A terrible realization is dawning upon the so-called free world. Cherished notions about freedom, embodied in terms such as free markets and free speech, are proving to be dysfunctional. Unless we can refine our notion of freedom, we do not have a model to offer other societies that we regard as less free.

Revelations about Facebook are the latest manifestation of dysfunction. Not only is Facebook lax about regulating the speech that takes place on its gargantuan platform, but the very nature of the platform is the result of an unregulated economic market. If Facebook is being used successfully to steal elections, spread false news, and infringe upon civil liberties, then it is undermining the very fabric of democracy.

Ironically, gossip can help to address some of these problems. You might think that gossip is the problem, if by that word we mean self-serving and often fallacious talk about others. But that’s not how gossip works in small-scale societies around the world. Consider the following example, drawn from an ethnography of the Melanesian island of Lesu[1].

There is much talk in the village because Tsengali’s pig has broken into Murri’s garden. Murri displays no particular anger but Tsengali is much annoyed because of all the talk that the incident has occasioned. So he announces that he will give a pig to Murri to stop the talk. But Murri tells him that this would be foolish “to eat a pig for nothing”. Instead, Murri declares that the incident has ended, and that there should be no talk.

In this example, a tiny rupture in the social order, caused by negligence and not mal-intent, results in an outpouring of gossip. The offender is so eager to salvage his reputation that he voluntarily offers to compensate for the damage that he caused. The person whose garden was damaged could afford to be good natured about it because gossip was providing all of the social control that was needed to correct the situation.

We don’t need to travel to Melanesia for examples such as this. Here is a description of California cattle ranchers drawn from a book titled Order Without Law by Yale Law School Professor Robert Ellickson[2]:

The mildest form of self-help is truthful negative gossip. This usually works because only the extreme deviants are immune from the general obsession with neighborliness….People tend to know one another, and they value their reputations in the community. Some ranching families have lived in the area for several generations and include members who plan to stay indefinitely. Members of these families seem particularly intent on maintaining their reputations as good neighbors. Should one of them not promptly and courteously retrieve a stray, he might fear that any resulting gossip would permanently besmirch the family name.

This is how gossip has operated for our entire history as a species. In fact, our very capacity for language evolved in the context of small groups whose members were able to hold each other’s self-serving impulses in check[3]. Three ingredients are required for gossip to function as an effective regulator of behaviors deemed appropriate by a group.

  1. There must be shared norms of what counts as appropriate behavior.
  2. It must be possible to detect transgressions.
  3. It must be possible to escalate punishment if an offender does not rectify his or her behavior.

When these ingredients are present, the regulation of behavior takes place so spontaneously that we don’t even think of it as regulation. A formal legal apparatus isn’t required, which is why Ellickson titled his book “Order without Law”.

Nearly twenty years ago, I conducted a series of experiments on gossip involving responses to fictional scenarios[4]. Take a look at this story about two students disappointed with their exam grades:

Jane and Susan are waiting outside their biology class for the final grades to be posted. They have been best friends since high school. Both are hard-working students, well liked and trusted by their friends. They take their classes very seriously and each are working part-time jobs to supplement their academic scholarships. The grade in this class is particularly important, because the medical schools they have applied to have high standards. When the grades are posted, they see that they have just missed the cut-off for a grade that the schools find acceptable.

This story was constructed to portray both students as solid citizens and the outcome of their grades as consequential. It was followed by two versions of a short dialogue.

Version A

Jane: “This would be easier to take if I didn’t know that a group of students cheated.”

Susan: “Really? What do you mean?”

Jane: “They asked me if I wanted to join them! They stole a copy of the exam from the office the night before!”

Version B

Jane: “I bet we would have gotten a better grade if we sucked up to the professor the way those students in the front row did.”

Susan: “Really? What do you mean?”

Jane: “Oh, you know—always asking questions and pretending that they’re interested.”

In version A, Jane is gossiping about clearly deviant behavior in an apparently truthful manner. In version B, she is griping in a self-serving manner. Participants in the experiment read one version of the story and were asked to rate their approval of Jane, Susan, and the other students to whom Jane was referring on a scale from -1 (highly negative) to + 1 (highly positive). In the first version, the cheating students were evaluated negatively but Jane and Susan were evaluated neutrally. In the second version, it was Jane who took the reputational hit. These results suggest that it is acceptable to gossip in a truthful manner about a norm violation--but gossiping in a self-serving manner is itself a norm violation.

In a second experiment, the story was based on Ellickson’s scenario, in which the cattle of one rancher named Jim graze on the land of another rancher named Tom. In one version of the story, Jim is portrayed as irresponsible and Tom gossips about it to other ranchers in a truthful fashion. In a second version, Tom adds a little gratuitous negative gossip by calling Jim a drunkard. In a third version, Tom gossips with Jim present in the room rather than absent. In a fourth version, Tom elects to remain silent rather than gossiping. In a fifth version, Jim is portrayed as a responsible person who retrieves his cattle as soon as possible and Tom is portrayed as self-serving by misrepresenting the situation for his own gain.

The results were exactly as we predicted. Tom’s approval rating was mildly positive for gossiping in a truthful manner but he took a reputational hit for calling Jim a drunkard, for failing to gossip about a norm violation, and especially for gossiping in a self-serving fashion. The response to gossiping with Jim present in the room was ambivalent. Some of the raters thought that this was commendable, while others thought that it was needlessly confrontational and that it was better to gossip in the person’s absence.

