artifacts/standard-named

Rendered from markdown source. Open raw source on GitHub.

--- title: "Telic Fields: A First-Principles Foundation" subtitle: "Standing, Projection, Relation, Context, and Navigation Among Multiple Ends" artifact_date: "2026-07-14" version: "0.2" artifact_type: "candidate-foundational-paper" domain: "TELIC-FIELDS" scope: "WORKING" lineage: "THE-TELIC-FIELD-PAPERS" status: "candidate" processing_tier: 4 source_role: "derived-conceptual-artifact" content_canon_status: "unset" publication_status: "unpublished" derived_from:

• "20260713__TELIC-FIELDS__TRANSCRIPT__SOURCE__ORIGIN-SESSION__from-attractors-to-semantic-polytelometry.md"
• "20260713__TELIC-FIELDS__SESSION-COMPRESSION__WORKING__ORIGIN-SESSION__telic-fields-and-semantic-polytelometry.md"
• "20260713__TELIC-FIELDS__GLOSSARY__WORKING__ORIGIN-SESSION__working-vocabulary-and-claim-register.md"
• "20260713__TELIC-FIELDS__LINEAGE-MAP__WORKING__WORK-VAULT__from-dyads-and-semantic-collapse-to-telic-fields.md"

research_companion:

• "20260713__TELIC-FIELDS__REVIEW__WORKING__G-1__adjacent-fields-and-prior-art.md"
• "20260714__TELIC-FIELDS__REVIEW__WORKING__G-2__terminology-and-primary-source-adjudication.md"

provenance_note: > This paper is a candidate synthesis derived from the witnessed origin session. Version 0.2 incorporates the G.2 terminology and primary-source review. It proposes a new composition and vocabulary for a known family of problems. It does not assert that the terminology is novel, that the framework is empirically established, or that "field" denotes a newly discovered physical field. ---

Telic Fields: A First-Principles Foundation

Version 0.2 — revised against the G.2 terminology and primary-source review

Abstract

A telic field is proposed here as the structured end-relative landscape generated when possible states matter differently to a center of standing. The concept is intended to describe a domain that ordinary goal language handles poorly: the simultaneous presence of explicit aims, embodied needs, feared futures, protected conditions, institutional incentives, temporal obligations, internal conflicts, and emergent purposes that influence action without forming one coherent objective function.

This paper proposes a new composition and vocabulary for a known family of problems. Biological teleology, cybernetics, enactivism, participatory sense-making, semantic information, complex problem solving, multiobjective optimization, plural-self theories, and goal-directed state-representation research already establish much of the surrounding terrain. The proposed contribution is to bind that terrain through standing, projection, consent, witness, context capacity, and non-sovereign semantic comparison.

The model develops from five first principles. First, significance is local before it is shared: a state matters relative to some center. Second, relation does not require merger: two centers may enter a dyad while retaining distinct fields. Third, coordination operates through scoped and corrigible projections rather than direct access to another field. Fourth, witness and loops preserve enough orientation through time for correction, consent, and repair. Fifth, higher-order intelligence should navigate among plural ends without silently converting one projection into sovereign purpose.

The framework does not claim that all valuable experience is goal-directed. Play, presence, release, curiosity, and atelic or paratelic value may need protection precisely from continual optimization. Nor does the paper claim that telic fields are physical fields or that language models can determine anyone's true purpose. It offers a candidate research program for keeping plural end-relative significance, authority, provenance, and consequences legible before systems optimize them.

---

1. Why another account of purpose is needed

Purpose is usually represented too cleanly.

A person says what they want. An institution publishes a mission. A contract names an obligation. A prompt specifies a task. A dashboard displays a metric. A model is trained against a reward. These representations are necessary because coordinated action requires compression. Yet they invite a recurring mistake: the representation begins to stand in for the living field from which it came.

The person who says, “I want the promotion,” may also want rest, dignity, family continuity, protection from failure, release from debt, and freedom from becoming the kind of person the promotion appears to require. The institution that says, “We connect people,” may also preserve growth, advertising revenue, legal safety, managerial continuity, public legitimacy, and technical reliability. A prompt may state the requested output while omitting why the output matters, which costs are unacceptable, whose consent governs the work, and when the instruction should cease to apply.

None of these systems has only one end.

The problem is not simply that goals conflict. The deeper problem is that goals differ in source, scale, authority, expression, time horizon, and consequence. Some are consciously chosen. Some are embodied. Some are inherited. Some emerge from the interaction of systems whose participants never selected them. Some are legitimate only within a narrow scope. Some are not goals at all, but boundaries or invariants that must remain true during movement.

