We often describe our moment in history as one in which evidence is ignored, denied, or suppressed. That diagnosis is comforting. It suggests that the problem lies with bad actors who reject facts. What is more troubling—and more accurate—is that evidence is everywhere, but its origins and methods of generation – its grammars – have been lost.
In short, we have not stopped using evidence. We have stopped thinking with it.
In an earlier essay, I argued that much of public life now operates through decision-based evidence making: decisions are reached first, then justified retroactively by whatever data, anecdote, correlation, or expert view can be made to fit. What has become clearer since is that this practice is sustained not merely by cynicism or ideology, but by a widespread collapse in critical reasoning itself—a failure to distinguish among types of evidence, types of inference, and types of claims.
The reaction to the New Yorker’s December 8, 2025 article Against Science (sorry – paywall) —particularly the comments and secondary commentary it generated—offers a revealing case study. What was striking was not disagreement with science. It was the inability to articulate what kind of claim science was making in the first place, or what kind of evidence could legitimately support or refute it.
Correlation was treated as causation. Anecdote was elevated to counter-example. Statistical regularities were dismissed by personal exception. Moral discomfort was repackaged as methodological critique. At no point did many participants appear to ask the most basic question: what kind of knowledge is being asserted here, and by what mode of reasoning?
This is not an argument about trust in experts. It is an argument about epistemic illiteracy (i.e. not knowing about knowledge).
Modern societies once taught—explicitly—that evidence comes in different forms and serves different purposes. Inductive reasoning allows us to generalize cautiously from patterns. Deductive reasoning tests internal consistency. Analogical reasoning suggests similarity without proof. Statistical reasoning describes populations, not destinies. Causal reasoning requires mechanisms, not mere association. Scientific claims are provisional, bounded, and probabilistic by design.
None of this is intuitive. It must be learned.
What we are witnessing instead is a flattening of evidence into a single undifferentiated mass: everything is either “data” or “opinion,” “my truth” or “your truth.” Once evidence is stripped of its category, it becomes infinitely malleable. A correlation can be waved about as proof. A single case can nullify a trend. Suspicion of motive can substitute for rebuttal. And a political preference can masquerade as epistemological aka knowledgeable skepticism.
This is why decision-based evidence making thrives. It does not require dishonesty so much as indifference to method. If one does not know—or does not care—whether a claim rests on induction, deduction, analogy, deconstruction, counterpoint or causation, then any form of support will do. Evidence becomes decorative rather than constraining.
The danger here is not merely that bad policies are justified. It is that thinking itself becomes unserious. Public discourse fills with assertions that sound scientific but are not; critiques that sound rigorous but are not; and debates that are really contests of identity and affect, masquerading as disputes over facts.
The New Yorker article did not create this condition. It revealed it. When science is received not as a mode of disciplined uncertainty but as a set of talking points to be accepted or rejected wholesale, we should not be surprised that nuance collapses. A society that no longer distinguishes between evidence types cannot tell when it is being misled—or when it is misleading itself.
This is not a call for deference. It is a call for intellectual rearmament. Critical thinking is not a personality trait; it is a learned practice. It requires teaching people how claims are built, how evidence is weighed, and how conclusions are bounded. Without that scaffolding, evidence becomes raw material for belief rather than a constraint on it.
When evidence is promulgated without nuance, it does not enlighten. It inflames. And when critical thinking disappears, the problem is no longer that we disagree about the world. It is that we no longer know how to argue about it at all.
That is the real crisis—and it cannot be solved by more data or studies or hollow claims or conclusions that brag that they are data driven or evidence based.
Tell us about your thinking, your nuancing, your thought processes. Situate them in schools of thought and the best existential critiques of those schools. Tell us about your doubts and reveal uncertainties.
And please lose the conclusion that you just need more data and more studies to get better. The intellectual archaeologists of tomorrow may just read that with rueful smirks and think: Imagine that they thought more evidence and more data was all they lacked.