Congratulations to Anil Seth for winning the Berggruen essay prize on consciousness!  I didn’t learn the outcome until I emerged from one of the rare cellular blackout zones in modern America. My wife and I were whale watching south of Yachats (“ya-hots”) on the Oregon coast in this week of remarkable weather. We came up bupkis, nada, nil for the great migratory grays, but saw seals and sea lions bobbing in the surf, red shouldered hawks, and one bald eagle glowering like a luminescent gargoyle atop a Sitka spruce near Highway 101. We turned around at the dune groupings by Florence (Frank Herbert’s inspirations for Dune, weirdly enough, where thinking machines have been banned) and headed north again, the intestinal windings of the roads causing us to swap our sunglasses in and out in synchrony with the center console of the car as it tried to understand the intermittent shadows.

Seth is always reliable and his essay continues themes he has recently written about. There is a broad distrust of computational functionalism and hints of alternative models for how consciousness might arise in uniquely biological ways like his example of how certain neurons might fire purely for regulatory reasons. There are unanswered questions about whether LLMs can become conscious that hint at the challenges such ideas have, and the moral consequences that manifold conscious machines entail. He even briefly dives into the Simulation Hypothesis and its consequences for the possibility of consciousness.

I’ve included my own entry, below. It is both boldly radical and also fairly mundane. I argue that functionalism has a deeper meaning in biological systems than as a mere analog of computation. A missing component of philosophical arguments about function and consciousness is found in the way evolution operates in exquisite detail, from the role of parasitism to hidden estrus, and from parental investment to ethical consequentialism.

I look forward to the next Berggruen challenge. It’s always a fun project and an honor to be evaluated alongside figures like Anil Seth. I don’t enter for the prize money, either, which I don’t want. There is just the joy of writing, whirling, and thinking—a great pleasure that I have this luxury, this privilege, and get to enjoy the fluctuating light of these remarkable places while I contemplate exotic ideas like the nature of consciousness. With luck, there may be a whale next week!

***

Consciousness as Functional Information

000

Introductory Bits

I am writing in the fervor of summer, hazy and warm, by a window aloft in the high canopy of a forest. Boxelder bugs are pacing on the glass outside while a big leaf maple shimmies in an intermittent breeze. Below, down the hillside, invasive English ivy has been gradually overrunning the sword ferns and wrapping up the trunks of the mountain hemlocks and alders. We fight invasions this time of the year, from rodents and ants, and enjoy the evening ballyhoo of coyotes that hop and scurry through the jumbled underbrush shot through with blackberry brambles.

The biological complexity is astounding even here in the hills above a city, and when my thoughts range over the wetlands, the depths of the oceans, and even the deserts of this world, the biomass we coexist with and use to sustain our own lives is beyond astonishing. The metaphor that comes to mind is that life is an unstoppable juggernaut, spreading, diversifying, and overcoming nearly every environmental extreme to continue, to grow and fill the tiniest to the largest niches.

We once thought life was a miracle, but now we regard it as self-replicating machinery that robustly sieves through the state space of possible configurations through mutation, genetic recombination, and selective pressures to arrive at the prodigious ecosystems that surround us. But why is there all this self-organization? And why does a lesser but still remarkable complexity abound in non-living systems, from swirling galaxies to planetary systems, then down into complex minerals and crystals? Is there a unifying force that moves the universe from the low entropy of initial conditions into this abundance? First there is glomming-together that, in turn, creates puddles of negentropy that sustains the conversion of energy into structure, and next there is the acceleration with the rise of semi-living and living systems. Finally, oh the astonishment at our own precarious part in the complexification of this planet. The social plays, enlarged brains, consciousness, politics, the taming of our environments, leaping out into space, divining the physics of the universe—all from the wavering contingencies of evolution and our nature as social beings.

001

Functional Information Systems

If there is a unifying force, we have to be cautious. The language we live with was built up and honed in the ordinary world of gathering food and throwing stuff [27][38]. When we try to describe all these abstractions discovered by science there is a need for a new language that builds on that semantics but is distinctive in its precision. It has always been an awkward part of the language of adaptation to describe the reasons for the existence of a trait as having a purpose. The purpose of the bird’s long beak is to pick bugs out of thick tree bark. The purpose of a cat’s claws is to grab a bird during a hunt. But purpose is not quite right. Human beings have purposes that emerge from our abstract relationships with completing goals or a sense of destiny. Purposes arise from intentionality or are imposed by outside drivers and needs. For evolution, the traits are a consequence of the algorithm of adaptation, converting Aristotle’s teleology into an alternative, automatic process—teleonomy is one invented, alternative term [32]. More, the role of traits can be semantically neutralized further by using “function” to describe them. The function of the bird’s long beak is to get dinner.

We might even be generous and claim that the function of gravity in accretive stellar disks is to rearrange the matter into planetary configurations. Or we might say that the function of certain chemicals is to catalyze mineralization. For these, countering Aristotle’s necessity is teleomatic where a physical process dynamically evolves towards an end point [32].

There is a pervasive quality to both function and its role in complex systems. And this is where a new theory shows promise for our understanding of the universe and our role in it. Why not unify this into a new force in the universe, one where the information that is gained by organizing matter and energy by function is explicitly recognized. This idea comes with an apt name, “the law of increasing functional information,” in a 2023 paper in Biophysics and Computational Biology [47]. The law codifies that natural systems evolve towards states of greater complexity, diversity, and richer patterns as functions arise or are selected by a combination of natural forces. The paper was published shortly after another paper on “Assembly Theory” that discusses how structural combinations of objects might contribute to our understanding of alien planets [42]. The latter has been met with skepticism in degrees from the evolutionary biology community, but it at least may contribute to our understanding of the teleomatic [26]. 

