Sunday, March 01, 2009

The evolution of disgust

New reseaerch illuminates the path from "oral won't" to "moral don't".

Confronted with the worst excesses of human wickedness and moral depravity, we’re apt to respond not just with condemnation, but with deep and visceral revulsion. And the daily news provides all too many opportunities to observe the baseness of our fellow humans. When the horrendously brutal details of the short life of Baby P came to light last November, my mind struggled to understand not only how the child protection services at Haringey Council could have missed the abuse this poor child continuously suffered, despite 50 visits to his home over 2 years, but also how on earth anyone could possibly mete out such treatment to a defenceless baby (hardly an unusual thought, I appreciate). It’s incomprehensible, and revolting.

In such cases, our moral abhorrence or disgust is patently justified. Indeed, you’d suspect that anyone lacking such feelings on hearing this tragic story had a moral screw loose. More generally, moral revulsion has been advocated as a guide to moral judgment. On this view, there is wisdom in repugnance, which may express an intuitive understanding of actiosn, events and situations that the rational mind can’t fathom [1].

But moral disgust is a complex emotion. Perhaps more than any other, it is easily put to thoroughly immoral ends. Just as we physically push away disgusting food or objects in front of us, we emotionally and socially distance ourselves from those we view as disgusting. Research shows that people dehumanise extreme out-group members, such as vagrants, and are primarily driven by disgust when they do so [2]. Portraying social or ethnic groups in disgusting terms — as cockroaches or rats or even just as fat, greedy and greasy humans — is a frequent prelude to pogroms, ethnic cleansing, and genocide [3]. Disgust has driven attitudes to interracial sex, and today is a still a potent force in shaping attitudes to gay sex, with knock-on effects on views about gay rights, particularly the right to marry [4].

Far from being an infallible or even reliable moral light by which to find our way, moral disgust is frequently the source of bias, prejudice and hostility. The disgusting are seen as less than human, and treated accordingly (and as history reminds us, people are all too willing to make people appear disgusting by forcing them to live in filthy, squalid, humiliating conditions so as to justify the mistreatment they will subsequently face).

The complexity of disgust as a social and moral emotion is reflected in its development, both through evolution and in individual development. Historically and developmentally, moral disgust follows on the coat-tails of core disgust – the revulsion experienced when you seen bodily fluids, a rotting carcass or dog shit on your shoe (or on the tyres of your bike, as is more often the case for me). And core disgust is itself a cognitively complex emotion.

For a start, it is more than just distaste or a felling of aversion towards something. As Paul Rozin, one of the pioneers of research into disgust, points out, disgust is a much more cognitive and emotional reaction that simple distaste, and draws on an understanding what food is and where it comes from [5]. Animals dislike and avoid certain tastes, but don’t qualify as having a genuine disgust response. This claim needs a bit of unpacking. Imagine I show you a sterilised cock-roach, and then dip it in a glass of lemonade using clean tongs. Would you sip the drink? Probably not, even if you’re thirsty. The drink will seem contaminated, and disgusting.

The notion of contamination, which is a complex cognitive evaluation, is an important part of the human disgust reaction, and is clearly more than just distaste. The lemonade, after all, will taste exactly the same after the cockroach dipping (that is, delicious — at least for those who, like me, have a sweet tooth). It will even be safe. But the feeling of aversion directed at the cockroach gets transferred to the harmless drink (similarly, people are often reluctant to eat chocolates shaped like turds). This doesn’t happen among other animals, nor in children under 5 to 7 years of age.

Similar notions of contamination and transference as found in core disgust (indeed, are hallmarks of a genuine disgust response) have also been found in the moral domain. Just as we are repulsed by the prospect of consuming foods or drinks contaminated by elicitors of core disgust, we also fear moral contamination. Studies show that people feel a bit queasy about the idea of putting on a sweater worn by Hitler (even if carefully laundered). What’s more, being forced to recall our own moral misdemeanours produces an urge to physically cleanse ourselves in an attempt to wash away the moral stain on our character.

Some researchers have suggested that much talk of moral disgust is merely metaphorical. There is general agreement that moral judgments related to actions that involve elicitors of core disgust — faeces, bodily fluids and offices, and certain forms of bodily contact — have a strong disgust element driving them. (It’s little wonder that people get so vexed about sexual morality.) But when we say we’re disgusted by the venality and irresponsibility of investment bankers, are we really experiencing a visceral feeling of revulsion, or just using a verbal tag to show off our disapprobation of their actions?

