The editors say that this book is not meant to provide a concluding summary of the project they have been leading for several years, but it does contain useful syntheses of subject areas covered in their research. Cleary and Gibson give a good overview of ‘Connectivity in Atlantic Europe during the Bronze Age (2800–800 bc)’ starting with the origin and dissemination of the Beaker package. They point out regional variations and differences in, for instance, the reuse of megalithic tombs, though for the later phases of the Bronze Ages the links are to be seen more in the metalwork rather than in, for example, the patterns of land use.

The chapter by Bray on the chemistry of the metalwork warns about how ‘heroic’ pioneer studies such as those of the 1950s, while breaking new ground, have had to be modified as it became clear the original goals and methodologies were not giving the hoped-for results; that is, defining the sources of the metals either by trying to characterise the trace elements at the mines themselves (an approach associated with Pittioni in Austria) or by using top-down statistical analysis to define groups (the approach of the Stuttgart group of Junghans, Sangmeister et al). Though the impact of specific sources, of their rise and decline, can to a certain extent be documented, such as the mines at Ross Island and at Great Orme or the arsenic rich copper sources in Iberia, the stories are not simple. Bray underlines the need for metallurgists and archaeologists to work together in their thinking, for instance understanding the histories not only of individual objects but also the metal they are made of, with the evidence of the working and reworking that slowly alters the chemistry of the metal even when not mixed with scrap metal, something visible in the fall-off patterns as one moves away from the mines.

For the studies of DNA dealt with by an international team (Marina Silva et al), we are perhaps still in the ‘heroic’ stage where new data on ancient DNA have just started to flood the literature. The experiences of the metallurgists might prove a useful warning, as we are still at an early stage of analysis; research projects and interpretations have tended to concentrate on ‘grand narratives’ derived from linguistics, such as the origin and spread of Indo-European languages, using the models that dominated the literature up to the 1960s based on Culture Groups and migration. The models that I and others were developing then mainly talked of small groups interacting with one another, with ideas, goods and individuals moving in all directions continuously rather than the unidirectional waves implied by mass migrations. This pattern was then criss-crossed by networks that mark favoured lines of movements, such as along sea coasts like the Atlantic, along rivers or high ground routes, and can be for trade, religious purposes (eg pilgrimages) or social connections. Personally I prefer the bottom-up models where aDNA studies are looking at populations (eg cemeteries), rather than the grand narrative approach.

So, what does this all have to do with the ‘origin’ of the Celts and the Celtic languages? Possibly very little. The project has tended to ignore critics, and one must ask whether what we are seeing along the Atlantic is any different from what we might see along other favoured axes, say along the Rhône, the Danube or the Rhine. There are too many unquestioned assumptions like the need for a lingua franca along the Atlantic – there was much more trade around the Mediterranean, but common languages did not appear until major powers emerged like the Phoenicians/Carthaginians, the Greeks and especially the Romans, and non-Indo-European languages like Etruscan and Iberian survived for a long time while their speakers were going through similar processes of culture change as their neighbours who spoke Indo-European languages. What the editors refer to as the ‘traditional’ view of the origin and spread of the Celts from southern Germany signified by the ‘expansion’ of the so-called Late Hallstatt and La Tène ‘Culture Groups’ blatantly does not work either chronologically or geographically; in its present form it only goes back to the 1970s and 1980s with the maps by Pierre-Marie Duval and especially the Megaws, though Celtic from the east has a long history from the first biblical and linguistic models of the spread of peoples and languages. But the key piece of evidence, that ‘Tartessian’ is a Celtic language, despite what the editors say, has largely been rejected by linguists, most recently discussed by Sims-Williams (Reference Sims-Williams2020). While this project has produced much useful new data and is important for trying to bring various disciplines together and querying the ‘La Tène Celts’ (though that has long been under discussion), it has had little or no impact on central and eastern European Celtic studies.