Fitouchi et al. argue that puritanical morality arises from moralization of self-control failures, which are seen to characterize undesirable future cooperation partners. We appreciate the authors' comparisons and contrasts between their moral disciplining theory (MDT) and moral foundations theory (MFT; Graham et al., Reference Graham, Haidt, Koleva, Motyl, Iyer, Wojcik and Ditto2013; Haidt & Joseph, Reference Haidt and Joseph2004), and we grant that self-control is central to puritanical morality and does not correspond neatly to any one moral foundation. We also appreciate the authors' historical psychological approach (Atari & Henrich, Reference Atari and Henrich2022; Muthukrishna, Henrich, & Slingerland, Reference Muthukrishna, Henrich and Slingerland2021), which helps theorists better situate cross-temporal changes in puritanical morality, rather than examining just one era. We also agree that it is crucial to study puritanism and purity, as it is the least WEIRD (western, educated, industrialized, rich, and democratic; Henrich, Heine, & Norenzayan, Reference Henrich, Heine and Norenzayan2010) aspect of moral cognition (Atari et al., Reference Atari, Haidt, Graham, Koleva, Stevens and Dehghani2022a) and remains understudied compared with other moral concerns that may be more salient in WEIRD contexts.
Although we find Fitouchi et al.'s theory of puritanical morality compelling (though incomplete, as we'll show), we believe their account of MFT has missed some important areas of agreement. First, MFT agrees with Fitouchi et al. that moral systems, as cultural products, are best understood as aiming toward cooperation. Haidt (Reference Haidt2012, p. 66) defined “moral systems” as “interlocking sets of values, virtues, norms, practices, identities, institutions, technologies, and evolved psychological mechanisms that work together to suppress or regulate self-interest and make cooperative societies possible.” But where Fitouchi et al. suppose that moral cognition is based on a single computational device that computes fairness (or is it harm?) to infer a person's cooperation potential, MFT argues that when humans developed widely shared and socially enforceable understandings of how things “ought” to be, they drew on multiple pre-existing cognitive systems, including the attachment system, reciprocal altruism, coalitional psychology, status hierarchies, and the behavioral immune system. MFT also posits that cultural learning can regulate these evolutionarily prepared psychological mechanisms (Graham et al., Reference Graham, Haidt, Koleva, Motyl, Iyer, Wojcik and Ditto2013), which can produce cross-temporal and cross-regional differences in the endorsement of moral foundations. Do Fitouchi et al. and other monists believe that these evolved psychological mechanisms did not get recruited into morality because the harm system (or is it the fairness system?) was sufficient for cooperation in our evolutionary history?
We do say, as Fitouchi et al. note, that the last item in that list has an evolutionary origin different from the others. The first four foundations all grew out of interpersonal dynamics in ancient primate and mammalian societies, whereas purity cognitions have their origin in a physical world full of pathogens (Atari et al., Reference Atari, Reimer, Graham, Hoover, Kennedy, Davani and Dehghani2022b). Yet MFT distinguishes between the evolutionary origins (from which we infer the “original trigger” of an intuition) and the current function (which includes the current triggers). We say, in numerous places, that purity now functions to enhance group binding, for example, “if you think, as I do, that one of the greatest unsolved mysteries is how people ever came together to form large cooperative societies, then you might take a special interest in the psychology of sacredness…Whatever its origins, the psychology of sacredness helps bind individuals into moral communities” (Haidt, Reference Haidt2012, p. 149). Consistent with this idea, Dehghani et al. (Reference Dehghani, Johnson, Hoover, Sagi, Garten, Parmar and Graham2016) found, using both social network analysis and social psychological experiments, that purity concerns were the greatest predictor of moral homophily both online and offline.
Linking puritanical morality to cooperation as the one ultimate end or social function of morality does not mean there must therefore be only one proximate moral concern or moral calculator in the mind: harm (or is it fairness?). For example, the senses all evolved to bring information about the outside world into the brain and into consciousness. Therefore, by the target article's logic, there must be only one sense: sight (or is it hearing?). But both genetic and cultural evolution are utterly indifferent to parsimony, and psychologists should be wary of theories that offer extreme parsimony at the price of a worse fit with the phenomenon under study. Koleva and Haidt (Reference Koleva and Haidt2012) called this trade-off “Occam's chainsaw.” Further, the claim that MDT's account supports a plurality of monisms (based on harm, or fairness, or both) seems conceptually incoherent to pluralists like us.
