CHAPTER SEVEN
Nutshell
And there’s your book-length answer to the Silent Movie Problem we introduced in chapter one. The problem, recall, is the galaxy-sized chasm between the dialogue-free lives of our relatively recent ancestors and the brimming-with-dialogue lives of us in relatively recent times. In short, living as social animals without language seems almost inconceivable.
But we saw that the lives of social animals aren’t silent at all. Whether visual, auditory, or whatever, emotional expressions are constantly reverberating between social animals. And these expressions serve us better than any actual signaling system an expensive consulting firm might provide, no matter how clever they think they are.
Our emotional expression system gives us the crucial ability to talk the talk that matters most in life. To find compromises. To negotiate. To avoid having to fight about absolutely everything!
Because if you do have to fight about everything, you’ll tend to make sure that “everything” is basically nothing, and you’ll end up missing out on loads of potential scenarios where both you and the other party could both get something good out of it. The dead-eyed Nigels and Butches of the world miss out on countless juicy opportunities given that the only notion of “compromise” they have available is fight, fight, fight . . . which they’d very much like to avoid, avoid, avoid.
That’s why, after hundreds of millions of years of emotionally rich but masked creatures, some of us evolved to be unmasked. For the first time on Earth, some animals began expressing their emotions rather than keeping them inside.
But they were not expressing their emotions in the sense of, “Hey, you know what? Here’s the emotion I’ve got going on internally right now!” Instead, their emotional expressions are actually moves (we called them responses) in a negotiation. “I should get more, and I’m not kidding around.” That sort of thing.
But with any move we might make, we need to be prepared for the other party to give up on the negotiation and press the “and now we fight” button. And part of being prepared is having the right internal emotion. If I express aggressive, then I’m thinking to myself, What if you respond by throwing down the gauntlet that you’re always carrying around with you for some reason? I’d better be prepared to fight just in case. If I plan on actually getting what I just claimed I should get with my expression of aggressive (namely, more for me), then the internal emotion that’s a pretty darn good one for actually getting more for me in a fight is . . . aggression.
The emotional expressions end up associated with internal emotions of the same name because that’s the sensible emotion for us to have if we need to act.
We’ve Got the Moves
As we worked out in chapters two through four, there are two dimensions of responses or negotiation “moves.” That’s because negotiations themselves have two dimensions: the offer dimension and the tension dimension. This led to the two-dimensional array of emotional expressions in Figure 20, or the lower half of Figure 24, with eight qualitatively distinct emotional expressions.
Four of these emotional expressions come from considering the space in volley-of-offer terms: the horizontal offer dimension is straddled by aggressive and conciliatory, and the vertical tension dimension is straddled by serious and casual. But we can instead view the space in diagonal terms, so that a “move” in a negotiation is made by mentioning my opinions about how confident I am and how confident you are. In this “me so smart and you so stupid” confidence way of considering the space, making claims about my own justifiable level of confidence amounts to a dimension straddled by the emotional expressions pride and humility, and my claims about your confidence amount to a dimension straddled by respect and disdain.
Those first eight emotional expressions are just the eight compass directions of response moves in a negotiation. They are at the center of our compromise-finding machinery.
Bet Your Life
But none of those moves would be at all useful if they weren’t simultaneously poker moves. It’s the poker side of emotional expressions that tends to keep the parties somewhat honest.
As we saw in chapter six, each of these emotional expressions concerns bets. Horizontal shifts are bets (shifts in my favor) or unbets (shifts in your favor), and vertical shifts dial up or down the degree to which these shifts are bets. And they’re bets because we’re social: folks in the community keep track of whether I have been right or wrong and how disagreeable I have been. The Poker Game of Life undergirding honesty depends on simple, run-of-the-mill, much-complained-about gossip. (So, now you’ve got a perfect response when your friends tell you to stop being such a gossip.)
The fact that betting is happening at all gives the emotional expressions their confidence-related meaning in the first place. As we’ve seen, emotional expressions are built out of the sole ingredient of confidence (along with “me” and “you”), and they get their confidence meaning because they’re a bet. That’s all a bet can convey.
Were social animals able to mind meld and be certain of the other party’s negotiation response move, that two-dimensional space and its eight compass-directed emotional expressions would be the whole enchilada. There’d be no need for any other emotional expressions.
Here’s Your Receipt
But, alas, mind melds were not an option back then (or now, for that matter). So social animals needed to be able to confirm that they’d gotten the intended message via what we called an acknowledgment expression in chapter five.
My total emotional expression is the combination of an acknowledgment of the response move you just made, and a response of my own (my own move). It’s through these “you said that, but I say this” salvos that the parties become temporarily tangled with one another in a negotiation tango.
This led to a second two-dimensional space as shown in the outer ring of Figure 22. For each response move (in the inner part of Figure 22), there’s some acknowledgment that says, “You just made that response move.” We could think of the acknowledgment space as a second ply over the top of the response space (see Figure 24).
It might not have initially seemed that a simple expression of an acknowledgment could lead to an interesting and meaningful emotional expression. But—boy oh boy!—the resultant emotional expressions were readily identifiable and among the greatest hits in our emotional expressive repertoire. The horizontal dimension for acknowledgments has our favorite hit of all time, happy, although its opposite end has our least favorite, unhappy. My expression of these serves to acknowledge receipt of your expression of, respectively, conciliatory and aggressive. And the vertical dimension for acknowledgments, which concerns confirming receipt that you got serious or casual, is the home of the emotional expressions, respectively, surprised and relaxed.
And with that we had two two-dimensional spaces. In terms of the compass directions, that meant we had eight emotional expressions in each space, for a total of sixteen emotional expressions (as in Figure 24).
