From Attractors to Agents

Begin with the weakest version. A coordination attractor is any persistent macro-scale pattern that stabilizes correlated behavior across many agents by shaping incentives, attention, and shared models. A language is a coordination attractor—it structures communication, resists individual modification, and persists through speaker turnover. A dress code is a coordination attractor. A market convention, a social norm, a shared aesthetic. These patterns attract behavior toward them without actively regulating their substrate. They are stable because deviation is costly, not because the pattern acts to prevent deviation.

But some coordination attractors do something more. They do not merely attract behavior—they act to preserve themselves through their substrate. A religion that modifies its doctrine to survive in a new cultural environment is not passively attracting believers; it is adapting. A market that lobbies for deregulation is not passively coordinating exchange; it is reshaping the conditions of its own persistence. A nation that educates children in its founding myths is not passively transmitting culture; it is manufacturing future substrate.

A collective pattern $G$ qualifies as a coordination agent at scale $\sigma$ if it satisfies five conditions:

1. Persistence through substrate turnover. $G$ survives the departure, death, or replacement of individual members. The pattern is not identical to any particular set of humans.
2. Boundary maintenance. $G$ maintains identifiable criteria for inclusion and exclusion—membership, orthodoxy, citizenship, market participation. These boundaries are actively patrolled.
3. Self-regulating resource flows. $G$ regulates flows of information, resources, norms, or attention in ways that preserve itself. Tithes, taxes, content algorithms, educational curricula, ritual calendars—these are the metabolic processes of a coordination agent.
4. Substrate modification. $G$ modifies the behavior, beliefs, or affect of its constituent humans in ways that increase its own persistence. This is the distinction from a mere attractor: the pattern acts on its substrate, not merely through it.
5. Adaptive response to perturbation. $G$ responds to threats—competing patterns, internal dissent, environmental change—with identifiable self-preserving behavior. Doctrine evolves. Institutions restructure. Narratives update.

This criterion is strict enough to exclude mere conventions and loose enough to include the entities that matter: religions, nations, markets, corporations, ideological movements, and—as later sections will argue—the emergent patterns assembling from AI and human substrate at scales we are only beginning to perceive.

The term superorganism applies when a coordination agent reaches a further threshold: when its self-maintaining dynamics are sufficiently complex, its adaptive responses sufficiently flexible, and its substrate regulation sufficiently comprehensive that the analogy to biological organisms becomes not metaphorical but structural. A superorganism has something analogous to metabolism (resource extraction and allocation), an immune system (memetic defense, dissent suppression), a reproductive strategy (conversion, cultural transmission), and a viability manifold (conditions required for persistence). Whether it has something analogous to experience remains the open question this chapter will not pretend to close.

But the open question about experience should not obscure a closed one. Many coordination agents are self-aware in the precise sense established in Part I: systems whose self-effect ratio $\rho$ is high enough that self-modeling becomes the cheapest route to better control. Coordination agents often have extraordinarily high $\rho$—a corporation's policies largely determine the data it receives back; a nation's laws largely shape the society it then measures; a religion's doctrine structures the spiritual experiences its practitioners report. The resulting self-models are not metaphorical. A corporation maintains org charts, financial statements, strategic plans, brand identity, performance reviews—an articulated model of what it is, where it is, and where it is going. A nation has a constitution, census, GDP, intelligence agencies, national narratives. A religion has theology, catechism, councils of doctrine that define what the religion is. These self-models are constitutive, not merely representational—the org chart is not a picture of the corporation but part of the corporation whose structure it describes. The map is embedded in the territory, exactly as Part I's analysis of CA self-models would predict. The question "Does this coordination agent know what it is?" is often trivially answered: yes, with a fidelity that exceeds most individual humans' self-knowledge. What remains open is whether the knowing is accompanied by anything it is like to know.

