On Truth, Morals & Democracy

1. The Munchausen Trilemma

Any constructive argument attempts to determine truth. To participate in meaningful debate, parties must first agree on a system to establish what is true and what is false. While logic and reason can develop ideas from existing axioms, every argument claiming absolute certainty eventually collapses in one of three ways: circular reasoning, regressive reasoning, or dogmatism. This is the Munchausen trilemma. Because individuals only have their own perception and reasoning to assess reality, every truth claim is contingent on those means. This does not mean truth does not exist — it means certainty about it is unavailable. The question, then, is not how truth can be reached with certainty, but which methods best approximate it.

2. Reasoning, Evidence, and Consensus

Naive subjectivism holds that since no truth is absolute, every opinion is equally valid. This is technically irrefutable but practically useless: if every belief is equally true, none can guide action. A better view treats inquiry as fallible but non-arbitrary. Careful reasoning from evidence - proposing explanations, testing them against observation and revising them - produces beliefs that, while never certain, are reliably better than those formed without such discipline. Bridges built on careful engineering stand; bridges built on wishful thinking do not.

It should be noted that consensus among competent inquirers is a useful symptom of good inquiry, not its standard. This is a crucial distinction. Consensus amongst astrologers says nothing about the stars; consensus amongst astronomers does. Scientific consensus hinges not on the agreement itself but the underlying discipline of evidence and method. For matters too complex for individuals to verify, such as general relativity, epidemiology and climate science, expert consensus is the average person's best available guide. However, the epistemic work is done by the evidence, while consensus signals that the work has been done.

3. Why Inquiry Tends to Converge

The scientific method insists that reality pushes back against false beliefs in ways that accumulate over time. Successful predictions survive while those that fail are abandoned. Human psychology supports this convergence. People have evolved to become broadly truth-seeking, as friction inevitably arises when beliefs and reality diverge. Mortality ensures that those clinging to failed paradigms are eventually replaced. Societies may stray from the scientific method for prolonged periods, but the reality of a world that refuses to bend to belief tends to reassert itself.

This convergence should not be overstated. It works best where feedback is tight - physics, engineering, medicine - and more slowly in domains where the world pushes back ambiguously. Systematic errors often take centuries or even millennia to correct. Still, the pattern is visible enough across history to observe that disciplined inquiry, sustained across generations, tends to track reality.

4. Gravity Exists

Gravity is not constituted by belief in it. Objects with mass attract one another regardless of what anyone thinks and did so long before any conscious creature arose to notice. Our theories - Newton's, then Einstein's, and whatever comes next - are successively better approximations of that underlying reality. None achieves certainty, but each improves on the last by making better predictions that fit a wider range of phenomena. The Munchausen trilemma reminds us we cannot verify the correspondence between theory and reality from some neutral vantage. Yet certainty is not mandatory for truth to emerge, so long as that truth approximates reality well enough. The accuracy of such approximations, attained through the rigorous application of the scientific method, is what distinguishes gravity from phlogiston, and heliocentrism from flat earth speculation. Truth is mind-independent, and while access to it is fallible the best proxy is produced from disciplined reasoning about the evidence.

5. Moral Truths

Moral claims have the same status as empirical ones. Anti-realists typically deny this by arguing that morality, unlike empirical fact, has no mind-independent existence and therefore falls outside the scientific method. The objection misreads what realism requires. Moral realism does not demand that moral facts exist as an independent substance or be innately known to every human. It requires only that moral claims have truth-values not reducible to what anyone happens to believe. The grounding is the conscious experience of sentient beings. Suffering and flourishing are not opinions; they are states of actual creatures with intrinsic character. A being in agony is worse off than the same being at ease - not according to an observer, but for itself, as a feature of what it is like to be that being.

Sophisticated anti-realists press harder. Some argue that human moral intuitions are artefacts of ancestral selection pressures rather than responses to moral facts, and so cannot track any such facts even if they exist. Others hold that objective moral properties would be metaphysically strange in ways no natural property is. These are serious challenges, but both assume moral facts would have to exist independently of the creatures they concern. If moral facts are instead facts about the conditions of conscious creatures, the challenges dissolve. The evolved capacity to recognize suffering in oneself and others is not a distorting lens but direct access to the relevant evidence, and the properties in question are not metaphysically strange because they are biological. Intuitions shaped by evolution track moral facts precisely because both are rooted in the same underlying biology.

