∆ 3 SYNTHESIS — 1st Draft
At the start of Unit 2, my enquiry concerned the “originality” of the digital image: whether visual disintegration could give an image a final destination and carry it through a complete life cycle. This framed the image as a finite resource, in which permanence belongs to the origin and transience to the rendering. It soon became clear, however, that my interest lay less in physicalising the image than in defining the terms on which a finite image ceases to exist.
This shift was clarified through the concept of obstructed physicality. As Sbordoni argues, “representation is a principle that no longer explains the image and its technology away”, suggesting that the value of the image now lies in its reproducibility: images are converted into data so that viewing becomes quantifiable and extractable. Under these conditions, the gaze, which involves engagement and invites dialogue, has been replaced by the stare: endless, habitual, quantified and centred on the device. Obstructed physicality can therefore be understood less as an infrastructural change than as the prevailing condition of media consumption. This led to a more precise question: what conditions can an image set for its own viewing, so that it is looked at rather than consumed? I translated the qualities of the stare into counter-conditions (finitude, friction, unmeasurability, situatedness and reciprocity) and tested them through physical computing experiments.
Reading Paglen’s “Invisible Images” (2016) complicated this further. Images are increasingly made for machines: optimised to be clear, quantifiable and transparent enough to enter the dataset, they no longer simply represent the world but intervene in it. This prompted a second strand of research into the content of the image itself, asking how its subject matter might resist a machinic visual culture that favours seamless, high-resolution perfection.
Looking across both strands, I recognised a shared logic: each replaces machinic efficiency with human effort. In the first, the effort is imaginative: an incomplete or reduced image requires the viewer to fill in what is missing, keeping its meaning open. In the second, embodied in the imagotchi (Δ1), the effort is physical and temporal: the image resolves only when someone holds it and stays with it, and degrades when that attention lapses. Both enact what Paglen calls “deliberate inefficiencies”: images that slow down and ask something of the person looking.
Writing on Steyerl’s “In Defense of the Poor Image” (Δ2) exposed a tension I had not considered. By treating resolution as the measure of the image’s health, the imagotchi reproduces the class hierarchy of images that Steyerl critiques. Rich in resolution and singular in quantity, the image risks becoming a luxury object, defined by scarcity and exclusivity rather than care. At the same time, Steyerl’s “Mean Images” (2023) shows how mass production and constant circulation now feed the extractive economies of AI. Withdrawing a private image from circulation may protect it from extraction, but it may equally produce the rich image in a new form.
Triangulated, these projects reinforce my commitment to attention, situatedness and care, while untangling care from ownership, and loss from death. They leave me with a sharper question to carry forward: in a culture of hyperconnectivity and constant production, can an image move the viewer from stare to gaze while embracing, rather than resisting, its condition as a poor image? And can such care be shared without relying on circulation?
Leave a Reply