The Arab Spring: a simple compartmental model for the dynamics of a revolution (Q2453824)

In the paper there is formulated a new mathematical model for revolutionary events that aims to explain the dynamics of peaceful revolutions. For testing it is applied to the Arab Spring revolutions in Tunisia and Egypt, taking into account the influence of the Internet and new media on the visibility of the revolution and the ensuing reduced efficiency of censorship. Formally, the model can be presented as follows. Let \(r(t)\) be the fraction of protesters or revolutionaries in the population at time \(t\). The model is given by a single differential equation for \(r(t)\): \[ \dot r(t)=\frac{dr}{dt}=\underbrace{c_{1}v(r;\alpha)(1-r)}_{g(r)}-\underbrace{c_{2}p(r;\beta)r}_{d(r)}, \tag{1} \] where \(\alpha,\beta \in (0,1)\) and \(c_{1} ,c_{2} >0\) are parameters, and where the functions \(g, d:[0,1]\rightarrow \mathbb{R}^{+}\) are called the growth and decay terms, respectively, since they model the growth and decay of the fraction of protesters. Considering rapid revolutionary transitions on a short time scale (on the order of months), and expecting that the regime is very unpopular, some simplifying assumptions are made. Particularly, the model neglects demographic and other long-term effects. These two assumptions imply a constant population that can be divided into two compartments: the population participating (\(r(t)\)) and not participating (\(1 - r(t)\)) in the revolution. According to the second assumption, the fraction of the population potentially willing to join the revolution at time \(t\) is \(1 - r(t)\). The function \(p(r; \beta)\) takes the form of a step function, because the authors assume that the regime loses availability to punish once the number of protesters exceeds the regime's police capacity. The parameters \(c_{1}\) and \(\alpha\) describe aspects of grievance, but the parameters \(c_{2}\) and \(\beta\) can be related to aspects of political opportunity and state capacity. The mathematical classification of the different types of dynamical behavior that may occur in model (1) proceeds case-wise by considering the parameter regions \(\alpha +\beta =1\), \(\alpha +\beta <1\), and \(\alpha +\beta >1\). Thus, we confirm that approximating the rapid transitions in the visibility and policing terms by step functions is justified by greatly simplifying Eq. (1) and its analysis. The computational experiments on some countries that could potentially be classified according to the above interpretation are discussed. As the situation in a particular country evolves the regime may pass from one parameter region to another. The authors discuss how adoption of new media may affect the parameter region of a country. Simple models like the given one have the advantage of relying on just a few basic assumptions about individual and communal behavior.