Skip to main content

On strong standard completeness in some MTL\(_\Delta \) expansions

Abstract

In this paper, inspired by the previous work of Franco Montagna on infinitary axiomatizations for standard \(\mathsf {BL}\)-algebras, we focus on a uniform approach to the following problem: given a left-continuous t-norm \(*\), find an axiomatic system (possibly with infinitary rules) which is strongly complete with respect to the standard algebra This system will be an expansion of Monoidal t-norm-based logic. First, we introduce an infinitary axiomatic system \(\mathsf {L}_*^\infty \), expanding the language with \(\Delta \) and countably many truth constants, and with only one infinitary inference rule, that is inspired in Takeuti–Titani density rule. Then we show that \(\mathsf {L}_*^\infty \) is indeed strongly complete with respect to the standard algebra . Moreover, the approach is generalized to axiomatize expansions of these logics with additional operators whose intended semantics over [0, 1] satisfy some regularity conditions.

This is a preview of subscription content, access via your institution.

Fig. 1

Notes

  1. 1.

    As it is usual in the literature, the symbol \(\Delta \) denotes both the connective in the language and the corresponding operation in the algebra, as it happens with \(\wedge \) and \(\vee \).

  2. 2.

    The \(\Delta \) connective is used to prove semilinearity of the axiomatic system.

  3. 3.

    This is simply a subset, closed under substitutions, of \(\{ \langle \Gamma , \varphi \rangle : \Gamma \cup \{\varphi \} \subseteq Fm \}\). Notice that we allow rules with infinite premises \(\Gamma \).

  4. 4.

    Let us stress that we do not consider the smallest consequence operator closed under substitutions.

  5. 5.

    Let us point out that indeed this condition implies that for every natural number \(m \leqslant n\), it holds that either

    • \(\xi _m \in \Sigma _{n+1}\), or

    • \(\lnot \Delta \xi _m \in \Sigma _{n+1}\) and there is some \(\psi \in \Phi _m\) such that \(\lnot \Delta \psi \in \Sigma _{n+1}\).

    Notice that this last disjunction is capturing the intuition that \(\Sigma _{n+1}\) (and all its extensions) is a model of all the derivations in \(\langle (\Phi _{m}, \xi _{m}): m \leqslant n \rangle \).

  6. 6.

    This finiteness is employed to show that for each rule in the axiomatic system, there is at most a countable number of substitutions.

  7. 7.

    If for all two constants like above \(d_1 *d_2 \leqslant c\), applying residuation \(d_1 \leqslant d_2 \rightarrow c\) and so the supremum can be taken in the left side. Similarly, we get that \(\sup \mathcal {C}^-_a *\sup \mathcal {C}^-_b \leqslant c\) which contradicts the assumptions.

  8. 8.

    This can be also seen as a direct consequence of the fact that all \(L_*^\infty \)-chains are relatively simple.

  9. 9.

    For the reader interested in checking the details we suggest to start considering the following three elements in \(\varvec{A}\):

    $$\begin{aligned} t_1:=\left( \frac{1}{2}\right) _{q \in [0,1)_\mathbb {Q}} \quad t_2:=(\widetilde{q})_{q \in [0,1)_\mathbb {Q}} \quad t_3:=(q)_{q \in [0,1)_\mathbb {Q}} \end{aligned}$$

    and checking that all possible combinations of these three elements under \(\wedge , \vee , \rightarrow , \Delta \) are also elements in our universe A. Indeed, all difficulties to provide a general proof that A is closed under the operations are illustrated in the previous particular case.

  10. 10.

    With the notation \(x \leqslant r \uparrow \) we mean that \(x \leqslant \overline{c}^{\varvec{A}}\) for every \(c \in (r,1]_\mathbb {Q}\). In an analogous way, \(r \downarrow \leqslant x\) stands for \(\overline{c}^{\varvec{A}} \leqslant x\) for all \(c \in [0,r)_\mathbb {Q}\).

  11. 11.

    As expected, \(\Upsilon _i^f(\vec {\varphi })\) stands for \(\{\lambda (\vec {\varphi })\}_{\lambda \in \Upsilon _i}\), and \(\vec {e(\varphi )} = \langle e(\varphi _1), \ldots e(\varphi _n) \rangle \).

