Abstract
Genuine Paraconsistent logics \(\mathbf {L3A}\) and \(\mathbf {L3B}\) were defined in 2016 by Béziau et al, including only three logical connectives, namely, negation disjunction and conjunction. Afterwards in 2017 Hernández-Tello et al, provide implications for both logics and define the logics \(\mathbf {L3A_G}\) and \(\mathbf {L3B_G}\). In this work we continue the study of these logics, providing sound and complete Hilbert-type axiomatic systems for each logic. We prove among other properties that \(\mathbf {L3A_G}\) and \(\mathbf {L3B_G}\) satisfy a restricted version of the Substitution Theorem, and that both of them are maximal with respect to Classical Propositional Logic. To conclude we make some comparisons between \(\mathbf {L3A_G}\) and \(\mathbf {L3B_G}\) and among other logics, for instance \({\mathbf {Int}}\) and some \({\mathbf {LFI}}\)s.
Similar content being viewed by others
Notes
The hypothesis in the corollary are presented slightly different since they consider the notion of algebra, subalgebra and domain \(\{0,1/2,1\}\), instead of \(\{0,1,2\}\), however they are equivalent.
A multi-valued operator \(\circledast \) is a conservative extension of a bi-valued operator if the restriction of \(\circledast \) to the values of the bi-valued operator coincide.
A bivalent \(\lnot \)-interpretation is a function f that associates a two-valued truth-table with each connective of the logic, such that \(f(\lnot )\) is the classical truth table for negation.
References
Arieli, O., Avron, A., Zamansky, A.: Ideal paraconsistent logics. Stud. Log. 99(1), 31 (2011)
Arieli, O., Avron, A., Zamansky, A.: Maximal and premaximal paraconsistency in the framework of three-valued semantics. Stud. Log. 97(1), 31–60 (2011)
Arieli, O., Avron, A., Zamansky, A.: What is an ideal logic for reasoning with inconsistency? In: Twenty-Second International Joint Conference on Artificial Intelligence (2011)
Arrazola, J., Osorio, M., Ariza, E.: The N’5 logic. In: Proceedings of the 6th Latin American Workshop on Non-Monotonic Reasoning (2010)
Béziau, J.-Y.: Are paraconsistent negations negations? Lecture notes in pure and applied mathemathics, pp. 465–486 (2002)
Béziau, J.-Y.: What is paraconsistent logic. In: Batens et al.[6. Citeseer (2000)
Béziau, J.-Y.: Two genuine 3-valued paraconsistent logics. In: Towards Paraconsistent Engineering, pp. 35–47. Springer (2016)
Béziau, J-Y., Franceschetto, A.: Strong three-valued paraconsistent logics. In: New directions in paraconsistent logic, pages 131–145. Springer (2015)
Caret, C.: Hybridized paracomplete and paraconsistent logics. The Aust. J. Log. 14(1), (2017)
Carnielli, W., Coniglio, M.E., Marcos, J.: Logics of formal inconsistency. In: Handbook of Philosophical Logic, pp. 1–93. Springer (2007)
Coniglio, M.E., Esteva, F., Gispert, J., Godo, L.: Maximality in finite-valued Łukasiewicz logics defined by order filters. J. Log. Comput. 29(1), 125–156 (2018)
Da Costa, N.C.A., et al.: On the theory of inconsistent formal systems. Notre dame J. Formal Log. 15(4), 497–510 (1974)
Hernández-Tello, A., Ramírez, J.A., Galindo, M.O.: The pursuit of an implication for the logics L3A and L3B. Log. Univ. 11(4), 507–524 (2017)
Jaskowski, S.: Rachunek zdan dla systemów dedukcyjnych sprzecznych. Stud. Soc. Sci. Torunensis, Sectio A 1, 57–77 (1948)
Mendelson, E.: Introduction to Mathematical Logic. CRC Press, London (2009)
Pelletier, F.J., Urquhart, A.: Synonymous logics. J. Philos. Log. 32(3), 259–285 (2003)
Petrukhin, Y.: Proof theory for genuine paraconsistent logics L3A and L3B (a preliminary draft). Unpublished manuscript avaliable athttps://www.unilog.org/prize/Russia.pdf, (2018)
Priest, G., Tanaka, K., Weber, Z.: Paraconsistent Logic. In: Zalta, E.N. (ed), The Stanford Encyclopedia of Philosophy. Metaphysics Research Lab, Stanford University, summer 2018 edition (2018)
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendices
Appendix A: Some Proofs in \(\mathbf {L3A_G}\)
1.1 A.1: Proof of Lemma 1
Proof
- (a) :
-
\({\varvec{\vdash \lnot \lnot \varphi \rightarrow \varphi }}\).
