Profanity is common in everyday life, yet law enforcement often treats all swear words alike. Building on Adams (2024), we surveyed a large public sample (n=2,412) who evaluated profanity’s appropriateness, professionalism, impact on trust, and disciplinary deservedness across nine scenarios (n=9,874) varying in intent (positive, neutral, derogatory) and target (self/situation, colleague, public). Results aligned with prior research: any profanity aimed at the public, especially in a derogatory way, drew the strongest condemnation. Meanwhile, positive or neutral profanity toward oneself or colleagues was generally acceptable, though derogatory profanity at colleagues elicited moderate concern. These findings underscore the need for nuanced language policies, rather than blanket bans, to address truly harmful speech without penalizing harmless expressions.
Profanity appears across many social contexts throughout the life course, including childhood (Jay & Jay, 2013; Jay et al., 2006), academic settings (Generous, Frei, et al., 2015; Generous, Houser, et al., 2015), and the workplace (Baruch et al., 2017; Johnson, 2012). Efforts to stop its use have had limited success, as profanity functions as a core element of everyday language (Eliasoph & Lichterman, 2003; Generous, Frei, et al., 2015). For example, in counseling, adopting a client’s vernacular—including profanity—can improve rapport and treatment outcomes (Giffin, 2016; Wiley & Locke, 1982). More broadly, profanity may help express emotion and alleviate stress (Husain et al., 2023).
However, police agencies have taken an opposing approach towards profanity usage, with departmental policies sidestepping the issue entirely (Adams, 2024), or relying on unrealistic blanket bans, deeming it “always unprofessional” (Michigan Commission on Law Enforcement Standards, 2022). High quality research shows that officer profanity can worsen public perceptions in use-of-force situations (Doyle et al., 2024; Martaindale et al., 2023; Patton et al., 2017; Sharps et al., 2019). Yet these broad bans overlook the varied contexts in which profanity arises in the policing workplace.
Recent research has begun to untangle these nuances, with this paper’s predecessor investigating law enforcement and human resource executives’ perceptions of police profanity usage. Adams (2024) proposed and experimentally tested a framework by which to evaluate the use of an expletive, specifically the word “fuck”, based on its intent (positive, neutral, derogatory) and target (self/situational, colleague, public). Adams found significant differences in respondent perceptions of profanity use based on differing combinations of intent and target. Generally, “derogatory profanity targeting colleagues, and all forms of profanity targeting the public were considered less appropriate, less professional, more harming to public trust, and more deserving of disciplinary sanction” (Adams, 2024, p. 3).
The current study replicates the methodology of Adams (2024), applying the same pre-registered theoretical framework and experimental methodology in a public sample. Our findings align with the original: the public sees profanity targeting the public as unacceptable, especially when used derogatorily. We find the public is just as capable and even-handed as police and human resources executives at evaluating the harms of profanity use across differing contexts. These findings provide further evidence in support of narrowing the breadth of profanity policy in pursuit of greater policy effectiveness and workplace fairness.
Rightly or wrongly, profanity is firmly embedded in the culture of policing. Like any workforce, police are not excluded from the overall assertion that people are profane at work (Baruch et al., 2017; Johnson, 2012). While the specific frequency at which police use profanity is not empirically established, motivating examples abound. Van Maanen (1978) provides the classic account of police profanity as a cultural phenomenon, providing insight into the unique way police officers use profanity to categorize those they encounter – most notably “The Asshole.” Nearly thirty years later, Moskos (2009) enters the police workforce to produce ethnographic evidence that profanity remains a time-honored verbal skill among patrol officers in Eastern Baltimore:
“When I first came on this job, I couldn’t understand 90 percent of what these motherfuckers were saying…I speak three languages: English, bad English, and profanity!” (p. 62)
“Some people consider that a bullshit lockup. But fuck ‘em. I don’t see them locking up Al Capone…” (p. 84)
Profanity can obviously serve as a verbal weapon, but often serves as a casual filler rather than an outright attack. Frankfurt (2005) classifies this type of speech as “bullshit,” contending that bullshit is “one of the most salient features of our culture…Everyone knows this” (p. 1). In policing, these “backstage” (Goffman, 1959) exchanges can at times meet social or psychological needs, while at other times amount to officers “saying stupid shit for the sake of it” (Sausdal, 2020, p. 103). Profanity may also project a sense of composure amid tension and reinforce in-group bonds among officers (Fine & Corte, 2024; Newburn, 1995; Waddington, 1999). However, this seemingly meaningless “bullshit” may also marginalize certain officers, in that profanity use is more often and intensely used by males (Güvendir, 2015). On the one hand, this raises the potential of profanity use as a tool that pushes women, an out-group in the overly-masculinized atmosphere in policing (Workman-Stark, 2021), farther to the margins of police culture. On the other hand, Adams (2024) found that men and women in executive policing and HR roles did not vastly differ in their evaluation of police profanity.
The word fuck holds a singular status in the English language, widely regarded as the most profane term, carrying the “deepest stigma of any in language” (Read, 1934, p. 264). Its prominence as the quintessential profanity has made it a focal point in both linguistic and criminological research (Adams, 2024; Adams, 2016). Profanity research presents conflicting conclusions, as its usage has been associated with honesty (Feldman et al., 2017) and authenticity (Lie et al., 2024), as well as dishonesty (de Vries et al., 2018) and a lack of education or trustworthiness (Jay, 1999). This may reflect profanity’s highly context-dependent applications, ranging from expressions of frustration or joy to fostering camaraderie or exclusion. In the absence of context, profanity is not inherently harmful (Jay, 2009), and at times is even considered socially appropriate (Jay & Janschewitz, 2008).
Profanity serves distinct social and functional roles. Within occupational subcultures, shared vernacular—including profane language—can strengthen social bonds, fostering a sense of community and cohesion (Eliasoph & Lichterman, 2003; Fine & Corte, 2024). Functionally, profanity enhances recall, making statements containing expletives more memorable (MacWhinney et al., 1982). However, when directed at individuals outside the subculture, such as members of the public, profanity can amplify power imbalances and harm perceptions of police legitimacy (Fine & Corte, 2024), These effects are particularly pronounced when expletives are used insultingly during contentious interactions, as seen in research linking profanity use to negative outcomes in use-of-force incidents (Doyle et al., 2024; Martaindale et al., 2023; Patton et al., 2017; Sharps et al., 2019).
Yet not all uses of profanity in policing are inherently harmful or negative. Officers may employ profane language in ways that foster understanding or empathy. For example, officers using colloquial, profanity-laden language can connect more effectively with individuals who speak similarly in their daily lives (Dolan & Johnson, 2017; Moskos, 2009). The situational context matters: exclaiming “good fucking job” to a colleague, expressing empathy to a member of the public by saying “that’s fucking terrible,” or using a slur like “motherfucker” during an arrest are qualitatively different scenarios. Despite the shared use of profanity, these situations vary significantly in intent and target, raising questions about uniform application of departmental sanctions (Adams, 2024).
Excessive policy demands increase stress among police officers (Worden et al., 2024). Profanity rules contribute to this burden when they ban all swearing without regard to context (Adams, 2024), potentially forcing officers to monitor the minutiae of their speech in everyday practice. As police are more carefully and publicly monitored than ever before – through the use of body-worn cameras as just one example – this increased workplace surveillance is related to increased emotional tolls on officers (Adams & Mastracci, 2019). Blanket prohibitions often fail to distinguish casual profanity among peers from malicious profanity aimed at the public, or profanity used during a use-of-force incident, which can harm community perceptions (Doyle et al., 2024; Martaindale et al., 2023; Patton et al., 2017; Sharps et al., 2019).
In practice, departments rarely punish casual swearing, and legal challenges that hinge on officer profanity seldom prevail (Adams, 2024). While some field professionals may contend that this policy’s presence is necessary to regulate profanity that is considered undesirable, these haphazardly applied punishments only further complicate the nuanced category of unprofessional conduct policy in policing (Noble & Alpert, 2008). A lack of consistency in punishment application for behavior that is a de facto violation of departmental policy makes the disciplinary system unpredictable, which harms officers’ perception of departmental fairness and leads to negative outcomes for all parties involved (Worden et al., 2024). Maintaining fair policies is central to organizational justice (Wolfe & Lawson, 2020), which helps reduce misconduct and improves job satisfaction (Rosenbaum & McCarty, 2017; Wolfe & Piquero, 2011).
Existing research that supports profanity bans usually focuses on force incidents involving public-directed profanity (Doyle et al., 2024, 2024; Patton et al., 2017; Sharps et al., 2019). These scenarios, however, capture only one of many target-intent combinations (Adams, 2024), and are relatively rare (Alpert & Dunham, 2004; McLean et al., 2022). A more balanced policy approach—one aligned with the realities of police practice—can allow for consistent enforcement and more effective regulation of profanity.
