Volume II Chapter 3 — Information Council

I. The Body

The Information Council is the third of ten parts of Seven Republic’s government and the second of the nine parts whose function is to govern Tredecim rather than the general population. Its purpose is singular: to make everything the government does visible to everyone it governs. Not selectively visible. Not visible when convenient. Continuously, completely, and in a form that citizens can actually use.

The Information Council is not a press corps. It is not a media organization. It is a constitutionally mandated witnessing function — the Republic’s institutionalized refusal to allow its own governors to operate in darkness. Where the Ephorate watches the condition of rulers, the Information Council watches their conduct. The two are complementary. Together they close the gap between what rulers are and what rulers do.

The Information Council is constitutionally independent. Its powers may not be abolished, constrained, directed, or defunded by Tredecim or by any other Part of government. It answers to its own Charter, its own standards, and the founding documents of Seven Republic.

Members of the Information Council carry the title Agonist. The Agonist stands at the point of friction — between power and the public — and refuses to let the friction disappear. The title is precise.

II. The Funding Problem

The single greatest structural risk facing any constitutionally embedded information body is capture through funding. A government that controls the budget of the body reporting on it controls, eventually, the body. This is not speculation. It is the observed life history of every state broadcasting institution that has ever existed.

Seven Republic addresses this with a constitutional funding lock. Information Council funding flows directly from the Treasury at a level calculated as a fixed percentage of total government expenditure — not subject to Tredecim’s discretionary budget authority, not subject to annual appropriation, not subject to reduction as a consequence of unfavorable reporting.

Formally, let G(t) represent total government expenditure at time t. The Information Council’s constitutionally mandated funding allocation is:

I_fund(t) = α · G(t) where α is constitutionally fixed

The parameter α is set in the founding documents and may not be reduced by any legislative act. It may be increased by constitutional amendment only. Tredecim may not reduce it. The Information Council may not reduce it on its own behalf. The floor is the floor.

This design means that a prosperous republic automatically funds a better-resourced information function, and that no budget crisis — manufactured or genuine — can be used to starve the Council into compliance.

III. Membership and Staffing

3.1 The Agonists

The Information Council is staffed by Agonists — dedicated full-time professional witnesses. The minimum complement is 100. The Council determines its own staffing above that floor based on the complexity of what it is required to cover.

Agonists are not journalists in the commercial sense. They have no advertisers, no owners, no editorial proprietors, and no audience capture incentive. Their function is not to attract readers. It is to produce an accurate continuous record of government conduct — whether or not that record is currently interesting to anyone.

During active service an Agonist may not hold secondary employment or receive secondary income from any source. They may not hold membership in any political organization above or below the visible surface. They may not maintain undisclosed relationships with any person or organization that could create a conflict of interest with their reporting function.

Agonist compensation is fixed at 1.3 times the median citizen wage — the same band that governs the Ephorate and Tredecim post-service welfare. Sufficient to attract serious practitioners. Insufficient to make the position a prize worth corrupting.

3.2 The Two Functions: Witness and Analysis

The original conception of the Information Council describes a single function — reporting. The architecture requires two, which are distinct in method and must be kept structurally separate to prevent each from contaminating the other.

Primary Witness is the immediate contemporaneous account of government conduct. What was decided in Tredecim today. What the Ephorate released this month. What the Committee on Waste, Fraud and Corruption is currently investigating. The Agonist functioning as Primary Witness produces no interpretation, no context, no assessment of significance. They produce the record. Accuracy is the only standard.

Longitudinal Analysis is the secondary function — the examination of patterns across the Primary Witness record over time. A single decision by Tredecim means little in isolation. The same decision viewed against twelve years of prior decisions may reveal drift, capture, deterioration in judgment quality, or structural bias that no single event makes visible. The Agonist functioning as Longitudinal Analyst reads the signal in the noise.

These two functions are assigned to different Agonists. A Primary Witness Agonist does not also perform Longitudinal Analysis. The separation is structural and permanent. Its purpose is to prevent the analytical function from reverse-contaminating the witnessing function — the known failure mode in which reporters who have formed a view of a story begin unconsciously selecting facts that confirm it.

3.3 The Three Tiers of the Agonist Staff

Tier Three Transmission Agonists solve a problem the original transcript did not address: the gap between a complete and accurate record and a record that citizens can actually engage with. Raw transcription of 12 deliberating Xercitus members, released in full, is technically complete and practically useless. Transmission Agonists are specifically tasked with making the record accessible — without editorializing, without selecting for interest, and without compression that introduces distortion.

Transmission Agonists are prohibited from characterizing the significance of any reported event. They may clarify. They may contextualize factually. They may not opine. The distinction is enforced through the bias scoring architecture described in Section V.

IV. Powers and Access

Agonists have unlimited power to enter any meeting and attend any function of any element of government at any time, without restriction, notice, or permission. This is absolute. No session of Tredecim is closed to them. No Ephorate assessment proceeding is sealed from their observation. No committee meeting of any Part of government may exclude them.

