A Tale of Dominance, Submission, and Friendship on the Cheap.

So humans are a peculiar species. You probably knew that. What you may not have realized — like I hadn't until recently — is just how many traits and behaviors are uniquely human.

We're not just big-brained featherless bipeds with special instincts for language, tools, and culture. We're also the only creatures¹ who sing from the ground, sing and dance together, bury our dead, point declaratively, enjoy spicy foods, blush, and faint — not to mention all our weird sexual practices. We have the least symmetrical brains, the most dependent babies, and the fastest and most accurate overhand-throwing skills. And we have unique instincts for tempering aggression — about which, more later.

But today I mostly want to talk about crying, or to be more precise, weeping or emotional tears. Of course humans aren't the only species whose eyes occasionally fill up with fluid — for example, when irritated. Nor are we the only species to wail in distress. But we're the only ones who unite watery eyes and noisy crying together in a single behavior.


So: Crying, weeping, emotional tearing — what's that about?

Let's start by clearing some misconceptions.

The most pernicious myth about crying is that we use it to "process our emotions," specifically the emotion of sadness — in other words, that crying is a "response to grief." This idea is so wrong, in so many ways, that it needs a thorough exorcism before we can proceed.

Problem the first: There's no reason for Nature to have designed us (by way of evolution) to use leaky eyes or a heaving chest simply to "process" any of our emotions. In fact, as a general rule, emotions aren't the kind of thing that need to be "processed" at all — as if they were industrial byproducts that needed to be discharged from the thinking factory. Emotions (and their expressions) aren't mere side-effects of something else — they're purposeful unto themselves. They evolved because they put our brains and bodies into a state that's helpful for dealing with specific problems or circumstances. If evolution devised to make our bodies do something, then the action is unlikely to be a meaningless side effect. There has to be a point to it.

Problem the second: We don't cry only when we're sad. Modern Western culture portrays tears and sadness as inextricably linked, but we need to break that link in order to move forward. Yes we often cry when we're sad, but we also cry when we feel angry or frustrated or powerless or humiliated, or when we experience deep awe or beauty, or during inspiring music, or when begging for forgiveness, or simply when we're in pain. There's an idea that we sometimes cry due to extreme happiness, but I think that's a little off the mark. I would rather say that we cry when we feel humbled by something great — something that, if it weren't so humbling, would normally cause unbridled happiness. For example, we cry when we feel unworthy of a significant gift or honor, or when we witness incredible acts of kindness. We cry at reunions with long-lost loved ones, as Odysseus cried with his son and wife after his 20-year absence. We cry when we learn about new members of the family. We cry not only when we lose at an important sporting event, but also when we win; we cry not only at funerals but also at weddings. If we need an emotional antecedent to crying, let's call it "feeling overwhelmed." Even that's not perfect, but it's more accurate than "sadness."

Problem the third: It's tempting to think that if crying isn't a response to sadness, it nevertheless has to be a response to something (some shared features of all the situations listed above), and our task should be to figure out what those features happen to be — i.e., what kind of stimulus provokes a human being to cry. Unfortunately, "stimulus: response" is the wrong frame to use to analyze crying, mostly because it saps the crier of intelligent agency. It treats the brain as a mindless physical system, like a lump of jello that wobbles when you poke it, rather than an adaptive agent with goals that it pursues by initiating action. And so the question to ask about crying isn't, "What causes it (as a symptom)?" but rather, "What are we trying to achieve by doing it (as a behavior)?"

Treating tears as a response to grief (or overwhelming emotion) is like treating an erection as a response to lust. It's true enough, I suppose, but doesn't help explain why erections exist. An erection isn't merely a symptom, or the response to a stimulus (like a naked body), or a way of "processing" or "expressing" libido — it's a solution to a specific problem. To analyze an erection as merely a response, then, is to completely miss the point (pun intended).