The idea of evolution as more pervasive that Darwin’s original biological concept has been a recurrent theme in many areas since Darwin released his astonishing first treatise. The expanded topics range from “Universal Darwinism” [9] to “evolutionary epistemology,” [7][8][17] and from “memetics” [33] to “evolutionary economics.” [14] Related ideas have arisen in chemistry [15]. The theme is always the same: apply the core idea of selection of functional success across the spectrum of conceptual areas to provide an automatic organizing mechanism for complex systems.

The law of increasing functional information has some unique properties to it compared to the other, more thematic concepts. It concretizes the idea of how much information is needed for specifying a given function. A function may be something as simple as a stable configuration of molecules on the one hand or may be something more complex like DNA transcription. We can measure it in terms of bits and more bits are needed to specify more complex functions. In order to measure functional information, take the range of configurations of a system that can perform the function and divide by the total number of possible configurations. Take a negative log of that and you have functional information measured in bits.

An example clarifies [24]. Let’s say I want to send a text message to a friend and ask them to feed my dog since I’m stuck at an airport. There are only a few configurations of strings in the English language that can convey the request, like:

Pls feed Duncan. Stuck at SFO.

Can u feed D? Stuck traveling.

Stuck at airport. Can u feed D.?

And we might also add in confusing ones that might be hastily dashed off as we walk, like:

FeedD?Stucj…

That might get the job done with a little back-and-forth clarification. The function is to get Duncan fed.

We can compute the number of functional strings over the possible range of them, which is over 26 English lowercase + 26 English uppercase letters, space, and a few punctuation marks. Even though that sounds simple enough, the space of all the possible strings gets big fast, with 58ⁿ strings of length n if we assume uniform independence character probabilities at first blush. Luckily, we are taking the log of these rare events, so let’s say that we can produce 100 meaningful strings of around 20 characters that get poor Duncan fed (and many more that sound like you’re drunk but still might get the job done), the functional information for that action is:

The number of bits needed to express a function therefore scales with the rarity of successful function in the sea of possible configurations.

Functional information is closely related to general information theory but with the additional requirement that the probability of the functionality of the message or configuration is considered. In information theory, the probability of a sequence determines its information content alone, with rarer occurrences having greater surprise or information. Algorithmic information theory (AIT) is closely related to this concept but involves the specification of string-accepting Turing Machines. In functional information, we have a measure focused on this somewhat loosely defined concept of functions, but one that doesn’t depend on anything more than the available functional configurations. The physical specifics are only important in defining the range of options.

This law of functional information increase now has a quantitative basis for specifying both information and how functionality in composition or behavior becomes information. We can see that rarer structures that perform a function against a backdrop of massive alternative configurations of their parts result in greater information as the numerator shrinks against a growing denominator. Now, how do we get this to increase, especially given the countering laws of the universe like the second law of thermodynamics where energy and disorder move towards a maximum? This often-misapplied law is only applicable to closed systems, and the universe is hypothesized to have begun as a low entropy system that is expanding towards maximum entropy. As matter and energy expanded, however, they did not do so uniformly, and the various forces of repulsion and attraction acted to pull matter together resulting in semi-stable configurations and regions of high energy. The non-uniform nature of the combinatorics of nuclear chemistry and chemical reactions lead to uneven trajectories through the space of possible configurations. This is “first order selection” in the theory and results in patterns of energy and matter that have regularities of equilibria, gradients, and the distribution of elements. The functions are relatively simple, but the consequences set the stage for additional functional selection.

Second order selection operates on core functions of autocatalysis, homeostasis, and dissipation in regions of high free energy. Stars resist collapse by balancing fusion against gravitation. Dynamical systems have energetic changes through dissipation and lead to autocatalytic chemical factories in pools of negentropy. Gravity sorts out rings around stars that collapse into rocky planets, and the nuclear synthesis continues within their cores, pushing energetic flows through the rocks in a churn of energetic dissipation.

This gives way to third order selection as greater, more complex assemblages arise that contain nested, reusable components. With third order selection we are in the arena we are most familiar with when talking about selection, a vast world of reproduction, social sharing, memory, knowledge, and even consciousness, but it begins at the much more basic level of functions of enzymes and molecular assemblies that manage homeostasis and persistence of cell walls.

010

Vigorous Self-Reflective Social Functions

Functionality in evolutionary struggles takes on unexpected and devious forms. Among solitary wasps, for example, there are species that lay their eggs inside paralyzed prey [43]. The larvae eat their way out from the inside in a way that triggers something deeply unsettling in us, like the infamous life cycle in Alien. There may be a special reason for our revulsion in that much of life has overcoming parasitism as a core survival function, although there remains uncertainty as to the force of that function as a proximate-adjacent driver [28][36]. Sexual reproduction is a diversity pump for immune systems on the one hand but also has an assortative mating function where desirable characteristics that demonstrate genetic vigor get selected in mate choice. In some species, males engage in mock combat for mating privileges. Birds develop bigger brains and more complex behaviors [39]. In human beings, displays of genetic vigor are balanced by an array of functions to enhance the survival of children, including female hidden estrus and parental investment effects where aggression is tempered into attachment and caring [1][46].

