Is Using Profanity a Sign of Honesty?
Claims that swearing is a sign of honesty are highly questionable
Posted March 24, 2017
Profanity, the use of offensive and rude words, is naturally a controversial topic. There is some debate about what profanity reveals about a person’s character, more specifically, whether it is a sign of honesty or dishonesty. On the one hand, profanity is generally offensive; people who use swear words violate social norms of politeness, and therefore, some people see profanity as a sign of potential antisocial and deviant tendencies. On the other hand, others regard profanity as an authentic way of expressing strong emotions and therefore see it as a sign of candor and frankness. A recent paper attempted to settle this debate by examining the relationship between profane language and (dis)honesty. Based on three studies, the authors concluded that profanity is associated with greater honesty. However, closer examination of their results shows that only one of their studies supported their conclusion, albeit tentatively, while another study actually points to the opposite conclusion, and the remaining study was inconclusive. Hence, the question remains unsettled.
Using profane language is considered taboo in polite social contexts, is often found offensive, and when directed at a person in anger is considered abusive. On the other hand, people often use swear words to express all manner of strong, intense emotions. One study (Rassin & Heijden, 2005) suggested that even though people might not generally approve of profanity, under certain circumstances, using swear words may increase a person’s perceived credibility. In this study, people read testimony of a person who denied committing a crime. When the testimony included swear words, readers found it slightly more credible. The authors noted, of course, that under different circumstances though, swearing could decrease a person’s credibility. More importantly though, they acknowledged that this study could not determine whether swearing really is a sign that a person is being more truthful. On the other hand, some other studies suggest that people with “dark” traits, which are associated with dishonesty, swear more often than other people. For example, research has found that people who use more frequent profane language in daily life tend to be somewhat higher in narcissistic traits, such as arrogance, sense of superiority, exploitativeness and entitlement (Holtzman, Vazire, & Mehl, 2010), as well as lower in agreeableness and conscientiousness (although they also tended to be somewhat more extraverted) (Mehl, Gosling, & Pennebaker, 2006), suggesting that people who curse more often tend to be somewhat less concerned about following societal rules and conventions. Additionally, another study found that greater use of swear words on Twitter was associated with higher levels of both psychopathy (willingness to violate others’ rights) and Machiavellianism (cynical, manipulative tendencies). The traits of narcissism, psychopathy, and Machiavellianism are collectively known as the “dark triad” of personality, and are associated with greater dishonesty. Hence, these studies suggest that people who use profane language might be more dishonest in general, although they did not test this directly.
Despite this, a recent paper (Feldman, Lian, Kosinski, & Stillwell, 2017) suggested that perhaps profanity might be positively associated with honesty, because such language can be used to express one’s true feelings directly, and might therefore indicate greater candor and frankness. To test this, the authors performed three studies examining relationships between use of profanity and purported measures of honesty. The authors claimed that their findings support a relationship between profanity and greater honesty, but as I will show, there are some problems with their methods and conclusions.
The first study used an internet sample; participants answered questions about their use of profanity, and completed a measure of social desirability, the Eysenck Lie scale. The latter scale purports to measure how much a person will lie to create a socially acceptable impression of themselves by asking about whether the person engages in highly desirable but also hard-to-believe behaviors, such as always keeping their promises, never littering, and so on. The idea is that people who say they do these things don’t really do them but just want to give an impression of being more virtuous than they are. Hence, Feldman and colleagues considered that higher scores would be an indicator of dishonest responding. The result of the study was that people who engaged in more profanity generally scored lower on the Lie scale, which the authors interpreted to mean that they were more honest, at least in respect to socially desirable responding. Participants were also asked to rate their reasons for swearing on 1 – 5 scale, where higher scores indicated this reason applied more often. The most common reason was to express negative emotion, followed by habit. Other reasons included an expression of true self and being more honest about their feelings. Least common reasons were intimidating or insulting others. The authors argued that this indicated that people more often saw profanity as a way of expressing their genuine emotions rather than being antisocial and harmful.
However, the authors’ interpretation of this result is problematic because multiple studies have shown that the Eysenck Lie scale does not measure what it is supposed to. For example, although the statements on the Lie scale are supposed to be far-fetched, there is evidence that people who score highly on this scale are often telling the truth! Studies on adolescents have found that high scorers on the Lie scale are less rather than more likely to tell lies in general, while on the other hand, low scorers are more likely to engage in antisocial and delinquent behavior (Pearson & Francis, 1989). Other research suggests that high scores may indicate more conscientious behavior (Chapman, Weiss, Barrett, & Duberstein, 2013). Hence, the finding linking more use of profanity to lower Lie scale scores actually suggests that people who swear more often are probably more likely to be dishonest in general, contrary to the claims of Feldman and colleagues. Regarding the reasons people gave for swearing, both expressing one’s true self and being more honest about feelings had scores below 3/5, indicating these were infrequent reasons for swearing. intimidating or insulting others had even lower scores, below 2/5, indicating that these reasons more rarely apply, which probably reflects the taboo nature of cursing people. I would not read too much into these results regarding reasons because they do not consider individual differences. As I mentioned earlier, people with dark triad traits swear more often, and people with these traits engage in more antisocial behavior. Hence, reasons people most commonly swear probably depend on their personality to some extent.
