A study of truth predicates in matrix semantics

T. Moraschini

A study of truth predicates in matrix semantics

T0 review · 0 major / 5 minor · reviewed 2026-08-14 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read For the reduced matrix semantics of a propositional logic, truth is almost parametrically equationally definable exactly when the Leibniz operator is almost completely order-reflecting, and Leibniz-injectivity transfers from theories to arbitrary filters exactly for countable languages.

desk verdict Settles an open transfer problem in abstract algebraic logic with a clean cardinality divide; the proof structure is solid, with only minor typos and a terse 'almost' case to fix. read the letter →

arxiv 1908.01661 v1 pith:32VP6NH6 submitted 2019-08-01 math.LO

T. Moraschini This is my paper · ORCID

classification math.LO

keywords truthleibnizoperatorsetsdeductivefiltersalgebraicarbitrary

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Every propositional logic can be described by matrices: algebraic structures with a chosen set of elements declared 'true'. The family of all reduced matrices associated to a logic is the canonical semantics. A recurring question is whether the truth sets can be recognized by equations, for instance whether there is a set of equations whose solutions, for every choice of parameter values, are exactly the true elements. This paper studies that question through the Leibniz operator, a tool that measures when two elements of an algebra are indistinguishable by the logic's filters.

Earlier work by Raftery characterized the parameter-free version: truth is equationally definable exactly when the Leibniz operator is completely order-reflecting. The paper extends this to parameters, showing that 'almost parametrically equationally definable' corresponds to the same property on nonempty filters. It also proves a surprising collapse: with parameters, definability over all reduced matrices reduces to definability without parameters, so the new notion matters only for purely inferential logics, those without theorems.

The second main result concerns implicit definability, the condition that a matrix is determined by its algebra. Long known to be equivalent to injectivity of the Leibniz operator on filters, the open question was whether injectivity on theories alone suffices. The paper proves it does for logics in a countable language, and constructs a counterexample in an uncountable language. A new intermediate notion, 'small' truth sets, is characterized by order-reflection. Examples from semilattices and bilattices illustrate the hierarchy of conditions.

Extended reading notes

Core claim

Main positive theorem: 'For logics expressed in a countable language the (almost) injectivity of the Leibniz operator transfers from theories to filters over arbitrary algebras' (Theorem 5.6). Main negative theorem: there is a logic in an uncountable language whose Leibniz operator is injective over theories but not over filters (Theorem 6.7, Section 6). If the paper is correct, the transfer problem is settled: injectivity on theories implies injectivity on all filters exactly when the language is countable.

Load-bearing premise

The negative half rests on the standard AAL convention, fixed in Section 2 ('a countable set Var of variables'), that the formula algebra has countably many variables while the operation symbols in Section 6 are indexed by uncountably many reals. The proof of Fact 6.5 ends by deriving that each of uncountably many constants ci requires a distinct variable yi, contradicting countability of Var. If the variable set were allowed to be uncountable, this particular counterexample would not go through, so the claimed failure of transfer in general is tied to this cardinality convention.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Short version: this paper solves an open problem in abstract algebraic logic, and as far as I can tell it does so correctly. The transfer problem—whether injectivity of the Leibniz operator on theories forces injectivity on all deductive filters over arbitrary algebras—gets a clean cardinality-sensitive answer: yes for countable languages, no in general, with an explicit uncountable-language counterexample. That settles Raftery's Problem 1.

What is actually new: parametrized equational definability of truth sets, characterized in Theorem 3.9 by the Leibniz operator being almost completely order-reflecting on non-empty filters. Corollary 3.10 is the structural punchline: parametrized definability collapses to plain equational definability whenever the logic has theorems, so the parametrized notion only matters for purely inferential logics. The paper also adds two new rungs to the Leibniz hierarchy—small and almost small truth sets—with an order-reflection characterization (Theorem 7.3) and a matching transfer theorem (Theorem 7.4). The proof machinery is genuinely worked out: Lemma 5.5 extends a subalgebra while preserving reducedness for two filters at once, and Section 6's construction is elaborate, explicit, and coerced by a real cardinality contradiction. The Fregean protodisjunction/protoconjunction results give broad, natural examples, and the alternative proof of the Czelakowski–Jansana theorem is a genuine simplification.

