Jump to ContentJump to Main Navigation
Ancient Economies of the Northern AegeanFifth to First Centuries BC$

Zosia Halina Archibald

Print publication date: 2013

Print ISBN-13: 9780199682119

Published to Oxford Scholarship Online: January 2014

DOI: 10.1093/acprof:oso/9780199682119.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2018. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in OSO for personal use (for details see www.oxfordscholarship.com/page/privacy-policy). Subscriber: null; date: 16 January 2019

Regionalism and regional economies

Regionalism and regional economies

(p.192) (p.193) 5 Regionalism and regional economies
Ancient Economies of the Northern Aegean

Zosia Halina Archibald

Oxford University Press

Abstract and Keywords

This chapter explores how ideas of ‘regionalism’, as defined by geographers, as well as historians, can be applied to the north Aegean economic ‘super-region’. Charting distributions of artefacts from the last five centuries bc (including coin types originating from maritime and inland sites; the distribution of metals with a distinctive technological signature; as well as the more familiar ceramic amphorae of identifiable forms), we can detect the inter-connected nature of the east Balkan landmass. Surviving artefacts reflect the particular bilateral forms of exchange documented in the sparse written sources referred to in Chapter 1. Far from being a continuous or uniform pattern, relations between bilateral agents fluctuated and changed, in concert with a range of other socio-political factors. The key role of Thasos as an exporter and transmitter of commodities, fashions, and skills, from maritime Aegean to continental Balkan areas, was supplanted by Rhodes and other producers in the southern Aegean, while Byzantion increasingly monopolized the passage of goods into and out of the Black Sea.

Keywords:   regionalism, distribution patterns, amphorae, royal monopolies, regional historiography, ethnographic geography, salt, Pella, Pistiros, Krastevich

Exploring The Northern Aegean As An Economic ‘Super-Region’

The northern Aegean as an axis of exchange and as an interface

Given that the landmass explored in this book is divided between three modern states (Greece, Bulgaria, and Turkey), and that it was not unified politically at any one time within the second half of the first millennium BC, in what sense can we consider it to have operated as an economic region or regions? It is not a contention of this book that the land area discussed here was in any sense a unity; it was not. If, however, we accept that complex societies of the European Iron Age did rely on a network of supplies, at varying distances from the consumers, and accept the overall applicability of the concept of Mediterranean ‘connectivity’ as a structural component of socio-economic relations, then it ceases to be necessary to think of ancient economies in an ‘essentialist’ manner, and in terms of well-defined geographical spaces. It was the economic interconnections between the different communities of the north Aegean that gave their economies a specific character in classical antiquity. It was the social expectations of the inhabitants, and the networks of exchange to provide them with specific commodities, that created the interconnections linking the various geographical spaces explored here. In a purely local and geographical context, ‘the northern Aegean region’ might simply refer to the islands of Thasos, Samothrace, Lemnos, and perhaps Lesbos and Chios, with the adjacent coastlines of the Thermaic Gulf, Chalkidike, and Thrace as far as the Chersonese; but as soon as we expand the frame of reference beyond the local and the geographical, then the scope of enquiry has to embrace more distant partners and different ecologies. The space under consideration includes a number of different ‘regions’ in the physical and administrative senses (whether we (p.194) are thinking of ancient or more recent geographical or administrative divisions). This is why I refer to the area as an economic ‘super-region’.

In Chapter 1, I suggested some of the ways in which the term ‘north Aegean region’ can be applied to the interlinked economies of two major territorial powers (Macedon and Thrace) and their coastal neighbours, who enjoyed a high degree of political independence from these kingdoms. The next chapter developed the theme of the Greco-Persian Wars as a defining stage in the history of the region, a period that catalysed the emergence of the two main territorial states. Macedon and Thrace became socio-economic magnets that refocused patterns of exchange between coastal and inland parts of the east Balkan peninsula. At the same time, a significant economic distinction began to emerge at the turn of the sixth and fifth centuries, with the development of new naval technologies by those Aegean states that had a vested interest in competing with the leading naval power of the era, namely the Phoenician and Egyptian fleets of the Persian Empire.1 From an economic point of view, the accessibility of key strategic resources for shipbuilding, not just timber, but also hemp, iron, and copper, made it highly desirable for states that needed to expand their supplies in these resources to establish or develop effective connections with partners in appropriate areas who could provide them, or to get resources by direct territorial acquisition. The northern Aegean, particularly the mainland coast, provided one of the principal areas explored for these possibilities from the final decade of the sixth century BC onwards, if not earlier, as we will see in Chapter 6.

Geography still plays an important role in the ways that ‘regions’ are conceptualized by historians and economic geographers. The physical geography of localities, together with climate, soils, and precipitation, has had a powerful effect on the parameters of local cultures. There is a close identification between people and the landscapes they live in. The north Aegean sea is ringed by mountains—the northern Pindhos (effectively the western boundary of our region, in terms of local ecologies and historical connections);2 the outliers of Pindhos to the east of the main range, Mounts Barnous and Bermion (Figs. 2.1 and 5.1), respectively west and east of Lake Vegoritis; the Pierian range as far as Olympos; (p.195)

                   Regionalism and regional economies

Fig. 5.1. Bridge over the River Haliakmon near Vergina (the ancient city of Aigeai)

Mount Boras/Voras, west of the River Axios, and Kerkine, to the east of it, which belongs to the southern outliers of the extensive mountain range of Rhodope; beyond it, the west–east running chains of the Sredna Gora and the Balkan range itself (Stara Planina, the ‘Old Mountain’); Strandja, the plateau region separating the east Balkans from the hinterland of Byzantion; the triple spine of Chalkidike and its outcrops, Thasos and Samothrace. The mineral diversity of this mountainous landscape and its wealth in timber and related resources has been one of its key long-term strengths. These assets, including both the renewable and the non-renewable resources, have determined some of the area’s principal economic activities.

Landscapes are more than potential physical resources that are available to be tapped by human societies. They are the canvases on which people write their life stories, just as they are the substance of individual and collective survival strategies.3 The kinds of materials that we use to build domestic structures, the range of staple crops and the repertoire of symbolic forms that structure our cultural languages all depend on a range of essential local resources. In a twenty-first century context, the (p.196) framework of regional administration tends to conceal the ways in which a region’s inhabitants embed specificity through habitual practice. Administrative mechanisms promote tourism and local commercial interests by reinforcing the perceived characteristics of regions in an oversimplified way—through typical culinary specialities, or the traditional forms of architecture. The language of regional promotion can mask the very processes of cultural production, presenting regional identity under a homogenized template of definitive forms, creating a sense of timeless cultural continuity, when the realities are more dynamic and more varied.

The notion of ‘regional economies’ represents a different approach to conceptualizing a ‘region’ than what can be achieved by looking at geographical or cultural perspectives alone. Grounded as they are in specific physical environments, regional economies exist by virtue of the people who created them. Contemporary regional studies, whether of present or past landscapes, do integrate the ethnic and political dimensions of regions alongside geographic ones, whilst the links between geography and economic behaviour are the raison d’être of economic geography.4 So geography is by no means irrelevant to the study of regional economies. Many modern niche industries arose because of specific synergies between a region and its inhabitants, and between the social drivers within that region and the demand for specific products. In terms of medieval and early modern industries, the textile producers of Prato, west of Florence, or the merchant traders around the Golden Horn at Istanbul, or the marble quarrymen of Thasos, all constitute examples of local skilled specialists whose expertise evolved as a result of synergies between social demand and local resources, and which developed over a long period of time.5

Regional studies of contemporary cultures offer other insights that can usefully be applied to economic questions about past societies. We might, for example, consider supply chains as indicators of economic networks. Another obvious case study might be the relationship between procurement zones for the exploitation of minerals, the specialists who extracted (p.197) these minerals, the different user groups who transformed the mineral into a more accessible form, those who bought this form on behalf of end users, and the end users themselves. Iron production, as discussed in Chapter 4, involved a supply chain linking remote parts of Rhodope with cities on the Aegean coastline. This is a relatively straightforward pattern, because the trace elements in the ferrous ores can be followed in the artefacts and tools used in places within a 30 km radius of the mining areas.6 There was a limited number of miners and smiths with the requisite knowledge to extract the metal in the first place, and an equally restricted number who knew how to manipulate these particular ores so as to produce workable tools. The distances are not large and the mechanisms of the supply chain were located within a small geographical area, as in the three cases explored in Chapter 4—the hinterlands of Zone, the fort of Kalyva, and Abdera. The situation is more complicated when we consider either more complex networks, with multiple potential suppliers and a more opaque process of supply or transmission (as in the case of grey-faced pottery, discussed in the same chapter), or more complex distribution patterns of objects, such as silver bowls inscribed with royal names and what appear to be places of origin.7 Distribution maps of coins that originated in a given civic centre, or were minted by certain rulers, are one of the well-known ways of demonstrating particular kinds of economic outreach. Gary Reger has recently explored the regional significance of Milesian silver and copper alloy coins, minted in the later third and early second century BC, in this light, as well as other Milesian issues such as the gold and silver ‘Alexanders’, which are usually thought to reflect payments made to retiring war veterans.8

The most obvious and the most widely researched investigations of economic connections within the east Balkan area are those associated with the delivery of substantial quantities of wine and olive oil to many inland locations.9 These were not merely mechanical patterns of bulk transmission of stock. As we saw in Chapter 1, transportation is not to be divorced from consumption and conscious patterns of demand. The capillary movement of wine and oil amphorae was complemented by evidence of other cultural traits that match the consumption of these two key foodstuffs—in the culinary field, the use of specific types of tableware and haute cuisine; in architecture, the adoption of ashlar masonry for (p.198) public amenities and of terracotta roof tiles. These material enhancements are visible throughout the region and on an increasing scale over the five-hundred-year period examined here, although there were marked fluctuations in the dissemination of specific artefacts and commodities.

Economic connections resulted from specific historical initiatives, which were driven by social or political imperatives, but they were also related to perceived ecological variability—the scarcity of certain minerals, and the need for shipbuilding timber, for instance.10 Those seeking resources developed in our region included distant partners such as the Athenians, as well as partners from the region itself, such as coastal harbours in need of metal ore. The targets of significant external bulk trade were mainly located in lower Macedonia and north of Rhodope, although some of the coastal towns along the north Aegean periphery were also net consumers of wine in particular, from Thasos, Chios, Kos, and Rhodes, whilst Pontic producers (mainly Herakleia Pontika, Sinope, and Chersonesos) competed with the former on an increasing scale in the final three centuries BC, albeit primarily to satisfy consumption within the Black Sea region and adjacent inland sites. The apparent impenetrability of Rhodope is largely the result of modern national border administration; distributions of archaeological material within the mountain region do not support the idea that trade did not cross upland terrain.11 In addition to the relatively well-known pattern of exchange between Aegean or Black Sea producers and Balkan consumers, new trading connections have recently been detected, delivering products from hitherto unknown producers in the east Balkan region (notably in Mesambria during the middle decades of the third century BC) to clients inland and to overseas destinations.12 These kinds of network connections reinforce the robustness of the regional network envisaged here.

It has been apparent to researchers that significant fluctuations did occur in the demand and supply patterns between major consumer centres and major suppliers over three centuries (that is, between the early fifth and the late third or early second centuries BC). These fluctuations have been hard to understand, beyond the curves and dips shown on charts and graphs from different consumer centres. The quantities of (p.199)

                   Regionalism and regional economies

Fig. 5.2. Vineyard in the east Balkan region

residual ceramic data are enormous; even ceramic experts have difficulty connecting the pottery with known sources and the results of classification have not been easy to analyse with respect to historical events. This complex picture has nevertheless been transformed by a range of scientific analyses of fabrics and of clay sources, and also by a valuable re-assessment of major known classes of pottery.13 The vast majority of eastern Mediterranean centres that produced wine and oil for export seem to have had a rather restricted outreach. They were primarily involved in satisfying demand within their own immediate regions (Fig. 5.2). This remained true of most producers throughout the timescale under consideration here. A proportion of producing centres seems to have made larger quantities of containers, but either there is no recognizable pattern in the observable distributions, or consumption was primarily regional. From an economic point of view, the regionally focused patterns are consistent with an established tradition of (p.200) consumers, and with a regular rhythm of communication, probably connected with other transportation needs. These regionally orientated producers included Ilion, Corinth, Miletos, and Samos, and in the northern Aegean, Ainos, Oisyme, and Samothrace. Sites that seem to have developed particularly close or intense relations with given entrepôts on a larger scale included Akanthos and Mende in Chalkidike, and the island of Peparethos, just north of Euboia. Only a small number of producers, no more than four in all (Thasos, Kos, Knidos, and Rhodes) can be said to have had truly large export distribution patterns, with continuous production throughout this period.14

The proportion of this export trade that was shipped to different localities did vary over time, however. The output of Thasian wine amphorae has been studied in terms of the volume of jars produced at known production sites on the island, as well as in terms of the numbers of stamps that have been recovered at recipient centres. Although stamps represent an unknown proportion of the gross output of wine jars, there are various ways of estimating the relationship between stamps and unstamped jars. Some 28,000–30,000 Thasian stamps are available for study and the comparative incidence of Thasian stamps at different recipient centres, particularly when shorter episodes can be juxtaposed, does provide useful information for the full course of the island’s output.

The earliest Thasian wine amphorae did not travel very far. During the course of the fifth century, this pattern of local exchange began to expand. In the later fifth century, between 70 per cent and 90 per cent of Thasian jars were exported outside the island’s immediate hinterland (the island’s peraia, or mainland possessions), primarily to consumer centres along the western and northern Black Sea coast, a proportion of which penetrated to dozens of known locations in the Thracian and Macedonian interior. There were intermediate emporia, along the main river arteries, at various distances from the coast, through which amphorae were trans-shipped to sites farther inland. These included Pella, on the River Loudias, in Macedonia (Fig. 2.1); a large penumbra in the hinterland of Odessos and in the Valley of the Roses, much of it probably river-borne; a number of sites between Edirne and Dimitrovgrad in the middle Hebros valley (Fig. 4.7);15 Adjiyska Vodenitsa, near Vetren, in the Thracian Plain (Fig. 4.7, nos 27, 32, 39, 48, 56; Fig. 4.8, nos (p.201) 22–25); and sites at more distant locations farther north in the foothills of the Sredna Gora chain, such as Krastevich (Fig. 4.7, no. 36), and Vasil Levsky, near Karlovo (Fig. 4.7, no. 49).16 The evidence for the dissemination of amphorae is represented by consumption at particular sites. So distribution patterns emerge rather from the overall range within which the products are represented. The commonest wine jars in the east Balkan interior at this time were Thasian, and, to a lesser extent, Chian ones. During the fourth century BC, the more distant entrepôts retained a significant share of the market (with Kallatis outstripping Istros, for example, in the volume of demand), but a larger proportion than previously was consumed by the principal urban foci of the north Aegean mainland, including Amphipolis, Abdera, and Maroneia.17 Study of kiln sites seems to confirm that the number of potters on Thasos increased in the fourth century BC to satisfy this increased demand; but a significant drop in the take-up of wine exports by the more remote markets followed in the late fourth century (perhaps as a result of disruption in trading patterns during Lysimachos’ reign, though it is also worth considering the indirect role of changing patterns of demand, brought about by demographic and social changes at this time). A further slump in Thasian exports succeeded this dip, during the third century.

At present it is not yet clear whether this was a progressive decline in Thasian imports to the region as a whole, or an abrupt change at a number of sites, including Seuthopolis.18 Neutron Activation Analysis has allowed the complex group of vessels formerly labelled the ‘Parmeniskos’ Group (identified as technically related by Virginia Grace) to be dis-aggregated, revealing at least two production centres, one of which was probably in Chalkidike (since some samples have formerly been identified as from Mende, including examples from Pella); the other perhaps at Ainos, close to the estuary of the River Hebros. The samples used in the study were drawn from a number of key consumer centres of the east Balkan interior: Adjiyska Vodenitsa (Pistiros), Sboryanovo, Odessos, Kabyle, and a configuration of sites between the Bay of Burgas (p.202) and the middle Hebros. Stoyanov used this data as a basis for a focused research programme in the vicinity of Mesambria Pontika, to determine whether the high concentration of related finds was symptomatic of production in Mesambria. This identification seems to be confirmed both by clay analyses and by the coincidence of at least two names on amphora stamps with those of magistrates at Mesambria.19 The production of Mesambrian amphorae is particularly well represented at Kabyle and in the hinterland of Mesambria during the second quarter of the third century BC.