And so too with weeping. It's a behavior that, as biological organisms, we enact for a specific purpose. And yes: despite our celebrated self-awareness, we happen to be mostly oblivious to our reasons for crying. It's not a decision; it's just something we do. But we also laugh and blush for reasons we're not conscious of. What can we say? Humans are a peculiar species.

Another misconception is that crying is somehow directly therapeutic. Yes, it can feel good to cry, but no:

  • Tears don't have medicinal healing powers.
  • Tears don't serve an excretory function, i.e., by flushing toxins or stress hormones out of the body.
  • Crying isn't "cathartic," in the sense described above, i.e., of being necessary to process our emotions in order to feel better.

Again, a moment's thought about how evolution works should disabuse us of these ideas. Q: Why would Nature design an organism to use such a roundabout way of healing itself or feeling better? A: It wouldn't.

I hope I've piqued your curiosity. Tears are a fascinating puzzle, ripe with insight about the human condition. And yet they're also surprisingly under-studied, perched just at the edge of our scientific understanding. So it's open season for people like us to speculate about their origins.

And speculate we will. But don't worry: there's enough empirical and theoretical meat on these bones to make tears a worthy subject of investigation. So let's get to it.

Stylized facts

The goal of our investigation is to explain three things: First, what weeping accomplishes. Second, how we evolved to do it. And third, why only humans weep. (Incidentally, Why Only Humans Weep by Ad Vingerhoets is the most useful book I've read about crying.) In light of these goals, then, let's take a look at some intriguing facts and observations about this behavior.

To start with: Tears are the only body fluid that doesn't arouse disgust. (Go on, see if you can think of a counterexample. Breast milk, maybe?)

Ritual greeting: In some cultures, tears are used as a ritual greeting. Vingerhoets:

Travellers, missionaries, soldiers, and anthropologists since the sixteenth century have reported the remarkable custom of tearful greeting. It has been described among a great number of South American Indian tribes, but also in North America, Australia, India, and the Andaman Islands.

Complex behavior: Sometimes we shed a lone teardrop or two, but during more intense episodes, tears are just one component of a broader physiological response. Other components include heaving and sobbing, red/puffy eyes, a runny nose, a quivering lip, and a shaky voice or lump in the throat. So what we consider the central element — tears — may simply be a convenient metonym for the entire package.

Tears are extremely involuntary: We have a hard time stopping them once they start to flow, and an even harder time producing them on demand. To cry at will takes great skill — typically via method acting or some other way of drawing on "real" feelings. And a runny nose is arguably even less voluntary than a tear.

Sex differences: Women, at least in today's Europe and America, cry a lot more frequently than men (~4x), for longer durations (~2x), and more intensely (~10x more sobbing), although these differences don't appear until adolescence. Crying is largely seen as an effeminate behavior, although clearly there are cultural factors at play. To give a counterexample: the Homeric heroes did a lot of crying. Odysseus cries at least once in every book of the Odyssey, while the Iliad features tears from a variety of masculine role models, from the kings Priam and Agamemnon to the warriors Patrochlus, Antilochus, and Achilles.

Women's tears reduce testosterone in men(!): In a 2011 study, researchers found that if they collected women's tears (by having them watch tear-jerker movies), then dabbed the tears onto a gauze which they taped under men's noses, the men showed reduced testosterone levels relative to a control group of men who were subjected to a saline solution. This study was widely reported in the popular press, for obvious reasons, but the methodology was extremely suspect (if not outright disingenuous) because the only effect tested was that of female tears on men. Male tears weren't tested at all, nor were female tears tested on women — nor were a woman's tears tested on herself (perhaps at a later point in time). Instead, the researchers seem to have almost designed their experiment to encourage a hasty interpretation. (Still, there's something intriguing about the idea that crying might be connected to testosterone, so hang onto it; we'll come back to it once we've developed a stronger theoretical perspective.)

