{"id":"cad78f8a-cb5c-477c-bb14-66f1a2e8bef8","arxiv_id":"1908.01659","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"The variety of positive S4-algebras has a finite free algebra in one generator but is not locally finite, and exactly three nontrivial structurally complete varieties of positive K4-algebras exist.","lead":"This math paper studies a simplified version of modal logic that drops negation, and it classifies which of its algebraic varieties are structurally complete. It finds that only three nontrivial families qualify, a surprising contrast with ordinary modal logic.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Theorem 5.1's free-algebra computation is delegated to an unreported UAC check, and the printed term set duplicates □3□x; the input is therefore ambiguous. Recomputing C/Cg(Γ) and comparing it with Figure 1 should settle whether the classification rests on solid ground.","rationale":"The reader's weakest_assumption identifies precisely the step I find most load-bearing: the exact shape of the free one-generated positive S4-algebra, established in Theorem 5.1 through an unshown finite computation. My stress-test confirms that this is not a superficial gap. Every subsequent structural result, including the list of one-generated subdirectly irreducible algebras, the bottom of the subvariety lattice, and the structural-completeness trichotomy, depends on Figure 1 being the correct free algebra. Independently of the reader's point, the paper as printed contains an ambiguity in Equation (4): the term set Σ is supposed to have seven generators, yet as typeset it duplicates □3□x and does not include 3□3x, even though the surrounding proof and the Fact 5.2 diagram require seven distinct terms. This makes the UAC input ambiguous on its face and strengthens the need for a reproducible computation. I do not see a separate circularity or an internal inconsistency in the main classification arguments: the proofs are detailed, the use of Jónsson's lemma and the structural-completeness characterizations is coherent, and the trichotomy is plausible. The appropriate posture remains conditional acceptance: the central claim is likely correct, but the deferred computation and the term-list typo need to be fixed and independently checked before full acceptance. My agreement with the reader is complete on the weakest assumption; the typo is a concrete manifestation of the same reproducibility concern.","tokens_in":30170,"tokens_out":18203,"duration_ms":142902,"concrete_test":"Reproduce the computation underlying Theorem 5.1 with a corrected seven-term list, e.g. Σ = {x, □x, 3□x, □3□x, 3x, □3x, 3□3x}. Run the Universal Algebra Calculator (or an independent lattice algorithm) on the free 7-generated bounded distributive lattice modulo Γ = {⟨φ,ψ⟩ : φ(b) ≤ ψ(b) in Fact 5.2's diagram}, and compare the resulting quotient's Hasse diagram, size, and order with Figure 1. Additionally, verify each relation in Fact 5.2 semantically in an arbitrary positive S4-algebra. If the quotient matches Figure 1 with the corrected Σ, the conditional reservation is resolved; if it differs, the one-generated SI list in Section 6 and hence the trichotomy in Theorem 9.7 must be re-examined.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The classification of structurally complete positive K4-varieties (Theorem 9.7) inherits its one-generator picture from Theorem 5.1: the eleven one-generated subdirectly irreducible algebras in Section 6 are read off Figure 1, and both the cover analysis (Section 8) and the exclusion lemmas 9.4–9.5 use that list. The proof of Theorem 5.1, however, delegates the key step to a computation: it asserts that the bounded lattice reduct of the free one-generated positive S4-algebra is exactly the quotient C/Cg(Γ) of the free 7-generated distributive lattice, where Γ encodes the order relations of Fact 5.2, and says this 'can be checked mechanically, e.g. using the Universal Algebra Calculator [24]'. No input, output, or independent verification is supplied. The printed data are also internally inconsistent: Equation (4) lists seven slots but contains □3□x twice and omits 3□3x, although the tuple following it and the Fact 5.2 diagram require seven distinct terms including 3□3x (or some correction). With the input ambiguous and the computation unreported, the one-generator free algebra is not independently verifiable from the paper. Since this free algebra is the scaffolding for the classification, an error in Figure 1 or in the quotient would propagate directly into Theorem 9.7. This is a genuine conditional-acceptance concern, not a defect in the conceptual architecture: the surrounding proofs are detailed and the main trichotomy is internally coherent. Fact 5.2, which supplies the order relations, is also stated without proof, so both the relations and the subsequent quotient computation are load-bearing.