Because of severe resource and logistical constraints in large areas of Africa, disease surveillance systems need to maximize the use of information provided by livestock keepers and make correct interpretations of indigenous livestock knowledge. This paper describes the use of participatory epidemiology (PE) to compare the names, clinical signs and epidemiological features of cattle diseases as perceived by pastoralists and veterinarians. Using results from two previous studies with pastoralists in southern Sudan and Kenya, provisional translations of local disease names into modem veterinary terminology were used to develop a matrix scoring method for use with veterinarians. Matrix scoring data from pastoralists and veterinarians were then compared using simple visual comparison of summarized matrices, hierarchical cluster analysis and multidimensional scaling. The results showed good agreement between pastoralists' and veterinarians' disease names and diagnostic criteria. The matrix scoring method was easy to use and appropriate for use in under-resourced areas with minimal professional support or laboratory services. Matrix scoring could be used to assist livestock disease surveillance workers to design surveillance systems that make better use of pastoralist's indigenous knowledge and correctly interpret local disease names. The method should be combined with conventional veterinary investigation methods where feasible.