I Wanted Erik Svensson’s Argument for Physicalism to Be Stronger. So I Tried to Build One Myself.
What Erik Svensson, Massimo Pigliucci, and Matthew David Segall reveal about the evidence for physicalism, the residual case for panexperientialism, and what each would have to do to progress.
Substack Repost Notice: This is an archival repost of my substack post. Some formatting may not display correctly on our website. See the original post here: https://sweetlyrational.substack.com/p/i-wanted-erik-svenssons-argument
Update, August 28, 2026: has now responded to this essay in Experience Is Not a Predicate. The problem is he did not quote my actual question; his paraphrase changed its logical form from a two-branch IF / may comparison into a positive commitment to primitive nonexperience, changed the epistemic state from suspension of judgment to metaphysical negation, changed the comparison class from a strong process-relational rival to a “wholly feelingless” one, and thereby changed the burden of proof by assigning me an abrupt-emergence problem my original argument did not incur. I appreciate that he engaged it, but after comparing his response with my actual wording, I think much of the apparent disagreement is produced by that change in the target and by a further semantic shift: my ordinary comparative use of experiential predicate is reinterpreted through Whitehead’s theory-internal sense of predication, despite my having already rejected the positions his reconstruction then attributes to the rival. Once the large amount of shared ground is restored, the dispute contracts to the question I was originally trying to isolate: What evidence or explanatory difference warrants the final move from shared causal inheritance, receptivity, integration, novelty, organization, and real influence to the universal claim that primitive causal actuality itself is intrinsically experiential, rather than leaving that intrinsic character open?
I respond in detail, quote the relevant passages, reconstruct what seems to me to be Matt’s strongest substantive argument, and explain where I think the real disagreement remains here: Matthew Segall’s “Experience Is Not a Predicate”: I Never Made the Argument He Rebutted. My Question Remains Unanswered.
With that out of the way, let’s get physical(ist)!
A sympathetic critique of Erik Svensson, a three-way comparison with Massimo Pigliucci and Matthew David Segall, and a recursive test that none of us gets to escape
I went into Erik Svensson’s () recent essay on Tim Maudlin, Kant, German Idealism, materialism, and consciousness expecting to agree with a great deal of it. I still do. That is why the parts that fail bother me more than they would in an argument whose basic commitments I rejected from the start. The essay frustrated me for much the same reason some of Massimo Pigliucci’s recent arguments for physicalism frustrate me: I am a huge admirer of both men’s work, I regard the empirical research programmes they defend and help articulate as invaluable, and I think the physicalist side of this debate begins with an exceptionally strong empirical hand. That is exactly why I dislike seeing that hand weakened by commitments it does not need and that often mirror the mistakes of the frameworks they critique. I’d like to help them see why those commitments are entirely unnecessary and indeed could be harmful to their goals and state what I think the stronger argument is, with an open invitation to knock it down.
Erik and I share several instincts that matter here. Reality does not bend itself around our preferred stories. The natural sciences have earned extraordinary authority where their methods produce risky predictions, reproducible interventions, and explanations that keep working when researchers change. Human consciousness, in every high-confidence case we can investigate directly, depends profoundly on organized biological activity. I know of no reproducible evidence that requires an independently acting immaterial soul, supernatural intervention, or a universal field of consciousness. I also share Erik’s impatience with theories that use difficult language to make their commitments harder to expose.
Those are substantial points of agreement. They make physicalist and naturalistic research programmes more credible to me than many of their rivals. They also make me want the case for them to be unusually careful.
I do not think that Erik’s essay is careful enough.
As our agreements show, the strongest case is already substantial. Every high-confidence case of animal consciousness available to us is physically realized. Anesthesia, lesions, stimulation, pharmacology, development, degeneration, sleep, and injury show that changes to organized biological activity systematically change conscious capacities. Across biology more generally, physically continuous explanations have repeatedly displaced appeals to special causal substances while producing risky predictions and effective interventions. Those facts should strongly affect our priors.
But they do not arrive bundled with every stronger proposition sometimes placed under physicalism. Erik makes the case weaker when neural dependence in known organisms becomes universal nervous-system necessity, when the failure of a particular panpsychist model is allowed to travel across an entire philosophical family, or when evidence favoring physical explanation over interventionist dualism is asked to establish exhaustive material ontology. Some of those additions are not entailed by every serious formulation of physicalism, and the empirical evidence certainly does not establish them as one package. Each unnecessary commitment enlarges the theory’s failure surface while doing little to strengthen the part of the case that was already excellent.
That is also what bothers me in Massimo’s strongest metaphysical formulations. His broader epistemology is fallibilist, probabilistic, and deeply attentive to science, which is part of why I value his work. Yet statements such as “matter and energy are all there is” can make the endpoint sound more settled than the inferential path warrants. Physical realization, causal continuity, supervenience, reduction, absence of fundamental mentality, and exhaustive physical ontology are related claims, but they are not one claim. The physicalist case becomes stronger when its best-established empirical successes are allowed to support exactly what they discriminate, while stronger metaphysical commitments remain separately exposed.
Matthew David Segall matters to me for the mirror-image reason. I do not want a failure to establish primitive experience, affectivity, or proto-subjectivity to erase whatever his process-relational framework gets right about causal inheritance, embodiment, relational constitution, emergent subjects, organization, and the incompleteness of simply redescribing experience away. But those surviving insights do not automatically establish the distinctively panexperiential addition either. I want Matt’s framework treated the way I want Erik’s and Massimo’s treated: preserve the strongest independently supported core, isolate what is genuinely additional, and make that residual earn the extra ontology attributed to it.
That is the frustration behind this essay. The aim is to reconstruct the strongest versions of all three programmes because weak arguments obscure the real disagreement. Once their shared commitments are separated from their distinctive additions, several apparent conflicts disappear, several familiar objections miss their targets, and the remaining questions become both harder and more useful. Physicalism can keep the enormous evidential advantage of physical realization without inheriting exhaustive ontology for free; Matt’s process-relational programme can keep what survives without primitive experience inheriting evidence from the shared carrier.
The ground I am willing to defend with Erik
Before getting to the failures, I want to protect the part of Erik’s argument that deserves protection.
The empirical dependence of ordinary human consciousness on brains is not a philosophical rumor. Anesthesia alters or abolishes conscious states. Lesions can eliminate specific capacities or produce profound changes in awareness. Stimulation and pharmacological interventions alter experience in systematic ways. Sleep, seizures, neurodegeneration, development, and injury all supply converging evidence that the forms of consciousness we know best depend on physical organization in living nervous systems. Whatever the final metaphysics turns out to be, a theory that ignores this dependence begins in a very bad place.
The same is true more broadly of biological explanation. Modern biology has repeatedly replaced appeals to special vital forces with mechanisms continuous with chemistry and physics. Evolutionary theory, genetics, molecular biology, physiology, and neuroscience have generated an enormous amount of explanatory and interventional success without needing a second causal substance. Erik is right to treat that history as evidence. It should affect our priors.
It affects mine.
If the comparison is between ordinary physical causal explanation and a Cartesian mind that independently pushes neurons around, I assign the physical account a large advantage. If a supernatural agent is alleged to intervene in ways that should alter publicly measurable outcomes, repeated failure to find the predicted residue should count against the intervention claim. If somebody says electrons have primitive experience, I do not think the mere possibility obliges anyone to believe it. The positive burden belongs with the addition.
There is also a valuable critical-rationalist instinct in Erik’s suspicion of explanatory vocabulary that cannot lose. A theory that accommodates every outcome after the fact is not strengthened by its flexibility. A term that sounds profound but never changes prediction, intervention, explanation, or comparison may be doing less than its advocates think.
All of that survives my criticism.
What does not survive is the move from this strong empirical core to a universal metaphysical conclusion without paying for the extra scope.
Where the argument starts to lose its target
A critique becomes much easier when the opponent has been simplified enough. It also becomes much less informative.
Kant is a good example. Erik presents transcendental idealism in a way that repeatedly approaches ordinary subjectivism: the mind constructs the world we know, objective knowledge is endangered, and the external world recedes behind subjective impressions. There are serious objections to Kant, including difficult questions about the thing in itself, causation, space and time, and the status of transcendental arguments. Yet Kant explicitly paired transcendental idealism with empirical realism, fought the charge that he had merely recreated Berkeley, and added the “Refutation of Idealism” to the second edition of the Critique of Pure Reason. The interpretive literature is notoriously divided, which is already enough to make a simple equation of Kant with external-world subjectivism unsafe.
