In almost every respect Weeting Castle represents a challenge. To begin with, the limited remnants of the building do not at first sight suggest that a very full mental reconstruction of the medieval building is a realistic prospect (Fig. 1). However, as this paper is intended to demonstrate, enough survives to allow not just the general form, but also a good deal of the detail to be envisaged with some confidence. The designation ‘castle’ is another hurdle, encouraging a misleading expectation that Weeting should be understood in the context of fortified residences. How it should be categorized is far from straightforward. The notice which greets visitors to the site refers to it as ‘an early medieval manor house’. As we shall see, the building has many of the characteristics of the later medieval double-ended hall house, but there are two problems about regarding it as such. The first is that we categorize it in this way only with the benefit of historical hindsight: it is not likely to have been viewed by its builders as a contribution to an ‘evolutionary’ process. The second is a danger either that Weeting comes to stand as typical of the twelfth-century development of this type, or conversely to be regarded as a ‘freak’, a building which accidentally resembles later hall houses but really has nothing to do with them. The state of our knowledge is such that we cannot say, and indeed may never be able to say how unusual Weeting was. However, that need not stop our trying to see the building in its contemporary setting both as regards understanding where the ideas might have come from and the parallels we adduce in reconstructing it.