In everyday discussions as well as in psychological research, “empathy” has many different meanings. It has been described as an epistemic state – knowing what another individual is feeling, or as an affective state – feeling what another individual is feeling. Some see empathy as a discrete act – coming to experience the world as you imagine someone else does. Some see it as a process that blends attunement, judgment, and action.
There are also different kinds of empathy. Here we can briefly distinguish five: emotional contagion, emotional empathy, fellow feeling empathy, cognitive empathy, and a blend I call entangled empathy. One type of empathy seen in both humans and other animals is emotional transfer/contagion, sometimes also referred to as affective resonance. This is a spontaneous response to the feelings of another. Anyone who has lived with dogs will be familiar with this phenomenon. Dogs are emotional sponges – they often become stressed when their person is stressed, sad when their person is sad, joyful when their person is joyful. Other animals, as well as human infants and small children, also seem to experience these spontaneous reactions. Emotional contagion is a kind of mirroring that is relatively automatic.
There are less automatic, slightly more developed, versions of emotional empathy described in the literature. Karsten Stueber presents this type of empathy as “a pre-theoretical, quasi-perceptual capacity that enables you to recognize that someone is experiencing this or that emotion or that someone intends to accomplish such-and-such a goal” (2010, 21). This capacity, too, is something that humans share with other animals. Most helping behaviors, toward conspecifics or even other species, can best be described as a type of emotional empathy. I’ll mention specific helping actions below.
While we share emotional empathy with animals, that doesn’t mean that animals share with us the capacity to distinguish between the feelings of the person empathizing and the feelings or mental states of another, nor does it mean that animals project themselves into the emotional lives of others. In many social animals, for example, when an individual expresses alarm or fear, and sometimes even glee, that emotion can spread through a group and is often strongest among those who are close to the individual initially expressing the emotion. I’ve had the delightful experience of watching two chimpanzees in a tickle game, and as the chimpanzee who is being tickled laughs, others start laughing too, and come to join in on the fun. This fellow feeling sense of empathy is yet another variety, and some theorists, particularly those critical of the value of empathy in ethics or epistemology, have limited their understanding of empathy to just these sorts of experiences and reject its importance for ethics (Bloom 2016). When empathy is understood exclusively to involve an experience in which the agent loses herself in the emotions of another, might project her own feelings onto the other, or is unable to reflect upon the experience or engage in a cognitive perspective-taking process, then it seems reasonable to wonder what role, if any, empathy should play in an account of mature, reflective ethical engagement.
But for many working on empathy, empathic experiences are not thought to be simple emotional responses, but rather are reflective response from individuals who are able to differentiate themselves from others, who can knowingly simulate or take the perspective of another, and who can at least make rudimentary causal inferences. This is a type of cognitive empathy, and the primary difference between emotional empathy and cognitive empathy is that in the latter, the empathizer is not merely mimicking, reacting, or projecting onto the emotions of the other, but is engaged in a reflective act of imagination that allows her to gain understanding of another’s frame of mind. And, as with any attempt to understand another, one might do this well or not so well, and thus there is room for revising or correcting one’s empathy.
To illustrate how empathy can get things right, or sometimes wrong, consider the much-cited case of cognitive empathy discussed by Frans DeWaal:
Kuni, a female bonobo at the Twycross Zoo in England, once captured a starling. She took the bird outside and set it onto its feet, the right way up, where it stayed shaking. When the bird didn’t move, Kuni threw it a little, but it just fluttered. Kuni then picked up the starling, climbed to the highest point on the highest tree, and carefully unfolded the bird’s wings, one wing in each hand, before throwing it into the air.
(Preston and de Waal 2002: 19)
If Kuni is empathetically responding to the starling, her response does not seem to be emotional empathy, as she is not described as sharing the starling’s fearful emotion. Rather, she appears to be attentive to the needs or interests of the starling, and may even believe that the starling would be better off flying away. And her response does not illustrate fellow feeling empathy; Kuni seems to distinguish herself from the starling, apparently recognizing that the starling flies on wings and she does not. Kuni doesn’t jump with the bird off the highest branch. Yet, if this is a form of cognitive empathy, Kuni appears to have made some basic errors in her empathetic engagement. Throwing a bird who cannot fly off the highest branch of the highest tree is not a particularly sophisticated act of empathy! In the case of more apt cognitive empathy, the empathizer would seek to better understand the situation of the object of empathy. Empathy of this sort enables the empathizer to not only grasp the object’s state of mind or interests, but also the features of the situation that affect her and information about what led to someone in need of empathy being in that situation in the first place.