Humans are an ultrasocial and uniquely cultural species, and they cooperate in various ways and for multiple purposes (e.g., reproduction, parenting, coalition building, economic prosperity, security, etc.). Morality can involve more than one cooperative strategy and employ more than one psychological building block. Fitouchi et al. seem to agree that loyalty and authority are clearly related to cooperation, so they seem to acknowledge that there are four valid moral foundations: care, fairness, loyalty, and authority. Other researchers who take a morality-as-cooperation approach also maintain a pluralistic framework, just with a slightly different list of foundations (e.g., Curry, Mullins, & Whitehouse, Reference Curry, Mullins and Whitehouse2019). And cultural evolutionary models of cooperation maintain that there are multiple evolved mechanisms underlying cooperation, such as kin-based altruism, direct reciprocity, reputation, punishment, and signaling (Henrich & Muthukrishna, Reference Henrich and Muthukrishna2021).
We have shown throughout the development of MFT that moral pluralism is more consistent with the empirical literature than is moral monism (see Graham et al., Reference Graham, Haidt, Motyl, Meindl, Iskiwitch, Mooijman, Gray and Graham2018, sect. 4). Further, the target article's monist claims about harm (and/or fairness) are specifically contradicted by the empirical literature on self-control moralization. Combining historical and experimental approaches, Mooijman et al. (Reference Mooijman, Meindl, Oyserman, Monterosso, Dehghani, Doris and Graham2018) showed that moralization of self-control failures is most strongly related to, and facilitated by, group-binding concerns of loyalty, authority, and purity, much more so than care and fairness. And a recent investigation of the moralization of sensory pleasure found the exact same pattern (Goenka & Thomas, Reference Goenka and Thomasin press). If you limit morality to harm and/or fairness, then, you fail to capture the moralization of self-control that is so central to the target article's account.
We are happy that Fitouchi et al. and other monist researchers believe that “The idea of ‘purity’ transformed moral psychology” (Gray, DiMaggio, Schein, & Kachanoff, Reference Gray, DiMaggio, Schein and Kachanoff2022), and we could not agree more. In the past few years, several emerging lines of research, using predictive modeling of naturalistic data, have demonstrated the central role of moral purity in illuminating the powerful and destructive forces of morality. Here are four examples: (a) Pathogen prevalence is predictive of endorsements of moral purity, even after controlling for political ideology, and historically, when purity values become more salient, infectious diseases drop in subsequent years (Atari et al., Reference Atari, Reimer, Graham, Hoover, Kennedy, Davani and Dehghani2022b), (b) endorsement of purity, and other binding foundations, is predictive of US county-level frequency of hate-group activity, even after controlling for political orientation and socioeconomic status (SES) (Hoover et al., Reference Hoover, Atari, Mostafazadeh Davani, Kennedy, Portillo-Wightman, Yeh and Dehghani2021), (c) COVID-19 vaccination rates are negatively predicted by county-level endorsements of moral purity, even after adjusting for structural barriers to vaccination, and the demographic and religious make-up of the counties (Reimer et al., Reference Reimer, Atari, Karimi-Malekabadi, Trager, Kennedy, Graham and Dehghani2022), and (d) hateful rhetoric across contexts (from Nazi propaganda to hate speech on alt-right social media sites) and across 19 languages is strongly concomitant with purity language (Kennedy et al., Reference Kennedy, Golazizian, Trager, Atari, Hoover, Davani and Dehghani2022). As these examples show, removing purity from descriptive accounts of human morality would prevent us from understanding much of the “dark side” of moral convictions and concerns (Skitka & Mullen, Reference Skitka and Mullen2002).
Although we are convinced by Fitouchi et al. that Puritans highly moralize behaviors that are thought to reduce self-control, we think that the writings of Puritans make it clear that, like all people in all societies, they have more than one moral concern. The New England Puritan Cotton Mather once observed a dog urinating at the same time he himself was urinating. He was disgusted, and drew moral inspiration from his disgust. He later wrote this resolution in his diary: “Yet I will be a more noble creature; and at the very time when my natural necessities debase me into the condition of the beast, my spirit shall (I say at that very time!) rise and soar…” (Mather, Reference Mather1708). Various branches of Buddhism, such as Tantra, also use disgust (as when contemplating a corpse) as a means of making moral progress by breaking one's attachments to one's own body. It seems to us rather awkward to force these moral statements into the procrustean bed of cooperation (or harm, or fairness), rather than taking them at face value as concerns and cognitions about purity.
As with all moral phenomena, we think that moral puritanism is best understood through a pluralist lens that embraces the full range of moral concerns, including the group-binding, cooperation-enhancing concerns of loyalty, authority, and especially purity.