Soft spots, in proportion. The negative result relies on the standard convention that the variable set is countable; the true boundary is |L| ≤ |Var|, not countability of the language per se, and an uncountable variable supply would kill the counterexample. The paper states the convention explicitly in Section 2, so it is a caveat rather than a flaw. Theorem 5.6's 'almost' case is terse to the point of being under-specified: to use almost-injectivity on theories you need both inverse-image theories non-empty, which is only guaranteed by putting an element of G into the generating set; the fix is immediate. Remark 7.9 ends in a self-contradictory clause ('almost implicitly definable, but neither almost small nor almost implicitly definable') that is clearly a typo. I found nothing load-bearing wrong; the citation pattern is appropriate, with Raftery's theorem as the base and the new claims clearly marked.

Who is this for: anyone in AAL, and more broadly logicians who like seeing a natural phenomenon—transfer of a structural property—turn out to depend on cardinality. It deserves a serious referee. Send it to peer review; the referee should check Section 6 carefully and ask for the small fixes above.

Referee Report

0 major / 5 minor

Summary. The paper studies definability of truth sets in the reduced matrix semantics Mod*L of a propositional logic L, connecting truth-set definability conditions to order-theoretic properties of the Leibniz operator restricted to deductive filters. Section 3 characterizes almost parametrically equational definability by almost complete order-reflection of the Leibniz operator (Theorem 3.9) and shows that parametrized equational definability collapses to equational definability for logics with theorems (Corollary 3.10). Section 4 gives semilattice, lattice, and bilattice examples. Section 5 addresses the transfer problem: Theorem 5.6 proves that, for logics in a countable language, injectivity (and almost injectivity) of the Leibniz operator over theories transfers to filters over arbitrary algebras; Section 6 constructs a logic in an uncountable language where transfer fails (Theorem 6.7). Section 7 introduces small truth sets, characterizes them via order-reflection (Theorem 7.3), and proves an analogous transfer result for order-reflection in countable languages (Theorem 7.4).

Significance. If correct, the paper settles [32, Problem 1] by showing that the transfer of injectivity of the Leibniz operator from theories to filters holds exactly under a cardinality bound on the language, relative to the standard countable-variable convention. It also expands the Leibniz hierarchy by adding almost parametrically equational definability and small truth sets as new levels, with concrete separating examples in Sections 4 and 7. The proofs are unusually complete: central lemmas such as Lemma 3.6, Lemma 3.7, Lemma 5.5, and the Section 6 facts are proved rather than cited, and the main theorems are derived from the definitions with explicit inductions and cardinality calculations. The negative construction in Section 6 is intricate and appears internally coherent.

minor comments (5)

[§7, Remark 7.9] The final clause of Remark 7.9 is self-contradictory: it says a logic is "almost implicitly definable, but neither almost small nor almost implicitly definable". The second conjunct must be intended to state a different property, presumably "nor almost parametrically equationally definable". The same remark also contains the incomplete phrase "by Theorem the fact that..." with no theorem number, and the construction of L′ as having theories ThL ∪ {∅} needs clarification, since if the previous L has theorems then ∅ cannot be a theory of a structural logic.
[§5, Theorem 5.6] In the almost-injectivity case, the proof says only that it follows by restricting to non-empty filters, but the written argument chooses C generated by {a} for a ∈ F and then applies injectivity over non-empty theories to Γ = h⁻¹[F∩B] and Γ′ = h⁻¹[G∩B]. To guarantee that Γ′ is non-empty when L may be purely inferential, one should choose C generated by {a,b} with b ∈ G. The fix is immediate but should be stated.
[§7, Theorem 7.4] The proof invokes Lemma 5.5 to obtain B with ⟨B,F∩B⟩ reduced, but Lemma 5.5 is stated for a pair of reduced matrices and the arbitrary filter G is not assumed reduced. The application should either invoke the lemma with the pair ⟨A,F⟩,⟨A,F⟩ or state the obvious one-sided version of Lemma 5.5; as written the reader must supply this step.
[§3, Definition 3.2] The phrase "almost parametrically equationally definable" is used from the abstract and introduction onward, and the paper relies on the general convention that a class of matrices almost enjoys a property when every non-almost trivial member enjoys it. It would be clearer to spell out formally in Definition 3.2 or immediately after it that "truth is almost parametrically equationally definable in Mod*L" means that a single parametrized equational translation defines the truth set of every non-almost trivial reduced model.
[§6, Theorem 6.7] The negative result is tied to the convention, fixed in Section 2, that Var is countably infinite: the contradiction in Fact 6.5 uses the fact that uncountably many reals i require distinct variables y_i. The paper should state explicitly that the real boundary is |L| ≤ |Var|, so that the transfer theorem applies when the language has cardinality at most that of the variable set, rather than only when the language is countable.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

