Lived experience is not a reliable epistemology.
By: Helen Pluckrose
Published: Oct 20, 2025
In our polarised age, where warring narratives clash daily, simply asking someone how they know something is true can provoke intense hostility. The question is often seen as pedantic or insincere - as an attempt to undermine a cherished narrative rather than a genuine effort to understand it. Among activist tribes on the left, it may be taken as “gaslighting” or ‘wilful ignorance’ of marginalised experience; among dissident tribes on the right, as a “psyop” enforcing obedience to corrupt elites. In both cases, the reaction is usually sincere. The claim feels so obviously true within one’s own narrative that it seems impossible anyone could just be asking a question. The questioner must be an enemy.
But neither intuitive certainty nor righteous rebellion makes something true. We are all pattern-seeking creatures, quick to mistake intuition for knowledge. Thinking for oneself is essential, but it’s not enough. We must also test our thoughts against reality.
Anuradha Pandey recently shared a piece by Roger McFillin entitled, “Do You Have a Study for That?: The death of common sense in the age of citation worship.” She posted it to Substack with this comment which summarises the thesis of the piece.
I get skewered for observing patterns and connecting them without citing studies for every sentence. Educated people, in particular, habitually appeal to authority while sounding “rational”. But demanding gated papers for every claim isn’t nuance, it’s logically fallacious. Scientific findings begin with pattern recognition and the formulation of hypotheses. Reality is, obviously, messy, and not all causal variables can be separated. Just yesterday, a man said I can’t assume screen time is bad for society, even though the causal chain of screen time → lack of discipline and focus→lack of mastery → external locus of control → victimhood → lack of success is patently obvious if you spend 2 minutes thinking about it. Stop outsourcing your thinking to people incentivized to bury rude truths.
On the surface, this seems to be a communication problem that could be avoided by clarifying when one is conducting a thought experiment or floating an idea, rather than asserting something as true. We all observe patterns, though not always the same ones, and comparing our perceptions can be valuable. People generally ask for citations only when they believe you are making a truth claim and want to know your reasons.
The ongoing debate over screen use illustrates this well. Jonathan Haidt argues that social media harms adolescent mental health; Candice Odgers disputes this, finding no causal link in forty studies. My own experience makes Haidt’s stance plausible, but my own experience and intuitions are not the same thing as evidence. The issue clearly isn’t obvious to everyone and it clearly can’t be settled by “thinking about it for two minutes.”
Pandey is right to be wary of appeals to authority, but that differs from asking for evidence. The appeal to authority fallacy occurs when expertise alone is taken as proof: “You must accept this because Jon Haidt said so.” But Odgers is also an expert, and both back their positions with data. Those who wish to determine what is true must form conclusions on which is more credible by assessing that evidence. If one prefers to go with lived experience or intuition, this can be offered clearly as a personal reflection from experience so it does not prompt requests for supporting evidence.
It’s vital to separate two things: respect for truth - that which corresponds with reality - and respect for authority, which may or may not serve that concept.
In his section entitled “The Truth About Truth,” McFillin takes a strongly skeptical stance against expertise, writing,
I believe in science. Real science. The kind that questions everything and changes its mind when wrong. Not “The Science” that demands compliance while contradicting itself every six months. Remember when the experts knew for certain that lobotomies cured mental illness? When they insisted dietary fat caused heart disease while pushing sugar? When they told pregnant women thalidomide was perfectly safe? Here’s today’s version: Deadly. Lab-grown “meat” created from immortalized cancer cells? The future! Saturated fat from an egg? Toxic. Sixty ingredients you can’t pronounce in your “plant-based” breakfast bar? Whole food. SSRI’s save lives! Over 20 vaccine shots before age 5 is lifesaving healthcare! This isn’t science. It’s corporate theology dressed up in a lab coat.
I also believe in real science which is self-correcting but this reasoning is fallacious. One cannot demand that science change its mind when wrong and simultaneously fault it for new findings contradicting previous findings. Nor are such changes of position consistent with forced compliance. The examples of lobotomies, fat, or thalidomide would seem to show science correcting itself, rather than evidence that it can’t be trusted to do so. Dismissing all new claims of science simply because past ones have been wrong is no more rational than accepting them all because past ones were right. The only way to judge any claim is to test it against reality which requires empirical research. (Also ‘science’ does not speak this way)
McFillin goes on,
Here’s the beautiful thing about Substack, about independent media: We make the rules. I can share what I observe without genuflecting to academic gatekeepers. When I need to cite something, I will. When there’s a specific claim that is well studied, I’ll back it up. But my observations about the world? My analysis of patterns I’ve witnessed? My conclusions from decades of experience? Stating the obvious? Those don’t need institutional blessing. If that bothers you, there’s a comment section where you can perform your academic theater. Just know that demanding citations for obvious observations is the intellectual equivalent of asking for a permit to think. When someone demands a citation instead of engaging with your ideas, they’re not being responsible. They’re performing the same intellectual submission as putting pronouns in their bio or a syringe emoji in their profile.
