Aphorism of the Day: Every tyrannical system, to conserve itself as a system, will scapegoat even its king. So does drama masquerade as change.
.
So I’m reading and digging Paul Churchland’s most recent book, Plato’s Camera, while puzzling over David Chalmer’s latest at the same time, and I find myself thinking of Spinoza’s approbation against misconstruing the Condition in terms belonging to the Conditioned. In Part II of his Appendix Containing Metaphysical Thoughts he writes:
In this Chapter God’s existence is explained quite differently from the way in which men commonly understand it; for they confuse God’s existence with their own, so they imagine God as being somewhat like a Man and do not take note of the true idea of God which they have, or are completely ignorant of having it. As a result they can neither prove God a priori, i.e., from his true definition, or essence, nor prove it a posteriori, from the idea of him, insofar as it is in us. Nor can they conceive God’s existence. (The Collected Works of Spinoza, 315)
Given the analogical nature of human cognition, the reasons for this nearly universal error are quite clear: ‘men’ mined the information belonging to their own manifest image in their attempts to conceive God, simply because it was the most intuitive and readily available. Given this heuristic brush and informatic palette, they painted God in psychological terms, only possessing their features to the ‘nth degree.’ A personal God.
Spinoza catalogues and critiques the numerous expressions of this fundamental error in what follows, showing why the perplexities and contradictions that pertain to a personal God arise, and how these problems simply fall away if you subtract what is human from God. He was branded a heretic for his trouble, disowned by the Jewish community, and so reviled by Christians that some commentators believe that the following figure I want to consider intentionally expunged all traces of Spinoza’s influence from his own philosophy.
In philosophy of mind and consciousness research circles, Leibniz is typically mentioned with reference to his famous windmill example, which he uses to illustrate the now hoary conceptual gulf between doing and feeling. He writes:
One is obliged to admit that perception and what depends upon it is inexplicable on mechanical principles, that is, by figures and motions. In imagining that there is a machine whose construction would enable it to think, to sense, and to have perception, one could conceive it enlarged while retaining the same proportions, so that one could enter into it, just like into a windmill. Supposing this, one should, when visiting within it, find only parts pushing one another, and never anything by which to explain a perception. Thus it is in the simple substance, and not in the composite or in the machine, that one must look for perception. (Monadology, §17)
In a sense, the problem of Leibniz’s Mill simply turns Spinoza’s Sin on its head. The Mill cannot be the Condition, Leibniz is arguing, because he cannot fathom how it could generate the Conditioned, manifest ‘perception.’ In a sense, it captures the Hard Problem in a nutshell: how could all this ramshackle machinery generate the exquisite smell of turkey dinner on a warm, autumn afternoon, or anything else that we experience for that matter?
What does this have to do with reading Churchland? Well, Churchland wants to argue that cognitive science is guilty of committing Spinoza’s Sin, that too many are too prone to construe the Condition, neural function, by analogy to the Conditioned, psychology and language. So, for instance, in The Cambridge Handbook of Cognitive Science, you find Barbara Von Eckhardt explaining:
There is nothing even approximating a systematic semantics for even a fragment of [any mental representation system]. Nevertheless, there are ways to inductively infer to some global semantic features [any mental representation system], arguably, must have. One way is to extrapolate, via a form of ‘transcendental’ reasoning, from features of cognitive science’s explananda. (33)
In other words, Spinoza’s Sin is actually a Virtue: the explananda of cognitive science are nothing other than manifest features of cognition, what it is we generally think we’re doing (given what little we have to go on) whenever we cognize ourselves, others, and the world. So the idea, Von Eckhardt is saying, is to reason from the Conditioned, our manifest informatic palette, to the Condition, whatever will be eventually described in a complete representational theory of mind. She thinks, quite sensibly, that our manifest experience and intuitions are what need to be explained.
