Death Through a Maternal Lens

Instructor Introduction

Scientists generally assume that humans are the only species that comprehends the idea of death. In her paper Death through a maternal lens, Michelle Huang explores what the behavior of non-human primates towards deceased group members—with a particular focus on the responses of mothers and infants—can tell us about “death awareness” in other animals. She applies a compelling evolutionary lens to help explain when and why we see behavior that suggests empathy, sadness, compassion, and other emotions commonly associated with grieving the loss of a loved one. Her paper does not shy away from the uncomfortable reality that some of the behaviors other species engage in can be deeply repulsive, even immoral, when viewed through a human normative lens. However, she also does a masterful job of conveying the reasons that many primatologists feel that other primates have emotional reactions to the death of a loved one that feel very much like our own.

— Stacy Rosenbaum

Death Through a Maternal Lens:
The Effects of Social Bonds, Repercussions, and Benefits on Non-Human Primate Responses to Conspecific Death 

Overview of Death Awareness in Non-Human Primates

For a long time, people have believed that our ability to comprehend our own deaths defines us as a race—but what if we weren’t alone in this understanding? (De Marco et al., 2022). Primate thanatology is the study of how non-human primate species react to death, with an interest in what their behavior can tell us about how death awareness evolved in humans (Gonçalves & Carvalho, 2019). Insights on how our close relatives understand death may allow us to draw inferences on how humans began understanding death, giving us a new perspective on the evolution of our own behavior.

There have been many observed cases of non-human primates displaying seemingly compassionate behaviors when faced with conspecific death—from grooming a dying group member to visiting the death site of a close associate—resulting in some studies concluding that non-human primates may have an elementary semblance of death awareness (Fernández-Fueyo et al., 2021; Gonçalves & Biro, 2018; Watts, 2020). One theory for death awareness in non-human primates is the levels of death hypothesis, which argues that there are four levels of death awareness: irreversibility (permanency of death), universality (encompassing nature of death), cessation (termination of biological functions after death), and causation (factors that cause death). With humans capable of understanding all four, non-human primates are theorized to only be capable of comprehending one or two (Gonçalves & Carvalho, 2019).

The unique exploratory and empathetic behaviors some non-human primates display when faced with conspecific death may demonstrate the form of death awareness they hold. However, the rarity of these actions raises a question: under what conditions do such behaviors appear? To answer, I will focus on the most common situations under which these behaviors arise, namely those related to the deaths of a mother or infant. Through the behaviors of non-human primate mothers and infants when faced with each other’s death, as well as the responses shown by other adult individuals, I argue that occurrences of non-human primate compassion towards the dead and dying are highly dependent on the strength of the social bond between the agent and recipient of the behavior, as well as the possible repercussions and benefits to such actions. In this paper, I will first analyze examples of how maternal bond strength influences compassionate behavior in responses to death before moving on to the effects that possible repercussions and benefits may have on said behavior.

Infant Corpse Carrying (ICC)

The bonds between mother and infant are some of the strongest social bonds that can be formed in primate groups. As such, cases of infant deaths are commonly studied occurrences in the field of primate thanatology, as it is within the responses to these deaths that we may find a large amount of data on how different factors influence behavior surrounding conspecific death. One of such behaviors is the practice of infant corpse carrying (ICC). Observed across a wide range of primate species, this behavior consists of a mother carrying, grooming, and caring for the body of her deceased infant as if it were still alive (Gonçalves & Carvalho, 2019).

Through an evolutionary perspective, ICC makes little sense. With the corpse unable to cling like a live infant, ICC not only makes it difficult to forage but also drains energy, as carriers must locomote on three limbs instead of four (Gonçalves & Carvalho, 2019). Yet the cost is not just a drain on energy. Carrying durations can range from hours to several months, leading to longer-carried bodies mummifying in their mothers’ arms and attracting potential aggression from both predators and other group members with the smell (Gonçalves & Carvalho, 2019). With such negative downsides to ICC, one would expect for this behavior to be strongly selected against, yet its unusual prevalence across different primate species indicates that there must be a reason for mothers to perform ICC.