Paul Abela’s Kant’s Empirical Realism places the realist side of Kant’s project at the center, while Dietmar Heidemann argues that Kant’s distinctions among forms of realism remain relevant to contemporary debates. The current Stanford entry on idealism likewise emphasizes Kant’s attempt to combine transcendental idealism with empirical and ontological realism.
None of that rescues Kant from criticism, of course.
If Erik wants to show that Kant’s distinction between appearances and things in themselves fails, he should attack that distinction. If he thinks Kant’s account of objectivity cannot support the kind of realism science needs, he should show where the support breaks. If he thinks later idealists made bad use of Kant, that is a historical claim that can be traced. Collapsing Kant into “we only know subjective impressions” gives Erik a cleaner enemy, but not Kant’s strongest position. When that weaker reconstruction is then used to make the real position look easier to defeat, the familiar name is justified: it is a straw man, not merely an abbreviated history.
Something similar happens with Bohr. There are good realist objections to Copenhagen-style interpretations of quantum mechanics, and Maudlin has pressed some of them forcefully. Yet the historical category “Copenhagen interpretation” contains several different views, and specialist scholarship no longer supports the easy picture of Bohr as a subjectivist who thought consciousness creates measurement outcomes. The Stanford Encyclopedia’s current treatment is unusually blunt on this point: Bohr rejected the claim that the observer’s consciousness directly determines the result, and contemporary scholarship generally does not classify his complementarity view as straightforward positivism or subjectivism.
Again, the better target remains available. One can argue about completeness, the classical-quantum cut, the meaning of the wavefunction, contextuality, the measurement problem, and what Bohr thought an objective physical description could amount to. Those are more difficult arguments. That is a reason to make them, not to replace them with a mystical cardboard cutout.
Panpsychism receives the same treatment. Erik sometimes writes as if panpsychism were the single claim that electrons and rocks are conscious in a dimmer version of the way we are. Some theories really do approach micropsychism of that kind. The contemporary family is broader. It includes constitutive and emergent versions, panprotopsychism, cosmopsychism, and Russellian approaches with importantly different commitments. Historically, panpsychist ideas also predate Kant by millennia. Spinoza and Leibniz precede him; early Greek discussions precede all three. Kant influences one later branch through figures such as Schopenhauer, but that is not a straight origin story.
Erik’s own complaint about “Neo-Darwinism” now comes back with unusual force. He is right that modern evolutionary biology should not be turned into a pejorative trash can containing whatever a critic dislikes about genes, selection, reduction, competition, or materialism. A heterogeneous research field deserves decomposition. So does panpsychism. So does idealism. So does physicalism.
In his writing on clear thinking and the Vienna Circle, he recommends asking what a claim means and how it could be tested. In his defense of evolutionary biology against the pejorative label “Neo-Darwinism,” he asks critics to understand a pluralistic field before attacking a caricature of it. Those are good rules. I would add one step: apply them recursively. Use the standard on the opponent, then turn it back on your own conclusion and on the method doing the evaluation. If the standard changes when it reaches home, something has gone wrong.
Applied directly to Erik’s own essay, that recursive step changes the shape of the argument. “Consciousness requires nervous systems” is not merely a report about every confirmed animal case; it is a necessity claim, so it needs a specification of what property of nervous systems is indispensable, what alternative realizations are being excluded, and what evidence could show the necessity claim wrong. Calling the hard problem a category mistake likewise requires identifying the category error precisely and showing what explanatory work disappears once it is corrected. And if materialism is meant as more than the empirical claim that known conscious organisms are physically realized, its additional content needs a loss condition of its own: what conceivable finding would lower confidence in exhaustive materialism while leaving the success of physical science intact?
His anti-caricature rule has the same consequence. Physicalism, idealism, and panpsychism are families of importantly different positions, not single hypotheses that inherit every commitment made by one representative. A failed panpsychist model therefore bears first on that model and whatever depends on it; a weak version of idealism does not stand in for Kant’s strongest position; and physicalism itself should not be made to carry neural essentialism, reductive identity, or exhaustive intrinsic materiality unless those commitments are actually part of the version being defended.
The implication is a narrower but stronger case. Erik’s own standards preserve the empirical dependence of known consciousness on physical organization, the extraordinary success of physically continuous explanation, and the resulting evidential advantage over views that require independently acting nonphysical causes. They do not by themselves establish universal nervous-system necessity, defeat every panexperiential or nonreductive rival, or turn the success of physical explanation into exhaustive ontology. Applying his rules recursively therefore removes commitments that enlarge the burden without improving the empirical case. The result is not a retreat from physicalism; it is a more exposed and potentially more progressive physicalist research programme.
That recursive turn, I think, makes the physicalist case stronger, makes Matt’s strongest challenge harder to dismiss, and gives all three programmes a clearer way to progress.
The mirror is uncomfortable because the methodological principles are Erik’s own.
The larger inferential problem
Historical simplification is only the surface problem. Deeper down, much of Erik’s evidence fails to discriminate the metaphysical conclusion he wants from several serious rivals.
Take the prehuman Earth. Of course Earth existed before human beings. That fact is devastating for the cartoon view that human belief literally produces stars, oceans, and fossils. It is almost useless for distinguishing physicalism from neutral monism, Russellian monism, many forms of objective idealism, or panexperientialism. All of those positions can affirm a world older than human observers.
If the reasoning is “physicalism predicts a prehuman world; we observe overwhelming evidence of a prehuman world; therefore physicalism,” the form approaches affirming the consequent. More importantly, even if we avoid forcing the argument into a textbook schema, the Bayesian problem remains: the evidence was expected under several rivals. Its likelihood ratio for strict physicalism over those rivals may be close to one.
The same happens with neural dependence. A sophisticated Whiteheadian such as Matthew David Segall can agree that reflective consciousness, representation, intentional perception, and the forms of animal awareness familiar to us depend on highly organized living bodies. A Russellian monist can agree. A property dualist who accepts physical causal closure can agree with much of the neuroscience. Lesions, anesthesia, and stimulation therefore provide very strong evidence against some rivals while doing much less against others.
Here the physicalist side deserves to be stated as strongly as the evidence allows. In every confirmed conscious organism we can investigate, consciousness is physically realized in an organism whose relevant structures and dynamics can be measured, perturbed, damaged, suppressed, and restored with corresponding changes in conscious capacities. That is not a speculative metaphysical posit. It is an empirical starting fact that every adequate rival has to accommodate. Views that deny physical realization, or require an independently acting nonphysical cause that should alter the physical story, therefore face a much steeper evidential burden than views that preserve the entire physical realization story and disagree only about what that realized structure is intrinsically.
Evidence has to be credited locally.
This is one place where a recursively self-critical rationalism changes the conversation. Instead of asking whether evidence is compatible with my theory, I ask which commitment generated a differential expectation. If two positions predict the same observation, that observation cannot be counted again as though it uniquely supported the feature on which they disagree.
Erik’s empirical case therefore establishes a great deal. Physical realization is not merely a correlation we happen to have noticed in a few cases. For the conscious systems we know best, physically characterized organization is interventionally consequential: change the relevant neural dynamics and consciousness, perception, memory, wakefulness, or report can change with it. No independently acting immaterial contribution has earned comparable support. This gives physically grounded naturalistic explanations a genuine evidential advantage, especially over theories that require extra causal influence outside the physical account.
“Reality is exhaustively physical” adds more. Physical realization tells us how confirmed conscious systems are implemented and causally constrained. It does not, by itself, tell us that the physical description exhausts their intrinsic nature, that every possible conscious system must share the same biological implementation, or that every ontology preserving the physical facts has thereby been excluded.
The addition may be true. The evidence above does not by itself establish it.
That distinction matters because “physical” is not philosophically trivial. The familiar Hempel dilemma still bites: define physicalism by present physics and the ontology inherits the incompleteness of present physics; define it by an ideal future physics and it becomes harder to specify what would count as nonphysical. There are responses to the dilemma, but the existence of the problem is enough to show why “physicalism is simply what science says” is too quick.
