Moralization, even with clear-cut issues, is context-specific
Learning from a study on consent pledges, and thinking about never/no vs. less/sometimes moral frameworks.
Gantman et al. (2025) have a new paper: Moralizing Consent: Three Field Studies Testing a Student-Led Intervention at University Parties.1 The basic setup: at “two social clubs on a college campus in 2017, we randomly assigned moralistic vs. informational messages about consent, delivered at the party’s door.”
I really like the paper’s theoretical framing, and I think it has some nice lessons for the animal welfare movement.
(Forewarned: this is a research-heavy post.)
Are we aiming for recognition or amplification?
From the introduction (emphasis added):
Moralization is the process by which preferences become values (Rozin, 1999; Rozin et al., 1997; Rozin & Singh, 1999), or ideas increase in their degree of relevance to a person’s sense of right and wrong (Rhee et al., 2019). Moralization has been studied primarily in two forms: moral recognition and moral amplification (Rhee et al., 2019). Moral recognition occurs when an idea or behavior not previously thought to be relevant to the moral domain shifts into moral relevance. For example, a person who experiences moral recognition may have previously thought that smoking cigarettes is a matter of personal preference but now thinks of smoking as morally wrong. Moral amplification occurs when an idea or behavior previously thought to be relevant to a person’s moral values becomes more central to their understanding of right and wrong. For example, a person who experiences moral amplification may have previously thought that smoking was somewhat morally wrong but now thinks of smoking as extremely wrong and conflicting with their core belief in not harming others.
In my model of the path to a vegan world, Recognition is probably step one,2 while amplification helps propels us towards being vocal, ethical vegans: folks for whom our food choices are both a reflection of our moral values and important to our sense of who we are. But when we are designing interventions that talk about the harms of factory farming, which are we trying to achieve? I think making people watch Dominion is aiming for recognition. But if most people basically accept that factory farming is bad, perhaps we should be more focused on amplifying that belief or sharpening it. For instance, you might tell people about some of the crueler practices in cage-free settings for hens and ask them to sign a petition to ban debeaking and mandate smaller flock sizes.
Does moralization require a “broad consensus” and “an imperative to action”?
From the abstract: “We tested whether messages given to students just before they entered a party impacted their thinking about consent in moral terms—i.e., as a clear issue, with broad consensus, and an imperative to action.” This is interesting to me because the authors later write that the “vast majority of existing work on moralization processes concerns smoking and meat eating (Feinberg et al., 2019).” I am not sure how to square these two things away, because I do not think there is a broad consensus that eating meat is bad. In fact, per that Feinberg et al. paper, a “long literature demonstrates that conformity pressures deter individuals from behaving differently than others for fear of not fitting in, being the target of gossip, and being ostracized,” and most people eat meat. What if there’s a consensus in your culture at large but a totally different one in your subculture? How do different cues activate one frame or the other?
Likewise with imperative to action: what kind of action? Vegans take a never/no approach to animal products, and reducetarians take a less/sometimes approach. In the realm of consent, articulating a personally reductionist approach would be horrifying: imagine someone saying they strictly respect consent at home but not at restaurants.3 What makes it more palatable for animal products? I have some hypotheses, but I don’t have very good theoretical tools yet for thinking the question through.4 I think this question is important, tractable, and neglected,
Institutional support is institution-specific
The intervention took place at two social clubs. At the first, students had already taken the lead by instituting
a new consent practice that was public, mandatory, approved by the student membership of the club, and developed independently of the university administration (and in response to a high-profile sexual assault at a peer institution; Stack, 2016a). At the club’s weekly late-night parties, open to all university students, a student wishing to enter the party would be required to read aloud a definition of consent.
Then, serendipity:
While we were conducting our experiment with the first social club, another club invited us to replicate our experiment at one of their parties. This club had no practice comparable to the first club’s pledge that would suggest to students that it moralized consent—an ideal comparison context.
What they find (emphasis added):
in a context where there is an institutionalized practice highlighting the importance of consent [the first social club], a moralistic message (vs. an informational message) engages students’ thinking about consent in moral terms, possibly triggering what we now would call moral amplification. In contrast, in a party setting organized by a student group with no institutionalized practice regarding consent, students reported greater moralized thinking about consent in response to an informational message about the practice, which did not portray it as a moral issue, suggesting that here, the issue of consent was not already widely moralized…This raises the intriguing possibility that explicitly moralistic messages are less effective in places where moralized attitudes are less widespread or less uniform—or that moralistic messages are not as effective for moral recognition as they are for moral amplification.
This got me thinking about the social context of some place-based meat reduction interventions:
Malhotra et al. (2020) promoted plant-based options at a cafe on Stanford’s campus, an online ordering app, and a fine dining restaurant in Palo Alto, and found mixed results that I’d characterize as an overall null;
Çoker et al. (2022) used a dynamic norms message to promote plant-based options at “22 in-store restaurants of a major retail chain across the UK” and also found null results, at least in part because some treatment sites didn’t faithfully implement the intervention;
Piester et al. (2020) study sustainability labels and norms-based messages at university cafes and found an effect on some women but few men.
A key difference: Gantman et al. “met with student leaders of the social club, who expressed interest in testing new language in their pledge.” For the three papers mentioned above, I have no idea whether subjects care what a fine dining restaurant, retail chain, or university dining hall suggests they should eat. But i definitely believe that attendees at a college party care about the house rules and the norms they communicate. There’s a clear element of social hierarchy there. How do we activate the same processes/categories for eating choices, particularly for how ethical vegetarians think about non-meat animal products?
A few outstanding questions (and preliminary responses)
How does the distinction between never/no and less/sometimes map onto the recognition/amplification framework?
I think people get stuck on the path from recognition to amplification. They have moral insights and then don’t commit. I see it in myself. Getting folks unstuck, or reducing frictions, is one way of to frame my research agenda.
Do some institutional contexts militate towards one or the other?
Clear us/them distinctions create tribal identities, and social movements are probably more compelling (cultlike?) if members adopt strict never/no framings. On the other hand, restaurants are probably a good setting for reductionist approaches. (Sweetgreen is not my friend.)
Are lessons learned from never/no issues applicable to less/sometimes ones?
Yes, for two reasons. First, interest groups groups often find that building a winning a coalition requires big-tent thinking. Second, harm reduction/behavioralist strategies are often surprisingly applicable to even the most clear-cut moral issues (though it might feel distasteful).
But I’m still thinking this through.
For me it was learning the fate of the lobsters at the supermarket when I was small. I didn’t connect the dots to what I should and shouldn’t eat until much later though.
The reductionist approach is a lot more intuitive here if you zoom out, e.g.. consider the implications of furniture choices in the typical college student’s dorm room. Per Hirsch et al. (2019), “sitting on someone's bed evoked consent for many students, even though dorm rooms generally lack other seating options.”