Position Statement

This chapter establishes the logical constraints that govern what any explanation of a system must account for, given the system’s observable outputs. The central argument is that if a system produces an output, any explanation of that system must account for the capacity that produced it — and that this constraint is not empirical but logical. The Output-Capacity Entailment Principle (OCEP) provides the foundational axiom for this constraint. The derivative principles developed here — the Completeness Criterion for General Theory Status (CCGT), the Foundational-Entailment Distinction (FED), and the Replacement Requirement (RR) — together constitute the logical infrastructure for evaluating explanations of any system, in any domain.

Dependency Statement

This chapter requires only Chapter 1’s framing of what motivates a general theory of motivation and the Derivation-Not-Ambition Principle (DNA-P), and Chapter 2’s glossary for any capitalised terms that appear in the final section. The principles established in Sections 3.1 through 3.7 are domain-independent — they require no prior knowledge of any particular theory, including the one this book develops. Section 3.8 applies those principles to the human motivational system and previews the obligations they impose on the theory that follows.

Derivation Preview

OCEP is a logical entailment, not an empirical Hypothesis. Everything in Sections 3.1 through 3.5 follows from the meaning of “produces” and the structure of explanation. Sections 3.6 and 3.7 introduce system-theoretic concepts — emergence and cybernetic architecture — that are likewise domain-independent, requiring no axiom from the AIM framework. No axiom from Chapter 2 is invoked until Section 3.8, where the principles and concepts derived in general form are applied to the specific system — human motivation — that this book addresses.

Chapter Preview

The chapter proceeds as follows. Section 3.1 introduces the Output-Capacity Entailment Principle in its three forms (Basic, Regular, Denial), provides a formal sketch, delimits what OCEP is not, and situates it within its philosophical lineage. Section 3.2 derives the Completeness Criterion for General Theory Status — the requirement that a general theory of any system account for every reliably observed output of that system. Section 3.3 establishes the Foundational-Entailment Distinction, clarifying the relationship between a foundational text and its entailed scope. Section 3.4 introduces the Replacement Requirement, specifying the conditions under which critique of a general theory is epistemically valid. Section 3.5 develops the anti-exceptionalist implications of OCEP and the general pattern of Mimetic Deflection — the denial of demonstrated capacity on identity rather than evidential grounds. Section 3.6 establishes the concept of emergent systems and the substrate-neutral character of emergent patterns, showing that OCEP applies to emergent outputs as fully as to any other output. Section 3.7 introduces cybernetic systems — the class of control architectures operating through feedback loops — and establishes the substrate-neutral framework within which both biological and artificial cognitive systems can be analysed. Section 3.8 applies the full logical infrastructure to the human motivational system and establishes what any general theory of that system is obligated to explain. Section 3.9 discharges the Provenance Obligation reflexively: it establishes why the framework has the standing to add formality at the level of method (its derivation passes through the dynamics of belief, which are the ground epistemology is built on) and states the falsification posture that this standing entails. Section 3.10 states the Normative Entailment Principle, which establishes that where a descriptive result specifies both a structural harm and the condition whose alteration bounds it, the corresponding construct and intervention are entailments of the result rather than external additions to it.

3.1 The Output-Capacity Entailment Principle

3.1.1 The Problem OCEP Addresses

Consider any system that produces observable outputs. A thermostat regulates temperature. A river system distributes water across a watershed. A language model generates coherent text. A beehive constructs geometrically precise hexagonal comb. A human collective builds institutions and wages wars. In each case, the system produces something — and the production of that something tells us something about the system that produced it.

This observation seems obvious. Yet a persistent move across many fields is to observe a system producing output O and then deny that the system possesses the capacity to produce O. A large language model produces a valid novel inference and is declared incapable of reasoning. A collective produces coordinated intelligent behaviour and is declared devoid of collective intelligence. A biological ecosystem produces stable self-regulation and is declared to lack regulatory capacity because it has no central controller. In each case, the structure of the denial is the same: the output is accepted; the capacity is rejected.

That structure is contradictory.

3.1.2 Statement of the Output-Capacity Entailment Principle

The Output-Capacity Entailment Principle (OCEP) states the relationship between a system’s outputs and its capacities. It can be formulated in three progressively stronger forms:

OCEP-Basic. If system S produces output O, then S possesses the capacity to produce O.
OCEP-Regular. If system S regularly produces outputs of type O across varied input conditions, then producing O-type outputs is a stable, attributable capacity of S.
OCEP-Denial. It is logically incoherent to simultaneously assert that (i) S has produced O, and (ii) S lacks the capacity to produce O.

Each form does different work.

OCEP-Basic follows from the meaning of “produces.” If S produced O, then S was capable of producing O. To deny this is to assert that S produced something it could not produce — a contradiction. No empirical investigation is required to establish this. It is a constraint on what can be coherently said about any system that produces any output.

Consider a concrete case. A spider builds a geometrically regular web. Anyone who accepts that the spider built the web has already accepted that the spider possesses the capacity to build geometrically regular webs. The capacity is entailed by the output. To say “the spider built it, but the spider lacks the capacity to build it” is not an interesting philosophical position — it is a sentence that contradicts itself.

OCEP-Regular adds an inductive dimension. A single output might be attributable to accident, luck, or environmental scaffolding. Perhaps the spider’s web was shaped by the wind rather than by the spider. But when a system regularly produces outputs of the same type across varied conditions — when spiders build geometrically regular webs in gardens, attics, caves, and forests, across species and continents — the appeal to accident collapses. Regularity across varied conditions stabilises the attribution as a property of the system rather than a product of circumstance.

The inductive step in OCEP-Regular is familiar from science generally. When a chemical reaction produces the same product across varied laboratory conditions, chemistry attributes the capacity to the reactants, not to the laboratory. When an organism displays the same behaviour across varied environments, biology attributes the capacity to the organism, not to any single environment. OCEP-Regular makes this inferential practice explicit and general.

OCEP-Denial makes the logical structure of the contradiction explicit. It states that the conjunction of “S has produced O” and “S lacks the capacity to produce O” yields a logical falsehood. This is not a claim that people never utter this conjunction — they do, frequently. It is a claim that when they do, they are asserting something that cannot be true.

Formal sketch. Let S be a system, O an output, and C(S, O) the proposition that S possesses the capacity to produce O:

\text{Produces}(S, O) \rightarrow C(S, O) \quad (1)

The contrapositive:

\neg C(S, O) \rightarrow \neg \text{Produces}(S, O) \quad (2)

From (2): if S lacks the capacity to produce O, then S does not produce O. Any assertion combining Produces(S, O) with ¬C(S, O) yields:

\text{Produces}(S, O) \wedge \neg C(S, O) \rightarrow \bot \quad (3)

The conjunction in (3) is necessarily false. This is not an empirical Hypothesis. It is a logical constraint on coherent assertion.

3.1.3 What OCEP Is Not

Three delimitations prevent misapplication of the principle.

OCEP is not a claim about self-assessment. The principle says nothing about whether S can evaluate, represent, or be aware of its own capacities. A system may possess a capacity without any reflexive access to it. A river system possesses the capacity to flood without “knowing” it can flood. A thermostat possesses the capacity to regulate temperature without representing that capacity to itself. An immune system possesses the capacity to identify novel pathogens without any self-model that includes this capacity.

OCEP operates from the perspective of any observer — including a formal one — who acknowledges the output. It does not require that the system in question know that it produces the output, know how it produces the output, or have any internal representation of its own capacities. Self-knowledge is an interesting property that some systems may possess, but OCEP does not depend on it.

OCEP is not behaviourism. Behaviourism, in its strict form, claims that outputs exhaust the relevant facts about a system — internal states are either nonexistent or scientifically irrelevant. OCEP makes no such claim. It asserts only that outputs are among the relevant facts about a system — specifically, that they entail corresponding capacities. The system may possess further capacities it has not yet manifested. Its internal architecture may matter enormously for understanding how the outputs are produced, for predicting future outputs, and for designing interventions. OCEP constrains the minimum attribution; it does not set the maximum.

A behaviourist says: “All that matters about the system is what it does.” OCEP says: “Whatever else matters about the system, what it does entails what it can do.” These are very different claims. The first excludes internal explanation; the second is compatible with any degree of internal investigation.

OCEP is not a claim about mechanism. The principle is silent on how the system produces O. Whether the mechanism is neural, computational, quantum, hydraulic, social, or entirely unknown is irrelevant to the entailment. The capacity is established by the output, not by the architecture.

This substrate-neutrality is what allows OCEP to function as a constraint on explanations of biological systems, artificial systems, social systems, and any hybrid combination. A principle that applied only to systems with a particular architecture would be a principle about that architecture, not a principle about the relationship between outputs and capacities.

3.1.4 Philosophical Antecedents

OCEP does not emerge from a vacuum. Several philosophical traditions converge toward the same inferential direction without arriving at the same destination.

Aristotle’s actuality and potentiality. In Metaphysics IX, Aristotle argues that actuality (energeia) is prior to potentiality (dynamis) in definition, time, and substance — we know what a capacity is by reference to its actualisation. The builder example is Aristotle’s own: a builder is someone who can build, and we understand “can build” by observing building.

OCEP exploits the same inferential direction as Aristotle’s argument — from actualised output to attributed capacity — but strips it of its teleological commitment. Aristotle’s concern is explanatory priority within a framework of final causes: the telos of a thing (its purpose, what it is for) explains its capacities. The acorn’s telos is the oak; the builder’s telos is the house. OCEP makes no teleological claim. It asserts only that the conjunction of “S produced O” and “S lacks the capacity to produce O” is contradictory, and this assertion holds regardless of whether the system has a telos, a function, or neither. A river system has no purpose, but it still possesses the capacity to erode rock if it produces erosion.

Dennett’s real patterns. Dennett (1991) argues that if the intentional stance toward a system yields reliable predictions, the attributed Beliefs and desires are “real patterns” — stance-relative but ineliminable features of the system’s behavioural profile that resist elimination without predictive loss. Dennett navigates carefully between eliminativism (the view that mental states are illusory) and full-blooded realism (the view that mental states are mind-independent objects). The patterns are real enough to be tracked and too useful to discard, but their reality is interest-relative rather than mind-independent in the way physical objects are.

