Does information matter? Online discourse on the Yemenite children’s affair in Israel after release of archival documents (2024)

Abstract

Purpose

This study evaluates how publicly available archival documents shaped online discussions about allegations that thousands of children were kidnapped during the 1950s in Israel, known as the Yemenite children’s affair. It examines if access to historical records leads to more informed and rational public discourse, especially on social media.

Design/methodology/approach

Using content analysis, this study examines Facebook posts from media outlets, politicians, NGOs and public groups between 2016 and 2021 to understand how the Israeli State Archives’ release of over 300,000 documents affected support of the kidnapping.

Findings

Despite extensive archival information debunking the kidnapping theory, public opinion and discourse largely continued to support it. This suggests a complex interaction between information availability, preexisting beliefs, echo chambers and group allegiances, suggesting that access to factual data alone may not effectively challenge established beliefs in online public settings.

Research limitations/implications

Since data were collected only from Facebook, our conclusions cannot be generalized to other platforms. The study relies only on publicly accessible data and does not establish causality between exposure to information and shifts in opinion. Our findings show that disclosing archival information does not significantly benefit public political discourse on contentious topics but also point to the advantages of mediating information by politicians, NGOs and journalists.

Originality/value

As a unique case study, this research contributes to understanding the role of historical archives in digital-age public discourse. It highlights their potential and limitations in facilitating informed debate and deliberation, emphasizing the complexity of influencing established beliefs with factual data.

Keywords

Information theory
Archives
Public sector organizations
Digital communications
History
Documentation

Citation

Peled, R. and Yavetz, G. (2024), "Does information matter? Online discourse on the Yemenite children’s affair in Israel after release of archival documents", Journal of Documentation, Vol. ahead-of-print No. ahead-of-print. http://doi-org-s-99.er.library.nenu.edu.cn/10.1108/JD-01-2024-0010

Publisher

Emerald Publishing Limited

Background

The Yemenite children affair (henceforth, the “affair”) has been a topic of heated social and political debate in Israel for more than half a century. Its origins are in the period immediately after the foundation of the state in May 1948, when the Jewish community of some 700,000 doubled its population in three years. This massive absorption of immigration took place while the state was fighting a war of independence and was under a food-rationing austerity regime (Seidelman, 2020). A little over 50,000 of those immigrants came from Yemen. They arrived in dire physical condition after an arduous journey through the desert. One-third required immediate hospitalization, among them, many babies and children (Meir-Glitzenstein, 2011).

The immigrants were temporarily housed in large, cramped tent camps. Babies were accommodated in brick barracks to protect them from the weather and unhygienic conditions. Field hospitals were erected, but they were poorly staffed and equipped. Many children perished under these conditions. Because of the physical situation and lack of communication and means of transportation as well as due to certain religious practices, children were often buried before their parents were notified of their death.

Starting In the mid-1960s, rumours began to spread among the community of immigrants from Yemen that the children reported dead had actually been kidnapped by the establishment and handed over for adoption to families of European origin (which we refer to here as the “kidnapping theory”). Lending credibility to these rumours was the fact that the national registry was poorly maintained due to the general disarray that was prevalent at the time, and as a result, some deaths were not recorded. Years later, in the 1960s, draft notices that were based on these faulty records were sent out to some families whose children would have reached conscription age and suspicions began to arise. These were then intensified by political and social groups with strong anti-government sentiments (Birenboim, 2019). Public outcry led to three cycles of investigations: a governmental committee in 1966, a public committee headed by an appellate court judge in 1988, and a national inquiry commission headed by a supreme court justice in 1995 (Levitan, 2023). Together, they reviewed complaints about the alleged kidnapping of a little over 1,000 children and found decisive documentation as to the death of 95% of them. In some of the remaining cases, investigators failed to find sufficient documentation, and a few others involved children who were given up for adoption through court proceedings. No evidence of any kidnapping was found. An additional parliamentary commission headed by a Knesset member (parliament member; MK) of Yemenite origin, who was determined to reveal the truth about the alleged kidnappings, found no such evidence either (Levitan, 2023). Several graves were exhumed and where possible, genetic samples were drawn, only to prove that the identities of the children buried there matched those that appeared in the official records. Free DNA tests that were given to the families failed to identify any living person whose parents filed a complaint stating they were falsely reported dead.

Nevertheless, suspicion towards the establishment, stoked by radical political groups from the right and left as well as academics (mostly critical race theorists), kept the kidnapping theory alive with claims that there was a large-scale conspiracy to hide the evidence (Levitan, 2023). In response, the government passed a resolution requiring the Israel State Archives to release the nearly 400,000 documents collected over the course of all the investigations, which it made publicly available online in December 2016. None of the documents yielded any evidence of kidnapping, but rather than putting the affair to rest, this only led to new allegations that the documents might have been falsified. Much of the ongoing controversy surrounding this affair has taken place online on social media. Troves of new information have been made easily accessible online and continue to feed the debate, which has only intensified. Thus, the affairs presents a unique opportunity to examine the role of information in public discourse on social media.

Literature review

The role of information in public discourse

In Habermas’s (1989) influential theory, he posited that public discourse was a core feature of modern liberal democracies. Habermas (1989) also assumed that a rational and informed civil society could agree on certain factual statements, which he argued was the basis for social coordination and consequently legitimized law. Earlier, Arendt (1967) too emphasized the necessity of a shared factual base for any political discourse. Both were acutely aware that achieving rational public discourse required overcoming significant obstacles, but their theories implied that it was possible to attain at least to some degree.

However, in recent decades, scholars have wondered whether this is indeed the case. Vast research has shown that most of the public is politically ignorant or worse, clings to proven lies (Brennan, 2016), both of which are thought to significantly threaten discursive democracy (Hannon, 2023). Public discourse has come to be seen as fundamentally defective, because of ignorance and misinformation (Lepoutre, 2021).

It is argued that some level of political ignorance is actually rational, given the costs associated with acquiring information (Somin, 2015) and its limited use to citizens (Brennan, 2016). Even more surprising, it has been shown that the effects of being completely uninformed might be preferable to those of having only piecemeal information that merely confirms previously held biased beliefs. It is only the few most highly informed that seem to respond to it rationally (Achen and Bartels, 2016). Cognitive dissonance brings people to stick to prior beliefs and justify past actions instead of updating their beliefs when rational choice suggests doing so. This has been shown clearly in relation to partisan politics (Mullainathan and Washington, 2009; Gonzales etal., 2022) but also applies to other political matters, where for many new information that contradicts previous beliefs is not a reason to update those beliefs (and may even increase misperceptions) (Nyhan and Reifler, 2010). The public’s lack of ability to comprehend and apply certain categories of information suggests that easy access to some public information even may be harmful (Lessig, 2009).