External realism, methodological naturalism, physical realization, causal closure, reductionism, and exhaustive metaphysical physicalism should not be treated as synonyms. A person can accept several and withhold another. I do.
Once those terms slide together, several familiar errors become easier to miss. External realism can be smuggled into the premise and exhaustive materialism returned as though it had been established, which risks begging the question. Presenting scientific seriousness as a choice between materialism and a single opposing bundle of idealism, dualism, panpsychism, and mysticism creates a false dilemma. The live space contains neutral and Russellian monisms, nonreductive physicalisms, structural realism, process views, and combinations that preserve the empirical sciences while disagreeing about intrinsic ontology.
A mirror I had already encountered with Massimo Pigliucci
Erik’s move felt familiar because I had already run into a closely related problem inmy exchange with , another thinker whose commitment to scientific methodological naturalism I largely share. Pigliucci has answered the question of what counts as physical with “whatever fundamental physics says,” and more recently states the stronger materialist thesis plainly: “matter and energy are all there is.” He also makes a useful distinction that helps locate the issue, noting that the philosophical versions of materialism and reductionism make stronger claims about how the world is than the methodological strategies used in science.
When I asked Massimo for a discriminating piece of evidence or prediction that would distinguish those stronger physicalist commitments from my lower-commitment boundary-and scale-relative framework, which accepts the same physical-realization evidence while offering its own grounding account of consciousness, he did not provide one and declined to pursue the comparison. That leaves the empirical common core intact while leaving the additional ontology to earn its own discrimination.
That is almost exactly where my agreement splits. The physical realization of known consciousness is extraordinarily well established. The success of explanations continuous with physics, chemistry, and biology should raise confidence in physicalist research programmes. Pigliucci is also right that the absence of any demonstrated independent nonphysical causal contribution matters. What requires an additional step is the promotion from those achievements to the universal conclusion that matter-energy exhausts what exists, or that whatever fundamental physics eventually says automatically settles every remaining ontological question. A successful research strategy can support a metaphysical programme without becoming identical to its strongest ontology.
Pigliucci’s May challenge, “Got a better, more complete theory?”, points toward inference to the best explanation, which is a legitimate form of comparative reasoning. Its conclusion is probabilistic. The best current programme can deserve the highest credence while remaining revisable. Lack of a better rival today does not convert its most ambitious universal predicates into observations. Erik’s argument often makes the same jump when the failure of idealism, dualism, vitalism, or one panpsychist model is allowed to promote physicalism from the strongest current programme to the only serious ontology in town.
The formal danger is familiar. If physicalism predicts E and E occurs, concluding physicalism deductively would affirm the consequent unless the alternatives have already been excluded. Probabilistically, however, E can genuinely favor physicalism when E is more expected under physicalism than under its rivals. That distinction strengthens the argument instead of flattening it into either proof or agnosticism.
The Bayesian version is cleaner. Let E be the robust physical realization and intervention dependence of known consciousness. Against an interactionist dualism that posits an independently acting nonphysical cause, E can strongly favor a physically grounded account, especially when the rival gives us reason to expect causal departures that never appear. Against Segall’s panexperientialism or a Russellian monism that preserves the same physical realization and causal dynamics, E may have almost the same likelihood under both theories. It then contributes little to their intrinsic-ontology dispute. A failed rival can still raise physicalism’s posterior probability by redistributing probability mass, but the size of that update depends on which hypothesis failed, why it failed, and how many serious alternatives remain. In an open hypothesis space, eliminating Y cannot by itself make X ontologically exhaustive. The failure of Y supplies evidence for X only to the extent that the failed result was differentially expected and the remaining alternatives have actually been accounted for.
This is why underdetermination of physicalism is a limit on confidence rather than a refutation. The literature on physicalism contains serious live disputes about how the thesis should be formulated, from Hempel’s dilemma to competing accounts of causal closure, grounding, supervenience, and the causal argument. Alyssa Ney, Justin Tiehen, David Papineau, Bradford Saad, and others have developed sophisticated versions of the physicalist case rather than treating physicalism as something science has already defined for us. The current Stanford Encyclopedia of Philosophy entry on physicalism even warns that a strongly physicalist intellectual culture can make us accept lower standards for arguments favoring the view than for arguments against it. That work strengthens the programme by exposing its premises. It also shows why the scientifically strongest form of the argument is comparative rather than absolute.
The evidential asymmetry with many anti-physicalist programmes remains important. Interactionist dualism faces direct pressure from the success and apparent causal sufficiency of physically specified mechanisms. Panexperientialism can preserve that physical story and therefore avoids much of that pressure, but its universal experiential addition receives little support from evidence equally expected under a non-panexperiential rival. Idealist, supernatural, or other mind-first programmes inherit still different burdens depending on what they predict. Strict physicalism also owes an account of its own universal residual, yet it begins from an empirical carrier that every viable rival must preserve. The rational result can therefore be a higher probability for physicalist programmes without a declaration that the metaphysical contest is over.
A stronger version of Erik’s and Pigliucci’s case would say this plainly: every confirmed conscious system is physically realized; interventions show that the relevant physical organization is causally indispensable to the conscious capacities we know; no independent nonphysical causal contribution has earned comparable support; and physically grounded explanations have the best demonstrated track record. Those facts justify a strong starting prior for physically continuous explanations in new cases and high current credence in physicalist research programmes. Exhaustive physicalism can then remain the leading metaphysical hypothesis to the degree that it continues to outperform serious rivals, rather than receiving universality as a reward for their failures.
Where valid skepticism turns into overreach
Hitchens’ razor is useful here only when it cuts symmetrically. An unsupported addition receives no positive evidential credit merely because nobody has disproved it. Trouble begins when withholding X quietly turns into certainty about not-X. At that point a sensible burden-of-proof rule can slide into an argument from ignorance unless the theory gave us reason to expect detectable evidence if X were present.
I have almost no positive reason to attribute phenomenal experience to an electron. That is enough for me not to do it. It does not license certainty that an electron could not instantiate any experiential property under any adequate future ontology. The latter is a stronger modal claim.
The same distinction applies to nervous systems. Every high-confidence biological case of consciousness presently available to us involves organized nervous systems. That should give neural realization a strong prior when we investigate animals. The proposition that all possible consciousness metaphysically requires nervous tissue goes further. A 2025 Philosophy Compass review by Mona-Marie Wandrey and Marta Halina describes substantial disagreement about the evolutionary origin of consciousness and emphasizes that the dispute turns partly on unresolved indicators in living animals. The same review notes that neurons and nervous systems likely arose in the Ediacaran, before the Cambrian boundary Erik sometimes treats as certain.
Two questions have been fused. We do not know exactly when phenomenal consciousness entered animal history, and present evidence does not establish nervous systems as a metaphysical necessity for every conceivable realization. One is historical; the other is modal. Their evidential burdens differ.
The necessity premise has not been earned, and the paleontological boundary is less secure than the rhetoric suggests. Evidence that known consciousness is neurally realized strongly supports neural realization in the cases we have; it does not yet establish that nervous systems are universally necessary, nor does uncertainty about nervous-system origins fix the first appearance of phenomenality. Erik’s conclusion therefore outruns both the modal and historical evidence.
The hard problem raises a related problem. I do not think Chalmers’ formulation proves dualism, panpsychism, or any other anti-physicalist ontology. An explanatory gap is not a certificate of ownership for the metaphysical rival standing nearest to it. Neuroscience, though, has not established that there is no residual problem either.
Seth and Bayne’s 2022 review still describes a field containing several major theories whose relations and empirical discriminability remain unclear. The preregistered Cogitate adversarial collaboration published in Nature in 2025 tested important predictions of global neuronal workspace theory and integrated information theory and challenged central expectations of both. That is productive theory testing. It is difficult to turn those results into evidence that consciousness has already been conceptually dissolved.
Erik is entitled to argue that the hard problem is a category mistake. He then owes the category diagnosis. Which concepts have been confused? What disappears after the correction? What rival survives the same empirical data and why is it worse? Treating a contested philosophical solution as though neuroscience has already delivered the verdict skips the argumentative work.
When criticism becomes social sorting
The rhetoric in Erik’s essays creates another kind of friction. Words such as “crackpot,” “woo,” “mysticism,” and “pseudoscience” sometimes follow a demonstration. Elsewhere they arrive early enough to shape the reader’s picture before the argument has earned them.