The Whole Enchilada
Was that the whole enchilada? Not at all. As we saw in chapter five, any of the emotional expressions in one space can be combined with any of those in the other space. Then—bang! klonk! zam!—we’re overflowing in emotional expressions, eighty-one in all! It’s in these combinations where we find the myriad of celebrity emotional expressions we all know and love, like angry (close to unhappy and proud), thankful (close to happy and serious), and many more.
Figures 29 and 30 served as Rosetta Stones, providing a summary of the full space of emotional expressions emanating from our theory and their meaning in negotiation terms. This is what is universal across all social animals. Plus, it all emanates simply from first principles; i.e., it’s something we can derive from fundamental premises.
Where Are the Laboratory Experiments?
Tests of grand unifying theories like ours are not done via laboratory experiments. Grand unifying theories make predictions about large-scale patterns found across phenomena, in this case across the nature of emotional expressions, and that’s not something one can get at in the lab at all. How then does one test theories of this kind?
Imagine some really elegant grand unifying theory of English grammar, i.e., about the right and wrong way to say things in English. For any possible thing you might want to convey, the theory predicts the grammatically correct English sentence that would express it. How would one go about testing this theory about English grammar?
Native speakers of a language have very strong agreement on what is grammatically correct. As such, you can go a long way toward testing such a hypothesis while sitting in an armchair (the proverbial “armchair linguist”!). You don’t need to do some careful experiment with hundreds of subjects to figure out whether “I is going to the store” is grammatically correct. You just have to check, proposition after proposition, whether the supposedly grammatical English sentence that expresses that proposition is indeed grammatical. In an afternoon you could thereby test thousands of the theory’s predictions. You might also go on to do empirical tests across corpora of sentences found in books or newspapers as representative examples of grammatical speech, and check if they match what the theory predicts is grammatically correct. There are a variety of ways to test a theory of this sort. However, few to none of them will involve human subjects or stereotypical science experiments.
The grand theory of emotional expressions we’ve built in this book is very similar. It makes a grand suite of predictions: emotional expressions should have these particular four dimensions of meanings related to confidence; there are two two-dimensional spaces (two plies) layered on the same fundamental two-dimensional space; there are eighty-one qualitatively distinct emotional expressions that come in opponent pairs and have certain particular meanings related to negotiation; each emotional expression in the repertoire has some bets of reputation associated with it; each expression also raises or lowers the level of seriousness of the discussion, which is what modulates the extent to which the subsequent move by the other party has to make a large bet of reputation; and lots, lots more! When you pile all these up, they amount to many hundreds of predictions about emotional expressions as we expect to find them for social animals. Viewed another way, it creates one elegant but really, really big, peculiar, and tangled prediction concerning the nature and structure of emotional expressions.
The testing comes in showing that our real human emotional expression repertoire does seem to fit these wide-ranging predictions. Just like native speakers with English grammar, we’re all experts at emotional expressions and how they “work” in real life. Therefore, we can, at each stage of “testing,” just ask ourselves whether the predicted nature of emotional expressions from our theory matches the nature of our actual emotional expressions. For example, does an expression of pride actually amount, as the theory predicts, to asking for more, raising the level of seriousness, putting some of my social capital on the line, and making it more costly in terms of social capital for you to counter?
That’s what we’ve been working hard to show you throughout the book. We haven’t only been developing the theory, but also checking our shared wisdom on the actual nature of emotional expressions all along. The predicted theoretical space, structure, and behavior of emotional expressions emanates from the demands of a negotiation signaling system, and we have seen that it seems to really nicely fit the space, structure, and behavior of the actual emotional expressions we know and love.
Not All Sunshine and Roses
Quite a happy ending for social animals: natural selection gave them the emotional expression system they needed to “do” negotiation, avoid fights, and get more of nature’s bounty, including reaping the benefits of cooperation.
But the acquisition of this first language wasn’t all sunshine and roses. Once unmasked, there were suddenly lots of new forces to contend with. Our wills really do push, and with everyone’s masks off, everyone’s wills were reaching out and pushing onto the wills of others. And these pushes were real. Sure, they’re not physical, but they’re forces as real as the force your friendly neighborhood Mafia representative (remember Vito, from chapter three?) employs to get you to pay up without him ever lifting a finger.
But it’s still totally worth it! Why? Because most of us prefer to be buffeted around by a thousand psychosocial forces rather than endure the pain of physical force.
In switching to this new compromise-heavy way of life, however, there were profound changes in what it meant to be intelligent.
True Intelligence
When we get analytical and talk about intelligence, we have a tendency to mean it in the chess-playing or scientific-discovery sense. Intelligence in these contexts is about Kasparov and Einstein, logic and calculus, IQ tests and brain puzzles, and Spock and Data from Star Trek. Surely that’s part of what we mean by “intelligent.”
But that’s not what we always mean. After all, most of our lives aren’t spent on the nerdy mathy stuff. Most of the time we are—and our ancestors were—simply dealing with each other, chatting, politicking, scheming, cooperating, befriending, competing, vying, and lying. Mostly we’re maneuvering our way through the social landscape, ensuring our status, share, place, or love.
When we say someone is smart—and especially in normal, everyday conversation—we often mean they are witty, funny, sharp, clever, entertaining, political, cagey, crafty, shrewd, sly, wary, wily, devious, sneaky, circumspect, cunning, tricky, biting, stinging, confident, popular, or a blast. These “smart”-associated terms have social implications . . . about how smartly we control and conduct ourselves among all the complex cooperative and competitive interactions with family, friends, enemies, and strangers. Our everyday use of “intelligent” usually concerns social intelligence.
