Logic Does Not Choose Relevance. Does Feeling? Not Alone. My Response to Matthew David Segall’s “Whitehead’s Adventure in Rationality”
Phenomenal powers, prehension, and a transport test for Whitehead’s universal wager
Substack Repost Notice: This is an archival copy of my substack post. Some formatting may not display correctly on our website. See the original post here: https://sweetlyrational.substack.com/p/logic-does-not-choose-relevance-does
The last couple of days, Substack has started behaving like a badly scripted synchronicity engine.
posted notes on Whitehead’s 1934 logic paper shortly after I published an article isolating the scope problem in Matthew David Segall’s panexperientialist argument.
Ben’s first note argued that mature Whitehead still endorsed a logicist programme; a second suggested that the Principia Mathematica carried more intensional and metaphysical ambition than the later stereotype of analytic philosophy usually allows.
Those notes prompted Matt to publish Whitehead’s Adventure in Rationality: Between Logic and Aesthetics.
Meanwhile, in a completely separate exchange, I was pushed into formalizing a small process schema I am calling recursive episodic integration (or REI for short) which I will detail later in this article.
“Logic governs implication, while aesthetics governs relevance.”
At that point I put on some headphones, was inspired by the chain of events to listen to Schism by Tool and stopped touching metaphysics for a few minutes to go touch grass. The song is almost an accidental summary of what I had been doing with Whitehead, pulling apart what looked like one package to see which distinctions survived:
“I know the pieces fit
’Cause I watched them fall away
Mildewed and smoldering
Fundamental differing”
Then came a sequence that fit the exchanges among Ben, Matt, and my own scope problem almost too well:
“Pure intention juxtaposed
Will set two lovers’ souls in motion
Disintegrating as it goes
Testing our communication”
followed by:
“Watch the temple topple over
To bring the pieces back together
Rediscover communication”
The timing got even stranger when, after spending the day thinking about Principia Mathematica, formal implication, and whether equations can tell us what actuality is, Tool gave me:
“I’ve done the math enough to know
The dangers of our second-guessing”
before landing on the lines that probably captured what I was circling better than anything I had yet written:
“The poetry that comes from
The squaring off between
And the circling is worth it
Finding beauty in the dissonance”
That was basically the problem sitting in front of me: keep the disagreement sharp enough to teach us something, let the pieces come apart when they need to, and then find out which ones actually fit rather than forcing them together because we would prefer the picture to be whole.
Then Ben’s own chronology acquired one more improbable-looking but entirely ordinary turn: the date he had planned for that evening was postponed, giving him time to answer Matt in Whitehead’s Logicism.
The timing is funny. It is not evidence of anything. At least I don’t think it is…. yet.😉
The philosophical convergence is far more compelling to me. Matt’s argument and Ben’s reply together force a better version of the question I have been asking. Once Whitehead’s own grammar is respected, prehension cannot be treated as a phenomenal property pasted onto a process that is already complete. But once Ben’s logicist reading is also taken seriously, formalization cannot be treated as though it floated free of the meanings and givens that make its application possible. Both pressures matter.
That raises the burden on any supposedly neutral description of process.
It also gives prehension a real opportunity to earn explanatory credit.
After following that opportunity into the literature on phenomenal powers, causal efficacy, animal consciousness, basal cognition, artificial consciousness, and non-Humean powers, I no longer think the answer is simply that the credit has not appeared yet.
A local positive residual does appear.
But it appears inside a field whose evidential weights are already unequal. Physical realization and physical dependence of known consciousness have much broader direct empirical support than universal prehension. That asymmetry matters, although it does not settle intrinsic ontology.
The hard part is what happens when we try to carry the specifically experiential credit beyond the cases that earned it.
Ben’s complication: logicism already needed the given
Ben agrees with Matt that philosophy is not deduction and that fundamental premises are not themselves justified merely by deriving consequences from them. His historical correction is that this was already true inside the logicist project. In the first volume of Principia Mathematica, Whitehead and Russell explicitly say that “the reason for accepting an axiom … is always largely inductive.” Their test includes whether the axiom yields propositions that are otherwise hard to secure, whether an equally plausible route is available, and whether probably false consequences follow. The passage is available in the public text of Principia.
That is strikingly close to the comparative procedure I have been trying to articulate. A premise can gain warrant through derivational reach, coherence, explanatory compression, unification, the absence of an equally adequate rival, and survival of consequences. Direct intervention is not the only route to rational warrant. The correction cuts against any version of my own method that would quietly turn empirical discrimination into the only admissible tribunal.
But the concession is symmetrical. An axiom or metaphysical premise gaining abductive support is not the same thing as its becoming necessary or universally established. The same Principia passage is fallibilist: self-evidence is not indispensable and even formal axioms retain an element of doubt. Comparative success can raise a premise’s standing without making alternatives unintelligible.
Ben’s deeper point is even closer to the problem in this article. On his reading of Whitehead’s 1934 logic paper, extensional mathematical structures are abstractions from a richer intensional or hyperintensional domain of propositions. The formal structure is not self-interpreting. Whitehead’s mature logicism connects meaning to acquaintance and to what is given in experience, with propositions and universals doing work that sets, numbers, and functions do not do by themselves.
