Terahertz spectroscopy has been investigated as a technique for concealed explosives detection since spectroscopic signatures in common explosives were first identified almost 10 years ago. This paper explores the progress towards practical development of the technology and the physical basis of the challenges involved in creating a deployable person-borne IED detection system. We conclude that, despite the theoretical potential of terahertz to provide a safe, through-barrier spectroscopic detection technique, this is unlikely to be possible in practice. The relatively weak and broad explosives features tend to be masked by the combined effects of atmospheric water vapor absorption, barrier attenuation and scattering from both clothing and the target explosives. Imaging at the lower end of the terahertz frequency range, although not offering the same specificity, is a more promising technique for security applications.