Philosophy › Section H · Value theory › Topic H.1

The fact-value distinction

"The river is polluted" and "the river ought to be cleaned" sound closely linked, yet many philosophers argue that no amount of the first kind of claim can, on its own, prove the second. Whether that gap is real shapes every debate about values.

Two kinds of claim

A descriptive claim says how things are: "the water contains lead", "most voters supported the bill". An evaluative claim judges how good, bad, right, fair, beautiful or important something is: "the bill was unjust". Some evaluative claims are prescriptive, saying what someone ought to do: "the city should replace the pipes". The fact-value distinction is the thesis that these are different kinds of claim, so that facts about what is the case do not by themselves settle what is good or what ought to be. The distinction is widely used and also contested. Defenders and critics both appear below, and this course takes no side.

Hume's observation

In A Treatise of Human Nature (1739 to 1740), David Hume remarked that writers on morality often reason for a while in statements about what is, about God or human affairs, and then suddenly switch to statements about what ought to be, without explaining how the new kind of claim follows from the old. This is the is-ought problem, sometimes called Hume's guillotine. In logical terms: a valid argument whose premises are all descriptive seems unable to reach an evaluative conclusion, since the conclusion contains a notion, "ought", that none of the premises contain. Commentators dispute how strong a claim Hume meant: some read him as asking for an explanation rather than denying that one exists.

Worked example: finding the bridge premise

Argument: "Cycle lanes cut traffic injuries, so the council ought to build more cycle lanes." 1. Classify the premise: descriptive, a claim that research could test. 2. Classify the conclusion: prescriptive. 3. Check validity using "Arguments, validity, and soundness": as stated, the premise could be true and the conclusion false, for instance if the cost would close a hospital. 4. Supply a bridge premise: "the council ought to adopt affordable measures that cut injuries". 5. Re-check: now the argument is valid, but the bridge premise is evaluative. 6. Conclusion: the ought came from a value premise, which needs its own defence. The gap was not crossed; it was filled.

Moore's naturalistic fallacy

G. E. Moore, in Principia Ethica (1903), made a related but different claim. He called it the naturalistic fallacy to identify goodness with any natural property, such as being pleasant or being desired. His open-question argument runs: if "good" just meant "pleasant", then asking "this is pleasant, but is it good?" would be as empty as asking "this is pleasant, but is it pleasant?". Yet the first question is a real, open one. So, Moore concluded, goodness is a simple, non-natural property. Critics reply that some true identities are not obvious: "water is H₂O" was discovered, not seen in the word's meaning, so an open question does not prove that two properties differ.

The appeal to nature

A third idea is often confused with the first two. The appeal to nature argues that because something is natural, it is good, or because it is unnatural, it is bad. "Our ancestors ate meat, so eating meat is right" and "vaccines are unnatural, so they are harmful" both fit the pattern. Counterexamples are easy: disease and floods are natural; spectacles and anaesthetics are not. The pattern is a version of the is-ought gap, since "natural" describes and "good" evaluates. Keep the three apart: Hume's gap concerns inference, Moore's fallacy concerns defining goodness, and the appeal to nature is an everyday argument from what is natural to what is good.

Objection: promising

John Searle (1964) argued that an ought can follow from an is. 1. Jones uttered the words "I hereby promise to pay you five dollars". 2. So Jones promised to pay. 3. So Jones placed himself under an obligation to pay. 4. So Jones ought to pay. Each step, Searle claimed, follows from facts about the institution of promising, so a descriptive starting point yields a prescriptive conclusion. Critics, including R. M. Hare, replied that the steps work only for someone who accepts the institution's rules, and that accepting them is already an evaluative commitment. On that reading, a hidden value premise does the work, just as in the cycle-lane example.

Objection: entangled concepts

Some words describe and evaluate at once. Bernard Williams (1985) called them thick concepts: "cruel", "courageous", "rude", "generous". To call an act cruel is to report facts about intended suffering and to condemn it in the same breath. Alasdair MacIntyre (1981) pointed to functional concepts: from "this watch loses an hour a day" it seems to follow that it is a bad watch, because what a watch is for is part of the concept. Ideas of telos, or purpose, carry the point further. Defenders of the distinction reply that a thick term can be split into a descriptive part and a separate attitude, though critics doubt that the descriptive part can be stated without the evaluation.

Objection: values inside science

Hilary Putnam, in The Collapse of the Fact/Value Dichotomy (2002), argued that facts themselves are reached with the help of values. Scientists choose between theories that fit the data equally well by appeal to simplicity, coherence and explanatory power, which are judgements about what is better to believe. If those epistemic values are legitimate, he argued, values are not mere feelings set apart from facts. Some traditions draw no sharp line either: on many accounts the Sanskrit term dharma covers both the order of things and the duties that order assigns. Defenders reply that epistemic values differ from moral ones, and that the logical gap between is and ought can survive even if the two are often entangled in practice.

Worked example: sorting claims

Classify each claim. 1. "The bridge was built in 1932." Descriptive: true or false by the historical record. 2. "The bridge is ugly." Evaluative, an aesthetic judgement. 3. "The council should repair the bridge." Prescriptive, a kind of evaluative claim. 4. "Leaving the bridge to collapse would be reckless." Thick: it describes risk and criticises at once. 5. "Most residents think the bridge is ugly." Descriptive: it reports people's values, which is a fact about them, not a value judgement by the speaker. 6. Test for the last case: could a survey settle it? Yes, so it is descriptive, even though its subject is values.

Common mistakes to avoid

First, treating a claim about people's values ("many people think lying is wrong") as an evaluative claim; it is a fact about those people. Second, merging Hume's is-ought gap, Moore's naturalistic fallacy and the appeal to nature, which are three distinct ideas. Third, assuming that the distinction proves values are subjective; it concerns what follows from what, and moral realists can accept it. Fourth, thinking any argument with an ought conclusion is fallacious; it may be valid if it includes an evaluative premise. Fifth, assuming every philosopher accepts the distinction; Searle, Putnam and others reject it in different ways.

Key takeaways

Descriptive claims say how things are; evaluative claims judge value, and prescriptive claims say what ought to be done. Hume observed that oughts seem not to follow from premises that only describe, so value conclusions need value premises. Moore's naturalistic fallacy is the separate charge of defining goodness as a natural property, supported by the open-question argument and challenged by cases like water and H₂O. The appeal to nature reasons from natural to good. Critics of the distinction cite Searle's promising argument, thick and functional concepts, and Putnam's claim that science relies on values. The debate remains open.

Flashcards

Flashcard

Descriptive claim

A claim about how things are

"The water contains lead."

Test your understanding

Question 1 of 5

Which statement is descriptive rather than evaluative?

Question 2 of 5

"Regular exercise lowers blood pressure, so everyone ought to exercise." What does Hume's is-ought problem suggest about this argument?

Question 3 of 5

According to Moore, what is the naturalistic fallacy?

Question 4 of 5

Critics of the open-question argument point to the example "water is H₂O". What does it show?

Question 5 of 5

Why are words such as "cruel" and "courageous" used against a sharp fact-value distinction?

← Previous topicNext topic →