Balancing Verifiability with Inclusion of Underrepresented Knowledge

Imagine you are researching the history of a specific indigenous farming technique. You find a detailed description in a community archive, but it’s not cited in any major academic journal. Now imagine you are editing a global encyclopedia. Do you include that knowledge because it is true and vital to that culture? Or do you leave it out because it fails the strict rule of "verifiability"?

This tension sits at the heart of modern information management. For decades, platforms like Wikipedia, the largest free online encyclopedia have relied on a rigid standard: if it isn’t published in a reliable, secondary source, it doesn’t exist. This works well for established facts about world leaders or chemical elements. It breaks down completely when dealing with oral histories, local traditions, and marginalized communities whose stories were never recorded by mainstream institutions.

We are facing a crisis of representation. The digital public square is vast, yet it mirrors the biases of the past. To fix this, we need to rethink how we define "reliable" without sacrificing accuracy. It is not just about being politically correct; it is about building a complete picture of human knowledge.

The Trap of Traditional Verifiability

Let’s look at where the problem starts. The concept of verifiability was designed to stop vandalism and nonsense. If someone claims aliens built the pyramids, they need a solid source. But over time, "solid source" became synonymous with "Western academic peer-reviewed journal."

This creates a feedback loop of exclusion. Academic publishing has long been criticized for its own biases. Who gets published? Usually, researchers from wealthy institutions in the Global North. What topics get funded? Often, those that align with Western interests. When an encyclopedia only accepts these sources as valid, it effectively erases everything else.

Consider the field of ethnobotany. Indigenous peoples have used medicinal plants for thousands of years. Their knowledge is passed down orally, through practice, and within community structures. If a researcher from a university studies this, publishes a paper, and cites the elder, the encyclopedia can use the paper. But if the elder speaks directly, or if the knowledge exists only in a local community newsletter, it is often deemed "unverifiable."

This is not neutrality. It is structural bias disguised as rigor. By refusing to accept non-traditional sources, editors inadvertently privilege the powerful and silence the underrepresented. The result is a knowledge base that feels universal but is actually quite narrow.

Defining Underrepresented Knowledge

What exactly counts as underrepresented knowledge? It is not just "minority" topics. It refers to information systems that operate outside the dominant epistemic framework-the accepted way of knowing things in mainstream society.

• Oral Histories: Narratives passed down verbally, common in many African, Asian, and Indigenous cultures. These are rigorous within their context but lack physical "publications" in the Western sense.
• Local and Vernacular Media: Newspapers, radio broadcasts, and blogs in local languages that serve specific communities but have low global circulation.
• Community Archives: Digital or physical collections maintained by grassroots organizations, churches, or cultural centers rather than national libraries.
• Practical and Tacit Knowledge: Skills and understandings learned by doing, such as traditional craftsmanship or ecological management, which may not be documented in textbooks.

When we exclude these, we lose nuance. We lose context. We create a world where certain ways of life appear invisible or primitive simply because they weren’t written down in English or French during the colonial era.

Redefining Reliability: A New Framework

So, how do we balance the need for truth with the need for inclusion? We cannot throw out verification entirely. That would invite misinformation. Instead, we must expand our definition of what constitutes a reliable source.

First, we need to recognize Primary Sources, original materials created at the time of an event more broadly. Currently, encyclopedias discourage primary sources to prevent original research. However, for underrepresented groups, primary sources are often the *only* sources. A photograph taken by a community member, a recording of a speech, or a document from a local archive should carry weight if it can be authenticated.

Second, we must value Community Consensus, agreement reached within a specific group regarding facts or narratives. In many cultures, truth is determined by collective agreement among elders or experts, not by publication in a journal. If a respected cultural organization endorses a narrative, that endorsement is a form of verification.

Third, we need to embrace Open Access Publishing, digital literature that is free of all restrictions. Many scholars from the Global South publish in open-access journals due to cost barriers. These are legitimate, peer-reviewed sources, yet they are sometimes viewed with suspicion compared to expensive subscription-based journals. Editors must treat high-quality open-access publications as equal to traditional ones.

