Lara Fabian 9 Multifaceted Highlands: The Economy of Armenia and its South Caucasus Context I Introduction Starting after the collapse of the Achaemenid Empire, the rugged upland territory of the Lesser Caucasus stretching down into the Zagros mountains developed into the heartland of Armenia (map 1). It was the most geopolitically important of the borderland states that arose at the interstices of the Mediterranean and Iranian spheres in this period, eventually coming to form part of the so-called Parthian Commonwealth, and also briefly becoming a province of the Roman Empire.1 Alongside its neighbors to the north and northeast, K‘art‘li and Caucasian Albania respectively, it came to dominate the South Caucasus, the territory between the Black and Caspian Seas, and acted as an important fulcrum connecting both the Mediterranean and Iranian worlds to that of the steppe to the north. Its place in broader economic systems, however, remains elusive. The difficulty lies partly in the evidence itself, as sources are a challenge in this space. Nevertheless, although often patchy, both texts and material culture speak clearly for wide-ranging interactions that were either explicitly economic or had economic ramifications (e.g., on consumption practices). Rather, the difficulty is largely methodological. On the one hand, the multifaceted nature of the region – sitting as it did along key political, sociocultural, and ecological frontiers – makes it difficult to categorize these interactions, such that the connection between local practices and wider economic systems is often obscured. Moreover, this same character allows Armenia to be interpreted in wildly different ways, depending on the perspective of the researcher.2 This creates a range of images of the territory – on the question of Silk Roads in Armenia, for example – that are not just contradictory, but indeed mutually exclusive. I. Armenia, Frontiers, and ‘Silk Roads’ One of the characteristics of Armenia, at least as it comes to us presented through the lens of classical historiography and modern historical scholarship, is as an entrenched political frontier between the Roman and Iranian worlds. Indeed and more explicitly, the descriptor of Armenia as a ‘buffer state’ echoes through historical scholarship, 1 On the term “Parthian Commonwealth,” see de Jong 2013. 2 Traina 2021a for reflections on the methodological difficulties involved in the study of Armenia, as well as a call to focus on connectivities as the path forward. Open Access. © 2023 the author(s), published by De Gruyter. This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License. https://doi.org/10.1515/9783110607628-011 442 Lara Fabian appearing in conceptual pieces such as Luttwak’s Grand strategy of the Roman Empire and also in focused historical studies such as Sherwin-White’s Roman foreign policy in the East.3 The buffer-state model is inappropriate, as it captures neither Armenia’s specific political relationships with its neighbors nor the complexity of its internal organization.4 But more problematic is what the idea connotes about the place of Armenia in broader systems, namely that the territory was a “neutral zone”5 that lay somehow both at the fringe of but also beyond imperial political space. Another dominant projection of the space, however, this one found more often in the work of specialists on Armenia, places it not as an imperial fringe, but rather as a central zone of trade and connectivity. These discussions are common in the Armenian and Russian literature, but make their way into English as well, as in the case of the groundbreaking work of Manandyan from the middle of the twentieth century.6 They are generally anchored in the reconstruction of trade activities and routes based on analyses of texts from the medieval period. Drawing on itineraries that mention urban centers in Armenia and speculate about Armenia’s role in the silk trade, scholars developed a vision of the routes in which “a network of cities [were] strung like beads on a string of infrastructure, tautened at either end in accordance with the desires of distant urbanites in Venice and Guangzhou.”7 Such a picture is, to be sure, a stark oversimplification of the complex nature of medieval trade networks in Armenia, which were both less urban and more integrated with local commercial markets than the prevailing picture allows.8 Nevertheless, the image gleaned from the medieval sources has had currency beyond studies of medieval Armenia. It has also been retrojected into deeper antiquity, where it provides a framework for understanding the economic importance of Armenia as early as the Hellenistic period. Thus, Garsoïan writes, “the considerable number of Greek coins dating from [the Hellenistic period] found on the territory of the Armenian Republic testify to the prosperity brought to this region by its transit trade.”9 Among scholars specializing in the region, then, there is a tenacious sense that the Armenians must, somehow, have played an important role in trade – even if the precise nature of that role remains hazy. This impression can be traced, in part, to Strabo’s descriptions of the South Caucasus, which included the statement that “[the Aorsi in the North Caucasus] were thus enabled to transport on camels the merchan- 3 Luttwak 1976, 24; Sherwin-White 1984, 337. See a fuller perspective on Luttwak in the broader context of Mesopotamia in Gregoratti, ch. 10, I, this volume. 4 On which, see Traina 2021a. 5 Luttwak 1976, 24. Also note the inaccurate geography of Armenia and the rest of the South Caucasus in Luttwak’s map 1.2. 6 See the classic work of Manandyan 1954, with its English translation, Manandyan 1965. 7 Franklin 2021, 45–46. 8 See the discussion in Franklin, particularly her case study of local-scale processes in the Kasakh Valley (2021, ch. 4). 9 Garsoïan 1997, 49. Multifaceted Highlands: The Economy of Armenia and its South Caucasus Context 443 dise of India and Babylonia, receiving it from Armenians and Medes.”10 In the picture painted by Strabo, the Armenians are positioned quite literally as the middlemen who facilitated the movement of goods from the Near East and South Asia into the Pontic Basin and continental Europe. According to Strabo, the Armenians were tremendously wealthy, likely based on this trade as well as access to natural resources including gold mines.11 But, outside of specialist circles, there has been relatively little attention paid to Armenia’s role in trading activity in antiquity, with the region’s political role in Roman-Parthian jockeying receiving far more detailed attention. Of course, however compelling one finds the idea of a centrally networked Armenia, one must also concede that, according to most reconstructions of ‘Silk Roads’ in antiquity that have been proposed since the time of von Richthofen’s early investigations, the routes did not, in fact, pass through Armenia or the South Caucasus.12 The main routes west of Central Asia are said to have run south of the territory, along the so-called Great Khorasan Road toward Ekbatana and then into Mesopotamia. The ‘northern Silk Route’ posited in some reconstructions, running from Central Asia and terminating along the coast in the North Pontic, involving steppe pastoralists, is generally conjectured to have passed to the north of the South Caucasus.13 In this way, the South Caucasus can be understood, from the perspective of western historiography, as a characteristic example of a space that has been excluded from meaningful participation in the Silk Road discourse.14 I. A Roadmap for Understanding Armenia’s Position in Global Systems What, then, can the case of Armenia add to our contemporary understanding of a connected Eurasian world zone that we try to build, beyond the Silk Roads? This chapter takes as its project to answer that question through an exploration of how these two facets of Armenia – its frontier-ness and its connective potential – were in fact deeply intertwined. In keeping with this volume’s theoretical approach to frontier zones as spaces of innovation and opportunity, but also risk, I argue that a unique flexibility emerged in Armenia, at least in part because of the overlapping frontiers. 10 Strabo 11. 5. 8, trans. H. L. Jones. 11 See Strabo 11. 14. 10: “Of the riches and power of this country, this is no slight proof, that when Pompey imposed upon Tigranes, the father of Artavasdes, the payment of 6,000 talents of silver, he immediately distributed the money among the Roman army, to each soldier 50 drachmae, 1,000 to a centurion, and a talent to a Hipparch and a Chiliarch,” trans. H. L. Jones. 12 There are of course problems with envisioning a fixed system of such roads, see Weaverdyck, vol. 1, ch. 7, 271–274; Weaverdyck and Fabian, vol. 2, ch. 8.A, 365–366. 13 E.g., the set of connections suggested Yao 2012, fig. 1. 14 In this sense, it is in some ways similar to the wider Eurasian Steppe as discussed in von Reden, ch. 1, this volume. 444 Lara Fabian The contours of this flexibility emerged out of the preexisting local conditions in the space and the networks that developed over time here. With their help, political authorities in Armenia were able to keep their tenuous hold on sovereignty despite tremendous imperial pressure. In what follows, I build an argument about Armenia within global circuits that considers (economic) connectivity inside of Armenia as well as between the area and its neighbors. Throughout, I loosely use the categories of economic actors and tools developed in volume 2 to disaggregate this behavior.15 The examination begins (sec. II) with a brief exposition of the historical development of Armenia as a polity that was deeply entangled with both its Iranian and Mediterranean neighbors, which sets the stage for the patterns of socioeconomic interaction that are reflected in the space’s trajectory. To that end, section III considers the most explicit evidence for the region’s economic activity, with a particular focus on coinage and the phenomenon of local minting. This exploration demonstrates the multiple roles that coinage could have in both constructing and supporting connectivity, economically and ideologically. Then, in section IV, I turn to local sociopolitical frameworks that both participated in and capitalized on this connectivity, ranging from dynastic elites and the inner ‘court’ to the phenomenon of urbanism. Next, in section V, we turn our attention to the consumptive and particularly the productive capacity of Armenia itself, which reorients our view on this space to see it not simply as a place through which goods moved, but rather as a hub in its own right. Section VI, then, considers evidence for merchants and trade routes in Armenia, adding a spatial dynamic to our discussion of the various networks of connectivity at play here. And, finally, the chapter ends with a consideration of Armenia’s relationships to its northern neighbors – an oft-overlooked component of the regional system. The result is a treatment of Armenia that stresses its unique position within global networks. This attention to positionality situates the region at the nexus of a number of sociocultural and political spheres, which provided the ‘raw materials’ for constructing diplomatic, social, and trade relations (to name just a few) that could be exploited as local actors here reached beyond their borders. And ‘reach beyond their borders’ they most certainly did: Armenia’s growing involvement in wide-scale international affairs over the course of the Hellenistic and Roman-Arsakid periods testifies to Armenia’s participation in the marked upswing of connectivity that we see across Eurasia in this period. II Imperial Neighbors and Armenia’s Emergence Since our focus in this chapter lies not with the political dimensions of Armenia’s connections, but rather with its economic ones, the following introduction to regional 15 See Fabian vol. 2, 61–62; Weaverdyck, vol. 2, 339–340. Multifaceted Highlands: The Economy of Armenia and its South Caucasus Context 445 history provides only a cursory treatment of issues that are often both complex and poorly understood. The goal is to provide familiarity with the general historical progression and to set the backdrop for the internal developments discussed in this chapter, as this political history set important conditions for Armenia’s development, including economically. At the same time, these political dimensions are also critical for understanding the network positionality of Armenia, as they served as a key factor shaping other dimensions of connectivity. The most natural starting point for a narrative of Armenia is the rise of the Achaemenid Empire, which spread its power widely across the Zagros, deep into Anatolia, and all the way up to the Greater Caucasus Mountains.16 This period saw the establishment of long-running patterns of interaction that would shape life in the highlands for centuries to come. After the collapse of the Achaemenid Empire, rule in Armenia can be divided into a series of dynastic periods, as well as several intermediate periods of less solid control. Although the following description, as the chapter more generally, will use the standard dynastic divisions (e.g., Yervandid, Artašēsid, Aršakuni), these descriptions are not without problem.17 Chiefly, they tend to reflect retrospective perspectives on the development of rule. It is not clear to what degree these were meaningful divisions for residents in the highlands, nor do they reflect the contentious, and often fractious, nature of local authority. II. Sources for the History of Armenia Before we get to that historical narrative, we need to consider the evidentiary basis that this exploration rests on. In reconstructing the history of the South Caucasus, beyond the material and documentary evidence (including built architecture, numismatics, ceramics, inscriptions, etc.), we have two primary bodies of literary evidence in the form of transmitted texts. The first come from the Graeco-Roman tradition, and – outside of a small number of ethnographic or natural historical descriptions18 – tend to detail the place of Armenia in the context of the long-running armed conflicts that unfolded on the Hellenistic and then Roman-Arsakid borders.19 These present the Mediterranean perspective on this region, and provide an episodic view of the space, leaving us largely in the dark about local events in periods when they did not have bearing on external, primarily Roman, political or military interests. 16 Starting the story with the Urartians is not, however, uncommon; see, e.g., Payaslian 2007, 5. 17 See Khatchadourian 2007, n. 1 on this point. 18 Strabo and Pliny the Elder are both valuable sources for Armenia, see, e.g., Traina 2017. For a comprehensive treatment of Armenian history with detailed discussion of the sources, see Chaumont 1985; 1976. 19 See the contributions in Gazzano, Pagani, and Traina 2016 for an overview of part of this textual tradition. 446 Lara Fabian The second comes from the local chronicle traditions of Armenia and Georgia, particularly the text attributed to Movsēs Xorenac‘i’s, History of the Armenians (MX, Patmut‘iwn Hayoc‘), but a wide range of other texts as well.20 It is impossible, of course, to use the medieval accounts preserved in MX and other chronicle texts as straightforward historical testimony.21 Composed far after the events described within and subject to significant layers of editing and rewriting in subsequent periods, they offer only shadowy hints at historical events. Particularly problematically, some of the key passages from MX are riddled with chronological inconsistencies, with some events from the late Hellenistic period projected into the second century .22 Nevertheless, the accounts offer important clues about how local medieval societies understood their own pasts and contain kernels of history that survived their complex transmission chains.23 A key difficulty lies in trying to integrate the historical knowledge that can be gleaned from this source tradition with the far more positivistic, but also distant and biased, classical texts. The attempt, however, is worthwhile, particularly when combined with archaeological evidence to give more dimension. II. The Achaemenid Period in the South Caucasus The rise and spread of Achaemenid power that brought the Armenian highlands into the imperial system occurred in the context of extensive reorganization across the territory in the wake of the collapse of the Early Iron Age Urartian Empire centered around Lake Van in the sixth century .24 Politically, not just Armenia but also much of the South Caucasus then came to be incorporated to one degree or another into the satrapal system of the Achaemenid Empire.25 The date of the incorporation of the region into the Achaemenid sphere is not fully clear. There is reason to believe that a victory over Urartu can be dated to 547 ,26 while Darius I’s campaign against 20 Thomson 2014 for an overview. 21 For example, see the discussion of the use of a roughly parallel text from the Georgian tradition in Rapp 2014, 1–30 and the brief discussion in Schottky 2012, 242; Traina 2019, 23–24. 22 E.g., Movsē s Xorenac‘i (MX) 2.37 and following, on which, Toumanoff 1963, 283–284. 23 See also Traina 2021a for a recent call to reinvigorate interaction with these sources. For an example of this in the context of K‘art‘li, see the study of Schleicher 2021, especially ch. 1. 24 On the structure of Urartian power, see, e.g., Zimansky 1985. 25 Although the internal and external borders of Achaemenid power are fuzzy, as our reconstructions rest on an eclectic range of sources, chiefly Achaemenid Royal Inscriptions; Herodotus’ list of tribes; and later tribe lists written by the Alexandrian historians. The dahyu (satrapy) of Armenia was of central importance, including much of the Armenian plateau as well as the middle Kura (Khatchadourian 2016, 124), with the lower Kura perhaps forming the northern edge of the neighboring dahyu of Media, though there are debates about this: compare the positions discussed in Jacobs 2000. 26 See discussion in Rollinger and Kellner 2019. Multifaceted Highlands: The Economy of Armenia and its South Caucasus Context 447 the Scythians in 512  would have brought even more Achaemenid presence to the region.27 The Achaemenid period brought not only new political structures to the space, but also a range of new material practices. In the areas to the north of Armenia, the markers of Achaemenid spread are particularly visible in the form of massive administrative buildings that find their closest parallels in the Achaemenid heartland.28 In Armenia, where preexisting models of political authority based on the Urartian Empire perhaps had longer echoes, the quantity of explicitly satrapal architecture is more limited, but the traces of new material practices are nonetheless visible, particularly in the realm of luxury and elite goods.29 It seems likely, nevertheless, that both the material consequences and the later memory of Achaemenid rule in this territory had significant ramifications for later development. II. Post-Achaemenid Transformations The two-and-a-half centuries following the collapse of Achaemenid control ca. 330  are a murky period for regional history in the South Caucasus. However, in broad strokes, the collapse sparked a new phase of regional political-authority construction, which would eventually lead to the rise of a number of polities, including most importantly Armenia, centered on the highlands of what is today the Republic of Armenia and southeast Turkey, along with K‘art‘li (known in the classical corpus as Caucasian Iberia) and Caucasian Albania to the north. The story of local political consolidation is often framed as an immediate post-Achaemenid process, although the tempo of the developments across the region in fact seem to be rather varied, such that it would be incorrect to see all political consolidation as a direct product of the Achaemenid period.30 Meanwhile, to the south of Armenia, in the Zagros belt and its foothills in Northern Mesopotamia, a similar process was unfolding, although in a context of more direct interaction with Alexander the Great and his successors. Of central importance to Armenia were the polities of Sophene and Kommagene, whose dynasts were often closely related to or direct family members of those in Armenia. II.. Yervandid Armenia As noted, the end of the Achaemenid period brought, at first, few changes to internal Armenian power structures: The first dynastic house to rule Armenia in the Hellenis27 For this interpretation of the campaign, see Jacobs 2000. 