The central results are proved from the definitions of logic, deductive filter, Leibniz and Suszko operators, plus a small set of established background theorems. No data-dependent parameters or new postulated entities are introduced.

assumptions (4)

standard math ZFC set theory and standard cardinal arithmetic
Used throughout, notably for cardinality bounds in Lemma 5.5 and for the uncountability argument in Fact 6.5.
domain assumption The framework of logics as structural closure operators over an absolutely free term algebra with countably many variables
Stated in Section 2; all results are internal to this framework and the negative example in Section 6 depends on the countability of variables.
standard math Raftery's characterization of equational definability by complete order-reflection of the Leibniz operator
Invoked as Theorem 3.5 to prove Corollary 3.10; cited to Raftery [32, Theorem 28] rather than proved in this paper.
standard math For Fregean logics, the Suszko congruence equals the Frege relation Lambda_L Gamma
Stated in Section 4 as an equivalent definition and used in the proofs of Theorems 4.6 and 4.9; this equivalence is a known result in the AAL literature, not proved here.

how reviews work

0 comments

Cite this review

Pith. "Pith review of A study of truth predicates in matrix semantics." pith.science (2026). https://pith.science/paper/32VP6NH6

@misc{pith2026190801661,
  author       = {Pith},
  title        = {Pith review of: A study of truth predicates in matrix semantics},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/32VP6NH6}},
  note         = {Machine review of arXiv:1908.01661}
}

read the original abstract

Abstract algebraic logic is a theory that provides general tools for the algebraic study of arbitrary propositional logics. According to this theory, every logic L is associated with a matrix semantics Mod*(L). This paper is a contribution to the systematic study of the so-called "truth sets" of the matrices in Mod*(L). In particular, we show that the fact that the truth sets of Mod*(L) can be defined by means of equations with universally quantified parameters is captured by an order-theoretic property of the Leibniz operator restricted to deductive filters of L. This result was previously known for equational definability without parameters. Similarly, it was known that the truth sets of Mod*(L) are implicitly definable if and only if the Leibniz operator is injective on deductive filters of L over every algebra. However, it was an open problem whether the injectivity of the Leibniz operator transfers from the theories of L to its deductive filters over arbitrary algebras. We show that this is the case for logics expressed in a countable language, and that it need not be true in general. Finally we consider an intermediate condition on the truth sets in Mod*(L) that corresponds to the order-reflection of the Leibniz operator.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

35 extracted references · 25 canonical work pages

[1]

Albuquerque, J

H. Albuquerque, J. M. Font, R. Jansana, and T. Moraschini. Assertional logics, truth- equational logics, and the hierarchies of abstract algebraic logic. In J. Czelakowski, editor, Don Pigozzi on Abstract Algebraic Logic and Universal Algebra , volume 16 of Outstanding Contributions. Springer-Verlag, 2018

2018
[2]

S. V . Babyonyshev. Fully Fregean logics. Reports on Mathematical Logic, 37:59–77, 2003

work page 2003
[3]

W. J. Blok and D. Pigozzi. Protoalgebraic logics. Studia Logica, 45:337–369, 1986

work page 1986
[4]

