Reduce Information Manipulation, Improve Public Discourse, Strengthen Democratic Resilience

2023 Taiwan Information Environment Policy and Action Recommendations

  • Author = IORG
  • Published = 2023.2.23 12:00

A. Introduction

A.1 Goal

  1. Understand the impact of information manipulation on Taiwan’s elections, referendums, public discourse, and overall information environment since the 2018 elections and referendums.
  2. Provide recommendations to various sectors of Taiwanese society on reducing information manipulation, improving public discourse, and strengthening Taiwan’s democratic resilience.

A.2 Methodology

IORG has partnered with the National Democratic Institute, planned and implemented a year-long research program to compile research and observations on Taiwan’s elections and referendums between 2018 and 2021 via literature review, academic consultations, and focus group discussions.

1. Literature Review

This research program has reviewed both domestic and foreign academic literature across multiple disciplines, as well as news articles and research reports. This program has then identified factors impacting Taiwan’s election and referendum results and explores the interactive relations between information manipulation and the various factors.

2. Academic Consultations

This research program has invited 6 academic researchers for consultations. Feedback on research progress has been collected via offline and online communications, and incorporated into research design and report content. The academic researchers are from a variety of disciplines including information engineering, sociology, political science, and journalism.

3. Focus Group Discussions

This research program has carried out 10 focus group discussions and 2 individual interviews in 5 cities – Taipei, Taichung, Tainan, Kaohsiung, and Hualien – across Taiwan. 38 individuals from a variety of backgrounds relating tightly to public discourse have been interviewed, including legislators, legislative staff, party administrative staff from multiple political parties, news media workers, policy issue activists, civil society organization members, open-source contributors, and college student dissenters. Participants of the discussions and interviews have been selected via purposive sampling, the discussions were semi-structured and guided by objective-oriented open-ended interview questions. Interview questions were also adjusted by the interviewers based on responses from the participants.

B. Analysis

B.1 Information Manipulation – Cases and Impact

Literary review shows 5 factors affecting political decision-making of the people of Taiwan – government performance, party identity, candidate image, policy issue, and the information environment, and that these 5 factors proceed to impact election and referendum results in Taiwan. Furthermore, this research program has identified multiple cases of information manipulation through the aforementioned methodology between year 2018 and 2021, and determined the content of these cases to be sufficient in affecting public understanding of public affairs, including government performance, party identity, candidate image, policy issue, and the information environment.

Altogether, information manipulation indirectly affects Taiwan’s election and referendum results through 5 factors.

Between 2018 and 2021, information manipulation persisted, affected public understanding of government performance, party identity, candidate image, policy issue, and the information environment, and proceeded to affect election and referendum results in Taiwan. Table below is a partial list of 58 cases of information manipulation affecting the 5 factors between 2018 and 2021, including year, background, content, evaluation, and actors of the cases.

1. Government Performance

Year

Background

Case

Evaluation

Actors

References

2018

Local elections

Overstocked pineapple dumped into water pond

Distorted content

2018

Local elections

Fake farmers and fishermen sharing personal experience

Masqueraded source

  • TW news media

2018

Local elections

Cross-strait tension causes drastic price drop for fruits

Bad causality; Generalization

  • CCP

2018

Local elections

Government bullying school children and forcing masturbation practice

False

2019

Local elections

2 million tons of pomelo dumped into reservoir

False

  • CTI news anchor Wang Yucheng (王又正)

2020

COVID-19 pandemic early stage

TW government bans flights from Wuhan but not the US

False equivalence

  • Chao Shaw-kong (趙少康)

2020

Solomon Islands and Kiribati cut diplomatic ties with Taiwan

Government is incompetent in foreign affairs

Bad causality

  • TW news media

2020

US-TW arms sales

TW buys overstocked, scrap metal, overpriced weapon from the US

Source not credible; Generalization

  • CCP

2021

COVID-19 pandemic

AZ vaccine is problematic

Pending scientific verification

  • CCP
  • Anonymous Facebook Pages

2021

US-TW arms sales

Taiwan wants vaccine, not weapon

Generalization

  • CCP

2021

Taiwan applies to join CPTPP later than China

TW government is lazy, derelict in its duty, and a false player

No evidence

  • Anonymous Facebook Pages

2021

Referendums

TW gov promoting referendums violates administrative neutrality

False equivalence

  • Ko Wen-je (柯文哲)
  • Chao Shaw-kong (趙少康)
  • TW news media
  • CCP

2021

After referendums

Next year’s electricity shortage is a certainty

No evidence

  • Terry Guo (郭台銘)

2021

Clustered COVID-19 infections at the Taipei Agricultural

Products Marketing Corporation

Do not visit PX Mart storefronts in northern TW for they are all infectious

False

  • Netizens

2. Party Identity

Year

Background

Case

Evaluation

Actors

References

2018

Local elections

Taiwan is under an iron curtain and the media is covering up the truth

Distorted content

2018

Local elections

DPP is engaging in homosexuality

Bad causality

  • Chou Hsi-Wei (周錫偉)