Much optimism accompanied the emergence of the internet, which was expected to lead to a better-informed citizenry (Volokh, 1995). This was short-lived. Although the efforts needed to acquire information have lessened dramatically, much of the public remained indifferent to this progress and still opts for “rational ignorance,” which is defined as the choice not to know about matters of public-interest due to the effort required to know and the lack of benefit in mastering such knowledge (Brennan, 2016; Somin, 2015; Hardin, 2006).

In general, the optimism of the early days of the internet because of the belief it would make citizens better informed has been replaced with fear that the public is deluged with misinformation and is rendered unable to tell good information from bad (f*ckuyama, 2017). Habermas (2022) himself has recognized that the fragmentation of the public into many smaller “publics” jeopardizes deliberative public discourse. Fragmentation has strengthened group reliance and encouraged individuals to replace knowledge with allegiance to a partisan or identity group, which is a disservice to rational thinking (Hannon, 2023). It has created separate ecosystems, one teeming with disinformation that frustrates people’s ability to tell truth from falsehood, and another modelled on the more traditional concept of what rational public deliberation looks like (Benkler etal., 2018). Recent studies (Hood and Reid, 2018) suggest that when used for discourse about historical events, digital platforms can promote informed public engagement, contributing positively to the public discourse by grounding it in both historical fact and community narrative.

Information still matters. It is still sought by all to lend validity to statements in the public sphere. Lack of information generates faulty priority setting (Brennan, 2016) and negatively impacts voting choices in elections (Achen and Bartels, 2016). It is nevertheless much less certain that the effect of information today is like what it was assumed to be in the second half of the 20th century, especially with respect to online discourse.

Public discourse and social media

The Habermasian town square has been replaced in the 21st century by social media platforms such as Facebook, X (previously known as Twitter) and Instagram, where complex subjects are debated and deliberated (Amin etal., 2022). However, traditional media outlets continue to play a key role in determining the content that is ultimately reported on (Kong etal., 2022). Because of the increased ease with which individuals can express their opinions on online platforms, it is important to pay attention to their effect on discourse. For example, climate change, which is a global concern, has sparked intense debate in the digital realm. Using hashtags such as “#ClimateStrike” and “#FridaysForFuture” has facilitated the convergence of engaged individuals and advocates working toward a shared goal. Anexample of using social media platforms to transform public sentiment about climate change is the narrative associated with Greta Thunberg (Wang etal., 2020). Beginning as a modest-scale protest held in the vicinity of the Swedish parliament in Stockholm, its considerable traction on various social media platforms subsequently inspired similar protests in several places around the world (Xue etal., 2020).

The #MeToo movement, which was launched by Tarana Burke and quickly became famous around the world after Alyssa Milano tweeted the hashtag, is another example of how social media tools such as Twitter may be effectively used to raise awareness of broader societal concerns (Sharma etal., 2017). The courageous actions of women sharing their experiences of sexual harassment and assault ignited global discourse on social media and beyond (Pascual-Ferrá etal., 2022). One also witnesses the politicization of global public online debates. For example, the case of refugees seeking entry into Europe triggered sharp political discourse between supporters and opponents (Sutkutė, 2023). In the developing echo-sphere of digital media, journalists are increasingly assuming more the role of sense-makers than investigators (Agnistikova, 2022) as their audiences turn to them for confirmation of their pre-existing beliefs. In times of crisis, the already heavy use of digital media escalates even further, contributing to the rise of so-called alternative facts such as those that marked the anti-vax movement during the Covid-19 outbreak (Ling, 2020). The proliferation of alternative facts has led to more instances of misinterpretations and conflicts and to people being driven into echo chambers where they interact only with other people and content that reinforce their preexisting convictions (Terren and Borge-Bravo, 2021). The rise of echo chambers has made it harder to distinguish between truth and fiction on social media.

Conspiracy theories and information gaps

Although social media has facilitated the exchange of new ideas, it has concurrently generated an atmosphere favourable to the proliferation of falsehoods and disinformation (Douglas etal., 2019). The dissemination of the QAnon conspiracy theory during the 2016US presidential election notably illustrates this phenomenon (Hoseini etal., 2023). Originating with anonymous posters on online platforms, the QAnon conspiracy posits that high-profile individuals, including former President Donald Trump, were engaged in efforts to combat a global network involved in child trafficking. Politicians such as Trump have engaged with these conspiracy theories either unintentionally or deliberately, whether to disavow them or to provide ambiguous support for them (de Zeeuw etal., 2020). Another prominent example is the farfetched discourse about the correlation between 5G cellular network infrastructures and Covid-19 (Romer and Jamieson, 2020). Despite the absence of empirical evidence from the scientific community, these claims rapidly gained traction on popular social media platforms such as Facebook and YouTube, ultimately leading to the removal of 5G towers in some countries. There has been an observable increase recently in the popularity of urban legends that assert a causal relationship between vaccinations and conditions such as autism and other disorders (Papakyriakopoulos etal., 2020). The onset of the Covid-19 pandemic was accompanied by an infodemic, especially the extensive dissemination of misinformation (Bin Naeem etal., 2020). Widespread dissemination of misinformation makes it more difficult to advance public health programs and undermines public trust in scientific and government bodies. This trend potentially boosts the spread of inaccurate information such as deepfakes (manipulated videos generated by artificial intelligence) or misattributed tweets to influence public sentiment and undermine democratic mechanisms (Shahsavari etal., 2020). The notion of post-truth is continuing to weaken the trust in governments worldwide and to some extent confuse between scepticism and anti-establishment views. For example, the murder trial in Israel of Roman Zadorov, who was accused of murdering the teenager Tair Rada, generated various theories questioning his guilt. According to alternative perspectives that were amplified on social media, Zadorov was unjustly accused and Rada’s killing was caused by another individual (Grossman and Lev-On, 2023). Ultimately, Zadorov was retried and acquitted in March 2023. This case also illustrates how clashes of views between the public and the governmental establishment escalate even further on social media (Lev-On, 2023).