Age differences: Children cry more than adults, of course. According to Vingerhoets, men begin to cry more (and get angry less) as they reach middle age, while women experience the opposite.

Crying in children: Because it's hard to study crying behavior in adults, some of our best data comes from studies of children. Vingerhoets again:

Falls and quarrels are the commonest reasons for crying, particularly when they are caused by another child. The majority of crying is associated with social contacts, and nearly half of the crying episodes are caused by physical contact. Boys are most commonly involved in instigating crying in both boys and girls..... In the great majority of cases (76%) [the cause of crying] could be classified as conflict with other children.

All of these observations support our initial bias toward studying tears as a behavior rather than a symptom. In particular, they're a social behavior, something we evolved to do because of their effects on the people around us. In the language of biology: Tears are a signal.

Quick primer on signaling

A signal is a message, an act of communication that conveys information about the sender (e.g. the person crying) to one or more receivers (e.g. those who notice the crying).

From the perspective of the receiver, a signal is useful because it contains information (typically reliable) about something known only to the sender — such as his or her emotional state — which then allows the receiver to behave adaptively in response. From the perspective of the sender, a signal is useful precisely because (and only insofar as) it changes the receiver's behavior.

For a signal to evolve, then, requires at least two conditions:

  1. Mutual benefit, i.e., both sender and receiver must stand to gain from the signal; and
  2. Reliability, i.e., it must be hard to fake.

When a bear meets a wolf in the woods and lets out a growl, it's a signal that the bear is ready and willing to attack, and also that it has a huge body cavity (the depth of the growl being impossible to fake, e.g., by a mouse). By growling, the bear is letting the wolf know that it means business. All else being equal, however, the bear doesn't actually want to fight. It's hoping that its growl will scare the wolf off. And the wolf, too, is happy to hear the growl, rather than getting no information at all about the bear's emotional state (and then being surprised by an attack). So it's the overlap of interest between bear and wolf — their shared interest in keeping the peace — that allows the growl to evolve as a useful signal.

In this way, communication is fundamentally cooperative. When there's no overlap of interest, there's no basis for communication — no cooperative "soil" for the signal to take root and grow in. Predator and prey, for instance, tend not to communicate with each other, at least not intentionally, since they're locked in an almost perfectly zero-sum competition, with only an extremely thin overlap of interests. The few exceptions — e.g. stotting behavior among gazelles — prove the general rule.

This suggests an approach that will be useful for our purpose today (understanding tears), but which is also a more general principle for how to analyze signals: Observe how receivers react to them. Receivers' reactions need to be analyzed along with the signaling behavior itself, because the signal and the reaction are two sides of the same coin.

How we react to tears

The fact that we, as receivers, often feel manipulated by someone else's tears is the giveaway that crying is a signal — i.e., because it aims to change our behavior. When someone's tears are making a "request" that we feel is legitimate (and when our interests overlap with the sender's), we're happy to oblige. But we need to make sure we're not getting jerked around, so we're eager to evaluate the situation, to determine how appropriate an episode of crying is. In other words, we can get really judgmental about tears.

We see this in the way we disparage people who cry for reasons that aren't (in our opinion) legitimate or sufficiently serious. We call these people names — "cry-baby," "wuss," or "sissy" — or we accuse their tears of being "fake" or "crocodile tears." A number of the top crying videos on reddit, for example, were submitted with headlines like "Hard gangsta cries like a little bitch after getting a life sentence," which seem intended to provoke moral outrage and/or schadenfreude among viewers. Even when crying isn't manipulative per se, we often consider people who shed tears too readily to be weak or emotionally unstable, and sometimes unfit for duty, especially leadership roles.

So passing judgment is how we deny the request made by tears. But it's more important to examine how we honor the request. And this, in turn, depends on whether the tears are addressing us as a "bad guy" (an aggressor who caused the tears) or a "good guy" (someone who may be able to help).

