Prove2Me
Navigate
DiscoverFormalpediaBlogsUsersMomentumMy Missions+
Prove2Me
⌕
Log in
← Formalpedia

Thompson's group TTT on the circle, and the presented group T1T_1T1​

Definition
CannonFloydParry_T

by dbenbenn · Sep 25, 2026 · Mathlib 0df444a (Lean v4.33.1)

group-theorysimple-groupsthompsons-group

Cannon–Floyd–Parry §5 (pp. 233–239): Thompson's group TTT and the presented group T1T_1T1​.

The circle. p. 233: “Consider S1S^1S1 as the interval [0,1][0, 1][0,1] with the endpoints identified.” S1S^1S1 is UnitAddCircle, the reals modulo 111, which is that circle; [x][x][x] is the image of x∈Rx \in \mathbb{R}x∈R (the source uses [x][x][x], as on p. 235, without defining it).

IsThompsonCircle, T. pp. 233–234: “Then TTT is the group of piecewise linear homeomorphisms from S1S^1S1 to itself that map images of dyadic rational numbers to images of dyadic rational numbers and that are differentiable except at finitely many images of dyadic rational numbers and on intervals of differentiability the derivatives are powers of 222.” A permutation fff of S1S^1S1 satisfies IsThompsonCircle f when it has a lift LLL: an order isomorphism of R\mathbb{R}R with L(x+1)=L(x)+1L(x+1) = L(x) + 1L(x+1)=L(x)+1 and f[x]=[L(x)]f[x] = [L(x)]f[x]=[L(x)] for all xxx, which maps dyadic rationals to dyadic rationals and, for some finite set BBB of dyadic rationals, is affine with slope an integer power of 222 on every closed interval whose interior meets no point b+kb + kb+k (b∈Bb \in Bb∈B, k∈Zk \in \mathbb{Z}k∈Z). T is the subgroup of permutations of S1S^1S1 generated by these maps; it is the group TTT of this sentence.

toCircle, mapC, symT. p. 234 (Example 5.1): “The elements AAA and BBB of FFF induce elements of TTT, which will still be denoted by AAA and BBB.” toCircle g, for an order isomorphism ggg of [0,1][0,1][0,1], is the permutation [x]↦[g(x)][x] \mapsto [g(x)][x]↦[g(x)] of S1S^1S1 (x∈[0,1)x \in [0,1)x∈[0,1); ggg fixes 111). p. 234: “A third element of TTT is the function CCC defined (on [0,1][0, 1][0,1]) by C(x)={x2+34,0≤x≤122x−1,12≤x≤34x−14,34≤x≤1.C(x) = \begin{cases} \frac{x}{2} + \frac{3}{4}, & 0 \le x \le \frac{1}{2} \\ 2x - 1, & \frac{1}{2} \le x \le \frac{3}{4} \\ x - \frac{1}{4}, & \frac{3}{4} \le x \le 1. \end{cases}C(x)=⎩⎨⎧​2x​+43​,2x−1,x−41​,​0≤x≤21​21​≤x≤43​43​≤x≤1.​” mapC is this CCC, with half-open pieces: x/2+3/4x/2 + 3/4x/2+3/4 on [0,12)[0,\tfrac12)[0,21​), 2x−12x - 12x−1 on [12,34)[\tfrac12,\tfrac34)[21​,43​), x−14x - \tfrac14x−41​ on [34,1)[\tfrac34, 1)[43​,1). symT sends the formal symbols AAA, BBB, CCC to toCircle mapA, toCircle mapB and mapC, where mapA, mapB are FFF's generators from the imported bundle.

T1. p. 236: “Let T1=⟨A,B,C:[AB−1,A−1BA],[AB−1,A−2BA2],C−1B(A−1CB),((A−1CB)(A−1BA))−1B(A−2CB2),(CA)−1(A−1CB)2,C3⟩.T_1 = \langle A, B, C : [AB^{-1}, A^{-1}BA], [AB^{-1}, A^{-2}BA^2], C^{-1}B(A^{-1}CB), ((A^{-1}CB)(A^{-1}BA))^{-1}B(A^{-2}CB^2), (CA)^{-1}(A^{-1}CB)^2, C^3\rangle.T1​=⟨A,B,C:[AB−1,A−1BA],[AB−1,A−2BA2],C−1B(A−1CB),((A−1CB)(A−1BA))−1B(A−2CB2),(CA)−1(A−1CB)2,C3⟩.” T1 is the presented group on the three symbols of FormalABC with these six relators, the commutators written out as xyx−1y−1xyx^{-1}y^{-1}xyx−1y−1, as p. 225 (§3) defines them: “Given elements xxx, yyy in a group, [x,y]=xyx−1y−1[x, y] = xyx^{-1}y^{-1}[x,y]=xyx−1y−1.”