Public discourse and archival documents

Archives were traditionally seen as “the backdrop to all scholarly research,” which validate “appeals to ultimate truth” (Velody, 1998). Archival materials were characterized as impartial and “free from the suspicion of prejudice in regard to the interests in which we now use them” (Eastwood, 1994). As such, they could provide a truth-seeking public with a wellspring of reliable information. Postmodern theorists in the second half of the 20th century suggested a very different approach to archives, emphasizing, as they often do, their role as political control mechanisms. Michelle Foucault (1972) saw the archive as a “system of discursivity” that “establishes the possibility of what can be said”. Derrida’s (1996) influential essay “Archive Fever” also called attention to the limits of archival offerings and to understanding what shapes the records therein and how these records shape discourse.

Many scholars of archives followed suit, and their theories have gained popularity in recent decades. Terry cook described the transformation of archives over the past 150years from passive repositories of documents to active creators and shapers of memory, identity and communities (Cook, 2013). Robert Mcintosh described archivists as practicing “a politics of memory” (1998) and Verne Harris declared archivists to be “from the beginning and always, political players” and “active participants in the dynamics of power relations” (Harris, 2007). Several other scholars not only call for greater attention to the shadow these insights cast on archives’ alleged “impartiality” but embrace it’s political aspects, calling on archivists to play an active role in fights against injustice (Gilliland, 2016; Punzalan and Michelle, 2016). These theories have indeed fundamentally shifted discourse on archives in what has been recognized as “the archival turn” (Ketelaar, 2017).

Other scholars however, argue it is a commonplace that historical records do not tell the full story and that its political nature is long recognized by historians who regularly assess the value of sources, taking into account various considerations, including those human biases theorists point to (Tucker, 2004). Supporting much of the contemporary scholarship on archives, historian Yael Sternhell writing on Civil War archives, insists that even if they do not tell a full and accurate story, they do tell “a story” if fundamentally imperfect (2023). Historian Robert Darnton is more critical of the “archival turn”, writing that “words such as ‘facts’ and ‘truth’ make us uncomfortable”, but “facts will not go away” (Darnton, 2003). Archive-based historical research guided by truthfulness and produced with academic rigor still contributes to our “understanding of a time and a world” (Farge, 2013). Such history is still bound by pretextual existence and is not as creative as post- modernist theorists claim, writes Carolyn Steedman (2001).

Contemporary theories following Derrida also point to access to archives as a measure of democratization. In this sense it is important to recognize that governments in liberal democracies have gone to great lengths to make archives publicly available as early as the 19th century (Joyce, 1999). The recent digitization of archives, making them available to anyone connected to the internet, is a giant leap forward in democratizing knowledge. The internet and digitization are tools that confront the monopoly of knowledge by disseminating historical records to the general public (Malhotra, 2021).

In this paper we focus on one historical affair for which vast historical documentation is available to tell a story (if limited) and has been digitized to make it easily available to the public. We go on to look at the impact of democratizing this historical information as far as it can be inferred from social media discourse.

Research questions

RQ1.

How does the discussion of archival information in social media posts influence the likelihood of supporting the kidnapping theory?

RQ2.

How has the support for the kidnapping theory and inclusion of archive information in social media posts changed over time?

RQ3.

When discussing the kidnapping theory and the archival information, how do posts on pages of individuals, traditional media outlets, politicians, NGOs, and open public-interest groups differ from each other?

The answers to these research questions may lead to further understanding of the dynamics generated by the kidnapping theory and the effect of introducing historical records in public discourse.

Methodology

Our primary methodological tool in this study was content analysis, which we used to systematically and thoroughly explore a diverse range of texts, messages and information in their original context and derive reliable conclusions (Kyngäs, 2020; Mende, 2022). Content analysis also facilitates the discovery of novel perspectives and enhances comprehension of the phenomenon under investigation (the affair) (Krippendorff, 2004). Content analysis can be used with quantitative variables but can also incorporate qualitative analysis to enrich insights and suggest new patterns. The focus of this research and main case study is the online Hebrew public discourse on Facebook regarding the affair between the years 2016 and 2021.

We chose the Facebook platform for our study for several reasons. In Israel, Facebook was (and still is) one of the most widely used social media platforms, with 75% of adult users active daily (Bezeq, 2022). Thus we were able to tap into the specific dynamics and characteristics prevalent in online discourse within the Israeli context during the examined years. The platform’s prominence makes it an influential space for discussions and information dissemination among the Israeli population.

Procedure

We systematically exported all the relevant, accessible Facebook posts that included the Hebrew phrase for “Yemenite children” (“יַלְדֵי תֵּימָן”) that appeared within our specified time frame, using CrowdTangle. The software generated a CSV data sheet that included every published post from public pages, accounts with verified profiles, or open groups that met our criteria. Closed or private groups and accounts with personal profiles were not included in the dataset to protect the users’ privacy. A total of 4,300 posts was obtained and extracted onto a CSV sheet.

Sampling frame

In order to glean as representative a picture as possible, we collected data from 1 January 2016 through 31 December 2020, a five-year-long period beginning one year before the National Archives released the archival material related to the affair on their website. Anticipating that the public disclosure would significantly alter the online discourse of the case, we wished to compare side by side the discourse during the preceding year with that which followed over the four subsequent years. The content of the posts was analysed quantitatively using descriptive statistics. Figure1 shows the number of posts published per quarter over the five-year period.

Data analysis

We adopted a mixed-methods content analysis approach combining quantitative and qualitative data to investigate our research questions. To ensure as representative an analysis as possible, we examined every second post in the sample of 4,300, reducing its size to a total of 2,150 posts. Using the Duplicate function in Excel, 578 duplicate posts were subtracted, leaving us with a representative sample of 1,572 posts, which we further analysed using the designated codebook. The coding was performed manually by the authors and a trained third coder for two main dimensions. The quantitative data analysis was performed using IBM SPSS Statistics version 25.

Coding schemes

To test the coding reliability, we conducted an intercoder reliability test. The two authors double-coded a random sample (n=100), first for archive-related topics and later for kidnapping-related topics. In a reliability test, Krippendorff’s alpha indexes for both variables were highly satisfactory: 0.87 for the archive category and 0.84 for the kidnapping category. The first dimension of coding focusing on archive-related topics is described in Table1.

The second dimension of coding focusing on the kidnapping theory and examining whether a given post supported or confirmed it is described in Table2.