  12. 12.

    Observe that if component k of region \(R_i\) has only one point, these premises are never met. Thus, the rule associated to that pair of region and component is equivalent to the rule \(\overline{0} \vdash \vartheta \), and so, it will not be used in order to “compute” the value of the function in that point (this will be done with a different component or a book-keeping axiom, depending on the case).

  13. 13.

    As already mentioned before concerning operations with regions with smaller dimensions, at this point of the proof, if f was an operation such that \(\{d \in A: (\Upsilon _i^f)^{\varvec{A}}(\langle \vec {b}, d \rangle _k \in \mathrm{{Int}}_f(R_i) \text { for some } \vec {b} \in A^{n}\} = \{\overline{c_0}^{\varvec{A}}\}\) for some \(c_0 \in \mathbb {Q}_*^{\mathrm{{OP}}}\), the proof would finish here. Indeed, \(c_0\) in \(C_1\) by definition, and so trivially \( \overline{f}^{\varvec{A}}(\overline{c_0}^{\varvec{A}}, a_2) \leqslant \sup \{\overline{f}^{\varvec{A}}(\overline{c}^{\varvec{A}}, a_2): c \in [0,b]_\mathbb {Q}, \overline{c}^{\varvec{A}} \leqslant a_1\} \).

  14. 14.

    This set is non-empty: \(x_1 > 0\) because \((\Upsilon _1^g)^{\varvec{A}}(x_1, x_2) = \{\overline{1}^{\varvec{A}}\}\) and the constants are dense in A.

  15. 15.

    Nevertheless, the case of the left-continuous t-norm operation \(*\) has a more direct proof, not needing any of the , \(\mathrm {\vee M^{f U}_{\langle i,k\rangle }}\) nor rules, relying on the MTL-axiomatization of a residuated operation, as we saw in Lemma 4.7.

References

  1. Agliano P, Montagna F (2003) Varieties of BL-algebras. I. General properties. J Pure Appl Algebra 181(2–3):105–129

    Article  MATH  MathSciNet  Google Scholar 

  2. Blok WJ, Pigozzi D (1989) Algebraizable logics, vol 396. Memoirs of the American Mathematical Society AMS, Providence

    MATH  Google Scholar 

  3. Bou F, Esteva F, Godo L, Rodríguez R (2011) On the minimum many-valued modal logic over a finite residuated lattice. J Logic Comput 21(5):739–790

    Article  MATH  MathSciNet  Google Scholar 

  4. Cignoli R, Esteva F, Godo L, Torrens A (2000) Basic fuzzy logic is the logic of continuous t-norms and their residua. Soft Comput 4(2):106–112

    Article  Google Scholar 

  5. Cintula P (2004) From fuzzy logic to fuzzy mathematics. Ph.D. thesis, Czech Technical University in Prague

  6. Cintula P (2016) A note on axiomatizations of Pavelka-style complete fuzzy logics. Fuzzy Sets Syst 292:160–174

    Article  MathSciNet  Google Scholar 

  7. Cintula P, Noguera C (2010) Implicational (semilinear) logics I: a new hierarchy. Arch Math Logic 49(4):417–446

    Article  MATH  MathSciNet  Google Scholar 

  8. Cintula P, Noguera C (2013) The proof by cases property and its variants in structural consequence relations. Studia Logica 101(4):713–747

    Article  MATH  MathSciNet  Google Scholar 

  9. Cintula P, Hájek P, Noguera C (eds) (2011) Handbook of mathematical fuzzy logic (in 2 volumes), volume 37 and 38 of studies in logic, mathematical logic and foundation. College Publications, London

    Google Scholar 

  10. Esteva F, Godo L (2001) Monoidal t-norm based logic: towards a logic for left-continuous t-norms. Fuzzy Sets Syst 124:271–288

    Article  MATH  MathSciNet  Google Scholar 

  11. Esteva F, Godo L, Montagna F (2004) Equational characterization of the subvarieties of BL generated by t-norm algebras. Studia Logica 76(2):161–200

