Abstract
Our study uses a qualitative analysis of social media discourse on a Facebook page to demonstrate how the phenomena of affective polarization and dehumanization are manifested through participation in a homogeneous enclave, or echo chamber. We employ Northrup’s theory of identity in intractable conflict to show how users express their desire for psychological and physical separation from the other and use dehumanizing language that normalizes potentially dangerous levels of hatred during their participation on a Facebook page. This study contributes to our understanding of the link between identity, affective polarization, and dehumanization.
Social media has transformed the way we receive, consume, and produce information as well as form opinions in the social arena. Predictions of the impact of technology on democracy have included a utopian view of the internet as a virtual town hall (Papacharissi, 2002); a dystopian view of technology as leading to greater distance, verbal aggression, and the creation of echo chambers (Matuszewski & Szabó, 2019); and a more neutral view positing it is not the technology itself, but how it is used that matters, for example: [T]he Internet functions as an amplifier: It increases our potential for good and productive work as well as for inappropriate and immoral endeavors . . . Ultimately, a technology is only as useful as the intentions of its users. (Banschick & Banschick, 2003, p. 161)
Media scholars point to the importance of balancing social media participation to include interaction with both homogeneous and heterogeneous groups (Sunstein, 2017); yet research confirms the existence of echo chambers (Matuszewski & Szabó, 2019; Wolleback et al., 2019), increased polarization of political groups (Stroud, 2011; Tappin & McKay, 2019), and increased dehumanization of the other (Cassese, 2019; Pacilli et al., 2016). These trends have serious consequences for dialogue and democratic deliberation across important social and political issues. While social media participation may expose users to diverse viewpoints (Bakshy et al., 2015), exposure to alternative views may actually encourage social media users to seek out sources that validate pre-existing beliefs rather than engage in deliberation and reasoned civil discourse (Stroud, 2011; Wolleback et al., 2019). Indeed, studies of affective polarization empirically support that members of political groups tend to emphasize the distance between their in-group and members of the out-group.
Studying political polarization in the United States and based on a social identity perspective, Iyengar and Westwood (2015) defined affective polarization as “the tendency of people identifying as Republicans or Democrats to view opposing partisans [out-group members] negatively and copartisans [in-group members] positively” (p. 691). Their research confirms that affective polarization across parties has increased over time in the United States (Iyengar et al., 2019; Iyengar and Westwood, 2015). Confirming evidence of affective polarization in the United States and other countries further demonstrates that political party members express their desire for social and moral distance from the out-group and this can be connected to tendencies toward dehumanization (Cassese, 2019; Pacilli et al., 2016; Tappin & McKay, 2019).
Scholars have recently called for increased attention to affective polarization and dehumanization in the political arena, as traditionally these have been examined in the context of ethnic, race, and gender identity (Cassese, 2019; Pacilli et al., 2016; Tappin & McKay, 2019). The purpose of this study is to examine social media communication (Facebook posts and comments) to illustrate how the discourse in a homogeneous enclave, or echo chamber (Stroud, 2011) reveals affective polarization and dehumanization. Muddiman and Stroud (2017) demonstrate that social identity theory explains uncivil discourse along partisan lines in online comments. We contribute to this research linking social identity, uncivil discourse, and affective polarization by using Northrup’s (1989) theory of identity in intractable conflict to show how a homogeneous group of Facebook users demonstrate their desire for psychological and physical separation from the other and use dehumanizing language that normalizes potentially dangerous levels of hate during their participation on a Facebook page. Affective polarization and dehumanization pose a threat to civil discourse and dialogue, as this communication reduces the willingness to engage in productive discourse with political opponents needed for a healthy democracy. While we see no easy or obvious way that one can employ to reverse this trend, we use Northrup’s (1989) ideas about de-escalating intractable conflict to provide modest suggestions for moving identity groups toward empathy and away from dehumanization.
Literature Review
In this section we describe previous studies that have examined political discourse on Facebook and other media platforms and the formation of echo chambers, affective polarization, and dehumanization. We then introduce and describe Northrup’s theory of identity in intractable conflict as the framework for our case study—an analysis of political discourse among members of a Facebook page that was identified as a leading online platform in propagating hate-speech: “The Shadow.”
Social Media as a Digital Town Hall?
