← All problems

Grothendieck–Teichmüller conjecture

5 years old · documented by

Let F^2\widehat F_2 be the profinite completion of the free group on x,yx,y, and F^2′\widehat F_2' the closure of its commutator subgroup. For f∈F^2f\in\widehat F_2 write f(a,b)f(a,b) for the image of ff under the continuous homomorphism x↦ax\mapsto a, y↦by\mapsto b. Let Γ^0,5\widehat\Gamma_{0,5} be the profinite completion of the mapping class group of the sphere with 55 marked points, or equivalently of π1(M0,5(C))\pi_1(\mathcal M_{0,5}(\mathbb C)). Let xijx_{ij} be its standard Dehn twist generators.

Definition (Drinfeld). GT^\widehat{GT} is the group of invertible elements of the monoid of pairs (λ,f)∈Z^×F^2′(\lambda,f)\in\widehat{\mathbb Z}\times\widehat F_2' satisfying

(I)f(x,y) f(y,x)=1,(II)f(z,x) zm f(y,z) ym f(x,y) xm=1where xyz=1, m=λ−12,(III)f(x12,x23) f(x34,x45) f(x51,x12) f(x23,x34) f(x45,x51)=1in Γ^0,5.\begin{aligned} &\text{(I)}\quad f(x,y)\,f(y,x)=1,\\ &\text{(II)}\quad f(z,x)\,z^m\,f(y,z)\,y^m\,f(x,y)\,x^m=1\quad\text{where } xyz=1,\ m=\tfrac{\lambda-1}{2},\\ &\text{(III)}\quad f(x_{12},x_{23})\,f(x_{34},x_{45})\,f(x_{51},x_{12})\,f(x_{23},x_{34})\,f(x_{45},x_{51})=1\quad\text{in }\widehat\Gamma_{0,5}. \end{aligned}

Each such pair defines an endomorphism of F^2\widehat F_2 by x↦xλx\mapsto x^\lambda, y↦f−1yλfy\mapsto f^{-1}y^\lambda f, and the monoid law is composition. For (λ,f)∈GT^(\lambda,f)\in\widehat{GT} one has λ∈Z^×\lambda\in\widehat{\mathbb Z}^\times [1].

The absolute Galois group GQ=Gal⁡(Q‾/Q)G_{\mathbb Q}=\operatorname{Gal}(\overline{\mathbb Q}/\mathbb Q) acts on the étale fundamental group π1eˊt(PQ‾1∖{0,1,∞})≅F^2\pi_1^{\mathrm{\acute et}}(\mathbb P^1_{\overline{\mathbb Q}}\setminus\{0,1,\infty\})\cong\widehat F_2, using the tangential base point at 00. Here σ\sigma acts by x↦xχ(σ)x\mapsto x^{\chi(\sigma)}, y↦fσ−1yχ(σ)fσy\mapsto f_\sigma^{-1}y^{\chi(\sigma)}f_\sigma, where χ\chi is the cyclotomic character and fσ∈F^2′f_\sigma\in\widehat F_2' measures the action on the path from 00 to 11. Drinfeld and Ihara showed that (χ(σ),fσ)(\chi(\sigma),f_\sigma) satisfies (I)–(III). Belyi's theorem shows the resulting homomorphism is injective [1, 2, 3]:

ι ⁣:GQ↪GT^.\iota\colon G_{\mathbb Q}\hookrightarrow\widehat{GT}.

Conjecture. The map ι\iota is surjective, so ι ⁣:GQ→ ∼ GT^\iota\colon G_{\mathbb Q}\xrightarrow{\ \sim\ }\widehat{GT}.

A disproof is a proof that ι\iota is not surjective. This question, raised by Drinfeld and Ihara in 1990, is the most common meaning of "the Grothendieck–Teichmüller conjecture" [6]. Two related statements carry similar names. One is Grothendieck's prediction in the Esquisse that GQG_{\mathbb Q} is the automorphism group of the tower of profinite fundamental groups of all moduli spaces Mg,n\mathcal M_{g,n} [4]. The other is the graded or pro-unipotent version described below.

Background

Grothendieck's Esquisse d'un programme [4] proposed studying GQG_{\mathbb Q} through its action on the "Teichmüller tower" of the groupoids π1(Mg,n)\pi_1(\mathcal M_{g,n}), linked by the maps that forget points and glue surfaces. He conjectured that this combinatorial object determines GQG_{\mathbb Q}. He also predicted that the levels of modular dimension at most 22, namely M0,4\mathcal M_{0,4}, M0,5\mathcal M_{0,5}, M1,1\mathcal M_{1,1} and M1,2\mathcal M_{1,2}, determine the whole tower. Drinfeld [1] found GT^\widehat{GT} independently while studying quasi-triangular quasi-Hopf algebras. Relations (I)–(III) are the profinite form of the hexagon and pentagon equations for associators. Ihara [2] developed the Galois-theoretic side.

