Hume’s Is–Ought Problem and Its Solution

Is and Ought

David Hume (1739/1978, T 3.1.1.27) observes in the Treatise of Human Nature that moral philosophers often begin by speaking about facts. They write about what is the case, how people act, what properties they possess, or what conditions obtain in the world. Then, within the same argument, an “ought” or “ought not” appears. Hume notes that this transition often occurs without explanation. Texts begin with statements about what is and then move to statements about what ought to be, without explaining what justifies this new claim.

Hume thereby formulates the problem that later became known as Hume’s Law, Hume’s Guillotine, the is–ought problem, or the is–ought gap. An ought-statement makes a different kind of claim from a mere statement of fact. Anyone who derives a norm from facts must explain how the normative content enters the conclusion. Hume did not see how this transition could be achieved by logic alone. For him, there was a sharp dichotomy between is-statements and ought-statements. On the weak, linguistic-logical reading, this is trivial. Without an ought-operator that enables the transition from is to ought, no ought-statement can follow from purely descriptive premises. This follows from the conservativity of logic, as Charles Pigden (2010) correctly observed.

The reason lies in the different functions of the two kinds of statements. A factual statement describes a state of affairs. An ought-statement contains a normative claim. Anyone who moves from facts to norms must explain how the normative content enters the conclusion. Hume did not see how this transition could be achieved by logic alone. For him, there was a sharp separation between is-statements and ought-statements. This observation identifies a real error in many moral arguments. An ought-statement cannot simply be smuggled into an argument without explanation.

Hume’s Conception of Morality

For David Hume (1739/1978, T 3.1.1.6), morality is characterized by its effect on the will and on action. Moral judgments would therefore have to possess motivational force. He writes: “Morals excite passions, and produce or prevent actions.”

From this, Hume concludes that reason alone cannot produce moral judgments. Reason may recognize relations between ideas and matters of fact, but by itself it cannot generate an impulse to act. Since moral judgments, according to Hume, must move the will, moral rules cannot be conclusions of reason. Hume (1739/1978, T 3.1.1.6) writes: “The rules of morality, therefore, are not conclusions of our reason.”

Hume’s Attempted Solution

Hume then attempts to solve the problem through sentiment. According to him, moral judgments arise from moral feelings. A person considers an action or a character, understands the circumstances, and feels approval or disapproval. For Hume, this sentiment forms the core of moral judgment. He summarizes this in the formula that morality is felt rather than judged: “Morality, therefore, is more properly felt than judg’d of.” (Hume 1739/1978, T 3.1.2.1). Moral evaluation therefore comes from feeling. Reason merely helps identify the appropriate means for acting accordingly.

He summarizes this division of roles in the following famous sentence: “Reason is, and ought only to be the slave of the passions.” (Hume 1739/1978, T 2.3.3.4)

The passions therefore determine the direction of action, while reason merely organizes information and means and cannot generate the final practical impulse. For Hume, the transition to morality is accomplished by a feeling that evaluates an action or character and possesses motivational force.

The Limit of This Solution

Hume’s solution explains how moral judgments arise psychologically, but not why they are valid. From the fact that people feel approval or disapproval, no standard follows for determining whether those feelings are correct. Feeling explains the origin of the judgment, but it does not justify its truth.

If feeling itself determines the validity of feeling, a circle results. If Hume’s theory is only meant to describe how people actually judge, the normative question remains open. We then know that people have moral feelings. We still do not know why those feelings should be trusted.

Feelings do not possess the status of an ultimate standard. They are themselves natural phenomena in need of explanation. Moral feelings are evolutionarily developed proximate regulatory mechanisms (John Tooby and Leda Cosmides 2008). They arose because, under certain conditions, they promoted behavior that served “lineage fitness.” This explains their motivational force, but also their fallibility. Under changed environmental conditions in particular, moral feelings can mislead.

