A Trump supporter threatens to burn down Revolution Books in Berkeley, California (Screen cap).
Are Donald Trump's supporters driven by immoral hatred, or simply shaped by moral codes that most liberals or leftists don't understand? That question, phrased in more sophisticated and technical terms, has sharply divided political scientists ever since Trump's emergence on the scene -- and in fact long before that.
More than a year after the violence in Charlottesville, just as midterm campaign rhetoric starts to heat up for, it's a useful moment to examine what's wrong with even the most intellectually serious attempts at the mode of political analysis known as both-sides-ism. What does a more realistic assessment might reveal about ideological motivations?
This article first appeared on Salon.
About a month after Donald Trump’s election, long before we had heard the phrase "very fine people on both sides, I wrote a story here questioning the "good people on both sides" logic of Jonathan Haidt's "moral foundations theory,” or MFT, based on a paper by Jeffrey Sinn and Matthew Hayes, “Replacing the Moral Foundations: An Evolutionary-Coalitional Theory of Liberal-Conservative Differences.” It wasn’t the first time I’ve criticized MFT based on specific researchers' work (see here and here), but it was the most promising in terms of suggesting a compelling alternative approach. Now a new paper by Sinn goes a long way towards fleshing out that alternative and revealing its superior explanatory power.
In a nutshell, MFT claims that liberals think of morality in narrower, more individualistic terms (focused on such dichotomies as harm vs. care and fairness vs. reciprocity) than conservatives do. Conservatives are "good people" too, Haidt claims, but are concerned with a cluster of three "binding morality" issues, which he defines as "authority/respect," "in-group/loyalty" and "purity/sanctity."
It’s a tantalizing just-so story, satisfying to both conservatives, who feel liberals have disrespected them intellectually, and liberals, who want to be inclusive and fair-minded.
But Haidt’s key claim that “liberals and conservatives rely on different sets of moral mechanisms (or foundations)” is contradicted by evidence that “moral judgment involves a common template grounded in perceived harm,” as detailed in “The Unifying Moral Dyad,” by Chelsea Schein and Kurt Gray. Haidt also leaves a great deal unexplained, not least the question of where such values originally came from in our evolutionary past — as well as why and how they can veer into darkness, as they have clearly done with some of the support for Trump, and have done before in human history with distressing frequency and dreadful results.
In contrast, Sinn and Hayes explain conservative values in more realistic evolutionary terms, reflecting the navigation of different sorts of threatened harm. Haidt's “binding morality" idea "underspecifies conservatism," in a crucial sense, Sinn told me. As he sees it, that morality "arose because it steeled coalitions against adversarial out-groups,” which also “blends in practice with a darker orientation specifically motivated to dominate and exploit vulnerable out-groups." (Sinn and Hayes' theory carries the somewhat unwieldy label of "evolutionary-coalitional theory," or ECT.)
This analysis draws on two well-studied ideological attitudes that overwhelm all other factors when it comes to explaining bias or bigotry. The former, “binding morality,” reflects the influence of right-wing authoritarianism (RWA), while the darker motivations Haidt overlooks reflect "social dominance orientation" (SDO). In other words, according to this analysis xenophobia and out-group demonization are a feature of "conservative values," not a bug.
Haidt also "underspecifies liberalism," Sinn said, by giving it a limited, individualistic focus. But that’s only part of the picture, he noted. “Instead, ECT contends that liberalism reflects concern with a wider world, one that includes all people and extends morality to include the natural world.”
There’s been no response to that critique so far, Sinn told me. Which is disturbing, given the way that Donald Trump’s gross immorality, reflected in far too many of his followers, poses obvious problems for MFT. As Sinn explained, via email, MFT is now perceived as "established science," which makes it "tough to challenge." Why has that happened?
I think in part because it seems a more “progressive” or “enlightened” position to want to broaden the moral domain to recognize conservative morality as a real thing. Also, [MFT] tells a very easy to grasp, intuitive story about political differences (that seems non-judgmental) so it’s sort of hard to unseat. Ironically, I think liberals like it because it appeals to their universalism and tendency to want to respect diversity (in this case, ideological diversity).
But Sinn’s perspective doesn't require us to reject all conservative politics as immoral. Rather, it could guide us toward a more rational, equitable way of balancing competing ideological perspectives. Liberal values of universalism and respect for diversity can still be honored, but not in a framework built simplistically on a just-so story.
As Sinn and Hayes have argued in another recent paper, the values now identified as right-wing authoritarianism and social-dominance orientation may have provided evolutionary advantages in our past. But the dark side of those values, which we still carry with us as a species, is something we need to struggle with, overcome, and repurpose. We will need the insights and guidance provided by liberals' more universalistic view to guide us in that process.
In Sinn’s new paper, he builds a more comprehensive account of ideological motivations in two ways. First, he’s strengthened his criticism of MFT by drawing on a larger framework for mapping ideology known as the Schwartz Value Theory Circumplex.
