Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

An interdisciplinary approach to agriculture in central and southern Italy 202-103 BC Hoyer, Daniel 2008

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


24-ubc_2008_fall_hoyer_daniel.pdf [ 5.25MB ]
JSON: 24-1.0066628.json
JSON-LD: 24-1.0066628-ld.json
RDF/XML (Pretty): 24-1.0066628-rdf.xml
RDF/JSON: 24-1.0066628-rdf.json
Turtle: 24-1.0066628-turtle.txt
N-Triples: 24-1.0066628-rdf-ntriples.txt
Original Record: 24-1.0066628-source.json
Full Text

Full Text

AN INTERDISCIPLINARY APPROACH TO AGRICULTURE IN CENTRAL AND SOUTHERN ITALY 202-103 BC by Daniel Hoyer B.A., University of Toronto, 2006 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS in The Faculty of Graduate Studies (Classics) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) MAY 2008 © Daniel Hoyer, 2008  Abstract This paper looks at the agrarian economy of Italy in the middle Republican period, specifically between 202 and 103 BC. I am concerned primarily with how land was utilized and by whom it was exploited during this pivotal time, and with the interaction between rural and urban space in the Italian economy. Scholarship on Roman agriculture has largely been, unfortunately, polarized between 'historians' and 'archaeologists,' between a slave-production, latifundia model on the one hand and a focus on micro-regional variation on the other. This staunch division is quite frustrating, as there has been far too little dialogue between the two sides and since the schism works to the detriment of both, for each discipline can, I believe, inform the other to give a more complete picture of ancient life than either one in isolation. It is my intention with this paper, then, to bring an interdisciplinary approach to the study of Roman agriculture, to discover whether any union can be made between these two conflicting models of land use in the second century BC. My research is concentrated on a few areas of south-central Italy, namely south Etruria, especially the ager Cosanus, Pompeii, and parts of the Biferno Valley, as these areas are the richest in terms of available evidence, including the literary, epigraphic, and archaeological sources, and I believe offer the most intriguing possibilities for analysis. Using all of these tools I explore the labour used in agricultural production in these areas, the function of urban space and the preponderance of trade and exchange, and, lastly, look at the inter-regional situation trying to discern whether any 'global' agricultural program for Roman Italy emerges.  ii  Table of Contents ABSTRACT^ TABLE OF CONTENTS ^  ii iii  1. INTRODUCTION ^  4  2. AGRONOMISTS AND THE LATIFUNDIA MODEL ^  9  3. MODERN RECONSTRUCTIONS OF ANCIENT AGRARIAN PRACTICE ^  20  4. ARCHAEOLOGY AND THE MICRO-REGIONAL APPROACH^ 30 5. BIFERNO VALLEY AND THE SAMNITE ECONOMY ^  35  6. SETTLEMENT PATTERNS IN THE AGER COSANUS AND SOUTH ETRURIA ^  46  7. WINE AND GRAIN PRODUCTION IN CAMPANIA ^  61  8. SLAVE LABOUR AND THE GRAIN TRADE IN SICILY ^  72  9. THE ROLE OF PEASANTS IN THE ITALIAN ECONOMY ^ 78 10. CONCLUDING REMARKS AND AREAS FOR FURTHER RESEARCH ^ BIBLIOGRAPHY ^  94 100  iii  Chapter 1 Introduction Of the territory which the Romans won in war from their neighbours, a part they sold, and a part they made common, and assigned to the poor and impoverished citizens for their use, on payment of a small rent into the public treasury. ... Later, the neighbouring rich men, by means of fictitious personages, transferred these plots to themselves, and fmally held most of the land openly in their own names. Then the poor, who had been ejected from their land, no longer seemed eager for military service, and neglected bringing up children, so that soon all of Italy was conscious of the scarcity of freemen, and was filled with gangs of foreign slaves, by whose aid the rich cultivated their estates, having driven away the free citizens (Plut. Life of Tiberius Gracchus.8).  With this brief synopsis of the developments in land ownership in second century BC Italy, Plutarch means to highlight the apoplexy felt by a young Tiberius Gracchus when he saw the 'scarcity of freemen' in Italy on his way to the Spanish frontier. This model of land cultivation, emphasizing large, aristocrat-owned plots and a predominantly servile labor force, prevailed not only in antiquity, but has remained even in modern times. There can be no doubt that Plutarch exaggerated the plight of these free peasants in order to underline the chaotic period which emerged from Tiberius' attempted land redistribution programs. Yet, many scholars nevertheless readily accept the basic premise of his argument. Max Weber, for example, boldly asserts that "from Cato the Elder until Augustus, enterprises based on slave labor grew steadily larger and in fact became the dominant form of economic activity (Weber, 1988, 318);" A.J. Toynbee's influential Hannibal's Legacy too follows this tradition, claiming that peasant small-plot farming was replaced throughout the later centuries BC by "an outlay of capital in the shape of plantations" and the concomitant "gangs of slave plantation-hands (Toynbee 1965, 160)." Historians of the Roman economy have, thus, long employed Plutarch's model as the starting point of investigation, seeking in their accounts mainly to explain how, and possibly why,  4  this great change in Italian agriculture occurred, having usually taken as read that the model itself is valid. It is clear, surely, that the third century BC was a pivotal time in Italian history, which saw the growth of Roman power culminating in the Second Punic War. In the following century, after Hannibal's exit from the peninsula and the (albeit temporary) defeat of Carthage in 202 BC, the city of Rome became a political and cultural center for the entire Mediterranean region, and the Roman state received a great influx of money as well as slave labourers from its many military victories. This latter point is demonstrated unequivocally by the major slave revolts which occurred in Sicily at the end of the century. These were all quite significant events that, by the mid-first century BC, had helped to prompt drastic changes in Roman life, political and social as well as economic. How precisely these events transformed the Roman economy, however, is much less clear. For, while there have been many illuminating studies on the ancient Roman economy and on agriculture in particular, their findings have been largely contradictory and a clear picture of the historical situation has yet to emerge. It has been well documented that Rome's military hegemony, established fairly conclusively in 202 BC, and subsequent expansion throughout the Mediterranean provided the Roman state not only with new territory, but drew in also great wealth through plunder and an expanded tax base. In exploring the effects of these developments, unfortunately, the tendency by many historians has been to rely too heavily on the reports of the ancient historians which often make broad, sweeping generalizations that do not necessarily reveal the real historical situation. This is true especially in the area of agriculture, where, until very recently, it was almost universally accepted that this influx of money and land led to the aggregation of most of the farmland of Italy and Sicily into large slave-run estates, sometimes called latifundia. Many of the 5  ideas about this latifundia formation come from the works of the so-called agronomists: Cato's De Agri Cultura, Columella's De Re Rustica, and Varro's Rerum Rusticarum. These are all treatises on agricultural exploitation which stress the extensive use of large plots, cash-crop production (notably grapes and olives), and slave labour. Certain speeches of Cicero, such as his speech De Lege Agraria Contra Rullum, and Plutarch's biography of Tiberius Gracchus, which relates the politician's alarm at the extensive use of slave labour in second century BC Italian farming, are also often cited as corroborating evidence confirming that the descriptions of the agronomists do truly represent Italian agricultural practice. Recent archaeological work, however, has sought to disprove many of these claims, and important field surveys have been conducted in Italy revealing that small- and medium-sized agricultural holdings did indeed survive the second century BC, and in some cases even multiplied. Such evidence clearly belies the latifundia model of Republican land use, prompting several scholars to suggest that micro-regional variations in land use, rather than any unitary system, can best explain Roman agriculture. Thus, these scholars wish to purport that the idea of a slave mode of production, as many other scholars have stressed, simply did not gain widespread adoption in Italy until at least the early Imperial period, and that slave labour in the second century BC was used extensively only in certain, isolated areas, notably Campania and Sicily. There is, however, a limit to what the archaeological work can accomplish, and many questions remain unanswered. Why, for example, do our literary sources unequivocally stress estate formation, slave production, and peasant displacement? If, conversely, we are to abandon the latifundia  6  model based on our archaeological evidence, what then was the economic structure' in which smaller land-holdings operated? One of the main problems with trying to answer such questions is that research into Roman agrarian history is far too binary in its approach; scholars either concentrate on the archaeological sources or on the literary ones. This staunch division between literary and archaeological scholars is frustrating, however, as there is little dialogue between the two sides and the schism actually works to the detriment of both, for each discipline can inform the other to give a more complete picture of ancient life than either one in isolation. This thesis, therefore, seeks to bring an interdisciplinary approach to the study of Roman agriculture, reviewing a variety of different sources to hopefully allow us to discover whether a union can be formed between the two conflicting models of land use in the second century BC. Certainly there are many challenges inherent in this research plan, including the limitations of archaeological research generally; namely, difficulties in dating sites conclusively based on pottery finds, a problem particularly acute for the Republican period, as well as problems concerning land ownership, tenancy, and the preponderance of slavery, all issues on which the archaeological record is generally silent. Recognizing these limitations reveals an inverse problem, however, which is the temptation to over-emphasize inconclusive archaeological evidence to verify theories which are based primarily on literary or comparative evidence. I make, therefore, no claims at the outset, but have sought only to explore these areas, trying to develop a deeper understanding of what I believe to be a fundamental aspect of Roman history. For, while some works have sought to examine  Including questions of ownership versus tenancy, the status of peasants (whether 'free' landholders, debtbondsman, clients, etc.), and the type of cultivation (subsistence or for-market, mixed or specialized).  7  slave labour, others urbanism and trade, and still others farm-tenancy and wage-labour, few scholars attempt to develop a coherent picture that incorporates all of these branches of economic interaction. Yet it is only by an inclusive, interdisciplinary approach that that we can bridge the gap left in our understanding of this aspect of Roman history and start to provide more complete, nuanced answers to the questions remaining about agriculture in Republican Italy.  8  Chapter 2 Agronomists and the Latifundia Model As mentioned, the long prevailing view among scholars of Roman agricultural practice is that latifundia,' meaning essentially large, slave-staffed wine- or olive-producing villas, had become the normative mode of land exploitation in Italy during the second century BC. This notion, still supported by many modern scholars, is founded largely on ancient literary evidence, including descriptions of agriculture in Cicero's Verrines and in the bucolic poetry of Virgil and Horace. The clearest evidence of Roman farming practices, however, comes from the treatises of the agronomists, the Elder Cato. Varro, and Columella. Admittedly, Varro and Columella lived and wrote in the middle of the first century BC and AD, respectively, and it may be argued that their ideas do not apply to the second century BC, when Cato lived. The importance of all three writers, however, lies mainly in raising conceptual possibilities for how wealthy Roman aristocrats understood and practiced agriculture. It is true that these possibilities may or may not apply to all periods equally, but it is also the case that it is unclear how far any of these writers' ideas can be applied even in his own time, nor how or to what extent farming changed during these periods. It will be fruitful, therefore, to examine all three writers in some detail in order to get a sense of how farming was understood by different Romans at different times. It will be quickly seen, too, that these three writers present a very consistent message and that comparing the treatises is both reasonable and productive. This will not be an exhaustive survey of their writings by any means,  2'  Plantation' is a more familiar and roughly equivalent term, sometimes used in place of latifundia. Its modern meaning, however, particularly in reference to Colonial America, is highly anachronistic for Republican Rome not only for the differences in farm technology between ancient and modern times, but also differences in the terrain and the fact that the land which became plantations in America, unlike in Italy, was not previously farmed extensively by a free peasant class. I do not, therefore, use the term.  9  but highlighting some of the main points raised by these writers and examining how their writings have been utilized by modern scholars can help us understand Roman attitudes towards farming and allow us to check their ideas against the archaeological evidence. All three agronomists, although their works are more involved and ambiguous than is often suggested, wrote exclusively for a wealthy landowning elite and did, in fact, espouse many of the ideas associated with latifundia agriculture: namely, large plots worked mostly by slave labor with a focus on intensive cultivation, usually with the intention of selling the produce at-market, of particularly vines and olives. This is especially evident with Columella and Varro, although Cato shares many of these notions as well. Columella, for example, mentions vines and olives before any other crop ( 1.1.5), and throughout his writing shows a preference for vine cultivation.' Varro is not as forceful in his preference for vines, although this crop still features prominently in his work. Rather, Varro allows for a wide variance in the types of crops which might be profitably cultivated, stating clearly that "some spots are suited to hay, some to grain, others to vines, others to olive, and so of forage crops, including clover, mixed forage, vetch, alfalfa, snail clover, and lupines ( 1.23.1)." The important feature here is that Varro is assuming his readership will be in possession of a large plot of land which would be able to support a diversified crop harvest, thus expressing a focus on large-scale cultivation similar to Columella. Many modern commentators' on these passages point not only to the opulent size of the property, but also to the fact that both Varro and Columella wrote clearly with the assumption that  3  See especially the discussion of the vine's profitability at Book 3, §3.  4  See the following chapter for a more detailed account of modem interpretations of these writings.  10  slave labour would be widely employed in agricultural work, one of the defining features of latifundia agriculture. Varro, for example, lists slaves as part of the equipment essential for any farm, along with cattle and carts (1.17.1), while Columella mentions slave-quarters (servis cella) in his description of how a villa ought to be constructed (1.6.3). Each also frequently alludes to the duties of the head slave, or bailiff (vilicus), who is trusted with keeping charge of the rest of the servile workforce, implying again large plots worked by a proxy/overseer and not directly by the landowner. Columella makes this latter point quite explicit, in fact, as he feels it necessary to admonish his readers to be present on their farms as often as possible so that they might keep tabs on the vilicus. Such a warning obviously applies only to rich, aristocratic Romans who would have had other compelling duties besides farming, and he offers this advice realizing that it would be a difficult ideal for many to actually achieve "since political ambition often calls most of us away, and even more often keeps us away when called (1.1.18)." Although Cato is not as clear as the later writers, he nevertheless anticipated many of these same concerns. Cato as well addresses his advice exclusively to wealthy landowners and recommends large, diversified plots, giving as the ideal size for a villa 100 iugera which should comprise "all sorts of soils (1.1.7)," while even Columella, writing at least 200 years later, advises moderation in buying land and hints that 50 iugera would be a suitable size ( 1.3.10). Moreover, Cato likewise acknowledges that most of the day-to-day operations of the farm will be conducted by a slave vilicus (1.2.1) and, like Columella, feels compelled to encourage his readers to visit their farms as often as possible (1.1.1). The other essential feature of agriculture which must be stressed, again made most explicit by Columella and Varro, is the advantages which a villa accrues by being situated, as Columella puts it, "not far from the sea or a navigable stream, by which its products may be carried off and 11  supplies brought in (1.2.3)." This is important because it is an often overlooked element of the agronomists' prescriptions, for many modern scholars focus on the isolationism and economic 'autarky' that can be obtained by a latifundium where the wealthy owner, his family and dependents (including the slave labourers) actually consume most of the produce grown on the land themselves. While independence is certainly a feature of the large productive estate, it does seem that all three agronomists are more nuanced in their descriptions than is often allowed, and that they do in fact acknowledge that a property owner may not be able to meet all of his needs on his own. Indeed, both Columella and Varro repeatedly mention that the closer a farm is to a village, town, or even other large farms, the more securely it can function. Varro succinctly gives the reason for this, explaining that "many [landowners] have among their holdings some [properties] into which grain or wine or the like which they lack must be brought, and on the other hand not a few have those from which a surplus must be sent away (1.16.1)." This unmistakably refers to markets and implies an interdependence between any single villa and its environs, however we must, of course, be wary of retrojecting too much modern economic rationality to ancient practice. These statements only express, as J.R. Love notes, an economic reality "obviously at some remove from the ideal of autarky" espoused by other writers (Love 1991, 86), but do not necessarily equate to a capitalistic production of consumer goods, as some would argue.' The essential point here is that Columella and Varro allow for and even expect that an owner of agriculturally productive land will have to interact with others, either selling goods which were grown specifically for that purpose, selling goods which were unexpectedly in excess  5  See the following chapter for a discussion of the differing views of how 'modern' (read rational and capitalistic) the ancient Roman economy was.  12  of the producer's needs, or purchasing goods needed to satisfy the owner's requirements, whether business or personal. It is in abandoning the staunch insistence towards autarky, moreover, where Columella and Varro appear at first to differ most sharply from the earlier De Agri Cultura of Cato the Elder, written sometime between the 230s and 140s BC. Cato espouses a clear reticence towards relying on other producers or the import of any good, stating that a landowner always "should have the selling habit, not the buying habit (1.2.6)." He is even wary about business inputs generally, an attitude quite foreign to modern standards and, thus, somewhat perplexing for modern interpreters. He maintains that a farm "should be equipped as economically as possible... [since] however great the income, if there is extravagance only a little is left (1.1.6)." Modern capitalistic theory, on the other hand, believes that inputs into a business, such as farming, are necessary to increase productivity and quality, thereby driving up the income the produce can generate. This variance with modern practice has, I believe, encouraged some scholars to emphasize the differences between Cato's attitudes and more current theories, leading many to characterize his prescriptions as isolationist or autarkist. Cato himself promotes this reading, for surely he wanted his readers to follow the attitude towards farming that he himself explicates when, in his opening remarks, he declares that: to obtain money by trade is sometimes more profitable, were it not so hazardous; and likewise money-lending, if it were as honorable...on the other hand, it is from the farming class that the bravest men and the surest soldiers come, their calling is most highly respected, their livelihood is most assured and is looked on with the least hostility, and those who are engaged in that pursuit are least inclined to be disaffected (1.intro).  This is often taken, as indeed it was intended, as an exposition of the aristocratic Roman ideal of the independent and self-reliant soldier/peasant, unconcerned with profiteering or economic 13  speculation. This ideal, which verged on fetish or a "peasant cult" as Peter Garnsey calls it (Garnsey 1998, 95), was expressed by almost all ancient writers who dealt with agrarian concerns, most eloquently by Virgil and Horace.' While Cato certainly contributes to this idealization of the peasant farmer, his ostensible devotion to self-reliance fails when one considers the rest of the De Agri Cultura. It is quite clear when reading Cato's treatise, as A.J. Toynbee adeptly points out, that it was "written for the benefit of the very business men from whom the author affects, in his introduction, to be dissociating himself (Toynbee 1965, 297)." Cato's concern for the business side of land ownership is confirmed when, notably, he advises land owners to make an account of their assets, including cash, grain, fodder, wine, and oil (argentaria, frumentaria, pabulum, vinaria, olearia), taking a special inventory of what is saleable (1.2.5). After the accounts are made, he insists the landowner then "give orders that whatever may be lacking for the current year be supplied [and] that what is superfluous be sold," including worn out oxen and other chattel (1.2.6). Cato also undermines his principle of autarky when discussing gardens, stating that "near a town it is good to have a garden planted with all manner of vegetables, and all manner of flowers for garlands (1.8.2)." The implication is that selling garden-produce is best done at a local town market, so that the proximity will keep the transportation costs down and offset the highly perishable nature of such goods. This is a minor point and applies only to land near urban centres, as Cato is quick to point out, but it nevertheless betrays Cato's acceptance that sometimes a landowner will want to cultivate goods with the premeditated intention to sell them. Such a  6  Note especially the locus amoenus or golden age rhetoric in Virgil's Eclogues 4 and Horace's Epode 2.  