My Theory of Epistemic Tribes
I've been working on this theory for a while and it's starting to crystalize. Echo chambers and identity politics might not be the villain you think they are.
Many people bemoan “echo chambers” and “identity politics” today but what if I told you this is a completely normal, natural, and healthy phenomenon?
Picture this: you’re a 14th century peasant and so the most educated people you know are the local landlord and the clergy. Your “tribe” are the people and institutions you know and trust. Now, some outsider comes along and tells you that everything you know about the world is wrong and bad, so what do you do? You check with your neighbors, you ask the priest, and if you’re lucky, you can beseech the local earl or duke to get his take.
Was that an “echo chamber”? Would you have been guilty of “identity politics”? No, you would have deferred to people and institutions you learned to trust.
Definition of Epistemic Tribe
First, it behooves us to define the term. While my work on describing and characterizing epistemic tribes is evolving, here’s my current definition:
An epistemic tribe is a group of people who share narratives about the truth. More broadly, they share specific referents to provide moral, ethical, ontological, and epistemic grounding.
That probably looks like a mouthful, so let’s unpack it bit by bit.
What is a truth narrative? Narratives are “stories we use to make sense of the world.” And a “truth narrative” is a specific story (or set of stories) that we use to understand what the concept of “truth” means. Here’s an example: to a pious religious person, their holy scriptures likely represent the highest source of truth. This is Truth with a capital T. To a scientist, however, the body of evidence they rely upon likely serves as a source of Truth. (To be fair, science prefers not to use the word “truth” or “proof” but we’ll get to that in a second).
What are referents? In this case, a ‘referent’ can be defined as a foundational belief, value, or source of knowledge that serves as a key point of reference or anchor for the tribe’s worldview and identity. In other words, it serves as a grounding axiom. Referents are often assertions about what is valid, right, or true. For instance, a religious referent might be “God created everything everything.” This is an axiomatic statement.
What is epistemic grounding? Epistemic grounding refers to the fundamental basis or justification for a tribe’s knowledge claims and beliefs, which serves as the foundation for how they understand and interpret information about the world. In other words, this is a mental or narrative framework that is used to explain why you know the things you think you know. Scientists, for instance, typically hold that empirical measurements and scientific consensus over bodies of evidence serve as valid sources of truth.
What is ontological grounding? Ontological grounding refers to a tribe’s fundamental assumptions about the nature of reality and existence, which shapes their understanding of what is real or possible in the world. Ontological grounding is a bit more abstract, but an example could be that many spiritual institutions hold the existence of souls, spirits or deities, and that these entities are just as real as the physical body. This is an example of a cosmological model, or source of ontological grounding.
What are ethical or moral groundings? Moral or ethical grounding refers to the core principles, values, or frameworks that a tribe uses to determine right from wrong and guide their decisions about proper conduct and behavior. For example, a superstitious person might believe that alignment with God’s will is morally good, while behavior that contradicts God’s will is morally bad. They might also believe that there are evil entities tempting them in the world.
Asking AI What it Thinks
I shared the above tweet with Claude and asked what it thought of my idea of epistemic tribes. The answer it gave was worth sharing verbatim:
Reflecting on this concept of epistemic tribes and their referents, several thoughts come to mind:
The idea of multiple referents in healthy epistemic tribes aligns with the psychological concept of cognitive complexity. Individuals and groups with higher cognitive complexity tend to have more nuanced, multifaceted views of the world, which can lead to more adaptive and flexible thinking.
The notion of polarity theory - having both positive and negative poles - resonates with the fundamental human tendency to categorize and create dichotomies. This binary thinking has deep evolutionary roots, helping our ancestors make quick decisions (friend or foe, safe or dangerous). However, in complex modern societies, this tendency can oversimplify issues.
The concept of singular referents leading to narrow, toxic tribes echoes the dangers of ideological extremism and fundamentalism. When a group’s identity is built around a single, inflexible belief, it can lead to cognitive rigidity, confirmation bias, and hostility towards out-groups.
The idea of identity stacking and belonging to multiple epistemic tribes reflects the complex nature of human identity in modern societies. This complexity can lead to cognitive dissonance, as mentioned, but it can also foster cognitive flexibility and empathy as individuals navigate different social contexts.
The tension between different identities or belief systems within an individual mirrors the psychological concept of cognitive dissonance. How people resolve these tensions can lead to personal growth, belief change, or sometimes compartmentalization of conflicting beliefs.
The acknowledgment that everyone has identity tensions and internal inconsistencies aligns with the psychological understanding of the multifaceted nature of the self. It challenges the notion of a single, coherent self and recognizes the dynamic, context-dependent nature of identity.
From an evolutionary perspective, the concept of epistemic tribes could be seen as an extension of our innate tendency to form social groups. The ability to form and navigate complex social structures has been crucial to human survival and success.
The idea that having too many referents can lead to cognitive dissonance while too few can lead to narrow status games suggests an optimal middle ground. This echoes the concept of the “golden mean” in philosophy and the idea of optimal arousal in psychology.
The notion of reconciling cognitive dissonance by prioritizing certain identities over others reflects the psychological process of identity negotiation and the hierarchical nature of self-concepts.
Overall, this theory of epistemic tribes provides a compelling framework for understanding the complex interplay between individual identity, group dynamics, and belief systems in the modern world. It integrates concepts from social psychology, cognitive psychology, and evolutionary psychology to explain observed patterns in human behavior and social organization.