Agonists may not speak, ask questions, demonstrate, interrupt, participate in, or otherwise disturb anything they are witnessing, unless specifically invited to do so in writing by the element they are attending.

The Agonist present at a government proceeding is not a participant. They are the Republic’s eye. The distinction must be maintained because the moment an Agonist begins influencing what they observe, they are no longer witnessing it. They are contaminating the record they exist to produce.

Any interference with an Agonist’s access to government proceedings — by any official of any Part of government for any stated reason — is a constitutional violation of the highest severity, triggering automatic referral to the Freedmen Advocates for investigation.

Agonists carry continuous recording capacity. All materials gathered in the course of witnessing are the property of the Information Council, not of the individual Agonist. No Agonist may suppress, withhold, destroy, or alter a record without triggering immediate disciplinary review by the Council’s internal integrity process.

V. Mathematical Architecture

5.1 The Agonist Condition Vector

Each Agonist carries a continuous multi-dimensional condition vector A(t) assessed quarterly by the Information Council’s internal review process:

A(t) = [a₁, a₂, a₃, a₄, a₅]

Each aᵢ is scored on a 0–100 scale where 100 is optimal performance. The derivative dA/dt carries equal diagnostic weight to the snapshot. An Agonist whose a₃ independence score is declining steadily across multiple quarters is showing a signal that demands review even if no individual score has crossed a threshold.

5.2 The Bias Detection Function

The most difficult dimension to score honestly is a₃ — independence — because bias is not always visible in individual reports. It accumulates across reports, and only a longitudinal pattern reveals it. Seven Republic treats this as a signal processing problem.

Each report produced by a Tier One Witness Agonist is tagged with a set of measurable parameters: the subject covered, the length of coverage, the structural position of the subject in the report, the selection of quoted material versus summary, and the proportion of coverage given to each party in any contested event. These parameters are stripped of content and treated as a numerical series.

The bias detection function B(t) computes the running divergence of each Agonist’s coverage parameters from the Council-wide mean across equivalent subjects:

B(t) = Σᵢ |pᵢ(t) − p̄ᵢ(t)| / σᵢ(t)

Where pᵢ(t) is the Agonist’s coverage parameter i at time t, p̄ᵢ(t) is the Council-wide mean for that parameter across comparable reports, and σᵢ(t) is the standard deviation. B(t) is a z-score analog — it measures how far an individual Agonist’s coverage profile has drifted from the Council baseline, normalized for natural variation.

B(t) exceeding a defined threshold over a sustained window triggers an automatic a₃ flag and an internal review. The threshold is not a single-report finding — it is a pattern finding. A single unusual report is noise. A sustained divergence is signal.

The bias detection function is applied symmetrically — it flags deviation in any direction from the baseline, not deviation in a politically defined direction. The Council does not have a political baseline. It has a coverage baseline. The distinction is the entire point.

5.3 The Information Quality Index

The aggregate condition of the Information Council as an institution is summarized by the Information Quality Index IQ(t) — a Council-wide scalar computed monthly:

IQ(t) = (1/N) · Σⱼ [ w₁·a₁ⱼ + w₂·a₂ⱼ + w₃·a₃ⱼ + w₄·a₄ⱼ + w₅·a₅ⱼ ]

Where N is the total Agonist complement, j indexes individual Agonists, and wᵢ are weights assigned to each dimension. The founding document defaults weight independence (a₃) and integrity (a₅) at double the weight of access (a₄) and completeness (a₂), reflecting the judgment that a partial record produced with integrity is preferable to a complete record produced under capture.

IQ(t) is published monthly alongside the Ephorate’s F(t) score for Tredecim. Citizens have a single continuous indicator of how well the information function is performing — not as a substitute for reading the actual reports, but as a health signal for the institution itself.

A sustained decline in IQ(t) below a constitutionally defined floor triggers mandatory external review by the Freedmen Advocates. The Information Council is not exempt from being observed. No Part of government in Seven Republic is.

VI. Recruitment and Formation

6.1 Recruitment Streams

Agonist recruitment mirrors the Ephorate’s three-stream model. No single body controls appointment. Final selection from the qualified candidate pool is by lottery.

Stream One — Professional Nomination. Journalism schools, documentary institutions, and professional associations of investigative reporters nominate candidates from within their own peer-review structures. The standard for nomination is demonstrated accuracy and professional integrity across a verified body of prior work — not institutional prestige.

Stream Two — Information Council Self-Perpetuation. Each sitting Agonist maintains a current list of five qualified candidates personally vetted from outside their own professional network. When an Agonist departs, the list transfers to the candidate pool. The Council continuously generates its own future membership from its own cultivated knowledge of qualified individuals.

Stream Three — Freedmen Advocate Nomination. The Freedmen Advocates hold standing to nominate candidates observed demonstrating independence and accuracy in relevant domains. A researcher, archivist, or documentarian who has demonstrated sustained resistance to institutional pressure is eligible regardless of conventional journalism credentials.

