Page:English Historical Review Volume 35.djvu/331

1920 [he does not in fact say 'to Paris'] I found Gilbert again'. Thirdly, John's description of the teaching of Bernard of Chartres is so intimate that it can only be based on personal knowledge. Before examining the evidence about William of Conches and Gilbert of La Porrée it is worth while to note that Schaarschmidt introduces an unnecessary complication into the matter by suggesting that John spent part of these three years at Provins and Rheims. He forgets that John lived twelve years in France, and we are now concerned only with the three years beginning with his third year. His visits to these places may have been made in any of the last seven years of his stay in France, which he only just glances at. He introduces this precious bit of autobiography as an episode, to illustrate the way in which philosophical studies were pursued; and it is only as an afterthought that he speaks very briefly of his later years of study.

Schaarschmidt's opinion that the triennium was mainly passed at Chartres has been generally accepted. But a good deal of fresh material has accumulated since he wrote, nearly sixty years ago; and one inconspicuous source of evidence which was made public a few years earlier escaped his attention. This is a letter-book which throws light on the schools of Chartres in the first quarter of the twelfth century, and introduces us to Bernard and Gilbert and perhaps also to William of Conches.

A manuscript formerly belonging to the chapter of Chartres and now preserved in the public library of that town (no. 1029) contains at the end of a collection of the letters of Bishop Ivo, the famous canonist, a number of letters and epistolary specimens or formulae occupying the last sixteen leaves of the volume. Thirty-six of these letters were printed by Lucien Merlet in the Bibliothèque de l'École des Chartes for 1855.4th ser. i. The manuscript was then designated 2nd ser. 19. The manuscript was at that date assigned to about 1150, but the last catalogue places it at the end of the twelfth century. Not much later it received additions of papal bulls and letters of bishops derived from sources which point to a Venetian origin. But the manuscript itself is indubitably a Chartres book, and the private letters at the end show that it was in part compiled from the correspondence of a particular family closely connected with some member of the capitular body. That this family was that of Arnold, who was dean from at latest 1087 to 1119, is pretty clear; for the