Legitimacy

Type: Relation / Contestable Normative Justifiability

Placement note: Legitimacy belongs in Division VII because it evaluates the same institutional arrangements Authority, Power, Procedure, and Constitution structure, without itself granting, specifying, or governing them. It is the last entry in the institutional self-governance arc this division's dependency roadmap tracked: several already-published entries in this division name the gap it fills and explicitly decline to fill it themselves.

Working Definition: Legitimacy is the contestable normative property attributed to the existence, distribution, exercise, or continuation of a legitimacy_target: an Authority grant, a Power relation, or a particular Act performed under either. It holds, under a disclosed Standard, to the extent that the target's recorded grounds render it justifiable to the governed party it binds. A target's Authority may be formally granted, its acts procedurally Valid, and its criteria satisfied, while Legitimacy under a given Standard remains absent, partial, contested, or lost. A target may also rest on Power without any formal Authority grant at all, and wide agreement may still treat it as legitimate.

Function in the Grammar: Several published entries already name this gap without filling it. Authority's own Distinctions state that "an institution may grant authority formally and exercise it in ways that lack legitimacy." Validity's own Distinctions state that "a validly enacted rule may still be unjust or illegitimate." Procedure's own Semantic Constraints disclaim that conformance "establishes Legitimacy," and its Common Failure Modes name "procedural legitimacy laundering" directly. None of these entries supplies the vocabulary Legitimacy itself needs: what a legitimacy claim asserts, to whom it is owed, on what grounds it rests, under which Standard it is assessed, and how it can be partial, contested, or lost over time. Legitimacy supplies that vocabulary as an attributable, contestable Claim, borne by a specific legitimacy_target, an Authority grant, a Power relation, or a particular Act, addressed to the governed party it binds, resting on disclosed grounds, assessed under a disclosed Standard, and subject to the same epistemic discipline this grammar already requires of any other Claim.

Legitimacy's grounds are plural and disclosed rather than fixed by this grammar as a mandatory test. Recurring grounds include legality, whether the target's Authority or Power conforms to an applicable Rule, Constitution, or Procedure; justifiability, whether the basis for that Authority or Power is defensible by reasons assessable against a disclosed Standard, Value, or shared belief, independent of whether the governed party has in fact accepted them; and consent, whether the governed party has expressed or demonstrated acceptance through Participation, Recognition, or another attributable act. This entry treats that set as open and non-exhaustive rather than a checklist: no single ground is stipulated here as individually necessary or jointly sufficient, and different traditions, Standards, or forums may weigh legality, justifiability, and consent differently without any one weighting becoming this grammar's own doctrine.

Legitimacy is also a continuing, degreed relation rather than a snapshot. A target's Authority holds only within its own bounded Scope, Duration, and Conditions, and may lapse through expiry or unmet Conditions as readily as through Amendment or revocation; where it remains in force, a Validity finding is fixed within its stated kind, scope, and time. Legitimacy under a given Standard can erode, narrow, or become contested while the underlying Authority grant and its Validity remain completely unchanged: a governed party's assessment of whether a target's grounds remain justifiable can shift with new evidence, changed composition, or withdrawn acceptance, entirely apart from any formal act touching the target itself. See the Minimum Viable Test Case for a target whose Authority stays Valid throughout while assessments under two different Standards diverge, and for how a loss between two assessed times is represented as a derived transition rather than a status value.

This entry deliberately narrows Legitimacy's targets to Authority, Power, and Act, rather than treating an Institution or a particular Decision as a further, independent target type. Contemporary political-philosophy usage often treats institutions and their individual decisions as direct subjects of legitimacy assessment in their own right, not only as aggregates; the Stanford Encyclopedia of Philosophy's entry on legitimacy surveys both uses. This grammar's narrowing keeps Legitimacy reducible to the Authority and Power relations, and the Acts performed under them, that this encyclopedia already formalizes, rather than adding a fourth primitive target type whose own admission this entry does not separately argue for. What is lost is a direct way to assess an Institution's or a Constitution's Legitimacy without first decomposing it into its constituent Authority, Power, and Act instances; Core Relations states that reduction explicitly. A particular exercise, a specific enactment or an InstitutionalDecision, remains directly representable as an Act, distinct from the general Authority grant or Power relation it was performed under; a private, non-institutional decision is not, since it lacks the AuthoritySource an Act admissible as a legitimacy_target requires. See the Corporate governance row in Examples.

Formal Grammar Representation

<legitimacy-profile> ::= LEGITIMACY_PROFILE(
  id: <profile-id>,
  target: <legitimacy_target>,
  governed: <governed-party>,
  grounds: <ground-set>,
  scope: <scope>,
  interval: <time-interval>
)
  -- a purely descriptive record: which grounds are disclosed, for
     which target, governed party, scope, and interval. It carries no
     truth-value of its own; see Semantic Constraints

`l.target`, `l.governed`, `l.scope`, and `l.interval` denote the corresponding recorded field of the `LEGITIMACY_PROFILE` identified by `l`.

<legitimacy_target> ::= Authority(...) | Power(...) | ExercisingAct

ExercisingAct ::= Act(...) whose AuthoritySource resolves to an
  Authority(...) or Power(...) instance
  -- an Act admissible as a legitimacy_target must itself exercise an
     Authority grant or a Power relation; a purely private act with no
     such source is not a legitimacy_target. Where the Act is an
     institutional decision, it is more precisely an
     InstitutionalDecision(d, I), Decision's own qualified type, not a
     bare Decision(d); see the Corporate governance row in Examples
  -- the specific grant, causal relation, or particular exercise being
     assessed. Distinct from its holder: the Agent or Institution that
     holds or performed it, recoverable from the target's own actor,
     agent, or Actor argument. Ordinary talk of an Institution's or a
     Constitution's own Legitimacy names an aggregate claim about the
     Authority, Power, and Act instances that Institution's
     Constitution establishes and that Institution's acts exercise. It
     is not a further target type; see Core Relations

