imperialism, empire and the integration of the roman … · imperialism, empire and the integration...

12
Imperialism, Empire and the Integration of the Roman Economy Author(s): Greg Woolf Reviewed work(s): Source: World Archaeology, Vol. 23, No. 3, Archaeology of Empires (Feb., 1992), pp. 283-293 Published by: Taylor & Francis, Ltd. Stable URL: http://www.jstor.org/stable/124763 . Accessed: 14/01/2013 08:15 Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp . JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]. . Taylor & Francis, Ltd. is collaborating with JSTOR to digitize, preserve and extend access to World Archaeology. http://www.jstor.org This content downloaded on Mon, 14 Jan 2013 08:15:17 AM All use subject to JSTOR Terms and Conditions

Upload: ngocong

Post on 08-May-2018

239 views

Category:

Documents


5 download

TRANSCRIPT

Imperialism, Empire and the Integration of the Roman EconomyAuthor(s): Greg WoolfReviewed work(s):Source: World Archaeology, Vol. 23, No. 3, Archaeology of Empires (Feb., 1992), pp. 283-293Published by: Taylor & Francis, Ltd.Stable URL: http://www.jstor.org/stable/124763 .

Accessed: 14/01/2013 08:15

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

.

Taylor & Francis, Ltd. is collaborating with JSTOR to digitize, preserve and extend access to WorldArchaeology.

http://www.jstor.org

This content downloaded on Mon, 14 Jan 2013 08:15:17 AMAll use subject to JSTOR Terms and Conditions

Imperialism, empire and the

integration of the Roman economy

Greg Woolf

Competing models of the Roman economy

Empires are political systems based on the actual or threatened use of force to extract surpluses from their subjects. Imperial elites spend these revenues on the infrastructure necessary to maintain power, and retain a profit that is distributed to groups that are privileged by virtue of their place within the imperial hierarchy. Pre-industrial empires differed from other imperial systems in so far as they extracted surplus from economies that were primarily agrarian, by means of relatively primitive military and communi- cations technologies. As a result pre-industrial empires could not support large govern- mental institutions and so secured their power by promoting a community of interest among 6elites within the empire, and a sense of imperial membership based on participation in ruler worship and adherence to imperial cultural and symbolic systems. Economically, however, empires were first and foremost tributary structures, and much of the limited energy at their disposal was devoted to ensuring adequate supplies of cash, labour and agricultural produce from the areas under their control. Such an analysis fits the case of Rome well. The provinces were taxed, taxes were spent on the army and the capital, the supremacy of Rome was maintained more by the support of Romanized elites than by a substantial military or governmental establishment, especially in the early empire (Garnsey and Saller 1987: 20-40).

But if the broad nature of the political economy is agreed, the same cannot be said for the economy of the Roman world as a whole. Some see the political economy as the only link between a series of essentially local economic systems (e.g. Finley 1985), while others see the Roman world as a well integrated economic system, within which the political economy played an important role, but which was also unified by markets in land, credit and agricultural produce (e.g. Carandini 1986; 1989b). Intermediate positions might be envisaged. The Roman world might have consisted of a series of regional economies, only loosely integrated into a single system, or else a loosely and sluggishly integrated Roman economy might have existed alongside the political economy. In the absence of data on the levels of, and fluctuations in, prices, wages and interest rates for most of the empire, archaeology offers the best means of choosing between these alternatives. Long-distance exchange is important here, not because commerce played a central role in the ancient economy, but because the exchange of goods is the most visible symptom of contacts

World Archaeology Volume 23 No. 3 Archaeology of Empires ? Routledge 1992 0043-8243/92/2303/283 $3.00/1

This content downloaded on Mon, 14 Jan 2013 08:15:17 AMAll use subject to JSTOR Terms and Conditions

284 Greg Woolf

between regions. The existence of long-distance exchange is not in itself evidence for a commercial economy. Among currently competing interpretations we may distinguish at least four positions. First, that the political economy, together with the transport of private produce by its owners for their own use, might account for most long-distance exchange within the Roman world (Whittaker 1985). Second, that the operation of the political economy might also stimulate some other long-distance exchanges within the empire (Hopkins 1980; Wickham 1988). Third, that the existence of the empire might in various ways facilitate the limited growth of a mercantile economy (Carandini 1986). Fourth, that long-distance exchanges derived from features of the ancient world as a whole, such as urbanism and the uneven distribution of resources, and even the imposition of empire did not create an integrated economy (Duncan-Jones 1990: 30-58). Such thumbnail sketches do little justice to the sophistication of the works cited; all these writers have argued their views in numerous publications and there are many points of convergence and agreement between them. But the diversity of interpretative hypotheses currently available is clear.

