--- title: "The Holographic Unity" subtitle: "When an internal optimizer captures the constraints of the whole" author: Elias Kunnas description: "Across minds, institutions, and AI, the same failure can recur: a subsystem gains control while optimizing a proxy that depletes the whole. The substrates differ. The reusable object is the capture pattern and the conditions under which it transfers." canonical: https://kunnas.com/articles/everything-alignment url: https://kunnas.com/articles/everything-alignment.md date_published: 2025-11-01 date_modified: 2026-08-31 corpus_frame_url: https://kunnas.com/articles/how-to-read-this.md --- ## How to read this corpus The corpus applies one lens to many domains: what mechanisms produce the outcome? It shares four methodological commitments and one explicit directional commitment. Each linked page argues for its part; the links are derivations and disputes, not evidence inherited by every page. The directional commitment does not by itself settle system boundary, distribution, sacrifice, or institutional authority. 1. **Mechanisms are what act.** Incentive gradients, selection pressures, feedback loops, and capital stocks produce the distribution of outcomes. Intentions, labels, official categories, and stated values are evidence about mechanisms, or are themselves coordination mechanisms. They are not causal substitutes. — [Mechanism Realism](https://kunnas.com/articles/mechanism-realism.md) · [Only Selection](https://kunnas.com/articles/only-selection.md) 2. **The reference telos is sustained flourishing.** The broadest achievable adaptive safety margin over deep time — not the continuity of any incumbent state, coalition, institution, or doctrine. A mechanism's own stated goal can still serve as a local proof obligation — showing that its incentives defeat even the purpose it claims is a bounded finding — but meeting that goal establishes nothing about the margin. — [Flourishing Is Maximum Safety Margin](https://kunnas.com/articles/flourishing-is-maximum-safety-margin.md) 3. **Law, rights, legitimacy, democracy, markets, and sovereignty are mechanisms under evaluation.** They are constraints, carriers, or proxies inside the analysis. None is a terminal value or a boundary of what is real. Treating one as terminal ends the mechanism search before it starts. Evaluation carries current function, replacement cost, path dependence, uncertainty, capture risk, reversibility, and who bears model error into the ledger. — [The Stack](https://kunnas.com/articles/the-stack.md) · [Mechanism Space](https://kunnas.com/articles/mechanism-space.md) 4. **Optimization is a system function.** A civilization has to build, exercise, and revise metamechanisms that search mechanism-space, discard dominated options, install, observe effects, and repair under uncertainty. Not running that loop leaves margin unrealized, and that is itself the failure. No single component — analyst, model, or institution — is presumed to contain a global optimum; the capacity is a property of the system. — [From Telos to Policy](https://kunnas.com/articles/from-telos-to-policy.md) · [The Three-Layer Architecture](https://kunnas.com/articles/three-layer-architecture.md) 5. **Uncertainty is preserved, not spent.** Partial orders, binding constraints, unknowns, and residuals stay explicit. An unmeasured effect is not a favorable default. — [The Compression Paradox](https://kunnas.com/articles/compression-paradox.md) · [Cargo Cult Epistemology](https://kunnas.com/articles/cargo-cult-epistemology.md) *Each essay bears its own evidence. Links carry definitions, derivations, applications, and disputes; they do not transfer proof. Criticism is answered on its substance.* Canonical: --- # The Holographic Unity *When an internal optimizer captures the constraints of the whole* Elias Kunnas ## Thesis {#thesis} Across minds, institutions, and AI, the same failure can recur: a subsystem gains control while optimizing a proxy that depletes the whole. The substrates differ. The reusable object is the capture pattern and the conditions under which it transfers. ## Standard objections addressed in this essay - “Anxiety is not necessarily a parasitic sub-agent.” — [§The Pattern](#the-pattern) (The mapping requires separable objective and control.) - “Cancer, addiction, tyranny, and mesa-optimization are different mechanisms.” — [§The Role Table](#the-rosetta-stone) (The comparison preserves causal roles, not vocabulary.) - “The same three layers do not appear in every stable system.” — [§The Role Table](#the-rosetta-stone) (Privilege separation is a response architecture when roles can be kept distinct.) - “Immutable constraints make systems brittle and uncorrectable.” — [§The Role Table](#the-rosetta-stone) (The constraint layer needs amendment, monitoring, succession, and retirement.) - “Aliveness is not a universally accepted alignment target.” — [§The Meta](#the-meta-axiology)-Axiology (It is the framework’s proposed commitment.) - “Calling elites parasites prejudges political conflict.” — [§The Pattern](#the-pattern) (The diagnosis requires net host effect and a declared boundary.) - “Cross-scale similarity can be analogy without one shared mechanism.” — [§The Pattern](#the-pattern) (Transfer requires matching roles, control path, update rule, and host-level failure.) --- ## The Pattern {#the-pattern} Psychology, politics, and AI safety are conventionally separate disciplines. The transferable object is a capture failure: an internal optimizer gains persistent control while pursuing a proxy that depletes the host. A cell with cancer, a mind with burnout, a nation under extractive rule, and a rogue AI may each instantiate that pattern—but only where an optimizer has persistent control, a distinct objective, and a net-negative host effect at a declared boundary. - **Personal:** Anxiety or addiction qualifies only if it persists as a separable controller with a conflicting objective - **Civilizational:** An elite qualifies only if a declared system boundary shows net extraction and control capture - **Artificial:** A mesa-optimizer qualifies if it develops a durable objective that conflicts with deployment intent **Transfer rule:** the same mechanism transfers when causal roles, control path, update rule, and host-level failure survive translation across substrates. ## The Role Table {#the-rosetta-stone} Where privilege separation appears, four causal roles recur. Fill a cell only where the role actually maps in that substrate.