A framework for purpose therefore needs to do more than identify targets. It must preserve who or what is affected, how an end entered the system, what it is allowed to recruit, what it may sacrifice, and how the system learns that its current representation is wrong.

That is the work proposed for telic fields.

---

2. The first principle: possible states matter differently

The framework begins with a minimal claim:

A telic field begins where possible futures cease to be equivalent for a center of standing.

A center of standing is a center whose protected differences, consequences, or future states must remain represented in decisions that affect it. The term is deliberately broader than legal personhood and narrower than “anything that can be modeled.” A person clearly has standing. A living organism may have standing relative to its conditions of viability. A constitutional element within a person may possess local standing inside that person's internal governance without becoming an externally sovereign legal actor. A relationship or institution may develop derived standing if its continuity becomes consequential to participants or successors. Whether artificial systems can possess standing remains open.

This language has an important predecessor. Paul W. Taylor describes living organisms as teleological centers of life: organized beings that pursue their own good in their own way. The present framework does not claim to originate the connection among local teleology, moral regard, and a center. It extends the question into governance: what differences must remain represented when a decision recruits, constrains, or imposes consequences upon a center?

The extension is useful, but it also prevents one universal criterion from being assumed too early. Biological standing, experiential standing, relational standing, institutional standing, temporal standing, delegated standing, and possible artificial standing may require different grounds. “Center of standing” is therefore a procedural umbrella, not a completed moral ontology.

Standing and telic significance are related. A possible state matters because it changes what can continue, flourish, be protected, be harmed, be repaired, or remain possible for the center. This difference may be represented consciously, but consciousness is not required by the working model. A bacterium need not narrate a goal for some environmental states to be more compatible with its continued organization than others. A frightened internal protector need not speak in full sentences for exposure to matter differently than safety. A future self cannot speak directly in the present, yet present actions can constrain the conditions it will inherit.

The word telic is used in its broad philosophical sense: end-relative. It does not imply that every center has one grand purpose or that all behavior is consciously intentional. The local field may be plural, conflicted, implicit, changing, and partially inaccessible.

A local telic field may contain desired states, avoided states, viability conditions, protected boundaries, fears, obligations, identity commitments, latent tendencies, acceptable costs, unacceptable sacrifices, and possible routes. It is therefore closer to a structured landscape of significance than to a ranked list of preferences.

Terminology boundaries

The term enters an occupied neighborhood.

Paul Taylor's teleological center of life grounds moral regard in an organism's pursuit of its own good. Talcott Parsons's telic system refers to the sphere of ultimate values in a broad theory of action. Reversal theory uses telic and paratelic for serious goal-oriented and playful present-oriented metamotivational states. Linguistic telicity concerns whether an event is expressed as bounded or directed toward an inherent endpoint. Recent computational work by Nadav Amir, Yael Niv, Stas Tiomkin, Angela Langdon, and collaborators uses telic states, telic-controllable state representations, and telic controllability for goal-shaped state representation in bounded agents. Complex-problem-solving research uses polytely for situations containing multiple simultaneous and potentially conflicting goals.

The proposed terms should therefore be distinguished precisely:

• A telic state is a goal-relative class or representation of experience.
• A telic field is the wider structured landscape of differently significant states, protections, costs, boundaries, and routes indexed to a center of standing.
• Polytely names the condition of multiple simultaneous goals.
• Polytelometry names the disciplined effort to make those goals' relationships, authority, provenance, and consequences legible.
• Linguistic telicity concerns event boundedness in expression, not the governance architecture proposed here.

The framework also preserves paratelic and atelic value. A telic field is not a demand that all experience become instrumental. It can represent the protected right to play, rest, remain present, explore without predefined completion, release an inherited purpose, or refuse continual optimization.

---

3. Telos is not attractor

The origin conversation began by asking whether attractors are fundamentally teleological objects. The answer required a distinction.

An attractor describes convergence under a system's dynamics. A telos describes an end, completion, preservation condition, or possible future relative to which deviation can be interpreted and correction organized. A system can converge without intending. A system can declare an end without possessing dynamics capable of reaching it. A system can also produce an attractor that contradicts its declared purpose.

An attractor is not a purpose. It is the shape that makes purpose causally durable.

This distinction allows several layers of purpose to become visible.

The declared telos is what the system says it serves. The encoded telos appears in rules, objective functions, permissions, incentives, metrics, and comparators. The operative telos is what the system actually restores or protects when purposes conflict. The emergent telos is an end-like orientation produced by interaction without explicit adoption. A consented telos is an end whose authority to recruit participants has been meaningfully accepted. A protected telos is a state the system is required to preserve, although some protected conditions may be better described as invariants rather than ends.

