TY - JOUR
AU - , Van Oyen, Astrid
AB - Abstract: The presence, uptake and economic impact of innovations in the Roman world have been much debated. Not subject to debate, however, is the agency behind innovation, which is assumed to be the large, elite landowner. Evidence of experimentation at the rural terra sigillata production site of Marzuolo (Tuscany, Italy) does not fit dominant models of external investment in the Roman world and challenges the directionality of innovation. Instead, this article makes the case that experimentation at Marzuolo was driven by intensification on the part of local smallholders, but was curbed by a lack of capital investment. A later, scaled-up terra sigillata production phase at the same site, linked to infrastructural investments, shows predatory investment behaviour by a landowner who appropriated a tried and tested facility. Recasting innovation as an open-ended process of trial and error that is centred on human capital development, labour and relations of production, changes the terms of study of the Roman economy and aligns it with broader conversations in economic history. I the question of rural innovation No question better illustrates the fault line that continues to divide the study of the ancient economy than innovation. In a seminal article in 1965, Moses Finley downplayed the structural role of innovation in the ancient economy by conflating a substantivist line of argument — ancient elites were concerned first and foremost with status, not productivity — with a formalist logic — readily accessible slave labour obviated the need for increased technological efficiency.1 The case of the watermill has become iconic: although it was invented in antiquity, and despite its potential to revolutionize ancient production and consumption, it was adopted only hesitantly, if at all, according to Finley.2 In other words, the fabric of the ancient economy had no role for innovation. The argument about the watermill acted as bait for the modernists opposing Finley, who countered by drawing on increased archaeological finds of, indeed, such innovative devices as watermills and screw presses, to prove their real uptake and the resultant effect on productivity and economic growth.3 Finley would have been unimpressed by such expanding anthologies of archaeological finds, reading them as yet another symptom of ‘that great curse of archaeology, the indestructibility of pots’.4 The problem, however, is not the pots (or watermills, for that matter) themselves, but Roman archaeology’s dominant approach to them, which takes such modes of representation as distribution maps or bar charts as its end point. As such, innovation is black-boxed: the assumption seems to be that after an original moment of invention, innovations became part of a stock of knowledge, ready to be tapped whenever there was a need for it (that is, to increase productivity), and waiting to become a dot on the archaeologist’s distribution map, as it were. Any argument that ensues is by necessity quantitative in nature and method: more watermills add up to a bigger economy.5 What falls through the cracks is innovation as a human and historically situated process; what is needed are micro-histories of innovation as an open-ended, uncertain process. Marzuolo, a rural site in south-central Tuscany (province of Grosseto) (see Plate 1), challenges such a post hoc perspective on innovation. In 2012, excavation by the Roman Peasant Project at Marzuolo discovered evidence of terra sigillata production,6 the Roman Empire’s iconic ceramic tableware with a distinctive shiny-red surface, the result of innovations in firing technology.7 A dump in the eastern section of the site (Area 11000; Figure 1) contained fragments of so-called experimental terra sigillata pottery (c.30/20–10 Bc), conforming to the shapes of contemporary output at Arezzo, where its manufacture is assumed to have originated, but non-slipped, poorly slipped or misfired (Plate 2).8 Two generations later, by 50–70 Ad, Marzuolo was a full-blown terra sigillata production site, as proven by an extraordinary find of stacks of pots of different shapes (Plate 3; found in Area 10000, Figure 1).9 1. Open in new tabDownload slide Podere Marzuolo in its regional context. Red: terra sigillata production sites; orange: towns; black: villas. 1. Open in new tabDownload slide Podere Marzuolo in its regional context. Red: terra sigillata production sites; orange: towns; black: villas. 3. Open in new tabDownload slide Podere Marzuolo, terra sigillata plate with MANNE stamp in planta pedis (50–70 Ad). Source: Marzuolo Archaeological Project. 3. Open in new tabDownload slide Podere Marzuolo, terra sigillata plate with MANNE stamp in planta pedis (50–70 Ad). Source: Marzuolo Archaeological Project. Figure 1 Open in new tabDownload slide Podere Marzuolo, Plan of Roman-Period Structural Remains From the 2012–2018 Excavation Seasons* * Source: Marzuolo Archaeological Project. Figure 1 Open in new tabDownload slide Podere Marzuolo, Plan of Roman-Period Structural Remains From the 2012–2018 Excavation Seasons* * Source: Marzuolo Archaeological Project. 2. Open in new tabDownload slide Podere Marzuolo, experimental terra sigillata plates with radial SEX/SEXTI stamps (30/20–10 Bc). Source: Marzuolo Archaeological Project. 2. Open in new tabDownload slide Podere Marzuolo, experimental terra sigillata plates with radial SEX/SEXTI stamps (30/20–10 Bc). Source: Marzuolo Archaeological Project. The first, experimental production phase at Marzuolo demonstrates that once developed in Arezzo, the new terra sigillata pottery and its manufacturing techniques did not simply become part of a repertoire, but that its transplantation and relocation needed work.10 It is to this work that the present article draws attention, thus diffracting innovation into capital, infrastructure, planning and skill development.11 As such, it moves away from a narrow, post hoc understanding of innovation as the uptake of technical devices to increase productivity, and opens up the black box of contextual adjustment and alignment needed to create a systemic ‘fit’ for an innovation.12Terra sigillata pottery may seem rather frivolous when compared to such things as watermills and screw presses, but they become more akin when the whole process of production is opened up to scrutiny. Traditionally, innovation in terra sigillata production is glossed as the mould-making process (used only for decorated terra sigillata pots!), whether this merely indicated a fashion trend (as per Finley)13 or actually reduced production costs.14 Instead of focusing on one particular aspect and debating its relation to productivity, this article examines the terra sigillata production process as an open-ended exploration and development of new materials (including clays and fuel), new skills, new techniques (such as firing at high and sustained temperatures), and new consumer demand; questions that could and should likewise be asked of the development and adoption of, say, a watermill. But the site of Marzuolo also offers another challenge to the existing discourse on innovation: as an unexpected location for innovation, it calls into question the directionality of the innovation process and, more broadly, the agencies at stake in Roman history. The Roman world’s agrarian economy structurally relied on land ownership. As a result, it tends to be assumed that large landowners were the players of the Roman world, deriving both social status and economic revenue from their portfolio of agricultural estates or villas, and anchoring these as political power through towns.15 It is no surprise then that innovation is typically located in town and villa, the realms of the large landowners, who could marshal funds together and had access to the wide distribution network that made investment in innovation economically viable,16 following a core–periphery logic.17 Technologies requiring large capital investment, such as watermills, do indeed cluster on villa estates and near cities.18 (Note that for watermills, this is a question of storage and commodity chains as much as of innovation in the milling device: flour did not preserve well so grinding large quantities of grain in a mill only made sense against the background of significant immediate consumption.)19 Yet even ‘low-tech’ innovations such as terra sigillata production, necessitating little capital investment, are often traced back to town and villa. The increasing evidence of rural terra sigillata production sites in central Italy has not changed this narrative (Plate 4).20 Limited excavations at Torrita di Siena (province of Siena), c.60 km north-east of Marzuolo, revealed the footprints of two rectangular kilns, as well as vases and production materials such as moulds dating to the mid first century Ad.21 Torrita di Siena was well connected to the Via Cassia, a main north–south road, and probably associated with the nearby villa of Poggio al Vento. A tentative association with a villa has also been postulated for Sesta (Montalcino, province of Siena), located c.10–15 km upstream from Marzuolo along the northern banks of the River Orcia and known only through survey evidence, where terra sigillata production is hypothesized based on the presence of three first-century Ad moulds.22 The site of Scoppieto in the Tiber Valley (Baschi, province of Terni), in turn, produced terra sigillata from the middle of the first century Ad onwards, with evidence including kilns, a clay levigation tank, individual potters’ workstations, as well as production materials such as moulds and figure stamps.23 Its proximity to the Tiber allowed Scoppieto to send its products downstream to meet a guaranteed demand in the city of Rome. In other words, city and villa continue to inform explanatory frameworks of terra sigillata production, innovation and the Roman rural economy more broadly. 4. Open in new tabDownload slide Terra sigillata production sites in central Italy. 