Anonymity Online

The aim of our project was to discover the ways in which online anonymity, or the lack of it, effects behavior. In the physical world, anonymity plays a large role in what we do and how we perceive consequences. It can be observed that when people feel anonymous, that is unidentifiable, their actions become less constrained. People behind a mask become more aggressive, truthful and generally careless in the things they are willing to do or say. The logic is simple; if an action cannot be linked to an individual, than no individual can suffer the consequence, and if there is no consequence to a given action, then there is no reason to refrain from said action. We wanted to see if this principle applied to the various realms of the internet, where anonymity exists in varying degrees. The underlying question that our project tries to understand is, ‘can the devices of the internet provide a great enough feeling of anonymity to alter our behavior?’ In our philosophy class we have learned that all information submitted to the internet can be recorded and traced back to the individual, or at least his or her IP address, from which the information was submitted. In this sense there can be no true anonymity on the net. However, there can still be the perception of anonymity, and there can be varying degrees of anonymity. Our project tests the degrees and hopes to find a correlation between higher levels of anonymity and more careless behavior.

Our strategy for testing this hypothesis was very simple. We chose four different socially interactive websites from which we would generate a comparative analysis. We intentionally chose sites which we believed offered substantially different degrees of anonymity. On one end of the spectrum, we chose Facebook, which offers nearly no amount of anonymity at all. A user profile is typically composed of their real name, a number of pictures of their physical personage and even their likes and interests. The facebook user is as near a living, breathing, identifiable person as is possible on the internet. On the other end of the spectrum, we chose 4chan, which offers nearly absolute anonymity. On 4chan, the mechanism for interaction is a simple image board which is continually renewed as anonymous users post images and text and often text responses to those images. There are no fixed profiles on 4chan and posts are differentiated merely by a long number, so users have no idea who posts what. It is even against 4chan policy to reveal personal information in images or text. At levels of anonymity between facebook and 4chan, we chose to study Tumblr, an image board consisting of users with profiles that can be followed and Reddit, a similar platform also consisting of user profiles. To accomplish our comparative analysis, our method was to post controversial topics, which would be consistent on all four sites, so that we could compare the responses of the users. In doing so, we hoped to find a different variety of answers on each site which may be related to the level of anonymity that the site provides.

The execution of our project turned out to be more complicated than we initially thought. We obtained a considerable amount of data from Facebook and 4chan, but Tumblr and Reddit offered much greater resistance. On Tumblr, the likelihood that someone will respond depends heavily on how many followers the posting profile has. On Reddit, the likelihood for response depends on a rank system in which individual posts can be voted positively or negatively. If the post doesn’t receive enough upvotes quickly, then it will die in its infancy, as was the case in our experiment. Between Facebook and 4chan, however, we believe we had obtained enough data to form at least a vague conclusion. Our subjective analysis revealed that users on facebook were generally more likely to respond in ways which protect the identity and users on 4chan were more likely to post responses with a more pronounced disregard of offensive language and content. From the information gathered between these two poles, the one of total identity and the other of total anonymity, we conclude that in general the levels of anonymity afforded by the internet do have the ability to change the things we say and do.