You make the rules about what you say, yes. I defend that right and also your right to define the rules of your own epistemology. Reality will not bend itself to anybody’s rules, however, and it is still best ascertained by evidence. Generally, when people ask for sources, evidence is what they are after. You need to cite something reputable and evidence-based if you make a truth claim and you want people who care about what is true and believe this is determined by evidence to believe you. There really isn’t any way round that.
Your observations of patterns, lived experience and what appears to you to be obvious? If they are well-founded and they are not receiving institutional blessing, this is something that bothers me a lot! If they are unfounded and are not receiving institutional blessing, I am pleased, this would indicate that the system is working. I would need to know which it is and I don’t know how to establish that without asking you to back up claims with evidence.
If I am asking you to evidence something, it probably means I don’t think it is obvious, rather than that I think you require a permit to think. This appears to me to be a petulant and childish reaction. You absolutely do not require a permit to think, nor do I have any power to grant any such permit. The only permit I may issue to any idea is by granting it entry to my own inventory of likely truths. Perhaps what seems obvious to you does not seem obvious to me because I am ignorant of a particular subject, but if so, my request for sources indicates I am trying to become more informed.
I don’t know how to engage with an idea without asking whether or not there is any evidence to indicate that it is true. What would such an engagement look like? “That’s a nice idea you have there. Very pretty words. Nice rhythm in the syntax. Would go well with drums?”
Possibly the most revealing part of the piece is this anecdote:
Welcome to the world where “source?” has become the new “shut the fuck up.” Just last week, perfect validation of this phenomenon landed in my inbox. I’d written about therapists wielding unearned authority, and within hours, a distinguished mental health professional penned a 500-word dissertation demanding citations. The punchline? He spent three paragraphs describing how he’d been FIRED for refusing unethical policies, witnessed “foolish and ridiculous trends” destroy patient care, and watched the field deteriorate for decades. Then he demanded I provide “empirical research” to prove these problems exist. My brother, YOU are the empirical research. You just wrote your own citation. You’re standing in the rain, soaking wet, demanding a peer-reviewed study to confirm that water is, indeed, wet.
This does not sound as though the man was asking you to shut up. It sounds as though he was asking for evidence of specific claims made, having said he shares concerns about the problem of institutional capture affecting patient outcomes in therapeutic fields. (I also hear quite regularly from mental health professionals on this subject). This sounds entirely reasonable. It is not the case that anybody who shares your general concerns about an issue and also has lived experience of it will then accept every truth claim you make that supports your shared cause. It sounds as if that man had the mentality of a scientist and you have that of an ideologue. This impression is strengthened by the ending of the piece,
Stop asking permission to think. Start asking why they’re so desperate to control what you’re allowed to notice. RESIST
At root, this reflects a conflict between two epistemologies: one based on lived experience - personal observation and pattern recognition - and one grounded in empirical data. The first derives meaning from perception filtered through personal assumptions; the second seeks to minimise bias and confounding variables to approach truth as correspondence with reality.
Empirical methods are better at establishing what is true precisely because they mitigate the cognitive biases inherent in lived experience. This does not mean lived experience has no value. Our brains evolved to detect patterns and infer meaning in precisely this way to enable us to navigate the world. A person who cannot do this cannot function independently and keep themselves safe. But as a method for interpreting complex social phenomena, it’s deeply unreliable because people perceive things differently even from the same position. “Listen to women” or “listen to black people” sounds plausible until you actually do listen to a representative sample and realise their experiences and interpretations diverge widely. Then we have the option of picking the lived experience of those with whom we already agree and declaring it as authentic and dismissing the lived experience of those with whom we don’t as inauthentic. This typically requires arguing that the authentic ones are right because their lived experience corresponds with reality, but the most efficient way to determine this is to go straight to empirical research.
Lived experience is therefore a useful heuristic, often right but prone to error when filtered through motivated reasoning. Confirmation bias is universal: we humans instinctively trust evidence that fits our worldview and scrutinise that which challenges it. I experienced this editing Areo, where I agreed with only about a quarter of submissions I published. I had to guard against demanding higher evidentiary standards from those I disagreed with. Even so, an external analysis found Areo leaned slightly left. Bias is unavoidable, but it can be mitigated, chiefly by checking our intuitions against rigorous empirical research.
The conflict between lived experience and empirical research has been prominent in the recent culture wars. Both the postmodern left (the “woke”) and the post-truth right (populists) favour lived experience and distrust science - one because it’s “white, Western, and patriarchal,” the other because it’s “corrupt and elitist.” Each sees science as a tool of control, differing only on who the controllers are.