Churchland argues otherwise–or well, almost. Not only does the ‘linguaformal’ approach look increasingly unlikely the more we learn about the brain, it renders the obvious cognitive continuity between humans and animals very, very difficult to understand. In Plato’s Camera he paints a picture of cognition where Kant’s simple frame of timeless transcendental categories is smashed into a myriad of nondiscursive, neural ‘maps’ understood according to the formation and weighting of synaptic connections among populations of neurons possessing various, mathematically tractable structural predispositions. “Simply replace,” he writes, “‘single complex predicate’ with ‘single prototype point in high-dimensional activation space,’ and you have the outlines of the view to be defended here” (23).
Churchland, in other words, isn’t so interested in overthrowing the old order as he is in electing a new government. As radical as his account often seems, he still clings to certain boilerplate semantic assumptions, still sees the Mill representationally, which is to say, as a kind of content machine. Meaning, for him, remains something requiring a positive explanation. He argues that “deploying a background map of some practically relevant feature space, a map sporting some form of dynamical place marker, is a common and highly effective technique for monitoring, modulating, and regulating many practically relevant behaviours” (Plato’s Camera, 131). But even in the examples he provides, the homomorphisms he points out are all simply parts of larger dynamic systems, begging the question of why maps should be accorded pride of place in his account of cognition, rather than being relegated to one kind of heuristic tool among many.
Put differently, he ultimately succumbs to temptation and commits Spinoza’s Sin. Rather than, as BBT suggests, demoting ‘traditional epistemology’–treating it as a signature example of the way informatic neglect leads us to universalize heuristics, informatic processes that selectively ignore information to better solve specific problem sets–Churchland wants to dress it in more scientifically fashionable clothes.
Grasping the abject wickedness of Spinoza’s Sin requires an appreciation of the abyssal nature of the gulf between the Condition and the Conditioned when it comes to the question of human consciousness and cognition. One needs to understand, in other words, why the Mill has such difficulty fathoming itself as a Mill. Churchland, after all, is more than just a very, very intelligent man. He also possesses the imaginative capacity and institutional courage to make the analogical leap beyond linguaformalism–and yet, even still, he cannot relinquish certain intuitions regarding content…
Why?
Imagine a Mill designed to cognize environmental information, whirring and clicking in the dark. If you could peer through the gloom you would see loosely packed machinery, literally unimaginable in complexity, clattering away, wheel spinning wheel, cog rotating cog–swiss-watch complexities extending through impenetrable gloom.
Now imagine a flashlight, shining down across and penetrating into this machinery, illuminating and eclectic multitude of surfaces, the crest of a spinning wheel here, a length of strut there, the handle of lever, a corner of casing, on and on, a cobweb of fragmentary glimpses, become more and more fractional and dim the deeper the light probes the machine’s bowel. Peering, all you can see are shreds of machinery, a kind of inexplicable constellation in the black.
Now imagine that what’s illuminated represents the information accessible to conscious experience. Not only is information pertaining to the vast bulk of the machine inaccessible, information regarding the actual mechanical role of those parts somewhat illuminated is also out of reach–so much so, that even information pertaining to the lack of this information is missing. This means you need to cut out all those fragmentary, functionally distributed glimpses, then paste them into a singular Collage, transform a mishmash of perspectival distortions into one ‘manifest’ image. The informatic cobweb fills the screen, you could say.
Not so different from what-you-are-experiencing-this-very-moment-here-now.
Feed this information back to the Mill (whose machinery, remember, is primarily designed to trouble-shoot environmental information). Utterly blind to the vast amounts of information neglected, it takes the Collage to be sufficient–all the information accessed becomes all the information required. Since information drives distinction, its absence leverages the cognitive illusion of sufficient wholes–as I have written elsewhere, consciousness can be seen as a kind of ‘flicker-fusion’ writ large. Short of neuroscience, it has no real recourse to information that hales from beyond the Collage in its attempts to cognize the Collage. It is informatically encapsulated.
