Overall animal welfare assessment reviewed. Is it possible?

Several authors have concluded that scientists should not attempt to perform overall animal welfare assessment (OWA). They argue that scientists have continued to fail to make progress in this area and that value judgements are inherently involved in OWA for which science cannot provide answers. We take a more positive attitude toward OWA and argue that scientists should avoid creating a self-fulfilling prophecy. OWA is necessary for making actual moral and political decisions. Science has already accumulated much relevant information about welfare and this information should be applied in decision making. The task of OWA is to assess welfare based on knowledge of the biological needs of animals. Weighting of welfare relevant factors constitutes a problem. However, when scientists cannot provide empirical data to solve weighting issues, this does not mean that rational answers cannot be found, e.g. in the form of procedural rules. OWA is conceived as a problem of multi-criteria decision making with fuzzy information. If focuses on the descriptive aspect of welfare, i.e. on what the welfare status of the animals really is without taking an ethical stance. The welfare status of animals depends on their biology and on the way animals assess their own welfare. It does not depend on how it happens to be perceived by us. Even though OWA necessarily remains a human activity, it is not arbitrary, nor does it allow of multiple 'correct' answers. OWA is a descriptive activity that can achieve more and more accuracy as science proceeds