The second study had a rather novel design in that it used a software to program to analyze participants’ status updates on Facebook. This program, Linguistic Inquiry and Word Count (LIWC), is based on the premise that the words a person uses to communicate can provide clues to their psychological state, and in particular, whether they are likely to be lying. Specifically, the theory is that when people are lying they use fewer first-person and third-person pronouns (as if they are trying to mentally distance themselves from their lies), use more negative emotion words (because they may feel discomfort about lying), and use less cognitively demanding language (because keeping track of one’s lies takes mental effort). (This is somewhat of an over-simplification, but this is the gist of the method). Previous research found that this method was about 61% accurate in sifting truth from lies (Newman, Pennebaker, Berry, & Richards, 2003). This is less than a stellar success rate, but it is significantly better than chance, and somewhat better than most people do when they are trying to detect lies using their own judgment. In Feldman et al.’s study they focused on anxiety-related words rather than general negative words, because past research suggested that anxiety words may be more predictive of honesty than other negative emotion words. Additionally, profanity is related to use of anger words in particular, so analyzing the latter may have biased their results. As expected, they found that people who used more profanities in their status updates tended to be somewhat more honest in their updates in general according to the LIWC software. (This was a correlation of .20, indicating a modest association). Interestingly, males tended to use more profanity than females, but also seemed to tell more lies, the opposite of the general result. However, even when statistically adjusting for sex differences, the positive correlation between profanity and honesty still held.
I think that the result of Study 2 is interesting, although it probably should be interpreted cautiously. In a previous study using LIWC software to analyze the honesty of Facebook status updates (Feldman, Chao, Farh, & Bardi, 2015), the authors admitted that the measure is a crude one and that not having control over what topics people update their status about represents a methodological weakness. That is, people on Facebook can post status updates about any topic, and this introduces an element of randomness into attempts to analyze the content of these updates. For example, if someone posts something as an obvious joke, does this count as truth or a lie? Additionally, because anxiety words were used to assess honesty, the results might mean that people who used more profanity were less anxious rather than more honest. Hence, I consider that the results of study 2 probably provide tenuous support for Feldman’s et al.’s hypothesis, but I would emphasize the tenuous part.
Study 3 attempted to assess broader social implications of profanity by extending the analysis to statewide data. This was done by examining US state level integrity data, e.g. ratings of executive, legislative, and judicial accountability in each of the 50 states. Profanity data was derived from participants’ scores in study 2. American participants were sorted by their state of origin, to provide mean state-level estimates of profanity. There was a positive relationship between profanity and state-level integrity, indicating that states with higher profanity scores also had more integrity. This is interesting, but what does it mean? Does this mean that people in states with higher rates of profanity are generally more honest and therefore their government institutions have more integrity? Not necessarily. After all, participants in each state may not be particularly representative of the people who run the state institutions. In fact, the people running the state are not random individuals and they might differ in important ways from the general population who make up each state and those posting on Facebook. The authors briefly mention that testing relationships on different levels of analysis has special difficulties due to Simpson’s paradox, also known as the ecological fallacy. This means that relationships that apply at an individual level may not apply at a regional level for example, and may even be reversed. However, the authors, in my opinion, do not give sufficient consideration to this as a problem.
As an example of the ecological fallacy, consider the so-called "conscientiousness paradox," that I discussed in a previous post. At an individual level, people who are highly conscientious tend to be healthier and live longer, probably because they have healthier lifestyles, e.g. they exercise more, and drink and smoke less. However, paradoxically, people in American states with higher mean levels of conscientiousness have shorter life expectancies. Internationally, countries with high mean levels of conscientiousness tend to be poorer and have more public health problems. One possible explanation that I have proposed is that when considering individuals living in the same environment, highly conscientious individuals tend to be better off than their peers, all else being equal. However, when comparing regions or countries, one is comparing different environments. It is possible that harsh, dangerous environments might foster higher levels of conscientiousness as a survival strategy, whereas in better environments, people can afford to be more relaxed. Hence, while conscientiousness may be a sign of health at an individual level, at an aggregate level it may signify the complete opposite. Another example that may be even more relevant concerns the relationship between religiosity and use of pornography. At an individual level, people who are more religious tend to avoid viewing pornography. However, at an aggregate level, US states that have more religious believers tend to have more frequent internet searches for pornography compared to states with more people with no religion (Whitehead & Perry, 2017). This finding is difficult to interpret because internet searches are anonymous, and so it is unclear who these people in each state searching for pornography are, and what their attitude to religion is. Hence, the people searching for porn in these states might not be representative of the wider population.
