REVIEW 2 major objections 3 minor 3 references
On associative operations on commutative integral domains
T0 review · 2 major / 3 minor · reviewed 2026-08-14 · deepseek-v4-flash
Pith's one-line read The paper proves that any associative multilinear polynomial function over a commutative integral domain must be one of six explicit forms, which immediately classifies all associative operations on a two-element set.
desk verdict A small but real extension of Marichal–Mathonet to finite integral domains, built on a true but unstated uniqueness lemma; the Boolean proof is solid on its own. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The central object is a multilinear polynomial function, a polynomial in which each variable appears with exponent at most one. The machinery is the reduction of associativity to polynomial identities followed by coefficient comparison: with multilinearity, two multilinear polynomials that agree on all inputs have identical coefficients even over a finite integral domain, so the six normal forms from the infinite-domain theorem survive unchanged. In the Boolean application, the corresponding piece of machinery is the identification of every two-valued function with a unique multilinear polynomial over the two-element field.
What would settle it
Enumerate all $3^8$ ternary multilinear polynomials over the three-element field, check each for the associativity identity on all $3^5$ five-tuples for each of the nine pairs of insertion positions, and compare the associative ones against the six listed forms; any associative multilinear polynomial outside the list would refute the theorem.
Extended reading notes
Core claim
The paper's central claim is Theorem 3.2: an n-ary multilinear polynomial function $p : R^n \to R$ over a commutative integral domain with identity is associative if and only if it has one of the six forms already known in the infinite-domain theorem, namely a constant; the first projection $x_1$; the last projection $x_n$; $c + \sum_{i=1}^n x_i$; $\sum_{i=1}^n \omega^{i-1} x_i$ with $\omega \neq 1$ and $\omega^{n-1} = 1$; or $-b + a \prod_{i=1}^n (x_i + b)$ with $a \neq 0$ and the stated conditions on $b$. The proof route is to take the earlier infinite-domain classification, observe that its derivation only uses multilinearity after the initial step, and rely on multilinearity to make the coefficient comparisons valid over finite integral domains as well.
Load-bearing premise
The argument depends on the premise that multilinearity alone makes the earlier proof's coefficient-comparison steps valid over finite integral domains; if two different multilinear polynomials could agree at every tuple of a finite domain, the six forms might not be exhaustive.
Editorial extensions
If this is right
- Every associative multilinear polynomial function over a finite integral domain falls into one of the six listed forms, so no extra finite-characteristic examples appear.
- All associative Boolean operations of any arity are described: constants, projections, meet, join, addition modulo 2, and its complement, with the last two related by parity.
- The only n-ary associative operation on a two-element set that is not derivable from a binary associative operation is the odd-arity complement of parity addition.
- The primitive n-ary associative operations on a two-element set are exactly the unary and binary ones together with that odd-arity parity complement.
- The general description of associative operations on finite sets with at least three elements remains open, and this theorem implies that any new polynomial example over a finite field must be non-multilinear.
Reading between the lines
- A consequence the authors do not spell out is that any finite-field counterexample to a full classification must involve a polynomial with a squared variable, so the search for new n-semigroups lies strictly above multilinearity.
- The same normal-form strategy could be tried over commutative rings with zero divisors, where coefficient comparison is more delicate; testing a ring such as the integers modulo 4 would show how far the argument generalizes.
- If the adaptation claim in the proof is correct, the Boolean classification is a direct corollary rather than a separate phenomenon, making the detailed combinatorial proof in the second half a self-contained alternative route to the same list.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper studies associative operations that are defined by multilinear polynomials over commutative integral domains. After recalling Marichal and Mathonet's classification for infinite integral domains (Theorem 3.1), it states Theorem 3.2, asserting that the same list classifies associative multilinear polynomial functions over arbitrary commutative integral domains with identity. The paper then derives the classification of associative Boolean functions as a consequence, and independent of Theorem 3.2 it gives a self-contained elementary case analysis proving Andres's classification of two-element n-semigroups (Theorem 4.1). It also identifies, in Proposition 4.3, the primitive associative operations on a two-element set. The central mathematical claim is Theorem 3.2, with Section 4 providing an independent proof for the Boolean case.
Significance. If Theorem 3.2 is made fully rigorous, it is a clean and useful classification: it extends the Marichal--Mathonet description from the infinite-domain setting to the class of multilinear polynomial functions over arbitrary integral domains, yielding a uniform explanation of the Boolean classification as a special case. The paper's strengths are the self-contained and apparently correct case analysis for Boolean functions, the explicit identification of primitive operations on {0,1}, and the transparent reliance on the cited Marichal--Mathonet theorem. There is no circularity: the Boolean proof does not depend on Theorem 3.2. The main weakness is that the proof of Theorem 3.2 is not actually supplied; it rests on an assertion about the cited proof that needs a missing uniqueness lemma. Since that lemma is true and easy to state, the defect is repairable and does not affect the plausibility of the result.
major comments (2)
- [Section 3, Theorem 3.2] The proof of Theorem 3.2 is reduced to the sentence, 'The remainder of the proof only relies on this multilinearity.' This assertion is load-bearing and is not demonstrated. The Marichal--Mathonet proof for infinite integral domains begins with the one-to-one correspondence between polynomials and polynomial functions, a correspondence that fails over finite domains. To transfer the proof, one must state and prove the uniqueness lemma that a multilinear polynomial over an integral domain is uniquely determined by its polynomial function, i.e., if a multilinear polynomial vanishes on all of R^n, then it is the zero polynomial. The associativity difference of a multilinear p is multilinear, so this lemma would justify the coefficient-comparison steps; without it, Theorem 3.2 is not proved as written. The missing lemma is true (by induction on n, or by passing to the field of fractions), so this is a repairable omission rather than a counterexample, but the proof must be supplied.
