Category Archives: Censorship

A recent Australian defamation verdict [PDF] awarded US$40,000 in damages for an Google search result. Google warns this precedent could lead to censorship of the web.

In 2016 George Defteros, a lawyer from Victoria, Australia, asked Google to remove an article from its search results. The article from 2004 reported murder charges against Mr. Defteros that were later dropped. He later sued Google for defamation after the search giant refused to omit the article from its engine.

In 2020, supreme court justice Melinda Richards ruled that the article implied Mr. Defteros crossed a line from professional lawyer to a confidant and friend of criminals. This is due to the lawyer representing various gangsters in court.

Googles lawyers argue that a search engine is not a publisher because a hyperlink is not, in and of itself, the communication of that to which it links. The companys submission adds:

The inevitable consequence of leaving the court of appeals decision undisturbed is that Google will be required to act as censor by excluding any webpage about which complaint is made from its search results, even when, as here, the webpage may be a matter of legitimate interest to the substantial portion of people who search for it and is published by a reputable news source.

This isnt the first time Google has been on the receiving end of such lawsuits. The right to be forgotten, also known as right to erasure, is an EU rule that gives citizens the power to demand data about them be deleted. In the case of search engines, requesting that links to web pages that may contain sensitive personal information about them.

Google eventually won a case against the French privacy watchdogCommission nationale de linformatique et des liberts (CNIL). In 2015, CNIL ordered Google to remove search results containingdamaging or false information about a person. The agency wanted the results to be removed for Google search around the world. Googles victory meant it didnt have to apply the rule globally, instead only in the EU.

In this case, Google may be right. The Australian defamation verdict would likely result in more such cases and probably more of the global rule versus local rule for the removal of search results.

Read more from the original source:
Australian Defamation Verdict Causes Google to Cry Censorship - The Mac Observer

At the weekend, a court ordered the imprisonment of reporter Sedef Kabas, guilty of using a proverb "offensive" towards the president. In one year 79 journalists lost their jobs for their critical opinions. Another 56 were victims of violence and targeted attacks, dozens of programmes were suspended.

Istanbul (AsiaNews) - A Turkish court thisweekend ordered the remand in custody pending trial of the journalist Sedef Kabas, accused of insulting President Recep Tayyip Erdogan using a famous proverb.

On 22 January, at two o'clock in the morning, officers detained the famous reporter, locking her up in a cell at police headquarters in Istanbul. The next day she appeared before the judges, who ordered her arrest using an article of law that has put tens of thousands of people in prison before her in recent years.

During a TV broadcast on a station close to the oppositions (and in a subsequent tweet), the journalist used a Circassian proverb that reads: "The ox does not become a king because it enters the palace, ratherthe palace becomes a stable", making a comparison with President Erdogan's years in power. According to the court, Sedef Kabas expressed a 'vulgar insult' against the president and the institutional office he holds.

The case is just the latest in a long line of arrests, repressions, prison sentences and fines against critical voices in Turkey, further confirming Turkey's 153rd place out of a total of 180 countries in the Reporters Without Borders (RSF) World Press Freedom Index. In addition, the Independent Communication Network (Bia) released its 2021 annual report yesterday, according to which 35 journalists were sentenced by judges last year to a total of 92 years, six months and 24 days in jail.

In the last 12 months, at least 41 Turkish journalists have been imprisoned, bringing the total figure for the last five years to 270. Not only jail, but there is also the threat of dismissal for those who do not align themselves with official propaganda: in 2021 79 journalists lost their jobs, for a total of 807 in the period 2017-2021. There is also the axe of censorship, with at least 975 articles blocked in the last year and 5,975 in the last five years.

In addition to imprisonment, there are personal attacks and targeted violence against the press: in 2021, 56 journalists were victims of violence (141 in five years). One example is the death of Hazm zsu, who worked in a radio station in Bursa, who was killed in front of his home by a person who "did not appreciate" his judgments and comments.

Finally, there are the fines imposed on broadcasters and press organisations "not aligned" with government policy and official proclamations. The Supreme Council for Radio and TV (Rtuk) imposed 158 administrative fines and suspended 48 programmes, with total fines of 31,630,000 Turkish liras (more than two million euro).

The Turkish judiciary, at the instigation of the government, represses with particular force any voice critical or "defamatory" of Erdogan. Since 2014, the year of his ascension to the presidency, at least 70 journalists have been tried and sentenced to prison and fines for "insulting the president" under Article 299 of the Penal Code. The European Council has repeatedly asked - in vain - Ankara to cancel or at least amend the rule, which continues to be applied with extreme rigour and continuity.

Here is the original post:
TURKEY Arrests, fines, censorship: freedom of the press in Erdogan's Turkey - AsiaNews

Florida GOP lawmakers are working to expand provisions in the states Civil Rights Act to protect individuals from being subjected to certain instructional materials regarding race or sex in Floridas classrooms and workplaces, potentially leading to civil actions or administrative proceedings.

At issue is an ongoing effort from the DeSantis administration to dictate how race and other topics are discussed in schools, as well as an increasing effort to limit the freedom of private businesses to make decisions for their companies.

Rep. Bryan Avila, a Republican who represents part of Miami-Dade County and the sponsor of HB 7, says that the legislation is an affirmation that people will not be judged by characteristics such as race or sex.

This bill makes it clear, that in Florida, people will be judged as individuals by their words, their characters, and their actions, Avila said at a Wednesday House Judiciary Committee meeting. The bill passed 14 to 7 (with one vote missing), and with Democrats in opposition.

This bill cripples the ability for teachers to teach effectively, said Rep. Dianne Hart, who represents part of Hillsborough County, said at the Wednesday meeting.

Every teacher Ive ever encountered, does their job from not only an academic standpoint, but from a personal one, Hart said. It is their personal experiences that they use to make the curriculum come alive for their students. Even more so for the Black and Brown students on the topic of race and discrimination.

HB 7 expands the Florida Civil Rights Act of 1992, according to the bill analysis, which secures for all individuals within the state freedom from discrimination because of race, color, religion, sex, pregnancy, national origin, age, handicap, or marital status.

The bill affects areas of education and employment, saying that individuals should not be subjected to training or materials that espouse principles such as:

/Members of one race, color, national origin, or sex are morally superior to members of another race, color, national origin, or sex.

/A person, by virtue of his or her race, color, national origin, or sex is inherently racist, sexist, or oppressive, whether consciously or unconsciously.

/A persons moral character or status as either privileged or oppressed is necessarily determined by his or her race, color, national origin, or sex.

/A person, by virtue of his or her race, color, national origin, or sex bears responsibility for, or should be discriminated against or receive adverse treatment because of, actions committed in the past by other members of the same race, color, national origin, or sex.

/A person should feel discomfort, guilt, anguish, or any other form of psychological distress on account of his or her race, color, national origin, or sex.

These principles would also apply to students and school employees under the Florida Educational Equity Act, should the bill become law. There is aSenate version of the billalso moving through the 2022 legislative session.

Those are principles that I think each and everyone of us whether youre a Democrat, whether youre a Republican, whether youre an Independent I think everyone would agree that when you look at those principles, no one would disagree with any one of those principles, Avila said at the Wednesday committee meeting.

Ben Diamond, a Democrat who represents part of Pinellas County, is opposed to the bill.

What were prepared to do is to say that if a business is engaged in the perfectly lawful exercise of diversity training, and someone in the business feels a sense of guilt or sense of anguish or has some emotional reaction to that, they can sue. How is this helping our businesses in our state? Diamond said.

Much of the debate and public testimony centered around the bills effect on schools and whether it would curtail frank discussions about United States history and race.

Aliva said that the bill does not ban the teaching of historical facts about slavery, about sexism, about racial oppression, racial segregation, or racial discrimination.

But many of the Democratic lawmakers disagreed.

At issue is an ongoing effort to dictate how race and other topics are discussed in classrooms.

In June, the Florida State Board of Education approved anew rule that prohibits critical race theoryin classrooms, claiming that the theory distorts historical events and is inconsistent with the state boards approved standards. The new rule also banned materials from The New York Times 1619 project, which focuses the establishment of the United States from perspective of Black people.

Rep. Hart brought up this attack on Critical Race Theory in debate on HB 7 Wednesday.

Critical Race Theory is not even taught in K-12 schools. Its, of course, used in law schools to increase understanding of the implication of laws. So the question becomes: What is this really about?

Ida Eskamani, representing the group Florida Rising and Florida Immigrant Coalition, said during public testimony:

This legislation is a part of a dangerous and shameful nationwide agenda to censor discussions of race and gender equality in the classrooms and the workplace.

These bills dont just set back progress this nation has made in addressing racism and sexism, they also rob young people of a fact-based education and blatantly suppresses speech about race, gender, and our collective history, she continued.

