Are People More Disturbed by Animal or Human Suffering?

Assessing the Influence of Victims Species and Age on Empathy Jack Levin and Arnold Arluke
Northeastern University

Abstract
This research examines the widely held belief that people are more emotionally disturbed by reports of animal than human suffering or abuse. Two hundred and fifty six undergraduates at a major northeastern university were asked to indicate their degree of empathy for either a brutally beaten human adult or child versus an adult dog or puppy, as described in a fictitious news report. In a 2 (dog vs human) X 2 (infant or puppy vs adult) factorial experimental design, participants responded to one of four vignettes on a scale designed to assess their degree of empathy. We hypothesized that the dependence of victimstheir age and not specieswould determine participants level of distress and concern for them. However, results revealed a somewhat more complicated picture. The main effect for age but not for species was significant. In a significant interaction effect, moreover, we found significantly more empathy for victims who are human children, puppies and fully-grown dogs than for victims who are adult humans. In other words, age makes a difference for empathy toward human victims, but not for dog victims. We also found that female participants were significantly more empathic toward victims either human or animalthan were their male counterparts.

Introduction
It is widely believed that people are more troubled by animals being harmed, made to suffer, or killed than by comparable mistreatment of human victims. When reports of animal victims reach the news, they seem to overshadow traumas and tragedies that befall humans. Not a month passes without at least one media report of some tragedy befalling an animal followed by hundreds if not thousands of people decrying the suffering or death, many of whom have no firsthand contact with the victim. One recent example of this occurred in Austin Texas when police mistakenly shot at point blank range and killed a mans dog, Cisco, while investigating a report of domestic violence at the wrong address (ABC News, April 16, 2012). Friends of Ciscos owner started a Facebook page

called "Justice for Cisco" that has nearly 14,000 supporters. Hundreds of people have left messages of support, outrage and anger. One supporter said: "How heartbreaking and so uncalled for. Tears just fall for the fallen. So very sorry for your loss over a mistaken address." Caring more about the plight of animals than humans can seem wrong if not disturbing because it suggests misplaced priorities that favor the former over the latter rather than the reverse. Popular television series have capitalized on this inconsistency to build a characters moral flaws. For example, the mafia gangster Tony Soprano, of the enormously successful HBO series bearing his last name, was capable of extreme violence toward people but also of intense compassion for suffering animals. In one episode, Tony stands in the rain with his sick racehorse, Pie-Oh-My, and shows more warmth and concern for this animal than for members of his own family. Sitting next to the filly, Tony reaches out and strokes the horses neck, telling her that everything will be all right. While this sensitivity to animal suffering may humanize Tonys character and make him more likeable to viewers, it also underscores his violent imbalance when it comes to humans. News reports of animal harm and subsequent outpouring of distress for the victim sometimes results in invidious comparisons to alleged lesser concern for human victims. One such incident occurred in 1994, when Barbara Schoener was attacked and killed by a cougar while jogging on the American River Canyon Trail in a California recreational area (San Francisco Chronicle, 1994). The cougars fatal attack orphaned Schoeners children; rangers ultimately shot the animal, orphaning her cubs. The public seemed more concerned about the cubs than the children of the victim; donations for the cubs far

exceeded those for Schoeners children until a conservative talk show host got hold of the story and railed at this disparity (Rollin, 2011). Despite the popularity of this belief in our society, it would be wrong to assume that there will always be more emotional distress and sympathy for animal than human victims of violence. Scholars have noted our societys inconsistent treatment of animals and how this ambivalence translates into widespread indifference toward, if not approval of, their harm (Arluke and Sanders, 1996; Herzog, 2010). For some people, animals are viewed as property and treated like objects; their harm would inspire little if any concern. Even if viewed as more than mere objects, animal suffering still may not prompt as much concern as humans made to suffer similarly. When it comes to animal versus child abuse, the former lags far behind the latter in terms of the willingness of society to criminalize the violence, intervene to stop it, and provide resources to help the victim. Social science research has yielded mixed results in support of animal versus human empathy. Research on pet attachment suggests that some people have greater emotional connection to their animals than to people. If true, it is reasonable to assume that harm to pets would disturb pet owners more than harm to people. In one study (Archer, 1997), 65% of respondents replied positively to the statement about animals: I care for them more than most people I know. However, other studies of human-animal interaction suggest less sympathy and distress for animals than people. Studies that compare grief reactions to the loss of pets versus the loss of significant humans provide inconsistent results. While some studies have found grief levels to be similar for the loss of pets to the loss of human significant others (Gerwolls and Labott, 1994; Sanders, et al. 1985), others have found less