<governed-party> ::= the Agent, Agents, or community subject to the
  target's Authority, Power, or Act within the stated scope
  -- distinct from Accountability's Forum: a forum receives an answer
     from the target's holder. A governed party is who the target is
     exercised over, and to whom justification is owed. An Agent or
     community that recognizes, contests, or merely comments on a
     target without being subject to it is not a governed party for
     that target; see the International relations row in Examples

<ground-set> ::= { <ground>, ... }
<ground> ::=
    legality(basis: Rule | Constitution | Procedure)
  | justifiability(basis: Standard | Value | shared_belief)
  | consent(basis: Participation | Recognition | other attributable act)
  | <other disclosed ground>
  -- open and non-exhaustive. This grammar stipulates no fixed subset as
     individually necessary or jointly sufficient; see Semantic
     Constraints

GROUND_HOLDS(l: <profile-id>, g: <ground>, at: t) iff
  t ∈ l.interval
  the specific disclosed ground g is satisfied for the LEGITIMACY_PROFILE
  identified by l, at time t
  -- an independently assessable, Boolean, ground-by-ground fact. It is
     never itself "contested" or "partial": a claim that it holds is
     supportable or contestable through Claim and CONTESTS like any
     other Claim, not by GROUND_HOLDS acquiring a third value; see
     Formal Pattern
  -- t ∈ l.interval is required because at: t names the time the ground
     claim concerns, not the real-world clock time anyone happens to be
     asking the question. A profile recorded for [1987, 1987] cannot be
     asked whether its legality ground holds "at 2026": that is a
     question about a different, later profile, l_t1 in the Minimum
     Viable Test Case, not a retrospective reading of l_t0

MEETS_STANDARD(l: <profile-id>, s: Standard, respect: q, at: t)
  -- reused exactly from Standard's own entry, applied to the
     LEGITIMACY_PROFILE identified by l as the assessed Object. Boolean,
     as in Standard's own definition. q = s.respect throughout this
     entry; q is never a free choice made by whoever invokes
     MEETS_STANDARD, LEGITIMATE_UNDER, or LEGITIMACY_ASSESSMENT

LEGITIMATE_UNDER(profile: l, standard: s, under: g, at: t) iff
  t ∈ l.interval
  q = s.respect
  APPLIES_TO(Standard(s), LEGITIMACY_PROFILE(l), for: legitimacy_assessment,
             under: GoverningBasis(g), at: t)
  MEETS_STANDARD(l, Standard(s), respect: q, at: t)
  -- Boolean, matching MEETS_STANDARD's own type exactly. l identifies
     one specific LEGITIMACY_PROFILE, not a reconstruction from target,
     governed party, and scope alone; where several profiles share a
     target, governed party, and scope, l disambiguates which one this
     determination concerns. under: g names the GoverningBasis that
     makes Standard(s) applicable at all, per Standard's own APPLIES_TO
     signature; APPLIES_TO is never satisfied by a claimant's own say-so
     alone. t ∈ l.interval keeps at: t the time the profile's grounds
     concern, not an arbitrary later time a claimant might prefer to
     ask about; the Minimum Viable Test Case never assesses l_t0 at a
     time outside [1987, 1987] or l_t1 at a time outside [2026, 2026]
  -- a criterion-relative determination, not a free-standing fact. This
     grammar does not itself choose s or g; every conclusion discloses
     both. A claimant's own disclosure of which grounds it satisfies
     does not by itself make LEGITIMATE_UNDER hold, and neither does a
     Standard the claimant names without an independent GoverningBasis
     making it applicable; see Semantic Constraints

Claim(c, claimant: a, proposition: LEGITIMATE_UNDER(l, s, g, t),
      modality: normative, context: C)
  -- "the profile l is legitimate under s" is asserted as an ordinary
     Claim, in Claim's own sense, not a bespoke relation this entry
     invents. context: C is Claim's own generic context, distinct from
     this entry's <scope>; see Distinctions and Cross-References

ASSERTS(Agent(a), Claim(c))
CONTESTS(Agent(b), Claim(c), in: Scope(S))
  -- Claim's own predicate, reused exactly, unmodified

CHALLENGES_GROUND(Agent(b), Claim(c), ground: <ground>)
  -- names which specific disclosed ground, or which candidate ground
     absent from the claimant's own profile, the contest in CONTESTS
     targets. CONTESTS's own signature carries no ground-level detail;
     this relation supplies it without altering CONTESTS

CONTEST_FORUM(Claim(c), Forum(f) | no_constituted_forum)
  -- records whether a Forum with jurisdiction exists to hear the
     contest. Contestation under CONTESTS does not require one;
     Dispute's own addressability condition does

LEGITIMACY_ASSESSMENT(
  profile: l,
  standard: Standard(s),
  under: GoverningBasis(g),
  assessed_by: Agent(a),
  outcome: <legitimacy-outcome>,
  at: t
) iff
  t ∈ l.interval
  q = s.respect
  APPLIES_TO(Standard(s), LEGITIMACY_PROFILE(l), for: legitimacy_assessment,
             under: GoverningBasis(g), at: t)
  EVALUATES(Agent(a), LEGITIMACY_PROFILE(l), against: Standard(s),
            using: EvidenceSet(E), result: outcome, at: t)
  outcome = satisfied           => MEETS_STANDARD(l, Standard(s), respect: q, at: t)
  outcome = unsatisfied         => NOT MEETS_STANDARD(l, Standard(s), respect: q, at: t)
  outcome = partially_satisfied => s's own decision_rule states a
    partial-satisfaction condition distinct from MEETS_STANDARD's plain
    Boolean result; a Standard whose decision_rule is strictly binary
    cannot produce this outcome value
  -- under: g is not decorative. APPLIES_TO under that same basis is
     required for the assessment to hold; an EVALUATES fact performed
     against a Standard that does not APPLIES_TO under the recorded g
     does not satisfy this definition, and does not license
     LEGITIMACY_ASSESSMENT for that g
  -- assessed_by is always an Agent, matching EVALUATES's own signature
     and Accountability's own discipline that "Forum(f) does not confer
     agency ... merely by naming a collective recipient." Where a Forum
     performs the assessment, assessed_by names the Agent constituted to
     act for it, on_behalf_of that Forum in Accountability's own sense,
     not the Forum itself
  -- a named instance of Standard's own EVALUATES and MEETS_STANDARD,
     not an independent primitive
  -- t ∈ l.interval keeps the assessment's temporal subject the profile
     l actually describes; it does not constrain when assessed_by
     performs the EVALUATES act in the world, only which time the
     resulting outcome is about