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies positive modal algebras, i.e. the ⟨∧,∨,□,3,0,1⟩-subreducts of modal algebras, with emphasis on positive K4- and S4-algebras. It proves that the variety of positive S4-algebras is not locally finite although its free one-generated algebra is finite (Theorem 5.1, Corollary 5.4), describes the bottom of the subvariety lattice of positive S4-algebras up to height 4 (Theorem 8.6), and uses this information to classify structurally complete varieties of positive K4-algebras. The central classification result (Theorem 9.7) states that a nontrivial variety of positive K4-algebras is structurally complete if and only if it is hereditarily structurally complete, and this happens exactly for V(B2), V(C2), and V(D4). A companion characterization of passively structurally complete varieties is given in Theorem 9.8. The paper also provides algebraic and duality-theoretic tools, including a study of well-connected positive S4-algebras and splitting algebras.","tokens_in":30537,"tokens_out":8296,"duration_ms":83642,"significance":"If the results are correct, they give a remarkably rigid picture of structural completeness in the positive fragment: unlike the full-signature case, where there are infinitely many hereditarily structurally complete varieties of K4-algebras, the positive K4 case admits exactly three nontrivial structurally complete varieties. The paper contains substantial original work: the finiteness of the one-generated free positive S4-algebra, the description of the bottom of the subvariety lattice, and the structural-completeness trichotomy. The proofs are generally detailed, and the main conceptual architecture is coherent; no circularity is apparent, and the paper openly imports the external characterization Theorem 1.1. The main reservation concerns the load-bearing verification of the free one-generated algebra in Theorem 5.1, which is delegated to an unshown mechanical computation with ambiguous input data. Because the classification of Section 9 inherits the one-generator picture, this issue must be resolved before the central claims can be fully certified.","major_comments":[{"comment":"The proof of Theorem 5.1 delegates the crucial identification of the bounded lattice reduct of the free one-generated positive S4-algebra with the quotient C/Cg(Γ) to the statement that this 'can be checked mechanically, e.g. using the Universal Algebra Calculator [24]'. No input, output, or independent verification is supplied. This quotient and the accompanying Figure 1 are the scaffolding for the rest of the paper: the list of eleven one-generated subdirectly irreducible algebras in Section 6, the cover analysis in Section 8, and the exclusions in Lemmas 9.4–9.5 all depend on it. The authors should provide a verifiable certificate for the computation, for instance the UACalc input and output, or replace the mechanical check with a hand proof.","section":"§5, Theorem 5.1"},{"comment":"The printed term set in Equation (4) is internally inconsistent: it lists seven slots but contains □3□x twice and omits 3□3x. However, Fact 5.2 and the tuple used immediately after the equation require seven distinct terms, including 3□3x. This ambiguity directly affects the definition of Γ and hence the claimed quotient C/Cg(Γ). The set Σ and all subsequent uses of the tuple (x, □x, 3□x, □3□x, 3x, □3x, □3□x) must be corrected to a consistent list of seven distinct generators.","section":"§5, Equation (4)"},{"comment":"The assertion that there are exactly eleven one-generated subdirectly irreducible positive S4-algebras is justified by 'inspection' of Figure 1. This is acceptable only if Figure 1 is known to be correct. Given the unresolved verification of Theorem 5.1 and the ambiguity in Equation (4), the reader cannot independently reproduce the list in Figure 2. After fixing the generator set and the quotient computation, the authors should confirm that Figure 2 and the associated cover assertions remain unchanged.","section":"§6, Figure 2"}],"minor_comments":[{"comment":"In the sentence 'Clearly f preserves 0 and 1, since t0 = 0 and t1 = 0', the second identity should be t1 = 1.","section":"§5, proof of Theorem 5.1"},{"comment":"The proof first says 'Our goal is to show that A ≅ C4' but later concludes that the algebra is D4; the notation should be made consistent.","section":"§6, proof of Theorem 6.7"},{"comment":"In the proof of (ii)⇒(i), the line 'C2 ∈ H(B2)' should presumably read 'C2 ∈ H(B)', where B is the finitely generated subalgebra constructed in the preceding sentence.","section":"§9, proof of Theorem 9.8"},{"comment":"In the final sentence of the proof of item 1, 'V(A, D3)' should likely be 'V(A, Ca3)', since the claim concerns covers of V(Ca3).","section":"§8, proof of Corollary 8.5"},{"comment":"Fact 5.2, which supplies the inequalities encoded in Γ, is stated without proof. Since it plays a central role in defining the quotient, a short derivation of the displayed order relations would improve verifiability.","section":"§5, Fact 5.2"},{"comment":"The notation switches between An and A−n; the positive reduct should be introduced with a single consistent symbol and used uniformly throughout the example.","section":"§9, Example 9.9"}],"recommendation":"major_revision","confidential_remarks":"The central trichotomy is likely correct and the paper is a strong contribution, but the free-algebra computation in Theorem 5.1 is load-bearing and currently unverifiable. I would be willing to accept after the authors supply either the UACalc input/output or a hand proof, and after the generator list in Equation (4) is corrected."