Laws restricting the right to peaceful protest are large-scale legislators throughout the United States. Traditionally, constitutional scholars have developed this type of policy as a test of the balance between freedom of speech and public security. Is there an empirical basis to assume that the former is threatening the latter at this time? That is what this message is about.
This spring, the Governor of Florida overwhelmingly overcame one of the country’s most restrictive terms, proudly taking the country’s strongest anti-riot law enforcement measures.
“If you riot, rob, hurt others, especially if you hurt law enforcement during these violent meetings, you’re in jail.” – Governor Ron DeSantis
I don’t think sot is controversial. However, the presumed assumption – a violent demonstration is broad enough to require a strong dismissal – does not appear to be reflected in the data.
Dr. Erica Chenoweth is one of the country’s leading investigators of nonviolent demonstrations. According to his research team, a publicly available data set About the U.S. demonstrations I use here. They acknowledge that the data are incomplete and incomplete – but I think they work well in this analysis of ours. We use it to form our own scale using the Bayesian theorem.
But why Bayes’ theorem? I believe Bayes provides an excellent tool for this question because it maintains uncertainty in a way that often representative models do not. Unlike a mechanistic process that can only produce a narrow range of outputs, Bayes offers the flexibility to adapt to the fact that we don’t have a “perfect picture,” much like we naturally approach problems.
If I went to the demonstration tomorrow, I have no idea what will happen. I accept that uncertainty – but it doesn’t stop me from subconsciously thinking about what could go wrong. Tempers can shine. The person next to me could hit the police. The one behind me was able to blow up the window of an office building. We could have been shot – or we can all give our voice to the subject of our concern peacefully and without events. Everything is credible.
For those who have attended large demonstrations, we intuitively assume that while a few people may get out of hand, most follow socially acceptable behavior and are somewhat moderate. Bayes makes it useful that it is designed to include that assumption in the analysis; from the beginning, we have gathered new information, integrated it into our knowledge, and approximated a probability value that is closer to reality. We can start wild with different assumptions, but we end up pretty close to each other.
It is intuition. Let’s now define it in more detail with a Python implementation.