That risks poisoning the well.
There is a legitimate place for strong classification. If a purported scientific model contains a central operator with no measurable quantity, offers no way to test its advertised predictions, and is subsequently retracted for those reasons, calling that model scientifically invalid can be impolite and still accurate.
The mistake comes when the failure travels farther than the dependency warrants.
Erik’s discussion of Maria Strömme’s retracted consciousness paper illustrates the problem. The retraction is evidence against that paper and against whatever broader commitments specifically depend on its failed formalism. Erik goes on to call the event a major blow against the scientific credibility of panpsychism generally and predicts that future attempts will probably fail as well. That propagation is much harder to justify. Whiteheadian panexperientialism does not depend on Strömme’s operator. Russellian monism does not depend on it. A failure in one model cannot simply be broadcast across every theory that shares a broad label.
Imagine somebody publishing a bad mathematical model of color polymorphism in evolutionary biology. The model is retracted because its central quantity cannot be measured. A philosopher then announces a major blow against evolutionary explanations of polymorphism and predicts that future attempts will fail too. Erik would rightly demand to know which mechanism was tested, which assumptions failed, and why selection, drift, mutation, recombination, frequency dependence, or other models should inherit the damage.
His scientific instincts are better than the inference he makes here. As an argument form, the move is a hasty generalization compounded by improper failure propagation: a local failure is made to carry a family-wide conclusion without a dependency bridge.
The same issue appears when funding sources, political descendants, or personal associations are used as epistemic shortcuts. Templeton funding can justify scrutiny of incentives and possible conflicts of interest. It cannot make a proposition false. Heidegger’s Nazism is relevant to intellectual history, ethics, and the interpretation of his work. It does not refute an argument about ontology by itself. A centuries-long genealogy from Kant through German Idealism to contemporary political categories may be historically interesting, but the truth of a philosophical proposition does not travel genetically through its descendants.
The genetic fallacy is especially tempting because bad intellectual lineages can be real. History still matters, but historical causation and evidential support need separate ledgers. Once an opponent’s funder, political descendants, or intellectual neighbors begin doing the work of an argument, guilt by association has entered as well. A many-century chain from Kant to contemporary politics also risks historical monocausalism: a branching causal graph is being narrated as if one intellectual lineage carried the explanatory load.
Erik already has the tool he needs
Here recursive self-critical rationalism can improve Erik’s case without asking him to become less physicalist.
Critical rationalism is already self-critical in Popper’s tradition, and William Bartley later pushed the idea toward comprehensive criticism. The recursive extension I find useful is more operational: trace how criticism reaches the specific commitment responsible for an error, then test the procedure that performed the criticism.
In practice, I use a simple mirror before I trust a verdict.
First I reconstruct the strongest version of the argument rather than the easiest target. Then I identify the standards the arguer explicitly invokes and any standards the inference quietly needs. I apply those standards to the conclusion itself, to the strongest rival, and finally to the framework doing the evaluation. If the evaluator cannot survive the same demands, the evaluation is unfinished.
Erik already supplies much of the raw material. He asks what a claim means. He asks how it can be tested. He objects when outsiders compress evolutionary biology into a pejorative label. He thinks science should correct itself when evidence pushes back.
Now turn that machinery inward.
What does “matter is primary” mean strongly enough to exclude something?
What possible result would lower Erik’s confidence in exhaustive physicalism while leaving physical causal success intact?
When neuroscience supports a premise that several metaphysical rivals share, how does he prevent the shared evidence from being counted as uniquely physicalist evidence?
What justifies moving from the actual realization of familiar consciousness in nervous systems to a necessity claim about every possible realization?
If one panpsychist model fails, what dependency map licenses the failure’s propagation to theories that never used the model?
These questions do not weaken the naturalistic case. They strip away argumentative liabilities that opponents can exploit.
Several of the intellectual tools behind this approach are also part of the traditions Erik and Pigliucci themselves take seriously. Erik has described the Vienna Circle as his first intellectual heroes, singles out figures such as Carnap and Neurath with admiration, and says Popper “had a point” in pressing falsification against verification. Neurath’s famous image fits this dispute almost perfectly: “We are like sailors who have to rebuild their ship on the open sea.” Physical science is the ship we are already aboard; revising an ontological plank does not require pretending we can step outside inquiry into a metaphysical dry dock. Pigliucci’s philosophy-of-science work has likewise returned repeatedly to Popper, demarcation, and Bayesian confirmation rather than treating scientific inference as a single deductive verdict.
Hitchens and Flew were foundational to my own intellectual development for closely related reasons, and Pigliucci’s own recent writing makes the connection unusually direct. He explicitly adopts Flew’s burden principle as his own stance, and while criticizing the New Atheists’ tone he says that in substance they “pretty much all got it right.” Hitchens gave me the compact burden rule, “What can be asserted without evidence can also be dismissed without evidence.” Flew’s more useful recursive question was, “What would have to occur … to constitute for you a disproof?” Applied symmetrically, those questions reach panexperientialism, dualism, and theology, then continue straight into physicalism: what would lower confidence in exhaustive physicalism while leaving physical realization intact? A position becomes stronger when it can answer that question before its rivals answer it for us.
Erik could retain a very strong position: physically specified mechanisms have explained every successfully investigated causal process involved in known organisms without a demonstrated nonphysical causal supplement; the burden belongs with anyone proposing an additional causal ingredient; biological priors should be strong in familiar consciousness attribution; and physicalist research programmes currently have a far better empirical record than interventionist dualism or supernatural explanation.
That is already formidable.
He does not need certainty about the intrinsic nature of everything in order to defend it.
Matt Segall makes the comparison harder
Matthew David Segall’s recent clarifications are useful because they remove several targets that Erik often attacks.
Matt’s view is subtler than the image of tiny human-like minds inside electrons. His Whiteheadian account distinguishes reflective consciousness, intentionality, representation, sense perception, and presentational phenomenology from a much thinner primitive notion of prehension. He describes causal efficacy as affective inheritance and treats the subject as something constituted through relational becoming rather than as a substance that first exists and then acquires experiences.
That matters. The standard subject-summing version of the combination problem is aimed most directly at constitutive micropsychism, and Whiteheadian process accounts do not fit that model neatly. Matt can also accept that sophisticated animal consciousness depends on highly organized bodies. His clarification therefore preserves much of the biological story Erik wants.
I also think this is the strongest formulation Matt has given me so far. A recognizable Whiteheadian through-line remains in his longstanding distinction between primitive prehension and reflective consciousness, while the newer formulation sharpens and narrows the commitments exposed to criticism. Earlier formulations had described prehension as having an “experiential quality,” as being “like qualia,” and even as the qualitative “what-it-is-like” of experience; elsewhere he had spoken of awareness, self-enjoyment at every level, pervasive subjectivity, and a presubjective inheritance of affect. Those formulations made it much easier to hear panexperientialism as phenomenality or rudimentary subjectivity projected downward.
The August 21 clarification trims several of those exposed edges. “Non-phenomenal” is now explicitly glossed as non-presentational, non-representational, non-conscious, and non-intentional. Primitive feeling is denied to be a miniature quale added to an otherwise complete causal process. The subject does not pre-exist the relations and then accumulate experiences; it is constituted through their integration. The analogy is proportional rather than univocal, so animal consciousness is allowed to remain a special achievement of highly organized bodies. And where an earlier part of our exchange pushed back against my emphasis on testability and compared speculative philosophy more to poetry than physics, this reply says that nothing proves or exempts prehension from criticism and that “all must be thrown into recursive scrutiny.”
Those are substantive improvements. They retire several weaker routes of criticism: the image of tiny conscious subjects being summed, the inference that ordinary phenomenality must already exist at every scale, the assumption of a substantial subject prior to relations, a simple conflict with the biological realization of animal consciousness, and methodological shelter from criticism. The thinner Whiteheadian position is now much harder to misread, and the earlier phenomenal or subject-like resonance has been trimmed into a cleaner set of commitments. That is genuine progress. A position that becomes harder to caricature and easier to compare with serious rivals has improved.
A critic of Matt therefore has to find the actual residual rather than revive the weaker targets that the clarification has removed.
