What the F

by Benjamin K. Bergen

  The people who do correlational studies cleverly try to control for those other factors by measuring and including them in their models. So if you believe that exposure to violence in media could be a confounding factor—it correlates with exposure to profanity and could explain some amount of aggression—then you measure not only how much profanity but also how much violence children are exposed to. The two will probably correlate, but the key point is that you can measure exactly how much media violence correlates with child aggressiveness, and you can pull that apart in a statistical model from the amount that profanity exposure correlates with child aggressiveness. The authors of the Pediatrics study tried to do this. But to know that profanity exposure per se and not any of these other possible confounding factors is responsible for increased reports of aggressiveness, you’d need to do the same thing not just for exposure to media violence, as the authors did, but for every other possible confounding factor, which they did not. That means that the real culprits—the real causes for increased aggression, if in fact there are any—may still be out there.

  And once again we have to deal with the possibility of reverse causation. Suppose we find that people who swear more really are also more aggressive. Which do you think is more likely? That using more profanity causes people to be more aggressive, as the article posits? Or that being more aggressive causes people to use more profanity? I would bet on the latter. In fact, there’s even some experimental evidence of this. A study conducted at Keele in the United Kingdom asked whether feeling more aggressive would cause people to swear more.25 Researchers randomly assigned participants to play one of two video games, a golf game or a first-person shooter (a first-person shooter involves navigating through a virtual world and trying to kill people with weapons like guns, missile launchers, and knives). Participants were then asked to name as many profane words as they could in one minute. The result was that after the (violent, aggressive) first-person shooter, they were able to name significantly more profane words than after the golf game.

  The upshot is that although suggestive, the Pediatrics study is only lightly so. In the way that if I polled a bunch of schoolchildren about their favorite color and found that the ones who liked red were more likely to also say they played sports frequently, that would suggest that color preferences cause sports participation. So, very, very lightly.

  The current state of our knowledge is this. We have no meaningful evidence that profanity causes aggression, just as we have no evidence that it causes changes in emotional responsiveness. Correlational studies don’t help. To know for sure, you’d need to run an experiment, which is the best way to ask a question about causation. You overcome the many limitations of correlational studies by manipulating one thing and seeing whether it has an effect on some other thing. If you do your job well (and designing experiments is not easy), then you can conclude whether or not the thing you manipulated caused a change in the thing you measured. The example everyone is familiar with is randomized drug trials. You give people a drug or a placebo (that’s what you manipulate), and you measure the effects on their health. Because you’ve randomly assigned people to the placebo or the drug condition, a meaningful difference between the two groups in terms of health outcomes means that the drug caused a difference in health compared with the placebo. It can’t be the reverse because the arrow of time doesn’t work that way. And as long as you make sure that the only difference between the groups is whether or not they receive the experimental drug, then you know there aren’t any hidden variables. You can conquer correlation through experimentation.

  So an experiment about profanity and aggression might go something like this. You take kids and randomly assign them to conditions that expose them to more or less profanity. There are no other differences between the groups. And after sufficient exposure, you measure how aggressively the children in each group act in some social task. Maybe you have a bully take away a child’s favorite toy and see how the kid responds, or something like that. And you look to see whether the kids who, through no fault of their own, were randomly assigned to greater profanity exposure act more aggressively as a consequence.

  As you may already be thinking, there’s an ethical dilemma here, which may be the reason this study hasn’t ever been conducted. In order to run this study, there has to be a reason to think that the hypothesis might be true—that increased exposure to profanity will increase aggression. But increased aggressiveness could be harmful to the child and those around him or her. So if the study found the effect it was set up to test for, it would harm half the participants. And those participants are children, who in general you’d prefer not to harm. As a result, the risk would outweigh the potential benefit, making a study like this a nonstarter.

  I posed this problem to the students in my lab, and one of them came up with a clever way around this problem.b Because the experiment I described is ethically impossible, suppose instead you took the opposite tack. What if some people were already opting-in to exposure to some profanity-laden media. Say, people who preordered some video game with swearing in it or who decided to watch an R-rated movie. What if you could convince the software company or the movie distributor to randomly send users one of two versions that differ only in the amount of profanity used. One is the normal version, with all the profanity; the other, an expurgated version without profanity. And then after these people have watched the film or played the game, you measure their aggressiveness in some social task. Neat idea. And it’s totally ethical in that the people will be exposed to the video game regardless, and if the hypothesis is correct, then the half of them who get the expurgated version will experience lower risk by dint of being in the experiment. It’s a win-win (as long as the only hypothesis in play is that more profanity makes people more aggressive).c

  It was Tyler Marghetis, for what it’s worth.

