As part of my dissertation work on media ontology, I've been wondering about the ideal audience for an autographic reproduction. Gravid with information, it seems to presuppose a bionic sensorium: an artificial eye capable of processing wavelengths in the electromagnetic spectrum that are invisible to its human counterpart; an eye so acute that, like Borges' Funes the Memorius, it can perceive vineyards and grapes in a goblet of wine. Such an eye is delphic, temporally optimized for discerning photonic information about the past or future in the present. My point is that the excesses of the autographic are simply lost on the human visual system, which, as Rudolf Arnheim reminds us, is highly “purposive and selective.” Our eyes scan an image via a series of rapid movements or jumps called saccades, whose purpose is to continually update vision by registering a new area of the visual field onto the depressed part of the retina known as the fovea. Visual continua, then, are processed digitally and temporally: we sample an analogue image over time rather than perceive it _in toto_.
In an experiment jointly sponsored by the National Gallery, London and the Applied Vision Research unit of the University of Derby, scan-paths and fixations of participants looking at Paul Delaroche's 'The Execution of Lady Jane Grey” on a computer screen were recorded using state-of-the-art eye-tracking software. The experiment ingeniously demonstrates that the painting contains a number of visual lodestones that serve to direct the viewer’s attention. Figures and especially faces function as hotspots toward which the eye gravitates. The search patterns are so regular and iterative that they appear like grooves in vinyl or lines etched with acid. As conspicuous as the scan-paths are, however, the vast arid regions of untrod visual ground are even more so. There is an optical debris field here that is impossible to ignore.
Autographic reproduction, however, is patently indifferent to this discriminatory power of the human eye-- indifferent as well to its meaning-making powers. The human visual system is an interpolative machine capable of filling in missing detail, integrating visual information across saccades, and actively producing the world it perceives. Indeed the predictive and generative behavior of both eye and ear have spawned a generation of scientific R&D that computationally models visual and auditory processing using state-of-the-art probability algorithms, notably Hidden Markov Models (HMMs). Wordsworth’s poetic formulation antedates the computational one by almost two hundred years: “all the mighty world / Of eye and ear, both what they half-create, / And what perceive.”
Posted by karik at October 13, 2003 2:21 PM | TrackBackThere are more ideas in these three paragraphs than in most books I read, and the kind of interdisciplinary authority on display in every sentence is a glimpse of what a 21st century humanities can really be. I'm particularly struck by the description of the human vision system and the essentially digital nature of visual processing. There's a tendency, I think, to conceive of the digital as artificial and the analog as a state of nature; the analog/digital distinction is very much played out as a nature/culture binary. Here we _see_ (so to speak) that the eye is physiologically and fundamentally a digital signal processor.
Thanks, Matt :-) I'm reminded, too, of Donald Knuth: "Human eyes are inherently digital, made from individual rod and cone cells" (From _Digital Typography_ 1999).