While some observers optimistically heralded the internet and social media platforms as providing spaces for the sharing of multiple views and public opinions, several studies call this view into question. Given relatively low political participation overall, it comes as little surprise that young people in the United States typically avoid politics on social media (Thorson et al., 2014). Yet even in Israel, where political discourse is the norm, Mor et al. (2015) found that young people choose their language carefully to conduct impression management and protect themselves from negative repercussions.
Furthermore, when people do engage in political communication on Facebook, there is little evidence that it results in the debate of ideas. A study of politically active Facebook users found that users were less likely to participate when they perceived their friends as having different views, and weak tie relationships were often damaged by political interaction on Facebook (Grevet et al., 2014). Supporting the latter point, a study of Israeli Facebook users found that users were more likely to unfriend someone with opposing political views (John & Dvir-Gvirsman, 2015). Even when Facebook users self-select into a group designed for cross-cultural dialogue, at least one analysis demonstrates that users are more likely to respond with defensiveness than to engage in listening and increased understanding (Mor et al., 2016).
Beyond Echo Chambers to the Normalization of Hate Speech
Sunstein (2017) presents a cogent description of the polarization effect, in which participating in deliberative groups is shown to move people from more moderate positions to the polarized extreme. He shares the result of one study which found “Before members started to talk, many groups displayed a fair bit of internal disagreement. The disagreements were reduced as a result of a mere fifteen-minute discussion” (p. 67). If this dynamic is true of face-to-face groups, it is not difficult to see how this can happen in online settings, given the above-described tendencies toward selecting social media conversations and networks that support preexisting views.
Recent studies describe how participation in comment forums can escalate users to anger, which, when validated by other users, reinforces echo chambers, incivility, and discourses of hate (Baider, 2018; Ermida, 2017; Muddiman & Stroud, 2017; Wolleback et al., 2019). In a study of online comments posted in response to news stories including reference to the LGBT (lesbian, gay, bisexual, and transgender) community, Baider (2018) illustrates how these online comments express, shape, influence, and reinforce Greek Cypriot homophobic hate speech. Ermida’s (2017) linguistic study on online comments to a British newspaper story about a family’s request for welfare support further demonstrates how such comments can serve as a platform for expressing anger, contempt, discrimination, and prejudice. In an examination of the roles of anger and fear in social media use, Wolleback et al. (2019) found that the experience of anger, such as described by Baider and Ermida, furthers “chamber dynamics and trench warfare dynamics in the digital public sphere” (p. 1).
In a similar vein, several empirical studies have also shown that when political identity is salient and there are perceptions of moral distance from the out-group, animalistic dehumanization is more likely to occur in both Italian (Pacilli et al., 2016) and US (Cassese, 2019; Tappin & McKay, 2019) contexts.
These studies lead to a more pessimistic perspective on social media as a platform for civil discourse. If political parties are increasingly polarized, individuals are turning to validating media sources, and there are increasing perceptions of moral distance that lead to dehumanization, this trend does not bode well for meaningful engagement with the diverse views and voices necessary to a healthy democracy.
This review suggests that when social media users encounter alternative views, a common tendency is to feel threat to one’s views and identity, leading to defensiveness, either in the form of avoidance (John & Dvir-Gvirsman, 2015), attacking the other’s views (Mor et al., 2016), and/or affective polarization (Cassese, 2019; Iyengar & Westwood, 2015). These dynamics are consistent with a theory of the role of identity in intractable conflict (Northrup, 1989). In fact, the theory amplifies the concerns that social media participation may be a key influence on the normalization of hate speech, not only threatening deliberative democracy but contributing to increased affective polarization and dehumanization with a concerning potential for harm.
Northrup’s Model of Identity in Intractable Conflict
Terrell Northrup (1989) defines intractable conflict as prolonged conflict that is resistant to resolution and characterized by conflict-intensifying features unrelated to the initial issues in dispute, attempts by each party to harm the other, and poor or nonexistent communication between the parties (p. 62). Northrup suggests threats to identity are central to the process of conflict escalation toward intractability. When conflict involves a serious threat to our identity, it is likely to stimulate a defensive response and sets the stage for intractability.
Northrup describes that our response to identity threat depends on the nature of the threat and the source of the information. The information is easier to distort or dismiss when the source is an out-group member rather than a member of our own identity group. The motivation to protect our identity arises from “a fear of continued but meaningless or powerless existence, [which can be] as threatening to the self as a physical threat” (Northrup, p. 65).
