Moral disagreement can persist even when two people agree about what happened. One notices suffering. Another notices unequal treatment. Someone else sees betrayal, disrespect for a legitimate role, coercion, or the degradation of something held sacred. Facts and interests matter, but moral conflict can also begin with different features of an event acquiring moral weight.

Moral Foundations Theory (MFT) offers a psychological framework for studying this variation. It presents a pluralist account of the moral mind: people may share recurring moral sensitivities while learning to interpret and prioritize them differently. The framework has widened moral psychology beyond harm and justice, but its proposed foundations, causal claims, and questionnaires remain open to revision (Graham et al., 2011, 2013).

What Is Moral Foundations Theory

Moral Foundations Theory asks how people can share broad moral concerns while constructing different moral worlds. It is a descriptive psychological theory. It does not decide which belief is philosophically correct. It asks which kinds of social information people readily moralize, how those reactions develop, and why their relative importance varies (Graham et al., 2013).

Jonathan Haidt and Craig Joseph first described an intuitive ethics in which prepared sensitivities make some social patterns especially easy to evaluate morally. Jesse Graham, Haidt, and their colleagues later developed that proposal into Moral Foundations Theory and designed questionnaires for studying individual and cultural variation (Haidt & Joseph, 2004; Graham et al., 2011, 2013).

The word foundation can sound more rigid than the theory intends. A foundation is a proposed psychological system that directs attention toward a recurring social problem. It is not a commandment, a political identity, or a permanent trait. Cultures can build different virtues and institutions from similar sensitivities, and experience can expand, restrain, or reinterpret an initial response (Graham et al., 2013).

Four Central Claims of the Theory

Moral Intuition and Prepared Foundations

MFT proposes a prepared first draft of the moral mind. Prepared does not mean that infants possess six finished moral rules, or that genes mechanically determine adult judgment. It means that development may be organized before experience so that recurring patterns such as suffering, reciprocity, coalition, rank, coercion, and contamination become especially learnable and emotionally charged (Graham et al., 2013).

How Culture Shapes the Moral Mind

Families, religious communities, schools, laws, occupations, and peer groups teach people what counts as loyalty, which forms of authority deserve respect, which inequalities are acceptable, and what is pure or degrading. A small set of prepared sensitivities can therefore support many culturally variable virtues rather than one universal code (Haidt & Joseph, 2004).

Culture is part of the developmental process, not a decorative layer added after biology. It helps define which situations activate a concern, what the response means, whose interests count, and which practices reinforce it. Cooley’s earlier account of moral standards as socially transmitted group ideals provides useful historical context, although it does not establish MFT’s evolutionary claims (Cooley, 1922).

Intuition and Moral Reasoning

The social intuitionist account proposes that moral judgment often begins with an intuitive evaluation. Conscious reasoning may then search for an explanation that supports the judgment already felt. Haidt described this as a common pathway and a model to test, not an absolute law (Haidt, 2001).

Moral reasoning still matters. Reflection can expose conflicting intuitions, revise a judgment, and influence other people through conversation. The model therefore concerns the frequent sequence and social setting of moral cognition; it does not reduce every reason to a rationalization (Haidt, 2001).

Moral Pluralism

MFT rejects the assumption that every moral judgment reduces to one concern. Harm may dominate one dispute, while obligation, hierarchy, group commitment, sacredness, equality, deservingness, or liberty becomes central in another. Pluralism does not mean that every concern is equally important to every person or culture. It proposes more than one recurring source of moral evaluation (Graham et al., 2013).

The Six Moral Foundations

The current six-foundation model is best read as a research map. Each foundation names a family of concerns and likely triggers. It does not describe a personality type, a diagnosis, or a rule a person must obey (Atari et al., 2023).

The labels compress broad and culturally variable material. Authority concerns role and order, including questions about obedience to authority, but does not make every command legitimate. Purity can involve sacredness, bodily practices, contamination, or degradation; it is broader than cleanliness. Equality and proportionality are both fairness concerns, yet one emphasizes equal standing or outcomes while the other emphasizes contribution and deservingness (Atari et al., 2023).

