As we mentioned, Kant restricts having intrinsic value or being an end in itself to rational beings, but it is difficult to see why this should be so. Surely any sentient or conscious being has states that matter to it in a positive or negative way—pleasure matters to an animal in a positive way, pain or fear in a negative way. Since it can value what happens to it, it has intrinsic value. Given the logic of morality, we should extend our moral attention to those states that matter to it when our actions affect that being. So what if it can’t reason?—not all or even most of our moral attention focuses on reason vis a vis people. Most of it in fact focuses on feeling, on not hurting people physically or mentally, or helping them be happy or escape from suffering. So if human beings are ends in themselves, why not animals, since they too have feelings and goals that they value?
(Bernard E. Rollin, "Reasonable Partiality and Animal Ethics," Ethical Theory and Moral Practice 8 [April 2005]: 105-21, at 117)