28 Babaev, Gagoshidze, and Knauß 2007; Knauß 2000; Knauß, Gagoshidze, and Babaev 2013. 29 Khatchadourian 2016. 30 For treatments that are cautious about the process of development, see on K‘art‘i, Meißner 2000; and on Albania, Bais 2001; Fabian 2020. 448 Lara Fabian tic period was, in fact, a continuation of the line that had ruled Armenia as a satrapy in the late Achaemenid period, known as the Yervandid (Orontid, Eruandid) dynasty. The Yervandid dynasty took power in Armenia in the wake of Alexander’s victory, led by Mithranes, the pro-Alexander son of the former Achaemenid satrap in Armenia, who was given Armenia in recognition of his support of Alexander.31 A dynastic line for this fairly shadowy family can be traced over the course of the next centuries, but with many points of raw conjecture.32 Broadly speaking, the local Yervandid rulers appear to have been semiautonomous, but politically tied to the later Seleukid kingdom. However, the formal status of these Armenian dynasts and the nature of their arrangements with their kinsmen in Sophene and Kommagene and the wider Seleukid world is not entirely clear.33 II.. Artašēsid Armenia In any event, we know that Armenia was granted independence after the Treaty of Apamea between the Roman and Seleukid Empires in 188 , and that rule of Armenia was given to a new king, Artašēs.34 Although Artašēs called himself an Yervandid in inscriptions that will be discussed shortly,35 following both the classical and Armenian textual traditions, he is seen to have been the initiator of a new dynasty, the Artašēsid (Artashesid) line, which would rule Armenia until 12 . The traditional narrative depicts Artašēs as an effective consolidator of Armenian territory, such that under his rule, the territory extended all the way to the Caspian Sea.36 However, in reality the first century or so of Artašēsid rule is poorly reflected in the classical textual tradition and confusingly telescoped in the Armenian one, while archaeological sources are also sparse, limiting our ability to discuss the developments in detail. We find firmer footing at the beginning of the first century , when Tigranes II, known as Tigranes the Great (r. 95–56 ), came to power. Tigranes, who had been raised at the Arsakid court as a hostage, came to the throne in a period of regional instability, as infighting distracted the Parthian court. Benefiting from the instability, Tigranes II first turned his expansionist eye to neighboring Sophene, which he annexed within the first year of his rule. Through marriage diplomacy, he forged an alliance with and secured peaceful relations with Mithridates VI of Pontos (120–63 ), who was consolidating his power around the Black Sea coast. Then, in the wake of the death of the Arsakid king Mithridates II in 91 , Tigranes moved against the Arsakid 31 Lang 1983. 32 Toumanoff 1963, 277 ff. 33 See, e.g., Traina 2021b. 34 Strabo 11. 14. 15: “[Artaxias and Zariadris] joined the Romans and were ranked as autonomous, with the title of king,” trans. H. L. Jones. 35 Sec. III.1.1 below. 36 Manandyan 1965, 44. Multifaceted Highlands: The Economy of Armenia and its South Caucasus Context 449 Empire itself, rapidly conquering territories from the gates of Ekbatana across the Zagros highlands. Having become a regional power broker of considerable scale, he was offered control of Syria in 83  by local authorities, who appear to have seen him as a more stable option than the rapidly fracturing Seleukid Empire.37 The sources disagree about how deep into the Levant his rule extended,38 but particularly consequential was the capture of Antiocheia on the Orontes, one of the undisputed capitals of the Hellenistic world. Tigranes’ eventual rapid downfall was rooted in the story of Mithridates VI, whose aggressive expansion attracted Roman attention. Although Tigranes II appears to have tried to stay out of the fighting,39 he eventually decided to harbor the fugitive king Mithridates at his capital Tigranocerta in 71 .40 The Romans moved decisively against his royal seat and won a quick victory that ended with Tigranes ejected from Tigranocerta and forced to retreat further into the highlands. By the mid-60s , Tigranes found himself facing not only an emboldened Rome but also a newly aggressive Arsakid Empire. The final blow came in 66  when, having been defeated by Pompey the Great, he was stripped of all of his territorial expansions, although he ruled Armenia itself as a “friend and ally of the Romans” for another decade.41 The following half-century was filled with colorful episodes of Armenian triangulation between the Roman and Arsakid Empires. The Armenians, for example, played a role in the disastrous Roman defeat at Carrhae.42 Later, Armenian hesitancy in supporting military efforts under Marc Antony confirmed the sense that the Armenians were not trustworthy allies, with Artavadzes and his family eventually taken hostage by the Romans and the king executed.43 II.. The Rise of the Aršakuni This was the background that greeted the Roman emperor Augustus. Throughout his rule and into the Julio-Claudian period, the Romans sought repeatedly, and with little 37 Justinus Epitome of Pompeius Trogus (Just. Epit.) 40. 1. 1–3. 38 Josephus Antiquitates Judaicae (Joseph. AJ) 13. 16. 41; Bellum Judaicum (BJ) 1. 5. 3 to Appian Syriake (App. Syr.) 11. 8. 48. 39 Following the account of Manandyan 1963. 40 E.g., Appian Mithridateios (App. Mith.) 12. 82, 84; Plutarch Life of Lucullus (Plut. Luc.) 19. 2; 21. 1–2, 7–9. For a reevaluation of this episode and particularly the subsequent battle, Olbrycht 2021. 41 Cassius Dio (Cass. Dio) 36. 53. 6. 42 The Armenian king, Artavadzes, had promised to send troops to support the Romans, but following a Parthian incursion into Armenia, Artavadzes was unable or unwilling to send his cavalry. Instead, he entered into a marriage alliance with the Parthian king – a marriage that the two kings were celebrating in the Armenian capital when news of the Roman defeat at Carrhae, along with Crassus’s severed head, appeared as a stage prop in a performance of Euripides’ Bacchae (Plut. Crass. 33). 43 Cass. Dio 49. 39. 3–5. 450 Lara Fabian long-term success, to install Armenian kings who were favorable to them, often drawing their candidates from local ruling families.44 Despite some successes, the first half of the first century  in Armenia was marked by a wildly unstable sequence of kings installed by either the Parthians or the Romans, most of whom ruled for only a handful of years.45 The first century also saw the start of periodic wide-scale raids perpetrated by North Caucasian populations, generally referred to as the Alans in the texts, who, supported by local allies, carried out attacks on both Roman and Arsakid territories, as well as occasionally the Armenians.46 Thus, would-be Armenian rulers faced a complicated set of pressures that emanated both from the major Mediterranean and Iranian empires and from their own immediate neighborhood. Out of this confusing period arose the final dynastic family of the pre-Christian period: the Aršakuni (Armenian Arsakid). The seminal moment of this new phase in terms of international power games came in 66 , when Nero crowned the Parthian prince Tiridates II as the king of Armenia.47 This came after a difficult period for Roman interests in the East 48 and resulted in a unique political situation for Armenia, which has been termed a cosuzerainty of Rome and Parthia.49 Under this agreement, Arsakid kings had the right to name the Armenian king, who then had to be crowned by the Roman emperor. The result was an Armenia whose connections to the Parthian power structure were no longer mere hints, but had become explicit. The system was also not entirely successful in maintaining regional stability, evidenced by Rome’s short-lived creation of the province of Armenia (114–118 ).50 For these few short years under the emperor Trajan, Armenia became a province of the Roman Empire. Despite the generally hands-off Roman approach to the space, this was neither the first nor the last time that a Roman general or emperor set his sights on gaining direct control over Armenia. In the late first century , it was Antony who appears to have begun down to path toward annexing the territory, although he never carried out the plan in full, and his intentions have been debated.51 Then, in the third century , Caracalla also made moves toward an annexation of Armenia, although again it was not actually brought to fruition.52 In contrast, Trajan’s successful, if short-lived, annexation of Armenia, which is attested in both literary and epigraphic sources,53 did actually briefly bring the space 44 Ariobarzanes, a prince from Media Atropatene (r. 2 –6 ) is one such example. 45 Chaumont 1976. 46 On the Alans and their ‘incursions,’ see Bosworth 1977; Halfmann 1986; Tuallagov 2014. 47 On the symbolism of this, see Clark 2021. 48 E.g., the defeat of Roman troops at Rhandeia. 49 Ziegler 1964, 69–70. 50 Cass. Dio 68. 19. 2–5, 20. 1–3. 51 See Patterson 2015 for an overview of the debate as well as an argument for Antony’s intentionality in this matter. 52 Patterson 2013. 53 For the epigraphic evidence, see Speidel 2021. Multifaceted Highlands: The Economy of Armenia and its South Caucasus Context 451 directly into the Roman system, if only for a moment.54 After the failure of the province, we find few concrete details about local rulers for most of the rest of the second century , except for on the rare occasions when Armenian affairs become central to Roman imperial affairs, for example in the late second century , when Roman military action against the Arsakid Empire once again brought troops to the region under the reign of Commodus.55 II. The Ambiguity of Armenia: Enduring Political Frontiers As Traina has recently reminded us, the people of Armenia were discussed in the classical corpus as an ambigua gens,56 an ‘ambiguous people,’ a term which he argues we should understand as reflecting most specifically Armenia’s ambiguous allegiance with respect to either Rome or Iran.57 This ambigua gens carved out a space for themselves along a fractious political frontier, and rather than following the course of so many other frontier polities – eventual absorption into the more powerful neighbor – they managed to maintain a tenuous hold on autonomy. It is truly striking that, throughout this period stretching from the rise of Armenia in the Hellenistic period through the Roman-Arsakid proxy conflict in later centuries, Armenia managed to remain functionally independent with a few brief exceptions, despite the intense interest of both Roman and Iranian authorities in this space. This political ambiguity had a range of consequences for local life, which will be more fully explored in the rest of this chapter. But, put briefly, since the space was never comfortably, durably, and unambiguously integrated into a neighboring system, regional practices can in some cases differ considerably from norms in similar spaces that experienced tighter integration. At the same time, the fact that the Armenians had not just one imperial power to alternately collaborate with or push against, but indeed two, created a high degree of internal diversity, as different segments of society interacted with the ‘Iranian’ and ‘Mediterranean’ spheres. This, furthermore, rewarded (and therefore privileged) those who were in a position to develop flexible networks that were both robust enough and diverse enough to react to changing local circumstances. It is precisely these sorts of networks that are likely to be sites of innovation, given their diversity and the need for the constant evolution of new strategies for defending, maintaining, or improving one’s position. In this way, Armenia offers a classic example for the broad ramifications of frontier entanglements.58 54 55 56 57 58 See Section VI.4 for some of the consequences of Roman military presence in the region. Cass. Dio 71. 14; Corp. Inscript. Lat., III, 6052. Tacitus Annales (Tac. Ann.) 2. 56. 1. Traina 2021a, 15. As discussed by von Reden, ch. 1, and Hoo, ch. 2, this volume. 452 Lara Fabian III The Local Economic Framework Now, having situated the historical development of Armenia within both temporal and spatial contexts, we turn to the question of the economy. The idea of economic strength is inherent in Graeco-Roman perceptions of Armenia: One of the repeated tropes in descriptions of Armenia in the historical corpus concerns the wealth of the territory, which has a long history as a site of gold mining, although the scale of this activity in antiquity is unclear.59 Wherever it may have originated, the clearest figures to quantify this wealth in the period under discussion come from the time of Tigranes II and are reported by Plutarch. According to him, Lucullus took 8,000 talents of coin from the treasury when he captured Tigranocerta in 69 ,60 with Tigranes forced to pay another 6,000 talents to Pompey in 66 .61 Although these episodes both came in the aftermath of Tigranes’ expansion of the Armenian state, and likely reflect capital that he himself had captured in the course of his conquests, they nevertheless testify to the considerable raw wealth that could be concentrated in this borderland state. What, then, were the fiscal systems that surrounded the management of this considerable wealth? There is a lack of evidence for discussing the fiscal regime of Armenia in a meaningful way. The nature of the taxation regime or the mechanisms of its collection are impossible to reconstruct given current evidence. However, if we turn to the more restricted question of numismatic activity, it becomes possible to describe, at the very least, the region’s monetary networks. In what follows, I offer a survey of the limited evidence for the economic situation in Armenia generally, beginning by looking at a few tantalizing pieces of evidence for fiscal administration. Then, we turn to the question of coin use in the region. Here, one can consider two types of numismatic activity: the circulation and use of coinage in Armenia generally, and the minting of coinage by Armenian dynasts. This discussion provides direct evidence for the monetary networks in which Armenia was participating, and proxy evidence for the economic priorities that underpinned state production of coinage. This exploration provides the first concrete example of one of the flexible networks described in the introduction, in this case created by a powerful dynast in this frontier region, Tigranes II. The example demonstrates how even coins – material objects with relatively specific and self-explanatory purposes – could be operationalized in extremely specific ways, and within particular networks, in the frontier. III. Financial Administration in and of Armenia To speak broadly about the details of financial administration in Armenia in the period discussed in this chapter is nearly impossible. On fundamental issues such as 59 E.g., Kunze et al. 2011; Wolf and Kunze 2013. 60 Plut. Luc. 29. 3. 61 Plutarch Pompey (Plut. Pomp.) 33. 4. Multifaceted Highlands: The Economy of Armenia and its South Caucasus Context 453 the processes of taxation, the sources are nearly silent. There are, for example, a few faint hints of evidence that suggest the role of the king in the allocation of land as well as the presence of wealthy temple estates, but the mechanism behind these processes and what they meant for tax/rent income is fully unclear. Little more can be said for periods when Armenia became part of the Roman Empire, although it is clear that, following Trajan’s conquest and as was typical in the Roman world, a financial administrator was assigned to bring the territory into the Roman system. The clearest evidence for this comes in the form of a single inscription concerning the career of a Roman official whose name has been lost, but who is believed to have been the later prefect of Egypt, T. Haterius Nepos. According to the inscription, he had also acted as the procurator of “Armenia maior” during its brief existence.62 No further information exists concerning the nature of his work in the highlands, but his appointment demonstrates that Rome was following the general course of affairs concerning newly incorporated territories in this period and suggests that taxes collected locally were being redirected to Rome, for this short interlude in any event. A possible echo of this system is recalled in MX 2. 48, where the Armenian king is said to have had to pay double tribute to the Romans, who arrived on the borders of Armenia.63 Land and Record Keeping Considering the broader issue of landholding in Armenia, we find some earlier evidence for the active role of Armenian kings in land-right issues. The evidence comes in the form of Aramaic border stele from the time of Artašēs I, which may correspond to a practice attested in MX, whereby the king delineated the borders of specific towns and estates with stone markers.64 This has been interpreted as a way of minimizing border conflicts between those with claims to the lands.65 Other possible evidence for the continuation of this practice comes from a Greek-language inscription from either the first or third century , which describes how a king Tiridates (thought to be either Tiridates I or Tiridates III) granted a certain territory to a particular family.66 Much about the inscription is unclear, but for our purposes, it is sufficient to note that it continues and expands on the formulation of the older Aramaic boundary markers, adding in the further dimension of a local (elite) family as the recipient. 62 CIL XI 5213 = ILS 1338 (Fulginiae). See discussion in Speidel 2021, 137–139. 63 This is part of the passage of MX that contains significant chronological confusion, such that the king named is Artašēs. 64 On the stele, Perihanân 1971; Khatchadourian 2007. MX 2. 56. 65 Russell 1987, 96. 66 SEG 40.1316. There are significant problems surrounding this inscription. See Canali de Rossi 2004; Trever 1953, 273; Vinogradov 1990, 558–559. 454 Lara Fabian Excavations at the site of Artašat (one of the capital cities of Armenia discussed at more length in sec. IV.3.2) have also uncovered a tremendously large cache of between 6,000 and 8,000 bullae, which deserve mention here.67 The bullae come from several areas of the site, but the largest concentration comes from a structure near one of the city gates, which was dated to between the second century  and the first half of the first century . Recent work on this material suggests that it was likely related to a public archive of some sort.68 The use of seals in recording administrative transactions was widespread in both the Seleukid and the Arsakid Near East, and the practice is also attested in the eastern Caucasus.69 The presence of a public archive of this significant scale at Artašat is one small, but meaningful, data point to demonstrate the clear Armenian understanding of the details of normative administrative practice in the broader Near East and – more than that – their adoption of these practices. Unfortunately, until the corpora of sealings are studied more completely, the precise economic relevance of this material remains unclear. III. Coin Use Turning now to our best body of evidence for Armenia’s economic structures, we consider numismatic material. In order to understand the developments in the Hellenistic period, we must first take a step back and consider coinage in Armenia in preceding centuries. There is clear evidence for the circulation of, and indeed the minting of, coinage in the parts of the South Caucasus dating back to the sixth to fifth centuries , although the activity was largely limited to the Black Sea circulation network, for instance in costal Kolchis, several types of silver issues were minted, while imported coins also are attested.70 In contrast, from the inland highland reaches, and from the territory of modern Armenia specifically, a single poorly attested Achaemenid coin find from the city of Yerevan71 as well as a small handful of Athenian tetradrachms and silver fractions from Miletus72 represent the only coinage known from the area from before the fourth century . When we think of the economy of the Achaemenid Empire, where both coined and weighed silver were used in state transactions, the satrapy of Armenia appears to have been aligned more closely to the weighed-silver than coinedsilver tradition, despite its proximity to Asia Minor, a center of coined transactions.73 67 Khachatryan 1996; Manoukian 1996. 