Bering in Mind

Bering in Mind

A research psychologist's curious look at human behavior

Is killing yourself adaptive? That depends: An evolutionary theory about suicide



Most psychological science is the science of being and feeling like a human being, and since there is only one human being that I have or ever will have experience in being, it is not always clear to me where my career ends and my personal life begins. And this is especially salient to me right now because, like many other adult gay commentators and horrified onlookers, the raft of gay teen suicides in recent weeks has reawakened memories of my own adolescent battles with suicidal thought. There is so much I want to say about this, in fact, that I’ll be breaking this column up into two separate posts, for I’m reminded of the many illuminating theories and studies on suicide I’ve come across over the years that helped me to understand—and more importantly to overcome and to escape from—that frighteningly intoxicating desire to prematurely rid myself of a seemingly interminable hell.

If only I could have reached out and gotten hold of Tyler Clementi’s shirttail before he lunged off the George Washington Bridge, or eased my fingertips between the rope and the neck of thirteen-year-old Seth Walsh before he hanged himself from a tree in his backyard, I would have pointed out to them that, one day, they will find beauty even in this fleeting despair. I would tell them that their sexual orientation places them in the company of some of the greatest figures and secular angels in creative history—to name just a few, Michelangelo, Caravaggio, Oscar Wilde, Andy Warhol, Leonardo da Vinci, Marcel Proust, Jean Genet, Hans Christian Anderson and Tchaikovsky. Finally, I’d tell them about the scientific research and ideas that I’m going to share with you, razor-sharp reasoning by bright scholars that may have pierced their suicidal cognition just enough to allow them to breathe a little more easily through those suffocating negative emotions.

In fact, a scientific understanding of suicide is useful not only for vulnerable gay teens, but for anyone ever finding themselves in conditions favoring suicide. I say “favoring suicide” because there is convincing work—all tracing back to McMaster University’s Denys deCatanzaro’s largely forgotten ideas from the early 1980s—indicating that human suicide is an adaptive behavioral strategy that becomes increasingly likely to occur whenever there is a perfect storm of social, ecological, developmental and biological variables factoring into the evolutionary equation. In short, deCatanzaro has posited that human brains are designed by natural selection in such a way as to encourage us to end our own lives when facing certain conditions, because this was best for our suicidal ancestors’ overall genetic interests.

For good-hearted humanitarians, it may sound rather bizarre, perhaps even borderline insensitive, to hear that suicide is “adaptive.” But remember that this word means a very different thing in evolutionary terms than it does when used in clinical settings. Because natural selection operates only on phenotypes, not human values, even the darkest of human emotions may be adaptive if they motivated gene-enhancing behavioral decisions. It’s not that evolution is cruel, but as a mindless mechanism it can neither care nor not care about particular individuals; selection, after all, is not driven by an actual brain harboring any feelings about, well, anything at all. In no case does this sobering fact come into sharper focus than with the case of adaptive suicide. (I notice a similar reactionary confusion, incidentally, among “New Atheists” who bleat and huff in a Dawkinsian manner whenever they hear mention of the empirically demonstrable fact that religion is adaptive, something I’ll save for another day.)

Saying that suicide is adaptive may also sound odd to you from an evolutionary perspective, because on the surface it seems to fly in the face of evolution’s first rule of thumb, which is to survive and reproduce. However, as William Hamilton’s famous principle of inclusive fitness elucidated so clearly, it is the proportion of one’s genetic material surviving in subsequent generations that matters; and so if the self’s survival comes at the expense of one’s genetic kin being able to pass on their genes, then sacrificing one’s life for a net genetic gain may have been adaptive ancestrally.

Before we get ahead of ourselves, let’s first ease into the suicide-as-adaptation argument with a few nonhuman examples, which come mostly from the insect and arthropod worlds. Take male Australian redback spiders (Latrodectus hasselti), for instance, which seem content to be cannibalized by—to say the least—sexually aggressive female redback spiders during sex. Aside from putting a damper on an otherwise enjoyable act, being eaten alive while copulating would seem rather counterintuitive from an evolutionary perspective. But when biologists looked more closely at this spidery sex, they noticed that males that are cannibalized copulate longer and fertilize more eggs than males that are not cannibalized; and the more cannabilistic a female redback spider is, it turns out, the more desirable she is to males, even rejecting more male suitors. Another example is bumblebees (Bombus lucorum), a species that is often parasitized by invidious little conopid flies that insert their larva in the bee’s abdomen. Once infected, the bumblebee dies in about twelve days, and the parasitical flies pupate until their emergence the following summer. What’s interesting about this, however, is that parasitized bumblebees essentially go off to commit suicide by abandoning their colony and spending their remaining days alone in far-away flower meadows. In doing so, these infected bumblebees are leading the flies away from nonparasitized kin, increasing inclusive fitness by protecting the colony from infestation.