- (b) :
-
If \({\varvec{\vdash \varphi \rightarrow \psi }}\) and \({\varvec{\vdash \psi \rightarrow \sigma }}\) then \({\varvec{\vdash \varphi \rightarrow \sigma }}\).
By DT we have \( \varphi \vdash \psi \) and \( \psi \vdash \sigma \), then by Cut \(\varphi \vdash \sigma \), finally applying DT again we get \(\vdash \varphi \rightarrow \sigma \).
- (c) :
-
\({\varvec{ \vdash \big (\varphi \rightarrow (\psi \rightarrow \sigma ) \big ) \rightarrow \big (\psi \rightarrow (\varphi \rightarrow \sigma ) \big ) }}\).
- (d) :
-
\({\varvec{ \vdash (\lnot \lnot \psi \rightarrow \lnot \lnot \varphi ) \rightarrow (\lnot \varphi \rightarrow \lnot \psi )}} \).
- (e) :
-
\({\varvec{\vdash (\lnot \psi \rightarrow \lnot \varphi ) \rightarrow (\lnot \lnot \varphi \rightarrow \lnot \lnot \psi )}}\).
- (f) :
-
\({\varvec{\vdash \lnot \varphi \rightarrow \lnot \lnot \lnot \varphi }}\).
- (g) :
-
\({\varvec{\vdash (\lnot \varphi \wedge \lnot \lnot \varphi ) \rightarrow \psi }}\).
From \(\lnot \varphi \wedge \lnot \lnot \varphi \) as hypothesis and \(\lnot \lnot \varphi \rightarrow ( \lnot \varphi \rightarrow \psi )\), which is valid by WE. We obtain \(\lnot \varphi \wedge \lnot \lnot \varphi \vdash \psi \), then applying DT we get the claim.
- (h) :
-
\({\varvec{ G(\varphi ), \varphi \vdash \psi }}\).
- (i) :
-
\({\varvec{ G(\varphi )\vdash \lnot \varphi }}\).
- (j) :
-
\({\varvec{ \vdash \big ( \varphi \rightarrow \psi \big ) \rightarrow \big ( G(\psi ) \rightarrow G(\varphi ) \big ) }}\).
- (k) :
-
If \({\varvec{\varphi \vdash \psi }}\) and \({\varvec{ \sigma \vdash \xi }}\) then \({\varvec{ \varphi \wedge \sigma \vdash \psi \wedge \xi }}\).
- (l) :
-
If \({\varvec{G(\varphi )\vdash \psi , N(\varphi )\vdash \psi }}\) and \({\varvec{ \lnot \lnot \varphi \vdash \psi }}\), then \({\varvec{ \vdash \psi }}\).
- (m) :
-
\({\varvec{\varphi \wedge \psi \vdash \sigma }}\) if and only if \({\varvec{\varphi , \psi \vdash \sigma }}\).