True Unintelligence
In fact, the association between intelligence and social intelligence is even more apparent when we look at the connotations of words we use in English for not being intelligent. Consider annoying, grating, asinine, dopey, foolish, goosey, jerky, blockheaded, boneheaded, loggerheaded, thick, cloddish, doltish, dense, dim, dull, dumb, obtuse, slow, lumpish, unthinking, soft-witted, yokel-like, idiot, imbecile, nitwit, pinhead, simpleton, tomfool, and twit. Many of these could be applied to someone unintelligent in the analytical, IQ sense.
But let yourself imagine the sort of person these “stupid”-related terms naturally bring to mind. You likely won’t find yourself picturing just some socially adept person who can’t do math or chess. Rather, you’ll imagine someone who may or may not be good at math or chess, and may or may not have a high IQ, but is something of a social klutz—the one who bungles normal interpersonal protocol, says the wrong things at the wrong time, and gets picked last to be on a team not because they’re bad at sports (which they probably are), but because they’re kind of annoying to be around.
The android character Data from Star Trek may be a genius in the computational, nerdy sense, but, really, most of us would treat him as an idiot, albeit a savant. His social idiocy is his idiocy. That’s actually what makes characters like Data interesting—he has what we often claim is intelligence, but in every episode, he displays in some novel fashion that it’s not intelligence after all, or at least not the kind we really care about.
Data is the ultimate artificial intelligence success story. Yet it’s a resounding failure. And that’s because the “I” in “AI” has been aiming at the wrong thing! Artificial intelligence as we really want it won’t be of the nerdy, brain-puzzle kind. Real AI—with intelligence as we really mean it in our everyday, unreflective sense—will dominate you, buy your house for a steal, and get the girl. Real AI will elbow its way to be homecoming queen. Real AI will know how to handle itself in the toughest, most important situations we have to cope with: socio-competitive interactions or dealing with other people. When AI does that, we won’t be able to ignore it.
Outer Intelligence
And notice that we don’t care a bit about whether our AI feels anything inside. What we’d like from our AI is for it to get all our subtle emotional expressions and have the same itself.
Does Alexa (an AI device contemporary with this book) have internal emotions?
Don’t know, don’t care!
We want her to talk to us like social animals know how to do. We want her not to act like a hopelessly logical and an utterly daft dolt.
Recall that while emotional expressions are in some way associated with internal emotions, there’s a wide chasm between expressions and the stuff on the inside. An animal whose emotional expressions were a perfectly transparent view into their internal emotions wouldn’t be socially smart at all. This level of utter transparency would lack social caginess. The socially smart creature expresses what makes the most sense for it to express, whether or not there’s any internal emotion corresponding to it.
The “Social” Inside Intelligence
In fact, consider yet again the names for “smart” we discussed above. Not only are these words associated with social intelligence, but, when you imagine such a person, you think of someone with good emotional control (e.g., political, cagey, crafty, and shrewd).
The words for “stupid” (e.g., foolish, goosey, jerky, blockheaded, boneheaded, etc.) bring to mind someone who has a poor ability to put on the socially appropriate—ahem, smart—emotion signal . . . an inability to intelligently use the social language of emotional expressions. You can almost picture the stupid expression on their face.
Those who are socially smarter are better able to control the connection between emotional expressions and internal emotions. They’re playing poker rather than trying to be scientists reporting their experiments.
Intelligence Isn’t What You Think
It may initially seem strange that knowing how to navigate the wills of people in your social community can trump the more analytical, tried-and-true, IQ-related notion of intelligence. But this observation isn’t even unique to social intelligence.
Let’s entertain the following question: Who’s smarter, a house cat or a squirrel? A cat, right?
One might defend that answer in one of several possible ways. Maybe cats do better in scientifically controlled cognition tasks, like, say, anticipating what comes next and getting some reward. Or perhaps cats are easier to train than squirrels. Or maybe feline brains are bigger relative to their body size. Or—and this only works for cat lovers—just look into a cat’s eyes and you’ll see a deeper, wiser soul!
In those ways, cats really are smarter than squirrels, who are, after all, just tree rats.
Fine. But now let’s make the battle more interesting, albeit not necessarily fair. Let’s suppose we’re interested in who is smarter—house cats or squirrels—when in the trees. Imagine for fun you give your house cat (or that annoyingly pretentious Siamese your neighbor owns) no choice but to live permanently in the canopy, among the squirrels. “Who’s smarter now, my Felis catus friend?” asks Sciurus carolinensis.
In the battle for which is smartest within the trees, squirrels have a talent that trumps cats’ supposed greater IQ. That talent is simply their ability to negotiate tree branches. But because of this not-particularly-brain-oriented talent, for all practical purposes, squirrels are geniuses, and cats utter dunces—in the trees.
To help capture the intuition that squirrels really are smarter than cats in the trees, simply ask yourself, “Who looks smarter in the trees?” Don’t think about it too much, or you’ll start to fall into our collective mistake of imposing IQ-style intelligence on everything. Forget about cognitive tests, trainability, brain size, soul depth, and anything else you might presume underlies “intelligence.” Just go on the basis of appearances—exactly what your mom told you not to do if you’re a good person.
House cats can negotiate trees, a little bit. But they’ll never catch a squirrel in a tree. And they’ll never dance around and play chase across the branches and from tree to tree. A house cat is likely to get up there, but it’s just as likely to meow like a newborn kitten for hours once up there, asking you to come to its rescue. No firefighter has ever been called to help a squirrel down out of a tree.
Squirrels run circles around cats in the trees, and any advantages a cat might have in solving cognitive tasks just don’t matter. “Dendrite is to neuron as branch is to what?” asks the IQ test examiner. “Don’t know, don’t care!” says the squirrel, running off to partake in the afternoon pestering and tail biting of the “hopelessly idiotic” Siamese cat meowing in vain the puzzle’s answer, “Tree. TREE!” In the trees, squirrels are more agile, quick, coordinated, dexterous, you name it..