14  reading is, moreover, enhanced by Cato's recommendation that a farm be located near a navigable stream, by a well maintained road, or a flourishing town (1.1.3). This is, as shown, echoed in the writings of Columella and Varro, and suggests that Cato likewise believed some trade/exchange of goods would be required to maintain a healthy villa. The importance of Cato's concern for selling should not be overstated, however, and J.R. Love rightly reminds us that "it is clear Cato does not view the desirability of owning an estate single-mindedly from the point of view of business considerations (Love 1991, 77)." The De Agri Cultura may not, admittedly, often advise growing goods with the intent to sell them for a profit, but Cato's concern for saving costs and recuperating losses by selling off used equipment shows an unmistakable businessman's mentality and a recognition that some market interaction may, at least occasionally, benefit a prudent landowner. In short, he is not consistent in his adherence to autarky. What the agronomists present to modern scholars, then, is a fairly unified picture of the ideal utilization of land: large plots farmed intensively by mainly servile workers, under the supervision of a vilicus, producing one or two cash crops, usually wine or oil, to be sold at market. This is, essentially, the latifundia model onto which commentators throughout history have focussed the better part of their attention. There is, however, much in the agronomists' accounts that is left unclear, including two crucial aspects which must be discussed in detail. First, while undoubtedly Columella, Varro, and Cato all primarily deal with planted crops and emphasize the profitability of vines, olives, and grain above all else, these crops are not exclusive in their writings and other agricultural products are mentioned as well. Varro and Cato make repeated mention of pasturage, including herdsmen in their inventory of labourers (as at Cato I.10; Varro 1.18.1) as well as noting the usefulness of fodder crops for these animals (Cato 1.5.8; Varro 15  1.31.4). Columella, in his list of fundamental farm equipment, mentions both ox-herds (bubulci) and shepherds (pastores) (1.6.8), betraying that even in a cash-crop centric estate animal goods could be produced as well. Most scholars, since they are mainly concerned with the way in which the agronomists differ from the traditional peasant ideal, consequently tend to focus on the formarket production of cash crops, which is also, admittedly, the focus of the treatises themselves. We must nevertheless be wary about overlooking the fact that all three writers acknowledge to some extent the value of mixed farming, including small-scale garden crops, timber, and pasturing. P.W. de Neeve thus suggests that "market products" were produced on only one small part of Cato's ideal villa, while the rest of the efforts were more in line with "the older, predominantly self-supporting farms;" namely, small-scale cultivation geared towards internal consumption (de Neeve 1984, 78). In this way de Neeve accounts for the discontinuity in Cato's writing, espousing the principle of agricultural autarky while allowing for exchange with other producers. This is an interesting idea and may apply also to Columella and Varro, however it is somewhat specious as the writers themselves neither confirm nor deny this. Cato does, it is true, list different crops and for which soil type they are best suited (1.6), although it is not entirely clear whether he expects that all of the different types of soil would all be present in different areas of the same farm, or if he is simply explaining what to plant in the various soil types that one might encounter. The point I wish to stress here is that intensive, large-scale cultivation of vines and olives are not the sole focus of the agronomists' work. The other aspect that is too often overlooked by scholars working with these treatises is the preponderance of free-born peasant labour being utilized alongside the servile workforce. This is not to discount the importance of slave labour, which is clearly meant to be the dominant form of 16  labour on these large villas, yet all three agronomists are not nearly as exclusive in their focus on slaves as has often been suggested. Cato offers very concise proof of this, when he declares that the more popular a landowner is with his neighbours, the easier he can hire extra hands when needed (1.4.1), whom he later describes as 'day-labourers' (using both the terms operarius and mercennarius). de Neeve, while he does admit that tenancy existed in the Republican period, wants to undervalue its prevalence by claiming that, in a 'plantation' style farm which Cato advocates, "the need to keep costs as low as possible means that labour, one of the largest components of the costs, has to be employed as efficiently as possible... [so that] tenancy is not an attractive alternative to slavery (de Neeve 1984, 81)." Theoretically that may be so, but we have seen that Cato juggles a variety of concerns, one of them being the pragmatic realisation that one's own labour force may not always adequately fill one's needs and that, therefore, extra help may be sometimes required. This does not necessarily refer to, it is true, free-born peasant workers, but Cato's use of the term mercennarius to describe these workers suggests that they are not strictly servile but are being paid for their work, even if they be slaves of another and the money went into their peculium 7 rather than their active possession. Columella too recognizes that a landowner will have two types of dependants under his authority as dominus, both slaves (servi) and tenants (coloni) (1.7.1). Columella shows a clear preference for a servile work force, over which the master has complete control, unlike with tenants, although the fact that he feels compelled to argue the case shows, as D.P. Kehoe points  This was a sort of bursary that slaves were allowed to accumulate to purchase goods, their own freedom, or even to rent property. However technically under the control of their dominus, it was to be given to the slave upon manumission. See D.P. Kehoe (166-75) for a detailed discussion of slaves using their peculia to act as semiautonomous tenants.  17  out, that employing tenant labour was still a widely held practice in the first century AD, if not even the normative one (Kehoe 1997, 156). P. Rosafio notes also that the legal evidence, dating mostly to the second and third centuries AD, also makes it unequivocal that tenants and slaves were assumed to have coexisted on the same estate' (Rosafio 1994, 150). It is, of course, unwise to apply evidence of a late date onto the middle Republican period, but the evidence of the jurists provides a conceptual framework in which we can imagine how slave and tenant labour was interconnected. This later evidence, coupled with the fact that Cato specifically mentions mercennarii, suggests that it is at least possible that tenant labour was utilized extensively in the Republic, even on the grandest latifundia. It must be acknowledged here that the details offered by the agronomists do not necessarily reflect any general trends in Roman agriculture, but they raise some possibilities for the goals and concerns of a wealthy Roman who owned a large plot of land. As we have seen, the possibilities are numerous and do not all follow the latifundia model as strictly as has sometimes been claimed. It is a dangerous assumption, and one that has too often been made, that the descriptions of agriculture presented in these treatises truly represented normative Roman agricultural practice. As J.R. Love observes with Cato's ideal farm, which applies also to Columella and Varro, "we are not told how many actual farms of this type existed (Love 1991, 73)." Nor should we really expect such information, for the agronomists intended only to describe their own theories and observances about agriculture and were concerned with only a small segment of the Roman population, namely wealthy and politically active investors of large tracts of land. Often these 8  The jurist Ulpian, in Digest 50.15.4, for example, notes that in the census a landowner was obligated to specify not only the type of crops cultivated on his estate, but also the number of slaves and the number of tenants, with risk if he failed to do so of owing the tenants' taxes.  18  treatises have been utilized by modern scholars simply in order to verify a preexisting notion about latifundia formation and a 'peasant crisis' in the later centuries of the Republic, an idea emphasized by writers such as Cicero, Virgil, Horace, and Plutarch which has persisted into modern times. The agronomists themselves, however, only confirm the possibility that latifundia agriculture was present in a given area and at a given time, but nowhere verify to what extent this was the dominant form of land exploitation. Nor do they provide a wealth of information about the systems of Rome's agrarian economy, about which modern scholars need to provide their own theories and reconstructions.  19  Chapter 3 Modern Reconstructions of Ancient Agrarian Practice Relying largely on the information culled from the literary sources, most notably the agronomists, but also writers such as Plutarch, Cicero, Livy, and Virgil, modern scholars have attempted to reconstruct exactly how Rome's agricultural economy during the middle Republic truly functioned. One of the principal scholars in this field is A.J. Toynbee, who structures his Hannibal's Legacy around the premise that the two Punic wars, especially Hannibal's campaigns throughout Italy at the end of the second century BC, marked a drastic departure for Roman society in terms of economic and agricultural practice. As he puts it: small-scale subsistence farming, which...had been maintaining itself in the Peninsula for at least two thousand years past, had now become precarious...If an alternative economic system, or set of systems, were to present itself, and if the introduction of this promised to be profitable to the old Roman nobility or to the new Roman commercial and industrial class whose fortune had been made by the wars that had been the Italian peasantry's ruin, this ruined and uprooted peasantry would no longer have the strength to protect and preserve its ancestral way of life (Toynbee 1965, 105).  Toynbee further argues that the latifundia system, explicated by the agronomists as discussed in the previous chapter, was precisely this profitable new system, pursued and promoted by Rome's wealthy aristocrats, which did in fact precipitate the ruin of Italy's free peasantry. His theory rests on the devastating impact of Hannibal's campaigns against traditional peasant smallholdings as well as the Roman military strategy of 'scorched earth,' especially that promoted in 217 BC by the consul Q. Fabius Maximus Verrucosus as an effort to deprive Hannibal of supplies, both of which forced farmers to abandon their land as they sought shelter inside fortified areas. Toynbee's argument holds that the Roman propertied classes, after the war was completed, stepped into the ownership vacuum created by Hannibal's raids, Fabius' strategy, and Rome's expropriation of 20  'enemy' property and were, therefore, able to accumulate great tracts of land in peninsular Italy, Sicily, and parts of Southern Gaul and Spain. He contends, moreover, that they then refocused most of the agricultural output of these newly acquired properties onto cash-crop cultivation, particularly vines and olives, as opposed to traditional small-scale subsistence fanning (Toynbee 1965, 31). This argument is linked with the notion, expressed with especial forcefulness in Plutarch's biography of Tiberius Gracchus, that Rome became increasingly reliant on foreign imports of grain to feed its population, since most of Italy's arable land had been turned towards the intensive cultivation of such cash crops. Following this basic model, W.V. Harris, in his War and Imperialism in Republican Rome, argues persuasively that Rome's expansionist policies in the second century BC after the Second Punic War were just as important as Hannibal's impact on Italy. He notes particularly the period between 146 and 120 BC, when many states throughout the Mediterranean became subject to Roman hegemony and, equally important, were forced to pay tribute. Harris maintains that most aristocratic wealth was based in land even before the Second Punic War and, while Hannibal's devastation certainly was an important factor in the rise of latifundia, this new agricultural system could not have supported itself even in a limited form without the supply of new territory, slave workers, and capital9 which was provided by Rome's wars of conquest in the second century, all of which were controlled mainly by the landed aristocracy and often reinvested into property (Harris 1979, 79-81). These developments clearly favoured the wealthy, however Harris does not agree  9  Particularly the liquid property gained from booty during these wars, something which Harris rightly emphasizes as one of the chief benefits of Roman warfare to her military classes. All soldiers had a share in this plunder, but it is important to remember that the share given to the commanders (themselves from the noble, senatorial class) was considerably greater (Harris 1979, 75).  21  with Toynbee's grim declaration that latifundia creation led to the complete 'ruin' of peasant farmers. Rather, he notes that equites and publicani both benefited also from the profits of Rome's imperialistic policy, with contracts for public works projects by the former and the need for the latter's labour in these building projects and in manufacturing, making supplies for the everexpanding number of soldiers (Harris 1979, 94). Unquestionably, success in the many wars of the early second century offered the Romans enormous opportunity to expand and alter their economic system, of which they clearly took advantage. What exactly their agrarian economy became, however, is not nearly as clear. As we have seen, the precise nature of Roman agriculture, even as presented by Cato, Varro, and Columella, is difficult to extrapolate given our limited information. Although most scholars now accept Toynbee's and Harris' basic arguments that a succession of campaigns in the second century BC after the defeat of Hannibal drastically affected Roman society, particularly in the tendency towards latifundia formation, there are still several, serious disagreements in modern reconstructions of the economy which resulted. The conflict, moreover, hinges mainly on the question of how modern the ancient Roman economy was. J.R. Love captures this situation succinctly: the scholarly discussion on the general character of the ancient economy, and on the state of Roman agriculture in particular, has essentially taken the form of a debate between those who have argued that nothing remotely approaching capitalism could have existed in antiquity (the so-called 'primitivists'), and those who have readily found all kinds of parallels between ancient conditions and those of modern times (the 'modernizers') (Love 1991, 60).  This debate has been raging for some time and has been discussed in many careful accounts, including the very useful summary in Love's Antiquity and Capitalism. There is, therefore, no need to describe the arguments at great length here, as a brief overview of what is most relevant to 22  the thrust of this work will suffice. The central difference between the primitivist and modernizing lines, in my view, is how much rationality (in the modern accounting sense) a scholar is willing to project back onto ancient business practice. Moses Finley, for example, one of the most well known and influential of all scholars on the ancient economy, expresses his primitivist stance while trying to make sense of Cato's De Agri Cultura, particularly the somewhat antithetical introduction of that work discussed in the previous chapter. For while Finley acknowledges that, on the evidence of Cato, Roman farms of the second century BC were certainly "farmed for their cash incomes," he nevertheless asserts that the ancients were driven chiefly by a desire to preserve "independence from the market as buyers, from reliance on others for their own necessities (Finley 1985, 109-10)." In other words, even activity that to a modern observer could only be classified as 'economic' was governed in Roman times by other motivations and conditions, since, Finley argues, there was no "enormous conglomeration of interdependent markets" in accordance with which people could gear their economic activity (Finley 1985, 22). Put more simply, Finley and those who follow his arguments feel that "the idea of efficiency, increased productivity, economic rationalism and growth" are entirely recent historical developments in economic thought which did not exist for, and thus cannot be applied to, ancient practice (Finley 1965, 31). It is interesting, too, that the notion that there was a lack of rationality among Roman farmers is largely based on accounting practices in antiquity. Max Weber, an influential scholar who helped develop the primitivist view before Moses Finley, argues that double-entry bookkeeping' is a fundamental element of economic 10  Namely, a system of accounting where two columns of information are stored for every transaction a business undertakes, credits and debits, such that there is an equal sum of debits and credits for all transactions. This is the most effective way to discern one's profit, for the sum when the debits are subtracted from the credits gives an  23  rationalism and that, since such a system was clearly absent from ancient practice, the Romans simply could not have been economically rational (Weber 1968, 107-9). This argument applies not only to discussions of ancient accounting, however, but has enormous bearing on how we conceptualize the entire Roman economic system. For, if we are to believe that landowners such as Columella, Varro, Cato invested in property without any notion of productivity, efficiency, or rational accountancy, that they indeed sought independence from other producers and consumers, we must then posit an economic reality in which market considerations played no significant role. This is not to suggest that Weber and Finley argue that there were no 'markets' in antiquity," however they clearly wish to undervalue the importance of market systems as well as trade and exchange. This is, of course, commensurate with the extreme autarky often ascribed to the agronomists, particularly Cato. In the previous chapter it was shown how this ideal of self-sufficiency espoused by agricultural writers should not be applied too strictly to actual ancient practice, for these same writers certainly were aware of at least the possible need for interdependence among producers. The agronomists were also concerned with furnishing a profit from agriculture, including the concomitant issues of productivity and accounting, even if their actual practices are either unexpressed or do not meet the standards of modern business. Martin Frederiksen warns against reading the agronomists too literally, noting that "we must make allowance for the facts which we are not given, and are presupposed (Frederiksen 1975, 169)." This includes the availability and cost of labour, consideration of expenses in immediate figure of liquidity. Such an assertion would be completely implausible, for the abstract term 'market', as Finley himself was well aware, merely seeks to describe an area within which the phenomena of exchange may occur, and is notably "untranslatable into Greek or Latin (Finley 1985, 22)."  24  determining profit, and the expected sales price for the goods being produced, among other things. Frederiksen here is following the more modernizing approach, which tends to conceptualize the ancient economy as fundamentally similar to modern ones in terms of using capital inputs to increase productivity, having at least a basic sense of costs contra income, and producing goods specifically in order to sell them at market. Michael Rostovtzeff further expresses this view, detailing how a drive for profit motivated large landholders to invest resources (specifically capital and labour) into their villas with the goal of increasing the output, both quantitatively and qualitatively, of marketable goods. Unlike Finley, Rostovtzeff sees this motivation as a factor even for Cato "who insisted on the destruction of Carthage - a measure which can only be explained as the suppression of a rival in the production of wine and oil for the western market (Rostovtzeff 1960, 89)." While the claim that Cato's concern for primacy in oil and wine markets was the only impetus for the Third Punic War is somewhat exaggerated, it is important nevertheless to point out that many wealthy ancient landholders displayed some concern for and consideration of market systems. While this is no doubt the case, we must still be careful not to impose, anachronistically and baselessly, actual modern economic practices, such as double-entry bookkeeping, onto the Roman world. The position in the primitivist/modernizing debate most compelling is that explicated by Willem Jongman, who outlines his view of the ancient economy by claiming: the neglect of modern economic theory is mistaken, and that it can indeed be profitably used. My argument will, however, not be based upon the claim that specific historical conditions prevailed that warrant its use, but rather on the claim that economic theory can still operate in the absence of such conditions... It will also show that we no longer need to search for the presence or absence of economic rationality in the objects of our investigation... but can redefine economic rationality as a strategy of the researcher, rather than as a property of his object (Jongman 1988, 36).  25  Thus, no matter how 'wrong' a modern scholar believes Cato's calculations were, since he did not consider his overhead, the depreciation of equipment, or other factors a modern businessman would, he nevertheless strove towards profit and, therefore, the language and theory of modern economics will be extremely useful as a guide for evaluating and discussing the economic system within which he operated. Given that Roman producers of agricultural goods considered market forces and that we can interpret these forces within a framework of modern economic understanding, we are confronted, then, with the second factor which must be considered, namely the status of cities in Italy and the interaction of rural and urban space. The traditional view, in line with the primitivist stance and again expressed most persuasively by Max Weber and Moses Finley, imagines a 'consumer' model for urban economic life. Roman cities, it is argued, did not produce goods in any significant, rational manner, but were merely the site of the consumption of income and produce from the hinterland. Proponents of such a model "tend to minimize the importance of trade and exchange," claiming that the only significant export was of agricultural goods into large urban markets, and that this was mostly non-rational exchange (Laurence 1994, 51). Finley offers the clearest expression of this tendency when he asserts that "the widespread prevalence of household self-sufficiency in necessities was enough to put a brake on extensive production for export (Finley 1985, 138)." A consumer city is thus "dominated by an elite that derives most of its income from outside the city walls...[it] does not exclude the presence of urban manufacturing within the town, only such manufacturing is largely of local importance (Jongman 1988, 29)." These notions of a consumer city complement the latifundia model of agriculture in that they posit a wealthy, landowning aristocracy who dwell mostly in urban areas — the same aristocracy 26  Columella encourages to spend as much time at their villas as possible because they mainly stay in their city homes enjoying the fruits of their villas and the small-scale urban industry geared entirely for their benefit. The logical counter to this theory is, of course, a 'producer' city in which a large urban manufacturing class benefit from exporting crafts and manufactures to the surrounding hinterland and further afield, and use the income generated by these activities to purchase any goods they might need from the countryside. Such a manufacturing class, most scholars believe, is founded too closely on an early-modern European bourgeoisie who embodied a set of practices and values (social and commercial) far removed from the realities of Republican Italy.' The producer city approach has therefore been all but discarded, and seems to be adduced mainly as a theoretical counterweight to better explain the consumer city model. The problem with the consumer city approach, however, is that it undervalues the activity within the urban space itself, as it involves mainly the import of agricultural goods into urban centres for the profit and enjoyment of the wealthy holders of rural property. P. Garnsey aptly describes the gap left by such rurally-focussed theories: It is also necessary to recognize that there were income-generating transactions taking place in the 'internal' urban economy. ... For example, [property-owners] took a rake-off from transactions between private producers and retailers on the one hand, and retailers and urban consumers on the other, through rents charged on premises (Garnsey 1998, 71).  A far more compelling alternative to the consumer city model, one which better considers the type of intra-urban activity Garnsey describes, is that offered by Donald Engels in his Roman Corinth which he calls the 'service-city' approach. He envisions the interaction between rural and urban  See Willem Jongman's Pompeii (1988, 28-35) for a more complete discussion of how the economic situation of early-modern Europe differed from Republican Italy, in the context of a discussion of the consumer/producer city debate. 