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"What you should know: this paper gives a sharp trichotomy for structural completeness in positive K4-algebras.  Theorem 9.7 says a non-trivial variety is SC exactly when it is one of V(B2), V(C2), V(D4), and that SC and HSC coincide there.  That is new, it contrasts with the full-signature K4 case where there are infinitely many HSC varieties, and it is a real step forward for the admissible-rules program in positive modal logic.\n\nWhat it does well: the paper is careful and honest.  The proofs are detailed, the Priestley-duality setup is used correctly, and the results on the free one-generated positive S4-algebra (finite, with a displayed diagram) are genuinely useful.  The bottom-of-the-lattice description and the splitting lemmas are also well organized.  I found no circularity or hidden dependence on the author's own prior work.\n\nWhere it is soft: Theorem 5.1 identifies the lattice reduct of the free one-generated positive S4-algebra with the quotient C/Cg(Γ) and says the identification can be checked mechanically via UACalc.  No input, output, or independent verification is given.  On top of that, Equation (4) lists the seven terms with □3□x duplicated and 3□3x missing, so the input to the purported computation is ambiguous as printed.  Since Sections 6, 8, and 9 all read off the eleven one-generated subdirectly irreducible algebras and the covers from Figure 1, an error here would propagate directly into the main trichotomy.  I think the intended description is probably right—the diagram and Fact 5.2 make the intended seven terms clear, and the later arguments are coherent—but as a referee I would want the computation actually shown or a certified verification, plus a corrected term list.  Fact 5.2 itself is stated without proof, though it looks routine.\n\nThe central classification is not built on sand.  The architecture is solid and the unverified step is a finite, checkable computation rather than a conceptual gap.  With the computation supplied or independently reproduced, I would expect the main results to stand.\n\nWho this is for: readers working on admissible rules, positive modal logic, or varieties of modal algebras.  It is a subfield paper, not a field-redefining one, but it settles a concrete and previously open question.  It deserves a serious referee, and I would send it out rather than desk-reject.  Recommendation: accept after the authors either prove the quotient identification in the paper or provide the UACalc input and output, and fix the term list in Equation (4).","headline":"A genuinely new and mostly convincing classification of structurally complete positive K4-varieties, held back only by an unshown finite computation and a typo in the term set.","tokens_in":31009,"tokens_out":2554,"would_cite":true,"duration_ms":27444,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["03B45","03C05","08B20","08C15"],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper proves that only three varieties of positive K4-algebras are structurally complete, and in that setting structural completeness coincides with hereditary structural completeness.","keywords":["positive modal logic","modal logic","structural completeness","admissible rules","abstract algebraic logic","positive K4-algebras","free one-generated algebra","algebraization of Gentzen systems"],"falsifier":"Re-run the mechanical computation of $C/\\mathrm{Cg}(\\Gamma)$ from the generator set $\\Sigma$ and inequality set $\\Gamma$ and compare the resulting lattice with Figure 1; in particular, verify that it has exactly eleven subdirectly irreducible homomorphic images matching Figure 2. If the quotient or the list of images differs, Theorem 8.6 and hence the structural-completeness trichotomy of Theorem 9.7 no longer follow from the proof given.","tokens_in":29986,"feed_emoji":"🧮","tokens_out":15285,"duration_ms":131500,"temperature":0.7,"pith_summary":"The paper studies varieties of positive modal algebras—the algebraic counterparts of the positive fragment of modal logic, where negation is dropped and only $\\wedge$, $\\vee$, $\\square$, $\\lozenge$, $0$, and $1$ remain. Its central result is a sharp classification: a nontrivial variety of positive K4-algebras is structurally complete exactly when it is one of three explicitly named varieties, and in that setting structural completeness coincides with hereditary structural completeness. A