68 Schreiber 2021. 69 Babaev 1966. 70 Fabian 2019. For detail on the early coinages of the coast, see Tsetskhladze 1993; Dundua and Lordkipanidze 1983. 71 Pakhomov 1926, vol. 1, n. 1. 72 Mousheghian, Mousheghian, and Depeyrot 2000, 60, 69. 73 A better parallel for the Armenian situation comes from Media, where hoard evidence suggests a similar reliance on weighed silver; see Vargyas 2008. On Armenia, see Khatchadourian 2016, 128–135. Multifaceted Highlands: The Economy of Armenia and its South Caucasus Context 455 III.. Imported Coinage in the Hellenistic Period As with so much else in the Armenian highlands, the rise of the Hellenistic world and Armenia’s increasing involvement with this oikoumene brought changes, in this case, a slow but steady surge of coinage into the region. Nevertheless, the number of finds of imported Hellenistic coins dating to the first century and a half or so after Alexander the Great in the Armenian highlands remains rather limited.74 Evidence for the gradually widening scope of monetary interactions comes from Seleukid tetradrachms of the second century , which appear in the territory of modern Armenia in hoards as well as stray finds and other archaeological contexts.75 The gradual increase of imported coinage over the third and second centuries  suggests that the rise of coin use in Armenia was not an immediate consequence of contact with the Hellenistic states, but rather the product of a longer process of adoption and reconfiguration. III.. Artašēsid Minting Activity Clear evidence for the internalization of the idea of coinage and its intentional deployment can be seen in the decision of Armenian Artašēsid dynasts to begin minting their own coinage, a development that occurred in the late second century . Artašēsid coinage is quite recognizable on account of the specific type of tiara worn by the monarch and has been studied in a series of publications since the 1960s. These fundamental publications, however, have relied mostly on unprovenienced finds and coins from ‘hoards’ assembled by dealers for the market and which therefore do not reflect actual patterns of coin cocirculation, a fact which complicates our ability to date and understand the coinage.76 There are controversies about the date of the first Artašēsid coinage. In earlier literature, scholars generally attributed the earliest known Artašēsid coinage – a series of four silver types – to Tigranes I, a figure of unclear historicity, but according to the traditional chronology, said to have ruled ca. 123–96 .77 This attribution, 74 See references in Mousheghian, Mousheghian, and Depeyrot 2000. 75 The closing dates of the hoards containing these second-century coins (e.g., the Sarnakounk and Artašat 1972 hoards) are often considerably later, however, complicating our understanding of the monetary situation on the ground in the second century . 76 This is true, for example, for the material that Bedoukian worked with in assembling his groundbreaking publications on Artašēsid coinage, Bedoukian 1978; 1987. 77 See Foss 1986, 48–50 on some of the issues surrounding the coins attributed to Tigranes I. There are, furthermore, two coins that are debated in the scholarship, which have been used to argue for an even earlier start of Artašēsid minting, before the time of Tigranes I. These include a gold coin attributed to Zariadres, a Seleukid satrap who ruled in Sophene, and a silver one of Artaxias I (e.g., Marciak and Wójcikowski 2016, 89 n. 92). The former has, however, has been considered by other scholars to have been a forgery (Bedoukian 1964), while the attribution of the second is unclear. 456 Lara Fabian however, has recently been questioned quite extensively, with a range of later dates suggested, generally in the end of the first century .78 In any event, the most complete and best understood numismatic evidence for coinage minted by an Armenian dynast comes from the time of Tigranes II (r. 95– 55 ), when Armenian territorial control reached its maximum extent, stretching all the way from the Caucasus to Phoenicia. In the course of his conquests in Syria and Phoenicia, Tigranes came to rule over cities such as Antiocheia and Damascus. These cities, with their own active mints and coin-production traditions, appear to have had an influence on the physical coins produced under the Armenian dynast, from iconography to denomination systems. Moreover, the interaction also appears to have affected practices of coin use within Armenia, such that not just the form of coins but also their social functions changed. One particular example of this comes from the increasing use of copper coinage in the Armenian heartland, although the phenomenon is difficult to accurately date.79 The silver and bronze issues of Tigranes II therefore provide a particularly interesting opportunity to consider the ramifications of state expansion on the use and deployment of the economic tool of coinage within Armenian space. Tigranes minted a variety of coins at a range of mints, both in his Armenian heartland and in the newly conquered territory of Northern Mesopotamia and Anatolia. An examination of a few of those coins gives some purchase on the roles that they played in Tigranes’ fast-expanding state. One type (fig. 1) is a tetradrachm of the so-called Royal Title type (ΒΑΣΙΛΕΩΣ ΤΙΓΡΑΝΟΥ).80 These coins are generally believed to be from the mint of Antiocheia and are the most common of Tigranes’ tetradrachms. It is worth noting the extremely distinctive portrait style that Tigranes used for his self-presentation, wearing the Armenian Tiara, despite the many features of these coins that represent continuations from earlier Hellenistic models. Based on the large number of dies known for this type, it seems clear that it was struck in very large quantities.81 We can compare this tetradrachm to one of Tigranes’ drachms (fig. 2), once again with the characteristic portrait style. This is of the Imperial Title type with the legend ΒΑΣΙΛΕΩΣ ΒΑΣΙΛΕΩΝ ΤΙΓΡΑΝΟΥ, as are the vast majority of drachms; most speculate this means they were minted in Armenia, but see also discussion on overstrikes below. Many of the drachms have series of letters of contested meaning, which might have been regnal years. Based on findspots of the coins, it is thought that the drachms were more popular in Armenia itself, whereas the tetradrachms were more popular in the Levant.82 78 Nurpetlian 2010, 134–137 suggests Tigranes III; see also Foss 1986; Nercessian 2006. 79 See discussion below on the civic issues of Artašat and the controversies surrounding their chronology. 80 As opposed to the Imperial Title type ΒΑΣΙΛΕΩΣ ΒΑΣΙΛΕΩΝ ΤΙΓΡΑΝΟΥ, which are quite rare and are often associated with mints in Armenia itself; see Nurpetlian 2010, 123; Bedoukian 1968, 53–54. 81 Nurpetlian 2010, 124. 82 Nurpetlian 2010, 125–126. Multifaceted Highlands: The Economy of Armenia and its South Caucasus Context 457 Fig. 1: Silver tetradrachm of Tigranes II, 83–69 , Antiocheia. Obv. Head of the king, facing right. Rev. Tyche seated right, holding palm over figure of Orontes. Diameter 26 mm, 15.79 g. ANS 1977.158.723. © American Numismatic Society. Fig. 2: Silver drachm of Tigranes II, Tigranocerta. Obv. Head of king, facing right. Rev. Tyche standing right, holding a palm over figure of Orontes. 4.128 g. ANS 1944.100.62299. © American Numismatic Society. Fig. 3: Bronze coin of Tigranes II, 83–69 , Antiocheia. Obv. Head of king, facing right. Rev. Nike walking left, holding wreath. Diameter 19 mm, 6.38 g. ANS 1944.100.76970. © American Numismatic Society. Fig. 4: Bronze coin of Tigranes II, 83–69 , Damascus. Obv. Head of king, facing right. Rev. Tyche standing left, holding cornucopia. Diameter 18 mm, 4.6 g. ANS 1944.100.78025. © American Numismatic Society. Next, we have the bronzes (figs. 3 and 4). There are a variety of reverse images, which are thought to correspond to denominations, although this question remains complicated.83 Of particular interest are the bronze coins of Tigranes II that are overstruck on issues of Aradus. The undertypes on the Aradian coins date at the earliest to 146  and at the latest to 69 , strongly indicating that Tigranes acquired these coins during his conquest, as his withdrawal from Syria took place in 69 .84 Why did he overstrike, rather than reminting? One theory is that they were minted in the course of military campaigns to pay soldiers, explaining the need for expediency. Interestingly, however, these coins bear the so-called Imperial Title, which has often been associated only with coins minted in the Armenian heartland. Thus, either that legend was used in a broader geographic area, or Tigranes transported a significant 83 Evidence for this comes from the bronzes minted at the Damascus mint, which seem to have been minted under fairly centralized control. The three types known each fall into a specific weight-band, suggesting that the type helped the user to identify the denomination, on which see especially Foss 1986. 84 Nurpetlian 2010, 156 for an overview of the issues surrounding the overstrikes. 458 Lara Fabian number of these relatively low-value coins across many hundreds of kilometers of mountainous terrain, only to countermark them there. Considered broadly, Artašēsid coinage shows the gradual expansion of the numismatic habit in Armenia, while Tigranes’ multitude of issues pinpoints a particular period in which it came to the fore. Through this, one can demonstrate that, certainly from the first century  onward, Armenia was able to sustain a monetary economy. This entire process has recently been described by Nurpetlian as Tigranes introducing “the concept of coinage to Armenia.”85 III.. Coinage after the Artašēsids One of the striking features of local coinage patterns is just how quickly the flourishing of Artašēsid minting drops off. There is a decline in the volume of minting of imperial issues in the late Artašēsid and early Aršakuni periods, with a near-complete cesura between the late 70s  and the early third century.86 Foreign coins, however, continued to circulate in large numbers, including two types that came to dominate across the South Caucasus more broadly: early Roman Imperial denarii (chiefly the Gaius and Lucius [G-L] type of Augustus) and Parthian drachms of the so-called Gotarzes type, currently attributed to Artabanus II (ca. 10–23 ), Gotarzes II (ca. 40–51 ), and Artabanus III (ca. 80–90 ).87 III.. Coins as Connectivity Above, I told an explicit story about the spread of the numismatic habit in Armenia and the expansion of this practice. However, there is another story that this coinage can tell, which has less to do with monetary practice itself, and more to do with the ways that coins themselves came to serve as a tool of connectivity in late Hellenistic Armenia. In thinking about Tigranes’ coins, we can identify a series of groups of individuals with whom the coins interacted. Through the choice of portrait iconography, we can understand him addressing a group who recognized and accepted this particular, and quite non-Hellenistic, representative style as an indicator of political authority.88 On 85 Nurpetlian 2010, 147. 86 Kovacs 2016 for an overview; see Bendschus 2018 for recent discussion of some of the post-Artašēsid period. 87 For a recent overview of these coins and argument about the functions of the G-L denarii and Gotarzes-type drachms, see Sherozia 2008, 240. 88 On the frequency of this image across the Armenian and Anatolian highlands, see Marciak and Wójcikowski 2016. Multifaceted Highlands: The Economy of Armenia and its South Caucasus Context 459 the other hand, the features of the coins that are more conservative and connected to the cities in the Levant where we believe them to have been minted attest to the fact that these coins were also meant to communicate with people who knew and accepted that numismatic vocabulary. Finally, the fact that so many of these coins, minted in the Levant, ended up back in the South Caucasus attests to a local population in the highlands who was familiar enough with the idea of coinage to readily integrate this material. What emerges, then, is a different understanding of the work that coins were doing: not just instruments for conducting transactions, but rather material objects that, because of their physical form, played a role in both defining and supporting the very network across which transactions occurred. IV Sociopolitical Networks and Administrative Structures We now move one step forward in the question of what those networks looked like and how they functioned outside of the restricted realm of coinage. To that end, the following examination of local social and political organization gives us a wider angle to consider the question of how these communities interacted with their neighbors, and what ramifications those choices had for the shape of economic behavior in the space. We begin with a consideration of dynastic families and the inner ‘court,’ move on to elite culture more broadly, and finally address the role of urbanism and other forms of infrastructural development. IV. The Armenian Kings and the Court We start our investigation at the top of local hierarchy, with the dynastic monarchs and their associated court system. As discussed in volume 2, political elites in premodern states assumed a number of central economic roles in the deeply personal bureaucracies of the time.89 Sovereign rulers of course set the terms for high-level connectivity. But they also exerted considerable rule-setting pressure within the larger circle of elites. As with many other elements of life in the highlands, the court culture that developed in Armenia appears to have drawn on a combination of local traditions, Achaemenid innovations, and Hellenistic patterns, eventually incorporating Arsakid devel- 89 Fabian and Weaverdyck, vol. 2, ch. 3.A: 72–73 for an introduction to this topic in the context of the Hellenistic and Roman worlds. On these themes generally in the Iranian space, see Taasob, vol. 2, ch. 3.B and Fabian, vol. 2, ch. 12.B. 460 Lara Fabian opments as well. Of the myriad ‘external influences,’ the fundamental role of those emanating from Iran should be emphasized. In a paradigmatic statement, Thomson framed the relationship this way: “Originally the Armenians were not so much permeated by Iranian culture as examples of it.”90 This was perhaps nowhere more clear than among dynastic elites, and it shaped the fundamental conditions of the Armenian dynastic world. IV.. Dynastic Marriages in the Historical Record At the very top of the administrative system, and helping to set broad conditions for Armenian political relationships, we find an aggressive practice of marriage diplomacy that appears to have been normal in the space. This practice tied the ruling family of Armenia into a tight network of peer aristocrats drawn both from major imperial contexts and from the pool of local authorities in the polities ringing Armenia.91 We have a range of mentions of dynastic marriages from the Armenian historical tradition that may refer to the Hellenistic period, but they are very difficult to date. Most concretely, MX notes that King Artašēs gave one of his sisters to a King Mithridates, described as the bdeašx92 of the K‘art‘velian marshlands, but who was probably one of the Pontic Mithridates,93 demonstrating Armenia’s close relationships with the Pontic space. We are on firmer historical ground by the first century , when a variety of foreign sources report Armenian–Iranian dynastic intermarriage, through which the Artašēsid dynasty came to be closely entwined with various Parthian Kings of Kings. A clear instance of this is attested in one of the Avroman parchment documents found in 1909, which mentions in passing that one of the wives of a Parthian king was Aryzate, the daughter of a king of Armenia.94 She is thought to be one of the daughters of Kleopatra of Pontos and Tigranes, whose marriage had occurred in the context of the alliance between the Armenians and Mithridates.95 Another of their daughters is said to have married into the Atropatenian ruling family.96 90 Thomson 2004, 373, discussed in de Jong 2015, 125. 91 For general treatments of this phenomenon, see Dąbrowa 2018. See also Schleicher 2021, 488–496 for the situation in K‘art‘li. An expanded discussion of the following section can be found in Fabian 2021. 92 Often translated as ‘viceroy,’ or following Rapp, as ‘toparch,’ this was an office attested in both Georgian and Armenian as well as in the Sasanian world. On the institution, Garsoïan 1989, 516–517; Hewsen 1988–1989, 1990–1991; Rapp 2014, 62–71; Sundermann 1989. 93 MX 2. 11. On the misunderstandings and conflations of the Mithridates in this passage of MX, see also Gazzano 2016; Mari 2016. The question of South Caucasus interaction with the Pontic kingdom deserves more consideration. 94 Dąbrowa 2018, 77–78 on debates surrounding the dating of this parchment. Luther 2018 proposes a later dating of the parchment. 95 Just. Epit. 38. 3. 2, 5. 96 Cass. Dio 36. 14. Multifaceted Highlands: The Economy of Armenia and its South Caucasus Context 461 A range of classical sources document a later Artašēsid–Parthian dynastic marriage between Tigranes the Younger and the daughter of Phraates III of Parthia (r. 69– 57 ), which occurred in the context of internecine fighting between Armenian factions.97 A later Armenian–Parthian union was between the sister of the Armenian king Artavasdes (r. 55–34 ) and Pacoros, the son of the Parthian king Orodes II (r. 57–37 ).98 The wedding of these two was the staging ground for one of Plutarch’s most striking scenes: the decapitated head of the defeated Roman M. Licinius Crassus used as a prop in a production of Euripides’ Bacchae.99 Artavasdes had earlier betrothed one of his daughters to Deiotarus I, sitting on the throne in Armenia Minor.100 There were also other Armenian–Anatolian dynastic unions, as between Archelaos I Sisines of Kappadokia (r. 36 –15/16 ) and an unnamed Armenian princess.101 Their daughter Glaphyra went on to marry into the Herodian dynasty and passed along a claim to the Armenian throne to her son Tigranes V.102 In the tumult of the Roman–Parthian struggle for supremacy in Armenia, claimants to the throne, including Tigranes V and his nephew Tigranes VI, invoked these dynastic ties – but others, like Zeno Artašēs (r. 13 –34 ) assumed the throne without such ties. Although dynastic struggles continue throughout the first and second centuries , the frequency of dynastic marriage in the classical sources declines, although the Georgian tradition describes this as a time of ongoing Armenian–K‘art‘velian unions. There is some mention of dynastic politics among the Armenian leading families,103 although use of MX for this period is complicated by the telescoping of much of the relevant material.104 However, in MX 2. 