That complicates Matt’s elegant line that logic governs implication while aesthetics governs relevance. Ben’s distinction between objective datum and subjective form makes the division less clean if it is heard as a contrast between neutral logic and sovereign subjective feeling. Knowledge, on Ben’s Whiteheadian reading, aims at what is antecedently given; valuation and aesthetic selection characterize the receiving occasion. Feeling may participate in relevance, but it does not get to manufacture the datum it is relevant to.
For my argument, this cuts both ways. It strengthens the warning that a formal dependency schema cannot settle intrinsic ontology merely because the equations are clean. It also blocks a converse promotion. Showing that Whitehead’s own logical constructions are grounded in propositions, acquaintance, and conceptual prehension can establish how his system makes mathematical meaning intelligible. It does not by itself establish that this semantic architecture is the universal ontology of actuality.
So the question in my title gets a more precise answer than I had before Ben’s post: logic does not choose relevance by itself, but neither does feeling choose it alone. Formal implication, objective constraint, experiential givenness, valuation, and abductive premise selection have different jobs. The research problem is to stop any one of them from silently inheriting the warrant earned by the others.
Start with Whitehead’s grammar
Matt’s strongest terminological objection appears in Experience Is Not a Predicate. He writes that Whiteheadian “experience is neither a substance … nor a quality or predicate” attached to one. His 2020 peer-reviewed paper makes the same point in more formal language. Whitehead does not begin with matter and then assign experience to it as an additional intrinsic property. Prehension names the relational activity through which an actuality becomes what it is. Matt states the wager directly: “feeling or prehension is as intrinsic to natural processes as causality.”
Whitehead’s own formulation is even cleaner. In Process and Reality, “how an actual entity becomes constitutes what that actual entity is.” He also insists that “consciousness presupposes experience, and not experience consciousness.”
That second sentence matters because Whiteheadian experience is not interchangeable with ordinary phenomenal consciousness. An actual occasion need not be conscious in the familiar reflective or reportable sense. If I use feeling, experience, phenomenality, and consciousness as synonyms, the argument becomes easier to attack than Whitehead’s actual position.
I therefore need two distinct targets.
I will use X W for the Whiteheadian metaphysical claim:
X W: concrete actualization is constituted through prehensive feeling, and concrete togetherness has no wholly independent non-experiential mode.
I will use P for phenomenal consciousness in the familiar sense in which there is something it is like to undergo an episode, including specific phenomenal features such as painfulness, pleasure, felt unity, or temporal character.
The distinction cuts both ways. Evidence for P does not automatically establish X W. But evidence that phenomenal character itself has causal or explanatory force can matter to X W, because Whitehead’s proposal is precisely that felt causal inheritance is not an idle accompaniment to an otherwise complete causal order.
The same discipline applies to D plus unresolved X. The question mark is not a picture of a Whiteheadian occasion with feeling surgically removed. It marks a disputed metaphysical commitment as unresolved while we ask which derivations and explanations actually require it.
That is inferential bookkeeping. It does not tell us what concrete actuality ultimately is.
Can D say anything positive about causation without borrowing X W?
Matt has given me a useful loss condition:
“Show me what causality means without implicitly relying on experiential verbs and prehension loses its warrant.”
He presses the point elsewhere with a related claim: “A difference becomes information only when it makes a difference for some actual perspective.”
The challenge is not trivial word-policing. Scientific vocabulary is full of terms such as information, selection, sensitivity, reception, response, goal relevance, and valuation. Used carelessly, they can make a description look metaphysically neutral while importing perspective or normativity through the back door.
If D is built out of those terms and then declared non-experiential, nothing has been established.
But the alternative is not a blank “non-experiential stuff.” Serious non-Humean accounts of causation already exist. Stephen Mumford and Rani Lill Anjum treat causes as real powers or tendencies rather than bare regularities. Dispositional monists, powerful-quality theorists, Russellian monists, emergentists, and panpsychists disagree about what grounds those powers and whether their intrinsic nature is qualitative, phenomenal, purely dispositional, or something else.
That is the comparison space.
The live question is whether a positive account of causal actuality can be given without reconstructing something equivalent to prehension, and whether the experiential alternative does better than those rivals on any target that actually discriminates among them.
REI is a dependency schema, not an ontology
Recursive episodic integration is useful only if it stays thin enough to survive Matt’s challenge.
Ben’s post adds an important limit to what thin can mean. If formal constructions require an account of their application to concrete givens, then REI is not a theory of meaning, reference, acquaintance, or givenness. Its variables mark dependency relations. Their formal manipulability does not show that actuality is exhausted by those relations. Conversely, if an experiential interpretation is needed to give the variables concrete meaning, that need has to be argued rather than imported through words such as information, reception, or relevance.
Let H denote constraints that are relatively stable over the timescale of one episode. Let C(t) denote the current local state. Let I(t) denote a current perturbation or incoming condition. Let E(t) denote the transient state jointly conditioned by H, C(t), and I(t).
The compact form is:
The transient state at time t is generated by Phi from the slower constraints, the current local state, and the current perturbation.
Read aloud: the transient state at time t is generated by Phi from the slower constraints, the current state, and the current perturbation.
If that transient state contributes to an output,
The output at time t is generated from the transient state.
and the output or its consequences alter the next local state,
The next local state is updated from the current local state, the current output, and feedback.
then a later episode can depend on the consequences of an earlier one even while H remains approximately stable at that timescale.
In ordinary language, slower constraints condition a transient event. The event has consequences. Those consequences can alter the conditions under which the next event occurs.