One hypothesis that attempts to answer this question is the grief-management hypothesis, which states that ICC may be a function of relieving negative emotions that stem from the infant’s death (Gonçalves & Carvalho, 2019). Generally, primate mothers keep their infants close by, with some displaying a kind of “separation anxiety” (Fernández-Fueyo et al., 2021). Given that bereaved non-human primate mothers often search for their infant’s body after separation, they may still experience stress when unsure of the corpse’s whereabouts, causing mothers to carry their infants after death to avoid these emotions. These observations align with the fact that mothers who experience separation and infant death produce increased levels of glucocorticoids (stress hormones) (Fernández-Fueyo et al., 2021). This continuation of maternal behavior not only applies to carrying but also to the protection of their infant’s corpse, as mothers have been observed protecting their infant’s body from other group members and any animals interested in feeding on it (Anderson, 2011; Gonçalves & Carvalho, 2019; Kennedy, 2005). If the grief-management hypothesis is assumed to be true, the practice of ICC may imply that non-human primate mothers continue to be affected by the maternal bond they formed with their infants even after the latter’s death, causing them to perform redundant maternal behaviors.

Maternal Cannibalism

While much rarer than ICC, maternal cannibalism—the consumption of an infant corpse by its mother—has been observed across many different species, including bonobos, chimpanzees, orangutans, vervet monkeys, drills, and tufted capuchins (Watson & Matsuzawa, 2018). The practice is rarely straightforward. Some mothers perform both ICC and cannibalism, presenting conflicting urges. Others may eat bits of skin in an attempt to groom their infant, similar to the ingestion of ectoparasites found during grooming (Watson & Matsuzawa, 2018). Yet others appear to completely treat their infant’s body as nourishment (Kaigaishi & Yamamoto, 2023). At first glance, the latter form of maternal cannibalism appears to be a direct counterpoint against the concept that strong social bonds in non-human primates result in more compassionate behaviors towards the dying recipient. However, this practice may actually provide evidence that a strong bond between mother and infant is necessary for the development of maternal care.

One of the most unique cases of maternal cannibalism is that of a Japanese macaque female living within the Awajishima group (Kaigaishi & Yamamoto, 2023). Having given birth to a premature stillborn infant, she initially displayed maternal behaviors, such as licking the body and holding it to her chest. However, she soon began eating the body from the lower half, leaving nothing but the head, which she abandoned shortly after. Her behavior of eating her stillborn was not unique—out of 16 cases of maternal cannibalism observed in Old World monkeys, 14 involved stillborns (Kaigaishi & Yamamoto, 2023). Previous researchers theorized that maternal cannibalism may occur due to environmental stress, similar to cases in rodents and other animals. However, what makes this case strange is that the group the mother in question belonged to was both free-ranging and had a stable food source from caretakers. She was also an experienced mother, which removes the possibility of first-time-birth-related stress (Kaigaishi & Yamamoto, 2023). If not due to stress, then why?

One possibility is that this event of maternal cannibalism may have been an extension of placental consumption, a behavior commonly observed in Japanese macaques (Kaigaishi & Yamamoto, 2023). The benefits of eating her placenta—replenishing nutrients and preventing the attraction of predators—would also apply to eating her stillborn, which could have encouraged her to perform the act. However, the difference between the consumption of her placenta and her infant is her initial behavior towards the stillborn. She originally treated the stillborn as a child, attempting to groom and cradle it, but quickly changed to treating it as nourishment. This drastic shift in mindset may have been caused by a lack of “infant” cues from the body—its lack of fur, undeveloped structure, and absence of movement may have given the mother enough information to realize the “lifelessness” of her infant or to stop recognizing the body as an infant altogether. This idea was further supported by the mother’s lack of reaction at the removal of her infant by caretakers, a rare reaction as non-human primate mothers tend to show severe agitation when separated from their infant’s corpse (Kaigaishi & Yamamoto, 2023). 

