{"id":"5272dc33-1e8e-40b2-aacf-6f5781651a44","arxiv_id":"2507.19834","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"Empirical structure physicalism says current special sciences supervene on current and future physics through refinement of empirical structure, giving a new route past Hempel's dilemma.","lead":"This paper introduces a new way to define physicalism, based on how theories distinguish space-time regions by observable data rather than on what entities they say exist. It argues that this version escapes Hempel's dilemma about choosing current or future physics, and keeps mental theories dependent on physics in the future.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The case for future empirical structure realism presupposes that a future physics without spacetime can still be assigned empirical contents over current spacetime regions; the paper asserts this comparability (§7) without argument, so the meta-induction from past refinements cannot support Rf…","rationale":"The reader identified the unproven historical premise (Rp) as the weakest assumption. I agree that the road to Rf is the load-bearing segment, but the deeper problem is that the meta-induction from Rp to Rf cannot even be applied to arbitrary future physics without an additional, undefended comparability premise. The paper defines empirical adequacy with respect to space-time regions that are, in the last resort, GR submanifolds. A future physics that replaces spacetime would render e^F_R empty for all such regions, making the future empirical structure trivial and breaking supervenience. §7 waves this away by asserting that submanifolds supervene on the future locus concept, but that assertion is itself an instance of empirical structure realism — the very thesis the optimistic meta-induction is supposed to support. So the central claim, that empirical structure physicalism avoids the second horn, rests on a circularity unless the comparability premise is argued for independently. That is a concrete, fixable gap: the paper could adopt a more neutral characterization of regions, or add a separate argument that any future physics must recover the old empirical structure. As it stands, the verdict CONDITIONAL is correct: the framework is novel and the derivation is formally valid, but the second-horn solution is not established.","tokens_in":22602,"tokens_out":10218,"duration_ms":121696,"concrete_test":"Re-derive the transitivity step in §4 while making explicit the dependence of empirical structure on the space-time region parameter M. Verify whether the inference from Pc and Rf to Pf requires an additional premise that the current notion of 'space-time region' (GR submanifolds, per §7) supervenes on the future theory's locus concept. If the proof cannot go through without this premise, and §7 supplies it only by analogy to Rf, the argument for blocking the second horn of Hempel's dilemma is circular.","verdict_should_be":"UNCHANGED","load_bearing_attack":"To make Rf meaningful, the empirical structure of future physics must be definable over the same space-time regions as current physics. But the paper's definition of empirical adequacy (§3.1) requires a history-description to be localizable 'throughout and restricted only to R,' where R is a region of our actual world, formally a submanifold of GR spacetime (§7). If a future theory abandons spacetime, its history-descriptions may not be localizable to any such R, making e^F_R empty for every R and the future empirical structure the trivial partition. Then current physics would not supervene on future physics, and Pf would fail. The paper's response (§7) is that we can expect the submanifold notion to supervene on the future locus concept 'analogously to how current physics supervenes on future physics.' That analogy is precisely the conclusion to be established: it is an instance of empirical structure realism, not an independent premise. The optimistic meta-induction from Rp cannot license it, because Rp only records past refinements among theories that already shared a spacetime framework; it says nothing about a theory that replaces the framework outright. Thus, even if Rp were fully defended, Rf would still lack support, and the second horn of Hempel's dilemma would not be blocked. The paper's own admission that the formal account of 'space-time region' would fall back on GR submanifolds underscores that the comparison across future theory change is assumed, not derived.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper introduces a formal notion of empirical content and empirical structure, defines theory supervenience as refinement of empirical structures, and argues that empirical structure physicalism (the conjunction of currentist and futurist theses, Pc and Pf) avoids Hempel's dilemma. The key derivation is that Pf follows from Pc and future empirical structure realism (Rf) by transitivity of theory supervenience. Rf is defended by an optimistic meta-induction from past empirical structure realism (Rp). The paper further argues that empirical structure physicalism satisfies most standard desiderata for physicalism, is compatible with multiple realization, and can coexist with the pessimistic meta-induction.","tokens_in":22904,"tokens_out":5250,"duration_ms":67919,"significance":"If successful, the paper would offer a new reductive relationship between special sciences and physics that sidesteps Hempel's dilemma without committing to the ontology of current or future physics. The formal framework is precise, the transitivity observation is correct, and the paper is admirably explicit about which premises are asserted and which are merely supported by plausibility arguments. The distinction between empirical structure and formal content is a useful contribution, and the application to mental theories is clearly illustrated. However, the main thesis rests on an unestablished historical premise and on an unresolved comparability problem across future theory change, so the significance is conditional on further work.","major_comments":[{"comment":"The