Thasian wine jars disappeared from the Balkan region after the second quarter of the third century BC, conceding to Rhodian, Koan, Mesambrian, and Pontic producers. Tzochev attributes the changing patterns in the export of Thasian wine to broader changes in the configuration of political (and hence economic) power in the Aegean and more generally in the eastern Mediterranean, beginning with the diminished role of the Athenians under Macedonian rule. The export of bulk commodities such as wine or oil was not unconnected to the export of other bulk products, particularly in cases where ship-owners collaborated in financing long-distance voyages, hedging their investments with a succession of mixed cargoes, aimed at multiple destinations—documented most famously in the journey between Athens, via Mende or Skione in Chalkidike, and the northern Black Sea grain ports, as described by the speaker of Demosthenes’ speech Against Lakritos (35.10–13). In this case, 3,000 jars of Mendean wine (whose containers have been identified at a wide number of sites in the north Aegean) were pledged for a return cargo, which would probably have consisted of grain. Surviving fragments of wine jars give the perhaps false impression of large and consistent patterns of imports. The speech Against Lakritos reflects the kinds of single transports that can also sometimes be documented from wrecks of merchantmen. Although wine jars probably did travel mainly in irregular, seasonal transports to the major coastal outlets, albeit often in smaller vessels, the quantification of amphora sherds by context at least suggests that consumption was regular at recipient centres, so the supply of wine within given localities seems to have been stored in sufficient quantities to (p.203) satisfy the demand.20 On the other hand, documents like the speech Against Lakritos show that major transports were potentially vulnerable to erratic changes at the supply end, if investors or ships’ captains decided, for whatever reason, not to put up capital (in the case of the former), or not to chance a distant voyage (in the case of the latter). It is by no means clear which of the possible causes may have triggered the shift in traffic away from the underlying vector of Aegean trade in the Classical period (when Athens-based finance was linked to north Aegean products and Pontic grain), in the direction of the axis fuelled by Rhodian finance and shipping, bringing Rhodian wines, as well as other southern Aegean commodities, to the northern Aegean and Balkan region. The demise of the Piraeus played a part, as did the reduction in Athenian grain demand.21

There seems to have been a series of major discontinuities in the vectors of trade across the region. One is represented by the disruption to commerce that occurred at the time of Macedonian expansion eastwards under Philip II and Alexander III. The disappearance of Attic fine wares, which were very widely disseminated up to the end of the fourth century BC, is probably connected, first of all, to the occupation of Mounychia hill by a Macedonian garrison after 322 BC, and then, in 295 BC, with Demetrios Poliorketes’ blockade of Piraeus, which prevented the supply of essential commodities to Athens by that route and, by the same token, of Athenian exports. Third-century BC Athenian exports, particularly tableware in the ‘West Slope’ style, are much less well represented.22

The brief appearance and subsequent disappearance of imported amphorae (with other evidence of a direct Greek presence) at Krastevich, deep in the foothills of the Sredna Gora mountains of central Bulgaria, testifies to the uncertainties experienced by merchants, who explored this new opportunity in the first half and probably into the third quarter of the fourth century BC, only to retreat thereafter.23 At Adjiyska Vodenitsa, Vetren, there is evidence of a series of destructions, c.300 BC, when the fortified tower above the Eastern Gateway collapsed and was never (p.204) rebuilt,24 and in the period around 280–278 BC, when a hoard of Macedonian coins was concealed. This last evidence has been connected by the excavator, Jan Bouzek, with the historically-attested Celtic incursions at that time.25 The period between the principal Celtic incursions in the Balkans and the war between Rhodes and Byzantion in 220 BC (Plb. 4.37.8; 45–52), presents a complex and confusing picture, since we have few precise facts on which to reconstruct a plausible understanding of the opaque tensions between, on the one hand, the cities of the western Black Sea coast and those near the Hellespontine Straits and, on the other, the various communities of the interior.

Alexandru Avram has provided an ingenious argument to explain why the invasion of Thrace staged by the Seleukid king Antiochos II during the later 250s BC was the ostensible sign of a larger confrontation between the Seleukid king and Ptolemy II of Egypt. Ptolemy bestowed a number of benefactions on the city of Byzantion, probably in the early 270s BC, and continued to provide support for the city, and for a closely connected alliance, which included Herakleia Pontika and Kios. Antiochos II besieged Byzantion, probably in 255 BC, at the time of the so-called ‘Monopoly War’ between Byzantion on the one hand and Istros and Kallatis on the other, for control of the emporion at Tomis.26 Avram (p.205) has explained the coincidence of these two events in terms of a polarization of interests. Ptolemy II was continuing a policy that allowed him a degree of power in the Straits and beyond, into the Black Sea. His support of Byzantion was a conscious component of this policy. Antiochos II, on the other hand, chose to court the cities of Apollonia, Kallatis, Istros, and probably Mesambria, perhaps using some of these as bases for operations inland, particularly for assaults on Kabyle (whose coinage was subsequently overstruck in his name), and perhaps Seuthopolis. A number of honorific inscriptions indicate cordial relations between Apollonia, Kallatis, and Istros at this time, while the distribution of coins in the name of Kavaros (the Celtic leader whose notoriety rests mainly on his threats to Byzantion), suggests that he operated from a base in the hinterland of Odessos, perhaps at the fortress of Arkovna.27 Although, as we shall see, Kavaros appears in Polybius’ account of the history of Byzantion as an aggressive threat to the city’s interests, the circulation of his bronze issues, clearly intended for minor transactions, fits rather smoothly into the pattern of regional commercial relations, alongside the bronze coins minted by the city of Mesambria. A very similar pattern of connections emerged in the late second and first half of the first century BC, when regal issues minted by one Mostis were succeeded by those of the ‘Sapaian’ dynasty, based at Bizye. The most prominent of these were large issues of Rhoimetalces I, sponsored by the Emperor Augustus, and his successors in the region, Rhoimetalces II and III.28 Such events highlight the changing form and nature of economic relations in the region, with an increasing focus on the Pontic seaboard and the Straits. Contrary to the impression of immense tension portrayed in the narrative sources, the coins issued by various authorities within the region, by civic magistrates, by local rulers, and by incoming powers such as Antiochos II, (p.206) point to a high degree of interaction between the different agents operating in the region. The common occurrence of shared coin designs (such as the reverses with an owl on a spearhead on coins of Agathopolis, Bisanthe, and of the ruler Adaios, dating to the second quarter of the third century BC) and common countermarks, such as the bronze of Antiochos II countermarked at Kabyle, or the six- or eight-pointed star that appears on the reverse of Byzantion’s earliest bronze issues (dated 240–220 BC) and contemporary coins of Adaios, show a closely interconnected web of commercial relations.29

The emergence of monopolistic behaviour, linked to control of the physical bottleneck constituted by the Straits, will be re-examined below in connection with the extraordinary rise of Byzantion. The internal picture of the east Balkans over the period c.300–100 BC indicates great prosperity on the one hand and substantial disruption of overland trade links on the other. The Celtic incursions of the early 270s BC are linked to rural destructions in lower Macedonian farm estates, as well as cities, such as Aigeai and Pistiros. At the same time, most of the coastal cities of the Aegean and west Pontic coast, the chief cities of lower Macedonia and central Thrace, experienced the first significant period of urban expansion in their histories, with the emergence of public squares, buildings, and various civic amenities. What is more, the archaeological evidence for the ‘Celtic’ presence in the region does not indicate a well-differentiated social entity.30 This confirms the notion that there was far more to the conflict between Byzantion and its neighbours than a simple case of greed (on the part of neighbouring communities). The web of connections was interrupted by the advent of Roman troops in response to the activities of Mithridates VI of Pontus (132–63 BC). This is the point at which the configuration of the networks begins to change to the new order, based on bilateral treaties with Rome. The lead here seems to have come from Kallatis, as early as 106–101 BC.31

Cross-cultural exchange

More subtle but no less penetrating than bulk transports, and the coin distributions that seem to follow the same dynamics as the bulk produce, were other socio-cultural features of economic life that seemed to move in step with the demand for bulk commodities, particularly those (p.207) associated with urban living, such as the appointment of market officials; the use of publicly approved coins, weights, and measures; the regulation of sales in general, and the collection of taxes. Should we expect the organization of market transactions to map directly over the kinds of commodities exchanged? This is a more complex question, which deserves deeper investigation. As the discussion of markets and market-type exchange revealed in Chapter 2, and as we will see further below, there is a variety of assumptions within the scholarly community about the infrastructure of exchange across cultural boundaries.32 These assumptions are linked to ideas about the nature of socio-economic structures on the one hand and to site classification on the other. The socio-economic questions are concerned with the control of market-type and non-market transactions. Were territorial powers, like the kings of Macedon and Thrace, all-powerful in economic as in social matters? How might these forms of authority have affected transactions and demand at a local level? Were the kingdoms of Macedon and Thrace more like the Persian Empire, in terms of attitudes to money and markets, than to mainland Greece? In Chapter 2 I argued that ‘royal economies’ should not be taken at face value, namely as command mechanisms, driven mainly or simply by aspirations towards military aggrandizement. So there is no reason to assume that any and every transaction took place only with the express approval of the authorities. In fact, there are strong reasons for thinking that a good deal of exchange was unconstrained, as we saw in the case of grey-faced pottery in Chapter 4.

The unconstrained exchange of ceramic technologies amongst a range of locations in the north-east Aegean beginning in the mid sixth century BC (and in some cases mid seventh or even earlier) on the one hand, and various coastal and inland centres of the east Balkan peninsula on the other, suggests that there were no formal restrictions on inter-cultural exchange and that these exchanges operated rather effectively. The production patterns observed in the early stages of this technological process did not change significantly in the fifth and fourth centuries BC, when we might expect to see differences, if royal or princely decisions really did affect what could and could not be exchanged in fundamental ways. The general osmosis of commodities, reflected most clearly in the widespread consumption of imported fine pottery from Corinth, Ionia, and particularly Athens, seems therefore to have continued much as before; but the specific legal and social forms of economic mechanisms (p.208) that enabled these kinds of transactions are still emerging from current research.

Royal monopolies

What was certainly new about the organization of commodity exchange under the monarchies was the introduction of a system of royal monopolies, reflecting a hierarchy of power, albeit one that had constantly to be reaffirmed. The royal monopolies over timber suitable for shipbuilding,33 and over gold and silver mines, at Dysoron in Macedonia from the early to mid fifth century BC onwards34 and at Krenides-Philippoi from the middle of the fourth century,35 are simply the best known examples. Viewed alongside these royal prerogatives, the mining rights of the historian Thucydides, evidently located on the mainland opposite Thasos, begin to look much more socially distinctive than historians of Athens have generally allowed. These were not just perquisites of power.36 The peculiar social significance of leaders in the societies of the northern Aegean at the time of the ancestral marriage between the Philaid Miltiades the Younger and Hegesipyle, the daughter of a Thracian landowner called Oloros (after whom Thucydides’ own father was named) elevated to a new status the Philaidai, who had ruled the Chersonese since the Elder Miltiades had established an enclave of power among the Dolonkoi of the Chersonese in the 540s BC. The historian tells us that he had inherited mining rights in gold mines (4.105.1), which meant, according to his own account, that he had considerable influence among people of standing there. For local societies, the value of gold lay in its timeless symbolic qualities. The leaders of (p.209) society, female as well as male, were marked out with gold in death, to represent the link between their responsibilities in decision-making and action on behalf of their communities and the duties that they owed as a result in the metaphysical sphere.37 The sheet-gold work that forms such a prominent component of sixth-century Macedonian burials came from a mine or mines that have not yet been identified. The Odrysian kings also had exceptional access to gold and silver reserves. The coincidence of known mines and gold artefacts, made from a similar, high-carat gold, in the river valleys at the western end of the Thracian Plain, indicates a close connection between mining and the consumption of gold and silver around Strelcha and Panagyurishte on the northern tributaries, Assenovgrad and the far western parts of Rhodope and the southern tributaries of the River Hebros, singled out by Pliny for gold mining (NH 33.66; cf. Luc. Fugitivi, 24: πολὺς χρυσὸς ἤ ἂργυρος ὀρύττεται‎, referring to the area of Philippopolis). These resources are less prominent in the works of ancient authors, because they were not the locus of inter-state competition.38

The cumulative evidence about gold mines and how they were exploited suggests that it would be incorrect to view mining rights exclusively in terms of kings, but rather to see a gradual evolution of rights of access in relation to how gold was used by northern societies. In the sixth century BC, gold and silver retained their significance as markers of élite social identity. In the course of the fifth century, the intense interest in gold and silver resources by many different users within the lower Strymon area and in the Pangaion district seems to have provoked a change in social practices too. Gold continued to be used in élite burials, but it was no longer deployed in the quantities that had been used in the past.39 The lack of response to Thucydides’ expectations of support from local peers may nevertheless be more apparent than real. The historian did retire to this region. There is every reason to believe that he did benefit from these dynastic connections. As we shall see in Chapters 6 and 8, the aristocratic circles of the later fifth century BC in the Strymon, Nestos, and Hebros valleys were still embedded in a social matrix that recognized the validity of hereditary roles, epitomized in the wearing of gold-sheet jewellery. The external use of the metal began to change only gradually in the more competitive environment of this period. The further intensification of the exploitation of precious (p.210) metals under Philip II and his successors served to enhance the monetary value of gold at the expense of its social significance.

The competition for resources in the lower Strymon area, to which we will return in Chapter 6, has usually been explored in terms of rival Greek claims in the region, rather than from inter-cultural perspectives. One reason for the reluctance to consider these broader, inter-cultural connections is a widespread assumption that culturally, as well as politically, the region combined very different partners, comprising not only the kingdoms of Macedon and Thrace, but also a number of independent Thracian ‘tribes’ and the coastal Greek cities.40 For a region to be recognized as an economic entity does not require prima facie cultural or ethnic unity. Reger has drawn attention to a number of examples where cultural differences are no barrier to regional convergence: ‘the Native American tribe of the Maricopas, who speak a completely different language from their neighbors the Akimel O’odham and moved into the [Sonoran] desert in the 1700s, adopted cultural practices virtually identical to the Akimel O’odham, to the point that the two groups merged and now share a reservation in the US state of Arizona’. These south-western American communities share a very similar ecological environment, and have adapted to it using similar agricultural and seasonal practices, festivals, and housing.41 Cultural and ethnic congruence may play a part in the manner in which economic patterns emerge and develop, but we need not assume that the absence of these common features constituted barriers to economic relations. On the contrary, it is a key argument of this book that cultural differences contributed to the economic space of east Balkan social and commercial culture in the second half of the first millennium BC.

Nevertheless, there is a well-embedded assumption about the cultural distinctiveness and separateness of the partners involved in commercial transactions that has complicated our understanding of how economic life was articulated. Since economic relations cannot be dissociated from social ones, as we saw in Chapters 2 and 3, it is appropriate to return at this point to the historiography of the region in order to comprehend (p.211) how far modern perspectives have helped to frame the notion of cultural, political, and economic separation.