  The only drawback I can come up with is the following. Suppose exposure to profanity actually decreased aggression. For instance, suppose that people seek out profanity specifically in order to have an outlet to deal with feelings of aggression. And what if shielding them from profanity actually deprives them of this outlet, thereby causing them to find other outlets for their aggression? If there’s reason to believe this (and to date there’s effectively as much evidence for this position as for the reverse effect), then the study again becomes unethical. Because as soon as it’s possible that your manipulation will harm children, there had better be some really remarkable payoff for conducting the study. This imaginary study would not, for instance, cure cancer.

  The upshot is this: If exposure to profanity increases aggression through increased acceptance and use of profanity, that fact is in principle knowable. There’s an experiment to do. It’s complicated, but it’s doable. But the current state of our knowledge is basically that we have none. We don’t know—although there’s reason to believe it from other findings in language learning—whether increased exposure to profanity increases children’s use of profanity. We don’t know whether increased use of profanity increases aggression, and there’s as much evidence of this as of any other possible relation between these variables: that increased aggression increases profanity use or that increased profanity use decreases aggression. In other words, we have no idea.

  # $ % !

  So let’s take a step back. If there’s no evidence of profanity causing harm, then why would so many people—including researchers, journalists, parents, and legislators—believe it? Why would the American Academy of Pediatrics issue a press release describing this research? Why would it seem so plausible?

  I think two things are going on here. The first has to do with moral thinking. Many people believe profanity is immoral. Bad words are bad. We’ll get into why people think that in a later chapter, but I want to focus here on the cognitive consequences of thinking that something is bad. If you believe that something is good, you’re more likely to believe other good things about it, and if you think something is bad, you
’re more likely to believe other bad things about it. The positive version of this, known as the halo effect,26 has been studied for a century.27 The negative version, the horns effect (halo: angel; horns: devil), appears to apply to what you think about everything from people to products. Politicians use the halo effect; if they dress well, sound smart and approachable, and look friendly, then you’re likely to think they’re good at their job.28 Marketers use it too.29 People who like Apple Computer, Inc., for instance, are more likely to think good things about its individual products—that they’re high in quality or useful or well designed—than the products of a company they don’t know anything about. And something similar is probably going on with profanity. People think shit is a bad word. So they’re likely to believe that it can also be dangerous, cause harm, and have other negative attributes—more so than they’d believe these things for a “good” word, like please or thank you. The bar to convince people that sir or ma’am increases aggression would be far higher than a single correlational, self-reporting study.

  People might also be prone to believe bad things about profanity due to a second cognitive consequence of negative beliefs: rationalization. When people have unfounded moral beliefs, they often come up with explanations for them. Those explanations (“rationalizations” or “justifications”) may or may not be true, but they do make people feel better about their beliefs. If you believe something is bad, like profanity, but don’t really have a good reason why, then you might be strongly attracted to explanations that would justify your belief. If profanity increases aggression, that justifies your belief that profanity is bad. This sort of reasoning from back to front leads people to overlook logical weaknesses in an argument as long as it supports something they already believe.

  People who believe that profanity is bad are psychologically biased to believe a variety of other bad things about it, like that it increases aggression or numbs normal emotional responses. In essence, people who believe profanity is bad conclude that little minds are fragile things, perverted and deformed by the linguistic deviants around them. But this gives children too little credit. While the effects of abuse are real, there’s every reason to believe that children’s minds are resilient to profanity. The real fragility resides in the minds of those adults who are easily swayed to believe in the deleterious effects of profanity.

  10

  The $100,000 Word

  Kobe Bryant was the starting shooting guard for the Los Angeles Lakers from the time he was drafted straight out of high school in 1996 until he retired in 2016. Over those twenty years, he saw most everything the National Basketball Association (NBA) has to offer. Yet even he was still prone to emotional swings on the court. In a game against the San Antonio Spurs on April 12, 2011, Bryant was whistled for an offensive foul and then a technical foul. He demonstrably disagreed with the calls—as he came off the court, he punched a chair and threw a towel onto the ground. The producers of the live broadcast kept the cameras trained on the commotion, a fact he might not have been aware of when, from the bench, he yelled at referee Bennie Adams, “Bennie! Fucking faggot!” Analyst Steve Kerr immediately suggested to the producer on air, “You might want to take the camera off him right now, for the children watching from home.”1

  Even for a twenty-year veteran like Bryant, sports are emotional. Beyond the familiar thrill and agony of, respectively, victory and defeat, athletes experience anxiety about performance, pain from injury, anger at officials’ decisions, relief after prevailing in a must-win situation, and so on. As we’ve already seen, profanity is a privileged conduit to emotion. Add to the mix the intensity, pressure, competition, and spontaneity of sports, and it’s really no surprise that athletes swear.

  And they swear a lot. This probably wouldn’t stir up too much controversy if we were just dealing with enthusiastic fuck yeahs of victory and saturnine aw fucks of defeat. But we’re not. Oftentimes, as in Kobe Bryant’s case, the profanity belongs to the linguistic third rail of slurs. Fucking might not be hard to play off, but faggot is going to keep your publicist busy.

