{"id":"8dea1e91-9f5f-4dab-b997-63df17899761","arxiv_id":"2507.04479","paper_version":1,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Anti-exceptionalism about logic is not necessarily committed to scientific realism; antirealist formulations are viable and the realist framing is a contingency.","lead":"This philosophy paper argues that anti-exceptionalism about logic does not have to assume scientific realism: current versions do, but that is a habit, not a requirement. It uses examples from quantum mechanics to show that science itself leaves many choices open, so logic can be tied to science in antirealist ways too.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Most load-bearing concern: the paper has not yet shown that the quoted AEL accounts (abductivism, predictivism) are irreducibly realist; if they can be reconstructed as empirical-adequacy views, the claim that realism is a contingent presupposition of current AEL loses its footing.","rationale":"After careful reading, I think the paper is a modest and well-hedged conceptual contribution. Its strongest claim as summarized is that realism is not a metaphilosophical necessity for AEL. That claim is supported by the flexibility of AEL's continuity thesis and the existence of alternative philosophies of science; the constructive-empiricist sketch, though undeveloped, is a plausible route. The reader's ACCEPT is defensible. However, the most load-bearing step is the historical-interpretive claim that the cited versions of AEL are genuinely realist. The authors support this with quotes, but the quotes do not entail the mind-independence condition in Chakravartty's definition. This is an internal-consistency concern, not an appeal to outside consensus. It does not overturn the paper because the argument is explicitly qualified ('some relevant versions', note 10, 'we have not defended a specific antirealist approach'), and the authors' central goal—demotivating the necessary connection—can survive even if some 'realist' AEL authors can be reinterpreted. Still, the paper would be stronger if it either restricted its claim to authors who explicitly embrace logical realism or provided an argument that abduction/predictivism cannot be combined with antirealism. I therefore keep the reader's ACCEPT verdict unchanged, while flagging the same weakest assumption the reader identified.","tokens_in":13664,"tokens_out":11359,"duration_ms":141367,"concrete_test":"Take Martin and Hjortland's 'Logical Predictivism' as the test case. Rewrite its appraisal criteria using only 'empirical adequacy' and a deflationary truth predicate, deleting every assertion about a mind-independent logical reality, and run it on the paper's own example of classical versus paracomplete logics for the semantic paradoxes. If the rewritten criteria yield the same theory-choice verdicts as the original, then predictivism does not presuppose realism and the central claim must be qualified to a narrower set of authors. If the reconstruction is incoherent or loses the theory-choice outcomes, the concern is settled.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section 2's central interpretive move is that AEL's methodological core—abduction and prediction—'subscribes to a straightforward realist description' of logic and science. The evidence adduced is that these authors distinguish logica docens from logica ens, talk about the 'correct' account of validity, and take the aim of logical theory choice to be truth. This conflates a deflationary or instrumentalist use of 'correct' with full-blooded realism in Chakravartty's sense of truth about a mind-independent world. A constructive empiricist can perfectly well say a logic is appraised by whether it saves the accepted inference patterns and 'predicts' which inferences are accepted, without believing in a unique mind-independent domain of logical facts. Priest's logica ens might mark a subject matter, but it does not settle whether that subject matter is mind-independent. Likewise, Martin and Hjortland's logical predictivism is an epistemology of appraisal, not an ontology. If these reconstructions are viable, the paper's positive exegetical claim—that realism is a 'sociological contingency' currently entrenched in AEL—is not established; realism may be an optional gloss. Section 3's quantum underdetermination does not repair this, because it shows at most that one domain (identity of quantum objects) fails to decide classical vs Schrödinger logics; it does not show that those two logics are underdetermined as global theories of validity, nor that the cited authors' methodology requires realism. The paper's own hedge (note 10; 'some relevant versions') limits the damage but does not identify which current versions are the realist ones.