# Named and Counted, Not Blamed nor Accounted v5.1 #### A metrological instrument for measuring structural patterns — naming what had no name **B. Greenway** ThinkWell Labs Metrology · 2026 *Companion to the live instrument at toroid.fyi/sabbot and the Polymorphic Catalysis Toroid.* *Sometimes things are invisible until they are named.* *v5.1 errata (July 2026): a drafting-artifact inter-rater-reliability claim has been removed — no record of such a test exists and none was run; the band-threshold "dimension count" derivation is withdrawn; the IRS row is updated to the live instrument's corrected scores; distribution-package file names corrected. Changelog at the end.* ## Preface — Not an Accusation; an Observation **What this is.** A measurement tool; and the first honest attempt to use it. The claim is small to say and hard to earn: a kind of governing can be measured, not merely alleged. The kind in question exhibits consistent directional patterns in outcomes... patterns that appear under the guise of ordinary administrative change. When you break something you didn't mean to, you fix it. Nothing has been fixed. Most writing stops at the accusation holding only empty claims. This piece builds the instrument that will prove me wrong. Five signatures; one composite score; a way to watch how fast that score climbs, and whether the institutions climb together; a set of worked cases; and, by design, a control group of prior reforms the tool should rate *low,* because a tool that can only rate things high measures nothing at all. **What it is not.** Proof. The scores are judgments; anchored to dated public evidence, but judgments, and so far mostly one person's. The grimmest numbers here— coverage lost, revenue lost, deaths modeled— are forecasts, and they say so every time they appear. The authority is the evidence; nothing besides. The whole of it is built to be argued with. To be attributed to other causes. One rule carries the book, so I will set it down plainly. It measures; it does not pronounce. Where it says a thing happened, there is a record. Where it says a thing is projected, there is a model; not a body count. Where it scores, the score is an argument asking to be answered. Read it as so, and it will not mislead you. Lose that line, and it will; and so would I. ## On accusation, intent, and proof I will not pretend I came to this neutral. I saw a pattern; I believed it was deliberate; and I built this apparatus to learn whether I was wrong. That origin is not a confession to apologize for; it is the reason the method has teeth. A conviction is worth nothing until you build the thing that could refute it. So I built the controls that show prior reforms the instrument was supposed to rate low... and I watched it rate them low. An instrument that can acquit is the only kind whose convictions carry any consequence. So the book makes a deliberate choice about how far it will go. It names; it counts; it stops. It does not blame, and it does not account. That restraint is not timidity; it is the strategy. An accusation asserted is cheap, and easy to wave away: *prove intent.* An accusation a reader reaches on their own, because the evidence left no other door, is one they will defend as theirs. So I lay the evidence down and decline to read the verdict aloud. The verdict is the reader's to reach. Intent is handled in two registers, and the difference is the whole discipline. Where intent is on the record, I cite it; a strategy its own authors published is not an accusation but a quotation. The Two-Santa doctrine was written down. "Starve the beast" is a slogan its architects say in public. A bipartisan border bill was killed after its opponents said, on camera, to blame them for the killing. Naming these is reporting, not inference. But where intent is hidden, I do not assert it. I measure the structure, and show that it behaves with directional asymmetry: selective, resistant to its own repair, arriving everywhere at once. And I leave standing the one question such a structure raises: what accounts for patterns this consistently one-directional? An answer will arrive within you without me saying it. A word, then, on a word. *Structured* is used sparingly and on purpose, reserved for the places where avowed doctrine or the full evidentiary threshold earns it. Everywhere else the structure is said to *exhibit directional asymmetry,* or to be *structurally produced* — the same evidence, the same weight, but every strong claim carrying a receipt. This is not hedging. It is the difference between a measurement and an accusation; and the measurement, in the end, is the more dangerous of the two. The title is the whole posture in five words: *named and counted* is the author's work; *blamed and accounted* is the reader's. ## How to read this book The framework is named through a *linguistic-autopsy* method — coining precise terms (*Obfuscratic*, *Directional Asymmetry*) so that what was previously only felt can be specified and measured. The naming is not the proof; it is the precondition for the proof. Part I builds the concept and the five signatures. Parts II–VI build the measurement machinery and, crucially, the *negative controls* — the cases the instrument must score *low*, including a near-identical prior reform, because an instrument that can only return high scores is an accusation, not a measurement. Part VII applies it. The method the instrument embodies has a name and a specification of its own: *caecusoscopy*, the disciplined study of structural blind zones through the traces they leave downstream — language drift, behavioral shifts, incentive gradients, institutional deviations — without asserting the hidden cause directly. What follows is a caecusoscopic application: it studies a regime-character it cannot observe directly, through the measurable footprints it leaves. A live, interactive version of the instrument — clickable architectures, the comparison table, the temporal and resource-operation layers, and the diagrams — runs at **toroid.fyi/sabbot**. The book is the argument; the instrument is the apparatus. ## A note on the evidence tags Throughout, factual items carry one of three tags. **Occurred** — a documented event verifiable against the contemporaneous public record. **Projected** — a modeled or forecast outcome (a CBO estimate, a mortality model), legitimate as evidence but never presented as realized. **Contested** — magnitude, attribution, or characterization is disputed; whether that lowers confidence depends on the *quality* of the contestation, not its mere existence. Motivated denial does not downgrade a documented fact; motivated affirmation does not upgrade a contested one. A fourth tag applies where relationships, not events, are at issue. **Correlation** — a relationship has been measured between variables, without implying causal direction. Causal claims require additional evidence beyond correlation and are explicitly marked when made. ## A note on terminology and orthography This book coins terms, deliberately. New phenomena need new names; the alternative is to keep describing them with words built for something else, which is how they stay invisible. Three conventions govern the coinages, stated here so that what follows reads as intent, not error. **Method.** Terms are formed by a *linguistic-autopsy* procedure: a base word is stripped to its root and recombined with a classical morphological suffix that carries the intended category. The suffix *-archy* (Greek, *rule* — as in monarchy, oligarchy, anarchy) marks a structural **regime form**; the suffix *-cratic* (Greek, *governing* — as in democratic, technocratic) marks an **operational modality**. Thus *Sabbotarchy* names the regime form (rule by sabotage — the verdict-laden noun, reserved for the reader) and *Obfuscratic* names the modality (how such a regime operates: through systemic incoherence). *Directional asymmetry* is the third term and the only one that is not a suffix coinage, because it names neither a regime nor a modality but the **measured property** itself — the observable a non-suffixed descriptor is correct for. The distinction between the suffixes is load-bearing, not decorative, and the property's plainness is deliberate: the thing actually measured should wear the least theoretical name in the set. **Hybrids are acknowledged.