Different answers have been suggested throughout human history, by animal lovers, scientists, philosophers, and others who interact with animals, but the core question has proven hard to answer since we can neither obtain spoken answers, nor assume anthropomorphism. As a result, on the one hand society recognizes animals can feel pain, by criminalizing animal cruelty, and yet on the other hand it is far from clear whether we truly believe animals "feel" in a meaningful sense. Often expressions of apparent pleasure are ambiguous as to whether this is emotion, or simply innate response, perhaps to approval or other hard-wired cues. The ambiguity is a source of much controversy in that there is no certainty which views, if any, are "right". That said, extreme behaviorists would also say that human "feeling" is a meaningless, hard-wired response to external stimuli.
In recent years, research has become available which expands prior understandings of animal language, cognition and tool use, and even sexuality. Emotions arise in the mammalian brain, or the limbic system, which human beings share in common with other mammals as well as many other species. This presents both a scientific dilemma — how can we tell? — and a potential ethical one — if true, what does it mean?
The cautious wording of Beth Dixon's 2001 paper on animal emotion exemplifies this viewpoint:
Recent work in the area of ethics and animals suggests that it is philosophically legitimate to ascribe emotions to nonhuman animals. Furthermore, it is sometimes argued that emotionality is a morally relevant psychological state shared by humans and nonhumans. What is missing from the philosophical literature that makes reference to emotions in nonhuman animals is an attempt to clarify and defend some particular account of the nature of emotion, and the role that emotions play in a characterization of human nature. I argue in this paper that some analyses of emotion are more credible than others. Because this is so, the thesis that humans and nonhumans share emotions may well be a more difficult case to make than has been recognized thus far.In a similar tone, according to Jeffrey Moussaieff Masson:
While the study of emotion is a respectable field, those who work in it are usually academic psychologists who confine their studies to human emotions. The standard reference work, The Oxford Companion to Animal Behavior, advises animal behaviorists that "One is well advised to study the behaviour, rather than attempting to get at any underlying emotion".
There is considerable uncertainty and difficulty related to the interpretation and ambiguity of emotion: an animal may make certain movements and sounds, and show certain brain and chemical signals when its body is damaged in a particular way. But does this mean an animal feels—is aware of—pain as we are, or does it merely mean it is programmed to act a certain way with certain stimuli? Similar questions can be asked of any activity an animal (including a human) might undertake, in principle.
Because of the philosophical questions of consciousness and mind involved, many scientists have stayed away from examining animal emotion, and have studied instead, measurable brain functions, through neuroscience.
Research suggests that canines can experience negative emotions in a similar manner to people, including the equivalent of certain chronic and acute psychological conditions. The classic experiment for this was Martin Seligman's foundational experiments and theory of learned helplessness at the University of Pennsylvania in 1965, as an extension of his interest in depression:
A further series of experiments showed that (similar to humans) under conditions of long term intense psychological stress, around 1/3 of dogs do not develop learned helplessness or long term depression. Instead these animals somehow managed to find a way to handle the unpleasant situation in spite of their past experience. The corresponding characteristic in humans has been found to correlate highly with an explanatory style and optimistic attitude and lower levels of emotional rigidity regarding expectations, that views the situation as other than personal, pervasive, or permanent. Such studies highlighted similar distinctions between people who adapt and those who break down, under long term psychological pressure, which were conducted in the 1950s in the realm of brainwashing.
Since this time, symptoms analogous to clinical depression, neurosis and other psychological conditions have been in general accepted as being within the scope of canine emotion as well.