The Original Five Foundation Model

Classic MFT research used care or harm, fairness or reciprocity, ingroup loyalty, authority or respect, and purity or sanctity. The original Moral Foundations Questionnaire (MFQ) was designed around these five dimensions. In the authors’ large self-selected online sample, responses broadly followed this five-part pattern, although later research found important structural and cross-cultural problems (Graham et al., 2011).

The Revised Six Foundation Model

The Moral Foundations Questionnaire 2 separates the former fairness foundation into equality and proportionality. Its developers tested the 36-item measure across 25 populations and seven languages and reported evidence that the six scores could be compared meaningfully within those samples (Atari et al., 2023).

The revision shows how theory and measurement can change together. Earlier MFT work treated equality as largely covered by fairness or liberty. Later studies of the questionnaire across cultures distinguished equality from proportionality. The six-foundation model is more current, but it is not necessarily the last possible map (Graham et al., 2013; Atari et al., 2023).

Individualizing and Binding Foundations

Much early MFT research grouped care and fairness as individualizing foundations because they protect individuals from harm and unfair treatment. Loyalty, authority, and purity were grouped as binding foundations because they help organize people into larger communities (Graham et al., 2009).

The distinction is a useful summary of older results, not a second foundation system. It can hide differences within each group and does not map perfectly onto the revised six-foundation model. Equality and proportionality, for example, can support individual claims, group arrangements, or institutional rules depending on the context (Graham et al., 2013; Atari et al., 2023).

Liberty and Other Candidate Foundations

Liberty is often described as a sixth moral foundation in popular accounts of MFT, but that wording now creates confusion. The current MFQ-2 has six measured foundations because it divides fairness into equality and proportionality. Liberty or oppression remains a candidate foundation discussed in the broader theory rather than one of the six MFQ-2 scales (Graham et al., 2013; Atari et al., 2023).

Research on self-identified libertarians highlighted liberty as a concern not fully captured by the original five-foundation model. In a large volunteer web sample, libertarian participants placed especially strong emphasis on individual liberty, while their profiles were not captured well by a simple liberal-conservative line (Iyer et al., 2012). The study relied mainly on self-report and a nonrepresentative, politically engaged sample, so its group averages should not be treated as a timeless portrait of all libertarians.

MFT’s developers have also discussed efficiency or waste, ownership or theft, and other possibilities. A concern becomes a foundation only if it meets demanding conceptual and empirical criteria. Candidate status signals an active research question, not a completed addition to the model (Graham et al., 2013).

Why Moral Disagreement Becomes Difficult

Moral disagreement becomes especially difficult when people argue from different starting concerns without recognizing the difference. A statement intended as compassion may sound disloyal. A defense of order may sound indifferent to suffering. An appeal to equal outcomes may be heard as disregard for effort, while an appeal to earned reward may be heard as permission to preserve inequality.

Moral convictions also become attached to social identities, relationships, and institutions. A challenge can then feel larger than a disputed conclusion. It may threaten a person’s sense of who deserves protection, which obligations hold a community together, or what kind of person one is trying to be (Aquino & Reed, 2002; Rai & Fiske, 2011).

One Conflict Seen Through Several Moral Lenses

Imagine that an employee discovers that an organization concealed evidence of a product defect. Care directs attention to people who could be injured. Equality may emphasize whether less powerful customers bear risks that insiders avoid. Proportionality may focus on a company receiving benefits while shifting costs to others.

Loyalty can operate in more than one direction: silence may be framed as commitment to colleagues, while disclosure may be framed as loyalty to the organization’s stated mission or to the public it serves. Authority may highlight reporting duties and legitimate oversight, or conflict with a superior’s demand for secrecy. Purity may appear in language about corruption and institutional degradation. Liberty may focus on coercion or the freedom to speak.

The example does not determine the correct action, and MFT does not supply an ethical verdict. It shows why listing foundations is less useful than tracing how concerns, targets, and interpretations interact within one conflict.

Moral Reframing and Moral Translation

Moral reframing expresses a position through values that matter to the intended audience. Research reviewed by Matthew Feinberg and Robb Willer found that such messages sometimes increased support across political divides when advocates moved beyond the moral language most natural to their own side (Feinberg & Willer, 2019).