What is critical to take away from these examples is that the suicidal organism is not consciously weighing the costs of its own survival against inclusive fitness gains. Redback spiders and bumblebees aren’t mindfully crunching the numbers, engaging in self-sacrificial acts of heroic altruism, or waxing philosophically on their own mortality. Instead, they are just puppets on the invisible string of evolved behavioral algorithms, with neural systems responding to specific triggers. And, says evolutionary neurobiologist Denys deCatanzaro, so are suicidal human beings whose emotions sometimes get the better of them.

So let’s turn our attention now to human suicide. To crystallize his position, I present deCatanzaro’s “mathematical model of self-preservation and self-destruction” (circa 1986):

Ψi = ρi + Σbkρkrk

Where Ψi = the optimal degree of self-preservation expressed by individual i (the residual capacity to promote inclusive fitness);

ρi = the remaining reproductive potential of i;

ρk = the remaining reproductive potential of each kinship member k;

bk = a coefficient of benefit (positive values of b k ) or cost (negative values of b k ) to the reproduction of each k provided by the continued existence of i (-1 ≤ b ≤ 1);

rk = the coefficient of genetic relatedness of each k to i (sibling, parent, child = .5; grandparent, grandchild, nephew or niece, aunt or uncle = .25; first cousin = .125; etc.).

For the mathematically disinclined, this can all be translated rather straightforwardly as follows: People are most likely to commit suicide when their direct reproductive prospects are discouraging and, simultaneously, their continued existence is perceived, whether correctly or incorrectly, as reducing inclusive fitness by interfering with their genetic kin’s reproduction. Importantly, deCatanzaro, as well as other independent researchers, have presented data that support this adaptive model.

In a 1995 study in Ethology and Sociobiology, for example, deCatanzaro administered a 65-item survey including questions about demographics (such as age, sex and education), number and degree of dependency of children, grandchildren, siblings and siblings’ children, “perceived burdensomeness” to family, perceived significance of contributions to family and society, frequency of sexual activity, stability/intimacy/success of relations to the opposite sex, homosexuality, number of friends, loneliness, treatment by others, financial welfare and physical health, feelings of contentment, depression, and looking forward to the future. Respondents were also asked about their suicidal thoughts and behaviors—for example, whether they had ever considered suicide, whether they had ever attempted it in the past, or ever intended to do so in the future. The survey was administered to a random sample of the general Ontario public, but also to theoretically targeted groups, including elderly people from senior citizen housing centers, psychiatric inpatients from a mental hospital, male inmates incarcerated indefinitely for antisocial crimes and, finally, exclusively gay men and women.

Many fascinating—and rather sad—findings emerged from this study. For instance, the greatest levels of recent suicide ideation were in male homosexuals and the psychiatric patients, whereas the prison population showed the most previous suicide attempts. “It gets better,” sure, but we’re always at risk, and this evolutionarily informed model helps gay individuals to come to grips with that lamentable reality. But the important takeaway message is that the pattern of correlational data conformed to those predicted by deCatanzaro’s evolutionary model. Although the author offers the important disclaimer that “the observational nature of this study limits strong causative inferences,” nevertheless:

The profile of correlations agrees with the notion that suicidal ideation is related to conjunction of poor reproductive prospects and diminished sense of worth to family. Concordance of the data with the hypothesis is apparent in reliable relationships of reproductive and productive parameters to suicidal ideation.

One noteworthy thing to point out in such data is the meaningful developmental shift that occurs in the motivational algorithm. Whereas heterosexual activity is the best inverse predictor of suicidal thoughts among younger samples, this is largely replaced among the elderly by concerns about finances, health and especially the sense of “perceived burdensomeness” to family. A few years after this Ethology and Sociobiology report, a follow-up study in Suicide and Life-Threatening Behavior, conducted by an independent group of investigators seeking to further test deCatanzaro’s model, replicated the same predicted trends.

As persuasive as I find this model, I still had a question left unanswered by deCatanzaro’s basic argument, so last week I dropped him an email seeking clarification. Basically, I wanted to know how the suicidal patterns of contemporary human beings relates to those of our ancestral relatives, who presumably faced the conditions in which the adaptation originally evolved, but who in many ways lived in a very different world than our own. After all, even with guns, knives and drugs at our disposal, committing suicide is not always an easy thing to do, logistically speaking. In an article published earlier this year in Psychological Review, for instance, University of Rochester psychiatrist Kimberly Van Orden and her colleagues cite the case of a particularly tenacious suicidal woman:

Case #7 was described as being socially isolated when she attempted suicide with an unknown quantity and type of pain medication and also opened her wrist arteries. This action led to some degree of unconsciousness, from which she awoke … She then threw herself in front of a train, which was the ultimate cause of her death.