Quantitative findings

A chi-square test of independence was used to examine the association between information about the archive and support for the kidnapping theory. Analysis yielded significant results, χ²(2, N=1,572)=124.72, p<0.001, indicating that authors of posts including information about the archive were less likely to support the kidnapping theory than authors of posts without it. Table3 shows that authors of posts where the archival material was mentioned were slightly more likely to reject the kidnapping theory. Only 0.9% of the authors of posts who did not mention the archival material showed scepticism towards the kidnapping theory, compared with 2.5% of the authors who did mention it. However, even among those who discussed the archival material, few rejected the kidnapping theory, largely upending expectations of a significant correlation. Interestingly, reference to the archival material showed more of a correlation with support for the kidnapping theory than a rejection of it. Support was a little more than 82% among those whose posts had references to the archival material compared with nearly 57% among those whose posts did not. For the most part, the material did not convince participants that the kidnapping theory was likely groundless; they merely switched to the neutral camp. Of the 26% drop in the rate of support for the kidnapping theory in posts that mentioned the archival materials, less than 2% switched to the no group and 24% joined the neutral group. Overall, we can see that in all the posts, regardless of whether they discussed archival material or not, only 1.2% opposed the kidnapping theory, 22.2% were neutral about it, and 76.6% supported it.

Figure2 shows how both the percentage of posts supporting or opposing the kidnapping theory and the percentage of posts with information about the archive changed over time. Inthe last quarter of 2016, when the archival materials were published, the number of posts mentioning them spiked, accompanied by a significant overall dip in support for the kidnapping theory. But apart from that quarter, there was not much correlation between the two. The percentage of posts supporting the kidnapping theory remained at medium to high levels over time, whereas the percentage of posts including information about the archive remained at low to medium levels over time. The percentage of posts opposing the kidnapping theory remained very low throughout the examined period, regardless of the volume of posts mentioning the archival material.

Qualitative findings

In order to uncover and examine in depth the source of each sample content, we used qualitative methods. Among the categories of content creators that CrowdTangle identified, four were significantly involved in the public discourse regarding the affair: traditional media outlets (journalists and media channels), politicians and political parties, NGOs, and open public-interest groups. A fifth miscellaneous category comprising subcategories, each of which consisted of only one or two posts, also emerged (see Figure3).

Traditional media outlets

An analysis of the findings showed that traditional media outlets used their Facebook pages to take an active part in the public discourse. For example, 49 different posts dealt directly with archival materials, such as the following one from Channel 13, a very popular page in Israel with over 800,000 followers at the time of publication. In promoting a report that appeared on the show The Source (2019), they wrote, “Forget everything you knew about the Yemenite children’s affair: the facts versus the myths – who are interested in lying to you? The Source takes a deep dive into the archives”.

The kidnapping theory is directly referred to in 91 posts. On the competing Channel 12, whose number of followers is similar to Channel 13’s, there was a post that adopted the terminology and rhetoric of the kidnapping theory that was published on 6 June 2016: “Rina Matzliach has put the affair of the kidnapped Yemenite children back on the agenda, and in a special column, she calls to put an end to the affair and to treat this wound at its source”.

At the beginning of the video clip attached to the article, the reporter, Rina Matzliach, reads the following text: “In the last two months that I’ve been dealing with the affair of kidnapped Yemenite children, and I deliberately say ‘abducted’, I don’t just mean lost children, because these children were kidnapped from their mothers’ arms”. The printed media also adopted this rhetoric, taking an even stronger stance on this topic, as evident in the following example from the Haaretz newspaper, which has 300,000 followers, on 8 April 2016: “A new archive is currently being set up that some describe as ‘the establishment’s greatest fear’. This archive threatens to expose a mechanism for abducting toddlers operated by the State of Israel in its early days”.

Politicians

In contrast to the media, which extensively addressed this affair over the years, relatively few politicians (N=138) – whether ministers or Knesset members, from either the left or the right – posted any related content. One notable exception is former Likud MK Nurit Koren, who was particularly active in investigating the affair and even served as chair of the Special Committee on the Disappearance of Yemenite, Eastern, and Balkan Children. She created 58 (42%) of all the politicians’ content. Yet only 36 posts of the total posts in the politician’s group explicitly addressed the archive and its various materials. One of them was MK Koren’s from 22 November 2018:

The Cohen-Kedmi Commission skipped over thousands of important documents held by the Labor movement [Haavoda] that shed light on the disappearance of Yemenite children. Today I visited the Lavon Institute archives, a research institute for the history of the Labor movement in the Land of Israel. During the tour, I realized that the historical materials at the institute that teach about the life of immigrants during those years in transit camps can greatly help a large public that looks to the committee for answers. I also discovered new materials that have not yet been published.

The percentage of posts published by politicians that support the kidnapping theory or use anti-establishment terminology was 52.8% (n=73). Particularly notable is one of the posts with the highest total of interactions (n=3,862) in the entire sample, which was written by MK Shelly Yechimovich of the Labor party and posted on 8 July 2016. In it she adopted the anti-establishment kidnapping rhetoric and celebrated the recommendations of a government committee to give more prominence to the Mizrahi narratives in the school curriculum and to include teaching about the “abduction of the Yemenite children”.

The committee’s conclusions give hope for change and carry a spirit of freedom and equality. The demand for mentioning the tragic issues, such as the abduction of Yemenite children, in textbooks, even though the state has not yet recognized this tragedy, expresses the freedom demanded by the authors [of the report].

Nongovernmental organizations (NGOs)

NGOs such as Amram, whose activities centre solely on the affair, as well as other NGOs that have addressed the affair made up another significant group that participated in the online discourse. NGOs shared a total of 165 relevant posts during the study period, of which only 26 dealt directly with the archive and its materials. For example, the New Israel Fund, one of Israel’s largest foundations and particularly supportive of progressive civil society initiatives, posted on 28 December 2016, hours after the archival materials were uploaded online:

We welcome the disclosure of the hundreds of thousands of documents used by the state commission of inquiry to investigate the affair of Yemenite children and hope that the entire information surrounding the affair – which has roiled the public for more than 60 years – will be fully exposed. The uncovered materials, now accessible to the public on the State Archives website, reveal the difficult stories of many families. However, this is only partial information since the commission of inquiry did not thoroughly examine the state’s responsibility and the possibility that systematic abductions were carried out.

Of the total number of NGO posts, 71.51% expressed some form of support for the kidnapping theory or used rhetoric characteristic of those who support it. A significant number of these appeared on Amram’s Facebook page, such as the following from 15 June 2020 criticizing WIZO, a century-old women’s Zionist organization that provides health and welfare services to women and children, which is associated with the establishment.