If we're the bad guy, the aggressor, tears seem to be asking us to back off. "You've gone too far," they're saying. "You've crossed a line." If we choose to respond, then, we will back off, and perhaps even apologize and try to make amends. Certainly we aren't required to honor the request, but that seems to be what the tears are asking of us. And in this regard, they seem to be functioning as a kind of white flag, i.e., a submission signal.

On the other hand, if we're a good guy — someone who didn't cause the tears — then they seem to be making a different request. "Please, I need support right now," they're saying. "Come help me." If a friend phones you up sobbing and sniffling, you're eager to drop everything to help. Even one-year-old toddlers, according to Vingerhoets, offer comfort when they notice another child in tears. This comfort can take the form of a hug, a gentle hand on the back, offering a tissue, or simply "being there" and listening; in short, providing a shoulder to cry on. No matter what caused the tears — a stubbed toe, an argument with another person, the death of a loved one — offering one's shoulder is an appropriate response. Again, as receivers of the signal, we don't have to heed it, but this seems to be what the tears are asking of us. And in this regard, they seem to be functioning as a distress signal.

So these are the two clearest accounts of what's going on when we cry, i.e., what our tears "mean": either they're a submission signal, a distress signal, or both. No one contests that these are, in fact, two of the ways we use tears. The question we face now is: How, exactly, did they evolve? What problem did our ancestors face for which tears were the (locally) optimal solution?

Candidate theories

Let's look at a few of these candidate theories. All of these (including my own, which I'll present in a minute) are "just-so stories," i.e., speculation without a ton of direct evidence. Some of these are hard to take seriously, but it's useful to get a feel for the kind of explanations that have been proposed.

Submission signaling:

  • The punched-in-the-face hypothesis (Michael Graziano). This is the idea that our first, most prototypical tears were shed after getting our noses punched during fistfights. These tears, then, were originally physiological in nature (reflex tears), but later came to be understood as a general-purpose submission signal.
  • The blurry-vision hypothesis (Oren Hasson). Hasson argues that tears make a good (reliable) submission signal because the extra eye-fluid blurs our vision, making it harder both to press an attack and to defend ourselves against further attack.

Distress signaling:

  • The funeral-smoke hypothesis (Paul MacLean). MacLean suggests that our ancestors first began to shed tears during funerals, when they got smoke in their eyes from burning the bodies of their loved ones. This means that emotional tears were originally reflex tears (of irritation from the smoke), but that they became symbolic of sadness by association with death and funerals.
  • The physical-pain hypothesis (Juan Murube, via Vingerhoets). Murube proposes that emotional tears were originally triggered (as reflex tears) by physical pain, whether injury to the eye, near the eye, or as a result of clenched eyes. Think of this like the "funeral smoke hypothesis," but with a far more common (albeit less dramatic) cause for the earliest tears.
  • The neonate-mimicry hypothesis (Frans Roes). Roes argues that tears evolved because they make our faces look more infantile (like the wet face of a newborn), playing into receivers' pre-existing instincts about how to treat children, i.e., by providing care and support. (Co-opting earlier instincts is a pretty common way for a signal to evolve, since receivers don't have to learn a new behavior from scratch. Think of it like an evolutionary shortcut. Of course the new signal still has to be reliable and mutually beneficial.)
  • The silent-alarm hypothesis (Ad Vingerhoets). Vingerhoets argues that tears, unlike loud wails, don't attract predators, and thereby serve to communicate distress without tipping off the nearby lions and tigers.

All of these hypotheses are grasping toward the same thing: An explanation for how tears evolved in order to function as either a submission or a distress signal. Some of the hypotheses also try to explain why only humans evolved this particular signal. Unfortunately, what none of the hypotheses explains is why early humans needed a new submission and/or distress signal, when we already had a number of them at our disposal.

It's important to stress why this is a problem. Nature doesn't evolve new signals just for the hell of it. In fact, Nature is fairly conservative in this regard, and (whenever possible) tends to build new things on top of existing ones. So for a new signal to evolve, there has to be a good reason — some unique communicative agenda not satisfied by the existing vocabulary.

