- in quantum gravity i.e. string theory, one should have no global continuous symmetries; it is known in string theory that a symmetry current, a (1,0) or (0,1) tensor, is associated with any symmetry, and it can be multiplied by "del X" or "delBAR X" to create a (1,1) marginal vertex operator for a gauge boson, proving that the symmetry is local after all
- this constraint looks too fragile because one can immitate global symmetries by weak symmetries with a very small coupling constant which seem allowed
- therefore, it would be more natural if very weak couplings were forbidden, too
- more precisely, if a U(1) gauge group has too small a coupling, ...
This posting was never finished. Read the preprint hep-th/0601001.