2018

Non-specific

Green communist

False equivalence

  • Chiang Wan-An (蔣萬安)

2020

Before national elections

Chen Ju (陳菊) is corrupt

No evidence

  • CN media

2020

National elections

DPP is confiscating referendum results

False; Bad causality

  • United Daily
  • KMT
  • Anti-LGBTQ groups

2020

National elections

DPP is rigging the votes

No evidence

2020

Cigarette smuggling through TW presidential aircraft

Presidential Office condones corruption

No evidence

2020

Cigarette smuggling through TW presidential aircraft

Presidential Office: “Over-purchase”; news media: “Setting the tone”

Generalization

2021

HK Apple Daily forced to shut down

DPP has double standards by shutting down CTI and supporting HK Apple Daily

False; False equivalence

  • TW content farms
  • CCP

2021

Non-specific

Ta-green-ban

False equivalence

  • Huang Wei-Han (黃暐瀚)
  • Kao Hung-an (高虹安)
  • Lucifer Chu (朱學恆)

2021

Non-specific

Green-pig

Dehumanization

  • Huang Wei-Han (黃暐瀚)
  • Kao Hung-an (高虹安)
  • Lucifer Chu (朱學恆)

2021

TW President Tsai sent condolences over death of US President Biden’s pet dog

Taiwanese people are less worthy than an American dog

False equivalence

  • TW news media
  • CCP

2021

Referendums

Referendums are rigged

No evidence

3. Candidate Image

Year

Background

Case

Evaluation

Actors

References

2018

Local elections

Large number of lifestyle Facebook Groups and Pages shifted to support Han Kuo-Yu

Coordinated behavior

  • Anonymous Facebook Groups
  • Anonymous Facebook Pages
  • Interviews

2018

Local elections

Same-sex marriage will spread AIDS and the company founded by Tsai Ing-Wen will reap benefit

Conspiracy

2018

Local elections

Voting candidate A is supporting politician B

Bad causality; oversimplification

  • KMT
  • DPP

2020

National elections

You will not know if your father is a man or a woman if you elect the wrong person

No causality

  • Hung Hsiu-Chu (洪秀柱)

2020

National elections

Man marrying man, woman marrying woman, that is the law Chiu Yi-Ying (邱議瑩) forcibly pushed to pass

Generalization

  • Wang Ling-Chiao (王齡嬌)

2020

National elections

Tsai Ing-Wen is corrupt

No evidence

  • Han Kuo-Yu (韓國瑜)

2020

National elections

Tsai Ing-Wen’s doctoral thesis and degree are fake

False

  • Dennis Peng (彭文正)
  • Chinese Unification Promotion Party (中華統一促進黨)
  • CCP

2020

National elections

Han Kuo-Yu’s tree

climbing and treehole filling is not helpful to the prevention of dengue fever; Han Kuo-Yu’s brain has holes

Generalization; Lack evidence

  • Hsu Yung-Ming (徐永明)
  • DPP city councilors
  • TW news media

2020

National elections

Love River has changed its color, Han Kuo-Yu just wants to run for president and does not care about Kaohsiung

Bad causality

  • Taiwan Statebuilding Party (台灣基進)

2020

National elections

Han Kuo-Yu has an illegitimate child with Miss Wang from Xinzhuang

Unverifiable content

  • TW news media

2021

Recall election

Representative Huang Jie’s (黃捷) mother sells tea in China

False

  • Chinese Unification Promotion Party (統促黨)
  • CCP

4. Policy Issue

Year

Background

Case

Evaluation

Actors

References

2018

Referendums

Same-sex marriage will eradicate humankind

No evidence

  • Anti-LGBTQ group

2018

Referendums

Same-sex marriage spreads AIDS

No evidence

  • Anti-LGBTQ group

2018

Referendums

Same-sex marriage violates parental rights

No causality

  • Anti-LGBTQ group

2018

Referendums

Referendum cheat sheet folded with voting notice

Oversimplification

  • Anti-LGBTQ group
  • Miaoli County officers

2018

Referendums

Same-sex marriage will eliminate father, mother, grandfather, grandmother; children can only say “parents”

No Causality

  • Anti-LGBTQ group

2020

Same-sex marriage legalized

No sons or grandsons; unable to care for grandchildren; family line will end

No Causality

  • Anti-LGBTQ group
  • KMT legislative candidates

2020

National elections

Tsai government provides financial support to HK rioters

No evidence

  • Foreign accounts

2021

Referendums

Ractopamin pork import sacrifices people’s health

No evidence

  • KMT

2021

Referendums

Anti-ractopamine pork is equal to anti-American pork

False equivalence

  • DPP

5. Information Environment

Year

Background

Case

Evaluation

Actors

References

2018

Referendums

Repeated identical comments at anti-nuclear power Facebook Pages or individuals

Spamming

  • Nuclear power supporters
  • Interviews

2018

Referendums

Allocation of large resources in disseminating information manipulation via news media, social platforms, street campaigns, and interpersonal relations