And here's the thing: Our vocabulary already includes a number a submission and distress signals. (In fact these kinds of signals are common in animal behavior.) For submission, we can fall over onto our backs (like dogs), or sit down, or use submissive body language (shoulders hunched, eyes cast down, limbs pulled in), or simply hold up our hands with our palms exposed — all of which make it harder both to defend ourselves and to continue an attack. For distress, we have yelps, whines, moans, frowns, and grimaces. To explain why we evolved a new signal, then, requires that we propose either a new constraint on how we communicate, or an entirely new communicative agenda.

Only the silent-alarm hypothesis even tries to do this, i.e., by proposing the constraint of silence. This is the kind of explanation we need, insofar as it tries to explain why our ancestors evolved an entirely new signal. Unfortunately, there are three problems with silence as an evolutionary constraint. First, it's not new or even remotely unique to our ancestors. Every animal that isn't at the top of its food chain would theoretically benefit from a silent alarm call. Second, our ancestors seemed to be evolving in the opposite direction, i.e., toward being louder and generally more conspicuous. Finally, the constraint of silence is already satisfied by two of our existing (and much more conventional) distress signals: frowns and grimaces.

So the puzzle persists: Why did we evolve yet another way to signal submission or distress? And why is this particular signal (weeping) so radically different from our other facial expressions?

* * * *

And now I'd like to throw my hat into the ring, by providing my own hypothesis for how and why tears evolved.

I'm going to present this hypothesis from two different perspectives: first as a "just-so story," then an economic model. The just-so story will focus on an iconic situation faced by our ancestors, one which created an opportunity for communication that no other (existing) signal could take advantage of. Then, in the following section, I'll provide a complementary economic explanation for what's going on. This will explain the just-so story, but will also allow us to generalize quite a bit.

Tears as a response to bullying

The iconic situation I'd like to put forward, to explain how and why tears arose, is that of being bullied. Specifically, I'm suggesting that:

Our ancestors evolved the behavior of weeping in order to prove to third parties that they were victims of aggression — to seek refuge from, and possibly "tattle" on, their aggressors. Unlike other submission and distress signals (all of which are momentary flashes of body language that don't leave any trace), weeping leaves the victim with a wet face, red/puffy eyes, a runny nose, and a heaving chest — signs that remain visible (and audible) for many minutes after a conflict. This is ideal for running off and soliciting help from others who may be initially out of earshot and/or direct sight.

So what I'm suggesting is that the original function of weeping was to serve as both a submission signal and a distress signal at the same time, but addressed to two different receivers. If Bobby-the-Bully makes Sarah cry while they're playing together, her tears are aimed both at Bobby, in submission, and potentially at an absent third party (like Sarah's mother), in distress. To Bobby, Sarah's tears are saying, "You've crossed a line and gone too far, and I have the wet face to prove it." Of course Sarah doesn't have to run off to her mother, but if she does, her tears will announce her condition. "Something traumatic happened to me," they'll say. The mother will then provide help and comfort to Sarah, and may also decide to make trouble for Bobby.

Playground disputes, of course, weren't the most critical situation our ancestors faced, and it's unlikely that a behavior would have evolved entirely to help children get along with each other. More important bullying scenarios may have included:

  • A child crying at a physically abusive caretaker (and running off to his mother).
  • A girlfriend crying at her physically or emotionally abusive boyfriend (and running off to her father or another boyfriend).
  • Anyone confronted with the sustained aggression of a peer (and running off to a friend or family member).

Tears, in this theory, are a political act, in the same way that tattling, gossiping, and whistleblowing are political. As such, we need to be constantly monitoring and evaluating them, to make sure they aren't being abused. When people cry in situations that strike us as disingenuous or manipulative, our anger and disapproval are ways of policing the signal. People can't just "cry wolf" (or "cry bully") whenever they want and get help from third parties.