\(\square \)
1.2 A.2: Proof of Lemma 2
Proof
- N1 :
-
\({\varvec{\vdash G(\varphi ) \rightarrow \lnot \lnot (\lnot \varphi )}}.\)
- N2 :
-
\({\varvec{\vdash N(\varphi )\rightarrow \lnot \lnot (\lnot \varphi )}}.\)
- N3 :
-
\({\varvec{\vdash \lnot \lnot \varphi \rightarrow G(\lnot \varphi )}}.\)
- I1 :
-
\({\varvec{\vdash G(\varphi )\rightarrow \lnot \lnot (\varphi \rightarrow \psi )}}.\)
- I2 :
-
\({\varvec{\vdash \lnot \lnot \psi \rightarrow \lnot \lnot (\varphi \rightarrow \psi )}}.\)
- I3 :
-
\({\varvec{\vdash (N(\varphi )\wedge N(\psi ))\rightarrow \lnot \lnot (\varphi \rightarrow \psi )}}.\)
- I4 :
-
\({\varvec{\vdash (\lnot \lnot \varphi \wedge N(\psi ))\rightarrow N(\varphi \rightarrow \psi )}}.\)
- I5 :
-
\({\varvec{\vdash (N(\varphi )\wedge G(\psi ))\rightarrow G(\varphi \rightarrow \psi )}}.\)
- I6 :
-
\({\varvec{\vdash (\lnot \lnot \varphi \wedge G(\psi ))\rightarrow G(\varphi \rightarrow \psi )}}.\)
- C1 :
-
\({\varvec{ \vdash G(\varphi )\rightarrow G(\varphi \wedge \psi )}}.\)
- C2 :
-
\({\varvec{\vdash (N(\varphi )\wedge N(\psi ))\rightarrow N(\varphi \wedge \psi )}}.\)
- C3 :
-
\({\varvec{\vdash (N(\varphi )\wedge \lnot \lnot \psi )\rightarrow \lnot \lnot (\varphi \wedge \psi )}}.\)
- C4 :
-
\({\varvec{\vdash (\lnot \lnot \varphi \wedge N(\psi ))\rightarrow \lnot \lnot (\varphi \wedge \psi )}}.\)
The proof is analogous to the case of C3 but starting with \(\mathbf {L3A5}\) instead of \(\mathbf {L3A4}\).
- C5 :
-
\({\varvec{\vdash (\lnot \lnot \varphi \wedge \lnot \lnot \psi )\rightarrow \lnot \lnot (\varphi \wedge \psi )}}.\)
- D1 :
-
\({\varvec{\vdash \lnot \lnot \varphi \rightarrow \lnot \lnot (\varphi \vee \psi )}}.\)
- D2 :
-
\({\varvec{\vdash (G(\varphi )\wedge N(\psi ))\rightarrow N(\varphi \vee \psi )}}.\)
- D3 :
-
\({\varvec{\vdash (G(\varphi )\wedge N(\psi ))\rightarrow N(\varphi \vee \psi )}}.\)
The proof is analogous to the proof of D2 but using Pos6 instead of Pos7.
- D4 :
-
\({\varvec{\vdash (N(\varphi )\wedge N(\psi ))\rightarrow N(\varphi \vee \psi )}}.\)
- D5 :
-
\({\varvec{\vdash (G(\varphi )\wedge G(\psi ))\rightarrow G(\varphi \vee \psi )}}.\)
\(\square \)
Appendix B: Some Proofs in \(\mathbf {L3B_G}\)
1.1 B.1: Proof of Lemma 6
Proof
- (a) :
-
\({\varvec{\vdash \lnot \lnot \varphi \rightarrow \varphi }}.\)
- (b) :
-
If \({\varvec{\vdash \varphi \rightarrow \psi }}\) and \({\varvec{\vdash \psi \rightarrow \sigma }}\) then \({\varvec{\vdash \varphi \rightarrow \sigma }}\).
See proof of Lemma 1, part b).
- (c) :
-
\({\varvec{ \vdash \big (\varphi \rightarrow (\psi \rightarrow \sigma ) \big ) \rightarrow \big (\psi \rightarrow (\varphi \rightarrow \sigma ) \big ) }}\).
See proof of Lemma 1, part c).
