On the Limits of "Fact-Checking"-- Embracing Plurality (#plurality) Might Be the Only Way to Achieve Universality

(Translated from Traditional Chinese by ChatGPT 4. Link to the original post)

In my previous article titled "Is Shinzo Abe a 'Champion' or 'Saboteur' for Peace and Democracy? Heated Debates of His Legacy is Just the Tip of the 'Incommunicability Crisis' Iceberg (Link to the Chinese article)," I briefly touched upon the issue of "social polarization." Though I wanted to discuss the problem of "fake news" as well, I couldn't include it in the same article.

Fortunately, I recently had the chance to delve deeper into this issue while joining a series of in-depth discussions co-hosted by da0 and Web 3 for all. These discussions were prompted by Web 3 leader Vitalik Buterin's comments about a new feature on Twitter called "community notes," which hasn't been made available in Taiwan yet. From my non-technical perspective (please correct me if I'm mistaken), "community notes" utilize open-source algorithms that draw on collective intelligence to offer Twitter users diverse and constructive information. This function encourages conversation between people with differing beliefs by promoting mutually agreeable content and displaying less polarizing information. Its goal is to prevent social media platforms from becoming echo chambers.

Quite unexpectedly, most of us came to the realization that "fact-checking" has severe limitations and we must go beyond fact-checking if we are to alleviate social polarization and the harms that come with it.

Upon the release of Buterin's commentary, three influential figures in the da0/Web 3 for all community immediately relayed and extended his discussion with a new piece each day. Web 3 architect at the Ministry of Digital Affairs, Yen-Lin Huang ("mashbean"), initiated the discussion on August 21, 2023, asking: Amid polarized online discourse, does the truth become more evident the more it's debated?

In the tribal era, we only had to trust fellow tribe members of the same bloodline. But when stepping out of the tribe, whom should we trust? Especially in the digital world, with so many communities, who can we rely on? Or, throughout the collaborative processes of various tribes, how do large-scale collaboration tools like "community notes" build trust?

Digital collaboration tools need to cater to the "cross-border trust" demand since information threats have no boundaries. "Cross-border" encompasses various definitions of "crossing" such as international borders, languages, and cultures. Sometimes, shared linguistic habits act as trust barriers, like scam calls with a Chinese accent making Taiwanese residents cautious. On the other hand, users might be more easily lured by scammers of the same culture due to linguistic similarities. Under such circumstances, collaboration requires standards trusted by people from different tribes, which is legitimacy.

... Trust is fragile and dynamic. When tribes span different platforms and completely distinct user contexts arise, the implementation of "community notes" may face challenges.

...Instead of obsessing over the absolute truth, it's better to harness collective intelligence and seek consensus across differences. Top-down counter-scam strategies are hard to achieve effectively, as democratic countries are unlikely to have a "Ministry of Truth" like in George Orwell's "1984". Horizontal connections between tribes, based on equal relationships, are more feasible. Such digital public goods should be promoted by public resources. Moreover, collaboration tools cannot have a one-size-fits-all approach globally; appropriate tools vary by location and evolve over time.

The next day, Taiwan's prominent translator on democratic governance and public affairs, Weijen Liu, continued Mashbean's inquiries, pointing out that to combat online opposition, one should understand the information environment of the other party.

The parameters recorded by the core formula of "community notes" seem to combine to address two crucial questions when dealing with opposition:

Why does certain information evoke emotions in the friendly partners of the opposing camp?

What kind of world do the friendly partners of the opposing camp truly see?

...The STS (Science and Technology Studies) field has long reminded us that the likelihood of facts being revealed and the solidity of discourse are significantly influenced by social norms and political actions. This is particularly evident in public affairs. Causal inferences presuppose value norms, and people from different camps may have different causal interpretations of the same event. "Fact-checking" can uncover facts but might not necessarily unify various narratives.

...Understanding the information environment of our counterparts can help us empathize with their behaviors that we may find unacceptable. By putting ourselves in their shoes, we can comprehend why our sense of justice may provoke their anger or fear.

By distinguishing the "information environment" from the "actor," we can begin to change the information environment rather than attempting to "change" the other person.

Such an approach not only genuinely uncovers the real social value of "information relevance" but also avoids criticisms that public discussions often face, such as "being susceptible to review bombing" and "requiring social consensus."

Lastly, just before the intense in-person discussion session, co-founder of VolumeDAO, Dr. Bao-cheng Zhang (zysbot.eth)'s Marxist perspective came in — expanding on the "information environment" discussed by Weijen the previous day, highlighting the "objective historicity/conditions" behind all statements.

The most intriguing, and perhaps somewhat provocative part of Buterin's text, might be this statement: "it is better to let ten misinformative tweets go free than it is to have one tweet covered by a note that judges it unfairly."

... Some may find his statement strange, as it seems to support the dissemination of false information and extremist beliefs without any repercussions. However, the main focus is not on identifying inaccuracies, but on understanding the shift in epistemology. If hurtful rumors and slanderous comments were completely baseless and unfounded, originating from arbitrary and groundless imaginations, then they would be swiftly reported and removed. However, it's important to realize that everyone is influenced by their specific circumstances, which can include international relations, political systems, culture, and family dynamics. These historical factors create a platform for expressing views, causing certain statements to seem "reasonable" in some groups while being completely absurd in others. Therefore, if we fact-check and remove false information today, will it eliminate the underlying mechanisms or conditions? Probably not. Additionally, misinformation and extreme views can act as indicators that help identify the mechanisms behind them. If these signals didn't exist, we may be unaware of the concealed conditions that produced them, which can be even more dangerous.

