Mixed signals
25 11 2009Judith Donath gave a presentation for the Berkman Cooperation group that I went to on Monday.
Basically, Judith is interested in using insights from signaling theory to help create richer online communities and to facilitate online communication. She’s currently writing a book on that subject.
Signaling theory is a rapidly growing sub-discipline in evolutionary biology, but it has really fascinating implications and applications throughout the human social world. The basic idea is that organisms evolve in all kinds of ways to send signals, with the purpose of altering behavior of other organisms in their environment. The signals could be anything from a brightly colored tree-frog signaling that it is poisonous to avoid predation, to a peacock signaling its fitness to a mate, to a human signaling wealth via conspicuous consumption. Pheromones, birdsong, coloration, fashion — nearly everything in nature signals something to some organism.
Sexual selection can drive signals to some really absurd lengths. Peahens drove male peacocks to evolve completely gaudy and useless plumage (this photo is incredible) even though this actually puts the peacocks at a greater risk for predation. I remember as an undergrad reading that female bullfrogs will choose a mate based on the pitch at which the male bullfrog croaks, and evolution is pushing them toward a particular pitch and vibrato.
Because signals have survival value and are crucial within sexual selection, it is often worthwhile for organisms to signal dishonestly, which also sets of some really fascinating arms races. There are a number of non-poisonous butterflies, for example, that have adapted to match the color scheme of poisonous butterflies so that predators will ignore them. Judith gave the example of tigers, who claim territory and attract mates scratching trees to demonstrate their height. Taller tigers frighten away potential competitors and attract mates in this way. But some shorter tigers have discovered that they can jump and claw trees at a higher point than even the tallest tigers can normally reach, making the signal less reliable.
Humans have come up with all kinds of ways to signal that they’re more fit or attractive than they actually are (lies, make-up, surgery, conspicuous consumption, status markers, etc.), and these signals produce real social and biological consequences. Easy to copy social signals sometimes get paired with high sanctions to keep people honest (counterfeit money gets you sent to jail, faked diplomas get you fired, etc.).
Because dishonest signals are all over nature, and organisms (and humans) tend to develop signals that are harder to reproduce or falsify. Amotz Zahavi was the first to formalize this observation in the handicap principle, which states that in order for a signal to be reliable, that signal has to be costly, i.e. either risky or hard to replicate. In order for signals to have any adaptive benefit and become prevalent through evolution, they have to communicate something that is hard to fake. More recently, mathematical models have been developed to model and express concept more precisely.
The examples multiply almost endlessly when you start considering human signaling. Judith Donath devoted a large portion of her presentation to showing that fashion serves as a signaling system that communicates leisure, taste, wealth, attractiveness, one’s attitude toward risk/convention, and a number of other traits with social and sexual significance. Because keeping fashionable requires a lot of inside knowledge and money, it’s a pretty reliable indicator.
What’s interesting about this is that the fashion industry is able to very successfully capitalize on people’s underlying desire to signal these traits. Because novelty is so crucial to this process, however, fashion has a forever shortening half-life. For people who really care, even something from a year or a season ago is terribly dated and unappealing. In itself that’s a fine way to signal wealth and taste, but this signaling system produces a staggering amount of waste.
I want to give some more thought (maybe in a paper?) to what signaling theory means for the legal system. I can imagine a pretty interesting anthropological account of a primitive precursor to our legal system serving the purpose of keeping signals honest (money should indicate legitimate earning, it should be costly to replicate/imitate authority figures, etc). There are probably a lot of legal rules that could be explained sensibly through the signaling theory language, which could provide a lot of interesting sociological/biological context.
Categories : Ethological



