** A few terms join Latin roots to a Greek suffix. Such Greco-Latin hybrids are irregular by classical standards but well-precedented in working English (*television, sociology, automobile*); they are used here for clarity over purity, knowingly. The point of these conventions is not novelty for its own sake. A coined vocabulary reads as authoritative only when every coinage is consistent and owned; an unexplained anomaly reads as a mistake. Everything that follows obeys these rules. --- ## Contents - **Preface** — What this is, and what it is not - **On accusation, intent, and proof** — the measure-don't-pronounce doctrine, and the two registers of intent - **Part I** — The Framework: the Obfuscratic Asymmetry Signature and the five dimensions - **Part II** — The Scoring Rubric - **Part III** — Evidence Discipline: the anchor audit and the contestation-quality filter - **Part IV** — Temporal Dynamics: velocity, breadth-velocity, and onset-synchrony - **Part V** — Resource Operations: selective allocation, asset dissipation, harm-utility coupling (S6–S8) - **Part VI** — Validation and Negative Controls: can the instrument return a low score? - **Part VII** — Application: the back-scored architectures, the reform-lineage baseline, projected harm, the doctrinal substrate, and the two adversarial tests the instrument has yet to face - **Appendix** — Distribution note and the two-instrument stack --- # Where This Begins: The Case Against the Book Begin with the strongest form of the objection, because if it stands, nothing after it survives. Government fails. It fails constantly, under every party, in every administration; it fails from scarcity, from turnover, from the ordinary entropy of large systems run by tired people with too little time. Most of what looks like malice is fatigue. Most of what looks like design is the *absence* of design — no one steering, which from below reads exactly like someone steering badly. A reasonable observer watching an agency produce ruin should reach for incompetence first and keep reaching for it long after the drama tempts him elsewhere. The dramatic reading is almost always the weaker one. I concede all of it at the outset, and without reservation, because the book is worthless if it cannot survive the concession. So here is the one thing the concession cannot absorb. Ordinary failure is symmetric, and slow, and uncoordinated. It wounds the people it was meant to help and the people it was meant to spare; it arrives in one office before another, on no schedule anyone could keep; and it yields to cheap repair, because no one is standing guard over it. What this book measures is failure that breaks all three habits at once — failure that routes its damage in a single direction, that fights the free fix as hard as the costly one, that arrives across a dozen institutions in the same season as if to a calendar. That is not the shape of fatigue. Fatigue keeps no direction and no schedule. The residue that ordinary incompetence cannot account for — the direction, the inverted cost of repair, the synchrony — is the whole of what is named and counted here. Everything else, I hand the skeptic, freely. A pattern with no name cannot be measured; and what cannot be measured cannot be argued, cannot be budgeted, cannot be refused. The sense that something in public life is failing too reliably to be only failure is widely held and almost never pinned, and it stays unpinned for want of words exact enough to carry it. This book supplies the words, then the ruler, then the readings — and, by design, a set of cases the ruler is built to rate *low,* because an instrument that can only convict has measured nothing. Name the parts; count them. Not yet blamed, not yet accounted; that waits, and arrives only if the count earns it. Invisible things, once named, stop being invisible. That is the wager of everything after. --- # Part I — The Framework *The Obfuscratic Asymmetry Signature, and the five dimensions that measure it.* --- ## 1. Two failures this framework is built to end There are two ways the conversation about state dysfunction goes wrong, and they fail in opposite directions. The first failure is naming without measuring. An observer watches an agency produce ruin too consistently to feel like accident; he says the dysfunction is deliberate, manufactured, designed. He may be right. He cannot show it. No instrument exists to separate built failure from ordinary failure, so the claim collapses into hyperbole, and a critic answers that this is just normal government mess, or just one administration's politics, and the argument loops forever. Nobody settles it, because nobody has a ruler. The second failure is the mirror image: measuring without naming. Researchers count one outcome at a time — coverage losses under one program, deportation rates under another, audit-rate shifts under a third — and treat each as a discrete policy choice. The forest disappears into the trees. The pattern that connects a workforce purge, a benefits administration, a detention build-out, an enforcement teardown, a research compression, and a regulatory rollback stays invisible, because no one is looking at them as one shape. This book ends both failures with one move. It supplies the measurement primitives that let any single architecture be scored, then applies those primitives uniformly across many, so the cross-architecture shape becomes legible. Name the thing, then count it. That is the whole method, and the rest of Part I builds the apparatus that does it. ## 2. The names, and why the spelling carries the argument The apparatus names three things, and keeps them on three different rungs of the evidentiary ladder, because collapsing them is how a measurement turns into an accusation. **Directional asymmetry** is the measured property — the bottom rung, the thing the five signatures actually read. It describes a structure whose outputs diverge from its stated inputs non-randomly: the failure keeps a direction. It is a *what*: observable, scorable, and silent on why. **Obfuscratic** is the operational modality — *how* a directionally asymmetric structure sustains itself. From *obfuscation* plus *-cratic,* it names a regime that works through systemic confusion: framing-mismatch, incoherent public information, dysfunction worn as incompetence. It is an adjective; it modifies, it does not pronounce. **Sabbotarchy** is the regime-form — the top rung, and the one the instrument never climbs on its own. It names what a reader may *conclude is present* when directional asymmetry concentrates across all five signatures and operates obfuscratically: rule by sabotage. The word is deliberately frightening, and deliberately withheld. The instrument measures the property and the modality; whether the pattern warrants the regime-name is the reader's verdict, reached on the reader's authority — or not reached at all. The measurement therefore travels under the compound that names only what it measures: **Obfuscratic Asymmetry**, modality joined to property. *Sabbotarchy* waits above it, unspoken, for a reader to reach. An earlier revision of this book let the frightening noun drop entirely in the name of caution; that was an overcorrection, because it left the framework with no name for the thing it was pointing at and a mere property doing a noun's work. The three-tier restores the order — a measured *what*, a structural *how*, and a verdict-laden regime-name held in reserve — each on its own rung, none borrowing the others' authority. A transparently asymmetric regime would be recognized and resisted; an obfuscratic one with no underlying asymmetry would have nothing to hide. Property and modality move together or not at all. The regime-name moves only when a reader decides it does. Naming settles nothing on its own. It is the precondition for everything after: without precise words for the property, the modality, and the regime-form they may amount to, the measurement problem cannot even be stated. With the words in hand, and kept on their rungs, it can. ## 3. The five signatures Honest failure and the failure this book is about can leave identical wreckage; the size of the damage tells you nothing about which one you are looking at. The shape of the damage tells you everything. The