Northrup (1989) identifies four stages of intractable conflict: threat, distortion, rigidification, and collusion. As conflict progresses through each stage, de-escalation becomes less and less likely. The four stages are described below.
Threat
A threat occurs when a core sense of identity is criticized or invalidated. One party may criticize or condemn the other’s beliefs, actions, or values or each party may behave in ways that deny the legitimacy of the other. When information threatens our self-definition and identity, we adopt a defensive stance, resulting in movement to the next stage of escalation.
Distortion
The response to threatening information is often denial or distortion. One of the most common distortions is assimilation, in which new information is altered so as to be consistent with current beliefs (Wilson et al., 1992). An example would be a person denying the existence of global warming by noting increasingly cold winters in certain parts of the world. Another type of distortion is denying the sender’s legitimacy; for example, by suggesting the sender does not have credibility or expertise. Depending upon how threatening the information is to one’s identity, the information may be distorted to a greater or lesser degree.
Rigidification
As people encounter others whose beliefs threaten their identity, they begin to define those who are like them as a distinct in-group. This begins a concurrent process of differentiation, where people who were not initially perceived to be substantially different are now seen as the out-group. As more and more out-group members are identified, the threat to identity becomes more pronounced and there is a need to create greater distance from them, as described in affective polarization. In the rigidification stage, beliefs become fixed and more points of distinction are made such that issues that were not part of the original conflict are now seen as threatening and become central to the conflict. For example, someone who does not believe in climate change may begin to define all climate-change proponents as “liberal tree-huggers,” allowing themselves to both distort the potential credibility of the science and provide evidence of the many issues on which they disagree with the out-group.
Northrup describes several other events that occur during rigidification. Communication with the other is shut down and physical or social barriers are placed between the groups. Stories emerge that depict the out-group as stupid or lazy, while presenting the in-group as noble and virtuous. Perceived differences between groups are exaggerated while intragroup differences are minimized to present a cohesive front. At the extreme, a process of dehumanization occurs where members of the out-group are not only viewed as different, but as evil and/or less valuable (Northrup, p. 74).
Collusion
In this final phase, the conflict is maintained because the parties collude in prolonging it. The conflict has become institutionalized, even celebrated, with each side appointing certain members to roles that perpetuate the conflict (e.g., lobbyists). Each group selects leaders who adopt the most extreme positions to maintain the “party line” and ensure no concessions are made with the other group. At this stage, the conflict itself is part of each group’s self-definition and identity, and members may believe “as a member of this group I must hate the other group.” Northrup notes that this stage carries with it the ultimate paradox—the parties are actually cooperating in the perpetuation of the conflict.
The danger of rigidification and affective polarization is the concomitant move to dehumanization. History is replete with examples of how dehumanization allows humans to engage in atrocities against each other, such as Bromley (2011) recounts in his analysis of the genocide in Rwanda. Baider (2018) further points to the danger of ideological language that “feed[s] intense emotions that in turn affect social relationships and allow violence against the others, who are constructed as an out-group” (p. 76). Given these concerns, the aim of our study is to examine the discourse of affective polarization that takes place within a society embedded in an intractable conflict. Our study contributes to previous research by examining the discourse through which the internal escalation and fragmentation within a given society is expressed and further propagated. We focus on a highly popular extreme right-wing Facebook page in Israel, “The Shadow,” with the aim of examining the following research question:
RQ1. To what extent and how does the discourse of comments on the Facebook page “The Shadow” demonstrate affective polarization and dehumanization through statements of threat, distortion, rigidification, and collusion?
Method
Research Site
Our study focuses on a Facebook page that was identified both in 2016 and in 2017 as the leading digital platform in Israel in propagating hate speech: The Shadow. This page was founded on 28 April 2012 by Israeli rap singer, Yoav Eliasi, known by his stage name “the Shadow.” Since 2014, during the days of Operation Protective Edge in the Gaza strip, the Shadow became less famous for his music and more for his online political activism, which is associated with an extreme right-wing agenda. The Shadow’s Facebook page is highly popular, with 418,513 people liking it and 447,498 following it, as of December 2019. In 2017, when the study was conducted, the page was rated in a list of the top 20 most interactive Facebook pages in Israel, in terms of followers’ engagement.