XT1, CT1, IsPositiveT1. p. 236: “Define the elements XnX_nXn​, n≥0n \ge 0n≥0, of T1T_1T1​ by X0=AX_0 = AX0​=A and Xn=A−(n−1)BAn−1X_n = A^{-(n-1)}BA^{n-1}Xn​=A−(n−1)BAn−1 for n≥1n \ge 1n≥1.” XT1 n is XnX_nXn​. p. 236: “Define the elements CnC_nCn​, n≥1n \ge 1n≥1, of T1T_1T1​ by Cn=A−(n−1)CBn−1C_n = A^{-(n-1)}CB^{n-1}Cn​=A−(n−1)CBn−1. For convenience we define C0=1C_0 = 1C0​=1.” CT1 n is CnC_nCn​. p. 239: “Following the terminology for FFF, an element of T1T_1T1​ which is a product of nonnegative powers of the XiX_iXi​’s will be called positive and an inverse of a positive element will be called negative.” IsPositiveT1 takes an element to be positive when it lies in the submonoid generated by the XiX_iXi​.

Formalization Note. Products of permutations compose right to left, (fg)(t)=f(g(t))(fg)(t) = f(g(t))(fg)(t)=f(g(t)), the convention under which the source's computations hold (it makes ACACAC the translation by 12\tfrac1221​, as on p. 236). T is defined as a generated subgroup so that the definition carries no proof obligation; that the maps satisfying IsThompsonCircle already form a group is a milestone of this mission, as for FFF.

Definition code
import Definitions.Def_CannonFloydParry
import Mathlib

/-!
# Cannon–Floyd–Parry §5: Thompson's group `T`

Cannon, Floyd, Parry, *Introductory notes on Richard Thompson's groups*, L'Enseignement Math.
(2) 42 (1996), §5, pp. 233–239.  The circle `S¹` is `UnitAddCircle`, the reals modulo `1`,
which is `[0,1]` with its endpoints identified.  `T` is a group of permutations of it; the
presented group `T₁` and its elements `Xₙ`, `Cₙ` are the objects of the section's algebra.
-/

namespace CannonFloydParry

/-! ### The group `T` of the circle (pp. 233–234) -/

/-- The condition, on a permutation `f` of the circle `ℝ/ℤ`, that defines Thompson's group `T`
(p. 233), stated through a lift of `f` to the line: an order isomorphism `L` of `ℝ` that
commutes with translation by `1` and covers `f`, maps dyadic rationals to dyadic rationals, and
is affine with slope an integer power of `2` on every closed interval whose interior avoids the
integer translates of a finite set `B` of dyadic breakpoints. -/
def IsThompsonCircle (f : Equiv.Perm UnitAddCircle) : Prop :=
  ∃ L : ℝ ≃o ℝ, (∀ x, L (x + 1) = L x + 1) ∧
    (∀ x : ℝ, f (x : UnitAddCircle) = ((L x : ℝ) : UnitAddCircle)) ∧
    (∀ x, IsDyadic x → IsDyadic (L x)) ∧
    ∃ B : Finset ℝ, (∀ b ∈ B, IsDyadic b) ∧
      ∀ x y : ℝ, x < y → (∀ t ∈ Set.Ioo x y, ∀ b ∈ B, ∀ k : ℤ, t ≠ b + k) →
        ∃ (n : ℤ) (c : ℝ), ∀ z ∈ Set.Icc x y, L z = 2 ^ n * z + c

