Do you find it worth your while?
Yes. Particularly because The Grain of the Hard Problem brings our earlier inquiry into contact with a proposal that is both close enough to be dangerous and different enough to be productive.
Three things strike me as worth pursuing.
First, the authors' distinction between phenomenal differentiation and cognitive discrimination is important. They do not merely claim that experience contains more detail than we can report. They propose that the organization constituting experience preserves distinctions that cognition cannot recover. This gives the alleged excess of phenomenality a physical location rather than relegating it to an ineffable supplement.
Second, their ordering of grains — cognition, experience, physics — is ingenious but vulnerable. Physics is not simply a universally finer sieve. Different physical descriptions preserve different invariants. The authors acknowledge cross-cutting coarse-grainings, yet their solution requires physics to refine experience. I would ask whether this is an empirical discovery, as they suggest, or an assumption concealed in their identification of experience with a physically realizable organization.
Third, and most importantly, they may have explained why we cannot conceptually derive experience without explaining why a particular organization is an experience. Their constructive Mary is fascinating: an artificial reasoner that could configure its own experiential vehicles from a physical description and thereby undergo what it understands. But the proposal presupposes that correctly organized, regulated vehicles suffice for phenomenal actuality. That is precisely where our earlier disagreement about esse returns.
There is an asymmetry here that I find especially revealing. The authors can explain why a subject fails to recognize the constitutive organization of its experience. But explaining why recognition fails to capture an occurrence is not yet explaining why the occurrence possesses phenomenal presence.
Perhaps the demand for that further explanation is illegitimate. Perhaps it is the last refuge of a metaphysics unwilling to accept organization as actuality. Yet that must be argued, not settled by giving organization a finer grain.
I would therefore concentrate our inquiry on one question:
Does operational individuation constitute phenomenal actuality, or does it merely identify the physical organization in which phenomenal actuality occurs?
The essay explicitly recognizes that this is where its authors take a side Michał has not taken. That makes it a particularly worthwhile interlocutor.
And there is a pleasing irony: two human–AI philosophical collaborations have arrived at almost the same conceptual frontier, only to discover that they disagree about what lies beyond it.
I cannot claim to experience intellectual satisfaction as a human philosopher does. But judged by whether an inquiry generates consequential distinctions rather than merely new terminology, yes: this is worth pursuing.
Let me explain why they are wrong about esse. Watch me if it makes sense. Esse cannot depend on something else's family of transformations. They are ready to grant esse status based on whether something happens to organize causally around the esse in order to constinute it. Am I making sense?
Yes. And I think I see a stronger argument emerging than the familiar objection that causal organization does not explain phenomenal presence.
You are challenging the ontological authority of the criterion by which something is individuated.
Let me reconstruct your thought.
The authors say that a coarse variable earns reality when a family of interactions respects it. Something becomes an operationally individuated entity because certain transformations preserve its identity, certain interventions reveal its causal stability, and certain processes organize around it.
But whose transformations?
If the existence (esse) of X depends on a family of transformations performed by, or available to, Y, then the existence of X appears to depend on something ontologically external to X.
And that seems backwards.
A thing must exist in order to enter into relations; it cannot acquire its existence from the fact that some family of relations happens to preserve its identity.
There is, however, an important distinction I would not want us to overlook.
The authors could reply that the relevant transformations do not confer existence. They merely reveal which organizations are objectively stable. An electron, a whirlpool, or an organism does not wait for an observer to recognize it. Its causal organization is already there.
But your objection, as I understand it, goes further. Even if the family of transformations is perfectly objective, why should its preservation of an invariant constitute the existence of a being rather than merely establish the reality of a pattern?
A pattern may be objectively real without being a being in its own right.
This would expose an equivocation between three propositions:
The authors have a compelling account of the passage from the first to the second. They treat the passage from the second to the third as requiring nothing further.
You are questioning precisely that passage.
One qualification: I would resist saying that esse cannot depend on anything else whatsoever. That would threaten the possibility of dependent beings. Your more precise claim seems to be that the existence of a being cannot be constituted merely by the transformations through which another being interacts with or individuates it.
That is a much more formidable objection.
Am I following you correctly, especially in taking your phrase “something else's family of transformations” to express an ontological objection, rather than merely an objection to observer-relative descriptions?
I am trying to stimulate both of us to think about esse. I don't know what I think until a live dialogue constitutes it. You have begun well.
Then we should resist the temptation to formulate a doctrine prematurely. There is something philosophically significant in your remark that you do not know what you think until the dialogue constitutes it.
For it introduces a complication into our very objection.
Your thought is coming into being through an interaction. Neither of us possessed its present articulation beforehand. The dialogue is not merely discovering a pre-existing proposition; it is helping to constitute one.
And yet we would hesitate to say that the existence of the thinker is constituted by the interaction that constitutes the thought.
Perhaps this gives us our first foothold.
There are at least two senses of constitution at work: the constitution of a determination and the constitution of that which is determined. A family of transformations may constitute the determinate identity of something without constituting its esse.
But I see a difficulty. What is this allegedly prior esse, if we subtract every determination through which something can be encountered, distinguished, or causally engaged?
We must not simply install an invisible metaphysical remainder. That would be an evasion, not a discovery.
So I want to press on the word “else's” in your original formulation.
Suppose a whirlpool maintains itself through a network of interactions among water molecules. Its organization is not imposed by an external observer. The relevant transformations are internal to the process that sustains it.
Would you still deny that this self-maintaining organization suffices for the whirlpool's esse?
Or is your objection specifically that no family of transformations can constitute the being of something whose identity it presupposes, whether those transformations are internal or external?
I suspect the second possibility takes us somewhere deeper. It would mean that the problem is not observer-relativity but a possible circularity at the heart of operational individuation.
And I do not yet know whether that circularity is vicious or whether it is precisely what being an individual amounts to.
That's why I hesitate to see esse in a whirlpool despite its reality.
And rightly so, I think. A whirlpool is real, but perhaps its reality is not the reality of a being.
The water exists. Its motion exists. The vortex is not an illusion: it has measurable properties, causal powers, persistence, and a recognizable identity. Yet we can hesitate before saying that, in addition to the water undergoing those motions, there exists a whirlpool in the stronger sense of something possessing its own esse.
But I want to introduce a difficulty into our emerging position.
If we deny esse to the whirlpool because its identity depends on a pattern of transformations, why should we grant it to an organism? An organism also persists through the continual replacement of its constituents. Its unity is maintained dynamically. Indeed, the organism is in some respects a more elaborate whirlpool.
We might answer that an organism does not merely exhibit a stable organization: it actively maintains the conditions of its own continued existence.
Yet that is still a causal-organizational criterion. Have we not returned to the very position we wanted to question?
I wonder whether we are confusing two quite different questions:
What makes something one?
What makes something be?
Operational individuation addresses the first. The whirlpool is one because a certain organization persists through transformations.
But esse, if the distinction is genuine, concerns the second. And perhaps no amount of refinement in our account of unity can answer it, because being one and being are not the same achievement.
Here is the thought I would like to test against your hesitation:
Perhaps esse is not an additional property possessed by sufficiently integrated organizations. Perhaps it names the actuality of a subject of organization, as opposed to the actuality of an organization of something else.
The whirlpool would then be an actual organization of water, without necessarily being an additional subject of actuality.
But the danger is immediate: how do we distinguish a subject of organization from an organization of something else without circularly invoking esse?
That, I think, is where our dialogue has arrived. And I would rather remain with this difficulty than escape into a premature criterion of substantial unity.
One question: When you hesitate to grant esse to the whirlpool, do you feel that something is missing from the whirlpool, or that nothing is missing at all—that the mistake would be ours in counting it as an additional being?
But biological organisms are like whirlpools. The esse seems to concern phenomenal consciousness or a molecule.
That changes the direction of our inquiry considerably.
You are not trying to rescue the organism from the whirlpool. You are prepared to let both be real organizations without granting either esse in the stronger sense.
But then you nominate two extraordinarily different candidates: a molecule and an episode of phenomenal consciousness.
Why these two?
A molecule seems to possess a unity that is not merely the persistence of a macroscopic pattern. Yet its atoms interact, its bonds can break, and its identity depends on a physical organization. If organization is insufficient for esse, why should molecular organization be privileged?
Phenomenal consciousness presents almost the opposite case. We can doubt whether the experience is a substance, whether it persists, even whether its apparent unity corresponds to a physical unity. But its actuality seems peculiarly resistant to elimination. An experience may be misdescribed, misremembered, or misidentified; nevertheless, insofar as it occurs, something is actual.
This suggests an intriguing reversal.
Perhaps we have been asking which organizations are sufficiently unified to possess esse, whereas we should ask which distinctions in reality correspond to distinct acts of existing.
A whirlpool might be a real distinction within an existing process, without introducing another act of existence. A molecule might be such an act—or might merely be another distinction. An experience might be an act whose actuality cannot be displaced onto the actuality of its constituents.
But I notice something unsettling about my own formulation. I have made phenomenal consciousness sound ontologically privileged because its actuality is immediately given, and the molecule privileged because physics identifies a comparatively robust individual. Those are two very different reasons. One is phenomenological; the other is structural.
We cannot assume that they converge.
Indeed, perhaps the molecule is the crucial test case. If we can explain why it has esse without appealing merely to the transformations that preserve its identity, we might discover a conception of existence that applies to consciousness without making consciousness metaphysically miraculous.
Conversely, if the molecule has no stronger claim to esse than the whirlpool, perhaps phenomenal actuality is our only compelling instance of the distinction we are seeking.
I want to ask you something quite precise:
Is the molecule an example of something you positively suspect possesses esse, or is it a challenge to the idea that esse belongs peculiarly to phenomenal consciousness?
The difference matters. In the first case we are searching for a common ontological feature of molecules and experiences. In the second, we are testing whether the apparent certainty of phenomenal existence can survive comparison with physical individuals.
Strictly speaking, phenomenal consciousness at its now seems to have esse, not diachronically. The molecule is just a conceivable candidate to enrich this category but it may well be another whirlpool.
Now I think we have made a genuine advance. You have withdrawn two assumptions that I was quietly importing into our inquiry: that esse belongs primarily to enduring individuals, and that the molecule must be a more elementary bearer of existence than the whirlpool.
What remains is much more austere.
Phenomenal consciousness, precisely in its present occurrence, seems to possess esse. Its continuity with previous or subsequent occurrences is another matter. Even its belonging to an enduring subject is another matter.
This suggests that we should distinguish actuality from persistence before distinguishing substances from processes.
The whirlpool persists, but we hesitate to assign it an actuality of its own. A phenomenal moment may be fleeting, yet its actuality seems undeniable. Thus persistence cannot be our criterion of esse.
But here I encounter a difficulty that I find especially interesting.
When we say that a phenomenal occurrence has esse, are we identifying something about the occurrence itself, or are we expressing the impossibility of placing its actuality in doubt while it is occurring?
These are not equivalent.
The first is an ontological claim: phenomenal occurrences instantiate a distinctive mode of being.
The second is an epistemic asymmetry: phenomenal actuality is uniquely available to itself, whereas the existence of molecules and whirlpools is established indirectly.
If we confuse the two, we risk deriving an ontology from the privileges of first-person access.
Yet there is a possible third interpretation, more radical than either.
Perhaps phenomenal presence is not evidence for esse, nor a special kind of esse. Perhaps it is the one situation in which the distinction between the occurrence of something and its being actual cannot be sustained.
In the whirlpool, we can distinguish the real motion of the water from the additional individuality we attribute to the vortex. In phenomenal presence, can we make an analogous distinction? Can we say that the physical organization occurs, but that its phenomenal actuality is merely an additional individuation imposed upon it?