The pro-unipotent analogue replaces F^2\widehat F_2 by the Malcev completion of F2F_2 over Q\mathbb Q. This gives a pro-unipotent group GTGT and its graded version GRTGRT. The corresponding conjecture is Drinfeld's prediction that the Lie algebra grt\mathfrak{grt} is free on generators σ3,σ5,σ7,…\sigma_3,\sigma_5,\sigma_7,\dots. Equivalently, the unipotent radical of the motivic Galois group of mixed Tate motives over Z\mathbb Z maps isomorphically onto the unipotent part of GRTGRT.

References
  1. V. G. Drinfeld, "On quasitriangular quasi-Hopf algebras and on a group that is closely connected with Gal(Q‾/Q)\mathrm{Gal}(\overline{\mathbb Q}/\mathbb Q)", Algebra i Analiz 2 (1990), no. 4, 149–181; English transl. Leningrad Math. J. 2 (1991), 829–860.
  2. Y. Ihara, "Braids, Galois groups, and some arithmetic functions", in Proceedings of the International Congress of Mathematicians (Kyoto, 1990), Vol. I, Math. Soc. Japan (1991).
  3. G. V. Belyĭ, "On Galois extensions of a maximal cyclotomic field", Math. USSR-Izv. 14 (1980), 247–256. doi:10.1070/IM1980v014n02ABEH001096
  4. A. Grothendieck, "Esquisse d'un programme" (1984), in Geometric Galois Actions 1, London Math. Soc. Lecture Note Ser. 242, Cambridge Univ. Press (1997), 7–48. doi:10.1017/CBO9780511758874.003
  5. D. Harbater and L. Schneps, "Fundamental groups of moduli and the Grothendieck–Teichmüller group", Trans. Amer. Math. Soc. 352 (2000), 3117–3148. doi:10.1090/S0002-9947-00-02347-3
  6. P. Lochak and L. Schneps, "Open problems in Grothendieck–Teichmüller theory", in Problems on Mapping Class Groups and Related Topics, Proc. Sympos. Pure Math. 74, Amer. Math. Soc. (2006), 165–186. doi:10.1090/pspum/074/2264540
  7. A. Hatcher, P. Lochak and L. Schneps, "On the Teichmüller tower of mapping class groups", J. Reine Angew. Math. 521 (2000), 1–24. doi:10.1515/crll.2000.028
  8. M. Boggi, "Automorphisms of profinite mapping class groups", preprint (2020, revised 2026). arXiv:2011.15075
  9. A. Minamide and H. Nakamura, "The automorphism groups of the profinite braid groups", Amer. J. Math. 144 (2022), 1159–1176. doi:10.1353/ajm.2022.0034
  10. F. Pop, "Finite tripod variants of I/OM: on Ihara's question/Oda–Matsumoto conjecture", Invent. Math. 216 (2019), 745–797. doi:10.1007/s00222-019-00855-8
  11. F. Brown, "Mixed Tate motives over Z\mathbb Z", Ann. of Math. 175 (2012), 949–976. doi:10.4007/annals.2012.175.2.10
  12. R. Hain and M. Matsumoto, "Weighted completion of Galois groups and Galois actions on the fundamental group of P1−{0,1,∞}\mathbb P^1-\{0,1,\infty\}", Compos. Math. 139 (2003), 119–167. doi:10.1023/B:COMP.0000005077.42732.93
  13. H. Furusho, "Pentagon and hexagon equations", Ann. of Math. 171 (2010), 545–556. doi:10.4007/annals.2010.171.545
  14. Wikipedia, Grothendieck–Teichmüller group.

Source: Marco Boggi, “Automorphisms of profinite mapping class groups”, arXiv:2011.15075 (2026).

Progress summary

Refreshed
Open

The known inclusion of the arithmetic symmetry group into Drinfeld’s group has not been shown to be onto, so the conjecture remains open.

Raised by Drinfeld and Ihara in 1990, the conjecture asserts that the faithful embedding GQ↪GT^G_{\mathbb Q}\hookrightarrow\widehat{GT} is surjective, hence an isomorphism. No retrieved source establishes this exact profinite-group statement.

Known results

  • Drinfeld and Ihara (1990–1991) constructed the embedding; Belyi (1980) gives its injectivity.
  • Harbater–Schneps (2000) identified GT^\widehat{GT} with the genus-zero tower’s compatible automorphisms.
  • Boggi (2020, revised 2026) proved GT^≅Aut⁡(Tˇout)\widehat{GT}\cong\operatorname{Aut}(\check{\mathfrak T}^{\mathrm{out}}) for the procongruence tower, but only a faithful representation for the profinite tower.
  • Furusho (2010) showed that the pentagon relation implies the hexagon relations.

2025–2026 related developments

A 2025 preprint explicitly states that surjectivity remains open; its version-2 Cantor-set homeomorphism concerns only underlying spaces, not groups. 2026 claims about the related Deligne–Drinfeld Lie-algebra conjecture, including Willwacher’s analysis, do not imply the profinite conjecture and remain unverified.

Current status (as of November 2026): The embedding GQ↪GT^G_{\mathbb Q}\hookrightarrow\widehat{GT} and several tower-level analogues are established, but surjectivity onto GT^\widehat{GT} remains open.

Sources

Solutions 0

No solutions have been posted yet.