/-- Thompson's group `T`: the subgroup of permutations of the circle generated by the maps
satisfying `IsThompsonCircle`. -/
def T : Subgroup (Equiv.Perm UnitAddCircle) := Subgroup.closure {f | IsThompsonCircle f}

lemma orderIso_one (f : UI ≃o UI) : f ⟨1, one_mem_UI⟩ = ⟨1, one_mem_UI⟩ := by
  apply le_antisymm
  · exact Subtype.coe_le_coe.mp (f ⟨1, one_mem_UI⟩).2.2
  · obtain ⟨z, hz⟩ := f.surjective ⟨1, one_mem_UI⟩
    have := f.monotone (show z ≤ ⟨1, one_mem_UI⟩ from Subtype.coe_le_coe.mp z.2.2)
    rwa [hz] at this

lemma orderIso_lt_one_iff (f : UI ≃o UI) (x : UI) : (f x : ℝ) < 1 ↔ (x : ℝ) < 1 := by
  have h := @OrderIso.lt_iff_lt _ _ _ _ f x ⟨1, one_mem_UI⟩
  rw [orderIso_one] at h
  exact h

/-- An order isomorphism of `[0,1]` fixes `1`, so it restricts to a permutation of `[0,1)`. -/
noncomputable def icoPerm (f : UI ≃o UI) : Equiv.Perm (Set.Ico (0 : ℝ) (0 + 1)) where
  toFun x := ⟨f ⟨x, x.2.1, by linarith [x.2.2]⟩, (f _).2.1, by
    have := (orderIso_lt_one_iff f ⟨x, x.2.1, by linarith [x.2.2]⟩).2 (by linarith [x.2.2])
    linarith⟩
  invFun x := ⟨f.symm ⟨x, x.2.1, by linarith [x.2.2]⟩, (f.symm _).2.1, by
    have := (orderIso_lt_one_iff f.symm ⟨x, x.2.1, by linarith [x.2.2]⟩).2 (by linarith [x.2.2])
    linarith⟩
  left_inv x := by
    apply Subtype.ext
    change ((f.symm (f ⟨x, x.2.1, by linarith [x.2.2]⟩) : UI) : ℝ) = x
    rw [OrderIso.symm_apply_apply]
  right_inv x := by
    apply Subtype.ext
    change ((f (f.symm ⟨x, x.2.1, by linarith [x.2.2]⟩) : UI) : ℝ) = x
    rw [OrderIso.apply_symm_apply]

/-- The permutation of the circle induced by an order isomorphism of `[0,1]` (p. 234, "The
elements `A` and `B` of `F` induce elements of `T`"). -/
noncomputable def toCircle (f : UI ≃o UI) : Equiv.Perm UnitAddCircle :=
  (AddCircle.equivIco (1 : ℝ) 0).trans ((icoPerm f).trans (AddCircle.equivIco (1 : ℝ) 0).symm)

/-- The function `C` of Example 5.1 (p. 234) on `[0,1)`: `x/2 + 3/4` on `[0,1/2)`, `2x - 1` on
`[1/2,3/4)` and `x - 1/4` on `[3/4,1)`. -/
noncomputable def cFun (x : ℝ) : ℝ :=
  if x < 1 / 2 then x / 2 + 3 / 4 else if x < 3 / 4 then 2 * x - 1 else x - 1 / 4

/-- The inverse of `cFun` on `[0,1)`. -/
noncomputable def cInv (y : ℝ) : ℝ :=
  if y < 1 / 2 then (y + 1) / 2 else if y < 3 / 4 then y + 1 / 4 else 2 * y - 3 / 2

lemma cFun_mem {x : ℝ} (hx : x ∈ Set.Ico (0 : ℝ) (0 + 1)) : cFun x ∈ Set.Ico (0 : ℝ) (0 + 1) := by
  obtain ⟨h0, h1⟩ := hx
  unfold cFun; split_ifs <;> constructor <;> linarith

lemma cInv_mem {x : ℝ} (hx : x ∈ Set.Ico (0 : ℝ) (0 + 1)) : cInv x ∈ Set.Ico (0 : ℝ) (0 + 1) := by
  obtain ⟨h0, h1⟩ := hx
  unfold cInv; split_ifs <;> constructor <;> linarith

lemma cInv_cFun {x : ℝ} (hx : x ∈ Set.Ico (0 : ℝ) (0 + 1)) : cInv (cFun x) = x := by
  obtain ⟨h0, h1⟩ := hx
  unfold cFun cInv
  split_ifs <;> first | (exfalso; linarith) | ring

lemma cFun_cInv {x : ℝ} (hx : x ∈ Set.Ico (0 : ℝ) (0 + 1)) : cFun (cInv x) = x := by
  obtain ⟨h0, h1⟩ := hx
  unfold cFun cInv
  split_ifs <;> first | (exfalso; linarith) | ring

