Yes, this is the current consensus of the scientific community. During subduction of plates (mostly oceanic crust full of water under continental plates of lowe

Yes, this is the current consensus of the scientific community. During subduction of plates (mostly oceanic crust full of water under continental plates of lower density) the volatiles (mostly water) exits the subducting plate and lowers the viscosity and density of the overlying mantle material (viscosity describing the 'fluidness' of the material, higher viscosity = less fluid). The material will then rise up and there will be volcanism, recycling the material back to the surface. This is what you see in most major volcanic arcs, such as the Andes or Japan. The idea that plate tectonics and water are intrinsically linked (at given circumstances) is pretty solidified. The theory of why not such much; the idea above fits our current data best, but data from the planet's interiour is hard to come by and I expect some updates to the theory within a few decades. It's nice that this is currently not critical knowledge for humanity.