Now consider the suicide methods that would have been available to our ancient relatives in a technologically sparse environment—perhaps a leap from a great height where, if one weren’t successful, might have at least led to wounds sufficient enough for the person to eventually die from infection. Starvation. Exposure. Drowning. Hanging. Offering oneself to a hungry predator. Okay, so maybe there were more methods available to our ancient forebears than I realized. You see what I mean, though. Today, moving your fingertip but by a hairsbreadth is a surer route to oblivion than anything our species has ever known before; gun-owners might as well have an “off” button, it’s so simple now. (This is one of the many reasons that I don’t own a gun—deCatanzaro’s suicide algorithm is stochastic, which means that the figure it generates for a given individual is in a constant state of flux.) But deCatanzaro doesn’t see technological advances as particularly problematic for his adaptationist model. Fossils of suicidal australopithecines or early Homo sapiens aren’t easy to come by, of course. But, as he told me in his email response to my questions:

Evidence indicates appreciable rates of suicide throughout recorded history and in almost every culture that has been carefully studied. Suicide was apparently quite common in Greek and Roman civilizations. Anthropological studies indicate many cases in technologically primitive cultures as diverse as Amerindians, Inuit, Africans, Polynesians, Indonesians, and less developed tribes of India. One interesting old review was written by [S. R.] Steinmetz in 1894 (American Anthropologist 7:53-60). Self-hanging was one of the most prevalent methods of suicide in such cultures. There are also data from developed countries comparing suicide rates from the late 19th century through the 20th century. These data show remarkable consistency in national suicide rates over time, despite many technological changes. So, the data actually do not show a major increase in suicide in modern times, although this inference must be qualified in that there may have been shifts in biases in recording of cases.

Interestingly, the methods of suicide have changed much more than the rates. For example in Japan, hanging prevailed until 1950, after which pills and poisons became the primary method. In England and Wales, hanging and drowning were common in the late 19th century, but were progressively replaced by drugs and gassing. Motives may have been more constant than means (italics added).

I find deCatanzaro’s argument that suicide is adaptive both convincing and intriguing. But I do think it begs for more follow-up research. For example, his inclusive fitness logic should apply to every single social species on earth, so why is there such an obvious gap between frequency of suicide in human beings and other animals? Each year, up to 20 million people worldwide attempt to commit suicide, with about a million of these completing the act. That’s a significant minority of deaths—and near deaths—in our species. And there is reason to be suspicious that nonhuman animal models (such as parasitized bumblebees, beached whales, leaping lemmings and grieving chimpanzees) are good analogues to human suicide. In our own species, suicide usually means deliberately trying to end our psychological existence—or at least this particular psychological existence. And whereas most other accounts of “self-destruction” in the natural world seem to involve some type of interspecies predation or parasitical manipulation, human suicides are more often driven by negative interpersonal appraisals made by other members of our own species. In fact, Robert Poulin, the University of Otago zoologist who first reported on the altered behavior of those parasitized bumblebees, even urges researchers to use caution in referring to such examples as “suicide”:

The adoption of a more dangerous lifestyle by an insect that is bound to die shortly may be adaptive in terms of inclusive fitness, but no more suicidal than, for instance, an ageing animal taking risks to reproduce in the presence of a predator as its inevitable death approaches.

I've got a hunch that suicide, like fantasy-enabled masturbation, may require recently evolved social cognitive processes that are relatively unique—in this case, painfully so—to our species. There are anecdotes aplenty, of course, but there are no confirmed cases of suicide in any nonhuman primate species. Although there are certainly instances of self-injurious behaviors, such as excessive self-grooming, these are almost always limited to sad or abnormal social environments such as biomedical laboratories and zoos. Yes, grieving young chimps have been known to starve to death from depression in the wake of their mothers’ death, but there is no evidence of direct self-inflicted lethal displays in monkeys and apes. Perhaps Jane Goodall can correct me if I’m wrong about this, but as far as I’m aware, there are no cases in which a chimpanzee has been observed to climb the highest branch it could find—and jump.

I think part of the answer to this cross-species mystery can be found in another theoretical model of suicide, this one by Florida State University psychologist Roy Baumeister, which I’ve always viewed as the “proximate” level to deCatanzaro’s “ultimate” level of explanation for suicide. These are not alternative accounts of human suicide, but deeply complementary ones. While deCatanzaro explains suicide in terms of evolutionary dynamics, Baumeister zeros in on the specific psychological processes, the subjective lens by which a suicidal person sees the world. His model describes the engine that actively promotes the adaptive response of suicide. I should hasten to add that I don’t think either of them— deCatanzaro or Baumeister—necessarily see their models as being complementary in this way. I don’t even know if either is aware of the other. But this is how the two approaches have always struck me. Baumeister’s 1990 Psychological Review article on the subject, titled “Suicide as Escape From Self,” is, quite honestly, one of the most shockingly insightful manuscripts I have ever read, in any research literature.

And it’s that piece that I’ll kick off with later this week in “Part II” on the science of suicide along with other evolutionary tidbits. I’ll also discuss more current work, including some thoughts about why I believe modern schools place vulnerable adolescents, such as gay teens, at heightened risk of suicide simply by creating an artificial social environment of exclusively same-age peers, one in which specific pressure-points of ancestral conflict are bizarrely exacerbated. “It gets better” for gay teens only because we eventually get out of that unnatural zoo that is high school.


Want more Bering in Mind? Follow Jesse on Twitter @JesseBering, visit, or friend Jesse on  Facebook.

Image ©


The views expressed are those of the author and are not necessarily those of Scientific American.

Share this Article:


You must sign in or register as a member to submit a comment.

Email this Article