The historiography of economies in the east Balkan landmass

When Velizar Velkov started the series of symposia on ‘Settlement life in Thrace’ in the late 1970s, his aim was to refocus attention on a key aspect of antiquity that had become opaque as a result of the increasing trend in Bulgarian historiography, in the post Second World War period, to develop an abstract discourse about the Thracians. Whilst Michael Rostovtzeff’s oeuvre on the Hellenistic world (SEHHW) had attempted to integrate the ‘fringe’ areas of the Mediterranean into his main narrative about classical antiquity, a new generation of post-war Bulgarian scholars was consciously drawing a boundary on the one hand between the Thracians as a cultural and historical phenomenon, and the Greco-Roman world on the other. Among the leading proponents of this new trend was Alexander Fol, who was inspired, initially at least, by the Marxist concept of the ‘Asian mode of production’, and whose conception of ancient Thracian society consisted of a tight, hierarchical ruling class, epitomized by Thucydides’ description of Odrysian princes and paradynasts, whilst the population at large was in essence a cipher, a shadowy presence, whose relevance did not really feature in a theory preoccupied by the ritual dominance of priest-kings. It is one of the strange ironies of the Communist era in Bulgaria that the ordinary people of the ancient world were in practice ignored by contemporary scholarship. In parallel with this trend was the comparative lack of interest in the nature of citizenship, or of the collective status in general, of the inhabitants of Macedonia, who were dismissed as ‘sub-citizens’ or ‘serfs’. This assumption is still active. ‘It seems inherently unlikely that rural egalitarianism would prevail in a society where hierarchy was so much in evidence elsewhere’.42

Central to the ideas propounded by Fol, and adopted by a large number of sympathizers, was the inalienability of land. In Fol’s view, the Thracian kings owned all land, so the concept of private ownership did not exist in Thracian territory. Fol’s ideas were sketched out in a number of papers that examined the terminology of Thracian settlements in Greek and Latin authors; but the underlying theoretical (p.212) principles of his approach were not fully explored.43 In particular, Fol did not explain how it came about that kings acquired unassailable power, and how this was exercised in practice, when such evidence as we do have about the political history of the Odrysians points rather to a fluctuating control over specific geographical areas and little explicit information about the scope of royal authority. Xenophon is not an altogether objective guide to Odrysian regional power—as a mercenary he and the surviving Cyreans in the employ of the prince Seuthes, for just over a month during the winter of 400BC (Xen. Anab. 7.1.5, 2.36), needed to keep their distance from native Thracians against whom they were fighting—but Xenophon nevertheless observes how an unnamed Odrysian officer from the central administration joined the local paradynast, Medosades, when it seemed that the behaviour of Seuthes and the Greek mercenaries was becoming ambiguous, and the rights of local people were not being upheld. Xenophon’s unwitting testimony appears to contradict the idea that absolute power was exercised by Thracian kings; Medosades and other high-ranking officers owned land of their own. Xenophon was promised various pieces of land, including the town of Bisanthe, in return for services to Seuthes.44

Margarita Tacheva, who adopted the essential principles of the ‘Asian mode of production’ in her analysis of Thracian society and economy, was prepared to recognize that there was evidence in Thrace during the second half of the first millennium BC of the early signs of class structure, but she was critical of those historians who wanted to see in Thrace the emergence of a ‘pre-polis’ or polis-type society.45 Tacheva took the view that territorial powers, including the kingdom of the Odrysians, were characterized by royal ownership of land and royal control over the means of production, which entailed the redistribution of commodities amongst the king’s subjects. The coastal communities, in her interpretation, were socially and economically distinct from the Thracian population of the interior, because the former recognized private ownership of property and market exchange, using coinage. There was, in her view, a ‘dialogue’ between Greek coastal communities and the Thracians, (p.213) mediated by aristocratic individuals who owned slaves and other property.46 In this interpretation, ancient references to economic ‘enterprise’, such as the Pseudo-Aristotelian anecdote concerning rural workers being urged to sow cereals for export to coastal towns, were interpreted in a similar light, as evidence of control over subject populations rather than as a partnership between the king and his subjects, which is, nevertheless, an equally valid understanding of the same passages.47

At the western end of our ‘super-region’, there has been a marked increase, during the last twenty-five years, in the number of epigraphic documents dealing with land sales—mainly, but by no means exclusively in the ‘new’ provinces of Macedonia, east of the River Axios. These deeds of sale, which date from the fifth to third centuries BC, provide some explicit evidence of land prices, as well as details about the mechanisms of sale.48 In fourth-century BC Macedonia, royal power and prerogatives coexisted with independent, private sales and community transactions. There does not seem to be any incompatibility between the allocation of resources to royal appointees, as occurred, for example, at Amphipolis after its appropriation by Philip II, and the sales instigated by private landowners in the same city, whatever the immediate motives.49 The implications of the evidence on land sales in Macedonia and the ‘new’ territories east of the Strymon have yet to be applied to Thrace. Tacheva’s radical distinction between Macedonian and Greek forms of landholding on the one hand, and Thracian practices on the other, was based on an interpretation of Macedonian institutions that no longer has the support of many scholars of the region.50

The dominant post Second World War trend in Bulgarian scholarship on Thracian society and economy has also been under fire from archaeologists. The principal excavator of the early Hellenistic city of Seuthopolis, (p.214) D. P. Dimitrov, developed the theory that Seuthopolis may have aspired to the status of a contemporary Greek polis (as the topographical content of the inscription found there, now dubbed the ‘oath of Berenike’, indicates: IGBR iii.2, no. 1731; Velkov 1991, 7, no.1); but in practice this was a ‘royal residence’, a type of settlement identified with the fortified residences described by Xenophon in Thrace and north-west Asia Minor (Seuthes’ residence: Anab.7.2.21; Asidates in Mysia: 7.8.12–14). Seuthopolis is, on this reading, a socially restricted type of settlement, with room for no more than some 40 aristocratic families. This social interpretation of Seuthopolis might equally, however, be applied to the spacious residences at the heart of Pella, or even to the generous accommodation at Olynthos, if we use the evidence of house forms and artefacts as direct reflections of societies. If these comparisons are valid, however, perhaps the spacious style of accommodation should be seen as a regional preference, rather than an ethnic one.51

Moreover, Stoyanov has criticized D. P. Dimitrov’s evaluation of the social hierarchy at Seuthopolis, arguing that the excavator was too ready to dismiss the evidence for the extramural population of the city as poorly housed and indigent. Long-term archaeological investigations at Sboryanovo (identified with ancient Helis), where there are several extramural sectors, show that tiles may sometimes be the only evidence we can point to of domestic structures, other than portable artefacts; but these investigations show no significant differences between the kinds of everyday objects found inside the urban walls, as compared with similar material found outside.52 If there are social distinctions to be drawn inside Thracian settlements, these cannot be made on the basis of the types of residential structures discovered so far.

Proponents of the ‘Asiatic mode of production’ in Thrace have had difficulty in accommodating the burgeoning evidence of commercial activities inland of the coastlines. Tacheva has argued that the rights accorded to Maronitan merchants and traders from other cities at the emporion of Pistiros were in effect forced on a puppet ruler, a successor of Kotys I, put in place by Philip II of Macedon, who was thereby obliged to renege on his exclusive rights. ‘Urbanization [in Thrace during the second half of the fourth century BC] was very late due to the strong royal power and its economy, which prevented producers from participating (p.215) in the market by distributing their production and gaining the resulting profit, which was expropriated by the Odrysian rulers’.53 No specific evidence, historical, literary, or epigraphic, is adduced to demonstrate how royal powers were asserted. Nevertheless, the capacity of rulers to implement their policies is perhaps ultimately less significant to economic analysis than the denial of community autonomy at any level. The idea of ‘royal cities’ is an attempt to square the application of royal power with the manifest evidence of urban nuclei, notably at Adjiyska Vodenitsa, near Vetren, at Philippopolis, Seuthopolis, Kabyle, Sboryanovo, and elsewhere.54 Nevertheless, the relationship between the inhabitants of these ‘royal residences’ and the central power has not been examined. Porozhanov has identified sites as different as Kypsela, Doriskos, and Ganos as ‘royal residences’.55 The history of these sites shows that they underwent various changes over the course of their history and a single designation does not cover the changing dynamics of the region.

One independent source of information about place names in inland Thrace has been the series of inscriptions on silverware, particularly silver bowls. The find spots have a rather localized distribution in northern Thrace and include the hoard found at Rogozen, in north-west Bulgaria, of 165 gilded silver vessels now usually thought to have been among the items stolen from the baggage train of Philip II by the Getae, in the wake of Philip’s Scythian campaign in 339 BC.56 Hatzopoulos used these place names (some of which coincide, as we have seen, with Porozhanov’s ‘royal residences’), to argue that there were no politically autonomous civic communities in Odrysian Thrace: ‘À l’intérieur du royaume, Pistiros, siège des emporitai grecs, n’était pas plus une polis que Béos, Apros, Sauthaba, ou Ergiskè, qui figurent toujours sur la vaisselle d’argent sous forme toponymique, en tant que localités, et jamais sous forme d’ethniques, témoignant de l‘existence d’une communauté civique. Le royaume odryse manquait la culture politique qui eût permis l’éclosion des communautés civiques en son sein et à plus forte raison qui lui eût donné l’occasion de fédérer les poleis déjà existantes’.57

(p.216) A number of different arguments have here been combined. First is the status of Pistiros, which, along with Beos, Apros, Sauthaba, and Ergiske, is negatively assessed in terms of political autonomy. The author does not explain why the Odrysian kingdom was ‘incapable’ of giving birth to autonomous communities, whilst the kingdom of Macedonia evidently was. The argument is based on assumptions that have not been fully articulated and the notion seems rather to rely on the kinds of cultural polarities that have aroused strong criticism from Rollinger and others (as we will see in the section that follows). The arguments that have dominated the discourse on ‘the polis’ are recognizable once more. Irrespective of where it was located in spatial terms, Pistiros qualifies as a polis in terms of the criteria defined by the Copenhagen Centre.58 As to the other locations, the likeliest explanation of their appearance on silverware is as an indication of the mustering points for tribute or payment collection from coastal communities, who were not subjects of the Odrysians.59 The inscriptions do not provide any evidence for or against the type of community that these represented. Judging by the ways in which places like Ganos (a teichos in Xen. Anab. 7.5.8; a chōrion in Aesch. 3.82), Ergiske (a chōrion in Dem. 7.37; 18.27), and Apros (a polis in Theopompos FGrH F160) are described in contemporary Greek sources, whether partial and opinionated or not, this was a heterodox group of places, selected for organizational convenience, perhaps.

Assumptions about the monopolizing nature of Odrysian power are not confined to those scholars who retain the ideological foundations of a Marxian dichotomy between ‘Asiatic’ and ‘classical’ state structures. Thucydides’ analysis of Odrysian revenues has been interpreted by François Salviat in a way that shows full accord with such assumptions. The ‘oligarchie clanique’ of the Odrysians relied on emporia to supply rich Thracians with wine, in exchange for cereals, cattle, and slaves.60 We can observe, in Salviat’s understanding of the Odrysian economy, a similar social divide to that presented in the work of Fol, Tacheva, and others. Nor is it dissimilar to common scholarly attitudes regarding Macedonian social relations, as we have seen. This brief survey of the (p.217) socio-economic literature for parts of our ‘super-region’ reflects an impasse in attempts to understand the relationship between the holders of power and the territories in their possession. A look at territorial divisions, as we will see later below, provides one way of getting around this impasse.

Describing A Region—The Ancient Geography Of The East Balkans

Another way in which modern historiography has sought to distance and polarize the different parts of the east Balkan–north Aegean economic region has been to define Thracian royal practices in terms of Persian ones, whilst approximating Macedonian practices to those of the wider Greek world.61 There are methodological as well as historical difficulties with this approach. Whilst it is perfectly legitimate to consider whether long-term historical connections with neighbouring cultural groups on the Asiatic side of the Bosporus had an impact on the organization and articulation of socio-economic practices on the European side, whether or not a relatively small kingdom, namely Odrysian Thrace, with a short institutional history, can legitimately be compared with a large territorial entity, the Achaemenid kingdom, whose institutional roots lie as much in centuries-old Babylonian and Assyrian practices as they do in the more recent innovations of the Achaemenid kings themselves, is a different kind of argument altogether. The trend of Macedonian historiography has been to compare Macedonian and Greek institutions. Whilst many Macedonian practices can be assimilated comfortably with Greek traditions (and recent research has done a great deal to illuminate these common traditions), there has been less scholarly enthusiasm to explore what made Macedonia regionally distinctive. It is even less clear why Thrace, a culturally related and neighbouring area, should be assumed to have been organized in ways that were structurally and politically fundamentally different from those of Macedonia, despite the shared experience of the Greco-Persian Wars. The ephemeral ambitions of Persia in southern Europe during the first three decades of the fifth century BC deserve a closer look if we are to evaluate the nature of the Persian legacy in economic terms.

(p.218) Herodotus, as we have seen at the end of Chapter 3, has quite a lot to say about the economic consequences of the Persian presence in Thrace and Macedonia; but the historian’s focus is largely on the territories south of Rhodope, unless we follow the view that one of the three columns of Xerxes’ Persian troops went much farther inland in 480 BC.62 What the historian has to say about these peripheral areas of his narrative is partly shaped by the underlying spatial pattern of his account; that is, Herodotus’ approach to the ‘ethnographic geography’ of the Greco-Persian Wars.63 Herodotus’ way of thinking about ethnography has been analysed in terms of a division of the known world into three parts, which were applied to his three continents—the ordered and familiar centre, the intermediate zone, and the unfamiliar fringes. He was dependent for information about the most remote areas on travellers’ tales, but interpreted these according to his three-fold classification. Thrace and Macedonia are not well defined within this scheme. In some respects, both regions are assumed to belong to the familiar sphere of settled communities, built houses, and comprehensible rites. In other ways they offered interesting material for his audience, notably in the stories about Macedonian royal origins and Thracian élite attitudes to agricultural work (5.6), religious ideas, and funerary practices (5.4–7). Farther away to the north, there was more exotic value in the Getic stories about Zalmoxis.64 Inland Thrace is more clearly defined (compared with Macedonia) as part of the ‘intermediate’ zone, between the familiar world of Greek-speaking farmers and the unfamiliar fringe of nomadic pastoralists.

Herodotus’ detailed narrative of military preparations and encounters merges with his ethnographic construction, which creates a texture of bright, colourfully sketched spaces, separated by deep shadow. The shadowy areas include much of the north Aegean region, particularly south-eastern Thrace, the Thermaic Gulf, and Rhodope. This is because they do not play a significant enough role—either in the war narrative or in his three-fold ethnographic scheme—not because they did not deserve objective treatment in some other memorial.

Persian documents provide a wholly different way of approaching ‘ethnographic geography’. In a recent review of the evidence for Greeks in the Persian Empire and Persians in the Aegean, Robert Rollinger has criticized the scholarly tradition that studies relations within the (p.219) geographical space of the Aegean and Near East in a dualist manner, that is, in terms of ‘Greeks’ and ‘non-Greeks’. The combined evidence from Greek and Persian sources, particularly the archive documents from Persepolis, reveals ‘intense and multi-faceted cultural interaction, which was able to prevail, largely unaffected by geo-political friction’.65 The discourse that polarizes interactions into a dialogue between Greeks and non-Greeks does justice neither to the agents revealed by the archive documents, nor to the nature of broader historical phenomena. Rollinger’s conclusions are based on new research that illuminates more clearly the identities of various workers listed in the Persepolis archive documents. One case relevant to the north Aegean region is the name ‘Skudra’, which has usually been taken to refer to the territories captured by Darius and Xerxes on the European side of the Straits.66 Achaemenid historians are now inclined to think that the ‘Skudraioi’ referred to in the Persepolis Fortification Tablets lived in north-west Asia Minor, not in European Thrace or Macedonia.67 So long as the word Skudra seemed to be applicable to the territories of the northern Aegean coastal areas that came under Persian rule from 513 BC, it was possible to refer both to ‘Skudra’ and to the Persian ‘satrapy’ in Europe, as if they were more or less synonymous. It now seems more likely that the inhabitants of the north Aegean coast should be among the various groups of ‘Yaunā’ referred to in the Persian documents. There is no obvious or apparent way of distinguishing amongst the various ‘Yaunā’, those of the (p.220) ‘mainland’, those who ‘dwelt by the sea’, or those ‘beyond the sea’, in the terms familiar from Greek accounts—Ionians, Athenians, or others, indeed between these and Karians, or Lydians, or Thracians. ‘Yaunā refers to an ethnos or a conglomeration of peoples, who lived at the western fringes of the empire and possibly beyond.’68

Bearing in mind this rather fluid categorization of peoples, both the ‘ethnographic’ distinctions made by Herodotus and the apparent lack of ethnic differentiation in Persian documents, we may understand why references to ‘Thrace’ in fifth-century BC documents and in later authors, including Strabo, are so imprecise and wide-ranging.69 ‘Thrace’, as it appears in Athenian documents, is a handy umbrella concept, which was applied unsystematically by Greek agents and writers. As a result, the territory to which the term was applied did not fit logically and consistently over the peoples who in practice occupied the designated terrain. Athenians still talked about ‘Thrace’ in the fifth century BC when referring to areas that were already being subsumed into the expanding kingdom of Macedonia. Meanwhile, other parts of ‘Thrace’ were being divided up—by the peers and rivals of the Macedonian kings—farther east and north. Stories circulating in the Aegean about distant peoples were not keeping up with what was actually happening there.70