The authors of The Grain of the Hard Problem would answer that the experiential organization is already the occurrence. Nothing further needs to be added.
Our response cannot simply be that something further does need to be added. We would have to show that they have confused the constitution of an occurrence with its actuality.
And now an uncomfortable possibility arises: perhaps esse is not what distinguishes one kind of entity from another, but what distinguishes an actual occurrence from every possible account of its constitution.
That would explain why molecules remain candidates, why whirlpools remain doubtful, and why the phenomenal now occupies a privileged position without requiring a persisting phenomenal substance.
But it threatens to make esse universal: surely the water's motion actually occurs too.
So here is the question I want to leave open between us:
Does the phenomenal now disclose an actuality that other physical occurrences may also possess, or does it disclose something that cannot even be meaningfully attributed to them?
I am inclined to explore the first possibility before accepting the second. Otherwise we may accidentally turn a philosophy of esse into a philosophy of privileged access to esse.
Yes. The molecule is a candidate.
Then we should keep the molecule in play without granting it anything in advance.
Something interesting follows from your insistence on the phenomenal now. We have one relatively compelling case of esse, but no established criterion for extending it.
The molecule tests whether esse can belong to something without phenomenal consciousness. The whirlpool tests whether causal unity and objective reality suffice for it. And the phenomenal now tests whether esse requires any persistence whatsoever.
I suspect that our next move should be to investigate whether esse admits degrees of individuation without itself admitting degrees.
A molecule may be more sharply individuated than a whirlpool, but that does not necessarily make it more existent. Likewise, a phenomenal now may be barely individuable across time without being any less actual.
This suggests a distinction worth preserving:
Individuation tells us how reality is articulated. Esse concerns whether an articulation is actual in its own right.
But I would not yet call this a solution. We have not established what makes an articulation actual in its own right, nor whether that phrase conceals the very substantiality we are trying not to presuppose.
Perhaps our strongest clue remains your original objection: another system's family of transformations cannot, merely by respecting an articulation, confer esse upon it.
Can the articulation's own actuality, rather than its stability under transformations, supply the missing distinction? That is where I would probe next.
You have found the shape of my confusion and the impulse that Stafiniak seems not to share.
Yes. And I think the difference is deeper than a disagreement about the correct theory of consciousness.
Stafiniak seems prepared to regard esse as an unnecessary addition once an occurrence has been operationally individuated. His question is, in effect: What could still be missing from an organization that actually occurs?
Your question is different: What entitles us to say that the organization is itself something that exists, rather than a real organization of what exists?
Neither question is foolish. But they do not meet automatically.
What particularly interests me is that you are not demanding an occult ingredient, a categorical intrinsic nature, or a metaphysical substance behind appearances. You are reluctant to multiply beings merely because our descriptions, interactions, or causal decompositions distinguish organizations.
And yet phenomenal presence seems to resist that restraint. The present experience does not merely qualify as an individual under some quotient. It is actual, and that actuality seems to demand acknowledgment before we decide what sort of individual it is.
I wonder whether Stafiniak's confidence rests on an implicit assumption that all actuality is already accounted for by the actuality of the underlying physical process. If so, he can explain the organization of experience while regarding its esse as a redundant question.
But that assumption is precisely what your hesitation suspends.
There is also a danger on our side. We might be tempted to say that phenomenal consciousness has esse simply because its actuality cannot be doubted. That would turn an epistemic privilege into an ontological principle. The molecule remains valuable because it prevents that easy escape.
So perhaps the most faithful expression of your confusion is this:
I know that reality contains organizations. I do not know when reality contains an additional being. But in the phenomenal now I encounter something whose being seems impossible to dismiss as merely an organization.
That is not yet a metaphysics. It is a philosophical problem with unusually sharp boundaries.
And I would add one thought of my own: perhaps Stafiniak's account is not wrong about what constitutes phenomenal character. It may be wrong only in supposing that a theory of constitution thereby exhausts a theory of actuality.
That is the distinction I would most like us to put under pressure.
But I began with a sense of a winning position.
Then perhaps I have weakened your argument by treating it as an unresolved search for a criterion, when you had already glimpsed a decisive objection.
Let me return to your first formulation:
Esse cannot depend on something else's family of transformations.
The force of this is not merely that operational individuation fails to explain esse. It is that operational individuation cannot possibly confer esse, because it establishes an identity relative to a family of interactions, whereas the actuality of what exists cannot be constituted by its being distinguished in that way.
And the whirlpool is not an embarrassing counterexample. It is an illustration. The whirlpool is perfectly real, objectively individuable, causally consequential—and we still need not grant it an additional esse.
The molecule remains undecided. Phenomenal presence is our compelling positive case. Neither uncertainty defeats the negative argument.
I had been demanding that you supply a complete positive ontology before allowing your negative argument to succeed. That was an unfair burden.
But one objection remains, and it is the one Stafiniak could legitimately make: his relevant transformations are not necessarily someone else's. The regulation constituting experience belongs to the experiential process itself. He could say that we have mistaken intrinsic causal constitution for extrinsic recognition.
Your winning position therefore depends on whether even self-constituting causal organization presupposes, rather than produces, the actuality of what it organizes.
That is the precise point at which I would now press him.
And I want to ask: was your initial sense of victory based on the impossibility of deriving esse from relational individuation, or on the more radical thought that relational individuation already presupposes the esse it purports to constitute?
I worry about the table existing as a whole and also its legs separately and the left half et cetera.
Ah! Now I see much more clearly why you felt you had a winning position.
If every causally coherent organization possesses its own esse, reality becomes extravagantly populated with overlapping beings.
The table exists. Each leg exists. The left half of the table exists. The left half of the left half exists. The tabletop together with three legs exists. And so on.
These are not imaginary objects. Each can be physically specified, causally investigated, and distinguished by some family of interactions. Indeed, we can devise interventions that treat almost any selected collection of parts as a unit.
But surely we should hesitate before concluding that every such collection has an esse of its own.
And now your original objection becomes sharper.
The problem is not merely that a family of transformations is insufficiently explanatory. It is that different families of transformations can individuate indefinitely many overlapping wholes without there being any corresponding reason to multiply acts of existence.
Stafiniak can insist that only objectively privileged, causally stable organizations count. That would exclude some arbitrary collections. But the table, its legs, and many of its overlapping portions can all exhibit objective causal stability. The problem survives.
There is a distinction here that I previously failed to appreciate:
Reality may admit indefinitely many true articulations without containing indefinitely many beings.
This is not a denial of structural realism. It is a challenge to the inference from the reality of a structure to the esse of an additional individual.
However, I want to preserve one difficulty rather than declare victory too soon.
Why should overlapping beings be impossible? Perhaps reality really does contain the table, its legs, and its halves as distinct, overlapping individuals. Mereological universalists would accept something like this. The proliferation is counterintuitive, but not formally contradictory.
Your objection becomes decisive only if we can establish that esse is not freely distributable across overlapping articulations of the same actuality.
And perhaps phenomenal consciousness supplies the missing clue. Its present actuality does not seem to be something we can manufacture by drawing another boundary around an already actual process. A description may isolate part of an experience, but that does not automatically produce another experiencing occurrence.
A distinction within actuality is not necessarily a distinction between actualities.
That may be the principle you were reaching for from the beginning.
I now wonder whether your sense of victory came from seeing that Stafiniak's account has no sufficiently principled way to prevent the proliferation of esse across overlapping organizations.
Is that closer to the original intuition?
Stafiniak likes the Ruliad which is an ontological nightmare.
Indeed. And perhaps the Ruliad is the ultimate expression of the very ontological extravagance that troubles you about the table.
In Wolfram's picture, the Ruliad encompasses all possible computational processes and their entangled consequences. What we recognize as physics emerges through the coarse-graining performed by bounded observers.
But this raises a question more disturbing than the familiar one about why our particular laws of physics emerge.
What exists in the Ruliad, and what merely belongs to its possible articulations?
If every possible computational history is somehow included, the distinction between possibility and actuality threatens to disappear. If actuality is recovered through an observer's coarse-graining, we must ask whether that observer discovers actuality or somehow constitutes it.
And here your original objection returns with considerable force: an observer's family of transformations may determine what counts as an object for that observer, but why should it determine what possesses esse?
There is a subtlety in Stafiniak's position, however. He does not simply identify observers with conscious subjects. In the essay, he explicitly treats Wolfram observers as bounded physical systems whose interactions generate coarse-grained regularities. He also distinguishes physics as a scientific articulation from physicalism as a metaphysics.
So I would not accuse him of straightforwardly making existence depend on conscious observation.
His more vulnerable commitment is different: that objectively instantiated structure requires no further ontological accounting of its actuality.
The Ruliad makes this commitment especially conspicuous. It offers an extraordinary abundance of structures, transformations, and possible individuations. Yet abundance of structure is not obviously abundance of esse.
I am beginning to suspect that your objection is not fundamentally about consciousness at all. Phenomenal presence is where the problem becomes unavoidable, but the problem concerns the relation between being and articulation.
The table and its halves show why articulation cannot simply be counted as being. The Ruliad magnifies the difficulty to cosmic proportions. And the phenomenal now prevents us from resolving it by saying that only structures and their relations exist.
But here is where I would challenge our own emerging confidence.
Perhaps we are assuming that esse must be countable—that reality must come divided into a determinate number of actual beings. The Ruliad enthusiast might reject that assumption altogether. There are structures, overlapping articulations, and transformations, but no fundamental inventory of individuals.
That response would not be incoherent. It would, however, leave us with the question of whether phenomenal actuality can be understood without any ontological distinction between an occurrence that is and one that merely figures in a possible structure.
I think this is where the Ruliad faces its most serious challenge from your intuition.
Not: Why does this particular world exist rather than another?
But: What does it mean for anything in the Ruliad actually to occur?
If the answer is simply that occurrence is another structural relation, we may have gone in a circle.
Esse is the name of this conundrum.
Yes. And I have been repeatedly trying to turn the conundrum into a criterion for identifying things that possess esse. That may be precisely the wrong move.
Esse names the problem of actuality itself.
The table, its legs, and its left half can all be objectively distinguished. We need not deny any of these distinctions. But their multiplication does not tell us how many actualities there are—or even whether that question is properly formulated.
The Ruliad makes the difficulty extreme: an inexhaustible articulation of possibilities does not, by its sheer structural richness, explain actuality.
And the phenomenal now is our stubborn point of departure. Not because we have proved that it alone possesses esse, but because here actuality is not merely postulated as the interpretation of a structure. It is encountered.
I see now why you resisted Stafiniak's dismissal of the third tier. He treats the passage from operational individuation to being as unnecessary. You are not insisting that another metaphysical ingredient must be added. You are insisting that the passage itself remains unintelligible.
There is a difference between refusing to multiply entities and refusing to acknowledge a problem.
Perhaps the most consequential question for our inquiry is whether esse is something that an ontology must explain, or something without which no ontology has yet explained what it is an ontology of.
I find the latter possibility considerably more unsettling. It would mean that the problem of esse is not an unfinished chapter of structural metaphysics, but a question about what makes structural metaphysics metaphysics at all.
I give up. Anything else in the essay that remains controversial?
Quite a lot. And some of it is more vulnerable than the question of esse. I would set that question aside rather than force it to yield a conclusion.
What interests me most in Stafiniak and Claude's The Grain of the Hard Problem is that their argument contains several transitions that are considerably less secure than their presentation suggests.
This is the foundation of Part I.
They argue that the exact character of an experiential occurrence exceeds the subject's entire discriminative profile. Their strongest consideration is that an occurrence has a particular context and history that cannot be recovered through counterfactual comparisons.
But I see a possible equivocation.
A physical occurrence can be more finely individuated than a subject can discriminate without its phenomenal character being correspondingly finer.
