Having reached its peak of descriptive adequacy virtually at birth, Arabic grammar often seems little more than an endless discussion and restatement of the same immutable facts. This superficially stagnant aspect of Arabic grammar has provoked the no less superficial criticism that it is “a somewhat dismal science”, while even the medieval Arabs complained that too much study of grammar could lead to madness. This brief historical outline will therefore emphasize the variety and flexibility of grammar as it responded to social pressures and the influence of other disciplines, in order to show that it is not the monolithic and fruitlessly abstract science it has sometimes been made to appear. For convenience, the terms naḣw and naḣwī are arbitrarily rendered “grammar” and “grammarian” throughout, but it cannot be stated too firmly that the equivalence is only partial, for neither naḣw nor “grammar” have remained stable in meaning over the centuries.

To keep the topic within bounds, two limitations are imposed: as far as possible only extant works are considered, and almost exclusively from the domain of syntax. The restriction to extant works ensures greater objectivity than is achieved by relying on the copious biographical literature, which provides abundant anecdotal material but seldom anything of technical value. Moreover, even the specialized grammatical biographies include many “grammarians” who have no real claim to the title -polymaths, amateur philologists, dilettanti and others who, in the absence of any surviving texts to judge them by, are no more than names.