Sampling and Data Collection
We examined comments written by page followers during two week-long periods in 2017, between the 14th and 21st of April and between the 8th and 15th of December. The posts during these periods dealt with current events that had appeared in the news, prevalently focusing on violent incidents related to the Israeli–Palestinian conflict.
The first week in our dataset was chosen as a convenience sample for a research project conducted by the first author. The second week was then chosen as a validity check to determine whether similar content and themes would be found. An initial thematic exploration of the data guided by the grounded theory approach (Strauss & Corbin, 1998) suggested that similar themes appear in the posts and in comments during both periods. Thus, we combined the two data sets into one corpus for further analysis and coding.
The sampling process began with retrieving every post and comment published during the two periods mentioned above, using the Netvizz application (Rieder, 2013). During both periods, a total of 194 posts were published, and they received a total of 51,126 comments (see Table 1 for more details). Because of our interest in discourse specifically referring to members of the left wing, we conducted searches for the keywords: “left” or “leftist,” prevalently used in Israeli discourse to describe left-wing Israeli Jews (Shamir et al., 2017). We searched for these words using their correct spelling and other unintentional or intentional misspelling that are sometimes used as derogating phrases against left-wing supporters (e.g., the use of the number zero instead of one of the letters, a misspelling that has a negative connotation). This process left us with 171 posts and 3,108 comments, written by 2,048 users. As a final step of refining our corpus, we went through the comments and removed repetitions of the same comment and comments that contain fewer than four words. Following this procedure, we had a total of 2,870 relevant comments for analysis.
Average Metrics of Followers’ Engagement During the Sampling Periods.
Data Analysis
A qualitative, textual analysis guided by Northrup’s model was conducted to determine whether comments revealed evidence of threats to identity, distortion, rigidification, or collusion. The process began with an exploratory coding of comments from the research corpus according to Northrup’s four stages of intractable conflict. This procedure provided preliminary evidence for the relevance of Northrup’s model to our data. In the second step of our analysis, a subset of comments from the research corpus was translated into English and shared with the second author. Each comment was then discussed by the first two authors until an agreement was reached on the main criteria for categorizing it into one of Northrup’s model categories. Comments that fit more than one category were further discussed until an agreement was reached on their classification based on the most prominent claim of the comment. The first author also used a thematic process of analysis to determine whether there were sub-categories within Northrup’s model as well as disconfirming cases, which were discussed with the second author. This last step is recommended to reduce researcher bias and increase validity of the findings. These discussions also led to further refinement of the classification criteria until a saturation point was reached (Strauss & Corbin, 1998, p. 143) at which no further criteria for categorization were identified. This process resulted in an agreed-upon set of coding guidelines that were used to code the remainder of the data. An inter-coder reliability test performed by two native Hebrew-speaking trained independent coders, on 290 comments (10% of the entire corpus), achieved satisfying scores (Krippendorff’s α > .7).
It is important to note that this case study focuses on one political-ideological group: right-wing Israeli Jews. By presenting the examples that follow, we do not intend to imply that this group is more polarized or engages in more dehumanization than other political-ideological groups. We address this limitation of our dataset further in the final discussion.
Results
Our analysis reveals evidence for the first three stages in Northrup’s model: threat, distortion, and rigidification—all apparent in followers’ comments on the Shadow Facebook page referring to left-wing Israeli Jews. More specifically, we present below comments portraying left-wing Israeli Jews as a threat to the State of Israel and even the world, comments that illustrate two categories of distortion: distorting information and denying the legitimacy of the other; and in line with Northrup’s theory and with affective polarization, two forms of rigidification have emerged from our data: comments expressing a desire to further exclude left-wing Israeli Jews from the in-group and comments conveying dehumanization. We also found limited but clear evidence for the fourth stage of collusion in our data: in several cases, members of the left wing commented on the Shadow Facebook page. These comments were met with right-wing responses that made it clear that there are other places for left-wing individuals to voice their views, and there is no interest in further interaction or discussion with them in this venue.
All original posts and comments were written in Hebrew and translated into English by the first author. We include the date of each comment and the number of likes it received for additional context.
