Fighting Fire With Fire: Using Technology to Battle Cyber Bullying

Recently New York State has passed the Dignity for All Students Act with a goal of addressing the growing problem of bullying in high schools. Part of the plan for eliminating bullying involved a six hour training course of all school employees on how to properly avoid and address various bullying situation. I, being a volunteer high school coach, fell under this umbrella and attended the course a couple of weeks ago.

Predictably, part of the seminar focused on cyber bullying. Much of what the instructor was saying during this discussion caught my attention as it sounded awfully familiar to our discussions from class.  She explained that while cyber bullying isn’t something that happens frequently on school grounds, it is far more detrimental to the victim due to the public nature of the humiliation over the internet. I thought of this during our discussion of Professor Schacht’s interview on Digital Thoreau being spreading further than expected across the internet. This same wildfire concept applies to cyber bullying.

I dug further into the issue and found a CNN article that sparked my interest. According to the article, students at MIT were looking to formulate a program that would detect bullying language. It is similar to the programs Kirk Anne discussed in class that would find and count the amount of times a word appeared in an article. The problem with this solution is lack of a way to detect sarcasm. These programs only work by sifting through the words on the page without considering their intent. For example, instead of posting “that’s an ugly haircut” which would be flagged for “ugly,” a teen could instead post “wow, great haircut” and have the same malicious intentions.

Difficulties reading sarcasm over the internet is certainly not a new revelation. The term Poe’s Law was coined to explain that without a clear description of the author’s intentions, it is impossible to determine whether their expressions were “sincere extremism” or “parody extremism.” Of course, no one ever thinks that what they say has any consequence. Unfortunately this isn’t true with the internet being, as we discussed, open to the whole world. The case of Justin Carter, who was arrest for what he perceived to be sarcastic remarks, spent months in jail for what was read by many to be violent threats against his school.

This raises the question of the future of internet security. With bullying, harassment, and threats becoming so prevalent across the web, what kinds of restrictions should we expect to see in the future? If what we say is indeed sarcasm, wouldn’t censoring our work be infringing on our first amendment right to freedom of speech? And how, if at all, will this affect us as Digital Humanists?