While the Kuni anecdote is an example of cognitive empathy coupled with a mistake in understanding, recent experimental work with rats suggests that perhaps they engage in more apt forms of empathy. In one set of experiments, a rat was placed in a small plastic tube that they could escape from only if another rat who was not in the tube was able to open it. The “free” rats quickly figured out how to open the tube to release the trapped rat. In these experiments, releasing the rats did not accompany any reward other than the company of the other rat.
In another set of experiments, the “free” rat was presented with two tubes, one that contained chocolate treats, and the other that contained a trapped rat. The free rat would open both tubes, often releasing the trapped rat first so that they both could enjoy a treat (Ben-Ami et al. 2011). The free rat’s behavior looks like an effective form of cognitive empathy, because they were able to understand that the trapped rat didn’t like being trapped, and that a good way to resolve the trapped rat’s situation would be to open the tube.
In these experiments, the free rat certainly is engaged in successful helping behavior, but perhaps they are not displaying empathy for the trapped rat but rather are just interested in satisfying their own desires for companionship (of course, both empathy and a desire for company can be operating simultaneously). To control for this possibility, however, in a set of experiments conducted in Japan, the motivation for companionship was experimentally held at bay in order to determine whether the rat would respond empathetically. In these experiments, two rats were in separate translucent boxes joined together. One of the boxes starts to fill with water, and the rat in that enclosure becomes distressed. The rat in the dry area releases the rat who is distressed getting wet, even though they will not be together upon release. As it turns out, rats who themselves have experienced being soaked act more quickly to release the trapped rat. Importantly, the rats will also release the distressed, soaked rats when they have the option of opening a door with chocolate instead. The researchers report that the rat will free the soaked rat first, 50–80% of the time (Sato et al. 2015). Given that the rats who had previously experienced their enclosure filling with water were quicker to free another in a similar situation, this experiment also demonstrates how increased empathetic aptness is a function of knowledge – not simply an automatic emotional response.
Rats are highly social and sensitive animals, and it is not surprising to observe that they respond to alleviate the distress of others. Are rats then more empathetic than chimpanzees? Of course, it may be easier for the rat to empathize with another rat as opposed to a chimpanzee empathizing with a bird, but before we jump to conclusions based on one case of Kuni’s unsophisticated empathy, let’s explore some other experiments with chimpanzees designed to learn about various components of cognitive empathy. These experiments sought to determine whether chimpanzees can understand that others have mental states distinct from their own; can understand the distinct interests, desires, or goals of the other; and can successfully engage in helpful behavior that facilitates the other in satisfying those interests or reaching those goals. The initial work of this kind was done by David Premack and Guy Woodruff with Sarah the chimpanzee.
In Premack and Woodruff’s experiments, they videotaped humans trying to solve problems, like how to acquire a banana that is too high to reach, or how to turn a light on when there is no bulb, or how to open a can of soup. They made videotapes of a human actor in a cage trying to accomplish some task. In addition to the videos, still photographs were taken of the actor engaged in a behavior that solves the problems. Sarah was shown each video, until the last five seconds, at which point the video was put on hold. Sarah was then shown two photographs, only one of which represented a solution to the problem. For example, in the video in which the actor was trying to open a can of soup, Sarah was presented with a picture of a can opener and a picture of a hammer. The experimenter left the room, and Sarah selected one of the two photographs by placing her selection in a designated location. Sarah made the correct selection in 21 of 24 trials. So, it looked as though Sarah understood that the human actor was attempting to achieve a particular goal, understood that he faced a problem that he wanted to overcome, and was able to determine what would allow the actor to overcome the problem to reach his goal. To be able to do that, Sarah would have to attribute “at least two states of mind to the human actor, namely, intention or purpose on the one hand, and knowledge or belief on the other” (Premack and Woodruff 1978, 515).