But are squirrels smarter?
We could insist that, no, cats are still smarter. But on what basis? That their brains when on the ground seem more capable? That just violates the game we’re playing at the moment: that they’re all confined to living in the trees. For life within the trees, house cats lose the battle of smarts to the squirrels.
Oaf
Cognitive intelligence of course still matters. Among the squirrels, those with greater IQs may do better, all things equal. Among the house cats, those who do better on standardized tests may excel. But, as a class, all the squirrels are smarter within the trees than all those schmucks stuck with the bodies and brains of vainglorious house cats.
If this sounds like a stretch in the use of “intelligent,” consider that we even speak this way about people. Klutz, bumbling, clumsy, butterfingers, oaf, blundering, clod, gawky, galoot, ham-handed, dunderhead, all thumbs, lumpish, unwieldy, loutish, clown, ungainly, gangly, unhandy—although these monikers point more toward motor behavior, they have strong connotations for “intelligence” as used in the everyday, non-reflective sense.
Whatever intelligence is, it’s much more than just the cognitive stuff. It’s behavioral, social, and physical. After all, what makes squirrels smart in trees is in large part in their hands, claws, grip design, body shape, and tail, not in the brain.
At My Fingertips
To help further blur the distinction between brain and body as the source of “intelligence,” consider a case concerning grasping, one that’s very close to home for us humans and some other primates. One of us showed that the reason we get pruney fingers when wet is not some accidental (osmotic) side effect of interaction with water. Instead, the shape of the wrinkle patterns on our fingers when they get pruney is exactly the peculiar shape expected if they are designed to channel out water efficiently in wet conditions, to avoid slipping (or hydroplaning). Our pruney fingers are rain treads!
So, one can imagine two ways of building animals for dealing with grip in wet conditions.
The first kind of animal has a fixed body, and what makes some animals of that kind smarter than others in wet conditions is their more clever behavioral software for employing that body. When it’s wet, the smarter-in-wet-conditions creature implements the right sorts of behaviors (how to grasp, which places to grasp, etc.), ones that minimize the risks of slipping.
But an alternative is to let the brain and behavioral software be constant, and instead let the physical body change its form when it’s wet, transformer style. For primate—and our human—grip, we change our grip morphology itself, acquiring rain treads when it’s rainy or dewy outside, but otherwise being smooth for dry conditions (akin to smooth racing car tires on speedways).
For either of these alternatives, the animal is able to handle wet conditions. Animals get high “wet intelligence” either way. In the first case, because the animal is calling up wet-condition behavioral software, we have an inclination to call the animal “wet smart” (relative to an animal not able to handle wet conditions). But in the second case it seems a little strange to call the animal “wet smart,” because there’s nothing in its brain undergirding this capability. It’s all out at the periphery, in the “tires.”
This, however, is a sort of prejudice. It’s body prejudice!
So, yes, the squirrels are smarter than house cats in trees due, in large part, to non-brainy morphological features (like claws, body size, long twirly tails, etc.). But that’s okay. “Smarter is as smarter does,” cry the squirrels to the lumbering house cats droning on and on about their enormous, convoluted brains.
Intelligently Negotiating One’s Habitat
With squirrels in mind, can we get more specific about what makes them so tree smart? It’s how well optimized each of their steps from branch to branch is. A squirrel that is fine-tuned to step in one spot so it can then launch to the next can bound relatively effortlessly through its habitat. It’s through these highly evolved “tree steps” that squirrels negotiate a tree branch and navigate their way through the canopy.
So it’s not at all surprising to suppose that social animals can be smart by virtue of specialized mechanisms for optimizing their steps through their social habitat, not via their cognitive smarts, per se. What if a “step” for a social animal is a social interaction with another animal? Emotional expressions are moves for more efficient living in our chosen environments. That’s ultimately what we’ve been outlining and building toward in these pages. Squirrels negotiate tree branches, and social animals negotiate the wills of the other social creatures in their community.
As we have seen, each of these “social steps” that social animals take is undergirded by emotional expressions, and the adept use of these expressions is what allows a socially smart creature to more ably navigate the social milieu.
Cool Intelligence
Not only are the socially smart brilliant at the Poker Game of Life, but more often than not they’re raking in social capital. The socially smart usually are the high-reputation ones. In one common interpretation of “cool,” the socially smart ones are cool. In fact, look again at those many terms for “smart”: witty, funny, sharp, clever, entertaining, political, cagey, crafty, shrewd, sly, etc. Most or all of these words are intuitively associated with cool. You kind of want to hang out with such a person, right? And for the words for “stupid” earlier, look at them again: annoying, grating, asinine, dopey, foolish, goosey, jerky, blockheaded, boneheaded, loggerheaded, etc. These sound accordingly uncool. Definitely low rep.
Being cool ain’t easy. It requires getting a hundred things right within a social environment of judgmental lookie-loos. “Getting things right.” Now that sounds like what intelligence is all about. But if the intelligence that truly matters is social intelligence, can social intelligence capture the intuition that it’s about “getting things right”?
When folks usually think about social intelligence, they’re not inclined to imagine “getting things right.” But we’ve seen that emotional expressions are ultimately all about getting things right.
From its very beginnings in chapter two, it was two beliefs about the truth at opposite ends of a compromise axis “battlefield.” And the entire compromise-finding signaling system we derived from first principles is about who is right . . . or who is more right. Social animals who tend to be more correct end up accumulating greater piles of social currency over their many interactions.
Those with a better reputation (more social capital) are the ones you can usually rely on. When they make claims, they’re more likely to be correct. And they’ve received that social capital in some large part by virtue of large numbers of social interactions, each a disagreement that they settled via a negotiation mediated by emotional expressions that led to either agreement or to a fight if one party, in the poker sense, called.