Nikolaas Tinbergen (1951) showed, using supernormal stimuli, that an originally functional mechanism can be misdirected under changed conditions. The same applies to moral feelings. They may have been useful in certain environments and misleading in others. They must therefore be tested against a standard. That standard cannot itself be another feeling. It follows from the evolutionary function of those feelings.

Motivation and Truth

Hume’s account of motivation explains why certain moral judgments motivate action, but it does not explain why a judgment is true or valid. A statement can be true even if it motivates no one. An ought-statement can be objectively valid even if a particular subject feels no motivation to follow it. Motivation concerns the psychology of the agent. Validity concerns the statement’s relation to reality.

Reason and Motivation

Hume’s claim that reason cannot produce motivation is also untenable. Motivation arises from expectations and evaluations. Rational insight can change such expectations. If an argument changes someone’s understanding of the consequences of an action, it can thereby create a new readiness to act (Wolfram Schultz 2016; Kent Berridge 2009; Berridge & Robinson 2016; Salamone et al. 2018).

Happiness, life satisfaction, and eudaimonia belong to the biologically grounded feedback structures of action. They can motivate and stabilize behavior. No ultimate moral standard follows from this. Reason also remains embedded in the evolutionary structure. It allows moral feelings to be checked against knowledge of the world and assessments of consequences. It cannot generate motivation arbitrarily, but it can change expectations and thereby trigger or prevent actions.

Pure Reason

Attempts to derive moral standards from pure reason or from a priori principles can be found in different forms in Kant, Rawls, and Korsgaard (Kant 1785/2016; Rawls 1971/1999; Korsgaard 1996). These approaches fail as well. Purely logical and analytic inferences cannot expand the content of their premises (Schlick 1918/1925; Hahn 1933). If a derivation is to end with moral content, that content must therefore already be contained in its premises or concepts.

The choice of particular principles of reason does not solve this problem either. Whether one begins with consistency, universalizability, or autonomy, no objective moral standard follows from the principle itself. For any stipulated moral principle, competing principles can be formulated that are equally logically consistent yet lead to opposite conclusions. Pure reason therefore cannot determine which of them is valid. That requires examining which principles correspond to reality.

Ontology of Value Judgments

To solve the problem and identify where Hume and others went wrong, we must first analyze what values actually are. What makes something valuable to someone? Something becomes valuable to a person when, under given conditions, it helps that person achieve a goal, or achieve it better than available alternatives. Value therefore consists in a relation between what is evaluated, the entity for which it is valuable, a goal, and the relevant conditions. Mario Bunge formulates this relational structure precisely:

“a is valuable to organism b with respect to c under circumstances d, with goal e, and in light of a body of knowledge f.” (Bunge 1989; Bunge & Mahner 2004, p. 173)

A value judgment therefore asserts a real relation. Whether something is in fact valuable to someone with respect to a goal under particular conditions can be tested. A one-place concept of value such as “X is good,” by contrast, omits the relations that determine what the value of X consists in. Without specifying for whom, with respect to what, and under what conditions something is supposed to be valuable, no determinate state of affairs has been specified. Many ethical theories therefore reconstruct values incorrectly at the conceptual level by treating “good” or “valuable” as though it were a property that belongs to an object or action in itself.

This also allows Hume’s error to be identified more precisely. The motivational effect of a moral judgment does not determine its value. It describes what a judgment does psychologically to a person. The value judgment itself instead asserts a relation between something evaluated and a goal under particular conditions. Hume therefore conflates motivation with value.

The Ultimate Goal

“Nothing in biology makes sense except in the light of evolution.” (Dobzhansky 1973)

On this reconstruction, feeling and reason acquire their ethical status through their relation to the goal of living systems. Feelings provide proximate regulation. Reason tests, corrects, and can generate motivation. Both must be measured against the goal whose pursuit explains their evolutionary emergence.