There are clues that this isn’t so, at least in some cases. Studies such as those on Hitler’s sweater provide one reason to doubt the metaphor hypothesis. If disgust and the associated ideas of moral contamination are just metaphors, why do people object to donning Hilter’s now-clean sweater? There is also some evidence that core disgust and moral disgust elicit similar physiological responses. Whereas anger tends to cause the heart rate beat faster, disgust — prompted by looking at a gory image, say — makes it drop. So what happens when people view an extreme moral out group, like the Nazis? Does their bodily response suggest that they are getting mad, or feeling revolted?

Jonathan Haidt, who has worked with Rozin and built on his ideas, paired up with graduate student Gary Sherman to address this question, using a variety of nasty video clips, as well footage of Nazi rallies, which subjects viewed while their bodily reactions where monitored [6]. Haidt and Sherman found that not only did people report being disgusted by the Nazis, their heart rates also told the same story. Strikingly, those who heart rates dropped the most also reported greater clenching of the throat, another disgust-related muscular response. (Brain-imaging studies also suggest that similar brain regions subserve both core and moral disgust.)

All of which leads to the following evolutionary and developmental scenario for the emergence of moral disgust. Initially, animals evolved a distaste response that guided them away from poisonous or otherwise harmful foodstuffs. In humans, this distaste foundation was built upon to create the more complex and cognitively demanding ‘core disgust’ domain, whose primary elicitors are things that practically all of us find totally gross (shit, piss, puke, snot, puss – I use these decidedly non-euphemistic terms to fire up your disgust response!). Combine this with the notion of contamination and the transference of bad properties of one type of object or matter to another, and now history begins to count: in assessing whether to eat or drink something, it isn’t just a case of whether it looks nice or smells fresh; its provenance matters, as does its history of contact with other disgusting things (this, lamentably, can even include people).

This core disgust system took a long while to evolve, and takes time to emerge through child development. But once in place, it has been co-opted by our social and moral psychology to serve new ends — principally to distance ourselves from the morally odious. Whereas core disgust originally protected the body against oral incorporation of dangerous things, the expanded concept of moral disgust enables us to protect our moral selves, at the levels of individuals and communities, from moral contamination and corruption. As Haidt frames it, the guardian of the body has taken on a new role as a guardian of the purity of our souls.

The link between the two domains, the oral and the moral, is captured in colloquialisms expressing condemnation of moral transgressions, like “His behaviour left a bad taste in my mouth”. Of course, this isn’t a literal claim, but what underlies it? The idea of an evolutionary and developmental trajectory from “oral won’t” to “moral don’t” has recently been tested by psychologist Hannah Chapman and colleagues at the University of Toronto, who argue that disgust related to oral incorporation is indeed similar to that experienced during moral judgement [7].

In a paper just published in Science, Chapman et al. report on experiments in which they had subjects look at images designed to elicit disgust (they depicted dirtiness, faeces, and insects, for example) and also to taste some salty, bitter and sour liquids [7]. They then compared the facial expressions associated with these actions with those elicited by being treated unfairly. The focus on facial expressions of emotion derives from the Darwin-inspired research on the cross-cultural and universal expression of certain basic emotions, which includes disgust. For disgust, the canonical facial response involves wrinkling the nose and raising the lips.

They found that all three conditions (images, clips and unfairness in the economic game) caused subjects to raise their lips and wrinkle their nose in a disgust-type manner (associated with activation of the levator labii muscle region of the face). These responses were related to self-reported disgust, but less so for self-reported anger or sadness.

These are intriguing findings, and I agree with the authors’ conclusion that they are “consistent with the idea that in humans, the rejection impulse characteristic of distaste may have been co-opted and expanded to reject offensive stimuli in the social domain.” But interpreting their findings is not entirely straightforward.

In an accompanying commentary, Paul Rozin, Jonathan Haidt and Katrina Fincher highlight some of the problems. They describe a three-tiered model of disgust psychology (shown below). In this model, there are, going from top to bottom, stimuli (potential disgust elicitors), a disgust-evaluation system, and a disgust output response (which in turn has nonverbal, behavioural and physiological elements). Some stimuli, such as bitter-tasting drinks, feed straight into the disgust output response. They use the direct distaste pathway common to many animals — and just as animals do not engage a disgust-evaluation system, neither to these basic stimuli in humans. Other elicitors of core disgust, such as cockroaches and certain sexual acts, are processed by the disgust-evaluation system (lacking in animals and young children), which then activates a disgust response.