The Perception of Getting Things Right
Social intelligence is about getting things right, about being correct about claims, about being correct among the people.
But while social intelligence is all about being correct, that’s not quite . . . correct.
A more accurate way of describing it would be that social intelligence is all about being believed to be correct. If I have acquired a lot of social capital over time by virtue of making correct claims, then people believe I tend to be correct because I do tend to be correct.
But there are other, darker, ways to accumulate social capital through disagreements—and subsequently to be believed to be someone who tends to be correct—without actually having to be correct.
Bluff
Remember that the emotional expression system we social animals use is a variant of poker. One common way to get your opponent to fold is to bluff, to bet so much that your opponent doesn’t wish to risk matching your bet, and so folds. A fold amounts to agreement, whether he truly believes your hand is stronger or not. In an emotional discussion, a bluff is to disagree a lot, which thereby corresponds to a large bet of social capital.
For example, imagine after Tim asks for more zucchini bread, Mark replies, “How dare you? I should get twice what I last asked for!” Mark’s expression of pride might be just enough to get Tim to back down, judging it’s not worth making a scene and looking ridiculous. And that might be true even if Tim is 99.9 percent sure that Mark is bluffing!
The world has its share of bluffers and blowhards who get their way—and raise their social capital—not by being correct, but by getting their opponent to back down by placing a large bet.
And it’s easier to be a blowhard if you’re rich . . . rich in social capital, that is. If you’re already very high rep, then in nearly any disagreement you’re able to bet much more than others are willing to. A small bet from your point of view might require going all in for someone else, and they’re going to immediately agree with you if you are that disagreeable. They simply can’t afford to enter an argument with you. This happens especially often in conversations with your boss or your company’s CEO. In poker a person with a giant pile of chips is called the “chip bully”—having disproportionately more chips than the others gives you the power to bully them into agreement (i.e., folding).
But some folks are so good at the poker side of social discussions that they can bluff their way through life even before they are rich in social capital. They have a better than usual understanding of poker in their bones and can talk their way into—or out of—anything. They’re the “salesperson.” They know how to use their wit and charm to get their way.
Reverse Bluff
Another strategy in poker is to reverse bluff. The idea here is to make it look like you’re bluffing, even though you’re not. It’s a tricky maneuver, but if you can carry it off, you can coax the other party to bet a lot, and you’ll thereby transfer more of his social capital into your reputation stack.
This isn’t quite the same thing as a pool shark, but it’s close. Pool sharks play a round of pool with you and purposely look weak enough that you can probably beat them. They then suggest another game, at which time you’re willing to bet a lot—you’re obviously better, so why not? But the pool shark snookered you, and he promptly pockets his balls without you even getting a chance to lift your pool stick.
Another example of a reverse bluff might happen at a party where you meet a stranger and somehow end up talking about your yellow-naped Amazon parrot. “And he’s from the Amazon basin,” you add.
“Are you sure that species is from there?” he politely and casually asks.
“Uh-huh,” you haphazardly follow up.
“Because I think they might not be from the Amazon basin,” he adds, nonchalantly.
Now you’re thinking to yourself, I’m the one who has owned an Amazon parrot all these years! So, of course he’s from the Amazon basin. “Yeah, pretty sure my Amazon parrot is from the Amazon basin,” you reply, now somewhat loudly, capturing the attention of half the party. And can we move on, please, you say to yourself.
“Ah. I see,” he says.
Finally! You smile a little too smugly and hope to go grab another drink.
But as you begin to move, this annoying guy adds, “Because I had learned that Amazon parrots, despite their name, are actually almost entirely from the Mixed Mestizo region.”
What on earth is the “Mixed Mestizo” region? And who is this guy? Sounds like the typical pretentious know-it-all who shows up at these west-side parties. I doubt he even knows what the Amazon basin even means.
“You do know where the Amazon basin is, right?” And you hit just the right disdainful intonation that you can almost hear the partygoers inwardly crying out, Oooooh! No, he did not just ask that!
To that, the party guy calmly says, “When I was tracking neotropical parrots for my PhD, I was amazed to find that fully ninety-seven percent of Amazon parrots weren’t in the Amazon basin at all. Had an entire section on it in my dissertation.”
The entire room at this point is quiet. Everyone can almost hear the sound of you deflating like a popped balloon. You were snookered. What kind of jerk lets me stumble into a conversation about his dissertation topic and allows me to babble on without immediately mentioning it?!
Poker Personalities
So, that’s also part of the complex mix of poker strategies socially smart people use in real life to accumulate social capital and rise through the ranks. He feigned humility, hoping to trick you into expressing considerable pride and disdain, and accordingly placing considerable social capital on the line. You took the bait, and he could finish you off with one solid expression of pride, at which point you had no choice but to fold and leave all your pride on the table, so to speak.
Bluffers, bullies, salespeople, sharks, and surely hundreds of other “poker personality types” are out there.
And all these types, all these emotional expressions, all these bets and unbets, all these negotiation moves—they’re not human per se. They’re a social animal thing. The emotional expressive system we have derived is a general, first-principles consequence of the demands of languageless, potentially lying social creatures in need of the simplest signaling system for negotiating a compromise.
We therefore expect that the fundamental emotion language—and software used for playing it well—would be found among any social animals in any galaxy anywhere, not only among the social animals here on Earth or our rock band paraphernalia-manufacturing alien business associates.
Stats for (Brilliant) Dummies
Imagine writing a careful scientific paper that makes some conclusion. Maybe it concludes that “African elephants are larger than Indian elephants.” If your paper’s conclusion was just that, and nothing else, it would have almost zero value.
