1.6.1 Coming into Being
1.6.1.1 In early childhood, we develop a theory of mind: we learn that others have beliefs, desires, and perspectives like our own—but also different from our own (Agüera y Arcas 2025, 249). By modeling the world and internalizing its real patterns, we learn to predict others. To understand you, I have to model you as the kind of thing that models me (Seth 2022, 167). Those predictions become scaffolding for self-modeling. You become the kind of thing that predicts others, and in doing so, you begin to predict yourself (Dennett 2009, 345). In time, you grasp that you occupy a unique perspective. We’ll call this transition becoming a being: a set of closures that carries a point of view through time—an integrated perspective and the beginnings of a self-model that persists.
1.6.2 The Subjective and the Objective
1.6.2.1 A being’s unique perspective is structural. It arises from physically occupying a specific place in the environment—literally having a unique viewpoint on the world. It also arises from the unique history that brought you here (all those combinations and negations), along with the real patterns you’ve internalized: your body and mind. Each movement through time changes what it is like to be you, even if only in small ways. In this way, in our complex system, subjectivity is fundamental.
Nagel talks about “facts that embody a particular point of view” and notes that these facts “do not consist in the truth of human propositions expressible in human language” (Nagel 1974, 442). We can think of his facts as our real patterns—patterns that are not themselves propositional (1.5.5.3). We should take this inaccessibility not as something mystical, but as a consequence of complex systems: internal structures that cannot be fully exported into public language.
Hannah Arendt, in talking about what makes us human: “we are all the same, that is, human, in such a way that nobody is ever the same as anyone else who ever lived, lives, or will live” (Arendt 2018, 8). She goes on to say we would no longer be human if we left Earth, which she describes as the “very quintessence of the human condition” (Arendt 2018, 2). That is, each of us occupies a unique position within a specific envelope. To be a subject is to predict from that position. Importantly, there is no god’s-eye view, no being outside of the system, for any one being.
As a consequence, any move to objectify the situation pulls away from the very thing we are trying to describe. Nagel again: “If the subjective character of experience is fully comprehensible only from one point of view, then any shift to greater objectivity … does not take us nearer to the real nature of the phenomenon: it takes us farther away from it” (Nagel 1974, 444–445). We confuse this fundamental fact of inaccessibility with something non-material—but not everything material is accessible.
In the picture I’m sketching, subjectivity is not some ghostly substance. It is the form prediction takes when prediction must be made from somewhere.
Each being is a limited, historically contingent, and uniquely physically situated model of a world too complex to be represented by any single being. Its internal patterns are real, but they are not fully exportable. They can be lived, used, and partially expressed but not fully copied into public language. In this way, minds operate above the inductive threshold. They are private, high-dimensional, history-laden, and irreducible to rules—in short, partially predictable but not fully exportable (1.5.5.3).
This is why Nagel is right that objectification moves us away from the subjective character of experience. To describe a perspective from the outside is to remove the standpoint that makes it that perspective. But this does not mean subjectivity is nonphysical or mystical. It just means that some physical patterns are fully available only from within the organization that holds them.
1.6.2.2 Objectivity, in this account, enters later (similar to our description of logic in 1.4.2.2). It is not a view from nowhere, since no such view exists inside a complex system. Objectivity is what stabilizes across our views. Language lets beings share inferences from their internal patterns, compare them, criticize them, and correct them inside a public envelope. The objective is therefore not the opposite of the subjective but a lower-dimensional, serialized, and stabilized overlapping pattern shared between subjective perspectives.
In this sense, the path runs from subjectivity to intersubjectivity to objectivity. A private pattern becomes public only by being compressed into language; it becomes objective only by surviving correction across many perspectives and over time. Science, and knowledge more broadly, is not less objective because it is constructed; it is objective because it is constructed under unusually disciplined conditions of public correction. Wittgenstein again: “And the concept of knowledge is coupled with that of the language-game” (Wittgenstein 1969, 78).
Each of us carries patterns that align with this shared objective view and patterns that do not. That difference is not an error to eliminate. It is what makes each of us us, what keeps us talking, and what drives science and knowledge forward.