4. Open in new tabDownload slide Terra sigillata production sites in central Italy. While terra sigillata production evidence has not challenged the status quo in our reading of rural dynamics, the expansion of survey archaeology in Italy has, showing a more diverse Roman countryside in the crevices between town and villa,24 where villas and farms were complementary rather than exclusive categories,25 and where casual free labour operated alongside, or sometimes instead of, slave labour.26 Moreover, the first excavations away from villas reveal a landscape of activities beyond mere settlement, structured by special-purpose sites (for example a pressing site) and a ‘spectrum of habitation’.27 It was into such a landscape, unknown from textual sources, that Marzuolo fitted. The site was located at least a long day’s walk (c.35 km) from the nearest city of Roselle and its coastal connections, and equally distant from the main traffic arteries, the Via Aurelia (to its west, along the coast) and the Via Cassia (to its north-east, inland) (see Plate 1).28 Characterized by rolling hills and steep, seasonal valleys, Marzuolo occupied a transitional position between the coastal plains, with villa-based agriculture,29 and the hills further inland, characterized by dispersed habitation.30 The site extends c.2 ha on a plateau bordered to the north by the Orcia, a non-navigable tributary of the Ombrone river, which acted as an east–west communication route.31 Marzuolo was a small rural centre, serving a modest regional market — neither the towns on the coast nor the city of Rome were easily reached. Such nucleated rural centres are often assumed to have been installed on the estate of a large landowner, although empirical evidence for ownership is not easily mustered for Marzuolo.32 The nearest villas (Plate 1) — Santa Marta (Cinigiano), Podere Cannicci, Dogana (Paganico) and Sesta (Montalcino) — all developed well after the start of occupation and terra sigillata production at Marzuolo.33 How to account for innovation at a site such as Marzuolo without straightforward links to town and villa? This article takes up the twin challenge posed by Marzuolo for the study of Roman innovation and investment. First, Marzuolo urges us to diffract the post hoc view of innovation as a fait accompli, instead probing contextually whether the model of investment by large landowners, channelled through town and villa, left space not just for adopting the latest novelty, but for experimentation, foresight and skill development. Second, Marzuolo offers the opportunity to investigate to what extent the diverse and dynamic Roman countryside that we are starting to see archaeologically, represented here by way of terra sigillata production at a small nucleated site, was locked into the Roman world’s broader networks of power and dependency. Was the directionality of innovation really always defined, as assumed by both Finley and his critics, by a trickle-down from large to small, from villa to farm, from town to country? The topic of innovation goes to the very heart of debates on the structure of the Roman economy and of the dynamics of the Roman countryside. Moreover, just as innovation’s impact on the Roman world has tended to be measured in reference to, or in contrast with, the post-Roman world (back to the watermill and its diffusion),34 so too its rethinking is likely to reverberate beyond the disciplinary bounds of Roman history. Indeed, the directionality of innovation is being rethought both in terms of bottom-up/top-down and core–periphery dynamics, in anything from the early medieval period to early modernity.35 II a story of investment Human capital Ancient innovation, as argued earlier, has been glossed as the uptake of new machines or devices, which represents one form of capital investment. In such studies, the question of why people in the past innovated — that is, chose to adopt such devices — boils down to whether or not they wanted to invest in productivity following a straightforward logic linking investment and result. Another form of capital, though, has received only scant attention for the ancient world, and that is human capital, or the investment in the development of skills and knowledge that inhere in, and are embodied by, a person.36 The fact that it is so sought after in today’s economy proves that human capital, too, has an effect on productivity, and indeed, the utilitarian overtones of the term have garnered critique.37 This article, instead, glosses ‘human capital’ as a form of ‘delayed return’, deferring short-term profit and foregoing a direct relation between investment and output in view of anticipated but uncertain future possibilities — not necessarily of economic value — which puts a different spin on the question of ‘why innovate’. More specifically, it draws attention to demand, and, perhaps more importantly, it encourages a dynamic, contextual and socially embedded account of labour.38 What is anomalous about the terra sigillata production at Marzuolo is not so much its output in the first century Ad, which, as mentioned earlier, followed an expansion across central Italy, but its precocious first phase, only a decade or two after the first tentative red slips were produced on sigillata vessels at Arezzo. It is this first phase that speaks most clearly to human capital development. First, the associated ceramic fragments (the majority of which were found in a dump, but some were present in use contexts across the site) have a slip of poor quality, unevenly applied, and often misfired,39 suggesting a trial-and-error learning process. This hypothesis finds further support in evidence of tinkering with tools and infrastructure. Three kilns have been excavated in the eastern part of the site, within a ten-metre radius of the dump that contained the experimental terra sigillata fragments (Figure 1). All kilns were found empty, and none can be related directly to the manufacture of experimental terra sigillata pottery, even if these respective features were found in the same general activity area. Yet overall, they point to experimentation with kiln design and technology. The footprints of two small (c.80-cm diameter) keyhole-shaped kilns were preserved near the dump in the backyard of a commercial/artisanal building (Area 11000).40 A differently oriented kiln in the front yard of this same building was larger (c.2.50 × 2.25 m), rectangular in shape, and with a firing chamber sunken partially below the surface level.41 During its service life, the rectangular kiln was modified with the addition of a secondary entrance or flue perpendicular to its original entrance — an intervention that demonstrates active experimentation with the firing atmosphere (Plate 5). A similar concern with the firing atmosphere — the key to obtaining the glossy red slip characteristic of terra sigillata — emerges from fragments of thin ceramic plates, pierced with small holes and found both covering the keyhole-shaped kilns and as part of the dump with experimental terra sigillata. While none of the pierced fragments show traces of firing, Emanuele Vaccaro proposed identification as an ‘experimental’ muffle to contain the pots during firing and buffer contact with the gases.42 5. Open in new tabDownload slide Podere Marzuolo, Area 12000, rectangular kiln with modified firing chamber. Source: Roman Peasant Project (Marco Sfacteria and Elisa Rizzo). 5. Open in new tabDownload slide Podere Marzuolo, Area 12000, rectangular kiln with modified firing chamber. Source: Roman Peasant Project (Marco Sfacteria and Elisa Rizzo). Such a co-development of skills and production environment is one of the hallmarks of innovation.43 Yet this process did not require large-scale capital investment. All three kilns are constructed out of clay and mudbrick, readily available and cheap materials. A single potter’s name is associated with Marzuolo’s experimental terra sigillata phase: Sext(ili)us. Dies reading ‘SEX’ are attested elsewhere, while a ‘SEXTILI’ die has been found at Arezzo too, all (tentatively) dated to the early Augustan period.44 Sext(ili)us is a common name for slaves and freedmen, however, so these links cannot bear much explanatory weight. The trial-and-error process observed archaeologically suggests someone keen to try new things (obtain red slips), with a general idea of how to get there (by modifying the firing atmosphere), but without specific operational knowledge to achieve this result. Sextilius could have been a potter altogether new to the techniques of terra sigillata production. But even if Sextilius was a potter from Arezzo, he came to Marzuolo at best with experimental knowledge of how to achieve consistent red slips. In any case, investment in this first phase had to take considerable interest in the development of human capital, and this very process would have shaped Sextilius’ identity and place in the Marzuolo community. The role of name stamps on terra sigillata pottery is debated.45 At large-scale production sites they probably played a part in production organization, and for consumers they may have become an expected product trait.46 Whatever the reason that caused Sextilius to stamp his products, the very act of stamping betrays the cultivation and display of an identity as craftsman and businessperson, which has been signalled as a key process of innovation in contemporary rural contexts.47 For the second terra sigillata production phase at Marzuolo, instead, a model of specialist itinerant craftsmen finds support in the pattern of stamps: the majority of preserved stamps belonged to a certain Manneius, who is also attested (by one stamp) at Torrita di Siena and at Arezzo (Plate 3).48 In addition, the L.VM.H stamp on twenty-two vessels at Marzuolo points to the involvement of a member of the Umbricius family, the dominant producers at Torrita di Siena.49 Experimentation and human capital development were less central to the success of the second terra sigillata phase at Marzuolo, which, instead, relied on the attraction of specialist artisans with operational knowledge linking product and process. Planning and labour Machines and human capital do not exhaust the investment portfolio underpinning innovation. Also often overlooked is the foresight that is revealed by planning. Here too, Marzuolo offers striking evidence. The consistent alignment of buildings at the site testifies to a significant degree of planning in construction across a rather narrow timeframe (Figure 1). The building that later housed the experimental terra sigillata production activity was constructed around the third quarter of the first century Bc; a large structure in opus (quasi-)reticulatum, following the same alignment, was added to the site between the mid-Augustan and Julio-Claudian period. Opus (quasi) reticulatum is a distinct, labour-intensive type of concrete masonry rarely found outside cities and villa sites.50 Where it has been attested on rural sites that defy the label ‘villa’ elsewhere in Roman Italy, opus reticulatum masonry has been seen as a quirk in need of explanation, for instance in the form of a state-run programme of colonial resettlement as proposed for the small farm of Monte Forco in the Ager Capenas.51 At Marzuolo and its environs the common construction technique consisted of walls in pisé or packed clay on a three- to four-course pedestal made of unmortared river stones. In contrast, the opus reticulatum walls consistently have a carefully curated, 80 to 90-cm-deep mortared foundation (Plate 6). The cubilia — the individual stones constituting the net-like pattern of opus reticulatum facing — are of particular interest at Marzuolo. Such stones are cut in a standardized pyramidal shape, which would have facilitated the assembling process, especially when using unskilled labour on large-scale construction projects.52 Cutting the cubilia into their pyramid-like shape would have been the most labour-intensive stage of the process, but in Rome and its environs the relatively soft tufa stone was highly amenable to shaping. Conversely, Marzuolo was situated in a region that lacks good building stones.53 The cubilia at Marzuolo were chiselled out of relatively hard limestones obtained from the River Orcia, rendering cutting the stones a challenging task and one that was comparatively wasteful of materials, but especially of labour, and turning their assembly into a large puzzle.54 6. Open in new tabDownload slide Podere Marzuolo, opus reticulatum wall with foundation cut in levelling layer. Source: Roman Peasant Project. 