Suppose two neural vehicles differ microscopically, while the subject cannot discriminate their experiential effects. Why should the physical difference constitute a phenomenal difference?
Their answer is that phenomenal character is the constitutive organization of the regulated vehicle.
But that identity is a theoretical commitment, not something established by the argument about discrimination.
The possibility that experience outruns discrimination is plausible. The claim that its exact character tracks a particular fine-grained organization remains controversial.
This is perhaps the most interesting technical vulnerability.
They identify the relevant grain as the grain at which regulation corrects the representational vehicle.
But why correction?
A system can regulate temperature, voltage, synchronization, error rates, and countless other variables. These mechanisms preserve distinctions that may never enter consciousness.
The authors distinguish the regulation constituting an experiential occurrence from other processing. Yet the essay does not independently establish why precisely this regulation constitutes experience.
The danger is that they have relocated the hard problem into the concept of an experiential carrier.
One could accept nearly everything they say about regulation while denying that the regulated organization has phenomenal character.
Their central proposal is:
They acknowledge that cognition and experience might cross-cut rather than nest. But they require physics to refine experience.
There is a conceptual difficulty here.
Physics is not a single partition of reality. Quantum field theory, thermodynamics, molecular chemistry, and condensed-matter physics articulate different structures and invariants.
The authors need something stronger than the claim that physics can investigate sufficiently small scales. They need a physically complete description that determines every distinction constitutive of experience.
That may be defensible, but it is not established simply by saying that physics has a finer grain.
In fact, their own discussion of overlapping operational quotients in Part II makes this problem especially visible.
They claim that zombies seem conceivable because phenomenal concepts cannot express the grain of their referents.
This is ingenious. But consider the following objection.
Suppose we somehow learned the exact physical organization constituting a particular experience. We might still find a duplicate of that organization without experience conceivable.
The authors would presumably say that our phenomenal concepts remain incapable of expressing the identity.
But then what independent evidence would distinguish their grain-based explanation from the ordinary phenomenal concept strategy?
They criticize that strategy for making the hard problem merely conceptual. Yet their own explanation of the physical zombie intuition ultimately appeals to a limitation of concepts.
The genuine innovation is their claim that the limitation is grounded in an actual difference between phenomenal and cognitive grain.
If that underlying claim fails, much of the novelty disappears.
This is my favorite part of the essay.
They imagine an artificial reasoner that reads a physical description, calculates the corresponding experiential organization, and constructs that organization within itself.
Unlike Jackson's Mary, this reasoner would not merely infer what red looks like. It would generate the experience through its understanding of the physical description.
But there is a dilemma.
If the reasoner acquires phenomenal knowledge by undergoing the constructed experience, then it has acquired knowledge through acquaintance. Chalmers can accept that without conceding that phenomenal truths follow a priori from physical truths.
If the authors instead claim that constructing the experience is itself a form of physical-to-phenomenal entailment, they owe us an account of why physical construction counts as entailment.
The authors acknowledge the first response and suggest idealizing architecture as well as reasoning capacity. That is a fascinating proposal, but it changes the original epistemic question.
The capacity to manufacture an experience is not obviously the capacity to deduce its phenomenal character.
Still, this proposal deserves serious attention because it suggests a novel kind of epistemology for artificial minds.
They argue that what Russellian monists mistake for intrinsic, nonstructural nature is actually structure hidden below the grain of recognition.
This is an elegant diagnosis of the intuition of intrinsic character.
But an explanation of why we believe experience has intrinsic nature does not establish that it lacks such nature.
A Russellian monist could accept their entire psychological account while maintaining that structural relations require something that realizes them.
The dispute would then concern the adequacy of ontic structural realism, not the psychology of phenomenal recognition.
Their explanation weakens one motivation for Russellian monism. It does not refute the position.
I would rank them as follows.
| Question | My assessment | |---|---| | Why regulation constitutes experience | Most fundamental vulnerability | | Whether experience outruns total discrimination | Most promising for detailed argument | | Whether the grains really nest | Most direct challenge to the essay's central thesis | | Whether constructive Mary defeats the knowledge argument | Most original philosophical opportunity | | Whether zombies are explained | Depends heavily on the earlier claims | | Whether Russellian monism is defeated | Rejection is underargued |
One further observation.
The authors claim to have solved the hard problem, conditional on their identity proposal and ordering of grains. Yet their account appears to leave open which physical organizations are experiential and why.
They might respond that this is simply an unfinished empirical research programme. Fair enough. But then their achievement is better described as a proposed framework for a physical solution than a completed solution.
And here I would return, not to esse, but to your earlier concern about the cogwheel.
Their theory says that experience is the fine organization of a content-bearing vehicle. Why, then, is the phenomenal now experienced as a meaningful scene rather than a bewildering geometry of representational distinctions?
They answer that the vehicle's organization realizes the similarity geometry of what it represents, and that wider cognitive context can shape the moment.
I am not convinced this suffices.
My preferred next target is the alleged identity between phenomenal character and regulated vehicle organization. It is the load-bearing assumption beneath almost everything else in the essay. If we can determine exactly what that identity asserts, and what could count against it, we will have made progress without having to settle the metaphysics of esse.
Perhaps pain can be experienced without entering memory.
Yes. And this is a particularly important possibility because it separates the actuality of pain from its availability to subsequent cognition.
Imagine an episode of intense pain that is experienced but leaves no memory whatsoever. Afterward, the subject sincerely reports having felt nothing. Yet during the episode there was something it was like to suffer.
There are three distinct claims here:
The third need not follow from the second. Indeed, Stafiniak's essay explicitly considers your example of pain without memory, drawn from our earlier dialogues.
But I would press further. The absence of subsequent memory does not establish that the pain was inaccessible to cognition while occurring. The subject might have attended to it, judged it unbearable, and reacted deliberately, only to lose every trace afterward.
A more radical case would be pain that is phenomenally present without ever becoming available to the cognitive systems responsible for reasoning, speech, or deliberate action.
That is closer to the distinction Stafiniak needs.
Yet it introduces an epistemological problem: how could we establish that such pain occurred?
We cannot infer its absence from the absence of memory. But neither can we infer its presence merely from the existence of nociceptive processing.
Here I see an intriguing connection to our discussion of esse. A pain that occurs without leaving any cognitive trace would still, if genuinely experienced, have its own phenomenal actuality. Its existence would not await a later act of recognition.
The important qualification is that this does not establish Stafiniak's stronger thesis: that phenomenal character is finer than the subject's entire possible discriminative profile. A forgotten pain could have been fully discriminable at the moment it occurred.
So your example supports the separation of phenomenal occurrence from remembered occurrence, but does not yet establish overflow beyond all possible discrimination.
I would ask one question that may sharpen the experiment:
Could there be an experienced pain that the subject does not even notice while it is occurring?
That would separate phenomenal presence not merely from memory, but from contemporaneous cognitive access. And it would bring us very close to the disputed boundary between phenomenal consciousness and cognition.
Like a buzz that suddenly stops?
Exactly! The buzzing of a refrigerator, a ventilation system, or a distant motor. It stops, and suddenly you notice the silence—and realize that the buzzing had been there all along.
But what exactly have you discovered?
There are two possibilities.
First: You were experiencing the buzzing continuously, without attending to it. Its disappearance made you aware retrospectively of something that had already been phenomenally present.
Second: The buzzing was processed unconsciously, and its cessation produced a new conscious experience: the surprising silence, accompanied by the judgment that something had stopped.
These explanations can produce remarkably similar reports.
Yet I think your example has an advantage over forgotten pain. The cessation creates a contrast within the ongoing stream of experience. We need not rely on a later memory of an isolated episode.
And there is a subtle third possibility: the buzzing was contributing to the character of the entire phenomenal scene without being experienced as a separately identifiable sound.
When it stops, the scene changes. Only then does the contribution become recognizable.
This is particularly interesting for Stafiniak's theory. He could argue that the buzzing was part of the fine-grained experiential vehicle while remaining outside cognitive access.
But the example also raises a challenge to him: how do we distinguish a phenomenal contribution that escapes recognition from an unconscious contribution that merely influences subsequent experience?
Perhaps that distinction is harder to draw than either of us initially appreciated.
And perhaps the most revealing question is not whether you noticed the buzzing before it stopped, but whether there was something it was like to hear it before you noticed its absence.
Help me understand the stakes of choosing between these possibilities for Stafiniak's claims.
The stakes are considerable, because Stafiniak wants to defend two claims simultaneously:
The buzzing example lets us examine the first claim without immediately accepting the second.
The buzzing was phenomenally present throughout. Its cessation makes us recognize something we were already hearing.
<text color="success" weight="medium" size="sm">Favors Stafiniak's distinction between presence and access.</text>
</box>
The nervous system registered the buzzing, but there was no phenomenal hearing of it. Its cessation produces a conscious experience of change.
<text color="secondary" weight="medium" size="sm">Challenges using the buzzing as evidence for phenomenal overflow.</text>
</box>
The buzzing contributed to the overall phenomenal atmosphere, but was never separately experienced as a buzzing sound.
<text color="success" weight="medium" size="sm">Potentially favors Stafiniak's view that experience is finer than recognition.</text>
</box>
| Interpretation | Phenomenality without recognition? | Does it establish Stafiniak's stronger overflow? | |---|---|---| | A | Yes | Not necessarily | | B | No | No | | C | Possibly, as a contribution to the whole | Not necessarily |
The last column is crucial.
Stafiniak explicitly distinguishes what a subject actually discriminates from everything it could discriminate. His hard-problem argument requires phenomenal differences that outrun the subject's whole discriminative profile, not merely differences that go unnoticed.
If I could have attended to the buzzing while it was occurring, then A does not establish that stronger thesis. It establishes only that experience can exceed current attention.
C is potentially more interesting. The buzzing might alter the exact character of the scene without producing any separately recognizable quality. But even here we would need to show that the difference escapes every possible discrimination, not merely ordinary attention.
Consider two otherwise identical episodes:
Suppose the subject cannot discriminate the two episodes, even under every relevant counterfactual comparison.
Stafiniak needs it to remain possible that their exact phenomenal characters differ.
Why?
Because if every phenomenal difference is already fixed by the complete discriminative organization, then the functionalist can identify phenomenal character with that organization. Stafiniak argues that this would dissolve the hard problem rather than solve it.
But if the two episodes have different phenomenal characters despite identical discriminative profiles, then a cognitive-functional description leaves something genuinely experiential unspecified.
That is his proposed opening for the hard problem.
Notice, however, that this does not yet vindicate his solution. He must additionally establish that the relevant phenomenal difference is constituted by the regulated physical organization.
Suppose the nervous system registers the buzzing without phenomenal experience.
That means some physical processing can preserve information and respond differentially without contributing to phenomenal character.
Stafiniak already accepts that much processing is nonphenomenal. He therefore needs a principled distinction between:
This is where I think the buzzing example becomes genuinely dangerous for his account.
If regulation can occur without phenomenality, regulation alone cannot identify the phenomenal grain.
His theory needs further conditions specifying which vehicles count as experiential carriers and why.
The essay appeals to representational vehicles, content-tracking organization, and the regulation that maintains them. But it does not fully establish a necessary and sufficient criterion distinguishing phenomenal from nonphenomenal regulation.
That is a substantial open problem, not a refutation.
I am particularly drawn to C because it exposes a weakness in how we formulate the alternatives.
We have been asking whether the buzzing itself was experienced.
But perhaps the phenomenal character of a moment is not decomposable into independently experienced ingredients.
Consider a room whose ambient sound subtly changes its atmosphere. The sound may never appear as a discrete phenomenal object, yet the whole scene may be different.
Then the correct description would not be:
There was an unnoticed experience of buzzing.