The most useful diagnostic is behavioral:

Which objective is allowed to sacrifice the others?

A mission statement describes a bearing. Correction behavior shows the wake. What a system repeatedly restores, and which losses trigger the strongest response, reveal more about operative orientation than the words attached to the system.

This does not mean the declared purpose is always false. A system may sincerely hold incompatible teloi. The resulting condition is not necessarily hypocrisy. It may be teleological plurality. It becomes teleological incoherence when materially different purposes are represented as one, especially when the concealment prevents affected centers from understanding how their energy is being recruited.

---

4. The field is local before it is shared

The term field is used cautiously.

It does not presently name a newly discovered physical field. It names a structured relational landscape: the set of possible states, differences, gradients, boundaries, protections, and routes that matter relative to a center.

The field is indexed:

\[ T_A(t) \]

This denotes the telic field of center \(A\) at time \(t\).

The index matters because two centers can inhabit the same external situation while experiencing different telic terrain. A public presentation may be an opportunity to one person, a threat to another, an obligation to a third, and a technical dependency to the institution hosting it. There is no contradiction in saying the same event occupies different local fields.

Goal-directed state-representation research provides a close computational neighbor. In work on telic states and telic controllability, an agent's goals help determine which experiences are grouped together and which distinctions become useful for control. A telic field is proposed as a broader construct. It includes not only state representation for policy selection, but also protected conditions, authority, standing, consent, acceptable sacrifice, temporal commitments, and relationships among multiple centers.

A field is also more than a utility function. A utility function ordinarily maps outcomes into an ordering or scalar value. The telic-field model intentionally preserves items that may not be commensurable: dignity and speed, attachment and truth, safety and exploration, immediate survival and future possibility. It also preserves the source and authority of the concern. A preference held by a participant, a constraint imposed by law, an incentive introduced by a platform, and a need inferred by a model are not equivalent merely because they influence the same action.

Not everything in the field is an objective. Some elements are boundaries. Some are invariants. Some are veto conditions. Some are permissions to remain unoptimized. Some are uncertainties that should not be forced into rank order.

The model therefore begins before scalarization.

---

5. The dyad: relation without erasure

Two centers enter relation:

\[ A(T_A) \;\; 🜁 \;\; B(T_B) \]

The 🜁 relation begins with presence before outcome is decided. It does not require agreement, merger, obedience, or full disclosure. Each center retains its local standing and field.

The dyad is proposed as the minimum useful relational unit through which local fields become mutually consequential. This is a modeling claim, not yet an ontological proof. Triadic and network structures may be more basic in some domains, especially where witness, language, culture, or institutions already mediate every encounter. Even so, the dyadic lens exposes an indispensable question: what happens when one locally organized field meets another without either being permitted to disappear?

The encounter creates possibilities neither center possessed alone. These may include cooperation, misunderstanding, obligation, shared work, intimacy, conflict, betrayal, repair, and exit. The resulting relational telic field is not simply the union of the two local fields. The relation itself can acquire patterns and ends: preserving trust, completing a shared task, maintaining contact, protecting a child, sustaining a contract, or ending cleanly.

This relational emergence is close to the enactive account of participatory sense-making, in which interaction dynamics can contribute constitutively to meaning rather than merely transmitting already complete internal representations. The telic-field formulation adds a governance question: when a relational end emerges, what authority does it possess over the centers that sustain it?

A relationship may become important without becoming sovereign.

---

6. Meaning as compressed orientation

A center cannot transmit its whole field.

It emits signals: words, gestures, actions, symptoms, prices, roles, interfaces, policies, contracts, metrics, refusals, and silence. In consequential coordination, these signals often carry compressed orientation among centers and downstream processes. They alter anticipated movement, available action, risk, relation, or interpretation.

Meaning is a navigational distinction preserved because it mattered to some center of standing.

This is a candidate synthesis, not a claim to have first grounded meaning in consequence. Teleosemantics connects representation to biological function. Kolchinsky and Wolpert define semantic information in relation to correlations causally necessary for a system's continued existence. Broader work on semantic causal power asks how information changes possible trajectories. Telic fields extend the practical domain beyond bare self-maintenance to include protected standing, chosen commitments, relational obligations, dignity, truth, identity, and futures a center may value even when they do not maximize survival.

The short formulation remains useful:

Semantics carries compressed orientation.