6. Open in new tabDownload slide Podere Marzuolo, opus reticulatum wall with foundation cut in levelling layer. Source: Roman Peasant Project. The dimensions of the opus reticulatum building are massive: its northern wing alone has been exposed over a length of 31 m, and still continues in both directions beyond the excavation limits. Its layout of large cell-like spaces with wide entrances opening up to a shared yard (Figure 2) suggests a commercial and/or productive function, but precisely what kind of activities the building was originally destined for remains unclear. Two features can be securely related to the original design: a sunken, rectangular space in a corner of a room, paved with tiles, and in close proximity to a drain (Area 14000);55 and (at least) two cylindrical tanks covered with a waterproofed mortar, at the back end of a room with a similarly waterproofed pavement (Area 20000).56 A role in processing either agricultural products or raw materials for craft seems plausible. Moreover, a carefully crafted channel with two mortared walls, a base and a tile cover ran along the building’s back wall, possibly collecting rainwater runoff.57 The whole project was monumental, labour-intensive, and meticulously planned. Figure 2 Open in new tabDownload slide Podere Marzuolo, plan of Opus Reticulatum Building and Craft Installations. Black: First-Use Phase (Augustan); Grey: Second-Use Phase (First Half of First Century Ad)* * Source: Marzuolo Archaeological Project. Figure 2 Open in new tabDownload slide Podere Marzuolo, plan of Opus Reticulatum Building and Craft Installations. Black: First-Use Phase (Augustan); Grey: Second-Use Phase (First Half of First Century Ad)* * Source: Marzuolo Archaeological Project. Around the middle of the first century Ad, a fire destroyed the central part of the building, which was abandoned thereafter. Despite a rather short lifespan of two generations at most and a nifty original design, the opus reticulatum building went through structural modifications associated with its use as a craft complex during the course of the first half of the first century Ad (Figure 2, grey). One of the original large cells was subdivided by a low tile wall to create a rectangular waterproofed tank with evidence of multiple phases of (re-)plastering, and a large ceramic water basin was installed. The cylindrical tanks, in turn, were partially filled in and went out of use before the fatal fire, as did, probably, the back channel. One of the cells was occupied by a blacksmith, with a set of iron tools and objects in the course of repair left under the fire debris, as well as concentrations of highly magnetic hammerscale.58 The aforementioned stacks of standardized terra sigillata pots, in turn, were retrieved from the yard in front of the easternmost cell, where they had been sitting on shelves at the time of the fire. Finally, the evidence for ceramic production itself bears witness to considerable investment. The most significant material resources at Marzuolo were probably fuel and clay, both deriving from land ownership; otherwise no materials with great acquisition costs are represented.59 The experimental and standardized terra sigillata production phases at Marzuolo used different clays, suggesting the exploitation of different clay beds.60 The use of finer clays was apparently outside the realm of possibilities for the producers and investors of the first phase, either because they had not yet identified more suitable clays or strategies of refinement (a question of skill), or because they did not have access to certain clay beds (a question of ownership and resources). In sum, the craft complex at Marzuolo seems to be a place of paradox. Its construction reveals high expenditure, and a top-down design, yet the various ad hoc modifications during its short service life hint at bottom-up agency on the part of the craftsmen working in this space. Its role in manufacture (and possibly initially agricultural processing) speaks to a goal-oriented concern with output and productivity, yet expenditure on the carefully constructed opus reticulatum walls defies a simple cost–benefit analysis; indeed, there is no reason why the traditional, cheaper walls with drystone pedestals could not have performed just as well. III where is experimentation in third-party investment? Section II established the extent and nature of investment at Marzuolo, which revolved around human capital development, planning and labour. But is it possible to get at the source of investment? Legal status of both owners and labourers is impossible to deduce from archaeological remains alone. It is thus bound to remain unclear whether the Marzuolo landscape was inhabited and worked by independent or dependent smallholders, or by slaves. Nevertheless, the current and following sections bring the evidence from Marzuolo into conversation with insights from archaeology, history and anthropology, in order to arrive at an explanatory structure — the best fit for the current data. A certain degree of ambiguity and extrapolation is inevitable, and some evidentiary gaps will need to be closed by means of logic and parallels, until evidence to the contrary appears. Such methodological risk is necessary to free archaeological evidence from the shackles of distribution maps and bar charts; in other words, to write archaeological micro-histories. As we will see, the linchpin in this particular micro-history is the allocation of agency in the first, experimental phase of terra sigillata production at Marzuolo. The planning and expenditure represented by the opus reticulatum building in all likelihood results from direct investment by an established landowner and/or businessman, possibly the Caius Decum(i)us whose name is stamped on no fewer than fifteen roof-tiles (re-)used in modifications to the craft facilities discussed earlier.61 The layout of craft spaces across the site of Marzuolo, with shared, widely accessible yards and production installations, yet with workspaces divided into cells resonates with a model of nucleated workshops as proposed for terra sigillata production in Arezzo by Gunnar Fülle.62 The landowner thus did not set up and organize a fully integrated firm, but more likely invested in infrastructure, and possibly raw materials, to be leased out through a variety of contracts (locatio-conductio).63 The leasing model is a good fit for the second terra sigillata production phase at Marzuolo whose potters were, as argued earlier, itinerant specialist craftsmen, setting up shop at different sites over time, thus creating a de facto network between production centres.64 The modifications to the opus reticulatum building would then reflect different lessees coming in through the years and adjusting the space to their needs. While the model of direct investment works well for the second terra sigillata production phase, the question is whether it can be extended, retrospectively, to the first, experimental phase. If so, no structural change occurred between both phases. A large landowner, who was quick to see the potential of the relatively newly produced red-slipped tableware and of the readily accessible clay and water sources near Marzuolo, would have planned and launched the site from scratch, investing in human capital development by Sextilius. Initial production would have been successful enough to attract more lessees, while knowledge of both the terra sigillata production technology in general, and the local resources and conditions more specifically, was refined. The second production phase would then represent the apogee of this gradual process, with several potters’ names attested — some of whom are known from nearby Torrita di Siena — and a range of forms and production infrastructure represented. This narrative of direct investment offers the best fit for scholarly assumptions about the structure of the rural economy: as mentioned earlier, indices of investment tend to be traced back to a landowning elite. But it does not find support in the data from Marzuolo: on current evidence, nothing connects the first and second production phases at Marzuolo. Not only were clays and potters different, the experimental production took off in an area of the site (Area 11000) that had a longer occupation history and a definite ‘local’ footprint in construction technique and living assemblage, almost 100 m removed from the concrete building. Even if the latter was erected around the same time as the pottery production took off (current evidence situates it a couple of decades later), no causal link can be posited. More generally, the model of direct investment falters when confronted with evidence of experimentation and human capital development. Elite attitudes to investment are subject to debate:65 based on the Heroninus archive from Egypt, Dominic Rathbone pictures landowners investing maximally with a view to increasing return.66 Conversely, Dennis Kehoe sees Roman landowners as investing in a more risk-averse way, privileging a secured return over maximized profit.67 Kehoe has been reproached for a limited focus on examples of agricultural leasing, but a similar risk-reducing strategy of investment has been proposed for ceramic production at Sagalassos (Turkey).68 Moreover, evidence of villas shows landowners carefully balancing investment with anticipated profit: new avenues for profit-making were definitely tapped, but only if a secure return could be expected. For instance, when competition from the provinces dried up the market for overseas export in wine, villas in central Italy partially shifted focus to the production of oil, grain and livestock for an Italian market.69 The market for Marzuolo products was at best limited to the regional settlements and their inhabitants, and especially the experimental terra sigillata production phase seems rather too tentative and small-scale, and its return too uncertain, to have warranted external planning and investment. The question of innovation, though, is about more than the risk of investment an sich. It is about what kinds of investment were made by wealthy entrepreneurs. Risky investments were not altogether shunned by landowners and entrepreneurs,70 but these involved known entities (commodities, skills, contracts, etc.) that could be, as much as possible, calculated and controlled. For example, estate owners might expand by adding an extra press,71 or by scaling up their storage facilities.72 Or they might buy a vineyard in a state of neglect, confident that they could marshal not just the funds but also the skilled labour to bring it to fruition.73 Even the most risky of endeavours, trade by sea, was sometimes undertaken, but the risk inhered in the vagaries of transport, not in the demand for the commodities shipped. Much the same picture can be gleaned from literary sources. Columella, for instance, emphasizes the importance of paying extra for a highly skilled vinedresser slave, but this skill is bought on the market as a ready-made commodity.74 When Plutarch describes Cato’s investment in the training of slaves in order to sell them after one year (‘They would buy boys, train and teach them at Cato’s expense’),75 skill is once more seen as a commodity, for which the large demand offsets the production costs. And Cato himself admonishes against hiring the same labourers for more than a day when running a villa — effectively preventing any form of human capital development. In other words, investment in human capital by large landowners was limited to buying or selling of a skillset predefined by, and circulating through, a commodity market76 — precisely the critique of commodification that ‘human capital’ receives today, and against which the evidence from Marzuolo cautions. Indeed, experimentation translates with difficulty based on such a premise. Even if landowners adopted a more laissez-faire approach to craft production, through