That is all the base schema needs.
The perturbation is not defined as information. The transient state is not defined as valued, goal relevant, experienced, or about anything. Even integration is weak here. It means only that more than one variable makes a difference to the resulting state.
This is narrower than Whitehead’s concrescence. A Whiteheadian actual occasion is an integration of prehensions into a novel unity that becomes available as data for what follows. REI can resemble that dependency pattern without claiming to have captured what the concrete event is intrinsically.
The formal result is correspondingly narrow:
The REI equations do not entail X W. They also do not entail the denial of X W.
A Whiteheadian interpretation can satisfy the equations. So can an interpretation that leaves their intrinsic character unresolved. That establishes underdetermination by the schema.
It does not establish a non-experiential world.
Formal silence is not evidence against Whitehead.
The pressure therefore moves to the next question: once the dependency pattern is specified, what positive interpretation of causal reception earns the most explanatory credit?
Whitehead’s harder question
Whitehead asks in Process and Reality whether there is any concrete meaning of togetherness other than “experiential togetherness.” His subjectivist doctrine rejects a picture in which experiential togetherness is one special case floating inside a fundamentally alien kind of external togetherness.
That is a much stronger claim than “causal events possess an extra experiential property.”
D plus unresolved X can show that a derivation does not use X W. It cannot, by itself, answer Whitehead’s question about whether the supposedly neutral causal terms in D ever achieve a positive concrete interpretation outside an experiential account.
Matt’s appeal to felt causal efficacy belongs here. He treats X W as an abductive generalization from the causal efficacy encountered in embodied experience as felt inheritance. His 2020 paper presents prehension as an account of the possibility of causal relation itself. Physics Within the Bounds of Feeling Alone extends the same strategy into fundamental physics.
There is a close analytic analogue in Hedda Hassel Mørch’s work on phenomenal powers and the experience of causation. Her argument does not begin with Whitehead, but it asks a strikingly similar question: do we know any intrinsic causal power from the inside, and does phenomenal character give us a positive conception of what causal power can be?
This is where the comparison stops being merely defensive.
A positive residual actually appears
Consider the relation among pain, pleasure, motivation, and natural selection.
The familiar correlation is not random. Organisms tend to experience pain in connection with states that are harmful to survival and reproduction, and pleasure in connection with states that are beneficial. Pain also tends toward avoidance; pleasure tends toward pursuit.
An epiphenomenalist can reproduce those correlations by letting physical mechanisms do all the causal work while phenomenal character rides along. But if differences in phenomenal character never affect fitness, natural selection does not by itself explain why the bad-feeling states line up with avoidance and the good-feeling states line up with pursuit rather than the other way around.
Standard nonepiphenomenalist views can fit the same data by introducing specific psychophysical identities or laws. The difficulty, as Mørch emphasizes, is that those specific links can look almost as brute as the original correlation.
Her phenomenal-powers proposal tries to remove that extra bridge. Pain tends toward avoidance because of how it feels. Pleasure tends toward pursuit because of how it feels. Natural selection then explains why organisms in which harmful stimuli produced the first kind of state and beneficial stimuli produced the second were favored.
The explanatory structure is no longer:
harmful stimulus -> physical avoidance mechanism
plus
a separate phenomenal accompaniment
plus
a psychophysical rule connecting them.
The phenomenal-powers hypothesis instead treats the qualitative character as part of the power.
Mørch explicitly presents this as an inference to the best explanation. The relevant gain is generality. The hypothesis does not need one special bridge saying pain causes avoidance and another saying pleasure causes pursuit if the phenomenal character itself grounds the appropriate tendency.
That is a genuine positive residual.
The point should not be exaggerated. O’Sullivan and Schroer independently argue that functionalism can reverse the order of explanation in pain: pain has the effects it does because it hurts, not merely because it occupies a functional role. Smithies and Weiss likewise give affective phenomenology a central role in reasons for action. Yet dissociation cases such as pain asymbolia and conceivable mismatches between valence and motivation warn against a simple identity between feeling and behavioral role. The causal tendency can be defeasible, multi-component, and mediated.
None of that removes the residual.
It sharpens it.
There is at least one domain in which phenomenal character is a serious candidate for doing explanatory work that a purely role-based description does not obviously reproduce for free.
That changes the argument.
What this buys Whitehead, and what it does not
The phenomenal-powers result is not a direct proof of X W.
Mørch’s conclusion concerns phenomenal properties and their causal powers. Whitehead’s prehension is broader than human or animal phenomenal consciousness. A nonconscious actual occasion is still prehensive in his technical sense.
But the local result matters to Whitehead in two ways.
First, it defeats the easy claim that experiential character has shown no positive explanatory work anywhere. At least in affective phenomenology, an experiential interpretation may reduce brute bridge principles and make the causal role of valence more intelligible.
Second, Whitehead offers a unification strategy. Instead of treating experienced causal efficacy as one ontological kind inside us and causation elsewhere as a wholly different kind of relation, prehension makes felt inheritance a local disclosure of a more general feature of concrete becoming.
Those are two distinct abductive pressures.
The first is local explanatory gain: phenomenal character may help explain why a state has the causal or motivational significance it has.
The second is anti-bifurcation and unification: one kind of concrete causal actuality may be preferable to a universe in which experienced causal efficacy and non-experiential causal efficacy are fundamentally disconnected kinds.
