Abstract
In this chapter we ask to what extent our results transfer to logics weaker than normal. We investigate this question only for our main results and in cases where a transfer is easily possible, and leave more comprehensive investigations to further studies. The weakest kind of a modal logic in which the modal operator represents some operation which applies to a proposition is classical modal logic: the only genuin modal principle which is required in the minimal system of propositional classical logic, C a 0, is the equivalence rule (ER): A↔B/□A↔□B (cf. Segerberg 1971, ch. 1; Chellas 1980, part III). Investigations in classical modal predicate logics are rare (cf. Gabbay 1976). The minimal classical predicate logic of type 1, C a 1, is obtained from C a 0 by adding axioms and rules of nonmodal predicate logic (cf. ch. 2.4.1: ∀13 and ∀R). The Barcan formula does not belong to the minimal classical 1-logic because the assumption of constant domain and rigid designators in neighbourhood semantics (see below) does not imply (aBF) nor its converse. Three sucessive strengthenings of classical modal logics are of particular importance: the monotonic, the regular, and finally the normal ones. Here are the basic definitions.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Notes
A comprehensive study on modal predicate logics based on regular Kripke frames is Bowen (1979, ch. 1–4). See also Gabbay (1976, ch. 2), and Segerberg (1971, pp. 23fí).
According to Zermelo-Fraenkel set theory it is impossible that S,W E W, because this would imply - after replacing S,W by {{S}, {S,W}} according to the Kuratowski definition - the circular E -chain W E {S,W} E {ISM {S,W}} e W, which is excluded by the axiom of foundation. Thus, the definition guarantees that WnW’ = 0.
The reason for this is that GH is formulated in a syntactic way. If GH would be formulated in a semantic way (with L-consequence and L-validity [t] instead of L-deducibility and L-theoremhood [~]), this restriction could be dropped; but then another restriction would be necessary, namely compactness of L, for otherwise the semantic counterpart of proposition 8 could not be proved.
In Schurz (1991a, p. 74) I have formulated the main semantic theorem about GH in terms of characterizability, as in corollary 3, instead of in terms of closure of F(L) under Sep, as in the theorem 2 stated above. Now I realize that closure of F(L) under Sep is the more fundamental property, from which characterizability of L follows. On p. 75 of (1991), last paragraph, I have mistakenly claimed that the direction = of the theorem 2 given there — which corresponds to corollary 3 above — holds also if Sep(F(L)) is replaced by the broader class of is-ought separated frames for L — these are a.d.-frames which satisfy the conditions (a) and (b) stated below definition 14, but not necessarily condition (c). The mistake was located in the proof of theorem 2 G given on p. 74 of this paper: I forgot to show that the model M can be chosen in a way that the world a belongs to the world set W; but this can be shown only by assuming that M is an is-ought separated double.
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 1997 Springer Science+Business Media Dordrecht
About this chapter
Cite this chapter
Schurz, G. (1997). Generalizations. In: The Is-Ought Problem. Trends in Logic, vol 1. Springer, Dordrecht. https://doi.org/10.1007/978-94-017-3375-5_8
Download citation
DOI: https://doi.org/10.1007/978-94-017-3375-5_8
Publisher Name: Springer, Dordrecht
Print ISBN: 978-90-481-4795-3
Online ISBN: 978-94-017-3375-5
eBook Packages: Springer Book Archive