nucleation

LIT-500: Morality beyond the WEIRD: How the nomological network of morality varies across cultures

2 min read

Properties5
StatusActive — read in full 2026-10-02 (NOTE-398), from the APA accepted manuscript; worth reading as the current Moral Foundations measure (the MFQ-2) and the theory's current list of six foundations, with Fairness split into Equality and Proportionality, and as the place MFT answers morality-as-cooperation directly: Curry et al. 2019 "left out Purity as an important and less-WEIRD moral foundation". Its strongest evidence for that, Purity's association with cultural distance from the U.S., rests on the one subscale that fails its own measurement-invariance threshold. Its samples are online panels in 25 populations, not small-scale societies.
Tagsmoral-psychology, social-science, society-and-governance, religion, psychometrics
ExtendsLIT-499 — Moral Foundations Theory: The Pragmatic Validity of Moral Pluralism
Cited as “paper” byNOTE-398 — Atari et al. 2023, the MFQ-2
Cited as “source” byTHEORY-435 — In Moral Foundations Questionnaire responses, Care and Fairness are endorsed across the political spectrum, while Loyalty, Authority and Sanctity rise toward the conservative end, in every world region sampled, THEORY-539 — Moral judgment draws on several innately prepared intuitive foundations, at least Care, Fairness, Loyalty, Authority and Sanctity, each a response to a different adaptive challenge, and Sanctity descends from pathogen avoidance rather than from cooperation, THEORY-613 — The Moral Foundations Questionnaire-2's six factors, Care, Equality, Proportionality, Loyalty, Authority and Purity, fit across 19 nations, Equality and Proportionality are nearly uncorrelated in WEIRD nations, and the US split between individualizing and binding foundations appears in only three national networks

Graph View

Mohammad Atari, Jonathan Haidt, Jesse Graham, Sena Koleva, Sean T. Stevens, Morteza Dehghani (2023), Journal of Personality and Social Psychology 125(5), 1157–1188; preprint PsyArXiv q6c9r (2022) — DOI-10.1037/pspp0000470

Standing in the record

Filed on 2026-10-02 in the morality-as-cooperation batch (ADR-018). It is MFT’s current statement and instrument, and it extends Graham et al.’s 2013 chapter, LIT-499, in the way that chapter said it might: one “continent” turned out to be two, so Fairness becomes Equality and Proportionality. The split is taken from Rai and Fiske’s relationship regulation, LIT-507, which the paper credits. No anthology topic holds it. It carries no instruction for machine-learning practice; the MFQ-2 is widely used to probe language models, but that use is in other papers, not this one.

Its bearing on morality-as-cooperation, set out in NOTE-398:

  • It names the dispute. Curry et al. 2019 (LIT-047, read in NOTE-034) “left out Purity as an important and less-WEIRD moral foundation” (p. 64). The evidence offered is that Purity is the foundation most strongly associated with cultural distance from the U.S. and with religiosity.
  • That evidence is weaker than the sentence. Purity is the only one of the six subscales whose item intercepts fail the paper’s own invariance threshold (39.5% non-invariant against 25%), and the authors themselves say Purity means should be compared across groups “with caution” (p. 30).
  • The MFQ-2 still has no item about returning a favour. The paper defines Equality as including “balanced reciprocity” (p. 60), but all six Equality items are about equal income or resources (Appendix). The measurement half of LIT-047’s charge that MFT conflates reciprocity with fairness therefore stands for the new instrument too.
  • Its six foundations and MAC’s seven still do not nest. The MFQ-2 adds Proportionality, which MAC’s “fairness” (dividing disputed resources) only partly covers. It has no bravery or property scale: ownership is among the candidates the paper says it cannot yet measure (p. 64), and bravery is not mentioned.

NOTE-398 is the close reading of 2026-10-02, and it placed the work: Active. It is the instrument any new MFT result will use, and its cross-national design is a real advance on the MFQ-1’s English-language internet samples. Its populations are still online, educated and mostly urban, as the limitations section says.

Backlinks