Skip to main content
Log in

Continuous Accessibility Modal Logics

  • Published:
Journal of Philosophical Logic Aims and scope Submit manuscript

Abstract

In classical modal semantics, a binary accessibility relation connects worlds. In this paper, we present a uniform and systematic treatment of modal semantics with a continuous accessibility relation alongside the continuous accessibility modal logics that they model. We develop several such logics for a variety of philosophical applications. Our main conclusions are as follows. Modal logics with a continuous accessibility relation are sound and complete in their natural classes of models. The class of Kripke frames where a continuous accessibility relation has a magnitude characterizing its degree of accessibility is not modally definable, and this has unappreciated significance to completeness proofs for such logics, revealing a methodological advantage of using classical multimodal semantics over fuzzy modal semantics. There is a pseudometric space modal logic that is complete in the class of pseudometric spaces, a natural semantic setting for quantitative modal reasoning about similarity. There is a metric space modal logic that is complete in the class of metric spaces, a natural semantic setting for quantitative modal reasoning about neighborhoods and counterfactual stability. There is a real line continuous temporal logic that is canonical for real lines, a natural semantic setting for quantitative modal reasoning about time.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Similar content being viewed by others

References

  1. Koymans, R. (1990). Specifying real-time properties with metric temporal logic. Real-Time Systems, 2, 255–299. https://doi.org/10.1007/BF01995674.

    Article  Google Scholar 

  2. Alur, R., Feder, T., & Henzinger, T.A. (1996). The benefits of relaxing punctuality. Journal of the ACM, 43, 116–146. https://doi.org/10.1145/112600.112613.

    Article  Google Scholar 

  3. Montanari, A., & Policriti, A. Executing metric temporal logic. workshop on programming in temporal and non classical logics at international joint conferences on artificial intelligence 1997, August 1997. http://citeseerx.ist.psu.edu/viewdoc/citations?doi=10.1.1.50.1193.

  4. Baratella, S., & Masini, A. (2019). A two-dimensional metric temporal logic. Mathematical Logic Quarterly, 66, 7–19. https://doi.org/10.1002/malq.201700036.

    Article  Google Scholar 

  5. Wolter, F., & Zakharyaschev, M. (2003). Reasoning about distances. In Proceedings of the 18th international joint conference on artificial intelligence (pp. 1275–1280).

  6. Kutz, O., Wolter, F., Sturm, H., Suzuki, N.Y., & Zakharyaschev, M. (2003). Logics of metric spaces. ACM Transactions on Computational Logic, 4, 260–294. https://doi.org/10.1145/635499.635504.

    Article  Google Scholar 

  7. Wolter, F., & Zakharyaschev, M. (2005). A logic for metric and topology. Journal of Symbolic Logic, 70, 795–828. https://doi.org/10.2178/jsl/1122038915.

    Article  Google Scholar 

  8. Esteva, F., Garcia, P., Godo, L., & Rodríguez, R. (1997). A modal account of similarity-based reasoning. International Journal of Approximate Reasoning, 16, 235–260. https://doi.org/10.1016/S0888-613X(96)00126-0https://doi.org/10.1016/S0888-613X(96)00126-0.

    Article  Google Scholar 

  9. Sheremet, M., Tishkovsky, D., Wolter, F., & Zakharyaschev, M. (2007). A logic for concepts and similarity. Journal of Logic and Computation, 17, 415–452. https://doi.org/10.1093/logcom/exm007.

    Article  Google Scholar 

  10. Lando, T.A. (2012). Probabilistic semantics for modal logic. Phd thesis. University of California: Berkeley.

    Google Scholar 

  11. Baratella, S. (2019). A completeness theorem for continuous predicate modal logic. Archive for Mathematical Logic, 58, 183–201. https://doi.org/10.1007/s00153-018-0630-7.

    Article  Google Scholar 

  12. Godo, L., & Rodríguez, R. (1999). A Fuzzy modal logic for similarity reasoning. In G. Chen, M. Ying, & K. Y. Cai (Eds.) Fuzzy logic and soft computing, volume 6 of the international series on asian studies in computer and information science. Springer. https://doi.org/10.1007/978-1-4615-5261-1_3.