Premack and Woodruff’s work with Sarah was to try to determine whether chimpanzees had theory of mind (see the chapters in Part IV: Mindreading). One very odd criticism of this early work was that maybe Sarah was simply “empathizing” with the subject. Apparently, the concern was that empathy would somehow undermine theory of mind. Although they didn’t spell it out, presumably the thought was that empathy wouldn’t require any cognitive processes that built upon a distinction between one’s own state of mind and that of another. Premack and Woodruff tried to control for empathy by testing different actors using a former acquaintance of Sarah’s, to whom she showed no affection, and, alternatively, Sarah’s favorite caregiver. Sarah selected the right responses to solve the problem for the actor she liked and selected the wrong responses, failing to solve the problem for the actor she didn’t care for, at a highly significant rate. Premack and Woodruff suggested that this meant that she wasn’t putting herself into the position of the human but, instead, could recognize distinct humans, solving the problem for the human she liked and not solving the problem for the human she didn’t.
I have known Sarah for over a decade (although not when she was performing these particular tests) and have a sense of her distinct personality, and it seems to me that she is usually eager to help someone she likes and not help someone she doesn’t like. But what does this have to do with empathy? I find that the worry that she was empathizing as a deflationary account of her attribution of cognitive states to others to be misplaced, because attributing cognitive states to others is precisely what cognitive empathy requires. Indeed, what Premack and Woodruff were suggesting is that Sarah can engage in a complex act of imagination, in which she understands the position of the other who is facing a problem that is not one that she faces (a very human problem, no less, but one that she was familiar enough with given that she was raised among humans from a very early age). And she can determine what it would take to solve the problem. In addition, she might decide she doesn’t want to help someone she doesn’t like. We might say that she could empathize, but chooses not to complete her empathetic imagination process with action to solve the problem. Despite her refusal to help someone she doesn’t like, it is hard to deny that this is a fairly sophisticated cognitive process.
It appears that some chimpanzees, like the rats, are able to recognize when others are in need of help. In work done decades after the experiments with Sarah, young chimpanzees were asked to identify a human’s goal, and reliably and spontaneously helped a friendly human achieve his goal. While engaged in what appeared to be informal social interactions with the experimenter, the young chimpanzees were tested on their ability to respond to a nonverbal request for help. For example, when the experimenter was using a sponge to clean a table and dropped the sponge onto the floor, the chimpanzee he was interacting with responded to his gestural request to retrieve the sponge by picking it up and handing it to him (Warneken and Tomasello 2006). In Sarah’s case, she decided when and whether to help, but presumably, like the younger chimpanzees in the Warneken studies, was able to recognize that help was needed.
While there are a number of other studies showing that chimpanzees understand quite a bit about others’ mental states and are able to respond appropriately to others’ goals, intentions, and emotions, all of which makes sense given the complexity of the social relationships they have to navigate, there is still some skepticism about whether other animals are empathetic in anything like the way that humans can be. Yet when we look at the rat empathy experiments, it makes sense to describe the free rats as caring about the state that the distressed rats are in. They may or may not feel that distress, but they recognize the conditions that are the cause of distress, they understand what it might take to alleviate that distress, and they act to do so. Sarah, too, may or may not experience, affectively, the frustration the actor is exhibiting, but she understands that the actor has a problem to solve and she recognizes how to solve it, and when she is so inclined, she may help solve the problem.
Empathy that isn’t merely emotional empathy involves understanding the state of mind of the other, others’ physical or social situation, and their differing goals. I have argued that it also involves engaging in some action to help the other (Gruen 2015). The relationships that the empathizer is in also will play some role. This is what I call entangled empathy. Entangled empathy is an experiential process involving a blend of emotion and cognition in which we recognize we are in relationships with others, and are called upon to be responsive and responsible in these relationships by attending to another’s needs, interests, desires, vulnerabilities, hopes, and sensitivities. Relationships between interacting individuals are absolutely crucial to gaining insight into what other humans and other animals might be thinking (Andrews and Gruen 2014). We know that chimpanzees recognize relationships between individuals. Group members know the relationships between mother and infant, and relationships between males who form a coalition. They can identify familiar individuals, individuals from rival groups, and unknown individuals. They have a very keen sense of the quite distinct personalities of others in their social groups. They can even make determinations about whether to take seriously the vocalizations and displays of others. In experimental setups, they make choices based on individual differences; chimpanzees prefer to cooperate with partners who share rewards more equitably (Melis et al. 2009), and they know which partners will best help them to achieve certain tasks (Melis et al. 2006).
