Random Coincidences Or: the return of the Celtic to Iron Age Britain

is normally a random coincidence, and need not concern us a lot. If, however, similarities mount up, and we find several, perhaps even many, similarities between two separate things, matters start to change. We will start to wonder whether the two things might not actually somehow be related. Such a relationship might be functional, as in the case of axes or hammers, which are similar all over the world because they need to be in order to work properly. Or it might be structural, as in the case of crystals, which will always grow to very similar shapes because of the properties of the substance they are made of. Or, it might be genetic, like in the case of art styles, each of which produces similar things because the artists use certain techniques or motives that are inspired by similar earlier examples of that same style. In fact, this seems to be a general property of the universe we live in: if two separate things share many similarities, there usually is some kind of reason for why they do. And thus, tending to generalise this empirical observation as we do, if we find two separate things sharing many, or regular, or particularly striking similarities, we suspect that these are not the result of mere random coincidence, but due to some reason. In his article on Iron Age southern Britain in Different Iron Ages, J.D. Hill (1995a, 45) asked the question ‘What if the common assumption of a generalised ‘Celtic’ form of Iron Age social organisation or religion is a myth?’. From there he proceeded to deconstruct the model proposed by Barry Cunliffe (1983; 1984a–c) for Early and Middle Pre-Roman Iron Age societies in Wessex. He constructs an alternative model (Hill 1995a, 49–54) which, in his opinion, seems to be fundamentally at odds with what can be learned from ‘Celtic’ sources. He points to the essentially ideological orientation of Iron Age houses and enclosed settlements, with entrances almost exclusively facing eastwards, the essential boundedness of individual homesteads, and compares this to Marx’s Germanic Mode of Production (1964; Bonte 1977). He thus argues for an essentially tribal, egalitarian, and boundary-obsessed (Hill 1995a, 51) society in Early and Middle Pre-