Documentation

Linglib.Theories.Semantics.Modality.Temporal

Temporal Modal Evaluation #

@cite{abusch-1997} @cite{condoravdi-2002} @cite{kratzer-2012} @cite{werner-2006}

Modal bases and ordering sources are functions of both world and time (@cite{kratzer-2012} Ch. 4, @cite{condoravdi-2002}). This module extends Kratzer.lean with time-indexed conversational backgrounds and derives the static (time-independent) versions as a special case.

Core Extension #

Kratzer.lean defines ConvBackground := World → List (BProp World). @cite{kratzer-2012} Ch. 4 argues that this should be World → Time → List (BProp World): the modal base and ordering source can vary with the temporal perspective.

This distinction matters for:

Key Result #

temporal_eq_static: temporal modal evaluation reduces to standard Kratzer evaluation when the conversational backgrounds are time-independent.

Time-indexed conversational backgrounds #

@[reducible, inline]

A time-indexed conversational background maps (world, time) pairs to sets of propositions. This is the book's core extension: f(w,t) and g(w,t).

At different times, the same world may yield different sets of relevant facts (modal base) or normative standards (ordering source).

Equations
Instances For
    @[reducible, inline]

    Time-indexed modal base: what facts are relevant at (w, t).

    Equations
    Instances For
      @[reducible, inline]

      Time-indexed ordering source: what standards apply at (w, t).

      Equations
      Instances For

        Fix a time t to obtain a standard (time-independent) conversational background. This is the "temporal perspective" operation: evaluating the modal at a specific time.

        Equations
        Instances For

          Lift a time-independent background to a trivially temporal one (constant across time).

          Equations
          Instances For

            Lifting then fixing at any time recovers the original background.

            Temporal modal evaluation #

            Modal necessity evaluated at a world-time pair.

            ⟦must p⟧(w, t) = ∀w' ∈ Best(f(w,t), g(w,t), w). p(w')

            Equations
            • One or more equations did not get rendered due to their size.
            Instances For

              Modal possibility evaluated at a world-time pair.

              ⟦might p⟧(w, t) = ∃w' ∈ Best(f(w,t), g(w,t), w). p(w')

              Equations
              • One or more equations did not get rendered due to their size.
              Instances For

                Temporal ↔ Static bridge: temporal modal evaluation reduces to standard Kratzer when the backgrounds are time-independent.

                Temporal duality: □ₜp ↔ ¬◇ₜ¬p.

                Historical accessibility #

                Worlds share history up to time t: they agree on all facts prior to t. This gives the "branching futures" model: the past is settled, the future is open.

                Historical necessity: p holds at all worlds sharing w's history up to t.

                The history function (a TemporalConvBackground) serves as the modal base; the ordering source is empty (pure accessibility, no ranking).

                Equations
                Instances For

                  With empty history (no shared past), all worlds are accessible: historical necessity collapses to universal quantification.

                  Epistemic change over time #

                  theorem Semantics.Modality.Temporal.evidence_monotone {Time : Type u_1} (f : TemporalModalBase Time) (t₁ t₂ : Time) (p : BProp Attitudes.Intensional.World) (w : Attitudes.Intensional.World) (hSub : qf w t₁, q f w t₂) (hNec : temporalNecessity f (fun (x : Attitudes.Intensional.World) (x_1 : Time) => []) t₁ p w = true) :
                  temporalNecessity f (fun (x : Attitudes.Intensional.World) (x_1 : Time) => []) t₂ p w = true

                  Evidence monotonicity: if the evidence at t₁ is a subset of the evidence at t₂ (more evidence was gathered), then what was necessary at t₁ (less evidence) is still necessary at t₂ (more evidence).

                  More evidence → fewer accessible worlds → at least as many necessities.

                  Future-as-modal (Ch 4 bridge) #

                  "Will" as a circumstantial modal with empty ordering source. The future marker contributes modal force (necessity over a circumstantial base) without adding normative/stereotypical ranking.

                  This captures the unitary modal analysis: "will" does not decompose cleanly into force × flavor.

                  Equations
                  Instances For

                    Future-as-modal with empty ordering = simple necessity over the circumstantial base.