Domains of disgust. The schematic represents routes by which eliciting situations may trigger the disgust output program. Those that run through the disgust evaluation system--which includes appraisal of the elicitor, feelings, and contamination ideation--trigger the full disgust emotion. Solid lines represent routes through which an elicitor can activate the disgust evaluation-output program. Dashed lines (green) represent direct elicitation of the disgust output program. The dotted line (brown) represents a metaphoric, indirect route. (Image copyright of AAAS)

It is when we move beyond moral transgressions involving aversive substances to more purely moral issues that things get a bit more complicated. As shown in the figure for the case of unfairness, such abstract issues could, in principle, drive the disgust response by various routes. First, they could feed into the disgust-evaluation system, like many other elicitors of core disgust, and trigger a response. Alternatively, they could directly activate the disgust response, in the way bitters tastes do. Finally, moral transgressions related to fairness rather than bodily functions could be associated with concepts or verbal tags that in turn directly activate the disgust response. (We should also ask whether, in any of these cases, the disgust response is really just a metaphor — although they might represent different types of disgust, are they not all a species of genuine disgust? I leave you to ponder that puzzler.)

Chapman and colleagues suggest that the disgust experienced when on the receiving end of unfair treatment is in some sense “the same” as that arising from classical elicitors of core disgust (e.g., cockroaches). If the Rozin–Haidt–Fincher model is on track, then this claim would only be true if unfairness is processed through a disgust-evaluation system. And that question isn’t settled by the current study. What is needed, according to Rozin and colleagues, are experiments that “examine the effects of a variety of elicitors on a variety of dependent measures (e.g., contamination, appraisals, and feelings)”.

There are some other issues that also need to be ironed out. Previous research has suggested that different sorts of moral violations are linked to different sorts of moral emotions. The anthropologist Richard Shweder has suggested that moral concepts broadly cluster into three families [8]. The ethic of autonomy deals with individual rights, and issues of justice and fairness. The ethic of community is more focused on adherence of group norms related to social stability. And the ethic of divinity draws on notions of bodily and spiritual purity, and the sacred and profane (an ethical domain that has atrophied in Western societies — whether this is a cause for celebration of lamentation is up for debate).

Haidt and Rozin have previously argued that violations of the domains of community, autonomy and divinity typically lead to the moral emotions of contempt, anger and disgust, respectively (they call it, rather neatly, the CAD Triad Hypothesis) [9]. According to this scheme, the unfairness in this study should have produced subjects who were more angry than disgusted, but possibly a mix of the two. This is plausibly what happened: raising of the upper lip, which Chapman and colleagues used as a principal measure of disgust, is also activated by anger. And so perhaps it is simply that elements shared with the disgust response were activated, rather than a full-blown disgust response prodded in action by a disgust evaluation of unfairness.

In any case, these ongoing explorations and debates continue to reveal the complexity of our emotional and moral lives. What is so strange is how unaware we typically are of all that’s going on when we’re making evaluations about the good or bad, the awe-inspiring or abhorrent, the commendable or condemnable. From the outside, moral disgust can look very simple (not to mention simple-minded): you look, you go ‘Yuk!”, and you say it’s wrong. But this simplicity hides the machinations of an evolutionarily and developmentally complex, and quintessentially human, moral emotion.

1. Kass, L. R. (1997). The wisdom of repugnance. The New Republic 216, 17–26.
2. Harris, L. T. & Fiske, S. T. (2006). Dehumanizing the lowest of the low — neuroimaging responses to extreme out-groups. Psychol. Sci. 17, 847–853.
3. Glover, J. (1999). Humanity: A Moral History of the Twentieth Century (Random House, London).
4. Nussbaum, M. (2004). Hiding From Humanity: Disgust, Shame, and the Law (Princeton University Press, Princeton, NJ).
5. Jones, D. The depths of disgust. Nature 447, 768 (2007).
6. Haidt, J. & Graham, J. (in press). Social Justice Res.
7. Chapman, H. A., Kim, D. A., Susskind, J. M. & Anderson, A. K. (2009). In bad taste: evidence for the oral origins of moral disgust. Science 323, 1222-1226.
8. Shweder, R. A., Much, N. C, Mahapatra, M., & Park, L. (1997). The "Big Three" of morality (autonomy, community, divinity) and the "Big Three" explanations of suffering. In Morality and Health (Brandt, A. & Rozin, P. (eds.) 119–169 (Routledge, New York).
9. Rozin, P., Lowery, L., Imada, S. & Haidt, J. (1999). The CAD triad hypothesis: a mapping between three moral emotions (contempt, anger, disgust) and three moral codes (community, autonomy, divinity). J. Personality Social Psychol. 76, 574–586.