Thursday, January 19, 2012

Kant Meets Cyberpunk

In 1992, my first year of graduate school, I read William Gibson's cyberpunk classic Neuromancer and, by chance, Kant's Critique of Pure Reason at the same time. It seemed to me that the two were intimately connected, but various older grad students in my Kant class pooh-poohed my ideas about this and I lacked the intellectual confidence to pursue it farther.

But the thought has stayed with me. In Neuromancer, like in Tron, there's an artificial environment that one can travel in virtually. One "enters" it by jacking into a neural interface. Also like Tron, but unlike The Matrix, the artificial environment of Neuromancer substantially differs in its basic structural features from the real-world environment. Derived from early computer graphics programs, Neuromancer's cyberspace matrix is composed of lines of light arranged into geometrical figures in simple colors; space is experienced in discrete units and movement is in rectangular clicks. As I seem to remember having imagined it, and as we might as well imagine it for present purposes (though now looking through the text, this not accurate), everything is laid out rectilinearly and the only colors are simple primaries.

So now imagine that you were born jacked into such a matrix. You might think that objects were necessarily laid out in straight lines at right angles and possessed of only primary colors, and that space came in discrete units. But this would be a feature, not of things as they are "in themselves", but rather of how your mind processes the structured input it is given. We might even imagine (and maybe it's true) that a human mind that developed in such a matrix couldn't even conceive of curves, oblique angles, tertiary colors, or continuous space. For such a mind, objects as presented in the cyberspace matrix would be the only empirically available reality, and what we non-cyberspace-embedded folks consider to be the real world would be an incomprehensible "noumenal realm" behind those appearances. Conversely, we might imagine -- though it's impossible to depict vividly in a novel -- this matrix-grown strange baby to have new sensory modalities and new basic ways of cognizing the world that are unfamiliar to us, especially if its brain is artificially enhanced (a fourth "spatial" dimension for matrix-informational layout would be a conservative start).

The analogy to Kant is imperfect. Kantian purists will, I suppose, cringe at the comparison. Time, causation, three-dimensionality, and many other properties are shared by the Neuromancer matrix and the reality outside of the matrix. And the features of the matrix available to the embedded mind might not be given "a priori" in a strict Kantian sense (whatever Kant's sense is). I'm sure there are other important disanalogies too. But as a way of getting a toehold on the Kantian picture, I still rather like the comparison. We are born into naturally given matrixes that necessarily structure our experiential encounter with the world, and out of which we cannot break, even in imagination. All this that I see and hear is just user interface.

Such thoughts are doubly apt, perhaps, if we are actually already living in a giant computer simulation.


scitation said...

It's not a coincidence that Pat Churchland et al. wrote "A Critique of Pure Vision" in 1994, eh?

djc said...

agreed. i exploit this analogy in "the matrix as metaphysics". also this from the introduction to "the character of consciousness":

"These two poles call to mind another famous dichotomy. Eden corresponds at least loosely to Kant’s phenomenal world: the world of things as they appear. The Matrix corresponds loosely to Kant’s noumenal world: the world of things in themselves. We might think of Eden as a pure phenomenal world: if we were in Eden, nothing would have a hidden nature. But if we are in the Matrix, the world has a hidden nature that is not revealed to us in perception or even in ordinary science. Like the analogy with the scientific image, the analogy with the noumenal world is imperfect. The movie offers us an un-Kantian route to see the things in themselves: the red pill reveals their computational nature, which of course just raises the question of the noumena underlying the computers in turn. Still, through this lens, the film no longer seems to be an illustration of Cartesian deception. Rather, it is an illustration of Kantian (or perhaps Russellian) humility."

Eric Schwitzgebel said...

Right, Dave! Still, I like the Tron world or the Neuromancer world (especially as I had misremembered it) a little better because they make more vivid that there can be major structural differences between the noumenal and the phenomenal world. In the Matrix (at least the original movie; I haven't seen the sequels) the basic elements of perceptual experience remain the same between the worlds, despite the contents of perception differing (except a bit in the slow-down sequences).

I'd be interested to see some fiction that strives to make the structural differences even sharper. Greg Egan plays with attempting to portray multidimensional visual spaces as among the phenomenal options in Diaspora, I know, but he doesn't push it very far. If I ever decide to push a little more on my "Strange Baby" case, one of my goals is to try to get my head into the possibility of Baby's (or Grandbaby's) radically different perceptual phenomenology. I'm also trying to sneak up on the issue in my work on inverting goggles, distorting mirrors, and the problem of known illusion.