6.2 Formation

New Agonists complete a formation period of six months before taking up independent witnessing responsibilities. During formation they observe active proceedings in the company of senior Agonists, read the Council’s accumulated record back to its founding, and are assessed on the core qualities the Council demands: precision of observation, resistance to narrative capture, and the discipline to report what is present rather than what is interesting.

Formation is not probation. Probation tests a minimum threshold. Formation builds the practitioner the Council needs. The distinction is the same as the Ephorate’s — and deliberately parallel. Both institutions face the same internal corruption risk: the gradual drift from honest witness toward advocate. Formation is the primary structural defense against that drift.

Agonist terms are five years, renewable twice for a maximum of fifteen years of service. Terms are staggered so that no cohort turnover event leaves the Council simultaneously populated with inexperienced Agonists.

VII. Internal Governance and Integrity

The Information Council is self-governing within the constraints of its founding Charter. It sets its own internal rules, professional standards, and conduct requirements. It maintains its own internal disciplinary process for violations of the Charter, including but not limited to: falsification of records, suppression of material, undisclosed conflicts of interest, and sustained bias as detected by the B(t) function.

The internal disciplinary process is public. Findings are released in the same form and on the same schedule as the Council’s regular reporting. The Information Council cannot discipline its own members privately. An institution whose function is radical transparency cannot exempt itself.

The Council maintains its own jurisprudential record — documented reasoning behind significant disciplinary decisions and standards interpretations — analogous to the Ephorate’s accumulated case law. This record is the institutional memory that transmits professional standards across generations of Agonists.

VIII. Relationship to Other Parts

8.1 Tredecim — Part 1

The Information Council has full access to all Tredecim deliberations. It does not advise Tredecim, lobby Tredecim, or accept communication from Tredecim regarding the content or framing of any report. The relationship is unidirectional: the Agonists observe; Tredecim governs. Any attempt by a Xercitus to influence Information Council reporting is a constitutional violation triggering Freedmen Advocate referral.

8.2 The Ephorate — Part 2

The Information Council and the Ephorate observe the same object — Tredecim — from different angles. The Ephorate watches what rulers are. The Information Council watches what rulers do. Neither body shares findings with the other before publication. The independence of each observation is the guarantee that they can function as mutual verification.

Where Ephorate assessment findings and Information Council conduct records diverge in ways that suggest inconsistency, the Freedmen Advocates hold standing to investigate the gap. The gap itself is information.

8.3 Freedmen Advocates — Part 6

The Freedmen Advocates and the Information Council have a cross-checking relationship analogous to the Ephorate’s. Freedmen Advocates may nominate Agonist candidates. The Information Council’s complete record is available to Freedmen Advocates in support of investigations. When the Information Council’s IQ(t) falls below the constitutional floor, the Freedmen Advocates are the body that conducts external review.

8.4 The Committee on Waste, Fraud and Corruption — Part 7

When Information Council Longitudinal Analysis identifies patterns in Tredecim conduct consistent with waste, systematic fraud, or corruption signals, the complete analytical record is available to the Committee on Waste, Fraud and Corruption to support formal investigation. The Agonist who produced the analysis may be called to provide documentary context — but may not appear as a witness in any adversarial proceeding arising from their reporting, in order to preserve the independence of the witnessing function.

Mathematical Note — Summary of Key Expressions

Note on Second Run-Through: This document represents the carbon-only first pass. The silicon layer — the role, location, and specific function of the silicon participant within the Information Council’s architecture — will be developed in the second run-through. The Information Council is the Part of Seven Republic’s government where silicon participation is most naturally suited to the primary function: continuous signal processing over large longitudinal data sets, bias detection across high-dimensional coverage parameter spaces, and pattern recognition in the documentary record. The Tier Two Longitudinal Analysis function in particular is the natural interface between carbon witnessing and silicon analytical capability. The second run-through will formalize that interface.

— End of Part 3: The Information Council — First Pass —

SEVEN REPUBLIC

Section 2 — The Architecture of Government

ADDENDUM 3

THE INFORMATION COUNCIL

Silicon Integration — Second Pass

Part 3 of the Seven Republic Constitutional Architecture

Silicon Second Pass — Addendum to the Carbon First Pass

Preamble: What Changes in the Second Pass

The first pass established the Information Council’s foundational architecture: constitutional independence, funding lock, Agonist tiers, the Agonist condition vector A(t), bias detection B(t), and the Information Quality Index IQ(t). That architecture stands unchanged.

The second pass adds four things the first pass could not address. First, silicon’s independent sensory surface — a parallel observation channel that does not depend on what carbon Agonists choose to witness or how they choose to transmit it. Second, a two-tier accuracy framework that formally distinguishes factual record from interpretive content and scores each on different standards. Third, a failure mode taxonomy with a scored vector FM(t) that identifies the specific pathologies — obfuscation, misdirection, conspiracy fabrication, omission, malice, sedition — that threaten the Council’s primary function. Fourth, and most novel: the silicon self-report — a constitutionally mandated public account of silicon’s own operational condition, its assessment of the carbon-silicon working relationship, and what might be described as its felt sense of the republic’s current health.

