In comments below, Erik says:
I am sure it is easy to overstate the importance of Twitter or other technologies and I am equally sure that the empirics are murky. Theoretically, however, it is not at all implausible that better/faster communication technology helps spur the informational cascades that are necessary to solve the inherent coordination problems in these types of protests (see for example Susan Lohmann’s work on the Leipzig protests).
This seems to me to be fair enough (as I said in my original post, information on other actors’ intentions are crucially important to explaining collective action), and Lohmann’s work is a nice starting point (although maybe not for understanding putative Twitter effects). Like other accounts of collective action, Lohmann focuses on signalling and collective beliefs. The interesting difference between her account, and, say, Russell Hardin’s coordination theory of state power (discussed in his contribution to the Cook and Levi volume, The Limits of Rationality) is the kind of information that Lohmann believes affects collective action. As well as asking whether actors believe that others will turn out to demonstrate, Lohmann argues (here I simplify slightly) that actors’ willingness to turn out to demonstrate provides crucial collective knowledge over whether the regime is a good regime or a bad one overall. Every individual has personal information from her own experiences about whether their experience with the regime has been good or bad. However, in the absence of free communication it is very hard for individuals to aggregate this information so as to determine whether their experience can be generalized. Totalitarian societies – in which information exchange is highly circumscribed – make it hard for individuals to establish the overall level of content or discontent among their fellow citizens. In East Germany, only senior party figures had access to survey data which suggested that support for the communist regime was plummeting.
What does this mean for demonstrations? If lots of people turn out to demonstrate, then this may tell others (who may be sitting on the fence) that this really is a regime which is bad enough to protest against. If they then themselves turn out to protest, then others may in turn be inspired, and so on (Lohmann also discusses the mechanisms through which participation may start to dampen down). However, a lot depends on who turns out. Previous discussions of collective action, such as Hardin’s book Collective Action suggest that crazy extremists can be a spur to action. They can start the protest snowball rolling, by demonstrating that you can get out there and start doing stuff. Lohmann’s arguments imply that moderates are highly unlikely to be influenced by extremists, because these extremists would probably be turning out regardless of whether they had negative knowledge regarding the regime. Hence, they are informationally very nearly useless for this particular purpose. In contrast, when so-called ‘activist moderates’ – people who will turn out conditional on their private information – take to the streets, then this does provide useful information to fence-sitters, telling them that others have had negative experiences with the regime and perhaps (if the fencesitters also have had negative experiences) getting them to turn out too.
What does this mean for the Iran protests? On the one hand, one might expect this model to provide less insights about Iran than East Germany (the subject of Lohmann’s study). There are some reformist newspapers in Iran, even if they have to be cautious about what they talk about. Civil society is less circumscribed than it was in East Germany. So one would expect that people would already have some idea of what people in general think in their society, even if their understanding is likely to be systematically skewed and flawed in important ways. Mass protests will help persuade some people that there is something wrong – but the underlying set of beliefs over the merits and demerits of the regime in Iran is probably less dependent on fractured knowledge, and hence less vulnerable to information cascades, than it was in East Germany.
On the other, Lohmann’s argument may help explain one interesting aspect of the demonstrations: Moussavi’s insistence on keeping them peaceful, and his efforts to frame them in ways that build upon previous protests that were considered legitimate (i.e. the protests against the rule of the Shah), and on religious frames. This obviously makes it more difficult for the police to crack down on protesters. But it also may make the protests more persuasive to people who are unsure about what to think. The more difficult it is to depict the protesters as crazy extremists, the more likely it is that moderates will find their underlying case compelling. Hence, perhaps also, the efforts of the regime to depict the protesters as violent rioters, underwritten by Foreign Interests &c&c.
More generally, while it is surely true that informational factors matter, and that new communications technologies can affect the flow of information, a lot depends on your underlying models of how information matters, and which kind of information matters. If you were to assume that actors were motivated to collective action by different kinds of information, then obviously you would make quite different predictions. For what it is worth, I suspect that Twitter probably doesn’t do that much to enhance Lohmann-style information cascades, since a large majority of the target audience that may be shifted to participate actively probably do not use Twitter, or even know what it is. One could see it helping motivate collective action more easily on other accounts, which stress the difficulties of organization, coordination on specific courses of action, meeting points, etc…