five signatures below read that shape. Each is a trace left downstream, an asymmetry you can see and check, and each stops short of any claim about a person's private intent. ### S1 — Outcome-distribution asymmetry Every operation has a stated beneficiary; the population it is officially conducted to serve. It also has a realized beneficiary; whoever actually gains, once you follow the flow of value, position, and jurisdiction. The first signature measures the distance between the two. Honest policy produces a roughly symmetric distribution. The operation aims to help population X and, with the usual overruns and side effects, in fact helps population X. Symmetry is the fingerprint of intent matching outcome. Where the signature is present, the distribution turns sharply asymmetric: conducted under the banner of helping X, the operation routes value, position, or capture toward Y, a resource concentration vector, a procurement channel, a surveillance substrate, a wealth-concentration engine. The asymmetry is the tell that the stated reason was cover for the working one. Measuring it takes three steps. Name the stated beneficiary from the official documents and rationales; identify the realized beneficiary by tracing where the resources actually went and which populations actually gained or lost; then compute the divergence, qualitatively by the category mismatch and quantitatively by the magnitude of concentration. ### S2 — Domain-selective competence Every architecture performs many functions. Some face the public: taxpayer service, benefit delivery, records access, court compliance. Some are extractive: data fusion, contract execution, surveillance integration, enforcement against disfavored targets. Honest competence does both well; honest incompetence does both poorly. The signature is selectivity: catastrophic failure in the public-facing work, sitting beside surgical precision in the extractive work, in the same agency, in the same months, run by the same people, drawing on the same budget. A benefits line that cannot answer the phone while the data-mining that feeds cross-agency surveillance runs clean is not an overwhelmed agency. Both functions drink from one well. A gap that wide has nothing to do with scarcity and everything to do with which function the architecture prioritized. ### S3 — Anti-remediation response When a dysfunction is identified, fixes get proposed. Some are expensive; some cost almost nothing. The third signature watches how resistance to a fix changes with the price of the fix. Honest incompetence resists in proportion to cost: expensive repairs get refused because budgets are real, cheap ones get accepted. The signature resists cheap fixes as hard as dear ones, sometimes harder. When an architecture fights a transparency requirement, a single procedural change, a document release — corrections that cost essentially nothing — with the same force it brings to an expensive rebuild, the resistance has left budget logic behind. What is left is protection of the dysfunction itself. That inverted slope is closed to honest failure; only the guarding of the asymmetry produces it. ### S4 — Personnel selection asymmetry Every architecture decides who is hired, promoted, kept, sidelined, pushed out. The fourth signature reads those decisions on three axes: whether experienced technical personnel stay, whether ideological-alignment candidates rise regardless of competence, whether internal dissent survives. A functional organization keeps its expertise, advances on some mix of merit and alignment, and tolerates disagreement. The signature is sharp on all three: technical people exit or are pushed, alignment advances over skill, and dissent is suppressed before it can pool. What is selected against is competence itself; the capable are the ones who go. ### S5 — Information-architecture asymmetry The fifth signature is the cleanest reading of obfuscratic operation, and it carries the most weight. It measures the ratio between coherence inside the regime and coherence in the public information the regime puts out. A functional architecture is roughly symmetric: it knows what it is doing, and the public can roughly follow what it is doing. A confused architecture is incoherent on both sides at once. The signature is a stark asymmetry: very high internal coherence flanked by very low external coherence. Inside: coordination meetings, integrated implementation plans, aligned talking points, a sustained narrative apparatus. Outside: mutually contradictory framings issued from coordinated sources, systemic ambiguity around specific operations, denial and implementation running at the same hour. That outer incoherence is structural, and it is the cover under which the inner coherence proceeds. This signature carries the heaviest weight in the composite, because it is the one with no honest alternative. The other four can be argued down: resource constraint, ideological commitment, ordinary political churn. Coordinated coherence inside, systemic noise outside, has no such escape. Obfuscratic operation is most legible here. ## 4. The composite, and the bands The five signatures fold into one number per architecture. Each is scored zero to one hundred; the composite is the weighted average: > Index = 0.25·S1 + 0.20·S2 + 0.15·S3 + 0.10·S4 + 0.30·S5 The weights track how structurally specific each signature is. S5 sits highest, at 0.30, because its signature is the least available to any innocent reading. S1 follows at 0.25, the most directly observable. S2 at 0.20 is robust but takes more interpretive work; S3 and S4 sit lowest, at 0.15 and 0.10, because each carries the most alternative explanations, resource limits and ordinary turnover, and demands the most careful evidence to score honestly. The index sorts into four bands. | Band | Range | Reading | |------|-------|---------| | Genuine | 0–30 | Consistent with honest competence or honest failure; low asymmetry. | | Mixed | 30–70 | Some signatures present, not concentrated; more analysis needed. | | High-Asymmetry | 70–85 | The directional signature pattern is satisfied. | | Extreme-Asymmetry | 85–100 | The asymmetry is concentrated across all five dimensions. | The thresholds are calibration conventions on the weighted mean. (An earlier edition claimed they were derived from dimension counts — 70 as four-of-five in the high range, 85 as all five at once. That does not survive arithmetic under unequal weights: four dimensions at exactly 70 with the fifth at 30 produce composites of only 58–66. The claim is withdrawn. The thresholds are retained because the scored set's band membership survives every weight perturbation tested — including dropping the heaviest dimension entirely.) They can be revised as more architectures are scored. **Uncertainty.** These scores are not measurements drawn from a sampling distribution; they are one analyst's judgments, and a confidence interval computed over a sample of one would be fabricated precision — the exact failure Part III forbids. So no ± is attached to any single score. What *can* be reported honestly is computed, not sampled: under every ±0.05 shift of the weight vector each composite moves by at most about one point, and treating each dimension score as uncertain by ±5 puts roughly ±2.5 to ±5 points of soft error on any composite — enough to cross a band boundary near an edge. Cross-analyst replication has not yet been run; until it is there is no rater-to-rater variance to report, and that absence is itself a stated limitation. Read every number here as a *band-level* claim, never a point estimate. A score is shorthand for "this architecture sits in this band, pending a hostile rater"; it is never shorthand for a precise quantity. A point of language, carried from the front of the book and meant literally here. The signatures *measure.* The word *structured* is an *inference* drawn from what they measure, and it belongs at the top of the evidentiary ladder, away from the act of measuring. Where the text says an architecture *satisfies the high-asymmetry signature,* it reports a measured fact about asymmetry, selectivity, and incoherence. Whether to call the architecture *structured* or *sabotaged*, to attribute design, is the reader's step, and the book takes it only where avowed doctrine or full convergence earns it. Everywhere else the disciplined sentence is that the structure *exhibits directional asymmetry.