She is the sister of Democrat Rep. Anna Eskamani of Orlando. She noted that a school district in Central Florida recently canceled a professors lecture on civil rights because these policies are creating a climate of fear among historians.

According to the Orlando Sentinel from earlier this week: A Flagler College history professor planned to spend an hour Saturday teaching Osceola County teachers about the civil rights movement, his area of expertise.

But days before the workshop, the school district canceled the event because administrators wanted to vet the materials to make sure they did not run afoul ofFloridas new rule banning critical race theory, or CRT, in schools.

But supporters of these initiatives to limit how race is discussed in classrooms and in the workplace claim that certain teaching and materials espouse that a persons race or sex determine a persons character.

House Speaker Chris Sprowls, a Republican who represents part of Pinellas County, said in a written statement Wednesday:

These movements have tried to hijack the important conversation about race and use it as a pretext to attack institutions ranging from capitalism to the very idea of objective truth in the hard sciences.

Sprowls continued: They want to use the sins of the past to shut down dissent in the present. HB 7 ensures Floridas workplaces and schools are places where we can have healthy dialogues about race or diversity without losing sight that we are all, first and foremost, unique individuals.

Danielle J. Brown, Florida Phoenix

More here:
Are Lawmakers Seeking to Censor Discussions of Race and Gender in Classrooms and the Workplace? - FlaglerLive.com

WASHINGTON, January 28, 2022 A lack of cultural understanding by Facebook, Instagram, and other social media platforms is a prevailing reason for inaccurate content moderation in Iran, Middle East experts said.

Moreover, and they said, Irans proposed international internet replacement, the National Information Network, is dangerously close to coming into effect.

Speaking at a Thursday event of the Atlantic Council designed to draw attention to the current status of social media in Iran, a human rights expert said that Big Techs chronic misunderstanding of the Persian language leads to censorship of content that is either entertainment-based or posted by Iranian activists.

Panelists at the event also highlighted a new report Iranians on #SocialMedia, as the inspiration for the discussion.

Facebook needs someone who actually understands what is going on on the ground, claimed Simin Kargar, a human rights and technology research fellow at Digital Forensic Research Lab. Because the company dont employ or contract with such people, said Kargar, the platform and its sister Instagram are inappropriately censoring posts in the country.

Because of the platforms negligence in understanding and adapting to local concerns, the Iranian people are not benefiting from the internet.

And because Iran also heavily monitoring and censoring the internet within its borders, the Iranian people end up being hindered by the double-whammy of Iranian and Facebook censorship, Kargar said.

Mahsa Alimardani, a researcher with the human rights organization Article19, agreed that misconceptions due to language are a dangerous foe.She made this comment whenasked what America can do to help and whether American sanctions have played play a part in the rise in content moderation.

All panelists at the event said that while American sanctions against Iran impact the internet in the country, they are not responsible for what is currently happening in Iran.

However, Alimardani also blamed Meta, the new corporate name for the company that runs Facebook and Instagram, for improper and excessive content moderation.

She said Facebook currently flag anything related to the Iranian guard after the Trump Administration created a list of dangerous people that should be restricted on social media. She disagreed that the Islamic Revolutionary Guard Corps should be listed as a foreign terrorist organization.

The National Information Network, the new censored internet that Iran is currently working to implement, had been planned to launch in March. Alimardani said she believes that the release will be postponed because of disagreements about who within the government will control content moderation, and the impact the firewall could have on Iranian tech companies.

Alimardani highlighted the unique nature of the Iranian law that created the national internet. Instead of being voted on by the Iranian Parliament, the legislative body deferred action on the creation of a permanent national internet only until after an experimental period with the firewall, she said.

Yet the government has been pushing its own online streaming and video platforms. These platforms are part of the governments attempt to incentivize an Iranian national internet.

Essentially, said Kargar, the government is promising more bandwidth at a lower cost through the National Information Network. The new network is also appealing to Iranian consumers because the NIN will primarily be in the countrys major dialect.

Holly Dagres, a nonresident fellow with the Atlantic Councils Middle East Programs and the author of the Iranians on #SocialMedia, also spoke on the NIN. She said it would take Iran back to the Middle Ages, and also limit communication with other Iranians and with the outside world.

View original post here:
Facebook is Failing Iranians, and Iran's Leaders Are About to Launch a Censored Internet - BroadbandBreakfast.com

It is worth looking beyond a headline event from last week: Christian websites reported the US Baptist pastor John MacArthur had a sermon thrown off YouTube.

There is no such thing as transgender, Macarthur said in his Sunday sermon. You are either XX or XY, thats it. God made man male and female. That is determined genetically, that is physiology, that is science, that is reality, he said.

On the one hand, the reality of that lie and deception is so damaging, so destructive, so isolating, so corrupting that it needs to be confronted, but on the other hand, that confrontation cant exaggerate what already exists, which is a sense of feeling isolated in relationships.

The US-based conservative journalist Todd Starnes reported YouTubes response after he had uploaded a clip of the MacArthur sermon.Our team has reviewed your content, and, unfortunately, we think it violates our hate speech policy, YouTube wrote to me. Weve removed the following content from YouTube: There is no such thing as transgender. You are either XX or XY. Thats it. Pastor John MacArthur.

But a check of YouTube reveals that last Sundays sermon Such Were Some of You is on YouTube and linked to MacArthurs churchs Grace To You (GTY) site.

It is possible some words have been removed, the transcript is not yet available, but there is no hint it was censored on the GTY site.

It may be that only the Starnes clip has been removed. But the sermon itself is a very strong affirmation of a conservative Bible exegesis on homosexuality.

The concern is that Canadas Bill C-4 is broadly worded and could, in effect, ban biblical teachings on sexual ethics. Christian Broadcasting Network

But outside of the YouTube controversy, MacArthur was making news. Some 5,000 pastors preached alongside him on human sexuality on January 16.Their motivation? Protesting against the passing of C-4 Canadas new law banning sexual orientation gender conversion efforts. The concern is that Canadas Bill C-4 is broadly worded and could, in effect, ban biblical teachings on sexual ethics, and might even limit personal communications on the subject, the Christian broadcasting network reported.

Australian readers will see strong parallels with Victorias Change or Suppression (Conversion) Practices Prohibition Bill 2020 that is due to come into effect next month.

The Canadian Bill passed after two failed attempts with the unanimous consent of the Canadian House of Commons. Bill C-4 defines conversion therapy as a practice, treatment or service designed to change a persons sexual orientation or gender identity for example, repressing or reducing non-heterosexual attraction or sexual behaviour or repressing a persons non-cisgender gender identity.

But the Gospel Coalition Canada did not go along with MacArthurs approach. Gospel Coalition Canadas Paul Carter responded to the MacArthur mass preaching initiative: I have tremendous respect for Pastor John and rejoice in the fact that there are a number of initiatives intending to peacefully protest the potential abuse and misapplication of Bill C-4. However, while many pastors will no doubt participate in this particular initiative, others will have concerns due to the fact that the statement associated with this initiative concedes illegality.

Instead, Carter took part in reading a statement in church written for the Canadian Religious freedom summit, which included this key passage. The laws stated purpose is to outlaw conversion therapy. We strongly oppose the coercive and unscientific therapeutic practices the Bill was introduced to address. We appreciate and affirm the desire of parliamentarians to protect the vulnerable.

However, we are deeply concerned that the effective reach of the legislation could be extended far beyond its stated purpose. Because its definition of conversion therapy is vague, many are concerned that it could capture parents, pastors and counsellors who teach a biblical understanding of sexuality in a variety of situations. The Canadian Charter of Rights and Freedoms guarantees our freedoms of religion, conscience, thought, belief, expression and association. It is our prayer that the law will be applied and clarified as needed in such a way as to honour these Charter protections.

C-4 will be tested against Canadas Charter of Rights and Freedoms Charter, which declares freedom of religion, freedom of thought, freedom of opinion, freedom of expression and freedom of media communication as fundamental freedoms.

All initial assurances are that it does not intend to criminalize religious expression. Paul Carter

Carters prediction is that it will be difficult to see how a charge against a pastor for preaching on Genesis 1:27 or Galatians 5:22-23 could withstand a Charter challenge.

He adds: It is not clear that preaching on biblical passages espousing a biblical ethic of sexuality is now illegal. That matter has yet to be adjudicated in the courts. Bill C-4 nowhere uses that language and all initial assurances are that it does not intend to criminalize religious expression in any of the contexts suggested in the letter [from MacArthur and his allies].

The statement suggested by the Canadian Religious Freedom Summit specifically makes use of language drawn from the Charter and also makes it clear that there is no desire whatsoever, on behalf of the participants and signatories, to endorse or engage in coercive or abusive practices. The MacArthur statement expresses no such sentiment.

