REVIEW 3 major objections 4 minor 12 references
Automorphism towers of groups of homeomorphisms of Cantor space
T0 review · 3 major / 4 minor · reviewed 2026-08-14 · deepseek-v4-flash
Pith's one-line read For every full and flexible group G of homeomorphisms of Cantor space, and for the generalized Thompson groups T_{n,r}, the paper proves Aut(Aut(G)) = Aut(G), so the automorphism tower stabilizes at height one.
desk verdict The full flexible group theorem is clean and new; the T_{n,r} theorem is a dependency-rich corollary that needs verification. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The load-bearing mechanism for the full-group case is the interaction between fullness and small supports: an element of small support is one that acts as the identity outside a proper closed set, and the paper shows both that a full flexible group is generated by such elements and that any small-support element in the normalizer of the group is forced back into the group. A reconstruction theorem for locally moving group actions supplies the bridge from automorphisms to normalizers inside the full homeomorphism group. For the $T_{n,r}$ case the mechanism is the core of a bi-synchronizing transducer: when a transducer is synchronizing, long enough input words force the automaton into a strongly connected sub-transducer, its core, and the germ group at a point of the circle decomposes as a product of the core group with one or two integer shifts. The proof uses the core-product formula $\mathrm{Core}(A^h f^h) = \mathrm{Core}(A^h) \mathrm{Core}(f^h)$ to cancel cores of conjugated germs and force the conjugated element back into $T_{n,r}$.
What would settle it
For the $T_{n,r}$ claim, a concrete falsifier would be an explicit choice of parameters $n,r$, a circle homeomorphism $h$, and elements $A \in T B_{n,r}$, $f \in T_{n,r}$ for which $\mathrm{Core}(A^h f^h) \neq \mathrm{Core}(A^h)\mathrm{Core}(f^h)$; the cancellation step in Corollary 3.14 requires this equality for all such choices. For the full-group claim, a counterexample would be a full flexible group $G$ and a homeomorphism $h$ with $h^{-1}\mathrm{Aut}(G)h \subseteq \mathrm{Aut}(G)$ but $h^{-1}Gh \nsubseteq G$, which would falsify Lemma 2.8 and hence Theorem 1.1.
Extended reading notes
Core claim
The central claim, stated as Theorem 1.1, is that a full and flexible group G of homeomorphisms of Cantor space satisfies $\mathrm{Aut}(\mathrm{Aut}(G)) = \mathrm{Aut}(G)$. Here 'full' means that every homeomorphism of the space that locally agrees with G is itself in G, and 'flexible' means that G can move any small nonempty clopen set into any other. The proof identifies $\mathrm{Aut}(G)$ with the normalizer of G inside the full homeomorphism group of the Cantor space, proves that each element of that normalizer with small support is already in G, and uses the fact that G is generated by its small-support elements. For the groups $T_{n,r}$, which are flexible but not full, the paper proves the same equality as Theorem 1.3 by describing $\mathrm{Aut}(T_{n,r})$ as the group $T B_{n,r}$ of bi-synchronizing transducers and classifying the germs of its elements at points of the circle; from that classification it follows that any homeomorphism normalizing $\mathrm{Aut}(T_{n,r})$ also normalizes $T_{n,r}$ itself.
Load-bearing premise
For the $T_{n,r}$ half, the load-bearing premise is the as-yet-unpublished characterization $\mathrm{Aut}(T_{n,r}) \cong T B_{n,r}$ together with the core-product cancellation formula used in Corollary 3.14; if either input fails, Theorem 1.3 is unsupported.
Editorial extensions
If this is right
- Every group in the class of full flexible Cantor-space homeomorphisms has automorphism tower of height one.
- In particular, $\mathrm{Aut}(\mathrm{Aut}(G_{n,r})) = \mathrm{Aut}(G_{n,r})$ for the Higman-Thompson groups, and the same holds for the rational group $\mathcal{R}_2$, the Nekrashevych groups $V_n(G)$, and the transducer overgroups $V_n(T)$.
- For the generalized Thompson groups $T_{n,r}$, the tower also stops at height one, extending the known height-one results for Thompson's groups $F$ and $T$ to all valid parameters.