* ## 5. How the scoring works The instrument runs on a hybrid model: seed architectures scored against anchored evidence, plus the infrastructure to extend the same method to more. Each score on each dimension traces to specific evidence: a dated document, an agency report, a court filing, a budget score, a congressional record, an investigative account. The audit trail is preserved and meant to be replicable; an independent analyst can take the same evidence and confirm the score or revise it. That replicability is the point. It is also, candidly, the thing the instrument most needs and has least of so far. To date the scoring is largely one analyst's, and by the framework's own standard that places the cross-architecture findings at a strong candidate inference awaiting independent replication. The book says so wherever the claim is made. **Inter-rater reliability:** none yet. Cross-analyst replication is an open research program, and the framework's external validity rides on it. The rubric's pre-committed standard: two experienced analysts, same evidence, should land within about ten points; looser convergence means the dimension is measuring mood, not structure. (An earlier draft reported a preliminary two-analyst test here; no protocol, raters, or data for such a test exist in the project record, and the claim is withdrawn as a drafting artifact, not a result.) ## 6. What the first scores show The seed architectures cluster, and the clustering is itself a finding—albeit one bounded by a methodological caveat. Every seed scores above 70; most score above 85. A reader should hold one fact in view before reading anything into that: the domain, government-services redirection, was chosen because it was where the signature was predicted to show most cleanly. Therefore, the clustering of high scores in this domain is an expected result of targeted validation, not a random sample proving systemic prevalence. So the honest reading runs the other way around. The cases chosen for being clean tests came back clean, and the whole weight of *that* rests on whether cases chosen as honest reform come back low. They do. Part VI is where the instrument is made to prove it can acquit, and where the prior-administration baseline shows honest reform landing in the Genuine band on a larger workforce cut than any seed. Once the instrument is shown to discriminate, the clustering carries weight within the sampled domain. Honest policy disagreement would scatter architectures across the whole range. A shared signature, appearing across institutions at once, gathers them into the high and extreme bands instead. And it raises one question, only one: what accounts for patterns this consistently one-directional. The book leaves the question with you. ## 7. What the framework cannot do Four limits, said plainly, because an instrument that hides its limits is the kind that gets torn apart. The scoring is interpretive. Every score anchors to evidence, and turning evidence into a number is still judgment, on which two careful analysts may differ. The framework's robustness rides on whether the *rank-ordering* of architectures holds across analysts, not on whether the individual numbers match to the point. Cross-analyst replication is the open work. The dimensions are not independent. An architecture extreme on S5 tends to run extreme on S1, because systemic incoherence is usually deployed to guard a value-redirection. The weighting accounts for this; the dimensions still cannot be treated as clean separate axes. The framework separates the signature from honest failure; it does not, on its own, say *what kind* of design a high score reflects. An architecture at 90 might be built for wealth concentration, for jurisdictional capture, for population subjugation, for coalition consolidation. The signature reports that extreme asymmetry is present and stays silent on which. That resolution belongs to the operational-mode analysis the companion instruments supply. And the scoring is, so far, a snapshot. The signature is a developmental thing; architectures move toward and away from it over time. A single moment misses the trajectory. Part IV builds the temporal layer that begins to repair this, and until that layer is fully populated the limit stands. **Correlation vs. causation:** The framework measures correlations between structural features and outcomes. It does not, and cannot on its own, prove causal direction. Causal claims require additional evidence such as temporal precedence, mechanism documentation, and elimination of confounding variables. Where causal language appears, it is explicitly marked and supported by such additional evidence. ## 8. Where this leaves the instrument The framework holds one specific position, and it is worth saying outright. It is a procedure, not a verdict. It scores an architecture on the signature dimensions and returns a number, and it makes no claim that the architecture is sabotaged. The score is replicable, the evidence is anchored, the method is on the page. An architecture below 30 sits comfortably with honest competence or honest failure, and the framework presses nothing further against it. An architecture above 85 has satisfied a measurable pattern that, on the framework's own terms, is consistent with directional dysfunction operating obfuscratically. The step from that pattern to that verdict belongs to the reader. Its value is twofold. It holds the charge of built failure to the same evidentiary discipline as any other empirical claim. And it makes the cross-architecture shape countable, so the simultaneous gathering of many institutions in the high bands becomes a measured fact instead of a feeling. The framework is built to be argued with. Every score anchors to evidence; every anchor is auditable; every score can be revised when the evidence is challenged or new evidence lands. Its authority comes from the documentation behind it, and from nowhere else. It is meant to outlive any one analyst's use of it. That is the only way a measurement ever earns the right to be called one. --- [Parts II–VI — the scoring rubric, evidence discipline, temporal dynamics, resource operations, and the negative-control validation — are documented in full in the companion methodology files and abridged here.] --- # Part VII — Application: The Back-Scored Architectures Everything before this has been the building of the instrument. This is the part where it is pointed at the world and the readings are recorded. **Seven** architectures are scored across the five signatures, at three dated checkpoints, and the scores are set beside the reform lineage of Part VI so that the high readings can be read against the low ones the instrument is equally capable of producing. Then the resource operations of Part V are traced through the cases where money, force, and harm actually moved. Then the projected harms, kept to defensible ranges and tagged as projections. And then the doctrinal substrate — the published strategies the whole pattern sits against. The verdict is not written here. The verdict is never written here. What is written here is what the instrument measured. > **A note on the numbers.** The scores, the correlation, and the projected ranges in this part are frozen as of June 2026 for print. The live instrument at toroid.fyi/sabbot continues to update against fresh evidence, so its current figures may differ from the ones fixed here. Where this part and the instrument disagree, the instrument is newer; this part is the dated record of where the readings stood when the book went to press. > **On the numbers' precision.