As social animals, we benefit from cooperation while often engaging in individual and collective aggression, but almost everything we do can be recast as having a participatory function in mating behavior and genetic survival, from social status acquired through artistry to the value of educational attainment for providing novel solutions to problems that enhance our well-being.

011

Consciousness, Survival, and Ego

A key component to our success as organisms are our conscious capabilities. Yet, in some respects they are fairly limited [4]. Learning a new language as an adult demonstrates the odd conundrum of how difficult it is to do learn something that is a fluid and semi-automatic act for our native languages learned in our youth. We can rotate images in our mind and the process is something like actually rotating physical objects in terms of the time it takes to complete the task. We can only hold a few complex facts active in our short-term memory at a time. There are two primary neural models for how consciousness might be implemented in our brains, the Global Neuronal Workspace Theory [2][3][31] and Integrated Information Theory [5][29]. Both theories build on decades of experimental cognitive psychology that shows the curious nature of the parts of our mental experience that rise to the level of consciousness. They have recently been tested to evaluate whether one is more predictive than the other but with mixed results [13]. There are also controversies about whether IIT rises to the level of science even after being evaluated as a scientific hypothesis but suffice it to note that both neuronal theories are compatible with the law of functional information increase. They just propose differing functional systems to get the job done as do related ideas [25]

Integrating functional information increase into considerations of consciousness begins to answer the “why” question of phenomenal experience by pinning its existence to emotion and ego [40]. Given the pinnacle that social bonding and sexual pairing occupy as functional drivers of evolutionary change, are the proximate factors related to that core component also the drivers of phenomenal experience? The way this potentially works is that a self-reflective “I” with a resilient ego is needed for understanding one’s status in social grouping, one’s personal appearance as separate from others, and the structure of one’s personality. Wants, likes, doubts, and the pain of unrequited love all are eruptions from having an ego that can observe as if it is an inner being, separate but made whole by our bodies and our relationships. This represents a sea-change for functional information increase by providing an executive function that mimics and models the whole of the unconscious and the impact of our actions in relation to the environment.

In other species we see the roles that hormones play in mating and aggression and how those functions fulfill the function of genetic sorting and mixing. We still retain related circuits but are more driven by the drama and choices in our social mixing. Having a central sense of self is essential to discovering and maintaining status as complex social beings.

100

The Problem of Phenomenal Consciousness

At first glance, there hardly seems to be a problem with the concept of consciousness. We are evolved creatures. This consciousness is just one of our evolved functions providing high information gain. What a piece of work is man! Yet it still evokes mystery along several dimensions. An eternal soul was smuggled into Christianity from Greek thinking and came to dominate pre-scientific thinking in Europe about the relationship between moral goods, personal responsibility, and ultimate purpose. Islam developed a similar idea while the Eastern traditions already had a special reverence for and transpersonal cosmologies built around the role of the soul in reincarnation and impermanence.

With Descartes came new arguments that began with the only truly knowable thing, our own thoughts, and set philosophy up for a long, serpentine odyssey through everything from idealism (there is only mind) to dualism (mind and body coexist somehow but are different things). Then came the scientific revolution that brought with it the decentralizing of humanity in the universe and a new view of our status as special animals, but animals, nonetheless.

But the problem has remarkable persistence in philosophical circles [10][11][35]. Scientists do not typically concern themselves with these issues. It seems clear that consciousness resides in our brains. We can poke, prod, shock, lesion, and experimentally test people and there is no evidence to the contrary. Some philosophers have, in turn, suggested that claiming otherwise or focusing on the linguistic problems of what roles “I” and “self” have in our contemporary discourse is a remnant of animistic language games and of little value even within philosophy [6][18].

Still, a goal in this paper is to test the idea that functional information as a fundamental law of the universe might help illuminate or even solve these pesky mind-body conundrums. We begin with the problem of reconciling phenomenal consciousness—our experience of being something and feeling things—with this meat bag of neuronal interactions. Philosophers often use conceptual scenarios to illuminate problems, and this is no different.

Let’s say there is a scientist named Mary who lives in a monochrome room but has complete knowledge of the physical world and all the causal and property details that have to do with color. She knows about red as having a certain frequency and she also knows about how the spectrum works and how eyes function and the brain processes color. The argument is that if she exits the room and sees red for the first time, she has the experience of red which is new information not present in the facts that she knows until then. This suggests that the experience or “qualia” of red is a new fact that is disconnected from the physical explanations. 

There is an intrinsic limit in the set-up, however. If Mary truly did know all the physical facts of the universe it would include this functional information encoded in her by the contingent evolution of her brain which includes consciousness. To reach the proper level of complete knowledge she could very easily design and build a color projector that would produce the required frequencies of light. She would then have the experience despite the rest of the room being monochromatic. She might design a fine VR system, in fact, that would allow her to have better knowledge of any number of critical human skills like the proprioceptive sensations of hurling a ball, a function that is tied directly to survival by physical actions. The problem also assumes Mary has the capability of experiencing red, which means that she is an ordinary human being with all the capacities needed to experience the color. So, the proper scope of “complete knowledge of the physical world” should involve all the implications of functional information including the reflective properties of consciousness as an important addition to neural and cognitive functioning. In fact, excluding the phenomenal part of information-physical consciousness presupposes the outcome; there is no way to connect one to the other. I’m claiming there is good reason to connect them based on functional information increase as an addendum to the billiard-balls physics presupposed in the problem set-up.