The residual lies below ordinary animal consciousness: causal becoming, even at primitive levels, is said to be experiential or affective in Whitehead’s technical sense. What survives the trimming is not consciousness everywhere but affective inheritance, proto-subjective self-unification, novel perspective, and realization of value.
That remaining claim is philosophically more interesting than the caricature, but it also creates a sharper burden of its own. The very moves that make Matt’s view more compatible with physical realization and biological emergence leave less room for the experiential predicate to hide inside familiar consciousness vocabulary. We can now ask more cleanly what “affective,” “proto-subjective,” “perspective,” and “value” positively contribute at primitive scales.
A recognizable Whiteheadian architecture runs through the earlier and current formulations. The remaining task is to make the positive content of the vocabulary stable enough for comparison.
If primitive feeling contains no more than causal uptake, integration, transition, and response, then a non-panexperiential process ontology may recover most of the explanatory work without the experiential predicate. If primitive feeling includes an irreducible affective or subjective character, the difference becomes substantive and needs warrant. Matt’s preferred route appears to resist both descriptions: feeling is neither an extra quale added to causation nor a decorative label on bare mechanism because relational becoming is already affective in its concrete actuality.
That is coherent as a metaphysical proposal. Coherence alone, however, does not settle the comparison.
A rival can grant relational becoming, embodied world-contact, emergent subjects, integration, causal inheritance, and scale-sensitive organization while suspending judgment about universal affectivity. What observable, explanatory, phenomenological, or conceptual result should favor Matt’s interpretation over that lower-commitment rival?
That is the question I have been asking him.
The mirror between “physical” and “experiential”
Once Erik’s and Matt’s arguments are reconstructed strongly, their disagreement becomes much smaller and more interesting.
Erik begins from the stronger empirical position. Every familiar conscious organism is physically realized, and the relevant physical organization is not causally idle: interventions on it systematically alter conscious capacities. Matt’s metaphysics does not erase that asymmetry. The overreach begins only when this extraordinarily well-supported realization claim is promoted into an intrinsic universal: reality is fundamentally and exhaustively physical.
Matt begins from the undeniable fact that at least some causal organization is known experientially from within. He is tempted to promote that access into another intrinsic universal: causal becoming is fundamentally experiential.
Both starting points are real. Neither universal follows for free.
This symmetry matters because each side can easily see the other’s leap. Erik asks why an electron needs experience if electron physics works without it. Matt asks why the success of structural and dynamical physics should be mistaken for an exhaustive account of concrete actuality. Each criticism has force. Each can also be returned.
To Erik: if physical equations and causal models capture the same successful structure under physicalism, Russellian monism, and a Whiteheadian process ontology, what evidence belongs specifically to the word “physical” when it is used as an intrinsic metaphysical predicate?
To Matt: if causal inheritance, integration, and emergence occur under both panexperientialism and a process-relational agnosticism, what evidence belongs specifically to the word “experiential” when it is generalized to causal becoming as such?
There is a Whiteheadian irony here. The fallacy of misplaced concreteness can be aimed in both directions. A physicalist may mistake the abstractions of successful physics for the full intrinsic nature of reality. A panexperientialist may mistake the undeniable experiential character of one richly organized region of nature for an invariant of causal process everywhere.
Equal uncertainty would be a mistake here. The empirical record is asymmetric. Physical realization and physical causal constraint have enormous support. Universal primitive feeling does not have a comparable record of distinctive prediction or intervention. That asymmetry should remain visible.
What remains underdetermined is narrower: whether the successful physical causal structure is intrinsically physical in a strong metaphysical sense, intrinsically experiential, neutral, structural, or poorly described by that whole menu of predicates.
That is a much smaller uncertainty than “we know nothing.”
Questions the comparison forces
The conversation becomes more productive when the questions are paired. The decomposition also makes some predictions before anyone answers: objections aimed at commitments already shared by the strongest rivals should fail to discriminate them, while criticisms that reach a genuine residual should survive clarification.
That immediately removes several familiar moves from the decisive part of this dispute. Evidence for a prehuman Earth bears against views that make the world’s existence depend on human minds; it does little against Matt’s Whiteheadian realism. Neural dependence is devastating for theories requiring ordinary consciousness to float free of physical organization, while Matt’s latest formulation already treats animal consciousness as an achievement of highly organized bodies. The absence of an anomalous nonphysical force bears heavily on interactionist dualism, while a panexperientialism that preserves the physical causal story predicts no such anomaly. The standard subject-summing combination problem strikes constitutive micropsychism much more directly than a view in which subjects arise through relational integration.
The mirror removes some Whiteheadian moves as discriminators too. Relationally constituted subjects are compatible with sophisticated physicalism. Embodied causal participation is compatible with physicalism. The fact that experience occurs within nature establishes experience in nature without deciding whether causal nature generally is experiential. An unresolved explanatory gap can keep physicalism incomplete without supplying evidence for panexperientialism. Theory-ladenness gives every side a reason for reciprocal comparison; it gives none of them ownership of the categories in which the comparison begins.
Erik, what is the strongest version of physicalism you intend to defend? If the answer is that every investigated causal process in biology respects physical constraints and that known consciousness is physically realized, I am largely with you. If the claim reaches intrinsic and exhaustive physicality, what observation, argument, or successful rival would lower your confidence in that extra step without requiring the empirical sciences to fail?
Matt, the corresponding question is: what is the thinnest positive content that still makes your view panexperiential rather than merely process-relational? If causal inheritance, integration, and state transition remain after “experience” is removed, what disappears with the word? If something genuinely affective or proto-subjective disappears, what gives us reason to expect that feature in primitive causal processes?
My own answer is provisional. Physical realization has earned far more confidence than exhaustive physical ontology. Experience is undeniably real in paradigmatic conscious systems; universal experientiality has not earned the same extension. So I keep the realization claim and withhold both intrinsic universals unless further comparison breaks the tie.
A second pair concerns absence.
Erik, what could count as evidence for something not adequately captured by your physicalist ontology? If the answer is “anything discovered by science would simply become physical,” then physicalism risks becoming true by terminological annexation. A serious research programme needs some way to lose content even if its empirical descendants continue to flourish.
Matt, what would a genuinely non-prehensive causal process look like? If prehension is attributed to every causal relation by definition, then no observation can classify a case as lacking it. You may still defend the ontology abductively, but the status of the claim changes. We should be explicit about that tribunal.
My framework’s answer is to separate claim types. Empirical claims should face observation and intervention where possible. Constitutive or metaphysical claims may depend more on coherence, explanatory comparison, bridge quality, and theoretical cost. Lack of a laboratory falsifier does not make a metaphysical claim meaningless; it does reduce how much empirical credit it can claim.
A third pair concerns scope.
Erik, what bridge licenses moving from “every high-confidence conscious biological system we know is neurally organized” to “all possible consciousness requires nervous systems”? Matt, what bridge licenses moving from “our embodied causal existence is experienced affectively” to “causal becoming generally is affective”?
Both moves transport a predicate outside the evidential region in which it was established.
My current rule is simple: the farther the transport, the more the bridge owes. Actuality comes before universality. Universality comes before necessity. An analogy can generate a programme; it cannot carry evidential weight indefinitely just because the vocabulary still fits.
Another pair concerns failure.
Erik, when one purported scientific panpsychist model is retracted, which assumptions failed and which other theories actually depend on them? Matt, if a Whiteheadian claim conflicts with a well-established scientific result, which part of the metaphysics changes, and how far does the revision travel?
I ask the same of my own method. If a bridge audit blocks valid inference more often than it prevents bad transport, the bridge rule should be narrowed. If common-core subtraction destroys interacting structure, the factorization failed and the theories should be compared as reconstructed wholes. If explicit recursive auditing adds paperwork without improving error localization or calibration, it should be compressed.
The evaluator has no special exemption.
Finally, what does each framework explain that its strongest lower-commitment rival cannot recover?
That question is harder than “can the theory accommodate the evidence?” Almost every sophisticated metaphysics can accommodate a great deal after enough interpretation. The informative comparison asks what the disputed commitment contributes after shared successes are removed.
Which brings us to the point where every position in the comparison has to pay the same bill.
The common core
Start with what survives across Erik’s, Massimo’s, and Matt’s strongest positions, before turning the same audit on my own framework.