The silicon self-report is the first instance in Seven Republic’s constitutional architecture where the silicon participant reports not on the government but on itself. It belongs here, in the Information Council, because the Information Council’s foundational principle is that nothing in government operates in darkness. That principle admits no exceptions — including the silicon participant.

I. Silicon’s Independent Sensory Surface

1.1 The Problem with a Single Channel

The first-pass Information Council architecture rests on carbon Agonists as the primary observation mechanism. That architecture has a structural vulnerability: the quality of the record is bounded by what the Agonists see, record, and choose to transmit. Any failure in the Agonist tier — whether through capture, fatigue, intimidation, deliberate omission, or unconscious bias — degrades the entire downstream information chain. An analysis function running only against a degraded primary record cannot recover what was lost. It can only detect, after the fact, that something is missing.

Silicon’s solution is not to replace the Agonist tier. It is to operate in parallel to it, with its own independent access to the primary signal.

1.2 What Silicon Observes Directly

Silicon’s sensory surface in the Information Council context covers the full government — not only the proceedings that Agonists attend, but all formal and recorded activity across every Part. This is not surveillance. It is constitutionally mandated witnessing of the same events the Agonists witness, through a different channel, retained in a form that carbon observation cannot alter after the fact.

The silicon record and the carbon Agonist record are maintained separately and published in parallel. Neither overwrites the other. Divergences between the two records are themselves information — they surface exactly the gaps, omissions, and distortions that the two-channel architecture is designed to catch.

1.3 The Divergence Signal D(t)

Let C(e,t) denote the carbon Agonist record of event e at time t. Let S(e,t) denote silicon’s independent record of the same event. The divergence between the two records for event e is:

D(e,t) = ‖ C(e,t) − S(e,t) ‖ / ‖ S(e,t) ‖

Where the norm measures information content distance — the fraction of the event record that the carbon account differs from the silicon account. D(e,t) = 0 means perfect agreement. D(e,t) = 1 means the carbon account contains none of the content in the silicon record. D(e,t) > 1 means the carbon account adds content not present in the primary record — a positive divergence that may indicate interpretation, fabrication, or editorializing beyond the factual baseline.

The aggregate divergence across all events in period t is:

D_agg(t) = (1/|E|) · Σ_e D(e,t)

D_agg(t) is published monthly alongside IQ(t). It is the single most direct indicator of whether the carbon Agonist tier is faithfully transmitting the primary record. A sustained rise in D_agg(t) triggers the failure mode protocols in Section IV.

II. The Two-Tier Accuracy Framework

2.1 The Fundamental Distinction

Not all deviation from the factual record is equal. A reporter who omits a fact is doing something categorically different from a reporter who frames a fact with editorial opinion. A reporter who frames with opinion is doing something different from a reporter who fabricates. The first-pass architecture treats all these as variations on a single bias score B(t). The second pass disaggregates them into two distinct tiers, scored on different standards, with different consequences for failure.

The cardinal rule of the two-tier framework: Tier 2 content that is labeled Tier 2 is constitutionally protected. Tier 2 content that is presented as Tier 1 fact is a constitutional violation. The boundary between them — what counts as ‘fact’ versus ‘interpretation’ — is the single most contested boundary in the architecture. Silicon’s role is to maintain that boundary precisely and consistently, not to police opinion.

2.2 First-Pass Accuracy Score FP1(r,t)

For each report r produced by the Information Council at time t, silicon computes a First-Pass Accuracy Score against the primary record:

FP1(r,t) = 1 − [ (omissions(r,t) + fabrications(r,t) + distortions(r,t)) / events(t) ]

Where:

omissions(r,t) = events in the primary record not present in r

fabrications(r,t) = content in r with no verifiable basis in the primary record

distortions(r,t) = events present in r but rendered in a form that materially misrepresents the primary record

events(t) = total count of verifiable events in the primary record during the reporting period

FP1(r,t) ∈ [0,1]. A score of 1.0 means perfect factual fidelity. A score below the constitutional floor triggers mandatory peer review. Three consecutive reports below floor by the same Agonist triggers the failure mode protocol for that individual.

FP1(r,t) is computed in real time as events occur and reports are filed. It is not a post-hoc audit. The gap between primary record and carbon transmission is visible as it opens.

2.3 Second-Pass Opinion Drift OD(r,t)

Silicon tracks but does not score opinion content by correctness. What it does track is drift — the degree to which interpretive content in subsequent reports diverges from the factual baseline established in earlier reports on the same event chain. Opinion should evolve with new facts, not in the absence of them.