depression after the death of a pet than after the loss of a family member or close friend (Gage and Holcomb, 1991; Rajaram et al., 1993), suggesting somewhat less human emotional connection to animals than to people. It is unclear, however, whether these comparative findings on grief reactions generalize to sympathy and distress over animal versus human suffering. In one study, moreover, no significant differences were found when comparing the willingness of bystanders of violence to help either a human or an animal victim. Laner et al., 2001 found that bystanders were no more willing to help an adult woman, a 6-yearold child, or a 40-pound dog who were hypothetical victims of a violent attack. Faulty research design may explain why the above results are mixed. The problem with prior studies that compare a participants emotional response to human versus animal tragedy is that they crudely pit species against each other. Doing so may falsely attribute greater emotional response to a victims speciesanimal or human--when in fact the response may be triggered by specific attributes of victims regardless of their species. We hypothesize that when reading about a violent act, young age of the victim will cause greater empathy than adulthood, regardless of species. More specifically, the goal of the present study is to examine the extent to which, if at all, people's reactions of sympathy and distress to animal and human suffering are due to the victim's perceived level of vulnerability as characterized by the victims youth.

Methods

All participants were undergraduate students enrolled in introductory sociology and anthropology classes at a major northeastern university. The approximate enrollment in these classes totaled 380 students; 240 students completed and returned the survey instrument. The age range was approximately, 17-25. Most students were Caucasian, and a mostly equal mix of male and female students was present in the sample. Only those students who are 18 or older were asked to participate. There were no other inclusion/exclusion criteria. In a regular classroom setting, participants were recruited by asking the students present if they would participate in the study. It was made clear that student participation was completely voluntary and their acceptance or rejection to participate would not affect their standing at the university or in the class. Before receiving the instrument, verbal informed consent was obtained and participants were given an explanation for completing the instrument. All data were collected in the class period and in the classroom of the students during a regular lecture period. Both oral and written instructions indicated that all answers were confidential, and that participants could they could stop at any time, even if they had already started the experiment. Consenting participants were randomly given a fictitious newspaper article that described one of 4 vignettes (one-year-old infant, adult in his early thirties, a puppy, or 6year-old adult dog) that detailed an attack perpetrated against the victim. The vignettes differed only by the identification of the victim. All other conditions remained the same. Our vignette for the infant victim was as follows: Please read the following article taken from the Boston Globe October 16th, 2010: BOSTON-After a noticeable increase of attacks against residents of certain Boston

neighborhoods, Police Commissioner Davis has assigned a larger law enforcement presence to certain crime hotspots around the City. Last week, police investigators documented a total of 11 attacks on residents of the South End alone. One assault involved a one-year-old male infant who was beaten by an unknown assailant. Arriving on the scene a few minutes after the attack, a police officer reported there were no life-threatening injuries to the victim. No arrests have been made in the case. Participants were then asked to indicate their degree of empathy toward the victim in the vignette by their response on an Emotional Response Scale originally employed by Batson (2007). As shown in the Appendix, the measure consisted of 16 emotions to which participants responded by indicating their feelings on a series of 7 point rating scales e.g., from not at all sympathetic to extremely sympathetic. The ratings on the 16 scales were summarized to yield a total empathy score ranging from 7 (little empathy) to 112 (much empathy). Finally, as a check on the effectiveness of the manipulation, participants were asked to recall the age and species of the victim from the vignette they had received. The nine students who gave at least one incorrect response were excluded from the study.

Results
As shown in Table 1, main effects for age of victim and gender of participant, but not for species, were significant. More specifically, female participants (X-bar = 74.94) were significantly more empathic than their male counterparts (X-bar = 69.28), regardless of species (F = 4.34; P = .04). This result is consistent with the finding of an earlier study,

in which females were found to be more distressed than males regarding victimization generally (Angantyr et al, 2011). Not surprisingly, participants in the present study were significantly more empathic when the victim was an infant or puppy (X-bar = 78.22) than an adult (X-bar = 65.99--F = 20.23; P = .0001). At the same time, the interaction between age and species also yielded significant differences, but of an unexpected character. Based on Tukeys multiple comparison test, all of the age categoriesinfant (X-bar = 82.76), puppy (X-bar = 75.34), and adult dog (X-bar = 73.15)received significantly greater empathy than did the human adult victim (X-bar = 62.09). Only in comparison with the infant did the adult dog receive significantly less empathy than other victims (HSD = 8.85).

TABLE 1: SOURCE TABLE FOR ANOVA OF THE EMPATHY MEASURE Source Sum of Mean Squares df Square F sig
Species Age Gender Species X Age Species X Gender Age X Gender Species X Age X Gender Error 356.41 6898.87 1479.40 4331.49 302.60 109.79 56.28 76383.94 224 1 1 1 1 1 1 1 356.41 6898.87 1479.40 4331.49 302.60 109.79 56.28 341.00 1.05 20.23 4.33 12.70 .89 .32 .17 n.s. P = .0001 P = .04 P = .0001 n.s. n.s. n.s.

