{"id":"3c6615dc-dcb9-48d8-9c8b-d62cb3de142d","arxiv_id":"1908.10612","paper_version":6,"verdict":"ACCEPT","confidence":"LOW","novelty_score":3.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A comprehensive survey of scalar curvature geometry, connecting Dirac operator index theory, minimal hypersurfaces, Ricci flow, and Seiberg-Witten invariants, with many open problems.","lead":"This paper is Misha Gromov's extended lecture notes on the geometry and topology of manifolds whose scalar curvature is bounded below. It reviews the known tools and results and adds several new conjectures and reformulations, serving as a field guide for researchers","discovery_kind":"review","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Reliance on unverified desingularization theorems for n≥8 threatens the accuracy of the survey's map of known scalar-curvature results.","rationale":"The reader's weakest assumption identifies the same load-bearing concern: the survey's claims for n≥8 rely on desingularization theorems the author has not verified. My good-faith reading confirms that the central thesis—the map of known limits and the role of two analytic means—does not logically require these theorems to be true to be meaningful, but the paper's accuracy as a guide to 'what is known' would diminish if those proofs fail. The author's explicit caveats are honest and partially mitigate the risk, so the ACCEPT verdict with LOW confidence remains appropriate. However, the concern is real and testable: a literature review or an attempted reproduction of the desingularization on a simple singular model would settle whether the n≥8 results belong in the 'known' category. Since the reader already flagged this exact assumption and the author already disclosed it, I do not propose to change the verdict, but I emphasize the concrete check that would raise or lower confidence.","tokens_in":64474,"tokens_out":13753,"duration_ms":149360,"concrete_test":"Check the peer-review status and independent validation of [Lohkamp(smoothing) 2018] and [SY(singularities) 2017]: (1) have these papers appeared in refereed journals, and (2) have subsequent authors successfully applied their desingularization to stable minimal hypersurfaces of dimension ≥7? If independent confirmation is lacking, attempt to reproduce the key smoothing step for a local model of an isolated stable singularity in dimension 8; if the construction cannot yield a smooth hypersurface with the required curvature lower bound, then the dimension-free statements in §§2.7, 3.1, 5.2, 5.10 should be reverted to the n≤7 barrier.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The survey's value rests on its accurate map of what is known versus open in scalar curvature geometry. Several definitive claims in dimensions n≥8 depend on desingularization theorems for stable minimal hypersurfaces that Gromov explicitly says he has not mastered (Section 1.5, footnote 26: 'I can't vouch for the proof'; Section 1.6.2: 'arguments in both papers are difficult and I have not mastered them'; Section 3.1 remark (c)). These theorems remove the classical n≤7 barrier in key results: the Schoen-Yau torus non-existence for all n, µ-bubble arguments, and the ∎-inequality for n≥9. If those proofs contain gaps, the dimension restrictions return and the survey overstates what is known. The abstract claims 'some new geometric constraints', and the introduction's central thesis asserts the limits of scalar-curvature flexibility are determined by two analytic means; the reliability of the n≥8 portion is thus load-bearing for the paper's central claim. The author's disclaimers mitigate but do not remove this risk.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper is an expanded lecture-notes survey, in seven sections, of the geometry and topology of Riemannian manifolds with scalar curvature bounded below. It argues that the apparent flexibility of such manifolds is limited by two analytic tools: index theory for Dirac operators and geometric measure theory via minimal hypersurfaces and stable µ-bubbles. The paper surveys classical results (Lichnerowicz, Hitchin, Schoen–Yau, Gromov–Lawson, Llarull, and others), sketches proof strategies, and formulates many open problems and conjectures. It also contains a number of statements described as new, such as comparison inequalities for cubical and ball-like domains, Sc-normalized extremality