Gemma Robles

The "depth relevance condition" (drc) is a strengthening of the "variable-sharing property" (vsp). Deep relevant logics are logics fulfilling the drc, and Brady's DR is a key item in this class. The "qualified factorization principles" (qfp) are strong distribution principles. The qfp can be added to Relevance logic R without the result collapsing in a logic lacking the vsp. The aim of this paper is to show that DR (and any logic included in it) can be extended with the qfp, the drc being preserved. © 2016 Elsevier B.V., All rights reserved.

A logic is relevant if it is a set of formulas closed under adjunction, modus ponens and enjoys the variable sharing property. It is connexive if it has some or all of the versions of Aristotle’s thesis and Boethius’ thesis. It is connexively acceptable if it does not sanction such formulas as the negation of the self-identity axiom. Finally, it is strictly connexive if it does not validate some formula implying (or being implied by) its own negation. Now, it is known that addition of Aristotle’s thesis to a relevant logic does not necessarily result in an acceptable connexive logic. Then, the aim of this paper is to examine the prospects of defining relevant acceptable strictly connexive logics based upon weak relevant logics.

The class of the logics free from paradoxes of relevance determined by all natural implicative expansions of Kleene’s strong 3-valued matrix with two designated values is defined. These logics are free from paradoxes of relevance in the sense that they have the "variable sharing property". They have "natural conditionals" in the sense that the function defining them coincideswith the classical function when restricted to the "classical values", satisfies Modus Ponens and, finally, assigns a designated value to a conditional whenever its antecedent and its consequence are assigned the same value. These logics are defined by using a "two-valued" overdetermined Belnap-Dunn semantics. Thus, the interpretation of the three values is crystalline. The logics here introduced enjoy the properties customarily demanded of many-valued implicative logics except, of course, the satisfaction of the rule "Verum ex quodlibet".

Let us refer by MK3 to Kleene’s strong 3-valued matrix. An implicative expansion of MK3 is natural if the conditional function defining it verifies modus ponens, assigns a designated value to a conditional whenever it assigns the same value to its antecedent and its consequent, and, finally, it coincides with the classical conditional function when restricted to the “classical” values ????\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mathfrak {t}$$\end{document} and ????\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mathfrak {f}$$\end{document}. Two are the main results of this paper. (1) It is proven that, from the viewpoint of functional strength, there is only one 3-valued natural implication expansion of MK3 with the variable-sharing property, the logic we dub L3VSP\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${ }^{\text{VSP}}$$\end{document}. (2) It is shown that L3VSP\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${ }^{\text{VSP}}$$\end{document} is a significant and strong logic that can be seen from different perspectives, one of them being to consider it an expansion of classical positive propositional logic.

The present paper is inspired by Sylvan and Plumwood’s logicBM defined in “Non-normal relevant logics” and by their treatmentof negation with the ∗-operator in “The semantics of first-degree en-tailment”. Given a positive logic L including Routley and Meyer’sbasic positive logic and included in either the positive fragment of Eor in that of RW, we investigate the essential De Morgan negation ex-pansions of L and determine all the deductive relations they maintainto each other. A Routley-Meyer semantics is provided for each logicdefined in the paper.

The well-known logic first degree entailment logic (FDE), introduced by Belnap and Dunn, is defined with |$\wedge $|⁠, |$\vee $| and |$\sim $| as the sole primitive connectives. The aim of this paper is to establish the lattice formed by the class of all 4-valued C-extending implicative expansions of FDE verifying the axioms and rules of Routley and Meyer’s basic logic B and its useful disjunctive extension B|$^{\textrm {d}}$|⁠. It is to be noted that Boolean negation (so, classical propositional logic) is definable in the strongest element in the said class.

Belnap and Dunn’s well-known 4-valued logic FDE is an interesting and useful non-classical logic. FDE is defined by using conjunction, disjunction and negation as the sole propositional connectives. Then the question of expanding FDE with an implication connective is of course of great interest. In this sense, some implicative expansions of FDE have been proposed in the literature, among which Brady’s logic BN4 seems to be the preferred option of relevant logicians. The aim of this paper is to define a class of implicative expansions of FDE in whose elements Boolean negation is definable, whence strong logics such as the paraconsistent and paracomplete logic PŁ4 and BN4 itself are definable, in addition to classical propositional logic.

