Food web is a community network of trophic interaction. Undoubtedly, the food-web approach has been a powerful tool to understand ecological systems. However, there is a drawback too – it is not easy to identify completely the interspecific trophic links of real community, given many efforts to be paid and the dependence of estimated food-web structure on the amount of effort. Stable isotope analysis allows an accurate estimation of trophic position of species with relatively a few efforts, although it does not tell who eats whom. Higashi et al. (1989) presented a theoretical technique, called ‘food-web unfolding’, where the material stocks and flows within a food web are re-arranged according to “the number of mouths” that the focal material was passed through to allow transforming a complex food web into a simple food chain. In this talk I point out the relevance of food-web unfolding to trophic position and present theoretical idea of how food-web unfolding can be utilized to understand food web. Using food-web unfolding, food-web complexity can be degraded into three diversity components, the diversity of trophic range, diversity of trophic levels and species diversity within a trophic level, each of which has different contributions to ecosystem functioning.