<legitimacy-outcome> ::= satisfied | partially_satisfied | unsatisfied
  -- satisfied holds exactly when MEETS_STANDARD(l, s, q, t) is true
     under a Standard whose decision_rule is binary. unsatisfied holds
     exactly when that same MEETS_STANDARD is false. partially_satisfied
     is available only when s's own decision_rule separately defines a
     partial-satisfaction condition; a Standard that does not define one
     produces only satisfied or unsatisfied, collapsing to
     MEETS_STANDARD's own Boolean result. This is the only axis this
     entry defines as an enumerated value. Contestation of the
     underlying Claim is tracked through Claim's own
     InstitutionalDisposition (pending, contested, accepted, rejected,
     unresolved). Epistemic warrant of the assessment or the Claim it
     supports is tracked through Epistemic Status's WarrantStatus and
     ReviewStage. Neither is a peer of outcome, and neither is defined
     here; see Semantic Constraints

LEGITIMACY_LOST(standard: s, under: g, from_profile: l1, to_profile: l2) iff
  l1.target = l2.target
  l1.governed = l2.governed
  l1.scope = l2.scope
  l1.interval precedes l2.interval
  EXISTS a1, a2, t1, t2:
    LEGITIMACY_ASSESSMENT(profile: l1, standard: s, under: g,
      assessed_by: a1, outcome: satisfied, at: t1)
    AND LEGITIMACY_ASSESSMENT(profile: l2, standard: s, under: g,
      assessed_by: a2, outcome: unsatisfied, at: t2)
    AND t1 < t2
  -- a derived transition between two specific, named profiles sharing
     one target, governed party, and scope, assessed under the same
     Standard and the same GoverningBasis, not a peer of outcome. The
     two assessors a1 and a2 need not be identical; what must be shared
     is the Standard and the governing basis making it applicable, not
     who performed the assessment. Outcomes reached under different
     governing bases for the same Standard do not instantiate this
     relation either. l1 and l2 must be the same governed party;
     changing the governed party between two profiles does not
     instantiate this relation, see the Minimum Viable Test Case for a
     case this restriction rules out. "Undetermined" is not a listed
     outcome value either: it is the absence of any LEGITIMACY_ASSESSMENT
     record at all, not a status this grammar computes

Semantic Constraints

The expression LEGITIMACY_PROFILE(id: l, target: b, governed: g, grounds: R, scope: S, interval: T) is well-formed when all six fields are identifiable. LEGITIMACY_PROFILE is purely descriptive: it records which grounds are disclosed and, through GROUND_HOLDS, which of them are satisfied. It asserts nothing about whether the target is legitimate. LEGITIMATE_UNDER(profile: l, standard: s, under: g, at: t) is the criterion-relative determination; the syntax does not establish that it actually holds. Naming l rather than reconstructing target, governed party, and scope separately also disambiguates which profile is meant when more than one exists for the same target, governed party, and scope. LEGITIMACY_PROFILE(l) without an identified governed party is under-specified in the same way VALID(x) without a stated kind, criteria, and scope is under-specified in Validity's own entry: "this arrangement is legitimate," asserted without saying legitimate to whom or under which Standard, is an incomplete Claim.

A claimant discloses the grounds recorded in its own LEGITIMACY_PROFILE. Disclosure is necessary for auditability: it makes visible what is being claimed and on what basis. Disclosure alone does not make LEGITIMATE_UNDER hold. LEGITIMATE_UNDER(profile: l, standard: s, under: g, at: t) requires a Standard s that independently applies to profile l, APPLIES_TO(Standard(s), LEGITIMACY_PROFILE(l), ..., under: GoverningBasis(g), ...), and that the profile actually meets that Standard's own stated criteria, MEETS_STANDARD(...). The under: GoverningBasis(g) argument matters as much as the Standard itself: APPLIES_TO is Standard's own predicate, and Standard's own entry requires a governing basis, not merely a claimant's assertion, to make a Standard applicable to a given case. A claimant who names a Standard without an independent basis authorizing its use has not thereby made it apply. A claimant who discloses legality alone and a claimant who discloses legality, justifiability, and consent both produce a well-formed LEGITIMACY_PROFILE. They do not thereby produce the same LEGITIMATE_UNDER conclusion under a Standard that requires more than legality: the Standard, not the claimant's choice of what to disclose, determines whether the recorded profile suffices. A claimant cannot manufacture a favorable LEGITIMATE_UNDER result merely by disclosing a thin ground-set and naming a Standard with no supporting GoverningBasis; see the Minimum Viable Test Case for two Standards, one adopted and one merely proposed, reaching different conclusions from the same profile.

The ground-set is deliberately open. Legality, justifiability, and consent name recurring dimensions along which a Legitimacy Claim is typically supported or contested. They are not a mandatory three-part test this grammar imposes. A target may satisfy legality while justifiability remains contested. A target may command broad consent while resting on grounds a disclosed Standard finds unjustifiable. A target may hold a defensible justifiability basis while consent was never sought. Each of these is a coherent, reportable configuration of GROUND_HOLDS facts. None of them is, by itself, a computed LEGITIMATE_UNDER result absent a Standard; see the next paragraphs and Common Failure Modes for how satisfying one ground gets mistaken for satisfying all.

Legality, when disclosed as a ground, is supported by Validity and Procedure conformance without being identical to either. A target's Authority can be Valid and its governing Procedure fully conformed to while LEGITIMATE_UNDER(profile: l, standard: s, under: g, at: t) remains unsatisfied under any Standard s that also requires justifiability or consent, exactly as Procedure's own entry already states that compliance "does not exhaust substantive justification." Justifiability requires a disclosed Standard, Value, or articulated shared belief the governed party could in principle assess. An assertion that a basis is obviously justified, without naming what it could be tested against, does not satisfy this ground merely by being confidently made. Consent requires an attributable act. That act may take the form of Recognition, a Participation record showing effective involvement, an expressed agreement, or another disclosed act; these are alternatives, not a cumulative requirement, and their shared minimum condition is that some act by the governed party, not its absence, is what is being pointed to. A governed party's silence does not by itself satisfy the consent ground, especially where Participation would show VOICE_WITHOUT_DEMONSTRATED_EFFECT or no practical opportunity to object at all.

