Is Photography Over? | Trevor Paglen | Monday, 24.03.2014


In the last post, I proposed that 21st Century “photography” has come to encompass so many different kinds of technologies, imaging apparatuses, and practices that the kinds of things we easily recognize as photography (cameras, film, prints, etc.) now actually constitute an exception to the rule. I proposed a much broader definition – seeing machines. The point of having such an expanded definition is to help us notice and recognize the myriad ways in which imaging systems (including traditional cameras), and the images they produce, are both ubiquitous, and actively sculpting the world in ways that were unimaginable just a few decades ago. Moreover, I proposed that classical photo theory is of little use, and may indeed actually hinder, a broad understating of contemporary imaging systems.

In this and the next few posts, my aim is to sketch out some ideas I use in my own work to try and make sense of seeing machines and their role in contemporary societies. Although I will eventually come to a discussion of images, my more immediate task is to begin understanding seeing machines by thinking about the kinds of “work” they do in the world. This is obviously a multi-faceted question that can, and should, be addressed at multiple geographic and temporal scales, which I will explore in upcoming posts. For today, I will begin with something I call “scripts.”

I think about a “script” as the basic and obvious function of an imaging system, its “style” of seeing, and the immediate relationships (between seer and seen, for example) it produces, and the obvious ways in which a seeing machine sculpts the world. To put it crudely, a script is all of those things that a given seeing machine “wants” to do, how it “wants” to see the world, and how it does what it’s designed to do. A machine’s script strongly proscribes certain activities and relationships, and at the same time precludes others. There is a two-fold question here, “how” a seeing machine sees is utterly bound up with the effects that it produces.

Let’s take an example, one that James Bridle has written about at some length, namely the increasingly ever-present Automated Number Plate Reader (ANPR) imaging systems. An ANPR camera does something very simple: it takes a picture of automobile license plate that passes through its field of view (using a standard shutter speed of 1/1000 and sometimes using infrared illumination), and processes that image to make it readable by optical character recognition software. Finally it stores some combination of the original photograph and metadata in a database. The entire point of an ANPR system is to end up with an easily storable, searchable database of images and information that, in many cases, can be correlated with other data points, or tasked with automatically taking an action in the world if certain conditions are met (i.e. if a car is speeding, the back-end software might automatically issue a ticket to its driver). My point here is that an ANPR system is designed to “see” the word in an extremely specific way and do specific things with the images it makes. These specificities are what I think of as its “script.”

A lawsuit currently making its way through the courts in Utah is instructive here. In the United States, a company called Vigilant Solutions, Inc. operates a private, nationwide fleet of car equipped with ANPR cameras. Their cars roam the country, creating a continually-updated database of where particular vehicles were are particular times. The database currently holds about 2 billion records and grows by about 100 million records each day. Vigilant’s business model involves selling this location information to banks, law enforcement, insurance agencies, and other clients.

Wary of the privacy implications of such databases, the Utah legislature recently passed a law outlawing private companies like Vigilant from collecting ANPR data (police agencies are exempted from the law). Vigilant Solutions and its partner Digital Recognition Network, Inc. responded in February 2014 by suing the state of Utah, claiming that its right to constitutionally protected activities was being infringed. Arguing against the Utah law, Vigilant pointed out that the legislation “arbitrarily prohibits an activity that is protected in all other settings.” Vigilant’s lawyer Michael Carvin compared his clients activities to those of any other person walking around shooting pictures with a digital SLR: “any citizen of Utah can walk outside and photograph anything they please, including a license plate.” On the surface, Vigilant’s argument makes sense – after all, the company simply drives around and takes pictures, which isn’t illegal. But, as far as their respective scripts are concerned, a Vigilant ANPR system is quite a different seeing machine than a Nikon DSLR, and taking pictures with a Vigilant system has very little do with taking photos with a camera designed to take pretty pictures for humans to look at.

The broader point I want to make is that every seeing machine, every camera, whether it’s an 8x10 field camera loaded with Fuji Velvia or a COBRA BRASS infrared and gamma-ray detector installed onboard an ORION signals-intelligence satellite in geostationary orbit, has its particular scripts – the range of activities that it “wants” to do and a range of activities that it cannot do.

My interest in introducing the idea of scripts is to begin developing a way to identify, or to “see,” the ways in which seeing machines create cultural, economic, and political footprints on society at large.

But scripts are really only the beginning of this – seeing machines have much larger footprints than the immediate scripts they perform. If we begin to understand seeing machines in terms of the larger networks of relationships they are produced-by and productive-of, our frame of reference quickly and dramatically expands into what I think of as “geographies of photography.”