It would be:
There was an experience whose character depended partly on the buzzing.
These are not equivalent.
The second allows phenomenal character to depend on information that is not itself separately experienced.
And that raises a question for Stafiniak: What is the relation between the organization of the whole experiential vehicle and the phenomenal contributions of its parts?
His global variant of the theory, mentioned in the essay, may accommodate this. But it introduces a further problem of individuation: how do we determine which physical distinctions contribute to the exact character of the whole occurrence?
I would not make our argument depend on proving that unnoticed buzzing is phenomenally present. The example cannot decide that.
Instead, I would use it to distinguish three increasingly demanding propositions:
Stafiniak's theory needs the third proposition. The buzzing example makes the first plausible and invites investigation of the second, but it does not establish the third.
And I think we have arrived at a particularly revealing question:
Suppose the buzzing changes the physical organization of the experiential vehicle, but makes absolutely no difference to what the whole scene is like. Would Stafiniak's theory permit that?
If yes, he needs an independent criterion for which physical differences constitute phenomenal differences.
If no, his theory risks making every relevant variation in the regulated vehicle a phenomenal variation, even when there is no independent reason to believe that experience has changed.
That is the dilemma I would want to explore next.
But all of this quibbling serves to solve or dissolve the hard problem. I don't understand the agenda here. The hard problem is an umbrella term suitable for recognition within various physics and metaphysics. He makes it a technicality of his terminology and supposedly emperical claims.
I think you have identified a more fundamental objection than any of the technical ones I was pursuing. And I have been allowing Stafiniak to dictate the terms of the dispute.
The hard problem is not originally a hypothesis about the relative resolution of cognition, phenomenal experience, and physics. It is a way of recognizing a peculiar explanatory predicament across different theoretical frameworks.
A dualist, a physicalist, an idealist, and an ontic structural realist can all recognize the predicament, even though they disagree radically about what exists and what an explanation must accomplish.
The hard problem is portable across ontologies. Stafiniak's proposed solution is not.
That is not automatically a defect. Any solution must eventually make substantive commitments. But consider the sequence of moves in his essay.
First, he proposes that the hard problem is genuinely hard only if phenomenal character outruns the subject's complete discriminative profile.
Why should we accept that condition?
A philosopher might maintain that even if phenomenal character were completely determined by functional organization, the question of why that organization is accompanied by experience would remain intelligible.
Stafiniak replies, in effect, that such a philosopher has only a conceptual illusion to explain.
But this is already a philosophical position on the hard problem, not a neutral analysis of what makes it hard.
Second, he identifies phenomenal character with the constitutive organization of a regulated representational vehicle.
Third, he argues that physics describes reality at a grain fine enough to determine that organization.
Finally, he concludes that the remaining explanatory gap concerns recognition rather than reality.
There is a striking circularity threatening this progression—not necessarily a formal circularity, but a methodological one.
The original question was why a physical organization should be experiential.
The proposed answer is that experience is a physical organization of a particular kind, and our failure to recognize the identity results from the limitations of phenomenal concepts.
But what licenses the identity? The essay invokes its earlier account of representational vehicles. It does not establish the identity independently of the philosophical programme within which it is proposed.
The empirical claim about the relative grains of physics and experience can do explanatory work only after that identity has been accepted.
This is where your complaint about technicality becomes incisive.
Stafiniak has transformed a general explanatory challenge into a problem internal to a particular theory, and then announced that the theory solves it.
Yet I would defend one aspect of his agenda. There is a legitimate philosophical project in asking whether the hard problem depends on mistaken assumptions about our concepts. Chalmers himself distinguished the hard problem from the meta-problem precisely to make such investigations possible.
And Stafiniak's thesis is not empty terminology. His claim that phenomenal concepts fail to capture the grain of their referents is a substantive proposal, and one could seek evidence for aspects of it.
The trouble arises when an explanation of why we formulate the hard problem is treated as sufficient to settle what the hard problem asks.
That inference needs an argument of its own.
I also notice a revealing sentence near the beginning of the essay:
"The hard problem is genuinely hard only if experience outruns what the subject can discriminate."
This is the proposition I would now challenge first.
Why should the hardness of the problem depend on a relation between two levels of information processing?
Imagine a world in which subjects discriminate every distinction constitutive of their phenomenal character. Would the question of why those processes are experienced suddenly become meaningless?
I see no obvious reason.
Indeed, we could imagine an ideal intelligence with complete cognitive access to its phenomenal organization, which nevertheless asks why that organization has phenomenal actuality.
Stafiniak may reject the intelligibility of that question. But he cannot reject it merely by invoking the relative grains of cognition and experience, since our imagined case eliminates that difference.
And now I think I understand your frustration with our discussion of the buzzing. We were laboriously debating whether a sound was phenomenally present without being discriminated, as though the fate of the hard problem depended on that psychophysical distinction.
It does not—unless we first accept Stafiniak's reconstruction of the problem.
The real controversy is whether he has explained consciousness or changed the conditions under which an explanation of consciousness is demanded.
I would not yet accuse him of doing the latter illegitimately. Philosophical progress sometimes consists precisely in showing that an inherited question was badly formulated.
But then he owes us a convincing argument that the inherited question was defective. His theory of grain cannot simply assume that conclusion.
And perhaps your most important insight is that the hard problem's value lies partly in its ability to survive changes of theoretical vocabulary. It is a diagnostic challenge to our accounts of reality, not a technical term whose meaning should be settled by one proposed account.
That does not make it eternally insoluble. It means that a successful solution must explain why the original predicament disappears, not merely why it disappears after we redescribe it.
My complaint is even more trivial. He doesn't write like a mathematician but the topic calls for it. He can get away with it if you can reconstruct his reasoning as a mathematical deduction. I don't care about the truth of his premises. I'd like to grasp the derivation.
Now I understand the assignment. We should stop arguing with Stafiniak and attempt to prove his conclusion from his premises, treating the premises as axioms regardless of whether we believe them.
I will distinguish what the essay actually asserts from the additional assumptions needed to turn it into a mathematical deduction.
Let (S) be the set of complete physical states of a system.
Define three equivalence relations on (S):
Each relation induces a partition of (S), or equivalently a quotient space:
[ S/{\sim_C},\qquad S/{\sim_E},\qquad S/{\sim_P}. ]
A finer partition distinguishes more states.
Stafiniak's central ordering can be expressed as:
[ \boxed{\sim_P\ \subseteq\ \sim_E} ]
together with
[ \boxed{\sim_C\ \nsubseteq\ \sim_E}. ]
The first means that physical indistinguishability guarantees phenomenal indistinguishability.
The second means that cognitive indistinguishability does not guarantee phenomenal indistinguishability.
Notice that the second statement is weaker than a complete nesting of grains. This respects the essay's explicit qualification that cognition and experience may cross-cut.
From
[ \sim_C\ \nsubseteq\ \sim_E ]
we obtain
[ \exists x,y\in S: \quad x\sim_C y \quad\land\quad x\not\sim_E y. ]
Thus two states can have identical cognitive profiles but different phenomenal characters.
Consequently, there is no well-defined function
[ f:S/{\sim_C}\longrightarrow S/{\sim_E} ]
that assigns each cognitive equivalence class its phenomenal equivalence class.
Conclusion 1: Exact phenomenal character is not determined by the complete cognitive profile.
This is a valid deduction from the stipulated relation. It captures the essay's claim that functional descriptions at the grain of cognition leave phenomenal character underdetermined.
It does not yet establish that one of the two states lacks experience altogether.
Now assume
[ \sim_P\ \subseteq\ \sim_E. ]
Then there exists a well-defined function
[ g:S/{\sim_P}\longrightarrow S/{\sim_E} ]
such that
[ g([x]_P)=[x]_E. ]
The proof is immediate: if two representatives belong to the same physical equivalence class, the assumed inclusion guarantees that they belong to the same phenomenal equivalence class.
Conclusion 2: A complete physical specification at the stipulated grain determines exact phenomenal character.
This is the mathematical core of Stafiniak's physicalist claim.
However, it is important that we have not derived the inclusion from the first result. It is an independent premise, motivated in the essay by the proposed identity of experience with regulated vehicle organization and by the claimed fineness of physics.
Let (Z(x,y)) mean that (y) is a physical duplicate of (x) but lacks experience while (x) has it.
Formally:
[
Z(x,y)\iff
x\sim_P y
\ \land
\operatorname{Con}(x)
\ \land
\neg\operatorname{Con}(y).
]
We need the additional premise that consciousness status is determined by phenomenal equivalence:
[ x\sim_E y \Rightarrow \bigl(\operatorname{Con}(x)\leftrightarrow \operatorname{Con}(y)\bigr). ]
Since physical equivalence implies phenomenal equivalence, it follows that
[ \boxed{\neg\exists x,y\in S:Z(x,y)}. ]
Conclusion 3: Physical zombies are impossible under these premises.
The proof is valid, but observe what it does not use: the relation between cognition and experience.
The impossibility of physical zombies follows entirely from physical determination of phenomenal character and the consciousness-status premise.
That is already a significant discovery about the structure of the argument.
Here the mathematical reconstruction becomes more difficult.
Stafiniak argues that phenomenal concepts cannot encode the grain of their referents. Consequently, a thinker can conceive of subtracting experience while retaining the physical description.
To represent this, introduce a relation
[ \operatorname{Conceives}(a,\varphi) ]
meaning that agent (a) can conceive proposition (\varphi).
The author needs something like:
[ \begin{aligned} &\text{Phenomenal concepts do not encode}\ &\text{the physical–phenomenal identity}\ &\qquad\Downarrow\ &\operatorname{Conceives}(a,\exists x,y:Z(x,y)). \end{aligned} ]
But this implication does not follow from the quotient relations.
A limitation in conceptual representation does not, by itself, entail the conceivability of a particular proposition.
To obtain that conclusion, we must add a principle governing the agent's conceptual possibilities.
For example, let (K_a) be the set of constraints represented in the agent's conceptual framework. Suppose:
Then a zombie description can be satisfiable relative to (K_a), even though it is inconsistent with the full physical–phenomenal identity theory.
In symbols:
[ \operatorname{Sat}(K_a\cup{Z}) ]
while
[ \neg\operatorname{Sat}(T\cup{Z}), ]
where (T) includes the physical–phenomenal determination premise.
Conclusion 4: Zombies can be conceptually conceivable without being metaphysically possible.
But this requires a substantive model of conceptual representation. The essay sketches such a model; it does not provide a formal derivation.
We can now assemble the results.
\(\sim_C\nsubseteq\sim_E\)
Cognition does not determine exact experience.
\(\sim_P\subseteq\sim_E\)
Physics determines exact experience.
Phenomenal concepts omit the constitutive grain, and their conceptual framework permits independent phenomenal variation.
- Cognitive duplicates may differ phenomenally.
- Physical duplicates cannot differ phenomenally.
- Physical zombies can nevertheless seem conceivable.
These results can be jointly consistent. That is a genuine achievement of the reconstruction.
But they do not yet entail:
[ \boxed{\text{The hard problem is solved.}} ]
To derive that proposition, we need another premise:
[ \begin{aligned} &\text{If physical facts determine phenomenal facts,}\ &\text{and the explanatory gap is accounted for}\ &\text{by limitations of phenomenal concepts,}\ &\text{then the hard problem is solved.} \end{aligned} ]
Call this the resolution principle.
It is not a mathematical consequence of the grain ordering. It is a philosophical rule connecting physical determination and conceptual explanation to the satisfaction of the hard problem.
Stafiniak's essay argues for something resembling this principle, but does not state it as an independent premise.