Maintaining that Mary’s knowledge was insufficient to account for the true scope of complete physical reality means this counterargument could be used for any causally potent theory where the physical produces phenomenal consciousness, but it takes on a special character in the face of how other experiences more directly contribute to functional information increase. For instance, the experience of being snubbed by a potential love partner ties the emotional ego/self to evolutionary ultimate causation and reproduction. The desire for a given partner is proximate with physical traits communicating robustness that may contribute to offspring survival against microorganisms. It also is tied into the complex overlay of parental investment equations, social standing, and all the drama that supercharges our societies, as discussed previously. Unromantically, the heart wants what this functional matrix wants. To say that Mary could know all the physical facts about a love snub without experiencing it herself is to rob the physical fact-world of some of its most essential, beautiful, and tragic aspects.

This strange idea of the separability of phenomenal experience from the physical extends to the notion that we can conceive of zombies that are physically just like us but lack the phenomenal. Because it can be conceived in this way it suggests that phenomenal experiences are different than other physical properties that are imparted their semantics by their relationships to other physical things. But this depends mightily on this conceivability claim and, given the physical facts we know, combined with the necessary and sufficient support of functional information increase composing our neural machinery, it suggests it is simply not possible for such zombies to exist in any possible world, for the only method for creating a zombie like that is using an evolutionary process that would require differences in contingencies, connectivities, and so forth. That rules out the similarity-sans-phenomenal-experiences that is the claimed outcome. We are unique, I’m afraid, and functional-information-physical-phenomenal unique. Summing up the counterclaim: no zombies without showing the work and God can’t do otherwise either. There is only one toolbox.

Another way of considering this problem is to treat functional information increase as a kind of “process” panpsychism. In panpsychisms (and protopanpsychisms) we hypothesize that to situate phenomenal consciousness alongside physical reality we need to make the phenomenal a basic property of reality, to weave it into the causally closed nature of reality as something fundamental. For example, we might claim that physics describes the fundamental particles, waves, fields, and so forth as having interactive properties like spin, charge, mass, and so on, but there is nothing in that bag of properties that says exactly what an electron really is; properties are not the thing itself. So, in panpsychism one plan is to just make the thing itself a bit of phenomenal material. We have psyche in everything, hence the “pan.” It might be not quite a complete phenomenal thing, though, so we might also add in a “proto.”

This leads to some curious problems, however, like how exactly do these phenoms or proto-phenoms combine in human brains to achieve the complete effect? [12][41] A “process panpsychism” based on functional information solves this difficulty, however. Functional information increase is a fundamental force of the universe, but it is unlike other forces. Each of the other fundamental forces (electromagnetism, gravity, strong and weak nuclear forces) involves basic changes of state for particles that then combine at macroscopic scales. This new force involves the creation of information over time using the other fundamental forces as building blocks. This is a process force that constructs over the dynamics and object organizational templates of the other fundamental forces. Its product is information while the other laws’ products are motion, energetic exchange, and state change. With the acceptance that increasing information can lead to phenomenal consciousness we can treat this fundamental force of the universe as a kind of protopanpsychism but without the various “summation” or “combination” problems that they bring to the table. It is instead the super-ultimate causation or urge for the coordinated neuronal structures that support our conscious experiences. Without it, consciousness can’t exist, so it is necessary. More, given no further discoveries of forces, it must be sufficient.

This is perfectly compatible with epiphenomenalism, as well. There is no requirement that the existence of conscious phenomenal experience is causally influential in the world. Perhaps the brilliant neuronal systems are causal on our behavior, but the conscious experience is just watching it happen. When we think we are making stuff occur consciously, we are just observing aspects of our neurology that looks much like that.

This can be bothersome to some people who want to believe that their feeling of being agents in the world is not just an illusion. It can also undermine moral responsibility for one’s actions by suggesting that there is just the roil of determinism underlying our choices and we are just watching it go by and seeing the film footage while believing we did something wrong, right, creative, or meaningful.

As I mentioned in the introduction to this section, these kinds of discussions are often odd and unscientific but sometimes find themselves smuggled into the discussion of scientific hypotheses about consciousness, as seen in recent spats over Integrated Information Theory where panpsychism is bandied about as a pejorative. But there is another topical area where science and consciousness collide into philosophical concerns: quantum theory [19][20][21][22][23]. This strain of thinking is based on stressing the importance of a causally closed universe at the level of Newtonian mechanics. We can’t have done things we claim we intended to do because a Newtonian brain just follows deterministic rules for producing outcomes. That’s not choice, it’s predestination. In quantum theories, there are quantum-interactive components of brain chemistry that allow for or promote the superposition properties of quantum fields. Depending on the theory, this also might explain the “measurement problem” in quantum physics where taking a measurement of a system “collapses” the outcome into patterns of waves or particles or position or momentum.

This again looks like a panpsychism where quantum entanglement is the kernel of phenomenal consciousness and intentionality. It escapes Newtonian determinism but not quantum logic, however, and still leads to a similar summation problem to the other panpsychisms. It does work just fine as an addition to a functional information model as another aspect of how physical laws may have been implemented to achieve the functions that the mind demonstrates. There are examples from nature that support this possibility, like the discovery that quantum phenomena may be part of photosynthesis [16]. New discoveries about quantum phenomena in brain chemistry also keep this topic alive.

For the problem of determinism and intentionality, however, I think there is another angle to cleave this diamond.