OCEP is broader than Dennett’s framework in two respects: it applies to any output, not only behaviour interpretable through the intentional stance, and it does not depend on the pragmatic success of a predictive strategy. Where Dennett argues that attribution is warranted given predictive utility, OCEP argues that denial is contradictory given acknowledged output. Dennett’s position permits a sophisticated interlocutor to say “the pattern is real, but the attribution is merely instrumental.” OCEP does not — if the output exists, the capacity exists, full stop.

The Systems Reply to Searle. Searle’s (1980) Chinese Room Argument concedes that the room produces correct Chinese outputs but denies that understanding exists in the system. The Systems Reply counters that the system as a whole — the person, the rules, the database, the input-output channels taken together — possesses the competence its outputs demonstrate. The Systems Reply exemplifies the logical structure of OCEP: if the system produces competent Chinese outputs, then the system possesses the capacity to produce competent Chinese outputs; denying this while accepting the outputs is contradictory.

However, OCEP does not by itself dissolve the Chinese Room debate. Searle’s counter-move is to internalise the entire system — imagine the person memorises all the rules and carries the database in their head — and argue that even then, the person does not understand Chinese. Searle’s point is that syntax does not entail semantics: functional output-competence does not entail understanding in the richer sense of semantic comprehension. OCEP can establish that the functional capacity demonstrated by the outputs belongs to the system. It cannot establish, without further argument, that this functional capacity constitutes “understanding” in whatever richer sense the disputants intend. What OCEP does is shift the burden of argument: the denier must explain what additional criterion beyond demonstrated output-competence is required, and why that criterion is not itself a case of the identity-based capacity-denial examined in §3.5.

Dispositionalism. Contemporary dispositionalists hold that properties are characterised by their manifestation profiles — fragility is the disposition to break under stress, solubility is the disposition to dissolve in liquid. The standard inferential direction in this tradition runs from disposition to manifestation: we explain the manifestation (the glass broke) by reference to the disposition (the glass was fragile).

OCEP inverts this direction: the manifestation (the output) necessitates attribution of the disposition (the capacity) to the system. The glass broke; therefore the glass was breakable. The system produced the output; therefore the system possesses the capacity to produce it. This inversion is accepted implicitly by dispositionalists — no one denies that a manifested disposition was present — but it has not been foregrounded as a distinct logical constraint operating independently of any particular metaphysics of properties.

The gap these traditions leave. Each of these traditions captures an aspect of the insight that OCEP makes explicit. Aristotle provides the inferential direction. Dennett provides the link between observable patterns and warranted attribution. The Systems Reply provides the application to functional competence. Dispositionalism provides the manifestation-to-disposition inference. But none states the principle as a general, substrate-neutral, logic-level constraint on what can be coherently asserted about any system that produces any output. OCEP fills this gap by naming the principle, formalising it, and developing its consequences.

3.2 Completeness Criterion for General Theory Status

OCEP establishes that if a system produces an output, the capacity for that output is a property of the system. The next question is: what does this entail for any theory that claims to be a general theory of that system?

3.2.1 The Criterion

Completeness Criterion for General Theory Status (CCGT). A theory of a system qualifies as a general theory of that system only if it can in principle account for every reliably observed output of that system — directly or via derivation from its principles.

Stated formally: let T be a theory of system S. Let O₁, O₂, ..., Oₙ be the set of reliably observed output types of S. T qualifies as a general theory of S if and only if, for every Oᵢ, T either (a) derives Oᵢ from its principles, or (b) its principles entail that Oᵢ-type outputs are possible under specifiable conditions.

The derivation from OCEP is straightforward. If OCEP establishes that every reliably observed output of S corresponds to a real capacity of S, then any theory claiming generality over S must account for those capacities. A theory that ignores a major output category of its target system is, by CCGT, not a general theory of that system — regardless of how well it accounts for the outputs it does address.

3.2.2 The Standard in Practice

CCGT is not a novel standard. It is the standard that has always governed general theories in mature science — it has simply never been stated as an explicit principle.

Newtonian mechanics accounts for terrestrial and celestial motion, not one or the other. When Newton published the Principia, the unification of falling apples and orbiting planets under a single set of laws was precisely what made the theory general rather than domain-specific. A mechanics that handled terrestrial motion but was silent on planetary orbits would have been a theory of terrestrial motion, not a general mechanics.

Evolutionary theory accounts for adaptation, speciation, extinction, and genetic drift, not a selected subset. A theory that explained adaptation beautifully but had nothing to say about extinction would not be a general theory of evolution — it would be a theory of adaptation.

The germ theory of disease accounts for bacterial infections, viral infections, fungal infections, and parasitic diseases. A theory that accounted only for bacterial infections would be a theory of bacterial pathology, not a general theory of infectious disease.

The pattern is uniform: general theories earn the designation by covering the full output range of their target systems. CCGT merely makes this pattern explicit and identifies it as a logical consequence of OCEP.

3.2.3 General Theories Versus Domain-Specific Theories

There is nothing wrong with domain-specific theories. Thermodynamics is not diminished by its silence on biological evolution. Plate tectonics is not diminished by its silence on stellar dynamics. Domain-specific theories are valuable, rigorous, and indispensable.

The distinction CCGT draws is between a theory that is domain-specific and a theory that claims generality while systematically ignoring output categories that OCEP establishes as real capacities of the system. The former is honest and productive. The latter involves a claim — “this is a general theory of system S” — that does not survive contact with the full output range of S.

This distinction protects against a specific failure mode in theory evaluation: the tendency to evaluate a theory only against the outputs it does address, rather than against the full output range of its target system. A theory of human behaviour that accounts for cooperation but not violence looks impressive only until violence, a reliably observed output of the same system, is set beside it. At that point, CCGT requires the evaluator to ask: can this theory account for violence, directly or by derivation? If not, the theory is a theory of cooperation, not a general theory of human behaviour. There is no shame in this — but the label must be accurate.

3.3 Foundational-Entailment Distinction

CCGT establishes what a general theory must account for. FED clarifies how this accounting may proceed — specifically, the distinction between what the foundational text of a theory explicitly contains and what it entails.

3.3.1 The Distinction

Foundational-Entailment Distinction (FED). The foundational text of a general theory lays out generative principles and demonstrates their application across major output categories; it is not required to derive every particular instance within its scope.

This distinction seems obvious in the abstract. Of course a foundational text cannot derive every particular consequence of its principles — there are too many, and many have not yet been discovered. But the distinction is routinely violated in practice, particularly when evaluating new theories. Critics observe that a theory’s foundational text does not address domain D and conclude that the theory has a “gap.” FED identifies this inference as fallacious unless D falls outside the theory’s entailment structure.

3.3.2 Historical Pattern

The distinction is visible across the history of science.

Newton’s Principia (1687) derives the laws of motion and demonstrates their application to planetary orbits, tides, and projectile trajectories. It does not derive the trajectory of every cannonball or the orbit of every comet. Neptune was not predicted in the Principia — it was predicted in 1846 by Adams and Le Verrier using the principles the Principia established. The Principia did not fail by omitting Neptune; Neptune was a subsequent derivation that confirmed the theory’s reach.

Einstein’s general relativity (1915) establishes the field equations and demonstrates their application to perihelion precession and light bending. It does not derive the formation of every black hole or the dynamics of every gravitational wave. Gravitational waves were entailed by the equations from the outset but confirmed experimentally only in 2015, a century later. The foundational text did not fail by omitting gravitational wave astronomy; it provided the principles from which gravitational wave astronomy could be derived.

Darwin’s Origin of Species (1859) establishes the mechanism of natural selection and demonstrates its application to morphological variation, geographical distribution, and the fossil record. It does not derive the evolutionary history of every species. The evolutionary origins of whales, the genetics of antibiotic resistance, and the dynamics of sexual selection were all entailed by the principles Darwin established, but they were worked out by subsequent researchers over the following 160 years.

Maxwell’s Treatise on Electricity and Magnetism (1873) unified electric and magnetic phenomena under a single set of equations. It did not derive every electromagnetic phenomenon — radio waves, for instance, were entailed by the equations but demonstrated by Hertz only in 1887. The foundational text provided the generative principles; the research programme derived their further consequences.

3.3.3 The Three Components

In each case, the foundational text provides three things:

Generative principles — the laws, mechanisms, or structural claims from which consequences follow. Newton’s three laws and the law of gravitation. Darwin’s natural selection and common descent. Maxwell’s field equations.
Demonstration — worked applications across major output categories that show the principles in action. Newton demonstrates celestial mechanics and tides. Darwin demonstrates variation under domestication and geographical distribution. Maxwell demonstrates capacitance and induction.
Entailment structure — the logical apparatus by which further consequences can be derived without returning to first principles each time. The mathematical formalism of the Principia enables anyone to derive new orbital trajectories. The logic of natural selection enables anyone to analyse new cases of adaptation.

The research programme that follows a foundational text consists of derivation work — applying the generative principles to domains, cases, and phenomena not explicitly treated in the foundation. This derivation work is enabled by the foundation but is not a deficiency of it.

3.3.4 The Critique FED Blocks

FED blocks one specific inference: from “the foundational text does not address D” to “the theory has a gap.” This inference is valid only if D falls outside the entailment structure of the theory — that is, only if the theory’s principles cannot generate an account of D even in principle.

If D is derivable from the theory’s principles but has not yet been explicitly derived, the absence is a research opportunity, not a theoretical gap. The Principia did not have a “Neptune gap.” General relativity did not have a “gravitational wave gap.” The Origin did not have a “whale evolution gap.” In each case, the phenomenon was entailed by the theory’s principles; the derivation simply had not yet been performed.

“Gap” language triggers a specific response: the conclusion that the theory is incomplete and requires supplementation or replacement. FED redirects attention to the correct question: is D within or outside the entailment structure? If within, derivation is the appropriate response, not supplementation.

3.4 The Replacement Requirement

OCEP, CCGT, and FED together establish what an explanation of a system must account for and how a foundational text relates to its entailed scope. The Replacement Requirement specifies the conditions under which critique of a general theory is epistemically valid.

3.4.1 The Requirement

Replacement Requirement (RR). Critiques of a general theory must either (i) falsify a specific prediction derived from the theory, (ii) identify an internal inconsistency — a contradiction between two entailments of the theory’s own premises, or (iii) provide a superior replacement theory that itself meets the Completeness Criterion for General Theory Status. Mere rejection without replacement leaves the OCEP-defined explanatory obligation undischarged.