The distinct difference between the compassionate behavior of ICC and the comparatively indifferent behavior of maternal cannibalism shows that a preliminary bond between mother and infant, created through interaction on both sides, may be essential for the full implementation of maternal behaviors. As initiation of ICC is an indicator for the presence of a maternal bond, this claim is supported by the observation that 90% of Japanese macaques carried infants who died more than a day after birth, while only 25% carried stillborns and those who died within a day (Kaigaishi & Yamamoto, 2023). As such, this case of maternal cannibalism shows how drastic changes in behavior can be caused by the absence of a bond and, inversely, how a strong social bond facilitates compassionate responses to conspecific death.

Group Responses to Infant vs. Adult Conspecific Death

Similar to how the absence of a maternal bond can result in maternal cannibalism, a weak social bond between infants and non-mother adult individuals (for primate species that live in communities) may also result in a weak response to infant death. Generally, adult individuals appear unfazed by the deaths of others’ infants, which may stem from the differences in strength between the infant’s maternal bonds and the bonds of the infant and other group members (Gonçalves & Carvalho, 2019). Primates invest a lot of effort into offspring, as shown by incredibly long nursing periods compared to other animals. To ensure the survival chances of their infants, and by extension, genes, mothers must exert great effort into protecting, feeding, and teaching their children until they reach maturity (Kennedy, 2005). As a result, young infants spend most of their time with their mothers, with few chances to interact with the other members of the group. This lack of interaction results in weaker social bonds between the infant and other adults in the group, leading to indifferent reactions to the infant's death. Further support for this claim comes from the strikingly different responses shown to adult conspecific death, with most members of the group showing interest and performing exploratory behaviors (manipulating, smelling, hitting, grooming, peering, etc.) (Watts, 2020).

While most interactions with adult dying conspecifics end at curious and simple consolatory actions, a few uniquely compassionate behaviors between those with strong social bonds have also been observed. For example, in the Ngogo chimpanzee population, a male peacefully accompanied another wounded male before the latter’s death (Watts, 2020). In a group of captive chimpanzees, all individuals surrounded and comforted a dying elderly female with grooming and soft noises, showing subdued behavior for weeks after her passing (Anderson et al., 2010). Previously unseen caretaking behaviors—including embracing, sitting by, and alarm calling—were observed in a wild common marmoset male as he seemed to show compassion for his dying pair-bonded mate (Bezerra et al., 2014). In all of these cases, the recipients and agents of the observed behavior had strong social bonds comparable to those of maternal bonds: the Ngogo males were close associates and grooming partners; the small group of captive chimpanzees had spent a large amount of time together; and the common marmosets had been pair-bonded for more than ten years (Anderson et al., 2010; Bezerra et al., 2014; Watts, 2020). These rare examples, along with the behaviors shown by bereaved mothers, provide further evidence to suggest that the effects of strong social bonds on the appearance of empathetic behaviors can be applied widely to the responses of most non-human primates to conspecific death, rather than just those of mothers and infants.

Infant Responses to Maternal Death

Just as non-human primate mothers show intense behaviors when their infants die, so do infants when their mothers die. Most orphaned infants seem to go through two phases of behavior after the death of their mother: first showing agitation as they attempt to regain their bond with their mother and then entering a state similar to that of human clinical depression (Anderson, 2011). During this period, they display high levels of stress, performing actions such as holding “vigils” by their mother’s corpse, appearing distressed when touching the body, rocking, giving alarm calls, and showing lethargy and a lack of appetite (Gonçalves & Carvalho, 2019). One particularly significant case was that of Flint, an 8-year-old chimpanzee male who was particularly emotionally dependent on his mother. After her death, he appeared to lose his “will to live,” refusing to eat and play. Three weeks after his mother’s death, he died as well (Anderson, 2018). While death is not an uncommon result for orphaned infants, Flint’s case was unique in the intensity of his behaviors compared to those of other orphans—a difference that may be attributed to his unusually close bond to his mother, as well as the loss of both her protection and the social and emotional bonds he shared with her.

