Males sacrifice themselves for the sake of their offspring, whereas females are aggressive attackers. Biologists typically resort to biased language like this to describe sexual cannibalism, a new study concludes. Such stereotypes, the paper's authors argue, could leave scientists in the dark about certain aspects of animal reproduction.
Traditional gender stereotypes have colored animal behavior research before. Early scientists believed that mate choice was driven entirely by the male, for example.
They thought that competition between males spurred the evolution of traits such as gaudy feathers in peacocks; females were simply passive responders. Not until the mid-20th century did researchers began to more widely emphasize the role of female choice.
Those preconceptions continue to crop up in the scientific literature. A 2011 study in Animal Behaviour examined the words used by scientists to describe sexual conflict in animals  over 28 years between 1979 and 2007. Sexual conflict occurs when males and females have different reproductive goals and strategies. For example, males may seek to mate as many times as possible, whereas a female benefits from less frequent matings that don't stress her nutritional reserves. The team found that when scientists write about sexual conflict, they frequently describe males with active words ("intimidate," "coercion"), reserving more reactive words for females ("resistance," "avoidance"). This choice of language, the authors argued, reflects human sexual stereotypes and may be inhibiting the understanding of animal behavior.
Do the same biases hold true for descriptions of sexual cannibalism, in which one sexual partner eats the other immediately before, during, or after copulation? The practice is common in spiders and other arthropods, including praying mantises and crickets. The vast majority of the time, the female ends up with the full belly. The extra nutrition, biologists suspect, helps her lay and guard her eggs.
But the idea of large, predatory females attacking seemingly helpless males doesn't jibe with our traditional gender stereotypes. To find out whether scientists fall prey to such clichés, Emily Burdfield-Steel, a biologist at the University of St. Andrews in the United Kingdom, and colleagues reviewed 47 research studies published between 1984 and 2009 that examined sexual cannibalism in 30 arthropod species. Their results, published online this month in Animal Behaviour, indicated potential stereotyping in the sexual cannibalism literature . They found, for example, that female sexual cannibals were described with overwhelmingly negative language. Twenty-three of the studies used the words "aggressive" and "attack" to describe female behavior. Other common labels included "predatory," "voracious," and "rapacious." The most typical words for male behavior, meanwhile, were "escape," "sacrifice," and "avoid." These passive words were used to describe even those species in which the males were actively trying to get away.
"The word 'sacrifice' implies a level of nobility, like they're doing something selfless. But even for the males that are allowing themselves to be eaten, it's not selfless at all," Burdfield-Steel says. If his partner has better nutrition, she will lay more eggs, and the male will pass along more of his genes.
"The problem with these words is that they make moral judgments about the behavior of the animal," Burdfield-Steel says. "It might seem to human eyes vicious to eat your mate, but that is just the spider being the spider."
Another interpretation of these results, Burdfield-Steel says, is that this choice of language accurately portrays what is going on. Rather than reflecting the human femme fatale stereotype, female sexual cannibals really could be predatory and voracious. Without searching for this potential bias in their studies of sexual cannibalism, however, researchers will never know which is true, she says.
"I have sympathy for efforts to remove biases in science so was pleased to see this paper," says Patricia Gowaty, an evolutionary biologist at the University of California, Los Angeles. "That said, I think the authors missed an important point." One of the main goals of animal behavior research, she says, is to operationally define behaviors so that another scientist not involved with the research could read the definitions and use them to replicate the experiment and quantify the movements and actions of the animals. Many of the words identified by Burdfield-Steel and colleagues, such as "voracious," can't really be operationally defined.
The problem isn't the sexual stereotypes implied by these words, Gowaty says, but that other scientists can't apply them to observe the same behaviors. Scientists should avoid them because they are imprecise, not because of any potential bias. "I wonder what in the world some of these words are doing in any scientific paper."