  13. Mironov, A.M. (2005). Fuzzy modal logics. Journal of Mathematical Sciences, 128, 3461–3483. https://doi.org/10.1007/s10958-005-0281-1.

    Article  Google Scholar 

  14. Van Benthem, J. (2010). Modal Logic for Open Minds. Center for the study of language and information Stanford University.

  15. Blackburn, P., De Rijke, M., & Venema, Y. (2001). Modal Logic. Cambridge tracts in theoretical computer science. Cambridge: Cambridge University Press. https://doi.org/10.1017/CBO9781107050884.

    Google Scholar 

  16. McKinsey, J.C.C., & Tarski, A. (1944). The algebra of topology. Annals of Mathematics, 45, 141–191. https://doi.org/10.2307/1969080.

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Corresponding authors

Correspondence to Caleb Camrud or Ranpal Dosanjh.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Appendix A: Plantability Proofs

Appendix A: Plantability Proofs

Recall the definition of a plantable property from Section 3.3:

Definition 5

Let π be a frame property. Let \(\mathfrak {F} = (M, \{R_x: {x} \in [0,\infty ]\})\) be an arbitrary frame that has π. Let \(\left (\mathfrak {M}, {a}\right )\) be an arbitrary pointed model extending \(\mathfrak {F}\). Let \(\left (\mathfrak {M}^{\dagger },\langle {a}\rangle \right )\) be its planted model, with \(\mathfrak {M}^{\dagger } = \big ({M}^{\dagger },\{R_{x}^{\dagger }: x \in [0,\infty ]\}, {V}^{\dagger }\big )\).

Property π is plantable if and only if there exists a set of accessibility relations \(\{R_{x}^{\prime } : {x} \in [0, \infty ]\}\) such that:

  1. 1.

    for all \({x}\in [0, \infty ]\), \(R_{x}^{\prime } \supseteq R_{x}^{\dagger }\);

  2. 2.

    \(\big ({M}^{\dagger },\{R_{x}^{\prime } : {x} \in [0, \infty ]\}\big )\) has π;

  3. 3.

    \(\big ({M}^{\dagger },\{R_{x}^{\prime } : {x} \in [0, \infty ]\}\big )\) is well-founded; and

  4. 4.

    for all w,uM corresponding to w and u, respectively, and for all \({x} \in [0, \infty ]\), \(R_{x}^{\prime }(w^{\dagger }, {u}^{\dagger })\) holds only if Rx(w,u) holds.

Below we show that various properties considered in this paper are plantable.

1.1 A.1 Proof of Lemma 2

We will show that upwardly closed accessibility is plantable.

Let \(\left ({\mathfrak {M}}, {a}\right )\) be an arbitrary pointed model with upwardly closed accessibility, with \(\mathfrak {M}=\left ({M}, \{R_{x} : {x} \in [0, \infty ]\}, {V}\right )\). Let \(\left (\mathfrak {M}^{\dagger },\langle {a}\rangle \right )\) be its planted model, with \(\mathfrak {M}^{\dagger }=({M}^{\dagger }, \{R_{x}^{\dagger } : {x} \in [0, \infty ]\},{{V}^{\dagger }})\). Define \(\mathfrak {M}^{\prime }:=\left ({M}^{\dagger },\{R_{x}^{\prime } : {x} \in [0, \infty ]\},{V}^{\dagger }\right )\), where \(\{R_{x}^{\prime } : {x} \in [0, \infty ]\}\) is constructed as follows.

  • Definition of \(R_{x}^{\prime }\): For w,uM and \({x,y} \in [0, \infty ]\),

    $$ R_{x}^{\prime} := \{({w}^{\dagger},{u}^{\dagger}) : R_{y}^{\dagger}({w}^{\dagger},{u}^{\dagger}) \text{ holds for some} y \leq x\}. $$

In other words, the x-accessibility relations of \(\mathfrak {M}^{\prime }\) are defined to be the upwardly closed accessibility relations of \(\mathfrak {M}^{\dagger }\).