** No composite or dimension score here carries a confidence interval, because none is entitled to one: these are single-analyst judgments, not estimates from a sample. The honest uncertainty statement is band-level. No second rater has yet scored these architectures — replication is the open program — and the computed sensitivity bands (roughly ±2.5 to ±5 points) are enough to move a case near a boundary across a band line. Treat every figure below as a claim about *which band*, not about a precise quantity, and treat any score within a few points of a boundary as genuinely ambiguous between the two adjacent bands. ## The seven architectures, scored Each architecture is scored at three checkpoints — first quarter of 2025, third quarter of 2025, first quarter of 2026 — on each of the five dimensions, and the composite is the weighted average defined in Part II. The table below gives the current (2026-Q1) dimension scores and the composite; the earlier checkpoints are used in the velocity discussion that follows. | Architecture | S1 | S2 | S3 | S4 | S5 | Composite | Band | |---|---|---|---|---|---|---|---| | IRS dismantling | 94 | 95 | 89 | 76 | 92 | 91 | Extreme | | DOGE workforce reduction | 88 | 91 | 86 | 84 | 93 | 89 | Extreme | | Regulatory / civil-rights rollback | 88 | 85 | 86 | 84 | 92 | 88 | Extreme | | OBBBA / Medicaid | 92 | 79 | 82 | 73 | 89 | 85 | Extreme | | ICE / immigration enforcement | 87 | 88 | 81 | 79 | 86 | 85 | Extreme | | NIH / research capacity | 82 | 84 | 74 | 68 | 81 | 80 | High | | VA / veterans' services | 80 | 79 | 71 | 64 | 76 | 76 | High | Two features of this table carry more weight than any single row. The first is the spread: the composites run from the mid-seventies to roughly ninety, which means the instrument is *differentiating* among the cases rather than slamming every one to the ceiling. The VA architecture scores meaningfully below the IRS one; an instrument that could not tell them apart would be worthless, and this one tells them apart. The second is the floor. Two architectures sit in the high-asymmetry band rather than the extreme one, and they sit there because their readings really are lower — NIH and VA show real signature but not the cross-dimensional saturation that puts IRS and DOGE at the top. The instrument is grading, not pronouncing. ## Velocity: the same architectures a year earlier The static table hides the most telling fact, which is how fast these readings arrived. Run the composites back to the first checkpoint and the motion is stark. | Architecture | 2025-Q1 | 2026-Q1 | Change | |---|---|---|---| | OBBBA / Medicaid | 16 | 85 | +69 | | ICE / immigration | 65 | 85 | +20 | | IRS dismantling | 68 | 91 | +23 | | Regulatory / rollback | 62 | 88 | +26 | | NIH / research | 56 | 80 | +24 | | VA / veterans | 56 | 76 | +20 | | DOGE workforce | 75 | 89 | +14 | The OBBBA row is the one to dwell on, because it is the cleanest evidence in the book that the instrument is not simply scoring my politics. At the first checkpoint, before enactment, OBBBA scored **16** — squarely in the genuine-pattern band, below the reform lineage's NPR (≈19) and PMA (≈20) and two points above the Grace Commission (≈14). The instrument looked at the thing in its latent state and declined to call it a signature. Only as the structure took form — the realized distribution diverging from the stated one, the information architecture going asymmetric — did the score climb, and it climbed seventy points. A measure that wanted to convict would have convicted at 16. This one waited for the evidence and moved when the evidence moved, and the record of it moving is the record of an instrument doing its job rather than confirming its operator. Set against the reform lineage of Part VI, the separation is total. Grace Commission, 14. Reinventing Government, 19, on a *larger* workforce cut than DOGE. President's Management Agenda, 20. Three prior reforms clustered in the low genuine band; seven current architectures clustered from the high band into the extreme. The gap between those two clusters is the finding, and it is a gap the instrument was demonstrably able *not* to produce, since the lineage proves it can score a large hard reform at 19. ## The resource operations, traced The five-signature scores measure the shape of each architecture's dysfunction. The resource measures of Part V trace where the consequences flowed, and they are clearest in the cases that form chains across dimensions — where a blocked fix becomes a sustained crisis becomes a redirected resource. **Immigration runs S3 into S6 [Correlation].** The 2024 bipartisan border bill was killed after its opponents said, on the record, to blame them for killing it (S3, anti-remediation, a documented refusal of an available fix) [Occurred]. The crisis it would have addressed was thereby sustained, and roughly $170B in enforcement and detention funding flowed into the private-detention architecture that profits from the sustained crisis (S6, selective allocation) [Occurred]. The chain is documented at the structural level; the overreaching claim that the entire migration system was manufactured is one the instrument explicitly rejects, because migration itself is not a domestic creation. What is measured is narrower and firmer: the fix was refused, the crisis persisted, the money was redirected. **Venezuela runs S8 into S7 into S6 [Correlation].** Pre-capture pressure on a target population (S8, harm-utility coupling, where the pressure was structurally load-bearing to the negotiation posture) preceded the capture of Maduro in January 2026 (S7, expenditure of force and legality) [Occurred], after which Venezuelan oil was reopened to U.S.-licensed extraction (S6, the resource concentration vector): sanctions lifted, trading licenses issued, a fifty-million-barrel supply arrangement, with sale proceeds flowing through U.S.-licensed private channels rather than into Treasury accounts directly [Occurred]. The precise custody of the proceeds is the point to state carefully — control is exercised through the licensing regime, not through literal government accounts — and the chain is scored on that narrower, firmer fact. This is the gravest of the resource chains and the one scored most conservatively; it is tagged occurred at each link, and it is not characterized as a bailout, because the direction of flow is extraction rather than rescue. **The $1.776B audit-bar structure [Occurred/Contested]:** Not a settlement — a withdrawal. A president sued an agency he commands; the court signaled it might rule the case a sham, lacking genuine adversity under Article III; and two days before that ruling he dismissed his own suit *with prejudice* under Rule 41(a)(1)(A)(i), a self-executing notice available because the government had filed no answer [Occurred]. On the same day, by a one-page addendum signed by his own acting attorney general — formerly his personal defense counsel — the IRS was "forever barred and precluded" from auditing the family, its trusts, and its companies for any matter predating the agreement [Occurred]. The announced $1.776B "Anti-Weaponization Fund," drawn from the Treasury Judgment Fund, was subsequently abandoned amid multiple suits to block it [Contested]. The fund was the announcement; the audit bar was the capture, and it was taken by closing the courthouse door from the inside, after the door could no longer reach it. This is S3 run into S1 at its purest: the cheapest possible fix — letting a judge rule — refused at the highest possible cost, to protect a single realized beneficiary in perpetuity. The instrument scores the part that holds, not the part that was announced. Each of these is scored on the resource dimensions held outside the composite index, exactly as Part V requires, so that the gravest and most rhetorically charged measures do not silently inflate the headline numbers. They inform the reading. They do not drive the score. ## Projected harm, kept to what the models bear Some of the consequences here are not yet events; they are forecasts, and they are tagged as forecasts every time. I have refused throughout to convert projections into body counts, because a fabricated precision would discredit the real evidence faster than any critic could, and the discipline of Part III applies hardest precisely where the stakes are highest. The Cuba oil blockade, running from February 2026 [Occurred], is the case where projected harm is most acute and most disciplined. The structural facts are tagged occurred: a tanker blockade in place; on the order of eleven thousand children with surgeries suspended; international monitors warning that children are dying [Occurred]. The mortality figure — infant mortality reported by UN monitors as roughly doubled, toward 9.9 per thousand — is tagged projected and contested, and it is scored on the blockade's contribution as a major proximate