Coordinated behavior

  • Anti-LGBTQ group
  • Interviews

2018

Referendums

Financial support to Anti-LGBTQ groups in purchasing large amount of TV ads

Foreign interference

  • Conservative Church groups in the US
  • Interviews

2018

Local elections

Han Kuo-Yu receiving high level and repeated exposure

Spamming

  • TW news media
  • Interviews

2018

Local elections

Increased online volume for Han Kuo-Yu

Foreign interference; Hired accounts

  • Foreign accounts
  • Interviews

2018

Non-specific

Cross-platform amplification of information manipulation

Cross-platform media convergence

  • Social platforms
  • Prop news sites
  • TW media
  • CN media
  • Lin Chao-Chen 2022

2020

National elections

Hired personnel organizing online communities, hired accounts spreading specific content or other interactions, including information manipulation

Coordinated behavior; Hired accounts

  • TW political parties
  • Political marketing companies
  • Interviews

2020

National elections

Increased polarization

Targeted recommendation

  • Social platforms
  • Interviews

2021

Referendums

Mobilization of voters with large amount of simplified slogans and infographics

Oversimplification

  • Main political parties in TW

2021

Referendums

Criticizing DPP’s cyber army and abuse of power; assisting KMT in referendum promotion

Foreign interference

  • CCP

2021

Referendums

Multiple anonymous Facebook Pages sharing content farm articles within short period of time

Clustered posting

  • Anonymous Facebook Pages

B.2 Challenges

1. Taiwan’s information environment is affected by few privately-owned web platforms and is crowded with fragmented information. The public is affected by one’s self and partisan identity, depends on opinion leaders, lacks methods to verify the quality and evaluate the credibility of information. Development of good public discourse is therefore limited.

Taiwan’s information environment is free and diverse. The popularization of radio, television, and internet communication has enabled free sharing of information and new connections. Anyone can have a voice and information is more accessible than ever, including information that conforms to one’s identity but not necessarily trustworthy.

Nowadays, access to information and other various forms of interaction are highly concentrated on a handful of privately-owned, profit-driven web platforms. These platforms control visibility of information through targeted recommendation algorithms powered by personal information of their users and cases of abuse have occurred. On these platforms, anonymity is allowed, bringing both freedom and problems. Interviews show that mainstream web platforms are not designed and not suited for public discourse. Existing platform mechanisms could also obstruct good public discourse.

While the internet information environment thrives, information fragmentation follows, and the environment is crowded with simplified information. At the same time, the public is affected by one’s self and partisan identity, depends on opinion leaders, internet influencers, and other authorities with varying degrees of qualifications, lacks publicly credible information quality verification mechanisms, and feasible information credibility evaluation methods. These circumstances limit the development of good public discourse.

2. Public discourse is becoming more difficult and participation in public discourse is diminishing across sectors of Taiwan’s society, in particular public online discussions where aggression and emotion is common while interaction remains limited to affirming identity and position rather than presenting facts and debating opinions.

Public discourse is becoming more difficult in recent years, in particular discussions on public web platforms, interviews show.

Aggressive, emotional content on various web platforms is a common experience in our daily lives. In Taiwan, public discourse is easily reduced into binary opposition when it is related to political parties or elections, partisan or self identity often take precedence over, even replacing substantive discussions. Web platforms, in encouraging the gathering of users with similar preferences and positions, could have amplified such an effect. Interviews show that interactions on various web platforms often pertain to verifying personal identity, fortifying position, and responding to emotion, rather than presenting facts and debating opinions. Under circumstances of clear confrontation, politicians or policy groups tend to be unable to avoid declaring their policy position. When such a position differs from the mainstream, it is becoming more common that they suffer criticism from supporters who are more extreme, in turn making a group with a similar position more homogeneous and leaning towards extreme.

Participation in public discourse is diminishing across sectors of Taiwan’s society and public attitude towards participation is turning conservative, apathetic, and tiresome. Some interviews show a lower passion towards political participation while others show the public losing patience towards longform information. LGBTQ and ally communities show fatigue towards public participation after frequent mobilization in support of same-sex marriage legalization. In recent years, students show reluctance towards making judgment and declaring positions on policy issues due to lack of political efficacy from the lowering levels of confrontation in social movement, abundance in alternative information from popular audio-video entertainment social platforms, and school curriculums encouraging diverse information consumption and discouraging political discussions on campus. Rising levels of difficulty are causing civil society groups to avoid online engagement and instead host offline closed-door stakeholder meetings for policy advocacy. Furthermore, debates in city councils are decreasing, said city councilors during interviews, while local politicians tend to follow party lines and avoid declaring their own positions on policy issues when facing local constituencies. These circumstances reduce participation and inclusivity in public discourse, in turn diminish the legitimacy of democratic processes.