12  27  space as a 'symbiotic relationship' whereby agricultural goods are exchanged for services that can only be provided in an urban context (Engels 1990, 43). These include non-economic services, such as religious sites (shrines and temples), judicial and social space, but also 'secondary' economic services such as food, temporary lodging, bathing, entertainment, and other, similar features (Engels 1990, 57). Clearly, enjoying goods was not the only function of a Roman city, but, according to Engels, there were important activities which regularly occurred inside city walls and brought true profit for urban peasant workers and larger property owners. True profit in this sense is that opposed to the 'circular' exchange of urban manufactures and agricultural goods all produced by a single individual or estate, as the consumer city approach often assumes was the normative form of exchange. This 'service-city' approach is, admittedly, not too far removed from the consumer model, but Engels does well to stress that the exchange between the urban and rural sectors was more involved and equitable than most advocates of the consumer city allow and also to highlight many of the other important non-material services that were available in an urban environment, essential elements of any economy which are too often overlooked. It will be quite apparent by this point that such reconstructions of the Roman economy, whether primitivist or modernizing, whether following a consumer, producer, or service city model, are all based primarily on the ancient literary accounts and all assume fairly assuredly that a latifundia model of land cultivation was widely applied throughout Italy by the second century BC. 13 There is, surely, evidence supporting many of these claims, such as the existence of great  13 It must be admitted that most scholars, including Moses Finley, accept that there were regional differences within the Roman world in terms of latifundia cultivation, noting that the phenomenon was more developed in some areas, being particularly prevalent in Campania, south Etruria, Sicily, and parts of Gaul and north Africa. This is assumed to be due mostly to the fact that the land in these areas was simply the most productive for growing vines, olives, and wheat, and the lack of 'plantation' style agriculture in other areas was mainly the result of  28  numbers of slaves, the acquisition of large properties by the wealthy elite, and the existence of large scale market-oriented production of certain 'cash-crop' goods. There are, however, an equal number of questions begging to be asked: exactly how widespread was the phenomenon of latifundia agriculture, and why did it excel or stagnate in certain areas? What was the situation for  non-slave workers in the second century BC? What was being produced and consumed in cities, and who really benefited from such transactions? We turn next, therefore, to look at some of the archaeological work in Italy that has been performed more recently, much of which seeks to answer these questions and to belie many of the economic reconstructions discussed in this chapter.  their natural deficiencies rather than any economic, social, or political aversion to the practice. Latifundia production is, therefore, still seen as the normative, or at the very least ideal, mode of agricultural cultivation in all areas under Roman control.  29  Chapter 4 Archaeology and the Micro-Regional Approach Motivated by the influential studies of scholars such as Max Weber, Arnold Toynbee, and Moses Finley, whose theories on Roman agriculture support unhesitatingly the preponderance of large-estate, slave-worked, cash-crop farming evident in the prescriptions of the elder Cato and decried in Plutarch's famous tale of Tiberius Gracchus, archaeologists working in Italy have been recently trying to prove (or, more properly, to disprove) the presence of such large landholdings by looking at the record left on the ground. Timothy Potter has adeptly revealed that much of the impetus for this is the fact that contained inherently within the traditional latifundia approach "are assumptions that capitalistic farming is bad (and hence relevant to the preoccupations of modern society and Western Europe), and that the Roman experience is an effective demonstration of this (Potter 1987, 99)." The other crucial point which has been stressed recently is that many of the historians who focus on the literary record too readily ignore the acute chronological differences between the end of the second Punic War and the onset of Imperial rule. The desire to challenge these assumptions has driven archaeologists to take a fresh look at the question of how Republican agriculture functioned, and more often than not the answer has been that the latifundia model endorsed so forcefully for so many years has been grossly overstated. This has led to what might be called a 'revisionist' school of agrarian history, which stresses the micro-regional differences apparent in the diverse areas of the Italian peninsula over any unifying or universal economic system. The revisionist push began in the early 1980s when many different scholars working all over Italy came to the similar conclusion that, while there were no doubt many large estates, the 30  archaeological record shows clearly that smallholdings did not entirely vanish during the second and first centuries BC, suggesting that Toynbee's 'ruined and uprooted peasantry' may not have been as disrupted by the Hannibalic Wars as he thought. Peter Garnsey, who has done much important work in this quest to reevaluate the traditional theories, has sought to expose some of the problems of the latifundia model in light of these new archaeological discoveries, particularly as they relate to the plight of the peasant worker. He argues that free peasant labor was always employed as a compliment to slave labor on large villas, however he cautions against taking too limited a view about this class of farmer. Indeed, peasant farmers were not a heterogeneous group, but were composed, as Garnsey details, of tenant fanners, freedmen, wage-laborers, and independent owner/cultivators who worked their own small plots (Garnsey 1980, 34). Nor were these groups mutually exclusive, of course, as a peasant cultivator, for example, could supplement the income generated by his personal property with seasonal wage-labor. What this shows, then, is that the interaction between wealthy landholders, peasant farmers, slave labor and urban workers was much more complex and involved than the traditional approach would allow. Some enlightening and persuasive archaeological work has been conducted which supports Garnsey's theories, none more so than the survey archaeology performed in south Etruria, particularly the area around Roman Cosa. Stephen Dyson, in a 1981 article, seeks to categorize the different sites discovered in the ager Cosanus, distinguishing four 'classes' of farm in terms of size and level of production (Dyson 1981, 261-2). He concludes that in the middle Republican period, roughly the second century to the first half of the first century BC, there was a considerable equality in the number of each type of estate (Dyson 1981, 263). In other words, during this period, large estates had in no way taken over the landscape around Cosa, but medium- and small31  sized farms flourished as well. In the early Imperial period, too, although he notes a clear rise in the number of large estates, there still appear to be significant numbers of the smaller sites. The importance of this assessment cannot, I think, be overstated, for if we are to believe that peasant farmers survived the second century largely intact, in stark contrast to the typical supposition, then we must re-examine our entire reconstruction of agrarian and economic life in Republican Rome. Nor is this situation peculiar to the ager Cosanus, as very similar finds have been made elsewhere in Italy for this same period, including John Lloyd's important work in Roman Samnium 14 around the modern Biferno valley, an area often neglected by historians both ancient and modern, as well as parts of Northern Apulia, the Tiber river valley, and the area of Campania around Pompeii. D.W. Rathbone, remarking on some of this evidence, concludes that there is in fact a great deal of information to suggest that in Republican Italy generally "peasant smallholdings and slave-staffed villas were mutually dependent and in some ways similar modes of agricultural production (Rathbone 1981, 11)." This is supported not only by the existence of small farm sites alongside the larger estates, but by the need of slave-staffed villas to use hired peasant labor to supplement their work needs, a reality, as we have seen, accepted even by the agronomists. Needless to say, the conclusions reached from such archaeological accounts do not always agree, and the evidence concerning these matters is not well preserved. Indeed, a major obstacle with all of the 'revisionist' archaeology is recognizing the limitations inherent in the available evidence, which is nowhere more forceful than when ownership is concerned. For, while we can  14  The most concise display of his findings is found in Lloyd's 1995 article "Pentri, Frentani, and the beginnings of urbanization (c.500-80 BC)."  32  readily identify small farm sites around larger villas, it is impossible, given our limited evidence, to come to any positive conclusions about who exactly owned this land: were all the small sites independently owned, were they all owned by a single wealthy landowner and leased out, or were they, in fact, simply different 'fields' attached to the same single fundus? Even the epigraphic data, which is more abundant in some regions than others, has not been able to shed much light on these topics. Still, it is a problem which must be grappled with whenever the settlement history of a particular region is being discussed and, while it may be intractable, it should not discourage us from raising the possibilities (with of course an understanding of their respective probabilities) allowed by the evidence for how slaves and free workers, wealthy and poor, largeholdings and small were all connected within a coherent economic system. Unfortunately, few scholars have sought to connect all of the reports which have resulted from the recent trend towards a micro-regional focus, preferring to isolate their queries to only one region at a time. This is because, it seems, of the difficulties in dealing with the massive amounts of information such broadly-focused research would require, not to mention the fact that, as JeanPierre Vallat points out, not all archaeological surveys are conducted in the same manner or towards the same goals, so comparing different reports from different regions is extremely difficult (Vallat 1991, 10). There are, moreover, inherent biases which each distinct 'discipline' — history, archaeology, epigraphy, etc. — has towards the others. Notwithstanding all of these challenges, I will now attempt to give a brief overview of the important archaeological and epigraphic evidence that has come out of these regional reports and try to connect it with the views of the literary historians and our evidence from the ancient writers, to see whether the different schools of thought can be in any way married. I have selected several important areas in south and central Italy — the 33  Biferno valley, south Etruria, and Campania, and Sicily — closely examining the evidence from each of these areas for the settlement history and economic output during the middle Republic. It bears repeating that this is not meant as an all-encompassing or definitive treatise exploring the totality of available evidence, rather I am attempting to highlight some of the important issues and point the way towards further research that embraces the literary, epigraphic, and archaeological records.  34  Chapter 5 Biferno Valley and the Samnite Economy The history of the Samnite and Oscan peoples has long been seen as a clash of civilizations between the highly advanced political, social, and economic systems of Rome and the 'backwards' tribes dwelling in the eastern highlands. This is a tradition inherited from the ancient writers, who often sought to describe the Samnite culture as rough and inferior, thereby framing the Roman conquest of the area as an impressive, magnanimous military feat. Much archaeological work has been conducted recently that seems to suggest, rather, that many parts of Samnium and northern Apulia had developed fairly advanced urban and economic structures before the first century BC. Surely the Romans played a large part in this development, directly or not, but the picture beginning to emerge now is one where the part of central Italy east of the Apennines, while isolated to some degree from the west, was nevertheless integrated into the society and economy of the Roman world to a much more significant degree than has been traditionally thought. When discussing Samnium, too, we must distinguish between the upland Pentri peoples and the lowland Frentani, the two major Samnite tribal groups whose very different early histories shaped the society they each exhibited in the later third and second centuries BC. Both areas, as well as many other parts of the Italian peninsula, were greatly affected by ongoing conflict with Rome throughout the fourth and third centuries, specifically between 343 and 280 BC (Lloyd 1995, 182). The victorious Romans tried to solidify their hegemony with the establishment of colonies in Samnite territory, at Cales (334 BC), Fregelae (328 BC), Luceria (314 BC), Venusia  35  (291 BC), and Beneventum (268 BC), to name only a few.' The relationships forged by the Pentri and Frentani with the Romans were not, however, uniform, and recognizing the micro-regional differences within Samnite territory is essential to understanding the economic systems which developed there.  Upland Pentri The settlement of the Pentri people during this time was typified by hilltop Torts,' defensible structures built on high-altitude spots for the obvious military benefit. Such hillforts appear throughout Samnium, however there is a particular concentration in the upland areas, likely because here the terrain was more naturally suited for such structures. The impetus for this building seems to have been the ongoing conflicts with Rome throughout the fourth and third centuries BC, with the prevailing view being that these hillforts did little but provide temporary refuge for the dispersed rural populations in times of crisis, such as during the Punic Wars (Lloyd 1995, 190; Gualtieri 2004, 39). The Pentri, unlike the Frentani, opposed Hannibal and Carthage during the Second Punic War, choosing to support their Latin neighbors instead. This was not entirely without consequence, as parts of their land were raided by Hannibal in 217 BC, possibly as punishment for their support of Rome (Salmon 1967, 297). Still, the Pentri territory remained relatively unscathed after Hannibal's raids and both their hillfort settlements and the greater part of their rural territory seem to have survived largely intact. It is now beginning to become clear, moreover, that these hillfort settlements were more  15  16  For a more complete list, see Salmon 1967, 23. For a detailed discussion of this peculiar type of settlement, see S.P. Oakley's Hill-Forts of the Samnites  (1995).  36  than simply temporary shelter from military threats, but were important centers of habitation in their own right. John Lloyd has identified strong fortifications in the leading Pentri towns of Bovianum and at Monte Vairano," both of which fit the hillfort model and show clear signs of continuous settlement from the third century up to at least the Social War at the beginning of the first century BC. Monte Vairano was fortified by a three-kilometer stretch of outer wall, featuring three gates, several towers, and a series of smaller fortified sites near the central fort, probably connected with the larger center in some way (Lloyd 1995, 188-9). Pottery, tile, three kilns, some evidence of public and private building, religious sites, and "an array of artefacts from many corners of the Mediterranean world" have all been identified within the walls at Monte Vairano 18 (Lloyd 1995, 190). The pottery finds include over one hundred examples of Rhodian amphorae, with also some evidence for Knidian, Punic, and other southern Italian types (Lloyd 1995, 206). Precise dating for many of these finds is not always conclusive, but they suggest a developed longdistance trade by, at the latest, the end of the second century BC. Such evidence thus belies the traditional notion that "Samnium was an economic backwater" which imported "nothing of demonstrable overseas provenance (Salmon 1967, 76)." Nor was agricultural exploitation lacking altogether, although transhumance may have been, as has been asserted from antiquity, the predominant rural activity. Medium-sized and smaller farm sites appear frequently in the archaeological record in the upland plains of the Pentri territory, concentrated especially around the larger hillfort sites such as Bovianum and Monte  17  The ancient name of the settlement is unknown.  18  For a detailed list of these artefacts, see "Catalogue of Samnite and Roman Finds" in Graeme Barker's The Biferno Survey: The Archaeological and Geomorphological Record, 99-138.  37  Vairano. This evidence comes mainly from wells, springs, and some scattered finds of artefacts, probably from debris piles (Lloyd 1995, 194-6). It is, in fact, in this respect that the economy of the Pentri differs most sharply from the ancient accounts, particularly that of Livy, who purports that the entire wealth of Samnium was gained from the plunder of their neighbors.' Livy is, of course, exaggerating for effect, as he wanted to highlight the Roman 'liberation' of Samnite territory from the fear of continual raids, yet such accounts have long informed modern scholars who assert that the Samnites did not exploit their own resources, industrial nor agricultural, to any significant degree (Salmon 1967, 65). The work of John Lloyd in the Samnite uplands has, contrarily, revealed pottery, wells, storage jars, coin deposits, a millstone, and animal and plant remains, all of which are "strongly suggestive of domestic occupation and agricultural activities" in the third and second centuries BC (Lloyd 1995, 194). Good farmland in the mountainous highlands is, of course, relatively scarce, although the available evidence does point towards the existence of a minor amount of surplus agricultural production in the region, at least from certain isolated patches. The best studied example of this is the farm-site at Matrice, located roughly 15 km northeast of Monte Vairano. This farm appears to have been a small landholding with a 'cottage' dating from the late third century BC, which grew into a medium sized villa rustica by the end of the second century, a site comparable to the many contemporary examples excavated in northern Campania (Lloyd 1991, 183-4). The principal activities here appear to have been cereal production along with pig and sheep pasturage. The assumption is that this larger site grew by incorporating several of the smaller landholdings which surrounded the original farm, something which appears also throughout the uplands, although 19  As, for example, at 10.20.9; 10.31.2; 9.40.3.  38  unfortunately the evidence in other areas is sparse (Lloyd 1991, 184). Although it is a relatively large, conglomerated farm site, Matrice does not equate to any latifundia model, as evidence of slave labor in Samnium is non-existent nor was it a highly-productive producer of cash crops (Lloyd 1995, 191). Matrice was surrounded too by many smaller farmsteads, but we must be careful here to remember the limitations of this archaeological work, noting especially that the ownership of these farmsteads is nowhere explained by the available evidence. Thus, we cannot be too quick to place true peasant owner-cultivators at all of these smaller sites which appear in our record. Nor is the dating of these sites always certain, as we have only tentative dates based largely on the existence of so-called 'Campania style' black-gloss pottery (Lloyd 1995, 183). Most of the finds at Matrice, therefore, seem to be of local manufacture, coming, for example, from the kilns identified at Monte Vairano (discussed above), and may not correspond exactly to the dates associated with other examples of the style. It is still useful to note, however, that this type of pottery is the most commonly found in all the Samnite territory, suggesting that the area indeed did peak economically between the third and first centuries BC, but again the actual size and date for most of the sites is doubtful. Nevertheless, there does appear to be fairly stable, diverse settlement throughout the uplands during this time and, moreover, significant agricultural exploitation in the region. This latter point has changed our perception of the economic life of the Pentri, for it now seems that the hillforts which formed the backbone of highland civilization, while they definitely carried pragmatic military benefit, were also at the center of nucleated settlements of small and mediumsized farms and transhumance herders. These hillforts do not fit exactly with the traditional idea of a Roman town, vicus or oppidum, yet they appear to have been an important economic center for 39  the hinterland, providing a market for agricultural produce, producing some manufactured goods,' and perhaps even providing finishing for pastoral goods, such as wool-working. Such sites represent an "urbanizzione in progress," a type of settlement between rural and urban (Gualtieri 2004, 38). Although we are beginning to get a better sense of the agricultural productivity in the area, it is still unclear how precisely the transhumance industry of upland Samnium, stressed by the ancient writers,' functioned in relation to these hillfort settlements. It seems highly probable that textiles, especially wool and leather goods which were probably processed in the hillfort centers, provided much of the export balance in the exchange with Apulia and Magna Graecia, whose goods, as seen, show clearly to have been widely imported into the upper valley. Saepinum, for example, shows evidence for wool manufacture, but it is not unequivocal that this was an export production. Yet, as John Lloyd maintains, "it is clear that something was exchanged in return for the varied imports which reached the valley, and animals, animal products, and grain seem the best bet (Lloyd 1995, 207)." Lowland Frentani The Oscan Frentani of the lowlands had a different history than their upland cousins, particularly regarding their relationships with Rome. Livy (9.16.1) mentions a Roman expedition against the Frentani, which seems to have resulted eventually in a formal alliance between the two, including the establishment of the Roman colony of Apulian Lucera in 314 BC. The center of  20  above.  Suggested, for example, by the presence of kilns for pottery production at Monte Vairano, as noted  21  For an example in the ancient commentary, see, Livy 24.20.4. Lloyd notes the modern evidence for the importance of transhumance in the upper valley (Lloyd 1995, 204).  