Thus grammar, the study of the Latin language, represented the first stage of knowledge for mediaeval men. This was to remain true through all the mediaeval centuries; since all youths who became scholars had to learn the language before they could study what was contained in it alone. One may also say, and yet not speak fantastically, that grammar, the study of the correct use of the language itself, corresponded spiritually with the main intellectual labour of the Carolingian period. Alcuin’s attention is commonly fixed upon the significance of language, Latin of course. And the labours of his pupil Rabanus, and the latter’s pupil Walafrid, are as it were devoted to the grammar of learning. That is to say, they read and endeavour to understand the works of the Fathers; they compare and collate, and make volumes of extracts, which they arrange for the most part as Scripture commentaries; commentaries, that is, upon the significance of the canonical writings which were the substance of all wisdom, but needed much explication. Such works were the very grammar of knowledge, being devoted to the exposition of the meaning of the Scriptures and the vast burden of patristic thought. A like purpose was evinced in the efforts of the great emperor himself to re-establish schools of grammar, in order that the Scriptures might be more correctly understood, and the expositions of the holy Fathers. In fine, just as knowledge of the Latin tongue was the end and aim of grammar, so a correct understanding of what was contained in Latin books was the aim of the intellectual labours of this period. It all represented the first stage in the mediaeval acquisition of knowledge, or in the presentation or expression of the same; and thus the first stage in the mediaeval endeavour to realize the human impulse to know. The next course of the Trivium was logic; and likewise its study will represent truly the second stage in the mediaeval realization of the human impulse to know, to wit, the second stage in the appropriation and expression of the knowledge transmitted from the past. We have spoken at some length of the logical studies of Gerbert, and his endeavours to adjust his thinking and classify the branches of knowledge by means of formal logic.[454] Those discussions of his which seem somewhat puerile to us, were essential to his endeavours to formulate what he had learned, and present it as rational and ordered knowledge. Logic is properly the stage succeeding grammar in the formulation of rational knowledge. At least it was for men of Gerbert’s time, and the following centuries. Rightly enough they looked on logic as a _scientia sermotionalis_, which on one side touched sheer linguistics, and on the other, had for its field the further processes of reason. Thus Hugo of St. Victor, Abaelard’s very great contemporary, says: