The Righteous Mind Values

Moral Foundations Theory — How It Actually Works in Practice

Most people who come across Jonathan Haidt's framework from The Righteous Mind get stuck on the labels. They memorize the six foundations and then try to shoehorn every political argument into that grid. That approach breaks down fast because human morality doesn't fit neatly into tidy boxes. The framework is a diagnostic tool, not a classification system.

Understanding The Righteous Mind Values in Real Conversations

The six foundations Haidt identified are Care/Harm, Fairness/Cheating, Loyalty/Betrayal, Authority/Subversion, Sanctity/Degradation, and Liberty/Oppression. Most introductory material presents them as equal pillars. They are not equal. The first three carry most of the weight in typical Western discourse. The last three—Loyalty, Authority, Sanctity—are what Haidt calls the binding foundations, and they matter enormously in specific contexts even if they get ignored in general conversation. I spent years working on organizational culture assessments for tech companies. We ran conflict mediation sessions where two teams would literally cannot work together but neither could articulate why. One team would describe the other as "disrespectful." The other would call them "rigid." Under The Righteous Mind Values lens, you were looking at an Authority/Subversion clash versus a Liberty/Oppression trigger. Once we named it, the argument lost its emotional grip almost immediately. People could negotiate around the value difference instead of attacking character. Here is the part nobody tells you about applying this framework. The foundations activate in different combinations depending on context. A single policy decision can simultaneously trigger Fairness and Liberty concerns in one person while triggering Loyalty and Sanctity concerns in another. That means you cannot predict someone's moral response just by knowing their political affiliation. Individual variation matters more than group averages.

When I first tried using this to analyze our company's remote work policy debate in 2021, I hit a wall. The framework predicted that supporters of returning to office would emphasize Authority and Loyalty. They did. But it also predicted that opponents would emphasize Care and Fairness. They partially did, but a significant chunk was driven purely by Sanctity—people felt the office environment had become morally contaminated through forced interaction during a pandemic. That Sanctity response did not show up cleanly in any standard political mapping. It took three additional interview rounds to surface that particular activation pattern.

How to Apply This Framework Without Making It Worse

The biggest mistake people make is using the six foundations as a weapon. You will see this constantly in online discussions where someone identifies which foundation their opponent is appealing to and then dismisses that appeal as illegitimate. Calling someone's Sanctity concern "primitive" or their Loyalty concern "tribal" is not analysis. It is just replacing one moral foundation with another while pretending the first one is neutral. A more useful approach is mapping. When you encounter a moral disagreement, write down which foundation each side is drawing from. Do this before you form a response. The exercise takes about five minutes and it changes your reaction from defensive to analytical. You stop hearing "this person is wrong" and start hearing "this person is weighing a different foundation more heavily." That shift does not resolve the disagreement. It makes the disagreement legible. There is a practical bottleneck with this method that researchers have not fully addressed. The foundations are not independent. They interact. Loyalty amplifies Authority. Sanctity amplifies Care in certain cultural contexts. When you treat them as separate switches you can flip on and off, you lose predictive accuracy. A better model treats them as a weighted portfolio where each person carries a different distribution. Two people can score identically on Care but deploy it completely differently because their secondary weights are different.

I found that pairing the framework with a simple survey instrument helped. Instead of asking people whether they agree with abstract moral statements, ask them to rank which considerations matter most when they evaluate a specific scenario. Scenario-based responses are more reliable than trait-based self-reports. People lie about their traits. They rarely lie about what they actually did in a described situation.

Where the Framework Fails

The Righteous Mind Values model was built primarily from American political data. It does not generalize well to collectivist cultures where individual rights and personal autonomy are not the default moral frame. In those contexts, Loyalty and Sanctity are not secondary foundations. They are primary. The framework still applies, but the weighting is inverted and anyone using it without that adjustment will misread the moral landscape entirely. There is also the issue of measurement drift. The original Moral Foundations Questionnaire has been revised multiple times. The MFQ-30, MFQ-33, and the newer IMFG all produce slightly different factor structures. If you are comparing studies that used different versions, you are not comparing the same construct. This matters when you are doing longitudinal analysis or meta-reviews. The framework also struggles with moral innovations. New movements create new moral concerns that do not map cleanly onto existing foundations. Environmental ethics, for example, contains elements of Sanctity but also introduces obligations to non-human entities that the current model does not account for. Animal rights arguments trigger Care but also a form of Fairness that extends beyond human reciprocity. The model requires updates to stay useful, and those updates have been slow.

If you need a quick reference for the six foundations and the original questionnaire, the research is openly available through Haidt's lab at NYU. The academic papers are freely accessible through Google Scholar. The framework itself is not proprietary. What is proprietary are the commercial scoring tools built on top of it. Those cost money and are not necessary to understand or apply the core concepts. The most reliable way to use this framework is to treat it as a vocabulary, not a verdict. It gives you words for moral disagreements that usually remain wordless. That is its actual utility. Anything beyond that tends to turn into intellectual posturing.