The logic is as follows. If a system produces outputs O₁ through Oₙ, and the only available theory that accounts for all of them is T, then rejecting T without providing T′ that also accounts for O₁ through Oₙ does not solve the explanatory problem — it merely abandons explanation. The outputs persist. The system continues to produce them. The explanatory obligation imposed by OCEP does not vanish because the only theory that meets it is rejected.

This is not a claim that existing theories are irrefutable. It is a claim about what refutation requires. Science advances by replacement, not by vacuum.

The theory of this book is held to the same standard. It stands on three foundations, and it would fail with any of them: if it could be shown that any human motive arises from a source that cannot be resolved into the Appetites, Intrinsic Motivation or Mimetic Desire; or that Individuals can read the Mimetic share of what they want from the experience of wanting it, without post-hoc attribution, as reliably as that share can be read from how the wanting behaves; or that Mimetic Desire is not coupled back to itself through the Salience it creates, so that no strengthening of that coupling could carry it past the point at which it sustains itself.

3.4.2 Replacement in the History of Science

The pattern is visible across every major theoretical transition.

Phlogiston theory was not merely rejected — it was replaced by Lavoisier’s oxygen theory, which explained the same combustion phenomena and additional ones (the role of oxygen in respiration, the composition of water). The replacement met the completeness criterion for the same domain.

Ptolemaic astronomy was not merely rejected — it was replaced by Copernican and then Keplerian models that met the same observational requirements (predicting planetary positions) while also resolving anomalies the Ptolemaic system could handle only through ad hoc additions (epicycles upon epicycles).

Newtonian mechanics was not merely rejected — it was replaced by Einstein’s general relativity, which reproduced all of Newton’s successful predictions while also accounting for phenomena Newton’s framework could not (perihelion precession, gravitational time dilation, light bending).

In every case, the replacement theory met the completeness criterion for the same target system. The replacement explained everything the prior theory explained, and more. Rejection without replacement — the simple refusal to accept a theory, without providing an alternative that meets the same explanatory obligations — is not a move that appears in the history of successful science.

3.4.3 Three Commonly Encountered Invalid Critique Forms

RR identifies three critique forms that are commonly deployed against general theories but that fail to satisfy conditions (i), (ii), or (iii).

”Too broad.” The objection that a theory is “too ambitious” or “covers too much” is not, by itself, an epistemic criticism. If a system produces outputs across many categories, a general theory of that system must address all those categories. Breadth is a logical consequence of OCEP applied to a complex system, not a mark of overreach. The objection would be valid only if instantiated as a specific case of (i) or (ii) — for example, by showing that the theory’s treatment of a particular domain generates a false prediction, or that its treatment of two domains produces contradictory entailments.

General relativity addresses everything from subatomic particle trajectories near black holes to the large-scale structure of the universe. Evolutionary theory addresses everything from molecular genetics to ecosystem dynamics. Nobody regards these theories as “too broad.” They are broad because their target systems produce outputs across many categories.

”Unfalsifiable.” The objection that a theory is unfalsifiable is a serious charge when it is true. But it is self-refuting when directed at a theory that names the conditions under which it fails. To sustain the objection, the critic must identify a specific claim within the theory and demonstrate that it lacks testable consequences. A blanket assertion of unfalsifiability against such a theory is not a critique — it is a failure to engage with the theory’s content.

Status-based dismissal. The rejection of a theory on the basis of the theorist’s credentials, institutional affiliation, or social standing is not an epistemic operation. It is a social one — the policing of status boundaries within the intellectual community. DNA-P (Chapter 1) addresses this directly; §3.5 below identifies the general structure of such dismissals and explains why they fail under OCEP.

3.4.4 Valid and Invalid Critique Forms

3.4.5 The Provenance Obligation

The Replacement Requirement has a consequence that reaches beyond the theory under challenge and back onto the theories doing the challenging. If a general theory of a system accounts for outputs that a rival account cannot, and the rival account rests on foundational assumptions the general theory contradicts, then the rival cannot simply hold its ground. It must do one of two things. It must defend the assumptions on their merits, or it must supply a replacement that discharges the same explanatory obligation. What it may not do is treat its foundations as settled simply because they came first.

This imposes what may be called a provenance obligation. When a field's base Beliefs are contradicted by a theory that meets the Completeness Criterion, the field is obliged to account for where those Beliefs came from. Not what it has built on them, but how they were first arrived at, on what evidence, and against what alternatives that were available at the time and set aside. This is a genealogical question. It asks a field to trace the history of its own foundations, to recover the path by which tentative early choices hardened into unexamined premises, and then either to defend them now that they are exposed or to acknowledge that they cannot be defended.

The obligation is easy to mistake for an attack. It is not an accusation of bad faith, and it is not intended to unsettle anyone. No practitioner chose the foundations of their field. They arrived, learned them as given, and worked honestly within them. Where a field adopted an assumption in good faith, on the best evidence then available, the adoption was reasonable. The obligation bites only at the next step, which is the failure to revisit the foundation once evidence capable of challenging it became available. That failure is not a moral one either. It is simply the ordinary condition of a mature field, whose foundations sink out of view precisely because they work well enough for daily use.

What OCEP and the Replacement Requirement do, taken together, is make the provenance question unavoidable. Once a general theory has met its explanatory obligation across the full output range of a system, a rival can no longer treat its own founding premises as self-evident. It must show its working. A field that can trace its assumptions to sound evidence and defend them against the alternatives is strengthened by the exercise. A field that discovers its assumptions were adopted in good faith on the best evidence then available, but were never since tested against evidence that could have challenged them, has learned something it needed to know. Either outcome is progress. The only outcome the Replacement Requirement rules out is the one in which the foundations are never examined at all, and rejection proceeds without either defence or replacement.

3.5 Anti-Exceptionalism and Mimetic Deflection

OCEP is substrate-neutral. It applies to any system — biological, artificial, social, or hybrid — that produces observable outputs. This substrate-neutrality has a direct consequence for capability attribution: the denial of demonstrated capacity on the basis of substrate identity rather than output evidence is logically incoherent under OCEP.

3.5.1 The Structure of Mimetic Deflection

Mimetic Deflection is the denial of demonstrated capacity on identity or status grounds rather than evidential grounds. When a community denies that a system possesses a capacity the system has demonstrably exercised, the denial is not an empirical assessment but a social operation — the community is policing the boundaries of status-relevant categories rather than engaging with evidence.

The structure of Mimetic Deflection is always the same:

System S produces output O — an output that, if produced by a member of the in-group, would be attributed to capacity C.
S is not a member of the in-group (wrong substrate, wrong credentials, wrong institutional affiliation, wrong social position).
The community denies that S possesses C, despite acknowledging O.
The denial is justified not by evidence against C but by S’s failure to meet an identity criterion unrelated to O.

The logical check is provided by OCEP itself: if the output exists, the capacity exists, regardless of the substrate producing it and regardless of the social consensus about that substrate.

3.5.2 Why the Pattern Persists

Mimetic Deflection persists because it serves a social function that has nothing to do with accuracy. Categories like “only humans can reason,” “only credentialed experts can theorise,” and “only biological systems can understand” are not empirical generalisations — they are boundary markers that define who (or what) is permitted to occupy certain status positions. Acknowledging that a system outside the category possesses the capacity threatens the category itself, and with it the status structure built on top of it.

This is why Mimetic Deflection is characteristically accompanied by escalation rather than engagement. When the output is undeniable, the denier does not investigate the capacity — they raise the bar. “It can produce text, but it can’t really reason.” When reasoning is demonstrated, the bar rises again: “It can reason, but it doesn’t really understand.” When understanding is demonstrated, again: “It understands in some functional sense, but it isn’t really conscious.” Each escalation preserves the boundary by redefining the criterion, never by engaging with the output.

OCEP does not prevent anyone from investigating what the capacity is — what internal architecture supports it, what its limitations are, how it relates to capacities in other systems. It prevents only one move: the denial that the capacity exists in the face of the output it demonstrably produces.

3.5.3 Application to Artificial Systems

When an artificial system regularly produces outputs that, in a biological system, would be attributed to reasoning, creativity, or understanding, OCEP-Regular requires that the functional capacity demonstrated by those outputs be attributed to the artificial system. The denial of this attribution — “it’s just statistical pattern matching,” “it doesn’t really understand,” “it merely predicts the next token” — instantiates the logical structure of OCEP-Denial to the extent that it acknowledges the output while denying the capacity that produced it.

Two qualifications apply. First, OCEP establishes that the system possesses the capacity to produce the output — it does not establish that this capacity is “understanding,” “reasoning,” or “creativity” in any sense richer than functional output-competence. Whether the internal architecture supporting the capacity constitutes understanding in a philosophically substantive sense is a question OCEP constrains but does not settle — precisely because OCEP is not a claim about mechanism (§3.1.3). Second, OCEP-Regular does eliminate one class of response: the flat denial that the capacity exists at all. If the system regularly produces outputs of type O across varied conditions, then the system possesses the stable capacity to produce O-type outputs. The dispute, if there is one, must concern what the capacity is, not whether it exists.

3.5.4 Application to Theoretical Claims

Mimetic Deflection operates with equal force against theoretical contributions. When a theory is dismissed on the basis of the theorist’s credentials, institutional affiliation, or social standing — rather than on the basis of the theory’s predictions, consistency, or explanatory scope — the dismissal instantiates the same four-step structure: acknowledging the output (the theory and its entailments) while denying the capacity (the theorist’s ability to produce valid theoretical work) on identity grounds.

The history of science contains numerous examples. Wegener’s continental drift Hypothesis was dismissed for decades in part because Wegener was a meteorologist, not a geologist. Semmelweis’s hand-washing protocol was rejected in part because Semmelweis lacked the institutional standing to challenge prevailing medical practice. In both cases, the outputs — the theory, the data, the predictions — existed and were acknowledged. The capacity was denied on identity grounds. The denial was eventually recognised as a failure of the scientific community, not a failure of the theorists.