83, for example, we find a hint of later intermarriage between the Armenian dynastic family and their neighbors to the north, in the story of the marriage of the late third-century King Trdat the Great to a certain Ašxēn. Although her heritage was not named in MX, she was understood by later authors in the Armenian tradition to have been Alan, from the North Caucasus.105 Thus, throughout the entire period under investigation, we find intense high-level ties binding families across this space. For our purposes in thinking about economic processes, these networks had two fundamental effects. They established channels for formal political agreement, which stabilized cross-border relations. The multilateral nature of these ties is suggestive of a foreign policy that aimed to keep Armenia 97 Plut. Pomp. 33. 6; App. Mith. 104; Cass. Dio 36. 51. 1; 37. 6. 4. 98 Plutarch Life of Crassus (Plut. Crass.) 33. 1. 99 Plut. Crass. 32–33. 100 Cicero Epistulae ad Atticum (Cic. Att.) 5. 21 .2. 101 Cass. Dio 49. 39. 2. 102 Augustus Res gestae 27. 2. Sullivan 1990, 300. 103 MX 63. 104 See sec. II.1. 105 On this connection, see Thomson 1978, 233, n. 2. 462 Lara Fabian balanced between its neighbors. At the same time, these connections would have also influenced local consumption patterns, as will become clearer in the discussion of urbanism below. IV.. Social Structures of the Local Elite Below the level of the dynasts and their kin existed a structure of elite families whose organization appears to owe much to Iranian influence, although it is unclear whether these should be understood as practices that date back to the Achaemenid period or whether they were new innovations adopted under the Parthians. MX discusses the origin of the structure of the Armenian court explicitly in relation to a King Valaršak, who is presented as the founder of the Armenian Aršakuni dynasty, but whose chronology is confused and who is largely regarded as legendary or at minimum a composite.106 The information is, therefore, entirely impossible to date, and it is impossible to identify concrete developments in the court system over time. It also remains possible that some of what is described in MX reflects later developments that came about after the arrival of Christianity in Armenia and unfolded in the Sasanian period.107 Some general observations are, however, possible based on this later evidence. The most important sociopolitical structure within the Armenian space by the fourth century , and likely predating this, was the naxarar system, which was a particular type of ‘feudal’ organization based around prominent families, the naxarars.108 The word naxarar is itself a borrowing from Parthian (naxvadār) and was in wide use in Armenian texts.109 These extraordinarily powerful families held the rights to specific offices within the administrative system110 and also held, collectively as a family, unalienable control over specific tracts of land.111 The families were a spatially dispersed phenomenon and were not connected to urban centers – or more precisely, actively avoided urban centers in favor of their own fortified rural estates with control over associated territories.112 The classical 106 In MX, Vałaršak would have ruled at some point in the second century , long before the accepted beginning of the Aršakuni dynasty. On the general scholarly consensus of Vałaršak as a legendary composite, see, e.g., Chaumont 1986. 107 The most detailed information about this system comes from a fifth-century source from Pʿawstos Buzand. See Garsoïan 1976; Adonts 1970 for discussions. 108 For skepticism concerning continuity, Chaumont 1986. For a favorable view that sees the naxarar system as present throughout the Hellenistic period, see Russell 1987, 93–94. 109 On the term and its Parthian roots, see Schmitt et al. 1986, sec. 7c. 110 E.g., the commander-in-chief of the army, the sparapet, which belonged to the Mamikonean family, Bedrosian 1983. 111 Garsoïan 1976, col. 183. 112 Garsoïan 1984–1985. For more on the relationship of this structure to urbanism, see sec. IV.3. Multifaceted Highlands: The Economy of Armenia and its South Caucasus Context 463 texts hint at a similar system of distributed land control at several places. Most clearly, in Pliny, there is a description of the territory as comprising 120 administrative districts termed strategoi.113 While some reject an association between this system and that of the naxarars, it seems likely that this reflects a durable pattern of sociospatial organization in the highland zone, where control of specific valleys is a logical structural choice, although the precise spheres of activity of the naxarars almost certainly did evolve over time. Beyond the elite families, there were additional layers of hierarchy within the elite, at least in some periods. Chief, but also subject to considerable debate, is the office of the bdeašx, often translated as viceroy. This hereditary institution was said by Toumanoff to have evolved under Tigranes the Great, who it is noted was accompanied by four officials who have often been interpreted as bdeašx.114 The term is common not only in Armenian sources, but also in K‘art‘velian ones, where it is attested epigraphically from the first centuries . Although it is hard to be more specific about the function of these individuals in pre-Christian Armenia, it seems to point to hazy evidence for at least some stratification within a broader landscape of distributed power spread between a fairly wide class of elite families, who furthermore were closely tied to specific land claims. IV. Elite Culture Elite actors in a given society exert significant influence over the behavior and particularly consumption patterns of the wider community. In that sense, it makes sense to consider what we know of elite life in Armenia, although one must stress how restricted this sphere was. The material lays the groundwork for understanding the social landscape across which this type of taste-setting flowed, as it echoed beyond the realm of the upper elites and into society more broadly. IV.. Religion One place where the impact of this Iranian orientation is particularly strong (even if the evidence itself is faint) is in the role of Zoroastrianism in Armenia and across the South Caucasus.115 Pre-Sasanian Zoroastrianism, a religion with roots in the Iranian plateau, is, however, a slippery subject that raises a host of interpretive problems in its 113 Pliny Naturalis historia (Plin. HN) 6. 27. 114 Toumanoff 1963, 155. 115 For a recent review of the evidence, de Jong 2015. The key work on the subject remains Russell 1987. 464 Lara Fabian own right,116 and it would, of course, have existed alongside other more local religious traditions which are often fairly slippery.117 Turning to the physical evidence for religious practice in the territory of Armenia, we find a scarcity of material. With the exception of a massive Ionic building, often called a temple but of uncertain function, from the site of Garni, discussed below (sec. V.1), we have limited evidence for other evidence of ritual structures in this period.118 Elsewhere in the South Caucasus, particularly in the territory of K‘art‘li, the evidence is somewhat more convincing. Here, a series of structures have been uncovered that can be connected to fire-temple traditions, although the precise association is not certain.119 Nevertheless, the Armenian textual tradition provides some references to Zoroastrianism in religious practice in the pre-Christian period, although the temporal complexities inherent in working with this body of material do not allow for convincing descriptions of changes over time. The sources describe the most important deity of the pre-Christian period as one Aramazd, who is understood as the Armenian instantiation of the Zoroastrian Ahura Mazda and who also became syncretized with Zeus.120 According to Armenian sources, this Aramazd was worshipped at a temple at Ani, where, at least in one period, the brother of the king was the chief priest.121 Other oblique references to Iranian echoes in Armenian religion of this period come, for example, from repeated references to the sun and sun imagery in descriptions of the Artašēsid kings, with the sun serving as central in Zoroastrian belief systems.122 The Zoroastrian orientation of Armenia’s elites, and particularly of its dynastic families, is also spelled out explicitly in several places in the Graeco-Latin textual tradition. Cassius Dio reports that, upon his visit to Rome to be crowned during the rule of Nero, Tiridates I (r. ca. 52–58, 62–88 ) of Armenia announced himself as follows: Master, I am the descendant of Arsaces, brother of the kings Vologaesus and Pacorus, and your slave. And I have come to you, my god, to worship you as I do Mithras. The destiny you spin for me shall be mine; for you are my Fortune and my Fate.123 116 For the complexity of Arsakid-period Zoroastrianism, see de Jong 2008. On the issue of both Zoroastrianism and other concurrent religious traditions in K‘art‘li, see Schleicher 2021, 376–406. 117 See, e.g., Vardumian 1991 for an approach that stresses local characteristics of Armenian religion in this period. 118 See the description of Artašat below, where a temple precinct has been excavated, but not published, in recent years. The inscriptions from Armavir, which may have related to an oracular temple known from the literary tradition, also deserve note; see sec. IV.2.2. 119 The ‘fire temples’ are four-columned structures known from a variety of post-Achaemenid contexts (see Plontke-Lüning 2009), although their interpretation remains unclear and deserves additional work. In the K‘art‘velian context, the largest and best-published of these is from Dedop‘lis Mindori, on which Gagoshidze 1992; 2001; Furtwängler et al. 2008. On the phenomenon in K‘artl‘i more broadly, K’imšiašvili and Narimanišvili 1995. 120 Russell 1987, ch. 5. As with other questions of the Iranian ‘influences’ on local religion, this association is not accepted by all scholars. 121 Agathangelos §785; MX 2. 14, 53. 122 See, e.g., Russell 1987, 59, 67, 76, 104. 123 Cass. Dio 63. 5. 2, trans. E. Cary with minor adaptions. Multifaceted Highlands: The Economy of Armenia and its South Caucasus Context 465 This Mithras, who eventually entered the Roman world through the mystery religion of Mithraism, was in his origin a Zoroastrian divinity who was invoked as the witness of contracts, and it is to that divinity that Tiridates was referring.124 Further reports of this same visit recorded by Pliny the Elder mention that Tiridates was a mage who introduced Nero to “magian feasts.”125 A final elliptical piece of epigraphic evidence supporting the importance of beliefs associated with Zoroastrianism among the Armenian dynastic families comes from a Greek inscription found at the cliff-top site Garni, which will be discussed in more detail shortly. In the long-accepted reading of this inscription, a King Tiridates – often thought to be Tiridates I – refers to himself as the sun (Helios Tiridates), an uncommon title with echoes of Zoroastrian beliefs although a recent reconsideration of the piece and new edition of the text throws this reading and its association to Tiridates I into doubt.126 The religious life of the court in Armenia was, nevertheless, steeped in Iranian vocabulary and likely behavior, and cultural practices such as Zoroastrianism appear to have played a role in the religious life in the dynastic family, even if they occured alongside other local religious systems. It is difficult to estimate how wide the reach of these practices would have been within Armenian society outside the dynastic family, although MX states that the dynasts placed religious obligations on the naxarars, which suggests that the system likely had a broad reach across elite society.127 Although the precise implications of this for patterns of elite consumption area are difficult to sketch, based on scattered reports about the wealth of these Armenian temple precincts128 and what we understand about the close relationship between these places and ruling families, we might speculate that they acted as significant economic nodes – and ones that were closely embedded within aristocratic power structures. IV.. Literature, Learning, and Language at the Armenian Court At the same time, the world of the Armenian court as described by classical sources was a place suffused with the literary and artistic traditions of the Greek world, and indeed a space that participated in that world. Of course, it must be noted that Greek learning of this type was also familiar from the Arsakid court, where cultural ‘Hellenism’ was at times widespread. In the complex cultural space of Hellenistic-period 124 On the association of this moment with the Zorostrian Mithra, see Beck 2000, 167, n. 95. On the relationship between Mithra and contracts in the Iranian tradition, see Russell 1987, ch. 8. 125 magicis etiam cenis initiaverat, Plin. HN 30. 6. 126 On the inscription, SEG 45.1873 = Canali De Rossi 2004, 14 (no. 17). For the new reading see Bresson and Fagan 2022. 127 MX 2. 14. 128 See particularly Cicero Pro lege Manilia 9.23 on the temple precinct of Anahit. 466 Lara Fabian Mesopotamia particularly, such overlapping adoptions were likely the norm rather than the exception.129 Looking to earlier days of the Armenian dynasts, in the first half of the second century , under the Artašēsids, we find deployments of both Greek and Aramaic inscriptions in official contexts, demonstrating the range of acceptable linguistic possibilities in the context of this kingdom. On the one hand, the Aramaic language was clearly strategically deployed in the context of Armenian royal messaging. The clearest evidence comes from the inscribed boundary steles erected by Artašēs I particularly in the territory around Lake Sevan, testifying to the control of the territory.130 But Aramaic was not the only language in use. There is one group of seven Greek inscriptions, generally dated to the second century , that deserve consideration here.131 Inscribed on boulders near the site of Armavir, both the form and the contents of the inscriptions are curious.132 They preserve fragments of Hesiod and other elegiac poetry, a calendar that lists the months in the official Seleukid format, and a series of what might be understood as excerpts of official correspondence or accounts, including the description of a death of an Armenian king.133 One interpretation of these inscriptions is that they represent some excerpts from a royal archive. Beyond these curious rock inscriptions, there is evidence for the epigraphic use of Greek in both royal and other elite contexts not just from Armenia, but even from more northern reaches of the South Caucasus.134 Latin, in contrast, appears in far more restricted contexts relating to the Roman military.135 By the first century, evidence for Armenian participation in this literary/cultural world is fairly considerable, given the general paucity of information. We learn from Plutarch, for example, that the king Artavasdes “actually composed tragedies, and wrote orations and histories, some of which are preserved.”136 According to Plutarch, the Armenian court also hosted Greek intellectuals, although it proved to be a dangerous place for some of them. He accounts the fates of two anti-Roman Greek thinkers, Metrodoros of Scepis and Amphikrates of Athens, who ended up at the Armenian court, but who fell out of favor and met untimely ends.137 Armenia continued to play 129 Mairs 2014. 130 MX 2. 56, n. 64. 131 Russell 1987, 54–58. 132 See sec. IV.3.3 below, as well as fig. 6. 133 Russell 1987, 74–75; Santrot and Badalian 1996; Mahé 1996. 134 From K’art’li T. Qauxč‛išvili 1999–2000; and the single Greek inscription from Albania, Trever 1959, 340–341, pl. 36. 135 E.g., Speidel 2021, or the Latin inscription found near the Caspian coast, on which Braund 2003. See also Kéfélian 2016 on the possible role of the Roman Army in the transmission of loan words from Latin into Armenian. 136 Plut. Crass. 33. 2, trans. B. Perrin. 137 Plut. Luc. 22. On which episode, Traina 2016, 115. Multifaceted Highlands: The Economy of Armenia and its South Caucasus Context 467 host to intellectuals through the Aršakuni period, as testified by the presence of Iamblichus in the late second century .138 IV. Urbanism and Infrastructure Now, we turn from relatively intangible elements of elite society to a manifestly physical representation of social organization: the city. As already discussed, narratives of Armenia’s place within the Silk Roads on a conceptual level are often connected with the idea of thriving urban centers, which were said to serve as the hubs for both craft production and long-distance trade. And, as discussed in volume 2, urban centers were often key actors in economic activity. We face something of a challenge, however, when we begin to examine the evidence for urbanism in the Armenian highlands, and across the South Caucasus more generally, in the period under discussion. In short, the evidence is contradictory. Although there is some archaeological evidence for considerable urban agglomerations, in particular at the site of Artašat, the broader South Caucasus seems to have a rather thinner patina of urbanization than we find in other areas of the ancient world. And, as described above, the elite naxarar system was decided nonurban in its orientation. At the same time, both the classical and the Armenian textual corpora describe the region’s cities in fairly grand terms. In the following section, we will consider the balance of evidence for urbanism in the area and seek to understand the role that these urban centers played, whether as economic hubs in the classical sense, or in some other specifically local way. IV.. A “Caucasian Model” of Urbanism in the Bronze and Early Iron Ages Considering the longer history of the South Caucasus, we find that local patterns of urban settlement were often quite different from those in other regions of the ancient world. This has been noticed in the later Sasanian period, where Armenia seems to stand as an exception to the general drive toward urbanization.139 However, in earlier periods, the absence of cities is much more marked. This absence of urbanism, even in the face of clear indications of political consolidation, has led to speculation about a so-called Caucasian model of urbanism, in which a nonurban populace was ruled over by a military aristocracy who dwelt in hilltop fortresses.140 More recent analyses have pointed out the shaky evidentiary basis of these theories, and we would be right to carry this concern into our own much 138 On Iamblichus and the Armenian connection, Connors 2017. 139 Garsoïan 1984–1985. 140 Masson 1997. 468 Lara Fabian later period.141 At the same time, detailed archaeological work in the region of the Tsaghkahovit Plain in Armenia has demonstrated that political complexity in the Late Bronze Age there did not coalesce through consolidation of settled populations, but rather from the cyclical spatial practices of pastoralist populations.142 This provides at least some hints of a different underlying social structure in this region, which may have had ramifications in later periods. IV.. An Urban Boom? Artašat and the Phenomenon of the Capital City If the textual sources are to be believed, however, the Armenian highlands saw an uptick in urbanism in the Hellenistic and Arsakid-Roman periods. The trope of city foundation is embedded deeply in the local textual traditions, with both the Armenian and Georgian sources associating city foundation with kingship in a direct way. The classical sources, too, speak explicitly about Armenia’s cities and more specifically position the cities of Armenia as one of the features that linked this ‘distant’ corner of the world to ‘normative’ (from the perspective of the texts) Hellenistic and Southwest Asian patterns. Under the Yervandid dynasty in Armenia, the first capital city had been Armavir, a foundation with roots in the still earlier Urartian Empire,143 whose physical reality was connected to that of the Urartian fortress that had once occupied the site.144 With the rise of the new Artašēsid dynasty, a new capital city appears to have arisen, Artašat. The foundation of this city is granted something of an etiological myth in the classical tradition. Plutarch explains: It is said that Hannibal the Carthaginian, after Antiochus had been conquered by the Romans, left him and went to Artaxas the Armenian, to whom he gave many excellent suggestions and instructions. For instance, observing that a section of the country which had the greatest natural advantages and attractions was lying idle and neglected, he drew up a plan for a city there, and then brought Artaxas to the place and showed him its possibilities, and urged him to undertake the building. The king was delighted, and begged Hannibal to superintend the work himself, whereupon a very great and beautiful city arose there, which was named after the king, and proclaimed the capital of Armenia.145 Strabo, repeating the story about Hannibal, also stresses the beauty of the city and its ideal location.146 The connection to Hannibal, as well perhaps as the nature of the 141 142 143 144 145 146 E.g., Hammer 2014, 758–759. Lindsay and Greene 2013. Hewsen 1986. On this overlapping, see Khatchadourian 2007. Plut. Luc. 31. 