These hypotheses might be used in different ways. One method would be to use rival models predictively, to generate models that might be falsified or modified by confron- tation with existing data sets. For example, if the political economy was the motor for all long-distance exchange, we might expect evidence for exchanges between the tributary provinces and the capital or the frontiers, but not between tributary provinces. In practice, some such exchanges did take place, such as the transport of Pascual 1 wine amphorae from Tarraconensis in Spain to south-west France (Fitzpatrick 1985: 319-21; Tchernia 1986: 142-5, 1989: 531). Exchanges between the centre and the periphery cannot then account for all the long-distance exchanges in the empire and the model must be adjusted accordingly. This method is a powerful one, particularly if the predictions are framed so that their falsification does not depend on arguments from silence. But this paper will use a different technique. The distributions first of container amphorae, and then of other artefacts, will be used to suggest some very general conclusions about exchange systems and their development between the late Republic and early Empire. In particular, a contrast will be drawn between the kind of integration achieved at the height of Roman imperialism, and that maintained after expansion had virtually ceased.

The evidence of container amphorae

The potentials and problems of amphora studies have been much discussed (Paterson 1982; Peacock and Williams 1986; Tchernia 1986). Amphorae are thick-walled ceramic containers, used for much of the period, for transporting wine, olive oil and fish products. On the basis of their epigraphy, typology and petrology, and through the excavation of wreck cargoes, kiln sites and dumps, amphorae are the best guide to patterns of production, exchange and consumption in the ancient world. Problems remain. We are ignorant of the contents and origins of many amphora types, and in some areas, crucially in the east Mediterranean, amphora studies are still in their infancy. It is also likely that some amphora types have been wrongly or incompletely provenanced, and besides, amphorae were not the only form of container used in this period. All the same, the broad patterns of development can now be reconstructed, in particular for those amphorae used to transport wine.

This content downloaded on Mon, 14 Jan 2013 08:15:17 AMAll use subject to JSTOR Terms and Conditions

Imperialism, empire and the integration of the Roman economy 285

The first Roman wine amphorae developed from Hellenistic models in the mid-third century BC and were produced in Etruria, Latium and Campania (Will 1982; Emp6reur and Hesnard 1988: 25-30; Hesnard et al. 1989). Some of these Greco-Italic type amphorae were produced at Marseilles and possibly elsewhere (Laubenheimer 1989: 119), but the majority seem to be central Italian in origin. Widely distributed in the western Mediterranean, they first appear on southern French sites at the end of the third century BC, only becoming numerous in the first half of the second century (Goudineau 1983: 79-81; Tchernia 1986: 94-8). Greco-Italics are more common in Spain, but again only become common in the early second century (Nolla and Nieto 1989). A very few even found their way inland as far as Chateaumeillant in the Berry and Manching in Bavaria. More important, but as yet unquantified, were those exported to the eastern Mediter- ranean (Tchernia 1986: 98). Will (1989) argues that a centre of production near Brindisi supplied the east, but there is no firm evidence for the production of Greco-Italic amphorae in Adriatic Italy (Cipriano and Carre 1989: 89-91).

Greco-Italic amphorae were replaced by a series of new types in the mid-second century BC. One of them, the Dressel 1, represents the most widely distributed amphora type in the ancient world, its variants being transported throughout the Mediterranean, and over large parts of temperate Europe, where it is ubiquitous on settlement sites of the latest iron age (Panella 1981; Fitzpatrick 1985; Tchernia 1986: 74-87). Dressel 1 amphorae were produced between c. 140 and c. 10BC, often in the same kilns as had produced Greco-Italics in central Italy (Tchernia 1986: 42-8, 126-7). Recently it has been shown that some Dressel 1 were also produced in France, including at Lyon (Laubenheimer 1989: 116-18) and at least one variant, Dressel 1C, was possibly produced in Spain (Will 1979; Fitzpatrick 1987: 81-2). Petrological objections seem overturned by recent research

(Peacock and Williams 1986: 91-2; Comas et al. 1987). North-east Spain and south-west France certainly produced another variant, the Pascual 1 amphora, which was distributed throughout western France (Fitzpatrick 1985: 319-21; Peacock and Williams 1986: 93-5). Production seems to have spanned the period c. 50BC-AD 80, but the majority of examples are found in Augustan and Tiberian contexts (Tchernia 1986: 143-5). Produc- tion of amphorae on the east coast of Italy is less well understood, but Tchernia argues that it followed a similar course to production in the west, with Lamboglia 2 amphorae appearing and disappearing at about the same times as Dressel 1 (Tchernia 1986: 53-6, 68-74). Recent research has shown that these amphorae were produced in a number of centres on the west and north Adriatic coasts, but the chronology of their origins and their mutation into Dressel 6A amphorae is still unclear (Cipriano and Carre 1989). Lamboglia 2 amphorae are known in the west, but the main distribution centred on the Adriatic and the eastern Mediterranean proper.