Claim(c, ...) and LEGITIMATE_UNDER are kept separate for the same reason Claim's own entry keeps a claim and truth separate: an asserted Legitimacy Claim may be contested, wrong, self-serving, or later revised without changing what LEGITIMATE_UNDER actually computes under a given Standard. A LEGITIMACY_ASSESSMENT's outcome is a further, distinct fact from that Claim's own InstitutionalDisposition (pending, contested, accepted, rejected, unresolved, in Claim's own sense): an institution can accept a Legitimacy Claim as its operative position, Claim's disposition, without any LEGITIMACY_ASSESSMENT under a disclosed Standard ever having been performed, and a rigorous LEGITIMACY_ASSESSMENT can find an outcome of partially_satisfied for a Claim an institution has, separately, accepted.

CONTESTS reuses Claim's own signature, CONTESTS(Agent(b), Claim(c), in: Scope(S)), unmodified. It does not gain a ground or forum argument of its own. Which grounds a contest targets is recorded through CHALLENGES_GROUND, and whether a Forum exists to hear it through CONTEST_FORUM, both attached to the same contested Claim rather than folded into CONTESTS itself. GROUND_HOLDS remains Boolean throughout this process: a challenged ground is represented as a separate Claim asserting GROUND_HOLDS(l, g, t), itself contestable through the same CONTESTS predicate, not as GROUND_HOLDS acquiring a contested value it was never typed to hold.

CONTESTS does not require an available Forum. Where an institution constitutes one with jurisdiction over the target and matter, a Legitimacy Claim may be contested through Dispute or Appeal. Where none exists, which is common for a state's basic constitutional order or an institution's own founding arrangement, contestation still occurs, through argument, protest, withdrawn compliance, or later historical judgment. This grammar does not treat the absence of a forum as evidence that no genuine question exists.

Legitimacy's continuing, degreed character is represented across four independent axes, not one status field. Whether a target's recorded profile satisfies a given Standard is LEGITIMACY_ASSESSMENT's outcome: satisfied, partially_satisfied, or unsatisfied, nothing else. Whether the assessment or the Claim it supports remains epistemically sound is tracked through Epistemic Status's WarrantStatus and ReviewStage, attached to that assessment or Claim, not to Legitimacy's own vocabulary. Whether the underlying Claim is being actively disputed is tracked through Claim's own InstitutionalDisposition. Whether a target held one outcome at an earlier assessed time and a different outcome at a later one, for the same governed party and Standard, is LEGITIMACY_LOST, a derived transition between two LEGITIMACY_ASSESSMENT records, not a peer of outcome. "Undetermined" is not a value in any of these vocabularies: it is what obtains when no LEGITIMACY_ASSESSMENT has been performed at all, an absence of record rather than a computed state.

Legitimacy is relative to scope, interval, and governed party in the same way Validity is relative to kind, scope, and time. A target's Legitimacy to one governed party, or within one scope or interval, does not transfer automatically to a different party, scope, or interval. A finding that a target's Authority is Legitimate to a founding generation does not by itself establish Legitimacy to a governed party that generation never included, and does not by itself instantiate LEGITIMACY_LOST for that later party either, since that relation requires the same governed party at both assessed times; see the Minimum Viable Test Case.

Formal Pattern

Legality disclosed, justifiability contested:

  b1 = Authority(α, actor: incumbent_official, act_type: enact_policy,
                 institution: State_I)
  g1 = residents_of(State_I)

  l1 = LEGITIMACY_PROFILE(
    id: l1,
    target: b1,
    governed: g1,
    grounds: {legality(basis: constitution_K)},
    scope: policy_enactment,
    interval: current_term
  )

  s_legality = Standard(subject_type: legitimacy_profile,
    respect: justifiability, criteria: {legality},
    decision_rule: legality alone is sufficient,
    scope: policy_enactment, purpose: legitimacy_assessment,
    version: legality_sufficient_v1)

  ADOPTS(incumbent_official, s_legality, for: State_I,
    under: office_interpretive_authority, at: current_term)

  c1 = Claim(claimant: incumbent_official,
             proposition: LEGITIMATE_UNDER(profile: l1, standard: s_legality,
                                      under: office_interpretive_authority,
                                      at: current_term),
             modality: normative, context: State_I)
  ASSERTS(incumbent_official, c1)

  GROUND_HOLDS(l1, legality(basis: constitution_K), at: current_term)
    -- holds. The official's Authority is Valid and traceable to
       constitution_K

  CONTESTS(opposition_coalition, c1, in: policy_enactment)
  CHALLENGES_GROUND(opposition_coalition, c1,
    ground: justifiability(basis: no_disclosed_rationale))
  CONTEST_FORUM(c1, no_constituted_forum)
    -- no forum has jurisdiction to adjudicate general political
       justifiability. justifiability was never part of l1's disclosed
       grounds; CHALLENGES_GROUND may name a ground the claimant's own
       profile omitted, not only one already disclosed

  No LEGITIMACY_ASSESSMENT under any disclosed Standard has been
  performed. What is known is ground by ground: legality holds for
  l1's disclosed profile, and c1 is now contested on a justifiability
  ground that profile never named. Calling b1 legitimate or
  illegitimate, without naming which grounds, whose profile, and which
  Standard that label rests on, would assert more than this record
  supports.