What we really want to know is: How confident are you in your claim that African elephants are larger than Indian elephants? You wrote a scientific paper, so presumably there are data, and those data lead to some particular quantitative level of confidence. So tell us! Any worthwhile paper on elephant size differences will not just state the conclusion but state the confidence in that conclusion. “African elephants are larger than Indian elephants, and our data make us this statistically confident.”
Without confidence claims, the scientific literature would be a massive pile of uselessness.
And that helps illustrate just how important our emotional expressions are.
Emotional expressions are just our evolved way to signal confidence in our claims. Our ancestors were not statisticians. We didn’t gather data in Excel files that we could carefully analyze. Yet, our brains did gather data and evidence over time. I know stuff you don’t. And the stuff I know does give me some level of confidence in my beliefs. I may not have any idea how to consciously compute confidences from my conscious memory of my experiences, but my big, social-creature brain can nevertheless do it, and do it without requiring any conscious thought at all on my part.
Emotional expressions tap into our inner unconscious statistician. They allow us to communicate statistical confidence in the only way other social creatures can actually comprehend it. If you want to communicate your level of confidence, the best way to do it is using the language that we’ve evolved for millions of years for communicating stats: the language of emotional expressions. That’s how you get your claims—and your confidence in them—into the minds of others. Express your emotions, not your stats.
Rabble-Rousers
Just as there’s no market for movies that are all words and no emotional expressions, there are also no orators who speak like robots and nevertheless rouse the crowd. Why should the crowd be roused, after all, given that the speaker never communicated whether she’s confident about the claims she’s making? Maybe she’s actually meaning to say, “. . . and I’m not particularly sure about everything I’ve thus far said.” In the absence of any emotional tinge to the speech, there’s often no way to know.
An ever-so-slightly better speaker might maintain the same flat speech but throw in some uses of “I am certain that” and “I have little idea if it’s the case that,” and perhaps even some statistical gobbledygook. She’d at least be conveying the confidences that are desperately needed. The audience might find it informative, but no one’s very motivated to do much about it.
If, for example, the speaker is informing the audience about how the mayor has been spending all the tax revenue on overpriced whiskey, the audience is only likely to rush out with pitchforks if someone in the back suddenly stands up and yells, “That’s our whiskey, and—let there be no doubt—we’re gonna get it!” And imagine this sentence is dripping in emotion as you would expect it to be given that whiskey is at stake, something we can’t actually put in printed words. Now the crowd is on its feet. If that fellow in the back is so confident, well, then we’ll surely be able to get through the moat, over the electric fence, and past the Dobermans to the fortified cellar! Let’s go!
Carefully titrated words and numbers won’t get the crowd moving, because even if the crowd actually comprehends the meaning, the words and numbers don’t truly get into their inner unconscious computational system designed for dealing with confidences.
Leaders understand this in their bones. You can’t be some idealized, measured, square philosopher type and have anyone actually follow you. And you can’t communicate your confidence levels as a scientist would—that just doesn’t sink into our brains, which were never designed for language or statistics.
You communicate confidence the way we’re supposed to, the way our brains are expecting to hear it: via emotional expressions.
Free Emotional Expression
When we think about free speech or freedom of expression, most of us tend to imagine that the crucial bit is speech. People need to be free to say the sentences that need to be said.
But, as is beginning to hopefully be clear, we need to get over our love affair with sentences at the expense of the emotional side. As we’ve been discussing, sentences without confidences aren’t worth the paper they’re printed on. The confidences are what matter.
And it’s our emotional expressions that are our way of doing exactly that. It’s free expression—free emotional expression—that is the most important facet of free speech. Without the emotional expressive side, reputation isn’t put at stake. And if reputation isn’t put at stake, the pressure for honesty goes away.
And, furthermore, reputation itself goes away. The mechanisms underlying social networks depend on these transfers of reputation currency, and that’s the stuff of emotional expressions, not non-emotional speech.
So, if someone asks you to tone it down, it isn’t innocuous. The emotional side is crucial to the speech. If they censor your tone, they censor you, plain and simple.
Ignominious Anonymity?
The deficiencies of sentences, the passion of oration, and social life as a series of efficient “social steps” in an environment (like a squirrel’s facility in a canopy)—we’ve introduced a few new ideas here as we draw toward the end of our time together. They all lead us to our last examination: how emotional expressions function or fail to function in a social environment that would have been completely foreign to our ancestors—the digital sphere of online social networks.
In order for any social network (whether in person or online) to function well, individuals must be able to negotiate using emotional expressions, and they must be able to remember one another. The online social networks we all participate in—like Twitter, Facebook, and others—are notoriously poorly functioning. Folks often act like “trolls,” i.e., they behave like jerks. What exactly is wrong with social media that our behavior is so unbecoming compared to real life?
One common hypothesis is that it’s because many people on social media are anonymous. Their Twitter handle is a fake name, and no one knows who they really are. Shouldn’t that completely undermine the very possibility of a social network behaving properly? How can folks accumulate reputation, or lose it, if there are no determinate folks?
That, however, is a false intuition.
Anonymous folk may be anonymous in the sense that no one knows who they are in real life, but within the social media’s network, they’re a real entity, one that’s “in the game,” rising and falling in reputation just like all the non-anonymous accounts. The owner of an anonymous account might be anonymous, but if the account has grown to a million followers, the owner is just as interested as anyone with a million followers in making sure his or her account is respected, and accordingly tends to be no less (or not much less) well behaved than a non-anonymous account.
And even most of the supposedly non-anonymous accounts might as well be anonymous; nearly all of them are “mere names” to you, and it would make little difference if many or most are in fact fake names.
Not only are anonymous folks on social media generally behaving about as nicely as the non-anonymous, but anonymous folks in real life also tend to behave. Sure, the coffee barista or gas station attendant has a name and identity, but to you they get the label of “coffee barista dude with the sideburns” and “redhead chain-smoking gas station attendant,” and these “names” have as little connection to their non-work life as does a Twitter anonymous account to the owner’s non-Twitter life.