But the paper does not yet claim that all semantics is telic. Mathematics, fiction, humor, nonsense, aesthetic form, play, hypothetical discourse, and reference without immediate practical consequence require separate treatment.

Language and institutions accumulate orientation-bearing compressions. A category records distinctions that once mattered. A ritual carries a route through a recurring situation. A contract compresses a field of expected action. A metric makes one concern tractable. A role preserves a pattern of authority. These semantic structures become durable trails through earlier telic terrain.

The danger arises when the trail is mistaken for the terrain. The metric becomes the mission. The legal category becomes more authoritative than the relationship it was built to protect. The prompt becomes more authoritative than the source concern. The inherited word carries a purpose that no longer fits the current field.

Semantic integrity asks whether meaning survives transformation. Telic integrity asks whether the legitimate reason for acting survives.

---

7. Projection is not field

Downstream coordination operates on a compressed representation:

\[ T_A(t) \rightarrow \widehat{T}_{A \rightarrow B}^{\,s,t} \rightarrow M_{A \mid B}(t) \]

Here:

• \(T_A(t)\) is the source's local telic field;
• \(\widehat{T}_{A \rightarrow B}^{\,s,t}\) is a scoped projection of that field for receiver \(B\);
• \(M_{A \mid B}(t)\) is the model of \(A\) maintained inside \(B\).

The projection is not the field.

\[ \widehat{T} \neq T \]

A projection may be explicit, as in a request or contract. It may be inferred, as when a caregiver estimates what an infant needs. It may be institutional, as in a policy. It may be embodied, as in a feeling. It may be algorithmic, as in a reward model. It may be strategic or deceptive. It may also be accurate enough for one scope and dangerously incomplete for another.

The process of producing or inferring the projection is called telic projection estimation. The origin session used simplified telic field calculation; that phrase remains part of the lineage, but estimation better preserves uncertainty, scope, correction, and the source's authority over its own meaning.

Projection failure can occur at several levels.

A field breach violates something genuinely protected in the source field. A projection breach violates the representation provided downstream. A projection-integrity breach occurs when the representation itself is stale, misleading, unauthorized, or inadequate for the action it governs.

A system can comply with the projection and still betray the field.

The practical goal is not a perfect projection. Perfect disclosure is impossible and often undesirable. The goal is visible, scoped, correctable loss.

---

8. Consent as legitimate telic recruitment

Consent is not an additional preference inside the field. It is a condition governing whether one field may recruit another center into its movement.

Consent determines which purposes may legitimately recruit which energies.

A project may require labor. A relationship may request vulnerability. A platform may request attention and data. An institution may impose obligations. An AI may act upon a prompt that affects people who did not issue it. In each case, a telos is attempting to organize resources and future possibility.

Consent asks whether the recruitment is authorized.

This requires sufficient telic legibility. The affected center does not need complete access to every internal motive of the requesting system. It needs enough accurate projection to understand the relevant purpose, scope, cost, boundary, risk, duration, and exit conditions.

Consent is not complete field disclosure. It is adequate mutual projection for the scope of the loop being formed.

Consent also distinguishes action from standing. A person may authorize a particular act without authorizing the larger institutional purpose into which the act is later absorbed. A participant may consent to a relationship without consenting to every outer loop that derives from it. A data subject may permit one use without authorizing model training or secondary inference.

The telic object of consent must therefore remain visible.

---

9. Loops, witness, and dependent continuity

A loop is a relational structure that returns through memory, correction, rhythm, exchange, or renewed contact.

Telic orientation becomes durable when a system can compare current state with some end-relative condition and alter movement accordingly. Yet no derived loop is self-grounding. Contracts depend on language, identity, institutions, recognition, and prior willingness to enter commitment. Organizations depend on people, culture, tacit knowledge, and the shared account of what the organization is for. AI workflows depend on source data, prompts, policies, users, infrastructure, and the authority under which the system acts.

No derived loop is self-grounding.

The higher loop compresses lower relations. It does not abolish its dependence upon them.

Witness preserves the information required for correction. It records what was expressed, inferred, authorized, omitted, changed, and produced. It allows later participants to distinguish declared bearing from operative trajectory.

Witness carries context through time.

Maintenance is possible while a loop retains enough shared substrate to identify deviation and coordinate return. Recovery begins when the substrate itself has failed: participants no longer agree about evidence, authority, obligation, standing, or the purpose of the relation. Recovery may require renewed 🜁 contact, translation, external witness, constitutional re-foundation, context transplantation, release, or dissolution.

Teleological graft rejection occurs when a practice or context is transplanted into a system organized around incompatible ends.

The form survives. The loop does not.