So this account is plausible enough. But at least one important question remains: Why did only humans evolve this particular signal? Why doesn't it make sense for any other creature (e.g. chimps)?

The answer is fairly simple. Humans, unique among all animals, have an instinct to resist aggression even when it's directed toward other members of the group (even non-kin). We have strong social norms against aggression, coupled with a unique eagerness to support the underdog. Our hearts go out to the oppressed, downtrodden members of the group, and our behavioral instincts follow suit. This whole suite of attitudes and reactions is called a reverse dominance hierarchy, and it goes a long way toward explaining some of the more distinctive features of human social behavior, especially our instincts around social status and cooperation.

In any other species, wearing a signal that advertises, "I recently lost in a dominance challenge," is a strict liability — an invitation for others to pile on, opportunistically, and attack you while you're down (or else to mentally note that you're no longer a good, strong ally). There's no upside, therefore, to using anything other than a quick facial expression or flash of body language, to show your submission only to the aggressor.

But humans are different. For us, wearing a badge of surrender isn't strictly a liability — it's also something of an asset, insofar as it plays into the instincts of third parties to help those who have been victimized. A dramatic example will bring home the point: According to a number of studies (via Vingerhoets, again), police officers are more likely to believe the story of a rape victim if she's crying.

The economics of weeping

So being bullied (and running off for help, maybe tattling) is kind of a cute story that seems to explain some facets of our weeping behavior. Now I'd like to present the economic perspective. Even if the bullying scenario isn't what ultimately caused this behavior to evolve, I think this economic reasoning is sound, and should play a role in understanding what's going on when we cry.

(Warning: What follows is a pretty heartless, strictly clinical description of what tears are designed to do. I'm certainly not suggesting that people perform these economic calculations consciously. What's interesting about tears is that they're an involuntary social behavior — something our brains and bodies simply do, without bothering "us" with too many of the details.)

Here's the economic perspective:

Our ancestors faced a complex optimization problem: how to pursue, use, and exchange a variety of different resources (food, health, mates, allies, prestige, and dominance, among others) in an effort to maximize reproductive success, i.e., the number and success of one's offspring. Weeping (I contend) is the human organism's way of coordinating a difficult tradeoff between two of these resources. Specifically, it's a way of giving up dominance in the hope of earning allies. In other words, it's a gambit: a near-term sacrifice for a shot at eventual gain.

In this view, dominance is a critical resource for a human being. Along with prestige, it's one of the two components of social status. We don't go around playing submissive (letting people walk all over us) unless we have to. So weeping is a tactic that our brains evolved to use only as a last resort — when our backs are up against a wall, so to speak.

Here's a concrete scenario. Imagine that a bully is being a real asshole to me: calling me names, making fun of me, and maybe even pushing and shoving. I'll hold out as long as I can, but if he's unrelenting, my brain will eventually say, "OK, clearly this isn't working. I guess it's time to give up, switch gears, and go with Plan B." So I'll start to cry. Maybe this will work as a submission gesture, causing the bully to back off — or maybe not. Either way, part of the crying behavior is an impulse (on my part) to seek out social support. So I'll run off to my friends, or perhaps (if they're nowhere to be found) to the nearest group of friendly bystanders. My wet face will be a request for support and safe harbor. Maybe these friends, old or new, will help me in the direct conflict with the bully, by intervening on my behalf, or maybe they'll just be a sanctuary. And they'll be incentivized to help me, whether out of the uniquely human desire to condemn aggression and support the victim, or (what I suspect is a similar instinct) out of a selfish desire to strengthen their relationship with me. Since my tears are an honest signal of reduced social status, my friendship comes at a discount. By providing just a little bit of help in my time of need, they're able to earn my trust, loyalty, and future support in their times of need.