Carter declares that he and other pastors are hopeful that the church will not be drawn into an adversarial posture toward the LGBTIQA community while I am praying for the wise application or emendation of Bill C-4 such that abusive or coercive practices are forbidden while speaking the truth in love continues to be permitted.

He recognises it is possible that the day will come when preaching what the Bible says about human sexuality is banned and adds: If it comes when it comes I will count it an honour to suffer on behalf of Christ.

The militancy of the mass preachers protest was noted by Good Book Company author and Baptist pastor Andrew Roycroft:

The Canadian bill came into effect in early January. Its final effect may be decided by the courts, which will have the task of balancing religious expression versus the LGBTIQ2S (the Canadian initials which include the Two Spirit group) communitys desire to avoid efforts to change them.

Why not send this to a friend?

Share

See original here:
John MacArthur, YouTube censorship and 'conversion therapy' ban - Eternity News

In a world that has become about as polarized and biased as ever, it can be extremely difficult to find unbiased news sources.

Especially in the United States, those on the liberal left believe right-wing news sources like Fox News are terribly biased. Those on the conservative right are convinced liberal media sources distribute nothing but fake news.

So whos right? Which media sources are free from government or corporate censorship and actually report the facts?

Because of how polarized the country is, youll be lucky if you can find two people who agree on which news sources are the most unbiased. The reality is that there really is no such thing. There are only varying degrees of bias.

One of the best ways to determine lack of bias is by surveying the audience itself. In fact Gallup and the Knight Foundation did just that in 2017, surveying 1,440 Gallup panel members.

Whats astonishing is that the publications listed as some of the most biased by conservatives and the least biased by liberals included The Washington Post and The New York Times. Yet those were two at the top of the list of the least biased news sources published by author Paul Glader at Forbes.

This reveals that either Forbes, or Paul Glader himself, has significant liberal bias leanings himself. If you check who conservative writers say are the most unbiased publications, youll find the opposite to be true.

So whos right? How can you choose the most unbiased news sources when even those promoting allegedly unbiased sources are likely biased themselves?

Again, the check with the audience.

Using the Gallup/Knight Foundation survey itself, its easy to identify the least biased news sources. Its those that least upset both liberals and conservatives. The news sources rated in the middle of both lists represent the least bias.

Why is this the case?

Because journalists who rely on facts and evidence are least likely to ask loaded questions at a news conference (upsetting conservatives) or use offensive terms in their reporting (upsetting liberals).

Most journalists have probably read The Elements of Journalism by Bill Kovach and Tom Rosenstiel at some point in their study of journalism. Its well known as a guide for good journalism. The Ethical Journalism Network further boils the principles from that book into five core principles of journalism.

What are the characteristics of the most unbiased news sources? According to EJN:

So which news sources are actually unbiased? Which news outlets adhere best to the five core principles of journalism?

The AP is consistently featured on nearly every list of unbiased news sources. So it makes sense to list it first here.

The tagline featured at the top of the APs main page is Advancing the power of facts.

Youll notice that the language used in the news stories here even political ones are very neutral and non-inflammatory.

AP journalists focus on reporting facts citing authoritative sources and corroborating or disproving claims made by sources by providing evidence.

Just this week, youll even find one story about all of the fake news readers might have seen on social media. The AP then confirmed or debunked those claims using hard facts.

The Top Stories section of the AP website is an excellent place to get the latest news around the world. There is also an interesting Video section with news segments, and Listen section which is the AP Radio broadcasts, updated hourly.

PBS News is one of the few media outlets that remains immune from major accusations of bias.

Stories here feature both sides of every issue. When you read quotes from politicians or other major figures, youll read them with all of the important context provided. Context is a critical part of balanced reporting, and PBS News does it far better than most other media sources.

On PBS News youll find news categories like Politics, Health, World, Nation, Economy, and much more. It also features a Full Episodes section of all news broadcasts, a Podcasts section of their most popular audio news podcasts, and a Watch live link at the top of the page for watching current, ongoing news broadcasts.

If you need a U.S. focused source of news that you can trust to remain fair and balanced in their reporting, you cant go wrong with PBS News.

While NPR news is often featured on most lists of balanced news sources, it also has more accusations of being left-leaning than other news sources on this list.

These accusations are usually related to a couple of NPR journalists who are less adept at keeping their political leanings in the background. However most NPR journalists adhere strongly to all of the core principles of journalism, especially accountability.

NPR listeners and readers who reach out to NPR will find that journalists are excellent at responding to criticism. In many cases they will even correct stories or seek to provide better balance on the topics where they hear criticism or complaints from their audience.

NPR offers a tremendous variety of news content for their audience. Youll find categories on the site like Politics, Business, Technology, Science, and Race & Culture.

NPR also offers wonderful shows and podcasts, and even a music section focused on helping the NPR audience better appreciate existing, new, and alternative artists and music.

For the most part, when you explore NPR news stories aside from the occasional left-leaning comments or slant from some of the journalists for the most part youll find balanced and fact-based reporting there.

According to a 2014 Pew Research Study, 40% of CBS News audience are left-leaning, while only 20% are right-leaning. While this is often pointed to by political conservatives that CBS News has a left-leaning bias, the truth is that the remaining CBS News audience is center-aligned.

This means that the CBS News audience is much more politically balanced than many other news outlets.

Even when covering very controversial issues, CBS News uses balanced and neutral language in its reporting. Article titles found across the site are matter-of-fact, and reporting includes context as well as opinions from all sides of any debate.

Other major network news outlets like NBC or ABC are often accused by conservatives of featuring inflammatory anti-right headlines, CBS News more often goes unscathed from those accusations.

In fact in the Gallup/Knight Foundation Survey, CBS News was rated far higher by conservatives than CNN, MSNBC, and NBC News. Liberals rated CBS News higher than CNN, USA Today, and even The Washington Post.

Two media-bias rating services, Media Bias/Fact Check and AllSides, both rank BBC News in the center of news reporting.

This means that for independent news readers, BBC news reporting stands as one of the best options. This may be due to the fact that the BBC is based in Britain so remains uninfluenced by U.S. political influence.

U.S. news coverage on the BBC is surprisingly balanced. Headlines are matter-of-fact and avoid pulling punches against either side of the U.S. political spectrum.

BBC also offers an extremely wide range of news coverage, including world news, business, science, health, and even a reality check section that debunks fake news you might have seen on social media or on other news sites.

Both Media Bias/Fact Check and AllSides report Reuters as even less biased in its reporting than the BBC.

It doesnt take long browsing the Reuters website to see why this is the case. Article titles there are refreshingly neutral, and journalists there are heavily fact-based in their reporting.

Despite that neutrality, Reuters journalism doesnt pull any punches. Youll find U.S based stories there that take a hard line against corruption, unethical politicians on both sides of the aisle, and well-rounded coverage of every issue.

Reuters is an excellent source of news for business and markets, politics, and even technology and lifestyle issues.

Theres also a TV link in the header where you can watch all of Reuters past video news broadcasts. Select the flag in the header to change Reuters coverage to your own region of the world.

One surprising contender on the list of unbiased news sources, given the name, is Christian Science Monitor. Both AllSides and Media Bias/Fact Check list Christian Science Monitor as center-balanced and unbiased.

This news organization has built a reputation over the years as providing extremely fair, in-depth exploration into important issues in the world today. Unlike the more biased media outlets out there, Christian Science Monitor works hard to bring in perspectives from all sides.

You wont find article titles on this site that seek to malign or unfairly cover one political side or another. Stories dive into the most relevant issues that matter to the audience, providing full context and drawing in multiple sources to corroborate all facts.

The site content is mixed with both text and video news content. Youll also find podcasts, photos of the week, and even a section devoted to book reviews.

Original post:
7 Unbiased News Sources Free From Censorship

DUBAI, United Arab Emirates (AP) The United Arab Emirates announced on Sunday that it will no longer censor films released in cinemas, the countrys latest effort to boost its brand as a liberal hub attractive to foreigners.

Instead of cutting sensitive scenes that could offend traditional Islamic sensibilities, the Emirati Media Regulatory Authority will introduce a new 21+ age category for viewers.

The movies will be screened in cinemas according to their international version, the authority said in a Twitter post.

Censors in the UAE, like elsewhere in the Middle East, have long removed scenes in cinematic releases that show nudity, homosexuality, sex and other content deemed inappropriate sometimes leading to plot holes.

Foreigners outnumber locals nearly nine to one in the federation of seven sheikhdoms. The diversity of culture and religion in the tourism-dependent country has at times been at odds with its Islamic laws and traditions.

But thats changing as the nation promotes its socially liberal environment to lure international workers. The government has reformed its Islamic legal code and next year will change its weekend to Saturday-Sunday to align with the Western businesses and markets.