The Role of Infanticide in Infant Corpse Abandonment

While there are many examples that show the link between strong social bonds and compassionate responses to conspecific death, there are also a few exceptions that oppose this pattern, one of which being the effects of infanticide—usually referring to the action of a male killing the infant of a female they wish to mate with—on ICC.

Although most non-human primate mothers carry the bodies of their infants for at least a few hours, in the cases of infanticides, mothers not only often fail to initiate ICC but may even abandon infants who are still living (Mora et al., 2024; Watts, 2020). In fact, a 2020 study found that out of nine eastern chimpanzee mothers in Gombe National Park who did not perform ICC, seven had lost their infants to infanticide (Lonsdorf et al., 2020). With such a large proportion, it’s likely that there is a reason behind this behavior. 

One possible explanation for this occurrence is the unawareness of death hypothesis, which states that non-human primate mothers perform ICC due to being unaware of their infant’s death (Gonçalves & Carvalho, 2019). As an infant who dies from infanticide has more cues on its body pointing to death, as well as important context surrounding their death (an infanticidal male’s attack), the traumatic nature of their death may help the mother come to a conclusion that her infant has died, resulting in a quicker abandonment of the body (Watts, 2020). However, non-human primate mothers who perform ICC display certain behaviors that may indicate they are aware of their infant’s death. Firstly, mothers tend to carry their live infants in an upright position, with the infant clinging onto their back or chest. However, those performing ICC often display abnormal carrying postures, such as slinging the body over their shoulder, hanging it around their neck, or dragging it behind them (Lonsdorf et al., 2020). Secondly, mothers performing ICC may sometimes put the body into dangerous situations, like a white-faced capuchin mother who showed no worries with completely submerging her infant’s body in water for an extended period of time. Interestingly, she was also seen preventing wasps from eating her infant’s corpse, demonstrating a continuation of protective maternal behavior even after showing awareness of her infant’s death (Anderson, 2011). Finally, the fact that some mothers carry their infant’s corpse until it has been mummified may provide further evidence against the unawareness of death hypothesis. Due to the fact that the hypothesis claims death cues cause the mother to realize her infant’s death and abandon the body, it would be difficult to argue that long-carrying mothers are completely unaware, especially with cues such as the visual decomposition and rancid smell of the body. 

Rather than unawareness, the answer to this question may relate to the choice the mother must make between her future progeny and the survival chances of her current child. When faced with the infanticide of her infant, the mother may choose not to perform ICC due to fear of retaliation from the infanticidal male (Gonçalves & Carvalho, 2019). Along with the threat to her safety, the stressful circumstances surrounding her infant’s death may also play a role in encouraging the mother to abandon her infant’s corpse, as the associated negative emotions might drive her to leave the area quicker (Fernández-Fueyo et al., 2021). As a result, non-human primate mothers may choose their safety, and therefore their future progeny, over the maternal bond they have with their current infant, especially if said infant has already died. Even if the infant is still alive, the mother may still choose to cut off her investment considering its lowered survival chances due to its injuries and likelihood of being attacked again (Gonçalves & Carvalho, 2019). This is not to say that the maternal bonds of non-human primates are so weak that abandonment occurs whenever the infant’s survival is threatened. In fact, in other cases of death, such as illness, the mother will carefully care for and support her child if it cannot cling by itself (Anderson, 2018). In the case of infanticide, it may simply be that the possible repercussions are severe enough to prevent the mother from displaying behaviors she might have shown under different circumstances—an example of how both consequences and social bonds strongly regulate the appearance of compassionate behaviors towards the dead and dying.