Grounds shifting for a different governed party, Authority unchanged:

  b2 = Authority(α, actor: founding_board, act_type: govern,
                 institution: Association_I, enacted_at: t0)
  g2_early = residents_at(t0)   -- overlapping with the founders
  g2_later = residents_at(t1)   -- fully turned over since t0

  l2_early = LEGITIMACY_PROFILE(id: l2_early, target: b2, governed: g2_early,
    grounds: {legality(basis: bylaws_1), consent(basis: founding_vote)},
    scope: governance, interval: [t0, t0])
  GROUND_HOLDS(l2_early, legality(basis: bylaws_1), at: t0): holds
  GROUND_HOLDS(l2_early, consent(basis: founding_vote), at: t0): holds,
    for g2_early specifically

  l2_later = LEGITIMACY_PROFILE(id: l2_later, target: b2, governed: g2_later,
    grounds: {legality(basis: bylaws_1)},
    scope: governance, interval: [t1, t1])
  GROUND_HOLDS(l2_later, legality(basis: bylaws_1), at: t1): holds
  consent(basis: founding_vote) is not disclosed as a ground of l2_later
    at all. g2_later never cast that vote. The ground does not carry
    forward to a different governed party merely because b2 is the same
    target.

  Authority(α) is unchanged between t0 and t1. No Amendment, no
  revocation. The delegation chain remains unbroken and Valid.

  Result: b2's Authority persists identically across t0 and t1. The
  disclosed ground-set supporting its LEGITIMACY_PROFILE does not:
  l2_early discloses two grounds for its governed party; l2_later
  discloses one, for a different governed party, without any formal act
  having touched b2 itself. Neither l2_early nor l2_later has been
  assessed under a disclosed Standard; what changed is which grounds
  are disclosed and for whom, not a computed outcome.

Core Relations

Typical Questions

• Which specific Authority grant, Power relation, or particular Act is the legitimacy_target, who is its holder, and to which governed party or parties is justification owed?
• Which grounds are disclosed: legality, justifiability, consent, or some other stated basis, and which remain undisclosed or unassessed?
• Does the target's formal grant, criteria-satisfaction, or procedural conformance hold, and is that being mistaken for LEGITIMATE_UNDER under some Standard?
• Is the ground-set being treated as a mandatory checklist rather than an open, non-exhaustive set of possible grounds?
• Under which disclosed Standard is LEGITIMATE_UNDER being assessed, and does that Standard actually APPLIES_TO this profile?
• What scope, interval, and governed party does this Legitimacy Claim actually cover, and is it being generalized beyond them?
• Is the outcome, warrant, contestation, and any loss over time being tracked as four separate facts, or collapsed into one status?
• Who is contesting the claim, on what grounds, and is there a constituted Forum with jurisdiction to hear the contest?
• Is the absence of such a Forum being treated as proof that no genuine legitimacy question exists?
• Is this Legitimacy Claim being confused with Accountability's directed answerability, or with Dispute's forum-addressability requirement?

Examples

Distinctions

Legitimacy ≠ Authority. Authority is the recognized power to perform binding acts, formally granted within a constitutional structure and traceable to a specific source. Legitimacy is the broader normative property of that grant, or of Power exercised without one, being justifiable to the governed party. An institution may grant Authority formally and exercise it in ways that lack Legitimacy; Power exercised without any Authority grant may still be widely regarded as legitimate. Authority answers who is recognized as having the right to act; Legitimacy asks whether that recognition, and its exercise, is itself justifiable.

Legitimacy ≠ Validity. Validity is the condition of satisfying stated criteria within a specified kind and scope; a validly enacted rule may still be unjust or illegitimate. Legitimacy may draw on a Validity finding to support its legality ground, but procedural or constitutional Validity does not by itself exhaust justifiability or consent, the two grounds Validity's own criteria-satisfaction test cannot reach.

Legitimacy ≠ Procedure. Procedure supplies a normative specification for a class of processes and states whether an actual Process conforms to it. Procedure's own Semantic Constraints already disclaim that conformance establishes Legitimacy: a fully conformant Procedure can produce a target whose justifiability or consent grounds remain absent, and a Procedure violation does not by itself settle that a target is illegitimate if other grounds independently hold.

Legitimacy ≠ Power, and ≠ Power's own "Legitimate Power." Power is the factual capacity to produce effects, whether or not authorized, recognized, or legitimate. Power's own entry defines a narrower technical predicate, Legitimate Power(p, I), holding exactly when Power and Authority are aligned for the same actor and domain. That alignment can hold, a target's Power and formally granted Authority may fully coincide, while this entry's broader Legitimacy still does not: an authority whose grant and enforcement power are perfectly aligned can remain unjustifiable to the party it governs.

Legitimacy ≠ Accountability. Accountability is a continuing, directed relation in which an attributable Agent must answer to a constituted Forum for a stated subject. It requires an identifiable Forum to be exercisable. Legitimacy assesses whether a target is justifiable to the governed party, addressed toward that governed party rather than a constituted recipient, and remains assessable and contestable even where no Forum for it exists at all.

Legitimacy ≠ Dispute. Dispute requires parties with standing, a specific matter, a Forum with jurisdiction, and a procedure before a conflict becomes institutionally addressable; without a Forum, a conflict is real but institutionally unaddressable in Dispute's own sense. A Legitimacy Claim does not require an available Forum to be a genuine, contestable claim: contestation through argument, protest, withdrawn compliance, or later historical judgment is a real form of CONTESTS, even where no institution exists that could hear it as a Dispute.

Legitimacy ≠ Standard. A Standard is an identifiable evaluative specification stating criteria, method, and scope for comparison. A disclosed Standard supplies the criteria against which LEGITIMATE_UNDER is assessed, but Legitimacy's ground-set is not reducible to any one Standard's criteria: justifiability may also rest on contested, undisclosed, or competing Value commitments a single Standard does not settle.

Legitimacy ≠ Justice. Justice concerns whether an arrangement's distribution of benefits, burdens, or treatment is fair on its own substantive terms. Legitimacy concerns whether a target's Authority or Power is justifiable to the governed party under a disclosed Standard, a narrower and differently grounded question. Much of the political-philosophy literature on legitimacy treats it as weaker than, or independently grounded from, justice; the Stanford Encyclopedia of Philosophy surveys this distinction directly. A target that is LEGITIMATE_UNDER some disclosed Standard may still be unjust by a further substantive standard this entry's justifiability ground does not itself supply. That ground names a disclosed Standard, Value, or shared belief as the test; it does not stipulate that the test must be a theory of justice, and this entry does not collapse the two questions into one.