Original post:
United Arab Emirates to end censorship of cinematic ...

Significance

We study the impact of crisis on information seeking in authoritarian regimes. Using digital trace data from China during the COVID-19 crisis, we show that crisis motivates citizens to seek out crisis-related information, which subsequently exposes them to unrelated and potentially regime-damaging information. This gateway to both current and historically sensitive content is not found for individuals in countries without extensive online censorship. While information seeking increases during crisis under all forms of governance, the added gateway to previously unknown and sensitive content is disproportionate in authoritarian contexts.

Crisis motivates people to track news closely, and this increased engagement can expose individuals to politically sensitive information unrelated to the initial crisis. We use the case of the COVID-19 outbreak in China to examine how crisis affects information seeking in countries that normally exert significant control over access to media. The crisis spurred censorship circumvention and access to international news and political content on websites blocked in China. Once individuals circumvented censorship, they not only received more information about the crisis itself but also accessed unrelated information that the regime has long censored. Using comparisons to democratic and other authoritarian countries also affected by early outbreaks, the findings suggest that people blocked from accessing information most of the time might disproportionately and collectively access that long-hidden information during a crisis. Evaluations resulting from this access, negative or positive for a government, might draw on both current events and censored history.

Scholars have long predicted that during crises or uncertain time periods, people will rely more on mass media for information relevant to their own safety and spend more time seeking out information (1). Increased attention to media during crisis has been shown empirically in democracies, such as during democratization in Eastern Europe (2), during the eruption of Mount St. Helens (3), and immediately after the September 11 terrorist attacks (46). Increased attention to the media presents opportunities for large changes in opinion or political socialization (2, 7), and crisis disruptions can also shift attention toward entertainment due to lack of mobility and boredom (8).

This paper identifies another effect of crisis: abrupt exposure to prior sensitive information blocked by governments. We examine the effect of crisis on information seeking in highly censored environments by studying the impact of the COVID-19 public health crisis on censorship circumvention in China. In January and February of 2020, COVID-19 cases in China were spiking, official news sources were slow to acknowledge the crisis, and many regions of China restricted movement. Using a variety of measures of Twitter and Wikipedia data, both of which are inaccessible within China, we show large and sustained impacts of the crisis on circumvention of censorship in China. For example, the number of daily, geolocating users of Twitter in China increases by up to 40% during the crisis and is 10% higher long term, while politically sensitive accounts gain tens of thousands of excess followers, up to 3.8 times more than under normal circumstances, and these followers persist 1 y after the crisiss end. Moreover, beyond information seeking about the crisis itself, we find that information seeking across the Great Firewall extended to information the Chinese Communist Party (CCP) has long censored, including information about sensitive historical political events and leaders.

Although just one of many crises, the global nature of the COVID-19 crisis makes this case a unique and important opportunity to compare information seeking during crisis in China to that in other countries that had similar COVID-19 outbreaks. To draw a comparison, we investigate the same patterns in countries with no censorship or in authoritarian regimes where the platforms we study are not censored that also experienced large outbreaks of COVID-19 cases soon after China. Consistent with other work on information seeking during lockdown in democracies (8), we find higher levels of engagement with online news media generally in comparison countries, but do not observe users seeking information about sensitive political topics unrelated to the crisis.

Together, these findings demonstrate that during crisis access to information fundamentally changes in autocracies in patterns that differ from democracies. Information spillovers originating from crisis could be especially pronounced when a regime has previously censored a large amount of political information and circumvention tools provide access to a wide variety of current and historical censored content. That information seeking during crisis spills over to unrelated and previously censored content in authoritarian contexts is related to previously studied gateway effects where the Chinese governments action to suddenly block a primarily entertainment website facilitated access to censored political information (9). However, our overall results and country comparisons suggest a broader implication: that the abrupt and wide-ranging consumption of hidden information may be a feature of censorship regimes themselves and can occur with or without contemporaneous government action to bring it about. This spillover effect is further robust enough that an ongoing crisis does not appear to distract from long-censored informationattention to information expands to include both the crisis and censored history. These results provide an important contribution to the literature on the impacts of crisis on authoritarian resilience and governance (1012).

While access to information the regime censors dramatically increases during crisis, note that we do not know the overall impact on public opinion. In the case of the COVID-19 crisis in China, access to blocked platforms facilitates access not only to censored information sensitive to China but also to the Western media, which contains a wide range of negative news about the United States and other democracies. It is generally difficult to infer true levels of support for authoritarian regimes (because of preference falsification) (1315), but we draw out the potential political consequences of increased censorship circumvention in this papers Discussion.

In many authoritarian countries, traditional and online media limit access to information (1619). While this control is imperfect, studies have shown that media control in autocracies has large effects on the opinions of the general public and the resilience of authoritarian regimes (2026), even though there are moments when it can backfire (9, 2732). Evidence from China suggests that media control may be effective in part because individuals generally do not expend significant energy to find censored or alternative sources of information.*

While many have studied the impact of information control in normal times in authoritarian regimes, less is known about information seeking during crisis. In democracies, information seeking intensifies during crisis, increasing consumption of mass media. Ball-Rokeach and Defleur (1) describe a model of dependency on the media where audiences are more reliant on mass media during certain time periods, especially when there are high levels of conflict and change in society. These findings are largely consistent with research on emotion in politics, which concludes that political situations that produce anxiety motivate people to seek out information (34). While in normal times information seeking is strongly influenced by preexisting beliefs, several studies have suggested that crisis can cause people to seek out information that might contradict their partisanship or worldview (7, 35), although they may pay disproportionate attention to threatening information (36).

Similar patterns may exist in authoritarian environments. Because the government controls mass media, citizens aware of censorship may not only consume more mass media that is readily available during crises, but also seek to circumvent censorship or seek out alternative sources of information that they may normally not access. For example, during the severe acute respiratory syndrome (SARS) crisis in China in 2003, Tai and Sun (37) find that people in China turned to Short Message Service (SMS) and the Internet to gather and corroborate information they received from mass media. Cao (38) shows an increase in censorship evasion and use of Twitter from China during regime-worsening events, such as worsening of trade relations between the United States and China and the removal of presidential term limits in the constitution in 2018.

Outside of facilitating access to information about the crisis, evasion of censorship during crisis could also provide information that has long been censored. In particular, a crisis could create spillovers of information, where evasion to find one piece of information facilitates access to a broad range of content. This phenomenon is related to the entertainment-driven gateway effect documented in ref. 9, where sudden censorship of an entertainment website (Instagram) motivated censorship evasion and thus facilitated access to unrelated political information. At the same time, crisis is a very different context than is sudden censorship of an entertainment website. Anxiety about the epidemic, perhaps especially when accompanied by boredom during quarantine and lockdown, could lead consumers of information to be more likely to seek out information that has long been censored after they have evaded censorship to better understand the trustworthiness of their government. On the other hand, the crisis itself may be sufficiently distracting to make them less likely to seek out unrelated and long-censored information. Further, crisis-induced spillover effects are more difficult for autocrats to avoid than gateways created through censorship of entertainment websites, which could be reduced by avoiding the initial censorship altogether or implementing less visible censorship. While the overall impact on the autocrat is unknown and could be outweighed by a successful, rapid government response to the crisis, such a gateway would strengthen the ability of consumers to read sources outside of China.

On 31 December 2019, officials in Wuhan, China confirmed that a pneumonia-like illness had infected dozens of people. By 7 January 2020, Chinese health officials had identified the diseasea new type of coronavirus called novel coronavirus, later renamed COVID-19. By 10 January, the first death from COVID-19 was reported in China, and soon the first case of COVID-19 was reported outside of China, in Thailand. As of December 2020, COVID-19 has infected over 91,000 people in China with over 4,500 deaths and at least 73.5 million people worldwide with over 1.6 million deaths.

While initial reports of COVID-19 were delayed by officials in Wuhan (39), Chinese officials took quick steps to contain the virus after it was officially identified and the first deaths were reported. On 23 January 2020, the entire city was placed under quarantinethe government disallowed transportation to and from the city and placed residents of the city on lockdown (40). The next day, similar restrictions were placed on nine other cities in Hubei province (41). While Hubei province and Wuhan were most affected by the outbreak, cities all over China were subject to similar lockdowns. By mid-February, about half of China780 million peoplewere living under some sort of travel restrictions (42). Between 10 January and 29 February 2020, 2,169 people in Wuhan died of the virus (43).

