Download PDF | Maria Duggan_ Mark Jackson_ Sam Turner , Ceramics and Atlantic Connections: Late Roman and Early Medieval Imported Pottery on the Atlantic Seaboard, Proceedings of an International Symposium at Newcastle University, March 2014.
163 Pages
Foreword
At some point between 1933 and 1935, R. E. Mortimer Wheeler suggested to C. A. Ralegh Radford that sherds of pottery he had excavated at Tintagel in Cornwall might be imported, starting a debate about the sources and function of exotic wares found at post-Roman sites in western Britain that has continued to date (Campbell 2007: 4; Radford 1935: 401). The Ceramics and Atlantic Connections Symposium, held at the School of History, Classics and Archaeology at Newcastle University in March 2014 marks another stage in this long narrative, but one, we believe, that signifies a new direction -a‘sea-change’ - in this research.
Although the overall quantities are not great, these sherds of imported ceramic wares attained a major significance in understanding connections between Britain and Europe in the centuries following the break with Roman Imperial control. John Hayes’ Late Roman Pottery, which has provided the typo-chronological framework for Late Antique archaeology across the Mediterranean, described the British instances as ‘...perhaps the most remarkable phenomenon in the history of exportation of the wares discussed here’ (Hayes 1972: 422). The ability to connect finds from the Athenian Agora or Carthage with the scant material culture of western Britain and Ireland provided not only a crucial dating tool, but an opportunity to draw lines of economic or diplomatic connection between these regions.
Nevertheless, as this volume makes clear, this exchange was not solely an Insular phenomenon. A surge in new ceramic data from across the Atlantic Seaboard, from the reports of recent rescue and research excavations, as well as new appraisals of older museum collections, has required revised understandings of long-distance links between the Mediterranean world and the wider Atlantic zone, and has highlighted the need for fresh dialogues on connections between sites situated along the Atlantic littoral of Europe.
The central aim of the Ceramics and Atlantic Connections Symposium was to consider recent and ongoing research on pottery imported to the Atlantic Seaboard between c. AD 400-700, reflecting an increased international interest in the archaeology and economy of this region in Late Antiquity and the early medieval period. Our hope was that the event would facilitate comparisons between groups of imported ceramics found at Late Antique sites along the Atlantic Seaboard, including western Britain, western France, Ireland, Portugal and north-west Spain. This pottery includes amphorae (particularly of East Mediterranean and North African origin) red-slipped finewares (African Red Slip and Late Roman C and D originating in the East Mediterranean), Gaulish finewares (including Dérivées-de-Sigillées Paléochrétiennes Atlantique/DSPA) as well as coarsewares of Atlantic production, not least ‘E-ware’. The ambition was not only to consider individual sites or assemblages, but to compare the distribution of ceramic types, relative chronologies and the differential compositions of these Atlantic assemblages, allowing a new assessment of patterns between the varied regions of the western seaboard.
Given the broad geographical scope of the research and the expanding distribution of the pottery, we were delighted that our invited speakers comprised archaeologists and ceramic specialists working across the Atlantic Seaboard, including representatives from Britain, Ireland, France, Spain and Portugal, all of whom had conducted recent research on this theme, and many of whom were early-career researchers. Our hope was that this event would provide an opportunity to share new ideas and fresh interpretations on this topic, and would enable communication between those working on Late Roman and early medieval pottery across the Atlantic zone, particularly those at a formative stage in their careers. We were also keen that this event would reinforce the strong history of ceramic research at Newcastle University - the institution has hosted many notable specialists in this field, including John Hayes, who held a Sir James Knott Fellowship at Newcastle between 1964 and 1966, and who credits Newcastle University as providing a grant towards the publication of Late Roman Pottery (Hayes 1972: xii).
This volume, arising directly from the symposium, comprises nine papers from the speakers - a stimulating presentation was also made on the important assemblage from Vigo by Adolfo Fernandez Fernandez. As well as covering new ground, the first three papers provide a more detailed research context for these discussions. Following his enlightening keynote at the symposium, Michel Bonifay highlights key questions that have directed research on the Atlantic systems, before presenting a detailed study of the North African component to this exchange - simultaneously extending the chronology under question and broadening the scope to consider African production centres, evidence from wrecks and key commodities. Ewan Campbell, whose research in Britain has allowed a comprehensive understanding of the Insular evidence - most notably changing perceptions of the Insular finds of ‘E ware’ and glass - provides a personal account of his introduction to the ‘Atlantic question’, summarising his predictions for the evolution of this debate, and positing more suggestions towards future directions in this research. Maria Duggan’s contribution describes a prior ‘Atlantic Symposium’ held in 1959 that also sought to collate knowledge on this theme, and discusses how, since this early phase, the availability of evidence has driven understandings of Atlantic exchange. This leads to a review of recent evidence emerging from south-west Britain.
