Kocurek, Jerzak & Rudolph 2020: Against Conventional Wisdom (c-monsters)

@cite{kocurek-jerzak-rudolph-2020}

Kocurek, A. W., Jerzak, E. & Rudolph, R. E. (2020). Against conventional wisdom. Philosophers' Imprint 20(22), 1–27.

Defining commitment

KJR argue that what they call Conventional Wisdom — the thesis that "truth at a scenario (counterfactual or otherwise) is evaluated relative to our (or the speaker's) actual linguistic conventions, even if those conventions diverge from the ones adopted in that scenario" (KJR p.2) — is FALSE. Certain embedded clauses can characterize conditions on the linguistic conventions in force rather than on the non-linguistic world. KJR call the embedding expressions c-monsters.

To accommodate c-monsters, KJR propose a substantial revision of the semantic architecture: replace worlds (as points of evaluation and objects of propositional attitudes) with world-convention pairs. Standard subject-predicate sentences then express something like diagonal content — propositions about whatever the conventions pick out for each predicate.

This stub encodes KJR's commitment to (i) rejecting Conventional Wisdom, (ii) world-convention pairs as evaluation points, (iii) the c-monstrous behaviour of certain embedding expressions.

K-G's disagreement (paper §4, p.24-25)

K-G accepts the same c-monster data (paper examples (4), (29)-(32) on Pluto-as-planet conditionals) but argues for a much less revisionary analysis. The c-monstrous behaviour is explained by positing a covert diagonalizer † that operates on the existing quotative interpretation function ⟨*⟩ — no replacement of worlds with world-convention pairs required. K-G's covert apparatus 𝔐 + † preserves Conventional Wisdom for ordinary subject-predicate sentences while giving c-monstrous readings only where † is present.

The two analyses make incompatible architectural commitments:

• KJR: worlds → world-convention pairs (semantic revision)
• K-G: worlds unchanged, add covert † at specific operator sites

This stub is sufficient to host the inequality theorem in KirkGiannini2024.lean.

Note on scope

Stub formalisation. Encodes (Conventional Wisdom)-rejection and the world-convention pair architecture. Does NOT formalise KJR's full semantics for counterfactuals over WC-pairs, nor the empirical data on embedding expressions identified as c-monsters.

Imports Theories/Semantics/Reference/Monsters.lean to mark Kaplan's thesis as the prior commitment KJR also reject (Kaplan's thesis is the no-context-shifting variant; Conventional Wisdom is the no-convention-shifting variant — KJR reject the latter).

structure KocurekJerzakRudolph2020.Convention (Pred Entity W : Type) : Type

A linguistic convention: an assignment of extensions to predicates (parameterized by predicate type and entity domain).

• ext : Pred → Entity → W → Prop

  The convention's extension function.

structure KocurekJerzakRudolph2020.WC (W Pred Entity : Type) : Type

KJR's evaluation point: a world-convention pair. Replaces the standard World-only architecture. Propositional attitudes hold over WC-pairs; conditionals shift the convention component.

• world : W
• conv : Convention Pred Entity W

def KocurekJerzakRudolph2020.diagContent {W Pred Entity : Type} (p : Pred) (e : Entity) (wc : WC W Pred Entity) : Prop

KJR's diagonal content for a predicate. At a WC-pair ⟨w, c⟩, the predicate p applied to entity e is evaluated using c's extension at w, NOT using the actual-world convention. This is where Conventional Wisdom fails.

Equations

• KocurekJerzakRudolph2020.diagContent p e wc = wc.conv.ext p e wc.world

def KocurekJerzakRudolph2020.ConventionalWisdom {W Pred Entity : Type} (c_actual : Convention Pred Entity W) : Prop

(Conventional Wisdom) — the thesis KJR reject. For all predicates, evaluations should use the actual-world convention c_actual, not the convention component of the evaluation point.

Equations

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

theorem KocurekJerzakRudolph2020.conventional_wisdom_fails {W Pred Entity : Type} [Inhabited W] [Inhabited Pred] [Inhabited Entity] (c_actual c_alt : Convention Pred Entity W) (p : Pred) (e : Entity) (w : W) (h_diff : c_actual.ext p e w ≠ c_alt.ext p e w) : ¬ConventionalWisdom c_actual

KJR's central claim: Conventional Wisdom fails. There exist convention pairs where the actual-world convention diverges from the WC-pair's convention, producing different extensions.

def KocurekJerzakRudolph2020.IsCMonster {W Pred Entity : Type} (embed : (WC W Pred Entity → Prop) → WC W Pred Entity → Prop) (c_actual : Convention Pred Entity W) : Prop

KJR's c-monster predicate. An embedding expression is a c-monster if it can produce readings on which the embedded predicate is evaluated using a NON-actual convention.

Equations

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

def KocurekJerzakRudolph2020.pluto_could_have_been_planet_via_convention_shift {W Pred Entity : Type} (planet : Pred) (pluto : Entity) (c_post2006 c_pre2006 : Convention Pred Entity W) (w : W) (h_post : ¬c_post2006.ext planet pluto w) (h_pre : c_pre2006.ext planet pluto w) : ∃ (wc : WC W Pred Entity), diagContent planet pluto wc ∧ wc.conv ≠ c_post2006

The Pluto example schematized. The counterfactual operator "could have been" applied to "planet" produces a c-monstrous reading where 'planet' is evaluated using a NON-actual convention (one that classifies Pluto as a planet, against the post-2006 actual convention). KJR's analysis: the antecedent of the counterfactual shifts the convention component of the WC-pair.

Equations

• ⋯ = ⋯