We use digital trace data to understand the effect of the COVID-19 crisis on information seeking. Table1 summarizes the empirical tests conducted in this paper. First, we show that the crisis increased the popularity of virtual private network (VPN) applications, which are necessary to jump the Great Firewall, downloaded on iPhones in China. We also show that the crisis expanded the number of Twitter users in China, which has been blocked by the Great Firewall since 2009. The crisis further increased the number of page views of Chinese language Wikipedia, which has been blocked by the Great Firewall since 2015. We also show that the areas more affected by the crisissuch as Wuhan and Hubei Provincewere more likely to see increases in circumvention.

Next, we show that the increase in circumvention caused by the crisis not only expanded access to information about the crisis, but also expanded access to information that the Chinese government censors. On Twitter, blocked Chinese language news organizations and exiled dissidents disproportionately increased their followings from mainland China users. On Wikipedia, sensitive pages such as those pertaining to Chinese officials, sensitive historical events, and dissidents showed large increases in page views due to the crisis. Finally, Comparison with Other Countries Affected by the Crisis shows that these dynamics do not occur on Italian, German, Persian, or Russian Wikipedialanguages of countries with similar crises but where Wikipedia is uncensored.

We show that censorship circumvention increased in China as a result of the crisis using data from application analytics firm App Annie, which tracks the ranking of iPhone applications in China. While most VPN applications are blocked from the iPhone Apple Store, we identified one still available on it. Around the time of the Hubei lockdown, its rank popularity increased significantly and maintained that ranking (Fig.1, Top).

Download rank of iPhone application in China: Facebook, Twitter, and Wikipedia. Data are from App Annie. Top intentionally omits the name of the VPN app and its precise ranking.

Concurrent with the increase in popularity of the VPN application is a sudden increase in popularity of Facebook, Twitter, and Wikipedia applications, as Fig. 1 shows. These increases indicate that those jumping the Firewall as a result of the crisis were engaging in part with long-blocked websites in ChinaTwitter and Facebook have been blocked since 2009 and Chinese language Wikipedia since 2015.

This finding is consistent with data we collected directly from Twitter and Wikipedia. Fig.2, Top shows the number of geolocating users in China posting to Twitter in Chinese in the time period of interest. Immediately following the lockdown, Chinese language accounts geolocating to China increased 1.4-fold, and postlockdown, 10% more accounts were active from China than before. Fig. 2, Bottom shows that the crisis also coincided with increases of new users, indicating that increases are due to new users and not dormant ones reactivating. We provide a rough, back-of-the-envelope calculation for the absolute size of these effects. If there were 3.2 million Twitter users in China (44) prior to the COVID-19 pandemic and the 10% increase in usage applies generally to Twitter users (i.e., not just those geotagging), then 320,000 new users joined Twitter because of the crisis, including users who do not post or post publicly. We assess this estimate in SI Appendix, section 4 using the estimated fraction of posts in Chinese that are geotagged (1.95%) and the total number of unique Twitter users in our sample (47,389 users posting in Chinese and in China).

(Top) Number of unique geolocating users in China posting in Chinese. (Bottom) The fraction of active unique users who joined Twitter in the last 30 d. The decline in new users after the end of lockdown (Bottom Right) is driven by a decline in new signups after lockdown easing, rather than lockdown users leaving the site (they are no longer considered new after 30 d).

Data from Wikipedia on the number of views of Wikipedia pages by language match the App Annie and Twitter patterns.# We measure the total number of views for Chinese language Wikipedia by day from before the coronavirus crisis to the time of writing. Fig.3 reveals large and sustained increases in views of Chinese language Wikipedia, beginning at the Wuhan lockdown and continuing above pre-COVID levels through May 2020. Views of all Wikipedia pages in Chinese increased by around 10% during lockdown and by around 15% after the first month of lockdown. This increase persisted long after the crisis subsided. In absolute terms, the total number of page views increases from around 12.8 million views per day in December 2019 to 13.9 million during the lockdown period (24 January through 13 March) and up to 14.7 million views per day from mid-February through the end of April.

Views of Wikipedia pages in Chinese. Shown is the ratio of total daily views of Wikipedia pages in Chinese compared to December 2019 views (12.7 million views per day in December 2019). The beginning of the Hubei lockdown and the first relaxation of lockdown in Hubei are indicated in gray.

Whereas the data from App Annie and Wikipedia cannot distinguish between circumvention patterns within China, the geolocation in the Twitter data enables the examination of subnational variation. Circumvention occurred in provinces throughout China as a result of the Wuhan lockdown; Hubei, the most impacted province, experienced the most sustained increase in geolocated users.

Fig.4 measures the initial increase of Twitter volume on 24 January 2020, the day after Wuhans lockdown and the start of lockdown in 12 other cities in Hubei, in comparison to the average from 1 December 2020 to 22 January 2020 in each province in China (the x axis). The y axis measures how sustained the increase wasthe ratio of Twitter volume 30 d after the quarantine to the baseline before the outbreak. Hubei is in the top right corner of the plot: Twitter volume there doubled in comparison to the previous baseline, and the doubling persisted 30 d after the crisis. These estimates are drawn from polynomial models fitted to the daily number of users per provinceSI Appendix, Fig. A1 displays the modeled lines over the raw data for each province.

Increases in geolocated Twitter activity by province (modeled). Shown is the increase in geolocated Twitter users compared to the average number of geolocated Twitter users in a province before the Hubei lockdown. Estimates for 30 d after and day of lockdown are drawn from a five-term polynomial regression on the number of unique geolocated Twitter users per day after the lockdown. These province-by-province polynomials are displayed over the raw data in SI Appendix, Fig. A1.

To further validate that this increase in Twitter usage in China is related to the Wuhan lockdown, we collected real-time human mobility data from Baidu, one of the most popular map service providers in China. The decrease in mobility in 2020 is correlated with the increase in Twitter users across provinces in China, net of a New Years effect (SI Appendix, Fig. A3). However, as the crisis spreads, the demobilization effect disappears, while Twitter usage remains elevated. The overall increase in Twitter users across China 2 wk after the lockdown and beyond cannot be explained by further decreases in mobility or New Year seasonality (SI Appendix, Fig. A4). SI Appendix, section 3 presents more detail.

This subsection examines how the crisis impacted what content Twitter users from mainland China and users of Chinese language Wikipedia were consuming. Both Twitter and Wikipedia facilitate access to a wide range of content, not just information sensitive to the Chinese government. New users of Twitter from China might follow Twitter accounts producing entertainment or even Twitter accounts of Chinese state media and officials, who have become increasingly vocal on the banned platform (45). New users of Wikipedia might seek out only information about the virus and not about politics. If the crisis produced a gateway effect, we should see increases in consumption of sensitive political information unrelated to the crisis.

We use data from Twitter to examine what types of accounts received the largest increases in followers from China due to the crisis. For this purpose, we identify 5,000 accounts that are commonly followed by Twitter users located in China.** Materials and Methods and SI Appendix, section 2 detail how we identified these accounts.

We assigned each of the 5,000 popular accounts into one of six categories: 1) international sources of political information, including international news agencies; 2) Chinese citizen journalists or political commentators, which include nonstate media discussions of politics within China; 3) activists or accounts disseminating information about politics in the United States, Taiwan, or Hong Kong; 4) accounts disseminating pornography; 5) state media and political figures; and 6) entertainment or commercial influencers. Categories 1 to 3 are accounts that might distribute information sensitive to the Chinese government, such as international media blocked by the Great Firewall (e.g., New York Times Chinese and Wall Street Journal Chinese); Chinese citizen journalists and political commentators such as exiled political cartoonist Badiucao and currently detained blogger Yang Hengjun; and political activists such as free speech advocate Wen Yunchao and Wuer Kaixi, former student leader of the 1989 Tiananmen Square protests. Accounts in category 4 are pornography, which we consider sensitive because it is generally censored by the Chinese government, but not politically sensitive like categories 1 to 3. Accounts in category 5 include accounts linked to the Chinese government, including the governments news mouthpieces Xinhua and Peoples Daily, as well as the Twitter accounts of Chinese embassies in Pakistan and Japan. Category 6 is also not sensitive, as these accounts mostly do not tweet about politics, but instead are entertainment or commercial accounts or accounts of nonpolitical individuals.

We want to understand how the coronavirus crisis affected trends in follower counts of each of the six categories and, in particular, compare how the crisis affected the followings of categories 1 to 3 to those in categories 5 and 6. We therefore downloaded the profile information of all accounts that began following popular accounts in categories 1 to 3 and 5 and 6 and a random sample of popular accounts from category 4 after 1 November 2019. We then use the location field to identify which of the 38,050,454 followers are from mainland China or Hong Kong (see SI Appendix, section 2 for more details).

Because Twitter returns follower lists in reverse chronological order, we can infer when an account started following another account (46). For the accounts in the six categories, we compare the increase in followers from mainland China to the increase in followers from Hong Kong accounts relative to their December 2019 baselines; we chose Hong Kong because it is part of the Peoples Republic of China but is not affected by the Firewall. The ultimate quantity of interest is the ratio of these two increases. If the ratio is greater than one, then the increase in following relationships is more pronounced among mainland Twitter users compared to those from Hong Kong.

