Values obtained for both sedimentation and convergence rates in certain trenches are inconsistent with the presence of thick, undeformed sediments. We propose a steady-state model in which the geometry of the sedimentary prism in a trench could remain constant by having inward-advancing deformation of sediments along the inner wall exactly balance outward-advancing overlap of new sediments on the outer wall. However, if the advance of deformation into a trench is considered to be equal to or greater than the convergence rate across the trench, it is clear that significant thicknesses of undisturbed trench sediments should not exist and that a steady state cannot exist. Therefore, for trenches containing thick, undeformed sediments, the convergence of crustal plates is being accommodated by erogenic deformation away from the trench, and the oceanic lithosphere is, in fact, strongly coupled to the continental lithosphere. There may be no discrete structural expression of a subduction zone at shallow crustal levels. All plate convergence during the time of trench filling can be reasonably converted into an episode of deformation and uplift of the adjacent erogenic belt.
The convergent motions of lithosphere plates through geologic time are too great to have all convergence accounted for by orogenic deformation and uplift. A bimodal behavior of subduction zones seems necessary, having lithosphere coupling and orogenesis alternate with lithosphere decoupling and simple underthrusting in trenches. The two modes may not be exclusive nor may they be the only modes. But perhaps such modes explain the paradox of continuous sea-floor spreading versus episodic orogenesis.