Each of us is a node, a weight, a subject. Our collective, corrected output is what we think of as objective knowledge. We give more weight to beings—and to institutions and methods—that are historically good predictors, and we negate claims that repeatedly fail. The objective is not sitting out there fully formed, nor is it merely invented. It is continuously stabilized.
In a related spirit, Walker describes science as: “definitionally unstable because it is not an objective feature of reality; instead, it is more accurately understood as an evolving cultural system, bred of consensus representation and adaptive to the new knowledge we generate” (Walker 2025).
1.6.3 Phenomenological and Self Closure
1.6.3.1 At every instant, diverse data streams into your mind—visual, tactile, auditory, interoceptive—giving rise to the what it’s like of experience. We may think this data simply flows in, but much of what becomes conscious has already been processed, selected, and shaped. Awareness is often late to its own causes. Our perceptions arrive not as raw channels but as a unified scene—bound together into a coherent perspective through what cognitive science calls binding processes (Harris 2019, 25–26).
Let’s call this unification of perception into a single what it’s like (Nagel 1974, 436) field phenomenological closure. Many signals become one frame: a coherent point of view anchored to a body and location—a bit in itself, if you will. Phenomenological closure weaves a shifting constellation of perceptions—a real pattern internalized moment by moment—into what it is like to be you.
In this way, it resembles lexical closure. Lexical closure happens when words stabilize their roles by relating to other words. Phenomenological closure arises when perceptions stabilize into one unified presence. What it’s like is not an extra ingredient in the system. It is a consequence of a structurally situated viewpoint and the integration of its contents into a stable, experiential whole.
In this framing, the hard problem (Chalmers 1995) is not “why is there experience at all?” but rather under what conditions do systems achieve phenomenological closure? When does a system’s internal modeling become not merely predictive, but perspectival—delivered as a unified what it’s like frame?
To state the obvious, this closure requires a body in a world—a perspective within an envelope. As neuroscientist Anil Seth writes on the boundary of self and world: “What it means for something to exist is that there must be a difference—a boundary—between that thing and everything else. If there were no boundaries, there would be no things—there would be nothing.” And “living systems actively maintain their boundaries over time” (Seth 2022, 197).
Lastly, there may be a question about whether phenomenological closure depends on biological “wetness.” Your mind is bathed in neurotransmitters as neurons fire. Neurotransmitters like dopamine and serotonin regulate mood and emotion and may help give rise to the what it’s like feeling. One could imagine this as a requirement for phenomenological closure—the difference between wet (biological) and dry intelligence (AI), and thus the difference between being conscious and not.
In this view, hardware and software are deeply intertwined. As Seth says, “there’s no sharp separation between ‘mindware’ and ‘wetware’ as there is between software and hardware in a computer. The more you delve into the intricacies of the biological brain, the more you realize how rich and dynamic it is, compared to the dead sand of silicon,” and thus “brains are the kinds of things for which it is difficult, and likely impossible, to separate what they do from what they are” (Seth 2026). Or as Kauffman puts it: “The know-how is not outside that propagating organization. The know-how is the propagating organization” (Kauffman 2000, 111). This view creates a host of challenges for traditional computational functionalism, which talks of separate hardware and software.
1.6.3.2 Imagine phenomenological closure without persistence: there would be something it’s like to be you at each moment, but no moment would carry forward. To complete the picture, we need a mechanism of transmission across time. Let’s call this self-closure: the binding that stitches momentary perspectives into a single continuing agent—the same thing that acts, remembers, expects, and moves through the world.
Self-closure mirrors pragmatic closure. Pragmatic closure stabilizes meaning across a community through correction. Self-closure stabilizes identity across time through continuity constraints—memory, expectation, habit, narrative. It creates the envelope-through-time in which a being doesn’t merely have experiences, but persists as the one who has them.
Where will this leave AI that develops so-called world models? How much data will they need, and is there a relation between phenomenological and self closure? Does one accelerate or enable the other? We could imagine an AI that navigates the world but lacks the what it’s like of phenomenological closure; absent this closure, it may take a massive amount of data to learn to properly navigate the world.
1.6.4 Predicting the Self
1.6.4.1 What is striking is how much of what we’ve described is believed by neuroscientists to arise from the act of prediction.
