Hi all, and sorry for the delayed posting! Here are some thoughts on data, epistemology, power, and Baudrillard to spice up your Saturday morning.
Reading Hamish Robertson and Joanne Travaglia on the connections between the “data revolution,” which swept 19th-Century Europe as the agencies of a newly centralized and bureaucratized state set out to better understand and control rapidly expanding urban populations, and our own crisis of “big data,” I felt like I had found the response to Baudrillard I had been looking for. What I most appreciate in this piece is the fact that, while certainly critical of the ideological functions of contemporary big data discourses, Robertson and Travaglia resist the urge to simply jettison the phenomenon altogether, excising it from the province – even the possibility – of meaning. Rather, taking a more historiographic approach than Baudriallard (whose main historical reference seems to be his own now-revised opinions on the meaningful/less-ness of mediated images), they give themselves space to sit for a moment with the weird politicalities of big data, thinking carefully about 1) what ‘big data’ actually represents or allows us to access, and 2) the specific institutional formations (the disciplin-ification of the contemporary university, the sponsorship of the state, etc.) that helped it to take shape.
Where Baudrillard was emphatic in his belief that representational technologies like opinion polling represent absolutely nothing – that they are properly the objects of simulation, rather than of meaning; objects that narrow the field of agency and resistance to ironic subversion and mocking laughter – Robertson and Travaglia rightly note that, even if big data do not represent what we think they do, they nonetheless represent something, and this something is certainly something worthy of our consideration: “That a census or a social survey is a snapshot of the way our societies are regulated is rarely remarked on and instead emphasis is given to the presumed objectivity of the categories and their data. This is the ideology of the small data era in action – the claim that it is science and not society that we are seeing through such instruments.” In other words, even if the referent of population data is not the population itself, we are still dealing with reference and meaning; we are glimpsing not a population in its totality, but the various ways in which that population is defined, managed, and governed.
For those of us concerned with the contentious social and cultural topographies of data – say, those of us in this class – this is something I don’t think we can afford to overlook. Our data collection activities might not, in the end, tell us that much about the objects we want to study, but they do offer an important opportunity to glimpse the ways in which we as researchers imagine our objects, or more precisely, how we imagine them to be accessible and available within the parameters of academic knowledge production; to reflect on how we attempt to locate ourselves within (and potentially without) the conditions of our scholarly formation. These are questions of epistemology, about knowability. Quite appropriate, then, that Robertson and Travaglia seek appeal to Luciano Floridi: “Floridi writing on the philosophy of big data, has said quite specifically that the real big data problem we face today is less one of the quantity or quality of data or even technical skills but rather one of epistemology.” If, as Robertson and Travaglia note, “a great deal of social data is coercive in nature” – or at least ensconced within the particular ways of knowing we inherit from 19th-Century European social science, and thus intimately bound up with the pathologization, governance, and in many cases eradication of certain populations – I think it behooves us to aspire to more than Baudriallard’s mocking laughter when confronted with the massive and admittedly overheated discourse of big data. We need to find where our data sets touch the world, and consider what that touching might tell us about the politics of knowing, even and especially when it “fails” to represent the world.
FAMST248 - Data Visualization as Media Praxis 
Hi Tyler,
I very much agree with your observation that “even if big data do not represent what we think they do, they nonetheless represent something, and this something is certainly something worthy of our consideration.” Sometimes it almost feels like Baudrillard is this academic boogeyman that we as media scholars are constantly having to ward off with justifications for why media isn’t all worthless. But for me, I think the real boogeyman here is just the temptation to take scholarly shortcuts, or to mask easy assumptions with revolutionary language. I would even go so far as to say that making sweeping gestures such as declaring everything a simulation and then disavowing the value of those datalogical objects is a form of laziness, or an excuse not to think more deeply about data and its implications for our lived world. There’s a lot of fatalism in academia, particularly at a time when so much physical and cultural violence is happening, but at the end of the day I have to ask: what’s the point of fatalism? I believe that the best scholarship sees the work done by academics not as mere outsider commentary but as an active form in participation, and that fatalism, really, is a refusal to participate.
LikeLike
Hi Tyler and Lisa,
I really agree with a lot of what both of you are portraying here. I think that these questions are crucial ones that deserve more time and thought than often we give them in scholarly work. I’m going to push this fatalism observance a little further, though, and tie it back into some of the strands Tyler’s picking up on in his piece. I think that a more important question than “what’s the point of fatalism?” would be “why is fatalism in academia happening?” and “what are the institutional conditions of academia that propagate fatalism or breed these feelings of non-participation?” After all, I feel that everyone that makes the commitment to being an academic has some sort of call-to-arms in mind at the beginning of their careers, but there’s something happening within the institutions that leaves them jaded along the way. Is it the corporatized nature of publishing that pushes forward the “revolutionary”/fatalistic works as the best ones? Or is it insidious to the scholarly community itself? If so, what pressures are leading to this trend? Is it really laziness, or are there particular situations that breed fatalist thought? In any light, I think that tying this important line of inquiry back to the agenda that our authors set out on this week may provide some insight as to why this trend is occurring. Shifting the focus this way could point to institutional reasons for this trend in academic discourse, as well as shed some light as to perhaps some epistemological ruptures that speak to this disavowal of datalogical objects in our field.
LikeLike
Hi Tyler,
Really great insights and I really appreciate the way in which you not only spoke about Hamish Robertson and Joanne Travaglia’s article, but also pretty much summed up many of the problems associated with data discussed in the entire quarter by stating that “(…)even if the referent of population data is not the population itself, we are still dealing with reference and meaning; we are glimpsing not a population in its totality, but the various ways in which that population is defined, managed, and governed.” Not only that, you also provided a thought to what the agency behind data studies should be in “We need to find where our data sets touch the world, and consider what that touching might tell us about the politics of knowing, even and especially when it “fails” to represent the world.”
Using the words of Walter Benjamin, it seems that the distractibility of new media has been affecting scholarship that only focus their attention on peculiarities and difficulties of social media data to represent bodies instead of broadening the scope to include epistemology and what it means to reduce bodies into hashtags.
LikeLike