Imagine an individual who denies service to a same-sex couple while using linguistic techniques associated with politeness—optimism, apology, and reciprocity: “I’m really sorry, gentlemen, but you understand that while we would like to help you, we cannot serve you here.” Is this person being uncivil? Some scholars would say yes—they are conveying anti-democratic messages in their decision to discriminate. Would we see their statement as uncivil if they had used a homophobic slur? What about if they were a member of the political elite instead of a restaurant or business owner? Each of these elements—the speaker, substance, and tone of political communication—has been labeled a definitional element of incivility. Using a survey experiment, we isolate each element to examine how it influences citizens’ perceptions of incivility and behavioral outcomes that have been linked to exposure to incivility, such as political trust. In the experiment, participants were randomly assigned to read one of 12 tweets addressing either President Obama’s position on campus free speech (in a pro- or anti-democratic manner) or his decision not to attend the funeral of a prominent conservative politician. These tweets came from either a politician or a college student, and were framed in either a polite or an impolite manner. After reading the tweet, participants were asked about the extent to which the message was uncivil or anti-democratic, as well as about a range of attitudes and behaviors that are influenced by incivility. We find that the use of an impolite tone is the greatest predictor of an individual’s perception that the tweet was uncivil; however, the use of impolite rhetoric by politicians is more likely to shape views on political trust and political incivility more broadly than if the same language is used by a citizen. Ultimately, our findings suggest that studies of political incivility need to take more than simply the tone of communication into consideration when designing research. Beyond the academic repercussions, these results suggest that advocates for increased political civility should consider concentrating their efforts on political elites, as reducing elite incivility has the greatest potential to mitigate some of the negative behavioral consequences of political incivility.