Effects of Infant Age on ICC

In a similar fashion to how the threat of male aggression can prevent mothers from performing ICC, excessively negative disadvantages of carrying may also cause the same result. The maternal bond strength hypothesis argues that the initiation and duration of ICC depend on the mother’s bond to her infant. Considering this hypothesis to be true, one would expect to see mothers carrying older children for longer due to a greater time spent caring for and forming a bond with them (Gonçalves & Carvalho, 2019). However, many studies have observed the exact opposite. Non-human primate mothers have shown a higher probability of initiating ICC and carrying for longer with younger children, while the bodies of older children are abandoned quicker. This may reflect the increased difficulty associated with carrying an infant of larger size, as older infants—though they may have a stronger bond with their mothers—are much more energetically expensive to carry (Fernández-Fueyo et al., 2021). As a result, while mothers may be able to bear the relatively smaller expenses of carrying a younger infant, there seems to be a point where the energy and resource drains associated with carrying an older infant outweigh any possible benefits to said carrying. Due to the excessive burden, the mother may choose to prioritize her ease of survival over her desire to carry by either shortening the duration of ICC drastically or forgoing the behavior entirely.

Non-Mother Carriers

Although repercussions can discourage compassionate behaviors such as ICC, benefits may play the opposite role and encourage these actions. In a previous paragraph, I established that adult members of the group rarely show interest in the deaths of others’ infants. However, there are exceptions, as young females have been observed taking and performing ICC with other females’ dead infants (Gonçalves & Carvalho, 2019). Theoretically, these non-mother carriers shouldn’t have had a very strong bond with the infant yet still chose to take on the heavy burden of ICC. One hypothesis for this behavior is the learning-to-mother hypothesis, which states that young females without children of their own can practice maternal behaviors on infant corpses to prepare for motherhood. This experience would allow them to increase the survival chances of their own progeny, ensuring their expensive investments in offspring turn out well (Gonçalves & Carvalho, 2019). As such, even though there may be a very weak bond between the infant and the non-mother carrier, the appearance of this seemingly needless behavior can be explained by the benefits from practice outweighing the energy costs of carrying.

Conclusion

From the evidence given, it can be argued that the initiation of compassionate actions from non-human primates towards conspecific death is highly dependent on the strength of the social bonds between the agent and recipient of the behavior. When comparing the cases of ICC, where the mother protects and carries her infant’s body at her own expense, and maternal cannibalism, where the mother eats her infant to regain nutrients, it becomes clear that the behaviors of mothers who have a strong bond with their infant are strikingly different from those shown when said bond is absent. This claim is further supported by the rare observations of compassionate behaviors shown by adult non-human primates to dying individuals they shared close bonds with. However, the repercussions and benefits of the behaviors initiated by social bond strength must also be considered, as they can result in the exhibition of behavior, such as in non-mother carrying, or the suppression of behavior, as highlighted in the case of infanticides. The relationship between the drive, deterrents, and incentives of compassionate behaviors towards the dying in non-human primates may be a helpful starting point for comparisons to human evolutions of grief. Before then, however, future directions of research should focus efforts on gathering more standardized data about non-human primate responses to conspecific death across different populations, as current data is largely anecdotal, resulting in a greater risk of misreading and anthropomorphizing observed behavior. Although we have many differences from our non-human relatives, we still have much to learn from them—both about ourselves and the role we play in this world. 

References

Anderson, J. R. (2011). A primatological perspective on death. American Journal of Primatology, 73(5), 410–414. https://doi.org/10.1002/ajp.20922

Anderson, J. R. (2018). Chimpanzees and death. Philosophical Transactions of the Royal Society B: Biological Sciences, 373(1754), 20170257. https://doi.org/10.1098/rstb.2017.0257

Anderson, J. R., Gillies, A., & Lock, L. C. (2010). Pan thanatology. Current Biology, 20(8), R349–R351. https://doi.org/10.1016/j.cub.2010.02.010

Bezerra, B. M., Keasey, M. P., Schiel, N., & Da Silva Souto, A. (2014). Responses towards a dying adult group member in a wild New World monkey. Primates, 55(2), 185–188. https://doi.org/10.1007/s10329-014-0412-8

Chaves, Ó. M., Martins, V., Camaratta, D., & Bicca-Marques, J. C. (2020). Successful adoption of an orphan infant in a wild group of brown howler monkeys. Primates, 61(2), 301–307. https://doi.org/10.1007/s10329-019-00785-2