- (d) :
-
\({\varvec{ \varphi \wedge \psi \vdash \sigma }}\) if and only if \({\varvec{ \varphi , \psi \vdash \sigma }}\).
See proof of Lemma 1, part m).
- (e) :
-
\({\varvec{\vdash \big ( \varphi \wedge \lnot (\varphi \wedge \varphi ) \big ) \rightarrow \psi }}\).
- (f) :
-
\({\varvec{\vdash \big (\lnot \lnot \varphi \wedge \lnot (\varphi \wedge \varphi ) \big ) \rightarrow \psi }}\).
- (g) :
-
\({\varvec{ G(\varphi ), \varphi \vdash \psi }}\).
- (h) :
-
\({\varvec{ G(\varphi )\vdash \lnot \varphi }}\).
- (i) :
-
\({\varvec{ \lnot G(\varphi )\vdash \varphi }}\).
- (j) :
-
\({\varvec{ \varphi \vdash G\big (G( \varphi )\big ) }}\).
- (k) :
-
\({\varvec{ G\big (G(\varphi )\big ) \vdash \varphi }}\).
- (l) :
-
\({\varvec{ \vdash \big ( \varphi \rightarrow \psi \big ) \rightarrow \big ( G(\psi ) \rightarrow G(\varphi ) \big ) }}\).
- (m) :
-
If \({\varvec{ \varphi \vdash \psi }}\) and \({\varvec{ \sigma \vdash \xi }}\) then \({\varvec{ \varphi \wedge \sigma \vdash \psi \wedge \xi }}\).
See proof of Lemma 1, part k).
- (n) :
-
If \({\varvec{G(\varphi )\vdash \psi }}\), \({\varvec{N(\varphi )\vdash \psi }}\) and \({\varvec{\lnot G(\varphi ) \wedge \lnot N(\varphi )\vdash \psi }}\), then \({\varvec{\vdash \psi }}\).
\(\square \)
1.2 B.2: Proof of Lemma 7
Proof
- N’1 :
-
\({\varvec{\vdash G(\varphi ) \rightarrow D(\lnot \varphi )}}\).
- N’2 :
-
\({\varvec{\vdash N(\varphi )\rightarrow N(\lnot \varphi )}}\).
- N’3 :
-
\({\varvec{\vdash D(\varphi ) \rightarrow G(\lnot \varphi )}}\).
- I’1 :
-
\({\varvec{\vdash G(\varphi )\rightarrow D(\varphi \rightarrow \psi )}}\).
- I’2 :
-
\({\varvec{\vdash D(\psi ) \rightarrow D(\varphi \rightarrow \psi )}}\).
- I’3 :
-
\({\varvec{\vdash (N(\varphi )\wedge N(\psi ))\rightarrow D(\varphi \rightarrow \psi )}}\).
- I’4 :
-
\({\varvec{\vdash (D(\varphi )\wedge N(\psi ))\rightarrow N(\varphi \rightarrow \psi )}}\).
- I’5 :
-
\({\varvec{\vdash (N(\varphi )\wedge G(\psi ))\rightarrow G(\varphi \rightarrow \psi )}}\).
- I’6 :
-
\({\varvec{\vdash (D(\varphi ) \wedge G(\psi )) \rightarrow G(\varphi \rightarrow \psi )}}\).
- C’1 :
-
\({\varvec{ \vdash G(\varphi )\rightarrow G(\varphi \wedge \psi )}}\).
- C’2 :
-
\({\varvec{\vdash (N(\varphi )\wedge N(\psi ))\rightarrow D(\varphi \wedge \psi )}}\).
- C’3 :
-
\({\varvec{\vdash (D(\varphi )\wedge D(\psi )) \rightarrow D(\varphi \wedge \psi )}}\).
- C’4 :
-
\({\varvec{\vdash (N(\varphi ) \wedge D(\psi )) \rightarrow N(\varphi \wedge \psi )}}\).