The finding supports moral translation, not a guarantee of persuasion. Effects depend on the issue, audience, message, and context, and important moderators remain unresolved (Feinberg & Willer, 2019). A responsible reframing preserves the facts and the speaker’s actual reasons while making a genuine shared concern visible. It should not become a technique for disguising evidence or exploiting a group stereotype.

Moral reframing does not determine whether an argument is ethical. Feinberg and Willer (2019) note that moral language has also served destructive political projects. The ethical quality of a reframed argument therefore depends on its truthfulness, its aims, and the consequences it asks people to accept.

Moral Foundations and Political Differences

MFT became widely known through studies of political differences. In classic United States samples, self-identified liberals placed greater relative emphasis on care and fairness, whereas conservatives distributed concern more evenly across care, fairness, loyalty, authority, and purity (Graham et al., 2009). These were average associations within particular samples, not fixed moral profiles for every person.

The libertarian findings add another caution. Participants who selected that label strongly emphasized liberty and did not fit neatly into the same profile as either liberals or conservatives (Iyer et al., 2012). Ideology can organize moral language, but a label does not reveal how any individual will interpret a concrete dilemma.

Group differences also appear outside party identification. A 2026 study of United States management researchers and practicing managers found higher average scores among managers on equality, loyalty, authority, and purity. Average care scores did not differ meaningfully, and proportionality differed only slightly (Telkamp & Anderson, 2026). The result suggests that professional communities can develop different moral emphases. It does not establish that occupation caused the differences or that one group was more moral.

The Moral Foundations Questionnaire and Its Limits

The Moral Foundations Questionnaire asks people how relevant certain considerations are to moral judgment and how strongly they agree with moral statements. Scores summarize how strongly respondents endorse the measured concerns. They do not directly show that each score reflects an evolved mental system, predict conduct in every situation, or establish that respondents understood an item in the same way (Graham et al., 2011; Atari et al., 2023).

Researchers often test the theory through questionnaires, so the quality of those questionnaires matters. In a study spanning 27 countries, the short MFQ did not capture the same five-part pattern consistently. The newer MFQ-2 was developed with broader cross-cultural input and showed stronger support for six dimensions within the populations studied (Iurino & Saucier, 2020; Atari et al., 2023).

Even a well-fitting questionnaire remains one form of evidence. Behavioral choices, developmental observations, language, institutions, and moral responses in relationships may reveal aspects of morality that self-report items miss. Researchers should match the strength of a claim to the kind of evidence collected.

Limits and Competing Accounts

Evidence Across Cultures and Its Limits

MFT proposes that some moral concerns recur across societies, but proving that they are universal is difficult. A questionnaire may work differently across cultures because translations shift meaning, people use response scales differently, or the questions do not reflect how a community understands morality. The short MFQ-1 did not produce a consistent pattern across 27 countries. That result limits broad claims based on the questionnaire, but it does not show that the underlying moral concerns are absent (Iurino & Saucier, 2020).

MFQ-2 improved the cross-cultural design, yet its 25 populations and seven languages do not represent all human societies. The authors note that their participants were generally online and educated and that the project did not include traditional small-scale communities (Atari et al., 2023).

How Question Format Changes Results

Wormley and colleagues analyzed five samples totaling 464,229 respondents. They found that answers were influenced partly by how the questions were asked: participants either rated how relevant a concern was or indicated whether they agreed with a moral statement. When the researchers accounted for this difference, some reported links between foundation scores and other measures became weaker, stronger, or pointed in a different direction (Wormley et al., 2025).

This matters because mistaking a question format for a moral foundation can make relationships with ideology, cognition, or behavior appear stronger or different than they are. Problems with MFQ-1 do not automatically make the broader theory false, but they narrow the conclusions its original scores can support (Wormley et al., 2025).

Are Foundations Causes or Descriptions

A questionnaire can show that certain concerns tend to appear together and relate to other responses. It cannot by itself prove that each pattern comes from a separate evolved mental system that causes moral judgment. That stronger claim requires evidence from development, behavior, culture, biology, and evolution (Graham et al., 2013).

MFT’s developers proposed criteria for adding or revising foundations. They also acknowledged that parts of the theory remain underspecified and should be compared with competing models. Readers should therefore distinguish a useful descriptive framework from confirmed evidence about the mental mechanisms behind it (Graham et al., 2013).