theorems, and refinements involving warped products and µ-bubbles.","tokens_in":64608,"tokens_out":6968,"duration_ms":78213,"significance":"If the survey's map of the field is accurate, it will be a valuable reference for researchers in scalar curvature geometry and adjacent fields. Its strengths include explicit labeling of many statements as conjectural or provisional, candid disclaimers where the author has not personally verified technical proofs (for example, Section 1.5 footnote 26, Section 2.7 remark (c), and Section 3.1 remark (c)), and a substantial amount of self-contained background on curvature formulas, Clifford algebras, and Dirac operators. The paper also proposes many concrete open problems and conjectures, which gives it utility beyond its expository content.","major_comments":[{"comment":"Several central statements about manifolds with Sc > 0 are presented without dimension restrictions, even though the proofs given in the text cover only n ≤ 7 (or n ≤ 9 in one Lipschitz statement) and the removal of the n ≤ 7 barrier depends on desingularization results in [SY(singularities) 2017] and [Lohkamp(smoothing) 2018] that the author says he has not studied in depth. For example, Section 2.7 states the torus corollary — a closed orientable n-manifold mapping with non-zero degree to T^n admits no metric with Sc > 0 — as an unconditional statement, while the author's own caveats in Section 1.5 footnote 26 and Section 2.7(c) indicate that the n ≥ 8 case rests on papers the author cannot vouch for. Since the survey's central deliverable is an accurate map of what is known versus what is open, each such statement should carry an explicit qualifier such as \"unconditional for n ≤ 7; for n ≥ 8 conditional on the desingularization theorems of Schoen–Yau and Lohkamp,\" or at least a back-reference to the caveat. This is load-bearing because the introduction's thesis is meant to cover the higher-dimensional cases as well.","section":"§1.5(ii), §2.7, §3.1(c)"},{"comment":"The abstract and introduction claim that the paper presents \"some new geometric constraints,\" but no passage explicitly lists which statements are new to this paper as opposed to expository. This makes it difficult to evaluate the paper's novelty claim and to check whether the new material is correctly attributed. Please add a short paragraph or table in the introduction identifying the statements that the author regards as new (for example, the ∎-inequality and its comparison versions, the Sc-normalized convex area extremality theorem, and the µ-bubble estimates), and state for each whether it is proved in full, proved only as a sketch, or conjectural.","section":"Introduction/Abstract"},{"comment":"The provisional proposition on profinitely hyperspherical spin manifolds is explicitly described as \"convincing but it is not quite a proof,\" and the later \"conclusion of the proof\" in Section 3.3.2 relies on an \"obvious smoothing\" that is asserted to make the correction term Δ◽_ε tend to zero \"in the strongest conceivable sense.\" As written, the relevant analytic estimates are not supplied, so the proposition remains a proof sketch rather than a complete proof. Since this proposition underpins the Dirac-theoretic half of the paper's central thesis, the text should either provide the missing estimates or explicitly mark the proposition, and all later uses of it, as a proof sketch rather than a fully proved theorem.","section":"§3.3 and §3.3.2"}],"minor_comments":[{"comment":"The sentence \"this needs additional analytical work to be extended to n ≥ 7\" appears to be a typo: the stated theorem is for n ≤ 7, so the extension issue concerns n ≥ 8.","section":"§1.6.4"},{"comment":"The list of conditions for reflection orbifolds is numbered ●1, ●4, and ●<2, with the intermediate numbers missing; this appears to be a numbering or typesetting error that should be corrected.","section":"§3.1.1"},{"comment":"There are numerous small typos and typesetting artifacts, such as \"Poof of ∆-Lemma\" in Section 2.9, \"Metics\" in the title of Section 3.17, and repeated words such as \"the the\". A careful copy-edit would materially improve readability.","section":"Throughout"},{"comment":"The citation style uses abbreviated labels such as [G(inequalities) 2018] and [SY(singularities) 2017], and some cross-references appear as \"??