3–4, trans. B. Perrin. Strabo 11. 14. 6. Multifaceted Highlands: The Economy of Armenia and its South Caucasus Context 469 city (neither Greek nor Roman),147 earned the city the moniker “the Armenian Carthage” in Plutarch, which evokes a fairly grand image.148 Modern scholars have been skeptical about the historical accuracy of the city’s foundation account,149 and indeed the version of the city’s foundation preserved in the Armenian tradition does not include the participation of any outsiders. In this account, Artašēs built the city atop a hill that pleased him, where he also built a temple to Artemis (i.e., Anahit) and transferred the cult statue from the previous temple. Then he populated the city with “Jewish captives,” along with the riches of the former capital of Armavir, and further developed the city into a worthy royal seat.150 It is worth noting that, as with Armavir, Artašat was also built on a site that held connections to the earlier local dynastic power in the region. Khatchadourian has described the city as being “grafted atop the Urartian citadel that had been abandoned for over four hundred years.”151 The city became a durable seat of power for Armenia, serving as the capital throughout both the Artašēsid and Aršakuni dynasties, with the exception of a brief period under Tigranes the Great, who founded a short-lived capital, Tigranocerta.152 It also became known to Roman audiences. Although not explicitly named, it was also likely the city in which the defeated Crassus’s severed head was said to have appeared as a stage prop in a production of Euripides’ Bacchae,153 suggesting that it must have had fairly grand public accommodations such as a theater. Unsurprisingly, it shows up repeatedly in historical accounts of Roman military activity in the region.154 According to these accounts, it was apparently destroyed by Corbulo155 and then rebuilt.156 Later, it was destroyed again under Nero but rebuilt again with the help of a gift from the Roman emperor to King Tiridates of 200,000,000 sesterces, upon the occasion of his visit to Rome. He reportedly used part of the payment to entice Roman craftsmen to return with him to Armenia to help in the reconstruction efforts.157 In the end, the city was recognizable enough to earn a place in one of Juvenal’s satires.158 Located on a number of hills near the Aras river, the site of Artašat has been the focus of long-running archaeological excavations in both the Soviet and post-Soviet 147 On this point, see Russell 1987, 101. 148 Plut. Luc. 32. 3. 149 See however Russell 1987, 101, who leaves the question open. 150 MX 2. 49. 151 Khatchadourian 2007, 60. 152 App. Mith. 67; Plut. Luc. 25. 4. The location of Tigranocerta is not certain; for more, see below, sec. V.3.3. 153 Plut. Crass. 33. 1–4. 154 E.g., passing mentions at App. Mith. 15. 104; Cass. Dio 36. 51. 1, 36. 52. 1, 49. 39. 3; Florus 1. 40; Eutropius 6. 13. 155 Tac. Ann. 13. 41; Cass. Dio 62. 20. 1. 156 Cass. Dio 63. 7. 2. 157 Cass. Dio 63. 6. 6. 158 Juvenal 2. 170. 470 Lara Fabian Fig. 5: Main hills of Artašat, showing areas excavated in earlier Soviet-period campaigns. (Tonikyan 1992, 170, fig. C) period, although the publication of the site has been uneven.159 In Western scholarly literature, the site is most famous for the two Latin inscriptions that have been discovered at the nearby site of Pokr Vedi, including a large dedicatory inscription dating to ca. 116 , or the period of Roman direct control of Armenia.160 The areas of the site that were most excavated in the Soviet period – hills 1, 5, and 8 – have however yielded a variety of material for thinking about life in Armenia’s capital, although the chronological precision of the available information does not currently allow for a phased reconstruction or much in the way of a developmental narrative (fig. 5). Hill 1 appears to have been mostly military, while hills 5 and 8 contained evidence of domestic activity, craft production, and possibly administrative 159 See Khatchadourian 2008, 266, 270–271 for a history of work at the site. For starting points for past work, see Arakelyan 1982; Khachatryan 2005; and also reports of recent fieldwork, e.g., Gyulamiryan et al., 2021. 160 AE 1968, 0510. Multifaceted Highlands: The Economy of Armenia and its South Caucasus Context 471 contexts as well. The structures on these hills are built in a new architectural style previously unknown from the region, with the details of the masonry techniques indicating familiarity with practices in the Hellenistic world more broadly.161 The presence of hypocaust structures and a likely bathhouse on hill 8 speaks to the adoption of this specific cultural practice of bathing and its embedding within the urban fabric. More recent work in the lower town has uncovered a large temple complex along with elaborate bathing facilities,162 as well as tentative indications of a possible aqueduct, which potentially dates to the period of Roman direct control.163 Materials such as the marble statue of Aphrodite and small finds from the site of metal vessels, glass, and gold jewelry point to a rich material landscape and the presence of considerable wealth, at least among some segments of the population. Graffiti from the newly discovered bathhouse, furthermore, find close parallels from the Northern Mesopotamian site of Dura-Europos.164 What we find, then, is a significant and wealthy urban center that displays clear familiarity with both cultural and material practices from neighboring territories. At the same time, the site in its entirety remains somewhat enigmatic. Consider, for instance, its very layout: although it has been argued that the plan of this site has Hellenistic overtones,165 the quixotic fortified hilltops of Artašat appear to rest uneasily within the corpus of Hellenistic urban planning, seeming in some sense to find better parallels in the world of local fortified hill sites known from earlier periods in the South Caucasus, where dense military infrastructures were packed into contourhugging fortifications. IV.. Other Cities: Capitals and Otherwise Beyond Artašat, there is some other evidence for urban settlements in Armenia, a number of which are named in texts or appear on the Peutinger Table, an itinerary map thought to be based on a fourth-fifth century original. Zarehewan and Zarišat, for example, are two cities known from later periods that have been said to have Hellenistic roots, which have been associated by some with the rise of caravan trade.166 However, the best evidence for exploring the phenomenon of urbanism comes from the other capital cities of the period. Other than Artašat, the archaeologically best-attested city from the period is the preceding capital of the Yervandid dynasty: Armavir. Here, a complicated stratigraphy 161 Invernizzi 1998; Tiratsyan 1979. 162 Khachatryan 2007. 163 Lichtenberger, Zardaryan, and Schreiber 2021. 164 The current state of publication precludes a closer analysis of these; see Khachatryan 2010, 46– 47, 49–50. 165 Tonikyan 1992. 166 Eremian 1953, 11–14. 472 Lara Fabian Fig. 6: Greek inscriptions #4 and #5 from Armavir (Trever 1953, fig. 75). makes the interpretation of the Hellenistic-period layers exceedingly complicated. However, both some scattered architectural traces and more abundant small finds, including imported ceramics, attest to the special status of the place.167 A series of seven rock-cut Greek inscriptions already mentioned deserve note here, as they were carved on the cliffs near Armavir (fig. 6). Although both the language of the inscriptions and the content of some of them display considerable familiarity with the Hellenistic world, the overall practice of placing inscriptions on natural cliff faces is one that has parallels (though imperfect) in the Iranian world, rather than the Mediterranean one, and which is also attested locally in the Urartian period. The relationship between these inscriptions and the city itself is not entirely clear, although they have been frequently associated with a oracular temple that is believed to have been at the site.168 167 Khatchadourian 2007. 168 Trever 1959, Russell 1987. Multifaceted Highlands: The Economy of Armenia and its South Caucasus Context 473 The third capital of Armenia to discuss is Tigranocerta, founded according to the literary sources as the monumental new center of Tigranes the Great’s expanded Armenia and captured nearly as quickly. In contrast to the other two capitals, which are attested both textually and archaeologically, the identification of Tigranocerta is the subject of more debate. One association that has become standard in recent years connects the city with the site of Arzan, located in southwestern Turkey,169 although other candidates still have their adherents.170 While the physical location of this city remains inconclusive, there is a passage concerning it in Strabo that deserves consideration, as we think about the transformation of Armenia under Tigranes the Great: But Tigranes, the Armenian, put the people in bad plight when he overran Cappadocia, for he forced them, one and all, to migrate into Mesopotamia; and it was mostly with these that he settled Tigranocerta. But later, after the capture of Tigranocerta, those who could returned home.171 The idea of a brand-new city, inhabited by those captured from adjacent territories, brought along with their wealth and skills into the new royal seat, and then just as quickly destroyed and scattered, seems to capture something about urbanism in these highlands more generally. It is not that cities did not exist in this space: they assuredly did; and as the evidence from Artašat demonstrates, they could be quite grand. Evidence of the presence of temple centers that grew into cities in their own right provides further evidence for the range of urbanism that existed in this space.172 And yet, this was not a world of cities in the same way that much of southwest Asia was. Everything from the shape of urban fabrics to the rapid building of new capitals in quick succession suggests a certain amount of experimentation. Nevertheless, the recurring Armenian textual accounts of urban foundations as a prerogative of kingship – and indeed, as an expectation associated with the office – perhaps give a clue about how urbanism was functioning in this space. We might best see urbanism as an outgrowth of political power, rather than the more frequent formulation, whereby political power emerged from cities. Finally, in this discussion of cities and urbanism, it is important to end with a note about nonurban settlements. The majority of the population, of course, comprised neither the upper elite group discussed before, nor even a more general urban elite. They were, instead, a dispersed rural population spread across the region’s valleys and highland plateaus. Given the scarcity of archaeological work on these spaces, however, we can say little about how they would have lived. 169 Several sites have been proposed, but for the association with Arzan, see Sinclair 1995; 1996; Marciak 2017. 170 Particularly a site between the Kura and Aras rivers excavated by Petrosyan, see Petrosyan 2021. 171 Strabo 12. 2. 9, trans. H. L. Jones. 172 On which, see sec. VI.I below. 474 Lara Fabian IV. Foreign Militaries and Their Innovations In considering the sociopolitical networks operating in Armenia, I have concentrated on those developing out of local, homegrown actors. But in thinking about the forces swirling in the highlands, it would be a mistake to forget about those brought by foreign powers, and particularly their militaries. Particularly in the realm of infrastructure, there are some hazy indications of significant developments during the brief period of Roman direct control in Armenia, which are in keeping with the pattern in other areas that came under Roman auspices. There is, for example, a single Roman milestone found in the foothills of Mount Ararat that dates to this threeyear period, and that is suggestive of rapid movement toward the expansion of Roman road systems into the highland territory.173 Another much-discussed Latin largescale inscription attests to the participation of Legion IV Scythica in the construction of some large structure near the site of Artašat during precisely the same period.174 IV. Social Flexibility in the Highlands This section has covered significant territory. I began with two of the most powerful actor categories in the region: dynasts and the upper elite. Using the vector of ‘elite culture,’ I considered how social practices inside these groups testify to their patterns of connectivity. Then, using the idea of cities and urbanism, I painted a picture of an Armenia that was at once conversant with, but also distinct from, the world of cities that stretched across the Mediterranean and southwestern Asia. Finally, I briefly noted specific evidence from periods of intense Roman interest in this space to consider how the imperial power sought to leverage infrastructural power to bring Armenia into closer alignment with Roman norms. When we combine the elements of this story, we see is that local political authorities were interacting widely and forming long-lasting connections with their neighbors but were nevertheless performing power, authority, and culture in their own often quite divergent way. This duality – the familiarity with external norms, but the divergent choices made here – is a clear consequence of the geopolitical positionality of Armenia. It furthermore demonstrates how Armenia could have played an active part in these multiple worlds, without ever quite seeming to have done so, thus enabling the diverse historical projections that have been applied to the region, as noted at the start of this chapter. 173 Speidel 2021, 146–149. 174 AE 1968, 510. See recent discussion in Speidel 2021, 139–141. Multifaceted Highlands: The Economy of Armenia and its South Caucasus Context 475 V Production and Consumption Next, we come to a more concrete question: What was produced and consumed in this space? And particularly, what was being produced and consumed in ways that tied Armenia into regional networks? In the following discussion on consumption, I wish to focus on several examples that demonstrate the variety of local and ‘foreign’ products that were being consumed. Rather than seeing these products as indices of foreign trade and contact, as they are often interpreted, I rather used them to demonstrate the choices that were available to consumers in Armenia, and the choices they themselves made as producers. V. Elite Consumption In considering the question of the consumption behavior of Armenia’s elite, our evidentiary material is fairly limited. Systematic excavation of domestic sites has been limited, precluding our ability to discuss the material indicators of elite life in this context, and textual sources are short on material detail. There are, however, several categories of material that we can bring to bear on the question. In principle, what follows considers architectural spaces that can be connected to local elites, and particularly to dynastic families, giving us a sense of the vocabulary of local power as expressed through the built environment. An important piece of evidence in this discussion is the Ionic structure at Garni (fig. 7), one of the few archaeological monuments of the South Caucasus from the period under discussion that has made its way into mainstream textbooks of classical archaeology, no doubt in part on account of its strikingly Mediterranean appearance: The building with a likely connection to the dynastic family ruling the highlands is noteworthy for its clearly classical character. Besides its unique-for-the-region form, the structure is also important as it is one of a very small number of structures that we can associate with the dynastic family of Armenia, although the Armenian texts reference and hint at palaces that undoubtedly must have existed.175 The absence of other examples of royal architecture is made more striking by the fairly widespread presence of structures that can be credibly associated with state power in the Achaemenid period.176 The Ionic structure from Garni, located atop a dramatic cliff, has aroused interest for centuries.177 Many details about this structure – from its identification as a temple 175 MX 2. 49. 176 Ter-Martirossov 2001. On the presence of similar structures across the South Caucasus, Knauß 2000; Knauß, Gagoshidze, and Babaev 2013. 177 It attracted, for example, the attention of foreign travelers Porter and de Montpéreux, who sketched its ruins in the nineteenth century. Ambitious plan for its reconstruction later lay at the heart of the work of the Russian Imperial Archaeological Commission in the South Caucasus in the 476 Lara Fabian Fig. 7: Reconstructed Ionic temple from Garni. © De Agostini Picture Lib., W. Buss, akg-images. to its date of construction – have been debated. Leaving aside the question of its function, stylistic analysis suggests a date in the first or second century , and it shows many affinities to stone-working traditions in Roman Asia Minor.178 The richly decorated hexastyle peripteral building is generally termed a temple dedicated to Mithras, but it may in fact have been a funerary monument connected perhaps to the Armenian royal family, in parallel with similar funerary monuments from Anatolia.179 The site of Garni has strategic importance. It is considered to be that of Castellum Gorneas, mentioned by Tacitus as a well-protected fortress that served as a place of refuge for Armenian kings and their allies.180 It furthermore served as the summer residence for the kings of Armenia, according to the medieval textual tradition.181 The structure was not alone atop the cliffs of Garni. A bath complex that might date to the third or perhaps fourth century  sits near the Ionic building. This complex is small, but richly decorated, featuring a water-themed mosaic sporting Nereids 1880s, although the actual reconstruction of the building remains did not proceed until the midtwentieth century. On this history, see Khatchadourian 2008, 251–258. 178 Wilkinson 1982 suggest a date in the last third of the second century . See also the brief treatment of this structure by Maranci 2018, 26–27. 179 For the interpretation as a funerary monument, see Wilkinson 1982. 180 Tac. Ann. 12. 45. 