Unit Review: Team Encoded Objective Law and Selective Transparency 2.3B.2-R

Key Structural Points

• Design without enforcement is decoration — a team charter on paper is not a charter in practice, and correct behavior must be the structural path of least resistance rather than a matter of individual vigilance.

• Encoded objective law at team scale embeds moral constraints in process: recurring high-stakes decisions become checklists, consequential acts require visible authorization through signatures, and an append-only audit log makes the decision record immutable.

• The enforce-in-code decision tree distinguishes decisions that warrant structural enforcement (clear conditions, testable criteria, high stakes, just default, attestation-verifiable) from those that should remain discretionary but transparent (published criteria, documented reasoning, appeal process).

• Selective transparency inverts the common default: process is public, persons are private, and exceptions are logged with written justification, scope, duration, and sunset — the team-level expression of privacy by design (1.2.4) and the asymmetry principle.

• Attestation without surveillance is the mechanism that makes selective transparency operational — claims about fairness, compliance, and consistency can be verified through spot-check auditing and published attestations without monitoring individuals.

• The dispute path provides structured recourse in four stages: informal resolution (forty-eight hours), mediated review (five days), binding decision with written reasoning (ten days), and appeal on narrow grounds — connecting to bounded interdependence's recourse requirement (1.2.6).

• Under AI conditions, all requirements intensify: errors are systemic (affecting many people), transparency is harder (no human can explain the system's reasoning), and auditability demands version tracking, behavioral drift monitoring, and immediate rollback capability.

• The four AI load cases — screening, measurement, automation, and decision drift — identify the specific structural risks when team decisions are mediated by computational systems, with structural requirements for each.

Philosophical Lexicon

• append-only state — A system design in which decisions and log entries, once recorded, cannot be silently altered, deleted, or backdated — ensuring the decision record is auditable by design. (RO)
• asymmetry principle — The structural commitment that institutions must be transparent and persons must be opaque — derived from the inherent power differential between institutional and personal positions. (RO)
• bounded interdependence — Resilient Objectivism's design discipline governing two structurally distinct domains: for voluntary cooperative arrangements, a design recommendation — clean exit, meaningful recourse, and data portability — grounded in the builder's own self-interest, not an entitlement owed to a departing participant; for the exercise of public force, a categorical requirement that legitimate authority stay bounded to retaliatory force against initiators. One of RO's nine design patterns. (RO)
• consent theater — Mechanisms that present the appearance of consent without satisfying the structural conditions for genuine consent — capacity, information, absence of coercion, and recourse. (RO)
• privacy by design — The architectural commitment that persons remain selectively opaque to power; systems default to privacy and require justification for exposure, not the reverse. One of RO's nine design patterns. (RO)
• encoded objective law — The embedding of moral constraints into system design — through code or protocol, not merely law, habit, or virtue — so compliance is structural rather than dependent on individual virtue. One of RO's nine design patterns. (RO)
• selective transparency — The three-tier institutional disclosure discipline: processes and rules public by default, operational details auditable on request, personal information protected by design — the team-level expression of privacy by design and the asymmetry principle. (RO)

Unit Glossary

• attestation — A verification mechanism confirming a claim without exposing underlying data — the decision-maker certifies the process was followed, and the certification is verified through spot-check audit.
• decision checklist — An ordered list of criteria or steps that must be completed before a decision is authorized — ensuring the decision-maker considers what should be considered.
• default classification — The pre-set transparency status of an information category: public (available to all), auditable (available on request), or protected (accessible only to the person and designated stewards).
• dispute path — The structured process for addressing violations: informal resolution, mediated review, binding decision with written reasoning, and appeal on narrow grounds.
• enforce-in-code decision tree — A five-question diagnostic determining whether a recurring decision warrants structural enforcement or should remain discretionary but transparent.
• exception log — A record of every instance where the default transparency classification was overridden, documenting justification, authorization, scope, duration, and sunset.
• integrity officer — The role responsible for auditing team processes, investigating violations, mediating disputes, and conducting quarterly privacy attestations.
• signature — Visible authorization for a consequential decision, recorded in the audit log with date and timestamp, creating accountability traceable to a named person.
• transparency matrix — The documented specification of what information is public by default, auditable on request, and protected by design — replacing ad hoc decisions about visibility.

Concept Check

• Why does the article claim that design without enforcement is decoration? What specific structural mechanism makes the difference between a charter on paper and a charter in practice?

• The enforce-in-code decision tree has five questions. Which question directly applies the asymmetry principle from Phase I's privacy by design (1.2.4), and why is that question the appropriate gatekeeper for structural enforcement?

• The transparency matrix inverts the common default. Explain why treating process as public and persons as private is structurally necessary — not merely ethically preferable — for a team practicing encoded objective law.

• How does attestation without surveillance work for verifying hiring fairness? What data is needed, what is not needed, and what structural mechanism prevents attestation from degenerating into rubber-stamping?

• The dispute path specifies four stages with time targets. What structural function does the time target serve — and what would happen to the team's accountability if the stages had no deadlines?

• Under AI conditions, the article claims that all requirements intensify. For the screening load case specifically, why must the person affected by an AI-mediated decision be told the system was involved — even if a human reviews the decision afterward?

• The AI sidebar identifies decision drift as a structural risk. Why is immediate rollback capability a structural requirement rather than a nice-to-have — and what happens to people affected by decisions made under a drifted system?

• What would be missing from a team's enforcement practice if it built the enforce-in-code decision tree but not the transparency matrix? Which of encoded objective law's structural requirements would fail?

Points to Ponder

• Think about a recurring decision in a team or organization you know. Is it enforced in code (checklist, signatures, logging) or discretionary? If discretionary, has anyone ever disputed the outcome — and was there a structural mechanism for resolution, or did it depend on informal negotiation?

• Consider the transparency default in your current work environment. Is individual performance data more visible than decision-making process? If so, what structural consequence does that reversal produce for the people being measured?

• The dispute path requires written reasoning at the third stage. In your experience, how often are consequential team decisions accompanied by written reasoning that the affected person can read and contest? What would change if they were?

• The AI sidebar argues that AI can draft but cannot decide. In practice, how often does an AI-generated recommendation become the effective decision because the human reviewer defaults to acceptance? What structural mechanism would prevent that defaulting?

To Learn More

• Ostrom, Elinor. Governing the Commons: The Evolution of Institutions for Collective Action. Cambridge: Cambridge University Press, 1990. Develops the design principles for self-governing institutions — including monitoring, graduated sanctions, and conflict resolution mechanisms — that provide the empirical grounding for the enforce-in-code decision tree's distinction between structurally enforced and discretionary decisions.

• Nissenbaum, Helen. Privacy in Context: Technology, Policy, and the Integrity of Social Life. Stanford, CA: Stanford University Press, 2010. Provides the theoretical framework for contextual integrity — the principle that privacy norms are tied to the social context of information flow — which grounds the transparency matrix's three-tier classification and the exception logging framework.

• Pasquale, Frank. The Black Box Society: The Secret Algorithms That Control Money and Information. Cambridge, MA: Harvard University Press, 2015. Documents the structural consequences of opaque algorithmic decision-making in finance, search, and reputation systems — directly relevant to the AI sidebar's four load cases and the argument that AI-mediated team decisions require stricter transparency and auditability.