Moving beyond Britain, Amanda Kelly and her collaborators use the comprehensive scientific analysis of a red-slipped sherd found in south-east Ireland to consider the important Irish connection to the Atlantic exchange routes, as well as the function of the imports within regional networks of consumption and redistribution. Heading south, the next two papers consider the vitally important evidence emerging from Atlantic regions of France. Joachim Le Bomin presents a detailed ‘state of knowledge’ of published examples of Mediterranean amphorae and fineware imported to western Gaul between the 3rd and 7th centuries, comparing broad geographical and chronological patterns within this dataset, and highlighting the long-term importance of the Bordeaux region. David Guitton’s study focuses on western Gaul between the Loire and Gironde, adding highly significant evidence from Saintes and Poitiers, and widening the debate with essential discussions of other Late Antique ceramic groups, including ‘céramique a |’éponge’, ‘DSPA’ and cooking wares.
While the symposium consciously focused on ceramic evidence as the key material source of data, José Carlos Sanchez Pardo’s article expands on this to consider textual and archaeological evidence for artistic, cultural, economic and ecclesiastical connections between the various regions of the Late Antique and early medieval Atlantic, with Galicia acting as a key nexus. The increasing evidence from the Iberian Peninsula is emphasised by the two remaining papers, which present important examinations of ceramics recovered from excavations in Portugal. José Carlos Quaresma’s comprehensive chronological study of pottery from recent excavations in Lisbon again includes crucial evidence of coarse wares and technological change, while considering the city’s long-term connection to Atlantic exchange. Finally, Ana Patricia Magalhdes, Inés Vaz Pinto and Patricia Brum take us to the south-western coast of Lusitania, focusing on imported amphorae, finewares and lamps recovered from the latest deposits studied at the major fish-salting production site of Trdia - prompting questions of the chronological extent of occupation and production in the area.
Conclusions: the future of Atlantic connections
It is testament to the continued significance of the imported pottery, and to the growing interest in the archaeology and economy of the western seaboard within Late Antique Europe, that the Atlantic is named as a region of interest to the Roman and Late Antique Mediterranean Pottery Series (Cau, Bonifay and Reynolds 2011: v). We are, therefore, extremely pleased that the proceedings of the Newcastle Symposium will allow the sites and ceramic assemblages of the late Roman and early medieval Atlantic Seaboard to be highlighted in the volumes of this prestigious series.
It is clear, however, that the Ceramics and Atlantic Connections Symposium and these published proceedings do not represent a final statement on the Atlantic in Late Antiquity. On the contrary, despite nearly a century of interest in the imported wares and associated Atlantic productions, we still have much to discover about the mechanics of Atlantic shipment, the nature of commodities moving in both directions along the western seaboard, and the socio-economic and political ties that underpinned these networks. Our aim was that the stimulating discussions at Newcastle would continue and be extended. As such, we have been encouraged to witness the contacts fostered from this event and the ongoing communication between the participants. We would hope to see these discussions broadened to encompass other artefact groups, including further discussions of glass, metals and organic commodities, and for the scope to be broadened geographically, including into Atlantic regions of Africa.
Finally, we are confident that the Newcastle Symposium and this resulting publication will not be the end of collaboration in the Atlantic Seaboard region, but represents the start of a new phase of co-operation and communication across this zone. We expect that there will be future events on this theme, extending the number and range of participants, and hope that this might not be the only Ceramics and Atlantic Connections Symposium, but the first of many.
Maria Duggan, Sam Turner and Mark Jackson, Newcastle University
Acknowledgements
Sincere thanks are due to all speakers for joining us at Newcastle University for the Ceramics and Atlantic Connections Symposium to share their research and interpretations. Particular thanks can be offered to Michel Bonifay for the keynote address on the evening preceding the symposium, and to Ewan Campbell for the opening talk of the following day. These enlightening presentations established a clear context for the subsequent discussions and we are very happy to include articles from both within this volume.