One clear lesson has emerged from experimental research – police profanity in use-of-force incidents negatively impacts public perceptions of those interactions (Doyle et al., 2024; Martaindale et al., 2023; Patton et al., 2017; Sharps et al., 2019). Although the public may not be familiar with specific departmental profanity policies, such violations signal a disconnect between policy and officer conduct. When such actions go unpunished, they undermine perceptions of organizational justice, which, in turn, damages the department’s legitimacy (Sunshine & Tyler, 2003; Tyler, 2004).
Public perceptions of police legitimacy are critical for securing compliance and public approval (McLean & Nix, 2021). These perceptions are shaped largely by personal experiences, with even a single interaction significantly influencing attitudes toward a department or policing more broadly (Gau & Brunson, 2010; Nix et al., 2015; Skogan, 2006). Key factors such as officer demeanor, response time, follow-up, and ease of contact are especially influential. Dissatisfaction with even one of these factors dramatically increases the likelihood of an overall negative assessment (Bradford et al., 2009).
Positive public perceptions of police also serve practical interests, such as addressing recruitment challenges and rebuilding public trust following the events of 2020 (del Pozo et al., 2024; McClure et al., 2023). This has inspired a multitude of efforts to enhance police transparency in an effort to improve public perceptions of police (Kochel & Skogan, 2021; Schafer, 2013). To improve transparency and community relations, departments have implemented reforms like body-worn cameras (Stoughton, 2017; White et al., 2018), civilian review boards (Adams et al., 2024; McGregor, 2016), and increased focus on community policing strategies (Koslicki et al., 2021). However, the effectiveness of these efforts varies. For instance, civilian review boards may counter-intuitively result in net-negative public perception outcomes for departments when disagreements between boards and executives arise, overshadowing potential benefits from improved perceptions in cases of alignment (Schiff et al., 2024).
Given the current effort’s reliance on the original (Adams, 2024), it is worth quickly revisiting the original design and findings. Adams (2024) conducted a pre-registered, 3 x 3 survey experiment with a mixed design that included both within- and between-subject conditions. Respondents were exposed to four of nine possible vignettes, each depicting an officer's use of the word fuck under varying conditions: one of three intents (positive, neutral, or negative) and one of three targets (self, colleague, or public). After each vignette, respondents evaluated the acceptability, professionalism, impact on public trust, and appropriateness of disciplinary action, both from the agency’s perspective and their own.
The findings revealed that perceptions of profanity varied significantly based on intent and target. Profanity directed at the self or situation, particularly when neutral or positive in intent, was viewed more favorably than profanity directed at colleagues or members of the public, especially when the intent was negative. These perceptions also differed across demographic characteristics, with males generally holding more favorable perceptions of profanity use than females, as well as increases in career length and education correlating with more negative perceptions.
These results underscore the nuanced nature of profanity use, suggesting that blanket bans on officer profanity are overly simplistic and fail to account for critical contextual differences. Adams’ policy recommendations align with McWhorter’s (2021) argument that language policies should reflect these subtleties. Specifically, policies should focus on curbing derogatory profanity aimed at colleagues and any profanity directed at the public, while permitting lenience for self- or situationally targeted profanity, particularly when it carries neutral or positive intent.
However, the study’s sample—composed exclusively of human resource and law enforcement chief executives—limits the generalizability of its conclusions to the public. While these findings indicate that such nuanced policies may gain traction among policymakers, they do not address whether similar views are held by the general public. That is, the results may be convincing that internally such policy guidance is welcome, those same executives may worry that the public would not be as understanding and nuanced in their approach to police profanity as neutral, or even beneficial, normal workplace utterances. This gap forms the basis for the present study, which seeks to replicate and extend Adams’ work using a broader public sample.
We replicate the original pre-registration for Adams (2024),1 creating a 3x3 classification of the different intents and targets to provide nine unique situations in which an officer could potentially use the word fuck in practice. Targets are categorized by whether they are aimed at self/situation, colleagues, or the public, and intents are defined by their functionality, whether they are used in a derogatory, positive, or neutral manner. These categories are combined to form the nine unique scenarios presented in Table 1.
Target Definitions
Self or Situation Directed: When profanity was primarily referencing the officer themselves, or when describing a particular event, circumstance, or state.
Colleague-directed: When profanity was primarily directed at a colleague.
Public-directed: When profanity was primarily directed at a member of the public.
Intent Definitions
Derogatory: Encompasses terms used to belittle, offend, or express disapproval. This category of profanity is characterized by its intent to insult, criticize, or demean an individual or a group. These terms are often employed to assert dominance, express frustration, or convey negative sentiments, reflecting the complex emotional landscape of law enforcement interactions.
Positive: Contains language used to express solidarity, encouragement, or positive affirmation. Such profanity intends to foster rapport, demonstrate approval, or cultivate an atmosphere of mutual understanding. Central to this category is the intent behind the use of these terms: to communicate positive sentiments or support, regardless of the recipient.
Neutral: Includes terms that articulate emotions, attitudes, or feelings, typically used to underscore or augment a statement, or to offer an informative depiction of a person, situation, or state. This category is marked by its use of profanity in a manner that is neither overtly positive nor derogatory, often serving as a linguistic tool to enhance communication or express sincerity.
The intent and target framework were embedded in a survey experiment, systematically varying the target (self/situational, colleague, or public-directed) and intent (derogatory, positive, or negative) of the profanity used in a policing context. Differentiating our approach from recent work (Doyle et al., 2024; Martaindale et al., 2023), we avoid high-profile, risk-laden contexts such as arrests and use of force. Each respondent was randomly exposed to four of the nine conditions available, thus creating both within and between measures. For each condition, respondents were asked to rate the appropriateness, professionalism, impact on public trust, and recommended discipline for the scenario they reviewed. Vignettes followed a prompt that provided consistent background for each scenario presented to respondents:
“You will be presented with four real situations, although the names have been changed, and asked for your opinion on each situation. Following each scenario, you will be asked your opinion on four brief questions that ask about the appropriateness, professionalism, impact on public trust, and appropriate level of discipline.”
Following the above prompt, the first of four (of nine possible) randomly selected vignettes were presented, and respondents were asked to respond to four outcome questions. General descriptions of each of the nine scenarios are listed in Table 1.
For example, a respondent presented with the combination of Target{Self/Situation} and Intent{Derogatory} would have been given the following vignette following the common spur given above:
Officer Smith had pulled over a speeding car. As he approached the vehicle, he realized they had forgotten his ticket book back at the station. With his words recorded on his body-worn camera, Officer Smith said to himself, "Man, I'm such a fuck-up."
The situation came to the police agency through a random review of body-worn camera footage. There have been no complaints at this time.
Alternatively, a respondent could be presented with the combination of Target{Colleague} and Intent{Positive}, which would result in the following vignette:
Officer Smith's partner, Officer Jones, skillfully diffused a potentially volatile situation during a traffic stop. Impressed, Officer Smith, with his words recorded on his body-worn camera, said, "Jones, you handled that fucking brilliantly."
The situation came to the police agency through a random review of body-worn camera footage. There have been no complaints at this time.
A complete report of all nine vignette combinations is provided in the appendix to this report.
Table 1: Notional Theory Table
| Derogatory Intent | Positive Intent | Neutral Intent |
Target: Self/Situational | Insulting oneself; Criticizing a situation | Self-motivation; Appreciating a situation | Casual remark; Describing a situation; emphasis |
Target: Colleague | Insulting a colleague | Praising a colleague | Discussing a work-related matter; emphasis |
Target: Public | Insulting a member of the public | Praising a member of the public | Discussing a situation with a member of the public; emphasis |
We test the original hypotheses set out in Adams (2024), with one exception – we do not test the outcome “How severe would the disciplinary action be, according to your agency’s policy, for this officer.” This outcome has no meaning in the context of a public sample. All hypotheses were pre-registered (see Adams, 2024, for pre-registration documentation).
Perceived Acceptability: “In your opinion, how acceptable was this officer’s use of profanity at work?” Participants responded on a Likert scale from 1 (completely unacceptable) to 5 (completely acceptable).
H1A: The use of profanity when self-directed or situation-directed will be perceived as more acceptable than when other-directed (colleague or public).
H2A: Positive or neutral intent in profanity use will be perceived as more acceptable than derogatory intent.
Perceived Professionalism: “In your opinion, how professional was this officer in using profanity at work?” Participants responded on a Likert scale from 1 (completely unprofessional) to 5 (completely professional).
H1B: The use of profanity when self-directed or situation-directed will be perceived as more professional than when other-directed (colleague or public).
H2B: Positive or neutral intent in profanity use will be perceived as more professional than derogatory intent.
Impact on Public Trust: “If made public, how do you think this officer’s use of profanity would affect public trust in the police?” Participants responded on a Likert scale from 1 (greatly reduces public trust) to 5 (greatly enhances public trust).
H1C: The use of profanity when self-directed or situation-directed will be perceived as having less negative impact on public trust than when other-directed (colleague or public).
