We know very little about the details of this remarkable revolution, because Scotland had no voice in the 12th century, none of her chroniclers being earlier than the end of the 14th century. As regards the subject which concerns this book, the building of castles, there are only one or two passages which lift the veil. A contemporary English chronicler, Ailred of Rievaulx, in his panegyric of David I., says that David decorated Scotland with castles and cities.[1005] In like manner Benedict of Peterborough tells us that when William the Lion was captured by Henry II.'s forces in 1174, the men of Galloway took the opportunity to destroy all the castles which the king had built in their country, expelling his seneschals and guards, and killing all the English and French whom they could catch.[1006] Fordun casually mentions the building of two castles in Ross by William the Lion; and once he gives us an anecdote which is a chance revelation of what must have been going on everywhere. A certain English knight, Robert, son of Godwin, whose Norman name shows that he was one of the Normanised English, tarried with the king's leave on an estate which King Edgar had given him in Lothian, _and while he was seeking to build a castle there_, he was attacked by the men of Bishop Ranulf of Durham, who objected to a castle being built so near the English frontier.[1007] But even if historians had been entirely silent about the building of castles in Scotland, we should have been certain that it must have happened, as an inevitable part of the Norman settlement. Robertson remarks that the Scots in the time of David I. were still a pastoral and in some respects a migratory people, their magnates not residing like great feudal nobles in their own castles, but moving about from place to place, and quartering themselves upon the dependent population. There is in fact no reason for supposing that the Keltic chiefs of Scotland built castles, any more than those of Wales or Ireland.[1008] But the feudal system must very soon have covered Scotland with castles. The absence of any stone castles of Norman type has puzzled Scottish historians, whose ideas of castles were associated with buildings in stone.[1009] In 1898 Dr Christison published his valuable researches into the _Early Fortifications of Scotland_, in which for the first time an estimate was attempted of the distribution of Scottish _motes_,[1010] and their Norman origin almost, if not quite, suspected. His book was quickly followed by Mr George Neilson's noteworthy paper on the "Motes in Norman Scotland,"[1011] in which he showed that the wooden castle is the key which unlocks the historians' puzzle, and that the motes of Scotland are nothing but the evidence of the Norman feudal settlement.