People who live with dogs usually understand the complexities of these relationships well. We know that dogs form different sorts of relationships with particular dogs, and this will impact a whole range of behaviors – for example, how they communicate intentions, how they read signals (in play, for example), and even who they show concern towards. (See, e.g., Horowitz 2010 or Hare and Woods 2013.)
By ignoring relationships, investigators assume it is possible to step outside of the social or to detach from the experiences of our particular embodiments and deny that we are entangled with others. It is indeed odd that most of the studies on empathy don’t mention the particular relationships, when it seems so important given that the ability to empathize functions as a key resource in social navigation. It is an especially important skill in the absence of language. Being able to empathize with others in one’s group allows one to learn what is expected and what is prohibited, and the safest way to interact with others. Empathy is important for understanding group norms. This is true of most social beings, like chimpanzees and dogs and rats.
Other animals may not have the exact kinds of empathy that humans are capable of, but sadly, humans continually fail to engage their empathy. We’re too often like Kuni, lifting up someone’s broken wing and thinking they should be able to fly, or ignoring those who are suffering altogether. Many people just don’t want to think too far beyond their living rooms, workplaces, and neighborhoods. When people do care, that care is often limited to those closest to them or most like them but not beyond, so whatever empathy they have is truncated. Empathy is also something we are taught to “get over” or grow out of. We learn to quash our caring reactions for others, and our busy lives and immediate preoccupations provide excuses for not honing our empathetic skills. Yet not recognizing that we are in complex relationships that require entangled empathetic responses, towards other humans as well as animals, is a mistake. It minimizes our agency, weakens our imagination, and undermines compassion. To correct it, perhaps we could learn a thing or two from empathetic rats.
Lori Gruen. 2015. Entangled Empathy: An Alternative Ethic for Our Relationship With Animals. New York: Lantern Books.
Heidi Maimbom (ed.). 2014. Empathy and Morality. Oxford: Oxford University Press.
Carl Safina. 2016. Beyond Words: What Animals Think and Feel. New York: Henry Holt.
Andrews, Kristin, and Lori Gruen. 2014. “Empathy in Other Apes” in H. Maimbom (ed.) Empathy and Morality. Oxford: Oxford University Press,193–209.
Ben-Ami, Inbal, Jean Decety, and Peggy Mason. 2011. “Empathy and Pro-Social Behavior in Rats”. Science, 334(6061): 1427–1430.
Bloom, Paul. 2016. Against Empathy. New York: Ecco Press.
Gruen, Lori. 2015. Entangled Empathy. New York: Lantern Press.
Hare, Brian, and Vanessa Woods. 2013. The Genius of Dogs. New York: Plume.
Horowitz, Alexandra. 2010. Inside of a Dog. New York: Scribner.
Melis, Alicia P., Josep Call, and Michael Tomasello. 2006. “Chimpanzees Conceal Visual and Auditory Information From Others”. Journal of Comparative Psychology, 120: 154–162.
Melis, Alicia P., Brian Hare, and Michael Tomasello. 2009. “Chimpanzees Coordinate in a Negotiation Game”. Evolution and Human Behavior, 30: 381–392.
Premack, David, and Guy Woodruff. 1978. “Does the Chimpanzee Have a Theory of Mind?”. Behavioral and Brain Sciences, 1(4): 515–526.
Preston, Stephanie D., and Frans de Waal. 2002. “Empathy: Its Ultimate and Proximate Bases”. Behavioral and Brain Sciences, 25(1): 1–20.
Sato, Nobuya, Ling Tan, Kazushi Tate, and Maya Okada. 2015. “Rats Demonstrate Helping Behavior Toward a Soaked Conspecific”. Animal Cognition, 18(5): 1039–1047.
Stueber, Karsten. 2010. Rediscovering Empathy. Cambridge, MA: MIT Press.
Warneken, Felix, and Michael Tomasello. 2006. “Altruistic Helping in Infants and Young Chimpanzees”. Science, 311(5765): 1301–1303.