40  Frentani life was the city of Larinum, which lay near the Biferno river whose valley produced, as it still does, the most fruitful agricultural land in Samnium. Larinum quickly became the principal city in the area and appears to have had friendly contact with Rome from the end of the fourth century onwards. This contact was, perhaps, maintained through the intermediary of Apulian Lucera, very close to Larinum itself (Lloyd 1995, 182). Polybius (3.101) mentions that Larinum was raided by Hannibal in 217 BC, however, as with the Pentri, this did not have too serious an impact on settlement nor on their relations with Rome. The lower valley does not exhibit the hillfort settlement of the Upper, probably because of the more amicable relationship with Rome and because the lowland terrain was simply not suited to such defensive installations. It does exhibit, however, a highly developed urban-rural economic interaction from the early third century BC, coinciding, not coincidentally, with the alliance to Rome and the Roman colony of Apulian Luceria. Larinum itself was clearly the hub of all lower valley economic activity. By the early third century we see a well planned and extensive street grid system, the type familiar in other large Roman towns of the same period, as well as significant monumental building. Later in the century we see also religious buildings, mosaic pavements, an "artisan quarter" for small-scale urban manufacture, and even a large atrium house (Lloyd 1995, 197). Lloyd also has identified evidence for widespread 'multi-cultural' interaction at Larinum, as many inscriptions and coins from this period show Oscan, Latin, and Greek side by side, belying the isolation and backwardness usually associated with Samnite settlement. It is interesting too, as John Lloyd rightly emphasizes, that precisely as "Larinum was emerging as a town, several communities in the middle and lower valley seem to have declined," likely because a large percentage of their populations went streaming into 41  the boom-town of Larinum (Lloyd 1995, 197). Villages elsewhere were not entirely extinct, but Larinum definitely emerged during the third century as the focal point of lower valley habitation, politically, socially, and economically. Additionally, and in sharp contrast to the reports of ancient historians, the evidence points to a concomitant increasing of rural density in the Biferno valley around Larinum, mostly with small farmsteads and some medium-sized holdings comparable to that at Matrice (Barker & Lloyd 1981, 302). Here again there is a deficiency in the evidence, with little provided by the smaller sites, most likely the result of mere archaeological providence. All that we can ascertain is that many small sites did exist, with the black-glaze 'Campania' pottery suggesting dates mainly from the third and second centuries BC (Lloyd 1995, 200). These sites, of course, represent a wide range of activity and exploitation and it is difficult to ascertain with any certainty what sort of goods were produced on any farm or by whom. Nevertheless, the fertility of the land, the appearance of at least traces of numerous concentrated farm sites, and the size of Larinum by the end of the second century all suggest extensive agricultural and pastoral exploitation in the lower valley. From this evidence, John Lloyd has constructed the persuasive theory that the lower valley had, by the early second century BC, developed an agricultural economy geared towards surplus production and, thus, trade (Lloyd 1995, 201). Again, the evidence is sparse, but scattered finds of loom-weights along with the literary sources which stress the importance of Samnium's transhumance and pasturage industries suggest that the lower valley probably did control an extensive wool and leather trade, before and after Roman colonies were established (Lloyd 1995, 204). Livy (22.54.2) mentions the Apulian textile trade with Rome, and it is likely that the Samnites had a similar industry and may have traded extensively with Apulia itself. Small-scale intra-regional trade in 42  legumes and cereals also seems to have been active in all periods. Larinum and her territory were able to develop more quickly and on a larger scale than other Samnite towns largely because of the Biferno valley's excellent fertility. Productive land which can produce regular surpluses, as this land near the Biferno river is and was, prompts the development of an urban center where these surpluses can be consumed and traded for other manufactures; the need for labor to produce these manufactures and to construct the city's buildings, it seems, can explain the fact that many smaller villages in the valley lost population to Larinum while that city was growing in response to the extensive cultivation of its hinterland. At Larinum, then, increasing agricultural production and urbanization occurred concomitantly and in a mutually re-enforcing manner The development of a significant urban community at Larinum was, thus, "a crucially important step" in the rise of the complex market-exchange economy which emerged in the lower Biferno valley during the third and second centuries BC (Lloyd 1995, 208). Integrated Samnite Economy While we must certainly respect the inter-regional (or micro-regional, to use the popular phrase) differences within the Samnite territory, which are indeed striking, this does not itself preclude the notion that there was significant economic interaction in a broader context. Even E.T. Salmon, who mostly under-emphasizes the Samnite economy, is impressed with the number of major roadways that appear already in Samnium's first encounters with Hannibal (Salmon 1967, 20). Obviously, communication between the coasts, crossing the rough Apennines, was difficult in antiquity, which, I believe, gave rise to the tendency by ancient writers of ignoring or relegating Samnium as an isolated backwater. Focusing only on communication with the city of Rome,  43  however, misses the importance of north-south roadways connecting the upper and lower valleys of Samnium both to each other as well as with the lands to the north and Oscan and Magna Graecia to the south, for which there is substantial evidence (Salmon 1967, 20-23; Barker 1981, 294). The existence of these lines of communication, moreover, works in concert with the evidence, mentioned above, of trade between the various Samnite settlements and their neighbors along the Adriatic, especially Apulia, and further supports the notion that at least some parts of Samnium had developed a robust exchange economy. As T.W. Potter explains, local markets, such as at Larinum, were both "a prerequisite in the development of larger-scale commercial enterprise" and often went "hand-in-hand with the spread of colonies," the kind we have seen in many parts of Samnium during the fourth and early third centuries (Potter 1987, 160). This, again, should be examined in relation to similar phenomena in the surrounding areas, such as the large centuriated colony at Taviolere in northern Apulia, where there is evidence for olive and vine cultivation (Potter 1987, 101). Centuriation in the area from the beginning of the Roman period, corresponding to the appearance of Roman hegemony in Samnium, led likewise to an increase in rural productivity as well as the rise and development of large urban centers (Smith 1967, 206). It is likely that the cities of Taviolere also were local market centers for the local goods and for the transport of goods to wider markets; it is extremely tempting to postulate robust economic interaction between Larinum and northern Apulia, which would go a long way to clear up our understanding of how the economy in a supposedly marginal area like Samnium truly functioned, although, so far as I know, no trading relationship can be positively identified between Taviolere and Larinum. Still, it is important to at least raise such possibilities and point the way for further, potentially lucrative research. 44  We can assert with certainty, however, that the economic activity which developed around and through Larinum was a diverse one, a mix of rural and urban exploitation, of small and large scale production, and of local and distant trade. The 'pseudo-urban' centers of the upper valley, such as Monte Vairano, Saepinum, and Bovianum, were comparable economic units, just on a smaller scale. Surely the productivity of agriculture in the upper valley could never have matched that of the lower Biferno valley, nearer the coast, nor were the lines of communication as easily navigable in the mountainous highlands. These 'urbanizzione in progress' hillfort settlements did, nevertheless, exhibit some of the same essential features of this centralized economic system, including the existence of small farms near the town, medium-sized holdings further afield, and some urban manufacturing. All of the recent archaeological evidence for these areas, then, when considered together, seems to suggest a highly complex economic system that contained both the isolated, micro-regional aspects often stressed by ancient and modern commentators as well as a significant level of inter-regional interaction, likely among the different peoples of Samnium, but certainly with the neighboring areas and possibly through them with the wider Roman world and the capital herself.  45  Chapter 6 Settlement Patterns in the Ager Cosanus and South Etruria The defining feature about settlement in the ager Cosanus is, as Daniele Manacorda has remarked in an essay surveying the land use in the area, its diversity (Manacorda 1981, 44). The archaeological record shows clearly that farmsteads of various size coexisted in the ager Cosanus throughout the Republican period, and even into the early Imperial period. We also see a fairly lively mixture of Romans and 'native' Etruscans co-inhabiting the region from the early third century BC. This seems to hold true for many areas in Roman Etruria, but I have chosen to focus on the ager Cosanus simply because, as is so often the case in any investigation into the ancient past, its settlement history by chance happens to be especially suited to the purposes of this paper; established as a Roman colony in 273 BC, the roughly 550 km 2 area flourished throughout the Republic, yet faded in importance and population in the Imperial period. The area, therefore, shows a rare purity in its Republican-era fords (Dyson 1981, 79). Frank Brown reminds us that there is scattered evidence for Imperial-era occupation in the city of Cosa, however there was a substantial drop in population after 70 BC, 22 with the result that the latter settlement left only a minor mark in the archaeological record (Brown 1980, 74). Cosa is located roughly 100 km north of Rome, in a naturally advantageous position along the Albegna river, a navigable stream whose beds produce good arable land (Dyson 1978, 255). It was founded as part of a series of Roman colonies established in the late fourth and third centuries  22  The city did, however, flourish again in late antiquity, though this latter settlement did not damage the Republican-era evidence extensively.  46  BC after Rome had successfully defeated the old Etruscan power centers during several battles.' One of the most significant of these battles took place in 280 BC when the Roman consular army led by Tiberius Coruncanius defeated a combined force from Volsini and Vulci (Brown 1980, 1). Cosa was selected as a site for a Roman colony shortly after this, largely for strategic reasons: to act as a buffer between Rome and the still hostile Etruscan territory further north and to afford Rome control over Tyrrhenian sea-routes, for the area possesses an excellent natural harbor in the channel between the promontory Mount Argentarius and two small islands just offshore. It is clear, however, that these were not the site's only benefits, for the fertility of the land along the Albegna river played a large role in determining the location as well. As would be expected, Roman headway into Etruria had a significant effect on Etruscan culture. Timothy Potter explains that: as the frontiers of Rome's territory were gradually extended during the fourth and third centuries, so was the structure of Etruscan life gradually eroded. It began with the decline and in some cases the abandonment - of many of the Etruscan and Faliscan towns that had for so long formed the focus of each pre-Roman pagus or tribal subdivision. In some cases, new towns were created, like Falerii Novi: but much more often the inhabitants of the nucleated sites were dispersed into the countryside, to found new farms and raise the level of the rural population. It was a deliberate policy of decentralization, aimed in part at raising levels of agricultural production. (Potter 1979, 93)  The 'Romanization' of the ager Cosanus, however, deviated from this pattern somewhat. While we see an attempt by the Romans to raise rural population and, subsequently, productivity, the ager was only partially centuriated when it was initially founded as a Roman colony (Dyson 1978, 259;  23 This same colonization is mirrored throughout central Italy at this time. See, for example, the discussion of the Roman colonization of Samnium in the previous chapter.  47  Attolini 1990, 144). This suggests only partial colonization,' while the rest of the land was left for its native occupants or was simply unused. Indeed, most of the evidence from Cosa suggests that pre-Roman settlement in the area was actually quite sparse (Dyson 1978, 258). Rather than supplant Cosa's native inhabitants, then, the Romans were mainly content to occupy land which had been left unused or at least under-cultivated (by Roman standards), drastically raising the rural population density without altogether disrupting the native habitation, scanty as it was. Admittedly, some of these calculations are problematic as the scarcity of pre-Roman settlement that we see around Cosa could simply reflect a paucity in the archaeological record left by the early inhabitants, not necessarily indicating an actual absence of habitation. Surely we should expect that the small, poor Etruscan farms of the pre-Roman period would have left little archaeological trace. In fact, we do have limited evidence for scattered Etruscan settlement in the area between the seventh and fifth centuries BC, mainly from cemetery finds, however after the fifth century there is "an as yet inexplicable blank" in occupation until 273 BC (Brown 1980, 8). It may be that the Roman colonization was so disruptive that it annihilated all traces of the previous inhabitants, however this seems unlikely, especially given that the limited centuriation we see in that initial phase of Roman settlement. Until we prove otherwise, then, we must assume on the basis of the evidence we currently possess that the Romans sent to colonize the ager Cosanus in 273 found the land largely unoccupied. Thus, the initial effects of the 'Romanization' of Cosa were mainly in increasing the rural population density and turning many previously uncultivated areas into productive land. During the middle of the third century BC, however, Roman incursions were made into  24  Centuriation being, of course, the typical feature of any Republican agricultural colony, and its appearance in an area is crucial to determining the extent of actual colonization.  48  Etruscan territory which did have a serious impact on the native settlement, centering around the destruction of Falerii Veteres during the Faliscan War of 242 BC. Nearly 15,000 Faliscans were killed in the war and in 241 BC Falerii Veteres, their chief city, was replaced by Falerii Novi. (Potter 1979, 99) The new city was deliberately constructed as a physical representation of Rome's new hegemony in the region as well as her administrative capital: the name of the new city, its location (founded right next to the site of the destroyed Falerii Veteres), and its large size and architectural features 25 made the entire city "an impressive symbol of Romanitas (Potter 1979, 99)." This city is, admittedly, not in Cosa's vicinity, though it is an important example of Rome's policy towards the newly-conquered Etruscans and deserves some attention. Indeed, the building of Falerii Novi occurred after a massive depopulation of Etruscan inhabitants from Falerii Veteres and the surrounding area, and the new Roman presence must have had a tremendous impact on the local society. Still, the replacement of one city by an entirely new one seems to have been the exception rather than the norm in Rome's colonization of Etruria. Nor was Falerii Novi the result merely of Roman political and military strategy. Rather, the Romans seem to have considered the economic implications of their actions as strongly as the other factors, as was the case with Cosa. It is indeed striking, as Simon Keay et al. point out, that the Romans chose not to simply leave the area depopulated once Falerii Veteres had been captured, opting instead to build a new city on nearly the same site as one which had been at the center of a major rebellion (Keay 2007, 232-4). They posit the compelling argument that the most likely causes of this decision were the natural economic advantages of the location, and that the new city was therefore not simply a major  These include massive perimeter walls which reached over six meters high in some places, with regularly spaced turrets (Potter 1979, 98). Soon after it was built, a large long-distance roadway, the Via Amerina, was constructed right through the heart of the city, further enhancing its central place in the region. 25  49  defensive installment. The archaeological remains from the site bear this out as well, as a large forum with its concomitant public buildings appear from the earliest stages of development, as well as a Roman style porticoed theater, all of which seem to have played an economic or cultural role rather than a military one (Keay 2007, 231). Roman colonization in Etruria continued after the Faliscan War and the establishment of Falerii Novi, notably with the founding of a centuriated colony at Saturnia in 183 BC as well as a second wave of colonization at Cosa in 197 BC (Livy 32.2.7). Unlike at Falerii Novi, these secondcentury colonies seem to have brought about mainly an increase in the rural density, urban population, and agricultural productivity, not entailing a complete disruption of the native habitation (Attolini 1990, 144). As at Cosa, centuriation around Saturnia appears to have been limited to the east, while to the north and west of the city the land was probably left to 'native' populations. Unlike Cosa, however, Saturnia was the site of a formerly important Etruscan city. It is possible, then, that the limited centuriation reflects an intentional aim by the Romans to focus colonization on unoccupied areas, being careful not to disturb the land that was productive before Roman occupation. Archaeological work has shown hints of this as well, for many of the farmsteads of the fifth and fourth centuries BC around Saturnia, sites much larger than the normal centuriated plot, remained intact after 183 (Attolini 1991, 142-3). Either this represents continuity in native habitation, or the original plots were simply turned over to new Roman occupants. Unfortunately, trying to distinguish between Etruscan and Roman habitation on these sites represents one of the greatest challenges with settlement archaeology, for the settlement record is simply silent about the individuals who actually lived on these farms. Yet, since there is evidence to "suggest the continuity of indigenous property (Attolini 1991, 144)," namely that many sites 50  persist between the pre- and post-Roman periods, it seems possible that there was partial continuity also of the indigenous occupants.' A complete replacement of Etruscan populations would, indeed, have been unnecessary unless the land was needed by Roman colonists, which does not seem to have been the case. It appears, rather, that there was ample territory which was unused or underused by the pre-Roman settlers. The picture of Etruscan settlement that this evidence paints, even considering the great impact of Falerii Novi and Rome's assertion of political and economic hegemony throughout the area, is that the Romanization of Etruria did not entail the complete dissolution of native settlement and culture. What we see rather is a shift in the focus of the area's political, social, and economic organization towards the city of Rome, yet with some Etruscan elements being retained. The complexity of Rome's incursion into Etruria is borne out also in the intricate system of roads which traversed the region.' Many Etruscan roads were already in place by the fourth century BC, mainly small paths connecting the important towns of Etruria. Just as there was some continuity in farm sites between the Etruscan and early Roman periods, so too many of these Etruscan roads were maintained and exploited by the Roman settlers. These older roads were, surely, "as relevant to the Roman landscape as the Etruscan," however with new Roman inhabitants farming the land who had strong ties to the rest of the Roman world, "there gradually emerged a need for roads that were specifically designed to afford access to central and northern Etruria and further afield (Potter 1979, 101)." Several large roadways were thus constructed during the middle  26 Potter (1979, 94) believes also that, at least in the early phases, Romanization did not completely disrupt indigenous populations. 27  See figure 2.  51  and later Republican period, not so much as alternatives to the smaller, local roadways as complimentary to them. The Via Aurelia, for example, constructed probably sometime in the early second century BC, ran near the Albegna river valley in Cosa's territory to provide communication between Rome and her Etruscan colonists, including the transportation of goods. Several other important roads were constructed during this period for similar reasons: the Via Clodia, which incorporated and expanded upon a preexisting route and ran from Rome up to Saturnia; the Via Amerina, which passed through Falerii Novi en route to Umbria; and the Via Flaminia, which traveled into Umbria and up to the Po River valley at Rimini  28  In his analysis of this road building, Timothy Potter emphasizes that many of these new Roman roads, particularly the Viae Amerina and Flaminia, seem to have deliberately by-passed many large Etruscan cities, notably Veii. It does seem that Rome intentionally sought to isolate the old centers of Etruscan power for strategic purposes, but we must remember that not all Etruscan towns were bypassed and that many of the smaller pre-Roman roads were maintained and, in some cases, as with the Via Clodia, renovated and expanded. While certainly trying to isolate some of the most powerful Etruscan centers to prevent any large uprising against their authority, it seems that the Romans were equally interested in building up the communication routes of Etruria. Thus, both local and long-distance routes were created or fortified to connect the areas of Etruria now under Roman control both to themselves and to the city of Rome and the rest of Roman Italy. The larger roads connected the chief Roman hubs — Rome, Cosa, Saturnia, Falerii Novi, Rimini, etc.— and by-  28  The date of the Via Clodia is unknown, but I am persuaded by Harris' argument that it was constructed around 183 BC, the same time that its terminus, Saturnia, became a Roman colony (Harris 1971, 166-7). The Via Amerina was built along with Falerii Novi, shortly after 242 BC. The Via Flaminia is firmly dated at 220 BC (Potter 1979, 102-4).  52  passed the smaller or strategically isolated places like Veii. The many smaller, local roads which were left intact or expanded connected the towns in between. At Cosa, for example, there is evidence in the Republican period of a small, local trunk road which leads to Saturnia, linking the two Roman settlements (Dyson 1978, 259). Its precise route or date is unclear, but the remains that appear are closely aligned with many Republican era sites in the area, suggesting that it was in use by at least the early second century. Between this small road, the Via Amerina, and the Via Clodia, then, Cosa, Saturnia, Rome, and all spots in between were able to maintain regular contact. A similar situation can, I believe, be postulated for many areas throughout Roman Etruria. Returning to the rural settlement, it was mentioned previously that the most striking characteristic about the ager Cosanus is that small- and large-sized farms co-existed in the area throughout the Republican period. This is particularly noteworthy because it both supports and contradicts the traditional notion of Italian settlement history, and it has indeed been the source of much recent controversy among scholars of Republican agriculture. On the one hand, traditional accounts have been bolstered by the appearance of large wine-producing villas in some parts of Etruria, including the ager Cosanus. On the other, those who wish to discredit the latifundia formation model of Roman agricultural development cite the evidence which reveals a prevalence of small sites in the area alongside the larger ones. Frank Brown, in this latter vein, suggests that even the larger estates that did exist "could not, in any sense, be described as domains, latifundia, of the kind ancient writers are so fond of lamenting (Brown 1980, 71)." Still, these larger properties cannot be discounted or suppressed, and we must try to rectify the fact that both arguments seem to have some archaeological basis. Nowhere is this more clear than in the case of the well-preserved villa at Sette Finestre. 53  This large estate in the ager Cosanus, made famous by the reports of Andrea Carandini, is probably the ancient site which most closely resembles what modern scholars mean by the term latifundia; the Sette Finestre estate was a massive villa geared mainly for the intensive cultivation of vines, containing a large and sumptuous main domus as well as roughly 500 iugera of productive territory, including rural buildings suited for storage, equipment, and slave quarters. Further supporting the idea that the region was a major producer of wine is the fact that large quantities of Dressel 1 amphorae have been found in the area around Cosa, a considerable percentage' of which bear the stamp of a single producer, Sestius (either stamped SES, SEST, or SESTI), who is sometimes 3° thought to be the owner of the Sette Finestre villa (Potter 1987, 108). The identification of Sestius as owner of the Sette Finestre estate is specious, admittedly, but even if not the case it is undeniable that Sestii amphorae are found all over northern Italy and even into southern Gaul and Spain, suggesting that some of the wine producing villas of the ager Cosanus, likely Sette Finestre itself, were extremely large and successful enterprises (Attolini 1991, 148; Panella 1981, 71). Indeed, all of the evidence from Sette Finestre fits the latifundia model to a tee, and the estate is often used as the archetypal example of the phenomenon. In an attempt to account for the small farmsteads which still survived in large numbers in the third and second centuries BC, proponents of the latifundia model note that the Sette Finestre estate which appears in our record  29  Roughly 10 percent of the approximately 800 fragments which bore stamps (Brown 1980, 71).  