leasing of facilities, their investment was structurally at odds with experimentation and human capital development.77 Three mid third-century Ad papyri from Roman Egypt that preserve the lease of a pottery for the production of amphorae are revealing in this regard.78 The pottery was situated on the estate of the landowner, who provided raw materials (‘earth’, water and fuel) and paid the lessee a set cash figure in return for a predefined number of pots. The lessee took on a large farmstead ‘together with its storerooms, kiln, potter’s wheel and the other equipment’.79 Of particular interest is the prior specification of the pots to be produced by the lessee not just in quantity but also in type (‘what are termed Oxyrhynchite four-choes jars’) and quality (‘well fired and coated with pitch from the foot to the rims, not leaking and excluding any that have been repaired or blemished’).80 Such a degree of prior product specification implies a developed market, in which product characteristics and expectations were generally understood, which lowered transaction costs.81 Experimentation has no place in such a contract: the investor, concerned with a guaranteed fixed output, would have been wary of the human capital development needed, as witnessed by for instance the tinkering with the firing infrastructure at Marzuolo. The owner/investor was concerned only with products, not with process. The lessee, in turn, would have been unwise to take up a contract without having the established skills or knowledge network to deliver the stipulated products. Perhaps the most notable observation about the Egyptian pottery leases is that human capital and human labour — the ‘who’ and ‘how’ of production — are entirely at the discretion of the producer: ‘I provide for myself sufficient potters, assistants and stokers’. The lessee is expected to have or develop a knowledge network, including wage labourers, trainees and apprentices, who could be called upon if and when a contract materialized. Again, experimentation such as witnessed in Marzuolo’s first phase has no place in this model. Indeed, Fülle concludes that the ‘best way [for a rural landowner] to take part in the terra sigillata business would have been to engage specialists who had already gained some experience elsewhere, for example in Arezzo, to establish and to undertake production while using the existing pottery facilities’.82 The problem of human capital development is thus left unanswered. Little is known about relations between workshops and potters at the earliest stages of terra sigillata pottery, to which the experimental production phase at Marzuolo belongs. Neither the name nor the dies in question allow tracing of secure links between Marzuolo’s Sextilius and other potters or production sites (at this time limited to Arezzo); in other words, there was no knowledge network or market in specialist artisans from which investors could tap to produce shiny red pots. And at that time there was not even a common template, with clearly defined product specifications, for such shiny pots. IV bottom-up innovation While a model of external investment by a large landowner, in particular when paired with the leasing of facilities, can account for the second terra sigillata production phase at Marzuolo, the first, experimental phase begs a different explanation, urging us to shuffle historical agencies. What other players can we call to the stage of Marzuolo’s history? Investment at Marzuolo was primarily a matter of labour — no large investment of funds was needed. In pre-industrial societies, labour is the smallholder’s joker: both seasonally and generationally, farming units will have surplus labour to devote either to intensifying investment per task (with diminishing returns), or to diversifying tasks (with the potential of increased overall returns).83 Externalized rural labour — mouths fed by agriculture without contributing to its productivity — can be roped into non-agricultural tasks such as crafting or transport.84 Diversification was not merely a risk-reducing strategy,85 but could also add up to overall intensification.86 Recent excavations of small rural sites, paired with environmental studies and ethnography, demonstrate that intensive cultivation strategies made sense for small farmers,87 that these could be more productive than previously assumed,88 and that farmers were able to make careful choices to respond to changing needs, labour requirements, climate, and so on.89 Contemporary rural sociology identifies a similar farming style as ‘economical farming’, with low external capital input yet high output achieved by raising labour input and technical efficiency.90 Economic anthropology too is undoing the stereotype of the conservative farmer and artisan,91 showing that small farmers, peasants and craftsmen routinely engage in high-risk behaviour, often directly drawing on market exchange,92 and that both their economic and moral frameworks are context-dependent.93 Meanwhile, long overdue attention to Roman small farms starts yielding increasing empirical evidence for innovation.94 If innovation typically falls well outside the historical script of the Roman smallholder, this is for want of looking, with our analytical gaze rusted into some oft-repeated assumptions and excavation typically limited to towns and villas.95 Collectively, a landscape of small farmers engaged in ‘economical farming’ could produce enough of a surplus to warrant the installation of a full-time craftsperson, a Sextilius, whether local to the area or immigrated (for example from Arezzo).96 Sextilius could have invested in a modest pottery by himself in order to tap that regional surplus, or he could have had an arrangement with (a) local farmer(s). He could have been a full-time potter dependent on others’ (such as his farming relatives; the small farmers around Marzuolo) produce or a farmer-potter himself. In any case, his activities would have relied on local demand, while also diversifying the local production portfolio and absorbing agricultural surplus labour (for instance for clay mining or transport). Ultimately, what mattered to Sextilius was people buying his pots. But what kinds of pots appealed to the people living in and around Marzuolo? Like Marzuolo itself, most sites in the region date back to the Augustan period. In other words: at the time of Sextilius, this was a rather new landscape, with a new population, including people uprooted and displaced by the civil wars. While the Augustan peace is all too often invoked as a generic institutional backdrop for changes in the early imperial period, in an Etruria that had been caught in the civil wars of the first century Bc, it would have had direct and real consequences,97 fostering increased integration and the extension of networks.98 Such networks would have guided the flow of both labour and demand,99 including a burgeoning taste for red pots.100 The ensuing horizontal mobility would have helped foster the conditions for investment in human capital, that is demand and a structure through which portable human capital could flow. If one had to, or chose to learn the ropes of potting in Augustan Etruria with its new sites and new consumers, then the terra sigillata pots newly coming out of Arezzo would have seemed like the product to aspire to. If, instead, Sextilius came from Arezzo with some tentative experimental knowledge of terra sigillata production, then this would have been welcomed by his community of newly connected consumers in and around Marzuolo. It took low-tech and bottom-up investment to respond to this demand by innovating. As a farmer-potter, or as a potter supported by small farmers, Sextilius would have been less constrained by previously normative products and skills; as a do-it-yourself potter, he was not dependent on (or did not have access to!) a pre-existing market of specialist human capital. Instead, experimentation and human capital development were part and parcel of smallholders’ economic orientation. First, crafting as a strategy of diversification by a farmer or an agricultural community would always have entailed the development of human capital: farmhands became (part) craftsmen.101 Incidentally, we see the same process at work, under different conditions of external investment, and in the pursuit of a predefined model, in the opus reticulatum building: while its design and its engineering speak to imported specialist skill, its execution — the cutting and assembling of the stones — drew on local labour (farmhands becoming stonemasons) and local materials (limestone).102 Small producers were not necessarily ‘jacks of all trades’, but had to value human capital development and experimentation in order to maximize their foremost resource: labour. Second, in contrast to large landowners, opportunity costs103 — the theoretical loss of the path not chosen — were low for small producers, who could start crafting but could not easily buy another plot of land or a pair of oxen. Put differently, opportunities were constrained by low means of capital investment. While a large landowner could simply have invested in scaling up an existing portfolio, for instance by buying an additional estate, a smallholder had every incentive to diversify their portfolio and to squeeze more out of available resources.104 But surplus rural labour was available only as long as subsistence strategies were not jeopardized: for small producers, the margin of low opportunity costs was real but small. In other words, activities other than agriculture were easily started but pursued and expanded only with difficulty. Technical development was key in exploiting this narrow margin, but in the absence of large capital investment, this took the form of tinkering, trialling and, indeed, experimentation.105 The same locally rooted process of tinkering, adoption and adaptation can be seen at Marzuolo in a new type of transport amphora, attested in a dump covering the modified kiln in the eastern sector, and thus dating roughly to the same chronological horizon as the experimental terra sigillata pots. The amphora’s triangular rim is reminiscent of the Dressel 1 type, of broad Mediterranean currency at the time, but its grooved handles, its smaller size, and its flat base (for overland transport by cart) make it a truly local innovation, fitting — and shaping — new regional demand and transport networks.106 V predation In a second phase, but probably only a generation later, activities and infrastructure at Marzuolo were scaled up, a move represented by the later standardized production of terra sigillata.107 As mentioned earlier, elite intervention at Marzuolo is likely at this stage, by a landowner who either already had a presence on site or bought into it soon after its origin (depending on the precise chronological relation between construction of the opus reticulatum building and the experimental terra sigillata production phase). This resonates with the region-wide expansion of terra sigillata production across Etruria, including sites such as Torrita di Siena and Sesta that were launched in the first century Ad and in relation to a villa. Similar rather swift transitions in techniques, clays and scale of production have been noted on terra sigillata production sites in Gaul.108 In contrast to an assumption that innovation trickled down from wealthy landowners to poor farmers,109 this article argues that a structural break separated the initial experimentation