  12. Givant S, Halmos P (2009) Introduction to boolean algebras. Undergraduate texts in mathematics. Springer, Berlin

  13. Hájek P (1998) Metamathematics of fuzzy logic, volume 4 of trends in logic. Studia Logica Library. Kluwer Academic Publishers, Dordrecht

    Book  Google Scholar 

  14. Haniková Z (2014) Varieties generated by standard BL-algebras. Order 31(1):15–33

    Article  MATH  MathSciNet  Google Scholar 

  15. Jenei S, Montagna F (2002) A proof of standard completeness for Esteva and Godo’s logic MTL. Studia Logica 70:183–192

    Article  MATH  MathSciNet  Google Scholar 

  16. Montagna F (2007) Notes on strong completeness in Łukasiewicz, product and BL logics and in their first-order extensions. In: Algebraic and proof-theoretic aspects of non-classical logics: papers in honor of Daniele Mundici on the occasion of his 60th birthday, LNCS, vol 4460. Springer, pp 247–274

  17. Pavelka J (1979a) On fuzzy logic. I. Many-valued rules of inference. Zeitschrift für Mathematische Logik und Grundlagen der Mathematik 25(1):45–52

    Article  MATH  MathSciNet  Google Scholar 

  18. Pavelka J (1979b) On fuzzy logic. II. Enriched residuated lattices and semantics of propositional calculi. Zeitschrift für Mathematische Logik und Grundlagen der Mathematik. 25(2):119–134

    Article  MATH  MathSciNet  Google Scholar 

  19. Pavelka J (1979c) On fuzzy logic. III. Semantical completeness of some many-valued propositional calculi. Zeitschrift für Mathematische Logik und Grundlagen der Mathematik 25(5):447–464

    Article  MATH  MathSciNet  Google Scholar 

  20. Takeuti G, Titani S (1984) Intuitionistic fuzzy logic and intuitionistic fuzzy set theory. J Symb Logic 49(3):851–866

    Article  MATH  MathSciNet  Google Scholar 

  21. Vidal A, Esteva F, Godo L (2016) On modal extensions of product fuzzy logic. J Logic Comput. doi:10.1093/logcom/exv046 (in press)

  22. Vidal A, Godo L, Esteva F (2015) On strongly standard complete fuzzy logics: \({MTL}^{Q}_*\) and its expansions. In: Alonso JM, Bustince H, Reformat M (eds) Proceedings of the IFSA-EUSFLAT-15, Gijón, Spain. Atlantis Press, pp 828–835

Download references

Acknowledgments

The authors are thankful to an anonymous reviewer for his/her comments that have helped to improve the final layout of this paper. Vidal has been supported by the joint project of Austrian Science Fund (FWF) I1897-N25 and Czech Science Foundation (GACR) 15-34650L and by the institutional support RVO:67985807. Esteva and Godo have been funded by the FEDER/MINECO Spanish Project TIN2015-71799-C2-1-P and by the Grant 2014SGR-118 from the Catalan Government. Bou thanks the Grant 2014SGR-788 from the Catalan Government.

Author information

Affiliations

Authors

Corresponding author

Correspondence to Amanda Vidal.

Ethics declarations

Conflicts of interest

The authors declare that they have no conflict of interest.

Human and animal rights

This article does not contain any studies with human participants or animals performed by any of the authors.

Additional information

This paper is dedicated to the memory of Professor Franco Montagna, an excellent researcher and better person to whom these authors will always be indebted to. He has been a pioneer in many research areas (not only in MFL), and among such pioneering contributions we can find his work on infinitary axiomatic systems for standard $$\mathsf {BL}$$ BL -algebras. The present paper has been inspired by the novel ideas introduced by Franco on that topic.

This is an expanded and fully revised version of the conference paper (Vidal et al. 2015). In particular some proofs have become more cumbersome than were presumed there.

Communicated by A. Di Nola, D. Mundici, C. Toffalori, A. Ursini.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Vidal, A., Bou, F., Esteva, F. et al. On strong standard completeness in some MTL\(_\Delta \) expansions. Soft Comput 21, 125–147 (2017). https://doi.org/10.1007/s00500-016-2338-0

Download citation

Keywords

  • Mathematical fuzzy logic
  • Left-continuous t-norms
  • Monoidal t-norm logic
  • Infinitary rules
  • Standard completeness