We are very grateful for all of the delegates who came to join us at Newcastle, not only for adding their expert opinions to the thought-provoking debate, but for helping to create a relaxed and supportive environment for discussion. The group comprised c.30 archaeologists and historians and included several ceramic specialists, including Kevin Greene, James Gerrard, Paul Bidwell, Paul Tyers, Eniké Hudak, Malcolm Lyne and Athanasios Vionis. We were particularly delighted to welcome John Hayes back to Newcastle University for the event. We were also very pleased to be joined by Fiona Pitt (Plymouth City Museum & Art Gallery) and Bill Horner (Devon County Council), who have both encouraged research on the imported ceramics found in the South West of Britain.
The event was funded with the assistance of the School Research Committee of the School of History, Classics and Archaeology at Newcastle University, and we are thankful for this support. We are also grateful for all the support and assistance in organising the day from the School, particularly within the School Office and Research Support team, and the Cluster for Interdisciplinary Artefact Studies at Newcastle University. Further thanks must be given for organisational assistance from Aurora Camafio, and for the warm welcome offered by Frances McIntosh and Andrew Birley during visits following the event.
We are most grateful to the speakers for their prompt submissions for these proceedings, their subsequent correspondence and engagement, and for their patience and support in bringing together this publication. We are also extremely grateful for the editorial support of Michel Bonifay, Paul Reynolds and Miguel Angel Cau in bringing this volume to press within the prominent RLAMP series.
Ceramics and Atlantic connections AD 250-700: the African perspective
Michel Bonifay Aix Marseille Univ, CNRS, CCJ, Aix-en-Provence, France michel.bonifay@univ-amu.fr
Rarely in the field of archaeology and history was so much owed by so many scholars to so few sherds... Since the discovery of the first fragments at Tintagel at the end of the 1930s, a total of 350 (MNV) Mediterranean table wares and amphorae are now registered (Campbell 2011: Table 12), among them 32 African Red Slip Wares (ARS) and possibly 5 African amphorae (Duggan 2016: 72-73, figs 3.3 and 3.4), the rest being in vast majority of Eastern Mediterranean origin. Even if there are not many, these sherds evidence trade connections between the Mediterranean and the British Isles in a very late - as well very short- span of time, between c. 475 and 550 AD.
Sixty years after the first classification of these imported Late Antique Mediterranean wares (Radford 1956), the same three issues are debated again and again (well summarised both by Wooding 1996: 43, and Campbell 2007: 127; see also Reynolds 2010: 108-111, and note 392):
1. ‘By which routes did these wares reach Britain’? ‘Was this direct traffic from the Mediterranean’ or ‘were the wares transhipped at any point’?
2. ‘Why did these voyages take place’? ‘What cargoes did the imports represent and what was traded in return’? ‘Who carried out the trade’ and ‘what was [its] scale’?
3. ‘What was the chronological range of the maritime ventures which brought [these wares]’?
Provisionally concluding a multi-decennial baton change on this matter between - among others - Ralegh Radford, Charles Thomas, and Ewan Campbell, Maria Duggan recently (2016) proposed to update Campbell’s model (2007) for explaining the arrival of Mediterranean wares in late Roman and post-Roman Britain. To achieve this, alongside a re-appraisal of the available documentation, she referred to recent publications in northern Portugal (Quaresma and Morais 2012) and in Galicia (Fernandez 2014), as well as works in progress in the Atlantic French coastline (see David Guitton’s and Joachim Le Bomin’s papers in this volume).
Inthe short text that follows, we will consider the possible contribution of the African wares to this new model, even if this material is a very minority fringe of what is already a small assemblage of ceramics found in the British Isles.
1. Direct or indirect routes
After having constituted the ‘doxa’ for more than 55 years, ‘the ceramic evidence for direct contact between early medieval Britain and the Byzantine world can finally be discounted’ (Duggan 2016: 358). This new hypothesis of an indirect trade is due to the discovery at Vigo of huge amounts of Mediterranean wares, matching very well with the more reduced sample found in Britain. Therefore, Vigo was plausibly acting as an emporium or a hub for the redistribution of the Mediterranean cargoes northward, playing the same role as Cadiz or Seville did at earlier times. For the period 475-550 AD, no other place in the north-western Iberian Peninsula has provided the same quantity of Mediterranean imports, either in Portugal or elsewhere in Galicia - even if other rias, like A Coruna, deserve to be explored further.