30  The Sette Finestre often is identified with the Sestius family, largely because it is a large, wine-producing villa in south Etruria. This identification is by no means certain, however, as there are other large villas in the area as well. The fact that there are several candidates for the Sestius villa further underscores that this area could support several large, latifundia style estates. -  54  dates to roughly the middle of the first century BC and actually "seems to have been one of the last [villas] to be constructed (Attolini 1991, 149)." The implication is, therefore, that widespread latifundia formation did in fact occur, only its origin was particular to the later Republican period. The many small farmsteads which are evident in the record for the early and middle Republican period, it is thought, were increasingly and inevitably swallowed up by the larger and wealthier villas beginning in the mid-first century BC. There are, however, two important objections that must be raised against this conclusion. First, even in the middle decades of the first century BC, when Sette Finestre was at its peak, the surrounding area still featured many smaller sites. Dyson, refuting some of Andrea Carandini's conclusions about the Sette Finestre villa, notes that the proponents of the latifundia model rely too heavily on: the description of Roman intensive, slave-based farming of the 2nd century- early 1st century BC found in writers such as Cato. An overview, however, of the archaeologically reconstructed agricultural world of the late Republican-early Imperial ager Cosanus when Sette Finestre flourished shows a more complicated landscape. There diverse types of farms continued in close proximity (Dyson 1981, 82).  The second problem is that villa formation is not limited to the first century BC; a few larger villas have been identified in the ager Cosanus even during the middle Republic, notably the wineproducing estate at Le Grotte belonging to L. Domitius Ahenobarbus (Dyson 1978, 260). Thus, the development of the Sette Finestre villa was not so peculiar to its time. It must be acknowledged that the dating of any site in the ager Cosanus is by no means definitive, and is based largely on the amphorae record. Dating from amphorae is, of course, not an exact science, nor do we everywhere have a sufficient quantity of finds to make definitive conclusions about chronology. It does seem safe, however, to assert that the available evidence suggests at the very least the existence throughout the Republican period of smaller sites alongside large villas, regardless of when exactly 55  any individual site existed. If, then, we are to conclude that small and large holdings occupied the ager Cosanus together at the same time, we must then attempt to explain how such disparate elements could have functioned in a single economic system. The most fruitful answer to this problem is provided by looking at the site occupation around Cosa within the framework of the central place theory. This theory argues that the location where a product is produced in relation to its intended market controls the profit gained on that product. Different goods, of course, have different rates of return, being either 'high order' goods which fetch a large price and carry a low transportation cost, or 'low order' goods which function in the reverse. The higher the 'order' of the goods, the further away from the destined market they can be profitably produced, since they are luxurious enough of an item to demand a high enough sales price to surpass the transportation costs.' The limit to this is governed by the economic principle of diminishing returns, which holds that the maximum limit that any product can be produced away from its intended market is the distance at which the costs of transportation will equal the profit garnered from the sale, thus furnishing a zero return for the producer. Donald Engels summarizes well the result of the exchange and interaction between these two levels of goods, explaining that "a regional marketing network arises in response to the supply and demand of high- and low-order products in a given area (Engels 1990 , 174)." The evidence indeed suggests that just such a 'regional marketing network' developed around the ager Cosanus and that, moreover, recognizing this economic interplay enables us to understand better the function and interconnectedness of the varied rural sites that appear in the archaeological record.  31  Other costs, such as labor, equipment, etc., are assumed to be also covered by the sale price.  56  Indeed, the settlement history in the ager Cosanus during the height of Cosa's prosperity in the second century BC shows some very interesting patterns, including a noticeable concentration of sites within a 6 km arc to the east of the city (Dyson 1978, 259). 32 The most obvious reasons for this concentration, as Stephen Dyson records, is that: to the east and slightly north of Cosa lay some of the most productive land; because 6 km still afforded fairly easy access to the urban center; and that many of these sites were situated near the roadways running through this part of the territory, which included the Via Aurelia and the smaller road which ran north-east up to Saturnia, discussed above?' Moreover, the larger villas from all periods are concentrated around the coast where the Albegna river opens up to the Tyrrhennian Sea or around Cosa's harbor (Dyson 1978, 261). According to Stephen Dyson, the evidence suggests that the larger villas around Cosa, at least during the Republican period, were not particularly productive economic units, but were "residential country places (villae maritimae) accessible by sea to the wealthy coming up from Rome" which exploited the view offered by a sea-side location (Dyson 1978, 261). This is possible, but given our information about the productive capacity of these villas,' it is unlikely that this was their only purpose. As discussed above, the co-existing motives a wealthy Roman might have for investing in land, including the social prestige of a sea-side resort as well as the land's productivity. It seems  33  The precise routes of these roads are not entirely certain, and this conclusion is, thus, not defmite, but again the traces of the roadway that are evident and its assumed route seem to correspond to the concentration of sites. (Dyson 1978, 259) 34  The Ahenobarbus estate, for example, shows signs of mixed cultivation, with an emphasis on wine production (Manacorda 1981, 45).  57  rather that, as Daniele Manacorda believes, such large estates were placed where they were in order to take advantage not only of the view, but also of the elaborate and highly effective system of roads in the area as well as the navigability of the Albegna river and Cosa's natural harbor. All of these factors would have kept transportation costs down for 'high-order' goods such as wine and enabled producers like Ahenobarbus and the owners of the Sette Finestre estate to exploit markets further afield, notably Gaul, Spain, and Rome (Manacorda 1981, 46). Attolini et al. agree that the large wine-producing estates which began to emerge in the second century BC were built in spots where they could maintain access to communication routes, in addition, of course, to fertile soil (Attolini 1991, 145). This would explain, I think, the fact that most of the larger sites in the ager Cosanus were concentrated around the coast or near the river. The amphorae record also supports these claims, as large concentrations of Dressel 1 amphorae have been excavated at the port of Cosa, most of which are dated to the second century BC, when the Dressel 1 amphora took over from the older Greco-Italic types, a change which seems to "signal the beginning of the commercial wine production which was to become the most important resource of the coastal area (Attolini 1991, 147)." Such evidence fits the notion that large wine-producing estates sprouted up around Cosa starting in the late third century BC, but the smaller sites function within this same model as well. The central place theory holds that small farms would have provided lower-order goods for a more local consumption than the wine produced in the area, and this is backed up by evidence for the emergence of a small-scale urban market at Cosa at the same time as the long-distance wine trade began to flourish. The city and hinterland of Cosa lost many of its inhabitants during the Hannibalic Wars which, coupled with Rome's desire to have productive colonial holdings in 58  Etruria, seems to explain the call for a second wave of colonists in 197 BC (Brown 1980, 33). This new phase brought about a further Romanization of the city, including an updated and expanded forum and, notably, several plain "utilitarian" buildings around it, namely a series of atriums and small chambers (Brown 1980, 35). Frank Brown has identified in these rooms signs of metalworking and pottery manufacture, and many of the other chambers appear to have been shops (Brown 1980, 35). This reveals, it seems, the emergence and development of an urban market center sprouting up in Cosa at the beginning of the second century BC, likely in response to the increased occupation and productivity of the hinterland. Cosa, then, acted as a market center to distribute locally made lower-order goods, including produce from the smaller farms sites in the area as well as the urban manufactures noted above. As Timothy Potter puts it, urban markets often sprout up in response to an upturn in agricultural output, which is itself a phenomenon associated with Roman colonies, providing "the ways and means for the dissemination of goods and produce (Potter 1987, 160)." It is therefore not surprising that we see market space and other commercial buildings being built in Cosa at the same moment when a new wave of colonization affected an increase in the area's agricultural occupation and productivity. The best theory to explain settlement in the ager Cosanus seems to be that wine, a principal 'high-order' product in all ancient economies, was produced by the large villas situated along the coast, while lower-order goods were produced inland and traded in the city center, perhaps along the local road systems to other Etruscan towns as well. Not only were the larger estates concerned with market interaction and long-distance trading, but even the smaller sites appear to have engaged in some market activity, albeit on a more limited scale, and thus neither seems to be entirely primitive in their economic output. However, it is still not entirely clear how extensively 59  these smaller plots were cultivated and to what extent they participated in market exchange. Nor can we assert that the situation of the ager Cosanus should be taken paradigmatically for the central Italy generally, although it is noteworthy that several other Roman-era sites in Etruria reflect a similar pattern, such as Saturnia and Luna, which is situated up the Tyrrhennian coast from Cosa. Helen Patterson, moreover, believes that these same regional trade networks developed also amongst the towns lying in the Tiber river valley. She argues, rightly, that the large-scale export of goods to Rome along the Tiber had dramatic side-effects on the other aspects of the producer territory's economy, as we see from the ager Cosanus (Patterson 2004, 64). In particular, she feels that a robust export trade of wine and oil in the area encouraged small-scale farming and urban manufacturing to provide the goods needed for local consumption which were not being met by the larger estates, because the latter focused most of their resources into cash-crops. This is not the place for a lengthy discussion of agriculture in the Tiber valley, however it is certainly interesting to note that there may be stable grounds for comparison with southern Etruria. What we are concerned with here is mainly how the archaeological remains of certain areas in Italy fit with traditional theories, and how we can amend our understanding of the agrarian economy to fit this new information. To this end Campania and Sicily need to be explored in detail, as they are the two areas most heavily associated with latifundia agriculture.  60  Chapter 7 Wine and Grain Production in Campania Many scholars rightly stress that the Roman world during the Republic was in no way uniform and that we must recognize micro-regional differences in settlement history and economic systems. In this vein it has often been argued that, while the dominance of latifundia production can be disproved or at least questioned in some areas of Italy, as we saw with parts of Etruria, it was significant in the areas of central-western Italy just south of Rome, particularly parts of Campania, as well as throughout Sicily. A.J. Toynbee once again leads the way in this charge, arguing that by 200 BC "the Central Italian lowlands were lost to the traditional peasant economy and were won for the slave-plantation form of commercial agriculture (Toynbee 1965, 252)." Essentially his argument is that much of Campania, particularly in the more mountainous regions, was by the early third century BC given over to large-scale wine production. This idea can be seen already in antiquity, when, for example, in his Georgics Virgil spends a great deal of effort to promote Italy's viticulture over the rest of the Mediterranean, in particular the unmatched quality of Campanian wine (Virgil, Georgics 2.96). The idea of Sicily being Rome's 'breadbasket' also has a long tradition, and it is widely held that grain-producing latifundia developed there at a very early stage. Indeed, the island is often presented as the clearest example of slave-staffed latifundia agriculture anywhere in the Roman world. For Sicily, as for Campania, however, it is not enough to rely on the ancient literary sources alone, as the actual settlement histories and agricultural economies of these regions are not as straightforward as they are sometimes presented. A.J. Toynbee sees the destructive impact which the Hannibalic Wars had on central and south-eastern Italy as the impetus for latifundia formation in Republican Campania. He places great 61  emphasis on the 'scorched earth' policy of consul Q. Fabius Maximus Verrucosus, whose strategy from 217 until 211 BC was to deprive Hannibal of his Italian allies and provisions. To accomplish this, Fabius ordered that all communities in the area destroy their farms and stockpiles of food and he issued decrees that all Romans should seek shelter in fortified towns while Hannibal was campaigning, with of course the threat of severe punishment for non-compliance. Fabius' efforts, while quite successful in their military purpose, had, according to Toynbee, the side effect of clearing away many smallholders from their land. This problem was exacerbated also with Hannibal raiding farms to supply his own troops, by the punishments meted out by Rome on those towns which did ally with the Carthaginians, and with the simple loss of manpower from this protracted period of fighting. Toynbee believes that, by the time Hannibal was expelled from Italy, it was "as if the Italian Peninsula had been devastated by a hurricane that had torn up the trees, and by a following forest fire that had burnt off the shrubs and the grasses....And, since Life, like the rest of Nature, abhors a vacuum, this was an invitation for some new form" of social and economic organization, which is how he believes latifundia first came to be in the region (Toynbee 1965, 101). The archaeological evidence supports Toynbee's ideas to a certain extent, as we see something approaching latifundia agriculture in parts of Campania. This is true especially of the territory around Pompeii, where a great number of sites show some of the characteristics of productive latifundia (Rostovtzeff 1957, 57). Campanian soil was noted in antiquity, as it still is today, for its excellent fertility and for the superior quality of the wine it produces, and many of the sites that have been excavated in Pompeii's hinterland have produced evidence for large-scale wine  62  making, as well as evidence that much of the labor was conducted by slaves.' However, it must be stressed here that, although it is always easy to exaggerate the importance of Pompeian evidence, there are not really any true latifundia in the area, and the sites that have been identified as villas are packed tightly in a tight area and are, on the whole, quite small; nowhere near the size or productive capacity of the contemporary estates in south Etruria, for example. It also must be born in mind that the sites excavated around Pompeii are mainly of later Julio-Claudian or early Flavian date and it is, therefore, somewhat inappropriate to retroject the agricultural economy of these villas to earlier Republican times. Still, the argument made by Toynbee is that many of the Pompeian villas of the first century AD actually began to be formed in the second century BC. This was allowed to progress, it is argued, because through Hatunbal's campaigning in central Italy, the expulsion of many peasants from their land through military obligations, and the devastation caused by Fabius' scorched earth policy, land became available for wealthy Romans to purchase in fairly large plots (Toynbee 1965, 31). We must be careful with accepting this analysis too readily, however, for there is the question of whether Pompeii's agricultural industry in the second century BC was indeed a tradeoriented cash-crop production. This is difficult to prove, and scholars in the past have too often relied on the mere fact of nature that Campanian soil is extremely fruitful and throughout recorded history has furnished regular surpluses. Martin Frederiksen has argued, for example, that because of Campania's remarkable fertility its agricultural economy "had long since developed beyond a simple peasant or pastoral stage...[and that] specialist and expansive forms of commercial  35  At villas 14, 25, 33, and 34, for example, slave quarters have been identified (Jongman 1988, 113-7).  63  enterprise may well have made their appearance before the Second Punic War (Frederiksen 1959, 108)." Nor is this the only view of Pompeian agriculture, as Willem Jongman takes the opposite approach and argues that the fertility of land around Pompeii was actually utilized primarily for small-scale grain cultivation (Jongman 1988, 143). To resolve this confusion, it is revealing to take a closer look at the archaeological evidence for large, productive villas around Pompeii and to scrutinize what they actually reveal about the agricultural economy of the second century. Although none of the sites around Pompeii are large enough to be classified as a latifundium, roughly 100 properties have been discovered around Pompeii which do exhibit the properties of intensive wine and oil production, though on a smaller scale than the type of agriculture the agronomists describe. 36 One such estate, villa #34 by Rostovtzeffs numbering, has been excavated from the area just south of Pompeii in the Sarno river plain, where 34 wine storing dolia were found, along with a large center for wine vintaging, slave quarters, and what appears to be stables for cattle (Jongman 1988, 115). Many other sites in the vicinity also reveal dolia, vintaging facilities and the remnants of where stakes where placed to prop up grape vines. Several of these sites as well have fairly sumptuous domus in addition to their agricultural facilities, with wall paintings and mosaic tiling found at villas #13, 25, and 27 (Jongman 1988, 117). This combination of ostentatious domestic buildings with productive agricultural space, often termed the villa rustica, demonstrates clearly the multi-purpose nature of land ownership which, as we have seen, is stressed in the accounts of the agronomists. Villa 13, for example, is the well preserved Boscoreale estate, a productive property which featured not only two wine presses and storage dolia, but also an olive press, threshing floor, and a bakery beside the main residential complex. In 36  See figure 4.  64  addition to the decorations of the domus, a large hoard of over 1,000 gold coins and fine silverware were discovered near the site, all suggesting a very wealthy and productive estate (Potter 1987, 96-7). The Boscoreale villa, then, as with many of the other sites in the area, was a comparatively large mixed-crop estate which clearly centered on wine production, but cultivated oil and wheat as well. It is also interesting to note that the distribution of these large villas around Pompeii follow a similar pattern to that displayed in the ager Cosanus, as they are mainly concentrated around the Sarno river or in relation to roadways, likely owing to the transportation afforded by these features for the goods that the villas produced (Jongman 1988, 123). This latter inference is admittedly specious, for although the natural inclination is to assume that most of the intensive wine production we see in the archaeological record was for exchange or market-sale, the small size of these villas suggests rather a small-scale production for mainly local consumption. Confusing the matter further is that several of the properties, while certainly much smaller than the estates of south Etruria, were extremely wealthy, like the Boscoreale estate. This suggests that the villa flourished economically, and it is most likely that much of the wealth came from the sale of its produce, though the owners of such properties certainly would have made money from other, unrelated ventures, such as the leasing of property.' Many of the villas in the area likely did engage in a significant level of wine production and exchange throughout Italy, though it must be noted that there was not a large number of such villas and that those which existed did not constitute the majority of Pompeii's territory, much of which was likely devoted to small-scale, mixed-crop  37  See below for a detailed discussion on leasing in Republican Italy.  65  production. The amphorae record does, in fact, support the conclusion that the villas in this area did engage in the production of wine for exchange. Pompeian amphorae are found all over Pompeii and elsewhere, and are particularly concentrated in certain deposits around Ostia, representing, for instance, nearly 30% of the finds at the Augustan era deposit of La Longarina.  