181 MX 2.51. Multifaceted Highlands: The Economy of Armenia and its South Caucasus Context 477 and other maritime tropes, along with a Greek inscription.182 Although the Ionic structure is unique in the architectural repertoire of the South Caucasus, the bathhouse is not. A series of bathhouses have been uncovered from near the royal seat of K‘art‘li in the north, which display clear associations between bath complexes and key sites of dynastic power.183 As in the case of the Ionic structure, the bathhouses of the South Caucasus demonstrate clear familiarity with Roman architectural practices. Research on the mosaic from Garni as well as another bathhouse mosaic from K‘art‘li, meanwhile, has identified some similarities to the mosaic tradition know from the site of Antiocheia, but also idiosyncratic characteristics.184 The point to emphasize in considering the site of Garni, and what it says about hyperelite consumption practices in Armenia, is that it reflects deep familiarity with and access to Anatolian and Northern Mesopotamian craft traditions and architectural practices. We must imagine that the construction of these structures required the presence of craftsmen either brought in from abroad or with significant experience and training in these spaces. And yet, at the same time, the site as a whole does not sit entirely neatly into either a Hellenistic or Roman paradigm, at least insofar as we might generally understand these terms. As pointed out by Versluys, this type of eclecticism in combining elements from Mediterranean and Iranian cultural vocabularies is a feature of Armenia’s neighbor, Kommagene.185 Although the evidence is less plentiful in the case of Armenia, it would be logical to expect a similar general process. Rather than seeing buildings such as the Garni Ionic building as clear markers of intensive Roman presence in the region, then, we should understand them as indicators of the profound Armenian familiarity with the world beyond its borders, and as the product of intentional decision-making about how to construct an architectural metaphor for expressing royal power. Although the dramatic structures of Garni attract our attention, it is also important to think about what is missing from this picture, namely either royal residences or any evidence for the built environment of the naxarar families. As Garsoïan has convincingly argued, the normative world of these lordly families was not the city, but rather “fortresses, forests and mountains.”186 The image of elite life as depicted in the later textual tradition rested heavily on images that are reminiscent of the Persian ideal of the ‘paradise’: a rural ideal with a long history in the Iranian tradition, and in fact with archaeological antecedents in the South Caucasus in the form of the Achaemenid palace complexes, which seem to have drawn on this mode.187 Architectural traces of these structures in our period are not yet known, but when 182 For an accessible English-language introduction to this building and its excavation, see Arakelyan 1968. See also Wilkinson 1982. 183 A royal necropolis at the site of Armazis-q‛evi and the apparent royal complex at Armazis-c‛ixe. 184 Odišeli 1995; Eraslan 2015; Wages 1986. 185 Versluys 2017. 186 Garsoïan 1984–1985, 76. 187 Canepa 2018, 350–351. 478 Lara Fabian we think about consumption systems as a whole, we should not forget about these spaces, which must have served as distributed, rural hubs of elite performance strewn across the highlands. V. Non-elite Consumption Given the extremely limited scope of domestic archaeology in the South Caucasus, the object-scape of Armenia as reflected in finds from archaeological excavation from mortuary contexts provides our most direct evidence for the shape of Armenian local consumption patterns beyond the hyperelite. Inside of this object-scape, identifiable imports have always received particular attention, particularly among Western scholars.188 Prestige items such as signet rings, glass vessels, and metal vessels have been used to demonstrate the expansion of luxury goods in the space. This attention is, to a certain extent, both understandable and justified. And yet it also sets up a problematic dichotomy between ‘local’ and ‘import’ when, in fact, there is considerable evidence for the local production of some categories of material that, based on autopsy alone, would be judged to be imports. One tangible example of this can be found in a type of pottery known from archaeological excavations in the region, a red-slipped ware that is reminiscent of eastern sigillata from the Mediterranean world. On the basis of visual autopsy, vessels made of this fabric would perhaps be classified as imported, but we now know, thanks to petrographic analysis, that there was local production of this ware type in the territory around Artašat.189 Although the scale of this production activity is impossible to judge, it acts as a cautionary reminder about the significant gaps that still exist in our interpretation of local material culture, which have particular ramifications for our examination of nonelite consumption patterns. V. Extraction and Production for Export In addition to the consumptive potential of the Armenian highlands, which would have made the space a draw for regional trade in its own right, we ought to also think about what the Armenians could have offered to regional and transregional markets: the products of this territory that could have traveled outward. The classical texts provide evidence for a number of natural resources that were exported from Armenia into the wider ancient world, and it is on those that we now focus. The list includes a broad array of primary products, ranging from gold, to a specific stone type used in polishing, to an alkaline clay used medicinally, Armenian 188 E.g., Knauß 2006 on the Achaemenid and early Hellenistic periods. 189 Fishman 2016. Multifaceted Highlands: The Economy of Armenia and its South Caucasus Context 479 bole. In order to understand the economic impact of local primary production in wider systems, we will examine two specific products from Armenia in more detail: natron, a salt used in the production of glass as well as other types of commercial processing, and horses, specifically a particularly prized breed of horse known as the Nisaean horse. Unfortunately, we know very little about the specifics of how these products might have moved out of Armenia. So, to complement these two explorations, we will then consider a detailed example of a trading process concerning Armenia’s eastern neighbors on the shores of the Caspian, the Caspii. Here, we learn details about the production of isinglass, a product derived from the swim bladders of sturgeon that was used as a surgical glue, giving us a glimpse into how complex these trading chains could be. I do not select these examples because I believe that they were the central export products from Armenia or the South Caucasus in terms of quantity; this is a question that we cannot answer on the basis of available data. Rather, inherent characteristics of these very different products, as well as the type of supply chains they participate in, provide an interesting pair for thinking about the ramifications of local resource extraction in broader economic systems, despite the fact that the scale of the extraction and export may in fact have been quite low. V.. Horses Horses were a resource that was widely available in the ancient world, and especially in the broader Black Sea space and of course the Steppe. And yet, despite the multitude of horses theoretically available, Armenia came to be associated in the classical corpus with a specific, and specifically wonderful, type of horse, the so-called Nisaean horse.190 There is some substantial mythology around these horses, which are associated in the textual tradition with Central Asia, Media, and the highlands more broadly. Furthermore, a connection between the Graeco-Latin and Chinese sources has often been made, whereby the Nisaean horses of the classical texts are associated with the “heavenly horses” prized in ancient China.191 Whatever the truth of these associations, Strabo clearly notes that the territory of Armenia is well suited to rearing horses in general, and one of the possible homelands of the Nisaean horses. These horses, as Strabo explains, had once played an important role in international economic ties between Armenia and the Achaemenid kings: The Armenians used to send 20,000 foals a year as tribute to the Achaemenid 190 Strabo 11. 14. 9. About the question of whether these were actually the same Nisaean horses as known from elsewhere in the ancient world, there has been much debate; see Anderson 1961, 22–23. This debate existed also in antiquity, Strabo 11. 13. 8. See also discussions of these horses in Morris, vol. 2, ch. 4, sec. VII.1.2; Fabian, vol. 2, ch. 12.B, sec. II. 191 Strabo 11. 14. 9. 480 Lara Fabian Empire.192 Xenophon provides an even more specific explanation of how the Achaemenid horse-tribute collection system in Armenia functioned. In this account, one particular Armenian village that Xenophon passed through was responsible for raising 17 horses a year that were to be passed on to the Persian king.193 The fact that both a horse and riding clothing appear in the canonical representations of the Armenian delegation on the Apadana reliefs from the Achaemenid capital of Persepolis194 – reliefs generally interpreted as depicting tribute-bearers – suggest that the broad contours of this association are likely correct. Thus, we can speculate that, in the Achaemenid period at least, horses served a political-economic function for the Armenians, articulating their relationship with the Achaemenid central authority and serving as a medium of transaction for meeting their tribute requirements. But also from an internal perspective, horses and horse culture appear to have held considerable social importance. Horse trappings are a frequent grave good across the South Caucasus broadly speaking and diachronically. Even more striking, one finds a scattered practice of horse burial present in the South Caucasus even in the Hellenistic period,195 in a tradition that stretches back to the Bronze Age.196 This speaks to a position of privilege for the horse within local belief systems – a logical state of affairs for a space where military power and horse power were intimately connected. We do not have direct evidence for the role of the horse in either tribute or trade after the Achaemenid period. However, the fact that cavalry from the region is cited as a decisive feature in regional conflicts into the Arsakid-Roman period attests to the continued relevance of these animals in international affairs.197 Whether the Arsakid dynasty took over the Achaemenid pattern of importing horses from this territory is not known. It is also possible the shift in military organization ushered in under the Arsakids led to the levying of mounted warriors rather than steeds themselves: that is, the focus may have shifted rather to the ‘export’ of mounted fighters. The same might have been true in the Roman case. We see, for example, the importance of Armenian cavalry during Marc Antony’s 37  ill-fated Parthian campaign. However, we hear nothing more about Armenian mounted fighters until the early second century , when Arrian describes the role that these troops had in defending Kappadokia from steppe pastoralists surging westward.198 Despite the many points of uncertainty, we do know from Strabo, but also from later accounts, that the Armenians continued to be associated with these horses throughout antiquity. This association 192 193 194 195 196 197 198 Strabo 11. 14. 9. Xenophon Anabasis (Xen. Anab.) 4. 5. 24, 34–36. Khatchadourian 2016, 217, n. 4. Nachmias et al. 2021. Pogrebova 2003. Plut. Ant. 8. Arrian Acies contra Alanos. Multifaceted Highlands: The Economy of Armenia and its South Caucasus Context 481 should be understood as a faint echo of the role that these animals could, and occasionally did, play in equipping regional armies. V.. Natron The second example treated here concerns a very different type of resource: natron. Natron, a critical ingredient in the production of glass and other industries that used vitrified materials such as glazed ceramics, is a fascinating commodity because of its extremely limited availability. Unlike horses, which could in principle be raised in a great variety of environmental zones, natron salts are known from a relatively limited number of sites in the ancient world, most notably Wadi Natrun in Egypt.199 However, another natron source is mentioned by Strabo, who, in the course of a passage on the natural resources of Armenia, cites the waters in and around Lake Van (located in eastern Turkey) as “natron-producing” (nitritis).200 The dominance of Egyptian natron in the ancient glass industry has been widely accepted. Glass production in the ancient world was a complex and segmented production process, with the primary production of raw glass – the energy-intensive process whereby sand was transformed into glass ingots – occurring at a relatively restricted number of sites, concentrated largely in the Eastern Mediterranean.201 The glass ingots were then shipped far and wide to local workshops, where finished vessels were produced. A well-developed supply chain appears to have emerged to supply natron, used as a flux to lower the melting point of silicates, to primary glassproduction facilities in the Eastern Mediterranean.202 However, that dominance of Wadi Natrun natron does not preclude the exploitation of the alternative Anatolian natron source mentioned by Strabo and known from other later sources.203 One hint at the possible exploitation of these sources comes from investigations of glass production from northern Mesopotamia and Anatolia in the Late Bronze and Iron Ages. Of particular interest is the glass from Urartian fortresses, located in the Armenian highlands and with a significant center of power around Lake Van. It was long assumed that this locally made glass was produced not with natron, but rather with another possible flux derived from plant ash. Recent investigations, however, have demonstrated that an alkaline flux was, indeed, used.204 Given the extraordinarily close proximity of these Urartian sites to Lake Van, it is very tempting to hypothesize that they were relying on the local natron resources. 199 On Wadi Natrun, Shortland et al. 2011. 200 Strabo 2. 14. 8. 201 Jackson et al. 2018 for a discussion of the Roman glass industry. See also Weaverdyck and Fabian, vol. 2, ch. 8.A, sec. VI.3 for a brief discussion of the unusual characteristics of this production system. 202 Shortland et al. 2006. 203 Strabo 11. 14. 8. 204 Dardeniz 2015, 196–197. 482 Lara Fabian Indeed, the alternative hypothesis – that they were importing natron from far-distant Egypt – seems rather unlikely. Thus far, there is no concrete proof that the Lake Van natron was being used in glass production either in the Bronze or Iron Ages or in the period we consider here.205 However, the ability to recognize different natron sources chemically or isotopically is still relatively new, and only a narrow range of vessels have been studied to date. One wonders, for example, about the famous Parthian green-glazed ceramics: Could Anatolian natron possibly have played a role in this major Iranian ceramics industry? V.. Isinglass Expanding our vision from Armenia to the east, along the banks of the Caspian, we find a range of both material and textual sources that suggest that there was organized trade activity in that space as well. One particularly valuable testimony, which describes both a trade route and a specific trade product – isinglass – is found in Aelian’s De natura animalium, a collection and compilation of anecdotes about natural history written in Greek in the third century . In this passage, likely derived from a Hellenistic source, Aelian details trade between the Caspii, one of the peoples located on the Caspian coast in the eastern Caucasus,206 and Ekbatana: I have heard that in the land of the Caspii there is a lake of very wide extent, and that in it there occur large fishes which are called Oxyrhynchi. Now the Caspii hunt them and after salting, pickling, and drying them, pack them on to camels and transport them to Ecbatana.207 The passage goes on to describe the process of producing an extremely valuable secondary product from the internal organs of the “large fishes,” i.e., Caspian sturgeon. Once dried and processed, sturgeon viscera, and particularly their air bladder, can be processed to create a ‘fish glue’ made from isinglass. This substance has a wide range of medical and pharmacological uses that were known in the ancient world, where it was a valuable commodity.208 Thus, this passage provides a general description of (1) a group of traders; (2) a commodity; and (3) a shipping destination. Although it does not enumerate the precise route along which the goods moved, and our inability to locate the Caspii with any 205 And indeed, there is good reason to believe that many of the finished glass vessels found from the South Caucasus were produced in Levantine workshops, see Shortland and Schroeder 2009. 206 The Caspii appear in a number of accounts of the eastern Caucasus, beginning with that of Herodotus, 3.92–93. See Bais 2001, 51–52 for an overview of the sources that mention this group. 207 Aelian 17. 32, trans. A. F. Scholfield. 208 Scarborough 2015. Multifaceted Highlands: The Economy of Armenia and its South Caucasus Context 483 accuracy prohibits much speculation, the passage is helpful in thinking about both the patterns and the scale of regional trade activity. The transport of fish is, of course, difficult to substantiate archaeologically, so we are unlikely to find evidence of this particular activity in material culture. However, assuming that the general flow of materials from the eastern Caucasus into northwestern Iran was not limited to fish, we can search for other, more durable, materials moving along these channels. On this point, recent research on early Sasanian glass from Gilan has posited a South Caucasus source for some of the material.209 This material postdates the period under discussion here and relates to the relatively isolated corner of Gilan, rather than the far more central Ekbatana. However, it does demonstrate the possibility of fairly extensive and specialized trade networks and production processes centered on the South Caucasus, serving northwest Iranian markets. VI Trading Centers and Routes in the South Caucasus Having considered a few classes of objects that were moving in this space, we next turn to the question of loci for trading activity, and routes along which the materials may have moved. Unfortunately, our ability to talk about merchants who would have been active in these spaces is hampered by the sources we have available. However, we can interrogate the physical spaces that facilitated the connectivity necessary for the movement of objects in the territory. This serves to connect the discussion thus far in this chapter back to fundamental questions about the place of Armenia within Eurasian systems. VI. Trading Centers In the classical literary corpus, there are a handful of brief references that explicitly situate the Armenians within long-distance commercial networks. Our earliest detailed source on the region who includes such information is Strabo, who, because of his own familial ties to the Black Sea region, should be understood as a unique reference. He speaks of the involvement of the Armenians in trade in several places, although always in passing. One interesting reference places Armenian merchants in the sanctuary city or “temple state” of Komana Pontika located in Pontos and dedicated to Artemis, which 209 Simpson 2015. 