3. Information manipulation persists, maintains high volume, and degrades the foundation of trust for public discourse in Taiwan. Sources are both domestic and foreign while forms encompass more than falsehood and include manipulative behaviors and environmental mechanisms.

Information manipulation has persisted and maintained a high volume in Taiwan’s information environment, interviews show. Negative state of public discourse during election campaigns is now the new norm, said civil society organizers during interviews. Foreign and domestic, various forms of information manipulation is crowding Taiwan’s information environment.

After the 2018 elections, many political parties and politicians undertook significant effort in organizing online communities, inspired by specific candidates during the 2018 elections, said political workers during interviews. Strategies were formulated to target various social groups in order to increase and improve overall volume and image of the person or party. During the 2020 elections, multiple government agencies used large amounts of simplified infographics and memes to promote government performance, which could have contained information manipulation, interviews show.

Taiwan’s news media further overflows the information environment with sensational, low-quality content due to its profit-driven nature, lack of fact-checking mechanism and culture, and existing negligence to truthfulness, rigor, and independence of its reporting. Excessive and repetitive reporting of specific candidates during the 2018 elections indirectly impacted election results, interviews show. Traditional media outlets, in transitioning to web platforms, maintain influence over its audience while content becomes more sensational and more frequently contains information manipulation. Local media organizations, in various symbiotic relations with local governments, representatives, and private corporations, further compromise the independence and trustworthiness of the news.

Online anonymous accounts spread aggressive content regardless of political stance. Anonymous extreme content attracts attention, amplifies confrontation, and impairs public discourse while being unaccountable. Some moderators of anonymous Facebook Pages publish content knowing that it could contain information manipulation. Additionally, low-quality articles from unverifiable sources are published by content farms and then frequently shared by clusters of anonymous Facebook Pages. Fixed combinations of content farms and Facebook Pages acting recurrently in coordinated fashion can artificially increase volume of specific topics on web platforms.

Political parties, politicians, news media, and other content producers often adjust their production and marketing strategies conforming to platform mechanisms and audience preferences, prioritizing clicks and other indicators for user interaction. This behavior could easily create information manipulation.

Foreign information manipulation continues to propagate in Taiwan’s information environment. PRC state agencies and CCP media outlets spread information manipulation during elections and other important events in Taiwan, reinforcing negative perceptions of Taiwan’s democratic governance and international democratic alliance. Specific personalities and media outlets in Taiwan amplify CCP content while CCP also amplify content from specific Taiwanese media, celebrities, “island netizens (岛内网民),” anonymous Facebook Pages, and hijacked accounts, in order to expand its sphere of influence. Additionally, interviews show conservative church groups in the US provided financial support to for anti-LGBTQ groups in Taiwan to purchase TV ads during the 2018 referendums.

Information manipulation evolves continuously and comes in various forms and degrees of severity. Not only falsehood, problematic source, content, inference, as well as manipulative behaviors and environmental mechanisms, are degrading the foundation of trust essential for reasonable public discourse.

Category

Source

Content

Inference

Behavior

Environment

Manipulative

  • Masqueraded source
  • Fake fact-checking
  • False content
  • Distorted content
  • Grafting
  • Bad subtitle
  • Deepfake
  • No evidence
  • Lack evidence
  • No causality
  • Generalization
  • Oversimplification
  • False equivalence
  • Logical error
  • Conspiracy
  • Coordinated sharing
  • Coordinated behavior
  • Hired accounts
  • Manipulated accounts
  • Hijacked accounts
  • Bots
  • Phishing
  • Personal data abuse

Suspicious

  • No source
  • Source unverifiable
  • Anonymous source
  • Bad source
  • Source not qualified
  • Source not credible
  • Island netizen
  • CCP propaganda
  • Content unverifiable
  • Out of context
  • Pending scientific verification
  • Pending judicial investigation
  • Bad causality
  • Bad authority
  • Appeal to fear
  • Appeal to hate
  • Appeal to distrust
  • Appeal to emotion
  • Warmongering
  • External governance
  • Authoritarian narrative
  • Recurring rumor
  • Spamming
  • Clustered posting
  • Authoritarian interference
  • Authoritarian collaboration
  • Authoritarian initiation
  • CCP interference
  • CCP collaboration
  • CCP initiation
  • CN-RU collaboration
  • Cross-platform media convergence

Alerting

  • Government propaganda
  • Dehumanization
  • Nicknaming
  • Gender stereotype
  • Claiming long-term effect
  • Claiming straightforwardness
  • Claiming truthfulness
  • Claiming media cover-up
  • Claiming new regulation
  • Call to share
  • Link provided
  • Appeal to authority
  • Universal statement
  • US skepticism
  • JP skepticism
  • TW skepticism
  • Foreign interference
  • Targeted recommendation
  • Targeted marketing

4. Information manipulation affects understanding of public affairs, obstructs reasonable public discourse, intensifies confrontation and polarization, and indirectly affects decision-making of politicians and political parties, elections and referendum results.