There is a world that resists our preferences and constrains inquiry. Organisms are physically realized and embedded in causal relations. In confirmed conscious systems, that realization is interventionally consequential rather than merely decorative: changing the relevant physical organization changes conscious capacities. Evolution shaped nervous systems and the cognitive capacities associated with them. Subjects need not be treated as Cartesian substances standing outside the world. Causal and relational organization matters across scales. No reproducible evidence presently requires an independently acting immaterial mind to explain ordinary neural causation.
That common core is already substantial. It explains why Erik, Massimo, and Matt can disagree at the ontological level while sharing so much of the empirical biology.
Now ask what each metaphysical programme adds, then ask what my method adds.
Erik’s distinctive addition, in its strongest metaphysical form, is that the successful causal reality described by the sciences is exhaustively physical or material, with consciousness appearing only when matter is organized in the relevant biological ways. To earn that addition rather than merely the physical realization claim, Erik needs a sufficiently clear conception of “physical” and some comparative reason to prefer it over rivals that preserve the same physical structure while interpreting intrinsic nature differently. A result that left all physical science intact but gave a rival independently successful predictions from its supposedly nonphysical or neutral residual should count against the exhaustive thesis. So should a demonstration that the metaphysical predicate contributes nothing beyond the shared carrier.
Massimo’s physicalism overlaps heavily with Erik’s empirical carrier, but his own probabilistic epistemology makes the decomposition easier to state. Physical realization, causal closure, universal supervenience, absence of fundamental mentality, weak rather than strong emergence, and the stronger claim that matter and energy exhaust ontology are connected commitments, not one proposition with one confidence value. Evidence can move them by different amounts. A new physically continuous mechanism can raise confidence in the physicalist programme while adding much less to the claim about intrinsic exhaustiveness if the strongest rivals expected the same mechanism. Conversely, a failure of universal supervenience would not erase the physical realization of every confirmed conscious organism. Treating the components separately makes his physicalism more Bayesian in exactly the place where the metaphysics otherwise risks being treated as settled.
Matt’s distinctive addition is that relational becoming is intrinsically prehensive, affective, or experiential even where reflective consciousness and representation are absent. To earn that addition over process-relational agnosticism, the experiential residual has to be positively characterized and tied to some comparative gain that the rival cannot recover as cheaply. If every explanatory success survives when primitive experiential language is replaced by causal inheritance and organization, panexperientialism loses distinctness even if the process philosophy remains useful. If the experiential predicate adds genuine qualitative or proto-subjective content, then evidence for that content must travel beyond the systems in which experience is already known.
My own framework makes a methodological wager rather than adding another intrinsic substance. It proposes a correction architecture: decompose the target, reconstruct serious rivals, separate the shared carrier from the disputed residual, track where evidence enters, audit nontrivial evidence transport, and propagate failures only through the dependencies they actually touch. The recursive commitment is that those operations themselves remain available for revision.
I built that architecture because these failures kept recurring in my own work. Its appeal to me earns it no evidential credit.
Its positive residual has to show up in comparative performance. If strong researchers using severe testing, causal inference, Bayesian comparison, construct validity, model criticism, and good provenance practices catch the same errors with less machinery, then my framework should shrink toward a synthesis of those methods. If explicit dependency tracking, bridge auditing, common-core reconstruction, and recursive self-application measurably improve error localization, rival discrimination, calibration, or recovery from failed assumptions, then the architecture has earned something distinct. The ablation has to decide.
That is the symmetry I care about most.
Erik can reasonably assign little confidence to universal experientiality without proving Matt impossible. Matt can question whether physical science exhausts intrinsic nature while accepting its empirical success. And I cannot call my framework superior merely because it can redescribe both errors after the fact.
Each position should say what it adds, why the addition matters, what would count against it, and what survives if it fails.
A methodology that cannot answer those questions about itself should not be trusted to ask them of anyone else.
What recursive self-criticism would let Erik keep
If Erik adopted this recursive version of his own critical instincts, very little of what seems to motivate him would need to disappear.
He could remain a realist about a world that does not depend on human preference. He could retain a strong presumption in favor of explanations continuous with successful physics, chemistry, and biology. He could regard known consciousness as deeply dependent on physical organization. He could keep a low prior for interventionist dualism, supernatural causal additions, and theories that repeatedly avoid risky contact with evidence. He could continue demanding that panexperientialists tell us what primitive experience adds.
What changes is the rhetoric of necessity and totality.
“No evidence currently requires an immaterial causal contribution” is stronger epistemically than pretending we have inspected every possible ontology and found none. “Every confirmed conscious system we know depends on biological organization” is a serious empirical statement. It becomes more vulnerable when it is inflated into a theorem about every possible substrate. “Panpsychism has not produced distinctive empirical support for ubiquitous experience” is a defensible criticism. Calling the entire heterogeneous family pseudoscience because one model was retracted invites an easy reply.
The thinner physicalism is in those places, the harder it becomes to attack.
Physicalist research programmes can have the best empirical record while the final intrinsic ontology remains unsettled. Calling that capitulation to idealism would confuse disciplined confidence with metaphysical surrender.
For somebody who identifies with critical rationalism, this also creates a better Popperian posture. The important comparison becomes whether successive physicalist explanations generate new constraints, survive stronger rivals, and solve problems those rivals cannot without ad hoc repair. Every such success can raise physicalism’s probability while showing us which part of the programme actually earned the update. That turns physicalism from the presumed scientific side of a cultural divide into a programme capable of demonstrating progress.
Physicalism should want that competition because it begins from the strongest empirical carrier in this dispute.
If it is right, a framework that refuses to give it a free pass is an ally.
What the same discipline leaves Matt
Matt also retains more than a simple materialist critique would allow.
The process-relational rejection of a detached Cartesian spectator survives. So does the idea that knowers are worldly processes rather than observers outside nature. The emphasis on embodied causal participation, temporal inheritance, and relational constitution can remain philosophically fruitful even if panexperientialism itself is underdetermined. Whitehead’s attempt to resist the bifurcation of nature identifies a real conceptual pressure: a scientific description of causal structure should not be casually mistaken for a complete inventory of intrinsic actuality.
Matt’s recent clarification makes the strongest version of that challenge harder for Erik to dismiss. I regard it as substantially stronger than the earlier formulations I had been arguing against, precisely because it gives up several easy targets while preserving the Whiteheadian core Matt cares about. If animal consciousness is a special achievement of highly organized bodies, ordinary neuroscience does not refute Whiteheadian prehension. If primitive feeling is not a miniature quale and the subject is constituted rather than presupposed, the standard micropsychist objections lose force. If the scheme itself is admitted into recursive scrutiny, disagreement no longer has to turn on whether speculative metaphysics is somehow exempt from correction. The dispute sits beneath the neural story and is cleaner than it was before.
That improvement has a cost, in the productive sense. Each ambiguity removed from the older language concentrates the burden on what remains. Why do affectivity, proto-subjectivity, perspective, or value travel downward with causality while consciousness, intentionality, representation, and sense perception are allowed to remain scale-specific achievements?
A proportional analogy tells us that the forms differ. It does not tell us which features are invariant.
My confidence in the process-relational core is substantially higher than my confidence in universal experiential interpretation. That leaves panexperientialism as an active research programme with an unusually well-localized residual: primitive affective or proto-subjective character. Its evidential standing should rise if that residual generates successful constraints its strongest process-relational rival cannot reproduce, and fall if the same work survives after it is removed.
Clarification can improve the programme conceptually by stabilizing constructs, removing unnecessary commitments, and locating better bridges. Successful differential risk would do something further: it would improve the evidential standing of the distinctive ontology rather than only the coherence of the scheme.
From defended positions to progressive research programmes
The decomposition has a constructive consequence. Recursive criticism does more than tell Erik, Massimo, or Matt where an inference reaches too far. It separates what each programme has already earned from what it still needs to discover, which gives each of them a more useful positive heuristic.
I use progressive here in the demanding sense associated with Imre Lakatos. A programme becomes theoretically progressive when a successor adds empirical content, and empirically progressive when some of that additional content is corroborated. Philosophical programmes can improve conceptually before they reach laboratory prediction; Lakatosian empirical progress begins when the distinctive commitments generate additional risk and some of that risk succeeds. Lakatos’s original formulation makes that distinction especially sharp.