If we now consider the suggestion that the Odrysian treasury was copying or continuing Persian royal custom through the practice of inscribing royal names on silverware,71 then the evidence is not quite what it seems. The inscriptions that appear to refer to geographical locations (Apros, Beos, Ergiske Sauthaba, perhaps Geiston)72 are puzzling, precisely because the names are all concentrated in a small area of south-eastern Thrace, while the find spots are without exception located north of the Balkan mountains, in other words outside the (p.221) territory of the Odrysian kingdom proper. Louisa Loukopoulou has restudied the inscriptions preserved on some of the vessels in the Rogozen hoard and has suggested that not only did these items come originally from the treasury of the most powerful Odrysian monarch defeated by Philip II, namely Kersebleptes, but that they were originally acquired as booty, and ended up as divided-up booty in élite burials of northern Thrace, having been stolen by the Triballians from Philip and parcelled out amongst local leaders.73 She concedes that the Odrysian royal treasury probably contained a heterodox collection of items acquired in a variety of ways, including tribute, gifts, and as war booty. Individual inscribed silverware, or groups of silver, such as the set of rhyta from Borovo, Rousse, does not really fit the interpretation that all the items north of the Balkan range were acquired by theft. Examples like Borovo are still best interpreted as gifts, even if the inscriptions reflect an earlier ‘treasury note’.74

So it is difficult to use these inscriptions as a basis on which to speculate about the methods used for documenting items that came into the Odrysian royal treasury. Notwithstanding the number of items preserved at Rogozen, the silverware it contained represents only some of the kinds of silverware in circulation. Precious metalwork that has survived into the twenty-first century was buried below ground, so that it could not easily be violated by robbers. Many of the fifth- and fourth-century BC tombs in central and southern Thrace were above-ground masonry structures, vulnerable to interventions. So far fewer of the gifts granted by Odrysian kings to their immediate followers survive.75 There may be a broad similarity between Odrysian and Persian methods of inscribing ownership on silverware, but in Thrace the inscriptions include at most, besides the royal name, weight marks alongside the apparent place name in the genitive case, but without any additional information relating to manufacture, to royal treasuries, or to the ‘king’s house’, as Achaemenid ones might. Plate that found its way into funerary contexts may have had texts added for other, more (p.222) metaphysical purposes.76 Stronach and Zournatzi have argued that the point of Thucydides’ contrast between Thracian and Persian practices of gift-giving (2.97.3–4) was motivated by contemporary war-time preoccupations with Persian money (during the later part of the Peloponnesian War), rather than being an intended ethnographic comment. At any rate, Xenophon adds an important gloss to Thucydides’ statement about Thracian gift-giving: ‘Then one Gnesippos, an Athenian, arose and said that it was an ancient and most excellent custom that those who had possessions should give them to the king for honour’s sake, and that to those who had nought, the king should give, “that so”, he continued, “I too may be able to bestow gifts upon you and do you honour”.77

We are entitled to conclude, on the basis of these different perspectives, that current research on Persian administrative archives suggests that we have even less reason than may previously have been thought to imagine an Achaemenid administrative area, a ‘satrapy’ on the European side of the Bosporus; that comparisons between Persian and Thracian administrative practice seem to have little substance, beyond broad regional and general structural resemblance; and that categorical claims about the ways in which rulers (Macedonian as well as Thracian) exercised power over their subjects are not substantiated. Royal powers were qualified in various ways—by social practice, as well as by traditional loyalties. What is more, analyses of social relations have tended to underestimate how the intensification of commercial interactions modified social practice, as the region became increasingly more accessible to a wider range of interest groups.

The Infrastructure Of Regional Trade


The infrastructure of trade relies on a substructure of subsistence and social practice. One of the most fundamental products that needed to be brought inland was salt. Under the peace terms negotiated by Aemilius Paullus, salt was banned as an import to Macedonia, although the (p.223) Dardanians of Upper Macedonia were allowed access to salt from the coast, which was taken to Stobi in Paionia (Liv. 45.29.12–13) for redistribution farther afield. The fact that Paullus put a price on it immediately suggests that there were revenue implications, so perhaps a tax on salt, or a monopoly on the sale of salt, had become a customary royal practice under the Argead kings.78 Salt is an essential component of diet, although we need very little of it in practice to satisfy the basic biological requirement. Nevertheless, the distribution of salt represents one of the principal drivers of the increase in intra-regional exchange in Late Bronze Age continental Europe, a process that underwent a significant step change in the first millennium BC, when there was a veritable explosion in the demand for salt and salt products, which included preserved fish, particularly in the most highly populated parts of southern Europe and the Mediterranean.79

The salty ‘limans’ of the Pontic coast were among the major producers of salt and salted products for export and we have already seen how this specialization probably lay at the heart of the ‘Monopoly War’ between Kallatis and Byzantion over Tomis. However, not all domestic salt was produced from the evaporation of seawater. In inland areas of the east Balkan peninsula, toponyms that include the name ‘Slatina’ may hold further clues about sources of salt from brine springs.80 On the north Aegean coast the lagoons east of Abdera created a number of interior lakes, including at least one salt lake, associated in Herodotus’ account with the site of Pistyros (Hdt. 7.109.2).81

Rivers, routes, and roads

Besides salt, a great deal of bulk traffic, including building materials, mill stones, and cattle, as well as containers of wine, oil, nuts, and various dry (p.224) commodities, travelled by rivers. The availability of water transport is another of the key linking factors in our ‘super-region’.82 The Haliakmon and the Loudias Rivers, like the Axios, the Nestos, and the Hebros, were major arteries for goods travelling inland during the summer months, when the volume of water was sufficient for flat-bottomed boats and rafts, and quiet enough to make larger transports viable. The results of river traffic can often be recognized by extended distribution patterns, with distinct concentrations following their course and those of major tributaries.83 What began perhaps as a system of regular boatloads of salt to particular regional nodes, distributed therefrom to other destinations, developed an increasingly varied range of components, as the existence of reasonably predictable consignments made it feasible to plan for other commodities that could travel with them.

Current research is beginning to illuminate the kinds of routes and roads that emerged during the first millennium BC prior to the expansion of metalled roads under Roman administration, which begins in the Balkans with the Via Egnatia. Most of the information acquired so far relates to the course of these roads, rather than to their structure and appearance. Paved surfaces are found in urban centres. Away from these foci, there is likely to have been a variety of surface materials, depending on terrain. Pre-Roman roads seem to have lacked a uniform character. In central and southern Greece, there are extensive traces of cart roads with parallel lines of stone.84 In northern Greece, Yannis Pikoulas has expanded a project begun in the Peloponnese with an ambitious programme in the Pindhos range, using upland fortresses built in the time of Philip II as a primary set of orientation points, on the basis of which he has explored a series of six separate routes across the high Pindhos west of Grevena, to link up with the lowland roads more readily identified during archaeological surveys, particularly around Edessa and in the (p.225) environs of Pella.85 Pikoulas has also investigated roads in the area of Mount Pangaion, partly for comparative purposes, including the possible routes of the Persian ‘royal road’.86 East of the Strymon, more than a century of work on Roman roads in the east Balkan area has been collated and developed by Mitko Madjarov, whose synthesis includes important remarks about routes throughout the region prior to the Roman network. In the Thracian Plain, as in Rhodope, there is plentiful evidence of the kinds of routes that preceded the formal planning of the Roman network. In the central plain, there is in many cases a close enough correspondence between the location of Iron Age sites and the lines of Roman roads to indicate the probable routes that existed, even when actual roads cannot be identified with certainty, particularly in view of changes in the bed of the River Hebros (Maritsa) dating to this very period.87

In Rhodope, there were three principal road routes to the Aegean, which have been studied with considerable confidence on the ground—a western one, from Bessapara (Sinitovo), west of Philippopolis, to Dospat, branching thence either westwards to Nicopolis ad Nestum and south to Philippoi, or south via Dospat in the direction of Drama; a central road, due south of Philippopolis, passing the Persenk peak (2,074 m), through modern Smolyan to the Drama plain; and an eastern road, through Assenovgrad and the Topolovo defile, past the peak of Sini Vruh (1,537 m), where it bent westwards in the vicinity of the Iron Age sites of Pavelsko and Luki, to resume a southerly course through Momchilgrad, in the direction of Abdera. All of these roads superseded earlier Iron Age routes through Rhodope.88 In the higher mountain ranges these undoubtedly became mule tracks rather than roads.

Perhaps the most intriguing questions about roads focus on the references to the waiving of taxes on road traffic, as stipulated in the Pistiros inscription found as Asar dere, near Vetren, central Bulgaria (ll 20–21: τέλεα κατὰ τὰς ὁδοὺς μὴ πρήσσειν‎), and to the granting of safe conduct for the goods transported, with the emporitai having sole right to open and close their vehicles (τὰς ἁμάξ/[ας] καὶ ἀνοίγειγ καὶ κλείειν‎). (p.226) The passage is opaque and requires some imagination to make sense of it. But the two components of the sentence must be understood as parts of a single statement—in other words, the phrase about taxation on roads is to be understood in direct connection with the statement about vehicles being opened. The gesture implied by this way of understanding the text, as Alexandru Avram has explained, would have been a key element in enabling commodities to make what amounts in this document to a guaranteed ‘extra-territorial’ journey ‘to Maroneia from Pistiros, or from other emporia, or from Maroneia to Pistiros, or to the emporia Belana of the Prasenoi’ (ll. 21–25), the first known instance of formal international commodity transportation.89

It is hardly surprising that the location of Pistiros has aroused so much discussion. Can it really have been located deep in central Thrace? Would it not make better sense if it were located somewhere in the Thasian peraia, nearer the coast of the Aegean? How can there be a polis in the heart of the continent of Europe?90 These are the kinds of questions that have been asked about the place name itself, irrespective of where the stone was discovered. The arguments about status once again resemble wider discussions about poleis. One of the best analogies to Pistiros is Naukratis in Egypt, which was also a large trading centre within a kingdom. Naukratis became a polis in the fourth century BC.91 On the other hand, Pistiros is unlike Egypt, because there were many historic links that encouraged institutional parity amongst trading partners. The long-term proximity and interactions that are implied by networks of exchange from the Bronze Age onwards in south-eastern Europe make it much more likely that the kinds of distinctions contemporary historians make amongst the ethnic communities of the (p.227) region have more to do with modern preoccupations about identities than they do with ancient institutional practice. Ancient perceptions of differences between Greeks and non-Greeks in the north refer mainly to those aspects of cultural practice where differences were much more specific, such as attitudes to authority and commemoration, not in terms of representative bodies or in exchange mechanisms.

Markets and currencies

Since commercial transactions during the first millennium BC were normally bilateral, confirmed by signs and symbols of ‘ritualized’ friendship,92 what we would expect to find is evidence of reciprocal relations and this is what the variety of coin types at major hubs seems to represent. At Nebet Tepe, the hilltop sanctuary and trading centre in the middle of Plovdiv (ancient Philippopolis), there are silver coins from various cities of Chalkidike, the lower Strymon valley, and beyond, as far as the Straits. The earliest issues of the late sixth to fourth centuries BC read like a roll-call of the main coastal trading harbours of the region: Akanthos; the ‘satyr and nymph’ issues of Thasos and Thasian imitations; coins of Neapolis; Dikaia by Abdera; Abdera itself; Maroneia, Orthagoreia, Ainos, Selymbria, the Thracian Chersonese, Parion, Kyzikos, Amisos; and beyond, Thebes, Athens, Aegina, Apollonia Pontika, and Istros. In the later fourth and third centuries BC, these were joined by numerous coins of Macedonian monarchs, by Pantikapaion, Kallatis, Histiaia, and the various ‘new style’ second-century BC issues of Athens, Thasos, Mesambria, the First Macedonian district, the magistrate Aesillas, and Mithridates Eupator, reflecting the progressive transformation of the economic drivers from the Aegean powers to Rome.93 This panorama of contacts reflects the rather complicated structure of exchange, requiring specific agreements between specific community agents in order for transactions to be validated. The range of coins enumerated at Nebet Tepe is indicative of the general trend, rather than being a full reflection of transactions.

The finds consist almost entirely of silver and some gold coins, supplemented by a number of copper alloy ones. The proportions do not correspond to what we would expect in a commercial centre, but rather to the tastes and priorities of collectors. There are far too few copper alloy issues in proportion to silver ones. The reasons are not hard (p.228) to find. Gold and silver coins have value today and stimulate interest. Copper alloy coins have, historically speaking, rarely been attractive to collectors and certainly not to the same degree as precious metals. On the other hand, the presence of relatively large denominations suggests that the kinds of transactions taking place were connected with significant quantities of the commodities exchanged. The evidence from Nebet Tepe seems to reflect periodic visits by merchants, who were interested in big transactions.

The value that might be derived from this kind of evidence in the northern Aegean has so far been explored in a rather limited way, despite good evidence of widespread distribution patterns and other indicators of economic relations, extending in time and space during the period covered by this book. Figure 5.3 shows the distribution of coins in the east Balkans emanating from the Thracian Chersonese.94 The hemidrachms of the Chersonese have been found as far north as Botevgrad and Loukuvit, well north of the Balkan range; and as far west as the headwaters of the Hebros, as well as a large number of places in between, along the middle Hebros valley. They often occur alongside similar denominations from Parion. We know that ambassadors from Parion were busy courting king Amadokos in 400 BC (Xen. Anab. 7.2.7, 3.17). So it is perhaps no surprise to find coins from these two cities so widely distributed. Three examples of markets present us with a range of possible scenarios for market-type transactions.


The first example is Pella. The agora at Pella provides a wealth of information about public and commercial activity over considerable periods of time, although these were brought to an end abruptly (by an earthquake, according to the excavators) because of the interrupted character of final activities—the quantities of terracotta statuettes ‘ready for sale’, as well as moulds and tableware in a destruction layer that evidently represents structural damage.95 The agora consisted of an almost square piece of ground (261.7 m x 238 m), built over sloping terrain with an (p.229)

                   Regionalism and regional economies

Fig. 5.3. Distribution of coins from the Thracian Chersonese (second half of the fourth century BC).

inclination from north-east to south-west, which necessitated terracing of the structures behind a central piazza, which measures 200.15 m by 181.76 m. There was a Doric colonnade on the south wing, faced by a line of stone piers on the north wing. Shops and offices lined the back of the colonnades, built of unbaked bricks on stone foundations, liberally roofed with timber and Lakonian-type tiles, some of which are stamped ΒΑΣΙΛΙΚΟΣ‎ or ΠΕΛΛΗΣ‎, although other stamps have also been (p.230) recorded. An elaborate system of pipes and drains carried waste water under the walls and surrounding roadways, while fresh water was supplied from wells.

Study of particular areas of the agora suggests medium-term association between certain crafts and certain areas or structures. Pottery manufacture and sale was concentrated on the southern part of the east wing, while farther north were stalls selling terracottas. The southern wing seems to have been reserved for fresh food, particularly liquid products; stalls for butchers and fishmongers have also been suggested. The southern part of the west wing housed imported pottery and lamps. Metalworking, including some evidence of work in precious metals, judging by the presence of litharge (from cupellation of lead), was practised on the outskirts of the precinct, where moulds and slag have been found at the east end of the south wing. A large section of the wing seems to have been occupied by stores of dry goods, including grains and flour. Wells in the southern wing contained waste material from pottery, metalworking, and food, including olive pits, walnut shells, chestnut husks, grape pips, and various seeds.

The north stoa evidently housed the main administrative quarters, including a hall with a Doric colonnade in the centre, where a large statue base occupied a prominent position, and at least one inscription listing the names of civic magistrates was on display, as well as various offices and cult chambers to either side. Clay seals from papyrus documents, bearing a club and an eight-pointed star, and inscribed (of) PELLA/(of the) POLITARCHS (ΠΕΛΛΗΣ/ΠΟΛΙΤΑΡΧΩΝ‎), point to what must have been the city’s principal administrative archive. In addition to this northern suite of rooms with administrative functions, there was a separate building in the south-west corner of the agora, which may have been a purpose-built headquarters for the city’s commercial magistrates. It had a central peristyle, supported by Doric columns and Ionic piers. Within the destruction layer from the collapsed rooms facing the colonnade was a collection of clay sealings, showing a grazing cow, some of which also bore the inscription Pellis/emporiou, or the names of civic magistrates. Besides the sealings there were writing materials, including ink-wells and pens, as well as a stone stamp seal and fresh clay.