De Marco, A., Cozzolino, R., & Thierry, B. (2022). Coping with mortality: responses of monkeys and great apes to collapsed, inanimate and dead conspecifics. Ethology Ecology & Evolution, 34(1), 1–50. https://doi.org/10.1080/03949370.2021.1893826

Fashing, P. J., Nguyen, N., Barry, T. S., Goodale, C. B., Burke, R. J., Jones, S. C. Z., Kerby, J. T., Lee, L. M., Nurmi, N. O., & Venkataraman, V. V. (2011). Death among geladas ( Theropithecus gelada ): a broader perspective on mummified infants and primate thanatology. American Journal of Primatology, 73(5), 405–409. https://doi.org/10.1002/ajp.20902

Fernández-Fueyo, E., Sugiyama, Y., Matsui, T., & Carter, A. J. (2021). Why do some primate mothers carry their infant’s corpse? A cross-species comparative study. Proceedings of the Royal Society B: Biological Sciences, 288(1959), 20210590. https://doi.org/10.1098/rspb.2021.0590

Goldsborough, Z., Van Leeuwen, E. J. C., Kolff, K. W. T., De Waal, F. B. M., & Webb, C. E. (2020). Do chimpanzees (Pan troglodytes) console a bereaved mother? Primates, 61(1), 93–102. https://doi.org/10.1007/s10329-019-00752-x

Gonçalves, A., & Biro, D. (2018). Comparative thanatology, an integrative approach: exploring sensory/cognitive aspects of death recognition in vertebrates and invertebrates. Philosophical Transactions of the Royal Society B: Biological Sciences, 373(1754), 20170263. https://doi.org/10.1098/rstb.2017.0263

Gonçalves, A., & Carvalho, S. (2019). Death among primates: a critical review of non‐human primate interactions towards their dead and dying. Biological Reviews, 94(4), 1502–1529. https://doi.org/10.1111/brv.12512

Kaigaishi, Y., & Yamamoto, S. (2023). A report of stillbirth and subsequent maternal cannibalism observed in a free-ranging group of Japanese macaques at Awajishima, Japan. Primates, 64(6), 573–578. https://doi.org/10.1007/s10329-023-01091-8

Kennedy, G. (2005). From the ape’s dilemma to the weanling’s dilemma: early weaning and its evolutionary context. Journal of Human Evolution, 48(2), 123–145. https://doi.org/10.1016/j.jhevol.2004.09.005

Lonsdorf, E. V., Wilson, M. L., Boehm, E., Delaney-Soesman, J., Grebey, T., Murray, C., Wellens, K., & Pusey, A. E. (2020). Why chimpanzees carry dead infants: an empirical assessment of existing hypotheses. Royal Society Open Science, 7(7), 200931. https://doi.org/10.1098/rsos.200931

Mora, J. M., Pérez Carrillo, J. F., Carrillo, L., & López, L. I. (2024). Comparative Behavior of the White-faced Capuchin (Cebus imitator) in Response to Deceased Conspecific and Non-Conspecific Infants. Caribbean Journal of Science, 54(1). https://doi.org/10.18475/cjos.v54i1.a16

Van Leeuwen, E. J. C., Mulenga, I. C., Bodamer, M. D., & Cronin, K. A. (2016). Chimpanzees’ responses to the dead body of a 9‐year‐old group member. American Journal of Primatology, 78(9), 914–922. https://doi.org/10.1002/ajp.22560

Watson, C. F. I., & Matsuzawa, T. (2018). Behaviour of nonhuman primate mothers toward their dead infants: uncovering mechanisms. Philosophical Transactions of the Royal Society B: Biological Sciences, 373(1754), 20170261. https://doi.org/10.1098/rstb.2017.0261

Watts, D. P. (2020). Responses to dead and dying conspecifics and heterospecifics by wild mountain gorillas (Gorilla beringei beringei) and chimpanzees (Pan troglodytes schweinfurthii). Primates, 61(1), 55–68. https://doi.org/10.1007/s10329-019-00735-y