Numerous studies have examined the electrochemical reduction of CO (COR) to oxygenates (e.g., ethanol). None have considered the possibility that oxygen in the product might arise from water rather than from CO. To test this assumption, (CO)-O-16 reduction was performed in (H2O)-O-18 electrolyte. Surprisingly, we found that 60-70% of the ethanol contained O-18, which must have originated from the solvent. We extended our previous all-solvent density functional theory metady-namics calculations to consider the possibility of incorporating water, and indeed, we found a new mechanism involving a Grotthuss chain of six water molecules in a concerted reaction with the *C-CH intermediate to form *CH-CH((OH)-O-18), subsequently leading to (O-18)ethanol. This competes with the formation of ethylene that also arises from *C-CH. These unforeseen results suggest that all previous studies of COR under aqueous conditions must be reexamined.