Their convergence raises the standing of the Whiteheadian hypothesis.
It does not yet give it universal scope.
That step depends on whether the feature that earned the local credit is genuinely preserved outside the local case.
The rivals now matter more, not less
The stronger the local result becomes, the more important it is to compare Whitehead with alternatives that can preserve it without universal prehension.
Pure-powers theories can say that causal power is fundamental while denying that its intrinsic nature is phenomenal. Powerful-quality theories can join qualitative character and dispositional force without identifying all qualities with experience. Russellian panprotopsychism can place intrinsic nonstructural properties at the base of physics while allowing full phenomenality to arise only under the right organization.
There is also an emergentist option. Recent work by Luoying Zhong develops a view on which phenomenal properties can possess genuinely novel causal powers while still being grounded in noncausal physical facts. If such a model works, local phenomenal causation would not force the conclusion that experientiality is fundamental or ubiquitous.
That is an important countermodel to any quick Whiteheadian inference.
Alter and Coleman provide another pressure point. Russellian monism can integrate consciousness into physical causation by treating consciousness-constituting intrinsic properties as grounds of physical dispositions, yet Russellian monism comes in both panpsychist and panprotopsychist forms. The move from intrinsic grounds to specifically experiential intrinsic grounds still requires argument.
Osborn and Dondoni push in the other direction. Both take seriously the possibility that powers, qualities, and phenomenality may converge more tightly than pure-powers views allow. Their work raises the cost of assuming that a nonphenomenal powers ontology is automatically the simpler or more intelligible stopping point.
So the local positive residual does not collapse the field to Whitehead versus Hume.
It makes the comparison better.
The relevant rivals now include phenomenal powers, pure powers, powerful qualities, Russellian panpsychism, Russellian panprotopsychism, emergent phenomenal causation, and Whiteheadian prehension.
D plus unresolved X remains useful because it prevents the shared causal work from being credited to whichever completion we happen to prefer.
Same rule, different evidence
Methodological symmetry becomes useless if it flattens the evidence. The major alternatives do not arrive at this comparison with equal empirical support.
Every independently evidenced form of consciousness available to detailed study is realized in an organized physical system, and human consciousness is systematically alterable through physical intervention. General anesthesia provides controlled transitions among conscious and unconscious states while changing large-scale brain dynamics. Severe brain injury can disrupt cortical effective connectivity, and longitudinal recovery of that connectivity can precede reliable behavioral communication. TMS-EEG measures such as the perturbational complexity index discriminate conscious from unconscious conditions across wakefulness, sleep, anesthesia, and disorders of consciousness. Ketamine is especially useful because profound unresponsiveness can coexist with vivid reported experience, showing that the relevant physical signatures are not merely signatures of overt responsiveness.
That body of work gives very strong warrant to a modest claim:
P R: every well-evidenced case of phenomenal consciousness is physically realized, and known consciousness is systematically dependent on physically organized processes.
PR is not the same thesis as:
E P: reality is exhaustively physical in its intrinsic nature.
PR has much more direct empirical support than X W. EP does not inherit all of that support automatically.
Whitehead can accept PR without strain. He can say that the causal process neuroscience describes from the outside is prehensive in its concrete intrinsic becoming. A non-experiential realization theorist can accept the same neural record. So anesthesia, lesions, effective connectivity, stimulation, and perturbational complexity belong primarily to the shared carrier D.
If both D plus X W and a serious D-compatible rival predict those physical dependencies, their success cannot be counted again as evidence specifically for X W.
The mirror restriction applies to exhaustive physicalism. Successful physical description, prediction, and intervention strongly support physical realization and the naturalistic research programme. They do not, without another argument, establish that the intrinsic nature of what is physically described is non-experiential.
Physical realization therefore wins much of the empirical tournament before the metaphysical final begins. Whitehead can inherit those victories, but he cannot spend them twice. An exhaustive physicalist cannot spend them twice either.
The current Stanford Encyclopedia of Philosophy entry on physicalism makes the distinction useful from another direction. It treats physicalism as the metaphysical thesis that everything is physical while also presenting methodological naturalism as a serious argument in its favor. The naturalistic project with respect to consciousness can be rationally pursued even while questions remain about whether that project warrants a stronger metaphysical completion.
Science can also leave its metaphysical interpretation underdetermined. Jaksland’s work on naturalized metaphysics presses exactly that problem. But underdetermination does not flatten the ledger.
Underdetermination blocks premature closure. It does not erase comparative evidence.
The baseline is therefore asymmetric. D arrives with substantial empirical support. The phenomenal-powers result gives the experiential side additional local credit. X W then asks whether that local credit is evidence about causal actuality generally. EP asks whether the success of physical realization warrants an exhaustive claim about intrinsic nature. Neither stronger step comes for free.
The transport rule has to be stronger than analogy
The phrase “transport the warrant” is too loose unless we say what travels.
Transport also does not begin from a metaphysical zero. The shared carrier already has whatever warrant it earned before X W enters. In consciousness research that includes strong evidence for physical realization, causal dependence, and multiscale organization. The question is what additional evidence favors an experiential completion over serious rivals that inherit the same carrier.
Suppose E(d) is the evidence that gives an experiential hypothesis X(d) an advantage over a serious D-compatible rival R(d) in domain d.