Information manipulation continues to spread in Taiwan’s information environment, affecting its audience in terms of understanding of public affairs. Interviews show negative social impact due to the continued existence of information manipulation, including the impediment of the spread of trustworthy information, obstruction of reasonable public discourse, and intensification of confrontation and polarization.

Additionally, domestic and foreign information manipulation affects decision-making of politicians and political parties in creating dilemmas in public communications, bias in formulating policies based on narrow or extreme opinions, and ineffectiveness in reflecting the diversity of one’s supporters.

5. Sectors of Taiwan’s society encounter various difficulties in responses and are unable to effectively reduce information manipulation and improve public discourse.

Polls show that a majority of people in Taiwan expect proactive government actions in reducing information manipulation. However, interviews show such actions have achieved limited results. Some interview participants raised concerns on government regulation on information manipulation as infringement of freedom of expression. Others warned of the possibility of government action causing backlash, creating an impression of government abuse of power, that the punishment is disproportionate or driven by corrupt pursuit of personal or partisan gain. Additionally, political workers have said during interviews that there is a lack of wide acceptance amongst lawmakers towards legislative responses to information manipulation, resulting in a lack of progress at the legislature.

Fact-checking is an integral part of the infrastructure in reducing information manipulation. However, fact-checkers face various difficulties. While the concept of fact-checking is nowadays widely known, the credibility of fact-checking reports and organizations is not yet fully established. The contemporary multiplicity in informational channels and formats increases the cost and creates predicaments for fact-checking, said fact-checkers during interviews. Additionally, interviews also show that fact-checking cannot resolve every kind of information manipulation.

Information literacy is a key capacity in evaluating credibility of various information and participating in public discourse. However, information literacy education lacks support in schools. Before the current “108” Curriculum Guidelines, related subjects were almost absent in the 12 years of national basic education between age 6 and 18. Out of the 3-year high school education, 1 class – approximately 1 hour – was allotted to media and public opinion formation, said educators during interviews. It has been 3 years up to August 2022 since the current Curriculum Guidelines came into effect in September 2019. Under the current Guidelines, media and information literacy is amongst the 9 core competencies, and must be incorporated into classes in each discipline. However, difficulties in advancing information literacy education persists due to lack of performance verification from authorities, lack of motivation for teachers and schools, lack of training, professional knowledge, and connections with experienced civil society groups for motivated teachers, and common student mentality that prioritize exams, decouples school education and everyday lives, and the habit of no politics on campus.

B.3 Opportunities

1. Public and government awareness of information manipulation is largely in place

A majority of the people and various government agencies in Taiwan are already aware of and alerted by information manipulation, interviews show.

For the people, internet rumors have existed for a long time. COVID-19-related information manipulation circulating since the start of the pandemic from Wuhan concerns personal well-being and further increased public awareness on the subject. Some interview participants said that falsehood is now part of popular culture while others mentioned the diminished effectiveness of politicians persuading voters using simplified slogans due to increased awareness.

For the government, despite actions resulting in varying effects, the central government has set up principles and guidelines and proposed new laws and amendments in response to information manipulation. Meanwhile, both central and local authorities have put in place services providing timely clarification to the public. These actions can be interpreted as government awareness and institutionalized response to information manipulation.

2. Some politicians and political staffers are aware of the negative effect of their behavior obstructing public discourse and have yet to reform; political figures advanced public discourse through open government mechanisms that were discontinued and not yet replicated elsewhere

Politicians have said during interviews that it is unavoidable to cater to user preferences when engaging with communities online. In the pursuit of exposure, adjustments to content production and management strategies can be made to the effect which they know could bring about information manipulation. Political staffers have also shared their experiences participating in online public discourse, in admission that political stance and personal identity of other users would affect whether they consider these opinions.

Interview participants have shared and recognized their positive experiences participating in a shadow city council. The council, being an open government mechanism created and fully-participated by the mayor, created incentives for participants to express their opinions on public policies, in turn facilitated public discourse in their own communities. However, interview participants have also said that the council was discontinued by the next administration, and that they have not observed any other similar example henceforth.

3. Fact-checking could help to reduce information manipulation in some situations while not in others; Ukrainian journalists and Hongkong pro-democracy activists continue to disseminate fact-based information and maintain trust under extreme circumstances

Fact-checking could help to reduce information manipulation on public platforms. Past cases show reduced amounts of related information manipulation after the publication of fact-checking results on public platforms. Alternatively, false information reoccurs on closed platforms regardless of fact-checking. Additionally, research shows that fact-checking can calibrate misunderstandings on public platforms, at the same time reinforce existing beliefs on closed platforms and be counterproductive to building information literacy.