- Every automorphism of $\mathrm{Aut}(G)$ is induced by conjugation by an element of $\mathrm{Aut}(G)$; equivalently, $\mathrm{Out}(\mathrm{Aut}(G))$ is trivial for these groups.
- Any homeomorphism of Cantor space that normalizes $\mathrm{Aut}(G_{n,r})$ is rational and induced by a synchronizing transducer.
Reading between the lines
- The germ-and-core strategy may extend to other flexible-but-not-full groups of homeomorphisms of the circle or Cantor space whose automorphism groups admit a similar transducer or germ description; this is an extension to test, not a claim of the paper.
- Because the tower stops at height one for the rational group $\mathcal{R}_2$, the open question whether $\mathrm{Aut}(\mathcal{R}_2) = \mathcal{R}_2$ can only fail through an outer automorphism already present in $\mathrm{Aut}(\mathcal{R}_2)$; no new automorphism can appear at the next level.
- If the unpublished input for $T_{n,r}$ is later replaced by a published proof, the same theorem would give a direct route to proving triviality of the outer automorphism group of $\mathrm{Aut}(T_{n,r})$ without redoing the germ analysis.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper proves that for every full and flexible group G of homeomorphisms of Cantor space, Aut(Aut(G)) = Aut(G) (Theorem 1.1), and that for the generalized Thompson group T_{n,r}, Aut(Aut(T_{n,r})) = Aut(T_{n,r}) (Theorem 1.3). The full-flexible proof uses Rubin's reconstruction theorem to identify Aut(G) and Aut(Aut(G)) with normalizers in the homeomorphism group of Cantor space, then shows that any homeomorphism normalizing Aut(G) must already normalize G; since full groups are generated by small-support elements, the automorphism tower stabilizes at height one. The proof for T_{n,r} instead analyzes germs of elements of Aut(T_{n,r}), using the identification Aut(T_{n,r}) is isomorphic to T B_{n,r} and the core of a synchronizing transducer, and concludes that any homeomorphism normalizing Aut(T_{n,r}) also normalizes T_{n,r}. Applications are recorded for Higman-Thompson groups, the rational group R_2, the Röver group, Nekrashevych groups, and the groups V_n(T).
Significance. If Theorem 1.1 holds, it gives a uniform and conceptually simple proof that automorphism towers of a broad class of Cantor-space groups stabilize at height one, unifying and extending earlier results for the groups G_{n,r} and R_2. The full-flexible half is a genuine advance: Lemmas 2.6 through 2.8 and Corollary 2.13 are coherent and self-contained modulo Rubin's theorem, and the paper's identification of small-support elements as the engine of the argument is elegant. The T_{n,r} half aims to extend Brin and Guzmán's work to the full two-parameter family, but its proof is substantially less self-contained: it rests on the unpublished preprint [9] for the identification of Aut(T_{n,r}) with T B_{n,r}, and on an unproved core-product formula. The paper openly notes the limitation that Lemma 2.6 does not extend to arbitrary compact Hausdorff spaces, which is a helpful and honest boundary statement.
major comments (3)
- [§3.4, Lemma 3.13] The proof of Lemma 3.13 contains the load-bearing assertion, stated as 'It is not hard to verify', that the product of two elements of T B_{n,r} has core equal to the product of the cores and that the Z×Z parameters add exactly. This product formula is not proved, and Corollary 3.14 later uses it to cancel Core(A^h) and conclude that f^h has trivial core. If the core map is not exactly a homomorphism to T O_{n,r}, or if the product of cores is not strongly connected, the conclusion that f^h lies in T_{n,r} fails. Since Theorem 3.15 depends on exactly this step, the formula needs a detailed proof or a precise citation to a published source.
- [§3.3, Theorem 3.6] Theorem 1.3 depends on the identification Aut(T_{n,r}) is isomorphic to T B_{n,r}, quoted from the author's unpublished preprint [9], which is listed as 'In Preparation'. Corollary 3.14 and Theorem 3.15 also use the product structure on T O_{n,r} from [9, 1]. No proof of Theorem 3.6 is included, so the T_{n,r} half of the paper is currently conditional on an external unpublished result. The manuscript should either include a proof of Theorem 3.6 in an appendix, replace the reference with a peer-reviewed published version, or explicitly state Theorem 1.3 as conditional on [9].