Fitouchi et al. argue that puritanical morality arises from moralization of self-control failures, which are seen to characterize undesirable future cooperation partners. We appreciate the authors' comparisons and contrasts between their moral disciplining theory (MDT) and moral foundations theory (MFT; Graham et al., Reference Graham, Haidt, Koleva, Motyl, Iyer, Wojcik and Ditto2013; Haidt & Joseph, Reference Haidt and Joseph2004), and we grant that self-control is central to puritanical morality and does not correspond neatly to any one moral foundation. We also appreciate the authors' historical psychological approach (Atari & Henrich, Reference Atari and Henrich2022; Muthukrishna, Henrich, & Slingerland, Reference Muthukrishna, Henrich and Slingerland2021), which helps theorists better situate cross-temporal changes in puritanical morality, rather than examining just one era. We also agree that it is crucial to study puritanism and purity, as it is the least WEIRD (western, educated, industrialized, rich, and democratic; Henrich, Heine, & Norenzayan, Reference Henrich, Heine and Norenzayan2010) aspect of moral cognition (Atari et al., Reference Atari, Haidt, Graham, Koleva, Stevens and Dehghani2022a) and remains understudied compared with other moral concerns that may be more salient in WEIRD contexts.
Although we find Fitouchi et al.'s theory of puritanical morality compelling (though incomplete, as we'll show), we believe their account of MFT has missed some important areas of agreement. First, MFT agrees with Fitouchi et al. that moral systems, as cultural products, are best understood as aiming toward cooperation. Haidt (Reference Haidt2012, p. 66) defined “moral systems” as “interlocking sets of values, virtues, norms, practices, identities, institutions, technologies, and evolved psychological mechanisms that work together to suppress or regulate self-interest and make cooperative societies possible.” But where Fitouchi et al. suppose that moral cognition is based on a single computational device that computes fairness (or is it harm?) to infer a person's cooperation potential, MFT argues that when humans developed widely shared and socially enforceable understandings of how things “ought” to be, they drew on multiple pre-existing cognitive systems, including the attachment system, reciprocal altruism, coalitional psychology, status hierarchies, and the behavioral immune system. MFT also posits that cultural learning can regulate these evolutionarily prepared psychological mechanisms (Graham et al., Reference Graham, Haidt, Koleva, Motyl, Iyer, Wojcik and Ditto2013), which can produce cross-temporal and cross-regional differences in the endorsement of moral foundations. Do Fitouchi et al. and other monists believe that these evolved psychological mechanisms did not get recruited into morality because the harm system (or is it the fairness system?) was sufficient for cooperation in our evolutionary history?
We do say, as Fitouchi et al. note, that the last item in that list has an evolutionary origin different from the others. The first four foundations all grew out of interpersonal dynamics in ancient primate and mammalian societies, whereas purity cognitions have their origin in a physical world full of pathogens (Atari et al., Reference Atari, Reimer, Graham, Hoover, Kennedy, Davani and Dehghani2022b). Yet MFT distinguishes between the evolutionary origins (from which we infer the “original trigger” of an intuition) and the current function (which includes the current triggers). We say, in numerous places, that purity now functions to enhance group binding, for example, “if you think, as I do, that one of the greatest unsolved mysteries is how people ever came together to form large cooperative societies, then you might take a special interest in the psychology of sacredness…Whatever its origins, the psychology of sacredness helps bind individuals into moral communities” (Haidt, Reference Haidt2012, p. 149). Consistent with this idea, Dehghani et al. (Reference Dehghani, Johnson, Hoover, Sagi, Garten, Parmar and Graham2016) found, using both social network analysis and social psychological experiments, that purity concerns were the greatest predictor of moral homophily both online and offline.
Linking puritanical morality to cooperation as the one ultimate end or social function of morality does not mean there must therefore be only one proximate moral concern or moral calculator in the mind: harm (or is it fairness?). For example, the senses all evolved to bring information about the outside world into the brain and into consciousness. Therefore, by the target article's logic, there must be only one sense: sight (or is it hearing?). But both genetic and cultural evolution are utterly indifferent to parsimony, and psychologists should be wary of theories that offer extreme parsimony at the price of a worse fit with the phenomenon under study. Koleva and Haidt (Reference Koleva and Haidt2012) called this trade-off “Occam's chainsaw.” Further, the claim that MDT's account supports a plurality of monisms (based on harm, or fairness, or both) seems conceptually incoherent to pluralists like us.
