How Civility is Exemplified in Controversial Conversations

One of the most surprising findings from our initial campus-wide survey was that students seem to believe that our campus falls as much short of the ideal in civility as in courage. While an overwhelming majority (88.6%) of students believe that interrupting is a negative behavior, for example, almost half (48.4%) say that it occurs frequently on campus, and 11.2% admit that they personally interrupt frequently. These numbers closely parallel those for choosing not to speak (perhaps the most problematic form of flight), and in both cases the gap between “ideal” and “real” scores was more than 1.6 on a 5 point scale. These were two of the three biggest gaps; of seventeen behaviors with gaps of 1.0 or more, three were associated with courage and five with civility. (There were, however, more total items associated with civility than with flight.) These findings would suggest that civility and courage are both much more difficult challenges than common ground and objectivity, both of which generated significantly smaller gaps (see Table 18, Conversational Behaviors with Large Gaps Between Ideal and Real, and Table 29, Civility Behaviors in Initial Survey and Campus Laboratories).

Yet in our observations we saw minimal evidence of campus incivility. Most of the classes and events we observed had one or no instances of interruption, and insulting language was virtually non-existent. At the same time, as noted above, we repeatedly encountered silence and reluctance to disagree. Indeed, our campus culture with respect to interrupting appears to differ markedly, even from other context in the upper Midwest. One research observed more interrupting in a single workshop for church professionals than he had in a dozen or more campus conversations, yet the participants in the workshop did not experience their conversational style as uncivil.

It is possible that most of the incivility at CSB/SJU takes place in places that are difficult for researchers to observe—late at night in the residence halls, for example, or perhaps at bars or “party houses.” We anticipated this possibility, and so on our initial survey we asked students if they were more likely to participate in positive or negative ways, or to avoid controversial conversations, in a variety of specific venues. As it turned out, students were reluctant to identify any specific context as a bad place for conversation. While an overwhelming 83.2% said that their participation was more positive in “classes with fewer than 25 students” as an especially good place for conversation, only 22% said they were more negative in “classes with more than 25 students,” indicating either that they were neither positive or negative, that they avoided controversial conversations in these classes, or that they had little experience with them. Likewise, most students said they participated positively “classes in my major” but not especially badly in “classes outside my major” as neither positive nor negative. A similar pattern holds true for non-class settings: “hanging out with friends” and “with people who agree with me” were rated very positively, while “at my job,” “student organizations,” “sports teams,” “with people who disagree with me,” “religious or spiritual events,” “public events,” and “electronic discussions” all received equivocal ratings. In virtually every setting, students were more likely to say that they avoided controversy than that they behaved negatively, and the only venue that received more negative than positive ratings was “classes with more than 25 students”—and this was precisely the venue where we observed rampant flight but minimal incivility (see Table 30, Conversational Differences among Venues).

Nevertheless, there is anecdotal evidence that informal settings are marked by more incivility than formal ones. When we share our findings with students, many respond with anecdotes about some particularly egregious form of verbal attack that occurred in a dorm room. Yet these anecdotes may suggest merely that even isolated instances of incivility tend to linger in the memory. Indeed, one of the challenges in coming to terms with the problem of flight is that even pervasive flight behaviors are inherently unmemorable, while a handful of uncivil behaviors can have a big emotional impact.

This distinction is especially evident in the surveys we conducted within our campus laboratories. Consistently, students perceived more incivility the further removed they were from a conversation, while they perceived more flight the closer they were to the conversation. I noted above that on our initial survey the gap between “ideal” and “real” scores was roughly equally for “interrupting” and “choosing not to speak.” But when students completed surveys at the end of individual class periods, they reported levels of “choosing not to speak” slightly higher than the “real” score in the initial survey (3.26 versus 3.21) and levels of “interrupting” that were dramatically lower (1.91 versus 3.40). Similarly, students reported very low levels of “raising one’s voice” and very high levels of “listening attentively” in individual class periods. When they were asked to evaluate their experience of the course as a whole at the end of the semester, on the other hand, the scores for “choosing not to speak” inched downward while those for “interrupting” moved up, though they still reported significantly more of the former. They were also asked to report on how much they personally had engaged in each behavior during the semester, and they acknowledged slightly more “choosing not to speak” and substantially less “interrupting” than they had ascribed to themselves at the beginning of the semester (see Table 21, Flight Behaviors in Initial Survey and Campus Laboratories, and Table 29, Flight Behaviors in Initial Survey and Campus Laboratories).

This last point suggests another difference between civility and courage: students are much more likely to acknowledge that they personally engage in flight behaviors than uncivil behaviors. On the initial survey, for example, the gap between “ideal” and “personal” scores for choosing not to speak was almost three times that for “interrupting,” even though the gaps between “ideal” and “real” scores were similar. As we suggested above, the greater social stigma attached to incivility may lead to greater self-reporting bias. But on the basis of our observations, it seems more likely that students are fairly accurate in their self-perceptions: they are much more likely to remaining silent during a controversial conversation than to interrupt or use insulting language. But when they think about controversial conversations in general, they are as likely to dwell on the incivility of a few of their peers as on the silence of the majority (see Table 18, Conversational Behaviors with Large Gaps between Ideal and Real).