This Sunday, as every year, there will be an awareness day for the abduction of children from Yemen and the Balkans. The upcoming awareness day will be marked by a demand that WIZO take responsibility for its part in the affair. A series of documents, data, and testimonies link WIZO to the affair of the children of Yemen, the East and the Balkans. The accumulated knowledge on the subject indicates that the organization conducted itself in a racist manner towards the parents of the children; parents who insisted on inquiring about the fate of their children were thrown out.

Public groups

In addition to examining official Facebook pages of organizations, personalities, politicians and media organizations, we also looked at how information was disseminated within various interest groups that were set up as public groups on Facebook. We found 1,341 posts about the affair that were shared across Facebook in these groups over the period of the study.

Some of the groups focused exclusively on the affair, while others were free ranging. The groups’ size in the sample ranged from a minimum of 50–60 members to hundreds of thousands. The largest, with 160,000 members at the time of the study was Home Design Recommendations and Tips from Semi-Free Electrical Appliances Ltd., which was mostly concerned with rather mundane issues such as home improvements and discounts. But three posts with references to the affair also appeared there, including the following from 18 March 2019: “When the state was established, Yemenite children and children from the Mizrahi communities were kidnapped. Why do you think they don’t expose the files from the archives and examine who is responsible for this crime?”

The public groups that exclusively focused on the archive and its materials published 174 posts, including one from a public political group called Friends Who Love Benjamin Netanyahu on 9 August 2020:

Thousands of files (Rabin’s assassination and the Shamgar Commission, Yemenite children) “died” in the State Archives’, according to the sources. And they are surely ‘alive’ on News 1. The lies of the left and the deep state are exposed. Messiah now! Urgent!

As many as 67.18% of posts in these groups expressed some support for the kidnapping theory (n=901). In another group consisting mainly of Netanyahu supporters, Likud Led by Benjamin Netanyahu, with 19,761 members at the time of the study, one member wrote on 22 December 2019: “I hope that the State Attorney’s office and the supreme court will begin to act like honest people and not like corrupt right-wing haters. We have not forgotten the left’s hatred of the right up until 1977, from the abduction of children from the Mizrahi communities, especially Yemenite children who were sold to the United States for experiments, and another part of the abductees that was sold for adoption”.

Discussion

We examine the premise that information can influence public discourse positively; that is, that it can improve public deliberation and make it more rational and productive (Bozdag and van den Hoven, 2015), bringing together even individuals who are not highly informed to make better decisions (Landemore, 2012). This premise, which lies at the core of democratic theories of public deliberation, has been recently challenged (Ahlstrom-Vij, 2012; Brennan, 2016; Hannon, 2023).

As our study shows, despite the formidable number of documents that three separate inquiry commissions examined, which showed that no systematic kidnapping was carried out, the impact on the discourse was limited and short-lived. This finding is in line with the literature that is sceptical of the public’s ability to rationally use information in public discourse (Ahlstrom-Vij, 2012; Brennan, 2016; Brown, 2018; Bagg, 2018; for a somewhat more optimistic view see Kube, 2023).

With respect to R1, our findings in Table3 demonstrate that the archival information had a limited effect on support for the kidnapping theory: two-thirds as many authors of posts mentioning archival material as those who do not mention it support the kidnapping theory, this difference is nearly entirely accounted for by an increase in posts that are neutral about the kidnapping theory and are more descriptive in nature. The increase in the number of posts actually repudiating the kidnapping theory was very low and statistically insignificant, and the overall number of posts in this group remained very low. We conclude that the discussion of the archival material in Facebook posts somewhat complicates an otherwise simplistic one-sided discourse, but it fails to prompt people to draw the virtually inevitable conclusions that emerge from the documents.

The minor effect of the archival materials on discourse seems to stem from two reasons. First, discussing them was largely avoided. Even though the release of the more than 300,000 documents was the only significant development in the affair in many years, they were ignored even in many of the media reports. Thus, even the limited impact of discussing the archival material is diluted when looking at the overall discourse, since it includes only a small number of such discussions (rightmost column in Table3). This finding supports the claim that public discourse is largely irrational and that participants are unaware of relevant information and often prefer to be so. Much of the writing about political ignorance of citizens attributes it to citizens’ overall indifference to matters they would be well advised to know about (Brennan, 2016). What is further evident here is that people avoided information with immediate implications for a topic they chose to express their opinions about.

The second possible reason the archives had so little affect has to do with the irrationality of many of those who participated in discussing the affair. As we have shown, only very few of those reached conclusions based on the information, while most continued to support the kidnapping theory and some were neutral. This illustrates how “politics seems to drive our beliefs about the facts instead of the facts driving policy”, a reality which is attributed to the allegiance of the participants to their prior groups (Hannon, 2021). This is especially evident in online debates, in which interlocutors tend to engage with information more passionately, as was evident during Covid-19 (Voinea etal., 2023). Looking at the two ecosystems side by side, we see how the one where the discourse was “rationally deliberating”, as described by Benkler etal. (2018), is much smaller than the one where people’s ability to tell truth from falsehood was frustrated. Our findings show that avoidance of engaging with information and engaging with it irrationally together diminish the effect of new information on one’s position vis-à-vis the kidnapping theory. This is an alarming finding for those who believe that information can foster rational public deliberation in general, much less so deliberation on social media.

The findings for RQ2 (Figure2) show how the release of the archival material and its presence in the public discourse changed over time, rather than whether those who make references to the archival material are more or less likely to support the kidnapping theory. Around the time of the release of the archival information, the percentage of posts supporting the kidnapping theory dipped significantly. It is likely that this correlated with the fact that the media coverage about the disclosure – much of which reported on the lack of a “smoking gun” that supported the kidnapping theory – took a neutral position. However, scepticism towards the kidnapping theory was short-lived. Not long after the online publication of the historical records, support for the kidnapping theory bounced back to its previous levels. It then rose and dropped over the next two years with very little correlation to the discussion of the archival material. After two years of mixed trends, interest in the archival materials subsided and support of the kidnapping theory began growing gradually, eventually exceeding even those levels seen before the records were made available online. This pattern may also indicate that the effect of the archival information that we found in response to R1 occurred mostly around the time the documents were released. An explanation for this may be that at that time new and more information-focused participants joined the discussion, who were interested in the event of the online publication itself. As interest in the event subsided, the regular participants in the discussion, those with stronger allegiances to the groups participating in the debate and more emotionally invested in it, returned to dominate the Facebook discourse of the affair. By 2019 and 2020, any impact of the archival information seems to have faded, while support of the kidnapping theory returned to the same level it was before hundreds of thousands of documents pointing in a completely different direction were made available to the public. These findings suggest that Foucault’s claim according to which the archive limits “the possibility of what can be said” are not necessarily realistic. At least in the Yemenite Children Affair the historical records do not carry much weight and have not limited or significantly altered public discourse.

Our third research question (RQ3) focused on possible differences in the way traditional media, NGOs, politicians or Facebook public-interest groups discussed the affair and the archival material on their Facebook pages. We see that none strongly objected to the kidnapping theory, but we identify differences between supporters and those who were neutral. Politicians and media outlets tend to be somewhat more sceptical of the kidnapping theory and prone to being influenced by the archival information. We suggest two (not mutually exclusive) explanations. First, unlike the utterances of NGOs and public-interest groups, statements by traditional media and politicians also have a presence outside social media, in traditional print and electronic media, which are less heated and polarizing outlets. Another possible explanation is that when it comes to politicians and traditional media, two major actors in any theory of public deliberation, at least some members of these groups (even if clearly not all of them) show at least some adherence to codes of rational and fact-based discourse or public accountability. As the qualitative analysis shows, many of their posts too, were not aligned with the historical records, yet to the extent that the records influenced the discourse at all, it was more evident among traditional media and NGOs than other groups. This finding gestures toward what might be a better model for publication deliberation, specifically one that gives greater weight and a greater role to the participation of “institutions that work as cognitive artefacts gathering and processing this information for the group” (Landemore, 2012). Nevertheless, at the same time it shows that these actors too were largely irrational participants in the debate.

The limitations in our study should be acknowledged. First, our study, which focused on Facebook because it was the primary Israeli social-media platform during the examined period, may not be generalizable to other social media platforms. Second, due to CrowdTangle’s privacy policy, our data collection was restricted to public pages, public groups, and accounts with verified profiles. Finally, our findings cannot establish a causal relationship between the impact of discussing the archives on support for the kidnapping theory. It is possible that a small portion of the discussants changed their view once they were presented with contradicting data. Or perhaps participants who had avoided voicing unpopular views felt encouraged to do so once supporting had surfaced. Still other explanations are possible and are worthwhile exploring.

We suggest further research including a public opinion survey to find out where the public at large stands on the affair. The results could be used together with our findings here to compare with the views of online discussants. Given our findings showing certain differences in how information is referred to by the media and politicians, on the one hand, and NGOs and Facebook group members, on the other, it would be interesting to also examine how academics, who presumably adhere more strictly to codes of rational discourse, compare.

Our study contributes to the body of knowledge concerning the relationship between public discourse and the dissemination of governmental information of historical and factual data. Our findings show that some disillusionment with public deliberation theories in general and in online setting in particular is warranted. Our research also supports some scepticism as to whether making historical records easily accessible indeed contributes to the online discussion of historical affairs, at least when those affairs are the focus of heated debate. However, though limited, we did observe some effect; that and the fact that certain actors were more given to influence than others leave room for some hope. Finding ways to build on these more positive results should also be the subject of future research.

Figures

Figure1

Total number of posts per quarter (N=4,300)

Figure2

Fluctuation in percentage of posts supporting kidnapping theory and percentages of posts with information about the archive over time

Figure3

Posts by page category (N=2,150)

Table1

Coding scheme for archive-related topics

Are archive-related topics mentioned?	Description of content	Examples
Yes	Specifically addresses archival or historic documents regarding the affair and mentions related archival materials	“The Knesset approved in the second and third reading the bill to open adoption records in the case of the kidnapping of children from Yemen, the East, and the Balkans” (posted to the public Facebook group Yemani Baneshama (Right-Wingers at Heart) on 4 July 2018)
No	Mentions the affair and its ramifications, without referring to the archives and its materials	“The case of the Yemenite children: the open wound of the State of Israel” (posted to the Globes newspaper’s public Facebook group on 25 April 2017)

Table2

Coding schemes for support of kidnapping theory

Supporting kidnapping theory	Description	Examples
Yes	Expresses clear support for the theory that the children were kidnapped in an organized manner	“I have documents that indicate that 1,670 children were kidnapped in total and in my estimation the number is higher” (posted to The Daily Agenda’s Facebook page on 18 July 2016)
No	Refers to the affair without supporting the theory	“The conspiracy of the kidnapped Yemenite children serves the old message of the extreme left: the Zionist establishment is racist and corrupt” (posted to Mida Magazine’s Facebook page on 21 July 2018)

Table3

Percentages of posts supporting kidnapping theory with respect to mentions of the archive

	Information about the archive
	No		Yes		Total
Support kidnapping theory	n	%	n	%	n	%
No	11_a	0.9%	9_b	2.5%	20	1.2
Neutral	210_a	16.7%	147_b	40.9%	357	22.2
Yes	1,032_a	82.4%	203_b	56.6%	1,235	76.6
Total	1,253	100%	359	100%	1,612	100

Note(s): Categories marked with different subscript letters differ significantly in column proportions from each other at the 0.05 level

References

Achen, C. and Bartels, L. (2016), “Democracy for realists: holding up a mirror to the electorate”, Juncture, Vol.22 No.4, pp.269-275, doi: 10.1111/j.2050-5876.2016.00873.x.

Agnistikova, O.I. (2022), “Strengthening the presence of subjective journalistic voice in the news as a trend of the digital environment”, Vestnik of Kostroma State University, Vol.28 No.4, pp.86-92, doi: 10.34216/1998-0817-2022-28-4-86-92.

Ahlstrom-Vij, K. (2012), “Why deliberative democracy is (still) untenable”, Public Affairs Quarterly, Vol.26 No.3, pp.199-220.

Amin, S., Alharbi, A., Uddin, M.I. and Alyami, H. (2022), “Adapting recurrent neural networks for classifying public discourse on COVID-19 symptoms in Twitter content”, Soft Computing, Vol.26 No.20, pp.11077-11089, doi: 10.1007/s00500-022-07405-0.

Arendt, H. (1967), “Reflections: truth and politics”, The New Yorker, Vol.43 No.25, pp.49-88.

Bagg, S. (2018), “The power of the multitude: answering epistemic challenges to democracy”, American Political Science Review, Vol.112 No.4, pp.891-904, doi: 10.1017/s0003055418000527.

Benkler, Y., Faris, R. and Roberts, H. (2018), Network Propaganda: Manipulation, Disinformation, and Radicalization in American Politics, Oxford University Press.

Bezeq (2022), “The digital life”, Bezeq Internet Report for, Bezeq.

Bin Naeem, S., Bhatti, R. and Khan, A. (2020), “An exploration of how fake news is taking over social media and putting public health at risk”, Health Information and Libraries Journal, Vol.38 No.2, pp.143-149, doi: 10.1111/hir.12320.

Birenboim, S. (2019), Khatifata Shel Hahistoria [The Kidnapping of History], Hashiloah, Jerusalem, Vol.13, pp. 95-114.

Bozdag, E. and van den Hoven, J. (2015), “Breaking the filter bubble: democracy and design”, Ethics and Information Technology, Vol.17 No.4, pp.249-265, 10.1007/s10676-015-9380-y.

Brennan, J. (2016), Against Democracy, Princeton University Press, NJ.

Brown, É. (2018), “Propaganda, misinformation, and the epistemic value of democracy”, Critical Review, Vol.30 Nos 3-4, pp.194-218, doi: 10.1080/08913811.2018.1575007.

Cook, T. (2013), “Evidence, memory, identity, and community: four shifting archival paradigms”, Archival Science, Vol.13 Nos 2-3, pp.95-120, doi: 10.1007/s10502-012-9180-7.

Darnton, R. (2003), “How historians play God”, European Review, Vol.11 No.3, pp.267-280, doi: 10.1017/s1062798703000279.

de Zeeuw, D., Hagen, S., Peeters, S. and Jokubauskaite, E. (2020), “Tracing normiefication: a cross-platform analysis of the QAnon conspiracy theory”, First Monday, Vol.25 No.11, doi: 10.5210/fm.v25i11.10643.

Derrida, J. (1996), Archive Fever: A Freudian Impression, University of Chicago Press, Chicago, IL.

Douglas, K.M., Uscinski, J.E., Sutton, R.M., Cichocka, A., Nefes, T., Ang, C.S. and Deravi, F. (2019), “Understanding conspiracy theories”, Political Psychology, Vol.40 No.S1, pp.3-35, doi: 10.1111/pops.12568.

Eastwood, T. (1994), “What is archival theory and why is it important?”, Archivaria, pp.122-130.

Farge, A. (2013), The Allure of the Archives, Yale University Press, New Haven, CT.

Foucault, M. (1972), The Archaeology of Knowledge, Pantheon Books, New York.

f*ckuyama, F. (2017), The Emergence of a Post-fact World, Project Syndicate, available at: http://www-project-ums-syndicate-org-s-99.er.library.nenu.edu.cn/magazine/the-emergence-of-a-post-fact-world-by-francis-f*ckuyama-2017-01 (accessed 3 January 2024).

Gilliland, A.J. (2016), “‘Dead on arrival’? Impartiality as a measure of archival professionalism in the twenty-first century”, available at: http://escholarship-org-s-99.er.library.nenu.edu.cn/uc/item/5m03n8cp 29 March 2024).

González, T.A., Capozza, F. and Granic, G.D. (2022), Can Cognitive Dissonance Theory Explain Action Induced Changes in Political Preferences? (No. 9549), CESifo, Munich.

Grossman, N. and Lev-On, A. (2023), “Social media jurors: conceptualizing and analyzing online public engagement in reference to legal cases”, Crime, Law and Social Change, Vol.79 No.3, pp.223-240, doi: 10.1007/s10611-022-10047-w.

Habermas, J. (1989), The Structural Transformation of the Public Sphere (translated by Burger, T.), MIT Press, Cambridge, MA, pp.85-92, 180.

Habermas, J. (2022), “Reflections and hypotheses on a further structural transformation of the political public sphere”, Theory, Culture and Society, Vol.39 No.4, pp.145-171, doi: 10.1177/02632764221112341.

Hannon, M. (2021), “Disagreement or badmouthing? The role of expressive discourse in politics”, Edenberg, E. and Hannon, M. (Eds), Political Epistemology, Oxford University Press, pp.297-317.

Hannon, M. (2023), “Public discourse and its problems, Politics”, Philosophy and Economics, Vol.22 No.3, pp.336-356, doi: 10.1177/1470594x221100578.

Hardin, R. (2006), “Ignorant democracy”, Critical Review, Vol.18 Nos 1-2, pp.179-195, doi: 10.1080/08913810608443656.

Harris, V.S. (2007), Archives and Justice: A South African Perspective, Society of American Archivists, Chicago, IL.

Hood, C. and Reid, P. (2018), “Social media as a vehicle for user engagement with local history: a case study in the North East of Scotland”, Journal of Documentation, Vol.74 No.4, pp.741-762, doi: 10.1108/JD-12-2017-0167.

Hoseini, M., Melo, P., Benevenuto, F., Feldmann, A. and Zannettou, S. (2023), “On the globalization of the QAnon conspiracy theory through telegram”, Proceedings of the 15th ACM Web Science Conference 2023, ACM, pp.75-85, doi: 10.1145/3578503.3583603.

Joyce, P. (1999), “The politics of the liberal archive”, History of the Human Sciences, Vol.12 No.2, pp.35-49, doi: 10.1177/09526959922120234.

Ketelaar, E. (2017), “Archival turns and returns”, in Gilliand, A.J., McKemmish, S. and Lau A.J. (Eds), Research in the Archival Multiverse, Monash University Publishing, available at: http://library-oapen-org-s-99.er.library.nenu.edu.cn/bitstream/handle/20.500.12657/31429/628143.pdf?sequence#page=237 (accessed 29 March 2024).

Kong, D., Chen, A., Zhang, J., Xiang, X., Lou, W.Q.V., Kwok, T. and Wu, B. (2022), “Public discourse and sentiment toward dementia on Chinese social media: machine learning analysis of Weibo posts”, Journal of Medical Internet Research, Vol.24 No.9, pp.511-525, doi: 10.2196/39805.

Krippendorff, K. (2004), “Reliability in content analysis”, Human Communication Research, Vol.30 No.3, pp.411-433, doi: 10.1111/j.1468-2958.2004.tb00738.x.

Kube, T. (2023), “Factors influencing the update of beliefs regarding controversial political issues”, The Journal of Social Psychology, pp.1-23, doi: 10.1080/00224545.2023.2253981.

Kyngäs, H. (2020), “Qualitative research and content analysis”, The Application of Content Analysis in Nursing Science Research, pp. 3-11.

Landemore, H. (2012), The Mechanisms of Collective Intelligence in Politics Collective Wisdom: Principles and Mechanisms, Cambridge University Press, pp.251-289.

Lepoutre, M. (2021), Democratic Speech in Divided Times, Oxford University Press, Oxford.

