substantive node: cloture on the measure. Threshold holder; 41 senators. Hardness marginal rather than absolute — many possible blocking coalitions, but party discipline makes one of them near-permanent, so it behaves as designated in practice.
Attribution cost is low but non-zero: unlike WTO consensus, a failed cloture vote is recorded and roll-called. Default direction is toward inaction — a bill dies without anyone acting.
Adjudicated node (F26): the Senate Parliamentarian applying the Byrd rule to reconciliation measures. This is the case that forced the adjudicated holder type into the ontology. The Parliamentarian is not a party exercising preference; she applies a stated test determining which provisions may travel on the privileged track. In February 2021 a minimum-wage provision was ruled non-compliant and stripped from the reconciliation vehicle, which is the clearest single instance of an adjudicated node determining substantive policy outcomes. The meta-node is real and lower-threshold — the Parliamentarian is appointed by the Majority Leader and has been dismissed before, and rulings can in principle be overridden — which is precisely why the adjudicated node must be recorded with its meta-node attached.
Maintenance nodes: budget resolution slots (a finite number of reconciliation vehicles per fiscal year), calendar and floor time, committee gatekeeping.
Geometry (F25). No horizontal axis exists — there is no parallel legislature to defect to. Displacement runs:
Stage 4 does not break. The reviewer’s predicted failure mode did not occur.
4d. Channel deformation — the case’s principal methodological yield. Reconciliation’s Byrd-rule admission test permits only provisions with non-incidental budgetary effect. Consequence: policy is redesigned to be fiscally legible rather than effective. Regulatory approaches become tax credits; mandates become subsidies; structural reform becomes spending. The substance is not merely delayed — it is transformed by the shape of the surviving channel.
The international cases hid this entirely. MPIA does the appellate job in the appellate form, so deformation cost was zero and the stage never noticed the variable existed.
5c benchmark: no published institutional benchmark exists. Analyst-constructed alternatives — the gap between polling majorities and enacted output, and the chamber’s population-to-seat ratio against its own stated representational premise — are usable but contestable, and are marked as lower-weight accordingly. This is a genuine evidentiary step down from the IMF case.
5d spillover — new and important. The Senate itself is not visibly hollowing. It functions, confirms, funds, and passes reconciliation bills. The costs land elsewhere: on executive-branch legitimacy (policy by unilateral action, reversed each administration), on the judiciary (asked to resolve questions the legislature did not), and on federal-state coherence. A Stage 5 run confined to the blocked organ would have found a healthy institution. This is why the spillover term is now mandatory: in captive systems the bill is presented to a different body than the one that incurred it.
6.1 Constitutive channel enumeration — where v0.2 fails.
| Channel | Threshold | Status |
|---|---|---|
| Textual amendment of Rule XXII | 2/3 present and voting | Blocked |
| Precedent via point of order (“nuclear option”) | Simple majority | Open, and repeatedly used |
| Standing order / organizing resolution | Varies | Partially open |
| Privileged track (reconciliation) | Simple majority, Byrd-constrained | Open with deformation cost |
The precedent channel changes the operative rule without touching the text: a point of order is raised, the ruling is appealed, and the appeal is settled by simple majority — establishing a precedent that governs practice notwithstanding the plain meaning of the written rule.
It has been used repeatedly. In November 2013 cloture on most nominations was reduced to a simple majority; in April 2017 that was extended to Supreme Court nominations; further precedents followed in 2019; and in September 2025 the chamber established, on a party-line vote, that unlimited nominees may be confirmed en bloc without the previously required 60-vote threshold. Commentary on the 2025 action notes the pattern explicitly — narrow, justify, normalize — with each majority citing its predecessors’ precedents when pushing further.
SCA v0.2 would have returned sealed on P5, on the textual reading of Rule XXII’s two-thirds requirement. That verdict is falsified by four documented instances of the rule being changed by simple majority. This is the framework’s second false negative and the first falsified by event rather than inspection. Hence F22 and gate G6.
6.3 Preference-revelation test — the reviewer’s hypothesis, confirmed.
A constitutive channel exists at a threshold the blocked coalition already meets. It goes unused for legislation. Therefore the rule is not the obstruction. The correct verdict on P1, P2 and P3 is soft-sealed [masked]: what blocks them is the unwillingness of members of the nominal majority to use an available channel — a position with its own stated reasons, including the expectation of reciprocal use by a future majority and a substantive commitment to the rule’s minority-protective function.
That is a materially different diagnosis from “procedurally impossible,” and it is the diagnosis participants on both sides have incentives to avoid: it denies the majority the excuse of impossibility and denies the critics the claim of a locked institution. The 2022 rules vote is the cleanest available evidence, since it was the coalition itself, not the rule, that came up short.
6.6 Rigidity migration. Where would pressure land if the legislative threshold were lowered? Onto the remaining maintenance nodes — floor time, calendar, committee gatekeeping, and the adjudicated node — and onto reciprocal reversal across control changes. The 2025 commentary anticipates exactly this: targeted changes accumulating into institutional drift, with each change inviting reciprocal changes by future majorities.