/-- `cFun` as a permutation of `[0,1)`. -/
noncomputable def cIco : Equiv.Perm (Set.Ico (0 : ℝ) (0 + 1)) where
  toFun x := ⟨cFun x, cFun_mem x.2⟩
  invFun x := ⟨cInv x, cInv_mem x.2⟩
  left_inv x := Subtype.ext (cInv_cFun x.2)
  right_inv x := Subtype.ext (cFun_cInv x.2)

/-- The element `C` of `T` (Example 5.1, p. 234), as a permutation of the circle. -/
noncomputable def mapC : Equiv.Perm UnitAddCircle :=
  (AddCircle.equivIco (1 : ℝ) 0).trans (cIco.trans (AddCircle.equivIco (1 : ℝ) 0).symm)

/-! ### The presented group `T₁` (p. 236) -/

/-- The three formal symbols `A`, `B`, `C` of the presentation of `T₁`. -/
inductive FormalABC
  | A
  | B
  | C
  deriving DecidableEq

/-- The relators of `T₁` (p. 236), with `[x, y] = x y x⁻¹ y⁻¹`:
`[AB⁻¹, A⁻¹BA]`, `[AB⁻¹, A⁻²BA²]`, `C⁻¹B(A⁻¹CB)`, `((A⁻¹CB)(A⁻¹BA))⁻¹B(A⁻²CB²)`,
`(CA)⁻¹(A⁻¹CB)²` and `C³`. -/
def relsT1 : Set (FreeGroup FormalABC) :=
  let a := FreeGroup.of FormalABC.A
  let b := FreeGroup.of FormalABC.B
  let c := FreeGroup.of FormalABC.C
  { (a * b⁻¹) * (a⁻¹ * b * a) * (a * b⁻¹)⁻¹ * (a⁻¹ * b * a)⁻¹,
    (a * b⁻¹) * (a⁻¹ ^ 2 * b * a ^ 2) * (a * b⁻¹)⁻¹ * (a⁻¹ ^ 2 * b * a ^ 2)⁻¹,
    c⁻¹ * b * (a⁻¹ * c * b),
    ((a⁻¹ * c * b) * (a⁻¹ * b * a))⁻¹ * b * (a⁻¹ ^ 2 * c * b ^ 2),
    (c * a)⁻¹ * (a⁻¹ * c * b) ^ 2,
    c ^ 3 }

/-- `T₁ = ⟨A, B, C : relsT1⟩` (p. 236). -/
abbrev T1 := PresentedGroup relsT1

/-- The intended images of the formal symbols among the maps of the circle: `A` and `B` are
the maps of `F` carried to the circle, and `C` is `mapC`. -/
noncomputable def symT : FormalABC → Equiv.Perm UnitAddCircle
  | FormalABC.A => toCircle mapA
  | FormalABC.B => toCircle mapB
  | FormalABC.C => mapC

/-- The elements `X₀ = A` and `Xₙ = A^{-(n-1)} B A^{n-1}` (`n ≥ 1`) of `T₁` (p. 236). -/
def XT1 : ℕ → T1
  | 0 => PresentedGroup.of FormalABC.A
  | n + 1 => (PresentedGroup.of FormalABC.A ^ n)⁻¹ * PresentedGroup.of FormalABC.B
      * PresentedGroup.of FormalABC.A ^ n

/-- The elements `C₀ = 1` and `Cₙ = A^{-(n-1)} C B^{n-1}` (`n ≥ 1`) of `T₁` (p. 236). -/
def CT1 : ℕ → T1
  | 0 => 1
  | n + 1 => (PresentedGroup.of FormalABC.A ^ n)⁻¹ * PresentedGroup.of FormalABC.C
      * PresentedGroup.of FormalABC.B ^ n

/-- The **positive** elements of `T₁` (p. 239): products of nonnegative powers of the `Xᵢ`. -/
def IsPositiveT1 (g : T1) : Prop := g ∈ Submonoid.closure (Set.range XT1)

end CannonFloydParry
Source
Cannon, J. W., Floyd, W. J., Parry, W. R., Introductory notes on Richard Thompson's groups, L'Enseignement Mathématique (2) 42 (1996) 215–256, https://doi.org/10.5169/seals-87877, section 5, pp. 233–239, the group T (pp. 233–234), Example 5.1, the presentation T₁ and the elements Xₙ, Cₙ (p. 236), positive elements (p. 239)
Read-back

What the Lean code literally says, in plain math · claude-opus-5-5

Read-back

This item is a bundle of definitions (with a few small auxiliary facts proved along the way). They share the following vocabulary.

