Reversible-deactivation radical polymerization (RDRP) in the presence of Cu-0 is a versatile technique that can be used to create well-controlled polymers with complex architectures. Despite the facile nature of the technique, there has been a vigorous debate in the literature as to the mechanism of the reaction. One proposed mechanism, named supplemental activator and reducing agent atom transfer radical polymerization (SARA ATRP), has Cu-I as the major activator of alkyl halides, Cu-0 acting as a supplemental activator, an inner-sphere electron transfer occurring during the activation step, and relatively slow comproportionation and disproportionation. In SARA ATRP slow activation of alkyl halides by Cu-0 and comproportionation of Cu-II with Cu-0 compensates for the small number of radicals lost to termination reactions. Alternatively, a mechanism named single electron transfer living radical polymerization (SET-LRP) assumes that the Cu-I species do not activate alkyl halides, but undergo instantaneous disproportionation, and that the relatively rapid polymerization is due to a fast reaction between alkyl halides and "nascent" Cu-0 through an outer-sphere electron transfer. In this article a critical assessment of the experimental data are presented on the polymerization of methyl acrylate in DMSO with Me6TREN as the ligand in the presence of Cu-0, in order to discriminate between these two mechanisms. The experimental data agree with the SARA ATRP mechanism, since the activation of alkyl halides by Cu-I species is significantly faster than Cu-0, the activation step involves inner-sphere electron transfer rather than an outer-sphere electron transfer, and in DMSO comproportionation is slow but occurs faster than disproportionation, and activation by Cu-I species is much faster than disproportionation. The rate of deactivation by Cu-II is essentially the same as the rate of activation by Cu-I, and the system is under ATRP equilibrium. The role of Cu-0 in this system is to slowly and continuously supply Cu-I activating species and radicals, by supplemental activation and comproportionation, to compensate for Cu-I lost due to the unavoidable radical termination reactions. With the mechanistic understanding gained by analyzing the experimental data in the literature, the reaction conditions in SARA ATRP can be tailored toward efficient synthesis of a new generation of complex architectures and functional materials.