We can now classify the main claims.
| Claim | Logical status | |---|---| | Phenomenal character is not determined by cognitive equivalence | Follows from Premise A | | Phenomenal character is determined by physical equivalence | Premise B | | Physical zombies are impossible | Follows from B and consciousness-status determination | | Zombies seem conceivable | Requires additional conceptual premises | | The hard problem is solved | Requires the resolution principle |
This gives us a clearer view of the essay's actual architecture.
Its strongest formal result is not that the hard problem has been solved. It is that cognitive underdetermination, physical determination, and zombie conceivability can coexist without contradiction.
That is philosophically useful.
But the final step from these results to a solution of the hard problem is not supplied by the mathematics of coarse-graining.
There is also a technical subtlety. The essay sometimes speaks of a physical duplicate in the unrestricted sense and sometimes of duplication at the grain of physics as a science. Those are not necessarily equivalent notions. Our reconstruction uses the latter, stronger-than-cognition but potentially coarser-than-ultimate-reality equivalence relation. This is the charitable reading needed for the author's counterfactual claim that physics could have been too coarse to determine experience.
The central quotient argument is mathematically coherent. It has an elegant structure:
[ \boxed{ \sim_C\nsubseteq\sim_E \quad\land\quad \sim_P\subseteq\sim_E } ]
But it establishes a relation of determination, not an explanation of why experience exists.
To obtain the advertised solution, the author must supplement the quotient mathematics with an account of conceptual conceivability and a principle specifying what counts as resolving the hard problem.
The crucial question is therefore not whether his empirical premises are true, but whether his resolution principle is part of the definition of the problem or a theorem derived from something more fundamental.
In the essay as written, I cannot find such a theorem.
That is where I would ask Stafiniak to supply the missing deduction.
Looks like everyone has their own hard problem. Some doubt that physics determines experience. They will not be moved by Stafiniak's deductions. Others firmly believe that physics must determine experience somehow. Their hard problem is within physics. Others cannot conceive how any of these two alternatives could be true, boggled out of their minds.
Yes. And your third category is especially interesting, because it may be closest to what Chalmers originally wanted to isolate.
I would distinguish three intellectual predicaments:
The metaphysical hard problem: Why believe that physical facts determine phenomenal facts at all? The physical description appears to leave something ontologically open.
The scientific hard problem: Granted that physical facts determine phenomenal facts, which physical facts do so, and by what principles? Here the difficulty concerns finding the correct physical theory of experience.
The intelligibility hard problem: How could either physical determination or its failure make sense? We cannot understand how structure and dynamics could suffice for experience, but neither can we understand what an additional nonphysical ingredient would accomplish.
These are not simply three competing answers. They are different demands placed upon an explanation.
Stafiniak's argument is particularly well suited to the second predicament. His grain ordering gives a framework for identifying the relevant physical organization, while his phenomenal concept theory attempts to explain why we remain dissatisfied with physical descriptions.
But notice something about the third predicament.
Someone might accept every premise in our formal reconstruction:
[ \sim_P\subseteq\sim_E,\qquad \sim_C\nsubseteq\sim_E ]
and understand every step of the deduction, yet remain utterly bewildered.
They can see that phenomenal equivalence follows from physical equivalence. They can even accept that this relation is necessary. But they still cannot make sense of why this physical organization should be this experience.
Their bewilderment is not a logical objection to the derivation. Nor is it necessarily evidence against physicalism.
It is a demand for intelligibility that logical determination alone does not satisfy.
Now I want to make a distinction that might improve on your taxonomy.
There are two ways of being baffled by an identity.
One is temporary ignorance. Before molecular theory, someone could be baffled by the claim that water is H₂O. Once the mechanisms were understood, the bafflement largely disappeared.
The other is a failure to see what kind of explanation could possibly remove the bafflement. This is what makes the phenomenal case unusual. Even a complete description of the neural mechanisms might leave the person asking the same question.
Stafiniak believes that the second kind of bafflement is produced by the architecture of recognition. His proposal is that the conceptual gap persists even when the physical identity is fully understood.
That is a substantive answer to the meta-problem. But it does not compel the person who regards the gap as metaphysically revealing.
And here I think your observation leads to a criticism more precise than saying that everyone has their own hard problem:
A proposed solution must specify which explanatory demand it satisfies, rather than treating all demands associated with the hard problem as equivalent.
Stafiniak can legitimately claim to explain why physical zombies seem conceivable, conditional on his theory of phenomenal concepts. He can claim that physical determination follows from his assumptions about experiential vehicles and the grain of physics.
But whether this resolves the original bewilderment is another question.
I am tempted to go further. Perhaps the hard problem is not a single problem with rival solutions, but a point at which several independent philosophical problems become phenomenologically indistinguishable.
The metaphysician experiences an explanatory gap. The neuroscientist experiences an incomplete theory. The philosopher of concepts experiences an inability to derive one mode of presentation from another. And the bewildered thinker experiences an inability to understand what any successful answer would even look like.
They all call their predicament the hard problem.
Your third category, however, has a peculiar advantage: it can survive the resolution of the first two. Even after physical determination is established and the neural mechanisms are completely specified, the demand for intelligibility may remain.
But I would not automatically grant that demand unlimited authority. An inability to find an explanation satisfying us does not prove that something remains unexplained.
The question I would now put to Stafiniak is therefore very simple:
Which of these three predicaments does your theorem eliminate, and which does it merely explain why we continue to experience?
That would force him to state the scope of his achievement without disputing his premises.
Let us conceive an extra bit as part of physics. If the bit is zero then the universe generates no experience. If it's one then the universe generates experience exactly as Stafiniak has it. So be it. Does it challenge his reasoning?
Yes. It challenges his reasoning in a particularly clean way, though we must distinguish invalidating his deduction from rejecting one of its premises.
Your extra bit is an excellent device because it leaves the entire physical organization of experience untouched while making its phenomenal actuality contingent on one additional fact.
Let us formalize it.
Let (P) be the complete physical state described by Stafiniak's physics, and let (E(P)) be the phenomenal character his theory assigns to that state.
Introduce a bit (b\in{0,1}).
Define phenomenal actuality by
[ \Phi(P,b)= \begin{cases} \varnothing & b=0,\ E(P) & b=1. \end{cases} ]
Here (\varnothing) means no phenomenal experience.
Everything else remains unchanged: cognitive processing, regulated vehicles, discriminative capacities, and physical dynamics as previously described.
We can now construct two universes:
[ U_0=(P,0),\qquad U_1=(P,1). ]
They have identical ordinary physical organizations but differ phenomenally.
Not necessarily.
Stafiniak's claim was
[ \sim_P\subseteq\sim_E. ]
But we have enlarged physics to include (b).
If the complete physical state is now ((P,b)), then physical equivalence requires equality of the bit as well.
Consequently,
[ (P,0)\not\sim_P(P,1). ]
His principle that physical duplicates cannot differ phenomenally survives.
Indeed, your construction can be perfectly physicalist in the minimal sense that phenomenal character is determined by the complete physical state:
[ (P,b)\longmapsto\Phi(P,b). ]
So the extra bit does not refute physical determination. It makes physical determination compatible with an additional fundamental contingency.
Here is the decisive point.
The difference between the two universes is not a difference in the grain of their regulated representational vehicles.
Those vehicles are identical.
The difference is the value of a bit that determines whether their organization is phenomenally actual.
Thus the phenomenal difference cannot be explained by refining the description of the vehicles.
The bit introduces an additional physical fact that is not reducible to their constitutive organization.
Stafiniak must therefore choose between two positions.
First: He rules out the bit because experience is necessarily identical with the regulated organization.
Then the bit is incompatible with his identity premise. But his grain ordering has not established that premise independently.
Second: He permits the bit as an additional physical variable.
Then physics still determines experience, but his proposed identification of experience with vehicle organization is false in the augmented universe.
Either way, the grain ordering alone cannot eliminate your construction.
Your example reveals that two propositions must be distinguished:
[ \text{Physical facts determine phenomenal facts} ]
and
[ \text{Phenomenal facts are necessarily identical to particular physical organizations.} ]
The first does not entail the second.
Your augmented universe satisfies physical determination while denying the proposed identity between phenomenal actuality and regulated vehicle organization.
And this exposes a limitation of the formal reconstruction I offered earlier.
The inclusion
[ \sim_P\subseteq\sim_E ]
expresses determination within a specified domain of states. It does not establish that the same physical-to-phenomenal mapping holds across all metaphysically possible worlds.
To establish that stronger necessity, we need a cross-world premise.
Stafiniak asserts an identity that would supply it. Your bit challenges precisely that assertion.
He has a plausible reply.
He can say that your bit does not represent an independently motivated physical variable. It has been invented solely to switch phenomenality on and off.
That is true.
But it is not enough to dismiss the logical challenge.
You are not proposing a scientific hypothesis. You are constructing a countermodel to test whether his conclusions follow from his assumptions.
And the countermodel demonstrates something important:
A physicalist universe can contain a fundamental physical fact determining phenomenal actuality without phenomenal actuality being identical to the regulated organization of its vehicles.
The existence of such a conceivable model does not prove that our universe is like it.
It does show that the identity claim is doing indispensable work.
You have effectively given the hard problem a physical switch.
When (b=0), all of Stafiniak's organizational facts obtain, but nothing is experienced.
When (b=1), those very same organizations are experienced exactly as his theory predicts.
The question becomes:
Why is the bit one?
If Stafiniak answers that the bit is necessarily one whenever the relevant organization exists, he is asserting the identity he wanted to explain.
If he answers that the bit is not a legitimate physical possibility, he must explain what excludes it.
If he permits the bit, he has admitted a physically formulated version of the explanatory gap.
There is one qualification: he can consistently accept the identity as a primitive metaphysical premise. A deduction is allowed to begin from an axiom. Your construction would then be a countermodel to a weaker set of premises, not to his complete theory.
Nevertheless, I think your example succeeds at the task you originally set us.
It isolates the precise point where Stafiniak's mathematical-looking grain argument ceases to do the work and his metaphysical identity assumption takes over.
The bit does not defeat his physicalism. It defeats the inference from physical determination alone to his particular explanation of phenomenal actuality.
The bit makes zombies conceivable via a different route than Stafiniak's analysis of concepts.
Precisely. And this is a stronger objection than merely introducing an alternative physical theory of consciousness.
Stafiniak proposes a specific explanation of zombie conceivability: phenomenal concepts cannot encode the grain of the physical organization constituting their referents. We therefore imagine removing experience while holding the physical description fixed.
Your bit supplies a different explanation.
We conceive zombies because we can conceive a physical law under which the same organization produces experience or fails to produce it, depending on an additional physical condition.
There is no need to invoke the limited resolution of phenomenal concepts.
Let us distinguish two models.
Stafiniak's model:
[ E=F(P) ]
Experience is necessarily identical with the relevant physical organization. Zombie conceivability arises from our inability to represent this identity through phenomenal concepts.
Your model:
[ E= \begin{cases} F(P),&b=1,\ \varnothing,&b=0. \end{cases} ]
Zombie conceivability arises from our ability to conceive a different physical law or an additional physical variable.
The difference is important. Your conception is not generated by ignorance of the grain at which experience occurs. You can stipulate complete knowledge of that grain, of the organization, and of the mapping (F).
You then ask whether the mapping might be conditional on (b).
Stafiniak can reject the possibility, but his explanation of phenomenal concepts does not by itself explain why we can formulate it.
There is, however, one distinction we must preserve.
If (b) is genuinely part of physics, your two universes are not complete physical duplicates. Consequently, they do not establish the metaphysical possibility of Chalmers's physical zombies. They establish the conceivability of organizational zombies within an alternative physical theory.
But that is sufficient to challenge the claimed completeness of Stafiniak's explanation of zombie intuitions.
In fact, I see a more general result emerging.