Examining the African ware assemblage of this period from Vigo, it is obvious that the table wares and amphorae originate from different regions of Africa. This is particularly clear for the ARS which show some products from central Byzacena (ARS C5: Fernandez 2014: 148-150) and the Sidi Khalifa workshop (Hayes 87A/88 and 88: Fernandez 2014: 174 and 210), alongside a vast majority of products of the Carthage region workshops: El Mahrine forms EM 18 and 53 (Fernandez 2014: fig. 99-100), Oudhna forms Hayes 91C and 96-99 (Fernandez 2014: fig. 99, 104106), and ‘atelier X’ forms Hayes 103 and 104 (Fernandez 2014: fig. 108-114). As the circulation of the different ARS productions is relatively partitioned in northern Africa at this time, we can presume that different cargoes arrived at Vigo from different African ports, or from other - non African - western Mediterranean hub ports.
The question of whether Bordeaux could have played the same hub role alongside, or instead of, Vigo in late Antiquity must be considered in regard to the successive period AD 550-620/30. The distribution of E ware during this period, probably originating from the BordeauxSaintes region, is considered as evidence for contacts between this port and the British Isles. Nevertheless, Adolfo Fernandez rightly observed that during this period the assemblages of ARS are not similar in Bordeaux and Vigo. While the classic late forms Hayes 90B, 105, and 109A are attested in both cities (Fernandez 2014: figs 9798, 115-116 and 118; Bonifay 2012: fig. 2), the difference is made by the marked presence at Vigo of very late products of the Sidi Khalifa workshops (Fernandez 2014: fig. 119121), which are completely absent in Bordeaux. So, even if no Mediterranean ceramic imports are attested in the British Isles after the mid-6th century, Vigo remained attractive for the Mediterranean seafarers until at least the first third of the 7th century. Nevertheless, it was challenged by Bordeaux which perhaps benefited from the reopening of the traditional fluvial/terrestrial road of the ‘Gaulish isthmus’ and by the fact that from 536 onwards the whole of southern Gaul was under Frankish rule. In this case, we could even think of a more segmented transport linked with Marseille, the main Mediterranean door of the Merovingian kingdom.
2. Cargoes and return cargoes
‘The link between the imported pottery in Britain and tin - as proposed in Radford’s original publication - can be reaffirmed’ (Duggan 2016: 346). Even if other commodities have been suggested, such as slaves or wood, it is credible that the main engine of this indirect trade was the need of the Mediterranean countries for tin. The main sources of this metal are located in Cornwall, alongside other minerals available in Wales (silver and lead) and in southern Ireland (copper). This need could explain why different trade flows converged on Vigo, well evidenced by the rich panorama of Mediterranean table wares and amphorae. The fact that only part of these assemblages is found in Britain forces us to consider the nature of the demand in the British Isles.
Amphora arrivals in Britain mainly comprise Eastern Mediterranean LR 1 (107 MNV) and LR 2 (72 MNV) types. With, potentially, as low as 5 examples, we do not have any clear traces of African oil, fish, and wine in postRoman Britain. An alternative explanation could lead us to imagine that the African foodstuffs were packaged or transferred in barrels in order to facilitate their transportation. We can think to an earlier period and the example of Vindolanda, well evidenced by Elise Marliére and Josep Torres Costa (2005), where the ratio between the quantities of wine transported in amphorae and in barrels is of 1 to 122. Nevertheless, even if the barrel was not unknown in Africa, it is always dangerous to reason on absence, while on the other hand African amphorae are not numerous enough at Vigo for thinking about a change in packaging during a break in shipment. The lack of demand could be a better explanation. Demand could also simply explain the high proportion of LRA 2 within the total amphora assemblage in comparison with the Mediterranean ratios, if we admit that oil was the content (Karagiorgou 2001: 146-149); but in this case why not African oil?
