Why Facts Alone Can Never Tell You What to Do

The question of whether science can settle moral disputes has become more pressing in recent years, partly because a prominent neuroscientist argued that it can. Sam Harris, in his 2010 book The Moral Landscape, made the case that questions about morality are actually questions about wellbeing, that wellbeing is a feature of conscious creatures that can in principle be measured, and that science is therefore the right tool for determining moral truth.

This is a significant claim, and it challenges one of the most entrenched assumptions in modern thought: that science tells us about facts, while values are a separate domain where science has nothing to say. If Harris is right, that assumption is wrong, and the implications are substantial. If he's wrong, we need to understand exactly why.

Hume's Is-Ought Gap

The standard philosophical response to Harris's position begins with David Hume, who noticed something in the 18th century that has never been satisfactorily answered. Hume observed that arguments in moral philosophy have a tendency to slide, at some point, from statements about what is the case to statements about what ought to be done. This transition, he thought, was never justified. You cannot derive an "ought" from an "is."

The is-ought gap, also called Hume's guillotine, is the logical point that no set of purely factual premises can entail a moral conclusion without the addition of at least one moral premise. Science tells you that sugar causes tooth decay. It cannot tell you, on its own, that you ought not to eat sugar. To get from the fact to the prescription, you need a value (something like "you ought to preserve your health") that is not itself a scientific finding.

Applied to Harris's argument: science might be able to tell you which actions maximize wellbeing. But it cannot tell you, without a prior moral commitment, that you ought to maximize wellbeing. That foundational commitment is a value, not a finding.

Harris's response to this objection is essentially to say that the commitment to wellbeing is so basic that questioning it is incoherent, like questioning whether we should care about truth when doing science. This is an interesting move but it does not fully close the gap. There have been serious moral frameworks, including some ascetic and some honor-based traditions, that do not treat wellbeing as the fundamental value. Dismissing them as simply confused does not engage with the philosophical challenge.

Moral Realism and Its Alternatives

The deeper question underneath Harris's argument is about the metaphysical status of moral claims. Are moral facts real in the same way physical facts are? Or is morality something else, a projection of human attitudes, a social construction, an evolutionary adaptation that feels objective but is not?

Moral realism is the view that there are objective moral facts that exist independently of what anyone thinks about them. Torturing children for entertainment is wrong, on this view, not because most people disapprove of it, but because it actually is wrong, in a way that does not depend on human attitudes. If everyone suddenly came to believe it was acceptable, they would simply be mistaken.

Moral realism has intuitive appeal. When we argue about moral questions, we behave as though we are trying to get at the truth, not just expressing preferences. When someone says that slavery was wrong, they do not mean "I disapprove of slavery." They mean slavery was actually wrong, even when most people in slaving societies accepted it.

But moral realism faces serious challenges. Where do moral facts exist? Physical facts are grounded in the physical world. What grounds moral facts? How do we come to know them? Our standard ways of knowing things, perception, reasoning from evidence, seem to underdetermine moral conclusions. People with access to the same facts persistently disagree about moral questions. That pattern of persistent disagreement is harder to explain if moral facts are objective and in principle knowable.

The main alternatives to moral realism are various forms of anti-realism. Error theory holds that moral claims purport to be objective but there are no objective moral facts, so all moral claims are false. Expressivism holds that moral claims do not describe facts at all but express attitudes. Relativism holds that moral claims are true or false relative to cultural or individual standards.

Each of these positions captures something but also faces serious objections. Error theory implies that when abolitionists said slavery was wrong, they were saying something false, which seems absurd. Expressivism struggles to explain moral reasoning and the logic of moral arguments. Relativism implies that we cannot coherently criticize practices in other cultures, which most people find unacceptable.

What Science Can and Cannot Contribute

Setting aside the question of whether moral facts are objective, there are ways in which science genuinely does inform moral reasoning, even if it cannot replace it.

Science can tell us about consequences. If you are trying to decide whether a policy is just, knowing its actual effects on human wellbeing is morally relevant. Historical moral progress has often been driven partly by better empirical understanding: the recognition that members of previously excluded groups have the same capacity for suffering and flourishing as those who excluded them was partly an empirical correction.

Science can also reveal when moral intuitions are based on factual errors. People's intuitions about which animals deserve moral consideration have often tracked superficial similarity to humans rather than actual capacity for suffering. Learning that fish experience pain in ways more similar to mammals than previously thought is morally relevant information, even if it does not settle the question of what we should do.

What science cannot do is provide the foundational values that give moral reasoning its direction. It cannot tell you whether to prioritize equality or liberty, individual rights or collective welfare, present generations or future ones. These are genuinely contested value questions, and the contest is not the result of factual ignorance. People who know the same facts hold different values because values are not derived from facts alone.

The Moral Importance of Getting This Right

Why does it matter whether science can answer moral questions? Because the answer determines who gets to speak authoritatively on the most important questions we face.

If Harris is right, moral expertise is a form of scientific expertise, and moral authority belongs to those who understand human wellbeing best. This is a centralizing move that makes philosophers, ethicists, and humanists somewhat redundant and concentrates moral authority in empirical researchers.

If the is-ought gap is real and moral facts are not straightforwardly scientific facts, then moral reasoning requires a different set of skills: the ability to reason carefully about values, to weigh competing considerations, to understand the history of moral thought, and to engage seriously with the lived experience of those affected by moral decisions.

The honest answer is probably that both matter. Good moral reasoning requires good empirical information about consequences. It also requires value commitments that science alone cannot provide. The mistake is to think that either one is sufficient without the other.

Hume identified the gap. We are still working out what to build across it.