The logic E4 is related to Brady’s BN4 in a similar way to which Anderson and Belnap’s logic of entailment E is related to their logic of the relevant implication R. In ‘A companion to Brady’s 4-valued relevant logic: the 4-valued logic of entailment E4’, quoted in this paper, three alternatives to BN4 and another three to E4 are summarily introduced in a couple of pages as the only alternatives containing Routley and Meyer’s basic logic B, provided some conditions are fulfilled. The aim of this note is to prove BN4 and its three alternatives are the same logic up to some point, since they are functionally equivalent to each other, as it is the case with E4 and its three alternatives, which are also functionally equivalent to each other; to some extent, E4 is superior to BN4, since the latter is functionally included in the former, but not conversely.

The basic quasi-Boolean negation expansions of relevance logics included in Anderson and Belnap’s relevance logic R are defined. We consider two types of QB-negation: H-negation and D-negation. The former one is of paraintuitionistic or superintuitionistic character, the latter one, of dual intuitionistic nature in some sense. Logics endowed with H-negation are paracomplete; logics with D-negation are paraconsistent. All logics defined in the paper are given a Routley-Meyer ternary relational semantics.

The logic DHb is the result of extending Sylvan and Plumwood’s minimal De Morgan logic BM with a dual intuitionistic negation of the type Sylvan defined for the extension CCω of da Costa’s paraconsistent logic Cω. We provide Routley–Meyer ternary relational semantics with a set of designated points for DHb and a wealth of its extensions included in G3DH, the expansion of G3+ with a dual intuitionistic negation of the kind considered by Sylvan (G3+ is the positive fragment of Gödelian 3-valued logic G3). All logics in the paper are paraconsistent.

Routley-Meyer ternary relational semantics was introduced in the early seventies of the past century. RM-semantics was intended to model classical relevant logics such as the logic of the relevant conditional R and the logic of Entailment E. But, ever since Routley and Meyer’s first papers on the topic, this essentially malleable semantics has been used for characterizing more general relevant logics or even non-relevant logics. The aim of this paper is to provide an RM-semantics with respect to which Łukasiewicz 3-valued logic Ł3 is sound and complete. Ł3 is understood as the set of all valid formulas in Łukasiewicz 3-valued matrices MŁ3. In this sense, leaning on previous work by us, Ł3 is axiomatized as an extension of Routley and Meyer’s basic positive logic B+, labelled Ł3. And the RM-semantics for Ł3 is actually defined for this particular axiomatization of Ł3. The result presented in the paper is interesting from the Universal Logic perspective, in the sense that it connects Łukasiewicz many-valued logics and similar systems to relevant logics from the point of view of the latter, the 3-termed relational point of view, in particular.

As it is well known, in the forties of the past century, Curry proved that in any logic S closed under Modus Ponens, uniform substitution of propositional variables and the Contraction Law, the naïve Comprehension axiom trivializes S in the sense that all propositions are derivable in S plus CA. Not less known is the fact that, ever since Curry published his proof, theses and rules weaker than W have been shown to cause the same effect as W causes. Among these, the Contraction rule or the Modus Ponens axiom, for example, are to be noted. But, moreover, as Brady has proved, even the Generalized Modus Ponens axiom or the Generalized Contraction rule give rise to “Curry’s Paradox” under the same circumstances as W does. In some previous work by us, “weak relevant model structures” are defined on “weak relevant matrices” by generalizing Brady’s model structure MCL built upon Meyer’s Crystal matrix CL. We have proved that wr-ms only verify logics with the “depth relevance condition”. The aim of this paper is to show how to falsify gMPa and gRW in certain wr-ms. In particular, it will be shown that gMPa is falisfied in any wr-ms and gRW in any wr-ms verifying Routley and Meyer’s basic positive logic B+.

Minimal Negation is defined within the basic positive relevance logic in the relational ternary semantics: B+. Thus, by defining a number of subminimal negations in the B+ context, principles of weak negation are shown to be isolable. Complete ternary semantics are offered for minimal negation in B+. Certain forms of reductio are conjectured to be undefinable (in ternary frames) without extending the positive logic. Complete semantics for such kinds of reductio in a properly extended positive logic are offered.