Skip to content
Closed
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
52 changes: 52 additions & 0 deletions Cslib/Foundations/Logic/Connectives.lean
Original file line number Diff line number Diff line change
@@ -0,0 +1,52 @@
/-
Copyright (c) 2025 Thomas Waring. All rights reserved.
Released under Apache 2.0 license as described in the file LICENSE.
Authors: Thomas Waring
-/

module

public import Cslib.Init
public import Mathlib.Order.Notation

/-! # Notation classes for logical connectives -/

@[expose] public section

namespace Cslib.Logic

/-- Class for implication. -/
class HasImpl (α : Type*) where
/-- Implication. -/
impl : α → α → α

@[inherit_doc] scoped infixr:25 " → " => HasImpl.impl

/-- Class for conjunction. -/
class HasAnd (α : Type*) where
/-- Conjunction. -/
and : α → α → α

@[inherit_doc] scoped infixr:35 " ∧ " => HasAnd.and

/-- Class for disjunction. -/
class HasOr (α : Type*) where
/-- Disjunction. -/
or : α → α → α

@[inherit_doc] scoped infixr:30 " ∨ " => HasOr.or

/-- Class for negation. -/
class HasNot (α : Type*) where
/-- Negation. -/
not : α → α

@[inherit_doc] scoped notation:max "¬" a:40 => HasNot.not a

/-- Instantiate negation for types with implication and a bottom element. NB: this has a lower
instance priority to account for proposition types with inbuilt negation. Possibly it should be
a `def` instead? -/
instance (priority := 900) instNotImplBot (α : Type*) [HasImpl α] [Bot α] : HasNot α where
not a := a → ⊥

end Cslib.Logic
170 changes: 170 additions & 0 deletions Cslib/Foundations/Logic/Model.lean
Original file line number Diff line number Diff line change
@@ -0,0 +1,170 @@
/-
Copyright (c) 2025 Thomas Waring. All rights reserved.
Released under Apache 2.0 license as described in the file LICENSE.
Authors: Thomas Waring
-/

module

public import Cslib.Foundations.Logic.Connectives
public import Cslib.Foundations.Logic.InferenceSystem
public import Cslib.Logics.Modal.Basic
public import Cslib.Logics.Propositional.NaturalDeduction.Basic

/-! # Semantics for logical systems -/

public section

namespace Cslib.Logic

/-- Objects of type `β` carry a forcing relation with the proposition type `α`. -/
class ModelClass (α : outParam Type*) (β : Type*) where
/-- `Forces b a` has the intended semantics "`a` is valid in the model `b`". -/
Forces : β → α → Prop
Comment thread
thomaskwaring marked this conversation as resolved.

scoped notation "⊨[" b "] " a => ModelClass.Forces b a
Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Why not write b ⊨ a? See also my comment about the notation for ForcesAt.


/-- Objects of type `β` carry a forcing relation worlds of type `γ` and the proposition type `α`. -/
class ParamModelClass (α : outParam Type*) (β : Type*) where
Param : β → Type*
Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It is not clear to me why we need this generality. Can you give a natural example of the use of ParamModelClass?

Note also that ParamModelClass can be implemented using ModelClass by substituting a dependent sum type for β, namely, Σ b : β, Param b, which means that the only thing that ParamModelClass gives you is the bundling of Param.

/-- Forcing relation associated to each parameter. -/
ForcesAt (b : β) : (Param b) → α → Prop
Comment thread
thomaskwaring marked this conversation as resolved.

scoped notation w " ⊨[" b "] " a => ParamModelClass.ForcesAt b w a
Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I find the order in which the arguments appear in the notation unintuitive: a is a "world" in model b, and yet w appears before b in the notation. Wouldn't it be better to write something like b, w ⊨ a?


instance ParamModelClass.toModelClass {α β : Type*} [ParamModelClass α β] : ModelClass α β where
Forces M A := ∀ w : ParamModelClass.Param M, w ⊨[M] A

/-- Bundled interpretation function. -/
class HasInterp (α : outParam Type*) (β : Type*) where
/-- Type carrying interpretation. -/
Ground : β → Type*
/-- Interpret a proposition in a model. -/
interp : (b : β) → α → Ground b

/-- An `InterpModel` consists of an interpretation function, and a set specifying which
interpretations are considered valid. -/
class InterpModelClass (α : outParam (Type*)) (β : Type*) extends HasInterp α β where
/-- The set of valid interpretations. -/
filter (b : β) : Set (Ground b)
Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Rather than Set, perhaps something like CompleteAtomicBooleanAlgebra will provide more generality?


instance InterpModelClass.instModelClass {α β : Type*} [InterpModelClass α β] : ModelClass α β where
Forces b a := HasInterp.interp b a ∈ filter b

namespace HasInterp

class AlgebraicAnd (α β : Type*) [HasInterp α β] [HasAnd α] [∀ b : β, Min (Ground b)] where
interp_and_eq (M : β) (x y : α) : interp M (x ∧ y) = interp M x ⊓ interp M y

class AlgebraicOr (α β : Type*) [HasInterp α β] [HasOr α] [∀ b : β, Max (Ground b)] where
interp_or_eq (M : β) (x y : α) : interp M (x ∨ y) = interp M x ⊔ interp M y

class AlgebraicImpl (α β : Type*) [HasInterp α β] [HasImpl α] [∀ b : β, HImp (Ground b)] where
interp_impl_eq (M : β) (x y : α) : interp M (x → y) = interp M x ⇨ interp M y

class AlgebraicNot (α β : Type*) [HasInterp α β] [HasNot α] [∀ b : β, Compl (Ground b)] where
interp_not_eq (M : β) (x y : α) : interp M (¬ x) = (interp M x)ᶜ
Comment on lines +56 to +66
Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I don't understand these definitions. What are those lattice-theoretic operations like and ? What sort of algebraic structures do you assume Ground b to have? How are Ground b related to filter b?