H2C: Positive or neutral intent in profanity use will be perceived as having less negative impact on public trust than derogatory intent.
Disciplinary Action: “In your personal opinion, how severe should the disciplinary action be for this officer?” Participants responded on a Likert scale from (1) no sanction, (2) verbal coaching, (3) written warning, (4) significant sanction such as time off, and (5) termination of employment.
H1D: The use of profanity when self-directed or situation-directed will be perceived as warranting less severe personal opinion-based disciplinary action than when other-directed (colleague or public).
H2D: Positive or neutral intent in profanity use will be perceived as warranting less severe personal opinion-based disciplinary action than derogatory intent.
The sample (n=2,412) for this study was drawn from a third-party commercial listserv (Mailers Haven, 2023), comprising 903,570 email addresses of heads of households in South Carolina. Prior experimental work has relied on similar distribution lists (Boehme et al., 2024). Surveys were distributed via Qualtrics beginning on August 6, 2024, with periodic email reminders. Data collection concluded on October 17, 2024. This corresponds to a response rate of less than 1% when calculated using AAPOR RR2 standards (AAPOR, 2022). However, beta testing suggested that only approximately 20% of emails reached respondents’ direct inboxes, indicating that the effective response rate for delivered emails is likely higher. Despite this adjustment, we cautiously interpret these figures. Among those who accessed the survey, 87% completed it.
The final sample was 51.4% female, 75.5% White, and 11.5% Black. Most respondents identified as politically conservative (44%) or moderate (36%), were 55 years of age or older (57%), held at least a four-year college degree (55%), and were married (62%). Nearly half (49%) were employed, with fewer than 10% reporting a history of victimization by crime and 30% reporting contact with police in the past 12 months.
Balance tests for experimental conditions, reported in Appendix A, indicate no significant differences across covariates, confirming successful randomization via Qualtrics. Chi-square tests and t-tests similarly found no statistically significant imbalances.
While the sample slightly overrepresents White respondents and underrepresents Black respondents compared to state demographics (United States Census Bureau, 2022) it aligns closely with South Carolina’s population in terms of gender and political affiliation. Relative to national demographics, the sample closely mirrors the national proportion of Black respondents and gender distribution. Despite these demographic variations, prior research supports the external validity of online survey techniques (Patten & Perrin, 2015). Nonetheless, we interpret generalizability with caution. And our design ensures a high degree of internal validity.
Full sample statistics are provided in Table 2, and additional details on balance across experimental conditions are presented in Appendix A.
Table 2: Sample Descriptive Statistics
|
| N | Pct. |
Sex | Male | 1079 | 43.0 |
| Female | 1291 | 51.4 |
Race | Asian | 22 | 0.9 |
| Black | 290 | 11.5 |
| Alaskan Native | 1 | 0.0 |
| Hispanic/Latino | 46 | 1.8 |
| Middle Eastern Descent | 6 | 0.2 |
| Native American | 13 | 0.5 |
| Two or More Races | 61 | 2.4 |
| White | 1897 | 75.5 |
| Other | 38 | 1.5 |
Age | 18-24 | 20 | 0.8 |
| 25-34 | 133 | 5.3 |
| 35-44 | 318 | 12.7 |
| 45-54 | 474 | 18.9 |
| 55-64 | 566 | 22.5 |
| 65+ | 867 | 34.5 |
Political Beliefs | Very liberal | 79 | 3.1 |
| Liberal | 258 | 10.3 |
| Moderate | 905 | 36.0 |
| Conservative | 874 | 34.8 |
| Very conservative | 234 | 9.3 |
Education | No high school degree | 12 | 0.5 |
| High school degree | 187 | 7.4 |
| Some college but no degree | 434 | 17.3 |
| 2-year college degree | 349 | 13.9 |
| 4-year college degree | 762 | 30.3 |
| Postgraduate degree | 632 | 25.2 |
Marital Status | Married, living with spouse | 1568 | 62.4 |
| Divorced | 279 | 11.1 |
| Separated | 31 | 1.2 |
| Widowed | 150 | 6.0 |
| Single, never married | 282 | 11.2 |
| Domestic partnership | 60 | 2.4 |
Employment | Full time | 1063 | 42.3 |
| Part time | 173 | 6.9 |
| Temporarily laid off | 13 | 0.5 |
| Unemployed | 36 | 1.4 |
| Retired | 884 | 35.2 |
| Permanently disabled | 79 | 3.1 |
| Taking care of home or family | 82 | 3.3 |
| Student | 12 | 0.5 |
| Other | 32 | 1.3 |
Income | < $30,000 | 142 | 5.7 |
| $30,001-$50,000 | 282 | 11.2 |
| $50,001-$75,000 | 378 | 15.0 |
| $75,001-$99,999 | 335 | 13.3 |
| $100,000-$149,000 | 453 | 18.0 |
| $150,000-$199,999 | 221 | 8.8 |
| $200,000+ | 234 | 9.3 |
| Prefer not to answer | 325 | 12.9 |
Victimization | No | 2151 | 85.6 |
| Yes | 227 | 9.0 |
Police Interaction | No | 1618 | 64.4 |
| Yes | 761 | 30.3 |
The independent variables of interest are the experimental conditions: target (self/situation, colleague, public) and intent (neutral, positive, derogatory), creating a 3 x 3 main effects design. Mixed-effects modeling was employed for analysis, due to each respondent’s responses to four randomly selected vignettes. To construct these models, the lmer package (Bates et al., 2015) in R (R Core Team, 2024) was utilized, and general form is presented below:
Equation (1)
The dependent variable, Outcome{i} for the i th observation, is modeled as following a normal distribution (mean = µ, variance = σ2). The mean is constructed as a combination of predictors, with αj[i] as a random intercept for participant j and observation i, and
The random intercept αj accounts for individual differences among participants and is modeled as following a normal distribution (mean = µ, variance = σ2). Following this main effects-only model, a second model was constructed to include interaction effects between target and intent. This model’s general form is presented below:
Equation (2)
The results strongly support the study’s hypotheses, demonstrating significant variation in public perceptions of police profanity based on its target and intent. These findings are summarized in Table 3, which reports the main effects of target and intent on perceptions of appropriateness, professionalism, public trust, and disciplinary action.
The target of profanity significantly influenced all outcomes. When profanity was directed at colleagues, it was perceived as significantly less appropriate (b=−0.284, p<0.001), less professional (b=−0.308, p<0.001), more harmful to public trust (b=−0.117, p<0.001), and deserving of stricter policy-based regulation (b=0.235, p<0.001) compared to profanity directed at the self or situationally. The effects were even more pronounced when profanity was directed at members of the public. In these cases, profanity was perceived as markedly less appropriate (b=−0.927, p<0.001), less professional (b=−0.845, p<0.001), more harmful to public trust (b=−0.447, p<0.001), and deserving of much stricter disciplinary regulation (b=0.818, p<0.001).
Profanity’s intent also shaped perceptions across most outcomes. When profanity was derogatory, it was viewed as less appropriate (b=−0.063, p<0.001), less professional (b=−0.066, p<0.001), and deserving of stricter disciplinary action (b=0.091, p<0.001) compared to neutral profanity. Conversely, profanity with positive intent elicited more favorable perceptions. Positive profanity was deemed more appropriate (b=0.119, p<0.001), more professional (b=0.099, p<0.001), beneficial to public trust (b=0.118, p<0.001), and less deserving of disciplinary action (b=−0.120, p<0.001) relative to neutral profanity.
The models explain a substantial proportion of variance, with conditional R2 values ranging from 0.538 to 0.639 across outcomes. However, the marginal R2 values, ranging from 0.063 to 0.152, suggest that random effects account for much of the explained variance. This finding aligns with expectations, given the diversity of respondents in a public sample compared to the more homogeneous sample of HR and law enforcement executives analyzed in the original study (Adams, 2024).
Overall, our results highlight the importance of considering both intent and target in the regulation of police profanity. While derogatory or public-directed profanity is perceived as highly inappropriate and deserving of strict sanctions, profanity with positive intent or directed at oneself or a situation garners significantly more favorable evaluations. These findings further validate the nuanced framework proposed by Adams (2024) and underscore the necessity of tailoring profanity policies to account for context-specific factors.