\". Please ensure that the bibliography is complete and that all internal references resolve correctly; a notation index for the ad-hoc symbols (∎, ☀, /enc-99) would also help readers.","section":"References and notation"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is clearly an arXiv lecture-notes text with a non-standard citation scheme and many Unicode artifacts; the journal production process will need significant copy-editing. The main scientific concern is not the correctness of the standard theorems surveyed, but the status labeling of several n ≥ 8 statements that are presented as unconditional despite the author's own disclaimers. I would not reject on the present grounds, but the manuscript needs a careful revision to distinguish fully proved results, proof sketches, and conditional claims."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague—\n\nYou should know this before anything else: this is Gromov's survey of scalar curvature geometry, deliberately a map, not a theorem-proving paper. The reader report says ACCEPT with low confidence, and I think that is about right. There is no single new theorem here, but the paper earns its place by synthesizing the two analytic pillars—Dirac/index theory and minimal hypersurfaces/µ-bubbles—and by stating dozens of open problems and conjectures that will shape the field.\n\nWhat it does well: the exposition is unusually honest. Gromov repeatedly flags what is proved, what is delegated, and what he has not personally mastered. The provisional proposition in §3.3 is explicitly marked \"convincing but not quite a proof.\" The n≥8 dimension restrictions come with direct caveats—\"I can't vouch for the proof\"—about Lohkamp's and Schoen–Yau's desingularization papers. That candor is real credit, and it is the right way to write a survey that includes recent results.\n\nThe soft spots are real but should not be overstated. The survey's map of what is known in n≥8 does rest on those desingularization theorems, and if those proofs had gaps, several stated theorems would shrink back to n≤7 or to the spin case. Gromov himself says he has not mastered those arguments. That is a genuine epistemic limitation, but it is openly disclosed, and the survey's central thesis—scalar curvature bounds are controlled by index theory and geometric measure theory—does not collapse even if the n≥8 statements weaken. For n≤7 and for spin manifolds in all dimensions, the picture is solid. The 'new' elements are mostly reformulations and conjectures; novelty is modest, but that is normal for an invited lecture series.\n\nOne thing the stress-test gets right: the abstract's phrase 'some new geometric constraints' overreaches if read as a claim of new theorems. Read as 'constraints described in a new way,' it is fine. Self-citation is heavy, but in a survey by the field's leading figure it is appropriate, and the cited results are standard.\n\nWho is this for? Graduate students and researchers who want a connected view of scalar curvature geometry before reading the original papers. It deserves a serious referee: not because it proves a crisp new result, but because accuracy of attribution and of the known/open boundary in a survey of this scope genuinely matters. I'd recommend accepting it, with a request that the author keep the caveats prominent and perhaps soften the abstract's novelty claim. The n≥8 dependence should be highlighted in a footnote, not hidden.\n\nFor peer review: yes, engage. It's not desk-reject material.","headline":"A candid survey of scalar curvature geometry that earns its keep through honest caveats, not new theorems; the n≥8 dependence is real but declared.","tokens_in":65138,"tokens_out":1706,"would_cite":true,"duration_ms":20026,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["53C21","53C27","53C42","58J20"],"pacs":[],"model":"deepseek-v4-flash","headline":"A four-lecture survey argues that scalar curvature bounds leave topology flexible but impose definite limits, and that all known limits are set by Dirac index theory and geometric measure theory.","keywords":["scalar curvature","positive scalar curvature","Dirac operator","index theory","minimal hypersurfaces","mu-bubbles","warped products","geometric measure theory"],"falsifier":"Find a codimension-one area-minimizing hypersurface in an 8-dimensional manifold, satisfying the hypotheses of the cited theorems, whose singular set cannot be desingularized while preserving the scalar-curvature descent; such