The Iron Age complex at Paule-Saint-Symphorien in Brittany is well known to prehistorians for its exceptional structures and its anthropomorphic stone figures, one of which holds a lyre and wears a torc. A 1988 rescue excavation led to a research programme from 1989–2013, in which Yves Menez and colleagues investigated approximately 5ha of the hilltop complex. Two Bronze Age tumuli, an early medieval residence, a motte-and-bailey and a chapel were also excavated—the last three not reported here. This clustering of ritual foci and centres of power is, in itself, of no small interest.

This meticulously researched monograph will ensure Paule's place among the key reference sites for Iron Age Europe. French and English summaries for each chapter, along with bilingual figure captions, open the volume to an international audience. Unlike many excavation reports, it sets out to be accessible, rejecting jargon and including devices such as human scales to help non-archaeologists. Although the 30 × 30cm format is unwieldy, it does allow plans, sections, photographs and finds to be viewed together. A plan showing how features were excavated (many were machine-dug) helps assessment of finds retrieval across the site.

Geology, limited stratigraphy and frequent reconstruction made for a technically challenging excavation (Chapter 1). Unpicking the complex archaeology was difficult, and aspects of the sequence are not beyond questioning. Chapters 2–7 present the six main phases and selected finds (mostly pottery), the illustrations grouped at the end of each chapter. From the outset, Paule was no ordinary site, and its initial layout set the scene for what followed. The centrepiece of the sixth-century BC site was a hall-like building, with cellar, cistern and souterrains at the back of an externally banked enclosure; around the edges were further ancillary structures (Phase 1). Outside were an annexe and two burial grounds with stone stelae. Around 300 BC, the site was rebuilt to a more rectilinear layout. An inner enclosure with rampart, two corner towers and four entrances housed two large buildings with souterrains; ancillary structures stood in a separate forecourt (Phase 2). Later, the defences of the residential area were doubled, and access restricted (Phase 3).

After a fire around 175 BC, the residence was reconstructed (Phase 4). Two quadrangular enclosures were created nearby, one accommodating a vast building compared by Menez to Roman horrea or medieval tithe barns. A three-sided structure outside the residence is interpreted as a stable but recalls courtyard structures seen as having public or religious functions at other centres. An outer defence was built, but much of its circuit is conjectural. In the later second century BC, the residence acquired a massive rampart with timber walkway (Phase 5). A forge, a well with a unique timber winding mechanism, and an unroofed ‘arena’ were later created inside, whilst new buildings outside suggest an enlarged population. In the Augustan period (c. 20 BC–AD 15), the site was dismantled, and the Bronze Age tumuli were transformed into a shrine or place of memory that endured well into the Roman period (Phase 6).

The review of the finds and structures (Chapters 8–9) reinforces the exceptional nature of the site. Italian wine was reaching Paule by the early second century BC; a gold pendant (or its platinoid gold) that originated in the Eastern Mediterranean was found in a first-century BC context. Crucibles, portable kilns, anvils and residues attest to the working of iron, copper-alloys and local alluvial gold; indeed, local minerals were probably a factor in the wealth of the site. An updated discussion of the four sculptures of ancestors or deities (deposited in Phase 5) advances the idea that their head bands might be schematised diadems or symbols of office.

For this reviewer, overviews of the pottery and absolute dating would have been useful. Whilst argued in detail, the length of Phase 1 sits rather oddly with the turnover of Phases 2–4. There was no radiocarbon dating, but Bayesian modelling of other sites suggests that ceramic dating often overplays the extent of continuities. For Phase 5 (dated 150–20 BC), the absence of chronological markers specific to La Tène D2 (for Dressel 1B amphorae, attributed to supply factors) could suggest that the occupation was waning well before the Augustan reorganisation of Gaul into civitates.

Chapters 10–12 assess the socio-political standing of Paule, drawing on comparisons from Gaul and beyond. Menez identifies a handful of other sites in Brittany whose morphologies suggest similar social trajectories. Comparing Paule to presumed high-ranking Iron Age settlements further afield, Menez argues that the occupants were on the top rung of the elite ladder. Whether the expanded Phase 4–5 settlement can be considered a counterpart of the fortified agglomerations (oppida) that developed across Late Iron Age Europe is more delicate, given its uncertain extent and density of occupation. There are analogies with bipartite ‘aristocratic farmsteads’ such as Batilly-en-Gâtinais, whilst the Trégueux complex and courtyard building in northern Brittany provide a regional parallel. A consistent message from these sites is that leading households were gaining dependents or retinues at this time.

Some readers will be uncomfortable with the use of medieval terminology such as ‘maison forte’ or ‘bourg castral’ to convey the rising wealth and power of Paule's occupants, even if their use brings the past to life more vividly. Scholars who argue against taking an overfamiliar view of the Iron Age may feel that the interpretation plays down some of Paule's more unusual features, from the early external banks to the late ‘barn’. Particularly in its developed stages, the complex could well have played a significant public or ritual role for the local area, to which the residential aspect was subordinate; one might even go so far as to suggest that Paule and Trégeux were forerunners of the large rural sanctuaries of Roman Gaul. Both lie near presumed civitas boundaries. The idea that many centres with assembly functions began at neutral meeting places could be another way to look at the Breton sites. Conversely, Paule may offer a model for the rise of British polyfocal complexes, such as St Albans, as Menez notes. Ritual and secular are inextricably entangled in pre-industrial societies—the perennial problem for archaeologists.

Enhanced by superb colour reconstruction drawings, the report largely succeeds in treading the difficult line between presenting the primary data and being inclusive. One frustration is the lack of lists of figures and tables, or index. Inevitably, in a work of this scale, there are errors, such as misplaced letters or numbers on illustrations, but these are minor blemishes. Menez and his colleagues are to be congratulated on an outstanding and insightful report that will be an enduring source of inspiration and debate about the nature of Iron Age societies.