Table 3: Mixed Effects Model: Main Effects
| Appropriate | Professional | Public Trust | Discipline (Policy) |
[Target] Colleague | -0.284 (0.018)*** | -0.308 (0.016)*** | -0.117 (0.013)*** | 0.235 (0.015)*** |
[Target] Public | -0.927 (0.018)*** | -0.845 (0.016)*** | -0.447 (0.013)*** | 0.818 (0.015)*** |
[Intent] Derogatory | -0.063 (0.018)*** | -0.066 (0.016)*** | -0.016 (0.013) | 0.091 (0.015)*** |
[Intent] Positive | 0.119 (0.018)*** | 0.099 (0.016)*** | 0.118 (0.013)*** | -0.120 (0.015)*** |
Intercept | 2.496 (0.021)*** | 2.343 (0.019)*** | 2.743 (0.017)*** | 1.648 (0.019)*** |
SD (Intercept Respondent) | 0.643 | 0.583 | 0.582 | 0.667 |
SD (Observations) | 0.700 | 0.608 | 0.499 | 0.567 |
Num. Obs. | 9874 | 9847 | 9853 | 9839 |
R2 Marg. | 0.148 | 0.152 | 0.063 | 0.141 |
R2 Cond. | 0.538 | 0.558 | 0.603 | 0.639 |
Following initial modeling, demographic controls collected in our survey were included in secondary models. Results show the demographic characteristics of the individual who witnesses officer profanity influences their baseline perceptions and opinions. Overall, these models fit the data slightly more accurately with conditional and marginal R-squared values seeing slight increases across most controlled models. Amongst the non-experimental controls, respondent race had a consistent negative effect to the overall acceptance of officer profanity, non-white respondents were associated with decreases in perceived appropriateness (b = -0.121, p < 0.001), professionalism (b = -0.082, p < 0.05), public trust (b = -0.229, p < 0.001), and raising call for disciplinary action (b = 0.168, p < 0.001). Respondent income had a small yet consistent effect on responses as well, with increases in income being positively associated with perceived professionalism (b = 0.015, p < 0.05) and trust (b = 0.014, p < 0.05) and negatively associated with the desire for disciplinary action (b = -0.019, p < 0.05). Finally, female respondents were associated with significantly lower levels of perceived appropriateness (b = -0.080, p < 0.01) and professionalism (b = -0.111, p < 0.001). Additional controls for education level and political affiliation were included; however, neither achieved significance in any model.
Public perceptions of police legitimacy and levels of support for police are of longstanding scholarly interest (Kochel & Skogan, 2021; Sunshine & Tyler, 2003; Tyler, 2004). We acknowledge this interest through pre-treatment legitimacy measures included in additional controlled models. The inclusion of this additional control did not shift the experimental results (as expected, see Blair et al., 2023), and generally demonstrate that those with high levels of perceived police legitimacy are slightly less damning of police profanity.2
Table 4: Mixed Effects Model: Main Effects with Controls
| Appropriate | Professional | Public Trust | Discipline (Policy) |
[Target] Colleague | -0.289 (0.019)*** | -0.311 (0.016)*** | -0.111 (0.013)*** | 0.242 (0.015)*** |
[Target] Public | -0.938 (0.019)*** | -0.854 (0.016)*** | -0.443 (0.014)*** | 0.829 (0.015)*** |
[Intent] Derogatory | -0.070 (0.019)*** | -0.069 (0.016)*** | -0.019 (0.013) | 0.097 (0.015)*** |
[Intent] Positive | 0.109 (0.019)*** | 0.095 (0.016)*** | 0.115 (0.013)*** | -0.117 (0.015)*** |
Sex: Female | -0.080 (0.031)** | -0.111 (0.028)*** | -0.003 (0.026) | -0.009 (0.030) |
Race: Nonwhite | -0.121 (0.039)** | -0.082 (0.035)* | -0.229 (0.033)*** | 0.168 (0.038)*** |
Political Affiliation | -0.032 (0.017)+ | -0.024 (0.015) | 0.003 (0.014) | 0.012 (0.016) |
Education | 0.004 (0.012) | 0.003 (0.011) | 0.003 (0.010) | -0.020 (0.012)+ |
Income | 0.014 (0.008)+ | 0.015 (0.007)* | 0.014 (0.006)* | -0.019 (0.007)* |
Intercept | 2.514 (0.218)*** | 2.369 (0.196)*** | 2.830 (0.187)*** | 1.615 (0.212)*** |
SD (Intercept Respondent) | 0.632 | 0.573 | 0.567 | 0.641 |
SD (Observations) | 0.698 | 0.601 | 0.493 | 0.563 |
Num. Obs. | 9273 | 9251 | 9262 | 9249 |
R2 Marg. | 0.164 | 0.170 | 0.088 | 0.165 |
R2 Cond. | 0.540 | 0.565 | 0.607 | 0.636 |
Interaction terms allow us to gauge how the effects of profanity’s intent differ across various targets. Main-effects models focus solely on individual predictors, holding all else constant, and thus do not capture how different combinations of intent and target might shape outcomes. To address this gap, we fit a second model that includes an interaction term between these two key variables.
Model fit improved notably with this interaction specification (see Appendix B for chi-squared tests). Full results can be found in Appendix D, and Figure 1 illustrates the joint effect of target and intent on the four dependent variables. Relative to the main-effects model, these interaction terms help clarify which combinations of profanity intent and target elicit stronger or weaker reactions.
Derogatory intent consistently ranks lower than neutral or positive, and its distance from the others tends to widen when profanity is targeted at colleagues. Self and public targets remained close together at the more acceptable (self-directed) and less acceptable (public-directed) ends of the results. Positive-intended profanity demonstrated a slight mitigating effect on the consistently negative perceptions of public-aimed profanity, as it received more positive perceptions than neutral or derogatory profanity.
These plots indicate that despite the mitigating or aggravating effects that intent may have on target, self/situational-directed profanity holds consistently higher levels of approval across all intents. Further, public-directed profanity does the opposite, as it consistently falls below other targets. Colleague-directed profanity is more nuanced, with derogatory intent drawing farther below positive or neutral intents, countering the clustered results of other targets. These results imply that the public’s sentiments towards public and self-directed profanity are rather straightforward: self-directed profanity is more acceptable, whereas public targeted profanity is unacceptable, and colleague-directed profanity should avoid derogatory intent.
Figure 1: Experimental interaction effects of {Target} x {Intent}
The public does not clutch its collective pearls when confronting police profanity, but instead display considered judgment that is sensitive to the target and intent of the profanity in use. Our analysis examined how profanity’s target (self, colleague, or public) and intent (neutral, derogatory, or positive) affect four key outcomes: perceived appropriateness, professionalism, public trust, and the degree of policy-based discipline suggested by respondents.
Replicating Adams (2024) with a new sample, we find strong alignment between public and executive judgments on the relative severity of profanity use. Approval varies sharply by intent and target, underscoring the need for context in evaluating police profanity. Clearer distinctions in policy-based discipline across contexts could enhance fairness, reducing perceptions of arbitrary enforcement and strengthening organizational justice (Wolfe & Lawson, 2020; Worden et al., 2024).
While the findings of previous police profanity research argue that profanity use by officers is always wrong and should not be permitted by policy (Doyle et al., 2024; Martaindale et al., 2023), the contextual elements of profanity usage are of utmost importance, and provide more guidance for assigning punishment than simple usage could. It is vital for policy to recognize contextual nuances in order to hold officers accountable for more harmful public-directed or derogatorily intended profanity use while also demonstrating leniency in the midst of harmless self-directed or positively intended profanity directed at the self/situation or colleagues. This profanity is not only quite frequent in the working world (Baruch et al., 2017; Johnson, 2012), but is considered ordinary and often lacking in meaning (Sausdal, 2020).
Results from this paper’s predecessor indicate that human resource and law enforcement executives are able to recognize the differences in harm between differing profanity use contexts, thus providing evidence of policymakers’ willingness to implement policies that recognize these nuances, as well as their importance for police practice (Adams, 2024). While executives may be more influenced by peer practices than public opinion (Adams et al., 2024), they share an interest in maintaining public trust (Sunshine & Tyler, 2003). Our results demonstrate that the public is likewise able to recognize this profanity nuance, showing strikingly similar results in overall approval and discipline desires. This finding suggests that more nuanced profanity policy, based on the framework suggested in Adams (2024), may be positively accepted by the public.
The interaction effects revealed in the experiment underscore the pivotal role of context in shaping perceptions of profanity. When directed at the public, respondents largely fail to differentiate between neutral and derogatory intent—both are perceived with similar disapproval. Although profanity with a positive intent receives marginally more lenient reactions, it remains significantly more condemned than profanity aimed at oneself or a colleague. In line with Adams (2024), our replication thus reinforces a broad consensus that officers should avoid using profanity in public-directed contexts. From the perspectives of chiefs, sheriffs, HR executives, and the wider community, such language is considered unprofessional, highlighting the need for policies specifically addressing profanity aimed at the public.
Direct interpretation of our results demonstrates a public indifference towards self-directed profanity, intolerance of public-directed profanity, and a largely intent-dependent opinion on profanity between colleagues. At the least, both Adams (2024) and our results coincide in a clear policy recommendation – derogatory profanity targeting colleagues, and any profanity targeting the public, should be restricted in all but the most exigent circumstances.
While the experimental conditions explain a substantial portion of variance in our models, other elements—such as tone, type of profanity, and the overall circumstances, as suggested by Adams (2024)—allow leeway for executive discretion in evaluating police profanity. Adopting a policy that distinguishes among various intents and targets of profanity is a meaningful first step, but its implementation must be handled judiciously. This careful approach ensures room for individualized review of each incident’s broader context. In doing so, agencies can maintain both positive public perceptions and internal trust in the fairness of disciplinary practices.
