Demonstrations #
A demonstration: an act that presents an individual in a context.
@cite{kaplan-1989} §IX: "A demonstration is, roughly, a visual presentation of a local object discriminated by a pointing." The demonstration provides the manner of presentation — different demonstrations of the same object constitute different modes of presentation.
- demonstratum : C → Option E
Given a context, yield the demonstrated object (if any).
- description : String
Description of the manner of presentation (for documentation).
Instances For
A demonstration succeeds in context c if it yields a demonstratum.
Equations
- d.succeeds c = (d.demonstratum c ≠ none)
Instances For
A vacuous demonstration: always fails (hallucination, empty pointing).
Equations
- Semantics.Reference.Demonstratives.vacuousDemonstration = { demonstratum := fun (x : C) => none, description := "vacuous" }
Instances For
A constant demonstration: always yields the same individual.
Equations
- Semantics.Reference.Demonstratives.constantDemonstration e desc = { demonstratum := fun (x : C) => some e, description := desc }
Instances For
True Demonstratives #
A true demonstrative: a demonstrative expression completed by a demonstration.
@cite{kaplan-1989} §XV: "that [pointing at Venus in the evening] is bright." The demonstration completes the character; without it, "that" is incomplete.
- demonstration : Demonstration C E
The associated demonstration
Optional sortal restriction (e.g., "that planet")
Instances For
The character of a true demonstrative: in each context, rigidly designate the demonstratum (if the demonstration succeeds).
If the demonstration fails, the content is a trivially rigid intension returning a default entity.
Equations
- td.character c = match td.demonstration.demonstratum c with | some e => Core.Intension.rigid e | none => Core.Intension.rigid default
Instances For
Principle 2 (@cite{kaplan-1989} §XVI): Complete demonstratives are directly referential — at every context, their content is rigid.
Different demonstrations in different contexts yield different contents.
If the demonstratum differs between two contexts, the contents differ.
Demonstrative Frege Puzzle (§IX.ii-iv) #
The Demonstrative Frege Puzzle: "that [Hes] = that [Phos]" is informative even though both demonstrations pick out the same object (Venus).
The two demonstrations constitute different manners of presentation. The identity is informative because the characters differ (different demonstrations), even though the contents coincide.
- demo₁ : TrueDemonstrative C W E
First demonstrative (e.g., "that" + pointing at evening star)
- demo₂ : TrueDemonstrative C W E
Second demonstrative (e.g., "that" + pointing at morning star)
- context : C
Context in which both demonstrations succeed with the same object
First demonstration yields the shared object
Second demonstration yields the shared object
The demonstrations themselves differ (different descriptions)
Instances For
In a Frege puzzle configuration, the contents coincide: both demos yield the same rigid intension at the shared context.
Bridges #
Wrap a TrueDemonstrative as a ReferringExpression for integration with
the Reference/Basic.lean types.
Equations
Instances For
Bridge to RSA Reference Games #
Demonstrations connect to RSA reference games:
- A demonstration is analogous to a pointing gesture in a reference game
- L0 (literal listener) ≈ attributive interpretation (resolves via description)
- S1 (pragmatic speaker) ≈ referential interpretation (chooses expression to single out the intended referent)
See RSA.FrankGoodman2012 for a reference game implementation.