Seth defines prediction as the core of who we are: “The experience of being me, or of being you, is a perception itself—or better, a collection of perceptions—a tightly woven bundle of neurally encoded predictions geared toward keeping your body alive. And this, I believe, is all we need to be, to be who we are” (Seth 2022, 154). What emerges is a living system actively modeling its world and body. We might say, “I predict myself, therefore I am” (Seth 2022, 201).
Friston’s free-energy principle offers a similar process of prediction, which he presents as a unified brain theory (Friston 2010).
1.6.4.2 What we think of as choosing may just be the negation of predictions. Seth again: “We have precisely one conscious experience out of vastly many possible conscious experiences,” and “each experience reduces uncertainty with respect to the range of possible experiences, . . . redness is redness because of all the things it isn’t, and the same goes for all other conscious experiences” (Seth 2022, 53).
We want to say we make choices, but we really make predictions—some more accurate than others—that are negated by the envelope of the world and our minds. Those not negated persist and are reinforced, becoming more robust.
What we think of as choosing may just be the space of potential predictions collapsing into a single trajectory—a prediction that survives rather than one chosen—a pruning of paths we call a decision (Agüera y Arcas 2025, 277). What we experience as choosing is the phenomenological signature of this narrowing.
Thinking is the evolution of thoughts through time.
Dennett, discussing this dynamic, writes: “The brain’s strategy is continuously to create ‘forward models,’ or probabilistic anticipations, and use the incoming signals to prune them for accuracy” (Dennett 2009, 169). Under this view, free will is the narrowing of an envelope that leads to action—a constriction until one prediction remains. It’s a process of negating possibilities, the result of which we call choice.
“Free will is the coupling of a human mind to otherwise random processes inside a brain” (Dyson 1988, 295).
1.6.4.3 Imagine the mind as a dynamic envelope continuously shaped by perception (incoming signals) and internal change (updated real patterns). The envelope’s current shape constrains what happens next: which predictions survive contact with the world and which are negated.
What’s strange is that the envelope’s bits appear to shape themselves. The system making predictions is also the system being predicted. This is the recursive process Douglas Hofstadter called a strange loop—a form of self-reference in which the mind climbs levels of abstraction and yet “returns” to itself as a closed cycle (Hofstadter 2007, 101–102). In essence, a self-referential prediction. This process echoes our earlier autocatalytic story: not molecules this time but neurons, bootstrapping into a process that sustains and revises itself.
Neurons (bits) form a mind (envelope) within which a series of closures occurs. Perception provides the perpetual prompt. Neurons fire. Predictions are made. Some persist, and those survivors become the next input. All the while, the mind is bathed in an ever-changing soup of neurotransmitters that determine which neurons may or may not fire. In this sense, emotions provide regulation and feedback, influencing what is reinforced and what is suppressed. Some connections strengthen, others weaken. The configuration of bits changes. What is possible shifts with each pulse of the clock. Our thoughts are a noisy, history-dependent walk through the adjacent possible.
Phenomenological closure gives the walk a unified what it’s like frame. Self-closure carries that frame forward, stitching moments into a continuing agent. The strange loop is what makes the whole thing reflexive: a system that not only predicts the world but continually remakes the predictor. The loop persists because it is a self-maintaining pattern of negation and reinforcement inside an envelope—a complex system.
1.6.4.4 Under this view, consciousness requires recursion—outputs must become inputs. It’s a process unfolding through time. To be aware of yourself as a self, you must observe yourself.
The process is reflexive: the output influences the next input. I leave myself a note on the nightstand to remember something in the morning.
The process is dynamic—the mind has plasticity. And unsurprisingly, there’s increasing interest in developing AI with “liquid neural networks”—systems that adapt in real time, much like the mind (Howarth 2025).
Because of the same earlier dynamics—noise, search-space limits, recursion—the process can’t be known ahead of time (1.2). To know your next thought, you have to think it. Hofstadter directly relates this feature of strange loops to Gödel (Hofstadter 1979).
The world provides the envelope that negates predictions. Conversely, we might think of dreaming as predicting without the world’s constraints to collapse our thoughts; thus, the process and its possibilities remain more open.
