The filtering void of consciousness (2)

And the limits to conscious intelligibility.

The previous post discussed the function of consciousness, and of self-consciousness. There are limits to the intelligibility of consciousness that are inherent in consciousness itself.

Self-awareness

Perhaps the most famous single line in Western philosophy is Rene Descartes’s cogito ergo sum: I think therefore I am. The full quote is:

dubito ergo cogito, cogito ergo sum.

I doubt therefore I am, I think therefore I am.

Descartes’ point is that we cannot doubt of our existence while we doubt. But that is true only if we are aware enough to doubt and be aware of our doubting. Which requires not merely thought, cognition, but thought about thought: metacognition.

His famous cogito ergo sum should really be:

scio cogito, ergo scio sum.

I am aware that I think, therefore I know that I am.

One of the key differences between Western Philosophy and Eastern Philosophy is that the former tends to consider consciousness only as part of looking outward, to our knowledge and experience of the world. Descartes epitomises this.

Eastern Philosophy tends to look inwards, to explore the inward experience and limits of consciousness. Zen thought in particular seeks to take us away from the distractions of self-consciousness. Zen monk Takuan Soho (1573-1645) cites a poem to that effect.

To think, “I will not think” —

This, too, is something in one’s thoughts.

Simply do not think

About thinking at all.

Chinese scholar Shao K’ang-chieh (1011-1077) telling us that “it is essential to lose the mind” is making a similar point. In modern psychology, this is about achieving flow states, aka being “in the zone”, where you just do without self-consciousness getting in the way.

Even though our self-consciousness can be distracting, very little of our actual thinking is conscious. We can focus our cognition on a problem, we can consider (and accept or reject) what our cognition comes up with, but the thinking itself is mostly not conscious. That’s why solutions to problems can “pop into our head” hours, days, even weeks later.

Hence also our capacity for self-deception. We tell conscious narratives about ourselves that we find emotionally satisfying but which come with no guarantee of accuracy. On the contrary:

The idea that introspective explanations are often confabulatory harmonises with much experimental work in social psychology. (Frankish, 2010)

The ability to consider and package information does not require our cognition in general be conscious.

Noah Buscher on Unsplash.

The void of consciousness

Just as much our thinking is not conscious, so our ability to consciously “look into” our thoughts is limited. This is because consciousness is not only an instrument of focus, it is also a filter. Both in a straightforward sense—if you focus on something, you are filtering out other things—and in a very difficult-to-think-about sense.

The difficult-to-think-about sense is that which we cannot consciously think about, so cannot consciously consider, due to being below the minimum limits of coherence that enable elements of cognition to be conscious. It is about the limits to what is intelligible to conscious thought such that it is expressible in language.

Thinking about what we cannot directly think about is hard. For we have to think around it.

To give a concrete example, you can’t see you your own eye balls, without a mirror. There is no reflect-itself, no see-itself, mirror for our own consciousness. Observing other people can tell us things about ourselves, but none of them is a mirror to our consciousness.

We can attempt to approach the limits to our ability to consciously think about thought, to “see” into the underpinnings of conscious thought, via the constituent elements of language. In language, the smallest element of sound that can change meaning is a phoneme. Ph is a phoneme—consider the difference between one and phone.

The smallest string of phonemes that can convey meaning is a morpheme. Phone is a morpheme. So is one.

Adding ph to one creates a different morpheme that is entirely unconnected to the first. Adding ing to phone, as phoning, changes the meaning of phone in a way that is connected to the original morpheme. It modifies the morpheme rather than creating a new one.

Now, consider thought. Suppose thoughts have elementary units of cognition, let us call them psynemes. They are the units of cognition that form morphemes.

How can we conceive of such elementary units of cognition? Well, we can’t. Certainly not consciously. A thought has to have a certain basic level of content-coherence in order to be conscious. If it is below that level of content-coherence, that coherence horizon, we cannot consciously “think” it, nor directly refer to it. The content of a thought has to be “large enough” to be visible to consciousness.

So, there is a coherence horizon below which thought cannot be conscious. Something has to have a sufficient level of coherence to be (consciously) referred to. Hence our inability to (consciously) break up fundamental concepts into more basic cognitive elements below that level of conscious coherence.

