A simple presentation for the mapping class group of an orientable surface (Q789781): Difference between revisions

The mapping class group of an orientable surface is the group of isotopy classes of orientation preserving self-diffeomorphisms of this surface which are the identity on the boundary. The main result of the paper is a presentation of this group for closed surfaces and for surfaces with one boundary component. The mapping class groups are very important both in low-dimensional topology and in the theory of Riemannian surfaces (but studied mainly by topologists). To write down a finite presentation of the mapping class group was a long standing problem (it is still unsolved for surfaces with many boundary components), in the solution of which many mathematicians were involved. But until this paper no solution was known. \textit{W. B. R. Lickorish} [J. Math. Proc. Camb. Philos. Soc. 60, 769-778 (1964; Zbl 0131.208)] proved that the mapping class group is finitely generated and \textit{J. McCool} [J. Algebra 35, 205-213 (1975; Zbl 0325.20025)] proved that this group is finitely presented. Lickorish had used the geometrical methods; his generators are Dehn twists. On the contrary, McCool's proof was based on combinatorial group theory. It is difficult to derive a concrete presentation from McCool's proof. At least, nobody succeeded in this. In the meantime, \textit{J. Birman} and \textit{H. Hilden} [Advances Theory Riemann Surfaces, Proc. 1969 Stony Brook Conf., Ann. Math. Stud. 66, 81- 115 (1971; Zbl 0217.486)] found a presentation of the mapping class group in the first nontrivial case of the closed surface of genus two. Entirely new ideas were introduced in this field by \textit{A. Hatcher} and \textit{W. Thurston} [see Topology 19, 221-237 (1980; Zbl 0447.57005)]. They reproved the finite presentability of the mapping class group by a two dimensional version of the Morse-Cerf theory (the Morse theory with parameters). Hatcher and Thurston had constructed a two-dimensional cell complex and an action of the mapping class group on this complex. The basic result is the simply-connectedness of this complex. From this result and elementary properties of the action a finite presentation can be derived. But the presentation so obtained is very complicated and indeed no presentation was explicitly written down by Hatcher and Thurston. In his computation of the second cohomology group of the mapping class group, \textit{J. Harer} (see the following review) found a smaller complex with the same properties as the Hatcher-Thurston's one (its simply- connectedness is derived from simply-connectedness of the Hatcher- Thurston complex). This complex is used by the author to derive a simple finite presentation. His generators are Dehn twists. Most of the relations are very simple (of well known form \(aba=bab)\), but three of them are very cumbersome. The proofs are difficult and technical, involving a lot of word computations. The next problem is to derive some consequences from this presentation. (As is pointed out in the paper, Harer also obtained an explicit presentation but his presentation was still rather complicated.)