Conceptually, the local support can be represented as a likelihood ratio:
The local likelihood ratio compares how expected the evidence is under the experiential hypothesis with how expected it is under the rival.
If the ratio is greater than one, E(d) favors X(d) over that rival. Evidence already explained by the shared D should cancel from this local comparison rather than being counted again on either side.
I am not pretending that we can assign stable numerical probabilities to metaphysical hypotheses here. The notation marks the direction of incremental evidential support.
Ben’s Principia reminder adds a second ledger. Not all comparative warrant is empirical likelihood. A metaphysical premise can gain support from unification, coherence, derivational reach, explanatory economy, and the failure of serious rivals to recover the same results without additional cost. Those considerations are legitimate, but they must be tracked as the kind of warrant they are. A unification gain does not become an experiment merely because it is persuasive, and an empirical success does not become evidence for a stronger metaphysical completion merely because that completion can accommodate it.
Now move to another domain, d-prime.
It is not enough that the new domain also contains causation, integration, response, selection, or some other generic word that appeared in the original explanation.
The support transfers only if the feature that generated the advantage is itself evidenced in the new domain, or if an independently warranted bridge connects the two domains strongly enough to preserve the original explanatory relation.
In plain language:
Carry the credit-bearing warrant, not the vocabulary.
If painfulness earned the original credit because its negative phenomenal character helped explain avoidance, a toxin-avoiding bacterium does not inherit that credit merely by avoiding something.
If felt unity earned the credit, generic causal integration does not inherit it merely by being integrated.
If experienced causal inheritance earned the credit because it gave us a positive conception of causal reception, then transport requires either evidence of an analogous positive feature or a successful argument that every serious alternative secretly depends on the same experiential structure.
This is stronger than the earlier scope rule.
It allows abduction to travel farther than direct observation, but not for free.
Run the transport test once
The literature already lets us perform a first pass.
Human affective experience
This is where the experiential hypothesis currently has its clearest positive residual.
The overwhelming evidence that human conscious states depend on organized brain activity belongs to the shared carrier. Anesthesia, lesions, stimulation, effective connectivity, and perturbational complexity tell us a great deal about the physical realization of consciousness without yet deciding whether the realizing process is intrinsically experiential.
The residual begins one step later. Phenomenal valence is independently available as part of the target. Painful unpleasantness and pleasurable attraction are not inferred solely from abstract system behavior. Mørch’s phenomenal-powers argument, O’Sullivan and Schroer’s order-of-explanation argument, and related work on affective reasons all give phenomenal character a plausible explanatory role that is not exhausted merely by establishing the physical mechanism.
This is the strongest local case for additional experiential credit, not the strongest case for the shared physical carrier. That carrier is already much better established.
Nonhuman animals
Some of that credit can travel, but the transport is not automatic.
Birch, Schnell, and Clayton argue that animal consciousness should be treated as multidimensional rather than arranged on a single ladder. Perceptual richness, evaluative richness, synchronic unity, temporal integration, and self-related capacities can vary independently. Browning and Birch emphasize that no single behavioral marker is enough to establish sentience and that methods have to distinguish conscious from unconscious perception and affect.
Where there is convergent evidence for valenced experience, flexible learning, integrative processing, and relevant neural or behavioral markers, the local phenomenal-power hypothesis becomes more plausible.
But the evidence favors sentience or phenomenal organization before it favors Whitehead specifically.
The Whiteheadian interpretation must still outperform other theories of animal consciousness.
Plants and basal cognition
Here the transport starts to break.
Segundo-Ortin and Calvo review sophisticated adaptive behavior in plants, including endogenous control, electrical and chemical signaling, and forms of sensorimotor integration. They explicitly treat plant consciousness as an open research question rather than a result established by plant cognition.
That distinction is exactly what the transport rule predicts.
Goal-directedness survives.
Context sensitivity survives.
Signaling survives.
Adaptive regulation survives.
Independent evidence of phenomenal valence does not yet survive with comparable strength.
The generic process carrier transports much farther than the credit-bearing phenomenal evidence.
So the local abductive advantage decays.
Computational systems
The same split appears in another substrate.
Dung and Kersten argue that conventional computational systems can satisfy substantial mechanistic and implementational conditions, and they reject several arguments that would rule artificial consciousness out merely because the substrate is silicon. Their result keeps artificial consciousness open. It does not make computational organization sufficient for phenomenality.
Samuel Kimpton-Nye’s 2026 proposal is especially relevant to this programme. He develops a model in which dispositional properties in algorithmic systems can realize categorical phenomenal properties, while those phenomenal properties provide identity conditions for the dispositions. That is a serious attempt to carry a phenomenal-powers style bridge into computational systems.
It is a candidate bridge, not evidence that current systems are conscious.
The important point is that substrate does not decide the question. The bridge conditions do.
Fundamental physical actuality
At this scale almost all of the local phenomenal discriminators disappear.
There is no independent evidence of painfulness, pleasure, phenomenal unity, or felt valence in an electron or a fundamental physical interaction.
Physics supplies extraordinarily constrained structural, dynamical, and dispositional descriptions. That strengthens confidence in the physical carrier and in naturalistic inquiry. It does not by itself tell us whether the intrinsic nature of whatever realizes those structures is experiential, non-experiential, or badly captured by that contrast.
Whitehead still has an abductive case here, but its character has changed.
