When SirAugustus Wollaston Franks, a Keeper at the British Museum,produced a catalogue of British decorated metalwork in 1863, hechose the term ‘Late Keltic’ to describe items of Iron Age date. Thephrase was used throughout the first edition of the British MuseumGuide to the Antiquities of the Early Iron Age (1905), but by thetime of the second edition (1925) the characterization was droppedbecause, as the preface cautiously noted, ‘There is some uncertaintyas to the existence or date of an earlier Keltic civilization in theseislands.’

This early eighteenth-century hypothesis has driven the debate eversince. Its longevity is truly remarkable and due in no small part tothe fact that linguists and archaeologists were for a long timeprepared to accept each other’s interpretations, each gainingreassurance by building upon the other in an uncritical circle ofmutually supporting assertion lacking firm foundation.