The question of ethics in animals is difficult. On the one hand, to argue that an animal has a sense of ethics means that the animal must have concepts of self, versus others, and must have the cognitive ability to recognize right and wrong. This seems unlikely for most animals, and extraordinarily difficult to test as a hypothesis.

Alternatively, you could make the case that the presence of behavior that we interpret as "ethically based", such as appeasement and punishment, indicates that a sense of ethics exists in animal societies. If trangressions from social norms are punished, the some concept of "wrongness", whether cognitive or not, must exist. If aggression against others is later counterbalanced by conciliation, then we have evidence of a behavioral balance that keeps the social group from unravelling following conflict among its members.

Suppose we accept that the boundary between science and philosophy lies at the point of the testable hypothesis. The most effective scientific arguments for ethics in animal societies still rely on defining behaviors as ethical based on their similarity to human behavior. Clever experiments may give some insight into the motivations that underly the behaviors, and by testing the right hypotheses a case for ethical behavior can be built. However, we still must always question the assumption that animal and human behaviors which appear to be similar have the same purposes.

For more on this, consider how conflict resolution functions in primate social groups.