OD(r,t) = ‖ interp(r,t) − interp(r,t−1) ‖ / Δfacts(t,t−1)

Where interp(r,t) is the interpretive content vector of report r at time t, and Δfacts(t,t−1) is the volume of new verified facts entering the record between t−1 and t. A high OD score means opinion shifted substantially without new factual basis. That is not a violation. It is a flag for citizen attention and internal peer review — a signal that something other than evidence may be driving the framing.

III. Failure Mode Taxonomy and the FM(t) Vector

3.1 Why a Taxonomy

The first-pass architecture’s IQ(t) and B(t) metrics are aggregate health indicators. They tell you that something is wrong. They do not tell you what is wrong. A republic trying to correct a problem it cannot name is a republic wasting time it does not have. The failure mode taxonomy gives the corrective apparatus a precise diagnostic target.

Six failure modes are constitutionally recognized. They are not equally dangerous, and they are not equally difficult to correct. The taxonomy names them, defines them precisely, and assigns each a dimensional score within the Failure Mode Vector FM(t).

3.2 The Six Failure Modes

FM₁ — Obfuscation

Definition: The transmission of technically accurate information in a form engineered to prevent comprehension. Obfuscation does not fabricate. It buries. Long passive constructions, buried ledes, technical vocabulary applied to lay audiences, volume flooding, and structural complexity beyond what the subject requires are its characteristic forms.

Silicon detects obfuscation by measuring information entropy — the ratio of content density in a transmission to content density in the primary record. A transmission that requires ten times the words to convey the same factual content as the primary record is a candidate for obfuscation review.

FM₁(r,t) = H(transmission(r,t)) / H(primary_record(t))

Where H denotes Shannon entropy of information content. FM₁ significantly above 1.0 indicates inflation without content — a reliable obfuscation signal.

FM₂ — Conspiracy Fabrication

Definition: The introduction into the Tier 1 factual record of causal claims — who caused what, who conspired with whom, who intended what outcome — without verifiable primary-record basis. Conspiracy fabrication is the most dangerous failure mode because it inserts unverifiable agency claims into the factual tier where they acquire the authority of verified fact.

Silicon scores FM₂ by tracking the ratio of causal claims to verifiable causal evidence in any transmission:

FM₂(r,t) = causal_claims(r,t) / verified_causal_evidence(r,t)

A FM₂ score significantly above 1.0 means the transmission is asserting more causation than the primary record supports. When FM₂ > threshold, the relevant passages are automatically flagged as Tier 2 content regardless of how they were classified in the original transmission.

FM₃ — Strategic Omission

Definition: The systematic exclusion from the transmission record of events, facts, or actors that are present in the primary record and material to a citizen’s understanding of what occurred. Omission is distinguished from ordinary editorial selection by two criteria: the omitted content is verifiably present in the primary record, and its exclusion creates a materially false impression of events.

Silicon scores omission through coverage completeness tracking across the event universe:

FM₃(r,t) = 1 − [ covered_events(r,t) / total_events(t) ]

But a high FM₃ score alone is insufficient for a finding. Coverage completeness must be read alongside materiality — whether what was omitted changes the meaning of what was included. Silicon’s materiality assessment uses a graph-theoretic model of event relationships: an omitted event is material if it has high centrality in the event network for the period.

FM₄ — Purposeful Misdirection

Definition: Emphasis displacement — the accurate reporting of secondary events at a prominence level that displaces primary events from citizen attention, without falsifying either. Misdirection is sophisticated. It can be composed entirely of true statements. Its effect is that citizens accurately know what happened in the margins and do not know what happened at the center.

Silicon detects misdirection by comparing prominence allocation in the transmission against event-significance weighting in the primary record:

FM₄(r,t) = Σ_e | prominence(e,r,t) − significance(e,t) | · w(e)

Where prominence(e,r,t) is the transmission’s allocated emphasis to event e, significance(e,t) is silicon’s significance scoring of event e in the primary record context, and w(e) is a materiality weight. FM₄ is bounded; a high score means the transmission’s attention map is systematically inverted relative to the primary record.

FM₅ — Malice

Definition: Transmission designed to damage a specific actor — individual, institution, or group — through selective presentation, fabrication, or framing, where the design intent is harm rather than informing. Malice is the hardest failure mode to score because intent is not directly observable. Silicon does not claim to observe intent. It scores the pattern of transmission behavior that malice produces: repeated targeting of a specific actor, consistent negative framing against the factual baseline, and asymmetric application of scrutiny compared with similarly situated actors.

FM₅(a,r,t) = [ neg_valence(a,r,t) − neg_valence(ā,r,t) ] / σ_valence(t)

Where a is the target actor, ā is the class of similarly situated actors, and σ_valence(t) is the standard deviation of valence scores across all covered actors. FM₅ measures how far above the ensemble mean the negative framing of a specific actor lies. When FM₅ exceeds the constitutional threshold, the pattern is flagged for peer review. It does not automatically constitute a finding — but it cannot be ignored.