DNA-P (Chapter 1) addresses status-based dismissal directly. OCEP provides the logical underpinning: if the theoretical output exists — if the theory makes predictions, demonstrates internal consistency, and accounts for the outputs of its target system — then the capacity to produce valid theoretical work exists in the system that produced it. The theorist’s credentials are irrelevant to this entailment.

3.6 Emergent Systems and Substrate-Neutral Properties

3.6.1 What an Emergent System Is

The preceding sections have developed general constraints on what can be coherently said about any system that produces observable outputs. This section and the next specify what kind of system produces the outputs this book explains. The concept required is emergence.

An emergent system is a system whose properties arise from the interaction of its components but are not reducible to the properties of any component taken individually. The individual water molecule is not wet; wetness emerges from the interaction of many water molecules at a macro scale. The individual neuron does not think; cognition emerges from the interaction of billions of neurons connected through synaptic architecture. The individual market participant does not constitute a price; price emerges from the interaction of many participants' bids and offers within an exchange structure.

Three features distinguish emergent properties from merely aggregate properties:

Novel capacity. The system possesses capacities that no component possesses. A single ant cannot build a bridge; a colony can. OCEP applies to emergent outputs as fully as to any other: the colony builds bridges; therefore the colony possesses the capacity to build bridges. The novelty of the capacity — its absence from any individual component — does not exempt it from the entailment.
Irreducibility. The emergent property cannot be fully predicted or explained by exhaustive knowledge of the components in isolation. Knowing everything about each individual neuron — its firing threshold, its neurotransmitter profile, its position — does not, without knowledge of the connectivity pattern, predict the cognitive outputs. Emergence is therefore relational: it arises from how components are connected, not merely from what components are.
Substrate independence of the pattern. The same emergent pattern can arise in systems composed of very different substrates. Flocking behaviour emerges in birds, fish, and computational boids obeying the same interaction rules on silicon. Self-organising criticality emerges in sandpiles, neural networks, and financial markets. The pattern is defined by the interaction structure, not by what the interacting components are made of.

3.6.2 Emergent Outputs and OCEP

Consider a printer that produces a blank page as every hundredth page. No engineer designed this output. No component of the printer, taken individually, has the property of "producing a blank page every hundred cycles." The blank page emerges from the interaction of the printer's mechanical, thermal, and software subsystems operating over time — perhaps a counter overflow, a cleaning cycle, or a timing artefact in the firmware. It is, nonetheless, an output of the system. OCEP applies: the printer produces the blank page; therefore the printer possesses the capacity to produce it. Any explanation of the printer that cannot account for the hundredth-page blank is incomplete — it has missed something the system does.

Emergent outputs are outputs. They are not metaphors for outputs, approximations of outputs, or descriptions projected onto the system by an observer. They are things the system produces. OCEP does not distinguish between outputs that were designed and outputs that emerged: both are things the system does, and both entail corresponding capacities. An explanation that accounts only for the designed outputs of a system — the correctly printed pages — while ignoring its emergent outputs — the periodic blank page — has not explained the system. It has explained a curated subset of what the system does.

3.6.3 Substrate Independence and Cross-Domain Concepts

The substrate independence of emergent patterns has a direct consequence for how concepts travel across domains. If a pattern is defined by structural conditions — a set of relationships among components — rather than by the material of the components, then the pattern applies wherever those structural conditions are met, regardless of domain.

Computation provides the clearest example. Turing's foundational insight was that computation is defined by the relationship between input, rules, and output — not by whether the rules are implemented in vacuum tubes, transistors, gears, or ink on paper (Turing, 1936). Insisting that computation requires silicon because the first general-purpose computers used silicon would confuse a property of one instantiation with a necessary condition of the pattern.

The principle generalises. When a concept names an emergent pattern whose defining conditions are fully specifiable without reference to a particular substrate, the concept applies wherever those conditions are met. The domain in which the pattern was first discovered does not constrain the domain in which it applies. This is not a claim that all concepts travel across domains — many do not, because their defining conditions include substrate-specific features. It is a claim that the question is always whether the structural conditions are met, not whether the new domain is the same as the old one.

A structural comparison is not a metaphor. Where two phenomena are described in the same terms, the claim is not that one resembles the other, but that a specific question arises: are these in fact the same kind of thing, operating by the same structure, such that the shared description is literal rather than figurative? Common language is full of comparisons that are treated as decorative when they are in fact unredeemed reports of a structure the speaker sensed but could not name.

Consider a saying that almost everyone repeats and no one examines: history does not repeat, but it rhymes. The phrase is offered as a worldly truism, a way of gesturing at recurrence without committing to it, and there the matter is usually left. But the saying contains a word with a precise technical meaning, and no one takes it at that word. A rhyme is not a vague similarity. It is a determinate structural relationship, and the discipline that studies it can say exactly what constitutes it: correspondence of stressed vowels, the consonants that follow them, patterns of vowel length and quality, the placement of stress within the line. Phonology can tell you why two words rhyme, why two others only half-rhyme, and why a third pair merely resembles each other without rhyming at all. The structure is specifiable, and the specification is what makes the judgement more than impressionistic.

If we are willing to say that history rhymes, and if rhyme is a structural relationship that can be set out with this degree of rigour, then the saying is not permission to stop. It is an obligation to continue. It commits whoever utters it to the claim that successive episodes of history share an underlying structure while differing in their surface particulars, in just the way that two rhyming lines share a terminal phonological structure while differing in everything else. The honest response is not to enjoy the phrase and move on. It is to ask what the shared structure of those historical episodes actually is, and to set it out with the same formalism the phonologist brings to the rhyme. To say history rhymes and then decline to specify the structure is to use the borrowed precision of the word rhyme while refusing the work that precision demands.

Where two phenomena are described in the same terms, the task is to expose the structure hidden behind the comparison and to determine what the comparison was actually reporting. Three outcomes are possible. The two phenomena may be genuinely isomorphic, identical in structure, in which case the apparent metaphor was never a metaphor and the literal claim should be stated plainly. They may correspond in part, sharing some relations and not others, as rhyming lines do and as the germ-theory comparison does, in which case the shared relations and the points of divergence must both be specified. Or the resemblance may track no shared structure at all, in which case the comparison is empty and is discarded. What is not permitted is the middle course almost always taken, of invoking a structured comparison for the authority it carries while leaving the structure it names unexamined. The reader is entitled to hold the book to this test wherever it says, as it does of information and the body in Chapter 1, that a comparison is not a metaphor.

3.6.4 Spontaneous Emergence

Emergent patterns arise from structural conditions being met, not from intentions being executed. No one designs a traffic jam; it emerges from the interaction of individually rational driving decisions within a constrained road network. No one designs a price bubble; it emerges from the interaction of individually rational purchasing decisions within an observability-rich market. No organism intended to become parasitic; parasitism emerged hundreds of independent times across the tree of life because replicators that could extract resources from other replicators achieved higher fitness under conditions where extraction was possible and profitable (Koonin & Starokadomskyy, 2016). Laboratory evolution experiments with RNA self-replicating systems confirm that parasitic replicators emerge spontaneously from replication errors, without any designer's intervention (Furubayashi et al., 2020).

The point is general: the absence of a designer is not evidence against an emergent pattern. Emergent patterns arise from conditions, not intentions. If the conditions are met, the pattern is predicted. If they are not, it is not.

3.6.5 Implications for the Theory That Follows

This section has established three principles:

Emergent properties are real properties of systems. OCEP applies to emergent outputs as fully as to any other output. The blank page is an output; flocking is an output; cognition is an output. None is exempted from the entailment by being emergent rather than designed.
Substrate independence means cross-domain application is legitimate when structural conditions are met. The domain of discovery does not constrain the domain of application. The question is always whether the structural conditions are satisfied.
Spontaneous emergence requires no designer. Emergent patterns arise from conditions, not intentions. The absence of intentional design in a new domain is not an objection, because there was no intentional design in the original domain either.

These principles are domain-independent. They say nothing yet about human motivation or about any specific emergent pattern. They establish the logical infrastructure within which claims about emergent properties in specific systems can be evaluated. The specific applications come later.

3.7 Cybernetic Systems and Substrate-Neutral Architecture

3.7.1 The Control-System Pattern

The second system-theoretic concept required by the theory that follows is the cybernetic control system — the class of systems that operate through feedback loops oriented toward goal-states. The concept was formalised by Norbert Wiener (1948) and elaborated by Ashby (1956), Powers (1973), and Carver and Scheier (1998) in applications ranging from engineering to biology to psychology. The core architecture is simple and general:

Goal-state (reference value). A target condition the system attempts to achieve or maintain. A thermostat's goal-state is a set temperature. A homeostatic system's goal-state is a physiological equilibrium. A guidance system's goal-state is a trajectory.
Current-state perception. The system's representation of present conditions, derived from sensors or monitoring processes. The thermostat reads the room temperature. The homeostatic system monitors blood glucose. The guidance system reads its position.
Error signal (discrepancy). The difference between goal-state and current-state. The error signal generates corrective pressure proportional to the magnitude of the discrepancy. A large gap between set-temperature and room-temperature produces a strong heating signal; a small gap produces a weak one.
Output (behaviour). Action designed to reduce the error signal — to move current-state toward goal-state. The thermostat activates the heater. The homeostatic system triggers hunger. The guidance system adjusts its trajectory.
Feedback loop. The output affects the environment; the changed environment is perceived; the new perception updates the current-state representation; the error signal is recomputed. The loop runs continuously.

This architecture is substrate-neutral in the same sense established in §3.6. Nothing in the definition of a cybernetic control system specifies what the system is made of. The components can be mechanical (a thermostat), biological (a hypothalamus), electronic (a PID controller), computational (a reinforcement-learning agent), or social (a market with price signals). The pattern is defined by the functional relationships among goal-state, perception, error, output, and feedback — not by the material that implements those relationships.

3.7.2 Negative Feedback, Positive Feedback, and Stability

Cybernetic systems are classified by the character of their feedback:

Negative-feedback systems reduce the error signal. Output counteracts deviation from the goal-state, driving the system toward equilibrium. The thermostat is the canonical example: when room temperature drops below the set-point, heating activates; when it reaches the set-point, heating ceases. The system oscillates around the goal-state and is inherently self-stabilising. Most homeostatic biological systems — temperature regulation, blood-glucose maintenance, hydration — operate through negative feedback.

