My dissertation research is primarily focused on developing an account of non-human animal welfare and then showing how such an account has ramifications for questions in applied ethics. The study of animal welfare has been a somewhat popular topic over the past couple decades within animal sciences. Animal scientists have been able to look at certain indicators, such as cortisol levels, to determine what things or events calm or stress animals. They have then been able to take this information and use it to create and implement more humane conditions for animals in labs and stockyards, such as the EU ban on battery cages to hold chickens.

As useful as such studies have been, they have generally excluded the more theoretical question of what constitutes welfare for a particular non-human animal. When we ask what it means for an animal to flourish, we generally want to know more than simply what conditions typically lead to flourishing for that animal, or what physical indicators are present when the animal is flourishing. We want to know what the nature of welfare is for the animal; what, at the theoretical ground level, makes animals' lives go better or worse for them? My dissertation is an attempt to answer this question.

There have been other attempts to answer this question, but my approach is novel because I reverse the typical methodological process, which is to determine the best theory of welfare, and then apply that theory to all welfare subjects (including animals). In contrast, I think that the appropriate way to determine animal welfare is to first look at the creatures in question and see which theory of welfare fits those creatures best, given the creature's cognitive and emotional capacities. I do this by first looking at the three major theories of welfare (mental state, preference-satisfaction, and objective list theories) and examining the typical objections to such theories, which are almost always objections in relation to humans as the main welfare subjects. I then consider whether the objections are as problematic when considering non-human creatures. I argue in many cases the typical objections are not as problematic for certain sets of creatures who possess certain cognitive and emotional capacities, while remaining problematic for other sets of creatures with different cognitive and emotional capacities. I ultimately conclude from this that no single common theory of welfare is a good fit for all welfare creatures. I argue for a hybrid theory of welfare: one should adopt a hedonistically-oriented theory of welfare when dealing with creatures with less sophisticated cognitive capacities, and as creatures become more and more cognitively sophisticated, one should give more emphasis to elements found in more preference satisfaction-oriented theory of welfare. Roughly, when dealing with creatures with very low levels or degrees of autonomy and agency, we should measure their welfare using a hedonic theory. As a creature increases in their degree of agency and autonomy, the satisfaction of preferences becomes more important in increasing the welfare of that creature.

This result has important implications for applied ethical questions. For example, theorists like Peter Singer have argued that non-human animals are not harmed as much by death as humans, since many non-human animals lack future-oriented preferences, and hence such desires cannot be frustrated. Since future-oriented preferences cannot be frustrated for nonhuman animals, but can be for persons, persons are deprived of much more when they die. But if I am correct, even if some non-human animal does fail to have future-oriented desires, this tells us very little about how bad their death is compared to the death of a person. The relevant comparison is not between the number of preferences a person has frustrated by death and the number of preferences a non-human animal has frustrated by death, but rather it should be between the number of preferences a person has frustrated by death and the amount of pleasure deprived to a non-human animal by death. Singer thus moves too quickly to claim that the death of a person is typically worse for the person than the death of a non- human animal is for that animal, and since this view informs his arguments for when it is permissible to kill nonhuman animals, such arguments become suspect.