Comments on the spontaneous strain and polarization of polycrystalline ferroelectric ceramics