Aesthetics of the Error – thoughts on the ‘generative' error
In Tim Barker’s Aesthetics of Error, the author discusses
the term "generative"s several time with regards to artists and errors. This is something that in my own research
practice is heavily reliant upon. An understanding of (and critical
discussion about) these terms is something that will be key to the
pertinence of 3D-Error.
Barker writes about numerous artists practices in his paper,
all of whom rely in some way on error. This manifests itself in a variety of
ways with glitch and noise in contemporary music, or errors in image encoding
in visual art practices. Barkers concerns (compared with that of Mako Hill)
appear to be aesthetic, interesting in the generative benefits that these
errors bring. When writing about VJ performance artist Jorge Castro, he
explains: “ A distinctly digital aesthetic is created here as the generative
qualities and idiosyncrasies of an error make obvious the processes of the
computer.” And although the aesthetic concerns of these works occur as a
document to various digital processes, Barker points out that the generative
nature of errors isn’t new, and isn’t refined to the digital age. “For
instance, at the beginning of the 1960s Franz Erhard Walter, experimented with
art “informel,” utilized errors and chance occurrences as generative tools.” As
I have previously mentioned in my paper Towards
a 3D-Error Methodology, many artists that work with errors do so as a way
in which a generative tool in their processes, not solely as a satisfactory
visual output, but for the purposes of bringing chance into their method.
Barker aligns this potentiality of chance with Deleuze’s theory
on the ‘virtual’. For Deleuze the virtual is an array of potential outputs,
actions or events which have not yet been actuated. Were various decisions or
event happen to create a happing in reality, there are a range of
potentialities which never actualize or have not yet been actualized. For Barker,
the error tap into the ‘virtual’ as artist and the machine are unaware of the
errant path the glitch will take them on. There is also a description of the
complex relationship the artist must nurture with the machine in which produce
errant results and thus tap into these potentialities. Barker explains that artists
“thus direct the machine toward a particular
operation, setting up the conditions for an error to emerge, but not actually
solely designing the aesthetics of the error.” Artists must learn the workings
of the machine, seeking out their limitations. This is important for a number
of reasons, A) to get an output at all. Many errors which halt the processes
and may produce nothing. B) to ensure the same results aren’t experienced each
time, ie. an errant but methodical path. C) To adapt the changing methods of
the machine, either learnt by the machine or adapted by the programmer. This is
particularly true of software, which can be updated and changed by developers,
automatically changed online or even on users computers. This would mean artist
have to re-learn methods to adapt to changes made to their generative process.
Barker also explains a more symbiotic approach for media artists in a vision on
how the generation of work isn’t purely the reserve of the artist themselves.
That the artist knows the generation is shared with the machine. The machine is
as much a part of the creation, and the proportion of the generation “In these
instances, as the unique generative qualities of error are actualized, the
artist can no longer be thought of as the sole creative force. Rather it is now
the artist’s role to provide the circumstances for an error to emerge.”
However, with direction from the artist, the output reveals the nature of the
machine. Barker goes on to point out that is the potential of error inivetably
exists in all technology, we cannot thing of machines as merely facilitating
predetermined paths. Thus nodding towards a theory of all processes and
technology as creative or generative.
I have particular interest in this relationship with the artist/machine as I have discussed in Towards a Generative 3D-Error Methodology. There is a learned and mutual generative effect of media art (an arguably all media output.) in which both artist and machine have generate errant potentialities and explore the ‘virtual’. Increasingly so, users are directed more and more heavily through “black boxes” with limited user mediation. Thus producing a greater significance of machine generativity. It is important therefore for artists to explore the political, commercial or technological reasons for this shift towards less user input, to know if there are deliberate manipulations of user data or merely errant ones. Reasoning for less user input could be for: ease of programming; for simpler, normative outputs; to be more commercially lucrative; for political reasons to protect the company from hackers and exposure to their intellectual property. There is also a drive for simplicity in technology in the hope that its ease of use will bring the company ubiquity amongst a wide range of users. This simplicity may not suit more creatives or activists seeking to create non-normative outputs and in return these output may not be a concern to the company. This discussion also leads into the the issue (as discussed in the Mako Hill article) about the need for error to expose these working and decisions in technology.
I have particular interest in this relationship with the artist/machine as I have discussed in Towards a Generative 3D-Error Methodology. There is a learned and mutual generative effect of media art (an arguably all media output.) in which both artist and machine have generate errant potentialities and explore the ‘virtual’. Increasingly so, users are directed more and more heavily through “black boxes” with limited user mediation. Thus producing a greater significance of machine generativity. It is important therefore for artists to explore the political, commercial or technological reasons for this shift towards less user input, to know if there are deliberate manipulations of user data or merely errant ones. Reasoning for less user input could be for: ease of programming; for simpler, normative outputs; to be more commercially lucrative; for political reasons to protect the company from hackers and exposure to their intellectual property. There is also a drive for simplicity in technology in the hope that its ease of use will bring the company ubiquity amongst a wide range of users. This simplicity may not suit more creatives or activists seeking to create non-normative outputs and in return these output may not be a concern to the company. This discussion also leads into the the issue (as discussed in the Mako Hill article) about the need for error to expose these working and decisions in technology.
Comments