Hence, even though mean state levels of profanity on Facebook may be positively correlated with state-level integrity, it does not follow that profanity is related to integrity in any obvious way. In the past, some scholars (e.g. Richard Lynn) have been tempted to assume that the character of society’s institutions reflects the average character of the individuals who comprise society, but the evidence (e.g. from studies of the “conscientiousness paradox”) suggests that this is not necessarily the case. For example, it would be naïve to assume that because Facebook users who swear frequently tend to live in high-integrity states that this means that their state legislators also use more profanity and this somehow signifies their honesty and incorruptibility. Of course, Feldman and colleagues have not suggested this, and I do not mean to imply any such thing. What I am saying is that their findings are difficult to interpret in a meaningful way, so the broader social implications of profanity are far from clear.
Can we say then whether profanity signifies honesty and candor? The evidence for this seems rather weak. The results of study 1 contradict this theory. Study 2 is interesting but due to the methods used, the results are tenuous. The results of study 3 deserve more investigation, but their social significance is unclear. Past research links profanity with antisocial characteristics. However, as Feldman and colleagues point out, the context in which profanity occurs needs to be considered to understand its meaning. Under some circumstances, swearing can be beneficial. For example, there is evidence that swearing can sometimes help people tolerate pain (Stephens & Umland). Perhaps under the right circumstances, profanity might be a sign of candor and sincerity, but there is currently not much scientific evidence to shed light on this. Hence, more nuanced research considering contextual factors is needed to understand this thorny issue.
 I personally had a chance to test this out once when I was passing a kidney stone a few years ago. A lot of swearing went on that night!
The Card Sharps by Caravaggio, 1594.
Chapman, B. P., Weiss, A., Barrett, P., & Duberstein, P. (2013). Hierarchical structure of the Eysenck Personality Inventory in a large population sample: Goldberg’s trait-tier mapping procedure. Personality and Individual Differences, 54(4), 479-484. doi:http://dx.doi.org/10.1016/j.paid.2012.09.024
Feldman, G., Chao, M. M., Farh, J.-L., & Bardi, A. (2015). The motivation and inhibition of breaking the rules: Personal values structures predict unethicality. Journal of Research in Personality, 59, 69-80. doi:http://dx.doi.org/10.1016/j.jrp.2015.09.003
Feldman, G., Lian, H., Kosinski, M., & Stillwell, D. (2017). Frankly, We Do Give a Damn. Social Psychological and Personality Science, 1948550616681055. doi:10.1177/1948550616681055
Holtzman, N. S., Vazire, S., & Mehl, M. R. (2010). Sounds like a narcissist: Behavioral manifestations of narcissism in everyday life. Journal of Research in Personality, 44(4), 478-484. doi:http://dx.doi.org/10.1016/j.jrp.2010.06.001
Mehl, M. R., Gosling, S. D., & Pennebaker, J. W. (2006). Personality in its natural habitat: Manifestations and implicit folk theories of personality in daily life. Journal of Personality and Social Psychology, 90(5), 862-877. doi:10.1037/0022-3522.214.171.1242
Newman, M. L., Pennebaker, J. W., Berry, D. S., & Richards, J. M. (2003). Lying Words: Predicting Deception from Linguistic Styles. Personality and Social Psychology Bulletin, 29(5), 665-675. doi:10.1177/0146167203029005010
Pearson, P. R., & Francis, L. J. (1989). The dual nature of the Eysenckian lie scales: Are religious adolescents more truthful? Personality and Individual Differences, 10(10), 1041-1048. doi:http://dx.doi.org/10.1016/0191-8869(89)90255-9
Rassin, E., & Heijden, S. V. D. (2005). Appearing credible? Swearing helps! Psychology, Crime & Law, 11(2), 177-182. doi:10.1080/106831605160512331329952
Stephens, R., & Umland, C. (2011). Swearing as a Response to Pain - Effect of Daily Swearing Frequency. The Journal of Pain, 12(12), 1274-1281. doi:10.1016/j.jpain.2011.09.004
Whitehead, A. L., & Perry, S. L. (2017). Unbuckling the Bible Belt: A State-Level Analysis of Religious Factors and Google Searches for Porn. The Journal of Sex Research, 1-11. doi:10.1080/00224499.2017.1278736