Holomorphic factorization of mappings into \(\mathrm{SL}_n(\mathbb{C})\) (Q764053): Difference between revisions

In the present paper, the authors solve Gromov's Vaserstein problem. Namely, they show that a null-homotopic holomorphic mapping from a finite-dimensional reduced Stein space into \(\text{SL}_m(\mathbb{C})\) can be factored into a finite product of unipotent matrices with holomorphic entries. After the Gromov-Eliashberg embedding theorem for Stein manifolds [\textit{Y. Eliashberg} and \textit{M. Gromov}, ``Embeddings of Stein manifolds of dimension \(n\) into the affine space of dimension \(3n/2 + 1\)'', Ann. Math. (2) 136, No. 1, 123--135 (1992; Zbl 0758.32012); \textit{J. Schürmann}, ``Embeddings of Stein spaces into affine spaces of minimal dimension'', Math. Ann. 307, No. 3, 381--399 (1997; Zbl 0881.32007)], this is another deep application of Gromov's holomorphic h-principle. \textit{M. Gromov} [``Oka's principle for holomorphic sections of elliptic bundles'', J. Am. Math. Soc. 2, No. 4, 851--897 (1989; Zbl 0686.32012)] extended the classical Oka-Grauert principle from bundles with homogeneous fibers to fibrations with elliptic fibers. The authors use the Oka-Grauert-Gromov-principle in its strongest form as elaborated by \textit{F. Forstnerič} [``The Oka principle for sections of stratified fiber bundles'', Pure Appl. Math. Q. 6, No. 3, 843--874 (2010; Zbl 1216.32005)] and \textit{F. Forstnerič} and \textit{J. Prezelj} [``Extending holomorphic sections from complex subvarieties'', Math. Z. 236, No. 1, 43--68 (2001; Zbl 0968.32005)]. Let us describe the result in more detail. It is well known that the group \(\mathrm{SL}_m(\mathbb{C})\) is generated by elementary matrices \(E+\alpha e_{ij}\), \(i\neq j\), i.e., matrices with 1's on the diagonal and all entries outside the diagonal are zero, except one entry. Equivalently, every matrix \(A\in \mathrm{SL}_m(\mathbb{C})\) can be written as a finite product of upper and lower diagonal matrices (in interchanging order). The same question for matrices in \(\mathrm{SL}_m(R)\), where \(R\) is a commutative ring (e.g. the ring of complex valued functions on a space \(X\) which are continuous, smooth, algebraic or holomorphic) instead of the field \(\mathbb{C}\), is much more delicate. For results in the algebraic setting, we refer to [\textit{P. M. Cohn}, ``On the structure of the \(\mathrm{GL}_2\) of a ring'', Publ. Math., Inst. Hautes Étud. Sci. 30, 365--413 (1966; Zbl 0144.26301); \textit{A. A. Suslin}, ``On the structure of the special linear group over polynomial rings'', Izv. Akad. Nauk SSSR, Ser. Mat. 41, 235--252 (1977; Zbl 0354.13009); \textit{F. Grunewald}, \textit{J. Mennicke} and \textit{L. Vaserstein}, ``On the groups \(\mathrm{SL}_2 (\mathbb{Z}[x])\) and \(\mathrm{SL}_2 (k[x,y])\)'', Isr. J. Math. 86, No. 1--3, 157--193 (1994; Zbl 0805.20042); \textit{D. Wright}, ``The amalgamated free product structure of \(\mathrm{GL}_2(k[X_1,\dots ,X_n ])\) and the weak Jacobian theorem for two variables'', J. Pure Appl. Algebra 12, 235--251 (1978; Zbl 0387.20039)]. We restrict our attention to the ring of holomorphic functions on a reduced Stein space \(X\). So, the problem amounts to factorizing a given holomorphic map \(X \rightarrow \mathrm{SL}_m(\mathbb{C})\) as a product of upper and lower diagonal unipotent matrices. Since any product of such matrices is homotopic to a constant map, one has to assume that the given map \(f\) is homotopic to a constant map, say null-homotopic. The main result of the present paper is a complete positive solution of Gromov's Vaserstein problem as posed in [Gromov, loc.cit.]: Let \(X\) be a finite dimensional reduced Stein space and \(f: X\rightarrow \mathrm{SL}_m(\mathbb{C})\) be a holomorphic mapping that is null-homotopic. Then there exists a natural number \(K\) and holomorphic mappings \(G_1, \dots, G_K: X\rightarrow \mathbb{C}^{m(m-1)/2}\) such that \(f\) can be written as a product of upper and lower diagonal unipotent matrices \[ f(x) = \left(\begin{matrix} 1 & 0\\ G_1(x) & 1\end{matrix}\right) \left(\begin{matrix} 1 & G_2(x)\\ 0 & 1\end{matrix}\right)\cdots \left(\begin{matrix} 1 & G_K(x)\\ 0 & 1\end{matrix}\right) \] for every \(x\in X\). As indicated above, the proof relies on an application of the Oka-Grauert-Gromov-principle to certain stratified fibrations. For this, a topological solution to the problem is required, i.e., a factorization of \(f\) into a finite product of unipotent matrices with continuous entries. That was achieved by \textit{L. N. Vaserstein} [``Reduction of a matrix depending on parameters to a diagonal form by addition operations'', Proc. Am. Math. Soc. 103, No. 3, 741--746 (1988; Zbl 0657.55005)] after some preliminary results by \textit{W. P. Thurston} and \textit{L. N. Vaserstein} [``On \(K_1\)-theory of the Euclidean space'', Topology Appl. 23, 145--148 (1986; Zbl 0611.18007)]. See [the authors, ``On the number of factors in the unipotent factorization of holomorphic mappings into \(\text{SL}_2(\mathbb C)\)'', Proc. Am. Math. Soc. 140, No. 3, 823--838 (2012; Zbl 1250.32009)] for an effecitive version of the main theorem with a bound on the number of factors \(K\) for the group \(\mathrm{SL}_2(\mathbb{C})\). The results of the paper under review had been announced in [``A solution of Gromov's Vaserstein problem'', C. R., Math., Acad. Sci. Paris 346, No. 23--24, 1239--1243 (2008; Zbl 1160.32017)].