Not only does anonymity not tend to lead to worse behavior, but not being anonymous doesn’t guarantee civility. Facebook is a great example of this. Your connections on Facebook tend to be disproportionately family, friends, and friends of friends. They’re real, and you are usually fairly certain who they are. Yet, despite the strong non-anonymity on Facebook, folks behave notoriously poorly. Disagreements in the comment section of posts tend to be absolutely venomous, and even worse than on Twitter, and that’s saying something!
Anonymity isn’t what makes social media such a toxic environment.
Emotionless Social Media
If the troll-like behavior on social media isn’t because of anonymity, what is the reason for it?
There are actually a number of reasons why social media fails to give us the functioning reputation networks that we would have typically found in our ancestral environments. For one thing, rather than the dozens or hundreds of individuals in our social network, social media has millions. That alone is a difficult scaling problem, and it surely underlies some of the failure of social media.
But a more basic reason for why social media is not well behaved is that one cannot properly emotionally express oneself on social media. In real life, whether anonymous or not, we can emotionally express ourselves, and—as we have seen—it is those emotional signals that mediate our interaction.
Social media has built for us massive social networks, but it’s not yet allowing us to utilize the very signals we evolved for social interaction. Bigtech social media companies somewhat grasp that their social networks need more emotionally expressive power. That’s why there are likes, dislikes, and in some social media sites the ability to put happy, sad, surprised, or a few other emoticons in reaction to a post.
But those are just the barest shadows of what’s needed. What’s minimally needed is the capability to emotionally express the two-dimensional space of responses. And probably to allow users the ability to acknowledge as well. That would allow the full four-dimensional range of emotional expressions.
Furthermore, to truly harness our human emotional expressions that undergird our transactions of social currency, social media would need to allow parties to negotiate back and forth (via emotional expressions), and either come to agreement or fight about it (which might be decided by, say, a vote among the community). That is, it’s not enough for me to signal any emotional expression at your post. You need to be able to argue back. Preferably this would be done via signals that feel natural to employ and natural to comprehend.
Social Currency Devaluation
But social media fails in another way. Even if social media got all the emotional signaling dialed in, it still won’t “work” if the network isn’t able to properly keep track of the social capital being put on the line via those emotional expressions, and to whom the social capital actually went to once the negotiation is complete (via agreement or via a fight).
In a tribe, gossip is enough to keep track of the rises and falls of reputation over time, and it’s remembered by many or all of the individuals in that tribe. But social media has millions in the network, not hundreds. And someone who lost in a hundred arguments with a hundred different folks, and who ought to have lost loads of social capital, might go totally unnoticed.
When that happens, it means the social currency has lost its value. If a currency lets you spend it, and then spend that same currency again, the community quickly devalues that currency. It’s a joke.
And once the social capital in that network loses its value, the incentives to be right about one’s claims dramatically collapse. That’s another reason why there are so many trolls and so much bad behavior on social media. They’re playing poker with the most plentiful and tasteless potato chips known to man.
For social media to actually work, when an interaction is completed, mechanisms are needed to help record what happened to the bets of reputation. Perhaps it’s explicitly recorded in some manner by an algorithm, and your reputation is shown as a number. Or—better—perhaps the interaction and its outcome is simply made visible to enough relevant people in the network that the community remembers how the reputations changed. That’s, after all, how it works in real, live social networks. We know in our bones the reputations of those in our community; we don’t usually look up reputations in a book.
Social Narratives
A community is a social community, then, only if social capital transactions can occur (which requires emotional expression) and be recorded and preserved. How do real-life social communities actually go about keeping track of the rises and falls of reputations?
As we mentioned earlier, there is no book in which to look up someone’s reputation. And there’s not one person in the community who decides the matter or even has the sole job of keeping track of all the reputations.
Instead, it’s done by roughly everyone.
And the way that folks remember all those swings in social capital is via something called a social narrative, which is something like a story of what happened over the history of the community. But, really, the key purpose of this story is that it helps everyone remember how reputations have changed.
Over the last week, for example, perhaps there were a dozen key changes in reputation due to a dozen fairly dramatic social interactions. The social narrative needs to record those social capital transfers. But rather than simply use a spreadsheet-like ledger, we human social animals do it via gossip.
“Doug was rude to Judy, and—man oh man!—Judy gave a great retort, and Doug had to back down and admit that the ‘wild boar’ he had killed was already dead when he found it. What a loser!”
A social narrative records the truth about what happened, but also how that bears on the reputations of those involved in arguing about the truth. To do that requires that it incorporate into the story the disagreements that occurred, who ended up wrong and right, and how disagreeable the parties were.
A social narrative is actually just a slightly more sophisticated way to talk about gossip. It’s gossip that spreads across the social network and keeps everyone apprised of the social capital fortunes and misfortunes of its members.
As base as gossip might seem, gossip undergirds why social currency works at all. If word doesn’t spread about the dramatic wild boar argument, then the dramatic wild boar incident might as well not have happened. Doug lost nothing; Judy gained nothing.
Social narratives, or gossip, are not only one of the key foundations of social capital—and thus of social networks—but allow social capital to be spent all over the network, and reputations swing without any central accountant to keep track of it all. As we mentioned, there’s no one person in charge of either bestowing reputation or even of recording it (where they could sneakily modify reputations of their friends and enemies).
Instead, social currency can be robustly spent all over the network using a distributed, decentralized system. And that might begin to sound like something else that’s become popular over the last few years: Bitcoin, cryptocurrency, and blockchain technology.
