The relationship between health and welfare

“Sickness and injury risks for outdoor cats are health, not welfare issues” (Legge et al. Reference Legge, Woinarski, Dickman, Nou and Dielenberg2026). This is a false summary of our argument. Physical health is absolutely one key determinant of animal welfare. Our argument, and that of the broader animal welfare science discipline for several decades, is that physical health is not the only factor contributing to an animal’s welfare state (Fraser & Duncan Reference Fraser and Duncan1998). Animal welfare by one influential definition is determined by the animal’s affective experience: how the animal feels (Hemsworth et al. Reference Hemsworth, Mellor, Cronin and Tilbrook2015). This subjective state is a product of many inputs that have been depicted in various contemporary models of animal welfare, e.g. the Five Domains model (Mellor et al. Reference Mellor, Beausoleil, Littlewood, McLean, McGreevy, Jones and Wilkins2020). Traditionally, greater importance has been placed on physical health than other aspects of animal welfare (e.g. social interactions, positive experiences, innate behaviours) and the same is true in arguments regarding cat confinement (Abbate Reference Abbate2020). This is why our paper focuses heavily on these other aspects of cat welfare as those most impacted by confinement (noting that cats’ physical health is also impacted by confinement, simply in other ways than outdoor living).

Another more recent paper, published after our original paper, and written by a panel of international animal welfare scientists, including one of the authors of the current paper, focuses on positive animal welfare and defines this as “the animal flourishing through the experience of predominantly positive mental states and the development of competence and resilience” (Rault et al. Reference Rault, Bateson, Boissy, Forkman, Grinde, Gygax, Harfeld, Hintze, Keeling and Kostal2025). In line with our original argument it adds, that “positive animal welfare goes beyond ensuring good physical health and the prevention and alleviation of suffering”. Furthermore, it emphasises the importance of “having choices and opportunities to actively pursue goals and achieve desired outcomes, according to species-specific and individual capabilities” (Rault et al. Reference Rault, Bateson, Boissy, Forkman, Grinde, Gygax, Harfeld, Hintze, Keeling and Kostal2025). It will be hard to argue that, on this definition of positive animal welfare, that nothing is lost in terms of welfare by preventing cats from roaming. Hence, our primary point: cat confinement involves welfare trade-offs.

The response of Legge et al. (Reference Legge, Woinarski, Dickman, Nou and Dielenberg2026) also continues to argue throughout that containing cats improves their welfare. In this way, it again assumes that physical health and safety are either synonymous with welfare or outweigh other aspects of welfare. It also claims that there is “abundant evidence” that any welfare risks can be minimised or completely overcome (and that we said this ourselves – we did not). While there have been empirical studies published regarding cat welfare and enrichment based on owner surveys (Hoff et al. Reference Hoff, Lundegaard, Møller-Petersen, Nielsen and Sandøe2026), they too demonstrate deficiencies in contained cat management and welfare.

Robust evidence in animal welfare science requires simultaneous examination of multiple animal-based measures; most commonly a combination of behavioural, physical, and physiological indicators (Hemsworth et al. Reference Hemsworth, Mellor, Cronin and Tilbrook2015). To our knowledge, there have been no studies to have done this for contained cats in general, none comparing contained to roaming cats, none testing the efficacy of enrichment in ameliorating welfare risks of confinement, and importantly, none examining the welfare state of cats that were previously allowed to roam and then contained. Ergo, there is a distinct need for integrated animal welfare science in this area, which utilises complementary animal-based measures, and is specifically designed to assess the effects of confinement.

Indoor life

“An indoor life is bad for the cat” (Legge et al. Reference Legge, Woinarski, Dickman, Nou and Dielenberg2026). We did not state that indoor life is categorically bad for cats, but that there is a trade-off to be made between different aspects of welfare. We highlighted the significant risks to cat welfare of a contained lifestyle as these are often underplayed. Also, there are possible negative physical effects on the welfare of containing cats, notably a higher level of obesity (Bjørnvad et al. Reference Bjørnvad, Mortensen, Tams, Jørgensen, Sandøe and Lund2026), which in turn may lead to other health problems. While some of these impacts may be managed to varying, somewhat unknown degrees by enrichment or other owner-driven interventions, some factors are either unlikely to be able to be replicated in an indoor environment or would take significant time and commitment by cat owners. The latter raises issues (as discussed in our original paper) of owner motivation, time, and competence to maintain a minimum welfare standard. Therefore, it is important that owners are appropriately educated about these risks, not falsely assured their cats will simply have better welfare because they are contained and safer. While Legge et al. (Reference Legge, Woinarski, Dickman, Nou and Dielenberg2026) rightly note that some confinement campaigns have promoted cat enrichment, many do not (as referenced in our original paper), and these are what concern us.

