I couldn't agree more.
> Usually what he's doing is giving the
> viewer some "plot devices" to peruse, which suggest that some form of
> narrative is going on. The viewer then mentally tries to turn the
> picture into a narrative, and the Surrealism comes from the obvious
> discontinuity of the plot line. I think it is the struggle to resolve
> and make sense of the discontinuities of an image, which gives rise to
> Surrealist experience. The imagination is stretched and new
> inferences about reality are deduced.
This is exactly right. Our attempts to create a virtual world should
be limited to providing the tools, or "plot devices" and allow the
user to define their own experience.
I don't tell you what to dream, nor you me. But we both share the
same real world, from which we both draw our images (plot devices)
that make up our dreams. It is likewise important that any plot
devices we supply be "editable". I should be able to select a clock,
and assign a melting attribute to it.
This tends to suggest that we would need to have a library of objects
(plot devices) that have an underlying attribute set that can be
turned on or off as selected by the user.
> The physical landscape is also often set up like a stage. Quite often
> the ground is just a flat plane leading to an essentially featureless
> back wall. To be sure, the stage is often huge because of distant
> mountains, tiny figures, and atmospheric perspective. But it's still
> fundamentally a stage, with a collection of props upon it. Sometimes
> the stage metaphor is explicit, as in "The Old Age of William Tell."
> Here a couple of pedestals hold up a cloth, which functions as a
> curtain for the naked actors in dialogue behind it.
I still believe the user should be given a blank canvass, which I see
as being a black void, representing the night and setting the scene
for the dream...
If I choose to be in a room, I set up some sort of bounding box on
which I can paint the walls. Naturally these walls could have their
own attributes, (melting, opaque, solid, moving, etc...).
> In much the same way that Hollywood will hand us an assault rifle or a
> breaking pane of glass, Dali hands us a number of recurrent symbols
> as plot devices. These become self-contained "widgets" for telling
> stories: drop them into a picture, and you have the basic material for
> a narrative.
Exactly. The "basic material for a narrative" should be just that,
the tools that the user can select from. Possibly with some sort of
dynamic inheritance, so my clocks, (which are melting) can inherit
the legs of a Stork, which I say are to follow me wherever I go.
> The melting watch is another plot device which appears over and over
> and over again. It is a self-contained narrative: it marks the
> ephemerality of the passing of time. That's basically what narrative
> is: marking the passing of time.
I think we need to be careful that we don't introduce our own
interpretation (however correct it may be), to the purpose of objects.
It should be the user who decides what an object means to them...?
> So what's the moral of all of this, for VR purposes?
>
> If you take symbols and situations that people recognize, and simply
> abut them together, you will get Surrealist experience. The viewer
> will start from the base of the well-known or the partially-known, and
> extrapolate to the unknown narrative that interconnects the standard
> plot elements. People will do this because they like to dream, they
> like to be told stories, and they like to feel that they are being
> told a story.
I think people could just as easily make nightmares from this. Dali,
and his fear of grasshoppers is typical of how a personalised fear
could be introduced. If you are fearful of grasshoppers and you fill
a bath with them, and then submerge yourself in them this would be a
very anxious experience for you!
> This would suggest that computer-generated Surrealist storytelling is
> possible.
Indeed it is.
> One needs to manually come up with a sufficient body of
> interesting symbols, and then devise algorithms which will present and
> rearrange those symbols for the viewer. The "art" in such an
> endeavor, would be to create algorithms which direct the viewer's
> symbolic construction process in an interesting way.
>
> Such algorithms bring us naturally into the discussion of _chance_ in
> the Surrealist process. What role should chance play? What is its
> aesthetic value? It is important to note that this topic was heatedly
> debated throughout the history of Surrealism. Early Surrealism
> advocated a complete abandonment of methodical approach; chance was
> king. However, Andre Breton indicates in his essay "What is
> Surrealism?" that later Surrealist theory was no longer adverse
> to method, and that the sheer reliance on chance more properly
> belonged to Dada, the art movement from which Surrealism sprang.
Whether we allow chance or not would depends on how much we want the
user to be in control of their world, and how much their world should
be the result of the interaction of the plot devices they use.
If we are to use Breton as a measuring stick, then we should note that
he (and other surrealists), *demanded* an art expressing either
desire in its raw state or unrestrained inner turmoil, or else a
clear desire to go beyond what is real. I fail to see how this could
not be enhanced by the presence of chance.
> For those of us who are interested in Surrealism and the possibility
> of computer algorithms, I think we need to read up on the theories of
> the latter-day Surrealists. I haven't done so yet, but I suspect they
> are most likely to yield techniques of practical value to VR world
> construction.
I agree that we should look into these works, but we must be careful
that we don't try to create the dream for the user. We need to just
supply the void world, and the "plot devices" and let the user define
the art...
Lee.
lee@giaeb.cc.monash.edu.au