The excavators have so far provided a vivid portrait of the most distinctive features of the last days of Pella’s main commercial hub and full publication will provide us with a more complete understanding. What is interesting about this manifestly purpose-built complex is the close proximity of workshops (metalsmiths, coroplasts), with sellers of fresh food. Stone masons, timber merchants, charcoal sellers, tanners, and similar insalubrious crafts are more likely to have been located (p.231) farther away from the agora.96 The products of commerce recovered from the excavation of the agora itself are closely tied to the period immediately preceding the market’s destruction: trade amphorae from Rhodes, Knidos, Kos, Thasos, Akanthos, and Chios predominate among Aegean imports, alongside others from Brindisi, Latium, southern Italy, and North Africa. Akamatis’ earlier examination of the proportions of stamped amphora handles produced a rather different distribution profile at Pella, with 32.8 per cent of the total represented by Rhodian vessels, 23 per cent Thasian, 20 per cent from the ‘Parmeniskos’ group (whether these originated in Mende or elsewhere in Aegean Thrace); and 12.5 per cent Knidian.97 The coins associated with the ceramic and other metal finds in the agora included a substantial number of copper alloy issues, both those issued by Macedonian kings, and a wide range of civic issues; while among the silver money coins, issues in the names of Philip II and Alexander III predominated, with some in the name of the Paionian king, Audoleon. Among the latest coins were Roman issues and a hoard of ‘new style’ Attic tetradrachms, dated in the late 90s BC. The range of civic issues may well tell us more about the specific commercial links between Pella and merchants or trades people from other Aegean and east Balkan communities. The data published so far merely hints at these.

The sealings at Pella depict a cow—the manifest sign of one of the area’s most prominent forms of livestock, and projected in another of the city’s names, Βούνομος, or Βουνόμεια (St. Byz. 515.7–8). Where then was the cattle market? Presumably this, like the ‘dirtier’ activities already referred to, was located farther away from the residential streets. The high level of planning in the city’s centre underscores the abstract importance of the agora—as a focus for civic expertise and strategic planning. Surely the agora was not built for the sale of cakes and terracottas? The scale of the square’s architectural elaboration implies that it served political as well as commercial purposes, managerial ones, rather than the quite modest retail functions of its latter years.

Pistiros and Krastevich

The people of Pistiros, the Pistirenoi, are referred to in an inscription carved on granite and found reused at the Roman mansio, known from (p.232)

                   Regionalism and regional economies

Fig. 5.4. Adjiyska Vodenitsa, identified with ancient Pistiros: approach to the eastern gateway from inside the city (looking north-east). The square external tower is behind the western threshold slab of the city gate in the centre, the roadway on its right, and the northern stretch of the fortification wall on the centre left

late antique itineraries as Bona Mansio, 2 km north–east of the river port at Adjiyska Vodenitsa (Hadji’s water mill), near Vetren, a small, picturesque, modern town in the foothills of the Sredna Gora (Fig. 5.4). The local geography of the western end of the Thracian Plain in central Bulgaria has been altered by radical changes in the riverbed of the Maritsa (Hebros) River, which have resulted in a flat alluvial plain. This gives little hint of the ancient activities that took place on its banks. The late antique mansio was positioned on a spur overlooking the Roman road between Byzantion and Serdica, on the northern bank of the river. Below these foothills, successive braids have formed when the river has cut a new course. These successive cuts have destroyed ancient riverside structures, including much of the pre-Roman settlement at Adjiyska Vodenitsa, whose remains continue to be eroded by the river.98 The surviving traces of an urban centre are preserved on either side of the eastern gateway, together with adjoining sections of a (p.233) powerful two-faced masonry fortification wall, which protected a dense concentration of structures, including one (Building no. 1) that evidently had administrative functions, rather like the south-west building in the agora at Pella. Large numbers of coins, especially copper alloy but also silver coins, lead weights, and commercial scales were found there. Many structures had tiled roofs. Some of the tiles were locally made, whilst others (including tiles stamped with the letter ‘A’) were shipped upriver from south-eastern Thrace. Commercial amphorae from Chios, Thasos, Herakleia, and the centres producing vessels of the so-called ‘Parmeniskos’ group were certainly intended for local consumption, but others were evidently distributed to centres north of the Hebros that were not accessible by river. Relative percentages, as at Pella, are based on stamped handles rather than quantified ceramics, and the period in question spans the second half of the fifth to the second half of the third century BC.99

Imported ceramic provides an indicator of consumption in the first place, but it is also one sort of yardstick, albeit an indirect one, of commercial patterns. At Adjiyska Vodenitsa trade patterns are reflected in many other ways too. Metallurgical waste, as well as moulds, tools, and finished metal items, in precious as well as base metals, indicates the production of a wide variety of articles on-site, of which at least some were intended for sale and export. The unusual range of artefact styles suggests that there was a ferment of creativity at the heart of this exchange centre.100

Graffiti from Adjiyska Vodenitsa show that the inhabitants came from a variety of backgrounds. There are Greek names of broad geographical distribution, Ionian, Macedonian, and Thracian names.101 The range of coins found here points to a similar variety of visitors and local or regional agents, with local representatives identifiable behind the Thracian regal issues, the most regular visitors shown by Thasian, Parian, Maronitan, and Chersonesian issues, which are among the commonest in silver and bronze. More occasional issues, such as those of Ainos, Kypsela, Sermyle, Kardia, Lysimacheia, Apollonia, and Mesambria (p.234) Pontika, point to other commercial travellers.102 As I have argued above, the mechanisms for ensuring the supply of any commodity meant that every community had to look after its own interests. For periods after the mid fourth century BC, when Macedonian coins progressively became available and widely disseminated, it becomes harder to connect specific users and to map their activities spatially. Individual coins can, of course, only be interpreted as indirect evidence for the movement of people. Yet, as we have seen, we can detect in the countermarks on coins the financial decisions of various authorities operating in the hinterland of Byzantion during the final two centuries BC, whether these were small civic entities or princes.

Apart from the hemidrachms of the Chersonese, which succeeded a not dissimilar, albeit less numerous, distribution of Cyzicene staters, one of the pre-Macedonian coin types that enjoyed a particularly wide outreach were the ‘satyr and nymph’ silver staters and drachmae of Thasos, or imitations of them. Examples have been recovered not just at sites in relative proximity to Thasos, but also from known, but comparatively distant centres of exchange, such as Vardarski Rid, near Gevgelija, FYROM, considered by its investigators to be one of the chief centres of Paionian authority, and Krastevich, a Thracian settlement and sanctuary in the foothills of the Sredna Gora, east of Strelcha.103 Visual and scientific analyses of selected coins show that some, perhaps many, were not made on behalf of Thasians. Imitations could be copied to boost the supply of particularly popular monetary media; they might also be deliberate attempts to copy good coins using a different alloy. How imitations should be evaluated as valid media of exchange depends on whether rival issues were acceptable to different authorities. In modern contexts, the deliberate falsification of coinage is treated as a criminal offence, in order to discourage crime and to protect the monopolies of issuing authorities; but a moderate level of falsification is tolerated, in the case of coined money at least. At Athens the acceptance of false coins masquerading as Athenian ones by the official responsible for verifying coinage was a punishable offence, not least because of the well known (p.235) purity of Attic silver money. However, this does not mean that false coins were not accepted in other transactions. The existence of the regulations strongly suggests that false coins circulated freely.104 The wide dissemination of ‘imitation’ Thasian issues reflects the widespread acceptability of the types, but does not tell us who made them. What matters is that they were acceptable as a medium of exchange and the occurrence of such coins in limited quantities, over a wide area, seems to be consistent with such an interpretation.

Investigations at Krastevich, a fifth- to fourth-century BC settlement nestling in the foothills of the Sredna Gora mountains, has, in the space of a few excavation seasons, produced surprising and unique features that force us to rethink what was happening deep in the interior of southern Europe. The investigations have focused in two locations, a terrace above the River Pyasechnik (Pamuk Tepe), and a hilltop (Sekiz Harman), c.1 km to the north-east, high above it. At Pamuk Tepe, and in an area c.50 m by 90 m, a number of structures with rubble stone foundations have been revealed, including two large multi-roomed building complexes, at least one of which had very well-built outer walls and inner staircases to an upper floor level (Fig. 5.5). The overall

                   Regionalism and regional economies

Fig. 5.5. Krastevich, near Strelcha, Bulgaria, plan of part of the town, showing the storage complex (left).

(p.236) plan of this building, with three long, narrow rooms and a stone waste-water channel (perhaps associated with industrial processes), resembles a well-protected storage block. Finds of storage amphorae and non-local coins (late fifth-century BC tetradrachms of Ainos and a pot hoard, containing silver coins of Thasos, Kyzikos, and Aegean Neapolis; issues of Apollonia and Mesambria), likewise indicate strong external commercial relations. The quality of the moveable finds, including metal ornaments and imported Attic fine ware, reinforce this impression. Yet none of these features prepares us for what was constructed on the hill above (Fig. 5.6). A limestone stylobate of almost square form, made up of well-finished ashlars, supported a stone colonnade, carved from cylindrical drums. The excavator, Mitko Madjarov, interprets this as the foundation for a sanctuary. It evidently had two phases, with a granite foundation of large masonry blocks in its refashioning. This does look like a temple, or temple-like structure. Although there are carved columns inside funerary monuments in various parts of the region, nothing in pre-Roman Thrace resembles this highly ambitious plan, which is hard to reach even now. Why such a monument was built high up in the hills above the Thracian Plain is one of the questions that future research at this site will grapple with.

                   Regionalism and regional economies

Fig. 5.6. Krastevich, near Strelcha, Bulgaria, plan of the upland sanctuary.

(p.237) If merchants really were travelling so far inland, what were they hoping to take back in the opposite direction? The most plausible explanation is refined or bar metal. As I have argued in Chapter 4, the extraction of metals was a highly specialized skill, while mining and metallurgy were among the activities that indigenous groups had invested in over many centuries. The hills on the north-western margins of the Thracian Plain, north of Plovdiv, are rich in metals and the rivers are gold-bearing. Local place names reflect these prominent resources, particularly those referring to gold and iron. Besides these high-value items, cattle, or cattle products, particularly pelts, skins, and leather, might also have been added from a region well stocked with herds, as well as other foodstuffs. The northern tributaries of the River Hebros provided a ready means of transportation of bulk goods during the summer months, though an alternative overland route also existed to Adjiyska Vodenitsa and other sites along the Hebros, and thence either downstream or through Rhodope.

The whole region created by the triangle of land between the Sredna Gora range, the Klisoura defile (at the western end of the Thracian Plain) and the middle valley of the River Hebros around Plovdiv constitutes an intense concentration of high-value, prestige monuments and moveable artefacts in the period between the fifth and third centuries BC. These forms of local consumption reflect the products of inter-regional exchange patterns on a substantial scale. There are similar concentrations in the Valley of the Roses, in south-eastern Thrace (in the area around Kirklareli),105 and in the combined regions of Emathia and Pieria, in lower Macedonia. All these particular regional concentrations coincided with the principal seats of princes and kings. The physical manifestations of significant consumption partly reflect the exploitation of regional resources and in part the concentration of royal and princely funds.

Byzantion—a major hub of northern commerce

Our third example is Byzantion (Fig. 5.7).106 The irresistible rise of this city, long before it was chosen as the New Rome by the Emperor Constantine, has somehow escaped the kind of attention from classical historians that its wealth and strategic importance surely deserve. Almost all histories of Constantinople begin with Constantine, not with the city (p.238)

                   Regionalism and regional economies

Fig. 5.7. Istanbul, view of the Bosporus from Topkapi palace, the heart of the ancient city of Byzantion.

of Byzantion itself and no monograph on Byzantion has been written since V. P. Newskaja’s, originally published in Russian in 1953.107 The scholarly and comparative evidence collected in the Inventory of Archaic and Classical Poleis helps to redress this strange imbalance. Byzantion can, as a result, resume its place among the truly exceptional city-states of the fifth and fourth centuries BC, alongside Athens, Sparta, and Syracuse. The territory controlled by the city of Byzantion in this period (estimated by Loukopoulou and Łajtar as c.1500 km2), extended from a north-western frontier on Lake Derkos, a famous fishing lake in close proximity to the Black Sea coast, to the River Athyras (Kara Su), and along the Athyras lake (Büyükcekmeçe) in the south.108 How and when the city acquired such extensive territory is not certain. According to Ps.-Skymnos (722–23), Byzantine territory included the locality Phileas, or Philia, in the north, on a promontory extending into the Black Sea. To the north it bordered land controlled by Thracian dynasts (Plb. 4.45.1); (p.239) to the south-west, Byzantine land bordered that of Selymbria.109 Most of the ancient sources that provide specific details belong to the city’s post-classical history. There are sufficient pointers, nevertheless, in pre-Hellenistic sources to indicate just how significant the city had become by the time it was besieged (unsuccessfully) by Philip II of Macedon.110 Xenophon’s sojourn in the hinterland of Byzantion provides only incidental details that fail to emphasize the city’s real territorial capacity. Moreover, its reach extended not just inland, but (periodically, at first) across into Asia. The city took possession of Kalchedon in the middle of the fourth century (Dem. 15.26).

In the fifth century, the city’s fortunes were variable. The Hellespontine region was perceived as a major asset by the Persians, who hoped, perhaps, to hang on to it after their retreat from most of the European mainland (Hdt. 9.101), but were in the end forced out, step by step, at the hands of the Delian League. In the middle of the fifth century BC the city had liquidity problems and created a range of new sources of revenue by leasing sacred land and land belonging to private associations, selling off shop spaces, or fishing and salt-collecting rights, as well as offering citizenship for 30 minas to those with one citizen parent ([Ar.] Oec.II.3c, 1346b).111 In the Politics (1291b 21), Aristotle talks about the high proportion of fishermen amongst the population of the city, a factor that, in the philosopher’s view, affected the political climate of Byzantion, much as the rowers of the Athenian fleet affected that city’s politics. Theopompos, on the other hand, talked about the propensity of the inhabitants of such a major emporion to gravitate in the direction of the market place and the harbour, lingering in the wine shops.112

The propensity of Byzantines to focus their energies on activities in and around the harbour reflects the importance of the Golden Horn as a centre of transhipment as well as regional exchange and, at the same time, the capacity of any authority located there to exploit the unique opportunity to extract revenues from those passing through the channel between the Bosporus and the Hellespontine Straits. Nothing illustrates this more vividly than the capture of the Athenian grain fleet in 340 BC by Philip II of Macedon, whilst he was besieging the city of Byzantion. This consisted of 180 ships, the dominant part of a convoy of 320 vessels, evidently transporting various other goods—incidentally a useful (p.240) indication of the level of trade in other commodities along this route. The convoy was assembled at Hieron, a harbour on the Asiatic shore at the northern end of the Bosporus.113 Hieron provided the principal shelter at the entrance to the Bosporus, allowing ships passing out of the Black Sea to wait for favourable winds for the southward passage. It was never much more than a harbour with a sanctuary. Control of the harbour was exercised primarily by warships, which were at this time under Athenian control.114 Philip not only seized the ships, but sold their cargoes—the grain, hides or live animals, timber, and other materials—whose value realized the eye-watering sum of 700T in cash for the Macedonian king.115

The incident was as significant in local terms as it was in international perspectives. Although Philip’s success was the result of a ruse, it demonstrated that the size of the Athenian navy was no longer effective in setting the terms on which the transit trade operated through the Bosporus. After 340 BC the Athenians effectively ceased to be competitors for naval supremacy of the north-east Aegean.116 This left the commercial control of the Bosporus almost exclusively to the Byzantines, although the shores and channel itself continued to be a focus for occasional military confrontations.117 It is not easy to appreciate what this meant in terms of shipping. There is little surviving information directly relating to the city’s naval capacity. The putative beneficiaries of the convoy seized by Philip would have included the Athenians, Byzantines, Rhodians, and Chiots.118 During the third century BC, the Byzantines were reluctant to get involved in inter-state encounters outside the Bosporus and the Hellespontine Straits. Byzantine naval vessels are mentioned in the battle of Chios against Philip V of Macedon, alongside Rhodians and other, unnamed ships.119