Hence the difficulty in conveying new concepts to people and our struggle to put thoughts into words. We can see our packaging-for-conscious-speech — our transmitting packaged information — glitching when we use the wrong name, term, word, etc. Likely because our packaging-for-conscious-communication is trying to keep up with our deeper cognition.

In explaining new concepts, we often proceed by example, metaphor and analogy — all indirect methods — precisely because the direct route of consciously breaking it up into elementary cognitive constituents that we can then directly refer to is not available to us. This is due to verbal communication being filtered through conscious use of language and conscious consideration of what to say. So the coherence horizon sets the bottom limit to what is consciously communicable.

It is clearly possible to convey new concepts to people. The use of examples, metaphors and analogies to convey concepts generally leads our minds to “get it”. That we are a highly imitative species probably assists our learning to “get around” the coherence horizon.

Nevertheless, conscious communication has to induce new concepts somewhat indirectly, due to the coherence horizon. Such communication has to assemble the psynemes we cannot consciously conceive, so cannot be consciously used in transmitting concepts.

There is thus a void in our consciousness. A minimum level of coherence is required for conscious thought. This limits what is consciously intelligible to us. The coherence horizon is also a reference horizon, as it limits what we can (consciously) refer to.

Implications of the void

There are lots of implications from this void in consciousness. It helps creates the mind-body problem, where things in the mind seem so very different from physical things and it is hard to connect one to the other.

The consciousness void is also a perennial difficulty for neuroscience. It makes it very difficult to connect activity in the brain to conscious thinking in any detailed way. How can neural signals be connected to any sense of experience, of consciousness, let alone that recursive awareness of experiencing that we call self-consciousness?

The consciousness void means that the program of Analytic philosophy of definitively delineating key concepts was never realisable. It makes it much easier for Philosophy to wander off into nonsense because its conceptual underpinnings are always opaque and its ideas are not systematically tested against structure in the way Maths, Science, Engineering or Law are.

Indeed, once we develop ways of systematically testing ideas against structure in some field — when we can move on from guessing, when we can test our guesses — the field exits from Philosophy. Though the field will still involve some philosophising, a certain amount of which is unavoidable in thinking in any general way about the world.

Without systematic tests against structure, there is no way to systematically ground concepts. As abstractions have no inherent, testable, grounding, it is hard to stop sliding across usages in argument.

In such circumstances, what is rhetorically powerful — so persuasive or otherwise congenial — tends to win over what is true. The less testing against structure there is, the more powerful is this effect.

One way to think of abstraction is as data compression. Unlike Science, Engineering and Law, which have structure-of-things setting limits of practicality, there is no inherent test in Philosophy of when to stop “pettifogging”, when one has as much data compression as one needs and no more. Indeed, the information-content and clarity of philosophical concepts can be highly variable.

Hence systems of thought which reject the constraint of tests against structure can select all the better for what is rhetorically powerful; for what is socially convenient; for generating social leverage and status. This is especially effective if all this can be moralised at the same time. The consciousness void is a great aid to self-deception: which can be very strategic self-deception.

The consciousness void — and the reference horizon it generates — enables the activist technique of using your language, your vocabulary, but not your dictionary. To use a word that most people take to mean one thing and use that to leverage the shifted meaning shared among the activists. That is how Equity becomes activist control over resources, Diversity becomes activist control over hiring and Inclusion becomes activist control over speech, assembly and association.

Types of knowledge

The consciousness void makes thinking about consciousness is hard. In particular, it makes developing anything replicating what conscious beings use their consciousness for difficult.

Consider the four types of knowledge: propositional knowledge, knowledge that (episteme); procedural knowledge, knowledge how (techne); perspectival knowledge, knowledge of (noesis); and participatory knowledge, knowledge in (gnosis). The first two can be algorithmically replicated by computation. The latter two are a rather different story.

To get computers to provide propositional knowledge, or to direct technical actions, is fairly straightforward. Enough computing power can substitute for conscious cognition fairly readily in those activities as they are straightforwardly algorithmic. Indeed, much training is about enabling folk to do things automatically, without being self-conscious, or even particularly conscious, about it.

To get machines to walk is harder, as we have to do it by creating some alternative to the mechanism of focus for self-maintenance that is consciousness. There is an algorithmic (so computationally replicable) element to human cognition, but neither consciousness nor self-consciousness is readily captured by such.

