Moral Foundations: Which?

About a decade ago I wrote at my old blog about moral foundations theory, the idea that people have different fundamental moral feelings which drive their ethical intuitions; I also explored this at the Weekly Wonders tumblr. Although I have grown in my thinking a lot since then, a basic version of Haidt’s idea has become essential to the furniture of my mind; it seems obviously, trivially true that something about our ethics is both given and idiosyncratic. Our moral foundations are given in that they are neither something we chose nor something we acquired through reason but rather something motivating us without our even knowing why; they are idiosyncratic in that different people have different moral motivations. Our moral judgements are always driven by normative feelings, a particular kind of motivation along with appetites and fears. I know better than to trust what seems obvious to me, but I really can’t see how moral foundations theory could be wrong.1

5C6xuP

“The deadly Franciscan Fist,” Matt Baume, flic.kr/p/5C6xuP

That said, I still have some questions and concerns regarding moral foundations theory which I’ve never gotten into. My friend Jon has asked me to write a bit about this sort of thing so he can share it with his students and I’m going to use that request as an excuse to finally get around to airing those concerns. These concerns can be described roughly speaking as “which?” and “whence?” Today I’ll talk about “which?”

Haidt proposes six moral foundations (1. care vs. harm, 2. fairness and reciprocity, 3. liberty, 4. loyalty, 5. authority and respect, and 6. purity and sanctity) but remains open that these six may not be the real ones. This is to his credit, especially as it seems that these six cannot possibly exhaust the moral options. His method for determining these foundations is somewhat suspect, grouping words for moral concepts together by similarity rather than beginning with observations of actual human behaviour or reasoning.

Fortunately, there are alternatives. Oliver Scott Curry outlined such an alternative, MAC, on Twitter in December 2018, deriving moral foundations from evolutionary and game theoretical origins of morality. MAC isolates seven moral foundations based on evolutionary theories for altruism which he calls (unhelpfully) kinship, mutualism, exchange, hawk, dove, division, and possession.2

MAC has several benefits. It distinguishes between kin and in-group altruism. It acknowledges that virtues like courage and humility are important to many people’s moral universes. It finds a place for the (wholly emotional, not at all reasonable) sense that property is a moral concern.

I am concerned, however, that MAC does away with purity as a moral concern. They justified this because “it is not concerned with how we treat other people” and is not based in an evolutionary-cooperative account of morality. The first justification is false on the face of it: purity and sanctity concerns reliably involve the exclusion of individuals and groups and are thus incontrovertibly concerned with how we treat other people. The second justification seems arbitrary at best, begging the question at worst: to reject purity as a moral foundation because it is not based on an evolutionary-cooperative account of altruism does not leave leave space for normative judgements which we might observe that are not truly altruistic. Perhaps this is simply my misunderstanding, but if normative feelings concerning purity or sanctity are not moral, then what are they? They are quite clearly normative. In other words, MAC seems to be using “morality” to denote a different spectrum of concepts than anyone else is; if that’s the case, are they really talking about morality, or are they really talking about something more specific, like altruism?

My view, of course, is that there aren’t “real” categories out in the world; MAC cannot be wrong to slice the world up this way, so long as they are attending to an actual observable pattern, but they can certainly be misleading if they conflate it with other observable patterns. “Normativity” seems to be as observable pattern as “evolutionary-cooperative altruism.”

For the moment, then, I will be assuming that the following moral foundations are operative, but I remain open to being corrected:

  1. Kin obligations (MAC: kinship)
  2. Group loyalty (MAC: mutualism)
  3. Reciprocity and responsibility (MAC: exchange)
  4. Strength displays (MAC: hawk)
  5. Humility and obedience (MAC: dove)
  6. Fairness and equality (MAC: division)
  7. Ownership (MAC: possession)
  8. Purity and sanctity

I will consider later whether there might be more moral foundations.


  1. I do disagree with several of Haidt’s elaborations of this work, for instance the idea that it is incumbent upon progressives to attend to moral foundations theory in order to improve political discourse, or that all moral reasoning is therefore post-hoc justification of moral intuition.
  2. As is so often the case in psychology, this group desperately needs to improve their terminology.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s