Fig.5 shows this ratio by category day. Relative to Hong Kong, the crisis in mainland China inspired disproportionate increases in the number of followers of international news agencies, Chinese citizen journalists, and activists (some of whom might otherwise, without exposure on Twitter, be obscure within China, especially ones who have been banned from public discourse for a long time)users who are considered sensitive and often have long been censored. In comparison, there is only a small increase in mainland followers of Chinese state media and political figures during the lockdown period and a slight decrease for nonpolitical bloggers and entertainers. Fig.6 reports the regression estimate for the relative ratio of number of new followers (akin to a difference-in-differences design with Hong Kong as control group and December 2019 as pretreatment period). The result is the same.

Increases in Twitter followers from China vs. Hong Kong by category. Shown is the gain in followers from mainland China compared to Hong Kong across six types of popular accounts, relative to December 2019 trends. Ratios here approximate the incidence rate ratios estimated in the models for Fig. 6. Each dot represents that category-days ratio. The blue lines indicate the moving averages, and the red lines represent the average during Wuhan lockdown. A value greater than 1 means more followers than expected from mainland China than from Hong Kong. Accounts creating sensitive, censored information receive more followers than expected once the Wuhan lockdown starts. Accounts that are not sensitive or censored, such as state media or entertainment, do not see greater than expected increases.

Increases in Twitter followers in China vs. Hong Kong by category (regression estimate). Incidence rate ratios shown are from negative binomial regressions of number of new followers on the interaction between indicator variables for in lockdown period and in mainland China, with December 2019 as control period and Hong Kong as control group.

We then demonstrate that the result does not depend on the choice of comparison group, and the relative increase starts no earlier than the Wuhan lockdown. SI Appendix, Fig. A6 conducts a placebo test by running weekly regressions, showing that the relative increase in followers in China starts precisely during the week of lockdown. In SI Appendix, Figs. A7A9 show that the same pattern holds with alternative comparison groups such as overseas Chinese in Taiwan and the United States.

Chinese government information operations on Twitter do not explain the results. Of the 28,991 accounts Twitter identified as belonging to a Chinese government information operation, none author a tweet in the 1,448,850 streamed geolocated corpus. To confirm this paucity, we then analyze the 14,189,518 tweets Twitter provided from the information operation accounts. Only 0.03% of those tweets are geotagged. Twelve of the 1.45 million tweets mention five information operation accounts. We then download tweets from 1,000 users from China and find zero mentions or retweets of the information operation accounts. We also find that none of these information operation accounts follow any of the popular accounts for which we collected followers.

SI Appendix, section 4 provides effect size estimates. There, we roughly estimate that around 320,000 new users came from China. Further, based on December 2019 follower growth rates, 53,860 excess accounts follow citizen journalists and political bloggers, 52,144 for international news agencies. By the end of the lockdown, citizen journalists and political bloggers benefit from 3.63 times the number of followers they otherwise would have had and activists from 2.97 times. Importantly, 8890% of the followers from China follow accounts in these categories 1 y later, and these rates are higher than for accounts which start following in the weeks after the end of the Hubei lockdown. In addition, SI Appendix, Fig. A10 shows that new users from China persist in tweeting at the same rates as those from Hong Kong and Taiwan.

To better understand patterns of political views in the Wikipedia data, we leverage existing lists (see Materials and Methods for additional details) to categorize the Chinese language Wikipedia views into three different categories: 1) Wikipedia pages that were selectively blocked by the Great Firewall prior to Wikipedias move to https (after which all of Chinese language Wikipedia was blocked), 2) pages that describe high-level Chinese officials, and 3) historical leaders of China since Mao Zedong. Whereas we would expect that a crisis in any country should inspire more information seeking about current leaders in category 2, only if crisis created a gateway to historically sensitive information would we expect proportional increases in information seeking about historical leaders in category 3 or information about sensitive events that were selectively blocked by the Great Firewall on Wikipedia prior to 2015 in category 1.

Fig.7 shows the increase in page views for each of these categories on Chinese Wikipedia relative to the rest of Chinese language Wikipedia. We find that the lockdown not only increased views of current leaders (purple), but also increased views of historical leaders (yellow) and views of pages selectively blocked by the Great Firewall (red). In SI Appendix, Tables A2 and A3 show specific pages disproportionately affected by the increase in views of Wikipedia. While pages related to coronavirus experienced a jump in popularity, other unrelated sensitive pages including the June 4 Incident, Ai Weiwei, and New Tang Dynasty Television (a television broadcaster affiliated with Falun Gong) also experienced an increase in page views.

Views of blocked, current leader, and historical leader Wikipedia pages in Chinese, German, and Italian. Vertical lines indicate the starts and ends of lockdown periods. See SI Appendix, Table A4 for specific dates. ZH, Chinese; DE, German; IT, Italian.

For more detail on this analysis as well as the Wikipedia pages that received the largest absolute and relative increases in traffic, see SI Appendix, section 6.

Since information seeking during crisis is common (1), we investigate Wikipedia data in other languages to explore how other countries were affected by the crisis. We show that the gateway effect of crisis on historically sensitive information is unique to the currently censored webpages in China. For comparison, we focus on Iran, another authoritarian country affected by COVID-19 that previously censored Wikipedia (but does not any longer), and Russia, an authoritarian country that does not censor Wikipediafor Iran, like China, we know which Wikipedia pages were previously censored (47). We also show data from democracies without censorship affected early on by the COVID-19 crisis, Italy and Germany.##

To make the comparison, we use lists of current leaders from these countries (based on office lists in the CIA World Factbook) (Materials and Methods) and create lists of historical leaders using de facto country leaders since World War II (see SI Appendix, Table A4 for a list of these titles and offices). All of these countries were affected by the crisis in late February or early March, and Italy imposed relatively stringent lockdowns. Therefore, we expect increases in information seeking for current leaders, as citizens begin to pay more attention to current politics as the crisis hits. However, none of these countries block Wikipedia. Information seeking about the current crisis therefore should not act as a gateway to information about historical events or controversies, as these pages are always available to the public.

Table 2 shows these results. While overall Wikipedia views and page views of current leaders increase in three of four comparison languages, only for Chinese language Wikipedia do historical leaders increase disproportionately and consistently throughout the whole time period. That is, we see an overall effect on information seeking throughout the world, including for historical leaders; for Chinese language Wikipedia, we see larger increases for historical leaders compared to Wikipedia page views in general. The small increases in historical political leader page views in German and Italian did not correspond with the start of the COVID-19 crisis or their respective lockdowns (Fig. 7).

During the lockdown period, Wikipedia views in Chinese increased relative to overall views for politically sensitive Wikipedia pages and political leader pages, as well as for historical political leaders

Further, we do not see increased attention to pages previously blocked in Iran (47) during the crisisWikipedia pages that can now be accessed without restriction in Iran.

In SI Appendix, section 6.2, we replicate these results for much larger sets of 1) historical leaders and 2) politically sensitive pages (pages related to the pre-https blocked pages in Iran and China and political opposition pages in Russia). We expand these sets of pages using Wikipedia2vec (48) and find that very broad information seeking about historical leaders and politically sensitive topics occurred only for Chinese language Wikipedia.

Crisis in highly censored environments creates widespread spillovers in exposures to sensitive, censored information, including information not directly related to the crisis. Like in democracies, consumers of information in autocracies seek out information and depend on the media during crisis. However, in highly censored environments, increased information seeking also incentivizes censorship circumvention. This new ability to evade censorship allows users to discover a wider variety of information than they may have initially sought, and users could also be particularly motivated to seek out accumulated, hidden information during a crisis. Our results suggest that informational spillovers produced by censorship evasion are a result of the structure of censorship and that they occur beyond government-induced backfire from sudden censorship of popular entertainment websites (9).

Public exposure to censored information during crisis is almost certainly not the intention of any regime with widespread censorship. However, the effect of this crisis-induced gateway to censored information on public opinion is unknown. In the case studied in this paper, surveys in China show increased support for the CCP over the course of the pandemic (and over the same time as large declines in favorability toward the United States) (49), even though we show that this increase in support occurs in conjunction with increased access to censored information. These findings could reflect favorable reactions to the governments pandemic policy response that may have overwhelmed negative impacts of access to censored information (50). Or the increase in support at a time of greater evasion of censorship could lend support to previous findings that access to Western news sources can counterintuitively increase support for the regime (51, 52). Studying the impact of evasion during the crisis on public opinion is left to future research. However, we include in SI Appendix, section 7 an exploratory analysis of the content posted by the popular accounts followed by our sample. While we see quite negative coverage of China on these accounts and coverage of sensitive topics such as human rights, the 1989 Tiananmen Square protests, and protests in Hong Kong, we also find that coverage of the United States by international news agencies was much more negative or neutral than positive, and the United States could have served as a favorable comparison for China and the Chinese governments handling of the pandemic.