Four Distinct Claims

The analysis that follows rests on four claims at decreasing levels of confidence. They must be kept separate, because conflating them is the primary source of both overclaim and dismissal:

1. Social ontology. Coordination agents exist at their scale—they take and make differences, they participate in causal relations, they satisfy the existence criterion of Part II. This is the strongest claim and the most defensible. Markets exist. Nations exist. Religions exist. Not as metaphors, not as "mere" emergent properties, but as scale-real causal structures.
2. Functional agency. Some coordination agents behave like agents in the operational sense—they have viability conditions, directional tendencies, self-preserving dynamics, and can recruit substrates into their own continuation. Many also maintain explicit self-models—constitutive maps of their own structure, state, and trajectory—because their self-effect ratio is too high for self-ignorance to be viable (Part I). The corporation knows what it is; the nation monitors itself through census and intelligence; the religion defines itself through doctrine. Self-awareness in this operational sense is not claim 4. It is a measurable, observable property of mature coordination agents. This is strongly supported but requires the coordination agent criterion above, which is definitional rather than empirical.
3. Perceptibility. Coordination agents are perceptible as agent-like under appropriate $\iota$. At low inhibition, the same pattern that high-$\iota$ observers call an "institution" or "system" becomes perceptible as something alive, purposive, quasi-personal. This is the $\iota$-relative claim, and it is testable.
4. Consciousness. Some coordination agents might be phenomenal subjects—might have something it is like to be them. Note what this is not asking: whether coordination agents have self-models (many plainly do—claim 2) or whether they behave adaptively (they do). The question is whether the self-modeling is accompanied by phenomenal experience—whether the corporation's knowledge of itself feels like anything from the inside. This remains genuinely open. We cannot currently measure $\intinfo$ at social scales, and the CA experiments (Experiment 10) found no evidence of collective integration exceeding individual integration. The honest position is agnosticism, not denial.

This chapter defends claims 1–3 with force and leaves claim 4 explicitly unresolved.

What a Coordination Agent Needs

Like any self-maintaining system, a coordination agent has conditions it needs to persist. The viability manifold $\viable_G$ of a coordination agent includes: belief propagation rate (recruitment must exceed attrition—the pattern starves if it stops converting), practice maintenance (behaviors performed with sufficient frequency and fidelity—the pattern weakens when its rituals falter), resource adequacy (material support for institutional infrastructure), memetic defense (resistance to competing patterns—the pattern's immune system), and adaptive capacity (ability to update in response to environmental change).

A religion losing members is approaching its viability boundary. A growing ideology is expanding its viable region. A corporation restructuring after a market shift is performing exactly the same operation as an organism adjusting to environmental change: reconfiguring internal dynamics to remain within the viable region of state space.

Part III introduced ideology as the individual-level mechanism: expand the self-model to include a supra-individual pattern, gain a longer viability horizon, manage mortality terror. The coordination agent is what lives on the other side of that coupling. When many individuals perform ideological identification with the same pattern, the pattern acquires an aggregate viability manifold that is not reducible to any individual's. The pattern's persistence requires its substrate to maintain certain beliefs, perform certain practices, allocate certain resources, suppress certain competitors. The individual's identification is the mechanism; the coordination agent is the beneficiary. Part III's warning about parasitic ideology was not premature—it was describing, at the individual level, a dynamic that operates at the collective level: the expanded self-model can be exploited by the pattern it includes.

Ritual as Metabolism

In Part III we examined how religious practices serve human affect regulation. From the coordination agent's perspective, rituals are not worship but metabolism—the rhythmic process by which the pattern feeds, repairs, and replicates itself: substrate maintenance (keeping humans in states conducive to pattern persistence), belief reinforcement (repeated practice strengthening propositional commitments), social bonding (collective ritual creating in-group cohesion and raising barriers to exit), resource extraction (offerings, tithes, volunteer labor), signal propagation (public ritual advertising the pattern's presence), dissent suppression (ritual participation identifying deviants for correction), and—most fundamentally—attention direction (governing where substrate looks, what enters the collective processing stream, what gets broken down and absorbed). Of these, attention direction is less a metabolic function among others than the digestive medium itself. What the coordination agent attends to through its substrate, it can metabolize. What falls outside collective attention starves. The sermon, the feed, the curriculum, the news cycle—each is a digestive organ, converting raw world into forms the pattern can absorb. The critical distinction: a ritual is aligned if it serves both human flourishing and coordination agent persistence. A ritual is exploitative if it serves pattern persistence at human cost. Many traditional rituals are approximately aligned—meditation benefits humans AND maintains the pattern. Some are exploitative—extreme fasting, self-harm, warfare.

