I’ve puzzled for some time about why people flirt. Students of (non-human animal) behavior know that males of many species aren’t typically subtle about signaling their interest in mating. A peacock displaying his tail feathers to a peahen is more or less saying, “My quality is this high, and I’m telling you this because I’d like to mate with you.” The males of other species sing, dance, strut, and generally make a big show to indicate their quality and their interest.
Set against this backdrop, humans seem, in a word, weird. People do signal to potential mates in various ways, displaying wealth and skills and so on, mirroring animal signals of quality. But at least some elements of human mating carry a certain ambiguity, especially regarding one’s intentions. “Do you want to visit the chili cook-off?” might reflect an intention for a casual, Platonic chat and a nice, hot bowl of chili. Or it might not.
As I say, I’ve puzzled over this for some time. So much so that from time to time I’ve conducted some field research on the subject, which led me to the conclusion that one important reason that people indicate their potential sexual intentions in a subtle way is that abandoning subtlety leads to drinks in one’s face, which in turn results in an unpleasant stinging sensation in the eyes. These informal investigations focused my attention on the issue of costs.
What are the costs of peacock-like signaling of sexual intent? In 2007, I had a talented group of undergraduates conduct a little study inspired by what we self-satisfyingly called the GIST model, standing for the Gradual Increase in Sexual Tension. One part of this model was the idea was that perhaps the costs of flirting had to do with third parties; if so, flirting should be more subtle when pairs were being observed. So, we had subjects chat with an opposite-sex partner in a chat room, either with or without a third party observer able to see the chat session. We got some small effects of this treatment – a bit more restraint when pairs believed (correctly) their chat session was being observed by another subjects – and a few additional undergraduate projects followed, many of which had similarly clever abbreviations and similarly small effects.
All of which leads me to a new paper (“Covert Sexual Signaling: Human Flirtation and Implications for Other Social Species”) that came out last week by Andrew Gersick and me in which we try to address this issue. Gersick is a student of the famous animal behavior scholars Robert Seyfarth and Dorothy Cheney, and has done some work showing that cowbirds sing songs of different “strength” depending on the social context.
A key part of the idea is that in various species, third party observers interfere with others’ mating attempts. This is unsurprising; male-male competition is more or less all about interfering with others’ mating attempts. Could this be driving the human subtlety of signaling one’s sexual intentions?
Perhaps. But as the drink-in-face example illustrates, there could be potential costs even if no one else is watching. For instance, in the context of a Platonic opposite-sex relationship, a (relatively explicit) signal of sexual interest might damage the relationship. Everyday experience – and When Harry Met Sally – suggests that an unmistakable sexual proposition in a previously non-sexual relationship tarnishes the relationship. In contrast, a subtle signal of potential sexual interest, if gently brushed aside, leaves both parties able to maintain the fiction that the interest was never signaled in the first place.
If explicit signals of sexual interest carry potential costs, either from the receiver or third parties, then less explicit signals might reduce or eliminate these costs. That raises the problem of how to indicate interest. One way to do this is to indicate interest in such a way that observers might not be able to infer with certainty that one’s intentions are, in fact, sexual. Ideally, one wants to signal in such a way that a receiver can detect that there might be sexual interest but in a way that is plausibly deniable.
One way to use this is to use “indirect speech.” When I ask if you want to go to the chili cook-off, there are (at least) two different questions I might be asking you. The surface meaning is whether or not you want to get some chili. A second meaning is a question about whether you’re interested in a date, a prelude to mating. If you are not the least bit interested in a date, you can interpret my remark as a simple invitation to chili, and indicate that you don’t like spicy food, or what have you. Further, if you challenge me by indicating that you’re engaged and so really shouldn’t be going to chili cook-offs, then I can innocently say that I was just being sociable, as opposed to making anything resembling a sexual overture. Indirect speech, and other ambiguous moves in social dynamics, allow parties to test the waters for interest while simultaneously minimizing potential costs imposed by the target or observers by leaving a safety valve of deniability.
We add one additional element, and by “we” I mean this piece was all Andy’s idea. Suppose it’s true that to be a “good” flirt, one has to be appropriately ambiguous. Considerably greater subtlety is required when one is trying to signal sexual interest, say, in front of a potential mate’s partner than in a context in which a pair is alone, with previously indicated mutual attraction. Using an appropriate level of overtness, then, itself becomes a kind of a metric of skill, and something on which potential mates are evaluated.
At the end of the paper, we identify a number of predictions for both sender and receivers of flirtation. Flirting is, in my experience, a difficult phenomenon to study, but I hope that this paper will stimulate additional work in this relatively sparse research area. If anyone out there wants to collaborate on some work testing these ideas, I’d be happy to discuss them. There’s this great chili cook-off we could go to…