Where and how subduction zones initiate is a fundamental tectonic problem, yet there are few well-constrained geologic tests that address the tectonic settings and dynamics of the process. Numerical modeling has shown that oceanic spreading centers are some of the weakest parts of the plate tectonic system [Gurnis M, Hall C, Lavier L (2004) Geochem Geophys Geosys 5:Q07001], but previous studies have not favored them for subduction initiation because of the positive buoyancy of young lithosphere. Instead, other weak zones, such as fracture zones, have been invoked. Because these models differ in terms of the ages of crust that are juxtaposed at the site of subduction initiation, they can be tested by dating the protoliths of metamorphosed oceanic crust that is formed by underthrusting at the beginning of subduction and comparing that age with the age of the overlying lithosphere and the timing of subduction initiation itself. In the western Philippines, we find that oceanic crust was less than ∼1 My old when it was underthrust and metamorphosed at the onset of subduction in Palawan, Philippines, implying forced subduction initiation at a spreading center. This result shows that young and positively buoyant, but weak, lithosphere was the preferred site for subduction nucleation despite the proximity of other potential weak zones with older, denser lithosphere and that plate motion rapidly changed from divergence to convergence. subduction initiation | tectonics | geochronology | Philippines | ophiolite