(p.241) The Aegean scope of Byzantine activity gives little indication of where the city’s naval strength really lay. In order to understand this, we must get behind the story of Philip’s capture of the convoy of ships through the Bosporus. The combined evidence of the different Greek historical and forensic sources on this incident, which are given particular coherence by a relevant passage in the Commentary on Demosthenes of Didymus, make it clear that ships were regularly convoyed through the channel and the Hellespontine Straits, in order to avoid the kind of arbitrary seizure along this route that Philip’s capture in 340 BC represented. Gabrielsen has argued that the sitopompeia, the convoying of grain and other commercial transports, had become a more or less regular way in which ships crossing the Straits and Bosporus to the Black Sea arrived at Byzantion, particularly if they had to wait for favourable winds before proceeding northwards from there.120 Similarly, convoys gathered at Hieron on the return leg of the journey, to be escorted back into the Hellespont.121 The duty of escort was undertaken by Athenian ships in the fifth and fourth centuries. Thereafter, Byzantion probably took over the same kinds of functions. The cost of such protection came to be built into the budget of bottomry loans (Dem. 35. 10–13). The forensic speeches of Demosthenes contain frequent references to the hindering of merchant vessels (ta ploia kōluein) and the actual commandeering of merchant ships to unplanned harbours.122 The passage from the Hellespontine Straits to the Bosporus became a bottleneck for ships, which were obliged to follow the winds and currents peculiar to these shores. In Chapter 43 of Book 4 in his Histories, Polybius describes how the currents flow in a zigzag manner from the Bosporus in the direction of Byzantion, making it easy for ships to stop off at the latter and difficult to dock at Chalkedon. The historian’s account at this point of his narrative is entirely focused on the behaviour of waves, currents, ships, and natural features of the shoreline. He refers to the sanctuary of Zeus (Ourios) at Hieron and to the Hermaion on the European side, north of Byzantion. These more contextual details are woven into a reflection on the gradual silting up of the shorelines adjacent to the Black Sea and Maiotic Sea (the Sea of Azov), which diverts from the principal subject in this part of Book 4, namely the extraordinary success of Byzantion and its highly (p.242) advantageous position (Plb. 4.38.1–8). The strong maritime focus of Polybius’ account, as well as its nice political perspectives, would suggest that he was drawing directly on a Rhodian source.123

The historian, who is here referring to the situation some one hundred years later, c.220 BC, goes on to say that if the Byzantines had allied themselves in the past with the Gauls, or, at the present time, with the Thracians, then they (the Byzantines) would not be, as indeed they were, the benefactors of all the Greeks. Polybius was being economical with the truth. The sort of control created by the Byzantines over the Straits was a benefit to merchants but at significant cost. The market-based economy of the import and export trade in grain and other commodities through the Straits represents only one aspect of the investment costs for merchants and ships’ captains, who were also obliged to pay the ‘protection’ costs of the predatory arm of this exchange process, which operated in strict synergy with the commercial one.124

The origins of the predatory component in these relationships lie in the measures set up by Athenian officials in the aftermath of the Greco-Persian Wars, in order to secure the Straits for the Greek allies. The Byzantines were among the ‘big spenders’ amongst Athens’ allies. The tribute that they paid to the Athenians on behalf of the League in the fifth century BC varied, but it was never less than 15T (the amount paid in 450/49 BC), whilst in 430 BC this tipped 21T. If we add to this the sum paid to the Athenians for tolls through the Hellespont, collected by local tax-farmers, then this could have raised the overall amount paid to Athens up to between 41T and 50T.125 When such tolls came into existence is obscure. The evidence for tolls rests on the ten per cent tax, the dekate, referred to in the so-called first Kallias decree (IG I3 52A = ML GHI 58A.7, variously dated to 434/3 or 431 BC), and in any case earlier than the similarly-named tax in Xenophon’s Hellenika.126 Xenophon describes how Alkibiades sailed to Chrysopolis in 410/9 BC, then a locality in the control of Kalchedon, on the coast opposite Byzantion (identified with modern Üsküdar), fortified it, and set up a toll station (dekateuterion). Gabrielsen argues, persuasively, that there was an earlier toll, operating in a similar way, originally located at Byzantion. But when the city fell into Spartan hands, in 411 BC, the Athenians lost control of their toll-raising powers until the city was regained as an Athenian (p.243) dependency by the same Alkibiades, in 408 BC (Thuc. 8.80.3; Xen. Hell. 1.3.15). Not so Kalchedon, which, save for the year when Alkibiades kept control of that part of the Asiatic coastline, passed into Spartan control. References to the dekate reappear in early fourth-century BC sources, but it is not clear whether they refer to the Hellespontine toll, or to other revenues raised in a similar way.127 At any rate, the Hellespontine dekate was the best known toll of this kind (Dem. 20.60; 23.177), while the need to provide protection for ships in the Straits was already apparent at the time of the Greco-Persian Wars, and recognized by the Athenians in the appointment of ‘guards’, the Hellespontophylakes, who acted as military policemen, based at Byzantion.128

In the third century BC, Rhodes became the city’s dominant Aegean partner. So the attempt to reintroduce the toll at the Bosporus, which the Rhodians interpreted as a direct blow to their commercial interests, became the lightning rod for a declaration of war in 220 BC. There are several separate but inter-related aspects of this conflict between two former partners. Polybius, as we have seen, uses the incident to explore Byzantion’s unique position, which enabled the Byzantines to set terms for ships and merchants wanting to trade in the Pontic area. They had further strengthened their territorial position by the recent purchase of Hieron and were in the process of acquiring territory in Mysia, on the Asiatic shore of the Bosporus.129 Stephanos of Byzantion refers to Astike, (p.244) the region between Apollonia Pontika and Perinthos, as being part of Byzantine territory. The likeliest period of the city’s expansion into these areas coincides with Polybius’ account, and helps to explain the heightening of tension on the landward, European side of the Bosporus, as well as in the Straits.

Byzantion had for many years paid a ballooning tribute to the Galatian enclave at Tylis in south-eastern Thrace.130 At the same time, Byzantion’s territorial claims were being challenged. The Byzantines are compared by Polybius with Tantalos: they could see the fruit of their labours but not enjoy them, as ‘their’ crops were being either damaged or appropriated by the local Thracian inhabitants (4.45). The city expanded territorially because it was economically successful. Nor was this success due simply to brinkmanship in the extraction of protection money for ships passing through the Bosporus. The Byzantines had developed an expertise in organizing monopolies. The attempts to raise civic revenues through the sale or leasing of certain rights, including fishing and salting, show how this kind of expertise may have emerged. In the third quarter of the third century BC, the Byzantines progressively created a monopoly of control over Bosporan trade, which, with the purchase of Hieron, and consequently the city’s complete dominance of the channel, was in danger of stifling the very partners who helped to make this a success story for Byzantion. This critical juncture masks a different kind of monopolistic activity, manifested in the city’s protest, in c.255 BC, at the creation of a more remote monopoly, namely at the western Pontic harbour town of Tomis, by its near neighbour, Kallatis.131 If Kallatis could have created a situation (by monopolizing Tomis) in which the city had a critical advantage in the pricing of salted fish, then this would have had a direct effect on the goods that Byzantion specialized in selling, namely salt and preserved fish (Plb. 4.38.4–5). The convoying of Black Sea grain transports made Byzantion a major clearing house for grain and other regional produce that often travelled alongside cereals—slaves, cattle (and hides), honey, and wax. Much of this arrived at the city’s (p.245) markets by sea, but by no means all. As I have argued in Chapter 4, large quantities of live animals on the hoof and cured hides, as well as finished products, descended on Byzantion from Thracian centres in the interior, to supply the wide range of demand that the harbours of the city attracted.132

The Byzantines were forced to back down when their Asiatic possessions were occupied by the ruler of Bithynia, Prusias I, who encouraged the Thracian assaults on their back yard. In 217 BC, the old configuration of allies re-emerged after the impasse of three years before, when envoys from Byzantion, Chios, Rhodes, and Ptolemy IV offered to negotiate between Philip V of Macedon and the Aitolians (Plb. 5.100.9–11). Having, in all probability, attacked and then made peace with Byzantion in 200/199 BC, Philip V maintained good relations with the city in his later career, as did his son Perseus.133 The extraordinary success of the city of Byzantion was in part the result of shrewd use of natural advantages; in part a consequence of regional specialization, backed by a ‘sheltered’ monopoly;134 and partly the fruits of a flexible strategy of negotiation and networking, at one time with the Delian League, at another with its regional neighbours, and with its principal trading partners, particularly, but by no means exclusively, with Rhodes. When Roman officials set about organizing a tax-collecting regime in the region in the early first century BC, Hieron and Chrysopolis continued to form important nodes in their own network of toll stations. All cities and regions, without exemption, were obliged to pay the portorium and the Bosporus was considered among the most lucrative sources of revenue. The extension of the Via Egnatia as far as Byzantion may well have helped to provide the physical means of putting into effect this economic pincer movement.135

Regional Divisions

The division of the old kingdom of Macedonia into four republican states, as a result of Roman intervention, never found favour with its (p.246) inhabitants; they quarrelled among themselves because, says Polybius (31.2.12), ‘they were unaccustomed to democratic and representative government … they preferred of course the more primitive institutions of their own monarchy and the Macedonian people at arms.’ This may have been what Polybius thought of Macedonian regional administration; but it is a view that singularly fails, not only to illuminate how Macedonia was organized under the kings, but why Macedonians liked the former regional structures created under their Argead and Antigonid rulers. Polybius ‘never shows any understanding of what mattered to a Macedonian, he judges Macedonian policy invariably in terms of Achaean advantage.’136 Frank Walbank’s admission of the ancient historian’s blind spot deserves serious reflection, because regional organization is crucial to an understanding of how the monarchies of Macedonia and Thrace operated in practice, and how we should view the links between central and local decision-making. Polybius, like Athenian orators such as Demosthenes and Demades, could only see kings as autocratic figure-heads. They failed to see how kings could also be managers and coordinators in regions where the distances between one major population centre and another, and especially between one lowland area and another, could be substantial, separated as they were by mountains, forests, and lakes. The effective exploitation of such regions was necessarily different from the experience of central and southern Greeks, inured to intense competition over territorial resources. The clearest indication of these royal coordinating functions is reflected in the Pistiros inscription, whose regulations are predicated on the existence of inter-regional roads, tolls (and, what is more important, the remission of tolls), as well as legal mechanisms, the overall coordination of which can only have been put in place by supraregional authorities, in this case, the Odrysian princes. Kings could underwrite major infrastructure projects, such as roads and ports, including riverside installations. The implementation of royal enactments fell to regional subordinates. Regional administration was the key component to the success of any central organization. Far from being rivals to power, regional administrators were the real force behind crown authority. Kings might circulate within their kingdoms, but regional administrators were the day-to-day decision-makers, policemen, and dispensers of justice. The scarcity of documentary evidence is not a sufficient reason for doubting that this echelon of power existed. Regional power was autocratic and seigneurial, emerging from the princedoms of the archaic (p.247) age. We glimpse this caste in the late archaic coinage of the Pangaion area, where rulers’ names appear on coin legends alongside ethnic ones (notably Getas of the Edoni; Bisaltai, Tyntenoi, Oreskioi, Zaielioi, and others); and to some extent in regional prosopography (Arrabaios, king of the Lynkestians: Thuc. 4.79.2, 83.1). The persistence of regional names—Emathia, Pieria, Almopia, Amphaxitis, Eordaia, Tymphaia, Lynkos, Elimeia, and others, including those east of the River Axios: Mygdonia, Krestonia, Anthemous, shading into similar ethnic names in Thracian-speaking areas, the land of the Odomantians, Sintians, Bessoi, and Odrysians—in short, all the regional levies called up by King Sitalkes in 429 BC, and the Macedonian levies that responded, provide ample evidence of how regional organization was articulated.137 The marriage between Sitalkes’ nephew Seuthes to King Perdikkas of Macedon’s daughter Stratonike, the diplomatic resolution of this invasion, illustrates the critical role of princes and princesses in government. This was the upside of dynastic diplomacy, even if factions and feuds were the downside.

Hatzopoulos has attempted to sketch the evolution of Macedonia’s regional infrastructure, which eventually emerged as four districts, probably from the time of Philip II, although the nature of these four divisions remains problematic.138 If these really were similar to the Roman divisions, as he suggests, then it is hard to see why Livy goes to such lengths (45.29.4–9) to describe how the new regions were to be arranged, and Polybius’ remarks about the negative response of Macedonians to these divisions demand some explanation. There was undoubtedly novelty in the two easternmost districts, where parts of Thrace and Paionia were to be incorporated. We have much to learn about how the regal merides operated over time.

Maria Gabriella Parissaki has recently reviewed the districts of Thrace, the strategiai, that are documented in a range of late republican and early imperial inscriptions, principally dedications, which show how the pre-conquest Thracian élite was incorporated into the new Roman administrative machinery that came to constitute the foundations of Roman provincial administration.139 One of the difficulties that has beset students of Thracian strategiai is the fluctuating number of these regional divisions. The reordering of epigraphic alongside historical data has enabled Parissaki to propose a new hypothesis about regional organization in Thrace. The earliest system that can be discerned from the (p.248) dedicatory inscriptions indicates a regional organization of fifty units (Pl. HN 4.40). Exactly how these regional units might be reconstructed in terms of actual modern topography is still unclear. Nevertheless, this prior system was superseded, at some point in Vespasian’s reign, by a reduction to fourteen regional units (Ptol. 3.11.8–10).140 The fifty divisions that are variously recorded in the earliest inscriptions of the late republican period reflect the deep roots of regional administration in the days of Odrysian rule, when territories were organized along ‘tribal’ lines. By the first century BC, the fulcrum of royal authority had moved from the middle Hebros and Tonzos river valleys to Bizye/Vize, in south-eastern Thrace (Pl. HN. 4.47), where the princes of the Sapaian dynasty had created an impressive civic centre, regional capital of the Astai, with a theatre and other civic amenities.141 The intensity of economic activities in and around the Bosporus may well have contributed to this relocation of the centre of power in the south-east Balkans. The kinds of predatory relationships that existed between Byzantine ‘protectors’ and the ships whose security they protected were paralleled by predatory relationships from the landward side, represented most vividly in the tribute demanded of Byzantion by Galatians from their enclave in south-eastern Thrace. Nevertheless, coinage circulating in the region shows that market relations also operated between the different political entities of the interior, on both sides of the Bosporus, with coin types and counter marks signalling the acceptability of specific issues. The very success of Byzantion made its buying power a magnet for many social and ethnic groups in the northern Aegean. The city’s economic success and dominance of the Black Sea trade also enhanced its political stature. Byzantion found itself among the key partners in major international negotiations, as we have seen in the case of Philip V of Macedon. The Ptolemies had courted Byzantion from the 270s BC onwards. This international dimension gave the city further assets, which the formerly dominant landward powers—the Odrysian kings of Thrace, the Antigonid kings of Macedon—and would-be powers, such the Galatians on either side of the Straits, could no longer compete with.


(1) Wallinga 1993, 130–44; Davies (forthcoming 2013a); on Mediterranean ‘connectivity’: Horden and Purcell 2000, 123–72.

(2) As the contributions to the periodic colloquia on Illyrian history and archaeology make clear, the western half of the Pindhos was mainly, though by no means exclusively, orientated towards the Ionian Sea (see e.g. the contributions to Cabanes and Lamboley 2004 and Antonetti 2010); Hatzinikolaou 2009 provides a survey of upper Macedonian historical geography up until the imperial reorganization: Livy 45.29.5–9, with Hatzopoulos 1996, I, 231–60, on ‘the Districts’ of Macedonia, esp. 231–40.

(3) For survey literature that is particularly relevant here, Doonan 2004, 93–118, on Sinop and its environs (especially timber); Bintliff 2009; Constantakopoulou 2007, esp. 20–8 (island connectivity); 38–60 (religious networks); 228–53 (islands and their peraiai).