structurally complete variety is one in which every admissible inference rule is already derivable, so the classification says that the admissible-rule structure of positive K4 logic is rigid: almost every variety of positive K4-algebras admits admissible rules that are not derivable. Along the way the paper shows that the free one-generated positive S4-algebra is finite, even though the variety of positive S4-algebras is not locally finite and the free two-generated algebra is infinite. This matters because it contrasts with full modal logic, where infinitely many hereditarily structurally complete K4 varieties are known.","feed_headline":"Only three positive K4 varieties are structurally complete","feed_subtitle":"Admissible inference rules are derivable in exactly these three cases, while all other positive K4 varieties admit non-derivable rules.","key_machinery":"The load-bearing object is the free one-generated positive S4-algebra shown in Figure 1. It is built from the seven modal iterates of a single generator under $\\square$ and $\\lozenge$ (the set $\\Sigma$ in the paper) by taking the free seven-generated bounded distributive lattice and quotienting by the inequalities in a set $\\Gamma$ read off from the order relations among those iterates; the paper states that a computer check confirms the resulting lattice is exactly the one drawn. This finite algebra supplies all eleven one-generated subdirectly irreducible positive S4-algebras in Figure 2, from which the covers in the subvariety lattice are computed. The companion machinery is the algebra $D_4$ satisfying $\\square\\lozenge x \\approx \\square x$ and $\\lozenge\\square x \\approx \\lozenge x$; Theorem 6.7 shows these equations axiomatize $V(D_4)$, and Theorem 9.6 uses projectivity of $D_4$ in $V(D_4)$ to push hereditary structural completeness through all subquasivarieties.","core_discovery":"On the paper's own terms, the main discovery is Theorem 9.7: for a non-trivial variety $K$ of positive K4-algebras, the properties of being structurally complete, being hereditarily structurally complete, and being one of $V(B_2)$, $V(C_2)$, or $V(D_4)$ are equivalent. Here $B_2$ is the two-element positive modal algebra with $\\lozenge x \\approx 0$ and $\\square x \\approx 1$; $C_2$ is the two-element positive S4-algebra term-equivalent to bounded distributive lattices; and $D_4$ is the four-element positive S4-algebra axiomatized by $\\square\\lozenge x \\approx \\square x$ and $\\lozenge\\square x \\approx \\lozenge x$. For positive S4-algebras the paper proves a tighter equivalence (Theorem 9.6): active, plain, and hereditary structural completeness all coincide, and each is equivalent to satisfying those two equations. The proof reduces the problem to the shape of the free one-generated positive S4-algebra, which is finite and explicitly described, and then to a description of the bottom of the lattice of subvarieties of positive S4-algebras. A separate result (Theorem 9.8) characterizes passively structurally complete varieties of positive K4-algebras as exactly $V(B_2)$ or those whose zero-generated free algebra is $C_2$ and which exclude the three-element algebra $D_3$.","pith_inferences":["One consequence the author leaves implicit: for every non-exceptional variety of positive K4-algebras, a concrete non-derivable admissible quasi-equation must exist, so searches for explicit admissible-rule bases can be safely directed at the three exceptional varieties.","The sharp contrast with the full-signature setting, where infinitely many hereditarily structurally complete K4 varieties are known, suggests that dropping negation dramatically coarsens admissibility; a testable extension would be to determine whether other positive fragments of transitive modal logics, such as positive Grz or positive GL, also have only finitely many structurally complete variet","Because the free-algebra description rests on a delegated computer computation, formalizing the quotient calculation with a proof assistant would convert the classification into a fully verified theorem and close the one gap a skeptical reader could point to.","The finite one-generated/infinite two-generated boundary suggests studying the n-generated free positive S4-algebras for $n \\geq 2$ to locate precisely where local finiteness fails, with $n=2$ already known to be infinite."],"forward_implications":["If Theorem 9.7 is right, the admissible-rule problem for positive K4 logic has only three possible answers: in $V(B_2)$, $V(C_2)$, and $V(D_4)$ every admissible rule is derivable, while in every other nontrivial variety some admissible rule fails to be derivable.","Structural completeness and hereditary structural completeness coincide throughout positive K4, so no variety there occupies the intermediate zone of being structurally complete while having a structurally incomplete subquasivariety.","For positive S4-algebras, active structural completeness, structural completeness, and hereditary structural