This was originally based on data from 20 countries, and first reported in 1992 by Shalom Schwartz. In a review article 20 years later, now drawing on data from 82 countries, Schwartz explained:
The theory identifies ten basic personal values that are recognized across cultures and explains where they come from. At the heart of the theory is the idea that values form a circular structure that reflects the motivations each value expresses. This circular structure, that captures the conflicts and compatibility among the ten values is apparently culturally universal.
Sinn proposes situating the ideologies he’s previously discussed — along with libertarianism — within the SVT circumplex framework like this:
Second, Sinn draws on a larger repertory of evolutionary processes — based on life history theory, signaling theory, parent-offspring conflict and more — to account for greater refinement of ideological differences, and strengthen the perspective of seeing ideologies change over time. In doing so, he lays out significant foundations for a much more robust analytical response to the experience of Trumpism.
Trump’s emergence doesn’t just challenge MFT’s presumptions about conservative morality. As described by a paper just presented at the American Political Science Association’s annual meeting (“The Problem of Donald Trump and the Static Spectrum Fallacy”), Trump’s
reconfiguration of conservatism (not historically unique, by any means) exposes an even more fundamental problem: how to account for ideological coherence and differences in the face of repeated, sometimes sudden and drastic, issue-level change.
Sinn’s new paper doesn’t directly address this, but his discussion of the multiple ways ideological attitudes can evolve over time represents a more promising perspective for making sense of what’s otherwise a deeply confusing political moment. In addition, Sinn drew my attention to another paper by the aforementioned Chelsea Schein and Kurt Gray, which points toward a more general evolutionary dynamic that can plausibly operate on a variety of time-scales, sometimes with great rapidity.
To help clarify some of the most basic insights in his latest work, Salon interviewed Jeffrey Sinn by email. Our exchange has been edited for length clarity.
Your new paper is titled, "Mapping Ideology: Combining the Schwartz Value Circumplex With Evolutionary Theory to Explain Ideological Differences." OK, so what is the "Schwartz Value Circumplex," and why is it a promising framework for making sense of ideological differences?
The Schwartz complex offers a very useful framework for understanding value differences. Its circular composition captures both alignments and tradeoffs that show up among different value positions. Adjacent values are complementary and correlate positively (e.g., power and wealth), values positioned across from one another are in conflict and correlate negatively. I find it promising for explaining ideological differences in that it captures both these convergences and conflicts.
You write that "Considering a possible libertarian orientation helps reveal SVT’s strength as a nomological network." What do you mean by that?
Because SVT can capture some of the psychological profile of libertarians that the rhetoric of libertarians misses. Libertarians emphasize self-direction and self-determination. So at face value, you might think that they would then want to liberate others (like liberals want to do). But libertarians aren’t typically very concerned about benevolence. They also tend to be folks who don’t back traditional values or conformity and tend toward being less empathetic and more selfish. It’s exactly the sort of profile that the Schwartz profile would suggest for them.
However, you can’t come out and say, “I’m just not really that empathetic or concerned about others and I’m sort of a hedonist.” It sounds better, it’s more morally defensible, to present yourself as very concerned about liberty and not being constrained by authority. In short, I think SVT reveals characteristics or a flavor of libertarianism that cuts away the veneer and gets at something closer to the actual set of values motivating libertarianism.
Ideological differences seem to tap into profoundly different assumptions about the way the world should work, and most ideological positions can be located within the logic of one of these 10 values and the differences among them. For example, conservatives tend to be more deferential to authority and power, and there’s an inherent tension there with the value of universalism (i.e., egalitarianism).
By contrast, MFT doesn’t tell you how its two principal factors (i.e., binding and individualizing morality) relate to one other. With Schwartz, there’s a visual explication of the underlying logical tension between wanting to recognize some people as superior and wanting to treat everyone as equal. MFT, in contrast, suggests you can be high (or low) [in every category], and so doesn’t capture these tradeoffs.
In the next section, "SVT and Morality," you write that "researchers might also consider whether SVT’s broader scope could actually absorb MFT’s constructs." You argue that MFT mistakes "moralized values for moral foundations." What's the distinction between those things, and why does it matter?
MFT implies that its constructs are “foundational” – that they have causal efficacy. It’s slippery, in that it doesn’t often make that claim explicitly, but you can see it clearly in the title of the 2009 flagship piece: “Liberals and conservatives rely on different sets of moral foundations.” Their data are correlation and don’t support this claim. A less speculative claim would be that liberals and conservatives moralize different sets of issues.
In other words, I would argue that "right-wing authoritarian" conservatives moralize deference to authority because it’s part of a broader strategy of maintaining cohesive, kin-based coalitions. It’s not as if they have a different internal moral sensor more finally attuned to authority issues. They make deference to authority a moral value because it feels right given the bigger overall strategy. You can see this in that they don’t moralize deference to scientific or professional authority as much as they do traditional sources of kin-based authority (e.g., ethnic, family or religious authorities).
What's the larger significance of this argument?