“Change the cloture precedent” → Stage 1 → reachable at simple majority → the loop does not terminate. Every prior case closed in one cycle because the fix was genuinely unreachable. Here it is reachable, so the recursion runs on into why it is not taken — which is a question for RDPE, not SCA.
Non-termination is now a diagnostic signal, not a defect: it is the signature of a masked verdict.
P4, nominations. Same chamber, same members, same sessions, differing only in the operative threshold — which is itself the product of the precedent channel. Confirmations proceed routinely; legislation does not. Stage 2 predicts this correctly, and the control is unusually clean because the treatment (threshold change) was applied to one probe class and not another within the same body, by the body itself.
Findings contributed: F22–F29; adjudicated holder type; Type 9 discipline criteria (applied retroactively to Case 3); gates G6 and G7.
| ID | Finding | Source | Disposition |
|---|---|---|---|
| F1 | Desires arrive as probe bundles with divergent verdicts | UNSC | Stage 0 |
| F2 | Veto density conflates node count with node hardness | UNSC | §1.2 |
| F3 | Cross-case precedent contamination | UNSC | Stage 2.5 |
| F4 | Type 7 added | UNSC | §1.4 |
| F5 | Displacement is contested, not static | UNSC | Stage 4c |
| F6 | Legitimacy accounting must be issue-partitioned | UNSC | Stage 5a |
| F7 | Soft-sealed verdict added | UNSC | §1.5 |
| F8 | Snapshot → time series | UNSC | Stage 5b |
| F9 | Maintenance nodes are a separate, softer layer | WTO | §1.2, Stage 2.2 |
| F10 | Default direction; inaction as weapon | WTO | §1.2 |
| F11 | Exploit conversion | WTO | Stage 4e |
| F12 | Attribution cost; consensus less accountable than veto | WTO | §1.2 |
| F13 | Rigidity migration | WTO | Stage 6.6 |
| F14 | Type applicability fields | WTO / IMF | §1.4 |
| F15 | Hardness as coalition distribution | IMF | §1.2 |
| F16 | Package space in threshold architectures | IMF | Stage 6.5 |
| F17 | Anticipatory non-tabling | IMF | Stage 2.6, G4 |
| F18 | Nested ratification | IMF | §1.2, Stage 2.7 |
| F19 | Type 9; norm-lock false negative | IMF | §1.4, Stage 6.4, G3 |
| F20 | Type 10; reinforcement through the permitted channel | IMF | §1.4 |
| F21 | Reframing rule promoted to core | All | §1.3 |
| F22 | Constitutive channel plurality; precedent as a lower-threshold route to rule change | Senate | Stage 6.1, G6 |
| F23 |
Preference-revelation test; masked verdict flag |
Senate | §1.5, Stage 6.3, G7 |
| F24 | Channel deformation — bypasses reshape substance to fit their form constraint | Senate | Stage 4d |
| F25 | Displacement geometry; Stage 4 re-axes in captive systems rather than breaking | Senate | Stage 4a |
| F26 | Adjudicated holder type, with mandatory meta-node | Senate | §1.2, Stage 2.8 |
| F27 | Type 9 discipline criteria; residual-category risk closed | Senate / review | §1.4 |
| F28 | Analyst-constructed benchmarks where self-quantification is absent; lower evidentiary weight | Senate | Stage 5c |
| F29 | Spillover — in captive systems the legitimacy cost is borne outside the blocked organ | Senate | Stage 5d |
Confirmed 4/4: Type 7; cost asymmetry; the reframing rule; the bypass design rule; issue-partitioned legitimacy; maintenance-node vulnerability.
Architecture-sensitive: hardness modelling; package space; observability of blocking events; nesting; displacement geometry; deformation cost; recursion termination.
Method failures found, cumulative: two false negatives (Type 9 norm-lock, Stage 6 sealed-verdict), two missing pathologies (Type 10, channel deformation), two missing actor layers (nested ratification, adjudicated holders).
MF1. Every architecture added to the validation set has broken something, and no architecture has yet been added without doing so. There is no basis for treating the set as saturated. G5’s floor of four is a floor.
MF2. The failure the fourth case actually produced was not the one predicted. The reviewer predicted Stage 4 would collapse where exit is impossible; instead Stage 4 re-axed and survived, while Stage 6 — the stage nobody flagged — turned out to rest on an unexamined assumption that amendment is the only route to rule change. The lesson generalizes: the stage most likely to fail is the one whose assumptions have never been stated as assumptions.
MF3. The v0.2 typology was drifting toward one new type per case. v0.3 adds none and instead converts both candidates into stage mechanisms and verdict flags. A framework validated by adding categories is not being validated. The bar in §1.4 exists to make that discipline explicit rather than aspirational.
MF4. Cost asymmetry is now doing more work than any other component and remains the least formalized. It is the v0.4 priority for that reason, not because a scored index is intrinsically better than prose.