The reflexive mind, the thinking-about mind, we are no closer to replicating now than when AI and machine learning began. The limits to conscious intelligibility, the coherence horizon, makes it hard to not only proceed, but to see how we might proceed while the reference horizon makes it hard to talk about.

With the limits to conscious intelligibility we are well into the territory of the famous passage from the Tao Te Ching:

The Tao that can be told is not the eternal Tao;
The name that can be named is not the eternal name.

Self-reference

The consciousness void, the coherence horizon, the reference horizon, generate various paradoxes of self-reference. Problems such as the Grelling-Nelson Paradox, Russell’s Paradox, Godel’s Incompleteness Theorems and Turing’s Halting Problem all revolve around the problems of self-reference.

The Grelling-Nelson Paradox is a matter of words referring to words, Russell’s Paradox is a matter of sets covering sets, Godel’s Incompleteness Theorems are about axioms referring to axioms and Turing’s Halting Problem about the computability of computability.

For, as noted above, the coherence horizon is also a reference horizon. The consciousness void blocks our ability to “see” into the process of reference. This makes it easier for what a particular word refers to, to shift during reasoning. Hence, this opaqueness in the basic operation of reference, of aboutness, can be used strategically, as when a term can mean something in ordinary usage and something else in activist usage.

Once we are in the realm of self-reference, we are attempting to apply one layer of the consciousness void to another layer of the consciousness void. To think around (but not directly into) the application to something that we also have to think around (but not directly into). It piles one layer of the coherence horizon on another, one layer of the reference horizon on another. Hence the difficulty in satisfactorily resolving various self-reference paradoxes.

The problems of the void of consciousness, of the coherence horizon, are difficulties with our thinking arising from the limits to conscious intelligibility. They are not a matter of fundamental structures of the universe, just the limitations of consciousness: in particular, its inability to fully see into itself, into its own foundations, due to there having to be a minimal level of coherence for something to be within the realm of what is conscious.

References

Books

Rene Descartes, Philosophical Writings, Elizabeth Anscombe and Peter Thomas Geach (eds & trans.), Nelson’s University Paperback, [1637, 1644, 1954] 1970.

James Franklin, The Worth of Persons: The Foundation of Ethics, Encounter, 2022.

Heather Heying and Bret Weinstein, A Hunter-Gatherer’s Guide to the 21st Century: Evolution and the Challenges of Modern Life, Swift, 2021.

John Man, Alpha Beta: How our alphabet shaped the western world, Headline, 2000.

Amaury de Riencourt, The Eye Of Śiva: Eastern Mysticism and Science, Honeyglen Publishing, 1980.

Kevin Simler and Robin Hanson, The Elephant in the Brain: Hidden Motives in Everyday Life, Oxford University Press, 2018.

David Stove, The Plato Cult and Other Philosophical Follies, Basil Blackwell, 1991.

Robert Trivers, The Folly of Fools: The Logic of Deceit and Self-Deception in Human Life, Basic Books, [2011] 2013.

David Weber, Ashes of Victory, Baen, 2000. (On Pp 324-7, there is a particularly clear discussion of phonemes and morphemes.)

Articles, papers, book chapters, podcasts

Thierry Coquand, ‘Type Theory,’ Stanford Encyclopaedia of Philosophy, first published Wed Feb 8, 2006; substantive revision Tue Sep 6, 2022.

https://plato.stanford.edu/entries/type-theory/index.html#ParaRussTypeTheo.

Keith Frankish, ‘Dual-Process and Dual-System Theories of Reasoning,’ Philosophy Compass, 2010, 5/10, 914–926.

Chris D. Frith, ‘The role of metacognition in human social interactions,’ Philosophical Transactions of the Royal Society B, 2012, 367, 2213–2223.

Panu Raatikainen, ‘Gödel’s Incompleteness Theorems,’ Stanford Encyclopaedia of Philosophy, first published Mon Nov 11, 2013; substantive revision Thu Apr 2, 2020.

https://plato.stanford.edu/entries/goedel-incompleteness/#SomForThe.

Takuan Soho, ‘The Mysterious Record of Unmoveable Wisdom,’ in The Unfettered Mind: Writings of the Zen Master to the Sword Master, (trans. William Scott Wilson), Kodansha International, 1987, 17-44.