38  Willem Jongman  does well to emphasize that, notwithstanding the misleading name, not all of the Pompeiian amphorae actually come from Pompeii. The explanation for the nomenclature is simply that most of the examples of this type have been found in or near Pompeii, likely for the obvious reasons of archaeological chance (Jongman 1988, 126). The epigraphic evidence is valuable here in distinguishing the provenance of some of these samples, as several dipinti on Pompeian amphorae actually suggest an origin in Surrentum. One amphora found at Ostia reads: "with Lucius Anneus and Marcus Trebellius as consuls, from Su(rr)entum (CIL.IV.5514)." CILIV.2556 (Vespasianic) and CILIV.9315 (Neronian) also read 'SUR' for Surrentum. It must be noted that these are only a few examples and that it is extremely difficult to determine the provenance and date for many of the samples which have been found, as the epigraphic evidence is much sparser than we would like. We must, nevertheless, conclude from the archaeological evidence that we do possess, which supports the ancient literary accounts, that a fairly active wine (and, to some extent, oil) trade from Pompeii and Surrentum to the wider Roman world was established by the middle first century AD, F-  at the latest. That this was also the case to a significant extent in the second century BC is, as mentioned,  38  See Jongman (1988, 125) for a distribution chart from this fmd.  66  usually assumed, however the evidence for this period is far less convincing. Again, the archaeological record is often simply silent about the past history of a large villa and its ownership. This lack of clear proof regarding the middle Republican period has led to alternate theories about the Pompeian economy during this time. One notable instance of dissent from the traditional latifundia school comes from Willem Jongman, who questions the prevalence of large villae rusticae not only for the third and second centuries BC, but in all periods. He asks how, if the majority of the land around Pompeii (and Surrentum) was devoted to wine and oil manufacture, were their citizens actually fed? (Jongman 1988, 98) Some of the staple goods consumed in these cities, surely, came from the large estates, for we saw at the Boscoreale villa some evidence for wheat cultivation along with wine and oil production. Yet Jongman thinks that the majority of grain consumed in Pompeii was actually grown on small farm sites, which he postulates lay mainly in the plain between the Sarno river valley and the more hilly area north of the city (Jongman 1988, 143). This is indeed possible, and perhaps even probable, especially given that, as we have stressed here, the wine-producing villas around Pompeii were small and few in number, leaving a great deal of territory on which these small farms could have operated. Unfortunately, small farms and unadorned 'cottage' residences are not prominently revealed by archaeological work, which means that their presence is as speculative as it is tempting. Even if we are to postulate significant levels of small-scale grain production from Pompeii, a more fruitful way to address Jongman's concern about the origins of the staple goods needed to feed Campania's population is to look at the rest of Campania. Indeed, one of the most crucial elements in Campania's agricultural economy, which is too often overlooked or under-represented in Pompeii-focused studies, is the importance of wheat production outside of Pompeii itself. This 67  was perhaps particularly salient in the second and early first centuries BC, as Martin Frederiksen suggests, "before Roman markets became dependent upon provincial supplies" for their grain needs, notably Sicilian (Frederiksen 1959, 124). Nicholas Purcell agrees with this assessment, noting that it is only in the Imperial period when provincial grain tithes came steadily into Rome that the Italian countryside was allowed to convert a majority of its land to wine production (Purcell 1985, 17). Pompeii, he believes, was one of the few cases where wine production was practiced in the middle Republic, but that this was not the case also in other areas, even the rest of Campania. This is worth stressing, for the evidence from Pompeii is, of course, excellently preserved and well studied, which often leads to exaggerating the importance of a city which was in many ways small and relatively unimportant. The peculiarity of Pompeii is borne out well in the ancient literary sources, for the Roman writers themselves mainly focused on the other parts of Campania rather than Pompeii, particularly Puteoli, one of Republican Rome's principal ports, and Capua, a major Campanian city much more prominent than Pompeii. Pliny the Elder, in his Historia Naturalis, praises the various wheat types which Campania produces, noting that the plains of Campania have "spongy" soil which is "as porous as pumice" and absorbs the abundant rainfall coming at the foot of the large mountain range, thereby allowing for excellent harvests (Pliny, 18.29.110). Pliny states quite explicitly that the best soil, and thus the most important area of Campania agriculturally is found in the triangular region between Puteoli, Cumae, and Capua, north and east of Pompeii (Pliny, 18.29.111). Even in the Republican period Campania was seen in this same light. Varro in fact mirrors Pliny's description of Campanian soil as porous, absorbent, and producing excellent cereal crops (Varro 1.20.4), and Cicero predicts the focus on Capua claiming the area to be "the capital of our public 68  inheritance, the Roman people's most valuable possession, our reserve of grain, our storehouse (Agraria, 7.21)." The vine cultivation from Pompeii, while it does seem to have been the prominent economic activity around that city, is thus revealed to be only a minor economic feature of the region as a whole and one based largely on the specific benefits of Pompeian territory, being situated on more hardy, hilly land at the foot of Mount Vesuvius. Many of these literary accounts, moreover, belittle the quality of wine produced around Pompeii as relatively harsh and unappealing (Pliny 14.4.22; Columella 3.2.10). In short, wheat cultivation in Campania was stressed in antiquity far more than wine or olive cultivation around Pompeii and Surrentum. To understand Campania's agricultural economy as the Romans understood it, then, we must look more closely at the grain producing land of the plain, particularly around Capua. Capua, along with neighboring Atella and Calatia, supported Hannibal against Rome from 216 until 211 BC, when the Romans effectively drove Hannibal from the region. Rome then reclaimed these cities and fulfilled Fabius' promise of severe punishment for supporting the Carthaginians; "the entire territory and public buildings were made the property of the Roman expanse of extremely fertile plain (Frederiksen 1959, 82)." The area was now ripe for Toynbee's latifundia building, for the fruitful land around Capua, according to Toynbee, "would have made plantation-agriculture, producing wine and oil for sale at Rome, a profitable enterprise (Toynbee 1965, 228)." Here however, unlike at the less important towns of Pompeii and Surrentum, exploitation centered on grain and other cereal crops rather than wine or oil. Toynbee suggests that this was not simply because of the natural benefits of Capuan soil, as the ancient commentators suggest, but the result of political considerations. He maintains that Rome's need for a steady supply of grain at the end of the third century BC, while it was still engaged with Hannibal 69  and the Carthaginians in south-eastern Italy, outweighed the potential profit to be made from largescale wine production in the area. The Roman government therefore took great measures to preserve many of the pre-Hannibalic peasant farms, hoping thereby to ensure that the production of grain which had been going on in the area for some time continued unabated. The penalty faced by Capuan citizens for opposing Rome, then, was not to lose their property, but simply to lose ownership of that property. For Toynbee believes that many Capuan peasants became tenants leasing predominantly the same land that they had owned before the Second Punic War, which had now been taken over by aristocratic Romans operating as absentee landlords (Toynbee 1965, 1236). The persistence of small peasant holdings around Capua, however, is seen by Toynbee as an aberration, who maintains that most of the Italian peasantry were indeed overrun by cheaper slave labor after the Hannibalic Wars. Still, this raises an extremely important point, for while it is clear 39 that Capua was expropriated by Rome after 211 and that several slave-staffed villas appear in the settlement record around Pompeii and Surrentum, it is not clear what happened to the peasants who were supposedly displaced by latifundia formation, either in Campania or in any other part of Italy. Toynbee's suggestion that many of the expropriated lands of Capua were leased back either to their original owners or to other Roman farmers is a very tempting explanation, however we simply do not have the archaeological or epigraphic verification that we need. Until some strong evidence emerges to explain precisely who was working the land in Capua, this explanation must remain merely speculative. Unfortunately, around Capua our evidence is particularly sparse, and the  39 •  Livy, for example, mentions that "all their land and houses were made public property of the Roman People (26.16)."  70  settlement history is poorly understood and cannot really shed any light on the situation. Thus, while the literary sources all unequivocally stress Capua's grain growing, it is still not clear what agricultural form this cultivation took, in the second century BC or even sometime afterward: slave-staffed latifundia, peasant-owned smallholdings, or quasi-dependant tenant fanners. What we can state with reasonable certainty is that the area around Capua and elsewhere in eastern Campania, as well as parts of north-western Apulia, was the subject of repeated attempts to redistribute land; most of the boundary marking cippi established under Tiberius Gracchus' land redistribution commission in 133 BC were found in that part of Italy (Toynbee 1965, 239), and even in the middle of the first century BC Cicero decries the Caesarian Rullus' attempts to found a colony around Capua for the underprivileged poor, coopting land held mostly by the wealthy classes (Agraria 5.16). P.A. Brunt likewise notes the strong correlation between civil unrest, land distribution programs, and military recruitment' in Etruria, Umbria, Apulia, Campania, as elsewhere (Brunt 1962, 73). These are not definitive proofs by any means, but they do suggest that, at least between the middle of the second and first centuries BC, much of the ager publicus in the area which had become available after the Hannibalic Wars was in fact owned by the noble classes, however again this reveals nothing about the size of the holdings or whether slave or tenant labor was predominantly used. In the following chapter we will look more closely at the evidence for latifundia agriculture in Sicily, the other area, along with Campania, which is most often presented as a clear expression of the practice.  40  Brunt argues here that high levels of recruitment in a certain area suggest a large population of dissatisfied rural smallholders.  71  Chapter 8 Slave Labor and the Grain Trade in Sicily As mentioned, the traditional approach to the agricultural history of Sicily is quite similar to that of Campania, with the common belief that both areas experienced widespread latifundia formation prompted by Rome's conflicts with Carthage. The economic history of Sicily deserves its own volume and the main concern of this thesis is Peninsular Italy, however by way of comparison it is of value to briefly look at some of the evidence for latifundia formation in Rome's first province. This is particularly so because the agricultural history of Sicily during this period actually shows many of the same patterns as were seen in Italy, contrary to the common account. Indeed, most scholars assert not only that latifundia agriculture was the dominant form of production in Sicily, but that this process began earlier here than anywhere in Italy, starting with the close of the First Punic War in 241 BC, but benefitting also from Rome's imperialist military expeditions in the early second century BC which allowed the noble classes to procure enough capital and slave labor to allow for the establishment of extremely large and productive rural villas. Thus, according to the influential account of Moses Finley, already in the early second century "Sicily was well on its way to becoming the classic land of large estates, the latifundia of the Romans (Finley 1979, 131)." That the agricultural economy of Sicily throughout the Roman period was dominated by latifundia, by slave labor, and by the intensive production of grain for export is widely accepted in modern scholarship. This account is supported also by some of the ancient literary sources, particularly the writing of Diodorus Siculus who asserts that the island's fields in this period were essentially overrun by slave gangs. As with Campania, however, more recent archaeological work, along with some contradictions in the accounts of the ancient commentators, has revealed that the actual 72  historical situation was far more complex than traditionally thought. In 263 BC a momentous event for Roman-Sicilian relations occurred, when the Syracusan king Hieron II abandoned his Carthaginian alliance and sided instead with Rome. Hieron was forced to compensate Rome for having formerly sided with Carthage, but was not heavily punished apart from this and was still allowed to retain control of Syracuse. Moses Finley notes that from the time of this alliance "the history of ancient Sicily is part of Roman history, with the most profound effects not only on Sicilian politics but also on its economy and civilization (Finley 1979, 113)." Through this alliance, the territory controlled by Hieron was able to prosper in a way unmatched by the western and southern parts of Sicily, which were severely punished for supporting Carthage against the Romans after the former's surrender in 241 BC. Thus, while much of the island suffered from heavy taxation, Roman political control, and the "commandeering and destruction of crops and cattle along with the killings, enslavements and physical ruin of many cities," Syracuse and its environs were left largely intact (Finley 1979, 116). Hieron's territory suffered a corn tithe paid to Rome which was not overly punitive nor disruptive, so that the city's economic and political stability do not seem to have been seriously compromised (Wilson 1990, 18-20). When Hieron II died in ca. 215 BC, however, the Syracusans allied once more with Carthage, eventually causing them to lose much of their economic and political independence after Rome's victory in the Second Punic War. The Romans at this time expropriated large tracts of property, turning much of the island's productive landscape into ager publicus, and applied the one-tenth corn tithe known in the Syracusan territory throughout Sicily. This tithe helped focus much of Sicily's agricultural output onto grain and may have actually increased production from earlier times, as Roman investors and the newly opened Roman markets supplied an excellent stimulus for export production (Wilson 73  1990, 20). Livy certainly held this view, recording that Sicilian grain shipments helped Roman soldiers withstand a siege at Tarentum in 210 BC (Livy 26.39,1), as well as feed Scipio's armies campaigning in Africa at the close of the war." The traditional account purports that a concomitant effect with the increased export production was the concentration of a majority of Sicily's new-found ager publicus into large slavestaffed units, it being assumed that such latifundia agriculture was the most efficient and profitable means to produce the large amounts of exported grain which the Roman markets encouraged. To support this notion scholars often rely on Diodorus Siculus, who details the two major slave revolts which Sicily experienced from roughly 135-2 and 104-100 BC. The First Slave Revolt was led by Eunus, slave of the wealthy landowner Damophilus, whom Diodorus claims was able to amass a huge force of discontented servile workers laboring in Sicily's many latifundia. Concerning this revolt Diodorus "paints a graphic if not always accurate picture of the conditions that led up to it the brigand-herdsmen, the administrative anarchy, the insolence and brutality of wealthy slaveowners (Green 1961, 10)." The implication inherent in such an account is that vast numbers of slaves must have been accumulating on Sicilian farms from the late third century BC, mistreated and neglected by the latifundia owners,' in order to account for the large numbers of slaves available to fill Eunus' ranks; for, according to Diodorus, at its height the rebellious slave force reached two hundred thousand strong (Diod. It goes without saying, of course, that these figures must be treated with extreme scepticism, but they nevertheless indicate that the revolt was seen as a major uprising and a significant threat to Roman interests. Apart from simply the 41  In 204 and 203 BC. (Livy 29.36.1; 30.3.2)  42  See for instance Diodorus 34.5.2.  74  numbers involved, however, there are some perplexing problems with Diodorus' narrative. The first major concern is that, as Gerald Verbrugghe points out, "in his account of the Slave Wars the main industry of the island is not cereal production, but animal husbandry (Verbrugghe 1972, 540)." Clearly, this does not fit the traditional 'breadbasket' approach to Sicilian agriculture. We might, of course, allow Diodorus some licence for exaggeration, particularly as slave herdsmen would have been probably the first to revolt given that they had more freedom of movement and greater access to weapons than their counterparts working on planted crops. Indeed, he mentions that Damophilus owned both cattle and grain-producing land, and we might expect a similar scenario for many of the wealthy landowners (Diod. In any event, Diodorus' main point would not be compromised, namely that the vast majority of the agricultural output of the island was produced by maltreated slaves. The other problem with Diodorus' account, which is more difficult to remedy, is that parts of it completely contradict other writers, particularly Cicero. In his Verrines, the statesman focuses almost entirely on the importance of grain cultivation on the island and states quite explicitly that much of Sicily's land was worked by independent smallholders: he worries about the plight of those "who cultivate single acres, who never cease from personal labour, of which class there were a great number, and a vast multitude among the Sicilians (ii in Verrem 3.27)." The accounts of these two authors seem on the surface equally valid, yet completely contradictory. If we were to put our faith in Cicero's narrative, for instance, claiming that Sicilian land was worked mainly by unfettered peasants up to the 70s BC, how could we then explain Diodorus' tale of the Second Slave Revolt from 104-100 BC, which, although not quite as large as the first, was still purported to have involved a substantial force of forty-thousand slaves? (Diod. 34.8.2) 75  Roger Wilson identifies this intractable situation, explaining: that there were plenty of small- and medium-sized holdings in Sicily is abundantly clear from Cicero, and we hear so much about them precisely because their owners were the principal target of Verres' attacks... But that as well there were plenty of huge estates run by gangs of discontented slaves is also self-evident from the Diodoran evidence - otherwise Sicily would never have experienced two serious slave wars (Wilson 1990, 21).  Archaeological evidence, unfortunately, has done little to increase our knowledge of the settlement history or economic situation of Sicily during this period. We are left, then, to conclude merely that Sicilian agriculture consisted of slave-staffed latifundia as well as peasant-worked smallholdings, both existing in the same areas and functioning in a single, complementary economic system. This is not, admittedly, a very bold assertion, but it does at least redress the tendency to exaggerate one mode or the other and treat the Sicilian economy as a monolith. Wilson reminds us too that the particular distribution of large and small sites would vary widely from region to region, depending on the area's particular political history and the fertility of its soil (Wilson 1990, 21). The exact nature of agriculture in Sicily thus remains largely a mystery. It is tempting to speculate a system of export-producing latifundia complimented by smaller plots which focused on more local markets, perhaps worked by tenants, similar to the regional trade network identified in south Etruria. Again, this is only a hypothesis, nor does it help to resolve the contradictions in the ancient authors about whether wheat, which is usually stressed, was the primary export, or if Diodorus is correct in implying that much of Sicily's output came from pasturage. The evidence for the ownership, dating, and productivity of Sicilian farms is simply as yet inconclusive or lacking altogether. What we can safely conclude, I think, is that the familiar notion of a displaced and ruined peasantry is a gross exaggeration, and that the effect of the Punic Wars on both Italy and Sicily's peasant populations was much more complex and deserving of greater scrutiny than scholars have 76  traditionally given it. In a rare exception to the traditional line Peter Garnsey notably cautions against discounting peasant labor in Rome's agriculture economy, and cites P.A. Brunt's analyses of "military manpower resources and recruitment practices... which show that the rural districts contributed a bulk of the very considerable number of recruits enrolled in the army in the first century BC as well as in the second," meaning that there were indeed significant number of citizen landholders who qualified for military service (Garnsey 1998, 108). Garnsey also reminds us that, although small sites were amalgamated into large latifundia in some areas of Italy, in other areas new peasant holdings were created, notably in centuriated, agricultural colonies (Garnsey 1998, 98). We have seen already examples of this around Cosa, Saturnia, and Apulian Luceria, and it is likely that Capua served a similar purpose, offsetting the larger wine-producing villas which were allowed to accumulate around Pompeii and Surrentum. Here again we must stress that this is purely speculative, however it offers an interesting framework to explain the existence of small farmsteads in areas where there were also several larger villas. Throughout Italy as a whole, therefore, and likely also in Sicily, the rural peasant population does not seem to have actually declined, even if it was shifted from one area to another through the creation of colonies. Truly, the role of these peasants in the agrarian economy of Republican Italy is poorly understood and requires closer scrutiny in its own right before we can come to any conclusions about the nature of the interaction between small and large holdings, or between wine and grain cultivation, in Campania, Sicily, or anywhere else.  77  Chapter 9 The Role of Peasants in the Italian Economy While in the past peasants have been largely ignored by Roman economic historians, one important consequence of the recent archaeological surveys which reveal a prevalence of small farm sites in many regions of Italy has been that greater attention is now being paid to the role of free, independent labor. It is generally assumed, as we have discussed above, that most of the small sites were worked by free Italians rather than slaves, so that many scholars have recently attempted to articulate a theoretical framework within which we might envision small and large farmsteads, free labor and servile, interacting in a single economic system. The most persuasive of these endeavors holds that the large cash-crop-producing villas and the small free-holdings of a given area were actually economically co-dependent. Willem Jongman, following this line, argues that the majority of the small grain-producing farms which he believes existed around Pompeii were actually owned by wealthy Roman investors who then leased the plots out to peasant or freedmen workers to cultivate, much like Toynbee's ideas for the leasing of Capuan land (Jongman 1988, 143). Many scholars also have used this argument to explain the settlement pattern in the ager  Cosanus, which we have discussed previously. For example, D.W. Rathbone suggests that "the slave staff could not satisfy the full labor requirements of the estate of Sette Finestre" so the owners of the villa resorted to hiring wage laborers from the pool of peasants who held small properties in the surrounding area (Rathbone 1981, 12). While it is very difficult to verify these hypotheses with our available archaeological or epigraphic evidence, it is consistent with the accounts given by Cato and the other agronomists, who all acknowledge the possible need for a villa owner to hire out temporary work (Cato, 1.4.1).  78  This labor could be, as Rathbone suggests, seasonal work used to supplement a villa's full-time slave staff during peak times of demand, however a villa owner could also benefit by leasing some land to tenant farmers who might grow staple crops used to feed the inhabitants of the wine- and olive-producing estates, the majority of whose land would have been devoted to producing those cash crops: what Jongman envisions around Pompeii. Even Andrea Carandini, who normally supports widespread use of slave labor and latifundia production, nevertheless agrees that these phenomena did not entirely wipe out the peasant workforce. He argues rather that small-scale artisan work and peasant farming produced goods for local markets which complimented the wider-market export production of the larger villas (Carandini 1981, 252-3). Daniele Manacorda holds similar conclusions about the large villa in the ager Cosanus owned by L. Domitius Ahenobarbus, discussed in a previous chapter. Manacorda argues that this villa specialized in producing wine using slave labor, but that Ahenobarbus also owned property which he leased out to peasant tenants (Manacorda 1981, 45). All of these are indeed intriguing possibilities for the role played by peasant labor, but we need still to look more closely at the actual evidence that we have for peasant labor and to try to gain a better understanding of who these peasants were and what kept them occupied. Rural Tenancy Certainly the role played by tenancy is one of the most intriguing factors to be considered regarding rural peasant labor, but one that has often been neglected. Obviously the idea that large numbers of peasants were employed as tenants has not often been championed, for the traditional notion that the Italian countryside was overrun with slave labor by the second century BC allows little room for free farmers. A.J. Toynbee, for instance, although he does suggest tenancy as a 79  viable explanation for Capua's grain producing farms, mainly stresses the displacement and ruin of the Italian peasantry in favor of slave-labor production. Moses Finley too argues that "the normal tenant was a man of few resources and without his own land, a failed peasant, a 'superfluous' peasant's son, or a dispossessed peasant," someone who did not feature prominently in Italy's slavestaffed villa economy (Finley 1985, 114). P.W. de Neeve and Pasquale Rosafio also both follow this line; each has produced a study on the role of tenant labor in Italy and each has concluded that the practice was negligible, at least during the Republic. de Neeve differentiates tenancy from other forms of contracted work, such as temporary wage labor and contracted work, arguing that true tenancy was a development mostly of the Imperial period. Rosafio likewise agrees that "in the postHannibalic period...the role played by tenancy was insignificant in the Roman economy" and supports the idea that tenancy is not really acknowledged before Columella in the middle fo the first century AD. (Rosafio 1993, 155) On the other side of the argument stand scholars like Lin Foxhall, who explains that disregarding tenancy as a prominent Republican activity is misguided and that we as economic historians need to: raise other possibilities for the interpretation of these small-scale and medium-scale sites, within the framework of a general discussion of tenancy as a source of labor in Roman farming. This analysis is based on three premises. The first is that tenants were a vital source of agricultural labor in the Roman world in all periods. Second, tenants operated within a framework of social and economic dependency relationships.... Third, not all tenants were subsistence farmers. A wide range of tenants can be both postulated and documented (Foxhall 1990, 97).  