Braid groups are linear (Q5915536): Difference between revisions

The question of whether the braid groups \(B_n\) (\(n\geq 2\)) are linear is an old one. The most famous representation, the so-called Burau representation, was shown by \textit{J. A. Moody} not to be faithful for \(n\geq 9\) [Bull. Am. Math. Soc., New Ser. 25, No. 2, 379-384 (1991; Zbl 0751.57005)]. It is now known that the Burau representation is faithful for \(n\leq 3\) and unfaithful for \(n\geq 5\) (the case \(n=4\) is still unsettled). In a previous paper [Invent. Math. 142, No. 3, 451-486 (2000; Zbl 0988.20023)], the author defined another representation \(\rho\colon B_n\to\text{GL}(V)\), where \(V\) is a free module of rank \(n(n-1)/2\) over a ring \(R\), and proved that it is faithful for \(n=4\). \textit{S. J. Bigelow} [J. Am. Math. Soc. 14, No. 2, 471-486 (2001; Zbl 0988.20021)] showed, using a topological argument, that \(\rho\) is faithful for all \(n\). In the present paper, the author exploits combinatorial properties of the action of \(B_n\) on \(\text{GL}(V)\) to give a completely different proof that \(\rho\) is faithful, and hence that all braid groups are linear.