In Taiwan, third-party fact-checking has been in operation in recent years and is now widely known. Interview participants have said that fact-checking reports started to appear in chat groups amongst family and friends and that fact-checking related to everyday life is the most accepted. Multiple public social platforms have preliminarily integrated fact-checking while third-party tools are still required on closed platforms, generating friction for the platform users.

Outside of Taiwan, fact-checking is being practiced by citizens and journalists confronted with greater challenges. In the face of state violence, Hongkong pro-democracy activists insist on spreading truthful information and avoiding unverified information that intensifies emotion. During the Russian invasion, Ukrainian news media continues to carefully fact-check government information, for falsehoods threaten safety and the lives of their fellow citizens.

4. Physical discussions are conducive to good public discourse and relatively feasible, working cases exist while lacking space, resources, methods, capabilities, incentives, and habit.

Physical discussions, compared to online public discussions, are relatively feasible and have a higher chance of becoming constructive, interviews show. Public discourse in physical space holds advantages over that in cyberspace. Face-to-face conversations have the benefit of physical presence through which the source of information can be identified, said some interview participants, while others pointed out that people tend to be more civil in face-to-face situations, causing less confrontation when in disagreement. Examples of physical discussions conducive to good public discourse, provided by local community organizers, include public policy book clubs hosted for local high school students and community spaces collectively maintained by the neighborhoods.

COVID-19 pandemic has forced citizens to gather online and the effectiveness of discussion suffered greatly, said policy activists during interviews, stressing the necessity of physical discussions, which is a focal point of their current work. Local community organizers also stated that the promotion of local public policy issues must return to the building of local communities and the maintenance of physical spaces.

Nevertheless, public discourse in physical space has its challenges. Physical discussions require space and other resources, as well as methodologies and capabilities for the participants. As some interview participants pointed out, lack of these elements and epidemic prevention measures would hinder the emergence of conversations. Additionally, physical discussions often attract people with similar information sources or political stances while having difficulties engaging with people who hold different opinions.

According to some policy activists during interviews, public discourse in physical space can energize extensive discussions on policy issues among local communities and facilitate systematic changes in local governance. Regardless of the success of one social movement, good public discourse leaves a legacy of experience with the civil society participants as well as positive changes in institutions.

Keys in promoting public discourse in physical space are open spaces, capabilities, incentives, and habit of participation. Moreover, proper utilization of public space could help in engaging people who hold different opinions, said interview participants. Lennon Walls appearing frequently in the democracy movements in Hongkong are good examples.

5. Start from education: Civil society action on reducing information manipulation and improving public discourse with information literacy education has achieved initial results. Expansion of teachers training and institutional support are in need.

Personal capacity of information literacy, which is the ability to discern various information manipulation, is the most essential and fundamental method in reducing information manipulation, interviews show. Furthermore, information literacy can and must be improved through deliberate practices.

Under the current Curriculum Guidelines, teachers are afforded more flexibility to expand public space on campus while students have higher incentives to participate in public discourse. Efforts in recent years by public-minded young teachers lead to increased levels of public participation among high school students, interviews show. Relative to the low level of participation at student dissenter clubs in universities, civil society organizers have pointed out the higher level of public participation among high school students, due to the general education offered at school, rather than the more specialized education for college students. Issues with local characteristics can be particularly effective in connecting with high school students who are more connected locally.

Nevertheless, teachers who implement information literacy classes or public discussions are in need of additional training for professional knowledge and skills that differ from pedagogy. Furthermore, institutional support from schools is much needed in order to help teachers in overcoming the habit of no politics on campus.

6. Trust is key: Maintaining trust relation, increasing emotional communication, appealing to common experience, identifying common interest help to establish connection, and reduce confrontation. Methods and processes of information literacy based on trust facilitate good public discourse.

Trust relations between individuals facilitate both expression of oneself and understanding of others, and are keys for our opinions to be considered and conversations to begin, said interview participants. Furthermore, appealing to common experience and identifying common interest helps to mitigate tension and reduce the opportunity of escalated confrontation between people with different political stances. Identifying common grounds before clarifying differences could increase opportunity for good public discourse.

When advocating for democracy in Hongkong, some civil society groups would prompt audience interest with Hongkong movies, dim sum, and other popular culture, before making connections to the current state of Hongkong’s society and its various social issues, said interview participants. Given initial results of this approach, interview participants have further recommended that activists should connect social issues with the everyday experience of the audience through concrete scenarios, in order to trigger thinking and conversations.

Conversations and consensus based on trust relations are not always based on trustworthy information. Interview participants have also mentioned various difficulties when introducing fact-checking into conversations with family and friends. Therefore, introducing the methods and processes of information literacy rather than the results of fact-checking, could be more effective in facilitating good public discourse.

R. Recommendations

To strengthen Taiwan’s democratic resilience, various sectors of Taiwan’s society must collaborate to reduce information manipulation, improve public discourse, and strengthen Taiwan’s information environment. From analysis and stakeholder feedback, IORG proposes the following action and policy recommendations to 9 sectors of Taiwan’s society: governments, political parties, politicians, news media, social platforms, research institutions, schools and educational authorities, educators, and individual citizens.