Gossip and Blockchains
There are, in fact, deep similarities between social currency and cryptocurrency. Both have (a) non-physical tokens and (b) no central authority recording interactions. Thus both have to worry about “double spending,” where Mark pays Tim some amount of currency, but because it isn’t recorded appropriately, Mark can spend it again.
For social capital, if I lose reputation in a social interaction with you, that “payment” of some of my reputation to you needs to be veridically recorded by the social network’s collective memory. But how? What if my buddies and I spread gossip that it is you who got humiliated in the argument, rather than me? Which of this week’s gossip gets validated and added to the end of the existing social narrative?
There are two ways this is handled in social networks, and each of these has a clear analogy for cryptocurrency.
1.Reputable Gossip: Higher reputation people tend to have truer gossip, and care more about the integrity of the reputation currency. So, if there are competing accounts for what actually happened this week, everyone votes, but those with greater reputation get bigger votes. That tends to help ensure that what really happened gets recorded as having happened. And “what happened” as much concerns what’s true as who was right. This is analogous to something called “Proof of Stake” within cryptocurrency, where a new “block” of transactions is validated by folks owning a lot of that coin.
2.Parsimonious Gossip: Gossip that most parsimoniously fits the existing tribe narrative (the “tribal blockchain”) is chosen, in some part because it’s easier for people to remember. One might think of the narrative being extended with a new story line elegantly explaining (after the fact) why the new transactions ended up as they did. Finding a highly parsimonious extension to the existing narrative is not easy, but it’s easy for others to see whether it’s a parsimonious extension; that’s key to it being a good, explanatory story or added “gossip block.” The winning gossiper’s story line is the one added to the narrative, and that gossiper gains some reputation for this service. This is analogous to “Proof of Work” in cryptocurrency. (Gossipers are like something called miners.)
A blockchain for cryptocurrency is the “narrative” of the many Bitcoin transactions that have historically occurred, and blockchains are notoriously difficult to change once created. That’s what allows cryptocurrency to be a currency at all. See Figure 36 for a summary of the analogy between social currency and cryptocurrency.
For the same reasons, social narratives are very hard to displace. That’s good in the sense that it has to be that way. If it weren’t, the bets of social capital wouldn’t be bets at all. But it can be bad because false narratives have a tendency to remain in place forever, something that’s deep at the heart of the madness of crowds and mass delusions.
Figure 36. An analogy between blockchains and social narratives.
Truth Stumble
In a sense, then, social media today just isn’t social at all, at least not in the relevant full “social animal” sense. In order to make social media truly social, and consequently not filled with trolls, we must reshape it so that it harnesses our social animal nature. We must inject social media with what it’s missing: (1) a full dose of emotional expressiveness, and (2) mechanisms allowing its members to keep track of the swings in reputation (our blockchain technology, so to speak). Until then, social media will remain broken, and we’ll be stuck with trolls, trolls, trolls.
As bad as trolls might sound, broken social media is much more dangerous. The real loser is truth itself.
You might think that science is about hypotheses and predictions and experiments and controls and analyses and statistics and confidence, but the real “genius” that makes science work is that it’s a decentralized network of scientists. Over time, some scientists rise in reputation by virtue of being able to show true and interesting things. They accordingly get more influence; others are more likely to listen to their claims. And some scientists lower in reputation, either by saying boring things or by repeatedly making claims that are false or incoherent.
At the end of the day, scientists don’t believe other scientists’ claims because they carefully slogged through all the data and did all the analyses. There are too many papers and too little time. We simply believe what they say, in some part because of some reputation they have garnered.
The entire process isn’t done by anyone, or any organization. It’s done in a distributed manner with no one in charge. It’s messy, but it tends to slowly find the truth, even if it sometimes spends years going in the wrong direction. As inefficient as it might be, it’s widely understood to be infinitely preferable to having some central authority determining which claims are true.
The truths that science comes to are just a tiny fraction of the truths that society as a whole comes to, though. The full range of truths depends on decentralized, distributed networks of people, all people, not just scientists. Over time, folks who tend to say true things tend to rise in reputation. That confirms to them that they’re right and encourages them. When social networks function well, they stumble toward the truth. Each of us humans, as social animals trying to avoid truth-determining fights, may be avoiding the truth on any given interaction, but the entire network does the truth finding for us. Inaccurately. Inefficiently. But eventually. It’s the best we have.
But in order for social networks to do their truth-finding work, they have to function properly. That includes allowing users the full range of emotional expressiveness to communicate confidences in the way we evolved to and ensuring that the network can actually keep track of social currency transactions. Without these measures, the currency becomes worthless, and the trolls take over. A network of trolls has no tendency to find the truth.
Furthermore, when social networks fail and lose their tendency to stumble toward the truth, centralized authorities feel more obliged to determine the truth for us. Fact-checkers, government authorities, and so on take it upon themselves to just give us the truth . . . but what they’re giving us is even more unlikely to be the truth and much more likely to be exploited.
Whatever the decentralized social network currently deems as true can sometimes be hit or miss. But its opinion about what is and is not true isn’t under the control of any one person or entity. It’s more like a Ouija board with a million fingers pushing on it—a Ouija board that actually reaches into a realm of discovery and finds truths over time.
Bottoms Up
Before you began this book, you might have deemed emotional expressions to be the stuff of folks who can’t get their emotions under control. Emotional expressions are for non-academic types, for the riffraff, for the non-rigorous and gauche among us. They’re wispy, flaky, pointless, and we’d be better off without this ancient baggage.
But over the course of this book we’ve seen that social animals require a “language” that allows them to engage in negotiations about anything that might come up. That needed language seems to have a whole heck of a lot of similarity to the emotional expressions we actually evolved to possess and have unpacked in this book. In this light, perhaps our emotional expressions are one of the greatest gifts natural selection endowed upon us. Without them, perhaps social life would have been, and would still be, impossible.