Sentience, Suffering, and the Case for Animal Welfare
Welfarist reasoning does not specify in advance whose welfare counts. Extending it beyond human beings requires an argument for where the moral boundary should be drawn, and the animal welfare field has largely converged on a single criterion to answer that question: sentience, the capacity to have subjective experiences, particularly the capacity to suffer.
Bentham's question
The philosophical starting point is usually traced to Jeremy Bentham, the founder of utilitarianism, who argued in the late eighteenth century that the relevant question for moral status was not "can they reason?" or "can they talk?" but "can they suffer?" This reframing matters because it locates moral relevance in a capacity many non-human animals plainly share with humans, rather than in capacities, like abstract reasoning or language, that are more distinctively human.
Singer's argument against speciesism
Philosopher Peter Singer extended this reasoning into a systematic argument in Animal Liberation (1975), one of the founding texts of the modern animal welfare and animal rights movements. Singer argued that giving less moral weight to a being's suffering purely because of its species membership is a form of unjustified discrimination, which he termed speciesism, structurally similar to racism or sexism in that it uses group membership rather than morally relevant capacities to determine whose interests count.
Singer's argument applies a principle of equal consideration of interests: like interests should be given like weight, regardless of whose interests they are. This does not imply that all animals should be treated identically to humans in all respects — a being's capacities still determine what its interests actually are, and a creature without the capacity for long-term planning does not have an interest in avoiding death in the same way a human with life plans does. What it does imply is that suffering itself, when it occurs, should not be discounted simply because it occurs in a non-human animal.
Welfare versus rights approaches
Two broad strategies for animal advocacy grew out of this reasoning. The welfare approach, which gives this site its terminology, argues for improving the conditions under which animals are kept and used — reducing suffering and increasing positive welfare, without necessarily opposing all use of animals in principle. The rights approach, associated with philosopher Tom Regan, argues that at least some animals have inherent rights that make certain uses impermissible regardless of how humanely they are conducted, treating welfare improvements as an inadequate response to what is, on this view, a rights violation.
The distinction matters practically: welfare-oriented reform focuses on standards, regulations, and measurable conditions, while rights-oriented advocacy focuses on ending certain practices outright. Much real-world animal advocacy work draws on both traditions without fully resolving the underlying philosophical tension between them.
Measuring welfare: the Five Freedoms
Because sentience itself cannot be directly observed, animal welfare science has developed indirect measurement frameworks. The most widely used is the Five Freedoms, first articulated by the UK's Farm Animal Welfare Council in 1979 and still a standard reference point: freedom from hunger and thirst; freedom from discomfort; freedom from pain, injury, and disease; freedom to express normal behavior; and freedom from fear and distress.
Later frameworks, such as the Five Domains model, restructured this into four physical/functional domains (nutrition, environment, health, behavior) feeding into a fifth mental domain representing the animal's overall subjective state — reflecting a shift in welfare science toward treating positive mental states, not just the absence of negative ones, as a legitimate object of measurement.
Empirical indicators
In practice, welfare scientists assess these domains through a combination of behavioral and physiological indicators: stress hormone levels, immune function, injury rates, and abnormal repetitive behaviors on the negative side; play behavior, exploration, and physiological markers associated with positive affect on the other. No single indicator is considered sufficient on its own, since behavioral and physiological measures can diverge — an animal may show few overt behavioral signs of distress while still exhibiting physiological markers of chronic stress, or vice versa.
Open questions
Substantial empirical uncertainty remains about the degree and character of sentience across different species, particularly outside vertebrates — questions about the welfare relevance of fish, cephalopods, and insects have become active areas of both scientific research and policy debate in recent years, with several jurisdictions extending legal welfare protections to some invertebrate species on precautionary grounds even where the underlying science is not fully settled.
This uncertainty mirrors, in an interesting way, the uncertainty explored in our article on AI and machine welfare: both fields face the difficulty of inferring subjective experience from external behavior and physiology, without direct access to whether or what a being actually experiences.