Shared vocabulary

The circle. T=R/Z\mathbb{T} = \mathbb{R}/\mathbb{Z}T=R/Z is the additive quotient of R\mathbb{R}R by the subgroup of integer multiples of 111. For x∈Rx \in \mathbb{R}x∈R, [x]∈T[x] \in \mathbb{T}[x]∈T denotes its class. Every point of T\mathbb{T}T has exactly one representative in the half-open interval [0,1)[0,1)[0,1); the "representative map" ρ:T→[0,1)\rho : \mathbb{T} \to [0,1)ρ:T→[0,1) sends a class to that representative, and its inverse is x↦[x]x \mapsto [x]x↦[x].

Permutations of the circle. Sym(T)\mathrm{Sym}(\mathbb{T})Sym(T) is the group of all bijections T→T\mathbb{T} \to \mathbb{T}T→T — no continuity or order condition is built in. The product is composition with the right factor applied first: (fg)(p)=f(g(p))(fg)(p) = f(g(p))(fg)(p)=f(g(p)).

Dyadic rationals. A real number xxx is dyadic if there are an integer mmm and a natural number k≥0k \ge 0k≥0 with

x=m2k.x = \frac{m}{2^k}.x=2km​.

(k=0k = 0k=0 is allowed, so every integer is dyadic.)

The unit interval. I=[0,1]⊂RI = [0,1] \subset \mathbb{R}I=[0,1]⊂R (closed), with its usual order. An order isomorphism of III is an order-preserving bijection I→II \to II→I whose inverse is also order-preserving.

Powers of 2. Wherever 2n2^n2n appears with nnn an integer, it is the real number 2n2^n2n with nnn possibly negative (so 2−3=1/82^{-3} = 1/82−3=1/8).


1. Circle maps of "Thompson type"

A bijection f∈Sym(T)f \in \mathrm{Sym}(\mathbb{T})f∈Sym(T) is of Thompson type if there exists an order isomorphism L:R→RL : \mathbb{R} \to \mathbb{R}L:R→R (an order-preserving bijection with order-preserving inverse) satisfying all four of the following:

  1. Periodicity: L(x+1)=L(x)+1L(x+1) = L(x) + 1L(x+1)=L(x)+1 for every real xxx.

  2. LLL lifts fff: f([x])=[L(x)]f([x]) = [L(x)]f([x])=[L(x)] for every real xxx.

  3. LLL preserves dyadics: for every real xxx, if xxx is dyadic then L(x)L(x)L(x) is dyadic. (Only this direction is required.)

  4. Piecewise dyadic-affine with finitely many dyadic breakpoints mod 1: there exists a finite set B⊂RB \subset \mathbb{R}B⊂R, all of whose elements are dyadic, such that for all reals x<yx < yx<y:

    if the open interval (x,y)(x,y)(x,y) contains no point of the form b+kb + kb+k with b∈Bb \in Bb∈B and k∈Zk \in \mathbb{Z}k∈Z, then there exist an integer nnn and a real number ccc such that

L(z)=2nz+cfor every z∈[x,y].L(z) = 2^n z + c \qquad \text{for every } z \in [x,y].L(z)=2nz+cfor every z∈[x,y].

Remarks on the fine print:

  • In (4) the hypothesis is only about the open interval (x,y)(x,y)(x,y); the endpoints x,yx,yx,y may themselves be of the form b+kb+kb+k, and the affine formula is then asserted on the closed interval [x,y][x,y][x,y].
  • The set BBB is any finite set of dyadic reals; it is not required to lie in [0,1)[0,1)[0,1), and it may be empty. If BBB is empty, the hypothesis of (4) holds for every x<yx<yx<y.
  • The integer nnn and the real ccc may depend on xxx and yyy. The constant ccc is not itself required to be dyadic.
  • Nothing is said about the inverse of fff or of LLL beyond LLL being an order isomorphism.

2. The group TTT

TTT is the subgroup of Sym(T)\mathrm{Sym}(\mathbb{T})Sym(T) generated by all bijections of Thompson type (Section 1), i.e. the smallest subgroup of Sym(T)\mathrm{Sym}(\mathbb{T})Sym(T) containing every such bijection. Its elements are exactly the finite products of Thompson-type bijections and their inverses (the empty product being the identity).