Our methodology is vastly similar to that of “Fuck: The Police,” and to a great degree its limitations are ours. While revered as the worst of the worst (Read, 1934), “fuck” is only one word out of a long list of profanities to choose from. This intuitively narrows the scope of the analysis, but due to the complex nature of language, using the most extreme example allows our findings to represent the worst possible result. However, it is plausible that different profanities or different conjugations of “fuck” may yield differing perceptions in both the public and law enforcement executives. Future research that expands the profanity framework to other expletives will be able to provide a wider picture of profanity perceptions in police practice.
Amongst the profanity framework originally proposed by Adams (2024) lie intensity and form, which refer to the emotions or situations described by the profanity, and the specific conjugation or linguistic structure of the words used. These have not been addressed in either study, not because they are unimportant, but because of the difficulty in operationalizing them experimentally. Defining a profanity’s intensity is difficult to represent in a text-based survey, and there are far too many conjugations of the word “fuck” to causally identify which ones have more adverse effects than others. For example, Adams (2024, Table 1) lists fifty “fucks in use” in his original study, prompting the question (p. 12) “if every ‘motherfucker’ could be good, bad, or indifferent, how can we hope to regulate it?” Future research can continue to operationalize more contextual factors to further explain the variation in perceptions of profanity.
A related key constraint of this study is its reliance on text-only scenarios, giving some pause in considering its ecological validity. Real-world perceptions of profanity often hinge on tone, volume, and inflection. Consider the difference between an officer quietly whispering, “What’s up motherfucker?” to a colleague at a briefing versus forcefully shouting it across the room while showing visible aggression. Although the sentence remains unchanged, the circumstances—and likely policy implications—differ substantially. Future research should follow the example of Doyle et al. (2024), which included audio in its replication of Martaindale et al. (2023). Incorporating auditory and other contextual cues would clarify how these additional factors shape reviewer perceptions, to uncover the unique properties that might be salient to policy review and public reaction.
This study uses a sample from the state of South Carolina. While American English serves as a common linguistic foundation across the United States, regional accents, slang, and cultural norms shape distinct vernaculars that may influence perceptions of language use. While it may be less plausible that a sample of North Dakotans would strongly approve of derogatory police profanity aimed at members of the public, mindfulness of the culture and language norms of the sample location is vital for proper generalization of our results. Consequently, careful attention to the cultural and linguistic context of South Carolina is crucial for interpreting these findings. Although aspects of our sample, such as sex and the representation of Black respondents, align with broader U.S. demographics, these results are most readily generalizable to South Carolina. Extrapolation to other states, the national population, or international contexts should be approached with caution. Addressing this limitation will require future research using random, representative samples from the populations of interest.
The flip side of transparency is exposure, and policing is often tasked with balancing these priorities. The distribution of risk is never neutral – with a tendency towards increasing transparency in policing, policy must distinguish between language that undermines public trust and language that does not. Overly restrictive policies risk alienating officers and reducing operational effectiveness, while insufficient regulation can further erode public confidence.
Our results extended prior work on police chiefs, sheriffs, and human resources chief executives to the general public. Beat-for-beat, our results map onto Adams (2024). The extreme similarities in magnitude and direction of effects in both studies provide convincing evidence in favor of a generally accepted practice aimed at constraining the worst excesses of profanity, while recognizing the shared human love of a good “fuck.” The results shared across Adams (2024) and here demonstrate that both policing and HR executive and the public all agree that profanity cannot be judged in a vacuum.
Neither ignoring profanity, nor blanket bans on its use are realistic or effective, and in the event of their strict enforcement, camaraderie-building profanity and derogatory profanity are subject to the same punishments Profanity alone is does not cause harm (Jay, 2009), rather, the contextual factors surrounding profanity largely dictate its positive or negative interpretation. Through the consensus demonstrated by law enforcement executives and members of the public, these contextual factors can be accounted for, and profanity policy can be tailored effectively, without undue costs to public trust.
The authors declare no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
The first and second authors contributed equally to this work. Both authors have the right to list themselves as the first author on their respective CVs.
AAPOR. (2022, November 17). Response rates calculator. https://aapor.org/response-rates/
Adams, I. (2024). Fuck: The police. Police Quarterly, 10986111241241750. https://doi.org/10.1177/10986111241241750
Adams, I., McCrain, J., Schiff, D. S., Schiff, K. J., & Mourtgos, S. M. (2024). Police reform from the top down: Experimental evidence on police executive support for civilian oversight. Journal of Policy Analysis and Management, n/a(n/a). https://doi.org/10.1002/pam.22620
Adams, I. T., & Mastracci, S. H. (2019). Police Body-Worn Cameras: Effects on Officers’ Burnout and Perceived Organizational Support. Police Quarterly, 22(1), 5–30. https://doi.org/10.1177/1098611118783987
Adams, M. (2016). In praise of profanity. Oxford University Press.
Alpert, G. P., & Dunham, R. G. (2004). Understanding police use of force: Officers, suspects, and reciprocity. Cambridge Univ. Press.
Baruch, Y., Prouska, R., Ollier-Malaterre, A., & Bunk, J. (2017). Swearing at work: The mixed outcomes of profanity. Journal of Managerial Psychology, 32(2), 149–162. https://doi.org/10.1108/JMP-04-2016-0102
Bates, D., Mächler, M., Bolker, B., & Walker, S. (2015). Fitting linear mixed-effects models using lme4. Journal of Statistical Software, 67, 1–48. https://doi.org/10.18637/jss.v067.i01
Blair, G., Coppock, A., & Humphreys, M. (2023). Research Design in the Social Sciences: Declaration, Diagnosis, and Redesign. Princeton University Press.
Boehme, H. M., Adams, I. T., Metcalfe, C., Leasure, P., & Nolan, M. S. (2024). Does scientific research change minds? Linking criminology and public perceptions of policing. Criminology & Public Policy, 23(1), 201–224. https://doi.org/10.1111/1745-9133.12644
Bradford, B., Jackson, J., & Stanko, E. A. (2009). Contact and confidence: Revisiting the impact of public encounters with the police. Policing and Society, 19(1), 20–46. https://doi.org/10.1080/10439460802457594
de Vries, R. E., Hilbig, B. E., Zettler, I., Dunlop, P. D., Holtrop, D., Lee, K., & Ashton, M. C. (2018). Honest people tend to use less—not more—profanity: Comment on Feldman et al.’s (2017) study 1. Social Psychological and Personality Science, 9(5), 516–520. https://doi.org/10.1177/1948550617714586
del Pozo, B., Rouhani, S., Clark, M. H., Andraka-Christou, B., Atkins, D., & Martins, K. (2024). Understaffed and beleaguered: A national survey of US chiefs of police about the post-George Floyd era. In CrimRxiv. https://doi.org/10.21428/cb6ab371.076db121
Dolan, H. P., & Johnson, R. R. (2017). The ‘language of the street’ fallacy. Dolan Consulting Group. https://www.dolanconsultinggroup.com/news/language-street-fallacy/
Doyle, M. K., Sandel, W. L., & Martaindale, M. H. (2024). Oh, @#%$!: A replication and expansion of Martaindale et al. (2023). Policing: An International Journal, ahead-of-print(ahead-of-print). https://doi.org/10.1108/PIJPSM-05-2024-0089
Eliasoph, N., & Lichterman, P. (2003). Culture in interaction. American Journal of Sociology, 108(4), 735–794. https://doi.org/10.1086/367920
Feldman, G., Lian, H., Kosinski, M., & Stillwell, D. (2017). Frankly, we do give a damn: The relationship between profanity and honesty. Social Psychological and Personality Science, 8(7), 816–826. https://doi.org/10.1177/1948550616681055
Fine, G. A., & Corte, U. (2024). Obscenity factories: Profanity and community in workgroup cultures. Work and Occupations, 51(3), 299–324. https://doi.org/10.1177/07308884241256101
Frankfurt, H. G. (2005). On bullshit. Princeton University Press. https://doi.org/10.2307/j.ctt7t4wr
Gau, J. M., & Brunson, R. K. (2010). Procedural justice and order maintenance policing: A study of inner‐city young men’s perceptions of police legitimacy. Justice Quarterly, 27(2), 255–279. https://doi.org/10.1080/07418820902763889
Generous, M. A., Frei, S. S., & Houser, M. L. (2015). When an instructor swears in class: Functions and targets of instructor swearing from college students’ retrospective accounts. Communication Reports, 28(2), 128–140. https://doi.org/10.1080/08934215.2014.927518
Generous, M. A., Houser, M. L., & Frei, S. S. (2015). Exploring college students’ emotional responses to instructor swearing. Communication Research Reports, 32(3), 216–224. https://doi.org/10.1080/08824096.2015.1052901
Giffin, H. (2016). Client’s experiences and perceptions of the therapist’s use of swear words and the resulting impact on the therapeutic alliance in the context of the therapeutic relationship. https://www.semanticscholar.org/paper/Client's-experiences-and-perceptions-of-the-use-of-Giffin/95a2c7ce602eee1cce7209586672bc32becf9cf2
Goffman, E. (1959). The Presentation of Self in Everyday Life. Oxford and Carlton: Blackwell Publishing.