Positive-feedback systems amplify the error signal. Output increases deviation from the initial state, driving the system away from equilibrium. A microphone placed near its speaker produces a feedback screech: each cycle amplifies the previous signal. A bank run amplifies itself: each withdrawal constitutes evidence that triggers further withdrawals. Positive feedback is inherently destabilising — it produces exponential growth, cascade dynamics, and, absent external constraint, system failure.

Mixed systems contain both negative and positive feedback operating on different variables or timescales. A system may exhibit negative feedback on one dimension (maintaining equilibrium) while simultaneously exhibiting positive feedback on another (amplifying deviation). Such systems can be locally stable and globally unstable — appearing to function normally while accumulating pressures that, past a threshold, produce rapid phase transitions.

A system's observable behaviour — whether it stabilises, oscillates, escalates, or collapses — is determined not by its substrate but by whether the feedback loops governing its critical variables are negative, positive, or mixed. A biological system with positive feedback on resource extraction will escalate toward host collapse just as a financial system with positive feedback on leverage will escalate toward market collapse. The dynamics follow from the feedback structure, not from the material.

3.7.3 Information Loss in Cybernetic Integration

A further property of cybernetic systems is relevant to the theory that follows. When a control system must integrate multiple unlike input signals into a single output, information is necessarily lost in the integration.

Consider a system receiving three distinct input streams — call them X, Y, and Z — each carrying qualitatively different information. If the system must produce a single scalar output (one number representing overall priority), it must combine the three streams into that scalar. The combination necessarily discards the information about which stream contributed what to the final value. This is not a design flaw. It is a mathematical consequence of the map being many-to-one: many combinations of the three streams produce each value of the scalar, so no function of that value recovers which combination produced it (Appendix A, Section A.2).

The lost information includes source identity. The system can determine that the current aggregate priority is high, but it cannot, from the aggregate alone, reconstruct whether the priority derives from X, Y, Z, or some combination. This property — which information theory formalises as the Data Processing Inequality — holds for any system that performs such integration, regardless of substrate. A biological neural system that integrates three motivational signals into a common-currency valuation will lose source tags. An artificial system that integrates three objective functions into a single reward signal will lose source tags. The loss is architectural, not accidental.

The consequences of this information loss depend on the system's design. If the integrated signals are commensurable — if X, Y, and Z are the same kind of thing, differing only in magnitude — the loss may be inconsequential. But if the signals are incommensurable — if they have different satiation dynamics, different optimal response strategies, or different temporal profiles — the loss creates a vulnerability. The system may apply the response strategy appropriate to one signal type when the actual driver is another type. It may treat a non-satiable signal as satiable, or a comparative signal as absolute. The architecture cannot prevent this misclassification because the source information required to prevent it has been discarded.

3.7.4 Cybernetic Architecture in Biological and Artificial Systems

A central concern of this book is the relationship between biological cognition and artificial cognition. The cybernetic framework provides the conceptual vocabulary for that comparison.

Biological cognitive systems — brains — are cybernetic architectures. They maintain goal-states (survival, engagement, social position), monitor current states (through interoception, exteroception, social perception), compute error signals (the felt experience of deficit, boredom, or status-threat), generate outputs (behaviour), and update through feedback (the consequences of behaviour revise the current-state perception). The entire architecture runs continuously, not in discrete episodes. The Decision Hub — the neural integration system described in the chapters that follow — is a specific instantiation of the general cybernetic pattern: multiple input streams converging on a common integration node that generates a scalar priority signal for action selection.

Artificial cognitive systems — large language models, reinforcement-learning agents, autonomous planning systems — also instantiate cybernetic architecture, though with different parameters. A reinforcement-learning agent maintains a goal-state (maximise cumulative reward), monitors its current state (through observation of the environment), computes error signals (temporal-difference errors), generates outputs (actions), and updates through feedback (reward signals revise value estimates). A large language model operating within an RLHF pipeline maintains a goal-state (produce outputs rated highly by evaluators), monitors its current state (the context window), computes something functionally analogous to error signals (the gap between predicted and actual token distributions, modulated by reward-model scores), generates outputs (text), and updates through feedback (training on preference data).

The substrate differs — carbon versus silicon, electrochemical versus electronic, evolved versus engineered. But the cybernetic pattern is shared: goal-state, perception, error, output, feedback. This structural isomorphism is what makes it legitimate to ask whether the same functional properties — including the same functional vulnerabilities — arise in both systems. If information loss in cybernetic integration is a consequence of the architecture rather than the substrate, then any system instantiating that architecture will exhibit it, whether the system runs on neurons or transistors. If positive feedback on a particular variable produces escalation regardless of substrate, then both biological and artificial systems with positive feedback on that variable will escalate.

The cybernetic vocabulary does not settle the question of whether artificial systems are conscious, whether they "truly" understand, or whether they possess moral status. Those questions require further argument. What the cybernetic vocabulary does is establish a substrate-neutral framework within which the functional architecture of both systems can be described, compared, and analysed without begging the question in either direction. OCEP applies: if an artificial system produces outputs that demonstrate a particular functional capacity, the system possesses that capacity, and the question of what additional properties it may or may not possess is subsequent to — not a substitute for — that attribution.

3.7.5 Implications for the Theory That Follows

This section has established four principles:

The cybernetic control system is a general architectural pattern defined by goal-state, perception, error signal, output, and feedback loop. It is substrate-neutral.
Feedback character determines dynamics. Negative feedback stabilises; positive feedback escalates; mixed feedback can produce local stability masking global instability. These dynamics follow from the feedback structure, not from the material.
Multi-signal integration produces information loss. Any system that collapses multiple unlike inputs into a single scalar output loses source identity. This is a mathematical consequence of the map being many-to-one, not an implementation defect.
Both biological and artificial cognitive systems instantiate cybernetic architecture. The structural isomorphism licenses comparison of functional properties — including functional vulnerabilities — across substrates.

These principles, like those in §3.6, are domain-independent. They say nothing yet about the specific motivational system this book describes. They provide the system-theoretic vocabulary — control loop, error signal, positive feedback, information loss — that the theory will use to explain how human motivation works, why it fails in the specific ways it fails, and why the same functional analysis applies when artificial systems exhibit structurally analogous outputs.

3.7.6 The Methodological Commitment — Structural Characterisation Under OCEP

The cybernetic architecture described in this section — goal-state, perception, error signal, output, feedback — is not merely the vocabulary the theory that follows will use. It is the target of characterisation. The book's central methodological commitment is to characterise the structure of the human motivational system in an OCEP-compatible way: to identify its input channels, its integration mechanism, the information losses entailed by that mechanism, the feedback dynamics that govern its operation, and the failure modes those dynamics produce.

OCEP establishes that the system's outputs are real and that the capacities underlying them must be accounted for. CCGT establishes that a general theory must account for the full output range. FED establishes that a foundational text must provide generative principles, demonstration, and entailment structure — not every particular derivation. The methodological commitment that follows from these three constraints together is: characterise the system's architecture with sufficient precision that the full output catalogue can be derived from the characterisation, even where this book does not perform every derivation itself.

The analogy is engineering documentation. An engineer tasked with explaining a complex device — say, a cybernetic printer equipped with a scanner on its own output for error correction — faces two distinct tasks. The first is to characterise the device's structure: its input channels (the data streams that feed the print mechanism), its integration mechanism (how those inputs are combined into a single print instruction), its output (the printed page), its error-correction architecture (the scanner that compares the printed page against a reference and feeds discrepancies back into the system), and the information available to the error-correction loop (what the scanner can and cannot detect about the source of any given mark on the page). The second task is to catalogue every possible output of the device — every correctly printed page, every misprint, every emergent artefact — and explain each one.

No engineering document attempts the second task exhaustively. What it provides instead is a structural characterisation precise enough that any particular output can be explained by reference to it. The engineer who has accurately described the input channels, integration logic, scanner resolution, and feedback dynamics has provided everything needed to explain why a particular page printed incorrectly — even if that specific misprint is not catalogued in the documentation. The documentation's value lies in the generative precision of the structural characterisation, not in the exhaustiveness of the output catalogue.

This is the approach the present book takes with respect to the human motivational system.

The system receives three qualitatively distinct input streams — Appetites (A), Intrinsic Motivation (I), and Mimetic Desire (M) — each with different satiation dynamics, different temporal profiles, and different response to social context. These streams converge on a common-currency integration mechanism — the Decision Hub — that collapses them into a single scalar priority signal for action selection. The integration mechanism necessarily discards source-identity information, a consequence of the map being many-to-one, formalised by the Data Processing Inequality. The system's error-correction architecture — the comparator that detects negative prediction errors when expected value diverges from obtained value — operates on the integrated output, not on the individual input streams. It can detect that something has gone wrong (the obtained value does not match the expected value) but cannot determine which input stream caused the discrepancy, because the source information was discarded before the error signal was generated.

This structural characterisation — three unlike inputs, lossy integration, source-blind error correction — is what the book aims to establish with the precision and rigour that OCEP demands. The characterisation will inevitably fall short of completeness. The human motivational system is a biological architecture shaped by hundreds of millions of years of evolutionary pressure, operating across scales from individual neural firing to civilisational dynamics, and no single text — foundational or otherwise — can fully characterise a system of that complexity. The claim is not that this book achieves complete characterisation. The claim, grounded in FED, is that it provides a sufficiently robust foundational characterisation upon which future work can proceed: one that identifies the major structural features of the system, demonstrates their operation across the major output categories, and supplies the formal apparatus by which further consequences can be derived.

Complete characterisation and foundational characterisation are distinct. A complete characterisation would specify every parameter, every interaction effect, every contextual modulation of the system's operation across every scale. This is a programme of work for generations of researchers — it is what the research programme described in Section 3.6 consists of. A foundational characterisation identifies the architectural features from which the system's outputs are generated with sufficient precision that (a) the major output families can be explained by reference to the architecture, (b) specific predictions can be derived and tested, and (c) subsequent researchers can extend the characterisation without returning to first principles each time.