Conclusion
Our results indicate that respondents were significantly less distressed when adult humans were victimized in comparison with human babies, puppies, and adult dogs. 7

Only relative to the human infant victim did the adult dog receive lower scores of empathy. These results provide partial support for the popular belief that people generally "care" more about animal suffering than human suffering. More specifically, individuals who view a crime scene tend to be more upset by animal than human victimization, unless the suffering is experienced by a human child. There are several possible explanations for these findings, none of which can be confirmed in the present study. Some scholars have suggested that people might have greater sympathy and distress when hearing about animal victims of violence than their human counterparts because animals are often thought of as innocent, helpless, and not culpable for being targeted for violence (Paul, 1996; Anganatyr, et. al., 2011). In their study of empathy for human and animal victims, Angantyr et al (2011) determined that their participants expressed the same degree of empathy for a baby as for a puppy. It is not surprising, therefore, that our respondents similarly felt sympathy

and distress for both human children and puppies as victims. However, unlike the results of Angantyrs earlier research which looked only at main effects of young members of both species (or interactions with gender of participants), we were able to examine the interaction between age and species. Our results suggest, in addition, that respondents were similarly concerned for adult dogs as victims. That is, only empathy toward the adult human was significantly lower than empathy expressed toward an infant, a puppy, or an adult dog. It may be that many people attribute qualities like innocence, dependence, and vulnerability to animals, regardless of their age, when compared to adult humans. In other words, dogs, whether young or adult, are seen as possessing many of the same qualities associated with human babies.

Of course, making similar attributions to both dogs and human babies may depend on the specific dog breed. In our study, breed was not specified in the experimental vignettes. Future research might examine whether the all dogs are babies perception is breed-dependent by using vignettes in which the identity of particular breeds of dogs is varied.

References
ABC News, April 16, 2012. Texas Man Claims Police Killed His Do When Responding to Wrong Address. ttp://abcnews.go.com/US/texas-man-claims-police-killed-dog-ciscoresponding/story?id=16150874#.T7kIXo6QkuI.

Angantyr, Malin, Eklund, Jakob, and Hansen, Eric M. 2011 A Comparison of Empathy for Humans and Empathy for Animals, Anthrozoos 24/4: 369-377.

Archer, John. 1997. Why Do People Love Their Pets? Evolution and Human Behavior 18:237-259.

Arluke, Arnold and Sanders, Clinton. 1996. Regarding Animals. Philadelphia: Temple University Press.

Batson, C. Daniel, Eklund, Jakob, Chermok, Valerie, Hoyt, Jennifer, Ortiz, Biaggio, 2007 An Additional Antecedent of Empathic Concern: Valuing the Welfare of the Person in Need, Journal of Personality and Social Psychology93/1: 65-74. 9

Gage, M. and Holcomb, Ralph. 1991. Couples Perception of Stressfulness of Death of the Family Pet, Family Relations 40:103-105.

Gerwolls, Marilyn and Labott, Susan. 1994. Adjustment to the Death of a Companion Animal, Anthrozoos 7:172-187.

Herzog, Hal. 2011. Some We Love, Some We Hate, Some We Eat. New York: Harper Perennial.

Kurdick, L. 2008. Pet Dogs as Attachment Figures, Journal of Social and Personal Relationships 25:247-266.

Laner, M., Benin, M., and Ventrone, N. 2001. Bystander Attitudes Toward Victims of Violence: Who's Worth Helping? Deviant Behavior 22:23-42.

Paul, Elizabeth. 2000. Empathy with Animals and with Humans: Are they Linked? Anthrozoos 13:194-202.

Rollin, Bernard. 2011. Personal communication. December 14. Boston, MA

New York Times, 1994. May 2.

10

APPENDIX: Emotional Response Scale For each emotion listed below, please circle the number that describes your feelings toward the person whose account you read. Do not worry if you are not feeling many of these emotions; only a few may apply to a particular account. Be sure to circle a response for each item.
not at all grieved moved shocked 1 1 1 2 2 2 3 3 3 4 4 4 5 5 5 6 extremely 6 6 7 7 7

offended 1 2 3 4 5 6 7 ____________________________________________________________ sympathetic perturbed compassionate 1 1 2 2 1 3 3 2 4 4 3 5 5 4 6 6 5 7 7 6 7

irritated 1 2 3 4 5 6 7 ____________________________________________________________ tender upset warm 1 1 1 2 2 2 3 3 3 4 4 4 5 5 5 6 6 6 7 7 7

disturbed 1 2 3 4 5 6 7 ____________________________________________________________

11

softhearted distressed concerned sad

1 1 1 1

2 2 2 2

3 3 3 3

4 4 4 4

5 5 5 5

6 6 6 6

7 7 7 7

12