Alternative Maps of Morality

Some theories dispute the need for several independent foundations. Dyadic morality proposes a more unified template: an intentional moral agent and a suffering moral patient. From this perspective, even apparently harmless purity violations may be interpreted through perceived agency and harm (Gray et al., 2012).

Relationship regulation theory organizes morality around four social motives: unity, hierarchy, equality, and proportionality. Its central question is not only what kind of act occurred, but what relationship people believe they are regulating and which obligations that relationship creates (Rai & Fiske, 2011).

These accounts overlap with MFT in places but organize the moral domain differently. MFT highlights plural intuitive concerns, dyadic morality looks for an agent-patient template, and relationship regulation begins with social forms and their obligations. The competition is productive because each theory directs researchers toward different evidence and different failure points.

Moral Concern From Salience to Conduct

MFT explains which kinds of concern become morally salient. That question is important, but it does not fully explain who receives moral consideration, whether morality is central to the self, or whether a person acts consistently with a standard. Several neighboring theories help separate these problems.

Moral expansiveness concerns the breadth of entities treated as worthy of moral concern. A person may care intensely about loyalty or protection while drawing the relevant boundary narrowly or broadly. Across six studies, the Moral Expansiveness Scale was related to, but not reducible to, moral foundations, moral identity, and other concern-related measures (Crimston et al., 2016).

Moral identity concerns how central moral traits are to a person’s self-concept and how that identity is expressed. Aquino and Reed identified internalization and symbolization as related dimensions and found associations with moral cognition and behavior across six studies (Aquino & Reed, 2002). The relationship between values and behavior is not automatic; knowing what someone values does not tell us how strongly being moral organizes that person’s sense of self.

Moral disengagement addresses another gap. Bandura described mechanisms that allow people to suspend the self-sanctions that ordinarily keep conduct aligned with moral standards. Harmful action can be recast as worthy, responsibility can be displaced or diffused, consequences can be minimized, and victims can be blamed or dehumanized (Bandura, 2002).

Taken together, these constructs distinguish four questions: What kind of concern is active? Who or what falls within its moral boundary? How central is morality to the self? What permits conduct to follow or depart from the standard? This four-question distinction is an editorial synthesis, not a validated unified model. Its value is to prevent a foundation score from being asked to explain more than it measures.

Using Moral Foundations Theory Carefully

When people agree about the facts but disagree about what is morally decisive, MFT offers a disciplined way to ask which concern is active and what makes it salient. It also directs attention to how culture shapes a concern’s meaning and who or what falls within its reach. This can widen inquiry when a dispute has been reduced too quickly to ignorance, selfishness, or a single ideological line.

The framework becomes misleading when six scores are treated as fixed types of people, complete explanations of moral behavior, or measures of moral worth. The original questionnaire has serious measurement and cross-cultural limits. The revised measure samples only part of human diversity, and competing theories organize the moral domain differently (Iurino & Saucier, 2020; Atari et al., 2023; Wormley et al., 2025).

A moral map earns its value by helping us notice terrain we would otherwise miss. MFT remains credible when used as an evolving framework for inquiry into obligation, violation, and the good, alongside evidence about identity, relationships, moral boundaries, and conduct.

References

Aquino, Karl; Reed, Americus II. (2002). The Self-Importance of Moral Identity. Journal of Personality and Social Psychology, 83(6), 1423–1440. DOI: 10.1037/0022-3514.83.6.1423.
(Return to Main Text)

Atari, Mohammad; Haidt, Jonathan; Graham, Jesse; Koleva, Sena; Stevens, Sean T.; Dehghani, Morteza. (2023). Morality Beyond the WEIRD: How the Nomological Network of Morality Varies Across Cultures. Journal of Personality and Social Psychology, 125(5), 1157–1188. DOI: 10.1037/pspp0000470.
(Return to Main Text)

Bandura, Albert. (2002). Selective Moral Disengagement in the Exercise of Moral Agency. Journal of Moral Education, 31(2), 101–119. DOI: 10.1080/0305724022014322.
(Return to Main Text)