By construction, \(\{R_{x}^{\prime } : {x} \in [0, \infty ]\}\) satisfies the first three conditions of being a plantable model. For the fourth condition, fix arbitrary w,uM, corresponding to w,uM, respectively. Suppose that \(R_{x}^{\prime }({w}^{\dagger }, {u}^{\dagger })\) holds for some \({x} \in [0, \infty ]\). By definition, there is some yx such that \(R_{y}^{\dagger }({w}^{\dagger }, {u}^{\dagger })\) holds. Since \((\mathfrak {M},{a})\) is bisimilar to \((\mathfrak {M}^{\dagger },\langle {a}\rangle )\), Ry(w,u) holds. Since \((\mathfrak {M},{a})\) has upwardly closed accessibility, Rx(w,u) holds.

1.2 A.2 Proof of Lemma 3

We will show that upwardly closed accessibility, reflexivity, symmetry, and additively transitive accessibility are jointly plantable.

Let \((\mathfrak {M},{a})\) be an arbitrary pointed model that has upwardly closed accessibility, reflexivity, symmetry, and additively transitive accessibility, with \(\mathfrak {M}=({M}, \{R_{x} : {x} \in [0, \infty ]\}, {V})\). Let \((\mathfrak {M}^{\dagger },\langle {a}\rangle )\) be its planted model, with \(\mathfrak {M}^{\dagger }=({M}^{\dagger }, \{R_{x}^{\dagger } : {x} \in [0, \infty ]\},{V}^{\dagger })\). Define \(\mathfrak {M}^{\prime }:=({M}^{\dagger },\{R_{x}^{\prime } : {x} \in [0, \infty ]\}, {V}^{\dagger })\), where \(\{{R}_{x}^{\prime } : {x} \in [0, \infty ]\}\) is constructed via transfinite recursion as follows.

  • Definition of \(R_{x}^{\prime }\):

    $$ {R_{x}^{\prime}} := \bigcup\limits_{{\alpha}\in{\text{ORD}}} R_{x}^{{\prime}{\alpha}}. $$
  • Base case: For w,uM and \({x,y}\in [0, \infty ]\),

    $$ R_{x}^{{\prime}0} := \Big\{({w}^{\dagger},{u}^{\dagger}) : R_{y}^{\dagger}({w}^{\dagger},{u}^{\dagger}) \text{or} R_{y}^{\dagger}({u}^{\dagger},{w}^{\dagger}) \text{holds for some} y \leq x \Big\}. $$
  • Successor case: For w,u,vM and \({x,y} \in [0, \infty ]\),

    $$ R_{{x}+{y}}^{{\prime}{\alpha}+1} := \left\{({w}^{\dagger},{v}^{\dagger}) : R_{x}^{{\prime}{\alpha}}({w}^{\dagger},{u}^{\dagger}) \text{and} R_{y}^{{\prime}{\alpha}}({u}^{\dagger},{v}^{\dagger}) \text{hold for some} {u}^{\dagger}\right\}. $$
  • Limit case:

    $$ R_{x}^{{\prime}{\alpha}} := \bigcup\limits_{\beta<\alpha} R_{x}^{{\prime}{\beta}}. $$

In other words, the x-accessibility relations of \(\mathfrak {M}^{\prime }\) are defined to be the upwardly closed and symmetrized accessibility relations of \(\mathfrak {M}^{\dagger }\), successively made additively transitive (which preserves upwardly closed accessibility and symmetry).

By construction, \(R_{x}^{\prime } {\supseteq } R_{x}^{\dagger }\) for all \({x} \in [0, \infty ],\) satisfying the first condition of being a plantable property. We show that \(\{R_{x}^{\prime } : {x} \in [0, \infty ]\}\) satisfies the remaining three conditions.

Lemma 1

\(\mathfrak {M}^{\prime }\) has all of upwardly closed accessibility, reflexivity, symmetry, and additively transitive accessibility.