Given this, the "public" in "public epistemology" can be extended to "objective," suggesting that statements have not just a subjective dimension but also encompass the "objective conditions" that enable them. However, doesn't this risk veering to another extreme — determinism, where one doesn't take responsibility for their words and simply blames the background conditions? Community notes and relevant algorithms precisely indicate that, even when expressing views influenced by certain environments, statements from different standpoints (and therefore different conditions) can still break free from deterministic constraints, intersecting at times to form consensus. The reason is simple: humans are still rational.

The impressive knowledge production, discussions, and iterations shown above are just a glimpse of what the da0/web 3 for all communities have to offer. Along with the three leaders on stage, there were also pre-Plurality Taipei meetings (informally known as "Plurality Taipei Summer Camp") and community-written "community notes on community notes" with numerous annotations taking place behind the scenes.

It has taken me some time to gather the energy to share my entire experience, as I am naturally slower. However, I am grateful to be part of a community that embodies the spirit of #Plurality. Despite my delayed pace, I feel welcomed and accepted without pressure or discrimination due to my status. This was evident when Buterin joined our "summer camp" and study session, and everyone treated him as a new friend to chat and sing karaoke with.

Although I cannot comment on the technical details of the platform, I did raise the following theoretical and philosophical points during the discussion.

A famous sociological saying known as the Thomas theorem states:

"If men define situations as real, they are real in their consequences."

Interestingly, based on different philosophical stances in the philosophy of science, the same statement can have various interpretations.

Positivists focus on the causality inherent in the "perception" itself. It's precisely because the "content of cognition" itself is the cause of many consequences, and "erroneous cognition" leads to "unfavorable outcomes" (whether all erroneous cognitions lead to unfavorable outcomes is another matter). Hence, through "fact-checking" and "debunking (misinformation)," they aim to eliminate the "cognitive factors" behind the "adverse effects."

Constructivists assert that "reality" is not an absolute objective truth but rather a construct. As such, no one perspective is more correct than another. Instead of striving for accuracy through debates, constructivists value diverse perceptions and seek consensus through shared vision and collective imagination to create a collective "construction."

A critical realist would probably add a segment to the Thomas theorem:

"If men define situations as real, not only are they real in their consequences, but they are also real in their conditions."

-- Once a situation is perceived as "real," not only are the consequences of that perception enough to make the situation "come true," but the "conditions behind this perception" are also "already there."

Returning to the context of "fake news" and "social polarization," we often overlook:

Due to their inherent limitations, language and concepts can only capture specific attributes of reality and cannot equate to reality itself. This echoes the ancient Chinese saying by Laotzu,

"The Tao (Way) that can be trodden is not the eternal Tao; The Name that can be named is not the eternal Name."

It can be incredibly difficult to accurately describe and convey certain experiences or emotions through language and concepts. There are times when situations or moments are simply "beyond words" or "indescribable," and even our best attempts at expression can fall short. This challenge is experienced by everyone, even those who are considered "intellectual" or "well-educated." We must acknowledge that the ability to verbalize/conceptualize implicit knowledge or non-textual things (e.g., our feelings, senses, experiences, aspirations), or to effectively communicate through language and concepts, is a privilege. While education may improve our chances of success, it is not a guarantee.

Therefore, when we label someone, especially our close family and friends, as a "fan of X" who constantly spreads "terrifying fake news," have we ever truly empathized and understood the experiences, feelings, aspirations, and desires that lead them to view this information as accurate, and even wish others to see the same truth?

Regardless of whether we agree with the information they disseminate, the experiences behind their actions are real, their feelings are real, and their aspirations are real. These experiences, feelings, and aspirations are the real conditions for their perceptions. Whether we agree or not, that information must have captured some reality they feel and experience, which otherwise remains unnamed and supports their genuine aspirations and desires.

I appreciate zysbot's point during the discussion, emphasizing understanding the social mechanisms behind cognition to help us achieve an "emancipatory interest."

In my words, If we disagree with certain information, what we should do is not "attack/dismantle/destroy" the lifelines or shores they've found in the overwhelming sea of information. Instead, we should embrace their genuine feelings, experiences, and aspirations. If their understanding stems from genuine experiences of suffering or injustice, we should address, change, or resolve those real situations. If their cognition reflects societal exclusion or marginalization, we should expand our discourse and demonstrate through actions that what we seek is mutual flourishing, not a zero-sum game.

This might sound overly idealistic. In reality, there are often conflicting and incompatible belief systems. But given the limitations of concepts and language, logically conflicting ideas don't mean the realities they refer to can't coexist. Practically, many conflicting beliefs and discourses might differentiate between short-term remedies and long-term solutions.

Which is more important, treating symptoms or addressing root causes? Both are essential. We need both immediate damage control and solutions that address the root causes of problems.

In static logic, the two might be incompatible. But in a holistic system, both might have their roles.

Regardless of whether we believe in universality at the level of truth/ontology, at the level of cognition-praxis or epistemology-normativity, by letting go of the obsession of "claiming a single universal truth," and humbly recognizing the limitations of our bodies, consciousness, and concepts, we might reduce disputes stemming from inevitably partial understandings.

We're fated as blind men trying to figure out what an elephant is. Embracing plurality, curiosity, listening, and expanding our understanding of different perceptions might be the only way to achieve universality.