argument for future empirical structure realism (Rf) is the optimistic meta-induction from (Rp), but the paper explicitly states that 'Length limitations prevent me from defending the historical premise here' and offers only illustrative examples (equant/epicycle, Newton/GR). Since (Pf) is obtained from (Pc) and (Rf), and (Rf) is what blocks the second horn of Hempel's dilemma, the central claim that empirical structure physicalism avoids both horns is conditioned on an unestablished empirical generalization. To make the main claim, the historical premise needs a systematic defense, or the conclusion must be explicitly weakened to a conditional claim.","section":"Section 5"},{"comment":"The definition of empirical content e^T_R in Section 3.1 presupposes that R is a region of our actual world, formally a submanifold of GR spacetime. Future empirical structure realism requires comparing e^{current}_R with e^{future}_R for the same R. If a future theory replaces spacetime itself, its history-descriptions may not be localizable to GR submanifolds, making e^{future}_R empty for every R and future structure trivial. The paper's response that the submanifold notion will supervene on a future locus concept 'analogously to how current physics supervenes on future physics' is precisely an instance of the thesis to be established, not an independent premise. The meta-induction from past theory changes cannot license this, because past changes such as Newton to GR preserved a spacetime framework. Without an account of how empirical contents are assigned across frameworks, (Rf) is not a well-defined a posteriori thesis.","section":"Section 7"},{"comment":"The definition of empirical adequacy depends on 'certain selected observable aspects of happenings', admissible idealizations, and an approximation level. These choices are not fixed by the theory, yet the supervenience relation and the physicalism theses are stated as categorical claims. Different choices can change whether e^T_R1 = e^T_R2, and hence whether one structure refines another; a sufficiently coarse choice of aspects can make any two theories supervene trivially. The paper should either provide a principled criterion for selecting these parameters or explicitly relativize (Pc), (Pf), and (Rf) to a parameter choice and show that the conclusions hold for natural choices.","section":"Section 3.1"},{"comment":"The plausibility argument for (Pc) appears to assume what it needs to show. The claim that 'no measurement method exists for current special sciences whose different outcomes would not manifest in a difference of empirical contents of current physics' is, in effect, the supervenience thesis restated in operational terms. The room example illustrates the thesis but does not provide independent evidence for it. Since (Pc) is a conjunct of (P), the paper should either present independent evidence or be explicit that (P) is defended only as a package whose overall plausibility is inherited from (Rp).","section":"Section 6.2"}],"minor_comments":[{"comment":"The empirical structure e^T is defined as a partition over all subsets R of M, but the examples and applications only consider connected, bounded space-time regions; the formal definition should either restrict to a well-defined class of regions or explain why arbitrary subsets are admissible.","section":"Section 3.3"},{"comment":"The multiple realization example conflates 'computer science supervenes on some physics theory' with 'the same special science is realized by different physical bases'; the fact that hypothetical GR-computers supervene on GR and quantum-sized computers on QM does not by itself show that current computer science is multiply realized, and this should be clarified.","section":"Section 6.3"},{"comment":"The reply to Kuhn losses dismisses explanatory losses as irrelevant, but since the historical premise (Rp) concerns empirical structure alone, the author should explain more explicitly why explanatory losses cannot indirectly affect judgments of empirical adequacy.","section":"Section 5"},{"comment":"The text contains numerous encoding artifacts and typographical slips (for example, 'T ˝ozs´er', 'H ¨orzer', and a duplicated 'that that'), which should be corrected in the final version.","section":"Throughout"}],"recommendation":"major_revision","confidential_remarks":"The stress-test concern about Section 7 is the most serious issue: without a theory-neutral account of regions or a separate argument for cross-framework comparability, the optimistic meta-induction cannot support Rf even if Rp were fully defended. I would encourage the editor to ask the author to address this point directly, along with providing a systematic defense of Rp or explicitly recasting the paper's conclusions as conditional."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Gyenis has written a serious and unusually clear paper. The core idea—define the empirical structure of a theory as the partition of spacetime regions by their empirical content, then say one theory supervenes on another if the second refines that partition—is genuinely new and gives a precise way to talk about empirical continuity across theory change. The transitivity derivation in Section 4 is correct, and the optimistic meta-induction is distinct from the pessimistic one in a way that is worth taking seriously. The paper also does a good job showing how empirical structure physicalism can coexist with multiple realization and with a kind of dualism.