Humans can set instrumental goals for themselves. They cannot freely choose an ultimate goal. Such a free choice would require libertarian free will capable of determining a goal independently of prior causes. A decision, however, is either caused by prior conditions or contains a random component. In the first case, the goal is not freely chosen. In the second, it is random (Schopenhauer 1839/1841, p. 25; Sapolsky 2023). A purely random choice could not give one goal objective priority over another either. If human goals are therefore not merely random, their origin and hierarchy must be reducible to a goal structure that was not freely chosen. A possible ultimate goal must therefore be reconstructable from empirical knowledge about the origin and development of life.

A distinction must therefore be made between teleology and teleonomy. Teleology explains goal-directedness through prior purposive setting. A goal exists because someone set it. Teleonomy, by contrast, denotes functional directedness that arises through natural processes without having been intentionally set beforehand. An organism can therefore be directed toward a goal without that goal having been established by a purposive agent (Pittendrigh 1958; Mayr 1974).

Since our goals, dispositions, and desires evolved just as our other biological traits did, an ultimate goal must also be explainable in evolutionary terms. Organisms whose heritable traits promoted the long-term persistence and reproduction of their genealogical lineage had a selective advantage. Corresponding goal structures could persist, while opposing ones disappeared together with their bearers (Darwin 1859; Dawkins 2016). I call this ultimate goal lineage fitness. It comprises the long-term persistence, reproduction, and adaptive success of one’s own genealogical lineage (Delor 2026).

Lineage fitness therefore satisfies the conditions of an ultimate goal. Living organisms are teleonomically directed toward it, while their other stable goals, desires, feelings, intuitions, and dispositions can be classified as proximate or instrumental mechanisms within this goal structure. Lineage fitness itself is not subordinate to any further goal and cannot be freely chosen by us. It therefore provides the objective standard by which other goals and value judgments can be assessed (Delor 2026).

This also allows the is–ought transition to be specified. If an action promotes lineage fitness under given conditions, then it ought to be performed under those conditions. The required ought-operator is not stipulated as a moral premise, but reconstructed from the relational structure of value and ought. The ultimate goal itself likewise comes from an empirical reconstruction of the goal structure of living systems. Facts about a goal and facts about the means that promote that goal can therefore yield an objective ought-statement (Delor 2026).

Conclusion

In conclusion, David Hume’s demand for an explanation of the transition from is to ought was entirely justified. His own proposal, however, does not solve the problem. Hume ultimately derives the normative significance of moral judgments from feelings and their motivational effects, thereby himself making a transition from facts to an ought. Moreover, those feelings themselves remain in need of explanation. There is also a category mistake. Hume assumes that a moral judgment must motivate, although motivation is a psychological effect and not a component of a value judgment.

The weak form of the is–ought problem remains unaffected. No ought follows by pure logic from purely descriptive statements if the required ought-operator is absent from the system. But no general barrier to a naturalistic ethics follows from this. If the ought-operator itself is reconstructed from real goal and promotion relations, and if the ultimate goal is likewise derived from facts about living systems, then no moral premise has been freely stipulated. The normative operator itself belongs to the theoretical structure reconstructed from facts. Ought-statements can therefore follow from facts about the ultimate goal and about which actions promote that goal under particular conditions.

Later, further barriers to naturalistic ethics were developed. These include, in particular, G. E. Moore’s (1903) so-called naturalistic fallacy. More recently, evolutionary debunking arguments have attempted to derive objections to the objective validity of our moral judgments from their evolutionary origin. These arguments also fail under closer examination. In my paper The Solution to the Is-Ought Problem based on Critical Rationalism and Evolutionary Theory, I reconstruct the solution outlined here in greater detail and address these and further objections to a naturalistic solution to the is–ought problem.

References

Berridge, K. C., Robinson, T. E., & Aldridge, J. W. (2009). Dissecting components of reward: ‘Liking’, ‘wanting’, and learning. Current Opinion in Pharmacology, 9(1), 65–73. https://doi.org/10.1016/j.coph.2008.12.014

Berridge, K. C., & Robinson, T. E. (2016). Liking, wanting, and the incentive-sensitization theory of addiction. American Psychologist, 71(8), 670–679. https://doi.org/10.1037/amp0000059

Bunge, M. (1989). Ethics: The Good and the Right (Vol. 8). D. Reidel Publishing Company. https://doi.org/10.1007/978-94-009-2601-1

Bunge, M., & Mahner, M. (2004). Über die Natur der Dinge. S. Hirzel Verlag GmbH.