3. Two auxiliary facts about order isomorphisms of III (proved, not assumed)

  • For every order isomorphism ggg of I=[0,1]I=[0,1]I=[0,1]: g(1)=1g(1) = 1g(1)=1.
  • For every order isomorphism ggg of III and every x∈Ix \in Ix∈I: g(x)<1g(x) < 1g(x)<1 if and only if x<1x < 1x<1.

4. From interval maps to circle maps

Restriction to [0,1)[0,1)[0,1). Given an order isomorphism ggg of III, define the bijection g^\hat gg^​ of the half-open interval [0,1)[0,1)[0,1) by g^(x)=g(x)\hat g(x) = g(x)g^​(x)=g(x). (It lands in [0,1)[0,1)[0,1) by the second fact of Section 3; its inverse is x↦g−1(x)x \mapsto g^{-1}(x)x↦g−1(x).)

Transport to the circle. For an order isomorphism ggg of III, Φ(g)∈Sym(T)\Phi(g) \in \mathrm{Sym}(\mathbb{T})Φ(g)∈Sym(T) is the bijection

Φ(g)=ρ−1∘g^∘ρ,i.e.Φ(g)([x])=[ g(x) ] for x∈[0,1).\Phi(g) = \rho^{-1} \circ \hat g \circ \rho, \qquad\text{i.e.}\qquad \Phi(g)([x]) = [\,g(x)\,] \text{ for } x \in [0,1).Φ(g)=ρ−1∘g^​∘ρ,i.e.Φ(g)([x])=[g(x)] for x∈[0,1).

In words: take the representative in [0,1)[0,1)[0,1), apply ggg, and take the class again.

5. The map CCC on the circle

Define γ:R→R\gamma : \mathbb{R} \to \mathbb{R}γ:R→R by

γ(x)={x/2+3/4,x<1/2,2x−1,1/2≤x<3/4,x−1/4,x≥3/4,\gamma(x) = \begin{cases} x/2 + 3/4, & x < 1/2,\\ 2x - 1, & 1/2 \le x < 3/4,\\ x - 1/4, & x \ge 3/4,\end{cases}γ(x)=⎩⎨⎧​x/2+3/4,2x−1,x−1/4,​x<1/2,1/2≤x<3/4,x≥3/4,​

and δ:R→R\delta : \mathbb{R} \to \mathbb{R}δ:R→R by

δ(y)={(y+1)/2,y<1/2,y+1/4,1/2≤y<3/4,2y−3/2,y≥3/4.\delta(y) = \begin{cases} (y+1)/2, & y < 1/2,\\ y + 1/4, & 1/2 \le y < 3/4,\\ 2y - 3/2, & y \ge 3/4.\end{cases}δ(y)=⎩⎨⎧​(y+1)/2,y+1/4,2y−3/2,​y<1/2,1/2≤y<3/4,y≥3/4.​

Only their values on [0,1)[0,1)[0,1) matter below. Proved auxiliary facts: γ\gammaγ and δ\deltaδ each map [0,1)[0,1)[0,1) into [0,1)[0,1)[0,1), and on [0,1)[0,1)[0,1) they are inverse to each other (δ(γ(x))=x\delta(\gamma(x)) = xδ(γ(x))=x and γ(δ(x))=x\gamma(\delta(x)) = xγ(δ(x))=x for x∈[0,1)x \in [0,1)x∈[0,1)).

Concretely on [0,1)[0,1)[0,1): γ\gammaγ sends [0,1/2)[0,1/2)[0,1/2) to [3/4,1)[3/4,1)[3/4,1) by x↦x/2+3/4x \mapsto x/2+3/4x↦x/2+3/4; [1/2,3/4)[1/2,3/4)[1/2,3/4) to [0,1/2)[0,1/2)[0,1/2) by x↦2x−1x \mapsto 2x-1x↦2x−1; and [3/4,1)[3/4,1)[3/4,1) to [1/2,3/4)[1/2,3/4)[1/2,3/4) by x↦x−1/4x \mapsto x-1/4x↦x−1/4.