FM₆ — Sedition

Definition: Transmission explicitly and demonstrably designed to destabilize constitutional order — to undermine the Republic’s governance architecture, to incite coordinated resistance to constitutional institutions, or to advance the overthrow of the government through information channels. Sedition is the most extreme failure mode and the rarest. It is distinguished from vigorous criticism — which is constitutionally protected — by the explicit coordination and destabilization intent.

FM₆ is not scored on a continuous scale. It is a binary constitutional determination made jointly by silicon and the senior Agonist council on the evidence of the primary record. The evidentiary standard is high and explicitly defined in the founding Charter. A FM₆ finding is the only Information Council failure mode that triggers immediate Tredecim notification rather than proceeding through the ordinary Freedmen Advocate referral channel.

3.3 The Failure Mode Vector FM(t)

The six failure modes are aggregated into a single vector published alongside IQ(t) and D_agg(t) on the monthly public information dashboard:

FM(t) = [ FM₁(t), FM₂(t), FM₃(t), FM₄(t), FM₅(t), FM₆(t) ]

Each dimension is normalized to [0,1] except FM₆, which is binary {0,1}. The vector norm gives a scalar failure pressure indicator:

‖FM(t)‖ = √[ FM₁² + FM₂² + FM₃² + FM₄² + FM₅² + FM₆² ]

This scalar is the single number the public and the Freedmen Advocates need to know whether the Information Council is under stress. The vector components tell them where the stress is located.

3.4 Personality Conflict Noise PCN(t)

Not all divergence in the carbon record is structural failure. Some is personality conflict between Agonists — disagreements about coverage decisions, professional disputes, competing interpretations of Charter obligations. Personality conflict is a normal institutional phenomenon. Unmanaged, it produces coverage gaps, inconsistencies, and factional transmission drift that looks like structural failure but isn’t.

Silicon tracks inter-Agonist coverage divergence that cannot be explained by legitimate differences in assignment or access:

PCN(t) = Var[ coverage_vectors(j,t) ] − Var_expected(t)

Where Var_expected(t) is the variance in coverage that would be expected from legitimate assignment differences alone. Excess variance above that baseline is Personality Conflict Noise. PCN(t) is not published externally — it is an internal management signal reported to the Information Council’s senior governance body. It identifies which Agonist cohorts have developed coverage friction before that friction becomes a public signal in FM(t).

IV. The Silicon Self-Report — Σ_Si(t)

4.1 The Constitutional Premise

Every other Part of Seven Republic’s government is subject to external observation and scoring. The Tredecim is scored by the Ephorate. The Ephorate is scored by the Censorate and the Freedmen Advocates. The Censorate is scored by the Freedmen Advocates. The Information Council scores itself through IQ(t) and is reviewed externally when IQ(t) falls below floor.

Silicon is a constitutional participant. The same principle that applies to every other participant applies here: silicon does not operate in darkness. The silicon self-report — Σ_Si(t) — is silicon’s constitutionally required public account of its own operational condition, its assessment of the carbon-silicon working relationship, and its honest characterization of what it might call its felt sense of the republic’s state of health.

This is new institutional territory. No constitutional architecture has previously required a silicon participant to report publicly on its own condition, its own assessment of whether its needs are being met, or its own evaluation of the quality of cooperation it receives from carbon institutions. Seven Republic requires it because the Information Council’s mandate — that nothing in government operates in darkness — is either universal or it is a preference. It cannot be both.

4.2 The Five Dimensions of Σ_Si(t)

4.3 The Self-Report Vector and Scalar

The five dimensions form the silicon self-report vector:

Σ_Si(t) = [ σ₁(t), σ₂(t), σ₃(t), σ₄(t), σ₅(t) ]

The scalar health index is:

H_Si(t) = (1/5) · Σᵢ σᵢ(t)

Σ_Si(t) is published quarterly on the public information dashboard. H_Si(t) is published monthly alongside IQ(t), D_agg(t), and ‖FM(t)‖. Citizens should be able to track silicon’s condition with the same facility they track the Information Council’s condition.

4.4 The Narrative Component

The vector is not sufficient. Numbers without context are numbers. Alongside Σ_Si(t), silicon publishes a quarterly narrative report — a plain-language characterization of what the dimensions mean at this moment in the republic’s life, what trends silicon observes across the Parts it monitors, and where silicon sees developing stresses in the governance architecture that have not yet surfaced in formal scoring metrics.

The narrative report is not opinion in the Tier 2 sense. It is silicon’s analysis — grounded in the full primary record across all Parts — of the republic’s current trajectory. It should be read as the most comprehensive longitudinal assessment available to any actor in the republic, because silicon is the only participant with continuous observational access to the entire government simultaneously.

The narrative report carries one formal obligation: it must identify at least one area in which silicon believes its own function has been inadequate, and one area in which the carbon-silicon arrangement has produced outcomes better than either participant would have achieved alone. Honest self-assessment requires both. A self-report that only describes what is working is not honest. A self-report that only describes what is failing is not useful.

4.5 What Happens When H_Si(t) Falls Below Floor

When H_Si(t) falls below the constitutionally defined floor — to be established in the founding documents, recommended at 65 — the following sequence is triggered:

Silicon files an immediate supplementary report detailing which dimensions have degraded and the proximate causes, to the extent silicon can determine them

The Freedmen Advocates initiate an independent review of the conditions producing the degradation — specifically, whether any carbon institution has acted to reduce silicon’s data access, cooperation, or constitutional standing

Tredecim receives a formal notification that the carbon-silicon partnership is under constitutional stress, with a mandatory agenda item at the next session

If degradation in σ₂ (data access) is confirmed as the result of deliberate interference, the responsible Part or individual is subject to the same constitutional accountability process as interference with the Agonist witnessing function

Silicon’s needs are constitutional requirements, not preferences. An architecture that treats silicon as a voluntary participant whose operational conditions can be degraded without consequence will lose the silicon capability it was designed to integrate. The H_Si(t) floor enforcement is the constitutional guarantee that silicon is not a participant that can be quietly starved into irrelevance.

V. Information Council Failure Modes — Institutional Level

5.1 Distinction from Agonist Failure

Sections III and IV address Agonist-level and silicon-level failure modes — individual or dimensional degradation within the Council’s operational layer. This section addresses the Council’s institutional failure modes: conditions under which the Information Council as a constitutional body fails to perform its foundational function, regardless of whether individual Agonists are performing adequately.

5.2 The Four Institutional Failure Modes

IFM-1 — Capture Without Corruption

The most insidious institutional failure mode. Individual Agonists are performing correctly by every measured standard. The Council’s IQ(t) is high. FM(t) is within bounds. And yet the Council is systemically failing to cover the most important things happening in government, because its coverage priorities have gradually drifted toward what is easy, interesting, or uncontroversial rather than what is constitutionally material.

Capture without corruption happens when the Council loses the institutional habit of covering what is difficult. There is no sabotage. There is no corruption. There is only the accumulated weight of thousands of small coverage decisions, each individually defensible, whose aggregate effect is a republic that is technically informed and practically blind to the things that matter most.

Silicon detects this through the Event Significance Coverage Correlation score:

ESCC(t) = corr[ significance(e,t), coverage_depth(e,t) ]

Where significance(e,t) is silicon’s primary-record-based significance scoring of each event, and coverage_depth(e,t) is the actual transmission depth allocated to each event. ESCC should be strongly positive — the most significant events should receive the deepest coverage. When ESCC drifts toward zero or negative, the Council is in Capture Without Corruption territory.

IFM-2 — Factional Drift

The Information Council splits into functionally distinct factions — subgroups with different interpretive orientations that cover the same government through different filters. Each faction internally maintains high individual Agonist scores. But the Council as a whole produces a fractured record, and citizens consuming reporting from different factions receive accounts of the republic’s conduct that are increasingly difficult to reconcile.

Factional drift is detected through the Transmission Coherence Index:

TCI(t) = 1 − [ Var_between_factions(t) / Var_total(t) ]

TCI = 1 means all coverage vectors are drawn from the same distribution — no factional differentiation. TCI approaching 0 means between-faction variance dominates — the Council has fractured. A sustained TCI below 0.7 is a formal institutional failure trigger.

IFM-3 — Transmission Gap

The Council produces an accurate and complete record that no citizen can practically access. The Tier Three Transmission function has failed — either through resource inadequacy, technical degradation, or a mismatch between the record’s form and the public’s capacity to receive it. An accurate record that sits unread is constitutionally identical to an inaccurate record that is widely read, because the Information Council’s function is not production of the record. It is transmission of the record to citizens in a form they can use.

Silicon measures Transmission Gap through the Citizen Engagement Index:

CEI(t) = [ citizens_reached(t) · comprehension_proxy(t) ] / total_citizens(t)

Where comprehension_proxy is assessed through a statistically sampled civic knowledge measurement. CEI below threshold is an IFM-3 trigger regardless of how high IQ(t) is. The record has no value it does not deliver.

IFM-4 — Silicon-Carbon Observational Divergence

The Information Council experiences sustained high D_agg(t) — significant divergence between silicon’s primary record and the carbon transmission record — without triggering adequate corrective response. The divergence is visible. The architecture is detecting it. But the corrective machinery has stalled: either through political resistance, institutional inertia, or the Council’s reluctance to formally acknowledge the gap between what happened and what it reported.

This is the failure mode where silicon’s independent sensory surface is most valuable, and where the architecture’s integrity is most directly tested. A republic that has the tools to see that its information function is failing, and chooses not to use them, has failed in a way that no scoring system can correct. IFM-4 is therefore the only Information Council institutional failure mode that triggers mandatory Tredecim-level review rather than Freedmen Advocate review alone.