Proof

To show that \(\mathfrak {M}^{\prime }\) has upwardly closed accessibility, we perform a simple proof by transfinite induction. For the base case, by construction, \(\{R_{x}^{{\prime }0} : {x} \in [0, \infty ]\}\) is upwardly closed. For the successor case, the inductive hypothesis is that, for fixed \({\alpha } \in {\text {ORD}}, \{R_{x}^{{\prime }{\alpha }} : {x} \in [0, \infty ]\}\) is upwardly closed. Fix w,vM and \({z^{\prime } \geq z}\in [0, \infty ]\). Suppose that \(R_{z}^{{\prime }{\alpha }+1}({w}^{\dagger },{v}^{\dagger })\) holds. Then there is some uM and \(x, y \in [0, \infty ] \) such that \(R_{x}^{{\prime }{\alpha }}({w}^{\dagger },{v}^{\dagger })\) and \(R_{y}^{{\prime }{\alpha }}({u}^{\dagger },{v}^{\dagger })\) hold and x + y = z. Since \(\{R_{x}^{{\prime }{\alpha }} : {x} \in [0, \infty ]\}\) is upwardly closed, there is an \({{x}^{\prime }} \geq {{x,y}^{\prime }}\geq {y} \in [0, \infty ]\) such that \(R_{{x}{^{\prime }}}^{{\prime }{\alpha }}({w}^{\dagger },{u}^{\dagger })\) and \(R_{{y}{^{\prime }}}^{{\prime }{\alpha }}({u}^{\dagger },{v}^{\dagger })\) hold and \(x^{\prime }+y^{\prime }=z^{\prime }\). It follows from the definition of the successor case that \(R_{{z}^{\prime }}^{{\prime }{\alpha }+ 1}({w}^{\dagger },{v}^{\dagger })\) holds. Therefore, \(\{R_{x}^{{\prime }{\alpha }+1} : {x} \in [0, \infty ]\}\) is upwardly closed. For the limit case, the proof is transparent.

To show that \(\mathfrak {M}^{\prime }\) is reflexive, note that \((\mathfrak {M}, a)\) is reflexive. By the definition of planted models, \((\mathfrak {M}^{\dagger },\langle {a}\rangle )\) is thus reflexive in 0-accessibility. It follows that \(\mathfrak {M}^{\prime }\) is reflexive in 0-accessibility. Since \(\mathfrak {M}^{\prime }\) also has upwardly closed accessibility, \(\mathfrak {M}^{\prime }\) is reflexive.

To show that \(\mathfrak {M}^{\prime }\) is symmetric, we again proceed by transfinite induction. For the base case, by construction, \(\{R_{x}^{{\prime }0} : {x} \in [0, \infty ]\}\) is symmetric. For the successor case, the inductive hypothesis is that, for fixed \({\alpha } \in {\text {ORD}}, \{R_{x}^{{\prime }{\alpha }} : {x} \in [0, \infty ]\}\) is symmetric. Fix w,vM and \({z}\in [0, \infty ]\). Suppose that \(R_{z}^{{\prime }{\alpha }+1}({w}^{\dagger },{v}^{\dagger })\) holds. Then there is some uM and \(x, y \in [0, \infty ]\) such that \(R_{x}^{{\prime }{\alpha }}({w}^{\dagger },{u}^{\dagger }) \text {and} R_{y}^{{\prime }{\alpha }}({{u}^{\dagger }},{{v}^{\dagger }})\) hold and x + y = z. Since \(\{R_{x}^{{\prime }{\alpha }} : {x} \in [0, \infty ]\}\) is symmetric, \(R_{x}^{{\prime }{\alpha }}({u}^{\dagger },{w}^{\dagger })\) and \(R_{y}^{{\prime }{\alpha }}({v}^{\dagger },{u}^{\dagger })\) also hold. It follows from the definition of the successor case that \(R_{x+y}^{{\prime }{\alpha }+1}({v}^{\dagger },{w}^{\dagger })\) holds, which means \(R_{z}^{{\prime }{\alpha }+1}({v}^{\dagger },{w}^{\dagger })\) holds. Therefore, \(\{R_{x}^{{\prime }{\alpha }+1} : {x} \in [0, \infty ]\}\) is symmetric. For the limit case, the proof is transparent.