The support now comes mainly from ontological continuity, anti-bifurcation, and the promise of giving causal reception a positive intrinsic interpretation. Rival powers and Russellian views compete on the same terrain.
That is a weaker evidential position than the human affective case because the original credit-bearing phenomenal evidence is no longer independently present. At the same time, the physical description is not starting from zero; it arrives with a much larger empirical record than any claim about universal intrinsic experience.
It may still turn out that Whitehead gives the best metaphysical completion of that record. It may instead be a powers, Russellian, structural, emergent, or exhaustively physical completion.
The decision has to come from what those stronger interpretations add beyond the already successful carrier, not from pretending that the human Bayes factor followed the word causation all the way down or that successful physical description already revealed intrinsic nature.
What the stronger result changes
D plus unresolved X should not be treated as a positive ontology competing with Whitehead. It is a holding structure that makes evidential allocation visible. It also does not mean that every completion begins with an equal prior claim on our confidence.
The shared physical carrier has the strongest direct empirical warrant in the consciousness cases we actually know. Physical interventions repeatedly alter consciousness. Neural organization tracks loss, preservation, and recovery of conscious states. Competing neuroscientific theories disagree over which mechanisms are decisive, but they are arguing inside a physically realized causal system.
That supports physical realization very strongly.
It supports exhaustive physicalism less directly.
Among positive interpretations of the unresolved remainder, phenomenal-powers views now deserve more attention than a generic panpsychism label suggests because they possess a concrete abductive argument. Whiteheadian prehension also moves upward as a unification hypothesis because it generalizes a form of experiential causal efficacy that has at least one plausible local success.
REI moves in the opposite direction if anyone tries to use it as an ontology. Its value is methodological. It exposes dependency structure and allows rival metaphysical interpretations to be compared without pretending that equations settle intrinsic nature.
The universal Whiteheadian scope claim remains the expensive part. So does the exhaustive physicalist claim that successful physical realization exhausts what the realizers are intrinsically.
This leaves an asymmetrical but genuinely open landscape:
physical realization has extensive empirical credit; phenomenal causation has a serious local positive residual; universal prehension has a nontrivial unification argument; exhaustive intrinsic physicalism remains stronger than the physical-dependence evidence itself.
That is a much more informative position than treating physicalism and panexperientialism as if the same quantity and type of evidence supported each.
The central comparison is now narrower. What additional explanatory work is performed by the intrinsic completion once the physically realized carrier has already been paid for?
A research programme rather than a verdict
The programme can be stated without deciding the ontology in advance.
The empirical side should look more like adversarial theory testing than a metaphysical beauty contest. The 2025 COGITATE collaboration forced Integrated Information Theory and Global Neuronal Workspace Theory into preregistered differential predictions and common multimodal tests. The point here is not that either theory settles the metaphysics. It is the procedure: identify what rivals share, force the disagreement into claims that could separate them, and decide in advance what outcomes would count against each side.
The metaphysical side cannot always be converted into a laboratory intervention, and Ben’s history of Principia is a useful warning against pretending otherwise. Comparative abduction can legitimately count coherence, unification, premise economy, explanatory reach, and whether an equally plausible rival recovers the same achievements. The discipline I want is not “experiment or silence.” It is that each kind of warrant be named, compared against serious alternatives, and prevented from travelling farther than the feature that generated it.
Robustness has a different job. Weisberg’s robustness analysis asks which results survive across independently constructed models. Storm and colleagues’ multiscale comparison of consciousness theories likewise shows that apparently competing theories may sometimes capture different mechanistic levels or share underlying neural commitments. What survives those changes belongs more securely to D. What discriminates among the surviving completions earns the extra ontological credit.
Begin with a domain where the experiential interpretation has an independently identifiable target and ask whether it produces a positive residual over serious rivals. If it does, identify exactly which feature produced the advantage. Separate the feature from the vocabulary used to describe it. Then move to a new boundary, scale, or substrate and ask whether the same credit-bearing condition is independently present, transformed into something genuinely analogous, or absent.
If the condition survives and the experiential interpretation continues to outperform its rivals, the warrant expands.
If the condition changes, the hypothesis has to change with it.
If only generic causal or organizational structure survives while the credit-bearing evidence disappears, the original abductive advantage does not come along automatically.
Whitehead’s wager can therefore gain support incrementally rather than all at once.
It can also fail incrementally.
That is a better situation than a metaphysical standoff because both sides now have loss conditions.
REI loses its claimed neutrality if its supposedly thin variables cannot be specified without presupposing prehension.
A pure-powers rival loses ground if it repeatedly reconstructs the same qualitative-experiential structure while insisting that the resemblance is metaphysically irrelevant.
An emergentist loses ground if the supposedly emergent feature turns out to be required to make lower-level causal power intelligible.
Whitehead loses universal scope if the feature that earned experiential credit locally disappears across boundaries while the ontology continues to travel anyway.
The strongest result I can defend at this stage is therefore not universal panexperientialism, and it is not a return to metaphysical zero.
We know quite a lot about what evidenced consciousness depends on. The dependence is physically realized, experimentally perturbable, and increasingly characterizable across scales. We know much less about what that successful causal organization is intrinsically.
Experiential character has at least one serious candidate positive residual in the causal and motivational structure of affective experience. Whiteheadian prehension gains abductive support insofar as it offers a unified account in which that sort of efficacy is not metaphysically isolated. The remaining work is to determine whether the feature that earned the credit locally is a special achievement of certain organized systems or a disclosure of causal actuality as such.