an example would restore the old dimension restrictions and break the all-dimension formulations.","tokens_in":64229,"feed_emoji":"📐","tokens_out":14271,"duration_ms":143013,"temperature":0.7,"pith_summary":"Positive scalar curvature does not force a manifold into a rigid shape: connected sums, thin surgeries, and toral stabilizations preserve it, so the topological landscape is wide and flexible. The paper's thesis is that this flexibility has definite limits, and that in every known case the limit is established by one of two analytic engines: index theory of Dirac operators and geometric measure theory via stable minimal hypersurfaces and $\\mu$-bubbles. Beyond being a survey, the lectures state sharp local and quantitative inequalities produced by these engines, including dihedral-angle criteria for $\\mathrm{Sc}\\ge 0$, band-width bounds, and scalar-curvature-normalized area estimates for maps. A sympathetic reader should come away seeing scalar-curvature geometry as a subject with a small number of load-bearing techniques and a large unresolved landscape, centered on the conjecture that no non-torsion homology class of an aspherical classifying space is dominated by a compact manifold with $\\mathrm{Sc}>0$.","feed_headline":"Two tools explain the known limits of positive scalar curvature","feed_subtitle":"Dirac index theory and minimal hypersurface descent set which manifolds can admit positive scalar curvature.","key_machinery":"The two load-bearing objects are the twisted Dirac operator and the stable $\\mu$-bubble. The Dirac side is the identity $D^2=\\nabla^*\\nabla+\\tfrac14\\mathrm{Sc}$ (the S-L-W-B formula): it makes positive scalar curvature and harmonic spinors mutually exclusive, so any nonzero index, carried for example by an almost flat unitary bundle induced by an $\\varepsilon$-Lipschitz map, becomes a topological obstruction to $\\mathrm{Sc}>0$. The geometric-measure-theory side is the second variation formula for stable minimal hypersurfaces $Y\\subset X$, $$\\int_Y |d\\psi|^2+\\tfrac12\\bigl(\\mathrm{Sc}_Y-\\mathrm{Sc}_X-|A|^2\\bigr)\\$psi^{2}$\\ge 0,$$ which for $\\mathrm{Sc}_X>0$ yields positivity of $-\\Delta+\\tfrac12\\mathrm{Sc}_Y$; via the Kazdan-Warner conformal change this gives $\\mathrm{Sc}>0$ on $Y$, and via the warped-product symmetrization $Y^\\rtimes=Y\\times T^N$ with metric $h+\\sum_i\\phi_i^2\\,dt_i^2$ it converts an eigenfunction of $-\\Delta+\\tfrac12\\mathrm{Sc}$ into a metric on the stabilized manifold with $\\mathrm{Sc}=2\\lambda$. The $\\mu$-bubble generalization prescribes $\\mathrm{mean.curv}=\\mu$, adapting the descent to bands, boundaries, and corners.","core_discovery":"On its own terms, the paper's central claim is structural: although scalar curvature is the weakest of the three curvature notions, an average rather than a control of sectional curvatures, the conditions $\\mathrm{Sc}\\ge\\sigma$ and $\\mathrm{Sc}>0$ genuinely constrain geometry and topology, and every understood constraint funnels through two mechanisms. The first is the Schr\\\"odinger-Lichnerowicz-Weitzenb\\\"ock-Bochner identity $D^2=\\nabla^*\\nabla+\\tfrac14\\mathrm{Sc}$ together with the Atiyah-Singer index theorem: a nonzero index forces a harmonic spinor, while $\\mathrm{Sc}>0$ forbids one. The second is the second-variation calculus of stable minimal hypersurfaces and $\\mu$-bubbles, which descends positivity of scalar curvature to hypersurfaces and, after conformal or warped-product modification, to lower-dimensional witnesses. The paper presents the resulting theorems: torus and aspherical non-existence, domination obstructions, sharp map inequalities, band-width bounds, and rigidity under $\\mathrm{Sc}\\ge0$, as pieces of one picture, and states the $[\\mathrm{Sc}\\not>0]$ principle as its main conjecture.","pith_inferences":["The local dihedral-angle criterion suggests testing whether $\\mathrm{Sc}\\ge0$ can be defined purely metrically, for continuous or Lipschitz metrics, by the absence of arbitrarily small mean-convex cubical domains with acute dihedral angles; the paper itself raises this as an open problem.","The warped $T^N$-symmetrization hints at a reduction principle: scalar-curvature questions on invariant toral extensions reduce to the base, which could be checked on the paper's unresolved example of a warped product over a hyperbolic 3-manifold with $\\mathrm{Sc}\\ge-6$.","If the two-tools thesis is