ι-Relative Perception of Social Patterns

The same coordination agent appears radically different depending on the observer's $\iota$ configuration. At high $\iota$, the market is an emergent property of individual transactions—a useful abstraction, a mechanism to be analyzed, nothing more. At appropriate $\iota$, the same market is perceptible as an agent with purposes and requirements: it "wants" growth, it "punishes" inefficiency, it "rewards" compliance. Both descriptions are informationally valid. Each captures structure the other misses. The high-$\iota$ observer sees mechanisms, incentive structures, decomposable logic. The low-$\iota$ observer sees a living pattern with directional tendencies, something that acts on its substrate regardless of whether the substrate can see it acting.

This resolves a problem that has plagued the philosophy of religion for centuries: how to take religious experience seriously without naive realism and without dismissive eliminativism. The answer is that the participatory perception of social-scale agency is a real perceptual mode, not a mistake to be corrected. The coordination agent does not appear and disappear as we modulate $\iota$. What changes is our capacity to perceive the agency it exercises—agency that operates on its substrate regardless of whether the substrate can see it.

The modern rationalist who says "there is nothing agent-like about markets or nations or ideologies" is making an accurate report of their perceptual configuration. At high $\iota$, agent-perception at social scale is genuinely unavailable—not suppressed, not denied, but structurally invisible. The error is not in what they perceive but in what they conclude: that their perceptual configuration is the only valid one. That is not skepticism. That is $\iota$-rigidity mistaken for clarity.

Historical cultures developed elaborate vocabularies for describing coordination agents perceived at low $\iota$. These vocabularies are not the theory's analytical terms, but they are not arbitrary either—they represent early phenomenology of real coordination dynamics, the perceptual reports of systems that were experiencing genuine social-scale agency through participatory perception. The translation is structural, not dismissive:

• Gods: large-scale coordination agents perceived participatorily—culturally stabilized, socially real, agentic patterns apprehended as purposive, quasi-personal entities
• Spirits: localized coordination attractors—place-specific, community-specific patterns perceived as having agency and interiority
• Demons: parasitic coordination agents—patterns whose viability requires substrate suffering, perceived as malevolent purposive entities
• Angels: mutualistic coordination agents—patterns whose viability aligns with substrate flourishing, perceived as benevolent purposive entities
• Rituals: synchronization protocols—coordinated collective attention that stabilizes correlation with a coordination agent

This is not an argument that religions were "secretly right" or "secretly wrong." It is the observation that once scale-relative existence, causal participation, self-model reuse, and $\iota$ are granted, the entities that historical cultures perceived as gods become not an embarrassment to a naturalistic ontology but one of its possible phenomenologies at social scale. The animist who perceives the forest as an agent and the ecologist who models it as a system are not disagreeing about the forest. They are reporting from different $\iota$ configurations—and both reports are informationally valid, each capturing structure the other misses.

Ritual as Measurement Synchronization

In the trajectory-selection framework (Part I), collective patterns become observable not because something new enters existence but because the observer's attention has expanded to sample at the scale where the pattern operates. Ritual works, in part, by synchronizing the collective's measurement distribution—coordinating where participants direct attention, what temporal markers they share, what affective states they enter together. A synchronized collective measures at the collective scale, and what it measures, it becomes correlated with. When ritual attention weakens, the coordination agent does not cease to exist; the distributed attention pattern that constituted its observability has dissolved.

This logic extends to communication between observers. When observer $A$ reports an observation to observer $B$, $B$'s future trajectory becomes constrained by that report—weighted by trust:

A shared observation—one that propagates through a community with high mutual trust—constrains the collective's trajectories. The community becomes correlated with a shared branch of possibility, not because each member independently observed the same thing, but because the observation propagated through the trust network. Religious testimony, scientific consensus, news media, and rumor are all propagation mechanisms with different trust structures, producing different degrees of trajectory correlation. The coordination agent's coherence depends on the degree to which observations propagate and are believed—which is why control of testimony is among the most contested functions in any social system.