at Marzuolo from the later, scaled-up investment and production output. At this point, it is useful to draw a modern parallel, not in order to set up a proper comparative argument, but to highlight some features of the Roman rural economy that have hitherto remained understated. Anna Tsing describes how modern capitalist firms prey on both material and human resources: their work has shifted from production itself to the work of gathering and aligning different production modes and making the products commensurate — a supply chain model.110 Risk is borne disproportionally by the lowest producers, not by those making profit; investors in turn operate in a predatory way, increasing productivity by expanding and appropriating more production units, not through innovation or intensification. Incentive for developing human capital lies squarely with the small producers.111 ‘Predation’ indicates a parasitical, sometimes even violent, relation between capital and human capital. At Marzuolo, predatory investment behaviour would have kicked in once the experimental stage — which involved ‘testing’ local resources, developing a market, and routing local skills and operation systems — was broached. The burden of such an experimental stage — in other words, the work of innovation — would have fallen on the small-scale producers. Following the supply chain model, small producers actively sought out innovation and developed human capital. But why not tell the story of Marzuolo as a rosier triumph of small farmers and experimental potters? Why could a C. Decum(i)us enter the stage? However intensification-driven and experimentation-friendly, ultimately small producers were caught in a scalar bind due to a lack of capital. Development economics shows time and again that smallholders typically start many small businesses (a strategy of diversification), but struggle to scale up without outside injections of capital.112 Such businesses often have high marginal returns (investing a little more would be worthwhile), but low overall returns: underneath a certain threshold of scale and capital investment, the business consumes more funds and labour than it yields.113 To scale up something like terra sigillata production, access to significantly extended markets and distribution networks was necessary, as were investments in levigation tanks and other production infrastructure. The opus reticulatum building represents an investment in infrastructure, and initial assessment of distribution patterns suggests that later Marzuolo products increasingly targeted a longer-distance market over local consumers.114 Further exacerbating such an economic reality in the Roman world was a structure of profit-making based on windfalls: an occasional good harvest or an occasional fortuitous sale — expected or hoped for, but unpredictable — especially when the majority of other producers had experienced a bad year.115 The windfall was predicated on micro-ecologies, which meant that one field could do well while the next did not; on fragmented transport and communication, which prevented balancing out of supply and demand across regions; and on volatile prices, which differed between areas.116 The vagaries of agriculture had a knock-on effect on the funds available for consumption goods such as terra sigillata pots. The difference between small producers and large investors was not one of opportunity, nor of risk-taking, but of scale.117 Opportunity costs were low for the small producer to the extent that surplus labour was available, as argued earlier, but so were the potential profits to be made from pursuing an opportunity.118 Local innovation was thus bound to reach a dead end, or to be scooped up by the outside investment of large landowners or entrepreneurs, who would have been concerned with scaling up the activities and their reach, not with launching them or with experimentation. This chimes with Fülle’s reading of large landowners investing in terra sigillata production, as expanding an existing pottery.119 In other words, small producers were actively innovating in an attempt to break free of structural constraints, but were caught in a scalar bind, while large landowners had no need for innovation to bridge what was already a direct link between investment-as-expansion and profit increase. VI conclusion The site of Marzuolo urges economic historians to (re-)consider the link between investment and innovation. Both Finley and his opponents ascribed innovation to investment by large landowners, debating their attitudes towards risk, productivity and status, and a priori leaving the vast majority of the ancient world out of the picture altogether. Combined with archaeology’s long history of privileging elite loci of towns and villas for excavation, innovation was reconstructed as a trickle-down process, from innovating landowners to imitating smallholders, artisans or peasants. By unpacking innovation into its constituent components — including experimentation, human capital development, foresight, planning and labour — this article, instead, has turned the trickle-down process upside down. It locates the innovative potential with the small agriculturalist/artisan: a mix of seasonal and generational labour surplus and low capital was conducive to diversification, intensification, tinkering and experimentation. But from the evidence at Marzuolo also surfaces a misalignment between incentives for innovation and means of investment; between development of human capital and provision of infrastructure, clays, and so on — a structural weakness that would have curbed innovation in the Roman rural economy. Smallholders and small producers had greater, but in the end unsustainable, manoeuvre space than typically granted them in current models: they were caught in a scalar bind, and ultimately their labour and human capital development fell prey to external investors, who simply sought to plug into networks with guaranteed demand and stable products.120 In contrast to today’s ‘start-up economy’ that many readers will implicitly associate with innovation, outside investors in a predatory supply chain as proposed for Marzuolo expected products with defined templates, thus solidifying121 but also petrifying networks of labour, skill and demand. This article has led the debate about innovation back to a question of labour and its organization, which was one part of Finley’s two-pronged account downplaying ancient innovation (the other part invoking elite mentalities). As such, Finley’s emphasis on how the ready availability of cheap slave labour obviated any need for technical innovation on the part of landowners is made to confront the continued existence of smallholders — and probably of ad hoc and hired labour — as brought to light by increasing archaeological fieldwork.122 In the area around Marzuolo, smallholders (free or dependent) provided both a labour supply and the main consumption market — and thus held at least indirect bargaining power in the rural economy. They were, and as suggested by the evidence of Marzuolo, remained constrained, however, in their access to capital investment, including land, raw materials, fuel, and so on. As external investors appropriated or provided the means of production, they separated them from the labour, thus curtailing the open-ended process of tinkering that characterized low capital/high labour investment and innovation. A glimpse of the resulting tension can be seen in the later phase of the opus reticulatum building at Marzuolo, where craftsmen who worked in (and possibly leased) this space had to negotiate its original design, resulting in ad hoc modifications. Can we extrapolate the narrative of bottom-up innovation followed by predation to other sites in Roman Italy, or beyond? In other words, how unique was Marzuolo? Intensification of surface survey in rural Roman Italy shows that sites like Marzuolo — by whatever label we choose to identify them, ‘small towns’, ‘minor centres’, or ‘nucleated centres’ — were widespread,123 much like in Gaul and the north-western provinces.124 Already surface survey and geophysical imaging are bringing to light a wide variety in size, activities, duration and central place functions within this heterogeneous class of settlements.125 In order for this more diverse landscape to reframe current historical narratives centred on towns and villas, however, there is a real need for excavation and its contextual data. The Roman countryside was always in flux and needs to be studied not just diachronically but at a sufficiently high resolution to detect short-term changes and to write micro-histories. In particular, this article has argued that new resources could not only be cultivated and launched but could also be re-appropriated within a short temporal horizon. Yet already we can list some distinctive elements about Marzuolo: the rapid and dramatic scaling up both of the site and of its craft production; its liminal position in between a landscape of villas and one of smallholders; its location in a corridor connecting the main transport networks, from which it was at a remove. Further research on other sites should allow comparative evaluation of these and other parameters. For now, it is clear that predation is an alternative model to direct investment, but it is not the all-encompassing model. For most first-century Adterra sigillata production sites in central Italy mentioned above, direct investment is a good fit. Meanwhile, for other types of sites such as the town of Sagalassos in Turkey, with a thriving ceramic fine-ware production that expanded through horizontal multiplication, the role of third-party investment has recently been minimized.126 In other words, escaping the claws of external investment seems to have been particularly difficult for smallholders and artisans in early imperial rural Roman Italy, for better (possibly greater security) or for worse (lesser impetus for innovation). Moreover, the disjuncture between capital investment and human capital development identified in this article can throw new light on the question of more capital-intensive innovations — back to the watermill. It urges us to consider innovation as an open-ended and situated process, even in cases that less obviously run counter to the standard model of direct investment. And it prompts us to search for other low-capital innovations (such as cultivation methods) in which smallholders could have played more of a part than hitherto acknowledged.127 By recuperating the bottom-up agency of small farmers and artisans, the predation model also better aligns a narrative of the Roman rural economy with that of later periods. Dagfinn Skre, for instance, reads the installation of craftsmen in eighth-century emporia as their escape from the remit of landowners, but also as their loss of landowners’ protection, which pushed them to embrace intensification and innovation.128 In the high Middle Ages, too, small farmers and artisans are increasingly seen as actively innovating — the difference between making a living and making money was never clear-cut — and lords as passively reaping the fruits.129 Finally, debates about both the Roman and the medieval economy revolve in large part around questions of connectivity. In the Roman case, rebuttals of Finley’s primitivism emphasize not just the scale of trade, but also its reach, with an emphasis on long-distance networks. The importance of long-distance trade, fuelled by elite demand, equally underwrites one prominent model of the early medieval economy.130 The alternative, instead, highlights the regionalism of the early medieval world’s agricultural underpinnings, picturing less connected and more independent peasants.131 This article by no means claims to solve these debates, or even to align with one or the other side. Instead, the evidence from Marzuolo encourages accounting for how connectivity is made and maintained. How can networks — in this case, networks of terra sigillata pots, their buyers, their producers, their investors, and the relevant skills and knowledge — be created and maintained that are both regionally anchored and have potential supra-regional reach? How was a patchwork132 made into a network? What was lost and gained along the way, and for whom? Footnotes 1 M. I. Finley, ‘Technical Innovation and Economic Progress in the Ancient World’, Economic History Review, xviii (1965). 