R.1 Reduce Information Manipulation

  1. Government should actively clarify policy-related falsehood.
  2. Government regulation of information manipulation should meet the constitutional requirements of due process. Elements of illegal content and behavior should be clearly defined, levels of severity and urgency should be set up, and regulations should be accordingly.
  3. Government should periodically and proactively disclose regulatory actions while providing appeal mechanisms.
  4. Political parties, politicians, news media, and social platforms should reduce the spread of simplified content and stop the amplification of extreme content.
  5. Social media platforms should stop the abuse of personal data.
  6. Social platforms should proactively disclose actions of user account restriction or content moderation as well as provide appeal mechanisms.
  7. Research institutions should continue to disclose scientific evidence of information manipulation.
  8. Schools and educational authorities should support educators in advancing various activities and curriculum in reducing information manipulation.
  9. Educators should proactively seek cooperation with civil society organizations with professional experience in promoting fact-checking and information credibility evaluation education.
  10. Various sectors of the society should work together in building open platforms tracking and observing political speech and news media reporting.
  11. Various sectors of the society should support civil society organizations with professional capacities and practical experiences and work together to build capacity of information literacy within various target groups of the society.
  12. Individual citizens should practice information credibility evaluation: deconstruction, verification, evaluation. Individual citizens should advance from discerning falsehood to evaluating credibility through deliberate exercises to gain capacity of information literacy, in order to reduce impact of information manipulation towards individuals and society.
  13. Individual citizens should recognize the unverifiability and unaccountability of anonymous information and engage with caution.
  14. Individual citizens should lower the dependency on opinion leaders, reduce the spread of anonymous information, simplified information, and stop the amplification of extreme content.

R.2 Improve Public Discourse

  1. Government, political parties, and politicians when publishing clarification or political promotion should prioritize the promotion of public discourse and maintain both clarity and contextual comprehensiveness.
  2. Government should implement open government mechanisms to expand public participation in politics, create demand for public opinion on policy and governance, and catalyze public discourse within various target groups of the society.
  3. Political parties and politicians should proactively seek cooperation with civil society organizations with professional capacities and practical experiences to create opportunities for conversations in physical space.
  4. Politicians should actively create opportunities of encountering diverse opinions when interacting with the public as exercises of improving public discourse.
  5. News media should implement fact-checking and strengthen investigative journalism in order to produce more credible information and promote reasonable public discourse.
  6. Social platforms should enhance two-way communications with users in order to work together on adjusting platform mechanisms to promote credible information and reasonable public discourse.
  7. Research institutions should continue the research on public discourse, as well as the popularization of research, and actively engage in conversations with civil society organizations and the public.
  8. Schools and educational authorities should support educators in creating public spaces and situations to exercise public discourse on campus, changing the habit of no politics on campus.
  9. Educational authorities should expand teacher training programs to include foundational knowledge of information literacy and personal data protection, as well as applied methodology for class plan design and implementation.
  10. Educators should actively create opportunities of encountering diverse opinions and spaces of public discourse for students, at the same time stay mindful of teacher-student power relations.
  11. Various sectors of the society should safeguard physical spaces on campus and in communities, gather resources and people, empower communities of active citizens with open collaboration, and engage in exercises with each other to promote reasonable public discourse.
  12. Various sectors of the society should work together to form consensus on the issue of information manipulation, publicly credible mechanisms of information quality verification, and feasible methods of information credibility evaluation.
  13. Various sectors of the society should work together to deliberate and develop alternative web platforms for public discourse.
  14. Individual citizens should recognize diversity in opinion as the norm in our democratic society and not eliminate those who disagree.
  15. Individual citizens should adopt credible information with verifiable source, truthful content, and proper inference and proactively participate in public discourse.
  16. Individual citizens when encountering others with dissimilar opinions should identify commonalities in life experience, hobby, interest, or identity in order to mitigate tension, decelerate polarization, and improve quality of public discourse.
  17. Individual citizens should improve emotional communications in order to maintain trust with friends and family members.
  18. Individual citizens should require government agencies, scholars, experts, and other institutions and opinion leaders to provide sufficient evidence and proper inference to support their views.