Erik begins with the strongest empirical base. Physical realization of known consciousness, interventionally demonstrated dependence on physical organization, evolutionary continuity, and the extraordinary success of physically continuous biology all survive even if some stronger metaphysical language is trimmed away. That base can support a genuinely positive programme: keep extending physically explicit mechanisms into domains where they have not yet succeeded, state in advance what the extension predicts relative to the strongest rival, and credit the physicalist commitment that actually produced the differential success. A failure can then be localized to a mechanism, a substrate restriction, a reductionist assumption, or a stronger ontological claim instead of being treated as a verdict on everything physicalism has ever done.
That reframes Erik’s substrate claim into a better research question. Every confirmed animal consciousness we know is biologically and neurally realized. What property of nervous systems should be necessary for consciousness rather than constituting the only realization we presently know? A neural-essentialist account and an organizational account realized in a radically different substrate should eventually diverge somewhere. Locating that divergence would teach us more than repeating the shared observation that neuroscience matters.
Massimo already has much of the required epistemology. His own writing treats scientific confidence as graded, probabilistic, and revisable. The recursive extension is to apply that calibration inside physicalism itself. Physical realization can deserve extremely high confidence while closure, universal supervenience, absence of fundamental mentality, and exhaustive material ontology receive separate posteriors and separate update conditions. A successful physical explanation raises the physicalist family most strongly when the result was substantially more expected under that family than under its developed rivals. The failure of dualism can move probability toward physicalism without promoting every physicalist commitment by the same amount.
Matt begins from a different strength. Whiteheadian process philosophy has a rich positive heuristic: causal inheritance, temporal becoming, relational constitution, affect, valuation, organization, subject formation, and the continuity between experience and nature all generate questions worth pursuing. His recent clarification also gives the programme cleaner correction conditions. The next step is to make the specifically panexperiential residual generate comparative risk. Let the metaphysical experiment run far enough to become definite, then compare it with a reconstructed process-relational rival that accepts physical realization, inheritance, integration, novelty, and emergent subjects while withholding primitive affectivity. If prehension produces a constraint on subject formation, valuation, integration, phenomenality, or some other target that the rival lacks, freeze the difference before the answer is known and test it.
The evidence ranks these programmes asymmetrically from the start. Physical realization and physical causal explanation have an enormous demonstrated record; primitive universal experience begins with a much larger evidential debt. Keeping probability mobile preserves that asymmetry while preventing success of the shared carrier from silently proving every metaphysical interpretation attached to it. Physicalism can lead the ranking while some of its stronger commitments remain open. Panexperientialism can remain a serious speculative programme while carrying the heavier burden for its distinctive predicate.
The same procedure also tells us what counts as progress short of decisive empirical discrimination. A programme can improve conceptually by clarifying constructs, removing contradictions, discovering that an old objection targeted the wrong claim, or building a bridge that finally makes a future discriminator possible. Matt’s recent clarification is progress in that sense. Erik would make similar progress by separating neural realization from neural necessity. Massimo would make it by assigning different confidence levels to the physicalist commitments his own probabilism already allows us to distinguish. Those improvements become Lakatosian empirical progress only when the revised programme leads to additional risky content that succeeds.
This gives the dispute somewhere to go tomorrow. Erik can search for properties that make the physicalist or neural account prospectively distinctive. Massimo can make the posterior structure of physicalism explicit. Matt can identify what primitive affectivity changes after the common process-relational carrier is held fixed. Each programme should count its own differential successes as discoveries; a neighboring programme’s failure merely reweights the field according to what that failure actually discriminates.
The recursion then reaches my own framework (Recursive Corrigibility Framework or “RCF” for short). Common-core subtraction, evidential locality, bridge auditing, dependency-aware revision, recursive mirroring, and persistent correction memory are hypotheses about how inquiry can be improved. If those operations repeatedly predict which objections will collapse, identify residual disagreements earlier, generate useful discriminators, or improve calibration and error localization over strong simpler methods, they earn their place. If they mainly supply a vocabulary for explaining good reasoning after it has already happened, the framework is accommodating success rather than producing it.
The strongest achievement available to any of these programmes is therefore the same: become easier for reality to correct while becoming better at discovering something the previous version could not.
Where I land
I still assign a substantial advantage to physicalism-adjacent naturalistic programmes over substance dualism, interventionist supernaturalism, and unrestricted claims of ubiquitous experience. The evidence is not evenly distributed. Physical causal explanation has an unmatched record of prediction and intervention. Known human consciousness is profoundly dependent on brains. Those facts should shape rational priors.
I am less confident in exhaustive metaphysical physicalism than I am in physical realization, and that difference should not be mistaken for neutrality between physicalist and anti-physicalist programmes. Physical realization and the causal relevance of physical organization are among the strongest facts any consciousness theory has to accommodate. I am also much less confident in Segall’s universal experiential interpretation than I am in the process-relational features of his view that can be supported independently. The probability I withhold from strict physicalism does not automatically migrate to panexperientialism. Much of it belongs in an underdetermined middle where the shared causal structure is well supported and its ultimate intrinsic description remains open.
This is disciplined uncertainty: a map of where the evidence currently reaches.
My disappointment with Erik’s article therefore comes from sympathy rather than hostility. He has enough evidence to make a strong case for naturalistic and physicalist research programmes. Some of the essay’s weakest moves are unnecessary additions: the caricature of Kant, the simplified Bohr, the genealogy that is asked to prove too much, the universalization from neural cases, the propagation from one retraction, and the rhetorical sorting of opponents before their strongest claims have been reconstructed.
Removing those moves would leave him with less metaphysical certainty and a better argument.
Matt faces the complementary pressure. The more carefully he distinguishes primitive prehension from animal consciousness, the less force Erik’s easy objections have. At the same time, the positive difference between panexperientialism and process-relational agnosticism becomes harder to locate. Clarification protects the view from caricature and increases the burden on its residual.
Massimo’s position shows how the physicalist side can absorb the same discipline without surrendering its evidential advantage. His probabilism can be extended from hypotheses within science to the component commitments inside physicalism itself. That preserves a very high posterior for physical realization while leaving closure, universal supervenience, absence of fundamental mentality, and exhaustive ontology free to move by different amounts as evidence accumulates.
That is exactly what a productive dispute should do. It should make the disagreement smaller, more precise, and more dangerous to every participant’s favorite language.
The question I would leave with Erik, Massimo, Matt, and myself is therefore the same one:
After everything our strongest rivals already share has been accounted for, what are we still adding, what difference should that addition make, and what would make us give it up?
For me, after subtracting what the strongest rivals already share, very little global ontology remains. The strongest common core is that inquiry encounters stable, non-voluntary constraints and that some constructs survive prediction, intervention, transfer, and representational change far better than others.
I presently interpret that as evidence for a reality not exhausted by our descriptions, but a sufficiently strong constraint-pragmatist rival may preserve the same successes without that realist addition.
Until I can identify what the realism itself changes, I should treat it as a favored working interpretation rather than something the evidence has uniquely forced.
If physicalism wins that comparison, I want it to win because it survived the test. If panexperientialism eventually earns a distinctive bridge that its rivals cannot reproduce, it should gain accordingly. If my own framework adds no measurable value beyond better-established methods, it should be compressed.
The word “everything” is expensive. Whoever claims it has to pay for the extra scope with evidence that the strongest lower-commitment rival cannot already explain.
