High T_c: Top down or bottom up?

We present and discuss recent tunneling results, which shed some light on the two contradicting approaches to the high-T_c problem. The "top-down" approach starts from the ansatz that the anomalous properties of the cuprates in their normal state above T_c need to be understood first, before the problem of superconductivity can be tackled successfully. On the other hand, in the "bottom-up" approach, both high T_c and the anomalous normal-state properties come about because of the vicinity of a quantum critical point. Our new tunneling results indicate that in the superconducting state, at some critical doping, the order parameter goes from a pure d-wave on the underdoped side to a complex one on the overdoped one. This symmetry breaking lends support to the hypothesis that there does exist a quantum critical point and, therefore, favors the "bottom-up" approach.