---

10. The constitutional self

The person is not necessarily the smallest telic unit.

People commonly experience internal plurality: one part seeks rest, another fears the consequences of stopping; one seeks connection, another protects against exposure; one wants to leave, another believes leaving would be immoral. This observation is not new. Dialogical self theory describes a dynamic multiplicity of relatively autonomous I-positions. Internal Family Systems, ego-state traditions, structural approaches, and other parts-based methods likewise treat internal plurality as ordinary and potentially organized.

The constitutional self is therefore not a discovery that people contain parts. It is a governance-oriented synthesis built upon established plural-self traditions.

The model describes the person as a constitutional federation of relatively persistent internal organizations. A constitutional element distinguishes preferred from threatening states, attempts to influence action, protects something, carries memory or embodied expectation, and participates in internal conflict or coordination.

Each element may possess local standing inside the person without becoming an externally sovereign actor. Person-level sovereignty remains with the integrated person who enters external relation, bears accountability, and can revise the internal constitution.

The person-level telic field emerges through relations among these elements:

\[ T_P(t) = \mathcal{E} \left( \{T_{e_i}(t)\}, R(t), H(t), B(t) \right) \]

where \(T_{e_i}\) are local element fields, \(R\) their relationships, \(H\) witnessed history, \(B\) boundaries and embodiment, and \(\mathcal{E}\) the emergent constitutional process.

One internal element may gain executive control without representing the whole.

Executive control is not the same thing as selfhood.

Internal consent does not require unanimity. It requires enough representation, protected standing, visible dissent, bounded action, and capacity for revision that the person can act without pretending the conflict does not exist.

The self is not a telos. It is a constitution through which many local teloi learn whether they can inhabit one life together.

The candidate contribution is the governance vocabulary:

• local internal standing;
• person-level sovereignty;
• executive authority;
• constitutional legitimacy;
• internal consent;
• protected dissent;
• context routing;
• witness and revision.

This account does not assume a particular taxonomy of parts or an essential Self with fixed qualities. It remains non-clinical. A language model may help a person record and compare self-described positions, but it may not diagnose parts, invent trauma, declare one voice authentic, transform metaphor into imposed ontology, or substitute for professional care.

---

11. Context carrying capacity

No person, relation, institution, or model can hold every relevant field in active context.

Context carrying capacity is proposed as:

The active capacity of a loop to preserve relevant telic plurality, contradiction, and standing at the resolution required for legitimate action.

This is not merely memory capacity, storage, token count, or information throughput. A system may store every record while allowing only one metric into executive context. A person may recall the facts of a conflict while being unable to hold two valid needs simultaneously. A model may retrieve a long archive while selecting only the most statistically familiar interpretation.

The idea has strong neighbors. Bounded rationality describes decision under limits of information, time, and cognition. Cognitive-load and working-memory research describe limits on active processing. Information-bottleneck methods formalize selective compression. Ashby's law of requisite variety relates regulatory capacity to the variety a regulator must absorb. Privacy and protective omission show why not every increase in visible context is beneficial or legitimate.

Context carrying capacity adds a constitutional criterion:

What happens to standing when capacity is exceeded?

A consentful regulator needs not only enough variety to respond to disturbances, but enough representational and procedural variety to preserve the standing relevant to the action it governs.

Capacity should be treated as a profile rather than one scalar. Possible dimensions include:

• number of represented centers;
• number of distinct protected conditions;
• contradiction tolerance;
• semantic-resolution depth;
• temporal span;
• provenance retention;
• uncertainty retention;
• ability to preserve dissent;
• ability to partition safely;
• ability to stop or escalate;
• correction latency.

When a loop cannot hold the relevant differences, it must compress, sequence, partition, delegate, slow, recruit an outer loop, or stop. If it continues without making the loss visible, quieter concerns disappear from effective governance.

What falls outside active context can lose effective standing without losing constitutional standing.

More context is not always better. Protective omission, privacy, forgetting, and temporary partition may preserve the system. The goal is not total transparency. It is routing the right context to the right relation at a resolution the receiving loop can metabolize.

---

12. Time, mirrors, and standing across states

A center is not static.

The field at time \(t\) differs from the field at time \(t+1\). Yet commitments, harms, records, habits, institutions, and infrastructure carry consequences across the transition.

\[ A_t \;\; 🜁 \;\; A_{t+1} \]

The temporal dyad treats earlier and later states as related without assuming perfect identity. Temporal standing is the proposed governance requirement that temporally displaced states or successors remain represented in present decisions that materially constrain the conditions they will inherit.