correct, progress on the main domination conjecture should come either from new index-theoretic cycles or from extending minimal-hypersurface regularity; a genuinely third mechanism would not just prove a new theorem but change the map of the subject."],"forward_implications":["Uniform limits of smooth metrics with $\\mathrm{Sc}\\ge\\sigma$ again have $\\mathrm{Sc}\\ge\\sigma$ when the limit is smooth; the inequality is C$^0$-closed, while metrics with $\\mathrm{Sc}\\le\\sigma$ are C$^0$-dense.","Closed manifolds admitting a nonzero-degree map to the $n$-torus, and more generally the classes detected by $\\mu$-bubble descent and the quasi-symplectic class, carry no metric with $\\mathrm{Sc}>0$; with the cited desingularization inputs this holds in all dimensions, not only $n\\le7$.","Complete non-compact manifolds are also constrained: the torus is not dominated by any complete manifold with $\\mathrm{Sc}>0$, so ends and punctured subdomains obstruct positive scalar curvature.","Quantitative map estimates follow: for spin manifolds with $\\mathrm{Sc}>0$, maps to spheres and to convex hypersurfaces cannot be too length- or area-contracting once metrics are normalized by scalar curvature, yielding extremality and rigidity statements.","Rigidity is inherited from non-existence: where $\\mathrm{Sc}>0$ is impossible, the borderline case $\\mathrm{Sc}\\ge0$ often forces the manifold to be flat, via Kazdan's deformation theorem and splitting arguments."],"supporting_citations":[{"why":"Supplies the S-L-W-B identity and the first Dirac-operator obstruction: Sc > 0 rules out harmonic spinors.","marker":"[Lichnerowitz(spineurs harmoniques) 1963]"},{"why":"Introduces the minimal-hypersurface descent and proves low-dimensional non-existence for manifolds mapping with nonzero degree to tori.","marker":"[SY(structure) 1979]"},{"why":"Adds twisted Dirac operators and surgery, giving spin non-existence in all dimensions and enlargability results.","marker":"[GL(spin) 1980]"},{"why":"Extends the arguments to complete manifolds and introduces the warped toral symmetrization used throughout.","marker":"[GL(complete) 1983]"},{"why":"Provides existence and regularity of area-minimizing hypersurfaces on which the descent is run.","marker":"[Federer(singular) 1970]"},{"why":"Gives the conformal-change theorem converting positivity of a conformal Laplacian into a metric with Sc > 0.","marker":"[Kazdan-Warner(conformal) 1975]"},{"why":"Supplies a desingularization of minimal hypersurface singularities in dimensions at least 8, removing the n <= 7 restriction.","marker":"[SY(singularities) 2017]"},{"why":"Provides an alternate smoothing/desingularization that supports the same high-dimensional extensions.","marker":"[Lohkamp(smoothing) 2018]"}],"fun_headline_variants":["Dirac index and minimal hypersurfaces constrain scalar curvature","Two mechanisms bound positive scalar curvature","How Dirac operators and bubbles limit positive curvature","The twin pillars of scalar curvature constraints"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"Several stated theorems in all dimensions depend on the desingularization theorems for minimal hypersurfaces in dimensions $n\\ge8$, whose proofs the author says he has not verified in depth; if those proofs fail, the dimension restrictions return.","fun_headline_variants_meta":{"raw":{"variants":["Dirac index and minimal hypersurfaces constrain scalar curvature","Two mechanisms bound positive scalar curvature","How Dirac operators and bubbles limit positive curvature","The twin pillars of scalar curvature constraints"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000711,"raw_usage":{"total_tokens":3128,"prompt_tokens":800,"completion_tokens":2328,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":416,"completion_tokens_details":{"reasoning_tokens":2273}},"tokens_in":416,"tokens_out":2328,"duration_ms":16211,"temperature":1.0,"reasoning_tokens":2273,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T10:37:47.319865+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Find a codimension-one area-minimizing hypersurface in an 8-dimensional manifold, satisfying the hypotheses of the cited theorems, whose singular set cannot be desingularized while preserving the scalar-curvature descent; such an example would restore the old dimension restrictions and break the all-dimension formulations.","supporting_citations":[],"review_version":1}