A network approach to the past requires social relations to be understood through archaeological objects rather than through interviews or direct observations; consequently, we are working with proxies of proxies. The leap from the identification of static network patterns to their explanation in terms of past social processes is hazardous. In her application of network theory to the archaeology of Bronze Age—and later—Italy, however, Emma Blake has clearly succeeded in bridging this difficult gap.

To avoid the theoretical problems associated with the concept of ethnicity, Blake approaches the formation of cultural groups in terms of regionalism. Chapter 1 reviews evidence that includes origin myths in the ancient Greek literature and geographical distributions of Iron Age material culture and language groups. These strands of evidence combine nicely to reflect the regional groupings known from the fourth and third centuries BC, but are not so informative in delineating the origins or early development of these groups, which most scholars date to the Early Iron Age or even the Final Bronze Age. In connection with this, in Chapter 3, Blake discusses the problems of studying identity formation from material culture and explains why she adopts an interactionist model—it is the interaction between groups of people (social networks) and the particular form of those interactions (the properties of the network) that determine the success of these people in identity formation (‘path dependence’, pp. 77–79).

In order to identify these social networks, Blake considers the circulation of imports and specialised objects (mainly bronzes) across the Italian Peninsula during the Recent and the Final Bronze Age 1–2. In particular, if two sites produce evidence for the same type of object—and they are less than 50km apart, or a day's sail in the case of coastal sites—Blake considers the sites to be connected. This is based on the assumption that “In Bronze Age Italy, where long-distance trade was limited, non-local objects would have circulated along local exchange routes. One may expect the co-presence of identical rare objects at nearby sites to be far from coincidental” (pp. 71–72). The Bronze Age networks that Blake creates form into a number of discrete clusters.

The Recent Bronze Age network divides into four main clusters (fig. 4.1): the Po-Apennine, Lombardy, Trentino and Southern. While demonstrating regionalism, these clusters seem quite different from the regional identities distinguished in Early Iron Age Italy. With the Final Bronze Age, however, the picture changes significantly (see fig. 4.5) and network clusters correlate much better with the areas that will be subsequently occupied by recognisable ethnic groups: Garda, Veneto, Apennine, Etruria, Basilicata and Apulia (although some other regions are missing or at least not clearly visible: Molise, Liguria, Campania, Calabria). But what is most interesting is that “those Bronze Age networks that demonstrate cohesion and dense interactions are located where well-defined groups emerge later, as in the case of the Etruscan and the Veneti, whereas the weaker, disconnected networks precede poorly defined groups in the same area, as in the case in Apulia and Basilicata” (p. 19). With regard to verifying correlations between Final Bronze Age networks and later regional affiliations, Blake tests her hypotheses statistically, treating distance and shared material culture as independent variables. Not surprisingly, sites that are closer together show very high probability of later regional affiliation, but more interestingly, regardless of proximity, the co-occurrence of specific objects also demonstrates a significant statistical relationship to the probability of later affiliation.

In Chapters 5–8, Blake considers the network clusters in further detail and traces their regional development, incorporating hoards and ritual deposits that were excluded from the original analysis and also removing the distance criterion so as to verify the influence of actual geographical proximity. For northern Italy, the network data fit quite well to the principal groupings of the Terremare and Canegrate cultures, with the emergence during the Final Bronze Age of smaller groups: around Lake Garda, as part of a broader circuit of long-distance transalpine connections, and in the Veneto, focused on the site of Frattesina. For west-central Italy, the network data show a lack of strong coherent groups during the Recent Bronze Age, with two clusters roughly corresponding to the Tolfa and Fiora groups emerging during the Final Bronze Age. Blake interprets Villanovan culture as an aggregative phenomenon rather than—as suggested by Bietti Sestieri— one that broke off from a more loosely defined group. Latium meanwhile appears as a peripheral zone “with no internally driven networks of its own” (pp. 178–79).

In the Marche, Umbria and the Apennines, there seems to have been a lack of regionalism during the Recent Bronze Age and Final Bronze Age. Possible explanations include transhumance practices that favoured supra-regional circuits, as well as the disruption caused by later migrations. In the southern sub-group, Blake identifies two alternative networks—a maritime circulation of Aegean pottery and a localised terrestrial circulation of rare metal objects (with partial overlap in Apulia)—but neither can be tied securely to any of the regional cultural groups. Generally, southern Italy shows weak regionalism during both the Recent Bronze Age and Final Bronze Age, which can only be partially explained by the presence of foreign groups and is best understood in terms of social relations.

In the final chapter, Blake considers the regionalism that can be detected in Italy during the Final Bronze Age 1–2. Where cohesive networks are observed, stronger ethnic groups will emerge during the Iron Age, such as in Veneto and Etruria. In the south, instead, where fragile networks are observed during the Bronze Age, ethnic groups of the first millennium BC are poorly defined. Meanwhile, in the Apennine region, a cohesive supra-regional network grouping northern Etruscans, Umbrians and Picenes, seems to resist the hypothesis of path dependence but can be explained in terms of the mobility and migration of cultural groups. Generally, Blake's approach demonstrates a high degree of consistency between the archaeological data, network clusters and the ethnic groups of pre-Roman Italy. As she admits, network patterns or behaviours can sometimes be explained in multiple ways and therefore the overall consistency recognised here is significant.

While not the only approach to ethnicity—I have the impression that the instrumentalist approach of Barth, Patterson and others is dismissed a little too hastily (p. 70)—Blake's work is innovative and establishes a convincing link between social practices and identity formation. The book provides a good example of the application of network analysis in archaeology—technically detailed but also simply and clearly explained. The theoretical framework builds on a detailed archaeological and historical foundation. It is not fully clear, however, why only imports (including the introduction of the donkey) and specialised products (mainly metal objects) are considered, but common pottery is not. The observation that Final Bronze Age regional patterning in material culture does not appear to be reflected in the ethnic and cultural groups of subsequent periods is significant but insufficient in its own right and begs explanation. Generally, although the bibliography is wide ranging, some of the Italian scholarship is overlooked; for example, Renato Peroni identified a Mediterranean metallurgical koiné some years before Claudio Giardino.

By way of conclusion, Blake compares the regionalism of the Bronze and Iron Ages to the administrative regions into which the emperor Augustus divided Italy at the end of the first millennium BC, and, later still, to the regionalism of medieval and modern times. As far as the Augustan regions are concerned, Blake shows how in general “the stronger groups were respected while the weaker groups were not. Thus [. . .] we can detect, in a shadowy way, the impression they must have made to those who encountered them” (p. 251). In this respect, it ought to be noted that the use of the Augustan regions as a source for earlier regionalism in Italy is not completely new, and Pallottino—quoted by Blake in other passages—could have been mentioned here as well. In relation to medieval and modern Italy, Blake notes the strong unifying power of the institutional and especially linguistic centralisation imposed by Rome, under which regionalism certainly existed but did not endure or re-emerge in its original form—Italian regionalism before and after the Roman Empire were two different and separate cycles.