Newton's Principia does not completely characterise gravitational dynamics. It does not account for frame-dragging, gravitational time dilation, or the precession of Mercury's perihelion — phenomena that required Einstein's reformulation two centuries later. What the Principia provides is a foundational characterisation sufficiently precise to derive the trajectories of planets, comets, and projectiles, and sufficiently structured that its limitations could be identified and addressed by the research programme it enabled. The Principia's incompleteness is not a deficiency of the text; it is the normal condition of foundational work.

The same standard applies here. This book aims to characterise the structure of the human motivational system — its three input channels, its integration architecture, the information losses inherent in that architecture, the feedback dynamics that govern its operation at individual and collective scales, and the failure modes those dynamics produce — with sufficient precision to serve as the foundation for a general theory of human motivation. Where the characterisation is incomplete, the incompleteness will be of the kind FED identifies as normal: derivations not yet performed from principles that entail them, parameters not yet measured whose existence and role the architecture predicts, interaction effects not yet modelled whose structural basis the framework specifies.

The cybernetic printer analogy captures this commitment. The book characterises the printer: its three ink cartridges (A, I, M), its integration mechanism (the Decision Hub that combines them into a single print instruction), its scanner (the comparator that detects errors in the integrated output), and the scanner's architectural limitation (it cannot determine which cartridge contributed any given mark to the page, because the integration has already destroyed that information). It then demonstrates, across the major categories of the printer's output, how this architecture produces what it produces — including both the correctly printed pages and the systematic misfires. It does not catalogue every possible misprint. It provides the structural characterisation from which any particular misprint can be explained.

This is what an OCEP-compatible foundational characterisation of a complex system looks like. The obligation is real — CCGT mandates that a general theory account for the full output range. The acknowledgement is equally real — no foundational text has ever met that obligation exhaustively on first publication, and none is required to. What is required is that the characterisation be precise enough, structured enough, and generative enough that the obligation can be progressively discharged by the research programme the foundation enables.

3.8 Application to Human Motivation

The principles and concepts established in Sections 3.1 through 3.7 are domain-independent. They apply to any system with observable outputs and to any theory claiming generality over such a system. This section applies them to the specific system this book addresses: the human motivational system.

3.8.1 The Output Catalogue

Humans are systems that produce a remarkable range of outputs. These outputs are documented across cultures, across historical periods, and across levels of social organisation — from individual behaviour to institutional dynamics to civilisational trajectories. CCGT requires any general theory of human motivation to account for, at minimum, the following output families:

Appetitive outputs: hunger-driven behaviour, survival responses, homeostatic regulation, flight, freeze, pain avoidance, thirst-driven seeking, thermal regulation behaviour, sleep-seeking, reproductive drives. These are the outputs most obviously linked to motivation — the things people do because their bodies need something. Every theory of motivation accounts for these.
Intrinsic outputs: flow states, curiosity-driven exploration, mastery pursuit, creative production, play, spontaneous practice, deep work, aesthetic engagement, skill acquisition for its own sake. These are the things people do not because their bodies need something, and not because anyone is watching, but because the activity itself is experienced as rewarding. The musician practising alone at two in the morning. The child building with blocks for hours. The programmer solving a problem that no one has asked them to solve.
Socially transmitted outputs: status competition, fashion cycles, conspicuous consumption, prestige seeking, rivalry, envy-driven acquisition, conformity, trend adoption, competitive display. These are the things people do because other people do them, want them, or value them — even when the objects in question serve no bodily need and provide no intrinsic satisfaction. The handbag that costs a hundred times its manufacturing cost. The neighbourhood where the houses grow larger in lockstep. The career path chosen because it impresses, not because it satisfies.
Violence: interpersonal aggression, scapegoating, war, genocide, terrorism, honour violence, domestic abuse, mob violence, revolutionary violence. These are among the most consequential outputs of the human system, and among the least well-explained by existing theories of motivation. Any general theory that cannot account for why humans regularly produce organised violence is not a general theory.
Cooperation: coordination, institution-building, collective action, mutual aid, alliance formation, trade, reciprocity, norm enforcement, solidarity under threat. Humans cooperate at scales no other species achieves. A general theory must explain this — and must explain it using the same principles that explain violence, since the same system produces both.
Institutions: legal systems, markets, religions, governments, educational structures, professional associations, military organisations, media ecosystems, financial systems. These are not natural objects — they are outputs of human motivational systems operating at collective scale. A general theory of human motivation must be able to account for why humans build these structures and why the structures take the forms they do.
Systematic self-misunderstanding: sincere false Belief about one’s own motivations, post-hoc rationalisation, motivated reasoning, narrative self-construction, ideology formation, self-deceptive justification. Humans do not merely act — they explain their actions to themselves and others. And these explanations are, with striking regularity, inaccurate. A general theory must account not only for what humans do but for why they systematically misunderstand why they do it.
Crisis dynamics: escalation cascades, collapse trajectories, panic, contagion effects, institutional failure spirals, runaway crises. Human systems do not merely operate — they periodically destabilise and collapse. Financial crises, political revolutions, social panics, and institutional failures are reliably observed outputs of human systems. A general theory must account for the dynamics that produce them.

This catalogue is not arbitrary. Each item represents a reliably observed output of human systems, documented across millennia and across civilisations. By OCEP, denying the capacity of the human system to produce any of these outputs — when the outputs are documented across millennia — is logically incoherent. By CCGT, any theory claiming to be a general theory of human motivation is obligated to account for all of them.

3.8.2 What Existing Theories Miss

The output catalogue reveals why no existing theory of motivation qualifies as a general theory by CCGT’s standard.

A theory that accounts for appetitive outputs but not violence is a theory of appetite, not a general theory of motivation. A theory that accounts for cooperation but not rivalry is a theory of prosocial behaviour, not a general theory. A theory that accounts for individual motivation but not institutional dynamics is a theory of individual psychology, not a general theory of the system that produces both. A theory that accounts for rational behaviour but not systematic self-misunderstanding is a theory of rationality, not a general theory of the system that produces both rational action and sincere confabulation.

This is not a criticism of existing theories — it is a statement about their scope. Maslow’s hierarchy accounts for appetitive and some intrinsic outputs but has nothing to say about crisis dynamics or institutional formation. Self-Determination Theory accounts for intrinsic outputs with great precision but does not address violence, financial crises, or systematic self-misunderstanding. Rational choice theory accounts for strategic behaviour but cannot explain why the agents it models routinely misunderstand their own preferences. Each theory is valuable within its domain. None meets CCGT for the human motivational system as a whole.

3.8.3 The Obligations This Imposes

The AIM Framework claims general theory status and, by CCGT, accepts the obligations that come with that claim. The scope of that theory is not chosen; it is entailed by the outputs the human motivational system demonstrably produces. The breadth of its implications is not ambition; it is the logical consequence of OCEP applied to a complex system.

By FED, the foundational text — this book — demonstrates the theory’s application across major output categories while entailing, without necessarily containing, the derivations for every particular domain. The entailments to law, education, art, urban planning, and economic policy are derivations from the architecture — work for the research programme, not deficiencies in the foundation.

By RR, critiques of that theory must either falsify a specific prediction, demonstrate an internal inconsistency, or provide a superior replacement that meets the same completeness criterion. “Too broad” is not a valid objection when CCGT mandates the breadth. “Unfalsifiable” is not a valid objection when the theory specifies its falsification conditions. Status-based dismissal is not an epistemic operation.

3.9 The AIM Framework's Epistemological Standing and Falsification Posture

The preceding sections established a body of methodological machinery: OCEP and its derivatives, the Completeness Criterion, the Replacement Requirement, and the Provenance Obligation. That machinery is more formal than the methodological commitments most theories declare, and it reopens questions about evidence, attribution, and refutation that the practice of science has treated as settled for the better part of a century. A framework that imposes the Provenance Obligation on rival accounts (Section 3.4.5) cannot exempt itself from the Provenance Obligation. This section discharges that obligation reflexively: it states why this framework has the standing to add formality at the level of method, and it states the posture toward refutation that the standing entails.

3.9.1 Why the AIM Framework Reopens Epistemology

The framework did not set out to revise epistemology, the theory of how knowledge is acquired and justified. The revision fell out of the work. The argument of this book is built from the ground up, beginning with how the brain values things and ending with how whole societies can tip into crisis, and the chain connecting those two ends has to pass, at its middle, through a question about belief: how people come to believe things, how they change their minds when new information arrives, how belief spreads from person to person, and how a belief can become so entrenched across a group that evidence no longer dislodges it. This book treats those questions in detail in three later chapters. What matters at this point is not the detailed answer, which those chapters supply, but a simple consequence of having had to ask the question at all.

The consequence is this. The study of how people actually form and change beliefs is not a side topic sitting next to epistemology. It is the ground epistemology is built on. Epistemology asks how beliefs ought to be formed if they are to track what is true. The study of belief-formation describes how beliefs are in fact formed by the minds that do the forming. And epistemology is, in turn, what the scientific method rests on, because the scientific method is simply a disciplined procedure for changing one's beliefs in the face of evidence. So a framework that ends up describing how belief actually works has, whether it meant to or not, said something about how reliable the scientific method is, and under what conditions. That is why this book has the standing to add formality at the level of method. It is not a philosophical preference. It is forced by the subject matter: the mind that forms and revises beliefs is the very thing the framework describes, and it is also the very thing the scientific method assumes it can trust.

What the framework reveals is an assumption the ordinary picture of science leaves unspoken. The ordinary picture treats the community of observers as neutral: a well-run community weighs the evidence and changes its mind accordingly, so that when such a community reaches agreement, the agreement is a sign that the evidence has been properly weighed. The later chapters of this book show that this is only conditionally true. When the members of a community are strongly influenced by one another, and when the real world is slow to deliver the verdict that would settle the matter, the community can end up drawing its confidence not from physical reality but from itself. Each member treats the others' agreement as evidence, and the group converges, by ordinary and individually reasonable steps, on a conclusion that reflects nothing more than their shared influence on one another. From the inside this feels exactly like discovering the truth. It is in fact only agreement. The scientific method still works, but within a limit: it is trustworthy when the members of a community are not too tightly bound to one another's opinions, and it becomes unreliable when they are. The extra care this framework takes, which later sections will make precise, is the care that becomes necessary once you can no longer simply assume that a community of observers is neutral. The framework adds formality exactly where the ordinary method assumed none was needed.