The last decades of the last century BC marked another series of changes in wine amphorae (Tchernia 1986: 126-57). As in the case of the changes described above, there were probably periods when several types were produced simultaneously and changes in form need not have corresponded to discontinuities in distribution. Dressel 6A seem to have evolved gradually out of Lamboglia 2 (Cipriano and Carre 1989). But the most important new type was the Dressel 2-4, produced in central and Adriatic Italy, north-western and southern Spain, southern France and possibly Britain (Peacock and Williams 1986: 105-6; Tchernia 1986: 127-9; Cipriano and Carre 1989: 91-2). Similar amphorae had long been produced in the vicinity of Rhodes and Cos, in coastal Cilicia and

This content downloaded on Mon, 14 Jan 2013 08:15:17 AMAll use subject to JSTOR Terms and Conditions

286 Greg Woolf

Alexandria in Egypt (Emp6reur and Picon 1989), and it was almost certainly the popularity of eastern wines that led to the adoption of the form in the west. But although the distribution of Dressel 2-4 was Mediterranean-wide, they are much rarer in temperate Europe than Dressel 1 amphorae. The number of production centres of Dressel 2-4 is greater, the proportion produced outside Italy is higher, but each production seems to have a much smaller distribution (Panella 1981). The same is true for Dressel 6A, which had a more restricted distribution than Lamboglia 2, and for Pascual 1, which are mainly concentrated in north-east Spain and south-west France.

Dressel 2-4 was the last wine amphora to be produced on a Mediterranean-wide scale. Even during the main period of its production, from the end of the last century BC to the mid-second century AD, it coexisted alongside other types, like Pascual 1 and Dressel 6A. The next generation of amphorae are regional productions. One of the best studied productions is that of Gaul (Laubenheimer 1985: 1989). A total of forty-six kiln sites are known in Narbonensis and five in central Gaul. These kilns had produced a few Greco-Italic and Dressel 1 amphorae and some Dressel 2-4, Pascual 1 and perhaps Dressel 7/11, but in the early first century AD a new series with flat-bottoms, suitable for transport by cart, began to be made. Laubenheimer has identified nine forms: of these, most were

produced by only one or two kilns, Gauloises 1 and 5 were made in several kilns in two localities, but Gauloise 4 was produced all over southern Gaul between the mid-first century and the third century AD, in a surprisingly standardized form. These amphorae were distributed primarily within Gaul, Germany, the Low Countries and Britain (Peacock and Williams 1986: 142-3).

Wine amphorae have received more attention than any other kind, generating competing histories of Roman wine (Purcell 1985; Tchernia 1986; Carandini 1989a; 1989b). The other major commodity carried in amphorae was olive oil, but no synthesis has yet been attempted despite much recent research of high quality (Blazquez and Remesal 1980; 1983; Mattingly 1988a; 1988b). In many respects, oil amphorae exhibit

comparable patterning to that produced by wine amphorae. Best known are the Dressel 20 amphorae produced in the valley of the Guadalquivir in southern Spain and distributed in large numbers in the Rhineland, France, Britain and in the city of Rome. Production started in the early first century AD and continued into the third century (Beltran 1970; Ponsich 1974; 1979; Mattingly 1988a: 38-44). Tripolitanian amphorae are less well studied, but production seems to have begun in the first century AD, and their distribution extended widely over North Africa and central Italy in the second and third centuries AD

(Peacock and Williams 1986:166-70; Mattingly 1988b: 31-6). Both Baetica and Tripo- litania started producing amphorae to export oil in the early first century and continued to supply regional markets into the third century; the regional scale of their distributions is comparable with that of wine amphorae like Pascual 1, Gauloise 4 or Dressel 6A. Similar distributions may have existed in the eastern Mediterranean (Riley 1979: 112-236), but as yet few of them are as clear as the western examples.

