Minimally congruential contexts: observations and questions on embedding E in K (Q691133)

The author is concerned with faithful embeddings of modal logical systems in other modal logical systems, particularly, with possible embeddings of the smallest congruential modal logic E in the smallest modal logic K. In general, there is a faithful embedding of a modal logical system S1 in another logical system S2 if and only if there is a translation function T mapping formulas of S1 into formulas of S2 in such a way that the translation of a formula A of S1 by T is a theorem of S2 if and only if A itself is a theorem of S1. Associated to the values of any of such translation function T, there is what the author calls ``the context'', which is just the general pattern that T makes use of for translating into S2 formulas of S1 preceded by the necessity operator. Different sorts of such translational patterns are possible, but clearly not all of them would form the basis for yielding a faithful embedding of one logic in another. Now, the author points out that an embedding of E in K has already been constructed. He briefly discusses the context associated to the translation function of this embedding and asks himself whether an alternative embedding of E in K can be found based on a translation function with an associated simpler context. He employs two criteria of simplicity, viz.: the maximal depth of embedding of the necessity operator in the formula A (which he calls the modal degree of A) and the number of occurrences of the necessity operator in the formula A (to which he refers as the modal complexity of A). He proves several negative theorems regarding certain contexts as possible candidates for translation functions that might embed E in K. Also, he proves other results regarding certain properties that a context might satisfy relative to a modal logical system. These properties include minimal monotony, minimal congruentiality and minimal normality. Those results are for the necessity operator taken as a context, in connection with the modal logical systems K, E, and EM. Other possible normal modal logical systems satisfying certain conditions are taken into account. The results in question are instrumental for the author's own project of simplifying the current embedding of E in K, since he thinks that certain particular context, explicitly stated in the paper, might yield the simplification he is searching for. According to him, his initial strategy for achieving this simplification was to show that such a context was minimally congruential in K and then conclude that certain fragment of K was precisely E. However, the latter requires that one of the results (stated in the second part of the paper) would not apply to any extension of E. But this, as the author himself notes, he has not proved. Rather, such a proof is left by him as an open problem. Thus, the main question with which the paper began is not solved, but rather a possible way to be explored for achieving a solution is offered. In addition to the aforementioned open problem, two other open problems are stated. One concerns the question of whether the context in question can embed E in K independently of a possible negative answer to the first open problem, and the other concerns the possibility of the existence of contexts minimally congruential in K.