Re: [-empyre-] multi-perspectival / cultural hegemony of space

Simon +,

Simon Biggs wrote:
> I disagree with you whole heartedly...but I will try to keep my temper ;)

i expected nothing less :)

indeed i think it requires such a renderer as you describe below to even
begin to approach non-cartesian spaces with versimilitude. but still we
are faced with the xy of the screen and there is no getting around that,
at least not at this juncture. and as i mentioned in a previous mail, we
are hopelessly hindered by the devices we use to measure our

i've always been very fond of medieval space, particularly bosch's
garden, where the top of the frame is back, the bottom front. looking at
it with contemporary eyes, one sees things stacked up like a pyramid
rather than converging into the distance. 

indeed medieval artists, as well as artists today, are faced with that
2d xy surface of the canvas, and in our case, the screen. it would be a
riot to create a medieval rendering system, where object placement is
algorithmically determined by cultural signifigance, dogmatic such as
that is. but it certainly can be argued that composition has always been
heavily weighted by the cultural signifigance of the things being
framed, regardless of the sophistication of the perspective.

i feel, arguably of course, that the compositional traits of medieval
painting have less to do with cultural signifigance and more to do with
a naiveté of spatial understanding (naturally the individual viewer
brings to the apprehension of the work their own bias - you tend to
place more emphasis on the cultural reading of a work, and i the
formal). bosch is a fine example.  being something of a transitionary
figure between late medieval and early renaissance, the garden
represents to me the first inklings of a "scientific" approach to a 3d
to 2d mapping. that things in the distance appear smaller has always
been generally accepted by artists, but getting the stuff in between
"correct" has been a challenge. i think of my own development as a
draftsman from age 6 to now, and it more or less parallels the
development of spatial understanding from the dark ages to the
renaissance. when i was a kid, things in the distance were at the top of
the page and small, things in front, big and at the bottom, and the most
important things were the biggest of all. i think this exemplifies
spatial naiveté, not artistic choice, with the exception of important =

if, as you suggest, fundementally hegemonic conventions determine how we
see things, and how we believe things actually are, it would in a rigid
sense, stand to reason that medieval society thought that things things
with signifigance are actually larger. i cant imagine that is what you
are suggesting, as i can't believe that the medieval person thought this
as well. 

cartesian space does not represent an arbitrarily arrived at world view,
determined not by how things are, but by the force of western greed and
self-rightousness (the axe you have the tendency to grind). if that were
so, it would stand to reason that the spatial systems employed in
chinese landscape painting would have worked equally well at putting a
person on the moon. without renaissance perspective, we would still
assume the moon was not terribly far away, made of cheese, or
what-have-you. you know, there prolly actually is this thing called
reality, that actually does play by certain rules, that we are endlessly
in the process of trying to understand, and that though we are all
limited by our cultural background, sometimes we just get it right. i'm
not trying to suggest that cartesian space is the "correct space" i'm
sayiing its a really really really useful space, a close approximation
of our everyday perception, regardless of the culture that produced it.


> Whilst I said it is difficult it is also entirely possible not only to
> imagine non-Cartesian modelling of space but also to write code to do it. It
> is only difficult because it requires you use your imagination to think
> outside a set of conventions that are so fundamentally hegemonic in
> determing how we see things that we think things are actually the way we
> think we see them.
> Plentiful examples of non-Cartesian single point spatial rendering models
> exist. I think here, for example, of how Medieval artists dealt with space
> and time in their work, often having different temporal aspects of a scene
> placed in the same spatial frame whilst simultaneously breaking up that
> spatial frame to allow for a rendering of the image that took account of the
> relative values of the objects contained within that spatial envelope. That
> is, Christ on the cross would loom larger than the figures around him, in
> defiance of our contemporary expectations of spatial representation, not
> because these artists were unable to get the relative scale of things right
> but because they were less interested in a "photographic" rendering of
> things and more intent on an imaging strategy that addressed the cultural
> value of things.
> The same logic is at work in the work of Picasso circa 1912 and in that of
> many artists of Western and non-Western cultures (Japanese and Persian
> spatial models are examples here - but so are Mondrian's "Broadway Boogie
> Woogie", Francis Bacon's tortured yet fluid spaces or even Visconti's single
> discontinuous take in the restaurant scene in "Death in Venice"). In fact,
> when seen in a larger history of visualisation strategies, the Cartesian
> approach has only a small part to play, having been dominant in Western
> culture for only a scant few hundred years. It just happens to map onto
> other culturally and historically powerful elements, such as the notions of
> ownership, the body politic, the definition of the individual in relation to
> this, etc, that have developed contemporaneously, which make escape velocity
> difficult to attain.
> The spatial systems you are arguing are inescapeable are only that if you
> choose to work within a certain world view and specifically choose to work
> with off the shelf technologies based on that world view.
> Some years ago - actually many years ago - I wrote a 3D modelling package (I
> have been trying to find the code, but I think it is in some storage format
> no longer readable and likely corrupted from years of lying around - in C++
> under Unix on a Sun mini-computer) which applied Einstein's notion of
> gravity determining the shape of space to a visualisation system. It used a
> conventional 3D axis calibrated space. What was unconventional was that the
> axial calibration was dynamic, and determined by the objects modeled in the
> space. That is, when creating an object you modelled not only its form (I
> never got as far as writing the code for surface characteristics) but also
> its mass or the mass of its component parts. The mass data was used to then
> non-linearly calibrate the units of measurement on each of the 3 axial
> dimensions, thus causing whatever you had modelled, whether a single object
> or not, to also have their form(s) modified further as the spatial envelope
> adapted to what it contained. The system was logically simple but the
> results quite complex and unexpected, being highly recursive, as various
> objects moving about the space interacted with one another and the space
> itself, causing things to change size and relative internal scale.
> Graphically it was pretty squishy and not at all Cartesian.
> Perhaps I will set myself the task of re-writing that code for the web. It
> was pretty simple and of course 3D modelling code is not that difficult
> anyway (so long as you forget about light as anything more than a Z-buffer
> determiner). The 3D engine in Babel, and other "3D" projects I've done, is
> literally 3 lines of maths, although that is then used by a lot of other
> code that is to do with behavioural programming and the like. However, the
> code that calculates the 3D aspect of the scene is almost nothing. A gravity
> based 3D modeller would also be very compact - measured in kilobytes I
> should think and no more than a few hundred lines.
> best
> Simon
> Simon Biggs
> Research Professor
> Art and Design Research Centre
> Sheffield Hallam University, UK
> _______________________________________________
> empyre forum

This archive was generated by a fusion of Pipermail 0.09 (Mailman edition) and MHonArc 2.6.8.