The Collage, in other words, is the Conditioned, the well from which our cognitive systems draw water whenever tasked with troubleshooting the Condition. Given the reworked Mill analogy above, it’s easy to see the peril of Spinoza’s Sin: From the informatic vantage of the Collage, the neurofunctional axis can only be indirectly inferred, never directly intuited. This is why the functional findings of cognitive science so often strike those without any real exposure to the field as so counterintuitive. Not only are we ‘in the dark’ with reference to ourselves, we are, in a very real sense, congenitally and catastrophically misinformed.
Pending a mature neuroscientific understanding, we are, in effect, the hostage of our metacognitive intuitions, and for better or worse, representation looms large among them. Churchland yields unwarranted pride of place to the homomorphic components of our heuristic systems, endows them with bloated significance, simply because metacognitive intuition, and hence tradition, mistakenly accords representations a privileged role. Because, quite simply, it feels right. It ain’t called temptation for nothing!
The Blind Brain Theory, as I hope the above thumbnail makes clear, affords the resources required to throw off the analogical yoke of the Conditioned once and for all, to subtract the human, not from God, but from the human, thus showing that–beyond the scope of a certain parochial heuristic at least–we just never were what we took ourselves to be.
And perhaps more importantly, never will be.
To explain subjective mental phenomena in terms of neural activity it seems to me that it is necessary to introduce some sort of representation of neural activity. Isn’t this what Churchland’s ‘maps’ are?
Churchland’s maps aren’t representations of neural activity, they’re representations via neural activity.
And this wrong because….?
The metaphor is just plain faulty – precisely because it is so intuitive, strangely enough. Churchland calls them ‘maps,’ but what he describes are various homomorphic neural recapitulations of invariant environmental features nested within vastly complicated neurofunctional systems. So then where does the ‘representational content’ lie? The fact is, each element of the system is as crucial to cognition as any other, so then why focus on the homomorphisms, transform them into the magical cogs, if not to flatter our neurofunctionally blind intuitions? Calling them ‘maps’ licences the equivocations he needs to make his account of cognition seem like the same o’ same o’ only understood scientifically.
It’s like giving an account of World War Two speaking almost exclusively about aircraft carriers.
Like you I am highly suspicious of intuitive interpretations of brain processes. What I’m not getting is what specifically it is about Churchland’s position that strikes you as an intuitive (vs scientific) interpretation.
BTW can you translate “homomorphic neural recapitulations of invariant environmental features nested within vastly complicated neurofunctional systems” into English?
Christ. Good thing they disbanded the jargon police! What I’m saying is that Churchland’s ‘maps’ are simply homomorphic cogs in much larger heuristic mechanisms – as he explicitly acknowledges at several junctures. I understand the intuitive appeal of maps, and I understand the intuitive appeal of mechanisms, but since the whole of the life sciences is erected upon the latter, why introduce this radical explanatory break when extending the life sciences into the realm of cognition? He himself talks about the way his account undermines the how/what dichotomy, the way it makes hash of the distinction between practical and theoretical knowledge (while vindicating the distinction between background and foreground knowledge), and yet his every consideration of how is bent on explaining the what, which is to say, how neural networks constitute ‘representational content.’ He ultimately wants his account to appeal to manifest intuitions regarding knowledge – to cater to prejudices borne of informatic neglect. He fills his head with innumerable little mechanistically interconnected ‘pictures of reality’ rather than, as I think is obviously the case even on his own account, innumerable little environmentally adapted heuristic processors.
In other words, he’s being a philosopher, worrying about the way his account fits with ‘Scientific Realism,’ when he really has moved past all that tangled, intuitive crap. ‘Environmentally adapted’ should be more than enough. This way, homomorphism can be exhausted by its actual neurofunctional role, without having to worry about some magical ‘correspondance’ to some magical ‘external reality.’
