Two recent developments have had a major influence on science as a whole. First, the ‘reproducibility crisis’ (https://www.nature.com/articles/533452a) refers to the realisation that many modern scientific studies are almost impossible to accurately reproduce. Second, emerging evidence suggests that science has become less innovative or ‘disruptive’ over time (https://www.nature.com/articles/s41586-022-05543-x). Given that scientific progress is built on the ideas of reproducibility, challenge, debate and argumentation, these are concerning developments.

I am an animal welfare scientist, but I also work in several other fields of science. Animal welfare science is a fledgling scientific discipline devoted to the experiences of animals. Some established paradigms in animal welfare may not be scientifically robust, but nonetheless are rarely challenged. One such approach is the use of the ‘Five Domains model’, a conceptual model designed to categorise the subjective experiences of animals, into an assessment tool. In our 2023 paper titled ‘Rethinking the utility of the Five Domains model’, my co-authors and I present a challenge to a popular but problematic approach to animal welfare assessment.

Use of the Five Domains model as a scientific animal welfare assessment tool relies on the use of expert panels to quantify and rank the welfare of animals subjected to different husbandry or management practices. Some problems are immediately apparent. It is not always clear how the experts are chosen, how they reach their conclusions as a group, what empirical science they consider, and how much uncertainty surrounds their findings. Referring to two tenets of the scientific method, how testable are the outputs of these assessments? How reproducible are their findings? My co-authors and I conclude that, at present, the answer to both questions is “not very.”

Much of the popularity of Five Domains assessments may be down to convenience. Animal welfare ‘science’ done this way can be quick, inexpensive and manipulable. But this manipulability creates the temptation for investigators working in contentious fields (and so much of animal welfare is contentious) to reach pre-determined conclusions. Opinion should not passed off as science, and there is a real risk of misuse for approaches that purport to be scientific but may be manipulated for political or public relations reasons. My colleagues and I conclude our paper by warning that “in the very worst-case scenario, the outputs of a Five Domains assessment may amount to nothing more than the opinions of the loudest or most determined member of an opaquely selected panel, expressed as numerical scores without measures of uncertainty.”

The general lack of constructive criticism in animal welfare science may be testament to the collegial culture in this discipline. I suspect that not everyone will appreciate having the flaws pointed out in an undoubtedly convenient approach, but challenge is needed for ideas to evolve and become refined. There is real risk to the perceived legitimacy of animal welfare as a scientific endeavour if we do not continue to test and update our ideas.

ABOUT THE AUTHOR

Jordan O. Hampton is a Research Fellow in the Faculty of Science at the University of Melbourne, and an Adjunct Lecturer in the Harry Butler Institute at Murdoch University. He is a veterinarian with a PhD in animal welfare and training in wildlife biology. He is one of the best-published authors in the burgeoning field of wildlife welfare and has contributed to numerous papers and books on wildlife management, welfare and ethics.