484 Lara Fabian is called a “noteworthy emporion” for the Armenians.210 The city was compared by Strabo to Corinth, since its lures – and presumably particularly the practice of temple prostitution – attracted “merchants and soldiers” who spent profligately, as well as large crowds who gathered for temple festivals. We know nothing more about which Armenians were operating here or the details of their activities,211 but Strabo’s association of such a sanctuary city with mercantile activities finds parallels in descriptions of other sites in Anatolia, for example Pessinous.212 The site of Komana, located on routes leading to the Anatolian heartland as well as the Black Sea, would furthermore have made a logical staging point for mercantile traffic moving across the Armenian highlands. Another similar wealthy temple precinct is associated with Armenia itself, the temple of Anahit (syncretized with Artemis) at Eriza, which is attested in both the classical and Armenian traditions.213 The texts make an explicit connection here between Iranian traditions and their practice in Armenia, as well as the mention of temple prostitution – making it quite similar to Komana. This temple precinct grew over time into a veritable city in its own right, and one with tremendous riches.214 One scholar of pre-Christian religion in Armenia, in his consideration of this type of temple, notes the explicit relationship between the religious and trade functions of this site and credits the expansion of trading activity for the flourishing of the site.215 Broadening our scope from the Armenians themselves to their neighbors in the South Caucasus, we find a handful of other references to emporia. In particular, two sites associated with the coastal territory of Kolchis come to mind. Dioskourias is noted as the “the common emporium of the tribes who are situated above it and in its vicinity,” while Phasis is, more explicitly, called the “emporium of the Kolchi.”216 These sites were integrated into a system of Black Sea trading hubs, many of which have roots in the Greek colonial period. In his discussion of the cities of Kolchis within this system, and of their economic basis, Tsetskhladze has commented on the prominent – nearly exclusive – focus on trade among ancient descriptions of Kolchian population centers.217 While it seems 210 Strabo 12. 3. 36. On the vocabulary of temple states in the context of Asia Minor, see Dignas 2002, 227. 211 Beyond the word emporion, Strabo provides little further description of trade activities at the site. 212 Strabo 12. 5. 3. 213 Strabo 11. 14. 16; Agathangos § 779–781; MX 2. 49. 214 Agathangos § 786. 215 Vardumyan 1991, 123–124. 216 Dioscurias: Strabo 11. 2. 16; Phasis: 11. 2. 17. For another mention of an emporion at Phasis, see Hippocrates Peri Aeron, Hydaton, Topon (Airs, Waters, Places) 15. 217 Tsetskhladze 1992, 238–239. The comparison here is drawn between cities of the Bosporan kingdom, which are described as multifaced hubs of different sorts of economic activities, and those of Kolchis, where the singular focus is on trade. Multifaceted Highlands: The Economy of Armenia and its South Caucasus Context 485 unwise to read too much into this feature of the textual corpus, it is entirely fair to note the uncommonly explicit connection that is drawn between urban hubs and trade in this space. We cannot speculate on what role the Armenians themselves might have played in these cities, but their presence would have come as no surprise. Beyond our general understanding of the close relations between local communities, evidence for Armenian interaction with Kolchian populations comes from another passage of Strabo. Here, he describes the existence of a temple precinct in the mountainous Moschian territory of the South Caucasus, which was “divided into three parts: one part is held by the Kolchians, another by the Iberians, and another by the Armenians.”218 Such shared cult sites could well have also served as the sites of periodic markets or other organized economic activity, as testified at Komana. One tantalizing, but entirely unclear, hint of the involvement of cult sites in economic transactions from the South Caucasus comes from the site of Vani, a settlement interpreted as a religious center in western Georgia. There is a bronze coinage associated with the city of Vani that seems to be minted in the broad pattern of bronze issues from the territories of Mithridates of Pontos, like issues from Dioskourias.219 Although it would be an exaggeration to argue that the cities discussed earlier did not play a role in trade systems, it nevertheless seems fair to point out that religious centers appear to have featured prominently in these networks. VI. A Route to India? A Route to the North? I end here with a discussion of several explicit references to the routes for traveling through this space. The first comes in Strabo’s extended description of the people dwelling in the South Caucasus and near the Caspian Sea in book 11, mentioned at the start of this chapter. Here, the Armenians are named as one of the participants in a long-distance trading system that brought “Indian and Babylonian” merchandise through Media, Armenia, and eventually to the Aorsi, a mobile pastoralist confederation living near the Caspian, likely in the North Caucasus.220 This passage of Strabo, along with a handful of other references to India, raises the tantalizing prospect that Armenia were not just active in trade in a regional way, but actually participated in a specific long-distance trade network stretching from India to the Pontic Basin/North Caucasus, which we might consider a discrete spur of the ‘Silk Road’ in the classical sense. The existence of such a route in the South Cauca- 218 Strabo 11. 2. 18, trans. H. C. Hamilton. 219 However, whereas those coins circulated broadly, the bronze issues from Vani did not, but instead appear to have been used almost exclusively at the single site; Dundua and Lordkipanidze 1983, 22– 41. For discussion and bibliography of these coins, see Tsetskhladze 1989. On the phenomenon of Mithridatic coinage more generally, see Saprykin 2007. 220 Strabo 11. 5. 8. 486 Lara Fabian sus has, unsurprisingly, attracted considerable excitement and been long debated in the specialist scholarship.221 This connectivity is theorized by many to have occurred via a link between the South Caucasus and the western reaches of Central Asia in the form of a shipping channel across the Caspian Sea. This is a prospect that is explicitly noted by Strabo, who describes the route as follows: Aristobulus says that the Oxus [Amu Darya] was the largest river, except those in India, which he had seen in Asia. He says also that it is navigable with ease, (this circumstance both Aristobulus and Eratosthenes borrow from Patrocles,) and that large quantities of Indian merchandise are conveyed by it to the Hyrcanian Sea [the Caspian Sea], and are transferred from thence into Albania by the Cyrus [the Kura River], and through the adjoining countries to the Euxine.222 The route was allegedly ‘discovered’ by Patrocles, a Macedonian general and representative of Seleukos I who supposedly circumnavigated the Caspian Sea. And Strabo’s is not the only mention of such connection. Pliny reports the following: Varro further adds that exploration under the leadership of Pompey ascertained that a seven days' journey from India into the Bactrian country reaches the river Bactrus, a tributary of the Oxus, and that Indian merchandise can be conveyed from the Bactrus across the Caspian to the Cyrus [the Kura River] and thence with not more than five days' portage by land can reach Phasis in Pontus.223 In both cases, the path noted for the movement of goods does not actually pass through the heartland of Armenia, instead traveling along the Kura River, then through Armenia’s northern neighbor K‘art‘li, to the coastal port city of Phasis. And yet, in broad strokes, the supposed path would put the movement of these goods within close reach of the Armenians. This transportation system, under the name the ‘Oxo-Caspian trade route,’ was first seriously discussed in the early twentieth century by Tarn, who viewed its existence as unlikely. Tarn’s thoughts on this route are worth discussing in some detail, as they set the tone for much later scholarship.224 His argument against the existence of such a route is based, on the one hand, on the fact that he saw no evidence for the existence of navigable rivers flowing from Central Asia all the way into the Caspian. He argued that the lack of frictionless riverine transportation would have made the route essentially useless, given the easier overland options.225 221 For recent discussions, see Braund 2002; Callieri 2001; Rtveladze 2010. 222 Strabo 11. 7. 3, trans. H. C. Hamilton. See also Strabo 2. 1. 15 for another account of this general route: The Oxus, which divides Bactriana from Sogdiana, is said to be of such easy navigation that the wares of India are brought up it into the sea of Hyrkania, and thence successively by various other rivers to the districts near the Euxine. 223 Plin. HN 6. 50 trans. H. Rackham. 224 Tarn 1901. 225 Tarn 1901, 25–26. Multifaceted Highlands: The Economy of Armenia and its South Caucasus Context 487 Leaving aside the question of the existence of a comprehensive path of navigable riverine routes in Central Asia, which remains a complicated one, one might note that the same general point about the difficulty of riverine transport could also be made for the South Caucasus. Here, Braund has argued that seamless east–west movement was interrupted by the Surami Ridge, which divided the coastal territory of the western Black Sea lowlands from the mountainous interior.226 We will return to the question of riverine transport in a moment. In arguing against such a route, Tarn furthermore points to another passage of Strabo, in which it is noted that there were “no vessels upon the [Caspian] sea, nor is it turned to any use.”227 The discussion of the route is therefore entangled with the question of whether the Caspian Sea saw maritime transport in this period. Strabo’s mention of the lack of ships on the Caspian, however, appears within a description of the Hyrkanians and other local populations near the Caspian that has a strong overtone of allegory – e.g., describing the bountiful harvest of local agricultural products that grew without any work from the residents. This context should shape our reading of this particular passage and makes it likely that the mention of lack of maritime exploitation was also part of this topos. There is considerable evidence for slightly later Sasanian-period maritime exploitation of the Caspian, and there is no reason to believe that this was a new innovation in the Sasanian period.228 Beyond the textual material, there is also some fairly limited archaeological evidence for such connections, with more coming to light in recent decades. The most obvious evidence includes a small number of Bactrian coins found at several sites in the South Caucasus: a single coin from a large hoard from the territory of modern Azerbaijan,229 and a hoard of six coins found in construction work in the city of Tbilisi in Georgia.230 Beyond this, there are from the region of modern Georgia a small number of glass beads that, based on their chemical signature, appear to be of South Asian manufacture, as well as some very limited evidence for silk textiles.231 Tarn ends his article on the Oxo-Caspian route by saying the following: “It appears to me that we are safe in saying that whatever trade came down the Oxus and across the Caspian was entirely in native hands during the whole period of Greek knowledge of this river; and that it was of no great extent.” On the first point – that the trade was conducted by local actors rather than by Greeks – one would be inclined to agree. On the question of extent, however, we would do well to be more cautious. It is not clear that one can argue that there was mass-scale movement of goods from Central 226 Braund 1994, 40–41. 227 Strabo 11. 7. 2. 228 See this argument in Callieri 2001, 540–541. On recent work on Sasanian harbor infrastructure, see also Rckavandi et al. 2008. 229 Babaev and Kaziev 1971; Thompson, Mørkholm, and Kraay 1973, no. 1737. 230 Pakhomov 1938, n. 319. 231 On the beads, Shortland and Schroeder 2009, 961–962; on the silk fibers, Kvavadze and Gagoshidze 2008. See also evidence presented in Schneider 2017, addendum 2. 488 Lara Fabian Asia into the South Caucasus. Rather, as our understandings of the complexity and segmented nature of ‘Silk Road’ trade has developed over the last century, we have come to realize that mass-scale mobilization of trade goods was likely more the exception than the rule. This was a system in which smaller-scale, but habituated, trade relationships played a central role. It is for precisely this reason that the lack of a complete riverine path either in Central Asia or in the South Caucasus should not bother us unduly. From the Eastern Desert, through Northern Mesopotamia, and across the Iranian Plateau, we find examples of movement of goods across harsh landscapes. Reframing our expectations within the context of this understanding, it becomes possible to see how the South Caucasus generally may well have served as an important segment within this larger system. Finally, there is an important point that is often overlooked from the passage of Strabo cited at the beginning of this chapter. The end point for the goods that moved through Armenia was not the Pontic, but rather the Aorsi, a mobile pastoralist group holding sway along the north Caspian coast: … for they [the Aorsi] held dominion over more land, and, one may almost say, ruled over most of the Caspian coast; and consequently they could import on camels the Indian and Babylonian merchandise, receiving it in their turn from the Armenians and the Medes, and also, owing to their wealth, could wear golden ornaments.232 Thus, in contrast to all of the discussion about east–west routes across the South Caucasus, Armenia actually serves in this reference as part of a system running south–north from Mesopotamia into the Eurasian Steppe. It is to this question – the northward connectivities of Armenia, that I now turn. VII On the South and North Caucasus In most of this chapter, I have focused on the web of connectivity stretching between Armenia and its imperial neighbors, first in the Hellenistic world and then later in the Arsakid and Roman ones. It would be a mistake, however, to overlook the neighbors to the north, dwelling along the northern slopes of the Greater Caucasus and their steppe frontiers. These patterns of connectivity are even more tenuous and are often missed for precisely this reason. They nevertheless deserve mention, as they likely represent an important component in regional dynamics. There are two core reasons for the lack of recognition of these patterns of connectivity. The first can be explained by another passage from Strabo, who states that there was a clear boundary between the nomads of the north and the sedentary polities of the Armenians, K‘art‘velians, and Albanians to the south: 232 Strabo 11. 5. 8, trans. H. L. Jones. Multifaceted Highlands: The Economy of Armenia and its South Caucasus Context 489 As for the Armenians, and the peoples who are situated above Colchis, both Albanians and Iberians, they require the presence only of men to lead them, and they are excellent subjects, but because the Romans are engrossed by other affairs, they make attempts at revolution ... whereas the nomads, on account of their lack of intercourse with the others, are of no use for anything and only require watching.233 But the more pervasive cause is an overly deterministic interpretation of landscape factors, with many generations of scholars seeing the Greater Caucasus range as a fundamental barrier to motion – a wall blocking paths of connectivity. This is, however, a mistaken impression that is poorly grounded both theoretically and materially, with the mountains acting as a far more porous ‘barrier’ than imagined.234 VII. Neighbors to the North Who, then, were these northern neighbors with whom the South Caucasian political authorities came in contact? The truth is that, despite significant advances in archaeological research in recent decades, our understanding of the communities of the steppe and steppe fringe in the North Caucasus remains exceedingly hazy. In our confusion, we are not alone: As Pliny the Elder noted in his discussion of the steppe around the Caspian Sea, “in no other part [of the world] is there greater incontinency among the authors, I think that it is because of the immense number of peoples and their nomadism.”235 Despite the bewildering number of discrete ethnonyms that appear concerning this space, there is a standard schematization that has been long used by both Soviet archaeologists and ancient historians more generally, which suggests a (relatively linear) sequence of successive steppe peoples inhabiting the zone. This process is said to start with the Scythians, an Iranian-speaking group with their heartland in the North Pontic, but thought to be present into Mesopotamia and Anatolia in the first half of the first millennium , alongside the Cimmerians. They were then, according to the standard description, displaced and replaced by Sarmatians, another group of East Iranian-speaking pastoralists who came to dominate the Western Steppe in the second half of the first millennium  while also maintaining ties further to the east.236 Then, finally, the Alans emerged, who are considered a late-stage Sarmatian community or federation, from whom the modern Ossetians trace their descent.237 This sequence of development has been criticized by scholars in recent years as being largely a modern invention, and the contours of variation inside of these groups 233 Strabo 6. 4. 2, trans. H. L. Jones. 234 Fabian 2017. 235 … nec in alia parte maior auctorum inconstantia. Plin. HN 6. 18. 236 Mordvintseva 2013b, 2013a, 2015 and Dan 2017 offer theoretical-historiographic perspectives on Sarmatian developmental narratives. 237 Shnirelman 2006 with reservations about the Ossetian-Alan identity discourse. 490 Lara Fabian remain highly debated.238 For our purposes here, it is enough to point out that the North Caucasus and the steppe that it bordered was the home to a diversity of interconnected steppe populations who shared many linguistic and cultural commonalities, above all reliance on pastoralism and (semi)mobile lifeways based on stockraising.239 VII. A Mythological Dynastic Marriage We have already considered a range of dynastic marriages that shaped the course of Armenia’s elite, and which were recorded in a range of historic and documentary sources. There is one critical one that has not been mentioned yet, which bears on the question of North–South Caucasus relationships: that of Artašēs I (the eponymous founder of the dynasty) to an Alan princess from the North Caucasus, Sat‘enik.240 This story is not recounted in the Graeco-Latin historical tradition, but instead comes to us through MX (Patmut‘iwn Hayoc‘), as part of the emic Armenian chronicle tradition.241 Despite the mythological nature of parts of this narrative, and a related one from the Georgian chronicle tradition discussed momentarily, the accounts are noteworthy for the worldview that they present – a worldview that reflects local understandings of regional relationships and history from the point of view of the late antique and medieval chroniclers. The MX recounts how Artašēs and Sat‘enik met during an Alan incursion across the Kura River, after the Alans had already made their way through K‘art‘velian territory. Artašēs’s army launched a successful campaign to repel the Alans and in the process captured an Alan prince, the brother of Sat‘enik. The princess, while attempting to negotiate for the release of her brother, caught the attention of Artašēs, who decided he wanted to take her as his wife. After an unsuccessful attempt to win over her father, Artašēs abducted the princess, lassoing her across the Kura. He subsequently paid a hefty bride price to the Alan king and was wed to the princess in a lavish ceremony, and they went on to have six sons. Intriguingly, a very similar story of a North–South Caucasus dynastic marriage sits at the heart of the etiological narrative of the first dynasty of Armenia’s northern neighbor, K‘art‘li, in the Georgian chronicle tradition. Here, the Lives of Georgia (KC 238 For an analysis of the constructed nature of much of the discourse, Dan 2017; Mordvintseva 2013b. 239 There is, however, a great degree of diversity within pastoralist practice: ‘Nomads’ is an oversimplification. For longue-durée perspectives on mobile pastoralist adaptation, Hammer and Arbuckle 2017; Honeychurch 2014. 240 MX 2. 50. On the chronological confusion and projection of this passage into the second century , see sec. II.1 above. See Fabian 2021 for a deeper exploration of this episode and related evidence explored in sections VII.2 and VII.3. 