In this paper, we prove some embedding theorems for LTL (linear-time temporal logic) and its variants: viz. some generalisations, extensions and fragments of LTL. Using these embedding theorems, we give uniform proofs of the completeness, cut-elimination and/or decidability theorems for LTL and its variants. The proposed embedding theorems clarify the relationships between some LTL-variations (for example, LTL, a dynamic topological logic, a fixpoint logic, a spatial logic, Prior's logic, Davies' logic and an NP-complete LTL) and some traditional logics (for example, classical logic, intuitionistic logic and infinitary logic).We will introduce a Gentzen-type sequent calculus GLT ω for a generalised first-order LTL (GLTL) in Section 4, and we will prove cut-elimination and completeness theorems for GLT ω using some embedding theorems. GLT ω has modal operatorsand it includes (first-order) LT ω as a special case.In the following, we will explain that the proposed new modal operators ♥ i , ♥ G and ♥ F can be regarded as generalisations of X, G and F. These operators are intended to characterise the axiom schemewhere K * is the set of all words of finite length of the alphabetWe now suppose that for any formula α, we have f α is a mapping on the set of formulas such that♥ G α then becomes a fixpoint of f α . This axiom scheme just corresponds to the so-called iterative interpretation of common knowledge. On the other hand, if we takewe can understand ♥ 1 and ♥ G , respectively, as the temporal operators X and G in LTL. The corresponding axiom scheme for the singleton case represents the LTL-axiom schemeSo the operator ♥ G can be regarded as a natural generalisation of G. Similarly, ♥ F can be regarded as a generalisation of F.
Infinitary extensions of LTLIn Section 5, we will introduce two Gentzen-type sequent calcului L ω and L − ω , which are infinitary extensions of LTL:-L ω includes a variant of dynamic topological logics; -L − ω , which is a subsystem of L ω , is an integration of both LT ω and LK ω . We will then prove cut-elimination theorems for L ω and L − ω and a completeness theorem for L − ω . There is no completeness theorem for L ω since a subsystem S4 ω of L ω is known to be Kripke-incomplete. S4 ω is an extension of both IL and the normal modal logic S4, and has been used as a base logic for formalising game theory (Kaneko and Nagashima 1997).