  Now, to close out this particular episode, Bryant later apologized and explained that he had spoken “out of frustration during the game, period” and that “the words expressed do not reflect my feelings toward the gay and lesbian communities and were not meant to offend anyone.” And to his credit, he’s refashioned himself as an advocate for LGBTQ rights and respectful speech. For example, when Jason Collins came out as the NBA’s first openly gay player, Bryant had this to say:

  I think him coming out was really brave. As his peers we have to support him, just rally around him and hopefully everybody else comes out and be themselves in who they are.2

  And when a pair of Lakers fans used a slur on Twitter, he called them out (in an economical 131 characters):

  Just letting you know @PacSmoove @pookeo9 that using “your gay” as a way to put someone down ain’t ok! #notcool delete that out ur vocab

  But his contrition was a day late and a dollar short. Actually, to be precise, it was a tad more than a dollar short. The NBA decided the day after the incident on the court to levy a fine for use of “offensive and inexcusable” words. The bill they sent Bryant: $100,000.

  Bryant is just one entry in a long list of basketball players punished for using slurs. In 2012, New York Nicks forward Amar’e Stoudemire was fined $50,000 for a tweet calling a fan a fag. In 2013 Roy Hibbert was fined $75,000 for using the expression no homo in a press conference. In 2015, Rajon Rondo was suspended for a game for calling a referee faggot. And so on. Simultaneously, the National Football League (NFL) has started penalizing players for slurs. The league’s officiating video, distributed to teams before the 2014–2015 season, stated, “The NFL will have ‘zero tolerance’ this season for players’ on-field use of racial slurs or abusive language relating to sexual orientation.”3 The penalty? Fifteen yards for the first infraction, as well as possible fines or other disciplinary action. San Francisco quarterback Colin Kaepernick was caught in the dragnet during the second game of that initial season, when he was fined $11,000 for ostensibly using the word nigger on the field.a

  The league eventually reduced the fine by half due to an absence of evidence that Kaepernick had actually uttered the slur: Sandritter, M. (October 15, 2014).

  On the surface, leagues and teams champion these policies as ways to ensure sportsmanship and to create safe and respectful environments. A more cynical interpretation sees them bowing to the same pressures as other entertainment organizations, like broadcast television and the film industry, blanching the entertainment product to the point where no viewer can find anything in it to object to. Regardless of the impetus, the consequences to players’ pocketbooks are real, and the message is clear: no slurs tolerated here. And for all of sports leagues’ documented evils (baseball’s history of segregation,4 football’s suppression of head trauma research,5 basketball and football’s exploitation of unpaid “amateur” college athletes),6 the antislur campaign in play in the major North American sports leagues is at least one thing that no one could possibly object to.

  Even if we accept that profanity in the aggregate doesn’t cause harm to children or adults, slurs seem like a different creature. They’re built to hurt. In this chapter, we’ll see how. But that doesn’t mean that banning them is the most productive approach. I’ll make the case below that even though slurs may cause harm, blanket policies outlawing them—like those adopted by the NBA and NFL and many other private and public organizations—can actually do more harm than good.

  # $ % !

  Slur-banning policies may at first blush seem reasonable for compelling psychological and historical reasons. To begin with, people find slurs more offensive than any other class of words, including other profanity. We can see this clearly from Kristin Janschewitz’s data, discussed in Chapter 1.7 You might recall that people were asked a number of questions about taboo and nontaboo words. I categorized the taboo words according to the now familiar
Holy, Fucking, Shit, Nigger Principle. Naturally, in doing so, I had to make some judgment calls. Is whore a slur or a sex-related word? (I classified it as a slur.) Does jism pertain to bodily effluvia or to sex? (I went with the latter.) But since these groupings are meant to give a general idea of how the language populates its profane vocabulary, let’s not get hung up on these questionable cases—they turn out not to make a difference for the big picture issue I want to extract. Look at what happens when you split profanity in this way. In the graph above, each word is a dot, and the mean offensiveness rating for each word is on the y-axis. So higher-up words are judged more profane on average.

  Each dot represents an English word in one of the four major categories of profanity. Slurs are judged far more offensive than any other group.

  As you can see, the offensiveness of the various words ranges broadly. There’s one single, lonely data point all the way on the top right, perched peerlessly at the zenith of offensiveness. That’s nigger. But although it’s an overall outlier, other slurs behave consistently with it. Directly below it is cunt, then fag, chink, and so on. The horizontal bars show averages by category, and you can see that on average slurs are judged more offensive than any other group by a full point or more. And this might actually understate how offensive slurs are. Some of the most offensive words in the other categories are arguably slurs—the list of most offensive Fucking words, starting at the top, are cocksucker and motherfucker.