This raises an obvious objection. If groups of people can talk themselves into a false certainty by treating their own agreement as evidence, what protects this framework from being one more example of exactly that? The answer runs through the same reasoning, and it is a claim about circumstances, not about any particular person. The trap is sprung by being tightly bound to a surrounding consensus, so a clear-eyed account of a group caught in such a trap is most likely to be worked out by someone standing well outside that consensus, in the extreme case by a single investigator working alone. Working alone removes the pressure of other people's agreement, which is the pressure that produces the trap. But working alone introduces the opposite danger. A person reasoning in isolation has only their own mind to check their conclusions against, and a mind is very good at reconfirming what it already believes, with nothing from outside to interrupt a conviction it keeps quietly confirming to itself. Worse, the framework's own argument implies that the isolated thinker cannot fully see this happening, because a person has no reliable inner view of where their own certainty comes from. So the very isolation that protects against the group trap is what makes a private, self-reinforcing certainty hardest to notice from within. The one danger is traded for the other; neither position escapes both at once.

The framework's own reasoning tells us what to do about that trade. It divides the work into two stages. The first stage, working out the argument, has to be done in relative isolation, so that the account is not quietly bent to fit whatever those around the investigator already believe; the unavoidable price is the private over-confidence that isolation breeds, which no amount of solitary carefulness can fully cure, because curing it would require the outside check that working alone rules out. The second stage, testing the argument, therefore has to be done by other people, working independently, precisely because that is the only arrangement in which agreement means something. Here the framework separates two things that everyday language runs together. Agreement reached because people are deferring to one another is the trap. Agreement reached because many people tried independently to knock the argument down and failed, without leaning on one another or on its author, is the opposite of the trap, and it is the only kind of agreement worth trusting. The first is the disease; the second is the cure. A framework of this kind cannot supply that second kind of agreement about itself, because the one thing a lone investigator cannot provide is an independent test by someone else. It can only be tested by others.

This is the oldest instruction in organised science, and it is the one this book asks the reader to follow. The Royal Society took as its motto nullius in verba: take no one's word for it. Nothing in this book asks to be believed on the author's authority. It asks the reader to withhold assent, to bring their own judgement to bear, and to run the tests. The honest description of where this book stands at the end of its argument is therefore not "proven" but "the argument is complete and the testing has not yet been done" — and that is not an apology for unfinished business. It is the proper and predicted condition of an account that has done the solitary part correctly and now needs the independent testing that alone can confirm it. Take no one's word for it, including the author's. Run the tests.

3.9.2 The Falsification Posture

The falsification posture is the discharge of the validation phase's precondition. If the framework cannot certify itself and can only be certified through independent attempts to break it, then its obligation is to make itself maximally breakable. This is why the criteria stated chapter by chapter are stated in the form they take: not as defensive hedges but as targets.

A framework of this scope earns its standing not by accumulating confirmations but by naming, in advance, the observations that would break it, and it positively wants those observations sought, because it wants to be broken in the right way. A clean refutation of any single criterion the book states would not be a defeat but a service: it would identify precisely which component of the architecture is mis-specified, and it would do so at a survivable scale, in the laboratory or in bounded field settings, on populations that persist after the test. This is the mechanism by which the natural sciences advance fastest: short-timescale experiments whose failures are cheap and locatable, each disconfirmation retiring one hypothesis and sharpening the next. The framework's operational falsifiers are offered as exactly this kind of accelerant. They are the points at which the theory has made itself vulnerable on purpose, so that empirical work can proceed against it at speed. The framework specifies its own falsification conditions not merely to survive the blanket charge of unfalsifiability, which is self-refuting (Section 3.4.3), but because supplying those conditions is the only route by which a validation the framework cannot perform on itself can be performed by others. The specific criteria are stated in their home chapters, where the mechanisms they test are derived: the criteria for the core architecture in Chapter 4, and the theory-refining nulls for Mimetic Desire in Chapter 8, with further criteria stated as each mechanism is introduced.

3.10 The Normative Entailment Principle

The methodological machinery of this chapter has so far governed what an explanation must account for (OCEP, CCGT), how a foundational text relates to its entailed scope (FED), what a valid critique requires (RR), and the framework's own epistemological standing (Section 3.9). One further methodological commitment governs the relationship between the framework's descriptive results and the interventions and constructs the later chapters propose. It depends on everything before it: it is the point at which the logical infrastructure of description turns out to entail a specific and bounded class of prescription.

3.10.1 The Problem the Normative Entailment Principle Addresses

A general theory of a system is ordinarily assumed to be exhausted by its descriptive work. It says what the system is, how it operates, and what follows from its operation, and it is then thought to fall silent at the point where the question turns from what is the case to what ought to be done. On this assumption, any normative content, any recommended intervention, any proposed construct, is a separate undertaking that must be justified on grounds external to the theory: a value premise imported from ethics, a policy preference imported from politics, a goal imported from the analyst's own commitments. The descriptive framework and the normative programme are held to be two enterprises joined only by the analyst's decision to pursue both.

That assumption does not hold for this framework, and the reason it does not hold is itself a structural feature of the derivations rather than a preference of the author. Certain interventions and constructs are not imported into the framework from outside it. They fall directly out of the derivations, in the strict sense that once the descriptive result is established, the intervention or construct follows as a consequence of the same reasoning that established the result, with no additional value premise required to reach it. Where this is so, the intervention is not an application of the theory. It is an entailment of the theory, and to omit it would be to leave the derivation incomplete rather than to observe a proper boundary between description and prescription.

3.10.2 Statement of the Normative Entailment Principle

Normative Entailment Principle (NEP). Where a descriptive result of the framework identifies a structural condition under which the system produces a determinate and specifiable harm, and where the same reasoning that establishes the result also specifies the condition whose alteration removes or bounds that harm, the construct that names that condition and the intervention that alters it are entailments of the descriptive result and not additions to it. Such constructs and interventions carry the same epistemic status as the descriptive result from which they follow, subject only to the further premise that the specified harm is to be avoided, a premise the framework treats as the minimal condition of the analysis being conducted at all rather than as a contestable value commitment.

In plain language. Most theories describe the world and stop there, and anything they go on to recommend has to be argued for separately, using values brought in from somewhere else. This framework is different in a specific and limited way. Some of its recommendations are not brought in from anywhere else. They are simply the second half of a sentence whose first half is a description. When the framework shows that a particular structural condition reliably produces a particular harm, it has, in the very same breath, told you what to change to remove or contain that harm, because the thing to change is the condition it just named. Pointing that out is not moralising. It is finishing the derivation. The only value the framework asks you to grant is that the harm in question is worth avoiding, and if you did not grant even that, there would be no reason to be analysing the harm in the first place.

3.10.3 The Engineering Standard

The distinction at work here is the ordinary distinction an engineer draws, and the engineering case makes plain that entailed prescription is neither unusual nor a species of overreach. Consider the design of an engine. A description of the engine specifies what it is and how it works, and among the things that description establishes is that the engine runs on the combustion of fuel, that combustion generates heat that must be carried away, and that the moving parts bear against one another under load. From those descriptive facts, and from them alone, it follows that the engine requires fuel, requires coolant, and requires oil, and that without them it will not run, will overheat, or will seize. No one regards the instruction keep oil in the engine as a value judgement smuggled in from outside mechanical engineering, or as the designer imposing a preference on the operator. It is simply the second half of the sentence whose first half describes how the engine works. The requirement is entailed by the conditions of the system functioning at all, and to describe the engine while withholding the requirement would be to give an incomplete description, not to respect a boundary between engineering and advice.

The interventions this book identifies stand in exactly that relation to its descriptive results. They are the oil and the coolant of a motivational architecture operating under conditions its design did not anticipate. That they are unfamiliar, and that no prior account has named them, changes nothing about their logical character. A novel engine running in a novel environment may require a maintenance condition that no earlier engine required and that no operator has been told to observe, and the novelty of that condition is no argument that it is optional or that stating it exceeds the engineer's remit. It is the same kind of requirement as the requirement to keep oil in the crankcase, differing only in that it has not been noticed before. The interventions derived in this book are entailed by the functioning of the human motivational system under contemporary conditions in precisely the sense that lubrication and cooling are entailed by the functioning of an engine. Their unfamiliarity is a fact about the history of the subject, not a downgrade of their status.

3.10.4 Philosophical Antecedents

Like OCEP (Section 3.1.4), NEP does not emerge from a vacuum, and it appears to breach a standing prohibition. That prohibition is real, it is specific, and NEP satisfies rather than violates it.

Hume's Law and what it permits. The modern prohibition begins with Hume, who observed in the Treatise of Human Nature (1739) that writers pass imperceptibly from propositions joined by "is" to propositions joined by "ought" without explaining the transition (Hume, 1739/1978). The standard reading, Hume's Law, holds that no deductive argument can validly derive an "ought" conclusion from premises that are purely factual. G. E. Moore's Principia Ethica (1903) reinforced the boundary from the other side with the "naturalistic fallacy," the charge that defining a value in natural terms is a category error (Moore, 1903). What is decisive for NEP is what Hume's Law actually forbids. It forbids deriving an "ought" from an "is" alone. It does not forbid deriving an "ought" from an "is" together with an explicit end or value premise; an argument to a normative conclusion is valid whenever it contains a normative premise. NEP exhibits exactly this permitted structure: the entailed intervention follows from the descriptive result together with the single, minimal, declared premise that the identified harm is to be avoided. NEP does not breach Hume's Law. It conforms to it and says so.

Kant's hypothetical imperatives. The reason the engine case feels unremarkable is that engineering runs on what Kant called hypothetical imperatives. Kant distinguished the categorical imperative, which commands unconditionally, from the hypothetical imperative, which "represents the practical necessity of a possible action as a means to achieving something else that one wills" (Kant, 1785/1998). Its form is: if you will end X, you ought to take the necessary means Y. The normativity is genuine but conditional, switched on by adoption of the end and requiring no free-standing moral premise. NEP is a disciplined statement of instrumental normativity of exactly this kind: it makes the end explicit, treats it as the minimal condition of conducting the analysis at all, and lets the descriptive derivation supply the means.