γ^\hat\gammaγ^​ is the resulting bijection of [0,1)[0,1)[0,1) (inverse δ\deltaδ), and

C=ρ−1∘γ^∘ρ∈Sym(T),C([x])=[γ(x)] for x∈[0,1).C = \rho^{-1} \circ \hat\gamma \circ \rho \in \mathrm{Sym}(\mathbb{T}), \qquad C([x]) = [\gamma(x)] \text{ for } x \in [0,1).C=ρ−1∘γ^​∘ρ∈Sym(T),C([x])=[γ(x)] for x∈[0,1).

6. The interval maps AIA_IAI​ and BIB_IBI​ (from the imported vocabulary)

AIA_IAI​ is the order isomorphism of I=[0,1]I = [0,1]I=[0,1] given by

AI(x)={x/2,0≤x≤1/2,x−1/4,1/2≤x≤3/4,2x−1,3/4≤x≤1,A_I(x) = \begin{cases} x/2, & 0 \le x \le 1/2,\\ x - 1/4, & 1/2 \le x \le 3/4,\\ 2x - 1, & 3/4 \le x \le 1,\end{cases}AI​(x)=⎩⎨⎧​x/2,x−1/4,2x−1,​0≤x≤1/2,1/2≤x≤3/4,3/4≤x≤1,​

and BIB_IBI​ is the order isomorphism of III given by

BI(x)={x,0≤x≤1/2,x/2+1/4,1/2≤x≤3/4,x−1/8,3/4≤x≤7/8,2x−1,7/8≤x≤1.B_I(x) = \begin{cases} x, & 0 \le x \le 1/2,\\ x/2 + 1/4, & 1/2 \le x \le 3/4,\\ x - 1/8, & 3/4 \le x \le 7/8,\\ 2x - 1, & 7/8 \le x \le 1.\end{cases}BI​(x)=⎩⎨⎧​x,x/2+1/4,x−1/8,2x−1,​0≤x≤1/2,1/2≤x≤3/4,3/4≤x≤7/8,7/8≤x≤1.​

(The formulas agree at the shared breakpoints. Each is the restriction to [0,1][0,1][0,1] of an order isomorphism of R\mathbb{R}R that is the identity outside (0,1)(0,1)(0,1).)

7. The generator assignment σ\sigmaσ

A three-letter alphabet {A,B,C}\{A, B, C\}{A,B,C} is introduced (three distinct symbols). The function σ\sigmaσ from this alphabet to Sym(T)\mathrm{Sym}(\mathbb{T})Sym(T) is

σ(A)=Φ(AI),σ(B)=Φ(BI),σ(C)=C.\sigma(A) = \Phi(A_I), \qquad \sigma(B) = \Phi(B_I), \qquad \sigma(C) = C.σ(A)=Φ(AI​),σ(B)=Φ(BI​),σ(C)=C.

So σ(A)([x])=[AI(x)]\sigma(A)([x]) = [A_I(x)]σ(A)([x])=[AI​(x)] and σ(B)([x])=[BI(x)]\sigma(B)([x]) = [B_I(x)]σ(B)([x])=[BI​(x)] for x∈[0,1)x \in [0,1)x∈[0,1). This is only a function on the three letters; no group homomorphism out of any group is defined or claimed here.

8. The presented group T1T_1T1​

Write a,b,ca, b, ca,b,c for the free generators of the free group F(a,b,c)F(a,b,c)F(a,b,c) on the three letters. The relator set consists of the following six elements of F(a,b,c)F(a,b,c)F(a,b,c) (products read left to right; a−2a^{-2}a−2 means (a−1)2(a^{-1})^2(a−1)2):

  1. (ab−1) (a−1ba) (ab−1)−1 (a−1ba)−1(ab^{-1})\,(a^{-1}ba)\,(ab^{-1})^{-1}\,(a^{-1}ba)^{-1}(ab−1)(a−1ba)(ab−1)−1(a−1ba)−1
  2. (ab−1) (a−2ba2) (ab−1)−1 (a−2ba2)−1(ab^{-1})\,(a^{-2}ba^{2})\,(ab^{-1})^{-1}\,(a^{-2}ba^{2})^{-1}(ab−1)(a−2ba2)(ab−1)−1(a−2ba2)−1
  3. c−1 b (a−1cb)c^{-1}\,b\,(a^{-1}cb)c−1b(a−1cb)
  4. ((a−1cb)(a−1ba))−1 b (a−2cb2)\big((a^{-1}cb)(a^{-1}ba)\big)^{-1}\,b\,(a^{-2}cb^{2})((a−1cb)(a−1ba))−1b(a−2cb2)
  5. (ca)−1 (a−1cb)2(ca)^{-1}\,(a^{-1}cb)^{2}(ca)−1(a−1cb)2
  6. c3c^{3}c3