- C’5 :
-
\({\varvec{\vdash (D(\varphi ) \wedge N(\psi )) \rightarrow N(\varphi \wedge \psi )}}\).
The proof is analogous to the case of C’4 but starting with \(\mathbf {L3B5}\) instead of \(\mathbf {L3B4}\).
- D’1 :
-
\({\varvec{\vdash D(\varphi ) \rightarrow D(\varphi \vee \psi )}}\).
- D’2 :
-
\({\varvec{\vdash (G(\varphi )\wedge N(\psi ))\rightarrow N(\varphi \vee \psi )}}\).
- D’3 :
-
\({\varvec{\vdash (N(\varphi )\wedge G(\psi ))\rightarrow N(\varphi \vee \psi )}}\).
The proof is analogous to the proof of D’2 but using Pos6 instead of Pos7.
- D’4 :
-
\({\varvec{\vdash (N(\varphi ) \wedge N(\psi )) \rightarrow N(\varphi \vee \psi )}}\).
- D’5 :
-
\({\varvec{\vdash (G(\varphi )\wedge G(\psi ))\rightarrow G(\varphi \vee \psi )}}\).
\(\square \)
1.3 B.3: Proof of Lemma 8
Proof
Let \(\varphi \in Form(\mathbf {L3B_G})\) and v be a three-valuation in \(\mathbf {L3B_G}\). The proof will be done by induction on the complexity of \(\varphi \).
Base Case: If \(\varphi =p\), with p an atomic formula, then \(Atoms(\varphi )_v = \varphi _v= p_v\), but we know that \(\varphi _v\vdash \varphi _v\), therefore \(Atoms(\varphi )_v\vdash \varphi _v.\)
Let us suppose now that for any \(\beta \in Form(\mathbf {L3B_G})\) with lesser complexity than those of \(\varphi \) the claim is true, namely \(Atoms(\beta )_v\vdash \beta _v.\) Now we proceed to analyse what happen with each connective in \(\mathbf {L3B_G}\).
Case \({\varvec{\lnot }}\): Suppose that \(\varphi =\lnot \psi \), we know that \(Atoms(\varphi )=Atoms(\psi )\) and by inductive hypothesis we have \(Atoms(\varphi )_v\vdash \psi _{v}\). Depending what is the valuation of \(\varphi \) we have one of the following cases:
-
1.
If \(v(\varphi )=0\), then \(v(\psi )=2\), \(\psi _v=D(\psi )\), \(\varphi _v=G(\varphi )=G(\lnot \psi )\) and \(Atoms(\varphi )_v \vdash D(\psi )\). By N’3 we have \(D(\psi ) \vdash G(\lnot \psi )\) then applying Cut we conclude \(Atoms(\varphi )_v\vdash G(\lnot \psi )\).
-
2.
If \(v(\varphi )=1\), then \(v(\psi )=1\), \(\psi _v=N(\psi )\), \(\varphi _v = N(\varphi ) = N (\lnot \psi )\) and \(Atoms(\varphi )_v \vdash N(\psi )\). By N’2 we have \(N(\psi ) \vdash N(\lnot \psi )\) then applying Cut we conclude \(Atoms(\varphi )_v \vdash N(\lnot \psi )\).
-
3.
If \(v(\varphi )=2\), then \(v(\psi )=0\) then \(\psi _ v=G( \psi )\), \(\varphi _v = D(\varphi ) = D(\lnot \psi )\) and \(Atoms(\varphi )_v\vdash G(\psi )\) and by property N’1, \(G(\psi ) \vdash D(\lnot \psi )\) therefore \(Atoms(\varphi )_v \vdash D(\lnot \psi )\).
No matter what be the value of \(\varphi \), we conclude \(Atoms(\varphi )_v\vdash \varphi _v\).