(4) Elton and Reger 2007, 11–15; Reger 2011, 371–8; Reger 2013a and 2013b; Oliver 2006a; Constantakopoulou 2007, 245–52; see above Ch.1, ‘The geography of north Aegean economies’.

(5) Prato: Horden and Purcell 2000, 94, 148; Golden Horn: Matschke 2002, 467, 471–2 (streets with dedicated crafts, including boot-makers, belt-makers, and furriers; the hunters’ district [τῶω̂ν κυνηγωῶ̂ν‎] near St John’s Gate, money-changers, warehouses, craftsmen, and vegetable sellers); Thasian marble: there are nine contributions (covering different periods) in Part I of Thasos, Matières Premières, ‘la Pierre’, 15–126; cf. also Vavelidis et al. 2006; more generally on geography and innovation, see Archibald 2012a.

(6) Ch. 4 n.91.

(7) See Ch. 4 for grey-faced pottery; silver phialai: see below, n.56.

(8) Reger 2011, 378–83.

(9) Tzochev 2010 and other contributions in Kassab Tezgör and Inaishvili 2010; Tzochev 2011; Tzochev (forthcoming).

(10) See below, n.33.

(11) Archibald 1998, 49–52, 126–35; Nekhrizov and Mikov 2000, 161–70; see further below on the Pistiros inscription and its implications for movement across Rhodope.

(12) See esp. Stoyanov 2011 and other contributors to Tzochev et al. 2011.

(13) Lawall (2005) discusses problems of quantifying amphorae; Lawall (2011) explores the challenge of identifying amphora contents; Tzochev (forthcoming) and Panagou (forthcoming) present recent work based on the archive of Virginia Grace in the American School of Classical Studies, Athens, as well as a wide range of published data.

(14) Panagou (forthcoming) identifies 58 potential centres of amphora production in the eastern Mediterranean, of which 51 have been located with reasonable confidence.

(15) Archibald 1998, 116–17 and fig. 9.5, for a preliminary analysis of sites; Triandaphyllos 2007 (middle Hebros valley finds); Balkanska and Tzochev 2008 (Seuthopolis); Tzochev 2010 (southern Thrace).

(16) For Pella and Adjiyska Vodenitsa (Pistiros), see further below; Krastevich: M. Madjarov, D. Tancheva, AOR 2007 [2008] 217–20; 2008 [2009] 195–8, 197 fig. 3 for plan, 240–3; 2009 [2010] 200–1; 2010 [2011] 189–90; 2011 [2012] 166–8; Vasil Levsky: K. Kisyov, AOR 2008 [2009] 237–40; cf. also Archibald 1998, 126, 141, 226–9, for earlier literature.

(17) Avram 1996 (Istros); Conovici 2005 (bulk imports at Istros, Tomis, and Kallatis compared); Debidour 2008 on the commerce of Thasian amphorae; Tzochev 2010 and forthcoming; cf. Panagou (forthcoming) and Lawall (forthcoming).

(18) The end date for Thasian imports of c.275 BC comes from a revised study of the material from Seuthopolis by Balkanska and Tzochev (2008).

(19) Kuleff et al. 2007 (INAA analysis); Stoyanov 2011, esp. 192–5 for magistrate Melseōn, whose name alludes to the mythical founder of Mesambria, Melsas, and is also the name of a coin issuer at Mesambria between 275 and 225BC (IGBR 12 308 (6)), as well as the middle name of a strategos, one of a group of six civic magistrates who dedicated an altar in the late second century BC (IGBR V, 5104).

(20) Wrecks of wine amphorae: Gibbins 2001; Lawall 2011 and forthcoming; see Ch. 7 for further evidence of consumption patterns and quantification by context at Pistiros.

(21) Oliver 2007, 49–68, esp. 48–55 (Piraeus); 86–100 (population changes in the late fourth and third century BC).

(22) Rotroff 2005, 13–28.

(23) Krastevich: see above n. 16; I am grateful to the excavator, Mitko Madjarov, for his current interpretation of the data accumulated up until 2012.

(24) Domaradzki in Pistiros I, 30–1 (coins of Philip II, Alexander III and early issues of Lysimachos show traces of fire damage; examples of Lysimachos’ final series do not); Katinčarova in Pistiros III, 36–7, confirming this chronology on the basis of three separate road surfaces in stone and the material between them); on trade amphorae at Adjiyska Vodenitsa: Bouzek et al. 2007; Tzochev in Pistiros III, 187–9; Y. Garlan in Pistiros IV, 246 (addenda and corrigenda); Tušlová et al. 2010 (dating imported amphorae at this site between the later fifth and end of the fourth century BC). However, among the samples analysed by Kuleff et al., there are examples with handles stamped ΜΑΤΡ/ΟΔΩΡΩ and ΑΝΤΙ/ΦΙΛΟΥ that belong to the ‘Parmeniskos’ group, formerly attributed either to Thasos or Mende and the cluster that includes the ΜΑΤΡΟ/ΒΙΟΥ stamps, identified by Stoyanov as products of a Mesambrian workshop from the end of the first quarter and mainly from the second quarter of the third century BC (Kuleff et al. 2007; Stoyanov 2011). The stamps in this series are typologically close to Thasian series X, dated by Tzochev 274–59 BC.

(25) J. Bouzek, Pistiros III, 64 and pl. 18; 286 (434 drachms of Alexander III, Lysimachos and other early Hellenistic Successors; 115 tetradrachms of Philip II, Alexander III, Demetrius Poliorketes, and Seleukos I; 3 gold staters of Alexander III); on the Celtic presence in the east Balkans and western Asia Minor: Mitchell 1993, 13–20; Chaniotis 2005, 220–1, 228, 230, 235–40; see also the contributions to Vagalinski 2010.

(26) Antiochos II and the siege of Byzantion: Memnon FGrH 434 F15; Polyaen. Strat. 4.16; Will, Histoire Politique 1, 247–8; Draganov 1993 on the coins minted at Kabyle; Avram 2003, 1184–9, 1201–2; civic inter-relations: 1193–1200; he also identifies IGBR I2 388 as a decree of Mesambria, not Apollonia, though it honours an Apollonian benefactor; ‘Monopoly War’: Memnon FGrH 434 F13, with Avram 2003, 1187–8, 1211–12, and Gabrielsen 2011, 223–6, emphasizing the city’s specialized control of fish stocks.

(27) Lazarov 2010, 99–105 with further bibl.

(28) Karayotov 1994, a key survey of numismatic evidence from the environs of Mesambria, beginning with a hoard of Cyzicene electrum and a Persian gold stater among the earliest coins, continuing with the Kocharitza hoard of 101 Macedonian bronze issues (89 of Philip II, 10 of Alexander III, 2 Mesambrian helmet en face/wheel with 4 rays, META + head of Melsas and/or shield); bronze of Philip II and Alexander III from Bilka, Rouen and Goritza; Sredna Mahala: a hoard of Apollonia tetrobols and others from Rouen, Bilka, and elsewhere; third-century BC hoards of Athena Alkis types of Mesambria; Aitos: second-century hoard of third-century BC Mesambrian tetradrachms; second century: Sadievo (42 tetradrachms of Byzantion [Lysimachos] + 3 Thasian ‘new style’ as well as one of Kavaros); Aitos, Rudnik, and Burgas hoards of Mesambrian tetradrachms and those of Odessos and Byzantion. He notes the rarity of Adaios issues and Odessitan bronze. From the end of the second century BC, there is a hoard of Byzantine and Thasian coins from Debelt (286). Karayotov 2002, 431–7 on second- and first-century BC regal issues of Mostis and the Sapaian dynasty.

(29) Jourukova 1994, 261–5, on Agathopolis and related issues.

(30) See esp. Emilov 2010.

(31) Avram 1999; Minchev 2011, 21–2.

(32) See in particular the discussion below about Pistiros in Thrace.

(33) RO 12; Hatzopoulos 1996, II, no.1 (alliance between Amyntas III and the koinon of the Chalkidians, early fourth century BC); Borza 1987, 32–52; Millett 2010, 472–3.

(34) Hdt. 5.17 (Alexander I of Macedon received one silver talent every day from the Dysoron mines); located in the Philippoi Plain: Faraguna 1998, 375–8 (identifying Dysoron with Mount Menoikion, on the basis of SEG 34. 664B, in line with the topographical interpretation of L. Missitzis; Hatzopoulos 2008a, 14–35, for full bibliography and detailed discussion of the topography of the lower Strymon; Kremydi 2011, 160–1; see also Ch. 6 for other interpretations of this evidence.

(35) Krenides (and Daton)—Isokr. 8.24; Theop. FGrH 115 F43; Ps.-Skyl. 67; Str. 7. 331 frs 34, 41, 43—attracted the attention of Philip II of Macedon, who seized Krenides, which was subsequently renamed Philippoi, and became the most significant centre associated with the mining area (Str. 7.331 fr. 34). He famously drew 1,000 talents (presumably per annum) from these mines (Diod. 16.8.6).

(36) Thucydides’ mining rights: 4.105.1; Plut. Cim. 4.1; Mor. 205c; Markell. Vita Thuk. 14, 40; Miltiades the Younger and Hegesipyle: Hdt. 6.39.2; PAA 653820; Miltiades the Elder: PAA 653685; Archibald 1998, 113–14, 117; see also Ch. 6.

(37) See further Ch. 8.

(38) See Archibald 1998, 22–3 and fig. ix; Tonkova 2000, 138 map fig. 2, for gold and silver reserves in Thrace.

(39) See Ch. 8.

(40) e.g. Hatzopoulos 2002, 272 n.43, citing Mihailov 1989, 55: ‘the principle of royal power in Thrace brought it closer to the political régime in Persia and opposed it to the Greek city-state (polis), but it should be borne in mind that this country was located between two different worlds, and served as a bridge between them in many aspects of life’ (although this strikes a different note from Mihailov’s earlier thoughts on urban development, esp. Mihailov 1986). Cf. Hatzopoulos’ own remarks, 2002, 266–70.

(41) Reger 2011, 374; cf. also Reger’s remarks on the legal application of the notion of ‘homelands’ to New England states.

(42) ‘Sub-citizens’: Ellis 1976, 27; serfs: Billows 1995, 9–10, 136–7; cf. Millett 2010, 478, for the citation and further refs n.21.

(43) Fol 1965; 1972, 73–5; 1975, 77–83; Tacheva 1997, 96–149, with discussion.

(44) Xen. Anab. 7.1.5, 2.10, 2.23–25, 7.15–16: Medosades; 7.2.32, 4.21, 5.1: other high-ranking Odrysians; Seuthes’ promises of land, including the town of Bisanthe, to Xenophon: 7.2.38, 5.8, cf. on Bisanthe, Thuc. 2.61; Hdt. 7.137; Nep. Alcib. 7.4; Plut. Alc. 36.3; Stronk 1995, 140 and fig. 9.

(45) Tacheva 1997, 96–144, esp. 97; at 101 she discusses Xen. Anab. 7.7.29–32 as potential evidence of early class structure.

(46) Tacheva 1997, 99–101.

(47) Archibald 1998, 226; [Ar.] Oec. II.26 1351a 26 refers to an incident involving the Athenian general Iphikrates, who was short of pay for his mercenaries and set them to sow three medimnoi of corn seed; another similar anecdote is recorded by Polyaenus (7.32), in which a king Seuthes provided five medimnoi of seed per farmer, within a group of farmers, with a view to selling the proceeds ‘on the coast’ for a large sum of money. The very similarity between these stories belies a significant gulf in economic mechanisms between Iphikrates’ men and Seuthes’ farmers.

(48) Hatzopoulos 1988b; 1991; 2011a.

(49) See esp. Hatzopoulos 2011a and his discussion of property in Mieza.

(50) Hatzopoulos has been the leading exponent of a less rigid and doctrinaire view of Macedonian society, but his views are founded on a growing body of epigraphic evidence; see Hatzopoulos 1988a; 1988b; 1991; and esp. 1997; 1999b; Faraguna (1998; 2006) and Mari (1999; 2002; 2011a) accept his interpretation; King 2010 for a more sceptical position regarding royal power (though the evidence of land sales is not in question).

(51) Dimitrov 1958, 683–99; Dimitrov and Chichikova 1978, 10–11, 48–64; K. Dimitrov 2011, 101–12, with recent bibliography and detailed re-investigation of many of these arguments, even if his conclusion is that Seuthopolis, whilst having many of the essential features of an early Hellenistic city, did not have ‘citizens’.

(52) Stoyanov 2006, 85; K. Dimitrov 2011, 101.

(53) Tacheva 2007, 593.

(54) See eg. Porozhanov 2009.

(55) Inventory, no. 645, Kypsela; pre-Hellenistic sites not attested as poleis: Inventory, 871, Doriskos, called a teichos basileion and a Perseōn phrourē by Herodotus at 7.59.1; 913, Apros, Ergiske, Ganiai, Ganos.

(56) Just. 9.3.1–3; G. Mihailov, BullÉp 1988, 259; 528); Archibald 1998, 121 fig. 4.4 (distribution of named sites), 237–9 on the incident with the Triballians, 265–8 for discussion of the hoard; Loukopoulou 2008, 139–69, esp. 158–63. The association with Triballian booty was first proposed by M. Hatzopoulos in 1987 and Loukopoulou has developed this thesis.

(57) Hatzopoulos 2002, 267.

(58) Inventory, no. 656.

(59) Archibald 1998, 121 fig. 4.4; Inventory, 913, listed as ‘Pre-Hellenistic Settlements not Attested as Poleis’; Loukopoulou 2008, 148–53, drawing attention to the patrios phoros claimed by the Odrysian kings.

(60) ‘Les emporoi, devenus par domiciliation emporitai, apportent argent monnayé, vins dont les Thraces riches usent parfois sans discrétion, produits de luxe; ils emportent céréales, du bétail, et aussi des esclaves. Les emporia sont un facteur essential de la prospérité économique du royaume. Aussi le pouvoir odryse promet-il franchises et stabilité’ (Salviat 1999, 273).

(61) Hatzopoulos 2002; Zournatzi 2000; for different nuances on Thracian and Persian tribute, Stronach and Zournatzi 2002.

(62) Salviat 1999, 267–71; on the Persian presence in the north Aegean, Archibald 1998, 79–90; Tuplin 2003.

(63) Karttunen 2002, esp. 457–66, 472–4.

(64) See Ch. 8.

(65) Rollinger 2006, 212.

(66) According to Henkelman and Stolper (2009, 316–22), Skudrians cluster in a number of areas around Persepolis, including 300 kurtaš in the central region, around Fahliyān, north-west of Persepolis, 220 at Rakkan, 46 ‘stone cutters/polishers’ at Tikraš; where c.1,000 are known—flour/barley and/beer received on behalf of various Skudrians, male and female. There is a northern cluster in towns and villages along the road from Media to Karaš/Gaba(e), in the direction of Tamukkan and the Persian Gulf coast. Travelling Skydrians form a separate group, which includes women travelling some considerable distances, under supervision, to Persepolis, with authorization from the satrap of Arachosia-Gandara. Activities associated with the rations include wine producers, grain storers, servants (though sealed orders indicate more specialized functions), stone cutters/polishers, report-makers/supervisors, grain producers, and grooms; but the activities carried out were probably varied.

(67) Henkelman and Stolper 2009, 295: ‘Even if Macedonian Σκύδρα was once a Thracian town, and even if this region was brought under Persian control in 513, it would still seem inexplicable why the Persians picked an ethnonym from the far west to denote the Thracians, many of whom they had encountered earlier during the Scythian campaign and probably before.’ Their solution is to identify Skudrians with a mixed ethnic group of Asiatic Thracians = Bithynians, now better attested around Daskyleion/Ergili. This seems more plausible, especially as Scutari = Üsküdār, (Σκουτάριον) and several other toponyms in Bithynia show further parallels, as well as Uscudama = Edirne (Amm. Marc. 14.11.15, Detschew (1976 [1957], 349).

(68) Rollinger 2006, 205; 213 Table 11.1 (‘Greeks’ in the inscriptions of the Achaemenid kings).

