If you’re like me, you spend too much time – way too much time – these days looking at polling data. I ran across some interesting remarks by Foucault on opinion yesterday, which I’ll share here as a technique of distraction. He makes them in the context of a 1976 conversation with J. P. Barou and Michelle Perrot (whose work on resistance to disciplinary power he favorably cites near the end of the conversation) that was published as the preface to an edition of Bentham’s Panopticon writings. It appears as “L’oeil du pouvoir” (D&E #195, pp. 190-207 in my 2 volume edition) and is translated in Foucault Live (=FL). For context, then, the conversation appears in the year after Discipline and Punish. It covers a range of topics, including Foucault’s own path to discovering the panopticon (initially via hospital architecture, which had the dual need to see patients and keep them physically separated to avoid the spread of disease).
Foucault suggests that Bentham’s original work enjoyed considerable uptake in revolutionary France. Barou sets Foucault up: “isn’t it surprising to realize that the French Revolution, in the persons of people such as Lafayette, favorably received the project of the panopticon?” (D&E 195). Foucault answers that Bentham “is the complement of Rousseau” because the “Rousseauian dream” that “animated the revolutionaries” was of a “transparent society, both visible and legible in each of its parts, which had no more obscure zones, no zones established by the privileges of royal power or by the prerogatives of such and such body, or indeed by disorder; that each, from the point that he occupies, is able to see the entirety of society; that their hearts communicate with one another, that the gazes encounter no more obstacles, that opinion rules, that of each on each [celle de chacun sur chacun]” (D&E 195/FL 230). Opinion thus functions similar to what Foucault will later identify as a function of pastoral power (omnes et singulatim), simultaneously totalizing and individuating. “Opinion” is an object of “police” power in Discipline and Punish (213), that power that is concerned with understanding and managing the minutiae of daily life.
In the interview, Foucault suggests that Bentham is “both this [Rousseauian dream] and its complete opposite” because he understands the problem of visibility “while thinking of a visibility organized entirely around a dominant and surveillant gaze” (D&E 195/FL 230). Thus you have joined the “lyricism of Rousseau and the obsession of Bentham.” Foucault then suggests:
“A fear haunted the second half of the 18th century: there is a dark space, a screen of obscurity which provides an obstacle to the complete visibility of things, of people, of truths. [One needs] to dissolve the fragments of night which were opposed to the light, to make it so that there is no more dark space in society; to demolish the black rooms where arbitrary political power, the caprices of the monarch, religious superstitions, conspiracies of tyrants and priests, illusions of ignorance and epidemics [all] fomented” (D&E 196/FL 231).
In this context, the then proposes that “the rule of ‘opinion’ that we invoke so often in our age is a mode of functioning where power will be able to be exercised on the sole condition that things are known and that people are seen by a type of immediate gaze [par une sorte de regard immédiat] that is collective and anonymous” (D&E 197/FL 232).
Hold on to the word "immediate" – it will be important in a moment. But for now, notice the extent to which this is a pretty good description of nights spent gazing at opinion polls, both the obsession with knowing, and the fear that there are (say) hidden Trump voters like there were last time. More light! This sort of epistemology is driven by a fundamentally paranoid logic, as Wendy Chun emphasized a while ago: once I use some sort of optical device to see beyond what my naked eye tells me, that opens the fundamental possibility that a better device would see more. See a molecule with the microscope? Is that really the smallest particle? After all, there were molecules lurking below the surface that you hadn’t seen before. Here the paranoia is inevitable given a sampling process. The poll represents underlying opinions, and the process of sampling invariably “sees” only part of the data, even when it’s done exactly right. And that’s before you factor in questions about whether people who don’t have landlines are proportionally represented, or whether young people who say they will vote will actually do so, and so on. That’s also before you get to issue polling, which I complained about before the last election, arguing that it worked to create populations and truths about them that were fully the product of the polling questions, making it (for example) impossible to express views that didn’t fit the categories the polling firm wanted to hear.
In any case, as anybody who works in data will tell you, our access to data is fundamentally mediated by the process of its collection and description, which means that the immediacy we want – to just see clearly whether Trump has enough hidden voters somewhere in Pennsylvania to pull off the upset. We have to content ourselves with something like the fivethirtyeight statistical runs: crunch all the polling data (including weighting it by quality of poll), simulate the election 40,000 times (there’s a vision of hell: 40,000 occurrences of this election!) and see what happens (as of this minute, Biden wins 88% of the time). But the human brain just doesn’t like that: a pair of papers, one in Philosophy and Public Affairs, and one in the Stanford Law Review, by David Enoch, Levi Spectre and Talia Fisher make the case that we prefer juridical reasoning in the sense that we can explain why somebody is wrong, rather than living with the statistical truth (in this case, that Trump could win and there’s nothing wrong with the model at all. Enoch Spectre and Fisher use the example of liability for a bus accident).
All of which leads to the following remarkable paragraph. In response to Perrot’s suggestion that Bentham seems to overstate the power of his panopticon, Foucault says:
“This is the illusion of almost all of the reformers of the 18th century who lent to opinion a considerable power. Opinion is only able to be good when it is the immediate consciousness of the complete social body; they believed that people would become virtuous by the fact that they were observed [regardés]. For them, opinion was a spontaneous reactualization of the social contract. They misunderstood the real conditions of opinion, of the media, a materiality which is caught in the mechanisms of the economy and power” (D&E 204/FL 238).
And:
“They also failed to understand that the media would necessarily be controlled by economic and political interests. They did not perceive the material and economic components of public opinion. They thought that public opinion would be just by its very nature, that it would spread on its own accord, and provide a kind of democratic surveillance. It was essentially journalism—a crucial innovation of the 19th century—that manifested the utopian characteristics of this entire politics of the gaze” (D&E 204/FL 238)
Of course as Foucault notes, journalism comes with its own politics, and different forms of it come with different politics. I’ll close with two tidbits. First, one large study found that false news traveled further, faster and deeper on Twitter than true news, almost no matter how one measured diffusion. It’s the materiality of the news infrastructure and the business models of the platforms through which opinion is constituted. Second, there appears to be a horrible nexus of Instagram, Facebook, QAnon and multi-level marketing schemes that’s contributing to the diffusion of that conspiracy: it turns out that QAnon and multi-level marketing share a lot in common, and social media tends to facilitate that. I can’t describe it better, but the article is worth a read. The paranoid logic is baked in!
Recent Comments