Case \({\varvec{\rightarrow }}\): Suppose that \(\varphi =\psi \rightarrow \sigma \). By inductive hypothesis we know that \(Atoms(\psi )_v \vdash \psi _{v}\) and \(Atoms(\sigma )_v \vdash \sigma _{v}\). Since \(Atoms(\varphi )_v= Atoms(\psi )_v \cup Atoms(\sigma )_v\) then by Mon we have \(Atoms(\varphi )_v \vdash \psi _v\) and \(Atoms(\varphi )_v \vdash \sigma _v\). Depending what is the valuation of \(\varphi \) we must consider the following cases:
-
1.
If \(v(\varphi )=0\) we have two possible scenarios \(v(\psi )=1\) and \(v(\sigma )=0\) or \(v(\psi )=2\) and \(v(\sigma )=0\), in any case we have \(\varphi _v=G(\varphi )=G(\psi \rightarrow \sigma )\).
-
If \(v(\psi )=1\) and \(v(\sigma )=0\) by Definition 8, \(\psi _v=N(\psi )\), \(\sigma _v=G(\sigma )\), \(Atoms(\varphi )_v \vdash N(\psi )\), \(Atoms(\varphi )_v \vdash G(\sigma )\), and by property I’5 we know that \(\vdash (N(\psi )\wedge G(\sigma )) \rightarrow G(\psi \rightarrow \sigma )\) then applying R-AND and MP we obtain \(Atoms(\varphi )_v\vdash G(\psi \rightarrow \sigma )\).
-
If \(v(\psi )=2\) and \(v(\sigma )=0\), then we have \(\psi _v=D(\psi )\), \(\sigma _v=G(\sigma )\), \(Atoms(\varphi )_v \vdash D(\psi )\), \(Atoms(\varphi )_v \vdash G(\sigma )\), besides by I’6 we know that \(\vdash (D(\psi ) \wedge G(\sigma )) \rightarrow G(\psi \rightarrow \sigma )\) again as in the previous case, applying R-AND and MP we have \(Atoms(\varphi )_v\vdash G(\psi \rightarrow \sigma )\).
-
-
2.
If \(v(\varphi )=1\), then there is only one case, namely \(v(\psi )=2\) and \(v(\sigma )=1\). We have that \(\psi _v= D(\psi )\), \(\sigma _v = N(\sigma )\), \(\varphi _v = N(\varphi ) = N(\psi \rightarrow \sigma )\), also we have that \(Atoms(\varphi )_v \vdash D(\psi )\), \(Atoms(\varphi )_v \vdash N(\sigma )\), and by property I’4 we know that \(\vdash (D(\psi ) \wedge N(\sigma )) \rightarrow N(\psi \rightarrow \sigma )\) then applying R-AND and MP we obtain \(Atoms(\varphi )_v\vdash N(\psi \rightarrow \sigma )\).
-
3.
If \(v(\varphi )=2\) then we have three possibilities, namely, \(v(\psi )=0\) or \(v(\sigma )=2\) or \(v(\psi ) = v(\sigma ) = 1\), in any situation we have \(\varphi _v=D(\varphi ) = D(\psi \rightarrow \sigma )\).
-
If \(v(\psi )=0\) then \(\psi _v=G(\psi )\), \(Atoms(\varphi )_v \vdash G(\psi )\) and by property I’1 \(G(\psi ) \vdash D(\psi \rightarrow \sigma )\) then by Cut we conclude that \(Atoms(\varphi )_v \vdash D(\psi \rightarrow \sigma )\).
-
If \(v(\sigma ) = 2\) then \(\sigma _v= D(\sigma )\), \(Atoms(\varphi )_v \vdash D(\sigma )\) and by I’2 we know that \(D(\sigma ) \vdash D (\psi \rightarrow \sigma )\) therefore using Cut we derive \(Atoms(\varphi )_v\vdash D(\psi \rightarrow \sigma )\).