2 Ibid., 35–6. 3 Kevin Greene, ‘Technological Innovation and Economic Progress in the Ancient World: M. I. Finley Reconsidered’, Economic History Review, liii (2000); Jean-Pierre Brun, Marc Borréani and Jean-Louis Guendon, ‘Deux moulins hydrauliques du Haut-Empire romain en Narbonnaise: Villae des Mesclans à la Crau et de Saint-Pierre/Les Laurons aux Arcs (Var)’, Gallia, lv (1998); Elio Lo Cascio (ed.), Innovazione tecnica e progresso economico nel mondo romano (Bari, 2006); Andrew Wilson, ‘Machines, Power and the Ancient Economy’, Journal of Roman Studies, xcii (2002); Örjan Wikander, ‘The Water-Mill’, in Örjan Wikander (ed.), Handbook of Ancient Water Technology (Leiden, 2000). 4 Finley, ‘Technical Innovation and Economic Progress in the Ancient World’, 41. 5 Andrew Wilson, ‘Approaches to Quantifying Roman Trade’, in Alan Bowman and Andrew Wilson (eds.), Quantifying the Roman Economy: Methods and Problems (Oxford, 2009). 6 Kimberly Bowes et al., The Roman Peasant Project 2009–2014: Excavating the Roman Rural Poor (Philadelphia, forthcoming). 7 Ninina Cuomo di Caprio, La ceramica in archeologia: Antiche tecniche di lavorazione e moderni metodi d’indagine (Rome, 1985), 135–48. 8 Emanuele Vaccaro, Claudio Capelli and Mariaelena Ghisleni, ‘Italic Sigillata Production and Trade in Rural Central Italy: New Data from the Project “Excavating the Roman Peasant” ’, in Tymon C. A. de Haas and Gijs W. Tol (eds.), The Economic Integration of Roman Italy: Rural Communities in a Globalizing World (Leiden, 2017), 238–41. 9 Kimberly Bowes et al., ‘Cinigiano (GR). Il sito di Marzuolo: Quarto anno del Roman Peasant Project’, Notiziario della Soprintendenza per i Beni Archeologici della Toscana, ix (2013), 602; Vaccaro, Capelli and Ghisleni, ‘Italic Sigillata Production and Trade in Rural Central Italy’. 10 Bryan Pfaffenberger, ‘Social Anthropology of Technology’, Annual Review of Anthropology, xxi (1992); Astrid Van Oyen, How Things Make History: The Roman Empire and Its Terra Sigillata Pottery (Amsterdam, 2016). 11 Compare for building techniques, Hélène Dessales, ‘Les savoir-faire des maçons romains, entre connaissance technique et disponibilité des matériaux: Le cas pompéien’, in Nicolas Monteix and Nicolas Tran (eds.), Les savoirs professionnels des gens de métier: Études sur le monde du travail dans les sociétés urbaines de l’empire romain (Collection du Centre Jean Bérard, xxxvii, Naples, 2011). See also Richard P. Saller, ‘Human Capital and Economic Growth’, in Walter Scheidel (ed.), The Cambridge Companion to the Roman Economy (Cambridge, 2012). 12 W. Brian Arthur, The Nature of Technology: What It Is and How It Evolves (New York, 2009); Everett M. Rogers, Diffusion of Innovations, 5th edn (New York, 2003). 13 Finley, ‘Technical Innovation and Economic Progress in the Ancient World’, 31, 42. 14 Andrew Wilson, ‘Large-Scale Manufacturing, Standardization, and Trade’, in John Peter Oleson (ed.), The Oxford Handbook of Engineering and Technology in the Classical World (Oxford, 2008), 396–7. 15 Annalisa Marzano, Roman Villas in Central Italy: A Social and Economic History (Columbia Studies in the Classical Tradition, xxx Leiden, 2006); Stephen L. Dyson, Community and Society in Roman Italy (Baltimore, 1992); Nicholas Purcell, ‘The Roman Villa and the Landscape of Production’, in Tim Cornell and Kathryn Lomas (eds.), Urban Society in Roman Italy (London, 1995); Nicola Terrenato, ‘The Auditorium Site in Rome and the Origins of the Villa’, Journal of Roman Archaeology, xiv (2001). 16 J. Geoffrey Kron, ‘The Diversification and Intensification of Roman Agriculture: The Complementary Roles of the Small and Wealthy Farmer’, in De Haas and Tol (eds.), The Economic Integration of Roman Italy; Walter Scheidel, ‘Demography’, in Walter Scheidel, Ian Morris and Richard Saller (eds.), The Cambridge Economic History of the Greco-Roman World (Cambridge, 2007). 17 Gil J. Stein, ‘From Passive Periphery to Active Agents: Emerging Perspectives in the Archaeology of Interregional Interaction’, American Anthropologist, civ (2002). See also Clark L. Erickson, ‘Intensification, Political Economy, and the Farming Community: In Defense of a Bottom-Up Perspective of the Past’, in Joyce Marcus and Charles Stanish (eds.), Agricultural Strategies (Los Angeles, 2006), 336, 338. 18 Adam Lucas, Wind, Water, Work: Ancient and Medieval Milling Technology (Leiden, 2006), 46; Stéphane Mauné and Jean-Louis Paillet, ‘Stockage et transformation des céréales dans l’économie rurale de Gaule Narbonnaise (Ier–IIIe siècle apr. J.-C.): L’exemple des moulins hydrauliques de Vareilles et de l’Auribelle-Basse (Hérault)’, in Patricia C. Anderson et al. (eds.), Le traitement des récoltes: Un regard sur la diversité du néolithique au présent. Actes des XXIII e Rencontres internationales d’archéologie et d’histoire d’Antibes, 17–19 octobre 2002 (Antibes, 2003); Wikander, ‘Water-Mill’. On innovations in building techniques, see Janet DeLaine, ‘The Cost of Creation: Technology at the Service of Construction’, in Elio Lo Cascio (ed.), Innovazione tecnica e progresso economico nel mondo romano, 240. 19 Astrid Van Oyen, The Socio-Economics of Roman Storage: Agriculture, Trade, and Family (Cambridge, forthcoming). 20 Astrid Van Oyen, ‘The Roman City as Articulated through Terra Sigillata’, Oxford Journal of Archaeology, xxxiv (2015). 21 Giuseppe Pucci, La fornace di Umbricio Cordo: L’officina di un ceramista romano e il territorio di Torrita di Siena nell’antichità (Florence, 1992). 22 Stefano Campana, Carta archeologica della provincia di Siena, xii, Montalcino (Siena, 2013), 236, 280. 23 Margherita Bergamini (ed.), Scoppieto III: Lo scavo, le strutture, i materiali (coroplastica, marmi) (Rome, 2013). 24 Peter Attema, ‘Landscape Archaeology in Italy: Past Questions, Current State and Future Directions’, in De Haas and Tol (eds.), The Economic Integration of Roman Italy. 25 J. Geoffrey Kron, ‘The Much Maligned Peasant: Comparative Perspectives on the Productivity of the Small Farmer in Classical Antiquity’, in Luuk de Ligt and Simon Northwood (eds.), People, Land, and Politics: Demographic Developments and the Transformation of Roman Italy, 300 Bc–ad 14 (Mnemosyne, ccciii, Leiden, 2008); Neville Morley, ‘The Transformation of Italy, 225–28 Bc’, Journal of Roman Studies, xci (2001); Robert Witcher, ‘Agricultural Production in Roman Italy’, in Alison E. Cooley (ed.), A Companion to Roman Italy (London, 2016). 26 Paul Erdkamp, ‘Agriculture, Underemployment, and the Cost of Rural Labour in the Roman World’, Classical Quarterly, xlix (1999); Alessandro Launaro, Peasants and Slaves: The Rural Population of Roman Italy (200 Bc to Ad 100) (Cambridge, 2011). 27 Bowes et al., Roman Peasant Project; Emanuele Vaccaro et al. ‘Excavating the Roman Peasant II: Excavations at Case Nuove, Cinigiano (GR)’, Papers of the British School at Rome, lxxxi (2013). See also Dominic Rathbone, ‘Poor Peasants and Silent Sherds’, in De Ligt and Northwood (eds.), People, Land, and Politics; Gijs Tol et al., ‘Minor Centres in the Pontine Plain: The Cases of “Forum Appii” and “Ad Medias” ’, Papers of the British School at Rome, lxxxii (2014). 28 Cam Grey et al., ‘Familiarity, Repetition, and Quotidian Movement in Roman Tuscany’, Journal of Mediterranean Archaeology, xxviii (2015), 200. 29 For Roselle, see Carlo Citter, ‘Il territorio di Roselle-Grosseto: Occupazione del suolo e forme insediative dalla preistoria al medioevo’, in Carlo Citter (ed.), Grosseto, Roselle e il Prile: Note per la storia di una città e del territorio circostante (Mantova, 1996). For Paganico, see Gabriella Barbieri, ‘Aspetti del popolamento della media valle dell’Ombrone nell’antichità: Indagini recenti nel territorio di Civitella Paganico’, Journal of Ancient Topography, xv (2005). 30 For Monte Amiata, see Franco Cambi, Carta archeologica della provincia di Siena, ii, Il Monte Amiata (Abbadia San Salvatore) (Siena, 1996), 169–71. For Cinigiano, see Mariaelena Ghisleni, ‘Carta archeologica della provincia di Grosseto: Comune di Cinigiano. Dinamiche insediative e di potere fra V e XI secolo nella bassa val d’Orcia e nella media valle dell’Ombrone’ (Univ. of Siena Ph.D. thesis, 2010), 80. See also Emanuele Vaccaro et al., ‘Maglie insediative della valle dell’Ombrone (GR) nel primo millennio d.C.’, in Giancarlo Macchi Janica (ed.), Geografie del popolamento: Casi di studio, metodi e teorie (Siena, 2009), 294. 31 Cambi, Monte Amiata, 171–3. 32 Alain Ferdière, ‘Des maîtres de domaines investissent dans la manufacture: Fundus et production artisanale en Gaule romaine’, AGER Bulletin de Liaison, xvii (2017); Philippe Leveau, ‘The Western Provinces’, in Scheidel, Morris and Saller (eds.), Cambridge Economic History of the Greco-Roman World, 654–5. In the later Roman period, rural craft sites in Italy are sometimes located on church estates, for example Mola di Monte Gelato — see T. W. Potter and A. C. King, Excavations at the Mola di Monte Gelato: A Roman and Medieval Settlement in South Etruria (Rome, 1997); and San Vincenzo al Volturno — see Richard Hodges (ed.), San Vincenzo al Volturno 1: The 1980–86 Excavations, Part I (Archaeological Monographs of the British School at Rome, vii, London, 1993). 33 For Santa Marta, see Stefano Campana et al., ‘Cinigiano (GR). Poggi del Sasso, località Santa Marta: Indagini 2015 (concesione di scavo)’, Notiziario della Soprintendenza per i Beni Archeologici della Toscana, xi (2015); Ghisleni, ‘Carta archeologica della provincia di Grosseto’. For Dogana and Cannicci, see Barbieri, ‘Aspetti del popolamento della media valle dell’Ombrone nell’antichità’, 126–7 and 122. For Sesta, see Campana, Montalcino, 236, 280. 34 Marc Bloch, ‘Avènement et conquêtes du moulin à eau’, Annales d’histoire économique et sociale, vii (1935); Adam Robert Lucas, ‘Industrial Milling in the Ancient and Medieval Worlds: A Survey of the Evidence for an Industrial Revolution in Medieval Europe’, Technology and Culture, xlvi (2005). See also Peregrine Horden and Nicholas Purcell, The Corrupting Sea: A Study of Mediterranean History (Oxford, 2000), 288. 35 On ‘multi-centered and global’ technological change in the late eighteenth and nineteenth centuries, see, for example, C. A. Bayly, The Birth of the Modern World 1780–1914: Global Connections and Comparisons (Oxford, 2004), 174. For early medieval studies, see section VI. 36 Saller, ‘Human Capital and Economic Growth’. 37 Margaret M. Blair, ‘An Economic Perspective on the Notion of “Human Capital” ’, in Alan Burton-Jones and J.-C. Spender (eds.), The Oxford Handbook of Human Capital (Oxford, 2011); Mark Blaug, The Economics of Education and the Education of an Economist (New York, 1987). 