To show that \(\mathfrak {M}^{\prime }\) has additively transitive accessibility, fix w,u,vM and \({x,y} \in [0,\infty ]\). Suppose that \(R_{x}^{\prime }({w}^{\dagger },{u}^{\dagger })\) and \(R_{y}^{\prime }({u}^{\dagger },{v}^{\dagger })\) hold. By the definition of \(R_{x}^{\prime }\), it follows that for some α,β ∈ORD, \(R_{x}^{{\prime }{\alpha }}({w}^{\dagger },{u}^{\dagger })\) and \(R_{y}^{_{{\prime }{\beta }}}({u}^{\dagger },{v}^{\dagger })\) hold. Fix one such α and β, and let γ be the larger of the two. By the definitions of the successor case and the limit case, it follows from reflexivity in 0-accessibility that if \(R_{x}^{{\prime }{\alpha }}({w}^{\dagger },{u}^{\dagger })\) holds, \(R_{x}^{_{{\prime }{\gamma }}}({w}^{\dagger },{u}^{\dagger })\) holds, and if \(R_{y}^{_{{\prime }{\beta }}}({u}^{\dagger },{v}^{\dagger })\) holds, \(R_{y}^{_{{\prime }{\gamma }}}({u}^{\dagger },{v}^{\dagger })\) holds. Thus, by the definition of the successor case, \(R_{x+y}^{_{{\prime }{\gamma }+1}}({w}^{\dagger },{v}^{\dagger })\) holds. From this it follows by the definition of \(R_{x}^{\prime }\) that \(R_{x+y}^{{\prime }}({w}^{\dagger },{v}^{\dagger })\) holds. \(\mathfrak {M}^{\prime }\) therefore has additively transitive accessibility. □

Lemma 2

\(\mathfrak {M}^{\prime }\) is well-founded.

Proof

\((\mathfrak {M}^{\dagger },\langle {a}\rangle )\) is well-founded. Here is the concern for \(\mathfrak {M}^{\prime }\). Planted models ensure well-foundedness by expanding one world’s continuum-many accessibility relations to a single world into one world’s single accessibility relation to continuum-many worlds. But the construction of \(R_{x}^{\prime }\) allows one world to access another through continuum-many paths. We must ensure that these paths don’t generate a set of accessibility relations with no minimum.

Definition 14

For all w,vM, a direct path from w tov is a path from w to v such that:

  1. 1.

    for each element of the path \({u_{1}^{\dagger }}\) before v, the next element \({u}_{2}^{\dagger }\) is such that, for some \({x} \in [0,\infty ]\), either \({R_{x}^{\dagger }}({u_{1}^{\dagger }},{u}_{2}^{\dagger })\) holds or \({R_{x}^{\dagger }}({u}_{2}^{\dagger },{u_{1}^{\dagger }})\) holds; and

  2. 2.

    no element appears in the path more than once.

The planted model \((\mathfrak {M}^{\dagger },\langle {a}\rangle )\) is just the tree-unraveling of \((\mathfrak {M}, a)\), save for self-0-accessibility. Since, by definition, no direct path has a world that appears more than once, self-0-accessibility cannot connect two elements of a direct path. So for the construction of direct paths, \((\mathfrak {M}^{\dagger },\langle {a}\rangle )\) is a rooted tree-like structure. From this it follows that, for all w,vM, there is exactly one direct path from w to v, and the path has a finite number of elements. We will henceforth call it the direct path from w to v.

Definition 15

For all w,vM, the length of the direct path from w tov is the sum of the indices of the accessibility relations connecting each pair of successive elements of the path.

More formally, let \(\langle {u_{1}^{\dagger }},{u}_{2}^{\dagger },\ldots ,{{u}_{n}^{\dagger }}\rangle \) be the direct path from w to v. Define {xi : 1 ≤ in} as follows.

$$ x_{i} := \begin{cases} x \text{such that } R_{x}^{\dagger}(u_{i}^{\dagger},u_{i+1}^{\dagger}) \text{ or } R_{x}^{\dagger}(u_{i+1}^{\dagger},u_{i}^{\dagger}) \text{ holds} & \text{ if } i<n; \\ 0 & \text{ if } i=n. \end{cases} $$

The length of the path is then:

$$ \lambda({w}^{\dagger},{v}^{\dagger}) := \sum\limits_{i=1}^{n} x_{i}. $$

Consider arbitrary worlds w,u,vM, and the direct paths from w to u, from u to v, and from w to v. Construct a possibly indirect path from w to v by concatenating the direct paths from w to u and from u to v, merging the last element of the former with the first element of the latter. The direct path from w to v is a subpath of the possibly indirect path from w to v, for otherwise there would be more than one direct path from w to v. Therefore, for all w,u,vM,

$$ \lambda({w}^{\dagger},{v}^{\dagger})\leq\lambda({w}^{\dagger},{u}^{\dagger})+\lambda({u}^{\dagger},{v}^{\dagger}). $$

