The results of several epidemiological studies of serum gamma-glutamyltransferase (GGT) led us to hypothesise that associations of GGT within its normal range with type 2 diabetes may reflect detrimental effects of xenobiotics found in the environment, such as persistent organic pollutants (POPs). Epidemiological observations showed that serum GGT activity within its normal range strongly predicted future type 2 diabetes; the predictability of diabetes from obesity was low with GGT at the low end of the normal range; and GGT showed a positive association with known markers of oxidative stress or inflammation. Experimental findings on cellular GGT suggest that serum GGT levels within the normal range may reflect oxidative stress related to the re-synthesis of intracellular glutathione; however, this interpretation is not completely satisfying because, in its role of regenerating intracellular glutathione, GGT activity should be antioxidative. Alternatively, serum GGT activity may reflect amounts of glutathione conjugates formed during the metabolism of xenobiotics. Accordingly, we postulate a two-part hypothesis: that the association of serum GGT with type 2 diabetes reflects exposure to POPs, as these substances, which have a very long half-life, may influence diabetes risk by residing in adipose tissue as endocrine disruptors; and that POPs or similar substances may interact with obesity to cause type 2 diabetes. Supporting this hypothesis, cross-sectional investigation of background exposure to POPs in the National Health and Nutrition Examination Survey showed relationships similar to those observed for GGT, including a powerful association with prevalent diabetes and no association between obesity and diabetes for very low POP concentrations. Our hypothesis can be tested in both prospective studies and toxicological studies.