Contrary to the normative emphasis on the rule of equality in the deliberation literature, little has been known about empirical consequences of the rule of equality, especially when applied in online discussions where political disagreement is prevalent. Given that hostile gender-related discussions have been noticeably increasing in South Korea, we investigated whether applying deliberative rules, especially cross-cutting exposure and equality, can improve gender-issue discussion quality and foster mutual understanding and healthy political engagement. For this purpose, we designed an online experiment involving moderated deliberations on the abolition of the national abortion ban via KakaoTalk, the most popular messenger platform in South Korea. The deliberative qualities of online discussions in terms of rationality and civility were assessed in a more objective and unobtrusive way: a content analysis of actual conversation transcripts. Participatory intention for gender issue-related activities and civic attitudes were also measured. Results indicate the equality rule can help to promote normatively desirable outcomes in discussions with disagreeing others while the positive effects of cross-cutting exposure were found limited. When combined with the rule of equality, hearing the other side meaningfully enhanced the deliberative qualities and participatory intentions of discussants.
Can online deliberation over a highly controversial issue engender democratic benefits that are predicted by the framework of deliberative theory, particularly vis-á-vis South Korea’s feminist movement? One of the primary agendas in the growing feminist movement in South Korea is a public reassessment of the legal status of abortion. In 2012, South Korean courts upheld the constitutionality of the legislation banning abortion, yet a recent public opinion survey showed that more than half of the population (51.9%) supported lifting the ban (
However, antipathy against feminism has been on the rise in male-dominated online communities, thus deepening fault lines between two gender groups in younger generations (
In what follows, we examined whether a well-structured online deliberation can bring about democratic benefits in a highly polarized context of gender-related issues. As South Korean gender-related dialogue is becoming even more hostile, especially through online platforms, it may be important for people to have some opportunities to deliberatively reflect on relevant issues. We explored whether applying some essential deliberative rules (such as cross-cutting exposure and equality) into political conversations on a messaging application can 1) improve the rationality and civility of discussions on abortion, 2) promote healthy political participation regarding gender-related issues, and 3) foster empathetic and tolerant attitudes toward people holding different opinions.
Our study contributes to the deliberation literature by focusing on one of the relatively understudied principles of deliberation: equality. Contrary to the normative emphasis on the rule of equality, neither the meaning of the rule has been coherently conceptualized nor have the effects of equality been systematically investigated. Here, we borrow the framework of Friess and Eilders (
We also contribute to the study of cross-cutting exposure or disagreement in general. By varying the level of disagreement to which an individual gets exposed, we improve previous approaches that rely on individuals’ perceived sense of disagreement (e.g.,
To explore the aforementioned relationships, this study designed an experiment on online deliberation via KaKaoTalk, South Korea’s most widely used messenger platform. In particular, its Open Chatting functions provide opportunities for anonymous users to freely participate in informal discussions over a diverse range of issues. It also enabled us to assess the deliberative qualities in a more objective and unobtrusive manner: a content analysis of discussion scripts.
In short, we examine whether cross-cutting exposure and the rule of equality—components required for deliberative discussions—can help us successfully achieve some desirable outcomes expected from the deliberative process. We empirically test some of the normative assumptions about the benefits of civic deliberation, rather than accepting such consequences as a given, with regard to one of the most critical gender issues in South Korea: legalization of abortion.