To prove that \(\mathfrak {M}^{\prime }\) is well-founded, we will prove that, for all w,vM, λ(w,v) is the minimum x such that \(R_{x}^{\prime }({w}^{\dagger },{v}^{\dagger })\) holds. Since \(\mathfrak {M}^{\prime }\) is reflexive in 0-accessibility and has additively transitive accessibility, for all w,vM, \(R_{{\lambda }({w}^{\dagger },{v}^{\dagger })}^{\prime }{({w}^{\dagger },{v}^{\dagger })}\) holds. We thus need to show that, for all x such that \(R_{x}^{\prime }({w}^{\dagger },{v}^{\dagger })\) holds, xλ(w,v). We show this by performing a proof by transfinite induction.

For the base case, fix w,vM and \({x} \in [0,\infty ]\). Suppose that \(R_{x}^{{\prime }0}({w}^{\dagger },{v}^{\dagger })\) holds. By construction, there is a yx such that either \(R_{y}^{\dagger }({w}^{\dagger },{v}^{\dagger })\) holds or \(R_{y}^{\dagger }({v}^{\dagger },{w}^{\dagger })\) holds. Since \((\mathfrak {M}^{\dagger },\langle {a}\rangle )\) is a rooted tree-like structure, it follows that the direct path between w and v is 〈w,v〉, which has length λ(w,v) = y. Therefore, xλ(w,v).

For the successor case, the inductive hypothesis is that, for fixed α ∈ord: for all w,vM and \({x} \in [0,\infty ]\), if \(R_{x}^{{\prime }{\alpha }}({w}^{\dagger },{v}^{\dagger })\) holds, then xλ(w,v). Now fix w,vM and \({z} \in [0,\infty ]\). Suppose that \(R_{z}^{{\prime }{\alpha }+1}({w}^{\dagger }{v}^{\dagger })\) holds. Then there is some uM and \(x, y \in [0, \infty ]\) such that \(R_{x}^{{\prime }{\alpha }}({w}^{\dagger },{u}^{\dagger })\) and \(R_{y}^{{\prime }{\alpha }}({{u}^{\dagger }},{{v}^{\dagger }})\) hold and x + y = z. By hypothesis, xλ(w,u) and yλ(u,v). Therefore, z = x + yλ(w,u) + λ(u,v) ≥ λ(w,v).

For the limit case, the proof is straightforward. □

Lemma 3

For all w,vM corresponding to w,vM, respectively, and for all \({x} \in [0,\infty ]\), if \(R_{x}^{\prime }\)(w,v) holds, then Rx(w,v) holds.

Proof

We perform a proof by transfinite induction.

For the base case, fix w,vM, corresponding to w,vM, respectively, and fix \({x} \in [0,\infty ]\). Suppose that \(R_{x}^{{\prime }0}({w}^{\dagger },{v}^{\dagger })\) holds. By construction, there is a yx such that either \(R_{y}^{\dagger }({w}^{\dagger },{v}^{\dagger })\) holds or \(R_{y}^{\dagger }({v}^{\dagger },{w}^{\dagger })\) holds. Since \((\mathfrak {M}^{\dagger },\langle {a}\rangle )\) is bisimilar to \((\mathfrak {M}, a)\), either Ry(w,v) or Ry(v,w) holds. Since \(\mathfrak {M}\) is symmetric, if Ry(v,w) holds, then Ry(w,v) holds. Thus, Ry(w,v) holds. Since \(\mathfrak {M}\) has upwardly closed accessibility, Rx(w,v) holds.

For the successor case, the inductive hypothesis is that, for fixed α ∈ord: for all w,vM corresponding to w,vM, respectively, and all \({x} \in [0,\infty ]\), if \(R_{x}^{{\prime }{\alpha }}({w}^{\dagger },{v}^{\dagger })\) holds, then Rx(w,v) holds. Now fix w,vM corresponding to w,vM, respectively, and fix \({z} \in [0,\infty ]\). Suppose that \(R_{z}^{{\prime }{\alpha }+1}({w}^{\dagger },{v}^{\dagger })\) holds. Then there is some uM corresponding to uM and \(x, y \in [0,\infty ]\) such that \(R_{x}^{{\prime }{\alpha }}({w}^{\dagger },{u}^{\dagger })\) and \(R_{y}^{{\prime }{\alpha }}({{u}^{\dagger }},{{v}^{\dagger }})\) hold and x + y = z. By hypothesis, Rx(w,u) and Ry(u,v) hold. Since \(\mathfrak {M}\) has additively transitive accessibility, Rx+y(w,v) holds, which means Rz(w,v) holds.