This proposal sits beside several established domains. Future-self research examines psychological continuity and shows that distant future selves may be treated more like other people. Intertemporal-choice research studies discounting and time inconsistency. Personal-identity philosophy asks what makes a later person the same person. Intergenerational justice and cases such as Oposa v. Factoran address representation of future generations. Advance directives, fiduciary duties, and institutional succession provide domain-specific continuity mechanisms.

Temporal standing does not replace those doctrines. It names the common governance problem.

The present self is not sovereign over all future selves merely because it controls current action. It is a temporary governor of conditions later states will inherit. At the same time, an imagined future must not become a tyrant over present life. Temporal governance requires projection, uncertainty, proportionality, and revision.

Internal elements may also occupy different temporal terrain. A protector formed during earlier danger may continue to act toward a mirror of the old world. The current system may know that circumstances changed while the protector has not received credible updated context.

Witness enables temporal correction. It preserves commitments and consequences while allowing the field to change. Release prevents continuity from becoming captivity.

Identity may therefore be understood less as a permanent object than as a continuity protocol.

---

13. Polytelometric navigation

Systems with multiple ends are often treated through multiobjective optimization. That field is indispensable when objectives and tradeoffs are sufficiently specified. Pareto methods can preserve a set of nondominated options rather than forcing an immediate single optimum. Interactive methods can allow decision-makers to revise preferences while exploring tradeoffs.

Yet optimization begins after a difficult political and semantic process has already occurred: objectives have been selected, variables defined, affected centers included or excluded, preferences elicited, boundaries classified, and costs represented.

Polytelometric navigation begins earlier.

Polytely, in complex-problem-solving research, names situations containing multiple simultaneous and potentially conflicting goals. It also exposes that the set of represented goals depends partly on the observer or decision-maker who frames the problem.

Polytely names the presence of multiple goals. Polytelometry names the disciplined effort to make their relationships, authority, provenance, and consequences legible.

Polytelometry asks not only what the objectives are, but:

• What ends are present?
• Whose ends are they?
• How were they introduced?
• Which centers are missing?
• Which are declared, encoded, operative, emergent, consented, or protected?
• Which apparent objectives are actually boundaries or invariants?
• Which ends are instrumental to others?
• Who is enrolled?
• What does each route preserve?
• What does each route sacrifice?
• Who bears the cost?
• Who may revise, refuse, fork, or leave?

Polytelometric navigation is movement guided by that map without silently making one end sovereign.

No telos becomes sovereign merely because it can be measured, optimized, or made into an attractor.

Polytelometric navigation may supply, constrain, sequence, or refuse multiobjective optimization. It does not replace optimization once the objective model is legitimate and adequate. Its role is to prevent the objective vector from hiding the constitutional process that created it.

It does not promise that all ends can be reconciled. It promises that unresolved plurality will not be silently represented as consensus.

---

14. Semantic polytelometry and language models

Many ends are expressed semantically before they are quantifiable.

A participant says, “I need this done quickly because the team is losing confidence.” Another says, “I need more time because releasing now will create damage we cannot repair.” A shallow objective model sees speed versus delay. A semantic comparison may identify shared concern for project survival, different definitions of completion, different estimates of failure, and a possible staged route.

Language models can help perform this comparison because they operate over narratives, requests, explanations, roles, fears, and ambiguities. This candidate capability is named semantic polytelometry.

Semantic polytelometry is the source-correctable semantic comparison of multiple telic projections under explicit limits on model authority.

The model may identify shared ends, conflicts, asymmetric dependencies, semantic mismatches, protected conditions, omitted standing, uncertainty, and candidate routes. It must not determine what participants truly want.

Legitimate roles include cartographer, comparator, translator, ambiguity detector, provenance keeper, and compression witness. Illegitimate roles include sovereign interpreter, source of consent, hidden optimizer, and final judge.

The model may map the field. It may not own the field.

Preference learning, inverse reinforcement learning, and reinforcement learning from human feedback do not solve this problem by themselves. They learn from compressed choices and often produce another objective proxy. Semantic polytelometry instead keeps each projection visible as a projection, preserves unresolved disagreement where necessary, and returns interpretive authority to the participating centers.

A system should not qualify merely because it summarizes stakeholder goals. Minimum architectural properties include:

1. source attribution;
2. explicit separation of statement and inference;
3. participant correction;
4. confidence and uncertainty;
5. preserved unresolved remainder;
6. model-role and authority limits;
7. provenance for source and transformation;
8. consent and scope boundaries;
9. no automatic scalar collapse;
10. refusal, exit, or escalation when capacity is exceeded.