W. J. Blok and D. Pigozzi. Algebraizable logics, volume 396 of Mem. Amer. Math. Soc. A.M.S., Providence, January 1989. 28 TOMMASO MORASCHINI ﬁnitely regularly algebraizable ﬁnitely algebraizable regularly algebraizable ﬁnitely equivalential algebraizable regularly weakly algebraizable equivalential weakly algebraizable assertional protoalgebraic truth equ...

work page 1989
[5]

W. J. Blok and D. Pigozzi. Algebraic semantics for universal Horn logic without equality. pages 1–56. Heldermann, Berlin, 1992

1992
[6]

W. J. Blok and J. Rebagliato. Algebraic semantics for deductive systems. Studia Logica, Special Issue on Abstract Algebraic Logic, Part II , 74(5):153–180, 2003

2003
[7]

Bou and U

F. Bou and U. Rivieccio. The logic of distributive bilattices. Logic Journal of the I.G.P .L., 19(1):183–216, 2011

work page 2011
[8]

Czelakowski

J. Czelakowski. Algebraic aspects of deduction theorems. Studia Logica, 44:369–387, 1985

work page 1985

Show all 35 references

[9]

Czelakowski

J. Czelakowski. Local deductions theorems. Studia Logica, 45:377–391, 1986

1986
[10]

Czelakowski

J. Czelakowski. Protoalgebraic logics, volume 10 of Trends in Logic—Studia Logica Library . Kluwer Academic Publishers, Dordrecht, 2001

2001
[11]

Czelakowski

J. Czelakowski. The Suszko operator. Part I. Studia Logica, Special Issue on Abstract Algebraic Logic, Part II, 74(5):181–231, 2003. A STUDY OF TRUTH PREDICATES IN MATRIX SEMANTICS 29

2003
[12]

Czelakowski and R

J. Czelakowski and R. Jansana. Weakly algebraizable logics. The Journal of Symbolic Logic, 65(2):641–668, 2000

2000
[13]

Czelakowski and D

J. Czelakowski and D. Pigozzi. Fregean logics. Annals of Pure and Applied Logic, 127(1- 3):17–76, 2004

2004
[14]

Czelakowski and D

J. Czelakowski and D. Pigozzi. Fregean logics with the multiterm deduction theorem and their algebraization. Studia Logica, 78(1-2):171–212, 2004

2004
[15]

Descalc ¸o and M

L. Descalc ¸o and M. A. Martins. On the injectivity of the Leibniz operator. Bulletin of the Section of Logic, 34(4):203–211, 2005

2005
[16]

J. M. Font. Abstract Algebraic Logic - An Introductory Textbook, volume 60 of Studies in Logic - Mathematical Logic and Foundations . College Publications, London, 2016

2016
[17]

J. M. Font, F. Guzm ´an, and V . Verd´u. Characterization of the reduced matrices for the {∧,∨}-fragment of classical logic. Bulletin of the Section of Logic , 20:124–128, 1991

1991
[18]

J. M. Font and R. Jansana. A general algebraic semantics for sentential logics , vol- ume 7 of Lecture Notes in Logic . A.S.L., second edition 2017 edition, 2009. First edition 1996. Electronic version freely available through Project Euclid at projecteuclid.org/euclid.lnl/1235416965

2017
[19]

J. M. Font, R. Jansana, and D. Pigozzi. A survey on abstract algebraic logic. Studia Logica, Special Issue on Abstract Algebraic Logic, Part II , 74(1–2):13–97, 2003. With an “Update” in 91 (2009), 125–130

2009
[20]

J. M. Font and T. Moraschini. Logics of varieties, logics of semilattices, and conjunction. Logic Journal of the IGPL, 22:818–843, 2014

2014
[21]

J. M. Font and T. Moraschini. A note on congruences of semilattices with sectionally ﬁnite height. Algebra Universalis, 72(3):287–293, 2014

2014
[22]

J. M. Font and V . Verd´u. Algebraic logic for classical conjunction and disjunction. Studia Logica, Special Issue on Algebraic Logic , 50:391–419, 1991

1991
[23]

M. L. Ginsberg. Multivalued logics: A uniform approach to inference in artiﬁcial intelligence. Computational Intelligence, 4:265–316, 1988

