A tale of four grammars (Q1850950): Difference between revisions

Categorical grammar dating back to [\textit{K. Ajdukiewicz}, Stud. Philos. 1, 1--27 (1935; Zbl 0015.33702); \textit{Y. Bar-Hillel}, in: Language and Information 61--74 (1964); Language 29, 47--58 (1953; Zbl 0156.25402)] assigns a type to each word of the language to each word of the language to be studied, say English, so that the sentencehood of a string of words can be determined by a calculation on their types, hopefully without recourse to further language specific grammatical rules, well-formedness of sentences being determined by grammatical rules not in the dictionary or after inserting additional words with appropriate types (e.g., whom). This paper considers relations existing between four dedective systems of categorical grammar as follows: \begin{itemize} \item[1.] the syntactic calculus [\textit{J. Lambek}, Am. Math. Mon. 65, 154--170 (1958; Zbl 0080.00702)]; \item[2.] classical bilinear logic [\textit{V. M. Abrusci}, J. Symb. Log. 56, No. 4, 1403--1451 (1991; Zbl 0746.03044); \textit{J. Lambek}, From categorial grammar to bilinear logic. Report from the Department of Mathematics and Statistics 91, No. 14. McGill University (1991)]; \item[3.] compact bilinear logic [\textit{J. Lambek}, Lect. Notes Comput. Sci. 1582, 1--27 (1999; Zbl 0934.03043); Grammars 4, No. 1, 21--39 (2001; Zbl 1007.03031)]; \item[4.] \textit{H. B. Curry}'s semantics calculus (positive intuitionistic propositional logic) [``Some logical aspects of grammatical structure'', in: Structure of language and its mathematical aspects. Providence, RI: American Mathematical Society. 133--173 (1961)]. \end{itemize} They correspond algebraically to \begin{itemize} \item[1.] a residuated monoidal category; \item[2.] a non-commutative *-autonomous category; \item[3.] a monoidal category in which each object, thought of as a bimodule, is of both s left and right adjoint; \item[4.] a cartesian closed category. \end{itemize}