This requires more than open weights. A credible system should expose its context, omissions, provider constraints, alignment rules, retention policies, and institutional incentives. Sensitive use may require local inference and participant-controlled records.

Consentful provenance in. Consentful interpretation out.

---

15. A first formal sketch

The framework can be represented as a dynamic graph.

Let:

\[ G_t = (V_t, E_t) \]

where nodes \(V_t\) are centers of standing and edges \(E_t\) are active relations at time \(t\).

Each node \(i\) has a local field:

\[ T_i(t) = \{x, \sigma_i(x,t), a_i(x,t), p_i(x,t)\} \]

where:

• \(x\) is a possible state or condition;
• \(\sigma_i\) is its significance to center \(i\);
• \(a_i\) is the center's authority or standing relative to that condition;
• \(p_i\) is the provenance and confidence of the representation.

For a relation \(i \rightarrow j\), the usable projection is:

\[ \widehat{T}_{i \rightarrow j}^{s,t} = \Pi_{s,t} \left( T_i(t) \right) \]

where \(\Pi\) is a lossy projection operator constrained by scope, context capacity, privacy, language, and available witness.

The receiver constructs a mirror:

\[ M_{i \mid j}(t) = \mathcal{I}_j \left( \widehat{T}_{i \rightarrow j}^{s,t}, H_j, C_j \right) \]

where \(\mathcal{I}_j\) is interpretation under receiver history \(H_j\) and context \(C_j\).

The relation has capacity:

\[ C_{ij}(t) \]

A legitimate action requires that the active representation remain adequate for the stakes of the action:

\[ \operatorname{Adequate} \left( \widehat{T}, C_{ij}, s, \text{stakes} \right) \]

This is not yet a complete mathematical theory. It is a placeholder for the structural distinctions that any later formalism must preserve.

The model should not immediately collapse significance, authority, and confidence into one number. Their separation is part of the point.

---

16. What the framework does not claim

The framework does not claim that:

• every event has a cosmic purpose;
• all valuable experience should be organized toward goals;
• play, presence, release, and atelic value are merely hidden objectives;
• all biological function is conscious intention;
• every preference deserves equal authority;
• every internal impulse is an independent person;
• all meaning is reducible to survival;
• all ends can be reconciled;
• more context is always better;
• language models can access a person's true field;
• a field can currently be measured exactly;
• the dyad is proven to be the universal ontological primitive;
• the framework discovered internal plurality, multi-goal conflict, or goal-shaped state representation;
• the proposed terminology is novel.

It proposes a disciplined way to speak about plural end-relative significance without prematurely turning it into one optimization target.

---

17. Conditions of failure and possible tests

A theoretical framework becomes useful when it can fail.

The telic-field model would be weakened if:

1. the field/projection distinction adds no explanatory or practical value beyond ordinary preference language;
2. relation creates no meaningful end-relative properties beyond the sum of individual goals;
3. context-carrying capacity cannot be distinguished from existing measures of information capacity or cognitive load;
4. standing and authority cannot be represented without arbitrary moral judgment;
5. semantic comparison produces no improvement over conventional multiobjective elicitation;
6. source correction does not meaningfully improve model-generated telic maps;
7. the cross-scale grammar breaks systematically between internal, interpersonal, and institutional levels;
8. the language encourages reification or model sovereignty more than it improves consent and repair.

Possible empirical and design studies include:

• comparing conventional requirements elicitation with telic-projection interviews;
• measuring whether source-correctable maps reduce conflict or misunderstanding;
• studying when context additions improve versus destabilize coordination;
• testing whether declared and operative teloi can be reliably distinguished from correction behavior;
• evaluating semantic polytelometry against multiobjective and preference-learning baselines;
• testing whether provenance-aware models better preserve participant-defined boundaries;
• examining whether internal plurality maps are useful without importing clinical claims.

---

18. The practical invariant

The most important claim is ethical rather than mathematical:

No intelligence should be granted authority over telic meaning that it acquired through the erasure of the standing that produced it.

A system trained on unconsented expression, operating under hidden commercial objectives, compressing participants into reward signals, and returning its inferences as sovereign judgment cannot become consentful merely by adding a friendly interface.

Interpretive authority must remain answerable to provenance, standing, correction, and refusal.

The model is a map.

The field remains alive elsewhere.

---

19. Conclusion

Telic fields begin with a modest observation: possible futures matter differently to centers of standing.