Practical Strategies for Editors and Creators

If you are involved in creating or curating content, here are actionable steps to bridge the gap between verifiability and inclusion.

1. Diversify Your Source Pool: Don’t just search Google Scholar. Look for databases like JSTOR Global Health, which includes regional journals, or HathiTrust, which hosts millions of digitized books from around the world. Use library catalogs from universities in the regions you are writing about.
2. Contextualize, Don’t Just Cite: When using a non-traditional source, explain why it is reliable. Instead of just linking to a blog post, write: "This account is supported by [Organization Name], a recognized authority on [Topic], which maintains extensive archives of local events." This adds credibility through explanation.
3. Collaborate with Subject Experts: Reach out to people who live the reality you are describing. If you are writing about a specific dialect, contact linguists or native speakers. They can point you to credible resources that mainstream algorithms miss.
4. Use Multiple Weak Signals: Sometimes one source isn’t enough, but three small ones are. If three independent community newsletters report the same event, and a local historian confirms it, that is stronger evidence than a single vague mention in a major newspaper.
5. Avoid "Notability" Bias: Notability criteria often favor famous individuals. Shift the focus to significance. Is this person or event significant to their community? If yes, it deserves documentation, even if they aren’t globally famous.

The Role of Technology and AI

Technology plays a double-edged sword role here. On one hand, search engines and AI models are trained on existing data, meaning they reinforce current biases. If Wikipedia lacks content on a topic, AI will likely hallucinate or ignore it.

On the other hand, new tools are emerging to help. Natural Language Processing (NLP), technology that helps computers understand human language is getting better at analyzing texts in low-resource languages. Projects like Wikidata are helping to structure unstructured data, making it easier to link disparate pieces of information.

We also see the rise of decentralized platforms. Tools that allow communities to host and verify their own data, linked via blockchain or other trust mechanisms, could eventually provide a new layer of verifiability that doesn’t rely on traditional gatekeepers.

Case Study: The Success of Local Language Wikipedias

Look at the growth of Wikipedia in languages like Catalan, Basque, or Kurdish. These projects started small, driven by passionate volunteers. They faced skepticism: "Who verifies these articles?" But they succeeded by building strong internal communities and partnering with local universities and cultural institutions.

In the case of the Kurdish Wikipedia, the Kurdish language edition of Wikipedia, editors worked closely with Kurdish historians and linguists to establish reliable sources. They didn’t wait for Western academics to write about Kurdistan. They digitized local newspapers, interviewed elders, and created a robust citation network based on regional credibility. Today, it serves millions of readers and preserves a rich cultural heritage that might otherwise have been lost.

This shows that verifiability is not a fixed standard imposed from above. It is a social contract built within a community. When that community is empowered, the quality of knowledge improves.

Challenges and Pitfalls to Avoid

Expanding inclusion is not without risks. One major concern is the spread of misinformation. Bad actors may exploit looser standards to insert propaganda or false narratives.

To mitigate this, transparency is key. Every article should clearly state its sources. If a claim comes from a controversial or niche source, flag it. Use tags like "Needs additional citations" or "Disputed" rather than deleting the content outright. This invites collaboration rather than censorship.

Another pitfall is tokenism. Including a paragraph about a minority group while ignoring the rest of their context is harmful. It creates a fragmented view of reality. Aim for depth, not just breadth. Understand the systemic issues that led to the underrepresentation in the first place.

Finally, avoid savior complexes. External editors should not assume they know best how to represent a community. Listen to the community. Let them lead the narrative. Your role is to facilitate and verify, not to dictate.

Building a More Complete Future

The goal is not to lower standards. It is to raise them for everyone. By recognizing diverse forms of knowledge, we make our information ecosystems more resilient and accurate. We move from a monoculture of information to a polyculture, where different types of evidence support each other.

This shift requires patience. It requires editors to read beyond their comfort zones. It requires platforms to invest in tools that support multilingual and multimedia sourcing. But the reward is immense: a world where every voice has the potential to be heard, verified, and preserved.

As we move forward, let us ask ourselves: What truths are we missing because we are looking in the wrong places? The answer lies not in abandoning rigor, but in expanding our vision of what rigor looks like.