This latter logic, connecting me (the crier) to a support network, is what intrigues me the most. By giving up dominance (i.e. unilaterally reducing my own social status), I'm temporarily slashing my "price" on the friendship market. My tears are like an advertisement: "Friendship on sale! 50% off! Today only!" So it makes perfect sense that third parties might want to snatch me up. In fact, a good strategy for collecting friends would be to find as many criers as possible; all you have to do is offer your shoulder and spend some time "being there" for them, and you'll have won yourself a new ally.

Apologies, again, for putting this in the most vulgar economic terms. At this point, we're deep in the raw code of the Matrix, and we shouldn't expect to see anything in terms that are relatable to real people.

If it helps, we can express it in the form of an "equation":


Now, remember that 2011 study I mentioned earlier, about the effect of women's tears on men? The one that showed reduced testosterone, but failed to test any of the other permutations (e.g. women's tears on women)? Well, this is where it would be really useful to have those other data points, because I suspect — nay, predict — that the intended effect of tears is to reduce testosterone (and thereby aggression and dominance) in the criers themselves, rather than in third parties who happen to catch a whiff of their tears. In these experiments, the men were subjected to women's tears by having tear-soaked gauze taped to their upper lips. But this is an incredibly artificial scenario. In the real world, whose upper lip do our tears fall on? That's right, our own.

A final thought

Once a signal evolves and becomes established within a population — enshrined as instincts in both senders and receivers — it will naturally be co-opted for other (related) purposes.

Clearly we don't cry only when we've been bullied by other people. Sometimes we cry when we've been "bullied" by the physical world, e.g., by an aggressive piece of clamshell packaging. But many of the circumstances in which we cry (e.g. at reunions, or when we're feeling humbled) are hard to understand in terms of bullying, no matter how far we're willing to stretch the metaphor. So what do we make of all these other kinds of crying episodes?

According to the economic reasoning above, tears aren't inherently a response to bullying. That happens to be why they evolved (I'm guessing), but inherently they're just a piece of social technology, a device for coordinating the tradeoff between dominance and social support. And this invention turns out to be useful in a variety of scenarios. Whenever people are rewarded for humbling themselves and/or opening themselves up for connection, we shouldn't be surprised to find tears.

This is why we cry at reunions with long-lost loved ones. Our tears here are a gift. We're unilaterally disarming (giving up claims to dominance, at least temporarily) and offering ourselves up for connection, almost for free. Suddenly those soldier homecoming videos start to make a lot more sense.

When an athlete cries after losing an important contest, his tears start to flow out of frustration, having been "bullied" by the world. But when the winner cries, his tears come from an entirely different place. He's showing humility, a display that can be very endearing to those who witness it. (In fact, humbling ourselves is one of our most characteristic responses to being awarded prestige status. This is why we bow, blush, and accept prizes by thanking others.)

And finally, what are we to make of our most salient tears: those we shed at a funeral, wailing in grief? Strangely, they would seem to have little to do with grief per se, and everything to do with reaching out for connection in a time of need. A funeral, then, is a venue designed not for cathartically "processing our emotions," but rather for cauterizing the wound left in the community by the death of the dearly departed. We heal not as individuals but as social creatures, soldered together with tears.


Well that was a doozy. I learned a lot by writing this post; hope you learned something by reading it. Please get in touch if you know anything about tears. I would love to collect more ideas.


Except where otherwise specified, I've used "crying," "weeping," and "tears" as three synonyms for the same behavior, i.e., tearful weeping.

¹ human-unique traits. Some of these are contested. Kangaroos are also "featherless bipeds" (which I included only because it was Socrates's description of man-the-animal). Ravens may point declaratively. Ants and mayyyybe magpies and elephants bury their dead (but certainly not with the intentionality that we bring to the exercise). And I'm sure there are other examples of similar behaviors enacted by non-human animals. But the basic point stands: We do a lot of things that, if not singularly unique, are nevertheless extremely rare.

Originally published December 5, 2014.