In the 25 years or so after the first clinicopathological descriptions of diffuse axonal injury (DAI), the criterion for diagnosing recent traumatic white matter damage was the identification of swollen axons ('bulbs') on routine or silver stains, in the appropriate clinical setting. In the last decade, however, experimental work has given us greater understanding of the cellular events initiated by trauma to axons, and this in turn has led to the adoption of immunocytochemical methods to detect markers of axonal damage in both routine and experimental work. These methods have shown that traumatic axonal injury (TAI) is much more common than previously realized, and that what was originally described as DAI occupies only the most severe end of a spectrum of diffuse trauma-induced brain injury. They have also revealed a whole field of previously unrecognized white matter pathology, in which axons are diffusely damaged by processes other than head injury; this in turn has led to some terminological confusion in the literature. Neuropathologists are often asked to assess head injuries in a forensic setting: the diagnostic challenge is to sort out whether the axonal damage detected in a brain is indeed traumatic, and if so, to decide what - if anything - can be inferred from it. The lack of correlation between well-documented histories and neuropathological findings means that in the interpretation of assault cases at least, a diagnosis of 'TAI' or 'DAI' is likely to be of limited use for medicolegal purposes.