While evaluations of responses to an ongoing crisis and comparisons to other governments responses to the same crisis may have benefited government officials in China in this particular circumstance (50), beyond these evaluations, increased access to historical and long-censored information, as documented here, has the potential to dampen positive or compound negative changes in trust and may also contribute to easier access to uncensored information about a government in the future. Natural disasters, including epidemics, tend to alter trust in government officials. When a policy response is perceived as efficacious, support for the level of government perceived to have directed the response increases (12, 53). On the other hand, neglectful responses can induce subsequent protest participation (11). In China, the average effect of natural disasters from 2007 to 2011 was to decrease political trust, and internet users have decreased baseline levels of political trust (53, 54). At the same time, political surveys in China suffer from preference falsification (1315), complicating our efforts to understand the political consequences of these events.

While the results here do not link the COVID-19 crisis gateway effect to the political fortunes of the Chinese government, they do suggest that a country with a highly censored environment sees distinctive and wide-ranging increases in information access during crisis. While in normal times censorship can be highly effective and widely tolerated, crisis heightens incentives to circumvent censorship, and regimes cannot rely on the same limits on information access during crisis, even for topics long controlled.

Download rank data for Facebook, Twitter, Wikipedia, and the VPN app come from application analytics firm App Annie (https://www.appannie.com), which tracks the popularity of iPhone application downloads in China. While most VPN applications are blocked from the iPhone Apple Store (and there are other means of obtaining VPNs), we identified one still available on it. VPN download rank shown in the text is for that VPN application. These data contain the ranking of an applicationfor Wikipedia, its rank within the Reference App categoryrather than the number of downloads. To protect the VPN application and its users, we do not disclose its name or the exact ranking.

For the Twitter analyses, we collected 1,448,850 tweets (101,553 accounts) from mainland China from 1 December 2019 until 30 June 2020. These tweets were identified using Twitters POST statuses/filter endpoint. Our analyses are limited to the 367,875 that were posted in Chinese (47,389 accounts that posted in Chinese, 43,114 that had names or descriptions in Chinese).

The Twitter follower analysis examines accounts that Twitter users from China commonly follow. To find those accounts, we randomly sampled 5,000 users geolocated to China. For each of these users, we gathered the entire list of whom they follow, their Twitter friends. From these 1,818,159 friends, we extracted the 5,000 most common accounts. We also selected only accounts that were Chinese language accounts or had Chinese characters in their name or description field to ensure that we were studying relevant accounts: those disseminating information easily accessible to most Chinese users. SI Appendix, section 2 provides more detail.

We downloaded the profile information of all accounts that began following these popular accounts after 1 November 2019. Because Twitter returns follower lists in reverse chronological order, we can infer when an account started following another account (46). We then use the location field to identify which of these 38,050,454 followers are from mainland China or Hong Kong (see SI Appendix, section 2 for more details). We downloaded all new followers of nonpornography accounts and all new followers of a random selection of 200 pornography accounts (the majority of the accounts were pornography). This sampling allows us to estimate the impact of the coronavirus on pornography while decreasing our requests to the Twitter Application Programming Interface.

Human mobility data are publicly available from Baidu Qianxi (https://qianxi.baidu.com/2020/), which tracks real-time movement of mobile devices and is used in studies of human mobility and COVID-19 containment measures (55). Our robustness checks use data across China during the Lunar New Year period in both 2020 and 2019. We extracted the data from the webpage, including the daily within-city movement index (an indexed measure of commuter population relative to the population of the city) as well as daily moving-out index (an indexed measure based on the volume of population moving out of the province relative to the total volume of migrating population on that day across all provinces in China). See SI Appendix, section 3 for more details.

Data on the number of Wikipedia page views are publicly available at https://dumps.wikimedia.org/other/pagecounts-ez/merged/. To better understand patterns of political views in the Wikipedia data, we use existing lists to categorize the Chinese language Wikipedia views into three different categories: 1) Wikipedia pages that were selectively blocked by the Great Firewall (https://www.greatfire.org/ maintains a list of websites censored by the Great Firewall) prior to Wikipedias move to https, after which all of Wikipedia was blocked; 2) pages about high-level Chinese officials (using offices listed in the CIA World Factbook, https://web.archive.org/web/20201016160945/ https://www.cia.gov/library/publications/resources/world-leaders-1/CH.html, excluding Hong Kong and Macau as well as the Ambassador to the United States); and 3) historical paramount leaders of China since Mao Zedong.

In comparing multiple languages and countries, we use the same offices listed in the CIA World Factbook to create lists of current leaders from Iran, Russia, Italy, and Germany (for office holders as of February 2020) and create lists of historical leaders using de facto country leaders since World War II. See SI Appendix, Table A4 for a list of these titles and offices, as well as the lockdown start and end dates used in the Wikipedia page view models displayed in Table 2. The list of Wikipedia pages blocked in Iran was published by Nazeri and Anderson (47).

In SI Appendix, section 6.2, we replicate the Wikipedia page view results for much larger sets of 1) historical leaders and 2) politically sensitive pages (pages related to the pre-https blocked pages in Iran and China and political opposition pages in Russia). We expand these sets of pages using Wikipedia2vec (48).

Incidence rate ratios for the follower analyses and the Wikipedia page view analyses are from negative binomial regressions. In the follower analysis, this models the number of new followers per day, with a separate model for each account category. Independent variables are in lockdown period and in mainland China, and the effect of interest is the interaction between these indicator variables (i.e., a difference in difference), with December 2019 as control period and Hong Kong as control group. The Wikipedia page view analyses use the same specification, reporting the coefficient for in lockdown period and in page set (current leader, historical leader, previously blocked) relative to December 2019 and relative to page views for the rest of Wikipedia. Observations are the total views per category by day. Figures displaying (log-scale) ratios of followers/Wikipedia page views approximate coefficients from these negative binomial regressions. Negative binomial regressions were estimated using the MASS library in R.

Increases in geolocated Twitter activity (unique users) by day and by province were modeled using a five-term polynomial regression (by day) for time trends after the Hubei lockdown and a mean without any time trend prior to lockdown (see SI Appendix, Fig. A1 for a province-by-province visualization of this model). The points in Fig. 2 are predicted values by province for the first day of lockdown and day 30 of lockdown.

We thank Thomas Qitong Cao, Lei Guang, Ruixue Jia, Susan Shirk, and Yiqing Xu in addition to participants at workshops at New York University, the University of Chicago, University of Southern California, and University of California, San Diego for helpful feedback. This work was partially supported by the National Science Foundation Grant 1738411.

Author contributions: K.-C.C., W.R.H., M.E.R., and Z.C.S.-T. designed research, performed research, analyzed data, and wrote the paper.

The authors declare no competing interest.

This article is a PNAS Direct Submission.

This article contains supporting information online at https://www.pnas.org/lookup/suppl/doi:10.1073/pnas.2102818119/-/DCSupplemental.

*Stockmann (24) provides evidence that consumers of newspapers in China are unlikely to go out of their way to seek out alternative information sources. Chen and Yang (33) provided censorship circumvention software to college students in China, but found that students chose not to evade the Firewall unless they were incentivized monetarily. Roberts (26) provides survey evidence that very few people choose to circumvent the Great Firewall because they are unaware that the Firewall exists or find evading it difficult and bothersome.

Source: New York Times, 15 December 2020. https://www.nytimes.com/interactive/2020/world/coronavirus-maps.html.

To protect the application and its users, we are not disclosing its name or the exact ranking.

Note that increase in popularity is not comparable across applications because popularity is measured in terms of ranks. More highly ranked applications (like Facebook and Twitter) may need many more downloads to achieve a more popular ranking.

SI Appendix, section 2 provides more detail, and SI Appendix, Fig. A1 shows trends per province.

#Wikipedia page view data are publicly available: https://dumps.wikimedia.org/other/pagecounts-ez/merged/. Note that these data do not track where users are from geographically; we use language as an imperfect proxy for geography.

While almost all provinces experience a sustained increase in Twitter volume, Beijing and Shanghai have an overall decrease in Twitter volume after the outbreak. We suspect many Twitter users in Beijing and Shanghai left those cities during the outbreak, which is corroborated by the Baidu mobility data we detail in SI Appendix, section 3.

