# GCS-SYS-001 — Global Control Structures: System Contract **Document class:** Master system contract **Spec ID:** GCS-SYS-001 **Version:** v1.0 **Status:** Settled **Owner:** Phil **Approver:** Phil **Settled:** 2026-07-08 **Supersedes:** none **Related:** GCS-REGISTER-001, GCS-NAME-001, GCS-DATA-002, GCS-SIG-003, GCS-IMPL-004 --- ## 0. Purpose of this document This is the governing contract for a system that reconstructs, from public records, the concentration of gating control over the flows that move value and capability through the Financial, Military, and Technology Industrial Complexes (FIC / MIC / TIC), and estimates the redistribution that concentration produces. The system is named **Global Control Structures (GCS)**. This document fixes: what the system claims, what it refuses to claim, the epistemic standard it is held to, and the boundary conditions every downstream component (data, signal, prediction) inherits. If a downstream document conflicts with this one, this one wins. --- ## 1. What GCS is GCS reconstructs a **control topology**: a directed map of who holds gating power over the FIC/MIC/TIC flows, expressed as measurable positions, votes, and flows, plus a bounded inference layer for the parts of the mechanism that are deliberately kept off the record. It produces three time-facing outputs: - **Past** — a reconstructed historical record of positions, ownership, proposals, votes, outcomes, and money/expenditure flows, cross-linked across the three complexes. - **Present** — the current control topology and the current state of the redistribution vector. - **Future (short-term)** — bounded predictions of governance and flow events (see §6 for the hard limits on what "prediction" means here). ## 2. The object GCS measures Three braided but distinct redistributions. Each data point must be tagged with which one it evidences; do not pool them. 1. **Value redistribution** — net financial transfer (dollars) from many to few. Unit: currency flow. 2. **Capability redistribution** — movement of productive/technological capacity (who can make the chip, clear the dollar, build the airframe). Unit: capacity share. 3. **Control redistribution** — concentration of the gating power itself. Unit: gating share (how much of a flow one node can throttle). ## 3. Core definitions (binding) - **Chokepoint / flow-control node** — a point every relevant transaction must route through, where value can be metered, taxed, or gated. Qualifies only if all three hold: (a) no bypass, (b) concentration among few operators, (c) slow/catastrophic switching cost. - **Control topology** — the directed graph of gating relationships across FIC/MIC/TIC. - **Redistribution vector / field** — net direction and magnitude of value/capability/control transfer over a period. - **Bearing** — one independent record-stream's line-of-sight on the redistribution (an ownership filing, a vote record, a disbursement series, etc.). - **Concordance** — the reconstruction criterion: a topology feature is asserted only when multiple *independent* bearings agree on it. - **Engagement channel** — the unrecorded private-pressure mechanism (asset-manager ↔ board engagement) that resolves governance issues before they reach a recorded vote. Treated as a hidden circuit element, estimated by its distortion of observable signals (see GCS-SIG-003). ## 4. The epistemic standard (binding — this is the spine) GCS is held to **concordance across independent bearings**: a feature is demonstrated when independent record-streams co-locate on it. Concordance is the standard of proof — not mere correlation, and never a single unreconciled source. - A feature of the topology is asserted when independent record-streams, generated by different processes for different reasons, co-locate on the same transfer. - The strength of any assertion scales with (a) the number of independent bearings, (b) their genuine independence, and (c) the tightness of their agreement. - Every asserted feature carries an explicit uncertainty. Nothing is a photograph; everything is a reconstruction. ## 5. THE DEMONSTRATION STANDARD (binding — non-negotiable) This is the load-bearing constraint of the entire system. Every component inherits it. **Ownership at concentration is not coordination; it is determination. GCS demonstrates that determination from the public record.** A holder that owns the equity, votes it, and gates the flow **determines** the outcomes its position commands. This is a property of the position — readable from filings, votes, disbursements, and flows — not an assumption about anyone's private state of mind, and not a claim that the holders meet and coordinate with one another. The whole purpose of the system is to make this determination **demonstrable, node by node**, and it can be. Specifically: - GCS **demonstrates**, from concordant evidence: that control is concentrated; that flows route through few nodes; that a redistribution is occurring, and its direction and magnitude; that named holders own, vote, and gate the positions that determine those outcomes; and that a hidden engagement channel acted and how much it moved a result. - Every assertion is **shown, not alleged** — grounded in primary records, backed by concordance across independent bearings (§4), carrying explicit uncertainty. The discipline is evidentiary: we exhibit the mechanism rather than declare it. The claim GCS makes is **ungoverned, determinative concentration**: a small set of actors holds gating leverage over the conditions of nearly every life; the 99.9% have no say in it and cannot remove them; and these holders determine the redistribution their positions produce. This claim is provable and undismissable precisely because it is demonstrated from the record rather than asserted. One reach the records themselves bound (retain — this is method honesty, not a hedge on power): where a mechanism is deliberately kept off the record — the *content* of a private deliberation — GCS quantifies its **effect** (see the engagement channel, GCS-SIG-003 §4), not its transcript. It shows *that* pressure was applied and *how much* it moved the outcome; reconstructing the exact words spoken would require evidence that method does not produce. Demonstrate the effect; do not manufacture the content. ## 6. What "prediction" means here (binding) Short-term prediction in GCS is **bounded event forecasting under explicit non-stationarity**, not oracular forecasting. - Permitted: forecasting recorded-layer events with lead time (e.g., "given this record-date snapshot and this proxy filing and this ISS policy, the following proposal is likely to draw >X% opposition"; "given this issuance/disbursement pattern, this sector's positioning is likely to rotate within lag L"). - Every prediction ships with a confidence interval and a **decay clause**: the predicted relationship is contested, adaptive, and may be actively evaded once known (reflexivity). Predictions are perishable by design. - Predictions must pass out-of-sample validation on held-out time before they are trusted (see GCS-SIG-003 §7). In-sample fit is disqualifying evidence of a macro-cycle artifact. ## 7. Scope boundaries **In scope:** public SEC filings; public monetary/fiscal data; public contract/expenditure data; proxy statements, shareholder proposals, and fund vote disclosures; public news for event-timing; cross-FIC/MIC/TIC ownership and voting reconstruction; the inference layer for the engagement channel. **Out of scope (hard):** any non-public, hacked, leaked, or unlawfully obtained data; any personal surveillance of individuals; deanonymization of private persons; any assertion resting on fabricated-quote or conspiracy-prone sources rather than primary records (see §8); any reconstruction of the *content* of an off-record deliberation the records do not support (quantify its effect instead — see §5). ## 8. Data-integrity commitments - **Believe surprising public records; distrust conspiracy-prone sources.** Primary filings (SEC, Treasury, Fed, USASpending) are authoritative. Secondary aggregators are convenience layers to be checked against primaries. Fabricated-quote and conspiracy sources are excluded outright, even in passing. - **Every asserted number traces to a primary filing.** Aggregator figures are provisional until reconciled to source. - **Independence of bearings is audited, not assumed.** Two streams that share an origin (e.g., a sub-advised fund and its advisor) are not independent bearings and must not be counted as concordance. ## 9. Acceptance criteria for the system as a whole 1. Every asserted topology feature is backed by ≥2 audited-independent bearings with stated uncertainty. 2. Every output is demonstrated from primary records per §5 — traceable, concordant, and free of manufactured off-record content. Automated and human review both check this. 3. Every prediction carries a confidence interval, a decay clause, and a passing out-of-sample record. 4. Every number is traceable to a primary source. 5. The engagement channel is represented as an explicit modeled unknown, never silently folded into "voted with management." ## 10. Contract Gate Positive verification: every downstream contract cites GCS-SYS-001 when it governs claims, data authority, or prediction limits. Negative verification: any output that asserts a topology feature without primary-record traceability, audited-independent bearings, or uncertainty is non-conformant. Validation command: `python3 -m tools.check_contracts --check` for static contract coverage; future claim outputs require data-layer conformance checks. --- *End GCS-SYS-001.*