When we used the Controversial Conversations Assessment Rubric to evaluate specific conversations, we were not surprised to find higher levels of civility than of courage. Indeed, civility received more “exemplary” ratings (42%) than any of our other dialogue components, and it was the only dialogue component that did not receive any “unsatisfactory” ratings. Moreover, the ratings would have been even higher had we not included in our definition a number of active behaviors that are relatively uncommon on our campuses. In order for a conversation to receive an “exemplary” rating, we specified, it was not enough for participants to be “consistently polite,” they also had to “actively encourage one another to speak by asking open-ended or clarifying questions, paraphrasing others’ statements, and identifying valuable insights.” Again and again, observers assigned “satisfactory” or “needs improvement” ratings to conversations that were flawlessly polite but lacking in active engagement and encouragement of other participants. Based on these observations, we would encourage CSB/SJU professors and others who lead conversations on campus to develop exercises that will help students draw one another out. More active civility, we believe, will go a long way toward mitigating the negative effects of occasional incivility (see Table 20, Controversial Conversations Assessment Rubric: Spring 2007 Results).

Though the priority should be on cultivating more active forms of civility, we do not believe that it would be wise to dispense entirely with conversational ground rules or other efforts to discourage incivility. There are two reasons for this. First, uncivil behaviors, even if they occur only occasionally, can have a disproportionately negative effect on the larger conversational culture. It is possible that the perception of incivility is one of the factors that contributes to flight, although on our initial campus-wide survey students who perceived especially high levels of incivility were not significantly more prone to flight than other students. These students did, however, report substantially higher levels of other negative behaviors including being asked not to speak by professors and peers, fearing physical or verbal attack, and choosing not to speak in the classroom to avoid conflict with professors. They were also less likely to say that faculty hold diverse views on controversial issues or that faculty encourage students to express diverse views.

In addition to diminishing the educational experiences of others, the small minority of students who are persistently uncivil experience substantial personal disadvantages. When we compared students with unusually high personal civility scores with those with low scores, we found a wide range of statistically significant differences. On the initial campus-wide survey, the highly civil students were more likely to say that they had changed their minds and learned new things from campus conversations, and less likely to say that professors had asked them not to speak on controversial issues (see Table 31, Significant Differences for Students Prone to Civil Behaviors, Initial Campus-Wide Survey).

Data from our campus laboratories suggests that highly civil students are significantly more capable of benefiting from courses dealing with controversial issues. At the beginning of the semester, the substantial majority of these students agreed that “I can learn a lot from students with whom I disagree,” that “it is important to participate in controversial conversations even when I do not have a strong opinion,” and that “I will work hard to protect the free speech of people with whom I disagree.” Only about a third to half of the less civil students agreed with these statements. Students were also asked if they found it helpful to engage in a variety of strategies for clarifying their views on controversial issues. The highly civil students were significantly more likely to identify each strategy as helpful, registering the largest gap (90% versus 38%) for “talking to people who are personally affected by the issue.” By the end of the semester, most of these gaps had widened. Reflecting on their experiences in the class, 87% of highly civil students and 36% of less civil students said that they had learned a lot from students with whom they disagreed; 68% and 28% respectively said that they had learned a lot from ideas in the texts with which they disagreed; 87% and 49% said that their views had become more complex; and 61% and 25% said that their views had changed (see Table 32, Significant Differences for Students Prone to Civil Behaviors, Campus Laboratory Surveys).

On many of these items, however, there were not statistically significant differences between highly civil students and students with personal civility scores close to the mean. Indeed, one might explain those differences by hypothesizing that students who see themselves as unusually civil may simply have sunny dispositions and be inclined to rate everything positively. The more important finding is that students with low personal civility scores—that is, students who see themselves as frequently engaging in uncivil behaviors—also report significant difficulties in benefiting from their college experience.

Since uncivil students harm both themselves and others, it seems wise to continue using discussion grounds rules and other teaching tools that stigmatize uncivil behavior. But it may be even more helpful to target these efforts more narrowly. The vast majority of our students are already civil enough to benefit substantially from college, but a minority are not. Many members of this minority, moreover, share their peers’ sense that civility is important, and they are conscious of their own shortcomings. Efforts to publicize the correlation between civility and educational benefit, and to assist these individuals in improving their conversational style, might bear significant fruit.

One might worry, finally, that students who are highly civil are also prone to flight behaviors, and that efforts to promote civility might have the unintended effect of exacerbating the problem of flight. We did not find conclusive evidence to support this worry. Indeed, on the initial campus-wide survey the most civil students were almost three times as likely to say that they “personally do not enjoy controversy” than uncivil students, and we found a statistically significant correlation between the personal civility composite score and the personal flight composite score. On the other hand, no such correlation appeared on the survey completed at the beginning of the semester by students in the campus laboratories. And by the end of the semester, the most civil students were significantly less likely to report that they had engaged in flight behaviors than their less civil peers. In any case, we found a much more dramatic inverse correlation between civility and competitiveness, a conversational component that we did not even identify until relatively late in our research process (see Table 33, Some Significant Correlations Between Personal Civility and Other Composite Scores).