Such speculations serve as a reminder that the volume's sections on England can also be of relevance to historians interested in medieval Wales, by providing comparative material and highlighting problems which need to be explored. If the specific discussions of Welsh agrarian society and economy are ultimately disappointing, this is because they are excessively cautious and narrow in their approach, thereby failing to address a number of fundamental questions. As a consequence, the reader gains little sense of the scale of changes in agrarian exploitation, rural settlement and social relations during the three centuries under consideration. Anyone seeking to attain a more rounded view of these developments will, however, be grateful for the data presented in this volume and be sure to come away from it with a keen appreciation of the need to recognise and explain a great diversity of local conditions. HUW PRYCE Bangor HISTORY AND LITERATURE. By Sir Keith Thomas. The Ernest Hughes Memorial Lecture 1988. University College of Swansea, 1989. Pp. 31. £ 1.50. In common with all those who had the good fortune to hear this sparkling lecture being delivered, this reviewer is delighted that it has now been made available in print. Not only because it will reach a larger audience, not only because it will give those who read it more time to ponder on the crucial issues it raises, but also because they will have available to them the range of references it provides-references which are extraordinarily wide, varied and up-to-date within so restricted a compass. The lecture was particularly appropriate as one in memory of Ernest Hughes, who liked to recall that, at the outset in Swansea, he had constituted in his own person the whole of the Faculty of Arts, and who, to the end of his life, retained his interest and delight in literature and drama as well as history. He would keenly have relished the author's robust but thoughtful and elegant defence of what history and literature have to offer one another. In a brief introduction, Sir Keith acknowledges the hostile contemporary currents against which he is swimming and illustrates-and deplores-the widening gap between those historians, who think that literature is at best peripheral and at worst positively misleading as an historical source, and the literary critics, who have become unrepentantly anti-historical in their perception of their studies. He then embarks on his central thesis, designed to counter the unnecessarily impoverishing influence both schools have had on the study of history and literature. His main thrust is understandably directed at an assessment of the value of literary sources to the historian. Although, as he himself recognized, there were other arguments that might have been adduced, he wisely concentrated on four and argued them with compelling clarity, subtlety and wit. First, literature is in essence no different from other written evidence employed by the historian and can be tested by his familiar methods of criticism. Second, literature is a form of social and historical action and so a legitimate