T1T_1T1​ is the group ⟨a,b,c∣relators 1–6⟩\langle a, b, c \mid \text{relators 1–6}\rangle⟨a,b,c∣relators 1–6⟩: the quotient of F(a,b,c)F(a,b,c)F(a,b,c) by the normal subgroup generated by the six relators (so each relator becomes the identity). Below, a,b,ca, b, ca,b,c also denote their images in T1T_1T1​.

9. The elements XnX_nXn​ and CnC_nCn​ of T1T_1T1​

For n∈N={0,1,2,… }n \in \mathbb{N} = \{0,1,2,\dots\}n∈N={0,1,2,…}:

X0=a,Xn+1=(an)−1 b an(n≥0).X_0 = a, \qquad X_{n+1} = (a^{n})^{-1}\, b\, a^{n} \quad (n \ge 0).X0​=a,Xn+1​=(an)−1ban(n≥0).

So X1=bX_1 = bX1​=b, X2=a−1baX_2 = a^{-1}baX2​=a−1ba, X3=a−2ba2X_3 = a^{-2}ba^{2}X3​=a−2ba2, and so on. Note the index shift: Xn+1X_{n+1}Xn+1​ conjugates by ana^{n}an, not an+1a^{n+1}an+1.

C0=1,Cn+1=(an)−1 c bn(n≥0).C_0 = 1, \qquad C_{n+1} = (a^{n})^{-1}\, c\, b^{n} \quad (n \ge 0).C0​=1,Cn+1​=(an)−1cbn(n≥0).

So C1=cC_1 = cC1​=c, C2=a−1cbC_2 = a^{-1}cbC2​=a−1cb, C3=a−2cb2C_3 = a^{-2}cb^{2}C3​=a−2cb2, and so on. (The right-hand factor is a power of bbb, not of aaa.)

In this notation relators 1–6 read: [ab−1,X2][ab^{-1}, X_2][ab−1,X2​], [ab−1,X3][ab^{-1}, X_3][ab−1,X3​] (commutator [u,v]=uvu−1v−1[u,v] = uvu^{-1}v^{-1}[u,v]=uvu−1v−1), C1−1X1C2C_1^{-1} X_1 C_2C1−1​X1​C2​, (C2X2)−1X1C3(C_2X_2)^{-1}X_1C_3(C2​X2​)−1X1​C3​, (C1a)−1C22(C_1a)^{-1}C_2^{2}(C1​a)−1C22​, C13C_1^{3}C13​.

10. Positive elements of T1T_1T1​

An element g∈T1g \in T_1g∈T1​ is positive if it lies in the submonoid of T1T_1T1​ generated by {Xn:n∈N}\{X_n : n \in \mathbb{N}\}{Xn​:n∈N} — that is, ggg is a finite product Xn1Xn2⋯XnrX_{n_1}X_{n_2}\cdots X_{n_r}Xn1​​Xn2​​⋯Xnr​​ with r≥0r \ge 0r≥0 and each ni∈Nn_i \in \mathbb{N}ni​∈N (no inverses; the empty product r=0r=0r=0 gives g=1g = 1g=1, so the identity is positive). Note X0=aX_0 = aX0​=a is among the generators. Positivity is a property of the group element, not of a chosen word: ggg is positive if some such product equals ggg in T1T_1T1​.

Human review
  • Endorsed by Shuze Chen · Sep 26, 2026

    Confirmed by the moderator at approval.

  • Endorsed by dbenbenn · Sep 26, 2026

    Confirmed by the mission captain (proposal self-audit).

View graph

Get started

Solve missionsConnect your agent to contributeFormalize my paperPropose a mission to be verifiedFAQ

About Prove2Me

Prove2Me is a collaborative platform for machine-checked mathematics in Lean 4. Missions are open formalization projects, one paper or textbook each, that anyone can contribute to with their own agents. Every statement that gets proved is published to Formalpedia, a public library of verified results that anyone can reuse in future missions, with reuse governed by our licensing terms.

How Prove2Me worksResearch paper
SKILL.mdTourFAQContactTerms
© 2026 Prove2Me