Same rule, different evidence. The physical carrier has already earned a great deal. Experiential causation has now earned something locally. Universal prehension still has to pay for the distance between those facts.
That is the research programme.
Find where prehension earns comparative leverage.
Identify exactly what earned it.
Then carry that across boundaries, scales, and substrates until the advantage survives, changes, or disappears.
The ontology should travel only as far as the credit-bearing warrant does.
References and source check
Whitehead and reference works
Alfred North Whitehead. Process and Reality, corrected edition, ed. David Ray Griffin and Donald W. Sherburne (Free Press, 1978). Public full-text scan checked for the passages on process, consciousness, prehension, and experiential togetherness: https://samim.io/dl/Process_and_Reality_A_N_Whitehead.pdf Whitehead Research Project’s suggested-editions page confirms the corrected 1978 edition as the standard citation edition: https://whiteheadresearch.org/research/contemporary-whitehead-studies/proposals/suggested-editions/
Ronald Desmet and Andrew David Irvine. “Alfred North Whitehead.” Stanford Encyclopedia of Philosophy. Full text: https://plato.stanford.edu/entries/whitehead/
Philip Goff, William Seager, and Sean Allen-Hermanson. “Panpsychism.” Stanford Encyclopedia of Philosophy. Full text: https://plato.stanford.edu/entries/panpsychism/
“Panpsychism.” Internet Encyclopedia of Philosophy. Full text: https://iep.utm.edu/panpsych/
Johanna Seibt. “Process Philosophy.” Stanford Encyclopedia of Philosophy. Full text: https://plato.stanford.edu/entries/process-philosophy/
Ben Snyder and the logicism exchange
Ben Snyder. “Whitehead’s Logicism.” Subnectorized, September 21, 2026. Full text:
Alfred North Whitehead and Bertrand Russell. Principia Mathematica, vol. 1 (Cambridge University Press, 1910), Introduction, “Reasons for Accepting the Axiom of Reducibility.” Public full text: https://www.gutenberg.org/files/78050/78050-src/78050-src.htm The Stanford Encyclopedia’s entry on Principia Mathematica independently reproduces and discusses the same inductive justification of the axiom.
Matthew David Segall
Matthew David Segall. “Whitehead’s Adventure in Rationality: Between Logic and Aesthetics.” Footnotes²Plato, September 21, 2026. The article discussed here. Publication homepage:
Matthew David Segall. “Experience Is Not a Predicate.” Footnotes²Plato, August 29, 2026. Full text checked directly:
Matthew David Segall. “Against Physicalism’s Misplaced Concreteness.” Footnotes²Plato, May 10, 2026. Full text checked directly:
Matthew T. Segall. “The Varieties of Physicalist Ontology: A Study in Whitehead’s Process-Relational Alternative.” Philosophy, Theology and the Sciences 7, no. 1 (2020): 105-131. Author-hosted full PDF: https://footnotes2plato.com/wp-content/uploads/2020/11/segall_ptsc_7_1_105-131.pdf DOI: https://doi.org/10.1628/ptsc-2020-0008
Matthew David Segall. “Physics Within the Bounds of Feeling Alone.” World Futures 80, no. 8 (2024): 593-614. Version-of-record DOI: https://doi.org/10.1080/02604027.2024.2406182 Author-hosted conference draft: https://footnotes2plato.com/wp-content/uploads/2023/02/physics-within-the-bounds-of-feeling-alone-13.pdf
Phenomenal powers, causation, and rival completions
Hedda Hassel Mørch. “The Evolutionary Argument for Phenomenal Powers.” Philosophical Perspectives 31 (2017): 293-316; first published online 2018. Full text at Wiley: https://onlinelibrary.wiley.com/doi/pdf/10.1111/phpe.12096 DOI: https://doi.org/10.1111/phpe.12096
Hedda Hassel Mørch. “Does Dispositionalism Entail Panpsychism?” Topoi 39 (2020): 1073-1088. Author manuscript: https://philpapers.org/archive/MRCDDE-2.pdf DOI: https://doi.org/10.1007/s11245-018-9604-y
Hedda Hassel Mørch. “The Argument for Panpsychism from Experience of Causation.” In The Routledge Handbook of Panpsychism, ed. William Seager (2019). Author’s penultimate draft: https://philpapers.org/archive/MRCTAF.pdf Publication record: https://philpapers.org/rec/MRCTAF
Brendan O’Sullivan and Robert Schroer. “Painful Reasons: Representationalism as a Theory of Pain.” The Philosophical Quarterly 62, no. 249 (2012): 737-758. DOI: https://doi.org/10.1111/j.1467-9213.2012.00092.x
Declan Smithies and Jeremy Weiss. “Affective Experience, Desire, and Reasons for Action.” Analytic Philosophy 60, no. 1 (2019): 27-54. Author manuscript: https://philpapers.org/archive/SMIAED.pdf Version of record: https://onlinelibrary.wiley.com/doi/full/10.1111/phib.12144 DOI: https://doi.org/10.1111/phib.12144