Lessig, L. (2009), Against Transparency, The New Republic, available at: http://newrepublic-com-s-99.er.library.nenu.edu.cn/article/70097/against-transparency

Lev-On, A. (2023), “Democratizing the discourse on criminal justice in social media: the activity for justice for Roman Zadorov as a case study”, Humanities and Social Sciences Communications, Vol.10 No.770, doi: 10.1057/s41599-023-02302-9.

Levitan, D. (2023), “The two ongoing narratives and current disputes on the affair of the ‘missing Yemenite children’”, in Friedman, E., Neubauer-Shani M. and Scham, P. (Eds) Israel, Polarization and Consensus-Building in Israel: the Center Cannot Hold, Routledge, London, pp.93-115.

Ling, R. (2020), “Confirmation bias in the era of mobile news consumption: the social and psychological dimensions”, Digital Journalism, Vol.8 No.5, pp.596-604, doi: 10.1080/21670811.2020.1766987.

Malhotra, A. (2021), “In what sense are archives political institutions?”, LSE Blogs, available at: http://blogs-lse-ac-uk-s-99.er.library.nenu.edu.cn/lseupr/2021/12/03/in-what-sense-are-archives-political-institutions/ (accessed 29 April 2024).

McIntosh, R. (1998), “The great war, archives and modern memory”, Archivaria, Vol.46, pp. 1-31.

Meir-Glitzenstein, E. (2011), “Operation magic carpet: constructing the myth of the magical immigration of yemenite jews to Israel”, Israel Studies, Vol.16 No.3, pp.149-173, doi: 10.2979/israelstudies.16.3.149.

Mende, J. (2022), “Extended qualitative content analysis: researching the United Nations and other international institutions”, Qualitative Research Journal, Vol.22 No.3, pp.340-353, doi: 10.1108/QRJ-11-2021-0127.

Mullainathan, S. and Washington, E. (2009), “Sticking with your vote: cognitive dissonance and political attitudes”, American Economic Journal: Applied Economics, Vol.1 No.1, pp.86-111, doi: 10.1257/app.1.1.86.

Nyhan, B. and Reifler, J. (2010), “When corrections fail: the persistence of political misperceptions”, Political Behavior, Vol.32 No.2, pp.303-330, doi: 10.1007/s11109-010-9112-2.

Papakyriakopoulos, O., Medina Serrano, J.C. and Hegelich, S. (2020), “The spread of COVID-19 conspiracy theories on social media and the effect of content moderation”, Harvard Kennedy School Misinformation Review, Vol.1 No.3, pp, doi: 10.37016/mr-2020-034.

Pascual-Ferrá, P., Alperstein, N. and Barnett, D.J. (2022), “Social network analysis of COVID-19 public discourse on twitter: implications for risk communication”, Disaster Medicine and Public Health Preparedness, Vol.16 No.2, pp.561-569, doi: 10.1017/dmp.2020.347.

Punzalan, R.L. and Michelle, C. (2016), “Critical directions for archival approaches to social justice”, The Library Quarterly, Vol.86 No.1, pp.25-42, doi: 10.1086/684145.

Romer, D. and Jamieson, K.H. (2020), “Conspiracy theories as barriers to controlling the spread of COVID-19 in the U.S”, Social Science and Medicine, Vol.263, 113356, doi: 10.1016/j.socscimed.2020.113356.

Seidelman, R. (2020), Under Quarantine: Immigrants and Disease at Israel's Gate, Rutgers University Press, NJ.

Shahsavari, S., Holur, P., Wang, T., Tangherlini, T.R. and Roychowdhury, V. (2020), “Conspiracy in the time of Corona: automatic detection of emerging COVID-19 conspiracy theories in social media and the news”, Journal of Computational Social Science, Vol.3 No.2, pp.279-317, doi: 10.1007/s42001-020-00086-5.

Sharma, E., Saha, K., Ernala, S.K., Ghoshal, S. and De Choudhury, M. (2017),“Analyzing ideological discourse on social media: a case study of the abortion debate”, in Dixon, D. (Ed.), Proceedings of CSSSA’s Annual Conference on Computational Social Science, New York NY, Association for Computing Machinery, doi: 10.1145/3145574.3145577.

Somin, I. (2015), “Rational ignorance”, Gross, M. and McGoey, L. (Eds), Routledge International Handbook of Ignorance Studies, Routledge, London, pp.274-281.

Steedman, C. (2001), “Something she called a fever: michelet, Derrida, and dust”, The American Historical Review, Vol.106 No.4, pp.1159-1180, doi: 10.2307/2692943.

Sternhell, Y. (2023), War on Record, Yale University Press, New Haven.

Sutkutė, R. (2023), “Public discourse on refugees in social media: a case study of The Netherlands”, Discourse and Communication. doi: 10.1177/17504813231188499.

Terren, L. and Borge-Bravo, R. (2021), “Echo chambers on social media: a systematic review of the literature”, Review of Communication Research, Vol.9, pp.99-118, doi: 10.12840/issn.2255-4165.028, available at: http://rcommunicationr-org-s-99.er.library.nenu.edu.cn/index.php/rcr/article/view/94

Tucker, A. (2004), Our Knowledge of the Past: A Philosophy of Historiography, Cambridge University Press, Cambridge and New York, NY.

Velody, I. (1998), “The archive and the human sciences: notes towards a theory of the archive”, History of the Human Sciences, Vol.11 No.4, pp.1-16, doi: 10.1177/0952695198011004.

Voinea, V., Marin, L. and Vica, C. (2023), “The moral source of collective irrationality during COVID-19 vaccination campaigns”, Philosophical Psychology, Vol.36 No.5, pp.1-20, doi: 10.1080/09515089.2022.2164264.

Volokh, E. (1995), “Cheap speech and what it will do”, Yale Law Journal, Vol.104 No.7, pp.1805-1852, 10.2307/797032.

Wang, X., Sirianni, A.D., Tang, S., Zheng, Z. and Fu, F. (2020), “Public discourse and social network echo chambers driven by socio-cognitive biases”, Physical Review X, Vol.10 No.4, 041042, doi: 10.1103/physrevx.10.041042.

Xue, J., Chen, J., Chen, C., Zheng, C., Li, S. and Zhu, T. (2020), “Public discourse and sentiment during the COVID-19 pandemic: using latent dirichletallocation for topic modeling on twitter”, PLOS ONE, Vol.15 No.9, e0239441, doi: 10.1371/journal.pone.0239441.

Corresponding author

Roy Peled can be contacted at: rpeled@colman.ac.il