-
If \(v(\psi ) = v(\sigma ) = 1\) then \(\psi _v = N(\psi )\), \(\sigma _v = N(\sigma )\), \(Atoms(\varphi )_v \vdash N(\psi )\), \(Atoms(\varphi )_v \vdash N(\sigma )\) and by I’3 \( N(\psi ) \wedge N(\sigma ) \vdash D(\psi \rightarrow \sigma )\) therefore using R-AND and Cut we obtain \(Atoms(\varphi )_v\vdash D(\psi \rightarrow \sigma )\).
Note that for each item, it was verified that: \(Atoms(\varphi )_v\vdash \varphi _v\). So in this case the claim is verified.
-
Case \({\varvec{\wedge }}\): Suppose that \(\varphi =\psi \wedge \sigma \). Analogously to the case of implication, we know that \(Atoms(\varphi )_v \vdash \psi _v\) and \(Atoms(\varphi )_v \vdash \sigma _v\). With base on the valuation of \(\varphi \) we must study the following cases:
-
1.
If \(v(\varphi )=0\) we have two possible scenarios \(v(\psi )=0\) or \(v(\sigma )=0\), in any case we have \(\varphi _v = G(\varphi ) = G(\psi \wedge \sigma )\).
-
If \(v(\psi )=0\), then \(\psi _v= G(\psi )\), \(Atoms(\varphi )_v \vdash G(\psi )\), and by property C’1 we know that \(G(\psi ) \vdash G(\psi \wedge \sigma )\) hence \(Atoms(\varphi )_v \vdash G(\psi \wedge \sigma )\).
-
If \(v(\sigma )=0\), using the same argument that in the previous case we have \(Atoms(\varphi )_v \vdash G(\psi \wedge \sigma )\).
-
-
2.
If \(v(\varphi )=1\), then \(v(\sigma )=1\) and \(v(\psi )=2\) or \(v(\sigma )=2\) and \(v(\psi )=1\). We have that \(\varphi _v= N(\varphi )=N(\sigma \wedge \psi )\).
-
If \(v(\psi )=1, v(\sigma )=2\) then \(\psi _v = N(\psi )\), \(\sigma _v= D(\sigma )\), \(Atoms(\varphi )_v \vdash N(\psi )\), \(Atoms(\varphi )_v \vdash D(\sigma )\) and by property C’4 we have \(N(\psi ) \wedge D(\sigma ) \vdash N(\psi \wedge \sigma )\) then using R-AND and Cut we conclude \(Atoms(\varphi )_v \vdash N(\psi \wedge \sigma )\).
-
If \(v(\psi )=2, v(\sigma )=1\) by a similar analysis to the previous case, now using C’5, we have \(Atoms(\varphi )_v \vdash N(\psi \wedge \sigma )\).
-
-
3.
If \(v(\varphi )=2\) then we have two possible situations, \(v(\psi )=2, v(\sigma )=2\) or \(v(\psi )=1, v(\sigma )=1\), in any situation we have \(\varphi _v = D(\varphi ) = D(\psi \wedge \sigma )\).
-
If \(v(\psi )=2, v(\sigma )=2\) then \(\psi _v = D(\psi )\), \(\sigma _v= D(\sigma )\), \(Atoms(\varphi )_v \vdash D(\psi )\), \(Atoms(\varphi )_v \vdash D(\sigma )\) and by property C’3 \(D(\psi ) \wedge D(\sigma ) \vdash D(\psi \wedge \sigma )\) then using R-AND and Cut we conclude \(Atoms(\varphi )_v \vdash D(\psi \wedge \sigma )\).
-
If \(v(\psi ) = v(\sigma ) = 1\) then \(\psi _v = N(\psi )\), \(\sigma _v = N(\sigma )\), \(Atoms(\varphi )_v \vdash N(\psi )\), \(Atoms(\varphi )_v \vdash N(\sigma )\) and by C’2 \( N(\psi ) \wedge N(\sigma ) \vdash D(\psi \wedge \sigma )\) therefore using R-AND and Cut we obtain \(Atoms(\varphi )_v\vdash D(\psi \wedge \sigma )\).