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper argues that anti-exceptionalism about logic (AEL), as formulated in the leading methodological literature centered on abductivism and predictivism, presupposes a unified and consensual scientific methodology and thereby inherits a scientific realist picture of logic. The authors claim that this realist commitment is a sociological contingency of the current debate rather than a metaphilosophical necessity. They support this by drawing parallels with disputes in the metaphysics of science and quantum foundations, where underdetermination at four levels (dynamical, ontological, metaphysical, and logical) is used to show that science does not straightforwardly deliver a unique true theory. They then sketch how a constructive empiricist antirealism could serve as an alternative background for AEL and list several benefits of an antirealist approach to logical theory choice.","tokens_in":13949,"tokens_out":6751,"duration_ms":72704,"significance":"If the central claim holds, the paper is a valuable metaphilosophical contribution: it disentangles a realist metaphysics from the methodological core of AEL, clarifies that continuity with science can be developed along antirealist lines, and opens a recognizable research program for antirealist AEL. The paper is carefully hedged (note 10; conclusion), explicitly credits prior related work by Erickson (2021), and does not overclaim that it has refuted realism or defended a specific antirealist approach. Its strengths are the clear textual engagement with Priest, Hjortland, and Martin, the explicit acknowledgment of alternative AEL formulations, and the constructive proposal to use science in a non-realist way. The main value is in reframing the debate rather than in new technical results, and the significance depends on whether the exegetical claim about current AEL can be made robust.","major_comments":[{"comment":"The central interpretive move, that abductivism and predictivism in AEL are 'openly realist' because they aim at 'the truth about a certain independently existing phenomenon,' is not forced by the quoted texts. Priest's distinction between logica docens and logica ens, and his characterization of logical facts as 'truths of the form that so and so follows from that such and such,' can be read deflationarily; a constructive empiricist can say that a logic is appraised by whether it saves the accepted inference patterns and predicts which inferences are accepted, without believing in a mind-independent domain of logical facts. Since the paper's claim that current AEL 'takes for granted' realism depends on this exegetical step, the authors should either provide direct textual evidence that Priest, Hjortland, and Martin reject antirealist reconstructions, or weaken the claim to the more modest thesis that the leading formulations suggest a realist gloss but do not require it.","section":"Section 2, pp. 5-6"},{"comment":"The quantum cases illustrate underdetermination within a single domain, namely interpretations of quantum mechanics and the identity of quantum objects, but they do not show that classical and Schrödinger logics are underdetermined as global accounts of validity, nor that the underdetermination transfers to logical theory choice in general. The paper's moral that 'AEL shouldn't be so quick in hoping that science is always teaching us clear lessons' (p. 11) is too strong: AEL's methodological claim only requires that science has a recognizable theory-choice process, not that it resolves all underdetermination. Note 10 already concedes that underdetermination is not a knockdown argument, but the section is presented as an instructive parallel; its scope should be stated more carefully, and the transfer from quantum-specific logical underdetermination to AEL's general theory choice should be argued explicitly.","section":"Section 3, pp. 8-11"},{"comment":"The paper's target is described in the abstract as 'some relevant versions' of AEL but in Section 2 as 'most of current literature.' Footnote 6 itself mentions Arenhart (2022), Tajer (2022), and Commandeur (2023) as non-realist AEL formulations, which undermines the generality of the central claim unless those alternatives are shown to be outside the relevant class. The authors should specify their target class (for example, methodological AEL centered on theory choice) and either engage with the cited alternatives or explicitly bracket them; otherwise the 'sociological contingency' claim is overgeneralized.","section":"Section 2, n. 6 and abstract"}],"minor_comments":[{"comment":"In the Priest block quote, 'logic docens' should be 'logica docens' for consistency with the surrounding text.","section":"p. 4"},{"comment":"In the box-statistics illustration, 'One indistinguishable particles is inside each box' should be 'One indistinguishable particle is inside each box.'","section":"p. 9"},{"comment":"Consider moving the discussion of non-realist AEL alternatives into the main text, since it bears directly on the scope of the paper's claim.","section":"p. 6, n. 6"},{"comment":"The claim that giving up the true logic 'deflates' the monism/pluralism debate is asserted in one sentence; a brief illustration would make the benefit concrete.","section":"p. 13"},{"comment":"The phrase 'correctly predicts what is actually valid' is ambiguous between a realist reading (validity as a mind-independent fact) and an antirealist reading (validity as what is accepted in a practice); the authors should disambiguate, since this ambiguity motivates the first major comment.","section":"p. 5"}],"recommendation":"major_revision","confidential_remarks":"This manuscript is an invited contribution to a Springer volume and has apparently already been through an editorial process. My recommendation is based on the philosophical content. The self-citations are to relevant prior work by the authors and are not excessive. The main risk is that the paper overgeneralizes from a small set of AEL authors; this is addressable by the major comments, which ask for either stronger evidence for the realist reading of the quoted authors or a more carefully qualified central claim."