Trolls, Conventional Signals, and Aristocrats...OH MY!!

In Judith S. Donath’s article Identity and Deception in the Virtual Community she focuses on types of deception along with ways to weed out trolls and impersonators using every day signaling techniques.  The two types of signals are assessment and conventional.  An example of an assessment signal is large horns on a stag, signaling strength for the stag and is almost impossible to imitate.  An example of a conventional signal is a person wearing a power lifter shirt.  The general public then assumes they are a power lifter, but it is easy to imitate.  Donath goes on to explain that conventional signal are used rather than the assessment signal because they “are often less costly, for both the signaler and the receiver” (4).  Later she goes on to show how conventional signals are used in the signature of Usenet through the example of a programmer’s signature.  This is a conventional signal but still has relatively costs to learn.  There are four separate steps to the signature, each testing to make sure the signer has the proper knowledge to post on the forum.   It would take a troll several hours/days/weeks (pending on their knowledge of the code at hand) to imitate such a signature.  Thus, the group of programmer’s keeps transaction costs high, while still using a conventional signal, for trolls looking to inappropriately stir the pot or steal the programmer’s social capital.

Making conventional signal’s cheap to the proper audience but expensive to impersonators reminds me of an Economic Colloquium I attended this winter by Doug Allen.  The lecture focused on the English aristocracy and the transaction costs involved in being an aristocrat.  In this lecture Doug explained that there were upwards of ten conventional signals that needed mastering in order to be considered an aristocrat loyal to the crown in England.  These signals carried high transaction costs to copy; like learning how to dance in a certain style.  If another aristocrat thought you were not loyal to the crown, they could challenge you to a duel and a proper aristocrat would have to agree.  If they denied the duel then it was obvious you were not loyal to the crown and exiled from the community.  If you agreed, you may be killed, but at least everyone would know you were a loyal to the crown.  Thus, through conventional signals the English aristocracy made it highly expensive and difficult to fake an aristocratic persona.

While these two examples are separated by centuries Allen’s argument reinforces Donath’s argument that conventional signals are cheap for the proper signalers/receivers but can be expensive for imitators.  If a child was raised in an aristocratic family and grew up with all these special rules, it would be simple to live by them.  However, if an adult tried to learn all of these rules, it would be time consuming and difficult to copy all the idiosyncrasies.  This is just like the signature that Donath describes.  If you are a proper member of the group it is easy to know the code because you live by the code.  However, it would be very costly to a troll to learn the code and pass as a proper member of the community.  In this way Allen’s argument augments Donath’s claim that while conventional signals are not always reliable, a community can make them cost enough to keep intruders out.