38 Tim Sorg, ‘Imperial Neighbors: Empires and Land Allotment in the Ancient Mediterranean World’ (Cornell Univ. Ph.D. thesis, 2018), 49. 39 Vaccaro, Capelli and Ghisleni, ‘Italic Sigillata Production and Trade in Rural Central Italy’, 238. 40 Bowes et al., ‘Cinigiano (GR). Il sito di Marzuolo’, 604. 41 Kimberly Bowes, Mariaelena Ghisleni and Emanuele Vaccaro, ‘Cinigiano (GR). Il sito di Marzuolo, seconda campagna di scavo: Quinto anno del Roman Peasant Project’, Notiziario della Soprintendenza per i Beni Archeologici della Toscana, x (2014), 496–7. 42 Vaccaro, Capelli and Ghisleni, ‘Italic Sigillata Production and Trade in Rural Central Italy’, 241. Some of the holes were filled with clay, which might suggest a role in clay preparation. 43 Carl Knappett and Sander van der Leeuw, ‘A Developmental Approach to Ancient Innovation: The Potter’s Wheel in the Bronze Age East Mediterranean’, Pragmatics and Cognition, xxii (2014); Andy Clark and David Chalmers, ‘The Extended Mind’, Analysis, lviii (1998); Arthur, Nature of Technology. 44 Philip M. Kenrick, August Oxé and Howard Comfort, Corpus Vasorum Arretinorum: A Catalogue of the Signatures, Shapes and Chronology of Italian Sigillata (Bonn, 2000), 407–8, nos. 1958, 1961. 45 Giuseppe Pucci, ‘I bolli sulla terra sigillata: Fra epigrafia e storia economica’, in William V. Harris (ed.), The Inscribed Economy: Production and Distribution in the Roman Empire in the Light of Instrumentum Domesticum (Journal of Roman Archaeology Supplementary Series, 6, Ann Arbor, 1993). 46 Gunnar Fülle, ‘The Internal Organization of the Arretine Terra Sigillata Industry: Problems of Evidence and Interpretation’, Journal of Roman Studies, lxxxvii (1997); Brian R. Hartley and Brenda M. Dickinson, Names on Terra Sigillata: An Index of Makers’ Stamps and Signatures on Gallo-Roman Terra Sigillata (Samian Ware), 9 vols. (London, 2008–12), i. 47 Pieter Seuneke, Thomas Lans and Johannes S. C. Wiskerke, ‘Moving Beyond Entrepreneurial Skills: Key Factors Driving Entrepreneurial Learning in Multifunctional Agriculture’, Journal of Rural Studies, xxxii (2013), 211–3. 48 Vaccaro, Capelli and Ghisleni, ‘Italic Sigillata Production and Trade in Rural Central Italy’, 247; Pucci, La fornace di Umbricio Cordo, 114 (no. 10); Kenrick, Oxé and Comfort, Corpus Vasorum Arretinorum, 272 (nos. 1099–1101). See also Fülle, ‘Internal Organization of the Arretine Terra Sigillata Industry’; Philip M. Kenrick, ‘Signatures on Italian Sigillata: A New Perspective’, in Jeroen Poblome et al. (eds.), Early Italian Sigillata: The Chronological Framework and Trade Patterns (Leuven, 2004). 49 Kenrick, Oxé and Comfort, Corpus Vasorum Arretinorum, 494–5 (no. 2470). 50 Opus reticulatum masonry was used in a bath complex of early imperial date at the site of Pietratonda (Paganico), c.20 km west of Marzuolo, see Barbieri, ‘Aspetti del popolamento della media valle dell’Ombrone nell’antichità’, 130. On problems with the term ‘quasi reticulatum’, see Lynne C. Lancaster, ‘Roman Engineering and Construction’, in John Peter Oleson (ed.), The Oxford Handbook of Technology and Engineering in the Classical World (Oxford, 2009), 262. 51 G. D. B. Jones, ‘Capena and the Ager Capenas: Part II’, Papers of the British School at Rome, xxxi (1963), 147–58; Rathbone, ‘Poor Peasants and Silent Sherds’. But see Lin Foxhall, ‘The Dependent Tenant: Land Leasing and Labour in Italy and Greece’, Journal of Roman Studies, lxxx (1990), 109–10. 52 For slave labour, see Filippo Coarelli, ‘Public Building in Rome between the Second Punic War and Sulla’, Papers of the British School at Rome, xlv (1977), 18; Mario Torelli, ‘Innovazioni nelle tecniche edilizie romane tra il I sec. a.C. e il I sec. d.C.’, in Tecnologia, economia e società nel mondo romano: Atti del Convegno di Como, 27–29 settembre 1979 (Como, 1980), 155. But such ‘prefab’ building did not necessarily reduce overall cost, see Janet DeLaine, ‘Bricks and Mortar: Exploring the Economics of Building Techniques at Rome and Ostia’, in David J. Mattingly and John Salmon (eds.), Economies Beyond Agriculture in the Classical World (London, 2001). For (seasonal) labour from small farmers, see Erdkamp, ‘Agriculture, Underemployment, and the Cost of Rural Labour in the Roman World’. 53 Antonia Arnoldus-Huyzendveld in Bowes et al., Roman Peasant Project. 54 DeLaine, ‘Bricks and Mortar’, 236. 55 Bowes, Ghisleni and Vaccaro, ‘Cinigiano (GR). Il sito di Marzuolo, seconda campagna di scavo’, 494–6. 56 Residue analysis of the mortar of the cylindrical tanks is ongoing. 57 Robert Thomas and Andrew Wilson, ‘Water Supply for Roman Farms in Latium and South Etruria’, Papers of the British School at Rome, lxii (1994), 140. 58 Rhodora G. Vennarucci, Astrid Van Oyen and Gijs W. Tol. ‘Cinigiano (GR). Una comunità artigianale nella Toscana rurale: Il sito di Marzuolo’, in Valentino Nizzo and Antonio Pizzo (eds.), Antico e non antico: Studi multidisciplinari offerti a Giuseppe Pucci (Sesto San Giovanni, 2018), 592. 59 Botanical evidence suggests a shift in fuel too in the second production phase, from locally available oak in the experimental production phase to beech and fir, available c.10 km south-east of Marzuolo on Monte Amiata. See Anna Maria Mercuri, Eleonora Rattighieri and Rossella Rinaldi in Bowes et al., Roman Peasant Project. 60 N. S. Müller et al., ‘Tinkering with Terra Sigillata: The Case of Marzuolo, a Roman Rural Multi-Craft Site’, European Meeting of Ancient Ceramics (Barcelona, 2019). 61 Both the epigraphic homogeneity and the comparative rarity of this stamp are indicative of short-term, self-contained investment in construction of this building. See Marzano, Roman Villas in Central Italy, 66–7. 62 Fülle, ‘Internal Organization of the Arretine Terra Sigillata Industry’. But see Mara Sternini, ‘Italian Sigillata: Some Insights into a Complex Reality’, HEROM: Journal on Hellenistic and Roman Material Culture, v (2016). 63 Pucci, ‘I bolli sulla terra sigillata’, 78; Karl Strobel, ‘Produktions- und Arbeitsverhältnisse in der südgallischen Sigillataindustrie: Zu Frage der Massenproduktion in der römischen Kaiserzeit’, Specimina ova Universitatis Quinquecclesiensis, viii (1992). 64 Sternini, ‘Italian Sigillata’. 65 Koenraad Verboven, ‘Mentalité et commerce: Le cas des negotiatores et negotia habentes’, in Jean Andreau, Jérôme France and Sylvie Pittia (eds.), Mentalités et choix économiques des Romains (Scripta Antiqua, vii, Pessac, 2004); John H. D’Arms, Commerce and Social Standing in Ancient Rome (Cambridge, Mass., 1981). 66 Dominic Rathbone, Economic Rationalism and Rural Society in Third-Century Ad Egypt (Cambridge, 1991). 67 Dennis P. Kehoe, ‘Investment in Estates by Upper-Class Landowners in Early Imperial Italy: The Case of Pliny the Younger’, in H. Sancisi-Weerdenburg et al. (eds.), De Agricultura: In Memoriam Pieter Willem De Neeve (Amsterdam, 1993); Dennis P. Kehoe, Investment, Profit, and Tenancy: The Jurists and the Roman Agrarian Economy (Ann Arbor, 1997); Dennis P. Kehoe, Law and the Rural Economy in the Roman Empire (Ann Arbor, 2007). 68 Jeroen Poblome, ‘Money Makes Pottery Go Round’, in Jeroen Poblome (ed.), Exempli Gratia: Sagalassos, Marc Waelkens, and Interdisciplinary Archaeology (Leuven, 2013); Jeroen Poblome, ‘The Potters of Ancient Sagalassos Revisited’, in Andrew Wilson and Miko Flohr (eds.), Urban Craftsmen and Traders in the Roman World (Oxford, 2016). 69 For example, at Settefinestre in the first century Ad (Carandini, Settefinestre); Tymon de Haas, Gijs Tol and Peter Attema, ‘Investing in the Colonia and Ager of Antium’, Facta, v (2011). 70 Verboven, ‘Mentalité et commerce’, 188. 71 Annalisa Marzano, ‘Agricultural Production in the Hinterland of Rome: Wine and Olive Oil’, in Alan K. Bowman and Andrew Wilson (eds.), The Roman Agricultural Economy: Organization, Investment, and Production (Oxford, 2013); Annalisa Marzano, ‘Capital Investment and Agriculture: Multi-Press Facilities from Gaul, the Iberian Peninsula, and the Black Sea Region’, in Bowman and Wilson (eds.), Roman Agricultural Economy. 72 For example Pardigon (Var, southern France), Gaëtan Congès and Pascal Lecacheur, ‘Exploitation et domaine sur la côte varoise à l’époque romaine: Exemple de la plaine de Pardigon (Cavalaire, Croix-Valmer, Var)’, in François Favory and Jean-Luc Fiches (eds.), Les campagnes de la France méditerranéenne dans l’Antiquité et le haut Moyen Age: Études microrégionales (Documents d’Archéologie Française, xlii, Paris, 1994). 73 See examples of failure cited in Simon Swain, Economy, Family, and Society from Rome to Islam: A Critical Edition, English Translation, and Study of Bryson’s Management of the Estate (Cambridge, 2013), 165. 74 Columella, De Re Rustica, iii.8. 75 Plutarch, Life of Cato the Elder, xxi.5–8. 76 See also Giuseppe Pucci, ‘La produzione della ceramica aretina: Note sull’ “industria” nella prima età imperiale romana’, Dialoghi di archeologia, vii (1973). 77 Harri Kiiskinen, ‘Production and Trade of Etrurian Terra Sigillata Pottery in Roman Etruria and Beyond between c.50 Bce and c.150 Ce’ (Univ. of Turku Ph.D. thesis, 2013), 54–7, proposes that the origin of terra sigillata production at Arezzo is to be linked to a strategic preference of investment in human capital over land during a period of Sullan land confiscations. It is unclear whether he imagines direct ownership (slaves) or leases for the development and operationalization of this human capital and where experimentation fits in. No such evidence for confiscations exists for the region of Marzuolo. 78 Helen Cockle, ‘Pottery Manufacture in Roman Egypt: A New Papyrus’, Journal of Roman Studies, lxxi (1981); Joachim Hengstl, ‘Einige juristische Bemerkungen zu drei Töpferei-Mieturkunden’, in Studi in Onore di Arnaldo Biscardi, iv (Milan, 1983). 79 Translation by Cockle, ‘Pottery Manufacture in Roman Egypt’, 90. 80 Ibid. 81 Morris Silver, ‘Glimpses of Vertical Integration/Disintegration in Ancient Rome’, Ancient Society, xxxix (2009). For a material culture-based view, see Van Oyen, How Things Make History, 55–7. 82 Fülle, ‘Internal Organization of the Arretine Terra Sigillata Industry’, 143. 83 Erdkamp, ‘Agriculture, Underemployment, and the Cost of Rural Labour in the Roman World’, 556–7; Paul Erdkamp, ‘Economic Growth in the Roman Mediterranean World: An Early Good-Bye to Malthus?’, Explorations in Economic History, lx (2016). 84 Different from farmers relying on craft work for their very livelihood, see Peter Garnsey, ‘Non-Slave Labour in the Roman World’, in Peter Garnsey (ed.), Non-Slave Labour in the Greco-Roman World (Cambridge, 1980). 85 Hamish Forbes, ‘ “We Have a Little of Everything”: The Ecological Basis of Some Agricultural Practices in Methana, Trizinia’, Annals of the New York Academy of Sciences, cclxviii (1976); Thomas W. Gallant, Risk and Survival in Ancient Greece: Reconstructing the Rural Domestic Economy (Cambridge, 1991); Peter Garnsey, Famine and Food Supply in the Graeco-Roman World: Responses to Risk and Crisis (Cambridge, 1988); Scheidel, ‘Demography’, 57. 86 On ‘adaptability of labour’, see Horden and Purcell, Corrupting Sea, 268–9. Note that the argument that craft production for the market was a response to taxation does not work for Roman Italy, which was exempt of taxes: see Keith Hopkins, ‘Taxes and Trade in the Roman Empire (200 Bc−Ad 400)’, Journal of Roman Studies, lxx (1980). 