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper does a clean job making a modest point: methodological versions of AEL (abductivism, predictivism) carry a realist flavor, but that link is not forced. The quantum parallel in Section 3 is the most useful part, because it concretely shows science not delivering unique answers at several levels, including the logic of identity. The authors are careful with hedges, explicitly disclaiming a knockdown argument and any defense of a specific antirealist position.\n\nTheir main contribution is framing realism as a sociological contingency rather than a metaphilosophical necessity. That framing is fair, but not brand new; they cite Erickson (2021) and earlier non-realist AEL work by themselves and others. What they add is the detailed underdetermination story and a clear constructive-empiricist alternative sketch.\n\nThe soft spot is the interpretive step in Section 2. They move from \"AEL talks about the correct logic and logica ens\" to \"AEL is committed to a correspondence-style realism about a mind-independent logical realm.\" A constructive empiricist can talk about correctness or \"what follows from what\" without believing in a unique mind-independent domain of logical facts. The paper leans on Priest and Hjortland quotes, but does not show those authors are realists in the Chakravartty sense. So the claim that realism is currently entrenched in most AEL is plausible but not established. The paper's hedge (\"some relevant versions\") cushions this, but it also makes the target less precise.\n\nSection 3's underdetermination does not fully repair this. The logical underdetermination they find is about reflexivity of identity in quantum objects, not about the global choice of logical consequence. It illustrates that science does not always decide logic, but it does not show that the cited AEL methodology requires realism. So the bridge is a bit loose.\n\nStill, the central message holds: if you want to bring logic closer to science, you do not have to be a scientific realist. That is worth saying clearly, and the paper says it carefully, with honest attention to what it does and does not prove.\n\nWho is this for? Philosophers of logic and anyone working on AEL. A general reader gets less out of it, but the quantum section is accessible enough. It deserves peer review and likely acceptance: it is honest, well-written, and makes a legitimate conceptual point, even if the novelty is partially anticipated. A referee should press for clarification on the realism attribution, but the paper is solid enough to be published after minor revision.","headline":"A useful, carefully hedged conceptual paper that makes a modest point about AEL and realism, with the quantum underdetermination parallel as its best addition; the exegetical claim about current AEL's realist commitments is plausible but under-supported.","tokens_in":14542,"tokens_out":3452,"would_cite":false,"duration_ms":36597,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Anti-exceptionalism about logic does not require scientific realism.","keywords":["anti-exceptionalism about logic","logical epistemology","scientific realism","antirealism","constructive empiricism","underdetermination","quantum mechanics","logical theory choice"],"falsifier":"A direct refutation would be to exhibit a formulation of AEL in the current literature that explicitly combines anti-exceptionalism with an antirealist philosophy of science, such as constructive empiricism, while still fitting the paper's own characterization of AEL; alternatively, showing that the quantum-mechanical cases of underdetermination do not transfer to logical theory choice would undermine the lesson.","tokens_in":13455,"feed_emoji":"⚛️","tokens_out":4583,"duration_ms":42665,"temperature":0.7,"pith_summary":"Anti-exceptionalism about logic (AEL) is the view that logical theories are chosen much like scientific theories, by abduction or by predictive success. This paper argues that the main current formulations of AEL quietly assume a realist picture of science: a unified method that aims at the true description of independently existing logical facts. It then argues that this realism is a sociological contingency of the literature, not a metaphilosophical necessity. Drawing on four levels of underdetermination in quantum mechanics, the paper shows that science itself does not supply a unique theory-choice method. It concludes that AEL can be rebuilt on antirealist foundations, such as constructive empiricism, and sketches what that would change.","feed_headline":"Logic's anti-exceptionalism need not inherit realism","feed_subtitle":"Quantum mechanics shows science yields no unique theory choice, so anti-exceptionalism could rest on antirealist foundations.","key_machinery":"The central device is the parallel between the metaphysics of science and the epistemology of logic, illustrated by a four-level ladder of underdetermination in quantum mechanics: dynamical (multiple quantum interpretations fit the same evidence), ontological (different accounts of what exists), metaphysical (individuality vs non-individuality of quantum objects), and logical (classical vs non-reflexive Schrödinger logics and quasi-set theory). Each level shows that the physics itself cannot settle the choice, and the logical level demonstrates the same underdetermination for logical theory choice, thereby demotivating the realist assumption that science delivers a uniquely true theory.","core_discovery":"The paper's central claim is that anti-exceptionalism about logic, as defended in the abduction-based versions of Priest and Hjortland and the prediction-based version of Martin and