The first thing that you need is usually power, which means a generator, as you will often be cut away from the regular logistical infrastructures. There are various reasons for this - the machine often needs to be set up in ruined or derelict buildings, in abandoned fields, or by stretches of rural freeway; it has to be absolutely mobile to function properly, and so you cannot rely on spatially fixed points of access to the power grid. So a generator and some method of moving it around are what you think about first once the decision has been made to put the machine to use. How much power will depend on what the plans are. You can get tiny man-portable models that will power a few lights and a heater and charge your phone, all the way up to enormous, humming industrial units designed to power entire building and built into shipping containers so that they can be moved from place to place by specialist trucking companies. If other people are involved you will also need to think about food and water, toilets, somewhere to sleep, somewhere to rest during operations, money for their time and their expertise. It is usually better to avoid this where possible, as something simple can quickly become very complex with the addition of other human workers.

Thankfully the machine has grown more and more autonomous over the years, as its processes and the products of its calculations have become both more complex and more abstract. These days it actually needs very little oversight. I remember that years ago it would function something like an architect or scene planner: you would be given a complex series of plans and blueprints, costings and contact numbers, and then you would do your best to act out all of the specifics with the aid of a small but competent staff. Things are very different now, although the final outputs from the machine are formally similar, even formally indistinguishable, to what was produced in the early days. What you are left with in each instance is a series of images. The machine could take no responsibility for their content, which often had a strangely arbitrary character, as though what was depicted could easily have been something else entirely - its only interest was their framing. The machine does not speak of course, it is simply a series of complex and complexly interrelated processes running in parallel, and it has no capacity to communicate. But I have noted its increasingly complex understandings of its role as a framing engine with great interest. I will try to explain.

Sometimes a young painter will have a few good ideas and no idea how to express themselves, and so will produce a series of fantastically interesting paintings that boil and twist beneath their surfaces; that are frustrated, neutered, bound, but that have some enormous vitality that remains trapped somehow beneath the actually-produced image. Sometimes the painters take the time to get better and are able to excavate and put to use this energy that would otherwise remain merely potential energy. Sometimes they remain stuck in this odd parallax position, producing paintings that are not themselves interesting, but which nonetheless have this frustrated, bound, constricted quality. Paintings that feel like pressure cookers or unexploded bombs, like bodies frozen into postures that do not suit them.

The machine had some of this feeling to it in the early days, as though it was struggling to express something true in its own nature. Since its nature is alien to me it is difficult for me to say whether it ever succeeded. As has been mentioned, the images themselves were never the focus of its attention, and they had then and still have something arbitrary about them. But the processes of framing grew exponentially more complex as the years passed and our professional reputation grew. In the beginning I could follow the logics without much difficulty. The image would find its proper context, and would be liberated in this. The frame was not the image, but it allowed the image to speak.

I remember watching over time as the machine began to insist in clearer and clearer terms that no useable distinction existed between process and frame. Everything that allows an image to come into being is a part of this holistic and disaggregated supporting infrastructure. This formulation was also comprehensible to me, although of course its implications were fractally complex, and I felt even then that they would need some arbitrary ex nihilo grammar or bounding to be useable. But this was how the machine began to understand its tasks over time. The conditions and affordances of the image would be incorporated into their technologies of display. The most interesting upshot of this shift, at least to my amateur sensibility, was that the machine began to understand itself as materially composed in these same processes; if there were logistical preconditions on the production of the image, then it was these exact preconditions that made up the body of the framing machine. My own assistant work during this period become less about setting up the small mobile devices (lights, heaters, projectors) that we used to compose a scene, and more about finding access points through which the machine could gain access to new and powerful temporary limbs. Power cabling, sewerage and water piping, gas mains, but also demographic information, the movement of populations, the distribution of food, medical care, income, etc. All of these became the constituent organs of a body, and this body would birth new bodies and grow new limbs as required. It would grow and shrink. It could grow to the size of a city for the fractions of a second that it required to make some small adjustment to one of its constituent processes, adjustments that I imagined would brush against the lives of millions of people. And then it would shrink down again, almost to disappearance, a series of directions or diagrams playing across the screen on my phone or computer. It was only after fully encircling its subject, its whatever-image, its arbitrary image, within thousands of these intangible enclosures, that the machine could produce its authentic frame. The operations were not dramatic; they happened quickly and in silence. I was responsible mainly for facilitating the material connections that we would need to use, and for the face-to-face work of handing off both image and frame to our clients once they were produced.


Now I am walking over soft, overgrown earth, trying to find level ground on which I will be able to set up the various heavy tripods and hard surfaces that the machine requires to be set up. It is late afternoon and there is no wind anywhere. The air is warm and smells strongly of grass and damp earth. A million tiny insects skim the surface of the thick mat of vegetation and catch the late sunlight with their bodies. Grass and weeds and wildflowers are violently interwoven, bursting upwards, crowding for room, stretching for the sunlight. In front of me is what we have come to frame, and behind it is an enormous lighting crane that we have had delivered to facilitate our work. A jeep with a generator built into its chassis sits next to the crane, and thick cables link the two vehicles. There are about eighty enormous stage lights attached to the tip of the crane’s enormous black arm, organised in parallel series on a steel grid structure. All of this is equipment has been abandoned on our instruction in this anonymous, neglected field that stretches away from the freeway, about forty minutes out from the outskirts of the capital. It does not take me long to set the machine up and hook it up to the generator. I am well practised. It sits inert in the long grass that is shading golden as the sun sinks lower, and I lie down next to it, waiting, a passive body, also inert, without a single thought in my head.


This is simply the best approximation that I can think of; there may be others that are more precise. The frame is like a tunnel or a pipe. The image is a vertical slice into the pipe. The machine builds these pipes and maintains them, and withdraws images from them at various lengths and times, like Palaeolithic core samples from sheets of arctic ice that stretch down kilometres beneath the surface. It arranges itself always with the intent to make more pipes, take more samples, arrange these correctly (in some series that I do not understand and cannot hope to reverse engineer), and finding new materials to pump through itself. The viscosity, acidity, toxicity etc. of these liquids and semi-liquids all change moment to moment, and each requires its own specialist infrastructure. The vertical slices take on the characteristics of whatever is pumped through the piping systems, which allows the images themselves to inherit some residue of viscosity, acidity, toxicity, etc.


It is darker now, and not as warm. I watch the sun slowly dip behind the horizon. As it disappears completely from view the lighting rig switches itself on with a great humming sound of electricity. What happens is this: for a given area around the subject, maybe twenty metres on each side, the light from above is exactly identical to that of the sunlight in the middle of the day. It comes down vertically, and there are no shadows. Around this bright, friendly zone the evening reasserts itself at a hard border. After ten seconds or so the light switches off, and I begin the slow, laborious work of packing the various pieces of the machine back into my car, which is parked not far, behind some low trees, tucked away from the sight of the road.

Commissioned for I Can't See a Thing/I'll Open That One, April 2023, K4 Gallery, Oslo.