Güvendir, E. (2015). Why are males inclined to use strong swear words more than females? An evolutionary explanation based on male intergroup aggressiveness. Language Sciences, 50, 133–139. https://doi.org/10.1016/j.langsci.2015.02.003
Husain, W., Wasif, S., & Fatima, I. (2023). Profanity as a self-defense mechanism and an outlet for emotional catharsis in stress, anxiety, and depression. Depression Research and Treatment, 2023(1), 8821517. https://doi.org/10.1155/2023/8821517
Jay, K., & Jay, T. (2013). A child’s garden of curses: A gender, historical, and age-related evaluation of the taboo lexicon. https://dx.doi.org/10.5406/amerjpsyc.126.4.0459
Jay, T. (1999). Why we curse. In Z.91. John Benjamins Publishing Company. https://benjamins.com/catalog/z.91
Jay, T. (2009). Do offensive words harm people? Psychology, Public Policy, and Law, 15(2), 81–101. https://doi.org/10.1037/a0015646
Jay, T., & Janschewitz, K. (2008). The pragmatics of swearing. 4(2), 267–288. https://doi.org/10.1515/JPLR.2008.013
Jay, T., King, K., & Duncan, T. (2006). Memories of punishment for cursing. Sex Roles, 55(1–2), 123–133. https://doi.org/10.1007/s11199-006-9064-5
Johnson, D. I. (2012). Swearing by peers in the work setting: Expectancy violation valence, perceptions of message, and perceptions of speaker. Communication Studies, 63(2), 136–151. https://doi.org/10.1080/10510974.2011.638411
Kochel, T. R., & Skogan, W. G. (2021). Accountability and transparency as levers to promote public trust and police legitimacy: Findings from a natural experiment. Policing: An International Journal, 44(6), 1046–1059. https://doi.org/10.1108/PIJPSM-04-2021-0062
Koslicki, W. M., Lytle, D. J., Willits, D. W., & Brooks, R. (2021). ‘Rhetoric without reality’ or effective policing strategy? An analysis of the relationship between community policing and police fatal force. Journal of Criminal Justice, 72, 101730. https://doi.org/10.1016/j.jcrimjus.2020.101730
Lie, D. S., Sung, B., Stankovic, M., & Septianto, F. (2024). How profanity in influences perceived authenticity and perceived helpfulness of online reviews: The moderating role of review subjectivity. Decision Support Systems, 178, 114144. https://doi.org/10.1016/j.dss.2023.114144
MacWhinney, B., Keenan, J. M., & Reinke, P. (1982). The role of arousal in memory for conversation. Memory & Cognition, 10(4), 308–317. https://doi.org/10.3758/BF03202422
Mailers Haven. (2023). Mailers Haven—Consumer, occupant, business & specialty data for mail shops and resellers. https://www.mailershaven.com/
Martaindale, M. H., Sandel, W. L., Duron, A., & Blair, J. P. (2023). @#%$!: The impact of officer profanity on civilians’ perception of what constitutes reasonable use of force. Police Quarterly, 26(2), 194–212. https://doi.org/10.1177/10986111221111332
McClure, K. A., Sweet, D. M., & Atkinson, D. J. (2023). Before and after George Floyd and Breonna Taylor: Citizen perceptions of a ‘reasonable officer.’ Psychology, Crime & Law, 1–25. https://doi.org/10.1080/1068316X.2023.2279335
McGregor, A. (2016). Politics, police accountability, and public health: Civilian review in Newark, New Jersey. Journal of Urban Health, 93(1), 141–153. https://doi.org/10.1007/s11524-015-9998-4
McLean, K., & Nix, J. (2021). Understanding the Bounds of Legitimacy: Weber’s Facets of Legitimacy and the Police Empowerment Hypothesis. Justice Quarterly, 0(0), 1–23. https://doi.org/10.1080/07418825.2021.1933141
McLean, K., Stoughton, S. W., & Alpert, G. P. (2022). Police uses of force in the USA: A wealth of theories and a lack of evidence. Cambridge Journal of Evidence-Based Policing, 6(3), 87–108. https://doi.org/10.1007/s41887-022-00078-7
McWhorter, J. (2021, February 8). Say no to police profanity. https://www.persuasion.community/p/say-no-to-police-profanity
Michigan Commission on Law Enforcement Standards. (2022). Basic training module specifications. Michigan Commission on Law Enforcement Standards. https://www.michigan. gov/-/media/Project/Websites/mcoles/TD-Webpage/basic_training_curriculum/ii-b-1.pdf? rev=74c97a3bee3e4a88af85bf353d6fc9fe
Moskos, P. (2009). Cop in the hood: My year policing Baltimore’s eastern district (Revised edition). Princeton University Press.
Newburn, T. (Ed.). (1995). Just boys doing business? (1st edition). Routledge.
Nix, J., Wolfe, S. E., Rojek, J., & Kaminski, R. J. (2015). Trust in the police: The influence of procedural justice and perceived collective efficacy. Crime & Delinquency, 61(4), 610–640. https://doi.org/10.1177/0011128714530548
Noble, J. J., & Alpert, G. P. (2008). Managing accountability systems for police conduct: Internal affairs and external oversights. Waveland Press.
Patten, E., & Perrin, A. (2015). Who’s left out in a web-only survey and how it affects results. Pew Research Center. https://coilink.org/20.500.12592/tf0m2j
Patton, C. L., Asken, M., Fremouw, W. J., & Bemis, R. (2017). The influence of police profanity on public perception of excessive force. Journal of Police and Criminal Psychology, 32(4), 340–357. https://doi.org/10.1007/s11896-017-9226-0
R Core Team. (2024). R: A language and environment for statistical computing [Manual] [Computer software]. R Foundation for Statistical Computing. https://www.r-project.org/
Read, A. W. (1934). An obscenity symbol. American Speech, 9(4), 264–278. https://doi.org/10.2307/452191
Rosenbaum, D. P., & McCarty, W. P. (2017). Organizational justice and officer “buy in” in American policing. Policing: An International Journal of Police Strategies & Management, 40(1), 71–85. https://doi.org/10.1108/PIJPSM-07-2016-0114
Sausdal, D. (2020). Police bullshit: Taking brutal police talk less seriously. Journal of Extreme Anthropology, 4(1), Article 1. https://doi.org/10.5617/jea.7360
Schafer, J. A. (2013). The role of trust and transparency in the pursuit of procedural and organisational justice. Journal of Policing, Intelligence and Counter Terrorism, 8(2), 131–143. https://doi.org/10.1080/18335330.2013.821738
Sharps, M. J., Torkelson, J. F., Hulett, D. L., Kuhn, M. L., & Sevillano, C. N. (2019). Police profanity and public judgments of guilt and effectiveness in officer-involved shootings. Journal of Police and Criminal Psychology, 34(1), 87–90. https://doi.org/10.1007/s11896-018-9268-y
Skogan, WesleyG. (2006). Asymmetry in the impact of encounters with police. Policing & Society, 16(2), 99–126. https://doi.org/10.1080/10439460600662098
Stoughton, S. W. (2017). Police body-worn cameras (SSRN Scholarly Paper 3094334). Social Science Research Network. https://doi.org/10.2139/ssrn.3094334
Sunshine, J., & Tyler, T. R. (2003). The role of procedural justice and legitimacy in shaping public support for policing. Law & Society Review, 37(3), Article 3. https://doi.org/10.1111/1540-5893.3703002
Tankebe, J. (2013). Viewing things differently: The dimensions of public perceptions of police legitimacy. Criminology, 51(1), 103–135. https://doi.org/10.1111/j.1745-9125.2012.00291.x
Tyler, T. R. (2004). Enhancing police legitimacy. The Annals of the American Academy of Political and Social Science, 593, 84–99.
United States Census Bureau. (2022). South Carolina—Census Bureau profile. https://data.census.gov/profile/South_Carolina?g=040XX00US45
Van Maanen, J. (1978). The Asshole. In P. K. Manning & J. Van Maanen (Eds.), Policing: A view from the street (pp. 221–238). Goodyear Pub. Co.