R.3 Strengthen Information Environment

  1. Government should ensure continuity or improvement of current mechanisms as well as implementation of civic participation on the legislative processes of information environment governance.
  2. Government should strengthen and expand government open data and other open government measures.
  3. Government should provide more credible public information on foreign aggression towards Taiwan’s physical, cyber, and cognitive spaces.
  4. Political parties should establish internal mechanisms in response to extreme opinions and supporters.
  5. Political parties and politicians should provide long-term stability with physical space and other resources in support of diverse and inclusive public discourse.
  6. News media and social platforms should establish public-private, multistakeholder, and binding collaborative self-regulatory mechanisms, including public accountability mechanisms for openness and transparency.
  7. News media and social platforms should proactively seek cooperation with research institutions, fact-checking organizations and communities, and improve fact-based shared understanding of platform ecology in order to rebuild public credibility.
  8. Social platforms should reinforce personal data protection.
  9. Research institutions should continue to strengthen scientific research capacities and advance information environment-related research in order to deepen fact-based shared understanding of the contemporary information environment and the interoperability of various actors within the environment.
  10. Research institutions should maintain independence in research activities and diversify research funding and other resources.
  11. Research institutions should proactively work together with various sectors of the society to develop open frameworks and promote multidisciplinary rigorous research.
  12. Research institutions should proactively contribute to scientific literacy education and advancements of credible research-based civic activism.
  13. Schools and educational authorities should create a friendly environment and provide sufficient resources in support of educators creating information environment-related activities and curriculum, as well as the institutionalization of such efforts in favor of sustainability.
  14. Educators should create class plans and materials with scientific research and credible information in order to advance information environment and digital citizenship education.
  15. Educators should promote activities and curriculum on emotional awareness, empathy, deliberation, expression, and dialogue.
  16. Various sectors of the society should credit information literacy in various social interactions such as school admission and hiring, creating a diverse range of incentives for acquisition, evaluation, and dissemination of credible information.
  17. Individual citizens should enhance understanding of the information environment based on scientific research and other credible information, including manipulative behavior and environmental mechanisms.
  18. Individual citizens should be attentive to their own political party being hijacked by extreme opinions.
  19. Individual citizens should pay attention to privacy, reinforce personal data protection, and require social platforms to reinforce protection of user data.
  20. Individual citizens should work together to safeguard the sustainability and openness of community physical space for public discourse.

Conclusion

This research program establishes the negative impact of information manipulation via literature review, academic consultations, and focus group discussions, which includes the following:

  1. Alter public perception, obstruct reasonable public discourse
  2. Exacerbate confrontation, polarization, damage trust in the society
  3. Manipulate decision-making of politicians and political parties
  4. Manipulate result of elections and referendums

This research program then lists 5 factors indirectly influencing Taiwan’s election and referendum results, 5 challenges facing public discourse in Taiwan, and 6 opportunities in strengthening Taiwan’s information environment. Lastly, this research program proposes 52 action and policy recommendations to 9 sectors of Taiwan’s society in order to reduce information manipulation, improve public discourse, and strengthen Taiwan’s information environment.

The issue of information manipulation has received wide attention from various sectors of Taiwan’s society since 2018. Nevertheless, gaps and differences remain. Consensus exists in the existence of information manipulation and lacks in the harm and the appropriate responses towards such phenomenon. Those who believe that information manipulation is severe might believe that government regulation and civil society responses are slow and weak, while those who believe otherwise might believe that government and civil society actions lack both urgency and legitimacy, therefore question their motivation.

The issue of information manipulation in Taiwan while being associated with freedom of expression and international relations is also connected to one’s partisan and personal identity. As a policy issue, the public discourse surrounding the issue of information manipulation faces similar challenges described in previous sections. In terms of reducing information manipulation, related public discourse is easily reduced to simplified pairing of certain actions with certain political stances or partisan preference, obstructing reasonable and substantive discussions. Additionally, terms such as “cyber army” (網軍), “information warfare” (資訊戰), and “cognitive warfare” (認知作戰) are being abused and used to attack those who hold different opinions. Ambiguity in the definition of these terms as well as a common lack of evidence or rigorous research when public figures address these topics are leading to more confrontation, damaging public trust in related fields, and weakening the resilience of our democracy.

The power of a democracy comes from its citizens and at the core of a democracy’s resilience is an informed citizenry. Public participation in politics is a strength and not a weakness of democracy. More participation grounded in credible information formulating public policy via a diverse range of channels can increase political efficacy, create a sense of ownership, and therefore strengthen democratic resilience. When it comes to the issue of information manipulation, everyone is a stakeholder. With greatly increased accessibility to information, information manipulation has a direct reach to all of us, affecting the everyday lives of every one of us living in this age of information. Therefore, the identification of problematic content, behavior, and mechanisms grounded on credible information at the individual level, and the formulation of community consensus and new societal norms through civic participation are essential processes in our self-correction as a democratic society. In these bottom-up processes, institutions must play a role in providing credible information and facilitating participation and consensus-building. Taiwan’s democracy now faces challenges from both outside and from within. We must form a rough consensus on strengthening democratic resilience and continue to connect and collaborate, realizing the strength of Taiwan’s free and vibrant civil society. That is our best way forward for a more perfect union.

Acknowledgment

IORG thanks the following researchers, as well as other researchers who wish to remain anonymous for their valuable contributions to this Policy Brief.

  • Dr. Austin Wang
  • Dr. Tony Ming-Hung Wang
  • Dr. Thung-Hong Lin
  • Dr. Chun-Ming Lai
  • Dr. Jia-Wei Liu