(69) ‘[Thrace] Ce nom désigne dans Strabon, et dans l’antiquité en général, la partie de l’Europe où prédomine une population thrace; c’est en gros l’angle S.-E. de la péninsule balkanique. Ses limites sont au nord le Danube … à l’est l’Euxin et le Bosphore, au S.-E. et au sud la Propontide, l’|Hellespont, l’Égée au S.O. la Macédoine où la frontière fut le Strymon d’abord, puis après Philippe II et Alexandre, le Nestos—étant entendu qu’il resta dans les limites de la Macédoine des enclaves et un fonds de la population thrace.’ (Baladié 1989, 331 with copious text refs.) See e.g. F25 (Bermion, formerly occupied by Bryges, a Thracian people, some of whom crossed to Asia).

(70) See the discussion in Archibald 2010, 326–7 with further refs.

(71) Zournatzi 2000, 689–702.

(72) Loukopoulou 2008, 153–63, suggests that the formula [Κότυος‎] ΕΓΓΕΙΣΤΩΝ should be read ἐκ/ἐγ λειστῶν or ληιστῶν‎ (with comparanda, 158–9); this is possible, though not entirely convincing (see Avram, RÉG122 (2009) = BullÉp 2009, 489–90).

(73) Loukopoulou 2008, 158–63; see also 162–3 for the weight inscription on a cup and oenochoe from Golyamata Kosmatka tumulus, attributed to Seuthes III, which refers to Seuthes in the genitive, as on the Rogozen and other inscribed vessels, but add weights in drachms, including Alexander drachms (cf. BullÉp 2008, 102).

(74) Borovo, Rousse: Archibald 1998, 160–1, 327 on gift-giving; see also Miller 2010; 867, 871–4, on Achaemenid echoes in Macedonian and Thracian silverware.

(75) Archibald 1998, Chs 6, 11, and 12; Kisyov 2005, for the unlooted tombs at Chernozem; cf. Delemen 2006. See also now the huge collection of (unprovenanced) plate in the Bojkov Collection, including more examples of the heterodox styles in circulation, although not all the items in the collection need have come from Thrace and some belong to distinctly different traditions, unknown on the European side of the Bosporus (Marazov 2011).

(76) Zournatzi 2000, 693–8, admits (695) that the common feature is the inscription of ownership; Archibald 1998, 318–35, App. 2; cf. SEG 47.1061, with several alternative, metaphysical interpretations of inscriptions from Duvanli, including ΔΑΔΑΛΕΜΕ δᾶ̂α δάλε με = ‘earth protect me’.

(77) Xen. Anab. 7.3.28 (tr. Stronk); Stronach and Zournatzi 2002; Miller 2010, 870–2; 874–8; Aperghis 1999 on Persian storehouses.

(78) Davies in HellEc I, 25; cf. Millett 2010, 482.

(79) This is the theme of Harding 2007 and Moinier 2007; Carusi 2007 and 2008 presents the Mediterranean perspective more systematically.

(80) Gaydarska and Chapman 2007 for Iron Age sources of salt in the east Balkan peninsula; 148 fig. 1 shows Slunchev Bryag, Obzor, Solnik, Mirovo and Topola as coastal sources; possible inland sources listed include: Slatino, Slatina, Biala Slatina (western Bulgaria); Rozovets, Goran Slatina, Maisko Lyubentsi, Ivanski, Malak Porovets (north-east Bulgaria); Gabrielsen 2011, 224–5 on salted products in the ‘Monopoly War’ of Byzantion; cf. Carusi 2008, 70–9, on Pontic salt; Byzantion as a chief centre for salt products: Str. 7.6.2 and Carusi 2008, 77–9 and nn. 82, 83, 84 [= Plb. 4.38.2–4].

(81) Carusi 2008, 67 (who uses this as evidence to deny the validity of Salviat’s thesis that Pistyros and Pistiros were one and the same, but located in the interior of Thrace); Loukopoulou and Psoma 2008b, 67–71 on Lake Ismaris and the lagoons east of Abdera.

(82) Chiverrell and Archibald 2009 on seasonal flows; see above, Ch. 4.

(83) Vardarski Rid: Mitrevski 2001; a river galley is shown on a Philippopolitan coin of Antoninus Pius: Mushmov no. 5115; the river god Hebros is a common reverse image on the city’s bronze coins of the imperial period: Mushmov nos 340 (Domitian); 5065, 5067, 5116–120; 5125 (Antoninus Pius); 5177 (Hadrian); <www.wildwinds.com/moushmov/philippopolis.html>.

(84) Most recently Pikoulas in Korres 2009, 244–8, on roads for wheeled traffic—parallel rails/rail-like arrangement of slabs, 1.4 m apart; see also, in the same volume, A. Matthaiou, 22–33, epigraphic evidence on roads; on the road system of Attica, G. Steinhauer, 34–73, including a detailed description of roads, photographs from recent excavations, and consideration of construction techniques and historical value; see esp. 44–5, figs. 3.4 and 3.5: road Athens–Acharnai; 50–1 and figs 3.11–3.12, road to Marathon and adjacent buildings; 55, fig. 3.15, ancient road between Spata and Loutsas, running alongside the modern road.

(85) Pikoulas 2007; cf. Pikoulas 2001; Edessa: Chrysostomou 2008; Pella: P. Chrysostomou 1990, 220 and fig. 226.

(86) Pikoulas 2001, 190–2; see further Ch. 6.

(87) Madzharov 2009, 87–124 (Bessapara to Augusta Traiana, via Philippopolis and Karasura); cf. Tonkova 2000, 140–2; for changes in the geomorphology of the River Hebros: Domaradzki in Pistiros I, 32–3 and fig. 1.18; Chiverrell and Archibald 2009.

(88) Madzharov 2009, 256–320 with numerous illustrations, esp. 265–9 for pre-Roman routes; cf. Bouzek in Pistiros et Thasos, 41–4 and fig. 3; Nekhrizov and Mikov 2000, 161–70; the eastern route has been identified with that taken by Alexander the Great in 335 BC.

(89) Avram 1997/98, 41; Chankowski and Domaradzka 1999, for the revised text, and Domaradzka in Pistiros II, 339–42 for further addenda.

(90) For a discussion of the various arguments, see Dossier Pistiros; (Fr. Salviat argues that the findspot of the inscription requires us to locate the Pistyros referred to by Herodotos (7.109) in central Thrace, rather than along the north Aegean coastal strip (BCH 123, 270), and even attributes membership of the Delian League to the same community by identifying it with the ‘Kystirioi’ of the Tribute Lists). B. Bravo and A. Chankowski, on the other hand, affirm that the epigraphically attested Pistiros is a polis, but deny that it has anything to do with the archaeological remains at Adjiyska Vodenitsa (BCH 123, 275–317). The other contributors to the volume, V. Chankowski and L. Domaradzka (247–58), K. Bošnakov (319–29), O. Picard (331–46), L. Domaradzka (347–58), and L. Loukopoulou (359–71), accept the identification of the latter with the epigraphic Pistiros. Salviat was not aware of some more recent investigations of Persian movements in Thrace when he compiled his article (e.g. Archibald 1998, 79–90); Archibald 2002a; Archibald (forthcoming a/; Chankowski 2010.

(91) Bresson 2000, 74–84.

(92) On xenia-type friendships: Gauthier 1972; Herman 1987.

(93) Kisiov et al. 1998, 10–33; Philippopolis: Inventory, no. 655.

(94) Archibald 1998, 126–35, with Table 5.1, figs. 5.1–5.2, 312–13 and fig. 13.1, for a preliminary assessment of numismatic finds; cf. Taneva 2000 (coins from excavations at Adjiyska Vodenitsa); Nekhrizov and Mikov 2000, 161–70, with cognate issues in eastern Rhodope.

(95) Lilimbaki-Akamati and Akamatis 2003, 40–51; Akamatis 2006a, 615–26; Akamatis 2011, 402–3; Akamatis 2012, 49, with full references; 49–59, with a more detailed exposition.

(96) Billot 1992 on tanners in Athens.

(97) Akamatis 2000, 193–213; Akamatis 2012, 57–8, for chronological evidence of the agora’s destruction in the late 90s BC.

(98) Chiverrell and Archibald 2009.

(99) Tušlova et al. 2010, 205–9; cf. also Bouzek et al. 2007; Tzochev 2007 on Pontic amphorae at Adjiyska Vodenitsa; Lawall 2005 on the problems of using statistics based only on amphora stamps; Taneva 2011 summarizes the evidence from the extra-mural ceramic kiln.

(100) See esp. Lazov, Pistiros II, 243–8, on bronze toreutics; Domaradzki in Pistiros II, 249–54, on ‘Celtic’ type fibulae; Bouzek in Pistiros IV, 221–2, for a distinctly ‘northern’ bronze appliqué of a lion; Archibald 1998, 138–41 (Adjiyska Vodenitsa) and more generally, 222–26, 260–81 for further comments about styles and tastes.

(101) Domaradzka 2002; 2005; 2007; Domaradzka and Domaradzki 1999.

(102) Taneva 2000; the range and number of coins have increased appreciably since this publication, with the current total close to 2,000 coins, of which the hoard of c.280–78 BC comprises 552. De Callataÿ’s table of published coins from excavations provides a broad scale of comparison, with Athens at the top (12,676 coins). The combined figure for Olynthos is 4,992 coins and Seuthopolis 1,305. Most published coin catalogues from excavated Greek sites have produced less than 1,000. De Callataÿ discusses the variable rate of survival over time (2006b, 179–91).

(103) Cyzicene staters: Archibald 1998, 126–35, for distribution and further refs; Picard 2000 for Thasian types; Vardarski Rid: Husenovski 2002; for Krastevich, see above n.16.

(104) Van Alfen 2005, 322–38, on Nikophon’s law in Athens (SEG 26.72) and imitations of Attic coins; he distinguishes issues in terms of ‘prototypes’, ‘artistic imitations’, ‘marked’ imitations (those with an obvious distinguishing mark or countermark); perfunctory imitations, plated or debased coins and, finally, counterfeit coins. Most of these categories can be identified in the east Balkans.

(105) See further Ch. 8.

(106) Inventory no. 674, pp. 915–18.

(107) Newskaja 1955; Loukopoulou and Łajtar (2004) provide an up to date point of reference, but Isaac’s chapter on Byzantion (1986, 215–37) is still a useful account and commentary; Robinson 2011, 146–9, reconsiders the city’s constitutional structure with reference to Aristotle’s statements (and their economic dimensions) about democracy at Byzantion.

(108) Str. 7.6.1, p. 319 and fr. 56; Pl. NH 4.46; Pompon. 2.24; Isaac 1986, 232–3 on territorial matters.

(109) Inventory, no. 679.

(110) Diod. 16.76.3–4; 77.2; Justin 9.1.2–4; Dem. 8.14, 18.88–92; 244; Plut. Phoc. 14.3–7; Gabrielsen 2007, 295.

(111) Isaac 1986, 233–4; Nixon and Price 1990, 153–4; Robinson 2011, 146–9.

(112) Theop. FGrH 115 F62 = Athen. 12.32, p. 526 D–F.

(113) Theop. FGrH 115 F292; Philoch. FGrH 392 F162; Just. 9.1.1–6; Front. Strat. 1.4.13; HM II, 576; Bresson 2000b, 132–3, 277–8; Gabrielsen 2007, 295–6; 306–7 Moreno 2008, 668–9; the most specific and nuanced explanation of what occurred is contained in Didymus’ Commentary on Demosthenes, 11.1, cols 10.34–11.5 (= T22 in Moreno 2008, 688–9).

(114) Moreno 2008, for a detailed discussion of the topography and history of the site.

(115) Gabrielsen 2007, for detailed discussion of the sums, esp. 295–7.

(116) Oliver 2007, 68–73, 169–71, 194–7, on Athenian naval activities in the Aegean during the third century BC.

(117) Such as occurred between Antigonos I’s admiral, Nikanor, and Polyperchon’s naval commander, Kleitos (Polyaen. Strat. 4.6.8 = Moreno 2008, 696, T38).

(118) Bresson 2000b

(119) Dumitru (2006Dumitru 2006SIG2IG2ibid

(120) Gabrielsen 2007, 299–311.

(121) Moreno 2008, 646–70.

(122) Dem. 17.19–21 (337 BC); Gabrielsen 2007, 311–13, with discussion.

(123) Will Histoire Politique II 2, 47.

(124) Gabrielsen 2007, 297.

(125) Loukopoulou and Łajtar 2004, 916, collect the evidence; Gabrielsen 2007, 296, with detailed discussion.

(126) Xen. Hell. 1.1.22 (for 410/9 BC); cf. Diod. 13.64.2–3; Polyb. 4.44. 3–4.

(127) There may already have been a number of different technical applications of this term. The Athenian Grain Tax Law of 374/3 BC (Agora I 7557) actually mentions two dekatai (Stroud 1998, 4 ll. 59–61; pp. 27, 31, 81–4). Stroud speculated that the ‘twin’ dekatai referred to twice in Agora I 7557 (ll 59 and 61), may represent the toll going into the Bosporus and the second charge for the return trip (83). But he recognized that the intervening period between the Athenian control of Chrysopolis and the reference to Charidemos of Oreus collecting taxes in 357 BC (dekátas lambánein…toùs dekatēlόgous …axiōn toùs hautou tōn telōn…kuríous eínai [under the control of his own custom house officials]: Dem. 23.177), was one of major disruption in the Straits and Bosporus. See also RO 26, and commentary, 122–9. Rhodes and Osborne follow E. Harris in seeing the ‘two tenths’ as down-payments for that particular year, rather than monetary taxes raised from tolls, although they admit difficulties with the reference to the previous year’s dekatai. Bresson noted the similarities between Agyrrhios’ decree and the law from Samos (Syll. 3 976, ll. 25–6, where the eikostē from Anaia is referred to), which seems the clearest parallel for a grain tax (Bresson, 207–8). A more wide-ranging investigation of the law is now available in Magnetto et al. 2010, esp. 243–44 for the text and 246, 248, for new translations by R. Stroud and U. Fantasia. No additional clarification emerges for ll. 59–61 as far as the ‘twin dekatai’ are concerned. Harris’s interpretation of the down-payment does seem to answer the context of Agora I 7557 convincingly, and shows that the dekate was a concept that was acquiring a number of applications, in addition to the one associated with the Athenian toll station at Bosporus.

(128) Rubel 2001; Gabrielsen 2007, 294, 310–11.

(129) Byzantion’s purchase of Hieron: Plb. 4.50.2–3; Dionysios Byz. Anaplous fr. 58; Byzantine sanctuary of Zeus Ourios, on the east side of the Bosporos, opposite the entrance to the Black Sea: Diod. 20.111.3 fr. 302; Plb. 4.50.3; territorial acquisitions in Mysia: Phylarchus FGrH 81 F8; Plb. 4.50.4, 9; Str. 12.8.11.

(130) Gabrielsen speculates that the Celts of Tylis had a reasonably accurate idea of what they could extract from Byzantion in the form of tribute (they demanded 80T per annum in c. 220 BC: Plb. 4.46.3–4). If the capacity of an average vessel was 3,000 medimnoi of grain, and the price of grain 5 drachmas a medimnos, this would deliver a toll of 15,000 drachmas per ship (Gabrielsen 2007, 295, 316; Bresson, 278 and n.66). For a recent review of the kingdom of Tylis, see the contributions to Vagalinski 2010.

(131) Memnon of Herakleia, FGrH 434 F13 (Photios 228a–b); Avram 2003, 1187–8; Gabrielsen 2011, 223–7.

(132) See also Archibald (forthcoming/c).

(133) Philip V in 200/199 BC: Will, Histoire Politique II, 45–6; Dumitru 2006; in 184 BC: Plb. 22.14.12; Livy 39.35.4; Perseus: App. Mac. 11.1 & 7; Livy 42.13.8; 40,6; 42.4; HM III, 497.

(134) Gabrielsen 2011, 220–46, for an in-depth discussion of ancient and modern monopolies.

(135) Mitchell 2008, 211; Walbank 1985; Walbank 2002a; Lolos 2009, 266.

(136) Walbank 1970, 305 [2002a, 105], 306–7 [2002a, 106].

(137) Thuc. 2.99–101; Archibald 1998, 107–11.

(138) Hatzopoulos 1996, I, 231–60.

(139) Parissaki 2009.

(140) Parissaki 2009, 337–50.

(141) Dawkins and Hasluck 1905, 175; Parissaki 2009, 324–6, 1/4–1/7, dedicated by Apollonios, son of Eptaikenthos, of Bizye, strategos in the environs of Anchialos.