Moore's Principia Ethica: Goodness, Value, and Right Action

Suppose someone tells you that a good life is a pleasant life. They might be making a claim about what gives a life value, or they might be proposing a definition: by "good," they mean "pleasant." Those are different claims. We could agree about what pleasure is and still ask whether a pleasant experience is worth having.

G.E. Moore builds Principia Ethica, first published in 1903, around that distinction. He argues that goodness cannot be defined by identifying it with pleasure, desire, or another property. His argument matters because a definition can appear to settle an ethical dispute before the substantive question has been considered. Calling something pleasant supplies a description; whether its pleasantness makes it good requires an argument.

The book also asks which things have intrinsic value and how judgments about value bear on what we ought to do. Moore's answer is not exhausted by a theory of moral language. His discussions of friendship, beauty, consequences, and the value of complex wholes give content to the distinction with which he begins.

Moore distinguishes a question about the meaning or nature of goodness from a question about which things possess it. If someone asks whether pleasure is good, an answer about pleasure's value does not automatically tell us what goodness is.

A useful comparison is a definition of a triangle. Explaining a triangle as a three-sided plane figure analyzes the relevant concept through its components. Moore asks whether goodness permits an analysis of this kind. His answer is that it does not: goodness is simple rather than composed of more elementary concepts (Principia Ethica, sections 6–8).

His comparison with yellow is meant to clarify simplicity. Someone who understands yellow can recognize what we are referring to, although listing its conceptual parts will not explain it as the definition of a triangle explains a triangle. Moore thinks goodness is similarly resistant to analysis. The comparison concerns their being simple notions; it does not make goodness a color or a property detected by eyesight.

This leaves room for several kinds of explanation. We can give examples, distinguish goodness from neighboring ideas, and discuss the circumstances in which something is good. What Moore denies is that these explanations amount to an analysis of goodness into something else. Indefinability therefore does not mean that ethical discussion has to stop.

Moore's challenge to proposed definitions appears in section 13. Imagine a proposal that identifies goodness with being desired. A person who understands the proposal can still ask whether what is desired is good. For Moore, this question expresses a genuine doubt rather than simply asking whether the desired is desired.

The reasoning depends on a contrast between an identity of meaning and a substantive connection. If "good" meant exactly "desired," then asserting that the desired is good would add nothing beyond saying that the desired is desired. Moore thinks ethical judgments do add something. We can understand both notions while considering whether they apply together.

The same challenge can be put to pleasure. The question whether pleasure is good seems intelligible even to someone who understands pleasure perfectly well. Moore takes the continuing intelligibility of the question as evidence that goodness and pleasure are different notions. His argument is commonly called the open-question argument.

The argument concerns the content of the question, rather than the psychological fact that someone can hesitate before answering it. People can hesitate because they are confused, uninformed, or inattentive. Moore needs the stronger claim that a competent understanding of the concepts still leaves a substantive question to answer.

This is also why an example cannot establish his conclusion by itself. Showing that one proposed definition fails would leave other proposals open. Moore claims the difficulty applies to any analysis of goodness into other notions. Whether he is entitled to that generalization is part of what a reader must assess.

Moore calls the mistaken identification of goodness with another property the naturalistic fallacy. His targets include accounts that identify goodness with pleasure or with what occurs in nature. His criticism concerns the attempted identification, rather than the mere suggestion that pleasure or a natural condition can be good.

Suppose, as an original illustration, someone says that a pleasurable friendship is good. That judgment need not commit Moore's fallacy. It attributes goodness to something. The different proposal that "good" simply means "pleasant" attempts to explain goodness away through the other property. Moore objects to that proposal even if some pleasures really are good.

The label can mislead. Moore also criticizes attempts to derive an account of goodness from a metaphysical reality beyond the objects of natural science. Moving from a natural property to a supposedly supersensible one does not answer his question if goodness is still being identified with something different (chapter IV). His account of indefinability is broader than a prohibition on using biological vocabulary.

A related distinction concerns the inference from a description to an evaluation. Saying that a behavior occurs in nature does not, on its own, establish that it deserves approval. That example can help reveal a missing evaluative premise, but Moore's analysis is specifically about identifying goodness with another property. Treating every questionable inference about what ought to happen as an instance of his precise argument blurs the issue.

