Fishing through, fishing up (i.e. the observation that mean trophic levels (MTLs) of catches are increasing with time) and the question of agreement between MTL from catches and surveys have been used to question fishing down the marine food web as well as the use of the marine trophic index as an effective ecosystem indicator. This has evolved into a crucial debate in marine ecology and fisheries management. We believe that fishing down, fishing through and fish- ing up the food web are not incompatible but only appear so because of confusion. Fishing down is an ecological process, which can be expected to either occur or not, i.e. it is a hypothesis which can be falsified and replaced by an alternative hypothesis that must also rely on ecological princi- ples (rather than technicalities). In contrast, fishing through and fishing up reflect technical issues related to biases in the available data and/or to fishing behaviour and thus do not falsify fishing down. The latter could be falsified if, for instance, in an ecosystem subjected to intense fishing, the biomasses of all or several large predators increase and these increases go along with an increase in their mean lengths for an extended period of time. Thus, MTL remains one of the most opera- tional indices available for testing fishing down. It is without doubt that, in various cases, MTL will not be effective because of various technical confounding factors, which can be clarified by local experts. Undoubtedly, MTL can be misused, usually when confounding effects cannot be properly disentangled, and thus it must be used and interpreted with caution.