Threat
Consistent with the first stage in Northrup’s model, the analysis of comments from our sample reveals that page followers define Israeli leftists as a threat to Israel. This notion is demonstrated in the comments below describing left-wing Israeli Jews as harming the country, leading it toward its end, and even as the enemy:
The leftists are our devil, because of their existence the country is being destroyed and the army weakened (13.12.2017; 1 like)
Whoever votes for the left-wing will bring about the end of the state (10.12.2017; 1 like)
There are those who have still not yet internalized that the left is the most dangerous enemy (10.12.2017; 4 likes)
May God take all the leftists, all the troubles in this country come from them. They give terrorists free reign (19.04.2017; 1 like)
Leftists are the main reason for the legitimization of terror in the world, and in particularly in Israel . . . (10/12/17)
The above comments show a strong link between presenting the other as a threat and the tendency to convey generalized distorted information about the other as exemplified very clearly in the claim “Leftists are the main reason for the legitimization of terror in the world.” This leads us to the second category in Northrup’s model, distortion.
Distortion
Consistent with the second stage of Northrup’s model, our analysis revealed a consistent pattern of distortion that appeared in two main forms: distorting the information or ideology of the other and denying the legitimacy of the other.
Distorting the Information or Ideology of the Other
The comments below illustrate distorted information by stating that leftist Israelis side with the Arabs and want to turn Israel into a Palestinian State. While we cannot know the full intentions or ideological stances of the Jewish-Israeli left wing, these comments depict them as universal when we know that people will vary along any ideological stance. Furthermore, it is unlikely that those in the left wing are anti-Semitic, desire to turn Israel into a Palestinian State, or want to destroy Israel, as they are accused of in the statements below. Finally, and unsurprisingly, these claims and assertions are not supported by public opinion–based research on attitudes of left-wing and right-wing Israeli Jews (see, for example, Nagar & Maoz, 2017; Shamir et al., 2017).
The left is full with conspiracies to turn the state into a Palestinian state . . . their punishment will be worse than the punishment that will be given to terrorists (10.12.2017; 1 like)
I will never understand how one could be a leftist, to be in favor of the Arabs in the state of Israel! (12.12.2017)
This is the true face of the leftists . . . the hatred eats them. They love the Arabs . . . they hate the right wing, this is the situation . . . In the elections for the Knesset, we will not forget and we will not forgive . . . (10.12.2017)
Our problem in this country is not the Arabs; it’s the leftists who are interested in the destruction of Israel! (10.12.2017; 5 likes)
The anti-Semites from the left-wing do not need proof of anything; they are just waiting for the destruction of the State of Israel!! (20.04.2017)
Denial of Legitimacy by Attributions of Mental Illness or Stupidity
The comments below convey delegitimization of leftists by referring to them as mentally ill or merely stupid. Leftists are presented as people with no logic who are unable to understand reality as it is, and thus their opinions should not influence Israeli society:
You’re right, how did we not think of that? Wow, being a leftist is simply a mental problem! I believe that in a few decades, they’ll recognize people who encourage those who want to slaughter them as mentally handicapped and they’ll be medically treated, as opposed to what’s happening today, where they run the media (12.12.2017)
The left has completely lost it, they need to take medication (14.04.2017; 4 likes)
The leftists are blind and are committing suicide. They want to take us with them in their death wish. It is important to be careful of these psychos . . . (18.04.2017; 2 likes)
I don’t know if I really want to know the answer to the question of whether the thinking of the left is due to infinite stupidity or infinite naivete (both possibilities scare me) (17.12.2017)
Ariel my leftist friend, when will you finally understand that they want you and me under the ground, when the hell will you understand? Why don’t you have the intelligence and brains to understand that you, the left, are making a terrible mistake, when will you understand, when??? (17.12.2017)
You stupid idiot, why didn’t God give you a brain, stupid leftists, why?! 
(10.12.2017; 2 likes)
The patterns revealed above of threat, distortion, and delegitimization further escalate into exclusion and dehumanization as shown in the next category, rigidification.
Rigidification
Consistent with the third stage in Northrup’s model, our data revealed two sub-categories of rigidification that are presented below.
Separation and Exclusion
The first sub-category includes comments that convey the desire for separation and exclusion, prevalently by implying that leftists are not really Jews (i.e., are not part of the same identity group) and therefore should not be allowed to remain in Israel.
