Archaeological knowledge of the non-elite Roman countryside based on academic excavations is virtually non-existent and highly dependent on the archaeological surface record produced in the many landscape surveys that have been carried out all over the Italian peninsula since the advent of systematic landscape surveys in the 1980s. Indeed, Bowes is only able to cite a handful excavations that have been carried out on rural sites that can plausibly be interpreted as the homesteads of Roman peasant families, a site category that, on the contrary, occurs frequently in the record of systematic surveys but not substantiated by excavation. The Roman Peasant Project, therefore, can be characterised as an undertaking that was geared towards the identification of peasant households in the Roman landscape. It opted to do so with a keen interest in the reconstruction of the environment in which they were set.
To this end Bowes and her multidisciplinary team of researchers selected a number of sites found in the surveys by landscape archaeologist Marielena Ghisleni, who compiled the archaeological map of the province of Grosseto in northern Tuscany between 2006 and 2009. The almost complete absence of the top tier of the rural hierarchy in the landscape surveyed by Ghisleni – the villa – made the study area attractive for Bowes's team as it offered the opportunity to study Roman peasant households in the context of their ‘locale’ without overt elite presence. Based on the small to modest sized artefact scatters mapped by Ghisleni, Bowes and her team expected therefore to be able to excavate the remains of peasant households, possibly with different wealth levels.
As it turned out, the archaeological excavations showed the rural smallholders’ use of the landscape to be much more complex than was expected, and only a few trenches provided convincing evidence of a domestic function. The majority of trenches pointed to seasonal occupation, sheds or stables and agricultural functions, including a (communal?) isolated oil press and a field drain. In other words, a quite different picture from what was envisaged in the project design.
Part 1 of the excellent report of the project introduces: (1) the project's profound historiographical interest in what the title of the book refers to as ‘the Roman Rural Poor’, but in the text consistently appears to be the Roman peasant household; (2) the project's landscape archaeological approach (comprising geophysics, excavation, artefactual, zooarchaeological and archaeobotanical analysis, and landscape studies); and (3) detailed excavation reports of the selected sites. Part 2 delves into the question where Roman peasants actually lived.
In Part 2 the project's landscape archaeological approach in combination with its anthropological underpinning (the concepts of ‘locale’ and ‘taskscape’) pays off as it paves the way for adopting the concept of ‘Distributed Habitation’. This concept does away with the static model of the single farmstead as central to the habitation and work environment of rural non-elite families. Instead, it posits a dynamic model in which the peasant farmstead figured as a permanent domestic unit within a range of specialised sites ‘set throughout the locale’, dedicated to the processing of products from the field, to animal husbandry, to water and land management facilities and to craft. Most likely these site types all functioned on a temporary / seasonal basis (p. 462).
The validity of the Distributed Habitation model is supported by the qualitative (by Bowes) and quantitative (by S. Collins-Elliott) analyses of the various phases of the excavated sites, of which, as stated above, only a minority could be attributed to the function of farmstead. It appears extremely useful to try and make sense of off-site scatters and scatters that do not correspond to household assemblages related to permanently occupied dwellings. This advances our understanding of locale and taskscape beyond the confines of the farmstead. On the basis of the observations of the environmental specialists that were part of the team, land use in the study area must be characterised as intensive rather than extensive. The evidence to support this notion has been gathered with much expertise and a keen eye for detail. The outcome, intensive agriculture, is in line with the chronology of the evidence from the excavations that gravitates on the period between the second century bce to late century first ce (besides a much less apparent late antique phase).
It is worth noting that the presence of late Republican and early Imperial fine wares and amphorae in the various excavated archaeological contexts in the study area points to a degree of integration of the peasant households into the regional market. One exponent of this phenomenon is the site of Marzuolo, which was also selected for excavation by Bowes and her team. Marzuolo is located on a terrace in the northern part of the study area and was discovered by Ghisleni in 2007. At the time of the survey there were three dense scatters of material visible (1–25 sherds / m2) composed of roof tiles, coarse and fine wares, much iron slag and mill-stones spread out over an area of c. 2.5 ha. Following initial geophysics and excavation by Ghisleni, the site was further explored in the RPP to test the assumption that Marzuolo represented a village with evidence for craft activities. While follow-up geophysics indicated the presence of an ‘agglomerated village with evidence for metalworking’, the major feature discovered was a series of archaeological contexts pointing to a ‘planned manufacturing centre for the production of Italic sigillata’ (p. 253). This was evident from the discovery of in situ stacks of terra sigillata of different productions in area 1 and from the presence in area 3 of three kilns (two so-called light-bulb shaped and one rectangular kiln). These were associated with the production of so-called ‘Experimental terra sigillata’, of which misfired fragments were found in a nearby dump. In the three excavated areas various buildings were uncovered with an origin around or just before 30 bce. Two out of the three areas gave evidence for the production of Experimental or Italic terra sigillata. Area 3 was established first with area 1 later in date, giving evidence (but without associated kilns) of the production (or storage?) of more developed phases of sigillata production, of around 50–70 ce. The early producers of this rural industrial place marketed their products locally, and its consumption was identified in the project's excavations at Pievina, Case Nuove and Tombarelle, and at four non-excavated survey sites. The second phase (50–70 ce) indicated an increase in scale and specialisation based on the quantities of vessels present. Marzuolo is furthermore thought to have functioned as a distribution centre for imported and local foodstuffs on account of an amphora deposit in Area 1, dated to 50–70 ce. The final phase of this remarkable site – that has continued to be under excavation – falls within the second to third centuries ce, when the area changes function from an industrial/commercial role to a rural vocation. On the basis of this important producer/marketing evidence in the ‘locale’, it does not seem too far-fetched to see the development of the non-elite agricultural landscape mapped by Ghisleni and partly excavated by Bowes and her team as a relatively short-term response to the increased market demand for agricultural produce in the late Republican and early Imperial period. The question that readers are left with is whether the late Republican to early Imperial peasant households brought to light in the project should be considered as representatives of the Roman rural poor. Finally, a methodological remark: according to Bowes the results of the project ‘emphasize the unreliability of surface survey for functional attributions, without extensive, open area excavation as a check’ (p. 449). One could, however, also state that (more) excavations are needed to inform the interpretation of systematic artefact surveys in order to increase the reliability of the interpretation of surface scatters. After all, even in the most ideal situations, only a small percentage of surface scatters in any project can be excavated considering the high costs and permits needed. Therefore, excavation projects targeting surface scatters from systematic surveys, as has been done by the Roman Peasant Project, are tremendously important for the advancement of site and off-site interpretations in archaeological survey. Bowes and her team are to be commended for this groundbreaking multidisciplinary project that has opened up new avenues in the archaeological study of non-elite rural society.