1988
[24]

Herrmann

B. Herrmann. Algebraizability and Beth’s theorem for equivalential logics. Bulletin of the Section of Logic, 22(2):85–88, 1993

1993
[25]

Herrmann

B. Herrmann. Equivalential logics and deﬁnability of truth . Ph. D. Thesis, Freie Universit ¨at Berlin, 1993. 61 pp

1993
[26]

Herrmann

B. Herrmann. Characterizing equivalential and algebraizable logics by the Leibniz operator. Studia Logica, 58:305–323, 1997

1997
[27]

Jansana and T

R. Jansana and T. Moraschini. Advances in the Theory of the Leibniz Hierarchy. Manuscript available on request, 2017

2017
[28]

Moraschini

T. Moraschini. Investigations in the role of translations in abstract algebraic logic. PhD thesis, University of Barcelona, June 2016

2016
[29]

Moraschini

T. Moraschini. On the complexity of the Leibniz hierarchy. Manuscript, December 2016

2016
[30]

Moraschini

T. Moraschini. A computational glimpse at the Leibniz and Frege hierarchies. Annals of Pure and Applied Logic, 169(1):1–20, January 2018

2018
[31]

Noguera and P

C. Noguera and P . Cintula. The proof by cases property and its variants in structural consequence relations. Studia Logica, 101:713–747, 2013

2013
[32]

J. G. Raftery. The equational deﬁnability of truth predicates. Reports on Mathematical Logic, (41):95–149, 2006

2006
[33]

J. G. Raftery. A perspective on the algebra of logic.Quaestiones Mathematicae, 34:275–325, 2011

2011
[34]

Rautenberg

W. Rautenberg. On reduced matrices. Studia Logica, 52:63–72, 1993

1993
[35]

Rivieccio

U. Rivieccio. An Algebraic Study of Bilattice-based Logics . Ph. D. Dissertation, University of Barcelona, 2010. E-mail address: moraschini@cs.cas.cz 30 TOMMASO MORASCHINI Institute of Computer Science, Czech Academy of Sciences, Pod Vod ´arenskou vˇeˇz´i 2, 182 07 Prague 8, C...

2010

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.

[1] [1]

Albuquerque, J

H. Albuquerque, J. M. Font, R. Jansana, and T. Moraschini. Assertional logics, truth- equational logics, and the hierarchies of abstract algebraic logic. In J. Czelakowski, editor, Don Pigozzi on Abstract Algebraic Logic and Universal Algebra , volume 16 of Outstanding Contributions. Springer-Verlag, 2018

2018

[2] [2]

S. V . Babyonyshev. Fully Fregean logics. Reports on Mathematical Logic, 37:59–77, 2003

work page 2003

[3] [3]

W. J. Blok and D. Pigozzi. Protoalgebraic logics. Studia Logica, 45:337–369, 1986

work page 1986

[4] [4]

W. J. Blok and D. Pigozzi. Algebraizable logics, volume 396 of Mem. Amer. Math. Soc. A.M.S., Providence, January 1989. 28 TOMMASO MORASCHINI ﬁnitely regularly algebraizable ﬁnitely algebraizable regularly algebraizable ﬁnitely equivalential algebraizable regularly weakly algebraizable equivalential weakly algebraizable assertional protoalgebraic truth equ...

work page 1989

[5] [5]

W. J. Blok and D. Pigozzi. Algebraic semantics for universal Horn logic without equality. pages 1–56. Heldermann, Berlin, 1992

1992

[6] [6]

W. J. Blok and J. Rebagliato. Algebraic semantics for deductive systems. Studia Logica, Special Issue on Abstract Algebraic Logic, Part II , 74(5):153–180, 2003

2003

[7] [7]

Bou and U

F. Bou and U. Rivieccio. The logic of distributive bilattices. Logic Journal of the I.G.P .L., 19(1):183–216, 2011

work page 2011

[8] [8]

Czelakowski

J. Czelakowski. Algebraic aspects of deduction theorems. Studia Logica, 44:369–387, 1985

work page 1985