Friday, March 07, 2008

Notes on Transmateriality

At the recent UTS symposium I gave a short presentation titled "After Inframedia: Presence and Transmateriality." The presence stuff I covered earlier, but the second idea - which I touched on very briefly in this 2003 paper - is much less developed. So here goes.

The relationship between matter and "information" or "the digital" has been a recurring theme in new media theory for more than a decade. We could sketch it very roughly as follows. In the early to mid 90s, as digital hype was gathering pace, artists and cultural theorists began to critique the apparent drive towards disembodiment in technoculture. Simon Penny's 1991 text "Virtual Reality as the end of the Enlightenment Project" is a good (and early) example, even if VR now looks a bit like a straw figure in these critiques. This critical project of grounding the digital in the material (and the body) has continued. In 2000 Felix Stalder wrote of the "ideology of immateriality" underpinning the so-called "new economy." Around the same time Katherine Hayles published a more complex investigation in How We Became Posthuman, asking "how information lost its body" but also considering the inevitably embodied effects of this supposedly immaterial stuff (this is well covered in her paper The Materiality of Informatics).

Hayles introduces a conceptual pair: inscription and incorporation. Inscription is "normalized and abstract ... a system of signs operating independently of any particular manifestation" [Posthuman 198]. Inscription refers to the properties of a text, for example, that can be transcribed without regard to its specific embodied manifestation - digital computation thus relies on inscription, in moving patterns of data through various substrates. Incorporation is its flip-side, referring to the inescapably embodied aspect of a sign. Both inscription and incorporation are verbs - practices or processes - rather than ontological states; and they oscillate, a bit like presence and meaning for Gumbrecht: "incorporating practices are in constant interplay with inscriptions that abstract the practices into signs" [199].

Today I came a cross a more recent paper by Matthew Kirschenbaum, who pursues this investigation into the materiality of the digital, and like Hayles is approaching it from the perspective of textuality. In “Every Contact Leaves a Trace” (pdf) Kirschenbaum critiques the neo-Romantic, screen-focused tendencies of digital textual theory that tend to emphasise ephemerality and instability. He uses digital forensics to moves us from the screen to the hard drive, showing exactly how data is embodied (as in this image: a magnetic force microscopy image of a hard drive surface, from Pacific Nanotechnology). In the process he introduces another pair of concepts: formal and forensic materiality. Formal materiality refers to machine-readable data that reveals material specificities - in Kirschenbaum's paper, the use of a hex reader to discover traces of not-quite overwritten game code on an old Apple II floppy disc.

Forensic materiality refers to the material residues or byproducts that mark out one digital instantiation as different to another; for example the physical instantiation of copies of a file on two different hard drives will be different due to the material specificities of the drives - as when a misaligned write head again leaves traces of overwritten data. Yet these files are, for the computers concerned, formally identical. As Kirschenbaum writes, this shows how

"computers ... are material machines dedicated to propagating a behavioral illusion, or call it a working model, of immateriality."
This really nails it for me. It's exactly the functionality of this immateriality that earlier critiques of the disembodied digital overlook. It is an illusion, but it's an illusion that (mostly) works, and so is easily maintained: this is a hard-working model.

I'm developing an idea of transmateriality (sorry about the coinage), that draws on Hayles and describes exactly the "conundrum" that Kirschenbaum poses here; but that also has, I think, some wider implications, specifically for the media arts. Briefly, it proposes that the digital is, of course, always and inevitably embodied; that concepts like "data" are functional abstractions for describing the propagation of material patterns through material substrates. But that at the same time these material patterns - and here I mean everything from optical pulses to hard disk substrates, luminous screens and speakers pushing air - these material patterns, and the sensations and aesthetics that result are profoundly shaped by data acting as if it were symbolic and immaterial. Transmateriality is an attempt to "ground" the digital without losing sight of its (let's say) generative capacities. It also seems to resonate with a lot of current work in the media arts - but more of that later.


Matt said...

Thanks for those comments. I'll look forward to following your project.

FWIW, my book, Mechanisms: New Media and the Forensic Imagination, is now out from MIT; it develops the ideas in that paper in a good deal more detail.

Gordon Monro said...

Two brief, and somewhat opposed, comments about materiality.

I'm specifically reacting to Kirschenbaum's paper.

Firstly, I am surprised that it is apparently news to Kirschenbaum's audience in 2005 (though obviously not to Kirschenbaum himself) that electronic media have a physical basis. A colleague of mine used to give lectures on the sophisticated error-correcting code used in audio CDs (the inclusion of the code in the CD standard dates from 1980). According to "The Mastering Engineer's Handbook" on Google Books, high-quality disks produce as few as 20 to 30 errors per second, while the error-correcting codes can deal with many more. Without the error correction, CDs would be unlistenable-to. DVDs have an even more complex error-correction system.

Also, the little transistors used in logic gates are engineered as far as possible to be either "off" or "on", but they can get into in-between states, known as metastable states. Electrical engineers worry about the logic pulses inside our computers, which are idealised as nice square waves, but which become distorted by stray capacitance and other things. Even the speed of light is an issue; 30 centimetres per nanosecond in a vacuum, less in a circuit. And I remember reading a long time ago a worry that as computer memories became more compact, they would be more likely to disruption by stray cosmic rays.

I suppose that I'm saying that it is amazing that computers, CDs and so on work at all. The illusion of immateriality is remarkably good!

Secondly, the illusion of immateriality is remarkably good, good enough in fact for much of computer science to be like pure mathematics, in that it can be based on axioms and doesn't have the tension of most sciences between a scientific model of a phenomenon and the actual phenomenon. A computer language functions exactly according to its grammar. It is entirely reasonable to rely on the formal properties of digital data; we just need to remember to back up our drives regularly!

Anonymous said...

I've been thinking about this for a while and I think I agree with the suggestion that digital technologies are in fact still material but, if you accept that to be the case, I think that makes particularly significant the notion that, particularly as a storage medium for various modalities, the digital often lacks an obviously haptic quality which allows us to engage with it's materiality.