Ghost variables are variables in program languages that do not correspond to physical entities. This special issue, based on a panel on “Race as a Ghost Variable” at the 2017 Meeting of the Society for Social Studies of Science, traces ideas of “race” in particular niches of science, technology, and medicine where it is submerged and disavowed, yet wields power. Each paper is a case study exploring ghosts that emerge through the resonance among things as heterogeneous as hair patterns, hormone (...) levels, food tastes, drug use, clinic locations, proximity to disaster, job classifications, and social belonging and suspicion, all of which vibrate with meanings accumulated over long racial histories. Together, the papers further elaborate methods and analytic models for identifying the operations of race—the relations and processes that make it, the effects that it has. A chief appeal of the metaphor of the ghost is that it brings the importance of history to the fore. Ghosts are simultaneously history and the present, not just an accretion of earlier experiences, but the palimpsest left when one tries to erase them. Sometimes faint and hard to discern, sometimes rambunctious and disruptive, ghosts refuse our attempts to simply move on. (shrink)