Waddington, P. (1999). Police (canteen) sub-culture. An appreciation. The British Journal of Criminology, 39(2), 287–309. https://doi.org/10.1093/bjc/39.2.287
White, M. D., Todak, N., & Gaub, J. E. (2018). Examining body-worn camera integration and acceptance among police officers, citizens, and external stakeholders. Criminology & Public Policy, 17(3), 649–677. https://doi.org/10.1111/1745-9133.12376
Wiley, D. A., & Locke, D. C. (1982). Profanity as a critical variable in counseling. Counselor Education and Supervision, 21(4), 245–252. https://doi.org/10.1002/j.1556-6978.1982.tb01708.x
Wolfe, S. E., & Lawson, S. G. (2020). The organizational justice effect among criminal justice employees: A meta-analysis. Criminology, 58(4), 619–644. https://doi.org/10.1111/1745-9125.12251
Wolfe, S. E., & Piquero, A. R. (2011). Organizational justice and police misconduct. Criminal Justice and Behavior, 38(4), 332–353. https://doi.org/10.1177/0093854810397739
Worden, R. E., Harris, C. J., & Kim, M. (2024). Disciplinary sanctions for police misconduct: An empirical analysis of sanction severity. Police Quarterly, 27(1), 80–108. https://doi.org/10.1177/10986111231174831
Workman-Stark, A. L. (2021). Exploring differing experiences of a masculinity contest culture in policing and the impact on individual and organizational outcomes. Police Quarterly, 24(3), 298–324. https://doi.org/10.1177/1098611120976090
Self or Situation Directed:
1. Derogatory Intent: Officer Smith had pulled over a speeding car. As they approached the vehicle, they realized they had forgotten their ticket book back at the station. Muttering to themselves, with their words recorded on their body-worn camera, they said, “Man, I’m such a fuck-up.”
2. Positive Intent: After maneuvering through heavy traffic, Officer Smith had successfully pulled over a speeding car. After driver left, and on their way back to their patrol car, with their words recorded on their body-worn camera, they said to themselves, “That was a fucking good job.”
3. Neutral Intent: Following a routine traffic stop, after the driver left, Officer Smith walked back to his car, with their words recorded on their body-worn camera, they said to themselves, “It’s a fucking beautiful day.”
Other-Directed (Colleague):
1. Derogatory Intent: During a traffic stop, Officer Smith’s partner, Officer Jones, accidentally dropped a piece of equipment. Irritated, Officer Smith, with their words recorded on their body-worn camera, said, “Jones, you’re a real fuck-up sometimes.”
2. Positive Intent: Officer Smith’s partner, Officer Jones, skillfully diffused a potentially volatile situation during a traffic stop. Impressed, Officer Smith, with their words recorded on their body-worn camera, said, “Jones, you handled that fucking brilliantly.”
3. Neutral Intent: Officer Smith was discussing the details of a traffic stop with Officer Jones. While describing the event, with their words recorded on their body-worn camera, Officer Smith said, “The driver was going at least 20 miles over the fucking speed limit.”
Other-Directed (Public):
1. Derogatory Intent: Officer Smith had pulled over a driver who was being uncooperative. Speaking to the driver, Officer Smith, with their words recorded on their body-worn camera, said, “Stop acting like a fucking idiot.”
2. Positive Intent: During a traffic stop, the driver immediately admitted to speeding and apologized. Appreciative of the driver’s honesty, Officer Smith, with their words recorded on their body-worn camera, said, “I fucking appreciate your honesty. It’s refreshing.”
3. Neutral Intent: While explaining the reason for a traffic stop to a driver, Officer Smith, with their words recorded on their body-worn camera, said, “You were going over the fucking speed limit.”
Appendix Table A: Balance Table by Target Treatment
Treatment Received
| |||||
Characteristic
| N
| Self, N = 3,285
| Colleague, N = 3,309
| Public, N = 3,289
| p-value
|
Sex | 9,455 |
|
|
| >0.9 |
Male |
| 1,435 / 3,141 (46%) | 1,443 / 3,179 (45%) | 1,429 / 3,135 (46%) |
|
Female |
| 1,706 / 3,141 (54%) | 1,736 / 3,179 (55%) | 1,706 / 3,135 (54%) |
|
Race | 9,468 |
|
|
| >0.9 |
Asian |
| 25 / 3,144 (0.8%) | 31 / 3,186 (1.0%) | 31 / 3,138 (1.0%) |
|
Black |
| 387 / 3,144 (12%) | 385 / 3,186 (12%) | 378 / 3,138 (12%) |
|
Alaskan Native |
| 1 / 3,144 (<0.1%) | 1 / 3,186 (<0.1%) | 2 / 3,138 (<0.1%) |
|
Hispanic/Latino |
| 64 / 3,144 (2.0%) | 58 / 3,186 (1.8%) | 62 / 3,138 (2.0%) |
|
Middle Eastern Descent |
| 7 / 3,144 (0.2%) | 8 / 3,186 (0.3%) | 9 / 3,138 (0.3%) |
|
Native American |
| 17 / 3,144 (0.5%) | 13 / 3,186 (0.4%) | 22 / 3,138 (0.7%) |
|
Two or More Races |
| 80 / 3,144 (2.5%) | 81 / 3,186 (2.5%) | 82 / 3,138 (2.6%) |
|
White |
| 2,512 / 3,144 (80%) | 2,557 / 3,186 (80%) | 2,503 / 3,138 (80%) |
|
Other |
| 51 / 3,144 (1.6%) | 52 / 3,186 (1.6%) | 49 / 3,138 (1.6%) |
|
Age | 9,482 |
|
|
| 0.8 |
18-24 |
| 32 / 3,144 (1.0%) | 24 / 3,191 (0.8%) | 24 / 3,147 (0.8%) |
|
25-34 |
| 194 / 3,144 (6.2%) | 165 / 3,191 (5.2%) | 173 / 3,147 (5.5%) |
|
35-44 |
| 418 / 3,144 (13%) | 422 / 3,191 (13%) | 431 / 3,147 (14%) |
|
45-54 |
| 625 / 3,144 (20%) | 634 / 3,191 (20%) | 632 / 3,147 (20%) |
|
55-64 |
| 747 / 3,144 (24%) | 764 / 3,191 (24%) | 746 / 3,147 (24%) |
|
65+ |
| 1,128 / 3,144 (36%) | 1,182 / 3,191 (37%) | 1,141 / 3,147 (36%) |
|
Political Beliefs | 9,370 |
|
|
| 0.9 |
Very liberal |
| 100 / 3,110 (3.2%) | 105 / 3,158 (3.3%) | 111 / 3,102 (3.6%) |
|
Liberal |
| 349 / 3,110 (11%) | 351 / 3,158 (11%) | 330 / 3,102 (11%) |
|
Moderate |
| 1,210 / 3,110 (39%) | 1,206 / 3,158 (38%) | 1,190 / 3,102 (38%) |
|
Conservative |
| 1,156 / 3,110 (37%) | 1,161 / 3,158 (37%) | 1,166 / 3,102 (38%) |
|
Very conservative |
| 295 / 3,110 (9.5%) | 335 / 3,158 (11%) | 305 / 3,102 (9.8%) |
|
Education | 9,473 |
|
|
| 0.8 |
No high school degree |
| 19 / 3,142 (0.6%) | 15 / 3,187 (0.5%) | 14 / 3,144 (0.4%) |
|
High school degree |
| 224 / 3,142 (7.1%) | 273 / 3,187 (8.6%) | 249 / 3,144 (7.9%) |
|
Some college but no degree |
| 579 / 3,142 (18%) | 571 / 3,187 (18%) | 582 / 3,144 (19%) |
|
2-year college degree |
| 453 / 3,142 (14%) | 474 / 3,187 (15%) | 459 / 3,144 (15%) |
|
4-year college degree |
| 1,010 / 3,142 (32%) | 1,024 / 3,187 (32%) | 1,004 / 3,144 (32%) |
|
Postgraduate degree |
| 857 / 3,142 (27%) | 830 / 3,187 (26%) | 836 / 3,144 (27%) |
|
Marital Status | 9,451 |
|
|
| >0.9 |
Married, living with spouse |
| 2,073 / 3,131 (66%) | 2,111 / 3,185 (66%) | 2,071 / 3,135 (66%) |
|
Divorced |
| 352 / 3,131 (11%) | 373 / 3,185 (12%) | 387 / 3,135 (12%) |
