This paper investigates experimentally the effectiveness of a new seismic design philosophy, in which soil failure is “utilized” to protect the superstructure. A physical model of a simple bridge pier is used as an example. Two alternatives are considered: one in compliance with conventional capacity design, with over-designed foundation so that the plastic “hinge” will develop at the bridge pier ; and one following the new philosophy, with under-designed foundation, moving the plastic “hinging” will develop in the soil or the soil foundation interface. The seismic performance of the two alternatives is investigated through 1-G shaking table testing using real records and synthetic motions as base excitation. It is shown that the performance of the new design concept can be advantageous : in cases where the conventionally designed system collapses, the new concept can survive the seismic motion with the “damage” being limited to residual deck drift and increased settlement.