While I agree with all of Lin Foxhall's premises, it is important to recognize that there are many challenges with trying to prove that Republican tenancy was widespread, most notably the facts that most tenants would have worked small plots leaving little archaeological record and that  80  tenants cannot always be distinguished in our sources from other forms of dependent labor.' Despite the difficulties that such an endeavor entails, it will nevertheless be rewarding to examine the evidence for tenancy that we do have and see to what extent we can reasonably apply the practice to Italy's agricultural economy in the second century BC. In his very informative study, de Neeve defines Roman tenancy as a legal contract brokered between the owner (locator) of a property and the leaser (conductor), whereby the conductor pays a certain fee for the right of usufruct on the property for a certain period (de Neeve 1985, 7). Confusing matters is that these contracts were often not written and that the period of tenancy was not standardized, however five-year leases seem to be most common. Not only that, but very similar contracts to tenancy agreements were made for wage labor, where again a locator agreed to pay a fee for a certain job performed by a conductor (operis faciendi) (de Neeve 1985, 10). It is to this latter relationship which de Neeve and many other scholars believe Cato was referring to at 1.4.1 where he mentions hiring temporary workers, being day-laborers rather than true tenants. De Neeve does admit, however, that wealthy landowners may have held plots far apart from each other so that the same person could own a latifundia worked entirely by slaves as well as smaller plots, unconnected to the larger property, which were leased out to peasant farmers (de Neeve 1985, 1102). This, he believes, would explain the appearance of tenancy in the evidence, especially the accounts of the agronomists and the legal texts. Still, de Neeve wants to say that tenant holdings were inconsequential compared to latifundia and that agricultural specialization after the  43  Even nominally free tenants often probably had economic or social obligations to the landlord in addition to the rent paid on the lease. See Foxhall (1990, 103) for the idea that tenants were typically freedmen of a landlord, who retained loyalty or some indebtedness to their former master. Compare also Garnsey (1980, 39) discussing the popular idea that many Imperial-era tenants were de facto slaves or 'serfs'.  81  Hannibalic Wars, namely the intensive plantation-style grain, olive, and vine cultivation as appears in Cato, precluded the possibility of tenancy: the need to keep down the costs of producing these cash crops in order to obtain a profit, and a desire to have absolute control over all aspects of production, meant that slavery was always the preferable alternative to tenancy. It must be stressed, however, that de Neeve's arguments about how the economic conditions of villa agriculture disallowed widespread tenancy are highly specious and are based largely on an anachronistic, modernizing approach to ancient farming; for he believes that, since Cato was, as we have shown, clearly concerned with business interests and since slave labor is in modern times understood to be less expensive than other forms, Cato therefore could not have considered employing hired laborers or tenant farmers to any significant extent. By no means am I suggesting that Cato did not advocate and assume a significant slave labor force for any large villa, but we saw that his actual prescriptions for agriculture allow for a wider variance in practice and a more diverse set of concerns than de Neeve and those scholars who follow his line of argument purport. Simply because Cato, as with Varro and Columella, addresses his work only to wealthy Romans investing in large, productive landholdings does not exclude investments in other forms of property, such as small plots leased out to tenants. Indeed, all of the agronomists imply that these wealthy landowners will have other concerns besides the maintenance of their villa, including political ambitions, keeping a luxurious residence, and other forms of money making. De Neeve I think too readily undervalues these other concerns, which could all be fulfilled by leasing: diversifying one's investments by holding land that was leased out to a conductor in addition to his larger villa would supply the locator with a steady income from the lease as well as assure productivity on land without personally overseeing it (a tenant in this case would serve a similar 82  function as a vilicus, but with the added benefit of having to pay for the usufruct); also, setting up a freedman or client as the conductor could enhance the landowner's status as patron, to be used for political purposes. All of this is admittedly somewhat speculative, as it is nowhere made explicit by any of our literary sources, yet it demonstrates that there are other scenarios that we can envision for wealthy Romans investing in land besides maximizing profit through the exclusive use of slave labor. While it is important to define the theoretical possibilities for tenant labor, however, we must always temper our analysis with an understanding that the actual evidence for the practice is scanty at best. One of the few places where we do have unambiguous reference to tenancy is in the legal texts, which are probably the most important evidence on the matter. A major problem with these texts, however, involves contending with the difficulties in dating, which are numerous and largely intractable. As Dennis Kehoe acknowledges, any scholar attempting to analyze these texts must worry about "the ambiguity concerning the extent to which the Digest and the Justinian Code genuinely preserve the writings of the classical Roman legal authorities (Kehoe 1997, 9)." Moreover, the few dates that we do have for the writings of these Roman jurists whose opinions make up the Digest are vague and at best places them no earlier than the late Republic, so they may not represent any third or second century BC practices at all. Nevertheless, recognizing some of the concerns with which the jurists were contending can help us understand some of the ways tenancy was conceived by Romans, and can perhaps define some of the possibilities for tenancy that we can use in analyzing the practice during the middle Republic. One of the most striking features that comes out of Kehoe's analysis of the legal sources in this regard is that the Roman nobility, unlike modern capitalistic ventures, sought profit not in 83  maximized gain but in the security of the return on an investment (Kehoe 1997, 77). Kehoe picks out as exemplary Digest 33.1.21 and 32.37, both of which deal with an inheritance of land and both assume a fixed annual income from the estate (Kehoe 1997, 100). Land was the preferable investment of the upper classes, as it provided a fairly predictable yield with little capital input after the initial purchase. This fits with the attitude of many aristocratic Romans who felt that moneylending and other forms of activity were too volatile and should, therefore, be shunned in favor of an agricultural villa." Under these constraints, tenancy would be a useful means for a landowner to supplement the profit gained from his own estate, for it provided a fixed return, namely the price of the lease, while at the same time deflected most of the risk involved in agriculture onto the tenant. It also seems that rents paid on leased properties were normally paid in cash, not in kind, further suggesting that stability and predictability in return was a chief concern for the propertied classes (Kehoe 1997, 148). It is important to note, too, that the legal sources bear out the same tenancy as we saw in the treatises of the agronomists, as it is generally assumed that most large landowners will be 'absentee landlords' and have multiple, dispersed holdings (Kehoe 1997, 145). Kehoe indeed argues quite persuasively that Columella's recommendation that landowners manage their own properties directly, rather than hand over land to tenants with little oversight, suggests that the latter was in fact the normative practice among busy aristocratic Romans by the first century AD. As much as realizing why tenancy was sought after by wealthy landowners, it is equally important to understand the benefits (as well as the drawbacks) of leasing property for a relatively poor Roman. Again, tenant farmers appear in our records as a diverse group, and the distinction between them is often quite vague. Even slaves, always a tricky subject in Roman law, were 44  Recall, for example, the introduction to Cato's De Agri Cultura.  84  sometimes treated as though they were peasants, whom Kehoe terms servi quasi coloni (Kehoe 1997, 166). Absentee landlords, it seems, often set up slaves as tenants on some parcel of their land, with the rent paid out of the slave's peculium, thinking that a slave acting as a tenant would treat the land better than if he were simply 'assigned' to tend to the property as a vilicus. Digest. 15.3.16, for example, deals with just such a consideration and makes it clear that the jurist saw the slave as a true tenant, akin in status to a free peasant, not simply part of the landlord's chattel.' Freedmen who had lingering obligation to their former masters would appear in this same light as well and have debt or other social/political obligations to the landowner. Many free-born peasants too had motivation to lease property, even those who may have owned some land of their own. As we have discussed throughout this study, Rome established centuriated colonies throughout Italy, wherever she extended her military dominance and political hegemony, from the later fourth to the second century BC. These colonial holdings were given largely to the veteran soldiers who had won the territory, or to other landless peasants and any 'surplus' population. Yet, most of the evidence suggests that these land allotments were quite small, usually under two iugera, and often in marginal land; the centuriated plots, then, would probably just barely maintain subsistence for a family of four (Brunt 1971, 194). While this is, again, somewhat speculative, Peter Garnsey is correct in asserting that "we have to suppose that colonist farmers supplemented their meager resources by working as tenants or laborers on neighboring farms (Garnsey 1998, 99)." This would indeed explain much of our ambiguous evidence, since it could not only account for the persistence of small sites alongside large farmsteads in areas that we would expect to be largely under the control of wealthy elites, but it also places tenant 45^• •  This is what slaves usually are seen to be, as, for example, in Varro 1.17.1.  85  relationships squarely into an economic system that we know to be dominated by the elite. Settling poor Romans onto plots that were just barely sufficient for subsistence afforded the Roman state a sizeable land-holding population from which it could draft soldiers to fulfill its military projects while at the same time providing a ready source of tenant and wage workers to supplement the labor and staple-goods shortfalls inevitable from slave-staffed villas which focused their energies onto cash-crop cultivation; "there was no humanitarian inspiration, therefore, behind these colonies. They served the interests of the state and of the wealthy landowners and war-lords who directed it (Gamsey 1998, 99)." Again, we must never lose sight of the fact that very little evidence positively confirms that tenancy was widespread in the Republican period and that much of the argument for the phenomenon is from silence. I do not think, however, that simply recognizing the limitations of the evidence should discourage us from examining tenancy as much as we are here, for it is worthwhile to offer tenancy as at least a possible explanation for the type of economic activity engaged in by peasants after the Hannibalic wars. Nor is rural tenancy the only possible explanation, for we need also to bear in mind that the Italian peasantry was never a homogeneous group but consisted of tenant farmers, freedmen employees, wage laborers, and free-born owner-cultivators and that there was an equally diverse mix in both the rural and urban environments. The same person, moreover, could exist in more than one of these categories at different moments throughout his life, further complicating the situation. While we certainly must acknowledge that small-plot farmers were a heterogenous group, it is equally important to stress that all of the different labor groups involved in Roman agriculture operated in a single economic structure and were largely interdependent. Peter Gamsey expresses this relationship well, noting that "the slave-owning landlord needed labor 86  from outside the farm for essential operations such as harvesting and possibly ploughing" and that this labor shortfall was met primarily by the "owner-cultivators and small tenant-farmers" who surrounded the larger villas and needed the work to supplement the meager income gotten from their own properties (Garnsey 1980, 3). Another part of the interdependence was the market interaction between a villa's goods and those from the smaller plots. Indeed, one of the few elements that clearly distinguishes tenant farmers from owner-cultivators is that the former group cannot rely on mere subsistence levels of production, for they have to cover also the rent paid on the lease. This means, then, that surplus production was an essential precondition for any sort of tenancy (de Neeve 1984, 65). Just as important, moreover, was a market for this surplus, without which surplus becomes simply waste. In an article discussing grain cultivation in Italy, Walter Scheidel aptly reveals where these markets may have appeared. While he agrees with the traditional notion that, even by the second century BC, "the city of Rome depended on the importation of large quantities of grain from overseas," he questions whether the same can be applied to all of Italy (Scheidel 1994, 159). Indeed, non-Roman Italians, particularly those living away from the chief harbors, did not benefit from these grain imports as much as those who lived in and around the city of Rome, but had to rely rather on their own resources. It is in this context that we must consider the rise of the large, slave-staffed villas which devoted their production to exportable goods, notably wine and oil, as we saw with parts of Etruria and Campania, and possibly also Sicily. In such a context, then, it would be prudent and natural for these villas to rely in some part on outside production, both from tenancy and from peasant owner-cultivators, for their grain needs. In other words, the very rise of slave-staffed villas would have created a market for the goods produced by other agricultural units. This is, indeed, a far cry from the primitivist school of the Roman 87  economy, which hardly allows any consideration to produce steady, anticipated, and marketable surpluses for even the wealthiest of landholders, let alone for relatively poor and small-scale tenant farmers. Yet we must go beyond the limits of the primitivist view, however without of course asserting a strict verisimilitude with modern practices, in order to account for the varied farm sites and complex economic interaction / interdependence that we see in all of our sources. Urban Peasantry  Another important facet of this intricate economic interaction is the urban sector, something we have looked at briefly in this study but which deserves some further attention here. Peter Garnsey sums up succinctly the general attitude of most scholars concerning the interaction between rural and urban economies: In emphasizing the predominance of rural property as a source of income, they are, indirectly, making a statement about the balance of trade between the rural and urban sectors of the economy. They are stressing the flow of goods from the rural to the urban sector. The cities are viewed as markets for agricultural produce. There was also a trickle of goods, consumer goods, back to the countryside from the cities. But on the whole, rural landowners pocketed the returns from sales of agricultural produce either directly in their capacity as agricultural producers, or indirectly through the rents they charged producers (Garnsey 1998, 70).  What we need to appreciate, then, is the amount of activity generated within the urban sector itself It was remarked before, discussing agriculture in the ager Cosanus, how urban markets often developed in response to increased production in their hinterland in order to provide "the ways and means for the dissemination of goods and produce (Potter 1987, 160)." Helen Patterson, who agrees with this notion, notes furthermore that rural production and urban manufacturing throughout the Italian peninsula, whether small or large scale, are closely related and often function in a symbiotic manner; namely, produce from the countryside is transported to a city and either exchanged for a good manufactured in the city or finished in an urban shop, as for example with the fullering of wool (Patterson 2004, 67). Put more simply, the economic division between town 88  and country in Republican Italy was simply not as pronounced as is often argued. That there was significant economic interaction between the rural and urban sectors seems fairly evident, but nevertheless many theories about the ancient economy that emphasize the landowning elite hold that this close relationship mainly benefitted the wealthy absentee landowners who owned villae rusticae in the hinterland, but who spent most of their time in cities. The urban market was, by this reckoning, a cyclical arrangement where wealthy urban dwellers consumed the produce grown on the farms which they themselves owned. In such a 'consumer city' model, goods produced on small farmsteads were consumed primarily by the inhabitants of that farm, essentially subsistence farming. Thus J.E. Skydsgaard suggests that the poor owners of Italy's small farms did not participate in market forces at all, and therefore "from the economic point of view they are of little interest (Skydsgaard 1980, 68)." This is based on the belief that small farms were not productive enough units to supply the stable surpluses necessary to contribute in a market system. This may have been the case in certain instances, but I do not think that we can assume a general lack of surplus agriculture on small farms throughout Italy, particularly if we are to believe the theory that tenancy 46 was fairly prevalent. In any case, focusing on the consumer aspects of urban life undervalues the output from cities, an underappreciated aspect of the Roman economy. Although scholars traditionally have focused on the rural economy, it is true that the same inflow of money in the third and second centuries BC which allowed wealthy Romans to invest in rural property also prompted many building and public works projects throughout Italy (Harris 1971, 41). H.C. Boren points out that the middle of the second century was a period of "unusual  46  See above for discussion of the idea that the need for a tenant farmer to furnish the cost of the lease presupposes a surplus.  89  spending at Rome" on temples, aqueducts, and similar massive building projects (Boren 1958, 894). This situation was mirrored in most of the significant urban areas, particularly the city centers of the many colonies which were founded by Rome in this period where, temples, bathhouses, market buildings, houses, and of course major road works were required to accommodate the new Roman inhabitants (Potter 1987, 73). All of these projects would, of course, have required massive amounts of labor, and there is little evidence suggesting that in the Republican period this workforce was predominately servile. Rather, it seems that such building projects were a major source of income for urban peasants and for rural dwellers hiring out their labor on a temporary or seasonal basis. The work needed by these massive building projects, then, was the urban equivalent to the seasonal labor required on large cash-crop producing latifundia, and it seems reasonable to assume that the opportunities were taken advantage of to a similar extent. Nor is it only construction that we see in Roman towns, but there is a great deal of evidence for metalworking, pottery manufacturing, and other urban craft work.' Many Italian cities, in fact, show numerous signs that the "urban economy was based upon small-scale workshop production (Laurence 1994, 55)." Even among those scholars who do acknowledge the importance of urban manufactures, however, many scholars want to place this industry squarely within the villa system, reinforcing the cyclic market interaction of wealthy landowners. J. Aubert, taking a 'primitivist' stance on the ancient economy, suggests in this vein that most of the exchange between farms and cities was not capitalistic, envisioning it rather as two different sectors of the same production unit; he believes that most large agricultural villas had a concomitant urban section "producing building  47  We have seen the evidence for pottery manufacture and metalworking in Cosa and for wool fmishing, millstones, and bakeries in Monte Vairano, Bovianum, and Larinum.  90  materials, containers, and other types of artifacts...[whose] primary purpose was to provide for the needs of the [rural] villa" to which it was attached (Aubert 1993, 73). This cyclical exchange may indeed explain some of the urban manufacturing that we see in many Italian cities, but I do not believe that it can fully account for all aspects of the urban economy. There was also, as David Engels adeptly points out, a wide variety of services provided in Roman cities that do not fit with Aubert's villa economy: bathing, lodging, food and beverage retail, religious sites, etc. It is, of course, very difficult to determine from our available evidence whether a shop in a city was independently owned and operated or was, in Aubert's model, owned by a wealthy property holder and worked on his behalf by a slave or freedman. Still, there is no evidence to suggest that these small-scale urban industries, whether manufactures or services, were conducted by slaves. Rather, it is probable that peasants accounted for the majority of the urban economy, whether acting as owner-operators, tenants, or perhaps workers contracted by villa owners, as Aubert might suggest. Indeed, in the same manner that wealthy landowners are seen investing in rural property which they leased out to tenant farmers, as discussed in this chapter, they also seem to have been active in owning and renting out urban property. Peter Garnsey notes that, interestingly, many Romans felt that renting urban property in fact furnished greater returns than rural, however carrying also greater risks (especially damage through fire, or rapid depreciation by other nearby building projects) (Garnsey 1998, 68). Admittedly, most of the literary sources for this urban leasing, as was the case with rural tenancy, are from the later first century BC and afterwards." Still, it must be the case that leases on urban space would have provided a wealthy Roman the same sort of fixed return as a lease on rural property, which we saw was one of the 48  See Garnsey (1998, 63-8) for a more detailed discussion of the sources, particularly Cicero.  