If one grants this, moral realism follows. Moral claims become claims about the conditions of conscious creatures: which actions produce suffering, which support flourishing, how the interests of different beings weigh against one another. These questions are empirical in principle, if staggeringly complex in practice, and are not made true or false by consensus. This requires no departure from a scientific worldview. Suffering is not a metaphysical posit but a biological phenomenon, the product of nervous systems, nociceptors, stress hormones, and neural activity that can be measured, compared, and in many cases manipulated. Flourishing, similarly, corresponds to identifiable physiological and psychological states. These states represent facts about creatures in the world, not projections onto it. What moral realism requires is only the further recognition that such states carry normative weight as a matter of what they are. Pain is not a neutral fact one happens to dislike; its badness is constitutive of what pain is for the creature enduring it.

On this view, slavery was wrong in 1700 even where consensus endorsed it, because it produced massive suffering in beings who were morally significant regardless of what was believed about them. Abolitionists were not ahead of their time in the sense of advocating a view that would become true once consensus shifted. They were correct, and consensus caught up, just like it did in the natural sciences. The same applies to the expansion of moral concern to women, other races, and animals. Each expansion tracks morally relevant properties that were always there to be recognized.

Consensus retains a role, but a demoted one. Moral consensus tracks moral facts imperfectly, valuable when those within it reason carefully from evidence about the experiences of those affected, and failing when such evidence is ignored or structurally inaudible. Modern examples such as factory farming show that consensus can harden against the interests of affected beings even as evidence of their suffering accumulates. Clearly, this is a failure of inquiry, not a feature of moral reality.

6. The Temptation of Epistocracy

If truth-tracking requires careful reasoning, should political decisions not give greater weight to better reasoners? The argument has real force but misreads what the framework implies. Most political questions are not purely epistemic, and much more concerned whose interests count and how to weigh them. On such questions, the affected parties have privileged standing. If moral facts are grounded in the experiential states of conscious beings, the testimony of those beings serves as the most substantial moral evidence available. A system that filters it out to achieve better reasoning has filtered out the very evidence this reasoning depends on.

"Better reasoning" is also not neutrally assessable. Every historical attempt at epistocratic gatekeeping has been designed around the interests of those administering it. This is structural, not contingent: the people designing the filter are the ones who benefit from its shape, and in practice every such filter has tracked proximity to the elite that designed it more reliably than reasoning ability. Even a well-designed filter would deliver less than epistocracy promises. Expertise is reliable in domains with tight empirical feedback, where bad reasoning is punished by failed predictions or collapsed bridges. Political judgment is not such a domain — its feedback loops are slow, its counterfactuals unavailable, its variables too many to isolate. Credentialed experts have consistently failed to outperform informed generalists at political prediction, and education tends to sharpen the defence of existing beliefs as much as the search for better ones, producing sophisticated justification more reliably than sound judgment.

Democracy's epistemic case does not depend on every voter being wise. Aggregation across many moderately-reliable judgments can outperform narrow expert judgment when errors are independent; restricting the franchise amplifies the correlated biases of the remaining pool. Universal suffrage forces policymakers to attend to everyone who can vote, surfacing information that would otherwise be ignored. Regular elections enable error correction, perhaps democracy's most significant epistemic feature. Any form of Epistocracy risks ossifying elite consensus, which history suggests is often wrong in ways the elite cannot see from within.

Many epistocratic proposals are well-intentioned: voters are often uninformed, and collective decisions sluggish. But the interventions with evidence behind them, such as civic education, better information environments, and deliberative institutions, expand meaningful participation rather than restricting it. Epistocracy treats the symptom by eliminating the votes rather than the ignorance. Universal suffrage should not claim everyone reasons equally well. Rather, it is the primary mechanism to ensure that evidence about how policies affect people's lives enters political decision-making. Far from the logical extension of evidence-based inquiry, epistocracy is its inversion.

Conclusion

Certainty is unavailable in any domain, empirical or moral. This does not collapse the distinction between truth and opinion. Reality pushes back against false beliefs, and disciplined inquiry, sustained across generations, tends to track it. Consensus is a useful symptom of such inquiry, not its standard, and its reliability depends on the feedback the domain provides — tight in physics, looser in politics, imperfect but real in morality. The framework that makes sense of empirical progress, a mind-independent world accessed fallibly through disciplined inquiry, makes sense of moral progress for the same reason: the suffering and flourishing of conscious creatures are facts about the world, not projections onto it. It also constrains the political arrangements we can defend. A system that filters out the experiential evidence of those affected by its decisions does not track truth more reliably; it simply tracks less of it. Morality is no more reducible to consensus than physics is, and democracy is not a concession to equal reasoning but a mechanism for keeping the relevant evidence audible.