Twitter communities are built on trust – sometimes too much trust. Recent XSS and XSRF exploits on Twitter have shown that the Twitter platform has been designed in a way that accidentally allows such trust to be used for evil purposes. My Fake Retweets experiment suggests that not all Twitter exploits are platform-level, architectural problems.

First things first: The real point of fake retweets is that they’re funny. What better way to make fun of your friends (or enemies) than to pretend to retweet stupid things that they allegedly said? I am not a performance artist, online or off.

Yet fake retweets do seem to say something worth saying about the medium. FRTs only work as a vehicle for jokes because there is a general assumption that all retweets are genuine. To some extent, this has nothing to do with Twitter. The only reason why jokes (or lies, or metaphors, or irony) work at all is because there exists a contrary convention that the jokester (liar, ironist) consciously flouts. In a world where people only tell lies, lies do not work in the same way that lies work in our world. I might lie about my dog eating my homework so that the teacher will give me an extension; but if there is no presumption in favor of truth, teachers will have no reason to grant an extension based on such a claim. (Echoes of Kant.) Jokes seem to work in a similar way: if we all spoke in puns all the time, for instance, then the utterance of a pun would have no element of surprise, robbing the joke of much of its value.

Thus the efficacy of fake retweets is at least in part an instance of a broader phenomenon. Anecdotally, though, it seems like Twitter is a particularly fertile yet underutilized environment for this kind of convention-flouting. With limited exceptions, people on Twitter generally seem to believe that everyone else is being genuine. There are some counterexamples, like Mark Sample‘s #MarksDH2010 or accounts like FakeAPStylebook. But both are either so absurd that no one could possibly think that they weren’t fake, or actively wear their fakeness on their sleeves with hashtags, or both. (This fact doesn’t necessarily take away from the funniness of the jokes in question. It just means that they don’t have the intent to deceive.) Aside from these sorts of extravagant Twitter charades, it’s hard to think of examples (from personal experience) where real live lying takes place on Twitter.

That’s not to say that my Fake Retweets were meant to deceive. (But they did. In one instance, a follower who commented on the fakeness of one retweet took another one as serious just a few minutes later.) The spirit of Fake Retweets in this case is to poke fun at friends, which means that my FRTs were mainly friendly and totally directed at friends. Yet I have to admit a little trepidation to the FRT, even with such benign content. There’s something about faking another person’s voice (perhaps especially in a community of academics) that seems to cross a sacred line.

The sense of violation in FRTs, it seems, is related to the fact that we all spend so much effort cultivating a specific persona via Twitter. Yet again, such cultivation is not a Twitter-specific phenomenon – surely there’s a sense in which personae are necessarily self-constructed – but it seems to be especially evident on Twitter. Maybe it’s because on Twitter, you control your own stream. In real life, all my eloquence and fashion won’t prevent the occasional piece of food in my teeth; in meatspace, there are infinite vectors for our self-constructed selves to get out of hand. Twitter, in contrast, has very few dimensions for self-presentation to run amok: tweets are finite in length and in number, you get to choose your avatar, you can spend hours crafting your 140-character pearls, you can even edit and delete mistaken tweets. The FRT threatens to cleave this controlled space, to taint our carefully manicured self-images.

The aspects of Twitter that make FRTs so uncomfortable aren’t necessarily bad things. Maybe the world would be a better place if it were as trusting as the Twitter community. (Though I wouldn’t want to be vulnerable to cross-site scripting in real life.) But certainly there is room for a little more skepticism when you see something come across your screen. Think before you click that link, before you believe that RT.