Previous geochemical and biomarker studies of the late Paleo-Mesoproterozoic propose a stratified world, with strongly reducing (possibly sulfidic) deep-ocean conditions overlain by an oxygenated surface-ocean and atmosphere. To investigate such a scenario, we look to the structure of the biogeochemical sulfur cycle. We present sulfur (32S, 33S, 34S, and 36S in sulfides) isotope data from the McArthur Basin (Barney Creek, Reward, Velkerri, and McMinn formations) that allows for a direct evaluation of the surface biosphere. We are interested in investigating the types of information that can be gained by including 33S and 36S. When the 34S/32S fractionations are small, the inclusion of 33S and 36S provides little additional information, but does provide ancillary evidence for relative isotopic homogeneity (with the internal consistency of 33S/32S and 36S/32S). When the 34S/32S fractionations are large, direct information about the fractionation mechanisms can be obtained, with the potential to distinguish the biological from abiological processes. For example, the reconstruction of the Roper Group suggests that seawater sulfate concentrations were high enough to buffer against spatial heterogeneities. Overall, our findings agree with previously proposed redox structure of the Proterozoic ocean, highlight contributions from the oxidative sulfur cycle, and outline a new tool for interpreting the state of the surface sulfur cycle.