Legitimacy ≠ Acceptance, recognition, or belief in legitimacy. Max Weber's account of legitimate domination treats legitimacy primarily as a sociological fact about belief: whether the governed in fact regard an order as legitimate, largely apart from whether that belief is itself warranted; see Intellectual Lineage. This entry's Legitimacy is instead a contestable Claim, assessable and challengeable on disclosed grounds under a disclosed Standard. A governed party's acceptance, Recognition, or stated belief that a target is legitimate may support the consent ground, but does not by itself satisfy LEGITIMATE_UNDER for a Standard that also requires legality or justifiability, and a governed party's belief can itself be manufactured, coerced, or uninformed; see Manufactured or coerced consent among the Common Failure Modes.

Common Failure Modes

Minimum Viable Test Case

Institution: Maple Court Homeowners Association
Constitution: bylaws_1, validly enacted in 1987, never amended

b = Authority(α, actor: current_board, act_type: assess_dues_and_enforce_covenants,
              institution: MapleCourtHOA)
  -- traceable through an unbroken delegation chain to bylaws_1.
     Authority(α) holds and is Valid under the bylaws' own terms and
     remains within its own Scope, Duration, and Conditions throughout

g_t0 = residents_at(1987)   -- overlaps with the founding vote
g_t1 = residents_at(2026)   -- fully turned over since 1987

s_legality = Standard(
  subject_type: legitimacy_profile, respect: justifiability,
  criteria: {legality},
  method: check whether the legality ground holds,
  decision_rule: legality alone is sufficient,
  scope: dues_and_covenant_enforcement, purpose: legitimacy_assessment,
  version: legality_sufficient_v1
)

s_beetham = Standard(
  subject_type: legitimacy_profile, respect: justifiability,
  criteria: {legality, justifiability, consent},
  method: check whether all three grounds hold,
  decision_rule: all three required; any missing or unsatisfied ground
    yields unsatisfied,
  scope: dues_and_covenant_enforcement, purpose: legitimacy_assessment,
  version: beetham_conjunctive_v1
)

l_t0 = LEGITIMACY_PROFILE(
  id: l_t0,
  target: b, governed: g_t0,
  grounds: {legality(basis: bylaws_1),
            justifiability(basis: founding_meeting_rationale_1987),
            consent(basis: founding_vote_1987)},
  scope: dues_and_covenant_enforcement, interval: [1987, 1987]
)
GROUND_HOLDS(l_t0, legality(basis: bylaws_1), at: 1987): holds
gc_legality_t0 = Claim(claimant: founding_board,
  proposition: GROUND_HOLDS(l_t0, legality(basis: bylaws_1), at: 1987),
  modality: factual, context: MapleCourtHOA)
ev_legality_t0 = Evidence(e: ev_legality_t0, offered_for: gc_legality_t0,
  institution: MapleCourtHOA, profile: bylaws_1_delegation_record)

GROUND_HOLDS(l_t0, justifiability(basis: founding_meeting_rationale_1987),
  at: 1987): holds
  -- the founding meeting's recorded minutes state a rationale for the
     dues structure, never revisited or withdrawn
gc_justifiability_t0 = Claim(claimant: founding_board,
  proposition: GROUND_HOLDS(l_t0,
    justifiability(basis: founding_meeting_rationale_1987), at: 1987),
  modality: factual, context: MapleCourtHOA)
ev_justifiability_t0 = Evidence(e: ev_justifiability_t0,
  offered_for: gc_justifiability_t0, institution: MapleCourtHOA,
  profile: founding_meeting_minutes_1987)

GROUND_HOLDS(l_t0, consent(basis: founding_vote_1987), at: 1987): holds,
  for g_t0 specifically
gc_consent_t0 = Claim(claimant: founding_board,
  proposition: GROUND_HOLDS(l_t0, consent(basis: founding_vote_1987),
    at: 1987), modality: factual, context: MapleCourtHOA)
ev_consent_t0 = Evidence(e: ev_consent_t0, offered_for: gc_consent_t0,
  institution: MapleCourtHOA, profile: founding_vote_tally_1987)

l_t1 = LEGITIMACY_PROFILE(
  id: l_t1,
  target: b, governed: g_t1,
  grounds: {legality(basis: bylaws_1)},
  scope: dues_and_covenant_enforcement, interval: [2026, 2026]
)

ADOPTS(current_board, s_legality, for: MapleCourtHOA,
  under: board_interpretive_authority, at: 2026)
  -- the board's Authority(α) includes interpreting its own governing
     documents; adopting s_legality for its own self-assessment falls
     within that Authority, though the adoption itself remains
     contestable, see below

PROPOSES(residents_petition_2026, s_beetham,
  for: legitimacy_assessment, at: 2026)
  -- proposed, not adopted by any authority within MapleCourtHOA. No
     provision in the association's own bylaws makes s_beetham
     applicable; its applicability rests on the general
     political-legitimacy convention residents_petition_2026 invokes,
     itself a further, contestable claim

c1 = Claim(claimant: current_board,
           proposition: LEGITIMATE_UNDER(profile: l_t1, standard: s_legality,
                                    under: board_interpretive_authority,
                                    at: 2026),
           modality: normative, context: MapleCourtHOA)
ASSERTS(current_board, c1)

GROUND_HOLDS(l_t1, legality(basis: bylaws_1), at: 2026): holds
gc_legality_t1 = Claim(claimant: current_board,
  proposition: GROUND_HOLDS(l_t1, legality(basis: bylaws_1), at: 2026),
  modality: factual, context: MapleCourtHOA)
ev_legality_t1 = Evidence(e: ev_legality_t1, offered_for: gc_legality_t1,
  institution: MapleCourtHOA, profile: bylaws_1_delegation_record)

consent(basis: founding_vote_1987) is not disclosed as a ground of l_t1
  at all. g_t1 never cast that vote. The ground does not carry forward
  to a different governed party merely because b is the same target.