cause, not as the sole cause, because a humanitarian collapse has many inputs and honesty requires apportioning rather than assigning the whole [Projected/Contested]. The OBBBA coverage and mortality figures are handled the same way: eleven to fifteen million in projected coverage loss (CBO estimate, 2034) is a CBO-class forecast, not a tally of the already-uninsured, and the mortality that tends to follow large coverage losses is given as a modeled range with its method attached, never as a number of named dead [Projected]. Call it restraint rather than squeamishness; it marks the difference between evidence and atrocity propaganda, and the instrument is worthless the moment it crosses that line. The projected harms are grave enough stated honestly. Stated dishonestly, they would forfeit the standing that makes them worth stating at all. ## The doctrinal substrate The architectures do not float free of stated strategy, and the final move of the application is to set them against the doctrines their authors published — which is reporting, not inference, because a quoted strategy is not an imputed motive. The nine-hundred-page implementation blueprint [Project 2025, Heritage Foundation 2023] [Occurred] is the densest internal-coherence anchor in the instrument; it is the document whose existence drives the S5 readings, because an operation that publishes a coordinated plan while its surrogates deny knowing the plan exists is the information asymmetry S5 was built to detect. The Two-Santa logic and "starve the beast," treated in Part V, supply the published rationale under which a deficit built deliberately becomes the argument for dismantling the programs on the other side of the ledger [Occurred - documented in conservative strategy literature since 1970s]. These doctrines do not prove that any specific architecture was run to their design. They dispose of the one objection that would otherwise sink the whole reading — that nobody would actually pursue such a thing on purpose — because the general strategy was pursued on purpose, in public, by people who signed their names to it. So the structure is measured, and the doctrine is quoted, and the two are kept apart on the page exactly as they are kept apart in the argument. The instrument detects a pattern of dysfunction that exhibits directional asymmetry: selective, resistant to its own repair, arriving across institutions at once. The public record establishes that designs of that general kind have an avowed constituency. The single question this leaves standing is the one the book has carried from its first page — what accounts for patterns this consistently one-directional — and it is a question the application lays fully open and pointedly declines to answer. ## What has been established The instrument has been built, controlled, and applied. It returns low scores on prior reforms, including one larger than the cases it scores high. It differentiates among the current architectures rather than condemning them uniformly. It scored the central fiscal architecture at 16 before convicting it at 85, which is the behavior of a measurement and not a verdict. It traces resource flows through documented chains while holding the gravest measures outside the headline number. It keeps every projection tagged as a projection and every forecast inside the range its method can bear. And it sets the whole pattern against doctrines that were published rather than imputed. That is the work. Seven architectures, named and counted. The thing they amount to — whether patterns this directional, this selective, this resistant to repair, this synchronized across institutions, is finally to be blamed and accounted to anyone — is not on this page, and was never going to be. The instrument names and counts. The reader, holding the count, is the one who decides what it weighs. That division of labor is the whole posture of the book, and it is kept to the end: the measurement is laid down complete, and the verdict is left, deliberately, in the only hands that can give it any weight — yours. --- ## The two tests the instrument has not yet faced Everything to this point shares one feature a fair reader should hold against it: every architecture the instrument scored high is one the author already opposed, and every architecture it scored low is temporally and politically remote from him. The negative controls of Part VI prove the instrument *can* acquit — but they acquit the distant dead. They do not prove it can convict a friend or acquit an enemy, and those are the only two proofs that separate a measurement from a mirror. A ruler its operator never loses to is indistinguishable from the operator's hand. So the instrument owes two tests it has not yet been made to face. They are stated here with their predicted outcomes, on purpose, so the predictions stand on the record *before* the scoring — and so that a failure to run them, or a refusal to publish what they return, counts against the framework exactly as it should. **The first test: convict a friend.** Point the five signatures — and S5 above all, internal coordination flanked by external incoherence — at an architecture the author is politically sympathetic to. Two candidates are clean: the pandemic-era public-health information apparatus of 2020–2022, and the federal–platform content-moderation complex of the same window. This is not an assertion that either is a directional-asymmetry structure. It is a narrower, falsifiable claim: the instrument *as specified* will plausibly score at least one of them into the High band, on S5 in particular — coordinated internal messaging beside shifting, self-contradictory external guidance; cheap corrections, a data release or a disclosure, resisted as hard as costly ones. If it does, and the score is published against the author's own sympathies, the framework has done the one thing no partisan instrument can. If the author reaches for reasons to score these low that he would never accept from a critic scoring the seed architectures low, the framework has failed this test, and the reader should say so plainly. **The second test: acquit an enemy.** Find one action of the opposed administration that, measured honestly, is ordinary symmetric incompetence, and score it Genuine or Mixed. The present distribution — seven live architectures, none below 70 — is exactly what a ceiling-slamming instrument produces, and the remote negative controls cannot answer the charge, because the bias risk lives *inside* the live domain; only a low score inside it answers anything. A candidate worth testing honestly is the tariff regime: broad, self-harming, falling on the opposed administration's own constituencies as heavily as on anyone's, with no clean realized-beneficiary divergence to drive S1. That profile predicts a Mixed score, possibly lower. If it scores high regardless, the instrument is slamming the ceiling, and the author is obliged to report that it failed to discriminate where discrimination was available. Both predictions are recorded as predictions — hypotheses, not findings — and neither test has been run at the time of this printing. That is the honest state of the instrument. The disagreement between the author and his own instrument, if it comes, is not an embarrassment to be managed; it is the proof the apparatus has been missing. Until the book contradicts its author somewhere a reader can see, the reader is right to suspect it has only ever agreed with him — and a measurement that only ever agrees with its maker has measured, among other things, him. --- # APPENDIX # Obfuscratic Asymmetry v0.1 — Distribution Package **ThinkWell Labs Metrology · B. Greenway · obfuscratic_sabbotarchy_final.zip** This package contains the **Obfuscratic Asymmetry — Deception Arcane** signature instrument and its accompanying methodology documentation. The instrument is a measurement apparatus for directional dysfunction regimes, scoring government-services-redirection architectures on five empirically-grounded signature dimensions to produce a composite Obfuscratic Asymmetry Index. The instrument is a *measurement apparatus*, not an argument. Its purpose is to allow observations of directional incompetence to be subjected to the same evidentiary discipline as any other empirical claim, and to make the