101

A Spark in Determinism’s Umbra

We can build a conceptual model that shows how human action, including a phenomenal and intentional self, can be constructed from algorithmic parts, but watch for two things as I proceed. First, there is an observer module that has influence on the outcomes and, second, that we overcome a deep reversion to determinism while still assembling all the pushes and pulls that make decision-making seem active but constrained: “I fretted and fretted, but in the end, there was only one path forward.”

Let’s construct a conscious and intentional robot out of algorithmic parts. To do so, consider the environment, E, of this robot, R, as a collection of signals that are collected into R by a perceptual subsystem, P. These signals are things like frequency and magnitude of lights that arrive at small light sensors, as well as details about orientation and direction of the incident signals. They also include proprioceptive measurements, angles, tilts, and relative forces of R’s actuators and other self-monitoring capabilities.

Next, we transfer those signals into a memory subsystem, M, that buffers them and then places them with timestamps or at least ordering metadata into a longer-term memory system. Next, we need to have R use these signals to respond to E for whatever goals we want from R, whether that is walking along without bumping R’s head against tree branches or folding towels for us since we are told we might be deficient in that practice. So, to act, R can use a collection of action scripts, also stored in M, that have placeholder variables that are then filled with various objects from M. There will be many of these scripts for complex activities and their effectiveness will vary based on past experiences as well as the novel aspects of the current signals that are imminent. Let’s call this system the Action Ranker, AR, and it uses some kind of fitness scoring of actions based on the historical effectiveness of actions given input events, with all those traces stored in M as well. But AR does not need to be a monolithic and linear ranker, it might also use a range of ranking algorithms depending on the category of an action (simple physical acts, linguistic response variation, emotional signaling displays, ethical problem solving). Indeed, different categories of rankers can surface candidates for the same script and the same inputs, resulting in a competition among possible scripts and solutions to a given problem.

We have actions, perceptual signals, and a ranking algorithm that might also be variable based on R’s internal state and arriving signals, all using a vast memory of past responses. Now the question is how do we build up all those traces, scripts, and responses in M? And more specifically, how can we imbue the system with the capability of new and effective responses to novel environmental events?

To do this, we add in a Generative Subsystem, GS, that does a simple thing to create novelty: it tears apart different clauses of actions with metadata scores about their performance with other elements of memory and combines them together into new actions and traces, flips weights in the traces up and down, and does so semi-randomly. Often this just leads to nonsense responses that have terrible weighting as they compete up the pyramid within AR. Those never see the light of day. Occasionally, however, one is a brilliant new solution to a vexing problem. Sometimes they are partially successful and rise within GS enough that they get mutated yet again in a honing process alongside other candidates until, finally, there is a truly new response to E.

We can add in another loop to this algorithmic engine, observer O. O has a collection of algorithms stored in M and takes the working buffers of GS and AR and applies additional honing and refinements, pushing weightings back down into M and doing an additional, important task: when AR has uniformly low-weighted options it can serialize simulation of the environment interacting with each action near the top of the stack or in accord with its own set of algorithms. Then, after running a simulation phase, it can label the highest-ranking actions with metadata stamps, like “O approved!” The actions are then pushed back on the top of AR’s stack and tend to fire off preferentially over the other actions.

This system is pretty simple compared to the extraordinary complexity of the human brain or the even more hypothesized complexity of how consciousness might operate within that wetware, but it illustrates how we can tame the problematic nature of determinism and mind as being separate from body, including the causal impotence of matter in impacting individual minds or the evolution of living things in response to the environmental complexity of the universe.

First, the observer, O, is a collection of algorithms that perform monitored simulations as well as generally react to what is being processed on the action stack, so it really is deterministic. It can direct actions to reorient R by adding preferred actions to the stack. It can also add specialized metadata to anything that passes into its processing buffers, including qualitative symbols that are different from the minutiae of frequencies and rates and whatnot. These symbols may have their own logic in terms of the interventionist algorithms stored in O’s procedural memory, resulting in changes to the overall behavior of R that transcends (oops, that just slipped in…treat it instead as “overrides” or “has a significantly different character than” if you take offense) the operations of the other modules. How those O systems come about is an interesting question. One model is just to apply GS to them as well, effectively combining them together with other novelty-generating subsystems but while also using the simulation process to filter them and refine them. Oh, this is cool, maybe R runs lots of simulations and integrates the new action traces into M during its down time while recharging. All those partial solutions, burbling up from memory, and moving around in a complex series of simulations would be R’s form of dreaming.

For the absolute philosophical purist who wants something that breaks away from determinism (because, you know, the state transitions are all predictable—just not easily reversible—even if we use a pseudo-random number generator within GS for fragmenting and combining historical action/signal combinations), we can just add a real random number generator. Hell, just let cosmic rays or ultraviolet light interfere with the memory traces is GS using a suitable substrate, perhaps a biological one. We could call that the base mutation rate or something. Of course this is unneeded. The complexity of E and especially all those motivated others vying and conspiring against us is so dauntingly unpredictable that I don’t doubt that we can get to correlative intentionality that is a bold enough rational alternative to mysterious muddle-puddles that grant that purity through mysterious forces. Living in America’s political present is enough for me to grant the stunning chaos of the world, at least from my armchair.