Leftists, come on . . .when will you understand that leftists aren’t Jews? (20.04.2017)
Us? Stupid leftist, you’re not a part of “us” . . . you have nothing no connection to the Jewish people (08.12.2017; 1 like)
First we’ll imprison all the leftist Jews, although I’m not even sure that they are Jews, because if they were they would have some emotion toward the country and to what’s happening around them . . . they incite and cause trouble to our soldiers and our country . . . wake up before it’s too late (08.12.2017; 5 likes)
There’s no need to get excited by what the left says/complains about, etc. To be a leftist in Israel is bordering on mental illness, or simply just not being a Jew. And that’s fine, we just don’t need to pay any attention to them, or give them rights, an ID, or any political economic or legal support. They should be removed from the country, from both its borders and its rights, for the reason that they are not citizens of the state of Israel (20.04.2017; 198 likes)
I’m in favor of deporting all the leftists and denying all their rights in the country (10.12.2017; 6 likes)
The harsh exclusion and even support for denial of rights expressed above is accompanied by animalistic dehumanization of left-wing Israeli Jews.
Dehumanization
The second sub-category of rigidification includes comments that dehumanize left-wing Israeli Jews by comparing them to dogs, insects, or a disease:
These filthy dogs get a lot of money from hostile elements to slander their own people. May the name and memory of all the traitors of Israel be blotted out. We should ask our elected officials to invoke a state of emergency, with the appropriate laws, so that we can take care of our enemies, including the leftists (10.12.2017)
A bunch of dogs all the leftists are traitors (10.12.2017)
The country needs urgently pest control . . . against the leftist cockroaches . . . microbes . . . vermin . . . left-wing viruses . . . and also a root canal treatment and uprooting of the leftist and of the Palestinians that are like terminal cancer . . . This is something only Putin can do!!! (9.12.2017; 1 like)
Another proof that . . . the anti-Semitic mentally ill left-wing . . . riddled with self-destruction . . . supporter of terrorism and fascist Islam . . . is the enemy!! Please treat the treacherous and stinking worms as they deserve! (21.04.2017; 30 likes)
Collusion
Our dataset included several comments made by left-wing Israeli Jews. The Shadow Facebook page followers’ responses to such comments were consistent with Northrup’s fourth stage of collusion. As shown in the quotes here below, the page followers demonstrated their commitment to maintaining the conflict and the clear separation between the two groups also on the concrete behavioral level—making it clear that left-wing participation is not welcome on this page:
This page is not for you . . . you can go and write on a leftists page, your response is not interesting . . . monkey (20.04.2017; 2 like)
Lior, the fact that you don’t like the Shadow’s opinion does not make it an incitement! You don’t want to see it? Then shut up and go and comment on the leftist pages that you like, this is a rightist page not a leftist.! you bored provocateur! Delusional and ungrateful leftist! Because of people like you terrorists get the inspiration to commit terror attacks and murders every day! You are just poor and ungrateful, what the Shadow tries to do (to open your blinded eyes) you won’t be able to do even in 500 years from now! (19.04.2017; 3 likes)
Disconfirming Case
Although the majority of comments in our sample demonstrate the escalation of conflict, there are a few comments—such as the comment here below—that reflect a more inclusive attitude toward the left and higher acceptance of a diversity of opinions.
You are stupid, the leftists want this country just as much as you do, they just believe in a different way, is this a reason to wish them murder? (19.04.2017; 16:41; 2 likes)
However, like other such comments, this comment triggered unsupportive reactions, as shown in the following thread of responses:
These leftists love it when somebody hurts them (19.04.2017; 16:44; 6 likes)
Oren, the leftists harm the country (19.04.2017; 16:58; 2 likes)
The leftists are our 11th plague [i.e., Plagues of Egypt] (19.04.2017; 17:11; 6 likes)
Discussion
The purpose of this study was to examine a case of a Facebook page whose followers include extreme right-wing members of Israeli society to examine to what extent and how social media discourse in homogeneous enclaves—echo chambers—conveys the connection between identity, affective polarization, and dehumanization within a political identity group. Using Northrup’s (1989) theory of identity in intractable conflict contributes to existing studies of these effects by demonstrating how the language of the Shadow’s Facebook page followers reflects stages of escalation, including perceptions of threat, distortion, rigidification, and collusion that delegitimizes and dehumanizes the out-group. Below, we elaborate our contributions by describing how the discourse of this Facebook page socially constructs the other and poses a threat to civil discourse and democratic deliberation. We conclude by describing Northrup’s ideas for de-escalating intractable conflict to modestly propose ideas for shifting parties toward empathy and away from dehumanization.
