Tragedy, Social Networking, and Online Collaboration

Our experiences of large-scale tragedies are somewhat different in the information age than they were previously.  The recent Boston Marathon bombing and the subsequent manhunt are the most recent examples of how our investment in the internet makes both real-time experiences during such events and the communal recollections thereafter more accessible and comprehensive.  The range of real-time narratives from affected areas, the range of conjecture in public discourse, and the collaboration amongst people working to understand what exactly happened have all increased and become more apparent.

Zuckerberg’s law can reveal how our experiences of tragedy are quite complex.  Zuckerberg’s law states “that we share twice as much every year as we did the year before” and manifests “a sweeping cultural force” (Buchanan, 2013).  Accordingly, Zuckerberg’s law expands the diversity of coverage in a given tragedy.  Those following the recent events facing Boston online were exposed to everything from available lines of police communication during the manhunt to a Bostonian’s article on Esquire of being under lockdown in the apartment of his one night stand the night before.  The availability of this range of narratives due to the mass amount of content we share online reflects the varied dimensions of a tragedy and how they affect those at the scene.

This contrasts the exchange of information during past tragedies.  Tragedies televised live, such as the Kennedy assassination, the Challenger explosion, and 9/11, introduced immediacy to media coverage of tragedy.  However, the proliferation of ICTs extended this immediacy in coverage to citizen journalism.  The Virginia Tech shooting was one of the first to exemplify this, as people read posts, tweets and texts from those on the Virginia Tech campus in real time (Halbrooks, n.d.).

This increased sharing also augments online speculation surrounding tragedy.  While many turned to the online profiles of the suspects in an attempt to find anything telling in hindsight, Buchanan (2013) notes that one’s online presence only shows one side of an individual, rather than giving a holistic sense of a user:

For all of the unbridled interiority that social networks suggest they convey, social media does not turn people inside out; it is a single side of a person, the face they put on for the rest of the world. It is worth looking at that face for the same reason we look somebody in the eye, hoping that some bit of what’s really inside slips through.

The growth of online content magnifies the amount of context that users can accrue in their searches.  However, due to the reductive nature of content supplied through one’s social presence, this can lead to more guesswork than fact-finding on the part of users.

Speculation is also seen in the online communities that tried to identify potential suspects through publicly available videos and photographs.  Reddit and 4chan users both tried to do so within the respective online communities of those sites.  Regarding Reddit specifically, Surowiecki (2013) claims that while user endeavors were ultimately unsuccessful, this only mirrors the conjecture that takes place during such investigations, rather than any deficiencies in their online approach:

You can certainly fault the Redditors for not recognizing the limits of their own knowledge and for jumping to conclusions (even if a good deal of that jumping was done by the national press). But this isn’t . . . specific to Reddit—on the contrary, official investigations fall prey to it all the time.

Thus, the speculative nature of online communities in these instances is no different than that of law enforcement, but the transparency of it on online platforms underscores the increased access users now have to the uncertainty surrounding such searches.  So, while online collaboration occurs during modern tragedies, it also plays into the widespread conjecture facilitated online in their aftermath.

All of these trends show that we experience tragedies differently as our investment in the internet as a platform for sharing increases.  It is important to note that this is not always favorable – in the case of the Boston Marathon bombing, “spammers are sending emails with links to either sites showing videos of the bombings” wherein “[a]s the unsuspecting recipient watches the video, his or her computer is infected with viruses” (Kleinman, 2013).  As seen in this example, many of the negatives are remarkably senseless, but at the same time, they are not unique to tragedies.  Spam, certainly, is considered commonplace.  Ultimately, it is through the trends considered here that the changing dynamics in public reactions to tragedy are the most pertinent and intriguing.

 

Works Cited

Buchanan, M. (2013).  Facebook, Twitter and the Tsarnaevs.  Retrieved from http://www.newyorker.com/online/blogs/elements/2013/04/facebook-twitter-and-the-tsarnaevs.html.

Halbrooks, G. (n.d.).  12 events that changed how media outlets cover news.  Retrieved from http://media.about.com/od/mediatrends/tp/12-Events-That-Changed-How-Media-Outlets-Cover-News.htm.

Surowiecki, J. (2013).  The wise way to crowdsource a manhunt.  Retrieved from http://www.newyorker.com/online/blogs/comment/2013/04/reddit-tsarnaev-marathon-bombers-wisdom-of-crowds.html?utm_source=dlvr.it&utm_medium=twitter.

Kleinman, A. (2013).  Spammers spread viruses in wake of Boston bombing.  Retrieved from http://www.huffingtonpost.com/2013/04/24/spammers-boston-bombing_n_3148270.html?ref=topbar.

Ned Prutzer

Ned Prutzer is a former CCT Graduate Student.

  • Dylan Bargteil

    I can’t locate it now, but I think like you say (and has apparently been the subject of some study) these intersections of social media and law enforcement really show how arbitrary and uncertain law enforcement procedures are. There was a great opinion piece by a technologist about transparency and randomness in law enforcement, but I unfortunately can’t find it or remember who the author was. If I remember though, I’ll be sure to pass it along.

  • mylisha richard

    This post is really great. Thanks for sharing this nice post with us. @ Love being an internet junkie