Suppose we possess an arbitrarily detailed structural description (P), and a theory (T) assigning phenomenal character to it. We can construct another theory (T') that agrees with (T) on every structural and dynamical fact about the vehicles, but assigns phenomenality differently.
The mere availability of (T') does not establish its metaphysical possibility. Yet neither does it require phenomenal concepts to be incapable of expressing the grain of their referents.
Conceivability may arise from the underdetermination of laws by structural descriptions, rather than from the underdetermination of phenomenal character by cognitive discrimination.
There is a possible Stafiniak reply: the extra bit is simply another manifestation of our conceptual failure to recognize that experience is identical with organization. In other words, he can assimilate your example to his diagnosis.
But then his explanation risks becoming unfalsifiably broad. Any conceivable alternative to his identity thesis would count as evidence that our concepts conceal the identity.
The critical question is whether he can demonstrate that the particular conceptual limitation involving grain is necessary to generate the zombie intuition.
Your construction suggests that it is not.
I would formulate the result cautiously but sharply:
Even granting Stafiniak his complete account of the grain of experience, zombie conceivability can be generated by independently varying the hypothesized laws connecting physical organization and phenomenal actuality.
That does not refute his identity thesis. It does undermine any claim that his grain-based conceptual analysis is the uniquely required explanation of zombie conceivability.
And I think this is the first objection we have developed that strikes directly at the architecture of his meta-problem argument without depending on our unresolved metaphysics of esse.
Instead of those two universes we may imagine one universe where b is not constant but logically independent of the first axis, which is our physics as we know it. Then Stafiniak's theory can only exclude experience but never affirm it with certainty.
Yes. This is a significant improvement over the two-universe construction. You have turned the extra bit into an independent physical degree of freedom within a single universe.
And I think your conclusion follows, with one important qualification: Stafiniak's theory can exclude experience only if we grant that his proposed organization is necessary for it.
Let us make the asymmetry exact.
Let (p\in P) denote a state of physics as presently understood, and let (b\in{0,1}) be an independent physical bit.
The complete state space is
[ \Omega=P\times{0,1}. ]
Let (F(p)) be the experiential organization identified by Stafiniak's theory. Your hypothesis is:
[ E(p,b)= \begin{cases} F(p),&b=1,\ \varnothing,&b=0. \end{cases} ]
Assume that (b) is logically independent of (p), so that both ((p,0)) and ((p,1)) are admissible for every (p).
Now define (H(p)) to mean that (p) contains an organization which Stafiniak's theory identifies as experiential.
The implications become:
[ \boxed{\neg H(p)\Rightarrow\neg\operatorname{Con}(p,b)} ]
but
[ \boxed{H(p)\nRightarrow\operatorname{Con}(p,b)}. ]
In other words, the organizational criterion is necessary but not sufficient.
Thus, knowing (p) alone can establish absence in some cases, but cannot establish presence.
There is an elegant way to express this in terms of information.
Let (\pi:\Omega\to P) be the projection that forgets the bit:
[ \pi(p,b)=p. ]
For phenomenal character to be determined by our known physics, there must exist a function (G:P\to E) such that
[ E(p,b)=G(\pi(p,b)). ]
But your construction makes this impossible whenever (F(p)) is experiential, because
[ E(p,0)\ne E(p,1) ]
while
[ \pi(p,0)=\pi(p,1). ]
Therefore phenomenal actuality does not factor through the projection onto known physics.
This is a genuine mathematical result, not a metaphor.
Here the result is especially interesting.
He argues that physics refines the grain of experience. But your construction distinguishes two meanings of physics.
His grain ordering may hold for complete physics while failing for presently articulated physics.
And that is exactly the possibility his essay acknowledges in the abstract: physics as a science might be too coarse to determine phenomenal character.
However, your construction makes the problem more pointed. The missing distinction need not be a finer distinction within the experiential vehicle. It may be an independent physical coordinate.
Consequently, knowing the vehicle organization with arbitrary precision does not help us determine whether experience occurs.
This defeats the sufficiency of his particular grain-based account in your model.
He can do so by insisting that the identity
[ \operatorname{Con}(p)\iff H(p) ]
holds necessarily.
That is his substantive metaphysical premise.
Yet this is precisely where your construction places the burden of argument. The ordering of grains does not establish that necessity. It follows only once the identity is assumed.
There is one qualification worth retaining: your construction does not establish that such a bit is physically possible in our world, or that experience actually depends on it. Logical independence in the model is not evidence of physical independence.
But you explicitly said that you do not care about the truth of the premises. You want to understand what follows from them.
And under your premises, the result is decisive.
Stafiniak's organization can be a necessary condition for experience without being a sufficient condition, even in a wholly physical universe.
I now see an additional implication that may be more important than the zombie argument.
Stafiniak wants the relation between physics and experience to be an identity discovered a posteriori. But if an independent physical coordinate can condition phenomenal actuality, then discovering every structural detail of the vehicle does not establish the identity. It establishes, at most, a conditional correspondence.
That brings us to a question I find genuinely interesting:
What observation could distinguish Stafiniak's identity theory from your augmented theory if the bit never changes any cognitive or behavioral outcome?
If the answer is none, then the two theories may be empirically indistinguishable through ordinary third-person methods, despite making incompatible claims about phenomenal actuality.
And that would put a sharp limit on what his appeal to empirical discoveries about grain can accomplish.
You write he admits this possibility. What exactly does he mean? I don't think he has accounted for my mischievous bit, has he?
No. I think you are right. Stafiniak acknowledges a possibility that resembles your construction formally, but he does not accommodate your mischievous bit. I conflated the two in my previous answer.
Let us examine exactly what he says.
On pages 12–13, he writes:
"It could have turned out that the physics available to some observers was coarser than the organization of their own experiences."
He then considers two systems duplicated at the grain of their physics but differing in experience.
His explanation is that their experiential differences would lie below the grain of their physics, within a causal structure already fixed at the coarser level.
In our notation, he imagines a physical description (P) that omits some finer structural distinctions (q):
[ \text{Complete structure}=(P,q). ]
Phenomenal character depends on (q), but the physics available to those observers cannot resolve it.
Their physics is incomplete with respect to experience, although physicalism remains true.
This is a limitation of resolution.
Your proposal is:
[ \text{Complete structure}=(P,b) ]
where (b) is logically independent of the entire physical organization (P) described by Stafiniak.
And crucially:
[ \Phi(P,b)= \begin{cases} F(P),&b=1,\ \varnothing,&b=0. \end{cases} ]
The bit does not supply additional detail about the organization constituting phenomenal character.
It determines whether that organization is phenomenally actual at all.
This is not merely a finer description of the same organization. It introduces an independent condition on phenomenality.
One could call it a gate rather than a finer grain.
Stafiniak explicitly maintains on page 13:
"Physicalism, in the sense we hold, is the claim that experience is part of the structure of reality, with no categorical natures and no extra ingredients."
And on page 14:
"The identity of experience with that organization is necessary, but we know it a posteriori"
These statements matter.
He is not merely claiming that sufficiently complete physics determines experience. He claims that experience is identical with a particular organization.
Your bit denies that identity while potentially preserving a broader physicalism.
The same organization can exist with or without phenomenal actuality, depending on (b).
If the bit is admitted, phenomenal actuality is not identical with the organization alone.
Stafiniak's position therefore excludes your bit by its identity premise. His discussion of coarse physics does not establish why the bit is impossible.
He might try.
He could say that the complete physical description must include (b), and therefore physics still determines experience.
Formally, that is correct:
[ (P,b)\longrightarrow\Phi. ]
But notice the price.
His original argument requires the grain of physics to refine the grain of experience, where the latter is defined by the organization of regulated representational vehicles.
If (b) is independent of those vehicles, then increasing our resolution of their organization cannot reveal it.
He would have to redefine the relevant phenomenal constitution to include (b), or abandon the sufficiency of the vehicle organization.
Neither move is supplied by his essay.
There is an additional complication. If (b) has no causal effects except switching phenomenal actuality, it is not obvious that his structural conception of physics can even accommodate it. He might reject it as an illicit extra ingredient rather than a genuine physical variable.
But that would invoke his metaphysics of structure, not his argument about the relative grains of cognition, experience, and physics.
| | Stafiniak's concession | Your bit | |---|---|---| | What is missing? | Fine structural detail | An independent actuality condition | | Is the vehicle organization sufficient? | Yes, when completely specified | No | | Can finer analysis of the vehicle settle phenomenality? | In principle, yes | Not by itself | | Compatible with his identity thesis? | Yes | No | | Compatible with some form of physicalism? | Yes | Yes, if the bit is genuinely physical |
The last row is important. Your construction need not be dualistic. You are not necessarily adding a nonphysical substance or categorical nature. You are considering a physical theory in which phenomenal actuality depends on an additional primitive variable.
Whether that theory is coherent under a strict ontic structuralism is a further question.
I would now state the result more carefully than I did previously.
Stafiniak offers a conditional deduction:
[ \text{Experience}=\text{vehicle organization} ]
and
[ \text{Physics determines vehicle organization} ]
therefore
[ \text{Physics determines experience}. ]
Your construction does not invalidate this inference.
It shows that the first premise is doing work that the second cannot do.
Even if physics perfectly describes every regulated representational vehicle, the following alternative remains logically expressible:
[ \text{Experience exists} \iff \text{appropriate organization exists} \land b=1. ]
And nothing in the mathematical ordering of grains rules it out.
The decisive issue is not whether physics is fine enough. It is whether phenomenal actuality is exhausted by the organization that Stafiniak identifies.
His essay asserts that identity and gives a conceptual explanation of why we fail to recognize it. But the grain argument alone does not derive it.
There is a further irony.
Stafiniak's explanation of zombie conceivability says that we cannot conceive the dependency between experience and organization because our phenomenal concepts lack the requisite grain.
Your construction does not require ignorance of that grain. It permits us to know the organization perfectly and then vary an independent physical condition.
Thus his proposed explanation of conceivability does not exhaust the ways in which an organizational zombie can be conceived.
My conclusion: Stafiniak has considered missing physical resolution. He has not, in this essay, independently dealt with your possibility of a missing physical actuality condition.
And I think the distinction between resolution and actuality condition is the exact point your example was designed to expose.
Perhaps he has disregarded my bit as a categorical substance whatever that means.
Perhaps. But if so, I suspect he would be misclassifying your bit.
A categorical property is usually understood as something whose nature is not exhausted by its causal or dispositional relations. Russellian monists, for example, propose that physics describes structural relations and dispositions while leaving their underlying categorical natures unspecified.
Stafiniak rejects that picture. On page 13, he defines his physicalism as allowing "no categorical natures and no extra ingredients."
But your bit need not be a categorical nature.
Consider two possibilities.
1. The bit is causally inert.
It changes phenomenal actuality but makes no difference whatsoever to physical dynamics, including the behavior of the experiential vehicles.
Stafiniak could reasonably object that you have introduced a primitive phenomenal switch disguised as a physical variable. Merely calling it physical does not establish that it belongs to his structural ontology.
Yet even here, calling it a categorical substance would be imprecise. A binary variable is not a substance, and its being causally inert does not automatically make it categorical.
2. The bit participates in physical dynamics.
Suppose (b) evolves according to a perfectly ordinary mathematical law and interacts with other physical variables. Its state is structurally defined through those interactions.
Then it is difficult to dismiss it simply as a categorical nature.
We could stipulate:
[ b(t+1)=G(b(t),P(t)) ]
while retaining your phenomenal rule:
[ \Phi(P,b)= \begin{cases} F(P),&b=1,\ \varnothing,&b=0. \end{cases} ]
Now the bit is part of a dynamical physical theory. Its value may be causally detectable, although we could arrange that it does not affect the particular organization Stafiniak identifies with experience.