Northrup’s Stages of Intractable Conflict: Social Construction of the Other
The depiction of the other illustrated in the Shadow’s Facebook page shows how members of an in-group socially construct the other as a dangerous threat. In this case, followers of the Facebook page describe the left as those who want to bring about the destruction of Israel. Notably, Shamir and Sagiv-Schifter (2006) point out that leftist Jews are seen as Palestinian sympathizers, and are therefore perceived as being farther from the in-group and are often targets of out-group hate.
The Shadow Facebook communication illustrates distortion in the denial of the legitimacy of the out-group. The language of the Shadow users’ comments describes those on the left as “anti-Semitic” and “Arab-loving.” Such comments employ the simplistic dualism “if you are not with us, you are against us,” and dismiss the possibility of having empathy for an oppressed group. Distortion is further evidenced in the Shadow users’ delegitimization by suggesting that leftist Jews are all either mentally ill or stupid. As predicted in studies of affective polarization, this distortion releases the Shadow followers from the need to interact directly with members of the out-group as seen in the words of one comment: “I don’t think that eyes of the left in Israel can be opened, it’s just hard for me to believe that that could happen . . . and it’s a shame.” This demonstrates how affective polarization can reduce the opportunity for civil discourse.
It is easy to see how delegitimizing the other as mentally ill, stupid, or otherwise incapable of reason allows the users of the Shadow Facebook page to separate themselves further from the left. While these data do not directly show increased affinity for the in-group, affective polarization is displayed in comments such as the following: “Us? Stupid leftist, you’re not a part of ‘us’ . . . you have no connection to the Jewish people.” The rigidification goes further with the separation, not just maximizing the psychological distance from the other, but calling for their complete annihilation. As has been described by others (Volkan, 1988), dehumanizing the other makes their extermination justifiable, as in the comment:
These filthy dogs get a lot of money from hostile elements to slander their own people. May the name and memory of all the traitors of Israel be blotted out . . .
The ease with which the Shadow Facebook followers call for extinction of human life raises concerns regarding the normalization of hatred in some social media sites.
One interesting counterpoint to dehumanization must be noted. In two comments in our sample, a member of the left has chosen to participate in the Shadow Facebook discussion. In both cases these comments generate responses from the Shadow followers, yet in contrast to dehumanizing them, the two responses we highlight here use the person’s first name in the response (comments are abbreviated for illustrative purposes below):
Lior, the fact that you don’t like the Shadow’s opinion does not make it an incitement!
Ariel my leftist friend, when will you finally understand that they want you and me under the ground, when the hell will you understand?
It is noteworthy that in the second example, the responder even includes the phrase “my leftist friend.” The use of these out-group members’ names suggests that it is more difficult to dehumanize another when we are speaking or interacting with them directly. This may provide some suggestions for de-escalation of intractable, identity-based conflict.
Implications: Movement Toward Empathy?
Admittedly, the literature described above in conjunction with the comments we found on the Shadow Facebook page do not paint an optimistic picture. In this section, we briefly describe Northrup’s three levels of intervention and how they might apply to political communication in the realm of social media.
Northrup’s Levels of Intervention in Intractable Conflict
Northrup (1989) describes three levels of intervention to help de-escalate intractable conflict rooted in identity. A Level 1 intervention is one in which a powerful third party forces a change in circumstances by calling for a temporary suspension of interaction or, in the case of war, a cease-fire. This is usually the role of the United Nations or a political figure with the clout or authority to help broker at least a temporary treaty. In the case of social media, the intervention would likely come from the platform managers, such as shutting down a Facebook page. We do not necessarily advocate this action, as offensive content is in the eye of the beholder, and shutting down a site would seem to take a side and possibly reinforce the conflict and existing power differences (see, for example, de Vries et al., 2015). Nonetheless, there is an interesting question regarding whether sites such as described here are normalizing hate speech, and whether social media should play a role in reducing this normalization by invoking a policy precluding dehumanization or calls for violence against another entire identity group.