This view also offers a lens for thinking about philosophical zombies. The zombie thought experiment asks us to imagine all the outward structure of consciousness without the inward what it’s like. But on the account developed here, this separation may be incoherent: if the relevant closures are genuinely present—phenomenological closure, self-closure (that is, recursive prediction within a history-bearing envelope)—then there may be no further ghostly ingredient left to subtract. The zombie intuition works only if consciousness is treated as an extra property added rather than as a closure achieved by a process through time (much like our description of intentionality in 1.5.4).
The real mystery, at least to me, is counterfactuals. How can the mind imagine things that never happen—examples not in the training data? What’s the mechanism? Is there, in essence, a “not” operator? Imagining counterfactuals is what makes the mind larger than the world. We can imagine worlds with different physical laws and different pasts. How do counterfactuals relate to the predictive nature of the mind? Are they just a form of prediction? This seems like a ripe area for further research, which is why the ideas found in constructor theory are so interesting—it’s working to rewrite physical laws in terms of counterfactuals (Deutsch 2012).
1.6.5 Criticality and Complexity
1.6.5.1 There’s a concept in physics called criticality, which refers to the threshold at which a system undergoes a phase transition or sustains a chain reaction. In our language, it’s the period right before closure—a point at which things are both stable and unstable, beginning to coalesce but not yet closed. It’s the time right before the ice forms.
This idea underlies the critical brain hypothesis, which posits that the “brain is always teetering between two phases, or modes, of activity: a random phase, where it is mostly inactive, and an ordered phase, where it is overactive … between these phases, at a sweet spot known as the critical point, the brain has a perfect balance of variety and structure and can produce the most complex and information-rich activity patterns” (Beggs 2023).
At the critical point, the brain is stable enough to process information but flexible enough to adapt quickly; in that sense, operating near criticality is highly efficient. A highly ordered brain is not equipped to adapt, and a brain that is too chaotic becomes dysfunctional. Think of the difference between being under anesthesia versus having an epileptic seizure (French 2026).
What’s fascinating about this theory is that a defining property of the mind may be its ability to oscillate between stability and instability—just as we described in our complex system (1.1.5.2). To be thinking, or even to be conscious, might be the process by which our minds teeter between stability and instability, order and randomness, structure and noise.
To be conscious, in this picture, may require operating near the threshold of closure: stable enough to bind to the world but unstable enough to revise understanding (i.e., to make predictions and internalize patterns).
1.6.6 Consciousness and Complexity
1.6.6.1 “The brain is always trying to find closure, to make its best guess” (Seth 2022, 145).
As described, consciousness is a form of complex system with a dual closure—one that creates a what it’s like, and one that propagates the self through time. In the case of a mind, the predictions are not about the environment but about the self—a process by which recursive self-predictions give rise to consciousness (i.e., a strange loop).
Consciousness must be generated through time because it requires predicting a future that is unknowable. This unknowability arises from noise, path dependence, and the nearly infinite combinatory space of the future (1.2). In essence, there would be no self without time and without noise. Because of this, descriptions of consciousness that measure certain states or values—like the quantity phi in Integrated Information Theory (Tononi 2004)—may provide a snapshot but can’t fully capture what it means to be conscious. It is a flow through time, not a state (Seth 2026). It’s a process that exists above the inductive threshold and one in which we again see the insights of the phenomenological philosophers.
There’s a deep insight here: to be alive is to resist decay and disorder—to resist entropy and the second law of thermodynamics (1.0.3). To be conscious is to be anchored in a process through time that degrades—that’s what gives rise to the need to predict.
We are not ghosts in machines; we are machines made ghostly by time.
If this is a requirement, then it raises the question of whether silicon can ever be conscious. A device that doesn’t degrade cannot be conscious by this definition. In this case, consciousness would be substrate-dependent—on one that degrades and that, in response, participates in some self-maintaining, entropy-resisting process (i.e., a complex one). If that degradation happens over millions of years instead of a hundred, does that satisfy the requirement? Would it just be less conscious? Or is consciousness deeply interwoven with the brevity of our lives? To be conscious would seem to require vulnerability to entropy, to require decay and also to struggle against it.
Viewing consciousness this way—as a type of closure that arises in a complex system through time—sidesteps requirements like we find in panpsychism (consciousness as a property of all matter) and in various other forms of dualism. Our description here mirrors the discussion on intentionality—again, not as a mystical substance, but as one that arises through closures in complex systems.