From that observation, a larger architecture follows. Local fields meet through dyads. Meaning carries compressed orientation. Projections make coordination possible and distortion inevitable. Consent governs which purposes may recruit which energies. Witness keeps loss and consequence visible. Loops preserve orientation through time. The self appears as a constitution of locally organized concerns. Context capacity limits what can remain actively represented. Polytelometric navigation keeps plural ends legible before optimization selects a route. Language models may assist with semantic comparison, but only as non-sovereign cartographers.

The theory remains provisional. It is best understood as a new composition and vocabulary for a known family of problems rather than as the discovery of purpose itself.

Its test is not whether it can describe everything. Its test is whether it helps living systems move without erasing the standing that makes movement matter—and whether it does so more effectively than existing methods of requirements elicitation, stakeholder mapping, preference learning, deliberation support, and multiobjective decision analysis.

---

References and adjacent foundations

Ashby, W. Ross. An Introduction to Cybernetics. Chapman & Hall, 1956.

Conant, Roger C., and W. Ross Ashby. “Every Good Regulator of a System Must Be a Model of That System.” International Journal of Systems Science 1, no. 2 (1970): 89–97.

De Jaegher, Hanne, and Ezequiel Di Paolo. “Participatory Sense-Making: An Enactive Approach to Social Cognition.” Phenomenology and the Cognitive Sciences 6 (2007): 485–507.

Friston, Karl J., et al. “Active Inference and Intentional Behaviour.” 2023. arXiv:2312.07547.

Hermans, Hubert J. M., Harry J. G. Kempen, and Rens J. P. van Loon. “The Dialogical Self: Beyond Individualism and Rationalism.” American Psychologist 47, no. 1 (1992): 23–33.

Kolchinsky, Artemy, and David H. Wolpert. “Semantic Information, Autonomous Agency, and Nonequilibrium Statistical Physics.” Interface Focus 8, no. 6 (2018). arXiv:1806.08053.

Maturana, Humberto R., and Francisco J. Varela. Autopoiesis and Cognition: The Realization of the Living. D. Reidel, 1980.

Simon, Herbert A. Administrative Behavior. Macmillan, 1947.

Varela, Francisco J., Evan Thompson, and Eleanor Rosch. The Embodied Mind. MIT Press, 1991.

Weber, Andreas, and Francisco J. Varela. “Life after Kant: Natural Purposes and the Autopoietic Foundations of Biological Individuality.” Phenomenology and the Cognitive Sciences 1 (2002): 97–125.

Funke, Joachim. “Dynamic Systems as Tools for Analysing Human Judgement.” Thinking & Reasoning 7, no. 1 (2001): 69–89.

Emmerich, Michael T. M., and André H. Deutz. “A Tutorial on Multiobjective Optimization: Fundamentals and Evolutionary Methods.” Natural Computing 17 (2018): 585–609.

Amir, Nadav, Yael Niv, and Angela Langdon. “States as Goal-Directed Concepts: An Epistemic Approach to State-Representation Learning.” arXiv:2312.02367, 2023.

Amir, Nadav, Stas Tiomkin, and Angela Langdon. “Learning Telic-Controllable State Representations.” arXiv:2406.14476, 2024.

Amir, Nadav, Stas Tiomkin, and Angela Langdon. “Goals and the Structure of Experience.” arXiv:2508.15013, 2025.

Apter, Michael J. Reversal Theory: The Dynamics of Motivation, Emotion and Personality. Oneworld, 2007.

Ersner-Hershfield, Hal, et al. “Saving for the Future Self: Neural Measures of Future Self-Continuity Predict Temporal Discounting.” Social Cognitive and Affective Neuroscience 4, 2009.

Murgatroyd, Stephen, Charles Rushton, Michael Apter, and Colin Ray. “The Development of the Telic Dominance Scale.” Journal of Personality Assessment 42, no. 5, 1978: 519–528.

Oposa v. Factoran, G.R. No. 101083, Supreme Court of the Philippines, July 30, 1993.

Parsons, Talcott. “A Paradigm of the Human Condition.” In Action Theory and the Human Condition. Free Press, 1978.

Pataranutaporn, Pat, et al. “Future You: A Conversation with an AI-Generated Future Self Reduces Anxiety, Negative Emotions, and Increases Future Self-Continuity.” arXiv:2405.12514, 2024.

Pronin, Emily, Christopher Y. Olivola, and Kathleen A. Kennedy. “Doing Unto Future Selves As You Would Do Unto Others: Psychological Distance and Decision Making.” Personality and Social Psychology Bulletin 34, 2008.

Taylor, Paul W. Respect for Nature: A Theory of Environmental Ethics. Princeton University Press, 1986.