end HasInterp

open ModelClass ParamModelClass InferenceSystem

variable {α β T} [ModelClass α β] [InferenceSystem T α]

def SoundFor (α β T) [ModelClass α β] [InferenceSystem T α] (S : Set β) : Prop :=
∀ (A : α), DerivableIn T A → ∀ M ∈ S, ⊨[M] A

lemma SoundFor.subset {S S' : Set β} (hS : S ⊆ S') : SoundFor α β T S' → SoundFor α β T S :=
fun h A hA M hM => h A hA M (hS hM)

def CompleteFor (α β T : Type*) [ModelClass α β] [InferenceSystem T α] (S : Set β) : Prop :=
∀ A : α, (∀ M ∈ S, ⊨[M] A) → DerivableIn T A

lemma CompleteFor.supset {S S' : Set β} (hS : S ⊆ S') :
CompleteFor α β T S → CompleteFor α β T S' := fun h A hA => h A (fun M hM => hA M (hS hM))

def IsCompleteModel (α β T) [ModelClass α β] [InferenceSystem T α] (M : β) : Prop :=
∀ (A : α), (⊨[M] A) → DerivableIn T A

def ParamModelClass.theory {α β : Type*} [ParamModelClass α β] {M : β} (w : Param M) : Set α :=
{A : α | w ⊨[M] A}

def ModelClass.logic {α β : Type*} [ModelClass α β] (S : Set β) : Set α := {A | ∀ b ∈ S, ⊨[b] A}

namespace Modal

structure BundledModel (Atom : Type*) where
World : Type*
model : Model World Atom

def Model.toBundledModel {World Atom : Type*} (M : Model World Atom) : BundledModel Atom :=
{World := World, model := M}

instance {Atom : Type*} : ParamModelClass (Proposition Atom) (BundledModel Atom) where
Param M := M.World
ForcesAt M w A := Satisfies M.model w A

example {World Atom : Type*} (S : Set (Model World Atom)) :
logic S = ModelClass.logic (Model.toBundledModel '' S) := by
simp [ModelClass.logic]
rfl

example {World Atom : Type*} (m : Model World Atom) (w : World) :
theory m w = ParamModelClass.theory (M := m.toBundledModel) w := by
simp [theory, ParamModelClass.theory]
rfl

end Modal

namespace PL

variable {Atom : Type*}

instance : HasAnd (Proposition Atom) := ⟨Proposition.and⟩
instance : HasOr (Proposition Atom) := ⟨Proposition.or⟩
instance : HasImpl (Proposition Atom) := ⟨Proposition.impl⟩
instance [Bot Atom] : HasNot (Proposition Atom) := ⟨Proposition.neg⟩

structure HeytingModel (Atom : Type*) where
H : Type*
[inst : GeneralizedHeytingAlgebra H]
v : Atom → H

instance (M : HeytingModel Atom) : GeneralizedHeytingAlgebra M.H := M.inst

def HeytingModel.interp (M : HeytingModel Atom) : Proposition Atom → M.H
| Proposition.atom x => M.v x
| Proposition.and A B => M.interp A ⊓ M.interp B
| Proposition.or A B => M.interp A ⊔ M.interp B
| Proposition.impl A B => M.interp A ⇨ M.interp B

instance : InterpModelClass (Proposition Atom) (HeytingModel Atom) where
Ground M := M.H
interp := HeytingModel.interp
filter _ := {⊤}

instance (M : HeytingModel Atom) : GeneralizedHeytingAlgebra (HasInterp.Ground M) := M.inst

instance : HasInterp.AlgebraicAnd (Proposition Atom) (HeytingModel Atom) where
interp_and_eq _ _ _ := rfl

instance : HasInterp.AlgebraicOr (Proposition Atom) (HeytingModel Atom) where
interp_or_eq _ _ _ := rfl

instance : HasInterp.AlgebraicImpl (Proposition Atom) (HeytingModel Atom) where
interp_impl_eq _ _ _ := rfl

theorem HeytingModel.sound [DecidableEq Atom] {T : Theory Atom} :
SoundFor (Proposition Atom) (HeytingModel Atom) T {M | ∀ A ∈ T, interp M A = ⊤} :=
sorry -- i have this in a branch

def Theory.lindenbaum [DecidableEq Atom] (T : Theory Atom) : HeytingModel Atom :=
sorry -- usual Heyting-algebra of propositions modulo equivalence

theorem Theory.lindenbaum_complete [DecidableEq Atom] {T : Theory Atom} :
IsCompleteModel (Proposition Atom) (HeytingModel Atom) T T.lindenbaum :=
sorry -- also in a branch

end PL

end Cslib.Logic
Loading