\n\nThe soft spots are where the reader's report says they are, plus one that matters more than the report suggests. First, the historical premise (Rp)—up to now later physics refined the empirical structure of earlier physics—is explicitly not defended. Length limitations are real, but this premise carries the entire optimistic meta-induction. The examples with Newton/GR and thermodynamics/stat mech are illustrative, not a defense. Second, and more worrying, is the comparability problem the stress-test note identifies. The whole framework defines empirical content relative to regions of 'our actual world,' formally submanifolds of GR spacetime. If future physics abandons spacetime, the empirical structure of future physics over those regions is not defined by the framework. The paper's Section 7 answers that we can expect submanifolds to supervene on the future locus concept 'analogously to how current physics supervenes on future physics.' But that is exactly what future empirical structure realism is supposed to establish—it is the conclusion, not a premise. So the optimistic meta-induction cannot license (Rf) for a theory that replaces the spacetime framework outright. The transitivity argument remains valid, but its reach is narrower than advertised.\n\nThe paper is honest about its limits: it says it argues for plausibility, not proof, of (Pc) and (Rp). But the abstract's claim to 'avoid both horns' overstates what the conditional derivation plus an undefended induction can deliver. For a reader working on physicalism or Hempel's dilemma, the paper is worth engaging; the formal notion of empirical structure may be useful in its own right. I would send it to peer review, primarily as a substantive proposal that needs to defend (Rp) and to say something principled about how empirical structure is defined across a change in the spacetime framework.","headline":"A genuinely new formal framework for physicalism with a valid transitivity argument, but the historical premise and the comparability of future theory change are not established, making the solution to Hempel's dilemma more conditional than advertised.","tokens_in":23389,"tokens_out":2818,"would_cite":true,"duration_ms":33084,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Empirical structure physicalism, the thesis that the current special sciences supervene on both current and future physics, avoids both horns of Hempel's dilemma.","keywords":["Hempel's dilemma","empirical structure","theory supervenience","physicalism","optimistic meta-induction","empirical content","multiple realization","supervenience"],"falsifier":"A concrete historical case in which a later, generally accepted theory of physics could not empirically distinguish a pair of space-time regions that an earlier accepted theory could distinguish, while no other accepted theory of that later period recovered the ability, would falsify past empirical structure realism and thereby remove the inductive basis for future empirical structure realism.","tokens_in":22375,"feed_emoji":"⚛️","tokens_out":7215,"duration_ms":71894,"temperature":0.7,"pith_summary":"This paper proposes a way to state physicalism that is immune to Hempel's dilemma, the argument that physicalism must be tied either to an incomplete current physics or to an unknowable future physics. The proposed thesis, empirical structure physicalism, says that the current special sciences supervene on both current and future physics, where supervenience means that the empirical structure of physics refines the empirical structure of the special sciences. Empirical structure is the partition of space-time regions according to which regions have the same empirical content under a theory. The paper's central derivation is that the future half of the thesis follows deductively from the current half together with future empirical structure realism, the claim that future physics will refine the empirical structure of current physics, by transitivity of supervenience. An optimistic meta-induction from the history of physics is then offered as strong support for future empirical structure realism, so that physicalism can survive future theory change without betting on the ontology of current or unknown future physics.","feed_headline":"Physicalism escapes Hempel's dilemma via empirical structure","feed_subtitle":"A supervenience relation on empirical structure lets physicalism avoid both horns of Hempel's dilemma.","key_machinery":"The central machinery is the notion of empirical structure together with the relation of theory supervenience. A theory's empirical content at a space-time region is the set of that theory's history-descriptions which are empirically adequate to the region, allowing idealization, approximation, and localization; the theory's empirical structure is the partition of all space-time regions into cells of equal empirical content. Theory supervenience holds when one theory's empirical structure refines another's, so every empirical distinction drawn by the lower-level theory is also drawn by the higher-level one. This relation is transitive, and transitivity is what makes futurist empirical structure physicalism a deductive consequence of currentist empirical structure physicalism and future empirical structure realism. Because empirical structure is defined without reference to the ontology a theory suggests, refinement can persist through theory change even when entities and terms do not survive.","core_discovery":"The paper argues that Hempel's dilemma can be solved by replacing the reductive relationship at the heart of physicalism: instead of relating entities or languages, empirical structure physicalism relates the empirical structures of theories. One theory supervenes on another exactly when the second theory's empirical structure refines the first's, meaning any pair of space-time regions the first theory can empirically distinguish is also distinguished by the second. Currentist empirical structure physicalism, the special sciences supervene on current physics, survives the incompleteness of current physics because it makes no claim that current physics is complete or that its ontology is final. Jointly with future empirical structure realism, transitivity of supervenience yields futurist empirical structure physicalism: the current special sciences remain empirically dispensable under any future physics that refines current physics' empirical structure. The paper further claims that future empirical structure realism is strongly supported by an optimistic meta-induction over the history of physics, and that this optimistic induction can coexist with the pessimistic