As far as the tableware is concerned, even if the presence of ARS is half that of the Phocaean Late Roman C wares, its typological range appears quite similar to Vigo. Large plates Hayes 103 and 104 are in the majority (7 MNV: Campbell 2007: fig. 7, A12-13, and Pl. 2) alongside Oudhna forms Hayes 91C-96-97 (8 MNV: Campbell 2007: fig. 7, A10, A9, A5) and some El Mahrine products (EM 18: Campbell
2007: fig. 7, A15 and 17); there is even one sherd of Sidi Khalifa Hayes 88 (Campbell 2007: fig. 7, A14). The only difference is the complete lack of ARS C5. Following only in part - some of Radford’s ideas about the Christian symbols on LRC dishes, one could suggest that a certain demand existed, among the very few Mediterranean tableware imported to Britain, for Carthage region products bearing cross decorations (see for example the cross with diamond ornament on the Hayes 103 dish from Cadbury Congresbury: Rahtz et al. 1992: fig. 116). However, the preference of post-Roman Britons for LRC tableware instead of better quality African products (think of ARS C5) remains unexplained. I am not sure that this has something to do with the chronology. I will hardly follow Evan Campbell’s interpretation of a slightly later African import flow (Campbell 2007: 138; Bonifay 2008: 1123), because all of the ARS found in Britain seems contemporary with the other Mediterranean imports. On the other hand, one sherd could be later than the rest: a possible Hayes 87B/109 from Tintagel (Radford 1956: fig. 14.8: the direction of the sherd has to be corrected), this form was attested at Vigo (Fernandez 2014: fig. 118) in ‘Horizon C’ of AD 560-620/30 (nevertheless this form is mostly late 6th century according to Reynolds 1987).
Finally, another point, very distinctive of the post-Roman Mediterranean imports in the British Isles, is the lack of lamps, neither the lampes galets from Syria found at Vigo (Fernandez 2014: fig. 194), nor the ARS lamps also present at Vigo (Fernandez 2014: fig. 193) and very popular everywhere in the Mediterranean. Perhaps the reason is that people had no need for these ceramic lamps, because they had glass lamps (Campbell 2007: 72-73 and fig. 41) or other means of lighting.
3. Changes through times
The trade of Mediterranean goods towards Britain probably never ceased during the Roman period but I am quite convinced, with M. Duggan, ‘that the western British ‘post-Roman’ imports do represent a separate dynamic, and that there was some break in the supply of pottery to Britain after the early-fifth century’ (Duggan 2016). Seeking to go further when considering the African imports, two main points deserve to be highlighted.
Firstly, the arrival of African wares seem to have been very scarce from the 1st to mid 3rd century AD. Checking the list published by D. Williams and C. Carreras (1995), it is difficult to find good examples of early and mid-Roman African amphorae in Britain. A few sherds of neo-Punic amphorae are noted (Williams and Carreras 1995: 242), of unspecified types (Arthur 1986: fig. 6.49, a handle). The four examples of ‘Mauretanian’ Dressel 30 found in Holborough are in fact 4th century Keay 1B amphorae (Jessup 1954: fig. 13, right). A Tripolitanian I sherd is noted from Lincoln, but not published. The presence of oil amphora type Africana I is difficult to assert: the presumed example from London Bishopsgate (Tyers 1984: fig. 2.3) is perhaps rather a variant of Ostia LIX (?), and the Southampton example is misidentified, being in reality an Eastern Mediterranean amphora Knossos 18 (Aylwin Cotton and Gathercole 1958: fig. 26.8). Africana II examples are late in date (mid-3rd century and beyond). The same can be observed when considering the vessels: only a few sherds of ARS Al (Hayes 2, 3B-C, 6A, 8A, 9A) are attested, all 2nd century forms, and only one sherd of African Cooking ware (Bird 1977). On the other hand, a series of Coarse Ware imitations of late ARS A and African Cooking Ware are known from York, Carlisle, and Cramond during the Severan period, probably linked with the African garrisons of Hadrian’s Wall (Swan 1992; Bonifay 2014).
Second observation, Britain does not seem to have been included in the mid-4th to early 5th century floruit of African trade in the Mediterranean. No discoveries of African amphora Keay 25.2 and spatheion type 1 are reported, which is very significant as these types are ubiquitous in the Mediterranean in the late 4th century and the beginning of the 5th century. It is also very significant that the sole example of ‘Keay 25’ amphora securely attested in Britain is a probable Spanish imitation of this type, found in the county of Kent (Callender 1965: Pl. Id), and close to the products of the El Mojon workshop near Cartagena (Berrocal 2012). ARS of the same date are also very rare: only one sherd of ARS D Hayes 67, and one or two fragments of ARS lamps are recorded by J. Bird (1977).
