Why do intelligent, well-informed people sometimes examine identical evidence and reach opposite conclusions—not through carelessness, but through careful reasoning? This puzzle sits at the heart of our most intractable disputes: climate policy, medical ethics, political philosophy, even interpretations of historical events.

The standard model of rational disagreement assumes that if we gather enough evidence and reason carefully enough, convergence will follow. But some disagreements resist resolution not because one party lacks evidence or reasoning capacity, but because the parties hold different fundamental assumptions about what counts as evidence in the first place.

These are cases of deep disagreement—disputes that cannot be settled by appealing to shared standards, because the standards themselves are what's contested. Understanding how such disagreements work, and what they imply for how we should engage with those who see the world differently, reveals something essential about the social nature of knowledge itself.

In his final work, On Certainty, Ludwig Wittgenstein introduced a concept that illuminates why some disagreements seem impossible to resolve. He called certain foundational commitments hinge propositions—beliefs that function like the hinges on which a door swings. The door of inquiry can open and close, but the hinges themselves must remain fixed for any movement to occur.

Consider how a scientist investigates whether a particular drug is effective. She relies on assumptions about experimental methodology, statistical inference, and the reliability of measurement instruments. These assumptions aren't conclusions she's derived from evidence—they're the framework that makes gathering evidence meaningful in the first place. She cannot simultaneously question whether randomized controlled trials produce reliable knowledge and use such trials to answer her question.

Hinge propositions vary across communities of inquiry. A traditional healer and a biomedical researcher may both be intelligent and sincere, yet operate with different hinges: different assumptions about causation, different standards for what constitutes healing, different frameworks for interpreting bodily experience. When they disagree about treatment, they're not simply weighing the same evidence differently—they're inhabiting different epistemic worlds.

This doesn't mean all frameworks are equally valid or that evidence is meaningless. Rather, it reveals that evidence only counts as evidence against a background of shared assumptions. When those assumptions diverge at the foundational level, more evidence cannot bridge the gap—because the dispute is about what should count as a bridge.

Takeaway

Some disagreements persist not because one side ignores evidence, but because the parties hold different foundational assumptions about what constitutes evidence—commitments that cannot themselves be justified within the inquiry they make possible.

Suppose you discover that someone you respect—equally intelligent, equally informed, equally careful in reasoning—has reached the opposite conclusion on a question you've thought about deeply. Should this discovery shake your confidence? This is the problem of epistemic peer disagreement, and it generates a genuine puzzle for how we should respond to intelligent opposition.

One influential response, called conciliationism, holds that discovering peer disagreement should lead both parties to reduce confidence in their positions. The reasoning seems sound: if someone just as capable as you, with access to the same information, sees things differently, doesn't that suggest your own judgment might be flawed? Treating your own reasoning as uniquely reliable seems like a form of epistemic arrogance.

Yet conciliationism faces serious problems in cases of deep disagreement. If the dispute extends to foundational assumptions—to what counts as good reasoning or relevant evidence—then the very notion of "epistemic peer" becomes contested. By whose standards are we measuring intellectual competence? Each party may recognize the other as intelligent while simultaneously believing the other's foundational framework is mistaken. The apparent symmetry dissolves upon closer inspection.

This reveals something important about expertise and disagreement. Being equally intelligent doesn't guarantee agreement, because intelligence operates within frameworks of assumption. Two brilliant architects can disagree about what makes a building beautiful, not because one reasons poorly, but because they're working with different aesthetic foundations. The disagreement is reasonable on both sides, which is precisely what makes it so difficult to resolve.

Takeaway

When intelligent people disagree persistently, the puzzle isn't always which one reasons better—it may be that they're reasoning well within different foundational frameworks, making the notion of neutral comparison itself problematic.

If some disagreements cannot be resolved through evidence and argument alone, how should we engage with those whose foundational assumptions differ from our own? One tempting response is relativism: declaring that since neither side can prove its framework correct, all frameworks are equally valid. But this conclusion doesn't follow. Acknowledging that we cannot prove our foundations correct from some neutral standpoint doesn't mean abandoning the commitment to inquiry, evidence, and reasoned argument within our frameworks.

A more productive approach involves what philosopher Robert Fogelin called recognizing deep disagreement as a distinct phenomenon—neither a failure of reasoning nor a sign that one party is irrational. This recognition itself can transform how we engage. Rather than endlessly rehearsing arguments that presuppose contested standards, we might instead focus on understanding the other framework from within, grasping why it seems compelling to those who hold it.

Such understanding doesn't require agreement. But it opens possibilities that pure argumentation forecloses. Sometimes we discover that apparent disagreements are actually differences in emphasis or priority rather than fundamental conflicts. Other times, we find unexpected common ground beneath surface-level disputes. Occasionally, deep immersion in another framework genuinely shifts our own foundational commitments—not through argument, but through a kind of gestalt switch.

Most practically, recognizing deep disagreement helps us distinguish disputes where more evidence can help from those where it cannot. This distinction matters enormously for how we structure institutions, educate students, and conduct public discourse. Not every disagreement is resolvable, and pretending otherwise generates frustration and bad faith on all sides.

Takeaway

When disagreement runs to foundations, productive engagement shifts from arguing for conclusions to understanding alternative frameworks—not to abandon your commitments, but to distinguish disputes where evidence can help from those where it cannot.

Deep disagreement reveals something profound about knowledge as a social enterprise. We do not reason from nowhere; we reason from within frameworks shaped by communities, traditions, and institutions. This doesn't make knowledge impossible—it makes knowledge situated, always achieved from somewhere rather than from an imaginary view from nowhere.

For knowledge-producing institutions—universities, scientific bodies, journalistic organizations—this suggests the importance of maintaining genuine intellectual diversity at foundational levels, not merely surface-level disagreement. It also suggests humility: the recognition that our own frameworks, however well-evidenced internally, rest on commitments we cannot fully justify.

The goal isn't to dissolve all disagreement or to pretend that frameworks don't matter. It's to engage with intellectual honesty about what evidence can and cannot do—and to remain curious about how the world looks through eyes that rest on different hinges than our own.