What patterns emerge from this brief survey? First, it is clear that a number of regions were producing substantial surpluses for distant consumption. Olive oil and wine were hardly luxuries, but they were not staples either: that dichotomy suits morality better than economics. Surpluses were created within the traditional range of agricultural produce, the Mediterranean triad together with supplementary foods like fish products. Demand

This content downloaded on Mon, 14 Jan 2013 08:15:17 AMAll use subject to JSTOR Terms and Conditions

Imperialism, empire and the integration of the Roman economy 287

for these products depended partly on their uneven distribution, partly on the frequency of both food crises and gluts in the Mediterranean basin (Garnsey 1988: 8-16), partly on the high proportion of the population who neither worked on nor owned the land, but also partly because some areas were held to produce higher quality produce than others. In these conditions, a market for grain had long existed in the East (Rathbone 1983b). But in the west the amphora distributions attest new levels of production. These productions began in the early first century AD, and their floruit coincided with the period when the highest proportion of the imperial population lived in cities.

Second, the distributions give some idea of the different scales of exchange and production. Between the widest distributions and the most local there was a continuum, but for convenience we can divide it into local, regional and empire-wide exchanges. Local distributions have been least well studied and hardly mentioned here, but are perhaps typified by those Gauloise amphorae that were produced in only one kiln site and distributed in the immediate vicinity. Regional distributions include types like Dressel 20, Pascual 1, Dressel 6A, Tripolitanian I and most Dressel 2-4 productions. Empire-wide distributions are represented by Greco-Italic and Dressel 1. Classified in this fashion, it is clear that the last category, the empire-wide distribution, is in fact very rare. Long-standing interest in Italian wine production may have obscured the extent to which its success was extremely atypical. Much more characteristic of the empire as a whole, although not of the period before it in the west, were regional distributions.

Finally, the chronology of the anomalous case of central Italian wine production is interesting. Large scale distribution in the west began in the early second century BC. Tchernia notes changes between Caesar and Augustus, with the appearance of Dressel 2-4 and Dressel 6A, but if these types had much smaller distributions than their predecessors, then those predecessors were already in decline. Within the Dressel 1 series, a shift took place from Dressel 1A to Dressel 1B from about 70 BC. At least in the north-west Dressel lB is distributed more widely and occurs in greater numbers than Dressel 1A (Fitzpatrick 1985). The anomalous distribution of Italian wine thus began slowly in the third century BC, reaching a peak in the third quarter of the last century BC before declining rapidly. The pattern follows the rhythm of Rome's imperial expansion.

Tableware, coins and other evidence

Container amphorae provide direct evidence of long-distance exchanges in the Roman world. The distributions of those goods not transported by amphora are much less accessible, although some sense of regional spheres of exchange can be gained from non-archaeological sources in the case of grain (Garnsey 1983: 119-21; Rathbone 1983b: 50-3). But further investigation of the integration of the Mediterranean economy depends on more indirect sources.

Ceramic tablewares offer one such source. Ceramics, even coarse wares, were traded long distances, but more often as space-fillers than as primary cargoes. Nevertheless, any imported ware still had to out-compete local productions, whether in terms of cost or perceived quality. Ancient cargoes were usually mixed, shippers picking up and putting down items at many of the ports visited in the course of 'cabotage' (Parker 1990: 342-3).

This content downloaded on Mon, 14 Jan 2013 08:15:17 AMAll use subject to JSTOR Terms and Conditions

288 Greg Woolf

As a result, the provenances of the most popular ceramics indicate neither the sources of the primary cargoes nor their destinations. What the distributions do show are regions linked by shipping frequent enough to allow tablewares to compete with each other in a regional market. High proportions of imported wares in the ceramic assemblages of Mediterranean cities themselves suggest a high level of exchange (Fulford 1987).

Roman tablewares evolved out of a widespread Hellenistic tradition of black-glaze ware. The so-called Campanian ware spread throughout the west Mediterranean at much the same time as Greco-Italic amphorae and from the same area (Morel 1980; 1981). Like the amphorae, this shiny black tableware expanded into areas where similar local productions had previously been dominant and both influenced subsequent production of those wares, and inspired imitations. Campanian A developed near Naples in the late third century, Campanian B was produced in Etruria from the early second century, and Campanian C appeared in Syracuse a little later, followed by productions elsewhere in Sicily, Africa, the Balearic Isles and imitations in Africa, eastern Spain and southern Gaul.