Hjortland, presupposes a broadly realist description of both science and logic: logical theories are candidates for the true description of validity, and theory choice aims at truth. Against this, the paper argues that the continuity between logic and science can be understood in radically different ways. The four levels of underdetermination found in quantum mechanics—the underdetermination of interpretation by physics, of ontology by interpretation, of metaphysics by physics, and of logic by both—show that science does not offer a single, consensus methodology. Since realism is just one philosophical gloss on science, the link from anti-exceptionalism to realism is not forced.","pith_inferences":["A natural extension, not developed in the paper, would be to specify what counts as the empirical adequacy of a logical theory independently of its truth.","The underdetermination argument could be tested on other scientific domains, such as chemistry or biology, to see whether a logical level of underdetermination transfers there as well.","If realism is indeed a contingency, the burden of proof shifts to realist anti-exceptionalists to justify why realism should be the default rather than a choice.","The paper's own footnote on underdetermination responses suggests that even a weaker, non-knockdown version of the point could still support the contingency thesis."],"forward_implications":["If AEL is disentangled from realism, anti-exceptionalists can appeal to a broader menu of philosophies of science, including constructive empiricism.","Logical theory choice can be reframed in terms of empirical adequacy rather than truth, so that multiple logics can be accepted without claiming to be the one true logic.","The debate between logical monism and pluralism is deflated, since the antirealist emphasis on understanding rather than truth removes the need to pick a single correct system.","Abandoning the 'one true logic' goal removes the need for an inaccessible realm of logical facts, such as Priest's logica ens.","Since quantum mechanics itself is underdetermined at multiple levels, appeals to science as a source of clear theory-choice lessons are weakened."],"supporting_citations":[{"why":"Provides the canonical characterization of AEL as the view that logic is continuous with science, which the paper argues presupposes realism.","marker":"Hjortland (2017)"},{"why":"Introduces the logica docens / logica ens distinction and the realist idea of logical facts that the paper identifies as a realist commitment.","marker":"Priest (2014)"},{"why":"Supplies the abductivist account of logical theory choice, one of the two main targets of the paper's argument.","marker":"Priest (2016)"},{"why":"Formulates the methodological version of AEL and logical predictivism, which the paper argues shares the same realist background.","marker":"Martin and Hjortland (2021)"},{"why":"Presents logical predictivism as the most detailed version of methodological AEL, directly relevant to the paper's critique.","marker":"Martin and Hjortland (2022)"},{"why":"Provides the constructive empiricist alternative that the paper suggests as a possible antirealist foundation for AEL.","marker":"van Fraassen (1980)"},{"why":"Establishes metaphysical underdetermination for quantum objects, which the paper adapts to motivate logical underdetermination.","marker":"French and Krause (2006)"},{"why":"Supplies the general recipe for realism that the paper uses to characterize the realist assumption in AEL.","marker":"Chakravartty (2017)"}],"fun_headline_variants":["Anti-exceptionalism in logic can do without realism","Quantum underdetermination frees logic from realism","Realism not a must for logic's anti-exceptionalism","Separating logic's anti-exceptionalism from realism"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument assumes that the abduction-based versions of Priest and Hjortland and the prediction-based version of Martin and Hjortland are representative of most current AEL literature and that these versions genuinely presuppose realism; if those authors can be reconstructed without realist commitments, the central claim weakens.","fun_headline_variants_meta":{"raw":{"variants":["Anti-exceptionalism in logic can do without realism","Quantum underdetermination frees logic from realism","Realism not a must for logic's anti-exceptionalism","Separating logic's anti-exceptionalism from realism"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000157,"raw_usage":{"total_tokens":1144,"prompt_tokens":787,"completion_tokens":357,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":403,"completion_tokens_details":{"reasoning_tokens":290}},"tokens_in":403,"tokens_out":357,"duration_ms":4052,"temperature":1.0,"reasoning_tokens":290,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T19:46:54.689326+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A direct refutation would be to exhibit a formulation of AEL in the current literature that explicitly combines anti-exceptionalism with an antirealist philosophy of science, such as constructive empiricism, while still fitting the paper's own characterization of AEL; alternatively, showing that the quantum-mechanical cases of underdetermination do not transfer to logical theory choice would undermine the lesson.","supporting_citations":[{"cited_title":"Scientific Realism without the Quantum","cited_arxiv_id":null,"evidence_quote":"Establishes metaphysical underdetermination for quantum objects, which the paper adapts to motivate logical underdetermination."}],"review_version":1}