As a matter of fact, most of the evidence of African imports in Roman Britain dates to a very short span of time, from the second half of the 3rd century to the first third of the 4th century. For example, Africana II D amphorae from coastal Byzacena are attested at Bedale, from Hadrumetum (Capelli and Bonifay forthcoming), and at Holborough (Jessup 1954: fig. 13 left: not the first half of the 3rd century because associated with four examples of 4th century Keay 1B amphorae), as well as at several other sites (Williams and Carreras 1995). Africana II C2/3 amphorae from Nabeul are present in London (Richardson 1986: 102, fig. 1.8), and Exeter (Holbrook and Bidwell 1991: fig. 89.7-8; Williams 1992: fig. 9.20). Keay 25.1/Africana III amphorae, early variant, possibly from Nabeul, were discovered in London (Richardson 1986: 102, fig. 1.9), and Exeter (Holbrook and Bidwell 1991: fig. 89.9, associated with a transitional variant of Africana IID/IIIA: fig. 89.11). The assemblages of London New Fresh Wharf and of Exeter, quite similar, have to be dated to the beginning of the 4th century, and can be compared with the cargoes of the Mediterranean shipwrecks of Pampelonne and Marausa (Bonifay and Tchernia 2012: 322 and Table 16.2).
4. Towards a single key for interpretation?
In the Mediterranean, this date range - between c. 250 and 330 AD - coincides with changes in the distribution of African amphorae. It is the period when Africana II B-D amphorae are associated with Spanish amphorae in several sunken cargoes scattered in the western part of the Mediterranean (Balearics, Corsica, Sicily), implying that these African containers were first transported to a port in Southern Baetica, before being loaded together with the Spanish containers on the way to - probably - Rome (Bonifay and Tchernia 2012: 320). I recently suggested that this segmented route could have been linked with the necessity for African traders to get merchandise that, at this time, was available only on the Atlantic coastline of Hispania. Furthermore, as in the pre-Roman past, before the opening of the Gaulish route through the Rhone valley and the via Agrippa, this merchandise could have been tin. Perhaps during the troubled second half of the 3rd century, this inland road was not practicable anymore, and the Atlantic route became active again. Following this hypothesis, African amphorae could be a sort of ballast for the African ships sailing to Spain in order to get the precious metal.
However, trying to link both these phenomena - the slightly higher proportion of African amphorae in Britain during the second half of the 3rd century and the frequent association of African amphorae and southern Baetican amphorae in some shipwrecks of the western Mediterranean - is highly hazardous, even if in the Mediterranean, as in the British Isles, the chronology of these facts extends well into the first third of the 4th century. Discoveries of shipwrecks in the English Channel suitable to support this hypothesis are not numerous. However, two wrecks found along the northern coastline of French Brittany were loaded with metal ingots.
The first one, sunken in the Sept Isles archipelago, contained 271 lead ingots, with inscriptions mentioning two Celtic tribes of Roman Britain, the Brigantes and the Icenes; unfortunately the wreck is not precisely dated, even if probably pertaining to the late Roman period (Hour 1987). The second one was explored in 2014-2015 close to Batz Island, near Roscofff, and delivered 5.5 tons of tin ingots, with pottery sherds dating back to the 3rd-4th century (Olivia Hulot, DRASSM, personal communication, and press information). Last but not least, the Guernsey wreck, dated to the end of the 3rd century, was loaded witha very diversified cargo including Spanish amphorae of types Dressel 30 and Tejarillo 1 (Rule and Monaghan 1993: fig. 61.1-2 and fig. 61.3) perhaps associated - the provenience from the wreck is not sure - with at least one African amphora of transitional type Africana IID/ IIA (Rule and Monaghan 1993: fig. 61.4: the graphic reconstruction is erroneous; the ‘hard orange-buff fabric’ could refer to a Nabeul production), It is as if the two first wrecks (Sept-Isles and Batz) evidenced outward trips, the British metal being transported southwards, and if the last one (Guernsey) evidenced a return trip, some Mediterranean goods being - by implication transported northwards.
