Maps that must be affine or conjugate affine: a problem of Vladimir Arnold (Q2209756)

The paper consists of three main results, two of which we will combine into one, by providing a second reading through the words in parentheses. Let \(F\) be an Archimedean ordered field and \(K\) be a topological division algebra over \(F\) (with the natural topology). An automorphism \(\sigma: F\rightarrow F\) which fixes \(F\) pointwise is called an \(F\)-automorphism. Let \(V\) be a finite-dimensional vector (projective) space over \(K\) with dimension (projective dimension) at least \(2\). Let \(T : V\rightarrow V \) be a map. If (i) \(T\) is injective, (ii) \(T\) is continuous at some point of \(V\) or \(F = {\mathbb R}\), \(\dim_F K < \infty\) and \(T\) is Lebesgue measurable, and (iii) there is some \(k\) with \(1 \leq k < \dim V\), such that \(T\) maps each \(k\)-flat to a flat; then \(T\) is a bijection and is \(\sigma\)-semiaffine (\(\sigma\)-linear) for some \(F\)-automorphism \(\sigma\) of \(K\). If \(K =F[\sqrt{-1}]\), then \(T\) is either affine (linear) or conjugate affine (conjugate linear). The third theorem provides a counterexample, and states that: For every \(n\geq 2\), there is a map \(T : {\mathbb CP}^n \rightarrow {\mathbb CP}^n\), such that: (i) \(T\) is surjective; (ii) under the map \(T\), each point of \({\mathbb CP}^n\) is the image of infinitely many points and, therefore, \(T\) is not injective; (iii) for all \(m \in \{1, 2, \ldots, n - 1\}\) and each \(m\)-flat, \(P\), of \({\mathbb CP}^n\) the image \(T[P]\) is an \(m\)-flat of \({\mathbb CP}^n\); (iv) for all \(m \in \{1, 2, \ldots, n - 1\}\), every \(m\)-flat of \({\mathbb CP}^{n}\) is the image under \(T\) of an \(m\)-flat of \({\mathbb CP}^{n}\). The conditions of continuity at one point or Lebesgue measurability are meant to prevent exotic solutions to Cauchy's functional equation. The first two results follow from what the authors believe to be stronger versions of the fundamental theorems of affine and of projective geometry. However, the authors cite only [\textit{J. F. Rigby}, Mathematica (Cluj) 10, No. 33, 369--383 (1968; Zbl 0184.23901)] and [\textit{A. Chubarev} and \textit{I. Pinelis}, Proc. Am. Math. Soc. 127, No. 9, 2735--2744 (1999; Zbl 0930.51002)] from the 20th century research literature regarding affine and projective transformations. It would have been desirable for the authors to engage with the results obtained in [\textit{D. S. Carter} and \textit{A. Vogt}, Mem. Am. Math. Soc. 235, 98 p. (1980; Zbl 0444.51004); \textit{M. Saltzwedel}, Kennzeichnung von Lineationen desarguesscher affiner Räume. Hamburg: Univ., FB Math. (1995; Zbl 0841.51001); J. Geom. 56, No. 1--2, 142--160 (1996; Zbl 0855.51003); Butzer, P. L. (ed.) et al., Karl der Große und sein Nachwirken. 1200 Jahre Kultur und Wissenschaft in Europa. Band 2: Mathematisches Wissen. Turnhout: Brepols. 287--296 (1998; Zbl 0960.51002); \textit{A. Brezuleanu} and \textit{D.-C. Rădulescu}, J. Geom. 23, 45--60 (1984; Zbl 0553.51010); Abh. Math. Semin. Univ. Hamb. 55, 171--181 (1985; Zbl 0595.51006); \textit{R. Frank}, Geom. Dedicata 44, No. 1, 53--66 (1992; Zbl 0767.51018)], so that the real novelties of the first two results could come to light.