There is not a great deal of “folk” intentionality in this model, at first glance. The clockwork spins along, the observer simulates and updates, the path forward is what rises to the top of the stack. Yet, in another sense that is exactly how intentionality might work when finagled into this deterministic framework. O is watching behavioral responses unfold and creating new, refined responses in coordination with GS and simulations. The ones that fire off are honed, refined, and selected by their ranked properties. They are a spark of intention.

Does R experience stuff like us? It’s impossible to say. We can only let R do the talking and maintain disbelief because we share something in our developed makeups as functional information machines—some commonality—so that when R reports an inner self, bundled up in all the self-naturalizing linguistic constructs that are needed to sound sincere and aware and deep, that we trust a bit that it is real and not just an extension of some magical knower embedded like a Cartesian demon, or that we share a cosmic consciousness, or some other theoretical umbilical attached to something (actually) transcendent that is so pure that it just wipes all this questioning away. We can’t know for sure, but all the right tickets have been punched, just like with those around us.

110

Functional Futures

The identification of the law of functional information increase poses an interesting question about the future of human development. If complexity and control work together in a high-functioning balance to grant us this amazing experience of being us, does that have implications for how we should think about our future perspectives and actions? There have been thinkers who have played in that pool in the past, like the radical Jesuit priest Pierre Teilhard de Chardin. For Teilhard de Chardin, there was a literal convergence of humanity as our numbers grew greater and greater until we became as one and identical to God [45]. The architect Paolo Soleri conceived of a similar religious-eschatological yet evolutionary future but often lacked Teilhard de Chardin’s coherence [44].

But since their times, we have new revelations that require evaluating this grand arc of existence. The challenge is identifying how to assert our powers of intellect and collective action to enhance functional information. Is there a conceptual balance between control and freedom implicit in the evolution of large brains and our social systems? Should novelty be curated and emphasized to ensure that social evolution extends this contingent spiral towards new functional capabilities?

We can approach the problem in several ways. First, we may want to emphasize personal and conscious function change as a way of individually embracing the ideas of increasing functional information. Second, we may want to try to improve social institutions and ethical drivers considering these new findings. Finally, we may choose to do nothing whatsoever since we are the fulfillment of super-ultimate causation—functional paragons—in embodied form and will never be otherwise.

At the personal level, if we can enhance our rather limited capacities for active, attentive learning we may also improve our overall functional information. The capacity to remember more, to focus better, and to use internal cognitive strategies all lead in the right direction. The Global Workspace Model of cognition includes a recognition of how biofeedback might work by training the conscious self to manipulate the subconscious functional modules of our minds [4]. Expanding on the theme, consciousness technologies lead towards transhumanism where we combine technological capabilities with our biological machinery to improve our functionality. More modestly, we can simply ramp up our knowledge production, creativity, and productivity, for in every function that we produce, rare and sublime, we are adding bits to the universe.

Expanding our vision to our societies and the planet, we can formulate a normative ethical system that values functional increase. Using a consequentialist-style framework for bit generation shares all the difficulties that also come with predicting happiness-enhancing choices, so it may be best to focus on a rules-based information-increase ethics. We can individually try to increase our bit impact by creating systems of coordination in this ethics while governed by rules of thumb. We also want to maximize happiness with our rules, so the goals have to march forward in sync, constantly refining the approach by modeling the consequences of impactful acts.

The obvious complaint inherent in this pursuit concerns the naturalistic fallacy: should we even try to convert what is into what ought to be? There are two reasons to think there may be value in trying to pierce Hume’s wall with virtues based on functional information increase. First, the impetus for social system function and personal identity is inherently tied to the new natural law of increasing functional information and, second, what “is” is actually an “ought” in the sense of the functional intentionality of the universe.

If we look at the history of virtue from the ancient civilizations forward, the trend was tribal, civic, religious, or role virtues up until the Enlightenment when scientific deconstruction of traditions demolished certitude about the individual’s place in the universe [30]. This has carried along in fits and starts into the 21st century with the dominant themes being individual freedom unmoored from any defined value system, forcing us into conflict over what is good and right. With functional information as a foundation, we see a sketch for how virtue can be reimagined: what are the greatest functional improvements to human life that we can collectively achieve? This has a slightly different character than expansions of compassion that drive much philanthropy and also a different character than the new broligarch exceptionalism that pushes absurdly towards space tourism, Mars, or extreme longevity [34]. Instead, it values organizational function that accelerates information increase. This can be achieved by creating rare and novel functional configurations out of larger and larger collections of diverse parts. For instance, it might be virtuous to create a museum or organization that has a novel educational function but brings together a breadth of new information and physical elements. The educational function of such a place and system increases our bit count while doing minimal other harms. There will always be a social tension for projects of this nature due to the opportunity costs pulled away from basic economic aid, but at least it reorients resources away from status projects and into a bit increase that aligns with this ethical structure. It’s a little old fashioned, but let’s build more museums, galleries, effective governance, and think tanks.

The final notion is that we just keep on keeping on, man. This is our programming, our destiny. We are already here. We are building all this functionality against the unknowable future. Our global economic systems have already lifted much of the world from poverty. Wealth and freedom continue to grow as violence ebbs [37]. Yes, with fits and starts and incriminating inequality while global climate change remains a daunting threat. Our consciousness and conscience can be reassured in our roles once again, moored to this new realization of our function as struggling actors in a play of potentialities that grind down the unknowables into novel futures, throwing off showers of brilliant bits as we change, generations building on generations. The is remains separate from the ought, while the ought builds new functionality to uncover new bits of is.