**We note that follower behavior is a useful window into user behavior and has advantages over other metrics in this context like the content of the new users tweets. First, merely following accounts is likely a less risky behavior than publicly posting content about politics, especially that related to China. That is, we expect users to self-censor their posts but not (to the same extent) whom they only follow. Second, tweet activity is right skewed in our data, which is common in social media data. The median account in the stream tweets twice, and the top 1% of active users author 40.3% of tweets. Analyzing tweets would therefore create a less complete analysis of user behavior than analyzing following relationships.

In June 2020 and September 2019, Twitter released datasets containing 28,991 accounts it identified as being part of pro-China information operation campaigns (https://transparency.twitter.com/en/reports/information-operations.html). Twitter granted us access to the unhashed version of the data they do not publicly release, meaning we had the information operation campaigns accounts actual screen names and user identification numbers.

Using data from https://www.greatfire.org/.

These lists are based on offices in the CIA World Facebook. We use this list for ease of comparisons with other countries and remove the Ambassador to the United States from each list. Chinas list is available here (and there are links to leaders of other countries on the same page): https://web.archive.org/web/20201016160945/ https://www.cia.gov/library/publications/resources/world-leaders-1/CH.html, excluding Hong Kong and Macau.

The June 2020 increase in China is due to the anniversary of Tiananmen Square protests. Our claim is not that only the COVID-19 crisis causes increases in views of sensitive content. That the same behavior is observed around another crisis event supports this papers argument.

##Like China, citizens in each of these countries speak languages relatively specific to their country, and therefore we expect most of the page views of Italian, German, Persian, and Russian Wikipedia to originate in Italy, Germany, Iran, and Russia, respectively.

More here:
COVID-19 increased censorship circumvention and access to sensitive topics in China - pnas.org

In early December 2021, Palestinians and Arabs representing a diversity of creative, activist and academic practice approached the board of Sydney festival after it was revealed the board had accepted $20,000 funding from the Israeli embassy for the presentation of Sydney Dance Companys realisation of Decadance, a work created by Israeli choreographer Ohad Naharin of the Batsheva Dance Company of Tel Aviv. The amount gave the embassy star partnership status with Sydney Festival.

We made three requests: divest from the star partnership, end all relations with the State of Israel, and remove any Israeli government emblem from Sydney festivals promotional material.

In arguing our case for divestment, we said Arab and Palestinian communities would not participate in a festival that does business with a state that stands credibly accused of war crimes and crimes against humanity, according to crimes defined in the Rome Statute of the International Criminal Court. In 2021, Human Rights Watch found Israel is committing crimes against humanity of apartheid and persecution.

We made it clear artists and arts organisations fundamental partners in any arts festival felt betrayed by Sydney festival. Finally, we pointed out this partnership denied artists an environment of cultural safety, leaving artists, creatives and companies with no choice but to withdraw.

Our arguments were rejected by the board on the grounds Sydney festival is a non-political organisation. In response, Palestinians and a cross-section of artists, arts organisations and communities publicly called for a boycott of the Sydney festival, inspired and guided by the global Palestinian Boycott, Divestment and Sanctions (BDS) movement, founded and led by Palestinian civil society.

The effusive response to the boycott call has been unprecedented, in fact historic. It is being cited as the most effective, creative and impactful campaign targeting complicit Israeli sponsorship of an international arts event in Australia, and indeed one of the most successful in the world.

The backlash to this artist-led cultural boycott has been predictable, indeed recycling arguments used in the 1980s against the boycott of apartheid South Africa.

One criticism in particular exposes how liberalisms conceits of free speech, marketplace of ideas, open debate and dialogue is weaponised against Palestinians to shut down their right to resist and to deny them permission to narrate as renowned Palestinian-American professor Edward Said famously argued in 1984.

According to New South Wales arts minister Ben Franklin, it is the boycott not the actions of Sydney festival which shut down specific creative voices simply on account of their nationality, acting as a kind of censorship.

In an opinion piece published in the Australian, federal arts minister Paul Fletcher described those involved in the boycott as Stalinist censors and Hamas useful idiots. Such contrived hysteria over the boycott stultif[ying] and suppress[ing] artistic and creative excellence, and laughable comparisons with Stalinist Russia, are amusingly desperate claims and demonstrate just how rattled Israels defenders are in the face of incontrovertible daily evidence of that states brutality.

The arguments are embarrassing and spurious. Organisers have repeatedly stated the cultural boycott aims at institutions not individuals, targeting complicity, not identity. There was never any attempt to shut down the actual production of Decadance. The target of the boycott call was Sydney festival as a cultural institution for its refusal to divest from its sponsorship and therefore its complicity with the State of Israel.

That Palestinians and their supporters are being forced to explain and restate the basis and terms of the boycott call, only to be ignored and misrepresented is a form of censorship itself. Whose voices are privileged: those who defend oppression or those resisting it?

Those arguing against the boycott claim boycotts burn rather than build bridges. At the first meeting with the board, artists made the crucial point bridges must be built on ethical and just foundations. A star partnership with the State of Israeli is one way to destroy these foundations and for this reason artists cannot, in good conscience, cross that bridge.

The boards refusal to listen to artists is a form of silencing.

The weaponising of censorship against the boycott is hollow because the ministers conveniently ignore questions of power and privilege. The power dynamics between artists and the board of Sydney festival, between marginalised communities and the monocultural establishment, between individuals and institutions are key critical points of reflection here.

What makes these censorship allegations even more disingenuous is the fact that in the same breath as Palestinians and their allies are accused of being censorious, opposition arts spokesperson, Labors Walt Secord called for legislation to cut off funding to arts organisations that participate in a boycott of Israel. Freedom of expression it seems is only afforded to those in power and with power.

Those who attack cultural boycotts in the name of free speech are invariably missing in action when Palestinians are routinely censored, bullied and cancelled for daring to speak their truth. Certainly they remain silent and indifferent to the violent suppression of Palestinian arts and culture, on the raids, lawfare and intimidation of Palestinian artists and artistic and cultural institutions.

This is precisely why the boycott of Sydney festival has been called and indeed, why it has been so impactful and effective.

Go here to read the rest:
Calling the Sydney festival boycott censorship is a disingenuous attempt by those in power to silence Palestinians - The Guardian

Internet access brings a lot of advantages such as enabling people to gain access to information with just a few taps of their finger. Many people are starting to call access to the internet a basic human right because of the fact that this is the sort of thing that could potentially end up allowing people to earn money as well as educate themselves in a manner that just hadnt been all that possible previously in human history due to technological barriers and limitations.

With all of that having been said and now out of the way, it is important to note that a lot of world governments dont really seem to care about that and are trying to block social media access. This can be a really big problem for the world, and a really unfortunate trend that has been noticed is that the level of social media censorship that the world is seeing is on the rise and there is a strong likelihood that it would get a lot worse before it gets any better.

This information comes from SurfShark which has been taking note of social media censorship over the last seven years. This research involved an analysis on the state of internet access and social media in all of the 193 countries that are recognized by the UN, and one thing to note is that this often involves preventing certain information from being spread on social media apart from restricting user access to said social media sites in the first place so this is quite a diverse issue.

The worst offenders when it comes to social media censorship are generally countries that are in Asia and Africa. These countries are somewhat more likely to have authoritarian rulers and dictatorial governmental regimes. Such forms of government give the state leeway to do whatever it would like to do in order to provide or take away internet access as they see fit. Hence, since there are no legal blocks that can prevent governments from censoring social media and blocking access to sites, they often move forward with it without any obstacles.

Another really concerning thing that this data reveals is that there is at least some kind of social media blocking that is occurring in around a third of the countries that exist at present. These blockages often center around things like elections and any type of political upheaval, and there are 71 countries that are either currently blocking some form of social media access or alternatively have done so at some point in the past. Most of these countries are in Asia and Africa, with South America also having a large number of them.

If we were to take a closer look at how these things work, it can be discovered that the vast majority of African countries do not allow unrestricted social media access to their citizens. Sometimes this can be relatively innocuous such as in the case such as Algeria blocking social media access during exam season so that students can focus on their studies. In other cases it can be more serious such as Nigeria completing banning Twitter when the new government came into power.

Asia is also a pretty bad offender in this regard. Perhaps the worst country in the world for social media use is actually China due to the reason that this country has blocked access to virtually all foreign social media platforms. Another example of terrible internet rights violations occurred in India, where the government completely blocked all forms of internet access in the disputed territory of Jammu and Kashmir after passing a controversial new law that suspended this territorys disputed status which caused widespread protests among the residents of that locale.

The fact that so many governments are adamant about banning social media is a strong indication of how powerful a tool it can be. It allows for the rapid dissemination of all kinds of information, and most governments that rely on the iron fist to maintain power would obviously not be all that happy about that and would want to restrict it whenever they can.

Go here to see the original:
Social Media Censorship is Getting Worse According to This Study - Digital Information World