Playing Sherlock Holmes
The Boston bombing shows the advantages and risks of crowdsourcing. Karan Lahiri writes.india Updated: Apr 23, 2013 15:07 IST
It takes a particular kind of nerd to be hunched over his computer screen in Cambridge, Massachusetts, writing about the investigation into the Boston Marathon bombers. I was compelled to write about this, however, because this was the culmination of a high-profile investigation where the Federal Bureau of Investigation (FBI) explicitly sought crowdsourcing assistance in identifying two suspects in the Boston Marathon bombing.
While speculation was already under way on sites like Reddit and 4chan, the online community began minutely analysing the photographs provided by the FBI. This was not really like sticking ‘Wanted’ posters on a wall. What made it different? It isn’t technology, in itself, that is crucial here — though the ubiquity of sophisticated devices that capture images does help.
The distinction lies in how technology allows us to interact (and the norms shaping those interactions), enabling data-collection, investigation and analysis, not by an individual or identifiable pockets of individuals, but by a broad, undefined community.
So — what if Sherlock Holmes was not one person but many people? Apart from getting data from diverse sources, there is the sense that harnessing group intelligence gives you access to this problem-solving resource that is larger than the sum of its parts. We may be looking at a future where an investigative, online ‘hive mind’ proves to be an effective detection tool. The internet has, after all, been used to nab criminals in the past, like Aidan Folan, who was identified online, in connection with suspected involvement in a New York mugging.
There is, however, a murkier aspect to all of this group-sleuthing. Accuracy is a problem, but that is an issue even in traditional, State-run detective work. Online finger-pointing, however, looks a lot worse. Consider, for a moment, the effect on someone being falsely accused in an online community.
For the record, this has happened in this case, as an innocent high school student’s picture at the marathon, which drew attention on Reddit, wound up on the front page of the New York Post. From this angle, crowdsourcing looks less like the work of citizen detectives, and more like mob vigilantism.
The fact that this effort involves lots of people, and is located online, makes the effects of an error more serious, given not only the size of the audience, but also the nature of the accuser, which is essentially a large group into which individual identities are partially subsumed.
The State, for the foreseeable future, will have a lead role in investigation and capture. This is desirable initially, at least until suspects are authoritatively identified, so that crowdsourced investigation is focused instead of speculative. A small minority may be driven by voyeuristic impulses. Even in the real world, people tend to gape and gossip around crime scenes.
There are, however, positive impulses that fuel the desire to contribute to justice, and those impulses become apparent in the self-regulation of the online community. Take, for instance the subreddit called Findbostonbombers, which adapted itself to optimise the pursuit of its mission.
The ‘rules’ of this forum, which evolved with the ‘investigation’, stipulate that racism will not be tolerated, that vigilante justice is not condoned and that personal information should not be posted. Therefore, the crowd itself transformed the rules of interaction, impelled not only by the overall purpose it set out to achieve, but also, perhaps, by the noble impulses that caused people to participate in the first place, creating a moral membership that binds the online political community.
Karan Lahiri is a lawyer and currently pursuing his LLM degree at Harvard Law School
The views expressed by the author are personal