- [§3.4, Corollary 3.14] Even accepting Theorem 3.6, the step 'From this it follows that A^h and A^h f^h have the same core' requires that equality of germs at a point implies equality of cores for arbitrary elements of T B_{n,r}, and the subsequent 'this is true precisely when f^h has trivial core' requires a cancellation argument in T O_{n,r}. These facts are asserted rather than proved in the manuscript. They are not consequences of Lemma 3.13 alone unless the core product formula is established exactly as stated, so the reasoning in Corollary 3.14 is not yet a complete proof.
minor comments (4)
- [Throughout] There are several typographical slips, including 'Propostion' in Proposition 2.11, 'Propositon' in Corollary 2.13, and 'antichians' in Section 3.2; these should be corrected.
- [§3.4, Lemma 3.13] In the displayed conditions of Lemma 3.13(1), the expression 'πB(τ(n−1)kp0)' should presumably read 'πB(τ(n−1)^k, p0)', and 'λB(τ′(n−1)^k, p0)' contains τ′ where τ appears to be intended.
- [§2.2, Theorem 2.9] The proof of Theorem 2.9 is only the sentence 'This is a straight-forward consequence of Lemma 2.8'; expanding this one-line proof would help the reader see exactly how the normalizer inclusions are converted into the automorphism-tower equality.
- [§3.3, Theorem 3.7] The paper states Theorem 3.7 from [9] without proof; if [9] is not yet available, a brief explanation of why Out(T_{n,r}) is relevant to the tower result would clarify the structure of Section 3.
Circularity Check
No circular derivation is present; the full-flexible theorem is self-contained, while the T_{n,r} theorem depends on an unpublished self-citation that is a correctness risk but not a circular reduction.
full rationale
The paper's main full-group theorem (Theorem 1.1) is derived self-containedly: Rubin's theorem (Theorem 2.10) is cited as an external tool, and Lemmas 2.6, 2.7, and 2.8 are proved in the text. Corollary 2.13 then follows from the normalizer characterization without fitting parameters or restating the conclusion as an input. No circular definition or fitted-input-as-prediction appears in this part. The T_{n,r} half is structurally different: Theorem 3.6, namely Aut(T_{n,r}) is isomorphic to T B_{n,r}, is imported from the author's own in-preparation paper [9], and Lemma 3.13's core-product assertion is stated with 'it is not hard to verify' rather than proved. These are load-bearing dependencies and genuine correctness risks, but they are not circular: the target claim Aut(Aut(T_{n,r})) = Aut(T_{n,r}) is not defined in terms of Theorem 3.6, and the proof still requires additional germ analysis in Lemma 3.13 and Corollary 3.14. The paper does not rename a known result as a new one, nor does it invoke a uniqueness theorem from the same authors to forbid alternatives. Thus the derivation chain is not equivalent to its inputs by construction; the appropriate finding is no significant circularity, with score 2 reflecting the self-citation dependency rather than a demonstrated circular step.
Assumptions & free parameters
assumptions (5)
- standard math Rubin's reconstruction theorem (Theorem 2.10): for locally moving groups on Cantor space, a group isomorphism is induced by a homeomorphism of the space.
- standard math McCleary-Rubin theorem (Theorem 3.10): an o-3-transitive group of orientation-preserving circle homeomorphisms has its automorphisms induced by homeomorphisms.
- domain assumption Aut(G_{n,r}) is isomorphic to B_{n,r} (Theorem 3.5).
- domain assumption Aut(T_{n,r}) is isomorphic to T B_{n,r} (Theorem 3.6).
- domain assumption The core product formula Core(A_h f_h) = Core(A_h) Core(f_h) holds and T O_{n,r} is a group.