In a third experiment, the story involved two teaching assistants named Pat and Adrian discussing a possible cheating event while proctoring an exam. In four versions, the person witnessing the event was 1) Pat (an eyewitness account), 2) a single trusted friend of Pat, 3) two students in the class who Pat did not know well, and 4) one student in the class who Pat did not know well. Participants in the experiment read the story and were asked to rate the credibility of the information. As predicted, the credibility ratings were ranked in the order 1>2>3>4. In other words, the raters were sensitive to the quality of the information. As part of the same experiment, we changed the story to be about someone observing the professor of the class coming to work with his pants inside out. As before, the quality of the information was varied from an eyewitness account to a single person who was not known well. In this case, however, the raters were not sensitive to the quality of the information, presumably because the consequences were not as important as identifying a cheater on an exam.

Several years after these experiments were conducted, my graduate student Kevin Kniffin (now a professor at Cornell University’s Dyson School of Business) had an opportunity to study gossip in a real-world situation[5]. He was conducting an ethnography of our university’s crew team and one of its members was proving to be a slacker, failing to show up for workouts and otherwise shirking his duties. There was an outpouring of negative talk about him, just as described for small-scale societies around the world. In addition, there was an outpouring of positive talk about the most exemplary members of the team, something that is often missed in discussions about gossip. Taken together, the positive and negative talk accentuated what counted as praiseworthy and contemptible behavior within the group. After the slacker quit the team, casual talk centered on neutral topics such as current events, joking around, and the movies.

These studies reveal that our use of speech in small group settings is highly moral, sophisticated, and sensitive to context. Not only do we hold each other to agreed-upon norms, but we also filter information for its accuracy when it is important to do so. The most highly regulated groups in the world are small groups and the speech that freely (=spontaneously) flows from our minds is an essential part of the regulatory machinery.

But that’s a far cry from the way speech operates on the Internet and many other modern settings. This is an example of evolutionary mismatch, which occurs when traits that evolved as adaptations to one environment become maladaptive when expressed in a different environment[6]. Evolutionary mismatches abound in modern life. Other examples include dietary mismatches leading to an obesity epidemic and immune system mismatches leading to an epidemic of inflammatory diseases. Recognizing collective speech pathologies as evolutionary mismatches is a giant step toward curing them.

A mismatch can be cured by identifying the proximate cause of the dysfunction and rectifying it with a further environmental intervention. For example, there is accumulating evidence that eye development results in myopia when people spend too much time indoors with low ambient light levels[7]. The solution is to spend more time outdoors or perhaps develop new forms of indoor lighting. Inflammatory diseases occur in environments that are overly hygienic[8]. The solution is to make sure that we are all inoculated with healthy microbiomes. Children might fail to develop executive functioning and social regulation skills if they are deprived of self-directed play[9]. The solution is to provide more opportunities for self-directed play. Notice that in each of these cases, the cure can be quite simple to implement, once the nature of the mismatch is understood.

Collective speech pathologies can be traced to a disruption of one or more of the three ingredients that were listed at the beginning of this essay.

1) The absence of shared norms of what counts as appropriate behavior.

2) An inability to detect transgressions.

3) An inability to escalate punishment if an offender does not rectify his or her behavior.

Is it possible that we might be able to cure collective speech pathologies afflicting so-called free societies, as easily as we can cure other mismatches? I am optimistic about this prospect. The biggest challenge is to get enough people to see the problem in the right way.

Much of the regulation can even be of the “order without law” variety, without the need for new laws, regulatory bodies, or changes in constitutions. However, governments and other leviathan institutions such as Facebook still have an important role to play in constructing a social environment that allows the three ingredients to be strongly implemented.

One of the most amazing facts about human nature is that almost any behavior can become the norm when it is regarded as good and right by members of a group and reinforced with the panoply of mechanisms that enable us to be such a highly cooperative species--including norms that reinforce the cardinal virtues associated with free societies. Protecting these virtues should not be confused with undermining the mechanisms that make all forms of human cooperation possible.

References
[1] Powdermaker, H. (1933). Life in Lesu: The study of a Melanesian society in New Ireland. Foreword by Dr. Clark Wissler. New York: Norton.

[2] Ellickson, R. C. (1991). Order without Law. Cambridge, MA: Harvard University Press.

[3] Boehm, C. (2011). Moral Origins: The Evolution of Virtue, Altruism, and Shame. New York: Basic Books.

[4] Wilson, D. S., Wilczynski, C., Wells, A., & Weiser, L. (2000). Gossip and other aspects of language as group-level adaptations. In C. Heyes & L. Huber (Eds.), Cognition and Evolution (pp. 347–365). Cambridge, MA: MIT Press.

[5] Kniffin, K. M., & Wilson, D. S. (2005). Utilities of gossip across organizational levels: multilevel selection, free-riders, and teams. Human Nature, 16, 278–292.

[6] Giphart, R., & Van Vugt, M. (2018). Mismatch: How Our Stone Age Brain Deceives Us Every Day (And What We Can Do About It). Robinson.

[7] Sherwin, J. C., Reacher, M. H., Keogh, R. H., Khawaja, A. P., Mackey, D. A., & Foster, P. J. (2012). The association between time spent outdoors and myopia in children and adolescents: a systematic review and meta-analysis. Ophthalmology, 119(10), 2141–51. http://doi.org/10.1016/j.ophtha.2012.04.020

[8] Rook, G. A. W. (2012). Hygiene hypothesis and autoimmune diseases. Clinical Reviews in Allergy & Immunology, 42(1), 5–15. http://doi.org/10.1007/s12016-011-8285-8

[9] Gray, P. (2013). Free to learn: Why Unleashing the Instinct to Play Will Make Our Children Happier, More Self-Reliant, and Better Students for Life. New York: Basic.