Although the answer to the question of “what is the most necessary condition for deliberation?” has been controversial (
Experiencing disagreement, inter alia, has been accentuated as one of the foremost conditions to be met for deliberation (e.g.,
Given that elements that constitute the deliberative quality of political conversations are manifold, we identified two main dimensions of deliberativeness based on some previous research: reason-giving (
Then, how can exposure to dissimilar opinions influence these two dimensions of deliberativeness, rationality and civility? Imagine a situation when you need to persuade others who manifest a different perspective from yours: we automatically look for legitimate reasons to justify our positions. As Larmore (
However, perceived differences among discussants may create a sense of conflicts and anxiety during a conversation, possibly making them less willing to discuss a given topic in a reason-giving manner (
A similar question can be raised with regard to civility. Above all, interacting with people of differing views is essential to a more sophisticated understanding of others (
As discussed so far, empirical evidence for the thesis that cross-cutting exposure actually increases deliberativeness remains relatively equivocal. Reflecting on these inconsistent predictions in prior research, our first research question has been formulated as follows:
On the other hand, a raft of empirical evidence has suggested the contrary; individually-encountered disagreements are positively associated with ambivalence and cross-pressure, and thus causing withdrawal from political participation (
Similarly, findings have been inconsistent with regard to whether exposure to disagreeing opinions can succeed or fail to engender civic virtues such as empathy and tolerance. As proposed by the contact hypothesis (
Against this backdrop, our inquiries on the effects of cross-cutting exposure on important democratic outcomes have been put forward as follows:
Is there another possible intervention that may help discussants to realize the assumed, but often obscure, benefits of deliberation in the real-world contexts? We argue that talking with an equality rule is essential when multifarious opinions are exchanged. In this regard, we attend to how the rule of equality moderates the effects addressed in RQ1, RQ2, and RQ3, assuming that some positive outcomes of cross-cutting exposure can be enhanced when combined with the equality rule.
Compared to the theoretical emphasis on the rule of equality as a normative precondition for proper deliberation (e.g.,
Equality norms have also been defined in the throughput or output of deliberation (
Equality can refer to a number of properties within deliberative settings, yet it has rarely been experimentally examined at the individual level. In this study, we focus on a particular input-related form of discursive equality: the provision of equal opportunities to speak during group discussions. By providing at least three opportunities to talk about their arguments and reasons, we sought to equalize the quantity of given speech from the participants, to the extent that the equality rule does not demotivate them from freely exchanging their thoughts. Ideally, such forms of discursive equality can be realized via active moderation and rule-enforcing by facilitators (
By pairing the experiment with a content analysis of conversation transcripts, we examine the effects of cross-cutting exposure moderated with the rule of equality (the input dimension) on deliberative equalities of shared discussions (the throughput dimension) and behavioral intentions and attitudes (the output dimension). We first inquire whether fulfilling both conditions, cross-cutting exposure and the rule of equality, can enhance deliberative qualities of political discourse, i.e., rationality and civility. Under conversational domination, one may find it difficult to reflect on a given issue in a balanced way (
Similar effects can be expected for the consequences of cross-cutting exposure on political participation and civic virtues. The rule of equality in deliberation can provide participants with an “equal opportunity to access political influence” (
Further, with the equality rule being a procedural principle, participants may realize that cooperation with disagreeing others is not impossible (
To the best of our knowledge, there is no previous research that empirically examined the influence of structured discussions among diverse
The experiment involved moderated deliberations on the repeal of the abortion ban via KakaoTalk Open Chat that provides chatting not only with users’ intimate contacts but also with anonymous others. Using Open Chat features, we originally constructed our basic experimental design as presented in Table
The original experimental setting (
Opinion diversity within a group | ||||
---|---|---|---|---|
2 pro-choice, 6 pro-life | 4 pro-choice, 4 pro-life | 6 pro-choice, 2 pro-life | ||
The rule of equality | No | Group 1 ( |
Group 2 ( |
Group 3 ( |
Yes | Group 4 ( |
Group 5 ( |
Group 6 ( |
Since opinion diversity was manipulated at the group level in our original experimental setting, the degree of cross-cutting exposure during group discussions varied among participants assigned to the same experimental conditions. For example, participants with a group-majority viewpoint (e.g., pro-lifers assigned to Group 1 in Table
To operationalize cross-cutting exposure as ego-centric diversity at the individual level, i.e., the level of disagreement one encounters during a group discussion, experimental groups were reconstructed after the post-test was completed (Table
The reconstructed experimental setting (
Cross-cutting exposure | ||||
---|---|---|---|---|
25% | 50% | 75% | ||
The rule of equality | No | Group 1 ( |
Group 2 ( |
Group 3 ( |
Yes | Group 4 ( |
Group 5 ( |
Group 6 ( |
While most of the study participants were undergraduates from major universities in Seoul, South Korea, some robust supporters of the abortion ban, who were seldom found in the university student sample, were recruited from recently-held abortion-related rallies. The topic of discussion was whether South Korea should lift its abortion ban. The entire discussion session was limited to 30 minutes, and only the 30-minute length of the discussion after the facilitator’s opening statement was considered for the content analysis. To ensure informed deliberation, a ten-page document on the current state of the abortion issue and major arguments and rationales on each side were delivered to each participant a few days before discussion. All participants were given a mobile voucher worth 5,000 KRW (approximately $5) for participation. Each participant used a pseudonym assigned by the researcher as a nickname to maintain confidentiality during the discussion.