Claiming that moralized values are instead “moral foundations” makes them part of the internal, authentic orientation of the individual in a way that suggests we need to respect those values, just as we respect immutable, inherent features like race or sexual orientation. It’s a rhetorical move unjustified by the data. You can see echoes of this in the rhetoric of the Christian right in talking about “religious freedom” – that is, the argument that I have to discriminate against Group X to authentically practice my religion.
In the section of your paper on right-wing authoritarianism and social-dominance orientation, you write that "RWA and SDO may reflect, respectively, slow and fast life-history strategies," and that the "slow-fast distinction may reveal another circumplex axis." Explain what you mean by this.
In brief, the "slow-life" orientation invests for the future and doesn’t build bridges, the "fast-life" orientation is opportunistic and short-sighted. The data (from another study by me and Hayes) suggest that right-wing authoritarians don’t see themselves as mean or prejudiced. They see themselves as reliable coalition partners, as altruists. Social-dominance folk, in contrast, aren’t really conservative in the social sense of the word (e.g., a Jesus-like concern for the poor, pious, altruistic, sexually restrained). They are interested in power. I have other data showing that SDO correlates with sadism and a preference for sadistic leaders. They overlap with right-wing authoritarians in sharing security and power values, but they also endorse hedonism (and self-interest) in a way RWA does not.
The Schwartz values likely break along "fast" vs. "slow" in orientation – i.e., half of the circle tends toward slow and half tends toward fast. That is, the fast-slow distinction may help explain why we see differences in the values people endorse.
In the section "Parent-Offspring Conflict and Ethnic Nepotism," you write that "the conflict between tradition/conformity and self-direction (or, more broadly, between social conservatives and liberals) may reflect, in part, parent-offspring conflict." What's the reasoning here?
The short story is that social conservatives are very concerned about restraining sexuality, and Schwartz doesn’t really get at this. That is, the labels of "tradition" and "conformity" don’t suggest anything about sex per se. The basic strategy is to build a cohesive community that makes the kids help each other and sacrifice for the larger family in a way that favors the genetic interests of the older generations (e.g., marrying someone in the tribe who will strengthen the tribe, vs. someone who will take your daughter or son away).
The self-direction more typical of liberalism is less interested in family and maintaining cohesive family structures, and more interested in controlling one's own sexuality to maximize one's own genetic fitness by marrying the individual with the best genetic material.
So a classic issue is whether an adult child can marry outside the tribe. If family is less central or has fewer resources (e.g., not nearby because of labor mobility, or less a source of prestige in the modern financialized world), then the offspring maximize their genetic fitness by mating with the best genetic fitness they can find -- the smartest or the healthiest -- regardless of that person's social background, religion, race, etc. Romeo and Juliet, in short, were liberals.
This paper is a theoretical overview that draws on a wide range of empirical studies to argue how they should be seen as fitting together. What's the most significant empirical test your theory could face?
It should be able to make finer distinctions than other theories. For example, and this isn’t in the paper, we should be able to distinguish the political issue positions of different types of liberals – those more driven by benevolence and universalism and those more driven by self-direction and stimulation. I think those scoring higher on the former should be more concerned with issues like child welfare and environmentalism, whereas the latter should be more concerned with things like access to family planning and legalization of drug use. Little ideology work has gone into explaining liberalism – it’s typically taken as a kind of default.
Your previous paper argued against MFT, in part by questioning how "moral" some "foundations" really are. This new paper seems to go even further than MFT. The SVT circumplex gives us more seemingly equivalent values. So which way are we headed: Do we regard all these values as equivalent or not? As you have previously argued, right-wing authoritarianism, combining in-group altruism and strong leadership, could provide an evolutionary advantage for hunter-gatherer tribes. But combined with social-dominance orientation, it threatens to drive us into a trajectory of mass extinctions, culminating in our own. Arguably, we can only avoid such a fate by shifting our balance toward more universalist values, which now have superior survival value.
So, putting my personal cards on the table, we need universalism or we’re doomed. At the moment, I’m inclined to accept arguments by moral dyad theory, which suggests there’s a specific mechanism that recognizes an agent causing harm and that this conceptual dyad is at the heart of perceiving an issue as moral in nature. This mechanism can get extended to other domains (by argument or experience) but this is core. In this sense, it’s just descriptive.
Now, ethically and philosophically speaking, “moral” seems to have a far wider reach, where people argue about what is moral and what is not, and advance claims they think are “true.” Not being a moral relativist, I do this, and I would be willing to argue in general for “universalism” at its core being more “moral” than some other orientations.
So neither Schwartz nor I are saying that Schwartz’s 10 values are definitely “good” or moral. I would argue that keeping immigrant children in cages is immoral. One issue I have with MFT is that it claims to be purely descriptive but often slides into demanding a recognition of conservative morality as "moral" in the normative sense. I think it’s wrong there.
On this issue and others, psychology bumps up against deeper metaphysical or ontological issues and sometimes has far more confidence than it should in its empirical methods. It has to inform those things, but it can’t offer a sufficient answer on its own.
Related Post