(BTW, I do make a couple references to your Matrix paper in my discussion of related issues near the end of the draft of "The Crazyist Metaphysics of Mind".)

Anonymous said...

>>We might even imagine (and maybe it's true) that a human mind that developed in such a matrix couldn't even conceive of curves, oblique angles, tertiary colors, or continuous space.

Sounds like a gimped human mind, to me. Did you mean that there might be a human with a mind that wasn't capable of conceiving of these things, or did you mean to suggest (quite radically, I'd think), that a human mind that developed in such conditions couldn't POSSIBLY conceive of such things? If you meant the latter, I'd think your suggestion is pretty darn implausible.

Carl said...

I find the Matrix analogy less than helpful in understanding Kant, because you can "get out" of the Matrix into "the real world." There's no possible route out into the world of noumena. What makes it the noumenal world is just that it is the world minus experience. All we can ever do is think it, but there's nothing there to imagine.

Kant's criticism of those he called "empirical idealists/transcendental realists" is that they tend to think in just this way: that there is a "real world" out there beyond our experiences, whereas the experiences themselves are only ideal. Kant calls himself a "transcendental idealist/empirical realist" because he wants to say that on the empirical level all there is is the experience of stuff. He's almost a naive realist! The table that we see is the objectively real table. But what about the table in itself? 'Oh, OK,' says Kant. 'There is something to the thought of a table in itself, but that's just an idea. It's not a possible object of experience.'

The Matrix analogy is more helpful for Plato's cave or Descartes' everything.

Eric Schwitzgebel said...

Anon 1/19: I was thinking on analogy of someone born blind being unable to conceive of visual phenomenology, and I was also thinking of the visual impoverishment of animals raised in visually deprived environments. Now, of course "conceive" comes in different strengths and can be applied to different facets of the situation. In some sense, I can conceive of there being sensory modalities radically unlike my own, with radically different phenomenal character; but in another sense, I can't really conceive of the phenomenal character of such possible modalities.

Eric Schwitzgebel said...

Carl: Of course that is one crucial difference. I was trying to tweak the thought experiment so that one was born into the matrix and incapable of escaping. As for Kant on what's "real": Well, he's pretty confusing about such matters, but I was thinking of the objects in the matrix (e.g., the cubes representing various programs one might try to hack into) as the empirically real objects for such a mind. So I hope the divergence from Kant is not *quite* as bad as you are suggesting.

However, to say that things in themselves are just an idea, as you seem to be suggesting, doesn't make a lot of sense to me. That sounds like making things in themselves the product of our minds, which seems a contradiction in conception. Of course *the idea* of the thing it itself is an idea; but that's a simple tautology.

John Jones said...

I don't know if there is a logic of incommensurables but if there was I'm sure it wouldn't allow an "interface" between Kant's phenomena and noumena, as they are inaccesible epistemologically speaking, one to to the other. It isn't a matter of there being some common ground between them that allows the construction of an interface. Their ontologies are different for one thing. This plays havoc with causation, as we can see in the brain/mind causation. For another, noumenons aren't hidden objects.

Eric Schwitzgebel said...

John: I agree that there are problems of this sort. For example, it's natural to think of the noumena as causing the phenomena, as in these sci fi scenarios. But since (if I understand correctly) Kant thinks causation belong only to empirical phenomena and isn't a feature of things in themselves, Kant can't really say that. So what the relation is, exactly, between noumena and phenomena for Kant is obscure to me.

Unknown said...
This comment has been removed by the author.
Eddy Nahmias said...

It's been a while since I read it, but I think the novel Flatland might also raise similar comparisons.

I always ask my students to consider why Neo would possibly have concluded that the "real" world was real once he left the Matrix. Unlike us, who wonder whether our world is real by simply imagining the *possibility* that we are in a Matrix, he experienced the *actuality* of having been in a Matrix. So, he should be even more skeptical of his knowledge of reality.

As such, I think the Matrix offers an illustration of Kant (perhaps on an overly simplistic reading)--there is simply no way to know the noumenal world that structures the reality we experience, even if one moves from one phenomenal world to another. Perhaps this gives us reason to be skeptical of the idea of noumenal reality.