Campanian ware was replaced suddenly by Arretine ware between 50 and 30 BC (Goudineau 1980; Pucci 1981). The new product was shiny red and often decorated, and formed the prototype for red glossy wares produced all over the Roman world for the next six centuries. Arretine was distributed widely throughout the Mediterranean basin and well into temperate Europe. The transition from Campanian to Arretine was made in north Italy, probably following eastern models. But production of similar wares, generically termed terra sigillata, spread rapidly. Branch workships of the Arretine firms and local potters had started producing terra sigillata in southern Gaul before 20 BC

(Bemont and Jacob 1986: 33). By the mid-first century AD, terra sigillata was produced at centres in northern Italy (Pucci 1980), Spain (Mayet 1980), north Africa (Tortorella 1988) and central France (Bemont and Jacob 1986: 121-71). Sigillata had been produced in Cyprus and Pamphylia from at least the last century BC (Gunneweg 1987). Most of these wares are ill understood, but it is clear that terra sigillata was produced in a large number of centres in each area, and that only a minority of these centres achieved a more than local distribution. But the distributions of the products of the major centres, like La

Graufesenque and Lezoux in France or Tricio and Andfijar in Spain, show very clear patterning. Unlike the empire-wide distribution of Arretine ware, the provincial terra sigillata productions were regional in scale. Tricio's wares are found throughout Iberia and Morocco, Andfijar supplied Baetica and Morocco and central Gallic products are concentrated north of the Alps, in central and northern France, Britain and the Rhineland. Most later productions have very local distributions with the exception of African Red Slip, which is distributed widely south of the Alps (Hayes 1972). It is possible that no sigillata was produced in Spain in the third century, while north of the Alps a series of local productions in eastern France disappears before 300 AD (King 1981). The amount of fineware supplied by successive centres was less and less, even within their more restricted distributions (Marsh 1981). But in the fourth century regional spheres of exchange once again appeared in the distribution of red tableware in Spain (Mayet 1980), northern France and Britain (Fulford 1989: 196-8), while the African products go from strength to strength (Fentress and Perkins 1989).

The overall pattern recalls that suggested by amphorae. Central Italian products achieved an empire-wide distribution for a while in the last two centuries of the Republic,

This content downloaded on Mon, 14 Jan 2013 08:15:17 AMAll use subject to JSTOR Terms and Conditions

Imperialism, empire and the integration of the Roman economy 289

but before the end of the last century BC, regional distributions have asserted themselves, including over areas of the west that had not participated in that tradition before Roman conquest. This picture is corroborated by other categories of evidence. The number of Mediterranean shipwrecks per century shows a gradual increase from the fifth century BC, suddenly increasing between the third and second centuries and peaking in the last century BC, before declining slowly to a nadir in the sixth century AD (Parker 1990: 336). The record is incomplete - more research has been conducted in the northern Mediterranean than in the south and much more around the coasts than in deep water, but the trend still suggests a temporarily high level of integration in the last century BC. Analysis of the silver coinages of the Mediterranean world, shows the process by which non-Roman coinages were progressively replaced by denarii from the early second century BC (Crawford 1985). But despite the creation of a uniform currency system, local coinages persisted under Roman control in Spain, Gaul and the Greek world in the late Republic, while closed currency systems existed under the empire in Asia, based on cistophoroi, and in Egypt on tetradrachms. The suggestion of regional exchange systems for coinage is backed up by Duncan-Jones' demonstration of differences between denarii contained in coin hoards in different parts of the empire (1989). Inter-regional exchange was clearly insufficient to eliminate variations created by irregular despatches of coin from the imperial mints, but within each region, the circulation of coinage created a common profile. Study of terracotta lamps suggests that each mark characteristically had a regional distribution (Duncan-Jones 1990: 48-58).

Conclusions

Models of the Roman economy, or of the impact of Rome on pre-existent economic systems, ought to be able to explain the patterning of material culture indicated above. Roman conquest incorporated new areas into systems of long-distance exchange, but for the most part exchange within the empire was rarely integrated above the regional level. Some empire-wide distributions of goods appear in the period between the second Punic War and the end of the Republic, but regional distributions reappeared in the last century BC. Roman imperialism, not the stable empire, provided the context for the maximum integration achieved in long-distance exchange systems.