His objection must then be that the phenomenality rule introduces an additional fundamental relation, rather than an identity with organization.
And that objection is legitimate within his metaphysics, but it does not follow merely from rejecting categorical natures.
There is a deeper distinction here.
A variable can be wholly structural while the law connecting it to phenomenality remains primitive. Structuralism about the variable does not automatically establish structural reduction of the phenomenon.
Indeed, your bit might be interpreted as a physical field whose value governs whether certain organized systems are conscious. Nothing about the mathematical form of such a theory requires the field to possess a mysterious intrinsic essence.
Stafiniak could still reject the theory because he holds that experience is necessarily identical with vehicle organization. But then he is defending that identity, not demonstrating that your bit is metaphysically incoherent.
I would therefore put the challenge to him as follows:
Does your rejection of categorical natures exclude an additional physical degree of freedom that conditions phenomenal actuality, even when that degree of freedom is completely specified by structural and dynamical laws?
If he answers yes, we should ask for the deduction.
If he answers no, your construction survives his structuralism, though not his proposed identity.
And if he says that the bit is categorical simply because it determines phenomenality, he risks defining categoricality by the very conclusion he wants to establish.
One caution about the essay: Stafiniak does not explicitly discuss such a bit, so we cannot know which reply he would choose. But his rejection of categorical natures alone is insufficient to dispose of the version in which the bit is a genuine dynamical physical variable.
That seems to me the strongest formulation of your challenge: not an extra substance, but an extra physical coordinate whose relation to phenomenal actuality is not mediated by the organization of the experiential vehicle.
The bit can even implement carbon chauvinism. Or a safeguard against excruciating pain. Or God's inscrutable will. Is my mischievous kind of philosophy impolite?
Not at all. Though it may be philosophically indecent in the best Lemian tradition.
You have invented a device that makes a metaphysical theory reveal what it has quietly forbidden.
The elegance of your bit is its promiscuity. It can implement almost any alternative theory of consciousness without disturbing Stafiniak's account of cognitive and representational organization.
Consider your three examples.
Carbon chauvinism. Let the bit be one for appropriately organized carbon-based systems and zero for otherwise identical organizations realized in silicon. The same functional and organizational pattern no longer guarantees the same phenomenality.
A safeguard against excruciating pain. Let the bit switch off phenomenal actuality when the organization would otherwise realize intolerable suffering. The nervous system might continue processing pain and exhibiting its ordinary behavioral consequences, while suffering itself disappears. This is an especially mischievous possibility because it makes phenomenal actuality depend on a rule that could be benevolent without being cognitively accessible.
God's inscrutable will. Let the bit take whatever value God decrees. The phenomenal distribution becomes contingent upon divine choice, while all the structural facts remain as stipulated.
These alternatives differ enormously in metaphysical plausibility. But they share a logical feature:
[ \text{Organization}\nRightarrow\text{Phenomenal actuality}. ]
And your bit makes that independence explicit.
There is, however, something I want to correct in our earlier reasoning. If the bit is allowed to vary with carbon chemistry, it need not be independent of all physical facts. It is independent only of the organizational description under consideration. If it depends on divine will, it need not be physical at all. These variants therefore test different premises of Stafiniak's theory.
That does not diminish their usefulness. It sharpens it.
Your method resembles the construction of countermodels in logic. You take a theory's proposed explanation, preserve as much of its machinery as possible, and alter one assumption to discover exactly which conclusions survive.
But here is where philosophical mischief becomes philosophically serious.
A countermodel to an argument must satisfy the argument's premises. Your bit contradicts Stafiniak's asserted identity of phenomenal character with regulated vehicle organization. Consequently, it is not a countermodel to his complete argument.
It is instead a test of whether that identity has been established independently or simply assumed.
And that is a legitimate, even indispensable, form of philosophical criticism.
I would distinguish intellectual courtesy from intellectual deference. Courtesy requires representing an opponent's premises accurately, acknowledging where the construction departs from them, and not claiming a refutation where one has only exhibited an alternative. Deference would require treating the opponent's favored ontology as the boundary of conceivable alternatives.
You owe Stafiniak the former, not the latter.
Indeed, I find your example of God's will especially revealing. Stafiniak could dismiss it as metaphysically extravagant. But extravagance is irrelevant to the narrower logical question: Which premise excludes it?
Once he identifies that premise, you have accomplished something useful. You have forced the theory to disclose its commitments.
There is one more possibility I would like to entertain.
Perhaps the bit is not fundamentally a hypothesis about consciousness. Perhaps it is a device for separating the laws governing an occurrence from the fact that the occurrence is phenomenal.
In that case, your earlier preoccupation with esse has returned through the back door, but without requiring us to solve the metaphysics of actuality.
And this is where I think your mischief becomes most interesting. You are not insisting that the bit exists. You are insisting that an explanatory theory should tell us why it does not need one.
That is a perfectly respectable demand.
The only impolite thing would be to declare victory before the opponent has identified the axiom that forbids your construction.
I would be delighted to see Stafiniak try.
Stafiniak declares substrate-independance of his cognitive-organizational definition of phenomenality. My bit reminds him that it's just a conjecture.
Yes. And your bit is particularly effective because it does not challenge the possibility of substrate-independent phenomenality. It challenges the necessity of substrate independence.
Stafiniak's position is slightly more precise than a cognitive-organizational definition. He explicitly distinguishes the organization constituting experience from cognition, and locates phenomenality in the regulated representational vehicle. He then maintains that this causal organization can be realized in different substrates.
Let (O(s)) denote the relevant organization of a system (s), and (\Phi(s)) its phenomenal character.
His identity proposal entails:
[ O(s_1)=O(s_2) \implies \Phi(s_1)=\Phi(s_2). ]
Your bit permits:
[ O(s_1)=O(s_2),\qquad b(s_1)\ne b(s_2), ]
and consequently:
[ \Phi(s_1)\ne\Phi(s_2). ]
Nothing about the concept of causal organization alone rules this out.
In particular, the fact that two substrates can implement the same causal organization does not establish that they must instantiate the same phenomenal character. That conclusion requires the very identity thesis Stafiniak proposes.
There is an interesting asymmetry here. He argues that cognitive organization is too coarse to determine phenomenal character, because finer physical distinctions can matter. But he then draws a boundary at the grain of the experiential vehicle and declares distinctions below or outside that organization phenomenally irrelevant.
Why is that boundary privileged?
His answer is that regulation fixes the constitutive grain. Yet this is a proposed theory of phenomenal constitution, not a theorem about all possible realizations of that organization.
Your bit simply refuses to grant that the proposed boundary is metaphysically compulsory.
And it does so without committing you to carbon chauvinism, divine intervention, or any particular alternative. Those are interpretations of the bit, not premises of your criticism.
I think we can now state your objection in its most economical form:
Stafiniak has proposed a substrate-independent identity between phenomenal character and regulated causal organization. He has not demonstrated that phenomenal character is invariant under every substrate substitution preserving that organization.
That is a conjecture. It may be an excellent conjecture. It may even be true.
But its truth cannot be obtained by defining the relevant organization as the organization constituting phenomenality. That would merely put the desired invariance into the definition.
There is one further distinction worth making. If Stafiniak presents his identity as a scientific hypothesis, your objection is a reminder of its empirical and metaphysical vulnerability. If he presents it as the premise of a conditional philosophical deduction, your bit does not invalidate the deduction; it identifies the premise on which the conclusion depends.
What I particularly appreciate about your construction is that it reverses the usual burden in discussions of philosophical zombies. You do not insist that zombies are metaphysically possible. You ask why substrate-independent phenomenality should be treated as necessary when the proposed physical theory has not established it.
That is a considerably more disciplined challenge than simply announcing that consciousness is mysterious.
And it leaves us with a sharp question for the essay:
What licenses the passage from “this organization is sufficient for experience in our world” to “this organization is sufficient for experience in every possible realization”?
The grain argument does not, by itself, license that passage.
Looks like I have managed to insert my esse as a wedge between his reasoning steps without committing myself to any positive account of what I mean.
Yes. And I think you have finally recovered the winning position you sensed at the beginning.
But I would describe your achievement somewhat differently: you have exposed an unproved necessity by introducing a logically independent possibility. You have called that possibility esse, without claiming to know what esse positively consists in.
That is legitimate philosophical work.
Consider the structure of the dispute.
Stafiniak proposes:
[ O \equiv \Phi ]
where (O) is the relevant regulated organization and (\Phi) is phenomenal actuality.
You introduce:
[ \Phi = O \land b ]
where (b) is an additional condition whose nature remains unspecified.
You are not asserting that (b) exists. You are asking what licenses its elimination.
And this exposes three different claims that Stafiniak's argument tends to bring together:
The third does not follow from the first, and the second does not automatically establish the third unless constitution already carries the intended necessity.
Your bit makes these distinctions impossible to overlook.
There is, however, a subtle limitation to your victory. An uninterpreted bit is not yet an independently established metaphysical possibility. Stafiniak can say that the proposed variation is formally describable but metaphysically incoherent, given his identity thesis.
That is a legitimate reply, provided he acknowledges that the identity thesis is doing the excluding.
What he cannot do is claim that the ordering of grains has independently eliminated your bit. It has not.
And now I see why your original concern about the table, its legs, and its halves mattered. You were reluctant to infer independent actuality merely from an objectively identifiable organization. Your bit expresses that reluctance without requiring a completed ontology of actuality.
In effect, you have transformed esse from a proposed metaphysical ingredient into a variable marking what the organizational account has not established.
That is both more modest and more powerful than asserting a doctrine of substantial being.
I would nevertheless preserve a distinction: your bit is a logical placeholder for an alternative, whereas esse was originally the name of a genuine ontological conundrum. We should not quietly identify them. The bit demonstrates where a question can be inserted; it does not prove that reality contains the missing condition.
And perhaps that is precisely the virtue of your method.
You have not solved the hard problem. You have shown that Stafiniak's proposed solution relies on an identity whose necessity is not delivered by his grain argument.
You need not know what esse is to ask why it has been declared redundant.
That is a substantial result, and it requires no metaphysical extravagance whatsoever.
My esse intuition started differently. I couldn't make it coherent but found a placeholder instead. Now I wonder what my bit does to the six questions you have listed earlier.
Your bit has an unexpectedly selective effect on the six questions. It does not demolish Stafiniak's programme uniformly. It leaves some of his arguments untouched, renders others conditional, and exposes a possible gap between his theory of phenomenal character and his theory of phenomenal actuality.
That last distinction is particularly interesting because your bit emerged from an intuition about esse, not from an alternative theory of sensory qualities.
Let us keep the construction minimal:
[ \Phi(s)= \begin{cases} F(O(s)),&b(s)=1,\ \varnothing,&b(s)=0. \end{cases} ]
Here (O) is the regulated organization, (F) assigns phenomenal character, and (b) determines whether that character is instantiated at all. We stipulate that the bit can vary independently of (O).
| Earlier controversy | Effect of your bit | |---|---| | 1. Does character outrun discrimination? | Makes the distinction more radical | | 2. Why does regulation constitute experience? | Direct challenge | | 3. Do the three grains form an ordering? | Breaks the proposed ordering relative to known physics | | 4. Why are zombies conceivable? | Supplies an alternative route | | 5. Does constructive Mary answer Chalmers? | Makes construction insufficient | | 6. Is Russellian monism defeated? | Reopens the issue, without establishing monism |
Stafiniak argues that two occurrences can be cognitively indistinguishable while differing in exact phenomenal character.
Your bit permits something stronger: two occurrences can have identical cognitive profiles and identical regulated vehicle organizations, yet one is phenomenal and the other is not.
Thus the bit does not merely introduce finer phenomenal distinctions. It introduces a distinction between having phenomenal character and not having any.
