A better proof of the Goldman-Parker conjecture. (Q2388875)

\(\mathbb{CH}^2\) is the 2-dimensional complex hyperbolic space, and \(\text{PU}(2,1)\) is its isometry group. A \(\mathbb{C}\)-reflection is an order 2 element of \(\text{PU}(2,1)\) whose action is conjugate to the action of \((z,w)\mapsto(z,-w)\). Let \(G\) be the reflection ideal triangle group acting on the hyperbolic plane \(\mathbb{H}^2\), with standard generators \(\iota_0,\iota_1,\iota_2\). A `complex hyperbolic ideal triangle group representation' of \(G\) is a representation which maps the generators to \(\mathbb{C}\)-reflections, and the products of pairs of generators to parabolic elements. \(\text{Rep}(G)\) is the set of such representations modulo conjugacy. This set was introduced by \textit{W. M. Goldman} and \textit{J. R. Parker} [in J. Reine Angew. Math. 425, 71-86 (1992; Zbl 0739.53055)], who showed it is naturally parametrized by the half-open interval \([0,\infty)\). Goldman and Parker proved that an element \(\rho_s\) of this space is a discrete embedding if \(s\in[0,\underline s]\) with \(\underline s=\sqrt{105/3}\). They conjectured that \(\rho_s\) is a discrete embedding if and only if \(\rho_s(\iota_0\iota_1\iota_2)\) is not an elliptic element of \(\text{PU}(2,1)\). The author of the paper under review, [in Ann. Math. (2) 153, No. 3, 533-598 (2001; Zbl 1055.20040)], gave a computer-assisted proof of this conjecture. In the present paper, the author gives a new proof of this conjecture, which is traditional (i.e. not computer-assisted).