|
Separated |
| 43 / 3,131 (1.4%) | 43 / 3,185 (1.4%) | 38 / 3,135 (1.2%) |
|
Widowed |
| 196 / 3,131 (6.3%) | 212 / 3,185 (6.7%) | 187 / 3,135 (6.0%) |
|
Single, never married |
| 380 / 3,131 (12%) | 372 / 3,185 (12%) | 373 / 3,135 (12%) |
|
Domestic partnership |
| 87 / 3,131 (2.8%) | 74 / 3,185 (2.3%) | 79 / 3,135 (2.5%) |
|
Employment | 9,468 |
|
|
| >0.9 |
Full time |
| 1,422 / 3,143 (45%) | 1,423 / 3,186 (45%) | 1,401 / 3,139 (45%) |
|
Part time |
| 223 / 3,143 (7.1%) | 234 / 3,186 (7.3%) | 233 / 3,139 (7.4%) |
|
Temporarily laid off |
| 17 / 3,143 (0.5%) | 16 / 3,186 (0.5%) | 19 / 3,139 (0.6%) |
|
Unemployed |
| 47 / 3,143 (1.5%) | 50 / 3,186 (1.6%) | 47 / 3,139 (1.5%) |
|
Retired |
| 1,157 / 3,143 (37%) | 1,197 / 3,186 (38%) | 1,166 / 3,139 (37%) |
|
Permanently disabled |
| 101 / 3,143 (3.2%) | 101 / 3,186 (3.2%) | 112 / 3,139 (3.6%) |
|
Taking care of home or family |
| 112 / 3,143 (3.6%) | 109 / 3,186 (3.4%) | 107 / 3,139 (3.4%) |
|
Student |
| 19 / 3,143 (0.6%) | 14 / 3,186 (0.4%) | 13 / 3,139 (0.4%) |
|
Other |
| 45 / 3,143 (1.4%) | 42 / 3,186 (1.3%) | 41 / 3,139 (1.3%) |
|
Income | 9,450 |
|
|
| >0.9 |
< $30,000 |
| 188 / 3,134 (6.0%) | 203 / 3,181 (6.4%) | 171 / 3,135 (5.5%) |
|
$30,001- $50,000 |
| 365 / 3,134 (12%) | 366 / 3,181 (12%) | 391 / 3,135 (12%) |
|
$50,001- $75,000 |
| 498 / 3,134 (16%) | 510 / 3,181 (16%) | 503 / 3,135 (16%) |
|
$75,001- $99,999 |
| 453 / 3,134 (14%) | 445 / 3,181 (14%) | 437 / 3,135 (14%) |
|
$100,000- $149,000 |
| 597 / 3,134 (19%) | 611 / 3,181 (19%) | 600 / 3,135 (19%) |
|
$150,000- $199,999 |
| 295 / 3,134 (9.4%) | 292 / 3,181 (9.2%) | 296 / 3,135 (9.4%) |
|
$200,000+ |
| 305 / 3,134 (9.7%) | 310 / 3,181 (9.7%) | 318 / 3,135 (10%) |
|
Prefer not to answer |
| 433 / 3,134 (14%) | 444 / 3,181 (14%) | 419 / 3,135 (13%) |
|
Victimization = Yes | 9,483 | 292 / 3,148 (9.3%) | 304 / 3,190 (9.5%) | 311 / 3,145 (9.9%) | 0.7 |
Police Interaction = Yes | 9,486 | 992 / 3,147 (32%) | 1,016 / 3,192 (32%) | 1,032 / 3,147 (33%) | 0.5 |
Intent | 9,883 |
|
|
| 0.4 |
Neutral |
| 1,129 / 3,285 (34%) | 1,105 / 3,309 (33%) | 1,095 / 3,289 (33%) |
|
Derogatory |
| 1,049 / 3,285 (32%) | 1,117 / 3,309 (34%) | 1,119 / 3,289 (34%) |
|
Positive |
| 1,107 / 3,285 (34%) | 1,087 / 3,309 (33%) | 1,075 / 3,289 (33%) |
|
n / N (%) | |||||
Pearson's Chi-squared test |
Appendix Table B: Main vs Interaction Model Comparisons
Outcome | N Param | AIC | BIC | LogLik | Deviance | Chi Sq | df | pValue |
Appropriate (Main) | 5 | 24684 | 24734 | -12335 | 24670 | - | - | - |
Appropriate (Interaction) | 9 | 24445 | 24525 | -12211 | 24423 | 219.55 | 4 | < 0.001 |
Professional (Main) | 5 | 21999 | 22049 | -10992 | 21985 | - | - | - |
Professional (Interaction) | 9 | 21807 | 21887 | -10892 | 21785 | 155.51 | 4 | < 0.001 |
Public Trust (Main) | 5 | 18921 | 18971 | -9453 | 18907 | - | - | - |
Public Trust (Interaction) | 9 | 18814 | 18893 | -9396 | 18792 | 219.55 | 4 | < 0.001 |
Policy Discipline (Main) | 5 | 21459 | 21509 | -10722 | 21445 | - | - | - |
Policy Discipline (Main) | 9 | 21332 | 21411 | -10655 | 21310 | 155.51 | 4 | < 0.001 |
Appendix C: Controlled Model: Police Legitimacy Latent Variable
The latent variable, Police Legitimacy, was constructed from the mean Likert scale response (1: Strongly Disagree to 5: Strongly Agree) rating respondent agreement/disagreement with the following six items about the police: 1) treat everyone equally, 2) clearly explain the reasons for their actions, 3) treat people with dignity and respect, 4) treat people fairly, 5 ) respect people’s rights, and 6) listen to suspects before making any decisions about how to handle a case (alpha = 0.96) (Tankebe, 2013). Due to the experimental nature of our methodology, including this control serves as a means to increase the precision of treatment effect estimates by reducing residual variance. Thus, while this variable’s inclusion may refine our treatment estimates, it does not affect the causal interpretation of the model.
| Appropriate | Professional | Public Trust | Discipline (Policy) |
[Target] Colleague | -0.289 (0.019)*** | -0.311 (0.016)*** | -0.111 (0.013)*** | 0.243 (0.015)*** |
[Target] Public | -0.938 (0.019)*** | -0.854 (0.016)*** | -0.445 (0.014)*** | 0.830 (0.015)*** |
[Intent] Derogatory | -0.070 (0.019)*** | -0.068 (0.016)*** | -0.019 (0.013) | 0.096 (0.015)*** |
[Intent] Positive | 0.109 (0.019)*** | 0.095 (0.016)*** | 0.116 (0.013)*** | -0.117 (0.015)*** |
Police Legitimacy | 0.064 (0.016)*** | 0.078 (0.015)*** | 0.141 (0.014)*** | -0.143 (0.016)*** |
Sex: Female | -0.082 (0.031)** | -0.114 (0.028)*** | -0.010 (0.026) | -0.002 (0.029) |
Race: Nonwhite | -0.093 (0.040)* | -0.046 (0.036) | -0.164 (0.033)*** | 0.103 (0.038)** |
Political Affiliation | -0.049 (0.017)** | -0.045 (0.016)** | -0.036 (0.015)* | 0.051 (0.017)** |
Education | 0.002 (0.012) | 0.000 (0.011) | -0.002 (0.010) | -0.015 (0.012) |
Income | 0.013 (0.008)+ | 0.014 (0.007)* | 0.012 (0.006)+ | -0.017 (0.007)* |
Intercept | 2.437 (0.100)*** | 2.223 (0.089)*** | 2.343 (0.084)*** | 2.105 (0.095)*** |
SD (Intercept Respondent) | 0.634 | 0.573 | 0.555 | 0.632 |
SD (Observations) | 0.698 | 0.602 | 0.493 | 0.563 |
Num.Obs. | 9267 | 9245 | 9258 | 9245 |
R2 Marg. | 0.159 | 0.169 | 0.107 | 0.177 |
R2 Cond. | 0.539 | 0.564 | 0.606 | 0.635 |
Appendix D: Mixed Effects Model: Main and Interactive Effects
| Appropriate | Professional | Public Trust | Discipline (Policy) |
[Target] Colleague | -0.243 (0.031)*** | -0.266 (0.027)*** | -0.134 (0.023)*** | 0.214 (0.026)*** |
[Target] Public | -1.104 (0.031)*** | -1.001 (0.027)*** | -0.605 (0.023)*** | 0.969 (0.026)*** |
[Intent] Derogatory | -0.081 (0.032)* | -0.098 (0.028)*** | -0.035 (0.023) | 0.072 (0.026)** |
[Intent] Positive | 0.004 (0.031) | 0.020 (0.027) | -0.033 (0.022) | 0.023 (0.026) |
Colleague x Derogatory | -0.277 (0.044)*** | -0.209 (0.039)*** | -0.078 (0.032)* | 0.195 (0.037)*** |
Public x Derogatory | 0.337 (0.044)*** | 0.306 (0.039)*** | 0.140 (0.032)*** | -0.144 (0.036)*** |
Colleague x Positive | 0.160 (0.044)*** | 0.084 (0.039)* | 0.127 (0.032)*** | -0.130 (0.036)*** |
Public x Positive | 0.190 (0.044)*** | 0.157 (0.039)*** | 0.333 (0.032)*** | -0.305 (0.037)*** |
Intercept | 2.540 (0.025)*** | 2.380 (0.022)*** | 2.800 (0.020)*** | 1.605 (0.022)*** |
SD (Intercept Respondent) | 0.642 | 0.582 | 0.582 | 0.667 |
SD (Observations) | 0.689 | 0.600 | 0.494 | 0.562 |
Num. Obs. | 9874 | 9847 | 9853 | 9839 |
R2 Marg. | 0.162 | 0.163 | 0.069 | 0.148 |
R2 Cond. | 0.552 | 0.569 | 0.610 | 0.647 |