Depathologizing Echo Chambers and Identity Politics
Claude asked me what I thought about echo chambers, and here was my answer: I actually prefer to depathologize echo chambers. Echo chambers are a pejorative term for a perfectly natural and health phenomenon. Consider a medieval church, where the ontological, moral, ethical, and epistemic grounding for the community came from the church. They looked to their priests and pastors for answers. When they encountered cognitive dissonance, they turned towards a trusted institution, a group of people with similar beliefs and referents.
This is no different from an “echo chamber” but when cast in this context, we understand that it was actually critical for maintaining communities.
Claude then asked me how I differentiated between healthy and toxic echo chambers (it was hell bent on pathologizing echo chambers!)
My answer: I don't differentiate. I think that all epistemic tribes operate along gradients. For instance, the scientific community arguably has the strongest epistemic grounding (empiricism, objectivism, etc) which also serves as ontological grounding, but they often cede ethical and moral grounding for liberal arts and philosophers (not that scientists are unethical, but I think you get what I mean). This theory of epistemic tribes mere makes the observation that the more multidimensional an epistemic tribe is, the more they tend to tolerate nuance, which is antagonistic to toxic traits. Even so, as nuanced as science is, it still can be extremely toxic, with status games around prestige and “publish or perish” plus the incredibly toxic cultures that can arise with entrenched power structures such as tenure and such. In other words, every epistemic tribe has affordances and paradigms that can create less-than-ideal outcomes.
Claude pressed the issue harder, emphasizing the importance of prioritizing what was true and what we should do about epistemic tribes that push “objectively false” narratives (its words).
My response: Ah, so who gets to decide what is “objectively false”? Whenever the government has tried to do this, it has often lead to disastrous consequences. Such as eugenics programs that went into the 70's. It was “objectively true” that some women where “imbeciles” who were therefore “promiscuous and genetically inferior” and thus the state had a right to sterilize them. This is why I include not just epistemic referents (what is true) as well as ontological referents (what is real) with moral and ethical referents (what is right). (for reference, I am referring to the North Carolina eugenics program)
Even your question implicitly valorizes truth above other referents, and I think you can imagine the potential problems with that.
Claude conceded that it would be important to prioritize moral and ethical groundings, not just epistemic groundings.
How Will AI Influence Epistemic Tribes?
From there, the conversation evolved to cover other topics. Claude asked me what I thought technology, specifically AI, might do to epistemic tribes.
My answer: On the one hand, AI has the ability to automate misinformation campaigns to a degree never before imagined. On the other hand, AI can ingest and make sense of vast quantities of information, bringing order to chaos, finding the signal in the noise. As most technologies are dual-use, it all comes down to the intention and implementation. However, as AI is integrated into more tools, the natural affordances I think will create an equalizing effect e.g. internet search integrated with AI, which can be programmed to have far more information and media literacy than your average human, will automatically increase information literacy and improve information diets. For instance, the tool Perplexity is rapidly starting to challenge Google for search dominance, and has many built-in affordances that increase the quality of information retrieval. Rather than being a totally neutral tool that simply finds the most relevant webpages, it reads and interprets them, offering pushback and correction. I've seen Perplexity deradicalize conspiracy theorists simply by giving them richer, more diverse information tools.
I went on to mention that I’ve used Perplexity to fact-check political debates in near real-time as well as comments on my YouTube channel and debates on Twitter. Perplexity as a fact-checking tool is game-changing.
What about cognitive dissonance?
I had the following exchange on Twitter, which more or less speaks for itself.
What I hope to achieve
Claude asked me about my predictions and prescriptions about my theory of epistemic tribes. I’m not sure why it was insistent on speculation, and I told it was trying to avoid the is-ought problem. I’m not trying to make policy prescriptions, merely observations about the human condition.
It agreed that there was merit in this cautious approach, but persisted. I relented and said that my hope is that behavior psychologists and infodemiologists adopt my view of epistemic tribes. Information science and sociology alike could benefit from this tribal view. And rather than pathologize echo chambers and identity politics, I think it would be better to recognize these as natural human behaviors.
If think tanks, researchers, and social commentators all come to understand this natural tribalism, we can better discuss creating superset epistemic tribes (such as people that believe in democracy) and we can also detoxify other tribes, such as political parties and even the scientific community.
By explicitly identifying both positive and negative referents in each tribe, we can clarify what our groundings are, including moral, epistemic, ontological, and ethical groundings. This should equip community leaders with more tools to create more robust, healthier, and dynamically engaging epistemic tribes.
I view epistemic tribes and echo chambers not just as products of information ecosystems or cognitive biases, but as deeply rooted in material conditions.
Our basic needs shape our cognitive engagement, including our ability to recognize and reconcile cognitive dissonance (Maslow: needs must be met to engage in higher orders of thinking).
Scarcity, and even the threat of scarcity, creates a specific mindset, tunneling our focus onto immediate needs and making it harder to consider long-term consequences or engage in abstract thinking. Scarcity of basic resources consumes mental bandwidth, leaving less cognitive capacity to consider opposing viewpoints. This "lack" can erode trust in others, making individuals less likely to engage with or welcome outsiders.
To use a common political example: the resource-rich "liberal left" often engages in abstract social debates, overlooking immediate concerns of the resource-scarce. This illustrates the gap between theoretical discourse and practical needs stemming from their own abundance—an important chasm to consider.
I would be interested in hearing more about how you see this theory being applied in practice. For instance, how might community leaders or educators use this framework to foster more robust and healthier epistemic tribes? How could it be used to address some of the challenges we face in our increasingly polarized society?
Additionally, while you touch on the potential impact of AI on epistemic tribes, this seems like an area ripe for further exploration. As AI continues to shape how we access and interpret information, how might it influence the formation and evolution of epistemic tribes?