Cooley, Charles Horton. (1922). Human Nature and the Social Order (Students’ edition). Charles Scribner’s Sons.
(Return to Main Text)

Crimston, Charlie R.; Bain, Paul G.; Hornsey, Matthew J.; Bastian, Brock. (2016). Moral Expansiveness: Examining Variability in the Extension of the Moral World. Journal of Personality and Social Psychology, 111(4), 636–653. DOI: 10.1037/pspp0000086.
(Return to Main Text)

Feinberg, Matthew; Willer, Robb. (2019). Moral Reframing: A Technique for Effective and Persuasive Communication Across Political Divides. Social and Personality Psychology Compass, 13(12), e12501. DOI: 10.1111/spc3.12501.
(Return to Main Text)

Graham, Jesse; Haidt, Jonathan; Koleva, Sena; Motyl, Matt; Iyer, Ravi; Wojcik, Sean P.; Ditto, Peter H. (2013). Moral Foundations Theory: The Pragmatic Validity of Moral Pluralism. In Patricia Devine and Ashby Plant (Eds.), Advances in Experimental Social Psychology (Vol. 47, pp. 55–130). Academic Press. DOI: 10.1016/B978-0-12-407236-7.00002-4; ISBN: 978-0-12-407236-7.
(Return to Main Text)

Graham, Jesse; Haidt, Jonathan; Nosek, Brian A. (2009). Liberals and Conservatives Rely on Different Sets of Moral Foundations. Journal of Personality and Social Psychology, 96(5), 1029–1046. DOI: 10.1037/a0015141.
(Return to Main Text)

Graham, Jesse; Nosek, Brian A.; Haidt, Jonathan; Iyer, Ravi; Koleva, Spassena; Ditto, Peter H. (2011). Mapping the Moral Domain. Journal of Personality and Social Psychology, 101(2), 366–385. DOI: 10.1037/a0021847.
(Return to Main Text)

Gray, Kurt; Young, Liane; Waytz, Adam. (2012). Mind Perception Is the Essence of Morality. Psychological Inquiry, 23(2), 101–124. DOI: 10.1080/1047840X.2012.651387.
(Return to Main Text)

Haidt, Jonathan. (2001). The Emotional Dog and Its Rational Tail: A Social Intuitionist Approach to Moral Judgment. Psychological Review, 108(4), 814–834. DOI: 10.1037/0033-295X.108.4.814.
(Return to Main Text)

Haidt, Jonathan; Joseph, Craig. (2004). Intuitive Ethics: How Innately Prepared Intuitions Generate Culturally Variable Virtues. Daedalus, 133(4), 55–66. DOI: 10.1162/0011526042365555.
(Return to Main Text)

Iurino, Kathryn; Saucier, Gerard. (2020). Testing Measurement Invariance of the Moral Foundations Questionnaire Across 27 Countries. Assessment, 27(2), 365–372. DOI: 10.1177/1073191118817916.
(Return to Main Text)

Iyer, Ravi; Koleva, Spassena; Graham, Jesse; Ditto, Peter H.; Haidt, Jonathan. (2012). Understanding Libertarian Morality: The Psychological Dispositions of Self-Identified Libertarians. PLOS ONE, 7(8), e42366. DOI: 10.1371/journal.pone.0042366.
(Return to Main Text)

Rai, Tage Shakti; Fiske, Alan Page. (2011). Moral Psychology Is Relationship Regulation: Moral Motives for Unity, Hierarchy, Equality, and Proportionality. Psychological Review, 118(1), 57–75. DOI: 10.1037/a0021867.
(Return to Main Text)

Telkamp, Jake B.; Anderson, Marc H. (2026). Value and Moral Divergences Between Management Researchers and Managers: Evidence and a Future Research Agenda. Applied Psychology, 75(4), e70120. DOI: 10.1111/apps.70120.
(Return to Main Text)

Wormley, Alexandra S.; Scott, Matthew; Grimm, Kevin J.; Cohen, Adam B. (2025). Measuring Morality: An Examination of the Moral Foundation Questionnaire’s Factor Structure. Personality and Social Psychology Bulletin, 51(2), 314–328. DOI: 10.1177/01461672231191362.
(Return to Main Text)