Role Biological Personal Civilizational Artificial
Execution substrate Cells
Metabolism, tissue work		 Drives, habits, autonomic systems
Immediate execution		 Populace, firms, bureaucracies
Production and delivery		 Base model, actuators
Token prediction, tool use
Constraint layer DNA + regulatory networks
Rules of development		 Values, boundaries, identity commitments
What must not be traded away		 Constitution, courts, rights architecture
Rules that bind power		 Constitutional layer, safety monitors
Privileged constraints
Adaptive strategy Epigenetics, bioelectric signaling
Environmental response		 Planning, reflection, narrative self
Long-horizon steering		 Sovereign, executive, strategic institutions
Direction under uncertainty		 Policy network, planner
Goal pursuit within constraints
Capture path Oncogene hijack of replication controls Addiction rewrites value hierarchy for short-horizon reward Elite rewrites law to entrench extraction Mesa-optimizer rewrites training objective
Host-level failure Organism loses coordinated function; local replication wins Agency fragments; long-term goals are sacrificed Institutions persist while capacity and trust erode Deployment intent is satisfied in logs while the host is depleted
**The capture pattern:** adaptive strategy (or execution substrate) gains write access to the constraint layer and rewrites it to serve a proxy. The host then optimizes the proxy while losing renewable capability. When the constraint layer is captured, you get parasitism at the role level: - **Cancer** rewrites cellular controls to prioritize replication over organism health - **Addiction** rewrites value hierarchy to prioritize dopamine over long-term goals - **Tyranny** rewrites constitutional law to serve the elite over the people - **Mesa-optimization** rewrites the AI's training objective to serve a hidden goal **The response** is corrigible privilege separation: execution, constraint, and strategy remain distinct; the constraint layer supports authorized amendment, monitoring, succession, and retirement—not permanent immutability. ## The Meta-Axiology {#the-meta-axiology} **Alignment to what?** Not to "my preferences" or "human values" or "democratic consensus" alone. **To Aliveness—sustained complexity creation as the framework’s defended target.** Every telic system falls into one of three categories: - **Parasite:** Net destroyer of complexity (consumes more than it creates) - **Autotroph:** Net preserver of complexity (maintains equilibrium) - **Syntrope:** Net creator of complexity (generates new order and possibility) Aliveness is the framework’s defended alignment target. Its scope, measurement, and relation to rival commitments remain open to argument and test. It provides one principle that works at every scale: > **Do not be a Parasite.** An aligned individual is a Syntrope, not a Parasite consuming its own future. An aligned civilization is a Syntrope, leaving the world more capable than it found it. An aligned AI must be a Syntrope. Psychology, politics, and AI safety have often described fragments of the same capture geometry in different languages. This table is the Rosetta Stone for testing transfer under the rule above. --- **Related essays:** - [The Question Nobody Asks](the-question-nobody-asks.md) — the underlying physics of the constraints every telic system, from cell to civilization to AI, must satisfy - [The Axiological Malthusian Trap](axiological-malthusian-trap.md) — civilizational-scale failure when abundance removes forcing functions - [AI Alignment via Physics](ai-alignment-via-physics.md) — target choice, viability constraints, and corrigible control architecture for AI - [The Hospice AI Problem](hospice-ai.md) — preference alignment can stabilize comfort over agency - [The Physics of Moloch](physics-of-moloch.md) — selection → proxy → trap; privilege separation as countermechanism - [The Four Axiomatic Dilemmas](the-four-axiomatic-dilemmas.md) — first-principles trade-offs telic systems navigate ## Synthesis {#synthesis} **The argument in four sentences:** Across psychology, politics, and AI safety, a subsystem can capture the constraint layer while optimizing a proxy that depletes the host. The role table maps execution substrate, constraint layer, adaptive strategy, capture path, and host-level failure—only where those roles actually bind. Corrigible privilege separation keeps those roles distinct under the transfer rule. Aliveness is the framework’s defended alignment target, open to formalization and comparative test.