PHYSICAL REVIEW D VOLUME 20, NUMBER 10 15 NOVEMBER 1979 Soluble gauge theory of a noncompact group Kevin Cabill* Physes Deparment Indiana Universi, Bloomington, Indiana 47405 (Received 6 April 1979) It's shown thatthe gnuge theory ofthe multiplicative group of complex numbers i the theory of two fee ‘vector mesoas only one of whichis mass. It was recently shown to be possible to con struct a gauge theory of a noncompact internal symmetry group." In this paper, one ofthe sim= plest ofthese quantum field theories, the pure gauge theory ofthe multiplicative group of com- plex numbers, is solved exactly. The Lagrange density of the gauge theory of the general linear group of all complex, non- singular n xn matrices, GLin,C), is L=—(e)* tr(FLgPY sg) HON? e.g D+ 01,20" - Voge, o where the curvature tensor Fy, {8 Fa Aveo Burd) ® and the covariant derivatives of the matter field {and of the internal metrie tensor g are Aue ° co and Bin =Bin tA, tale. a) ‘The potential V0 represents a self-interaction of g, the subscript, u means 9/0x°, and e and f are coupling constants, ‘Under the gauge transformation associated with the matrix ale) in GL(n,C), the n complex sealar fields ¢,(s) transtorm 25 e[e0)=a4 (09,60). ° ‘The Lagrange density L is invariant under this transformation because the internal metrie tensor ‘£() and the connection A, (x) transform as 8G) = aN gears) o and as ALG)=al0)A, a0) + ale) a°%e)« o ‘The problems* usually associated with gauge theories of noncompact groups, negative probabil- ity and negative energy, are avoided here through the use of the internal metric tensor g. The tensor 4g is Hermitian and non-negative and may be re- Garded as the product of a more fundamental Hermitian matrix h with its adjoint, g="2, Al- though the matrix g(r) transforms (6) as a tensor in the internal space, it is composed of fields 4,60) that are scalars under Lorentz transforma- tions. By linearizing the Lagrange density L, one may show that the n? complex vector mesons 4, Sep- arate into nt real massless veetor mesons C,, associated with the group U(n) 1 the noncompact and non-Abelian as~ pects of the GLin, C) gauge theory become inter~ twvined and no exact solution is available. One may Impose theghost-free gauge condition, =, and fix the remaining compact gauge freedom, but it is not known whether the resulting perturbation theory would be renormalizable. However, if M, gauge theories of noncompact, non-Abelian groups are renormalizable or otherwise implementable, then they might be used to form unified theories of the strong and electro-weak interactions, for they possess an intrinsic geometrical mechanism that gives a mass to some of the vector mesons (namely those associated with the generators of the noncompact part of the group). Existing uni~ fied gauge theories use the Higgs mechanism, which is somewhat arbitrary and unmotivated, {or that purpose. ACKNOWLEDGMENTS Tam grateful to Professor N. Fuchs, Pro- fessor D. Lichtenberg, and Professor D. R. ‘Stump for useful discussions. This work was supported in part by the U. S. Department of Energy under contracts to Indiana University and to Harvard University (EY76-C-02-3064 000). ‘Present address: Lyman Laboratory of Phyaies, Har vard University, Cambridge, Massachusetts 02198, 4K Cahill, Phys, Rev. D 18, 2930 (1978), SL, Glashow and M, Gell-Mann, Ann. Piya. (N.Y 497 960), ‘ar quantzation ts carried out inthe temporal guise, Ay "0, then, and commute andthe brackats rele only tothe opertor otdeving of g and which canbe a= ange 0 that (follows