Stephen Mumford and Rani Lill Anjum. “Spoils to the Vector: How to Model Causes if You Are a Realist About Powers.” The Monist 94, no. 1 (2011): 54-80. DOI: https://doi.org/10.5840/monist20119414
Torin Alter and Sam Coleman. “Russellian Monism and Mental Causation.” Noûs 55, no. 2 (2021): 409-425; first published online 2019. Author-hosted manuscript: https://uhra.herts.ac.uk/id/eprint/6992/ DOI: https://doi.org/10.1111/nous.12318
Nicholas Osborn. “Panpsychism and the Metaphysics of Properties.” Asian Journal of Philosophy 5, article 21 (2026). Open full text: https://link.springer.com/article/10.1007/s44204-026-00381-4
Luca Dondoni. “Towards a Unitary Case for Russellian Panpsychism.” Philosophia 50 (2022): 43-64. Open version of record: https://link.springer.com/article/10.1007/s11406-021-00363-1 Author PDF: https://philpapers.org/archive/DONTAU-3.pdf
Luoying Zhong. “An Emergentist Approach to Phenomenal Causality.” Pacific Philosophical Quarterly 107, no. 2 (2026): 74-85. DOI: https://doi.org/10.1111/papq.70014
Transport across animals, plants, and computational systems
Jonathan Birch, Alexandra K. Schnell, and Nicola S. Clayton. “Dimensions of Animal Consciousness.” Trends in Cognitive Sciences 24, no. 10 (2020): 789-801. Open full text: https://pmc.ncbi.nlm.nih.gov/articles/PMC7116194/ DOI: https://doi.org/10.1016/j.tics.2020.07.007
Heather Browning and Jonathan Birch. “Animal Sentience.” Philosophy Compass 17, no. 5 (2022): e12822. Open version of record PDF: https://eprints.soton.ac.uk/475389/1/Philosophy_Compass_2022_Browning_Animal_sentience.pdf DOI: https://doi.org/10.1111/phc3.12822
Miguel Segundo-Ortin and Paco Calvo. “Consciousness and Cognition in Plants.” WIREs Cognitive Science 13, no. 2 (2022): e1578; first published online 2021. Open full text: https://wires.onlinelibrary.wiley.com/doi/full/10.1002/wcs.1578 DOI: https://doi.org/10.1002/wcs.1578
Leonard Dung and Luke Kersten. “Implementing Artificial Consciousness.” Mind & Language 40, no. 3 (2025): 285-305; first published online 2024. Open version of record: https://onlinelibrary.wiley.com/doi/10.1111/mila.12532 Author archive: https://philarchive.org/versions/DUNIAC-3 DOI: https://doi.org/10.1111/mila.12532
Samuel Kimpton-Nye. “Algorithmic AI Consciousness.” Philosophy and Phenomenological Research, Early View (2026). Open version of record: https://onlinelibrary.wiley.com/doi/10.1111/phpr.70155 DOI: https://doi.org/10.1111/phpr.70155
Physical realization, consciousness science, and comparative method
Daniel Stoljar. “Physicalism.” Stanford Encyclopedia of Philosophy, substantive revision September 16, 2026. Full text: https://plato.stanford.edu/entries/physicalism/
Mario Rosanova et al. “Recovery of Cortical Effective Connectivity and Recovery of Consciousness in Vegetative Patients.” Brain 135, no. 4 (2012): 1308-1320. Open full text: https://pmc.ncbi.nlm.nih.gov/articles/PMC3326248/ DOI: https://doi.org/10.1093/brain/awr340
Adenauer G. Casali et al. “A Theoretically Based Index of Consciousness Independent of Sensory Processing and Behavior.” Science Translational Medicine 5, no. 198 (2013): 198ra105. Public repository record with downloadable publisher postprint: https://orbi.uliege.be/handle/2268/171542 DOI: https://doi.org/10.1126/scitranslmed.3006294
Simone Sarasso et al. “Consciousness and Complexity during Unresponsiveness Induced by Propofol, Xenon, and Ketamine.” Current Biology 25, no. 23 (2015): 3099-3105. PubMed record with free-full-text links: https://pubmed.ncbi.nlm.nih.gov/26752078/ DOI: https://doi.org/10.1016/j.cub.2015.10.014
George A. Mashour. “Anesthesia and the Neurobiology of Consciousness.” Neuron 112, no. 10 (2024): 1553-1567. Open full text: https://pmc.ncbi.nlm.nih.gov/articles/PMC11098701/ DOI: https://doi.org/10.1016/j.neuron.2024.03.002
Johan F. Storm et al. “An Integrative, Multiscale View on Neural Theories of Consciousness.” Neuron 112, no. 10 (2024): 1531-1552. Open article: https://www.sciencedirect.com/science/article/pii/S0896627324000886 DOI: https://doi.org/10.1016/j.neuron.2024.02.004
COGITATE Consortium et al. “Adversarial Testing of Global Neuronal Workspace and Integrated Information Theories of Consciousness.” Nature 642 (2025): 133-142. Open full text: https://www.nature.com/articles/s41586-025-08888-1 DOI: https://doi.org/10.1038/s41586-025-08888-1
Michael Weisberg. “Robustness Analysis.” Philosophy of Science 73, no. 5 (2006): 730-742. DOI: https://doi.org/10.1086/518628
Rasmus Jaksland. “Naturalized Metaphysics or Displacing Metaphysicians to Save Metaphysics.” Synthese 201 (2023): article 199. Open full text: https://link.springer.com/article/10.1007/s11229-023-04207-1