For the limit case, the proof is straightforward. □

1.3 A.3 Proof of Lemma 5

We will show that upwardly closed accessibility, reflexivity, symmetry, additively transitive accessibility, and D-bounded accessibility are jointly plantable for each \({D} \in [0, \infty ]\). The proof is the same as for Lemma 3 in Appendix A.2, except as follows.

Fix \({D} \in [0, \infty ]\). Stipulate that \((\mathfrak {M}, a)\) also has D-boundedness. Define \(\mathfrak {M}^{\prime }:=({M}^{\dagger },\{R_{x}^{D} : {x} \in [0, \infty ]\}, {V}^{\dagger })\), with:

  • Definition of \(R_{x}^{D}\): For w,uM and \({x,y} \in [0,\infty ]\),

    $$ R_{x}^{D}{} :={} \{({w}^{\dagger}{\kern-.5pt},{{u}^{\dagger}}):{\kern-.5pt} R_{x}^{\prime}({w}^{\dagger},{u}^{\dagger}) \text{holds, or} R_{y}^{\prime}({w}^{\dagger}{\kern-.5pt},{{u}^{\dagger}}) \text{holds for some} y \geq x \geq D\}. $$

In other words, the accessibility relations of \(\mathfrak {M}^{\prime }\) are defined to be the same as in Lemma 3 in Appendix A.2, except that they are D-bounded.

By construction, \(R_{x}^{D} \supseteq R_{x}^{\dagger }\) for all \({x} \in [0,\infty ]\). Also by construction, \(\mathfrak {M}^{\prime }\) will have D-bounded accessibility, preserving upwardly closed accessibility, reflexivity, symmetry, and additive transitivity. Again by construction, since \(R_{x}^{\prime }\) is well-founded, \(R_{x}^{D}\) is well-founded. So the first three conditions of being a plantable property are satisfied.

For the fourth condition, fix w,u,∈ M, corresponding to w,uM, respectively, and fix \({x} \in [0,\infty ]\). Suppose \(R_{x}^{D}({w}^{\dagger },{{u}^{\dagger }})\) holds. Then either \(R_{x}^{\prime }({w}^{\dagger },{u}^{\dagger })\) holds or there is a yxD such that \(R_{y}^{\prime }({w}^{\dagger },{{u}^{\dagger }})\) holds. In case \(R_{x}^{\prime }({w}^{\dagger },{u}^{\dagger })\) holds, then by Lemma 15, Rx(w,u) holds. In case there is a yxD such that \(R_{y}^{\prime }({w}^{\dagger },{{u}^{\dagger }})\) holds, then by Lemma 15, Ry(w,u). Since \(\mathfrak {M}\) has upwardly closed accessibility, \(R_{\infty }(w,u)\) holds. Since \(\mathfrak {M}\) has D-bounded accessibility, RD(w,u) holds. Since \(\mathfrak {M}\) has upwardly closed accessibility, Rx(w,u) holds.

1.4 A.4 Proof of Lemma 9

We will show that upwardly closed accessibility, reflexivity, symmetry, additively transitive accessibility, D-bounded accessibility, and coreflexive 0-accessibility are jointly plantable for each \({D} \in [0, \infty ]\). The proof is the same as for Lemma 5 in Appendix A.3, except as follows.

Stipulate that \((\mathfrak {M}, a)\) is coreflexive in 0-accessibility. By the definition of planted models, \((\mathfrak {M}^{\dagger },\langle {a}\rangle )\) is also coreflexive in 0-accessibility. From this it follows that \(\mathfrak {M}^{\prime }\) is coreflexive in 0-accessibility.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Camrud, C., Dosanjh, R. Continuous Accessibility Modal Logics. J Philos Logic 52, 221–266 (2023). https://doi.org/10.1007/s10992-022-09671-7

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10992-022-09671-7

Keywords

Navigation