In the three cases of conjunction we conclude that \(Atoms(\varphi )_v\vdash \varphi _v\).
-
Case \({\varvec{\vee }}\): Suppose that \(\varphi =\psi \vee \sigma \). Analogously to the previous case, we know that \(Atoms(\varphi )_v \vdash \psi _v\) and \(Atoms(\varphi )_v \vdash \sigma _v\). With base on the valuation of \(\varphi \) we must study the following cases:
-
1.
If \(v(\varphi )=0\) then \(v(\psi )=v(\sigma )=0\) from here \(\psi _v = G(\psi )\), \(\sigma _v= G(\sigma )\), \(\varphi _v = G(\varphi ) = G(\psi \vee \sigma )\) and \(Atoms(\varphi )_v \vdash G(\psi )\), \(Atoms(\varphi )_v \vdash G(\sigma )\) and by property D’5, \(G(\psi ) \wedge G(\sigma ) \vdash G( \psi \vee \sigma )\) then using R-AND and Cut we conclude \(Atoms(\varphi )_v \vdash G(\psi \vee \sigma )\).
-
2.
If \(v(\varphi )=1\), we have three possible situations, \(v(\psi )=1, v(\sigma )=0\) or \(v(\psi )=0, v(\sigma )=1\) or \(v(\psi ) = v(\sigma ) = 1\), in any situation we have \(\varphi _v = N(\varphi ) = N(\psi \wedge \sigma )\).
-
If \(v(\psi )=1, v(\sigma )=0\) then \(\psi _v = N(\psi )\), \(\sigma _v=G(\sigma )\), \(Atoms(\varphi )_v \vdash N(\psi )\), \(Atoms(\varphi )_v \vdash G(\sigma )\) and by property D’2 \(N(\psi ) \wedge G(\sigma ) \vdash N(\psi \vee \sigma )\) then using R-AND and Cut we conclude \(Atoms(\varphi )_v \vdash N(\psi \vee \sigma )\).
-
If \(v(\psi )=0, v(\sigma )=1\) by a similar analysis to the previous case, using D’3 we have \(Atoms(\varphi )_v \vdash N(\psi \vee \sigma )\).
-
If \(v(\psi ) = v(\sigma ) = 1\) then \(\psi _v = N( \psi \)), \(\sigma _v = N(\sigma )\), \(Atoms(\varphi )_v \vdash N(\psi )\), \(Atoms(\varphi )_v \vdash N(\sigma )\) and by D’4 \(N(\psi ) \wedge N(\sigma ) \vdash N(\psi \vee \sigma )\) therefore using R-AND and Cut we obtain \(Atoms(\varphi )_v\vdash N(\psi \vee \sigma )\).
-
-
3.
If \(v(\varphi )=2\) then we have two possible scenarios, \(v(\psi )=2\) or \(v(\sigma )=2\), in any situation we have \(\varphi _v = D(\varphi ) = D(\psi \vee \sigma )\).
-
If \(v(\psi )=2\), then \(\psi _v= D(\psi )\), \(Atoms(\varphi )_v \vdash D(\psi )\), and by property D’1 we know that \(D(\psi ) \vdash D(\psi \vee \sigma )\) hence \(Atoms(\varphi )_v \vdash D(\psi \vee \sigma )\).
-
If \(v(\sigma )=2\), using the same argument that in the previous case we have \(Atoms(\varphi )_v \vdash D(\psi \vee \sigma )\).
Accordingly to the previous analysis the proof is complete.\(\square \)
-
Rights and permissions
About this article
Cite this article
Hernández-Tello, A., Pérez-Gaspar, M. & Borja Macías, V. Axiomatisations of the Genuine Three-Valued Paraconsistent Logics \(\mathbf {L3A_G}\) and \(\mathbf {L3B_G}\). Log. Univers. 15, 87–121 (2021). https://doi.org/10.1007/s11787-021-00269-2
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11787-021-00269-2