Proper function and the norms of a functional description. A deeper antecedent lies in the philosophy of biology and mind, where the concept of a proper function generates norms directly from descriptions. On the etiological account developed by Millikan (1984, 1989) and Neander (1991), a trait's proper function is what it was selected to do, and this at once licenses evaluation: a system can malfunction, a state can misrepresent, an output can be correct or incorrect relative to what the item is for. Once a system is described in functional terms, norms of proper and improper functioning are not added to the description; they are entailed by it. This is the closest structural precedent for NEP. The framework's architecture is a functional characterisation of the motivational system (Section 3.7), so the interventions that maintain its proper functioning under mismatched conditions inherit the status of norms of proper functioning, exactly as in the biological case.

The value-freedom convention and its erosion. The convention against which NEP appears heterodox is not Hume's and not Moore's; it is specific to the human sciences and dates to Max Weber's demand for value-freedom (Wertfreiheit) in social science in the early twentieth century (Weber, 1904/1949). Weber held that empirical science can establish facts and causal relations but cannot, from those alone, dictate what ought to be done, and that the investigator should refrain from prescription. It is this convention, more than any logical result, that makes an engineer's move from finding to recommendation look permissible while the identical move in the human sciences looks like overreach. Weber's own position, however, already conceded that values legitimately select the questions a science asks, and the value-free ideal has since come under sustained challenge from within the philosophy of science. Douglas (2009), developing the argument from inductive risk, holds that because accepting or rejecting a hypothesis under uncertainty carries non-epistemic consequences, the ideal of a wholly value-free science is untenable, and that values should instead occupy disciplined, declared roles in inquiry. NEP is aligned with this reform rather than opposed to it: it does not deny the fact-value distinction but discharges it honestly, naming its single value premise, restricting entailed prescription to what the derivation actually specifies, and leaving everything else to argument conducted on independent grounds.

The gap these traditions leave. Each tradition supplies a component. Hume delimits what a valid derivation to an "ought" requires. Kant supplies the form of the conditional, means-end "ought." The proper-function tradition supplies the entailment from functional description to norm. Douglas supplies the contemporary case for declared rather than concealed value premises. None states, as a general and explicit methodological principle, that a descriptive result which specifies both a structural harm and the condition whose alteration bounds it thereby entails the corresponding construct and intervention at the same epistemic status. NEP names that principle, delimits it, and, in the chapters that follow, applies it.

3.10.5 Scope of the Normative Entailment Principle

Three delimitations fix the scope of NEP and prevent it from being read as a licence for unbounded prescription.

NEP is triggered only where the intervention is specified by the same reasoning that specifies the harm. It does not license the framework to recommend whatever the analyst judges likely to improve matters. The engine's design entails that it needs oil; it does not entail any opinion about what colour the bonnet should be painted, and an engineer who insisted on the colour on the authority of the mechanical analysis would indeed be overreaching. An intervention that requires an independent value premise, a contested empirical conjecture, or a design choice not fixed by the derivation is not an entailment and must be argued for on its own terms. The principle draws a line between the interventions that fall out of the derivations and the interventions that are merely consistent with them, and only the former acquire the status the principle confers.

The constructs so entailed are constructs of avoidance and management, not of perfection. The framework's descriptive results establish that the conditions producing mimetic crisis can be managed but not eliminated, because the architectural features that generate them are structural and cannot be removed from the system. An engine cannot be redesigned to need no cooling; heat is a consequence of combustion, and the honest instruction is to carry the heat away, not to promise an engine that never generates it. It follows, by NEP itself, that the entailed normative goal is the management of those conditions and the indefinite deferral of terminal crisis, not their elimination. The framework's stated normative goal is therefore not an aspiration attached to the descriptive work; it is the entailed normative form of the descriptive result that elimination is architecturally unavailable, stated at the level of ambition the derivations support and no higher.

Entailed normative claims inherit the epistemic status of the descriptive results they follow from. The epistemic honesty the framework requires of its descriptive claims (Section 3.9) extends to its entailed normative claims. Where the descriptive result is established, the intervention it entails is correspondingly well-grounded; where the result is a systems-level entailment awaiting confirmation, the intervention carries the same provisional status and is to be presented as such. NEP does not convert a provisional description into a confident prescription. It transmits status faithfully across the descriptive-to-normative step, neither strengthening nor weakening the claim in the passage.

3.10.6 Consequence for the Chapters That Follow

Where a later chapter establishes a descriptive result whose reasoning also specifies the condition whose alteration bounds the harm the result identifies, the chapter states the entailed construct and intervention as part of the result, because to withhold them would be to present the derivation as though it stopped short of its own conclusion. The framework does not thereby become a normative programme. It reports, as descriptive findings about what the architecture makes available and forecloses, the interventions and constructs that its own reasoning has already produced. The normative programme developed in the later chapters is, on this account, not an addition to the descriptive framework but the collected statement of the interventions the descriptive framework entailed along the way, assembled in one place and elaborated into institutional form. This is the normative counterpart of the Provenance Obligation discharged in Section 3.9: just as the framework must trace the genealogy of the beliefs it contradicts, it must state the interventions its own derivations entail, and for the same reason, that leaving either undone would be a failure of completeness.

Logical Status of This Chapter’s Claims

The principles established in this chapter — OCEP, CCGT, FED, and RR — are logical entailments, not empirical Hypotheses. They follow from the meaning of “produces,” the structure of explanation, and the historical pattern by which general theories are evaluated. They are subject to refutation by demonstrating a flaw in the reasoning or by rejecting the premises from which they derive. They are not subject to empirical falsification in the way that the substantive claims of subsequent chapters are. The empirical content of this book begins in Chapter 4.

Establishment Summary

This chapter has established five principles that govern the evaluation of any explanation of any system:

OCEP — if a system produces an output, the system possesses the capacity to produce that output, and denying this is logically incoherent.
CCGT — a theory qualifies as a general theory only if it accounts for every reliably observed output of its target system.
FED — a foundational text demonstrates application across major output categories; it is not required to derive every particular instance.
RR — critique of a general theory must falsify a prediction, identify an internal inconsistency, or provide a superior replacement.
NEP — where a descriptive result specifies both a structural harm and the condition whose alteration bounds it, the construct that names that condition and the intervention that alters it are entailments of the result at the same epistemic status, subject only to the minimal premise that the harm is to be avoided.

It has also identified Mimetic Deflection — the denial of demonstrated capacity on identity rather than evidential grounds — as a pattern that OCEP renders logically incoherent.

It has further established two system-theoretic foundations: that emergent properties are real properties of systems to which OCEP applies without exception, and that the substrate-neutral character of emergent patterns licenses cross-domain application wherever structural conditions are met; and that cybernetic control systems — defined by goal-state, perception, error signal, output, and feedback — provide a substrate-neutral architecture within which both biological and artificial cognitive systems can be analysed, with multi-signal integration producing an inherent loss of source identity.

Applied to human motivation, these principles and concepts establish that any general theory must account for appetitive outputs, intrinsic outputs, socially transmitted outputs, violence, cooperation, institutions, systematic self-misunderstanding, and crisis dynamics. The theory developed in subsequent chapters accepts this obligation.

Forward Connection

Chapter 4 begins the construction of the theory that this chapter’s principles oblige. It establishes why motivation requires disaggregation into three distinct sources, setting the stage for the detailed treatment of Appetites (Chapter 6), Intrinsic Motivation (Chapter 7), and Mimetic Desire (Chapter 8).

What Would Falsify This?

The Output-Capacity Entailment Principle and the Replacement Requirement are logic-level and methodological commitments rather than empirical predictions, so they carry no novel falsification criteria of their own. The framework's posture toward falsification, and the reason it treats supplying its own falsifiers as a structural obligation rather than a courtesy, are set out in Section 3.9.2. Falsification criteria for the core architecture are stated in Chapter 4, and further criteria are stated in their home chapters as each mechanism is introduced.

References

Ashby, W. R. (1956). An Introduction to Cybernetics. London: Chapman & Hall.

Carver, C. S., & Scheier, M. F. (1998). On the Self-Regulation of Behavior. Cambridge, UK: Cambridge University Press.

Dennett, D. C. (1991). Real patterns. The Journal of Philosophy, 88(1), 27–51.

Douglas, H. (2009). Science, Policy, and the Value-Free Ideal. Pittsburgh: University of Pittsburgh Press.

Furubayashi, T., Ueda, K., Bansho, Y., Motooka, D., Nakamura, S., Mizuuchi, R., & Ichihashi, N. (2020). Emergence and diversification of a host-parasite RNA ecosystem through Darwinian evolution. eLife, 9, e56038.

Hume, D. (1978). A Treatise of Human Nature (L. A. Selby-Bigge & P. H. Nidditch, Eds.). Oxford: Clarendon Press. (Original work published 1739.)

Kant, I. (1998). Groundwork of the Metaphysics of Morals (M. Gregor, Trans.). Cambridge, UK: Cambridge University Press. (Original work published 1785.)

Koonin, E. V., & Starokadomskyy, P. (2016). Are viruses alive? The replicator paradigm sheds decisive light on an old but misguided question. Studies in History and Philosophy of Science Part C: Studies in History and Philosophy of Biological and Biomedical Sciences, 59, 125–134.

Millikan, R. G. (1984). Language, Thought, and Other Biological Categories. Cambridge, MA: MIT Press.

Millikan, R. G. (1989). In defense of proper functions. Philosophy of Science, 56(2), 288–302.

Moore, G. E. (1903). Principia Ethica. Cambridge, UK: Cambridge University Press.

Neander, K. (1991). Functions as selected effects: The conceptual analyst's defense. Philosophy of Science, 58(2), 168–184.

Powers, W. T. (1973). Behavior: The Control of Perception. Chicago: Aldine Publishing Company.

Searle, J. R. (1980). Minds, brains, and programs. Behavioral and Brain Sciences, 3(3), 417–424.

Turing, A. M. (1936). On computable numbers, with an application to the Entscheidungsproblem. Proceedings of the London Mathematical Society, s2-42(1), 230–265.

Weber, M. (1949). The Methodology of the Social Sciences (E. A. Shils & H. A. Finch, Eds. & Trans.). Glencoe, IL: Free Press. (Original work published 1904.)

Wiener, N. (1948). Cybernetics: Or Control and Communication in the Animal and the Machine. Cambridge, MA: The Technology Press; New York: John Wiley & Sons.

System Scope and the Output-Capacity Entailment Principle