NOT ONLY A great historian, William was also, as we have seen, an omnivorous reader and an indefatigable collector of books for his abbey library. From his own statements and from textual evidence we know that he scoured the country between c.1115 and c.1140 looking for ancient volumes and rare works. The details of this large undertaking are mysterious and fascinating, provoking many questions. Did he purchase or otherwise obtain old or ‘second-hand’ copies? Did he have new copies made where the exemplar was held, or did he borrow the exemplar and have the copying done at Malmesbury? If the copying was done locally, how was it organized? What part did he himself play? Did he form a regular team of scribes to assist him? How did he set out his books? The mechanics of William's acquisition programme, seen in terms of these and related questions, are my concern in this chapter.

Let me first outline what has already been done in this direction. By the end of the nineteenth century William's own hand had been recognized in more than one book. The starting point for these identifications was (as it still must be) the autograph copy of the Gesta Pontificum, Oxford, Magdalen Coll. MS lat. 172. The work of Hamilton and Stubbs was corrected and amplified in 1938 by Neil Ker.Since then William's hand has been found in other manuscripts which are listed in the 1964 edition of Dr Ker's Medieval Libraries of Great Britain. Thus at the present time William's hand and its main characteristics are well known, although there is room for detailed work on his scribal habits and their changes during the course of his life. For instance, although William always wrote what may be described as a ‘scholar's’ hand, he had both formal and informal versions of it (Pls 1–2), using the latter alongside the former, for corrections and annotations. But he also wrote complete texts in the less formal style. To what extent, then, was his formal hand an improvement developed in the course of long practice? I shall touch upon this question below, although the scanty evidence makes an assured answer impossible.