meta-induction because it concerns retention of empirical structure rather than survival of entities.","pith_inferences":["The paper's actualist definition of empirical structure could be extended to empirical possibility, turning empirical structure physicalism into a modal thesis rather than a claim only about the actual world.","The optimistic meta-induction, if accepted, suggests a general criterion for theory comparison: one empirically refined theory can be said to be preserved by its successor even when their formal structures are incommensurable.","The refinement relation between theories could in principle be tested on experimental records by checking whether every pair of regions a special science distinguishes is also distinguished by physics at the same precision."],"forward_implications":["Empirical structure physicalism states physicalism without committing to the ontology of current physics or of any unknown future physics.","If current physics refines the empirical structure of current mental theories, then future physics will too, so mental theories remain empirically dispensable in the future.","Empirical structure realism gives a cumulative account of theory change that can coexist with the pessimistic meta-induction, since it tracks empirical structure rather than entity survival.","Because empirical structure is independent of a theory's ontology, the thesis is compatible with multiple realization, including exotic future computers.","The thesis satisfies most of the standard desiderata for a physicalism thesis and addresses the knowledge-argument scenario by decoupling empirical adequacy from phenomenal knowledge."],"supporting_citations":[{"why":"Formulates Hempel's dilemma against metaphysical physicalism, the target the paper generalizes and answers.","marker":"Crane and Mellor, 1990"},{"why":"Another formulation of the dilemma that the paper treats as the central challenge to theory-oriented physicalism.","marker":"Melnyk, 1997"},{"why":"Original linguistic physicalism version of the dilemma, which the paper contrasts with its empirical-structure approach.","marker":"Hempel, 1980"},{"why":"Supplies the notion of empirical adequacy that the paper generalizes into empirical content and empirical structure.","marker":"van Fraassen, 1980"},{"why":"Extends Hempel's dilemma to any deep-structure and changeable account of experience, motivating the paper's framework.","marker":"Firt et al., 2022"},{"why":"Provides the material theory of induction used to assess the strength of the optimistic meta-induction.","marker":"Norton, 2003"},{"why":"Argues the pessimistic meta-induction lacks a licensing material fact, which the paper uses to contrast its own optimistic induction.","marker":"Shech, 2019"},{"why":"Supports the expectation that future quantum gravity recovers spacetime submanifolds, used in arguing that future locus concepts will supervene on current spacetime structure.","marker":"Wüthrich, 2017"}],"fun_headline_variants":["Empirical structure supervenience resolves Hempel's dilemma for physicalism","Physicalism via empirical structure: a path around Hempel's dilemma","Optimistic meta-induction backs empirical structure physicalism","How theory supervenience on empirical structure escapes Hempel","Empirical structure: the key to physicalism and Hempel's dilemma"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The crucial unproven premise is the historical generalization that up to the present, later physics has refined the empirical structure of earlier physics; the paper explicitly says length limitations prevent it from defending this premise.","fun_headline_variants_meta":{"raw":{"variants":["Empirical structure supervenience resolves Hempel's dilemma for physicalism","Physicalism via empirical structure: a path around Hempel's dilemma","Optimistic meta-induction backs empirical structure physicalism","How theory supervenience on empirical structure escapes Hempel","Empirical structure: the key to physicalism and Hempel's dilemma"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00022,"raw_usage":{"total_tokens":1451,"prompt_tokens":952,"completion_tokens":499,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":568,"completion_tokens_details":{"reasoning_tokens":410}},"tokens_in":568,"tokens_out":499,"duration_ms":5713,"temperature":1.0,"reasoning_tokens":410,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T13:58:27.323812+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A concrete historical case in which a later, generally accepted theory of physics could not empirically distinguish a pair of space-time regions that an earlier accepted theory could distinguish, while no other accepted theory of that later period recovered the ability, would falsify past empirical structure realism and thereby remove the inductive basis for future empirical structure realism.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Formulates Hempel's dilemma against metaphysical physicalism, the target the paper generalizes and answers."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Another formulation of the dilemma that the paper treats as the central challenge to theory-oriented physicalism."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Original linguistic physicalism version of the dilemma, which the paper contrasts with its empirical-structure approach."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the notion of empirical adequacy that the paper generalizes into empirical content and empirical structure."},{"cited_title":"Hemmo, and O","cited_arxiv_id":null,"evidence_quote":"Extends Hempel's dilemma to any deep-structure and changeable account of experience, motivating the paper's framework."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the material theory of induction used to assess the strength of the optimistic meta-induction."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Argues the pessimistic meta-induction lacks a licensing material fact, which the paper uses to contrast its own optimistic induction."}],"review_version":1}