Various explanations have been proposed for the success of central Italian products in the late Republic. Some see these developments as reflecting the rise and fall of the slave mode of production (Giardina and Schiavone 1981; cf. Rathbone 1983a), with Italy eventually succumbing to competition from the provinces (Carandini 1989a). Provincial production of wine, oil and terra sigillata clearly did out-compete Italian produce in the provinces, but the balance of the evidence suggests that provincial wine and sigillata did not capture markets in Italy (Martin 1985; Purcell 1985; Tchernia 1986). But it seems preferable to see imperial expansion as providing Italian producers with a series of new opportunities (Hopkins 1978: 1-98). Italian producers invested the profits of empire in exploiting the demand for Roman goods created by Romanization, the communications networks centred on the growing city of Rome, the presence of Italian troops abroad and the logistical infrastructure set up by the Roman army. Some of these opportunities

This content downloaded on Mon, 14 Jan 2013 08:15:17 AMAll use subject to JSTOR Terms and Conditions

290 Greg Woolf

remained after expansion slowed down. But the rare degree of integration achieved by the Roman economy in the last two centuries BC was created by imperialism, not the infrastructure of a stable empire, and regional distributions reasserted themselves as soon as the cultural gap was closed (Tchernia 1989). How far the regional patterning discussed in this paper can be ascribed to the operations of imperial command economies is a matter for fierce debate. My own impression is that the tributary and redistributive operations of the political economy can provide at best only a partial explanation.

2.viii.91 Magdalen College

Oxford OX] 4A U

References

Beltran, M. 1970. Las anforas Romanas en Espafa. Zaragoza: Fernando et Catolico.

Bemont, C. and Jacob, J. -P. (eds) 1986. La terre sigillWe gallo-romaine. Lieux de production du Haut Empire: implantations, produits, relations. Paris: Editions de la Maison de Science.

Blazquez, J. M. and Remesal, J. (eds) 1980. Produccion y commercio del aceite en la antiguedads Congresso I. Madrid: Universidad Complutense.

Blazquez, J. M. and Remesal, J. (eds) 1983. Produccion y commercio del aceite en la antiguedads Congresso II. Madrid: Universidad Complutense.

Carandini, A. 1986. II mondo della tarda antichita vista attraverso le merci. In Societa romana e impero tardoantico III. Le merci, gli insediamenti (ed. A. Giardina). Rome: Editori Laterza, pp. 3-19.

Carandini, A. 1989a. L'economia italica fra tardc Repubblica e medio Impero considerata dal punto visto di una merce: il vino. Ricordando i tempi dello scavo ostiense, che sembrano cosi lontani. In Lenoir et al. 1989, pp. 505-21.

Carandini, A. 1989b. Italian wine and African oil: commerce in a world empire. In The Birth of Europe. Archaeology and Social Development in the First Millennium AD (ed. K. Randsborg). Rome: L'Erma di Bretschneider, pp. 16-24.

Cipriano, M. T. and Carre, M.-B. 1989. Production et typologie des amphores sur la c6te Adriatique de l'Italie. In Lenoir et al. 1989, pp. 67-104.

Comas, M., Martin, M., Matamoros, D. and Mir6, J. 1987. Un tipus d'amfora Dressel 1 de produccio Laietana. In Homenaige a Josep Estrada i Garriga, pp. 372-9.

Crawford, M. H. 1985. Coinage and Money under the Roman Republic: Italy and the Mediterranean Economy. London: Methuen.

Duncan-Jones, R. P. 1989. Mobility and immobility of coin in the Roman empire. Annali dell' Istituto Italiano di Numismatica, 36: 121-37.

Duncan-Jones, R. P. 1990. Structure and Scale in the Roman Economy. Cambridge: Cambridge University Press.

Emp6reur, J. -Y. and Hesnard, A. 1988. Les amphores hell6nistiques. In Ceramiques Hellenistiques et Romains II (eds P. L6veque and J.-P. Morel). Paris: Annales Litt6raires de l'Universit6 de Besancon, pp. 9-71.

Empereur, J.-Y. and Picon, M. 1989. Les regions de production d'amphores imp6riales en Mediterranee orientale. In Lenoir et al. 1989, pp. 223-48.

Fentress, E. and Perkins, P. 1989. Counting African Red Slip Ware. Africa Romana, 5: 205-14.

This content downloaded on Mon, 14 Jan 2013 08:15:17 AMAll use subject to JSTOR Terms and Conditions

Imperialism, empire and the integration of the Roman economy 291

Finley, M. I. 1985. The Ancient Economy (2nd edn). London: Hogarth Press.

Fitzpatrick, A. 1985. The distribution of Dressel 1 amphorae in north-west Europe. Oxford Journal of Archaeology, 4: 305-40.