completeness are equivalent, so checking the two equations $\\square\\lozenge x \\approx \\square x$ and $\\lozenge\\square x \\approx \\lozenge x$ decides all three properties.","The free one-generated positive S4-algebra is finite while the free two-generated one is infinite; hence the variety $PS4$ is not locally finite, and the finite/infinite boundary in generation rank is sharp.","There are infinitely many passively structurally complete varieties of positive S4-algebras and infinitely many that are not, so passive completeness does not collapse in the same way."],"supporting_citations":[{"why":"supplies the mechanical verification that the free one-generated positive S4-algebra's lattice reduct is the quotient $C/\\mathrm{Cg}(\\Gamma)$ drawn in Figure 1.","marker":"[24]"},{"why":"provides the classification of hereditarily structurally complete K4 varieties that the three-variety result in positive K4 contrasts with.","marker":"[57]"},{"why":"shows that the one-generated free S4-algebra is infinite, the fact that makes Theorem 5.1's finiteness result significant.","marker":"[56]"},{"why":"gives the Priestley-style duality for positive modal algebras used throughout the paper's representation arguments.","marker":"[11]"},{"why":"introduces positive modal algebras as positive subreducts of modal algebras, the objects of the whole study.","marker":"[19]"},{"why":"supplies the algebraic definitions and the free-algebra characterizations of structural completeness that Theorem 1.1 collects.","marker":"[4]"},{"why":"establishes the Gentzen-system presentation of positive modal logic that connects the variety of positive modal algebras to admissible rules.","marker":"[10]"}],"fun_headline_variants":["Exactly three positive K4 varieties are structurally complete","Positive K4: structural completeness iff one of three known varieties","The structurally complete positive K4 varieties: exactly these three","Positive K4 structural completeness characterized by three cases"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The classification depends on the computer-verified claim that the lattice reduct of the free one-generated positive S4-algebra is exactly the quotient $C/\\mathrm{Cg}(\\Gamma)$ drawn in Figure 1, a computation the paper does not display; the later list of eleven subdirectly irreducible algebras and the three-variety trichotomy are built on that diagram.","fun_headline_variants_meta":{"raw":{"variants":["Exactly three positive K4 varieties are structurally complete","Positive K4: structural completeness iff one of three known varieties","The structurally complete positive K4 varieties: exactly these three","Positive K4 structural completeness characterized by three cases"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000841,"raw_usage":{"total_tokens":3651,"prompt_tokens":921,"completion_tokens":2730,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":537,"completion_tokens_details":{"reasoning_tokens":2666}},"tokens_in":537,"tokens_out":2730,"duration_ms":21973,"temperature":1.0,"reasoning_tokens":2666,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T15:48:43.453165+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Re-run the mechanical computation of $C/\\mathrm{Cg}(\\Gamma)$ from the generator set $\\Sigma$ and inequality set $\\Gamma$ and compare the resulting lattice with Figure 1; in particular, verify that it has exactly eleven subdirectly irreducible homomorphic images matching Figure 2. If the quotient or the list of images differs, Theorem 8.6 and hence the structural-completeness trichotomy of Theorem 9.7 no longer follow from the proof given.","supporting_citations":[{"cited_title":"Freese, E","cited_arxiv_id":null,"evidence_quote":"supplies the mechanical verification that the free one-generated positive S4-algebra's lattice reduct is the quotient $C/\\mathrm{Cg}(\\Gamma)$ drawn in Figure 1."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"provides the classification of hereditarily structurally complete K4 varieties that the three-variety result in positive K4 contrasts with."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"shows that the one-generated free S4-algebra is infinite, the fact that makes Theorem 5.1's finiteness result significant."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"gives the Priestley-style duality for positive modal algebras used throughout the paper's representation arguments."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"introduces positive modal algebras as positive subreducts of modal algebras, the objects of the whole study."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"supplies the algebraic definitions and the free-algebra characterizations of structural completeness that Theorem 1.1 collects."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"establishes the Gentzen-system presentation of positive modal logic that connects the variety of positive modal algebras to admissible rules."}],"review_version":1}