There is a further qualification. An ethical theory might identify which things are good without claiming to define goodness. A hedonist could say that pleasure alone has intrinsic value while accepting that "good" does not mean "pleasurable." Moore's objection to a definition would not, by itself, refute that substantive claim. He needs a separate argument against pleasure being the only good.

Moore's positive position is usually described as a form of non-naturalism. He takes goodness to be real and irreducible to the natural properties of the things that are good. A natural description may tell us what something is like while leaving the evaluative judgment to be made.

This position should be distinguished from the claim that goodness is a separate substance. Moore is discussing a property, not an additional object hidden inside a pleasant experience. Nor is he treating "good" as an expression that merely reports the speaker's feelings. His account takes judgments of intrinsic value to make claims that can be true or false.

His use of intuition also needs care. In the preface he explains that calling a proposition intuitive concerns its inability to be proved by inference from other propositions. It does not mean that whatever first seems right must be correct. The distinction is about the grounds of a judgment, rather than a guarantee of its psychological certainty.

That leaves a serious question for the view: how should competing judgments of intrinsic value be assessed when argument reaches this point? Moore offers careful comparison and isolation of the objects being judged. A reader may still doubt whether those methods provide enough discipline. Indefinability does not itself answer the question of reliable knowledge.

Consider a possible objection to Moore's reasoning. Two expressions could present something differently even if they refer to the same property. If so, a meaningful question about their relationship would not necessarily establish that there are two distinct properties.

The familiar comparison with water and its chemical constitution can clarify the logical point. Learning a scientific identity can be informative; it need not be a repetition of something already understood from the ordinary word. This is an illustration of an objection to the argument, rather than a claim that Moore uses the comparison in Principia Ethica.

Applied to ethics, the objection asks whether a naturalist must offer an obvious synonym for "good." A naturalist might instead propose an identity requiring substantive investigation. Moore's argument against a transparent definition would not automatically defeat that different proposal. A further argument would be needed to show that goodness cannot be natural in that way.

This objection does not establish naturalism. It identifies a gap between showing that two concepts differ in meaning and showing that their objects must differ in reality. The reader has to decide whether Moore's argument crosses that gap successfully, rather than treating the phrase "open question" as a conclusive refutation of every rival account.

After distinguishing goodness from its proposed definitions, Moore separates questions about ends from questions about means. Something can be valuable because of what it produces, or valuable for its own sake. The reasons for these judgments differ (sections 15–17).

Consider an original example: a difficult conversation might be unpleasant but help repair a friendship. Its unpleasantness and its usefulness are distinct features. Judging the conversation useful requires a view about what it is likely to bring about. Judging friendship good for its own sake asks a different question, about the value of the relationship itself.

Moore calls attention to this difference because instrumental judgments depend on causal knowledge. Whether an action promotes something valuable cannot be settled merely by knowing what goodness means. We must also understand the action's likely consequences and the alternatives available.

An intrinsic judgment is not a prediction that something will improve another outcome. Its object is the thing considered as valuable in itself. Failing to distinguish these questions can make a defense of something's usefulness appear to establish its final value, or make an objection to an action's effects appear to settle the value of its aim.

The distinction does not require means and ends to fall into permanently separate categories. The same thing can have intrinsic value and also contribute to other goods. What matters is identifying which relation a particular judgment asserts.

Moore adds a complication: the intrinsic value of a whole need not equal the sum of the values its parts have considered separately. He calls wholes with this kind of relationship organic unities (sections 18–22). "Organic" here concerns value, rather than whether the thing is biologically alive.

Imagine, as an illustration, an experience of understanding a work of art. We can distinguish the object, the person's awareness, and the attitude taken toward it. Moore's principle warns against evaluating these separately and assuming their total gives the value of the entire experience.

The point is not simply that parts cooperate efficiently. That would concern the effects they produce. Moore's claim is that the complex whole can have an intrinsic value that is not obtained by adding independently assigned component values. The relation between its parts belongs to what we are evaluating.