Cite this review
Pith. "Pith review of Automorphism towers of groups of homeomorphisms of Cantor space." pith.science (2026). https://pith.science/paper/7EIUPK4O
@misc{pith2026190803815,
author = {Pith},
title = {Pith review of: Automorphism towers of groups of homeomorphisms of Cantor space},
year = {2026},
howpublished = {\url{https://pith.science/paper/7EIUPK4O}},
note = {Machine review of arXiv:1908.03815}
}
abstract
We show that for any full and sufficiently transitive (i.e. \textit{flexible}) group $G$ of homeomorphisms of Cantor space, $\mathrm{Aut}(\mathrm{Aut}(G)) = \mathrm{Aut}(G)$. This class contains many generalisations of the Higman-Thompson groups $G_{n,r}$, and the Rational group $\mathcal{R}_{2}$ of Grigorchuk, Nekrashevych, and Suchanski{\u \i}. We also demonstrate that for generalisations $T_{n,r}$ of R. Thompson's group $T$, $\mathrm{Aut}(\mathrm{Aut}(T_{n,r}))= \mathrm{Aut}(T_{n,r})$. In the case of the groups $G_{n,r}$ and $T_{n,r}$ our results extend results of Brin and Guzm{\' a}n for Thompson's group $T$, and generalisations of Thompson's group $F$.
Reference graph
Works this paper leans on
-
[9]
Feyishayo Olukoya, Automorphisms of the generalised Thompson ’s group Tn,r, In Preparation, 2018, pp. 1–35
work page 2018
-
[1]
Collin Bleak, Peter Cameron, Yonah Maissel, Andr´ es Nav as, and Feyishayo Olukoya, The further chameleon groups of Richard Thompson and Graham Higm an: Automorphisms via dynamics for the Higman groups Gn,r, 2016, submitted, pp. 1–89
work page 2016
-
[2]
Brin, The chameleon groups of Richard J
Matthew G. Brin, The chameleon groups of Richard J. Thompson: automorphisms an d dynamics, Inst. Hautes ´Etudes Sci. Publ. Math. (1996), no. 84, 5–33 (1997). MR 14410 05
work page 1996
-
[3]
Brin and Fernando Guzm´ an, Automorphisms of generalized Thompson groups , J
Matthew G. Brin and Fernando Guzm´ an, Automorphisms of generalized Thompson groups , J. Algebra 203 (1998), no. 1, 285–348. MR 1620674
work page 1998
-
[4]
Casey Donoven and Feyishayo Olukoya, Conjugate subgroups and overgroups of Vn, 2018
work page 2018
-
[5]
R. I. Grigorchuk, V. V. Nekrashevich, and V. I. Sushchans ki ˘ ı,Automata, dynamical systems, and groups , Tr. Mat. Inst. Steklova 231 (2000), no. Din. Sist., Avtom. i Beskon. Gruppy, 134–214. MR 1841755 (2002m:37016)
work page 2000
-
[6]
Australia n National University, Canberra, 1974, Notes on Pure Mathematics, No
Graham Higman, Finitely presented infinite simple groups , Department of Pure Mathematics, Department of Mathematics, I.A.S. Australia n National University, Canberra, 1974, Notes on Pure Mathematics, No. 8 (1974). MR 0 376874 (51 #13049)
work page 1974
-
[7]
Stephen H. McCleary and Matatyahu Rubin, Locally moving groups and the reconstruction problem for chains and circles 1 , 2005
work page 2005
Show all 12 references
-
[8]
117, American Mathematical Society, Providence, RI, 2005
Volodymyr Nekrashevych, Self-similar groups, Mathematical Surveys and Monographs, vol. 117, American Mathematical Society, Providence, RI, 2005. MR 2162164
2005
-
[10]
R¨ over, Constructing finitely presented simple groups that contain G rigorchuk groups, J
Claas E. R¨ over, Constructing finitely presented simple groups that contain G rigorchuk groups, J. Algebra 220 (1999), no. 1, 284–313. MR 1714140
1999
-
[11]
Appl., vol
Matatyahu Rubin, Locally moving groups and reconstruction problems , Ordered groups and infinite permutation groups, Math. Appl., vol. 354, Kluwer A cad. Publ., Dordrecht, 1996, pp. 121–157. MR 1486199 (99d:20003)
1996
-
[12]
Thompson, Notes on three groups of homeomorphisms , Unpublished but widely circulated handwritten notes (1965), 1–11
Richard J. Thompson, Notes on three groups of homeomorphisms , Unpublished but widely circulated handwritten notes (1965), 1–11. 11
1965
Reviewed August 14, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.