The asymptotic safety scenario for quantum gravity hinges on the existence of a nontrivial fixed point for the dimensionless Newton constant g N . We propose an affirmative answer to the question raised in the title provided a suitably improved perturbative framework is used. For any one loop renormalizable gravity theory one obtains flow equations which are uniquely determined by the coefficients of the powerlike and logarithmic divergences in the background covariant effective action. These flow equations exhibit nontrivial fixed points for g N and the dimensionless cosmological constant λ with respect to which the flow is asymptotically safe. The gauge and scheme dependence can be discussed analytically. Results for higher derivative gravity in minimal gauge are presented. Remarkably, spectral positivity of the Hessians can be satisfied along the entire flow, evading the traditional positivity problem. Dependence on O(10) initial data is erased to accuracy 10 −7 after O(10) units of the renormalization mass scale and the flow settles on a λ (g N ) orbit. September 17-19, 2008 Brighton, University of Sussex, United Kingdom c Copyright owned by the author(s) under the terms of the Creative Commons Attribution-NonCommercial-ShareAlike Licence.
Workshop on Continuum and Lattice Approaches to Quantum Gravityhttp://pos.sissa.it/
PoS(CLAQG08)005Fixed points from perturbation theory
Nontrivial perturbative fixed pointsPerturbation theory is as routinely applied by some as it is frowned upon by others, both not always thoughtfully. Before turning to gravity proper we offer some general remarks on what one can and cannot reasonably expect from perturbation theory (PT). The identification of a nontrivial fixed point in PT at first sight seems to be a contradiction in terms, alas it is not. By definition PT is a saddle point expansion in the loop counting parameterh introduced into the functional integral as the inverse prefactor of the bare action,h −1 S 0 . Forh → 0 one obtains, depending on the signature, a generalized Laplace or stationary phase expansion which may or may not coincide with the expansion in some 'small' bare reference coupling. In the presence of massless degrees of freedom the orders in the loop counting parameter also do not coincide with the orders in Planck's constant. The 'bare'h expansion can typically be shown to provide an asymptotic expansion of the properly regularized functional integral. After renormalization such proofs are exceedingly difficult and are available only in superrenormalizable or purely fermionic field theories. Off hand therefore the renormalized PT expansion is only a formal power series in the loop counting parameterh. An important advantage of PT however is that (in a perturbatively renormalizable field theory) the ultraviolet (UV) cutoff can termwise strictly be removed, independent of the nature of the coupling flow. The perturbatively defined coupling flow itself then provides a plausibility criterion [2] for the existence of an underlying 'exact' theory such that f...