Fitzpatrick, A. 1987. The structure of a distribution map: problems of sample bias and quantitative studies. Acta Rei Cretariae Romanae Fautorum, 25/26: 79-112.

Fulford, M. 1987. Economic interdependence among urban communities of the Roman Mediter- ranean. World Archaeology, 19: 58-75.

Fulford, M. 1989. The economy of Roman Britain. In Research on Roman Britain 1960-89 (ed. M. Todd). London: Society for the Promotion of Roman Studies, pp. 175-201.

Garnsey, P. 1983. Grain for Rome. In Trade in the Ancient Economy (eds P. Garnsey, K. Hopkins and C. R. Whittaker). London: Chatto & Windus, pp. 118-30.

Garnsey, P. 1988. Famine and Food Supply in the Graeco-Roman World: Responses to Risk and Crisis. Cambridge: Cambridge University Press.

Garnsey, P. and Saller, R. P. 1987. The Roman Empire: Economy, Society and Culture. London: Duckworth.

Giardina, A. and Schiavone, A. (eds) 1981. Societa romana e produzione schiavistica. Roma: Editori Laterza.

Goudineau, C. 1980. La ceramique Aretine. In Ceramiques Hellenistiques et Romains I (eds P. L6veque and J.-P. Morel). Paris: Annales Littdraires de l'Universitd de Besancon, pp. 123-30.

Goudineau, C. 1983. Marseilles, Rome and Gaul from the third to the first century BC. In Trade in the Ancient Economy (eds P. Garnsey, K. Hopkins and C. R. Whittaker). London: Chatto & Windus, pp. 76-86.

Gunneweg, J. 1987. Roman pottery trade in the eastern Mediterranean. Acta Rei Cretariae Romanae Fautorum, 25/26: 119-29.

Hayes, J. W. 1972. Late Roman Pottery. London: British School at Rome.

Hesnard, A., Ricq, M., Arthur, P., Picon, M. and Tchernia, A. 1989. Aires de production des grdco-italiques et des Dr. 1. In Lenoir et al. 1989, pp. 21-65.

Hopkins, K. 1978. Conquerors and Slaves: Sociological Studies in Roman History I. Cambridge: Cambridge University Press.

Hopkins, K. 1980. Taxes and trade in the Roman Empire, 200 BC-AD 200. Journal of Roman Studies, 70: 101-25.

King, A. C. 1981. The decline of Samian ware manufacture in the north west provinces: problems of chronology and interpretation. In The Roman West in the Third Century (eds A. C. King and M. Henig). Oxford: British Archaeological Reports, International Series 109 (i), pp. 55-79.

Laubenheimer, F. 1985. La production des amphores en Gaule Narbonnaise. Paris: Annales Litteraires de l'Universit6 de Besanqon.

Laubenheimer, F. 1989. Les amphores gauloises sous l'Empire. Recherches nouvelles sur leur production et leur chronologie. In Lenoir et al. 1989, pp. 105-38.

Lenoir, M., Manacorda, D. and Panella, C. (eds) 1989. Amphores romaines et histoire economique. Dix ans de recherche. Actes du colloque de Sienne (= Collection de l'Ecole Francaise a Rome 114). Rome: Ecole Frangaise de Rome.

Marsh, G. 1981. London's Samian supply and its relationship to the development of the Gallic Samian industry. In Roman Pottery Research in Britain and North-west Europe (eds A. C. Anderson and A. S. Anderson). Oxford: British Archaeological Reports, International Series 123(i), pp. 173-238.

This content downloaded on Mon, 14 Jan 2013 08:15:17 AMAll use subject to JSTOR Terms and Conditions

292 Greg Woolf

Martin, A. 1985. South Gaulish sigillata in Italy. Opus, 4: 125-39.

Mattingly, D. J. 1988a. Oil for export? A comparison of Libyan, Spanish and Tunisian olive oil production in the Roman empire. Journal of Roman Archaeology, 1: 33-56.

Mattingly, D. J. 1988b. The olive boom: Oil surpluses, wealth and power in Roman Tripolitania. Libyan Studies, 19: 21-42.

Mayet, F. 1980. Les sigille6es hispaniques: 6tat de la question. In Ceramiques Hellenistiques et Romains I (eds P. L6veque and J.-P. Morel). Paris: Annales Litt6raires de l'Universit6 de Besangon, pp. 281-301.

Morel, J.-P. 1980. La c6ramique companienee: acquis et problemes. In Ceramiques Hellenistiques et Romains I (eds P. L6veque and J.-P. Morel). Paris: Annales Litt6raires de l'Universit6 de Besangon, pp. 85-112.

