Today, William of Malmesbury is considered not only the most erudite and well-read historian of the Anglo-Norman age, but also one of the best travelled. Rodney Thomson, in his invaluable commentary on the Gesta Pontificum Anglorum, gives a list of some thirty-six places in England, Wales and Normandy definitely or presumably visited by the Malmesbury historian. There can be no doubt that much of the information in some of William's works, including the Gesta Pontificum, Gesta Regum, De Antiquitate Glastonie Ecclesie and Vita Wulfstani, derives from first-hand experience. Antonia Gransden considered William's work the prime example of ‘realistic observations’ and thought it ‘likely that most (perhaps all) of William's descriptions were based on personal observation, not hearsay’. Similarly, S. B. Wright claimed that ‘personal travel experience is omnipresent’ in the Gesta Pontificum and stressed ‘William's emphasis on eyewitness experience’. However, it is only for a handful of these places that William presents us with evidence that he visited them in person. In fact, in the Gesta Pontificum only once does William state that he personally saw the site he describes: the church at Sherborne, thought to have been built by Aldhelm. Of course, this limited direct testimony does not mean that we should take a hypercritical point of view and turn William into a sort of armchair scholar. For many places, such as Canterbury, Glastonbury and Worcester, the sheer volume and nature of his information make it almost certain that William had been there in person. But in other cases, close scrutiny of his text demonstrates William's heavy dependence on written sources.
The greater weight of written material can be traced, for example, in his account of regions of northern England. William Kynan-Wilson recently demonstrated that the descriptions of churches in Ripon and Hexham in the fourth book of the Gesta Pontificum (100. 22–3 and 117. 1) are almost entirely based on a single literary source, the Vita Sancti Wilfridi by Stephen of Ripon. The famous description of Roman remains in Carlisle – including an inscription of ‘Marii uictoriae’ (99. 3–4) on the front of a triclinium, although it has no parallels in known written sources – also exhibits, according to Kynan-Wilson, a large number of bookish topoi and cliches, for example, from the famous poem on Rome by Hildebert of Lavardin (c. 1055–1133).