cross-architecture signature pattern legible by applying the same measurement primitives uniformly across multiple targets. ## The framework name The apparatus names three things and keeps them on three rungs: - **Directional asymmetry** is the measured property: a structure whose outputs diverge non-randomly from stated inputs, favoring unstated outcomes. A *what*, observable and scorable, silent on why. - **Obfuscratic** (from *obfuscation* + *-cratic*) is the operational modality: a regime sustaining itself through systemic confusion and incoherence. An adjective — *how* it operates, not *what type* it is. - **Sabbotarchy** is the regime-form: rule by sabotage. The verdict-laden noun, reserved for what a reader may *conclude is present* when the property concentrates across all five signatures and operates obfuscratically. The instrument never asserts it. The measurement travels under the compound naming only what it measures — **Obfuscratic Asymmetry**, modality joined to property. *Sabbotarchy* waits above it for the reader. A transparently asymmetric regime would be recognized and resisted; an obfuscratic one with no underlying asymmetry would have nothing to hide; the regime-name moves only when a reader decides it does. The coinages were formed by the author through linguistic-autopsy procedure — stripping phonemic material from base words and recombining with classical morphological suffixes to expose previously-unnamed analytical predicates. ## Package contents ### Primary artifact **`sabbotarchy_deception_arcane.html`** — the signature instrument itself (deployed live at toroid.fyi/sabbot). Self-contained HTML file with embedded SVG visualizations and complete dataset. Opens in any modern browser without external dependencies. Total: ~2,000 lines, ~150KB. The instrument scores **seven** government-services-redirection architectures (DOGE federal workforce reduction, OBBBA / Medicaid administration, ICE/DHS detention apparatus, IRS dismantling, NIH research-funding compression, VA workforce restructuring, Regulatory/civil-rights rollback) on the five signature dimensions. Each score is anchored to documented evidence. The instrument computes the composite Obfuscratic Asymmetry Index per architecture and ranks all seven in a comparison table. ### Methodology documentation **`sabbotarchy_methodology_paper.md`** — formal methodology paper documenting the framework's analytical foundations, the bipartite naming structure, the five signature dimensions, the composite index, the hybrid scoring methodology, initial findings across the seven seed architectures, methodological limitations, integration with the Polymorphic Catalysis Toroid, and future directions. ~32KB. **`sabbotarchy_scoring_rubric.md`** — operational scoring rubric specifying the per-dimension scoring criteria, evidence requirements, scoring bands, and a worked example (IRS dismantling architecture). Provides the methodological precision required for analyst-extended scoring of additional architectures using the same framework. ~21KB. **`sabbotarchy_evidence_anchors.md`** — complete citation list for all evidence anchors referenced in the instrument, with URLs, access dates, and archival links where available. ~15KB. **Uncertainty and sensitivity analyses** — the weight-sensitivity results, error bands, and epistemics inventory live in the instrument's footer; a standalone reliability document will exist when the replication study runs. (An earlier edition listed an `asymmetry_uncertainty_analysis.md` containing inter-rater results; no such file or test exists, and the entry is corrected.) ## The five signature dimensions The framework rests on five measurable signature dimensions that distinguish directional dysfunction from genuine incompetence: **S1 — Outcome-Distribution Asymmetry** (weight 0.25). Divergence between the stated/policy-rationale beneficiary and the realized/documented beneficiary. Genuine policy produces symmetric distributions; directional incompetence produces distinct concentration toward unstated outcomes. **S2 — Domain-Selective Competence** (weight 0.20). Differential between public-facing function delivery and extractive function delivery within the same operational architecture. The signature is *selectivity* — surgical precision in operations of extraction or enforcement, sitting beside catastrophic failure in operations of public service, with the same personnel and resource pool available. **S3 — Anti-Remediation Response** (weight 0.15). Slope of resistance-intensity as a function of remediation cost. Genuine incompetence shows positive slope (expensive fixes resisted more than cheap ones — budget logic). Directional asymmetry shows flat-high or inverted slope (effectively-free fixes resisted as strongly as expensive ones — structural protection). **S4 — Personnel Selection Asymmetry** (weight 0.10). Three-axis pattern: technical-competence retention rate × ideological-alignment advancement rate × dissent-tolerance ratio. Asymmetric architectures concurrently produce low technical retention, high alignment advancement, and near-zero dissent tolerance. **S5 — Information-Architecture Asymmetry** (weight 0.30, the cleanest operational definition of obfuscratic operation). Ratio between internal information coherence (within the regime) and external information coherence (the public information ecosystem the regime produces). Obfuscratic architectures operate with extreme internal coherence flanked by systemic external incoherence. This dimension is weighted highest because the asymmetric internal-external coherence signature has the fewest alternative explanations available. ## The composite Obfuscratic Asymmetry Index Computed as the weighted aggregate: > Index = 0.25 × S1 + 0.20 × S2 + 0.15 × S3 + 0.10 × S4 + 0.30 × S5 Four bands: - **0–30 Genuine pattern** — consistent with genuine competence or genuine incompetence - **30–70 Mixed signature** — multiple dimensions present but not concentrated; further analysis required - **70–85 High-Asymmetry pattern** — directional signature pattern satisfied - **85–100 Extreme-Asymmetry pattern** — asymmetry concentration across all five dimensions **Uncertainty:** No score carries a confidence interval; single-analyst judgments do not have a sampling distribution to compute one from. The available variance data is computed sensitivity, not sampled: weight perturbations move composites by ≤~1 point, and ±5-per-dimension judgment uncertainty gives ±2.5–5 on the composite. No inter-rater data exists yet. Read all figures as band-level, not point-level. ## Initial findings The seven seed architectures produce a distribution clustered in the high and extreme bands: | Architecture | Composite Index | Band | |--------------|----------------|------| | DOGE / federal workforce reduction | 89 | Extreme-Asymmetry | | IRS dismantling | 91 | Extreme-Asymmetry | | OBBBA / Medicaid administration | 85 | Extreme-Asymmetry | | ICE/DHS detention apparatus | 85 | Extreme-Asymmetry | | Regulatory/civil-rights rollback | 88 | Extreme-Asymmetry | | NIH research-funding compression | 80 | High-Asymmetry | | VA workforce restructuring | 76 | High-Asymmetry | The clustering itself is a targeted validation. Because the domain was selected for predicted high signal, this clustering confirms the instrument's sensitivity to the signature it was built to detect, rather than claiming to prove systemic prevalence across all governance. The cross-architecture pattern matters analytically as much as any individual score. ## Government-services-redirection as the focused domain The seed architectures are drawn from the government-services-redirection domain by methodological design. This is the domain where outcome-distribution asymmetry is most cleanly measurable (the redirection of funds from public-serving functions toward extraction, surveillance, and procurement is the cleanest available proof signal), where signature dimensions are most concurrently observable, and where the diagnostic case is structurally strongest. Future versions of the instrument will extend within the government-services domain (SSA, Department of Education, EPA, USDA, HUD, SEC architectures all warrant scoring) and beyond it (corporate, foreign-state, civil-society architectures). The framework is regime-character-agnostic with respect to political affiliation; it applies wherever the signature pattern emerges. ## Companion to the Polymorphic Catalysis Toroid The Obfuscratic Asymmetry — Deception Arcane is the *measurement layer* of a two-instrument diagnostic stack. The companion *relational layer* is the **Polymorphic Catalysis Toroid** (distributed separately as `torusfilesfinal.zip`), which documents the operating-coalition architecture as a Three.js toroidal visualization with 349 documented operational cells, 130 actor-actor relationship cells, 50 wormhole backchannel cells, 17 polarity-tension cells across 136 actors and 118 targets. The integration is structural. Cell references in the toroid anchor measurement rows in the Deception Arcane. The toroid is the evidence layer; the Deception Arcane is the measurement layer. An analyst working with both instruments can: identify an operation of interest in the toroid; examine its cell-content for operational documentation; consult the Deception Arcane to see the signature score of the broader architecture; and either confirm or revise the score based on cell-specific evidence. The instruments are designed to support this iterative workflow. Each instrument is also separable. The toroid functions as a standalone relational instrument for analysts who want to understand operational architecture without invoking the directional-asymmetry diagnostic. The Deception Arcane functions as a standalone measurement instrument for analysts who want to apply the signature framework without engaging with the full toroid schema. ## Hybrid scoring methodology The instrument operates on a hybrid scoring model: - The seven seed architectures are scored by the author with documented evidence anchors and full audit trail - The infrastructure (HTML data model, panel rendering, comparison table) supports extension to additional architectures via the same evidence-anchored methodology specified in the scoring rubric - Cross-analyst replication is an open research program — the framework's external validity depends on whether independent analysts produce convergent rank-orderings of architectures The hybrid model is methodologically chosen. Pure author-scoring would limit external validity; pure methodology-scaffolding would limit demonstrative power. The hybrid produces both: a demonstrative initial dataset and infrastructure for the framework's continued development. **Inter-rater reliability:** not yet measured. Cross-analyst replication is an open research program; the pre-committed convergence standard is ±10 points per dimension. An earlier draft reported a preliminary rater test here; no record of such a test exists, and the claim is withdrawn. ## Operating philosophy The framework occupies a specific analytical position: **It is a measurement procedure, not an accusation.** It does not claim that any specific architecture is sabotaged; it provides the procedure by which an architecture can be measured on the signature dimensions and scored. An architecture scoring below 30 is consistent with genuine incompetence or genuine competence, and the framework provides no basis for further diagnostic claim. **It distinguishes measurement from verdict.** Naming Obfuscratic Asymmetry doesn't prove intent; signature analysis maps the shape. The framework treats obfuscratic operation as a measurement problem rather than a rhetorical accusation. **It is self-critical by design.** Each scoring decision is anchored to evidence; each evidence anchor is auditable; each score can be revised when new evidence emerges or when the analyst's interpretation is challenged. The framework's authority is derivative of the documentation behind it, not constitutive of any analyst's claim. **It is meant to outlive any specific analyst's use of it.** The methodology is documented for replication; the scoring rubric is operational; the evidence-anchoring discipline is reproducible. Other analysts can adopt the framework, score additional architectures, and contest the seed-architecture scores. The framework gains validity through use, not through assertion. **It acknowledges uncertainty.** No score carries a fabricated confidence interval; uncertainty is stated honestly at band level, with the computed sensitivity bands reported and the absence of replication stated plainly. Every projection is tagged. Every correlation is distinguished from causation. Every contested claim is marked. The framework's credibility rests on transparency about what it does not know as much as what it does. ## Citation If referencing this instrument or framework in scholarly or analytical work: > Greenway, B. (2026). *The Obfuscratic Asymmetry Signature: A Measurement Framework for Directional-Dysfunction Regimes.* ThinkWell Labs Metrology. Companion methodology paper and signature instrument (Obfuscratic Asymmetry — Deception Arcane). The originating insight — the bipartite construction *Obfuscratic Asymmetry* as a linguistic-autopsy compound naming directional dysfunction operating through deliberate obfuscation — is attributed to B. Greenway. The three terms — *directional asymmetry* (the measured property), *Obfuscratic* (the operational modality), and *Sabbotarchy* (the regime-form held in reserve for the reader) — were coined or reframed by the author. The Multimalocracy 42-mode discriminative-vocabulary taxonomy within which this framework operates is also attributed to B. Greenway and is documented in the companion Polymorphic Catalysis Toroid package (`torusfilesfinal.zip` → `multimalocracy_revision_proposal.md`). --- *Prepared June 27, 2026 (review completed June 28 local time, east of the International Date Line — both dates honest, noted so a Western-hemisphere reader does not misread the timestamp as future-dated).* --- **Sources:** - [Trump Administration Plans To End The (Economicsecurityproject)](https://economicsecurityproject.org/news/trump-administration-plans-to-end-the-irs-direct-file-program-for-free-tax-filing-ap-sources-say/) - [7t7q0 (Taxnotes)](https://www.taxnotes.com/featured-news/irs-shutters-direct-file-citing-cost-and-low-uptake/2025/11/05/7t7q0) - [Irs Chief Says Agency Plans To (Cnbc)](https://www.cnbc.com/2025/07/30/irs-chief-says-agency-plans-to-end-free-direct-file-program.html) - [Trump Plan To End Free Direct (Cbpp)](https://www.cbpp.org/blog/trump-plan-to-end-free-direct-file-program-and-rely-on-for-profit-tax-preparers-is-a-mistake) - [Taxes Irs Direct File (Nytimes)](https://www.nytimes.com/2025/11/06/business/taxes-irs-direct-file.html) - [Irs Would Eliminate Direct File Under (Federalnewsnetwork)](https://federalnewsnetwork.com/congress/2025/05/irs-would-eliminate-direct-file-under-trump-backed-budget-reconciliation-bill/) - [Pushback Against Trump Plan To End (Tax)](https://tax.thomsonreuters.com/news/pushback-against-trump-plan-to-end-irs-direct-file/) - [Irs Ends Direct File Shifts Focus (Journalofaccountancy)](https://www.journalofaccountancy.com/news/2025/nov/irs-ends-direct-file-shifts-focus-to-free-file-upgrades-and-private-sector/) --- ## Changelog v5 → v5.1 (July 2026) 1. **Inter-rater claim removed (four sites).** v4 introduced, in a "(v3.1)" revision block, a claim that "two additional analysts on three architectures showed convergence within ±12 points." v5 deleted that block's twin (a fabricated ±5–8 confidence interval) as "fabricated precision" but kept this one. No protocol, raters, scores, or data file for such a test exist anywhere in the project record — the cited `asymmetry_uncertainty_analysis.md` was never written — and the author has no means of having run it. The claim is withdrawn as a drafting artifact. The replacement is the truth: replication has not been run, and the computed sensitivity analyses are the only variance data. 2. **Band-threshold derivation withdrawn** (§ the composite index): the "four-of-five / all-five" dimension-count justification fails arithmetic under unequal weights; thresholds are calibration conventions, retained on robustness grounds. 3. **Grace Commission comparison corrected**: OBBBA's pre-enactment 16 is above Grace (≈14), below NPR (≈19) and PMA (≈20). 4. **IRS rows updated** to the live instrument's corrected per-dimension scores (94·95·89·76·92 → 91). 5. **Distribution-package description corrected** to the real file names and instrument size.