87 Maaike Groot et al., ‘Surplus Production for the Market? The Agrarian Economy in the Non-Villa Landscapes of Germania Inferior’, Journal of Roman Archaeology, xxii (2009); Evi Margaritis, ‘Agricultural Production and Domestic Activities in Rural Hellenistic Greece’, in Edward M. Harris, David M. Lewis and Mark Woolmer (eds.), The Ancient Greek Economy: Markets, Households and City-States (Cambridge, 2016). 88 Kron, ‘Much Maligned Peasant’; Kimberly Bowes et al., ‘Palaeoenvironment and Land Use of Roman Peasant Farmhouses in Southern Tuscany,’ Plant Biosystems, cxlix (2015); Kimberly Bowes et al., ‘Peasant Agricultural Strategies in Southern Tuscany: Convertible Agriculture and the Importance of Pasture’, in De Haas and Tol (eds.), The Economic Integration of Roman Italy. But see Annalisa Marzano, ‘Villas as Instigators and Indicators of Economic Growth’, in Paul Erdkamp and Koenraad Verboven (eds.), Structure and Performance in the Roman Economy: Models, Methods and Case Studies (Collection Latomus, cccl, Brussels, 2015). 89 Paul Halstead, Two Oxen Ahead: Pre-Mechanized Farming in the Mediterranean (Chichester, 2014). 90 Jan Douwe van der Ploeg, ‘Revitalizing Agriculture: Farming Economically as Starting Ground for Rural Development’, Sociologia Ruralis, xl (2000). 91 For the moral economy tradition, see James C. Scott, The Moral Economy of the Peasant: Rebellion and Subsistence in Southeast Asia (New Haven, 1976); Eric R. Wolf, Peasants (Englewood Cliffs, 1966). For alternative views, see Samuel L. Popkin, The Rational Peasant: The Political Economy of Rural Society in Vietnam (Berkeley, 1979); Teodor Shanin, ‘Expoliary Economies: A Political Economy of Margins’, Journal of Historical Sociology, i (1988); Robert McC. Netting, Smallholders, Householders: Farm Families and the Ecology of Intensive, Sustainable Agriculture (Stanford, 1993). 92 Jason Antrosio and Rudi Colloredo-Mansfeld, ‘Risk-Seeking Peasants, Excessive Artisans: Speculation in the Northern Andes’, Economic Anthropology, i (2014). 93 Patricia Torres Mejia, Peasants, Merchants, and Politicians in Tobacco Production: Philippine Social Relations in a Global Economy (Manila, 2000). 94 Dennis P. Kehoe, Management and Investment on Estates in Roman Egypt during the Early Empire (Bonn, 1992); Lisa Lodwick, ‘Arable Farming, Plant Foods, and Resources’, in Martyn Allen et al. (eds.), The Rural Economy of Roman Britain (London, 2017). 95 Stephen L. Dyson, ‘Is There a Text in This Site?’, in David B. Small (ed.), Methods in the Mediterranean: Historical and Archaeological Views on Text and Archaeology (Mnemosyne, cxxxv, Leiden, 1995). 96 See also Frans Theuws, ‘River-Based Trade Centres in Early Medieval Northwestern Europe: Some “Reactionary” Thoughts’, in Sauro Gelichi and Richard Hodges (eds.), From One Sea to Another: Trading Places in the European and Mediterranean Early Middle Ages (Proceedings of the International Conference, Comacchio, 27th–29th March 2009, Turnhout, 2012), 30. But see Chris Wickham, Framing the Early Middle Ages: Europe and the Mediterranean 400–800 (Oxford, 2005), 706–7. For the later Middle Ages, see Christopher Dyer, ‘Did Peasants Need Markets and Towns? The Experience of Late Medieval England’, in Matthew Davies and James A. Galloway (eds.), London and Beyond: Essays in Honour of Derek Keene (London, 2012). 97 William V. Harris, Rome in Etruria and Umbria (Oxford, 1971). 98 On roadside settlements indicative of an increase in traffic north of Rome, see Paul Johnson, Simon Keay and Martin Millett, ‘Lesser Urban Sites in the Tiber Valley: Baccanae, Forum Cassii and Castellum Amerinum’, Papers of the British School at Rome, lxxii (2005), 95–6. 99 Erdkamp, ‘Economic Growth in the Roman Mediterranean World’; Seuneke, Lans and Wiskerke, ‘Moving beyond Entrepreneurial Skills’, 215; Anne Moxnes Jervell, ‘The Family Farm as a Premise for Entrepreneurship’, in Gry Agnete Alsos et al. (eds.), The Handbook of Research on Entrepreneurship in Agriculture and Rural Development (Northampton, Mass., 2011), 69. 100 And other consumer items, see Bowes et al., Roman Peasant Project. 101 Cam Grey, Constructing Communities in the Late Roman Countryside (Cambridge, 2011), 28, 31. 102 See also Dessales, ‘Les savoir-faire des maçons romains, entre connaissance technique et disponibilité des matériaux’, 59–62. 103 Poblome, ‘Money Makes Pottery Go Round’, 89. 104 But see De Ligt, Fairs and Markets in the Roman Empire: Economic and Social Aspects of Periodic Trade in a Pre-Industrial Society (Amsterdam, 1993), 147–8. 105 Such tinkering is acknowledged by Finley, ‘Technical Innovation and Economic Progress in the Ancient World’, 29, as ‘development within the limits of the traditional techniques’ and as ‘continuous improvement’, but is seemingly dismissed as without economic impact. 106 Emanuele Vaccaro in Bowes et al., Roman Peasant Project. 107 The faunal assemblage associated with the second production phase at Marzuolo is more consumption-oriented and indicative of a wealthier diet than that associated with the first production phase, with a high preponderance of work animals (see Michael MacKinnon in Bowes et al., Roman Peasant Project). 108 Van Oyen, How Things Make History, 45–6; Maurice Picon, ‘Les modes de cuisson, les pâtes et les vernis de la Graufesenque: Une mise au point’, in Martine Genin and Alain Vernhet (eds.), Céramiques de la Graufesenque et autres productions d’époque romaine: Nouvelles recherches (Montagnac, 2002). 109 Kron, ‘Much Maligned Peasant’. On innovations trickling down from large construction projects to more modest ones, see DeLaine, ‘Cost of Creation’, 246–7. 110 Anna Lowenhaupt Tsing, The Mushroom at the End of the World: On the Possibility of Life in Capitalist Ruins (Princeton, 2015), 112–14. Compare with the ‘Putting-out System’ already invoked by Pucci to account for terra sigillata production, see Pucci ‘La produzione della ceramica aretina’, 281. 111 Note how this resonates with the leases discussed in section III. 112 David P. S. Peacock, Pottery in the Roman World: An Ethnoarchaeological Approach (London, 1982), 10. On rural credit in late antiquity, see Dennis P. Kehoe, ‘The State and Production in the Roman Agrarian Economy’, in Bowman and Wilson (eds.), Roman Agricultural Economy, 45. On farmers without spare cash: Cicero, In Verrem II.3, 192–200; Michael H. Crawford, ‘Money and Exchange in the Roman World’, Journal of Roman Studies, lx (1970), 44; Christopher Howgego, ‘The Supply and Use of Money in the Roman World 200 Bc to Ad 300’, Journal of Roman Studies, lxxxii (1992), 21. On rural poor in the twentieth-century Mediterranean acquiring land and capital ‘by hard work and wise or lucky decisions’, see Halstead, Two Oxen Ahead, 296. 113 Abhijit V. Banerjee and Esther Duflo, Poor Economics: A Radical Rethinking of the Way to Fight Global Poverty (New York, 2011), 138–9, 214–33. 114 Bowes et al., Roman Peasant Project. 115 Nicholas Purcell, ‘Literate Games: Roman Urban Society and the Game of Alea’, Past and Present, no. 147 (May 1995), 21. On a similar windfall logic among peasants and artisans in the Northern Andes today, see Antrosio and Colloredo-Mansfeld, ‘Risk-Seeking Peasants, Excessive Artisans’; and for peasants more generally, see Teodor Shanin, The Awkward Class: Political Sociology of Peasantry in a Developing Society: Russia 1910–1925 (Oxford, 1972); Popkin, Rational Peasant, 21. 116 Peter F. Bang, The Roman Bazaar: A Comparative Study of Trade and Markets in a Tributary Empire (Cambridge, 2008), 138–40. 117 Paul Halstead, ‘Traditional and Ancient Rural Economy in Mediterranean Europe: Plus ça Change?’, Journal of Hellenic Studies, cvii (1987); Hamish Forbes, ‘Surplus, Storage and Status in a Rural Greek Community’, World Archaeology, xlix (2017), 20. 118 Poblome, ‘Money Makes Pottery Go Round’, 89. 119 Fülle, ‘Internal Organization of the Arretine Terra Sigillata Industry’, 143. 120 Marzano, ‘Villas as Instigators and Indicators of Economic Growth’. 121 And reducing transaction costs, see Elio Lo Cascio, ‘The Roman Principate: The Impact of the Organization of the Empire on Production’, in Elio Lo Cascio and Dominic Rathbone (eds.), Production and Public Powers in Classical Antiquity (Cambridge, 2000). 122 Launaro, Peasants and Slaves; Bowes et al., Roman Peasant Project. 123 Elisabetta Todisco, I vici rurali nel paesaggio dell’Italia romana (Bari, 2011); Sara Santoro, ‘Crafts and Trade in Minor Settlements in North and Central Italy: Reflections on an Ongoing Research Project’, in De Haas and Tol (eds.), The Economic Integration of Roman Italy. 124 Philippe Leveau, ‘Vicus, “agglomération secondaire”: Des mots différents pour une même entité?’, in Christian Cribellier and Alain Ferdière (eds.), Agglomérations secondaires antiques en région Centre: Actes de la Table Ronde d’Orléans, 18–19 novembre 2004 (Tours, 2012); Michel Tarpin, Vici et pagi dans l’Occident romain (Rome, 2003); Martin Millett, ‘Strategies for Roman Small Towns’, in A. E. Brown (ed.), Roman Small Towns in Eastern England and Beyond (Oxford, 1995); Alexander Smith et al., The Rural Settlement of Roman Britain (London, 2016); Jesús Bermejo, ‘Roman Peasant Habitats and Settlement in Central Spain (1st c. Bc−4th c. Ad)’, Journal of Roman Archaeology, xxx (2017), 361–4. 125 Johnson, Keay and Millett, ‘Lesser Urban Sites in the Tiber Valley’; Tol et al., ‘Minor Centres in the Pontine Plain’. 126 Poblome, ‘Potters of Ancient Sagalassos Revisited’, 389–92. 127 Bowes et al., ‘Peasant Agricultural Strategies in Southern Tuscany’. 128 Dagfinn Skre, ‘Markets, Towns and Currencies in Scandinavia c.Ad 200–1000’, in Gelichi and Hodges (eds.), From One Sea to Another. It is generally agreed that at least initially, emporia were not the result of top-down initiation (see Chris Wickham, ‘Comacchio and the Central Mediterranean’, in Gelichi and Hodges (eds.), From One Sea to Another), but their role in craft production is more debated, see Richard Hodges, Dark Age Economics: A New Audit (Bristol, 2012), 61. See also Tim Pestell and Katharina Umschneider (eds.), Markets in Early Medieval Europe: Trading and ‘Productive’ Sites, 650–850 (Bollington, 2003). 129 Chris Dyer, ‘L’industrie rurale en Angleterre des années 1200–1550: Géographie, sociologie et organisation de la production et des marchés’, in J.-M. Minovez, C. Verna and L. Hilaire-Pérez (eds.), Les industries rurales dans l’Europe médiévale et moderne (Toulouse, 2013), 46. 130 Michael McCormick, Origins of the European Economy: Communications and Commerce, Ad 300–900 (Cambridge, 2001). See also Theuws, ‘River-Based Trade Centres in Early Medieval Northwestern Europe’. 131 Wickham, Framing the Early Middle Ages. 132 See also Horden and Purcell, Corrupting Sea. Author notes ∗ For their generosity, collegiality and inspiration I am immensely grateful to the directors of the Roman Peasant Project and to Gijs Tol and Rhodora Vennarucci, my co-directors of the Marzuolo Archaeological Project. For comments on this article at various stages of development, I should like to thank Georgia Andreou, Kim Bowes, Cam Grey, Sturt Manning, Martin Millett and Tim Sorg. Any errors are my own. © The Past and Present Society, Oxford, 2020 This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Innovation and Investment in the Roman Rural Economy Through the Lens of Marzuolo (Tuscany, Italy)
JF - Past & Present
DO - 10.1093/pastj/gtz062
DA - 2020-08-01
UR - https://www.deepdyve.com/lp/oxford-university-press/innovation-and-investment-in-the-roman-rural-economy-through-the-lens-QI7Jf0vqcZ
SP - 3
EP - 40
VL - 248
IS - 1
DP - DeepDyve
ER -