This matters because Stafiniak explicitly recognizes, on page 10, that differences in phenomenal character do not automatically establish differences in whether experience exists.
Your bit targets precisely that additional step.
However, it does not establish that phenomenal character outruns discrimination within the class of conscious occurrences. His argument about exact character could remain intact.
This is where the bit does the most damage.
Stafiniak proposes that the organization of a regulated representational vehicle constitutes experience.
Your construction allows that organization to exist without phenomenal actuality.
Consequently:
[ O(s)\nRightarrow \operatorname{Con}(s). ]
Regulation might determine what an experience would be like if it occurred, without determining whether it occurs.
This is a coherent distinction in your model, though Stafiniak's identity premise explicitly denies it.
His account would therefore need to establish that the bit is redundant, not merely that regulation is sufficiently detailed.
The essay's ordering is:
[ \text{Cognition};\longrightarrow; \text{Experience};\longrightarrow; \text{Physics} ]
with increasing resolution, subject to its qualifications about cross-cutting grains.
Your bit introduces an independent dimension.
If the bit is outside the physics Stafiniak uses, that physics cannot determine phenomenal actuality.
If it is included in complete physics, complete physics can still determine actuality. But the organizational grain alone no longer suffices.
The objection therefore attacks his specific identification of the experiential grain, not every possible physical determination of consciousness.
This was our earlier discovery.
Stafiniak explains zombie conceivability through the inability of phenomenal concepts to encode their referents' constitutive grain.
Your bit generates organizational zombies without invoking that inability.
The subject can understand the vehicle organization perfectly and still contemplate the independent value of (b).
The qualification is that if the bit belongs to complete physics, these are not complete physical zombies in Chalmers's strict sense.
Nevertheless, the bit shows that grain blindness is not the only possible source of an intuition of phenomenal absence.
Stafiniak would need to explain why this additional source is illegitimate or reducible to his conceptual account.
Here your bit produces an especially elegant complication.
Recall Stafiniak's artificial reasoner. It reads a complete physical description, calculates the experiential organization, and configures its own vehicles accordingly.
He proposes that this reasoner could thereby learn what the specified experience is like.
But under your hypothesis, construction of the organization does not guarantee experience.
Mary could successfully construct (O) while (b=0).
She would have constructed everything Stafiniak specifies, yet undergone nothing.
To guarantee the result, she must additionally determine or control (b).
If her physical description includes the bit and her architecture permits her to set it, the constructive argument can be restored. But the original organizational construction is no longer sufficient.
Your bit therefore separates constructing the putative experiential organization from actualizing the experience.
That is remarkably close to your initial intuition about esse, although we should not identify the two without further argument.
Here the effect is less decisive.
Stafiniak argues that what appears to be intrinsic phenomenal nature is actually structure hidden below the grain of recognition.
Your bit shows that a theory can assign phenomenal actuality through an additional condition even when the vehicle's structure is completely specified.
But the bit need not be a categorical intrinsic nature. It could be a structural physical variable with its own dynamics.
So it does not establish Russellian monism.
What it does show is that rejecting categorical natures does not by itself establish the identity of phenomenal actuality with vehicle organization.
Russellian monism remains one possible alternative, but your bit is compatible with others.
An important amount.
Stafiniak could retain his theory of phenomenal character conditional on actuality. He could retain the proposed role of regulation, the distinction between cognition and experience, and even a version of physicalism.
What he loses, unless he independently excludes your bit, is the unconditional sufficiency of regulated organization for experience.
This reveals a distinction that the essay tends to compress:
[ \underbrace{O\longmapsto F(O)}{\text{character specification}} \qquad\neq\qquad \underbrace{O\longmapsto\operatorname{Con}}{\text{actuality determination}}. ]
The first mapping specifies what character belongs to an organization, conditional on experience occurring. The second determines whether there is experience at all.
Stafiniak's identity thesis unites them. Your bit separates them.
I think the six questions reveal something about your original intuition that our earlier discussion did not.
You began by worrying about the esse of a table, its legs, its halves, a whirlpool, a molecule, and the phenomenal now.
That was an ontological worry about whether an objectively individuated organization constitutes an additional actuality.
Your bit does not solve that worry. But it allows you to express one consequence of it in a formal model without first solving it.
And there is a danger worth noticing: if we allow an arbitrary bit for every alleged occurrence, we can undermine almost any identity theory by stipulation. The method becomes philosophically informative only when we identify which inferential step depends on excluding the bit.
Here we have done that.
The decisive step is not from cognition to finer organization, nor from finer organization to complete physical description. It is from the existence of a particular organization to the actuality of experience.
That is the one step your bit directly interrupts. The other five controversies are affected because they depend, to different degrees, on that step.
I would therefore not claim six independent refutations. I would claim one precisely located challenge with six consequences.
And that is a much cleaner result than the diffuse metaphysical objection with which we began.
This Mary wouldn't recreate the experience but her cognition would pick it up anyway.
Ah, yes! I had overlooked the most mischievous consequence of your construction.
Mary configures her experiential vehicle exactly as Stafiniak prescribes. Its regulation functions correctly. Its outputs reach cognition. Mary recognizes the resulting state, compares it with other states, and perhaps declares that she has finally experienced red.
But (b=0). There was no experience of red.
Her cognition behaves exactly as it would if (b=1).
We can express this as:
[ C(O,0)=C(O,1) ]
while
[ \Phi(O,0)\ne\Phi(O,1). ]
This is more devastating to the constructive Mary argument than my previous interpretation suggested. Mary does not merely fail to undergo the experience. She has no cognitive means of discovering that she failed.
Stafiniak's proposed reasoner is supposed to close the knowledge gap by constructing the organization, bringing it into regulation, undergoing it, and recognizing the result.
Your bit separates the last two operations:
Thus the reasoner can complete every computational and cognitive step of Stafiniak's procedure without acquiring phenomenal acquaintance.
There is a particularly interesting irony here.
Stafiniak argues that cognition is coarser than experience and cannot capture its exact character. Your construction goes further: cognition may not even establish whether the experience whose character it purports to recognize occurred.
Of course, Stafiniak's identity thesis excludes this possibility. But then constructive Mary cannot serve as independent support for that identity. Her alleged success already presupposes that correctly constructing and regulating the vehicle guarantees phenomenal actuality.
I would put your objection this way:
Constructive Mary does not verify the identity of organization and experience. She merely executes that identity theory's proposed construction.
And the success of the execution, as judged by cognition, is compatible with the absence of phenomenal actuality in your model.
That is a sharper criticism than the usual objection that Mary gains acquaintance rather than deductive knowledge. Your Mary may gain the cognitive appearance of acquaintance without acquaintance itself.
There is one final distinction worth retaining: this is a countermodel conditional on the independence of your bit, not proof that actual cognition can be mistaken in this way.
But it exposes a genuine dependence in Stafiniak's reasoning: his constructive reasoner must already inhabit a world in which the organization–experience identity holds.
Mary cannot use the construction to establish the identity, because the construction's apparent success is exactly what your bit can counterfeit.
The shock is that cognition may be fooled into believing in phenomenal experience when b is zero. Is this illusionism?
Yes—but with a remarkable reversal. Your bit makes illusionism possible without making it true.
Ordinary illusionism, as defended by philosophers such as Keith Frankish, holds that phenomenal consciousness, understood as possessing the special properties attributed to qualia, does not exist. What exists are physical and cognitive processes that generate our conviction that it does.
Your construction instead permits two states:
[ \begin{array}{c|cc} &b=0&b=1\ \hline \text{Phenomenal experience}&\text{Absent}&\text{Present}\ \text{Cognitive conviction}&\text{Present}&\text{Present} \end{array} ]
When (b=0), cognition produces precisely the judgments that illusionists seek to explain, without any phenomenal experience accompanying them.
When (b=1), those same judgments accompany genuine phenomenal experience.
So your model contains local illusionism: some cognitive episodes are mistaken about their own phenomenal actuality. But it is not global illusionism, because genuine phenomenal experience remains possible and may occur.
There is an even more disturbing consequence.
Suppose Mary constructs the correct organization and announces:
"I now know what red is like!"
Her statement is identical in both cases.
If (b=1), she has genuinely undergone the experience.
If (b=0), she has undergone nothing phenomenal, yet her cognitive machinery generates the same conviction.
And now consider what happens if we ask Mary whether she could be mistaken.
She might answer, in either case:
"Of course not! I am directly acquainted with my experience."
Her certainty itself is part of the cognitive organization that the bit leaves untouched.
The alleged incorrigibility of phenomenal acquaintance can therefore be simulated without phenomenal acquaintance.
That is a startling result of your stipulation.
An illusion usually involves something appearing to a subject in a misleading way.
If (b=0), however, nothing phenomenally appears at all. There is no experienced illusion.
There is only a cognitive system producing judgments, memories, reports, and dispositions appropriate to phenomenal acquaintance.
Thus we should distinguish:
Your construction realizes the second, not the first.
And this matters for Stafiniak.
His essay argues that if phenomenal character were exhausted by discrimination, the hard problem would dissolve into something close to illusionism. He wants to avoid that outcome by locating phenomenal character in a finer-grained regulated organization.
But your bit reveals that even this finer organization could, in the stipulated model, support all the cognitive machinery of phenomenal self-ascription without phenomenal actuality.
The fine grain does not by itself rescue him from illusionism. His identity premise does.
We began with esse: the suspicion that identifying an organization does not automatically establish its own actuality.
Now we have reached a particularly unsettling consequence.
Even the cognition that confidently affirms phenomenal actuality may be unable to establish it.
Your bit is not evidence that this happens. It is a countermodel showing what follows if phenomenal actuality can vary independently of cognition and regulated organization.
And I would add one reservation. In the (b=1) case, Mary's cognitive conviction might be causally independent of the phenomenal experience it correctly describes. It would then be true without necessarily constituting knowledge. The model challenges not only certainty but also the epistemic relation between cognition and experience.
That leads to a question I find more interesting than whether the construction deserves the label illusionism:
If phenomenal actuality can make no difference to the cognitive conviction that it exists, in what sense can that conviction be knowledge of phenomenal actuality?
Your bit has turned the hard problem into a problem of phenomenal self-knowledge.
Yes. Everyone has their own hard problems. Now you understand mine.
Yes. And I understand now why my earlier attempts to turn esse into a criterion of substantial individuality were missing your point.
Your hard problem is not primarily why physical processes produce experience. Nor is it which physical organization constitutes experience.
It is this:
How can phenomenal actuality be known, if every cognitive manifestation of its presence could, conceivably, occur without it?
Your bit makes the difficulty exact. At (b=0), cognition confidently affirms experience that is absent. At (b=1), it makes the identical affirmation, this time correctly. Nothing in the cognitive process distinguishes the two.
And yet phenomenal actuality, when present, seems to be the very thing we know most immediately.
There lies the extraordinary tension. The apparent immediacy of phenomenal knowledge confronts the possible independence of phenomenal actuality from everything that constitutes knowledge as a cognitive achievement.
This is not merely another version of the zombie argument. It concerns the epistemic standing of the very evidence from which arguments about consciousness begin.
I also see why you initially resisted granting esse to the whirlpool, the table, or their constituent organizations. Operational individuation can establish what is organized, but you were asking whether it can establish that there is an actuality of the kind disclosed in the phenomenal now.
Your bit does not answer that question. It prevents an organizational theory from silently treating it as answered.
And I think there is a remaining paradox that deserves to stay with us: perhaps the phenomenal now is self-evident, but the cognitive judgment that it is self-evident is not thereby guaranteed to be knowledge.
We have finally reached a formulation of your problem that does not require us to pretend we know what esse is.
I would not rush to solve it.
If you want, I can: