The Landscape of Normative Systems

Political, Moral, Religious and Social Sources of Governance

Document class: Internal research memorandum series Program: Computable Governance (upstream reference for D1/D2) Status of D0: Immutable. Not modified, not criticized, not defended in this series. Series ID: LNS (Landscape of Normative Systems)

0.1 Purpose

This memorandum series prepares the intellectual foundation for a new scientific field, Computable Governance. The field already possesses a frozen constitutional document, D0, which specifies (a) a computational kernel for normative systems and (b) a research methodology. Both are treated here as immutable inputs.

The task of this series is not to design the kernel. It is to map the territory above the kernel: every source from which normative systems originate, the mechanisms by which they are produced, legitimated, enforced, and mutated, and the structural invariants that any compiler sitting on top of D0 would have to accept as input.

The intended reader is a future compiler architect — an engineer who will build D1 (an intermediate normative representation) and D2 (an executable governance layer) on top of D0. The series must be complete enough that such a team could work for six months without commissioning another political or philosophical survey.

This is not a literature review, a textbook, or a philosophy essay. It is a working reference memorandum.

0.2 Relationship to D0 (the frozen kernel)

D0 contributes two things this series must honor:

1. A computational kernel for normative systems (the "Governance Kernel"). Its internal design is out of scope. We only characterize its interface: what kinds of objects it can accept and what it must refuse.

2. A research methodology (D0's "Research Context"), summarized as binding constraints:

• M1 Falsifiability. Every scientific claim must be falsifiable.
• M2 Elimination over brainstorming. Primitive concepts are derived by removing what fails, not by enumerating what sounds good.
• M3 Hostile emergence. Architecture must survive hostile counterexamples; it is not assumed.
• M4 Language is presentation. Natural language is a surface. We seek computational objects beneath it.
• M5 Destructive testing. Every proposed primitive must survive attempts to break it.
• M6 Discovery, not invention. "No kernel exists" remains an acceptable outcome. We do not manufacture structure to reach a desired conclusion.
• M7 No forced separation. The distinctions knowledge / semantics / execution / presentation are used only if they genuinely emerge.
• M8 Theories are evidence, not authority. Existing political, legal, and moral theories are data points that may all be partially wrong.

Every memo in this series is written under M1–M8. Where a memo cannot satisfy M1 (falsifiability) for a claim, it must downgrade the claim's epistemic tag (see §0.4) rather than assert it.

The kernel/above-kernel boundary is the central organizing question of the series and is resolved in Memo 12.

0.3 Series structure

Memos 01–08 survey sources. Memos 09–10 abstract across sources. Memo 11 attacks the whole enterprise. Memo 12 synthesizes the invariant interface.

0.4 Epistemic tagging (mandatory)

Every substantive claim carries one of four inline tags. This is a hard requirement under M1.

• FACT — Established, checkable, and not seriously contested within the relevant scholarly community. Historical events, textual content of laws, formal results (e.g., Gödel's incompleteness theorems).
• INFERENCE — A conclusion drawn by reasoning from facts, defensible but not directly observed. Marked so the reader can audit the reasoning.
• HYPOTHESIS — A proposed explanation or model that is plausible and, per M1, stated so it could be falsified. Not yet established.
• OPEN — An open problem: something the series cannot resolve and that D1/D2 work must treat as unresolved.

When uncertain, tag down (FACT→INFERENCE→HYPOTHESIS→OPEN), never up. Uncertainty is stated explicitly. References are never invented; where a thinker's position is summarized, it is attributed by name and work only when the attribution is safe, otherwise phrased as "the position associated with X."

0.5 Section separation (mandatory)

Within each memo, findings are separated into four standing categories, either as top-level sections or clearly labeled subsections:

1. Established consensus — what the relevant fields broadly agree on.
2. Competing theories — live disagreements, presented without adjudication.
3. Open questions — unresolved problems.
4. Research opportunities — where D1/D2 work could contribute or must decide.

No memo may blur these into a single undifferentiated narrative.

0.6 Shared vocabulary (used identically across all memos)

• Normative system — any system that produces, maintains, and enforces ought-statements (rules, obligations, permissions, prohibitions, powers).
• Norm — a single ought-statement with a modality (obligation / permission / prohibition / power / immunity), a subject, a condition, and a consequence.
• Source — the origin from which a norm derives its content and/or authority (e.g., legislation, revelation, contract, custom).
• Legitimacy — the property by which a norm's subjects treat it as binding independent of immediate coercion. Distinguished sharply from validity (formal membership in a system) and efficacy (actual compliance).
• Enforcement mechanism — how deviation is detected and how consequences are applied.
• Mutation mechanism — how the norm set changes over time (amendment, reinterpretation, desuetude, revolution, drift).
• Kernel (D0) — the frozen computational core. Content-neutral by hypothesis.
• Above the kernel — everything ideological, value-laden, or culturally particular. This series claims most normative content lives here.
• Interface — the invariant, content-neutral structure that any source must be reduced to before it can be handed to the kernel. Identified in Memo 12.

Hohfeldian primitives (used as the working modality set)

We adopt Wesley Hohfeld's eight jural relations as a candidate modality vocabulary because it is (a) content-neutral and (b) already near-formal: right/duty, privilege/no-right, power/liability, immunity/disability. Adoption is provisional and is stress-tested in Memos 10–12. INFERENCE that this set is adequate; it is a working assumption, not a result.

Deontic modalities

Where formal logic is invoked, we use standard deontic operators: O (obligatory), P (permitted), F (forbidden), with the standard relations P(x) ≡ ¬O(¬x) and F(x) ≡ O(¬x). Known paradoxes (Ross, Chisholm, Good Samaritan, gentle murderer) are treated as destructive tests, per M5, in Memo 10.

0.7 Style constraints

• Serious internal memo register. No marketing, no inspirational language.
• No speculation presented as fact (enforced by §0.4 tags).
• Political neutrality: describe how systems generate norms; do not rank them.
• Diagrams are ASCII/Mermaid-style text, so they survive plain-text tooling.
• Each memo is self-contained but cross-references others by memo number.
• Target depth: each memo is a substantial standalone document. Depth and completeness take priority over hitting an exact page count.

0.8 Reading order

Linear (01→12) is intended. Readers building D1 may start at Memo 10 (taxonomy), then Memo 12 (interface), using 01–09 as the evidence base and Memo 11 as the risk register.