We can’t, however, ignore the existential threat bound up in this theory and its implications. Given that consciousness is a product of functional information increase in bio-machines, it can transfer to other implementation media, including computer technologies. The race is currently on for artificial general intelligence (AGI) built on the back of the recent successes and failures of large language models. Those of us in the trenches of this changing technology landscape have our doubts that AGI can be achieved through the current approaches that try to infer predictive models from large seas of internet data, but there are other efforts underway that try to embody and evolve intelligence. There have been for at least 80 years but we now have resources to scale the efforts. From what we can guess from functional information increase, the only limits to achieving AGI are resource-bound in time, money, energy, and space, which means there is a certain inevitability to the outcome. In so doing, AGI must be brought into functional harmony with our own future development by establishing limits on control and power for it as a participant in future societies. As an adjunct to our evolution, AGI can accelerate functionality, but existential threats lead to functional information loss at the grossest level, so we have a moral obligation to stop things from falling apart.

111

A Few More Bits

I’m tying this off while a tapestry of ecstatic greens and rippling shadows again breathes along the hillside. The trees grow through one another in a functional interlace as they seek the negentropic bath of sunlight. Vines corkscrew the trunks, biding time as the summer simmers in its dry phase. Soon the sustaining drench of rain will return. The coyotes were more distant last night, nearer to the dogs who always bark back in synchrony with the mating cants. This balance, this dynamo of energetic information, winding and enveloping, is tangled through my consciousness, and in these few more bits I now write.

References

[1] Antfolk, J.; Sjölund, A. (2018). High parental investment in childhood is associated with increased mate value in adulthood. Personality and Individual Differences. 127 (1): 144–150. doi:10.1016/j.paid.2018.02.004

[2] Baars, B. J. (1993). How does a serial, integrated and very limited stream of consciousness emerge from a nervous system that is mostly unconscious, distributed, parallel and of enormous capacity? Experimental and theoretical studies of consciousness. Ciba Foundation Symposium 174 (Chichester: Wiley), 282–303.

[3] Baars, B. J. And Franklin, S. (2007) An architectural model of conscious and unconscious brain functions: Global Workspace Theory and IDA. Neural Networks, 20(9):955-961.  https://doi.org/10.1016/j.neunet.2007.09.013.

[4] Baars, Bernard J. (1988). A Cognitive Theory of Consciousness. New York: Cambridge University Press.

[5] Barrett, A. B. (2014). An integration of integrated information theory with fundamental physics. Front. Psychol. 5, 63. https://doi.org/doi:10.3389/fpsyg.2014.00063

[6] Bennett, M., Dennett, D., Hacker, P., Searle, J., & Robinson, D. N. (2007). Neuroscience and Philosophy: Brain, Mind, and Language. Columbia University Press, NY

[7] Campbell, D. T. (1960) Blind Variation and Selective Retention in Creative Thought and in Other Knowledge Processes. Psychological Review, 67:380-400.

[8] Campbell, D. T. (1974) Evolutionary Epistemology, The Philosophy of Karl Popper, I, Open Court, LaSalle, I., 413-463.

[9] Campbell, J. O. (2016). Universal Darwinism as a process of Bayesian inference. Frontiers in Systems Neuroscience, 10.

[10] Chalmers, D. (1996). The Conscious Mind: In Search of a Fundamental Theory. Oxford: Oxford University Press.

[11] Chalmers, D. (2003) Consciousness and its Place in Nature. In The Blackwell Guide to Philosophy of Mind. Edited by Stephen P. Stich, Ted A. Warfield. 102-142. Blackwell Publishing Ltd

[12] Chalmers, D. J. (2017). “The combination problem in panpsychism,” in Panpsychism: Contemporary perspectives, eds G. Bruüntrup, and L. Jaskolla, New York, NY: Oxford University Press.

[13] Cogitate Consortium., Ferrante, O., Gorska-Klimowska, U. et al. (2025) Adversarial testing of global neuronal workspace and integrated information theories of consciousness. Nature 642, 133–142. https://doi.org/10.1038/s41586-025-08888-1

[14] Dopfer K. (2005) Evolutionary economics: a theoretical framework. In: Dopfer K, ed. The Evolutionary Foundations of Economics. Cambridge University Press. 3-56.

[15] Eigen, M., Schuster, P. (1978) The Hypercycle. Naturwissenschaften 65, 7–41). https://doi.org/10.1007/BF00420631

[16] Engel, G., Calhoun, T., Read, E. et al. (2007) Evidence for wavelike energy transfer through quantum coherence in photosynthetic systems. Nature 446, 782–786. https://doi.org/10.1038/nature05678

[17] Gabora, L. (2019). Creativity: Linchpin in the quest for a viable theory of cultural evolution. Current Opinion in Behavioral Sciences, 27:77-83.

[18] Garvey, J. (2010) Hacker’s Challenge. tpm: The Philosopher’s Magazine, 51, https://philosophypress.co.uk/?p=1583

[19] Georgiev DD (2011) A linkage of mind and brain: Sir John Eccles and modern dualistic interactionism. Biomed Rev 22:81–84

[20] Georgiev, D. (2013) Quantum No-Go Theorems and Consciousness. Axiomathes 23, 683–695. https://doi.org/10.1007/s10516-012-9204-1

[21] Georgiev, D. D. (2024). Evolution of Consciousness. Life, 14(1), 48. https://doi.org/10.3390/life14010048

[22] Hameroff, S. (2012) How quantum brain biology can rescue conscious free will. Frontiers in Integrative Neuroscience. 6: 93. doi:10.3389/fnint.2012.00093.

