Amidst ongoing declines in honey bee health, the contributory role of the newer systemic insecticides continues to be intensely debated. Scores of toxicological field experiments, which bee scientists and regulators in the United States have looked to for definitive causal evidence, indicate a lack of support. This paper analyzes the methodological norms that shape the design and interpretation of field toxicological studies. I argue that contemporary field studies of honey bees and pesticides are underpinned by a "control-oriented" approach, which precludes a serious investigation of the indirect and multifactorial ways in which pesticides could drive declines in honey bee health. I trace the historical rise to prominence of this approach in honey bee toxicology to the development of entomology as a science of insecticide development in the United States. Drawing on "complexity-oriented" knowledge practices in ecology, epidemiology, beekeeping and sociology, I suggest an alternative socio-ecological systems approach, which would entail in situ studies that are less concerned with isolating individual factors and more attentive to the interactive and place-based mix of factors affecting honey bee health.
Keywords: bee; colony collapse disorder; field study; insecticide.