This also does not mean that every whole is better than its parts. His principle permits differences in either direction. Nor does it imply that an object changes its own intrinsic value whenever it enters a different combination. The value of the combination is a further question.

This principle matters for his treatment of pleasure. If pleasure is a constituent of a valuable experience, it does not follow that the whole experience owes all its value to pleasure. Appreciating another person, understanding something beautiful, and enjoying that appreciation can form a complex object of evaluation. Removing the other features changes what is being judged.

Moore proposes considering things in isolation when we assess their intrinsic value. We ask what value something would have apart from its usual effects and accompaniments. The method tries to prevent a useful consequence from being mistaken for value in the thing itself.

His comparison between a beautiful unobserved world and an ugly one applies this method (section 50). He asks whether the beautiful world would be better even if neither world could be contemplated by a human being. His judgment challenges the restriction of all value to pleasure within human experience.

That is a thought experiment, not evidence about an actually uninhabited world. A reader can understand the comparison and dispute the judgment. Its role is to clarify what would have to be accepted or denied when deciding whether beauty has value apart from being enjoyed.

Isolation must also be used alongside organic unities. Separately examining the constituents of a whole does not remove the need to evaluate the whole itself. Otherwise the method would reproduce the very error Moore is warning against: attributing all the value of a complex experience to whichever isolated constituent first attracts attention.

In chapter VI, especially section 113, Moore gives a substantive account of important goods. He places personal affection and the appreciation of beautiful objects at the center of his view. The valuable things are complex states of consciousness and their relations to their objects, rather than a bare quantity of agreeable feeling.

This is why the book should not be reduced to a prohibition on definitions. Moore wants to distinguish worthwhile ends and to explain why the relationships and experiences involved matter. His account of organic unities makes room for the character of what is loved or appreciated to affect the value of the whole.

His evaluations remain claims requiring assessment. Indefinability alone does not establish that personal affection and aesthetic appreciation are the greatest goods. A person could accept his analysis of goodness while disagreeing with his ranking of ends. Keeping those arguments separate allows us to identify where the disagreement lies.

The same caution applies to the beautiful-world comparison. In section 113 Moore treats the mere existence of beauty as much less valuable than consciousness of beauty. His thought experiment therefore should not be used to portray his final view as indifferent to experience or to human relationships.

Moore connects right action with producing the best possible results among the alternatives available. This is a consequentialist feature of his account: the moral evaluation of action depends on what results it brings about (chapter V). Rejecting a definition of goodness as pleasure does not entail rejecting consequence-based reasoning.

Knowing what ought to be done consequently raises difficulties beyond identifying intrinsic goods. We may not know the remote consequences of an action, the effects of alternative actions, or how the relevant goods and evils compare. Moore discusses the limits this places on confident particular judgments.

His treatment of familiar rules belongs to that practical problem. A rule can guide conduct because experience gives us reason to think that following it generally supports better results. That does not make the rule's authority a definition of goodness or show that we can calculate every consequence of every action.

The distinction between a standard of rightness and a usable decision procedure is helpful here. Saying that the right action has the best results states a criterion. It does not supply an agent with complete knowledge of those results. A reader should assess both Moore's criterion and his account of what limited agents can reasonably know.

A careful reading can keep three questions separate: what goodness is, which things possess it, and what actions best promote it. Moore thinks confusion between them has distorted ethical argument. His own answers also need to be assessed separately.

Return to the suggestion that a good life is a pleasant life. Ask first whether it is a definition or a substantive evaluation. If it is an evaluation, consider what exactly is being valued: pleasure by itself, the whole experience containing it, or the consequences that follow. That exercise makes Moore's distinctions useful without requiring agreement with every judgment he makes.

The remaining challenge is to defend the connection between those distinctions and his positive position. The openness of a question does not relieve us of assessing the answer. Principia Ethica asks us to take value seriously enough to investigate it, including when an attractive definition seems to have settled the matter already.

• G.E. Moore, Principia Ethica, originally published in 1903; the linked transcription follows the 1922 reprint. Read sections 6–14 for indefinability and the naturalistic fallacy, sections 18–22 for organic unities, and chapters V–VI for action and intrinsic goods.