Does that make more sense? The thing to keep in mind is the way ‘representation’ is an artifact of informatic neglect. Our blindness to neurofunctional contexts led philosophers to conceive representationalism in the first place. How else explain transparency? Churchland is trying to cram an account that outruns the Epistemic Heuristic back into the Epistemic Heuristic.
“This way, homomorphism can be exhausted by its actual neurofunctional role, without having to worry about some magical ‘correspondance’ to some magical ‘external reality.’”
Thanks Scott. This makes much more sense now!
I’m reading “The Ravenous Brain” by Daniel Bor. He’s a neuroscientist with a background in philosophy. The horrible title aside, the book does a good job of explaining Bor’s take on the neural mechanisms behind seemingly mysterious things like attention and consciousness. If you get a chance you might want to check it out.
Definitely. Will do!
When you drive in the car the GPS map gives you a representation of space and time because objective reality outside your windshield shows you the roads and landmarks that it says you should be seeing and that you reached Point B after Point A and before Point C.
What reality does a neural map show you? Does it give you an aha! moment?
This actually illustrates the issue nicely, I think. Maps, like those on your GPS, are cognitive prosthetics, effective environmental ways to coordinate environmental interventions. This is so intuitive that it really seems to make life simple positing similar prosthetics in the brain. But the apparent functional prominence of the environmental map is actually an illusion generated by informatic neglect. Since all the neural machinery involved, even in interpreting a map as a map is neglected, let alone in ‘coordinating environmental interventions,’ the map seems to be the key to the whole kingdom, rather than just another cog in the machine. This is just one more example of the heuristic nature of environmental cognition, and the way we leave its scope of applicability behind the instant we bring it to the question of consciousness and cognition.
Is that a fair evaluation? Wouldn’t it be like showing you some roads, but the map is missing side roads and such? And more importantly, it doesn’t show the subways? Sure to take the part shown as the whole leads to even more entwined complexity of assumption, which is a danger, but surely if taken as merely a partial guide, there’s some paralel there to be seen? It seems to dismiss a potential baby step option for noobs like us to chew upon?
It really has to do with the way it flatters what is a deceptive way of looking at these kinds of things to begin with – like bringing a basket of chocolate to a Slimfast meeting. See my reply to Terry.
Well if his intended point is that these maps things describe the whole rather than a fragment (and possibly as if in hi res rather than blocky, kludgey pixels), then that’s poisoned the well already. But it seems the main problem is always that documentation potentially flatters. The risk that they’ll eat whatever you bring – even if you only bring yourself! Where is the line where a documents risk of this is too high and it should be dismissed?
Didn’t find time to fully digest this post before.
“Now imagine a flashlight, shining down across and penetrating into this machinery, illuminating and eclectic multitude of surfaces, the crest of a spinning wheel here, a length of strut there, the handle of lever, a corner of casing, on and on, a cobweb of fragmentary glimpses, become more and more fractional and dim the deeper the light probes the machine’s bowel. Peering, all you can see are shreds of machinery, a kind of inexplicable constellation in the black.
Now imagine that what’s illuminated represents the information accessible to conscious experience.”
This seems like an apt metaphor even for neuroimaging itself, though obviously what you’re trying to communicate has far more dramatic consequences for human understanding.
The rag and bone shop model of consciousness… Another human bubble popped!
P.S.
How do you intend exactly “representation”?
Do you mean the tendency to anthropomorphize everything and so represent things the way we perceive ourselves?
That seems a nice way to explain the repeating occurrences of patterns, and why we essentially may be just holding up mirrors whenever we think we are describing stuff outside. And gods and myths would be echoes, as in the theory in The Wayward Mind, all internal stuff misperceived and projected outside.
To me it’s always interesting to hear the interpretation from the physics side. You see, scientists are pretty convinced that they understand already the bigger part of it:
Yet he seems to dodge very carefully what happens in the brain, and the link between the brain and reality. Because that’s the real backdoor where everything is at stake, and I don’t think any of his claims can be valid as long the brain is a variable.