Northrup describes Level 2 intervention as a change in the communication patterns of the parties in an intractable conflict. Attempts at Track 2 negotiation between Jews and Arabs (Benziman, 2016) and reconciliation-aimed intergroup encounters (Ron & Maoz, 2013) are examples of this intervention. According to Northrup, changing the way parties interact might enable them to see each other differently, promoting listening, understanding, and reduction of seeing the other as a threat. For example, a study by Shnabel et al. (2013) demonstrated the effectiveness of an intervention that decreased competitive victimhood and increased forgiveness among Israeli Jews and Palestinians by inducing perceptions of both groups as common victims of an external threat. A second-level intervention in social media might promote direct engagement among members of different groups. As shown in our dataset, while the Shadow Facebook followers did not welcome leftist contributions, they also did not engage in dehumanization during direct interaction, as revealed by the use of first names in their response. While these responses fell short of constructive, previous research has shown that creating an online space where opposing parties engage with each other directly might curtail dehumanization (Kraut & Resnick, 2012). One example of a Facebook page that is currently promoting dialogue among diverse groups and communities in Israel is a site called “Dibur Acher” (different talk). On their “About” page it reads, Dibur Acher is an initiative which started as unique cooperation between Google Israel [and two Israeli NGO’s] . . . and its mission is to impact online discourse in Israel by creating and encouraging new initiatives in the digital sphere, which promotes an alliance between different communities in the Israeli society. (Authors’ translation from Hebrew)
While this page has only 5,663 followers (as of December 2019), there are several nongovernmental organizations that work toward dialogue and peace in Israel, and these may be additional sources for hosting facilitated online (and offline) interaction.
Another possible Level 2 intervention on social media comes from the design of strategic online interaction. Grevet et al. (2014) point out that while social media algorithms typically reinforce echo chambers by highlighting Facebook friends with common views, scholars of computer-supported collaborative work and human–computer interaction are investigating novel ways to design algorithms that expose users to diverse information and views. While existing studies do not support that exposure to diverse views increases understanding and openness, Grevet et al. (2014) demonstrate how Facebook can use algorithms to connect people who might not typically appear in one’s newsfeed, and this could be done strategically to highlight commonalities instead of differences, and by extension, could perhaps reduce affective polarization.
Northrup describes Level 3 intervention as one that occurs over time, given successful Level 2 intervention. The presumption is that at this stage, the parties’ identities have transformed such that they no longer see the other as a threat and maybe even see themselves as co-members of an identity group. Significant work needs to be done at Level 2 to reach this goal, which suggests movement toward empathy and away from dehumanization. Such work can help promote civil discourse and deliberative democracy.
Limitations and Directions for Future Research
One significant limitation of this study is its focus on one Facebook page founded as a place for extreme right-wing Israeli Jews to express their views and values. It should be thus emphasized that we do not assume that the followers of this page reflect all those who identify with more hawkish (right-wing) positions in the Israeli–Palestinian conflict, and we acknowledge the posts and comments on this page are not typical of Facebook users as a whole. The study is limited in that we were specifically interested in the phenomena of affective polarization and dehumanization, and this influenced our sample selection. Finally, it is important to note that we do not include a parallel Facebook page from the opposing political group in our analysis, and in this omission, we do not intend to convey one side as more engaged in dehumanization or hate than the other. On the contrary, we emphasize here the need for a further study that will also examine posts and comments on media platforms identified with left-wing Israeli Jews. While the findings from this study alone are not generalizable to social media as a whole, when examined in the context of the other studies reviewed here, it supports the claim that participation in online and social media may be more likely, in certain conditions and cases, to be used in the interest of self-validation, reinforcement, and polarization of existing beliefs than to expose ourselves to diverse views and values as called for by Sunstein (2017) and other proponents of deliberative democracy.
Future research should continue to examine mediated and face-to-face spaces in which direct interaction with others and exposure to diverse views is happening. We need to understand how individuals can be motivated to attend to diverse information and opinions as well as directly engage with those with opposing identities and viewpoints to inspire greater understanding and empathy and reverse the trend toward the normalization of dehumanization and hate. We encourage broader and deeper investigation of how social media platforms encourage or combat affective polarization and ethnographic accounts of successful dialogue on social media sites.
Footnotes
Acknowledgements
The first and the third author would like to thank the Smart Institute of Communication and the Truman Institute, both at the Hebrew University, for the support of this research project. The second author would like to thank the Lady Davis Foundation for their support, which made her time at Hebrew University possible.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship and/or publication of this article: This work was partially supported by the second author’s Lady Davis Foundation Fellowship.
