Abstract
Despite the clinical and basic scientific importance of neonatal seizures, a standard operational definition of them is lacking. Virtually all quantitative studies of phenomenology, prevalence, treatment efficacy, and prognosis have been undermined by inadequate or inconsistent criteria for the object of investigation. This is not so much a fault of the studies, however, as a consequence of the very nature of neonatal seizures themselves, which seem to defy universally applicable definitions and criteria for quantification. The main obstacles to formulating such definitions are reviewed, and directions for further research are suggested.