Darwin, C. (1859). On the Origin of Species by Means of Natural Selection, or the Preservation of Favoured Races in the Struggle for Life. John Murray.

Dawkins, R. (2016). The Extended Selfish Gene. Oxford University Press. (Original work published 1976)

Delor, F. (2026). The Solution to the Is-Ought Problem based on Critical Rationalism and Evolutionary Theory. Preprint. https://philpapers.org/rec/DELTST2

Dobzhansky, T. (1973). Nothing in biology makes sense except in the light of evolution. The American Biology Teacher, 35(3), 125–129. https://doi.org/10.2307/4444260

Hahn, H. (1933). Logik, Mathematik und Naturerkennen. Verlag Gerold.

Hume, D. (1978). A Treatise of Human Nature (L. A. Selby-Bigge & P. H. Nidditch, Eds.). Oxford University Press. (Original work published 1739)

Kant, I. (2016). Grundlegung zur Metaphysik der Sitten (D. Schönecker & B. Kraft, Eds.; 2nd ed.). Felix Meiner Verlag. https://doi.org/10.28937/978-3-7873-2878-9 (Original work published 1785)

Korsgaard, C. M. (1996). The Sources of Normativity. Cambridge University Press. https://doi.org/10.1017/CBO9780511554476

Mayr, E. (1974). Teleological and teleonomic, a new analysis. In R. S. Cohen & M. W. Wartofsky (Eds.), Methodological and Historical Essays in the Natural and Social Sciences (Vol. 14, pp. 91–117). Springer. https://doi.org/10.1007/978-94-010-2128-9_6

Moore, G. E. (1903). Principia Ethica. Cambridge University Press.

Pigden, C. (2010). On the triviality of Hume’s Law: A reply to Gerhard Schurz. In C. Pigden (Ed.), Hume on Is and Ought (pp. 217–238). Palgrave Macmillan. https://doi.org/10.1007/978-0-230-29688-6_19

Pittendrigh, C. S. (1958). Adaptation, natural selection, and behavior. In A. Roe & G. G. Simpson (Eds.), Behavior and Evolution (pp. 390–416). Yale University Press.

Rawls, J. (1999). A Theory of Justice (rev. ed.). Harvard University Press. (Original work published 1971)

Salamone, J. D., Correa, M., Yang, J.-H., Rotolo, R., & Presby, R. (2018). Dopamine, effort-based choice, and behavioral economics: Basic and translational research. Frontiers in Behavioral Neuroscience, 12, 52. https://doi.org/10.3389/fnbeh.2018.00052

Sapolsky, R. M. (2023). Determined: A Science of Life Without Free Will. Penguin Press.

Schlick, M. (1925). Allgemeine Erkenntnislehre (2nd ed.). Verlag von Julius Springer. https://doi.org/10.1007/978-3-642-50929-2 (Original work published 1918)

Schopenhauer, A. (1841). Über die Freiheit des menschlichen Willens. In Die beiden Grundprobleme der Ethik: Behandelt in zwei akademischen Preisschriften. Joh. Christ. Hermann’sche Buchhandlung. (Original work published 1839)

Schultz, W. (2016). Dopamine reward prediction-error signalling: A two-component response. Nature Reviews Neuroscience, 17, 183–195. https://doi.org/10.1038/nrn.2015.26

Tinbergen, N. (1951). The Study of Instinct. Clarendon Press.

Tooby, J., & Cosmides, L. (2008). The evolutionary psychology of the emotions and their relationship to internal regulatory variables. In M. Lewis, J. M. Haviland-Jones, & L. F. Barrett (Eds.), Handbook of Emotions (3rd ed., pp. 114–137). Guilford Press.