August 12, 2004

Disgust for accents : pre-adaptation or figure of speech?

Continuing our earlier discussion of whether people sometimes feel real disgust in reaction to the speech of others, Paul Bloom sent me an electronic copy of a book chapter in which he discusses some closely related questions. This is chapter six of his new book Descartes' Baby: How the Science of Child Development Explains What Makes Us Human, and it discusses the development of disgust in human infants and children. Bloom deals with the progression from "babies and toddlers [who] will happily play with, roll around in, and even eat substances that make their parents gag", to adolescents and adults who may say that they are disgusted by such abstract concepts as commercial greed or badly designed software.

I'm focusing here on just one aspect of Bloom's chapter, where he deals with the question of whether the same emotion of disgust is really involved all along this gamut. The background is work by Paul Rozin, April Fallon, Jonathan Heidt and others, arguing that "core-disgust is an emotion that makes people cautious about foods and animal contaminants of foods"; but "disgust has expanded ... to become not just a guardian of the mouth, but also a guardian of the 'temple' of the body, and beyond that, a guardian of human dignity in the social order." (I'm quoting here from a paper entitled "Body, Psyche, and Culture: The Relationship Between Disgust and Morality", Heidt, Rozin, McCauley and Imada, because it's available on line. It's not one of the sources that Bloom quotes, but the ideas are essentially the same).

Heidt et al. go on to place this expansion of disgust into an evolutionary framework:

If the heterogenous class of disgust elicitors is linked together by a set of shared schemata, then the elaboration of disgust, from core through socio-moral, may be explained by the mechanism of "preadaptation" (Mayr, 1960). Mayr suggests that the major source of evolutionary "novelties" is the co-opting of an existing system for a new function. We suggest that core disgust be thought of as a very old (though uniquely human) rejection system. Core disgust was "designed" as a food rejection system, as indicated by its link to nausea, its concerns about contamination, and its nasal/oral facial expression. Human societies, however, need to reject many things, including sexual and social "deviants". Core disgust may have been preadapted as a rejection system, easily harnessed to other kinds of rejection. This harnessing, or accretion of new functions, may have happened either in biological evolution or in cultural evolution (Rozin, 1976; Rozin, Haidt & McCauley, 1993). Human societies take advantage of the schemata of core disgust in constructing their moral and social lives, and in socializing their children about what to avoid.

In the cited chapter from Descartes' Baby,Paul Bloom argues that this goes too far. He quotes Paul Rozin explaining (in another paper) that disgust has developed "from a defense of the physical body to a more abstract defense of the soul":

Humans must eat, excrete, and have sex, just like animals. Each culture prescribes the proper way to perform these actions—by, for example, placing most animals off limits as potential foods and most people off limits as potential sexual partners. People who ignore these prescriptions are reviled as disgusting and animal-like. Furthermore, humans are like animals in having fragile body envelopes that, when breached, reveal blood and soft viscera; and human bodies, like animal bodies, die. Envelope violations and death are disgusting because they are uncomfortable reminders of our animal vulnerability. Finally, hygienic rules govern the proper use and maintenance of the human body, and the failure to meet these culturally defined standards places a person below the level of humans. Insofar as humans behave like animals, the distinction between human and animals is blurred, and we see ourselves as lowered, debased, and (perhaps most critically) mortal.

Bloom suggests that disgust is not nearly this "smart", or at least not this abstract:

Rozin’s theory is too conceptual, too cognitive. It misses the physicality, the sensuality, of disgust. It is just not such a smart emotion. Simply being reminded—intellectually—of the fact we are animals is neither necessary or sufficient for disgust. Humans breathe and sleep, after all, “just like animals.” But breathing and sleeping are not disgusting. Looking at a brain scan or an X-ray is a stark and striking reminder of our physical nature, but these are not disgusting activities. Ruminating that I will one day die—just like any other animal—might make me sad, but it does not normally disgust me. In general, being reminded of our animal nature is not, by itself, disgusting.

A more plausible view is that death, bad hygiene, body-envelope violations, and certain sex acts disgust us simply because we perceive them, at a basic sensory level, in much the same way we perceive rotten meat and decaying flesh.

Bloom grants that people often use the language of disgust in "highly abstract and intellectual" ways:

In just a few months, I heard the word "disgusting" used to describe

The president’s tax plan
Someone writing a negative review of a grant proposal because he disliked the applicant
The high cost of prepared spaghetti sauce

But he argues that this is a "metaphor", not a true pre-adaptation:

This all seems to indicate that disgust can be highly abstract and intellectual. But I am skeptical. My hunch is that in these statements “disgust” is a metaphor. Saying that we are disgusted by a tax plan is like saying that we are thirsty for knowledge or lusting after a new car. After all, if you actually observe people’s faces and actions during heated political or academic discourse, you will witness a lot of anger, even hate, but rarely, if ever, the facial or emotive signs of disgust.

One problem is that none of the terminology (emotion, disgust, metaphor, pre-adaptation, etc.) is very precisely defined here. In the case of "emotion" and "disgust", the whole point is really to try to figure out what the boundaries and subdivisions are. And I should think that the "accretion of new functions" whereby "human societies take advantage of the schemata of core disgust in constructing their moral and social lives" might be described as a "metaphor" and simultaneously as a "pre-adaptation".


In evaluating these questions, it would be helpful to have more precise definitions of the terms involved, and it would also be helpful to have experimental evidence about a number of independent indicators. We have the words that people use to describe their feelings; we have their self-reporting about states like nausea; we have their facial expressions. It also seems that "core-disgust" may have some reasonably well-defined neurological correlates: a 2003 meta-analysis by Murphy et al. of available functional imaging studies found disgust-related activity most often concentrated in the insula/operculum and the globus pallidus, in quite a different pattern from the activation for other negative emotions:

["Lateral OFC" is "lateral orbitofrontal cortex", and "RSACC/DMPFC" is "rostral supracallosal anterior cingulate cortex/dorsomedial prefrontal cortex".]

It should be possible to check whether more abstract intances of self-described "disgust" show the fMRI patterns associated with disgust, or anger, or both, or neither. In particular, we could check whether the visceral flashes of negative emotion that many people report feeling in response to disliked accents look like disgust, in fMRI terms as well as in terms of facial expressions and so on.

Sometimes, when people say "disgusting" to mean "something I don't like", I'm sure that the metaphor is as dead as a computer mouse. It's plausible that being annoyed by overpriced sauce is often a completely different emotion from being nauseated by the smell of rotten meat, even if people use the same adjective to describe it. But it's also plausible that moral revulsion at price gouging is sometimes be strengthened by resonance with "core-disgust". Looking empirically at facial expressions (as Bloom suggests) and at functional imaging data probably wouldn't settle these questions, but it might help move them to more interesting levels of uncertainty.

Anyhow, in considering the emotional valency of speech sounds, we shouldn't limit ourselves to disgust, or even to the broader set of negative emotions. We're a cheerful and optimistic bunch here at Language Log, and so we'd want to consider the emotional reactions to accent and voice quality in a broader perspective. Reactions can be positive as well as negative. Some have to do with the listener's basic emotional frame for the speaker: sexual attraction or repulsion, dislike and annoyance or warmth and benevolence, respect or its disdain, enjoyment or disgust. Others evoke fairly abstract stereotypes of the speaker -- as snooty or stupid or whatever -- that have an emotional loading. A bit of poking around has not turned up much literature that deals with this subject systematically, but I'll keep looking.


Posted by Mark Liberman at August 12, 2004 08:45 PM