CONTESTS(residents_petition_2026, c1, in: dues_and_covenant_enforcement)
CHALLENGES_GROUND(residents_petition_2026, c1,
  ground: consent(basis: never_sought))
CHALLENGES_GROUND(residents_petition_2026, c1,
  ground: justifiability(basis: no_disclosed_rationale_for_current_dues_level))
CONTEST_FORUM(c1, no_constituted_forum)
  -- MapleCourtHOA's own governing documents specify no forum with
     jurisdiction to hear whether the 1987 bylaws' continued application
     to a fully turned-over resident base is justifiable. Both
     challenged grounds go beyond l_t1's own disclosed profile, which
     named only legality

Authority(α) throughout: no Amendment, no revocation, and no lapse of
Scope, Duration, or Conditions. The delegation chain remains unbroken
and Valid at both t0 and t1.

Result:
  b's Authority is Valid identically at t0 and t1. The disclosed
  ground-set supporting its LEGITIMACY_PROFILE is not identical: l_t0
  discloses legality, justifiability, and consent for its governed
  party; l_t1 discloses legality alone for a different governed party,
  and c1 is now contested on two grounds neither of which l_t1's own
  profile disclosed, all without any formal act touching Authority(α)
  itself.

  No LEGITIMACY_ASSESSMENT under any disclosed Standard has been
  performed for either l_t0 or l_t1 yet. Calling l_t0 or l_t1
  "legitimate" or "illegitimate" without naming a Standard would assert
  more than this record supports. What the record supports so far is
  the ground-by-ground picture above.

  The petition also has no constituted Forum to bring it before. That
  absence does not resolve whether c1's challenged grounds are
  justifiable. It means the contest, so far, has nowhere institutional
  to go, exactly the configuration this entry's Semantic Constraints
  distinguish from Dispute's own Forum requirement.

Two Standards are now applied to l_t1, resolving what the record above
deliberately left open:

MEETS_STANDARD(l_t1, s_legality, respect: justifiability, at: 2026): true
  -- l_t1's only disclosed ground, legality, holds, and s_legality's
     decision_rule requires nothing further
APPLIES_TO(Standard(s_legality), l_t1, for: legitimacy_assessment,
  under: board_interpretive_authority, at: 2026)
LEGITIMATE_UNDER(profile: l_t1, standard: s_legality,
  under: board_interpretive_authority, at: 2026): holds
  -- both premises of the iff are satisfied; c1's own proposition,
     asserted under s_legality, is therefore true

EVALUATES(model_assistant, l_t1, against: s_legality,
  using: EvidenceSet({ev_legality_t1}), result: satisfied, at: 2026)
  -- ev_legality_t1 supports gc_legality_t1's proposition that
     GROUND_HOLDS(l_t1, legality(basis: bylaws_1), at: 2026); the
     ground fact and the Evidence supporting it remain distinct
LEGITIMACY_ASSESSMENT(profile: l_t1, standard: s_legality,
  under: board_interpretive_authority, assessed_by: model_assistant,
  outcome: satisfied, at: 2026)
  -- outcome = satisfied because MEETS_STANDARD(l_t1, s_legality, ...)
     is true and s_legality's decision_rule is binary

MEETS_STANDARD(l_t1, s_beetham, respect: justifiability, at: 2026): false
  -- l_t1 discloses no justifiability ground and no consent ground;
     s_beetham's conjunctive decision_rule requires both
APPLIES_TO(Standard(s_beetham), l_t1, for: legitimacy_assessment,
  under: general_political_legitimacy_convention, at: 2026)
  -- this APPLIES_TO claim is only as strong as residents_petition_2026's
     own proposal; MapleCourtHOA has adopted no governing basis making
     s_beetham applicable, unlike s_legality
LEGITIMATE_UNDER(profile: l_t1, standard: s_beetham,
  under: general_political_legitimacy_convention, at: 2026): does not
  hold
  -- MEETS_STANDARD fails, so the iff's right-hand side fails regardless
     of APPLIES_TO

EVALUATES(model_assistant, l_t1, against: s_beetham,
  using: EvidenceSet({ev_legality_t1}), result: unsatisfied, at: 2026)
  -- the same ev_legality_t1 is offered again; s_beetham's conjunctive
     decision_rule still fails since no Evidence supporting a
     justifiability or consent ground for l_t1 exists to offer
LEGITIMACY_ASSESSMENT(profile: l_t1, standard: s_beetham,
  under: general_political_legitimacy_convention,
  assessed_by: model_assistant, outcome: unsatisfied, at: 2026)

Comparison at t0, same Standard:

MEETS_STANDARD(l_t0, s_beetham, respect: justifiability, at: 1987): true
  -- l_t0 discloses legality, justifiability, and consent, and all
     three GROUND_HOLDS facts above hold; s_beetham's conjunctive
     decision_rule is satisfied
APPLIES_TO(Standard(s_beetham), l_t0, for: legitimacy_assessment,
  under: general_political_legitimacy_convention, at: 1987)
LEGITIMATE_UNDER(profile: l_t0, standard: s_beetham,
  under: general_political_legitimacy_convention, at: 1987): holds

EVALUATES(model_assistant, l_t0, against: s_beetham,
  using: EvidenceSet({ev_legality_t0, ev_justifiability_t0, ev_consent_t0}),
  result: satisfied, at: 1987)
  -- each Evidence record supports one ground's GROUND_HOLDS Claim; all
     three grounds s_beetham's conjunction requires are supported
LEGITIMACY_ASSESSMENT(profile: l_t0, standard: s_beetham,
  under: general_political_legitimacy_convention,
  assessed_by: model_assistant, outcome: satisfied, at: 1987)

LEGITIMACY_LOST(standard: s_beetham,
  under: general_political_legitimacy_convention,
  from_profile: l_t0, to_profile: l_t1)
does not hold: its own premises require l_t0.governed = l_t1.governed,
and g_t0 and g_t1 do not overlap, so that premise fails before the
satisfied/unsatisfied comparison is even reached. Treating this pair as
a loss would repeat the Scope, interval, or party overgeneralization
failure mode named above. A genuine LEGITIMACY_LOST instance would
instead require two profiles sharing one governed party, one satisfied
at an earlier time and the other unsatisfied at a later one; this case
does not construct that instance.