- [Section 3, paragraph after Theorem 3.1] The claim that 'the remainder of the proof only relies on this multilinearity' also requires a check that no later step in the Marichal--Mathonet argument uses infinitude of R, for example by equating coefficients of a polynomial that could be nonzero as a formal object yet vanish as a function on a finite field. The paper gives no such check and no lemma-by-lemma account of which parts of [2] are being reused. This is the same gap as in the previous comment, but the authors should address it explicitly by either providing the adapted proof or identifying precisely which statements of [2] are used after multilinearity is assumed.
minor comments (3)
- [Section 4, Theorem 4.1] In the statement of Theorem 4.1, the last two operations are printed as '+(n), +(n)' and the definition 'where +(n)(a1,...,an) := +(n)(a1,...,an) + 1' is circular as typeset; the second operation should be a separate symbol, presumably \bar{+}^{(n)} or \boxplus^{(n)}, with an unambiguous definition.
- [Section 4, Case 1.2.1.1] The displayed chain beginning '0 = (0000^{n-3}) = ...' is hard to parse for small n, especially n=2 and n=3; the word lengths and exponents should be corrected or written more explicitly.
- [Abstract and Section 3] The phrase 'extends Marichal and Mathonet's result on infinite integral domains' is potentially misleading: Theorem 3.2 concerns only multilinear polynomial functions, whereas Theorem 3.1 concerns all polynomial functions over infinite domains. It would be more accurate to say that the paper extends the classification to multilinear polynomial functions over arbitrary integral domains.
Circularity Check
No circularity: the finite-domain transfer reuses an external theorem and the Boolean proof is independent, with only a repairable proof gap.
full rationale
The paper's central result, Theorem 3.2, transfers Marichal and Mathonet's Theorem 3.1 from infinite integral domains to arbitrary integral domains under the additional hypothesis of multilinearity. This is not circular: Theorem 3.1 is an external, independently published result, not a self-citation, and it is not derived from the paper's own assumptions. The paper fits no parameters, renames no known result, and does not define its target notion in terms of the conclusion. The sentence 'The remainder of the proof only relies on this multilinearity' is a load-bearing assertion about the portability of the Marichal-Mathonet proof to finite domains, but it is an unsupported proof gap rather than a circular reduction. A rigorous transfer would require a uniqueness lemma for multilinear polynomial functions over integral domains, namely that a multilinear polynomial vanishing at all points is the zero polynomial; this lemma is true, it is not equivalent to the classification being proved, and its omission does not make the conclusion an input. The Boolean classification in Section 4 is proved by an independent, self-contained case analysis over {0,1} that does not rely on Theorem 3.2; it reproduces Andres's known classification with a new proof, which is legitimate independent content. No step in the paper reduces by construction to its own inputs, and no self-citation carries any load. Accordingly, the appropriate circularity score is 0.
Assumptions & free parameters
assumptions (3)
- domain assumption Marichal-Mathonet's Main Theorem correctly classifies associative polynomial functions over infinite integral domains.
- ad hoc to paper The proof of Marichal-Mathonet's theorem, once multilinearity is assumed, does not require the domain to be infinite and the coefficient comparisons transfer unchanged.
- standard math Every Boolean function on {0,1}^n is uniquely represented by a multilinear polynomial over GF(2).
Cite this review
Pith. "Pith review of On associative operations on commutative integral domains." pith.science (2026). https://pith.science/paper/OG7F3BSX
@misc{pith2026190802009,
author = {Pith},
title = {Pith review of: On associative operations on commutative integral domains},
year = {2026},
howpublished = {\url{https://pith.science/paper/OG7F3BSX}},
note = {Machine review of arXiv:1908.02009}
}
abstract
We describe the associative multilinear polynomial functions over commutative integral domains. This extends Marichal and Mathonet's result on infinite integral domains and provides a new proof of Andres's classification of two-element $n$-semigroups.
Reference graph
Works this paper leans on
-
[2]
Semigroup Forum 83, 241–249 (2011)
Marichal, J.-L., Mathonet, P.: A description of n-ary semigroups polynomial-derived from integral domains. Semigroup Forum 83, 241–249 (2011)
work page 2011
-
[1]
Andres, S.D.: Classification of all associative mono- n-ary algebras with 2 elements. Int. J. Math. Math. Sci. 2009, Art. ID 678987, 16 pp. (2009)
work page 2009
-
[3]
“2018 Mikl´ os Schweitzer”, Art of Problem Solving porta l, accessed June 11, 2019, https://artofproblemsolving.com/community/c771105 2018 mikloacutes schweitzer
work page 2018
Reviewed August 14, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.