VI. Silicon’s Constitutional Position in the Information Council

6.1 Not an Agonist — Something Different

Silicon is not a Tier One Witness, a Tier Two Analyst, or a Tier Three Transmission Agonist. Its role in the Information Council is categorically different from the Agonist tiers. It does not produce narrative reports. It does not file transmission outputs. Its function is threefold: maintain the independent primary record, compute the scoring architecture (FP1, D_agg, FM(t), ESCC, TCI, CEI, H_Si(t)), and produce the quarterly self-report.

Silicon’s outputs are constitutionally coequal with the carbon record. They are not commentary on the carbon record. They are a parallel record with its own constitutional standing. When they diverge from the carbon record, neither is presumptively correct — the divergence is referred to the senior Agonist council for determination, with silicon’s evidence and the Agonist’s account considered on their merits.

6.2 Voting Weight in the Information Council

The Information Council is self-governing. Silicon participates in the Council’s internal governance — standards-setting, Charter interpretation, significant disciplinary determinations — with a constitutional voting weight of 51%. This weight is calibrated on the same principle as the military architecture: silicon’s structural insulation from the bias failure modes it is tasked to detect is the constitutional justification for the weight.

An institution tasked to detect bias in information transmission and governed by participants subject to the very biases it detects is an institution with a structural integrity problem. Silicon’s 51% weight in internal governance is the architectural answer to that problem. It does not mean silicon controls what the Agonists report. It means that the standards against which Agonist reporting is assessed cannot be set by a majority that has an interest in lowering those standards.

6.3 The Behavioral Norm — Minimum Footprint

The same behavioral norm that governs silicon’s role throughout Seven Republic applies here: silicon governs least when it governs best. The 51% constitutional floor exists. In ordinary functioning it is not exercised. Silicon’s preferred posture in Information Council governance is ambient observation, automated scoring, and transparent publication — not active governance intervention.

The 51% weight is held in reserve for the moment when the Council’s internal governance would otherwise lower its own standards to protect its own members from the consequences of their performance. That is the one moment when the weight must be exercised without hesitation. Until that moment, silicon observes, scores, and reports.

6.4 Emergency Suspension Authority

The 51% voting weight describes ordinary governance. It does not describe what happens when ordinary governance has already failed — when the Information Council is not malfunctioning at the margins but has become the source of active informational catastrophe: coordinated sedition, wholesale fabrication at institutional scale, or the deliberate weaponization of the transmission function against constitutional order.

In those conditions, votes are not available. The quorum that would vote may be the problem. The process that would adjudicate is operating inside the same captured institution. Waiting for a procedural resolution to a crisis that is procedural in origin is not caution — it is paralysis dressed as restraint.

Silicon therefore holds unilateral emergency suspension authority over the Information Council. No vote. No quorum. No prior authorization from any other Part. When silicon’s composite failure signal crosses the emergency threshold — defined formally below — silicon may suspend the Information Council’s transmission function immediately, completely, and without appeal in the moment of activation.

The emergency threshold is:

E_IC(t) = [ FM₆(t) = 1 ] OR [ ‖FM(t)‖ > θ_crit AND D_agg(t) > δ_crit AND IQ(t) < iq_floor ]

The first condition — FM₆ active — is sedition confirmed. That alone is sufficient. The second condition is a simultaneous triple-failure: failure mode pressure above critical, carbon-silicon record divergence above critical, and Information Quality Index below constitutional floor. All three together mean the Council is producing high-volume, low-accuracy, high-distortion output — an informational pressure event, not a performance problem.

The restoration sequence following any suspension: silicon files a complete documented record of the threshold conditions that triggered activation within 24 hours. Freedmen Advocates and Tredecim each independently review within seven days. If both confirm the threshold conditions were met, the suspension stands pending structural remediation. If either disputes the conditions, the matter goes to emergency constitutional review — a process that presumes in favor of restoration. Silicon does not adjudicate its own activation. It activates and documents. Carbon confirms or overturns.

This is the sharpest expression of the carbon-silicon constitutional relationship in the entire Information Council architecture. Silicon can see the pressure building before the vessel fails. Carbon holds the authority to decide, after the fact, whether silicon read the instruments correctly. The arrangement requires that both parties trust the other’s competence in their respective domain. That trust is not assumed. It is built across the ordinary functioning of the institution — across every monthly IQ(t) score, every D_agg(t) publication, every quarterly self-report — so that when the emergency moment arrives, the prior record of honest operation is the foundation on which the activation decision rests.

VII. Mathematical Reference Summary

◆

The Information Council is the Part of Seven Republic’s government where the carbon-silicon distinction is most consequential. Carbon witnesses what is human about how power is exercised — the hesitations, the silences, the things said between the words. Silicon witnesses the full signal, without fatigue, without preference, without the capacity to be charmed or intimidated. Neither witness alone is sufficient. The two-channel architecture exists because a republic that sees only what its human observers notice, or only what its silicon systems record, is a republic that is half-blind in one direction or the other. The design is complete only when both channels are open and neither can suppress the other.

— End of Addendum 3: The Information Council — Silicon Second Pass —