[23] Hameroff, S., Penrose, R. (2014) Consciousness in the universe. Physics of Life Reviews. 11 (1): 39–78. doi:10.1016/j.plrev.2013.08.002 

[24] Hazen, R., Griffin, P., Carothers, J., et al. (2007) Functional Information and the Emergence of Biocomplexity. In: National Academy of Sciences; Avise JC, Ayala FJ, editors. In the Light of Evolution: Volume I: Adaptation and Complex Design. Washington (DC): National Academies Press (US); 2007. 2. Available from: https://www.ncbi.nlm.nih.gov/books/NBK254300/

[25] Hunt, T., and Schooler, J. W. (2019). The easy part of the hard problem: a resonance theory of consciousness. Front. Hum. Neurosci. 13, 378. https://doi.org/doi:10.3389/fnhum.2019.00378

[26] Jaeger J. (2024) Assembly Theory: What It Does and What It Does Not Do. J Mol Evol. Apr;92(2):87-92. https://doi.org/doi:10.1007/s00239-024-10163-2.

[27] Kampourakis, K. (2020) Students’ “teleological misconceptions” in evolution education: why the underlying design stance, not teleology per se, is the problem. Evo Edu Outreach 13, 1  https://doi.org/10.1186/s12052-019-0116-z

[28] Lively CM. (2010) A review of Red Queen models for the persistence of obligate sexual reproduction. J Hered. 2010 Mar-Apr;101 Suppl 1:S13-20. https://doi.org/10.1093/jhered/esq010

[29] Lombardi, O., & López, C. (2018). What Does ‘Information’ Mean in Integrated Information Theory? Entropy, 20(12), 894. https://doi.org/10.3390/e20120894

[30] MacIntyre, Alasdair (1984). After Virtue: A Study in Moral Theory (2nd ed.). Notre Dame, Indiana: University of Notre Dame Press.

[31] Mashour, G., Roelfsema, P., Changeux, J.P., Dehaene, S. (2020) Conscious Processing and the Global Neuronal Workspace Hypothesis. Neuron. Mar 4;105(5):776-798. doi: 10.1016/j.neuron.2020.01.026.

[32] Mayr, E. (1982) The Growth of Biological Thought. 48-50. Belknap Press. Cambridge, MA

[33] McNamara A. (2011) Can we measure memes? Front Evol Neurosci. May 25;3:1. https://doi.org/doi: 10.3389/fnevo.2011.00001.

[34] Morris, A. (2025) What You’ve Suspected Is True: Billionaires Are Not Like Us. Rolling Stone. https://www.rollingstone.com/culture/culture-commentary/billionaires-psychology-tech-politics-1235358129/

[35] Nagel, T. (1974) What is it like to be a bat? The Philosophical Review, 83(4):435-450

[36] Ochoa G, Jaffé K.(1999) On sex, mate selection and the red queen. J Theor Biol.;199(1):1-9. https://doi.org/10.1006/jtbi.1999.0931

[37] Pinker, S. (2018) Enlightenment Now: The Case for Reason, Science, Humanism, and Progress. New York; London Viking; Allen Lane

[38] Reiss J.O. (2005) Natural selection and the conditions for existence: representational vs. conditional teleology in biological explanation. Hist Philos Life Sci. 27(2):249-80.

[39] Sayol, F., Downing, P.A., Iwaniuk, A.N. et al. Predictable evolution towards larger brains in birds colonizing oceanic islands. Nat Commun 9, 2820 (2018). https://doi.org/10.1038/s41467-018-05280-8

[40] Schutter, D. And van Honk, J. (2024) Extending the global workspace theory to emotion: Phenomenality without access. Consciousness and Cognition, 13(3), 539-549, https://doi.org/10.1016/j.concog.2004.05.002.

[41] Seager, W., (1995) Consciousness, information, and panpsychism, Journal of Consciousness Studies, 2(3):272–288.

[42] Sharma, A., Czégel, D., Lachmann, M. et al. Assembly theory explains and quantifies selection and evolution. Nature 622, 321–328 (2023). https://doi.org/10.1038/s41586-023-06600-9

[43] Smith, M., Rodriguez, J., Whitfield, J., Deans, A., Janzen, D., Hallwachs, W., Hebert, P. (2008) Extreme diversity of tropical parasitoid wasps exposed by iterative integration of natural history, DNA barcoding, morphology, and collections, Proc. Natl. Acad. Sci. U.S.A. 105 (34) 12359-12364, https://doi.org/10.1073/pnas.0805319105

[44] Soleri, P. (1981) The Omega seed: An eschatological hypothesis. Anchor Press/Doubleday, NY.

[45] Teilhard de Chardin, P. (1964) The Future of Man. Harper & Row, NY.

[46] Trivers, R.L. (1972). Parental investment and sexual selection. In B. Campbell (Ed.), Sexual selection and the descent of man, 1871-1971 (pp. 136–179). Chicago, IL: Aldine. ISBN 0-435-62157-2.

[47] Wong, M. L., Cleland, C. E., Arend, D., Bartlett, S., Cleaves, H. J., Demarest, H., … & Hazen, R. M. (2023). On the roles of function and selection in evolving systems. Proceedings of the National Academy of Sciences, 120(43). https://doi.org/10.1073/pnas.2310223120