For all sessions, the facilitator announced the rules for discussion in the beginning (e.g., a 30-minute time limit) and regulated unpleasant language if needed. For the groups informed with the equality rule, the facilitator announced the predetermined order by which the participants should speak at least three times during discussion: at the beginning, in the middle (after 15 minutes from the start), and at the end. More specifically, participants in the groups where the rule of equality was enforced were shown the following messages three times during the discussion: “Now, please present your opinions about the discussed topic in the predetermined order: Frodo – Ryan – Neo – Jay-Z – Tube – Peach (names of Kakao characters). If you wish not to speak, you can say ‘no opinion.’”
Adapting the equality rule from Deliberative Polling (
The pretest questionnaire contained several items regarding demographic information and attitudes toward the abortion ban. Gender, ideological orientation, political interest, interest in gender issues, and political knowledge were measured and treated as control variables (see
Particularly, the deliberativeness of a political discussion was measured through a content analysis of the discussion scripts in the chat rooms from a total of 6 sessions. (See
The study tested whether the conditions of deliberation, i.e., cross-cutting exposure and the equality rule moderated in discussions, meaningfully enhance the deliberative quality of discussion and promote gender-related political participation and civic attitudes.
We explored whether encountering contrary opinions influences the deliberativeness of gender-related discussions (RQ1), and whether enforcing the rule of equal participation moderates the effects of cross-cutting exposure (RQ4). To test these inquiries, a series of analyses of covariance (ANCOVA) on rank-transformed measures
Effects of cross-cutting exposure and the rule of equality on the deliberativeness of abortion discussion.
Rationality | Civility | |||||||
---|---|---|---|---|---|---|---|---|
η2 | η2 | |||||||
Cross-cutting exposure (A) | 2 | 0.96 | .05 | .393 | 2 | 1.11 | .06 | .342 |
The rule of equality (B) | 1 | 5.08 | .12 | .030* | 1 | 2.25 | .06 | .142 |
Interaction (AXB) | 2 | 3.24 | .15 | .050* | 2 | 2.50 | .12 | .096^ |
Error (S/AB) | 37 | (166.82) | 37 | (184.72) |
**
Results for RQ1 indicate that the effects of cross-cutting effects on deliberativeness were limited; cross-cutting exposure did not entail a significant increase in rationality,
On the other hand, the effects of interaction between the level of egocentric diversity and the provision of equal chances of speaking (RQ4) were found modestly significant for the deliberative qualities of the discussion. Positive effects of the rule of equality on rationality were most prominent when the level of cross-cutting exposure was the highest, while the effects were reduced in more homogeneous conditions,
Rationality in different deliberative conditions.
We investigated whether exposure to opposing opinions toward the abortion issue and application of the rule of equality in such discordant settings can influence gender-related participation and civic virtues such as empathy and tolerance toward the opposing parties. The results are summarized in Table
Effects of cross-cutting exposure and the rule of equality on political participation and civic virtues.