Humans are an ultrasocial and uniquely cultural species, and they cooperate in various ways and for multiple purposes (e.g., reproduction, parenting, coalition building, economic prosperity, security, etc.). Morality can involve more than one cooperative strategy and employ more than one psychological building block. Fitouchi et al. seem to agree that loyalty and authority are clearly related to cooperation, so they seem to acknowledge that there are four valid moral foundations: care, fairness, loyalty, and authority. Other researchers who take a morality-as-cooperation approach also maintain a pluralistic framework, just with a slightly different list of foundations (e.g., Curry, Mullins, & Whitehouse, Reference Curry, Mullins and Whitehouse2019). And cultural evolutionary models of cooperation maintain that there are multiple evolved mechanisms underlying cooperation, such as kin-based altruism, direct reciprocity, reputation, punishment, and signaling (Henrich & Muthukrishna, Reference Henrich and Muthukrishna2021).
We have shown throughout the development of MFT that moral pluralism is more consistent with the empirical literature than is moral monism (see Graham et al., Reference Graham, Haidt, Motyl, Meindl, Iskiwitch, Mooijman, Gray and Graham2018, sect. 4). Further, the target article's monist claims about harm (and/or fairness) are specifically contradicted by the empirical literature on self-control moralization. Combining historical and experimental approaches, Mooijman et al. (Reference Mooijman, Meindl, Oyserman, Monterosso, Dehghani, Doris and Graham2018) showed that moralization of self-control failures is most strongly related to, and facilitated by, group-binding concerns of loyalty, authority, and purity, much more so than care and fairness. And a recent investigation of the moralization of sensory pleasure found the exact same pattern (Goenka & Thomas, Reference Goenka and Thomasin press). If you limit morality to harm and/or fairness, then, you fail to capture the moralization of self-control that is so central to the target article's account.
We are happy that Fitouchi et al. and other monist researchers believe that “The idea of ‘purity’ transformed moral psychology” (Gray, DiMaggio, Schein, & Kachanoff, Reference Gray, DiMaggio, Schein and Kachanoff2022), and we could not agree more. In the past few years, several emerging lines of research, using predictive modeling of naturalistic data, have demonstrated the central role of moral purity in illuminating the powerful and destructive forces of morality. Here are four examples: (a) Pathogen prevalence is predictive of endorsements of moral purity, even after controlling for political ideology, and historically, when purity values become more salient, infectious diseases drop in subsequent years (Atari et al., Reference Atari, Reimer, Graham, Hoover, Kennedy, Davani and Dehghani2022b), (b) endorsement of purity, and other binding foundations, is predictive of US county-level frequency of hate-group activity, even after controlling for political orientation and socioeconomic status (SES) (Hoover et al., Reference Hoover, Atari, Mostafazadeh Davani, Kennedy, Portillo-Wightman, Yeh and Dehghani2021), (c) COVID-19 vaccination rates are negatively predicted by county-level endorsements of moral purity, even after adjusting for structural barriers to vaccination, and the demographic and religious make-up of the counties (Reimer et al., Reference Reimer, Atari, Karimi-Malekabadi, Trager, Kennedy, Graham and Dehghani2022), and (d) hateful rhetoric across contexts (from Nazi propaganda to hate speech on alt-right social media sites) and across 19 languages is strongly concomitant with purity language (Kennedy et al., Reference Kennedy, Golazizian, Trager, Atari, Hoover, Davani and Dehghani2022). As these examples show, removing purity from descriptive accounts of human morality would prevent us from understanding much of the “dark side” of moral convictions and concerns (Skitka & Mullen, Reference Skitka and Mullen2002).
Although we are convinced by Fitouchi et al. that Puritans highly moralize behaviors that are thought to reduce self-control, we think that the writings of Puritans make it clear that, like all people in all societies, they have more than one moral concern. The New England Puritan Cotton Mather once observed a dog urinating at the same time he himself was urinating. He was disgusted, and drew moral inspiration from his disgust. He later wrote this resolution in his diary: “Yet I will be a more noble creature; and at the very time when my natural necessities debase me into the condition of the beast, my spirit shall (I say at that very time!) rise and soar…” (Mather, Reference Mather1708). Various branches of Buddhism, such as Tantra, also use disgust (as when contemplating a corpse) as a means of making moral progress by breaking one's attachments to one's own body. It seems to us rather awkward to force these moral statements into the procrustean bed of cooperation (or harm, or fairness), rather than taking them at face value as concerns and cognitions about purity.
As with all moral phenomena, we think that moral puritanism is best understood through a pluralist lens that embraces the full range of moral concerns, including the group-binding, cooperation-enhancing concerns of loyalty, authority, and especially purity.
Competing interest
None.