Misunderstanding and misattribution of the concept of rights

“Cats have rights, but other animals do not” (Legge et al. Reference Legge, Woinarski, Dickman, Nou and Dielenberg2026). We do not discuss the ‘rights’ of any animals in our paper. Indeed, none of us have in our writings tended to agree with this school of ethical thinking (Hampton et al. Reference Hampton, Warburton and Sandøe2019). This interpretation suggests a misunderstanding of the relationship and difference between animal welfare science and ethical theories. We do, however, consider that we adopt responsibilities to animals in different ways. Just as our responsibilities to cats do not negate our responsibility to act on the extinction crisis, the extinction crisis does not negate our responsibilities to cats. We most certainly do not consider that “the cat’s right to roam should somehow trump everything else” (Legge et al. Reference Legge, Woinarski, Dickman, Nou and Dielenberg2026), but that the risks to contained cat welfare need to be appropriately understood and recognised in order to make the ethical weighing of interests the authors promote.

Our paper focuses heavily on the welfare risks of cat confinement (while also noting the benefits to safety and some physical health aspects) as these are what have been largely underrepresented in such arguments previously. Unlike what our critics claim, we nowhere suggest and would not argue that allowing cats to roam carries more weight than preventing the negative effects caused on wildlife by the cats – quite to the contrary. We simply contend that confining cats involves trade-offs regarding cat welfare.

Shooting the messenger

It is stated that we “make a series of statements that are excessive and seem designed to offend other scientists working in this space…” (Legge et al. Reference Legge, Woinarski, Dickman, Nou and Dielenberg2026). This was not our intention, and we apologise for any personal offence caused. Our critique of win-win messaging is simply that, a critique of an approach that we consider to be risky for both public trust and the effectiveness of confinement campaigns. Our comments regarding messengers being “misleading”, “disingenuous”, or “deceptive” either related to the perception by the public (i.e. that they are perceived as being these things, not that they actually are) or were accompanied by the alternate explanation that those promoting the “win-win” message may not fully understand animal welfare science. By stating that scientists from one discipline may not understand the complexities of another discipline we meant no personal attack; simply a statement of inevitable fact. Conservationists may not (and should not be expected to) have an expert understanding of contemporary animal welfare science. When real-world problems such as these cross scientific disciplinary boundaries, it is practically necessary to draw upon the expertise of those different disciplines (Sekar & Shiller Reference Sekar and Shiller2020).

Lastly, Legge et al. (Reference Legge, Woinarski, Dickman, Nou and Dielenberg2026) appear to imply that any welfare risks associated with cat confinement are trivial, that “one may as well debate the choice of font on book covers whilst standing in the middle of a burning library”. However, contemporary Australians do not consider animal welfare to be a trivial issue (Sinclair et al. Reference Sinclair, Lee, Hötzel, de Luna, Sharma, Idris, Derkley, Li, Islam and Iyasere2022). It is possible to care about more than one value simultaneously, and herein lies the merit in ethical analysis (Lynch et al. Reference Lynch, Allen, Berger-Tal, Fidler, Garrard, Hampton, Lean, Parris, Sherwen, White, Wong and Blumstein2025). While we appreciate the urgency and scale of the extinction crisis in Australia (Legge et al. Reference Legge, Rumpff, Garnett and Woinarski2023), ploughing ahead with a potentially flawed approach with a determination to “get on with the task” (Legge et al. Reference Legge, Woinarski, Dickman, Nou and Dielenberg2026) is likely to backfire.

Our intention has always been to stimulate debate, not offend, and highlight the need for further cat welfare and behaviour change research. With shared goals in many respects, we consider the most productive way forward to be working together to broaden our understanding of this complex issue, refine approaches, and ultimately improve the lives of both cats and wildlife.