Thus, in the current state of our knowledge, the two main periods of (relative) affluence of African ceramics in Britain could well be explained by the same mechanisms: the reopening of the Atlantic route for the trade of tin and other precious metals. The southern port of transit could have changed over time - Cadiz or Seville in the 250-330s and Vigo in the 475-550s - as well as the goods transported to Britain as return cargoes: wine (Keay 25.1 amphorae) and fish sauces (Africana II C and D? amphorae) in the first period, tableware (mainly with Christian decoration?) in the second period. During the rest of the Roman period, African imports remained at a quite low level, and probably mainly used the inland routes through Gaul (Campbell 2007: 125-126, with bibliography), or the alternative, partly Atlantic route from Bordeaux and Narbonne, through the Gaulish isthmus (where Spanish imitations of Keay 25, as the one found in Kent, are abundant: see Le Bomin in this volume),
The question of when the Atlantic connection ended is widely discussed in this volume and cannot be resolved by a survey of the African imports alone. Nevertheless, it must be repeated that African imports are present at the possible transit ports of Vigo and Bordeaux until the first half of the 7th century. However, from the mid-6th century onwards, the former inland route perhaps benefited from an increase in frequentation due to the Pax Francorum, with first Marseille and later Arles, perhaps in the late 7th-beginning 8th century, being the main gateways for the African products. The Corbie charter tells us about the amount of oil, wine, and liquamen, perhaps of African origin, being transported toward Amiens after having passed the tolls of Marseille or Fos (Loseby 2000: 178179) Recent discoveries of abundant African amphorae of type Keay 8A in a context of the first quarter of the 8th century in Arles (Mukai et al. forthcoming) could be traces of these movements of goods through the Rhone valley at this time. It must be stressed that this latest type of African amphora is, for now, completely absent at Vigo and Bordeaux.
5. Conclusion
It might be objected that syntheses on African amphorae and tableware written at the end of the last century (Williams and Carreras 1995; Bird 1977), on which this note is based, cannot account for the real situation of African imports in Roman and post-Roman Britain. Nevertheless, even if a full review of the evidence could be crucial in the future, the most recent additions do not seem to fundamentally change the panorama (Campbell 2011: Table 12; Duggan 2016: Appendix C).
At least, the African ware evidence suggests that the rhythms of trade in Britain are quite different from those of the Mediterranean, except only when the Atlantic route is in a position to supply the Mediterranean with tin. In other words, the African imports in Britain were closely linked to the rhythms of demand not so much from the consumers in Britain, but from the Mediterranean traders in charge of supplying the Mediterranean cities with tin and other metals.
As a matter of fact, it is not clear whether the Empire or the post-Roman kingdoms and/or Church were involved in the commercialisation of the metal extracted from the British mines, or if the metal was part of the normal circuits of exchange (Andreau 1989: 112; Campbell 2007: 136-137). Perhaps the much diversified return cargoes would better support the second hypothesis.
It is probable that people in Roman Britain had no specific need for African foodstuffs, except when the African garrisons of Hadrian’s Wall demanded their usual cooking wares, soon locally imitated. Other required foodstuffs, like olive oil, could have been supplied through imports from other Mediterranean provinces, for example Spain, as is well attested by the amphora evidence (Carreras and Funari 1998). It seems that African imports reached Britain only when it was easy to make them arrive (due to more profitable shipments going southwards and generating return cargoes), but also in this case local demand played a major role. For example, African olive oil never benefited from local demand and/or was not chosen by the merchants for the supply of Roman and post-Roman Britain (Spanish oil, and then Greek oil, was preferred).
Of course, African imports remained peripheral to the phenomenon of post-Roman Mediterranean trade towards the British Isles between c.475-550. But, considering the huge distribution of African ceramics at the same time throughout the Mediterranean, I think it was useful to adopt this perspective in this note, if only to mark some differences with the rest of the Ancient World. Obviously, even if it is a common assumption, we never have to forget that our few sherds cannot account for all Ancient trade, that probably a lot of liquid foodstuffs were transported in barrels, and that the most valuable cargos (spices, silk, and papyri?), of which the Mediterranean wares are a simple and imperfect proxy, did not leave traces.
Acknowledgments : I warmly thank Maria Duggan to having invited me to the Newcastle conference in 2014, and for the accurate revision of my English text.
Link
Press Here
0 التعليقات :
إرسال تعليق