Alternatively, water can change the physical and chemical properties of rock to give it properties more convenient for plate tectonics; water would get into sedimentary rocks as a result of their formation, and ocean-floor sedimentary rocks may get subducted, dragging water-containing rocks into the Earth's interior.

I recall seeing something like this as a hypothesis for the origin of granite, but I don't recall the details.

However, this water hypothesis would not explain the original plate tectonics very well, because it would have had to start without water-containing sedimentary rocks.