91  primary concerns of the landowning class. For the leaser, moreover, the property could have provided space for urban craft work, manufacturing, and retail that show clearly in the record of many Republican cities as well as housing, if not a combination of all. Pompeii, naturally, produces the best and most convincing evidence for the urban economy. Many buildings in Pompeii can be identified by inscriptions as shops (tabernae), rented apartments, or other accommodation, precisely the type of urban property leasing Garnsey imagines. One inscription from the Insula Arriana Polliana demonstrates this quite well, as it advertises that "shops (tabernae) with their rooms (pergulae), equestrian apartments (cenacula), and the house, which are located at the Insula Arriana Polliana owned by Gnaius Alleius Nigidus Maius, [will be leased] from the first of July. For leasing, inquire of Primus, slave of Gnaius Alleius Nigidus Maius (CIL. IV.138)." There are several other examples like this, as well as inscriptions advertising a hotel' as well as food and beverage retailers. 5° It must be repeated here that, simply because the Pompeian evidence is so fortuitously well preserved, we cannot necessarily use the city as archetypal for the Roman world as a whole. It is interesting to note, however, that other cities exhibit some signs of the same urban activity as is seen in Pompeii: Rome and Ostia," for example, show the same sort of leasing arrangements as many of the insulae in Pompeii, on a much larger scale (although mainly at a date beyond the period with which this thesis is concerned); excavations at Cosa have revealed a forum flanked by what appear to be small  periods.  49  CIL.IV.807  50  CIL.IV.202; 241; 1679; 3423; 3485.  51  See Frier (1977) for a discussion of rented urban property in the late Republican to Early Imperial  92  political and commercial buildings; and even cities and the larger towns in the supposedly marginal area of Samnium reveal pottery manufacture, baking facilities, and what is postulated as a fairly active wool finishing industry. Pompeian evidence may indeed illuminate some of the possibilities for conceptualizing Rome's urban economy, however still much more work is needed to determine how widely we can apply Pompeii's example and to what extent this type of urban manufacturing and property leasing were developed in the middle Republican period. The focus of this study is on the agricultural situation, and the urban side of the economy deserves its own investigation. It is important, nevertheless, even in such a brief manner, to note some of the issues associated with understanding the economic output of a Roman city and point out aspects which demand further attention. Indeed, it is very easy to focus solely on agriculture, but in doing so one loses sight of the fact that the rural and urban sectors were inexorably linked in Republican Italy, and it is impossible to understand the one without having some conception of how it may have interacted with the other. Thus, on the basis of the limited evidence we possess, the most likely scenario seems to be that the urban economy, like the rural, was a complex mixture of small- and large-scale production of manufactures and a wide variety of services. In cities, moreover, perhaps even more so than on farmland, free, peasant labor played a major role in the economic output, likely as renters of property rather than as owners.  93  Chapter 10 Concluding Remarks and Areas for Further Research The archaeological evidence examined in this thesis clearly undermines many of the longheld suppositions about the Roman economy in the second century BC. Regional differences seem more apparant than it has traditionally been argued, while the prevalance of latifundia producing cash-crops for export, although they do appear unmistakably in some areas, is not as extensive as has often been claimed, starting with the ancient commentators themselves. Indeed, many archaeologists have recently sought to stress that no dominant settlement pattern jumps out from the physical evidence which we can apply uniformly to all of Republican Italy. It is interesting to point out, however, in spite of this unmistakable regionalism, that there are certain tendencies which seem to appear in all of the areas we have looked at, wherever Rome asserted her political and economic hegemony: settlement nucleated around a chief urban center which facilitated a market for locally grown produce, as well as the space for manufactures which were likely exchanged for that produce; the mixture of smaller farmsteads alongside larger productive villas which were often situated around communication routes; and a complex system of roads serving local and long-distance needs. We see this pattern, moreover, not only in the areas typically associated with fertility and which are generally seen as the centers of Roman agriculture, but it appears even in the supposedly marginal area of Samnium, however admittedly in a less developed form. It is also striking that we find a similar pattern in parts of Campania and even in Sicily, normally believed to be the bullwarks of latifundia agriculture. The persistence of small farmsteads after the Hannibalic period, which again has only recently received serious attention by economic historians, presents considerable problems in  94  developing a coherent picture of Roman agricultural practices. There is simply not sufficient evidence currently to explain precisely how small- and large- farmsteads coexisted within the Roman economy. Perhaps the most vexing issue within this discussion is trying to distinguish the role (or roles) played by non-servile peasants, for we know that, at various places and times, peasants exercised significant economic influence as: owner-cultivators of colonial property tenant farmers; wage labourers; as well as urban labour providing manufactures and services. All that we can safely conclude, as this study makes clear, is that the traditional paradigm of Republican agriculture, which wants to highlight latifundia formation and the use of servile labour within a 'primitive' or under-rational economic system, is no longer workable. Such theories do not fully encapsulate the diverse, intricate, and highly developed agricultural economy which we now know existed in second century BC Italy. This rejection of the old model, however, does not equate to a wholesale acceptance of its antithesis, namely studies which strictly examine the archaeological evidence and undervalue the importance of villa agriculture altogether. These theories leave much unexplained, particularly the repeated and severe political crises in the last hundred years of Republican rule, known through consistent accounts in the literary accounts and which were clearly intertwined with the economic situation of the times. For, in the 130s BC, in peninsular Italy, Tiberius Gracchus rose to political prominence backing a program of land reform which sought to redress some of the peasant displacement caused by slave-staffed latifundia. P.A. Brunt, picking up the traditional approach to Roman economic history from scholars such as A.J. Toynbee and Moses Finley, captures the economic side to these crises noting that they: began with the Gracchi and with agrarian reform, and agrarian reform remained a leitmotiv in the turbulent century that followed...the laws or bills of Lucius Philippus, Saturninus, Sectus  95  Titius and the younger Drusus, the settlement of the Sultan veterens. the proposals of Plotius, Rullus and Flavius, the agitation of Catiline, and the land allotments of Caesar, the Triumvirs and Augustus himself (Brunt 1962, 69).  Indeed, there is a consistent and unambiguous account from a variety of ancient literary sources that long periods of civil unrest were fomented by peasant landlessness, poverty, and disenfranchisement resulting from the widespread agglomeration of rural property and the rise in the use of slave labour in Italy after the Hannibalic Wars. There were, in fact, several land reform programs' and the Slave Revolts in Sicily. Yet, the archaeological record shows unmistakably that even in the areas of north-east Campania and western Apulia, where this latifundia formation was supposed to be the most drastic and where indeed many of the land reforms were focussed, smallholdings persisted through the Republican period in significant numbers. How, then, are we to reconcile these two conflicting accounts? One of the most intriguing ways of accomplishing this is, I think, is to focus even more attention on peasant labour in both the rural and urban economies of Republican Rome. We have stressed in this thesis that peasants were an extremely diverse group and that it is difficult to discern the differences between various occupations - tenant, smallholder, freedman, servus quasi colonus which may account for why we have so little clear evidence for how peasants participated in the Roman economy and why they have been largely ignored by modern scholars. Indeed, we are just now coming to understand the Italian peasantry and how important they were to the Roman economy, yet the matter demands still further investigation. For it seems now that Italian peasants were not completely bereft of land in the second century, as has often been argued, however many  52  Including those carried on by Tiberius' brother Gaius after Tiberius was murdered in 133 BC, as well as soldier resettlement conducted under the demagogic generals Marius and Sulla and the triumvirs. Caesar and his supporters also attempted redistribution programs similar to the Gracchi.  96  of their holdings likely did shrink and there was some displacement, allowing for the limited latifundia formation that we do see. Nevertheless, it is clear that peasants continued to have an important function in both the agricultural and manufacturing sectors throughout Italy in the third and second centuries BC. Nor were these sectors differentiated from each other; the allotments of land in centuriated colonies were often not sufficient for subsistence, forcing many peasants to lease additional land, hire themselves to larger villas as help, or seek employment in cities, aiding in this increase both of rural production and urbanization. Moreover, in the post-Hannibalic period, as we have discussed, Rome experienced an extreme economic boom from the imput of land, slaves, and booty from their many, successful campaigns. In Italy this boom allowed for many major construction and manufacturing projects, for which free peasants supplied much of the labour. What we should be seeking to explain the civil unrest that occured at the close of the Republican era, then, is not the single drastic and society-changing phenomenon that has often been sought in the past, and often found in the idea of latifundia formation, rather we need to develop more nuanced and complex explanations which encompas all of our available evidence. One of the potentially fruitful lines of thought to pursue in this regard is the notion that the economic boom which Rome experienced in the early and middle second century was significantly curtailed by the middle of the first century, when Roman control in the Peninsula had become firmly established and the pace of overseas military campaigns slowed. At the same time, with Rome pushing her frontiers even further into Spain, Gaul, Africa, and Egypt, not only were many more peasant holdings lost because of debt or death incurred on campain, but Roman markets also became exposed to new centers of production, altering the dynamic of market-exchange within Italy itself. This latter point has, indeed, often been stressed in more recent scholarship, as noted in this 97  thesis, by those who wish to assert that latifundia creation was widespread, however it began no earlier than the middle of the first century BC. It would certainly be interesting to find an agricultural pattern in the provincial areas similar to that which we identified in central Italy, namely a deliberate attempt by the Romans to increase rural occupation and productivity, with the concommitant developments of interconnected system communication routes and various urban market centers surrounded by both small- and large-holdings, similar to what we saw with southern Etruria, parts of Samnium, and possibly around Capua. This is not, of course, to ignore the regional differences throughout the Roman world, as certainly the particular topographic, climatological, social, and political considerations would vary widely between regions. Yet, might there be, nevertheless, some fundamental aspects of the Roman economic system which were universal and served to unite the different areas under Roman control, at least economically? It is, I think, fascinating that David Engels' study of Roman Corinth exhibits many of the same attributes that are seen in comparable places in Italy, although Corinth, by the time Rome asserted her authority in the Aegean, had long enjoyed a thriving economy in its own right. Nor is Corinth the only example of this outside of Italy and Sicily. John Bintliff notes a similar scenario in Roman Boeotia, detailing how an increase in rural population from roughly 300 to 100 BC led to an increase in agricultural production-for-export, similar to the marketable surplus of goods that was encouraged by the Roman colonizing process around Cosa (Bintliff 1991, 124). In Gaul too there has been seen an increase in rural population and production after the Roman conquest in the first century BC (Millar 1984(b), 1). This is not the place for a lengthy discussion of these areas or to what extent we may reasonably compare the different regions of Italy with other parts of the Mediterranean, however the apparant similarities between some of these sites should 98  serve to remind us of the need to extend the kind of interdisclipanary investigation undertaken in this report to the broader Roman world, trying to discover to what extent our discoveries here are peculiar to their respective areas or time frames, or if there are any similarities with the agricultural economies of Rome's many new provincial holdings at the end of the Republic. The research here has been fruitful in identifying some of the problems with how modern scholarship has dealt with Republican agriculture and in trying to offer some possibilities for how we might reconcile the contradictions in the traditional accounts. Examining all types of evidence, not limiting our investigation because of ideological prejudice, we have been able to gain a more holistic view of the situation, both in terms of the conclusions we can safely reach and the areas where we must admit the limitations of the evidence. Again, although we have found some very intereseting patterns of economic behaviour, much more work remains to be done on the agricultural economy throughout the Roman world.  99  BIBLIOGRAPHY Attolini, A. et al. 1991: "Political Geography and Productive Geography between the Valleys if the Albegna and the Fiora in Northern Etruria," in Barker, G. & Lloyd, S. (eds.) Roman Landscapes: Archaeological Survey in the Medeterranean Region. British School at Rome: London, 142-152. Aubert, J-J. 1993: "Workshop Managers" in Harris, W.V. (ed.) The Inscribed Economy: Production and Distribution in the Roman Empire in Light of instrumentum domesticum. Cushing-Malloy Inc.: Michigan, 171-182. Barker, G. and Lloyd, S. 1981: "Rural Settlement in Roman Molise" in Barker, G. & Hodges, R. (eds.) Archaeology and Italian Society: Prehistoric, Roman and Medieval Studies. B.A.R.: Oxford, 289-304. Barker, G (ed.) 1995: The Bifërno Survey: The Archaeological and Geomorphological Record. Leicester University Press: London. de Benedittis, G. 2004: "Bovianum, Aesernia, Monte Vairano: Considerazioni sull' evoluzione dell' insediamento nel Sannio Pentro" in Jones, H (ed.) Samnium: Settlement and Cultural Change. Center for Old World Archaeology and Art: Providence, 23-34. Bintliff, J. 1991: "The Roman Countryside in Central Greece: Observations and Theories from the Boeotia Survey (1978-87)" in Barker, G. & Lloyd, S. (eds.) Roman Landscapes: Archaeological Survey in the Medeterranean Region. British School at Rome: London, 122-31. Boren, H.C. 1958: "The Urban Side if the Gracchan Economic Crisis" in The American Historical Review V63 #4, 890-902. Brown, F.E. 1980: Cosa: The Making of a Roman Town. University of Michigan Press: Ann Arbor. Brunt, P.A. 1962: "The Army and the Land in the Roman Revolution" in Journal of Roman Studies V52, 69-86. Brunt, P.A. 1971: Italian Manpower 225 B.C. - A.D. 14. The Clarendon Press: Oxford. Carandini, A. 1981: "Sviluppo e Crisi delle Manifatture Rurali e Urbane" in Giardina, A. & Schiavone, A. (eds.) Societe: Romana e Produzione Schiavistica, VIII. Laterza: Rome,249260. Carandini, A. 1985: Settefinestre: Una Villa Schiavistica Nell 'Etruria Romana. Panini: Rome. 100  Dwyer, E. 2001: "The Unified Plan of the House of the Faun" in The Journal of the Society of Architectural Historians V.60, No.3, 328-43. Dyson, S. 1978: "Settlement Patterns in the Ager Cosanus: the Wesleyan University Survey, 19741976" in Journal of Field Archaeology V. No.3, 251-68. Engels, D. 1990: Roman Corinth: An Alternative Model for the Classical City. The University of Chicago Press: Chicago. Finley, M.I. 1965 "Technical Innovation and Economic Progress in the Ancient World" in The Economic History Review V.18, N1, 29-45. Finley, M.I. (ed.) 1979: Studies in Roman Property. Cambridge University Press: Cambridge. Finley, M.I. 1978: "Empire in the Graeco-Roman World" in Greece & Rome V25, Ni, 1-15. Finley, M.I. 1979: Ancient Sicily. Chatto & Windus: London. Finley, M.I. 1985: The Ancient Economy. The Hogarth Press: London. Foxhall, L. 1990: "The Dependent Tenant: Land Leasing and Labour in Italy and Greece" in The Journal of Roman Studies V.80, 97-114. Frank, T. 1927: An Economic History of Rome. Johns Hopkins: Baltimore. Frederiksen, M. 1975: "Theory, evidence and the ancient economy" in Journal of Roman Studies V65, 164-71. Frier, B.W. 1977: "The Rental Market in Early Imperial Rome" in Journal of Roman Studies V. 27-37. Garnsey, P. 1980: "Non-Slave Labour in the Roman World" in Garnsey, P. (ed.) Non-Slave Labour in the Graeco-Roman World. Cambridge Philological Society: Cambridge. Garnsey, P. 1998: Cities, Peasants and Food in Classical Antiquity: Essays in social and economic history. Cambridge University Press: Cambridge, 34-47. Green, P. 1961: "The First Sicilian Slave War" in Past and Present #20, 10-29. Gualtieri, M. 2004: "Between Samnites and Lucanians: New Archaeological and Epigraphic Evidence for Settlement Organization" in Jones, H (ed.) Samnium: Settlement and Cultural Change." Center for Old World Archaeology and Art: Providence, 35-50. 101  Harris, W.V. 1971 (a): "On War and Greed in the 2nd Century BC" in The American Historical Review V.76, N.5, 1371-1385. Harris, W.V. 1971 (b): Rome in Etruria and Umbria. Oxford University Press: Oxford. Harris, W. V. 1979: War and Imperialism in Republican Rome, 327 - 70 B.C. Clarendon Press: Oxford. Jongman, W. 1988: The Economy and Society of Pompeii. J.C. Gieben: Amsterdam. Keay, S. et al. 2004: "New Approaches to Roman Urbanism in the Tiber Valley" in Patterson, H. (ed.) Bridging the Tiber: Approaches to Regional Archaeology in the Middle Tiber Valley. The British School at Rome: London, 223-236. Kehoe, D.P. 1997: Investment, Profit and Tenancy. The University of Michigan Press: Ann Arbor. Laurence, R. 1994: Roman Pompeii: Space and Society. Routledge: London. Lloyd, J. 1991: "Farming the Highlands: Samnium and Arcadia in the Hellenistic and early Roman Imperial periods" in Barker, G. & Lloyd, S. (eds.) Roman Landscapes: Archaeological Survey in the Medeterranean Region. British School at Rome: London, 180-93. Lloyd, J. 1995: "Pentri, Frentani, and the Beginnings of Urbanization (c.500-80 BC)" in Barker, G. (ed.) A Mediterranean Valley Landscape Archaeology and Annales History in the Biferno Valley. Leicester University Press: London, 181-212. Love, J.R. 1991: Antiquity and Capitalism: Max Weber and the sociological foundations of Roman civilization. Routledge: New York. Manacorda, D. 1981: "Produzione Agricola, Produzione Ceramica e Proprietari nell' Ager Cosanus nel 1 A.C." in Giardina, A. & Schiavone, A. (eds.) Societa Romana e Produzione Schiavistica, VIII. Laterza: Rome, 3-54. Millar, F. 1984 (a): "The Mediterranean and the Roman Revolution: Politics, War, and the Economy" in Past and Present #102, 3-24. Millar, F. 1984 (b): "The Political Character of the Classical Roman Republic, 200-151 BC" in Journal of Roman Studies V 75, 1-19. Nappo, S.C. 1997: "Urban Transformation at Pompeii, 3rd - 2nd C. BC" in Laurence, R. & Wallace-Hadrill, A. (eds.) Domestic Space in the Roman World: Pompeii and Beyond. 102  Thomson-Shore: Michigan, 91-120. de Neeve, P.W. 1984: Colonus: Private Farm-Tenancy in Roman Italy during the Republic and Early Principate. J.C. Gieben: Amsterdam. de Neeve, P.W. 1985: "The Price of Agricultural Land in Roman Italy and the Problem of Economic Rationalism" in Opus V.4, 77-109. Oakley, S.P. 1995: Hill-Forts of the Samnites. British School at Rome: London. Panella, C. 1981: "La Distribuzione e i mercati" in Giardina, A. & Schiavone, A. (eds.) Society Romana e Produzione Schiavistica, VII. Laterza: Rome, 55-81. Patterson, J.R. 2004: "City, Territory and Metropolis: the Case of the Tiber Valley" in Patterson, H. (ed.) Bridging the Tiber: Approaches to Regional Archaeology in the Middle Tiber Valley. The British School at Rome: London, 11-28. Pirson, F. 1997: "Rented Accomodation at Pompeii: the evidence of the Insula Arriana Polliana VI 6" in Laurence, R. & Wallace-Hadrill, A. (eds.) Domestic Space in the Roman World: Pompeii and Beyond. Thomson-Shore: Michigan, 91-120. Potter, T.W. 1979: The Changing Landscape of South Etruria. Paul Elek Ltd: London. Potter, T.W. 1987: Roman Italy. British Museum Publications Ltd.: London. Purcell, N. 1985: "Wine and Wealth in Ancient Italy" in The Journal of Roman Studies V.7 5, 1-19. Rathbone, D.W. 1981: "Development of Agriculture in the `Ager Cosanus' during the Republic: Problems of Evidence and Interpretation" in Journal of Roman Studies V. ^10-23. Riggsby, A.M. 1997: 'Public' and 'Private' in Roman Culture: the Case of the cubiculum" in Journal of Roman Archaeology V.10, 36-56. Rosafio, P. 1994: "Slaves and Coloni in the Villa System" in Carlsen, J. Orsted, P. & Skydsgaard, J.E. (eds.) Landuse in the Roman Empire. 'L'Erma' di Bretschneider: Rome, 145-58. Rostovtzeff, M. 1957: The Social and Economic History of the Roman Empire. Oxford University Press: Oxford. Rostovtzeff, M. 1960: Rome. Oxford University Press: New York. Salmon, E.T. 1957: "Hannibal's March on Rome" in Phoenix V11, 153-63. 103  Salmon, E.T. 1967: Samnium and the Samnites. Cambridge university Press: Cambridge. Sauer, E.W. 2004: "The Disunited Subject: Human History's Split into 'history' and 'archaeology" in Sauer, E.W. (ed.) Archaeology and Ancient History: Breaking Down the Boundaries. Routledge: New York, 17-45. Scheidel, W. 1994: "Grain Cultivation in the Villa Economy of Roman Italy" in Carlsen, J. Orsted, P. & Skydsgaard, J.E. (eds.) Landuse in the Roman Empire. 'L'Erma' di Bretschneider: Rome, 159-66. Shaw, B. 2001: Spartacus and the Slave Wars: A Brief History with Documents. St. Martin's: Boston. Skydsgaard, J.E. 1980: "Non-Slave Labor in Rural Italy during the Late Republic" in Garnsey, P. (ed.) Non-Slave Labor in the Graeco-Roman World. Cambridge Philological Society: Cambridge. Smith, C.D. 1967: "Ancient Landscapes of Taviolere, Apulia" in Transactions of the Institute of British Geographers #41, 203-8. Toynbee, A.J. 1965: Hannibal's Legacy: the Hannibalic War's Effects on Roman Life. Oxford University Press: London. Vallat, J.P. 1991: "Survey Archaeology and Rural History" in Barker, G. & Lloyd, S. (eds.) Roman Landscapes: Archaeological Survey in the Medeterranean Region. British School at Rome: London, 10-17. Verbrugghe, G.P. 1972: "Sicily 210-70 BC: Livy, Cicero, and Diodorus" in Transactions and Proceedings of the American Philological Association V.103, 535-59. Ward-Perkins, B. 1981: "Luni: the Prosperity of the Town and its Territory." in Barker, G. & Hodges, R. (eds.) Archaeology and Italian Society: Prehistoric, Roman and Medieval Studies. B.A.R.: Oxford, 179-190. Weber, M. 1968: Economy and Society. Bedminster: New York. Weber, M. 1988: The Agrarian Sociology of Ancient Civilizations. (Frank, R.I. trans) Verso: New York. White, K.D. 1967: "Latifundia: a Critical Review of the Evidence" in Bulletin of the Institute of Classical Studies V.14, 62-74.  104  Wilson, R.J.A. 1990: Sicily Under the Roman Empire: the Archaeology of a Roman Province, 36 BC-AD 535. Aris and Phillips Ltd.: Warminster. Witcher, R. 2006: "Settlement and Society in Early Imperial Etruria" in Journal of Roman Studies V. ^88-123. Wood, E.M. 2002: "Landlords and Peasants, Masters and Slaves: Class Relations in Greek and Roman Antiquity" in Historical Materialism V10:3, 17-69.  Ancient Works Cato, M. Porcius De Agri Cultura. (Cato) Cicero, M. Tullius Orationes Verrinae. (Verres) Cicero, M. Tullius Orationes de Lege Agraria Contra Rullum. (Agraria) Columella, L. Iunius Moderatus Re Rustica. (Col.) Diodorus Siculus Biblioteca Storica. (Diod.) Livius, T. Annales ab Urbe Condita. (Livy) Plutarchus, Vita Ti. Gracchi. (Plut.) Plinius Secundus, Historia Naturalis. (Pliny) Polybius, Histories. (Polybius) P. Virgilius Maro, Georgicon. (Virgil)  105  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items