The illiberal populist right, however, has an advantage: they’re not entirely wrong. Many institutions of knowledge-production have indeed been ideologically captured by left-wing identitarian theories that constrain what can be researched and concluded. I have been addressing precisely this problem for many years from my position as an empiricist and rationalist on the liberal left. Where they go terribly wrong is not in saying that we have good reason to be concerned about the methodologies being used by people claiming to be experts whom we should be able to rely on to value truth as something that corresponds with reality. It is in becoming radically skeptical of the methods of science and empirical research itself and demanding everybody treat their politically biased lived experiences, observations of patterns and perceptions as authoritative and obvious. They are also prone to insisting there is something very morally wrong and suspicious about anybody who does not. (Yes, the postmodern left does this too). By being radically skeptical about almost anything that can be posited to have been claimed by “The Science,” they undermine those who are trying to draw attention to legitimate concerns about questionable or ideologically biased premises in areas like psychiatric medication or gender medicine. They also make people hesitant to accept life-saving medical treatment.
Institutions of knowledge production are expected to use methods that mitigate the flaws inherent in lived experience, but too often they don’t. When an institution becomes ideologically captured, it presents the lived experience or perceptual lens of one ideological group while failing to seek or include disconfirming evidence. A primary reason many have grown sceptical and resentful of expertise is that too much of it now operates within an ideologically biased lived-experience framework rather than through empirical research. Consequently, the problem is not with empirical methods themselves but with replacing them with the same perceptual lens and ideological bias that makes reliance on lived experience unreliable. We respect knowledge-producing institutions when we can trust them to do better than the average individual at assessing a variety of views rigorously and testing beliefs against reality.
This problem cannot be solved by relying further on lived experience nor by a battle between the postmodern left and the post-truth right to make their own worldview culturally dominant. Both camps make the same unwarranted leap: from recognising that “when institutions become biased by one observational lens, experts become no more reliable than political ideologues,” to believing that their own political ideologues are therefore more reliable than the experts. One could argue they are not always worse, but they lack the tools to be better. Meanwhile, rigorous independent researchers who offer valid criticisms, solid disconfirming evidence, and genuine solutions struggle to be heard amid the din.
None of this means that people who wish to write reflectively from their own lived experience, sharing their observations and the patterns they see, should refrain from doing so. Whether we feel unable to back up truth claims with empirical research because the available evidence is unreliable, or we simply wish to think aloud about patterns we notice and see if they resonate with others, we should feel free to do so, but with epistemic humility. When something cannot be supported by rigorous empirical research, frame it clearly as a perception, an observation, a lived experience, or a thought experiment rather than a truth claim.
I think it is particularly important that those of us who have any influence at all in the public sphere are conscientious about distinguishing between truth claims, arguments and speculations/observations. I work in a consulting capacity helping people facing an authoritarian ‘woke’ problem in their organisation and have written a couple of books on this subject. I also speak about this at conferences and individual events. I have also co-published several academic papers on the subject, (although this was part of a sting operation). In this capacity, I can be considered an expert. Consequently, I do often make truth claims about what these theories say and the impacts of them on workplaces without directly citing them, but it is nevertheless important that I can do so if asked to. Even if I think it should be obvious to everyone that, for example, diversity training does not improve workplace relations and can even make them worse, I should be able to show some evidence of this if I state it as a truth claim rather than as a hunch. Being asked for this should be understood as a perfectly reasonable request and not as an entitled demand that I ‘educate’ the less enlightened or an implication that I require a permit to think.
Much of what I write, however, is an argument from a liberal perspective. I argue that liberalism is the best system for advancing knowledge and resolving conflicts, or I explore what I think a good liberal response to a current event would be. Sometimes people confuse arguments with truth claims and ask me to provide evidence that liberalism is correct. I then have to explain that liberalism can be neither correct nor incorrect. It is a set of philosophical principles one adopts if one wishes to live in a liberal society. I can try to persuade someone that such a society is desirable, but if they genuinely do not want one, my arguments will not persuade them. When I write from this perspective, my responsibility is to reason well from clearly stated premises. I can suggest books an interested reader might enjoy, but there is no expectation to cite sources because the argument originates in my own thinking.
Occasionally, I write something that I think is true based on my own observations, but which I do not have any empirical evidence for. Then I will say something like “It is my perception that…” or “I suspect that…” to indicate that this is a personal perception, an observation of a pattern or a lived experience and not a truth claim so people should not take my word for it. This is partly a rather neurotic conscientiousness because I am afraid of telling people something that is not true and partly a sense of self-preservation, because I have a lot of nit-picky academic detractors. I could be biased by my own assumptions or only seeing part of the picture. I could be wrong. That’s not a confession of weakness; it’s just the reality of trying to think honestly in public.