(By the way, I also think it's helpful to get students thinking about what Neo would have thought about the reality of the "real" world, contra the Matrix world, had he awoken in it, not with a body and senses just like he had in the matrix, but as an ectoplasmic blog with radically different senses and no way to categorize or parse things in the new world in terms of concepts he'd gained in the matrix world.)

Eric Schwitzgebel said...

That's a nice angle on The Matrix, Eddy. I like it. Once question is whether the correct predictions of the authority dispensing the pills generate enough credibility to overcome the doubt engendered by the phenomenal world transitions.

And yes, Flatland/Sphereland is a nice comparison point too, especially if one can read them not as *just* about the consequences of differences in dimensionality.

Anonymous said...

I sort of get your point, but isn't the idea of an artificial environment that there isn't a 'noumenal realm' behind it, because it's an illusion? There may be a machine programmed to cause the illusion, but the phenomena you experience in the artificial world aren't representations of that machine or its program (causation is different from representation). I strongly doubt Kant wanted his arguments to show that we stand to the world in the same relation as someone suffering an illusion stands to the world.

I wonder if the cyberpunk analogies might be too vivid for their own good. Rather than providing toeholds for Kant they might fill the imagination so well as to crowd Kant himself out, but leave the impression that he'd been understood.

Arnold Trehub said...

Eric, you wrote:

"But as a way of getting a toehold on the Kantian picture, I still rather like the comparison. We are born into naturally given matrixes that necessarily structure our experiential encounter with the world, and out of which we cannot break, even in imagination."

The innate brain machinery that "necessarily structure[s] our experiential encounter with the world" is (I claim) the retinoid system. See *The Cognitive Brain* (1991), Ch. 4, "Modeling the World, Locating the Self, and Selective Attention: The Retinoid System", here:

See also "Space, Self, and the Theater of Consciousness", here:


"Evolution's Gift", here:

Eric Schwitzgebel said...

Arnold: Yes maybe so!

Anon Jan 23: "Illusion" is a tricky world here. The matrix in The Matrix is arguably illusion (though Chalmers doesn't think so), but the cyberspace matrix in Neuromancer is meant to veridically represent programming information in the world and enable you to interact with it productively, so it's more difficult to argue that it is illusion. It's more like that is how the information is designed to be experienced by us -- arguably analogous to color or taste on some mainstream secondary-quality views of such properties. At least that's what I'm inclined to think.

Anonymous said...

Anon Jan 23 again.

To my mind, a lot depends on whether the artificial environment is causally complete - if, in the artificial world, you starting cutting things up and asking how they work, would there be an answer? I think that if not (which I think is the case for the 'The Matrix' matrix, though I'm not sure), then the artificial world is illusory.

I would also say that if the artificial world is a simulation on a machine in our world, and you enter into it from the real world, then you are experiencing an illusion even if the artificial world is internally causally complete. (It's interesting that there could be an internal artificial-world cause of every phenomenon in the artificial world, as well as the external real-world causes, which are processes in the simulating machine.) I'm not sure on what grounds I'm drawing a line between that case and the case of someone born into the simulation as a simulated being. I'm tempted to say that the latter doesn't live in the simulation at all - instead, since all structures that can be simulated at all exist in a platonistic fashion elsewhere, it really lives in an entirely independent world somewhere far distant, and it is that which the simulation is simulating. When you enter the simulation from the real world, on the other hand, your consciousness is being used somehow in the simulation, which corresponds to your experiencing an illusion (possibly a voluntary one).

Eric Schwitzgebel said...

Anon 23/25: Interesting how these questions get weirdly tricky as you pull on them! The "causal completeness" thought is intriguing but raises difficulties of application. Assuming naive realism about the world of physical objects around us -- are they "causally complete"? Only in a relatively weak sense, it seems -- and then maybe that sense would be weak enough that the Neuromancer matrix or the Matrix matrix would qualify?

Anonymous said...

If you track down a copy of the old Shadowrun RPG Sourcebook "Virtual Realities", the first edition, there's a short story which occupies most of the last half of it which is just on this topic.

Eric Schwitzgebel said...

Thanks for the suggestion! I played Shadowrun a few times back in the day. I wonder if I was influenced by that without remembering. I'll check it out.