Political participation in gender-related activities | ||||||||
---|---|---|---|---|---|---|---|---|
η2 | ||||||||
Cross-cutting exposure (A) | 2 | 0.33 | .02 | .724 | ||||
The rule of equality (B) | 1 | 5.04 | .12 | .031* | ||||
Interaction (AXB) | 2 | 4.56 | .20 | .017* | ||||
Error (S/AB) | 37 | (134.90) | ||||||
Cross-cutting exposure (A) | 2 | 6.50 | .26 | .004** | 2 | 0.69 | .04 | .510 |
The rule of equality (B) | 1 | 0.10 | .003 | .755 | 1 | 1.29 | .03 | .264 |
Interaction (AXB) | 2 | 0.06 | .003 | .944 | 2 | 0.63 | .03 | .538 |
Error (S/AB) | 37 | (167.51) | 37 | (167.85) |
**
RQ2, which deals with the effect of cross-cutting exposure on political participation with regard to gender issues, did not entail any directional findings. Hearing the other side on the abortion issue did not significantly promote nor undermine political activism. On the other hand, the effects of interaction between the two deliberative conditions on political participation (RQ5) were statistically significant,
Gender-related political participation in different deliberative conditions.
Results for RQ3, regarding the improvement in civic virtues through cross-cutting exposure, did not support the normatively assumed benefits of deliberation. Although cross-cutting exposure influenced empathy,
We explored how some of the key normative conditions from deliberation theory can promote empirical benefits in the context of conflictual discussion over a gender issue via an online messaging application. Several research questions regarding the effects of cross-cutting exposure and the rule of equality on 1) the deliberativeness of gender-issue discussions, 2) participation in gender-related political activities, and 3) civic virtues such as empathy and tolerance and were proposed.
Experiencing a higher level of disagreement did not automatically transfer to a better quality of political discussion. However, the equality rule functioned as an important moderator: A greater level of cross-cutting exposure led to a higher level of rationality when the rule of equality was enforced. Individuals were more eager to deliver their arguments with the reasons that are relevant to the issue when equal opportunities to deliver their opinion were given. Further, our findings suggest that benefits from the interaction of the two aforementioned conditions are largely driven by the discussants exposed to the greatest level of disagreement while speaking. The equality rule helped these discussants argue with others in a more rational way. In this sense, a structured intervention to provide equal chances to speak for everyone may be crucial in relatively mixed-opinion interactions for ensuring participants, especially those who hold minority opinions, that they are being treated as reasonable discussants in a fair manner.
When the participants experience disagreement, they may react in two different ways: they would either actively persuade others on plausible grounds or rather remain silent or indifferent. When the other parties’ opinion is deemed as the majority and the discussed issues are socially controversial, the latter scenario seems more compelling (
Similar patterns arose regarding political participation related to gender issues. As stated in a recent meta-analysis (
For civic virtues, the role of the equality rule was found limited; it did not appear as a critical moderator for the effects of hearing the other side. Whereas encountering conflicting opinions decreased empathy toward the opposing side and failed to cultivate tolerance for disagreement, additionally enforcing the rule of equality did not make any meaningful changes.
While academics have long debated on the extent to which cross-cutting exposure is normatively meritorious (e.g.,
This study is not without limitations. We did not precisely illuminate the possible mediating process in which the key deliberation conditions influence democratic outcomes through the deliberativeness of political discussion. Furthermore, our sample displayed a relatively high level of interest in gender issues (
The equality rule in an online deliberation should be more conceptually refined and contextualized. There is a possibility that the equality rule in online settings might backfire because it is unnatural in online settings. From debriefing sessions, we also found that it is necessary to consider the unique characteristics of a given online platform in enforcing specific rules for deliberation. Another remaining question pertains to a more theoretical issue, namely, the desirable degree of cross-cutting exposure. The limited effects of cross-cutting exposure observed in this study may be due to the vague explication of the desirable level of hearing the other side. In this sense, more studies should be ensued to identify the optimal level of diversity one should encounter during a political discussion.
The data reported in this paper and additional supplementary materials, including the questionnaire, coding instructions for a content analysis, and results for supplemental analyses, can be found in each link. The supplementary files have also been cited in the main text.
We would like to thank the anonymous reviewers who suggested to use ANCOVA on ranks to address the problem of having a small sample size in our study.
The authors have no competing interests to declare.