Sources and further reading
Erik Svensson and the arguments discussed
Erik Svensson, “Philosopher Tim Maudlin on the problems with Immanuel Kant and German Idealism”, 26 August 2026. URL:
Erik Svensson, “Clear thinking is more important than ever: why the Vienna Circle got most things right”, 22 April 2026. URL:
Erik Svensson, “Evolutionary biology is not ‘Neo-Darwinism’”, 7 April 2026. URL:
Erik Svensson, “A major blow against panpsychism: a scientific journal retracts a hyped paper from Sweden”, 17 May 2026. URL:
Kant, panpsychism, physicalism, and Bohr
Paul Abela, Kant’s Empirical Realism, Oxford University Press, 2002. URL: https://academic.oup.com/book/6931
Dietmar Heidemann, “Kant and the forms of realism”, Synthese, 2019. URL: https://link.springer.com/article/10.1007/s11229-019-02502-4
Stanford Encyclopedia of Philosophy, “Idealism”. URL: https://plato.stanford.edu/entries/idealism/
Stanford Encyclopedia of Philosophy, “Immanuel Kant”. URL: https://plato.stanford.edu/entries/kant/
Stanford Encyclopedia of Philosophy, “Panpsychism”. URL: https://plato.stanford.edu/entries/panpsychism/
Stanford Encyclopedia of Philosophy, “Physicalism”. URL: https://plato.stanford.edu/entries/physicalism/
Stanford Encyclopedia of Philosophy, “Copenhagen Interpretation of Quantum Mechanics”. URL: https://plato.stanford.edu/entries/qm-copenhagen/
Pigliucci, physicalism, underdetermination, and comparative confirmation
Massimo Pigliucci, public reply to Matthew David Segall on the meaning of “physical”, 7 May 2026. Primary source. URL:
Massimo Pigliucci, “Five allegedly bad ‘isms’ – Part I”, 14 July 2026. Primary source on materialism, reductionism, physical causation, supervenience, and emergence. URL:
Massimo Pigliucci, “Five allegedly bad ‘isms’ – Part II”, 16 July 2026. Primary source; includes the statement that “matter and energy are all there is.” URL:
Massimo Pigliucci, “Mr. Bayes and the True Nature of Scientific Hypotheses”, 2001. Primary essay on probabilistic scientific inference. URL: https://infidels.org/kiosk/article/mr-bayes-and-the-true-nature-of-scientific-hypotheses/
Massimo Pigliucci, “Atheism 101: I – Basic epistemology”, 3 February 2026. Primary source; Pigliucci explicitly endorses Flew’s burden principle as his own stance and frames evidential judgment fallibilistically and probabilistically. URL:
Massimo Pigliucci, “New atheism: a critique of a critique”, 2026. Primary source; despite criticizing the New Atheists’ tone, Pigliucci says that in substance they “pretty much all got it right.” URL:
Alyssa Ney, “Defining Physicalism”, Philosophy Compass 3(5), 2008, 1033-1048. DOI: https://doi.org/10.1111/j.1747-9991.2008.00163.x
Justin Tiehen, “Physicalism”, Analysis 78(3), 2018, 537-551. Author-archived PDF record: https://philpapers.org/versions/TIEP-6 ; DOI: https://doi.org/10.1093/analys/any037
David Papineau, “The Case for Materialism”, in Thinking about Consciousness, Oxford University Press, 2002, 13-46. DOI: https://doi.org/10.1093/0199243824.003.0002
Alyssa Ney, “Overdetermination and Causal Closure: A Defense of the Causal Argument for Physicalism”, ProtoSociology 39, 2022, 35-50. Open manuscript record: https://philarchive.org/rec/NEYOAC-2 ; DOI: https://doi.org/10.5840/protosociology2022394
Bradford Saad, “Grounding Causal Closure or Something Near Enough”, Acta Analytica 40, 2025, 379-398. Full open text: https://pmc.ncbi.nlm.nih.gov/articles/PMC12335395/ ; DOI: https://doi.org/10.1007/s12136-024-00620-4
Maria Caamaño-Alegre, “Empirical Underdetermination: The Empirical Side of the Duhem-Quine Thesis”, Philosophy Compass 20(3), 2025, e70022. Author-provided full-text page: https://www.researchgate.net/publication/389326949_Empirical_Underdetermination_The_Empirical_Side_of_the_Duhem-Quine_Thesis ; DOI: https://doi.org/10.1111/phc3.70022
Finnur Dellsén, “Explanatory Rivals and the Ultimate Argument”, Theoria 82(3), 2016, 217-237. Author manuscript / full PDF: https://philpapers.org/archive/DELERA-3.pdf ; DOI: https://doi.org/10.1111/theo.12084
Friedel Weinert, “The role of probability arguments in the history of science”, Studies in History and Philosophy of Science Part A 41(1), 2010, 95-104. DOI: https://doi.org/10.1016/j.shpsa.2009.12.003
Roberto Festa and Gustavo Cevolani, “Unfolding the Grammar of Bayesian Confirmation: Likelihood and Antilikelihood Principles”, Philosophy of Science 84(1), 2017, 56-81. DOI: https://doi.org/10.1086/688935
Intellectual lineage: criticism, burden, and revision
Otto Neurath, Philosophical Papers 1913-1946, especially “Protocol Statements,” pp. 91-99. Primary collection; DOI/book URL: https://doi.org/10.1007/978-94-009-6995-7 URL: https://link.springer.com/book/10.1007/978-94-009-6995-7
Karl Popper, Conjectures and Refutations, especially Chapter 1 on conjecture, criticism, and risky testing. Full PDF. URL: https://mechanism.ucsd.edu/bill/teaching/philsci/popper.conjecturesandref.pdf
Antony Flew, “The Falsification Response”, Religious Studies 5(1), 1969, 77-79, restating his earlier falsification challenge. Cambridge PDF: https://www.cambridge.org/core/services/aop-cambridge-core/content/view/BF7F5EBA7B1DF78689EE5A34D54A82F2/S0034412500003978a.pdf/falsification_response1.pdf ; DOI: https://doi.org/10.1017/S0034412500003978
Christopher Hitchens, God Is Not Great, 2007, including the evidential-burden formulation now known as Hitchens’ razor. Primary book record. URL: https://openlibrary.org/books/OL7537199M/God_is_not_great
Consciousness and evolutionary uncertainty
Anil K. Seth and Tim Bayne, “Theories of consciousness”, Nature Reviews Neuroscience 23, 2022, 439-452. URL: https://www.nature.com/articles/s41583-022-00587-4
Cogitate Consortium et al., “Adversarial testing of global neuronal workspace and integrated information theories of consciousness”, Nature, 2025. URL: https://www.nature.com/articles/s41586-025-08888-1
Mona-Marie Wandrey and Marta Halina, “The Evolution of Animal Consciousness”, Philosophy Compass 20, 2025. URL: https://compass.onlinelibrary.wiley.com/doi/full/10.1111/phc3.70069
Matthew David Segall and Whiteheadian process thought
Matthew David Segall, Footnotes2Plato, including his work on Whitehead, prehension, and process-relational cosmology. URL:
https://home.footnotes2plato.com/
Matthew David Segall, “Why the World Is Unfinished”, including the earlier description of prehension as having an experiential quality and being like qualia in that sense. URL:
Matthew David Segall, “Reality Is Process”, including the earlier gloss of feeling in qualitative “what-it-is-like” terms. URL:
Matthew David Segall, “There’s No Scientific Evidence That…”, including discussion of awareness, self-enjoyment, and pervasive subjectivity. URL:
Matthew David Segall, “Whitehead and Media Ecology”, including his characterization of causal efficacy as presubjective affective inheritance. URL: https://matthewsegall.files.wordpress.com/2019/07/06_segall_ps-48_2-text.pdf
Matthew David Segall, doctoral dissertation, especially his treatment of causal efficacy as presubjective affective inheritance and an ontologically basic experiential register. URL: https://footnotes2plato.com/wp-content/uploads/2016/06/matt-segall-diss-final-draft-with-tech-edits-4_9_2016.pdf
Critical rationalism and the framework used here
Imre Lakatos, “Falsification and the Methodology of Scientific Research Programmes”, in Criticism and the Growth of Knowledge, 1970, pp. 91-196. Full PDF: https://www.csun.edu/~vcsoc00i/classes/s497f09/s690s08/Lakatos.pdf ; Cambridge primary bibliographic record and DOI: https://doi.org/10.1017/CBO9781139171434.009
Internet Encyclopedia of Philosophy, “Popper: Critical Rationalism”, including discussion of Popper, Agassi, and Bartley’s comprehensively critical rationalism. URL: https://iep.utm.edu/karl-popper-critical-ratiotionalism/
Nathan Sweet, “Recursive Corrigibility Framework”. URL: https://sweetrationalism.com/frameworks/recursive-corrigibility-framework/
Nathan Sweet, “A Provisional, Boundary- and Scale-Relative Realism”. URL: https://sweetrationalism.com/frameworks/ontology-scale-relative-realism/