Result:
  Under s_legality, adopted by the board for its own governance,
  LEGITIMATE_UNDER(l_t1, s_legality, board_interpretive_authority, 2026)
  holds: legality is all s_legality requires, and legality holds. Under
  s_beetham, only proposed by the petitioners and adopted by no
  authority within MapleCourtHOA, LEGITIMATE_UNDER(l_t1, s_beetham,
  general_political_legitimacy_convention, 2026) does not hold: consent
  and justifiability are both required and neither is disclosed. Both
  results are well-formed and both stand at once; they disagree because
  they are different statements with different standard and governing-
  basis arguments, not because either is mistaken. What changed between
  1987 and 2026 is which governed party is being asked about and what
  that party's profile discloses, not a single continuant's assessment
  moving from satisfied to unsatisfied under one Standard and governing
  basis.

Intellectual Lineage

This entry's three-part, non-exhaustive ground-set draws on David Beetham's account in The Legitimation of Power (1991), which separates a power relation's legitimacy into three independent dimensions: conformity to established rules, justifiability of those rules by reference to shared beliefs, and legitimation demonstrated through the expressed consent of subordinates. Beetham argues that a power relation is fully legitimate only where all three hold, and that each can fail independently, producing recognizably different kinds of illegitimacy: unlawful, unjustified, or delegitimated through withdrawn consent. This entry adapts his decomposition as an open typology of possible grounds rather than adopting his stronger three-part necessity claim: LEGITIMATE_UNDER(...) here does not stipulate that all three, or any fixed subset, must hold, leaving that further, contestable determination to whichever disclosed Standard performs the assessment, consistent with this entry's own discipline against adopting one theory of legitimacy as doctrine.

This entry's Legitimacy is deliberately distinct from Max Weber's account of legitimate domination (Herrschaft, in Economy and Society, 1922), which Authority's and Power's own Intellectual Lineage sections already draw on for the authority-power distinction itself. Weber treats legitimacy primarily as a sociological fact about belief: whether subordinates in fact regard an order as legitimate, largely apart from whether that belief is itself justified. This entry's Legitimacy is instead itself a contestable Claim, assessable and challengeable on disclosed grounds under a disclosed Standard. A governed party's belief can be manufactured, coerced, or uninformed, in which case Weber's descriptive criterion and this entry's grounds diverge sharply; see Manufactured or coerced consent among the Common Failure Modes and Legitimacy ≠ Acceptance, recognition, or belief in legitimacy among the Distinctions.

This entry's justifiability ground, and its refusal to treat legality or consent as sufficient by default, is closer to Bernard Williams' Basic Legitimation Demand (in In the Beginning Was the Deed: Realism and Moralism in Political Argument, 2005; see also the Stanford Encyclopedia of Philosophy's entry on Bernard Williams and Edward Hall, "The Basic Legitimation Demand: A Response to the Amoralist") than to any single theory of justice. Williams argues that a state's power over a subject requires an intelligible justification offered to that subject, a "critical theory test" distinct from asking whether the arrangement is independently just: a coercer who could not offer such a justification to the coerced, only a justification to third parties, fails the demand regardless of the arrangement's substantive fairness. This entry generalizes that structure, justification owed to the governed party rather than to an observer, without adopting Williams' specific critical-theory test as this grammar's own doctrine: a disclosed Standard may embody Williams' test, a theory of justice, or another normative framework entirely, and this entry does not choose among them. Williams' demand rejects legality as a default sufficient justification; it is not a reason no disclosed Standard could ever adopt legality alone, as s_legality does in the Minimum Viable Test Case, when a governing basis makes that Standard applicable. See Legitimacy ≠ Justice among the Distinctions for the boundary this departure preserves.

Cross-References

Required

• Authority — a primary target type whose formal grant Legitimacy does not reduce to
• Power — the other primary target type; Legitimacy applies to unauthorized as well as authorized capacity
• Act — the third target type, admissible only when its own AuthoritySource resolves to an Authority or Power instance; a particular institutional exercise, not the ongoing grant
• Validity — may support the legality ground without exhausting justifiability or consent
• Procedure — its own Semantic Constraints already disclaim that conformance establishes Legitimacy
• Standard — supplies the criteria, applicability test, and aggregation rule LEGITIMATE_UNDER and LEGITIMACY_ASSESSMENT are relative to; without a disclosed Standard, no outcome is computed
• Claim — the assertoric form every Legitimacy statement takes, and the vehicle through which both LEGITIMATE_UNDER claims and individual GROUND_HOLDS challenges are asserted and contested

Consequential

• Institution — the aggregate subject when Legitimacy is discussed at the level of a whole governing arrangement; reduces to its constituent Authority, Power, and Act instances rather than being a further legitimacy_target type
• Constitution — the aggregate subject alongside Institution; its non-derogable core exists partly to protect that aggregate Legitimacy from erosion through amendment
• Value — orients why a justifiability ground matters without settling whether it is met
• Accountability — a distinct directed relation; see Distinctions
• Dispute — a channel through which a Legitimacy Claim may be contested when a Forum with jurisdiction exists
• Appeal — a bounded, Forum-based variant of that same channel against an appealable disposition
• Jurisdiction — bounds which Forum, if any, may adjudicate a contested Legitimacy Claim, when a Forum exists at all
• Judgment and Decision — may resolve a Legitimacy dispute where an authorized Forum exists, a possible institutional consequence rather than evidence for or against LEGITIMATE_UNDER

Evidential

• Evidence — supports or undermines a claimed GROUND_HOLDS fact, exactly as Standard's own Evidence relation supports a MEETS_STANDARD conclusion
• Epistemic Status — supplies WarrantStatus and ReviewStage, the vocabulary this entry reuses to track whether a LEGITIMACY_ASSESSMENT or the Claim it supports remains epistemically sound, distinct from outcome and from Claim's own InstitutionalDisposition
• Recognition — may evidence the consent ground when the governed party accepts the target's holder
• Participation — its own coordinates, especially voice without demonstrated effect, may evidence consent's thinness even where formally elicited