Morel, J. -P. 1981. La Ceramique Campanienne: Lesformes. Rome: Ecole Frangaise de Rome.

Nolla, J. M. and Nieto, F. J. 1989. La importaci6n de anforas romanas en Catalufia durante el periodo tardo-republicano. In Lenoir et al. 1989, pp. 367-91.

Panella, C. 1981. La distribuzione e i mercati. In Societa romana e produzione schiavistica II (eds A. Giardina and A. Schiavone). Roma: Editori Laterza, pp. 55-80.

Parker, A. J. 1990. Classical Antiquity: the maritime dimension. Antiquity, 64: 335-46.

Paterson, J. 1982. 'Salvation from the sea': amphorae and trade in the Roman West. Journal of Roman Studies, 72: 146-57.

Peacock, D. P. S. and Williams, D. F. 1986. Amphorae and the Roman Economy: An Introductory Guide. London: Longman.

Ponsich, M. 1974. Implantation rurale antique sur le Bas-Guadalquivir I. Paris: Casa de Velazquez.

Ponsich, M. 1979. Implantation rurale antique sur le Bas-Guadalquivir II. Paris: Casa de Velazquez.

Pucci, G. 1980. Le officine ceramiche tardo-italiche. In Ceramiques Hellenistiques et Romains I (eds P. Leveque and J. -P. Morel). Paris: Annales Litt6raires de l'Universit6 de Besangon, pp. 135-57.

Pucci, G. 1981. La ceramica italica (terra sigillata). In Societa romana e produzione schiavistica II (eds A. Giardina and A. Schiavone). Roma: Editori Laterza, pp. 99-121.

Purcell, N. 1985. Wine and wealth in ancient Italy. Journal of Roman Studies, 75: 1-19.

Rathbone, D. W. 1983a. The slave mode of production in Italy. Journal of Roman Studies, 73: 160-8.

Rathbone, D. W. 1983b. The grain trade and grain shortages in the Hellenistic East. In Trade and Famine in Classical Antiquity (eds P. Garnsey and C. R. Whittaker). Cambridge: Cambridge Philological Society, pp. 45-55.

Riley, J.A. 1979. The coarse pottery from Benghazi. In Excavations at Sidi Khrebish, Benghazi (Berenice) II (ed. J.A. Lloyd). Tripoli Department of Antiquities: pp. 91-497.

Tchernia, A. 1986. Le Vin de l'Italie romaine. Essai d'histoire economique d'apres les amphores. Rome: Ecole Francaise de Rome.

Tchernia, A. 1989. Encore sur les modeles 6conomiques et les amphores. In Lenoir et al. 1989, pp. 529-36.

Tortorella, S. 1988. La ceramica Africana: un riesame della problematica. In Ceramiques Hellenistiques et Romains II (eds P. L6v&que and J. -P. Morel). Paris: Annales Litt6raires de l'Universit6 de Besanqon, pp. 279-327.

Whittaker, C. R. 1985. Trade and the aristocracy in the Roman empire. Opus 4: 49-75.

Wickham, C. 1988. Marx, Sherlock Holmes and late Roman commerce. Journal of Roman Studies, 78: 173-93.

This content downloaded on Mon, 14 Jan 2013 08:15:17 AMAll use subject to JSTOR Terms and Conditions

Imperialism, empire and the integration of the Roman economy 293

Will, E. L. 1979. The Sestius Amphoras: a reappraisal. Journal of Field Archaeology, 6: 339-50.

Will, E. L. 1982. Greco-Italic amphoras. Hesperia, 52: 338-56.

Will, E. L. 1989. Relazioni mutue tra le anfore romane: i ritrovamenti in oriente, alla luce dei dati ottenuti nell'occidente. In Lenoir et al. 1989, pp. 297-309.

Abstract

Woolf, Greg

Imperialism, empire and the integration of the Roman economy

While the political economy of the Roman empire is well understood, the nature of the economy as a whole is a matter of fierce debate. A range of competing hypotheses about the nature of the Roman economy are presented, and then distributions of amphorae and certain other artefact categories are used to suggest some broad patterns which successful theories of the Roman economy will have to address. In particular, it is argued that regional distributions of material usually constituted the highest level of integration achieved in the Roman world, and that larger-scale patterning was only created during the period of imperial expansion during the last two centuries BC.

This content downloaded on Mon, 14 Jan 2013 08:15:17 AMAll use subject to JSTOR Terms and Conditions