241 Sec. II.1. Multifaceted Highlands: The Economy of Armenia and its South Caucasus Context 491 [K‘art‘tlis Cxovreba]) recounts how the first king of the P‘arnavazad dynasty in K‘art‘li, P‘arnavaz (r. ca. 299–234  according to the chronicle tradition),242 marrying a woman from the region of Durżuket‘i (mod. Ingushetia/Chechnya).243 Although the KC is not as detailed about the events surrounding this union, the testimony is direct about the benefits of this alliance for P‘arnavaz’s son and successor, Saurmag. When Saurmag faced problems with other elite K‘art‘velian families, he was able to withdraw to his mother’s people in the north, and with their help and that of Alan kinsmen on his father’s side, defend his claim to the throne.244 Put directly, these two accounts present (from the perspective of their medieval authors) a picture of the ancient world in which marriages across the Greater Caucasus watershed sat at the heart of dynastic political authority in the South Caucasus. VII. Other Evidence for North–South Connectivity There are several other lines of evidence suggesting that local communities and political authorities in the South Caucasus were more closely interwoven with the steppe in the north than is generally acknowledged. Much of this evidence, however, relates to Armenia’s northern neighbor K‘art‘li, which seems to have been a central mediator in these relationships, and likely more tightly networked than Armenia itself. This evidence is, nevertheless, relevant for our understanding of the broader system in the highlands. I begin by briefly introducing two lines of evidence: (1) onomastics of elites from K‘art‘li, and (2) a small number of tamgas (symbols common in the steppe world) found in elite contexts/objects from K‘art‘li. I then conclude with an analysis of the role of steppe populations in the geopolitics of the South Caucasus based on the Graeco-Latin textual corpus. VII.. Onomastics and Tamgas In the first case, there are several key names of political authorities in the Georgian chronicle tradition, like Saurmag and K‘art‘am, that have possible East Iranian (i.e., Sarmato-Alanic) associations.245 These individuals appear in parts of the tradition that concern the Hellenistic/classical-period history of the kingdom (insofar as material 242 Following the regnal dates of K‘art‘velian dynasts proposed in Toumanoff 1969; see also Rapp 2009, 652, n. 18. Some argue for a later date for the consolidation of Hellenistic K‘art‘li, e.g., Meißner 2000, 188; Schottky 2012, 245–246, cf. Gagoshidze 2008, 3. 243 KC, ed. S. Qauxč‛išvili 1955, 24, 26–27. 244 KC, ed. S. Qauxč‛išvili 1955, 26, Z. 19–23, trans. Thomson 1996, 38–39. 245 See the discussion in Rapp 2014, 226–227. 492 Lara Fabian from the chronicles can be dated). Documentary evidence for similar onomastic connections comes from a small number of inscriptions from K‘art‘li. We find the name Zewaḥ on several large Armazian–Greek bilingual or Armazian inscriptions, and beyond this on at least three other smaller objects from graves in the region, as well as on an inscription at Tanais in the North Pontic.246 Even more explicit in its northern connection is an inscribed intaglio from Žinvali that likely dates to the third century , bearing a Greek inscription reading BAKOYR ALANA, generally read as “Bakour the Alan.”247 The name Bakour appears elsewhere in the Georgian tradition, where two kings called Bakour and Bak‘ar ruled between the early third and early fourth centuries respectively. This onomastic material should not be taken as clear evidence of ethnic affiliation. Firstly, names can mark a range of other identity formulations, such as political ties.248 And secondly, the timeframe of the scattered names is very broad. Nevertheless, it is notable that these names appear repeatedly in close connection to ruling families. In addition to this onomastic evidence, I also briefly mention the few tamgas that have been found in the South Caucasus. Tamgas and tamga-like signs are associated with Sarmatian contexts, where they seem to have served a range of functions including the marking of presence and of property ownership.249 Despite their steppic associations, finds of tamgas are also known from outside of the steppe proper, in settlements in the North Pontic for example, where they seem to have retained an associated with authority and administrative power.250 From the South Caucasus, symbols that seem to be tamgas appear above the top line of text one Armazian–Greek bilingual inscription from the site of Armazis-q‛evi.251 Although it is not entirely clear 246 From Mtskheta: on gemstone set into a belt (Apakidze et al. 1958, pl. XLV 3); on a gold ring (Apakidze et al. 1958, 79, fig. 135.16); and more ambiguously in an abbreviated form on a second ring (Apakidze et al. 1958, 48–49, fig. 19). The name, this time in Armazian, was also inscribed on a spoon found in later excavations at Bagneti (Apakidze 1973). For the name at Tanais, see Latyshev 1885– 1901, no. 2.447, l. 17. For discussions of the name, see Abaev 1949, 190; Chaumont 1975, 107; Wheeler 1977, 85; Braund 1994, 215. Another name from the monolingual inscription bearing a possible East Iranian etymology, Asparug (Gr. Aspauroukis), appears in a Greek form on a signet ring found during excavations at Mc‘xeta, where Asparug was named as the pitiaxēs (Apakidze et al. 1958, 29, fig. 4). See discussions of this name in Abaev 1949, 157–158; Duichev 1953; Schmitt 1985; Rapp 2014, 65, n. 162; Schottky 2016, 215–216. Following Schmitt and Schottky, this individual should not be connected with the later king Asp‘agur. 247 Ramishvili and Dzhorbenadze 1976. See also Braund 1994, 247; Balakhvantsev and Nikolaishvili 2010; Perevalov 2003. 248 For Roman names as a sign of political afiliation among the Caucasian elite, see, e.g., Braund 1993; Linderski 2007, 267, and also Linderski 2007, 273–276. 249 For surveys of tamgas in steppe contexts, see Iatsenko 2001; Drachuk 1975; Kuznetsov 2007; Solomonik 1959. For more recent overviews, see Kozlovskaya and Ilyashenko 2018; Voroniatov 2009; Muratov 2017, 187–190. 250 Kozlovskaya and Ilyashenko 2018, 172–173. 251 For assessments that see the marks as tamgas, see Preud’homme 2019, 1; Altheim and Stiehl 1963, 250; Wheeler 1977, 82–84. These symbols have, however, been read in different ways, with early interpretation suggesting their connection to an early Georgian alphabet, Apakidze et al. 1958, 72. Multifaceted Highlands: The Economy of Armenia and its South Caucasus Context 493 that these are in fact tamgas, the association is supported by the presence of unambiguous tamgas below the text on a lesser-known inscription found in western Azerbaijan, bearing a Greek funerary epitaph.252 Here, there is no doubt about the nature of the symbols; they find precise parallels in the North Caucasus.253 This stele had been reused in a later grave, just like the Armazis-q‛evi example. Tamgas have also been identified on a range small finds from across the northern South Caucasus: on ceramics from the eastern Caucasus;254 and accompanying an inscription in Greek on a silver vessel from Mc‘xeta.255 The association of these tamgas with Greek or Armazian inscriptions mirrors the context of some tamgas in the North Pontic.256 Such an association suggests that one function of the tamgas may have been to increase the legibility of the inscriptions in the South Caucasus. VII.. Historical and Ethnographic Accounts Beyond the material evidence, textual accounts from the classical tradition furnish evidence about the social and military frameworks in the region, and in this context for the relationship between the North and South Caucasus. Despite the many missing or muddled details in these accounts, they provide important evidence for social networks in the space. One clear example of this comes from the question of how ‘nomadism’ is handled by the ethnographies. I have already cited the programmatic passage of Strabo, which draws a clear difference between the South Caucasus polities and their nomadic northern neighbors. But this dichotomy dissolves when Strabo considers ethnic or cultural dimensions of society, particularly in the northern reaches of the South Caucasus. For example, noting the many varied peoples in the mountains above the Black Sea, Strabo says, “the greater part of them are Sarmatians (Sarmatae), but they are all Caucasian (Caucasii).”257 Moving to the east, he describes the highland inhabitants of K‘art‘li and Albania as kinsmen of their Sarmatian neighbors to the north.258 And This argument echoes the recurrent association between tamgas and various other alphabets and protoscripts: see discussion in Manassero 2013. 252 Trever 1959, 340–341, pl. 36. 253 Iatsenko 2001, 76. 254 Iatsenko 2001, 76 n. 19. 255 Two possible tamgas appear at the center of a silver plate bearing an inscription marking its transfer from a king Flavius Dades to a pitiaxēs Bersouma; Apakidze et al. 1958, 60–63, pl. LIV. On the identity of Flavius Dades, there has been disagreement, see Balakhvantsev 2005; Braund 1993; Linderski 2007; Melikishvili 1959, 56–58. For a recent argument for a late (fourth century) date, see Coert and Schmitt 2019. The Greek text below Bersouma, reading MAKEDONI, also deserves note, Linderski 2007, 270. 256 Kozlovskaya and Ilyashenko 2018, 177–180 discuss tamgas found with inscriptions. 257 Strabo 11. 2. 16, trans. H. L. Jones 1917–1932. 258 Strabo 11. 3. 3; 11. 4. 5. 494 Lara Fabian these genetic ties were not merely curiosities, but instead provide the justification for military collaboration between communities, particularly when faced with outside pressures.259 Beyond Strabo, the classical corpus offers descriptions of Alan/Sarmatian military actions unfolding in the South Caucasus, which are often characterized by a combination of close collaboration and fierce infighting. In particular, there were three episodes of North Caucasian forces moving into the South Caucasus in the first two centuries , one in 35 , a second in ca. 72 , and a third in 135 ,260 echoes of which can also be found in the Georgian tradition.261 To see this, it is worth spending a moment on the earliest episode of 35 . Accounts of this episode are preserved in Tacitus and Josephus, describing the chaotic aftermath of the death of the king of Armenia, Artašēs III, in 34 . The accounts agree that the Roman emperor Tiberius, looking to slow Arsakid expansion, wanted to set proxy fighters against the Arsakid interests in Armenia. He gained local support for this plan through the offer of enticements to various parties. In Josephus’s less detailed version, bribes are offered to the kings of both the K‘art‘velians and the Albanians. Although the two monarchs refuse to fight themselves, they open the “Caspian Gates” (passes through the Greater Caucasus mountains), allowing the “Alans” to pass through their territory and fight on side of the Romans. Tacitus’ account, in contrast, describes a more complicated political situation. Here, Tiberius convinces the K‘art‘velian king, P‘arasmanes, to go out and rally support to fight against the Arsakids. P‘arasmanes then brings both the Albanians and the nomads (described here as “Sarmatians”) to his side and launches an attack on the Arsakid positions, again facilitated by K‘art‘velian control of key routes south. Tacitus notes, however, that not all of the Sarmatians were allied with P‘arasmanes, as the group in fact took bribes from multiple sides. The dynamic nature of alliances that one can glimpse in this final historical account offers a bit of insight into how complicated the intraregional relationships must have been in this space, and how much energy it must have taken to manage these relationships. Thus, when we add the northern factor back in to the story, we find another axis of political negotiation in which Armenian dynasts would have been involved. There was not a sharp, clear dividing line between the various groups that constituted local societies – it was instead a more intertwined story of collaboration and conflict. If we combine this observation with the role of northward trade activity discussed above, it becomes one with economic relevance. In this case, however, evidentiary biases make it temptingly easy to ignore this story, despite its significance. The intensely 259 Strabo 11. 4. 5. 260 35 : Tac. Ann. 6. 33–35; Joseph. AJ 18. 96–97. Ca. 72 : Joseph. BJ 7. 244–251. 135 : Cass. Dio 69. 15. 1–3. 261 Although it is difficult to relate this episode to a specific period: KC, ed. S. Qauxč‛išvili 1955, 33– 54, also Toumanoff 1969, 2–3, 12–13. Multifaceted Highlands: The Economy of Armenia and its South Caucasus Context 495 local and ephemeral nature of these relationships has rendered them largely invisible in the classical authors, who present a picture on a rather more global scale; but also in the archaeological record, where a host of issues concerning both the categorization and the interpretation of available data hampers the development of a holistic understanding. The dual observation – of the importance of these connections, and of their near invisibility in the historical and archaeological record – must give us pause as we think about the project of telling the story of a connected antiquity. VIII Conclusion I began this chapter with a discussion about the multiple scholarly perspectives that can be brought to bear on the question of Armenia in the period under discussion: Was it a ‘neutral’ buffer state? Or was it a central space of trade? I have taken the view that what sits under both of these interpretations is, fundamentally, a story of the multiple axes of connectivity that can develop new types of infrastructure – and then create meaning in complex, new ways – particularly in frontier spaces such as Armenia. The word ‘frontier’ is of course problematic: The ancient residents of this space are unlikely to have understood their homeland as a frontier – they would rather have seen it as the center of their world. However, following a theoretical approach developed in the course of this project that articulates a model of the frontier as heuristic, one finds that there is much benefit to be drawn from this framing.262 The heuristic understanding of frontiers calls attention to issues of dynamic, relational interactions in spaces where difference meets. These are spaces of opportunity, but with that opportunity comes risk and often danger. In the case of Armenia, local authorities and residents more broadly had to contend with the intersection of overlapping frontiers that converged in their area. In this case, that situation led to the development of a type of flexibility that local residents and especially political authorities leveraged to cope with the changes the buffeted their homeland. When viewed this through these dual facets of frontier-ness and its connective potential, Armenia’s place within Eurasian systems becomes clearer. Armenian actors were extraordinarily well networked within their local systems, maintaining diverse and wide-ranging ties with neighbors that furthered political, social, and economic goals. Over the course of the Hellenistic period, this led to the development of a variety of infrastructural developments that fostered economic ties joining Armenia into broader networks. And yet, despite the profundity of these interactions, and despite the wider taste-setting that developed within this environment and which brought Armenia into supraregional systems, it also maintained its uniqueness in 262 Weaverdyck and Dwivedi forthcoming. 496 Lara Fabian ways that confound traditional approaches to imperial integration. This is most explicit in the social structure and particularly the naxarar system that it developed over the course of time, but it also applies to the question of urbanism and even religion. Rather than seeing this as an exception, it is better to understand this sort of particularism as one of the benefits of life along imperial interstices, where there was a plurality of choice. Life in such a space, moreover, created unique opportunities for trade activity, both in terms of the products that moved and the routes along which they moved. In this way, the frontier context of Armenia and the story of its economic connectivities are so deeply entangled that one cannot be understood without the other, and they provide a powerful counterpoint to the more traditional narratives of economic development through integration that are pervasive across studies of ancient Eurasia. References Abaev, V. I. 1949. Осетинский язык и фольклор [Ossetian language and folklore]. Vol. 1. Мосвка: Издательство АН СССР. Adonts, N. 1970. Armenia in the period of Justinian: The political conditions based on the Naxarar system. N. G. Garsoian (trans.). Lisbon: Calouste Gulbenkian Foundation. Altheim, F. and R. Stiehl. 1963. Die aramäische Sprache unter den Achaimeniden. Vol. 1. Frankfurt: Klostermann. Anderson, J. K. 1961. Ancient Greek horsemanship. Berkeley, CA: University of California Press. Apakidze, A. M., G. F. Gobedzhishvili, A. N. Kalandadze, and G. A. Lomtatitdze. 1958. Мцхета. Итоги археологических исследований. Археологические памятники Армазис-хеви по раскопкам 1937–1946 гг. [Mtskheta. Results of archaeological research. Archaeological monuments of Armzis-Khevi excavated in 1937–1946]. Тбилиси: Издательство Академии Наук Грузинской ССР. Apakidze, G. 1973. “ʩʨʪʰʴʬʯʰʹ ʱʲʫʬʱʨʸʰʹ ʹʨʸʱʵʼʨʪʰ” [A sarcophagus of Baginet‘i, Kldikari] ˃ʬʪʲʰʹ ʳʬʪʵʩʨʸʰ 33, 59–65. Arakelyan, B. N. 1968. “Excavations at Garni, 1949–50.” In H. Field (ed.), Contributions to the archaeology of Armenia. Cambridge, MA: Peabody Museum. – [Arakelian]. 1982. Арташат I. Основные результаты раскопок, 1970–1977 гг. [Artašat I: The main results of excavations, 1970–1977]. Ереван: Издательство АН АрмССР. Babaev, I. A. 1966. “Глиняная булла, найденная в Мингечауре” [A clay bulla found at Mingəçevir] In Azərbaycan tarixi məsələləri: Aspirant məqalələri külliyyatı, 5–10. Bakı: Azərbaycan SSR Elmlər Akademiyası nəşriyyatı. Babaev, I. A., I. Gagoshidze, and F. S. Knauß. 2007. “An Achaemenid «palace» at Qarajamirli (Azerbaijan). Preliminary report on the excavations in 2006.” Ancient Civilizations from Scythia to Siberia 13.1, 31– 45. Babaev, I. A. and S. M. Kaziev. 1971. “Кабалинскии клад монет эллинистической эпохи” [The Gabala hoard of coins of the Hellenistic Era]. Нумизматика и Эпиграфика IX, 16–32. Bais, M. 2001. Albania caucasica: Ethnos, storia, territorio attraverso le fonti greche, latine e armene. Milan: Mimesis. Balakhvantsev, A. S. 2005. “Царь Флавий Дад” [King Flavius Dades]. Нумизматика и эпиграфика 17, 44–55. Multifaceted Highlands: The Economy of Armenia and its South Caucasus Context 497 Balakhvantsev, A. S. and V. Nikolaishvili. 2010. “‘Алан Бакур’ из древней Иберии” [“Alan Bakour” from ancient Iberia]. Российская археология, no. 4, 130–135. Beck, R. 2000. “Ritual, myth, doctrine, and initiation in the mysteries of Mithras: New evidence from a cult vessel.” The Journal of Roman Studies 90, 145–180. Bedoukian, P. Z. 1964. “Gold forgeries of Tigranes the Great of Armenia.” Museum Notes (American Numismatic Society) 11, 303–306. –. 1968. “A classification of the coins of the Artaxiad dynasty of Armenia.” Museum Notes (American Numismatic Society) 14, 41–66. –. 1978. Coinage of the Artaxiads of Armenia. London: Royal Numismatic Society. –. 1987. Armenian coin hoards. Los Angeles, CA: Armenian Numismatic Society. Bedrosian, R. 1983. “The Sparapetut‘iwn in Armenia in the fourth and fifth centuries.” Armenian Review 36.2, 6–46. Bendschus, T. 2018. “Tigranes V. und seine Münzen. Armenische Herrscherlegitimation des frühen 1 Jhdts. n. Chr. im Spiegel der numismatischen Zeuginisse.” Numismatische Zeitschrift 124, 9–54. Bosworth, A. B. 1977. “Arrian and the Alani.” Harvard Studies in Classical Philology 81, 217–255. Braund, D. 1993. “King Flavius Dades.” Zeitschrift für Papyrologie und Epigraphik 96, 46–50. –. 1994. Georgia in antiquity: A history of Colchis and Transcaucasian Iberia, 550 BC–AD 562. Oxford: Clarendon Press. –. 2002. “Indian traders at Phasis: Neglected texts on ancient Georgia.” In M. Faudot, A. Fraysse, and E. Geny (eds.), Pont Euxin et commerce: la genèse de la “route de la soie,” 287–295. Besançon: Presses universitaires franc-comtoises. –. 2003. “Notes from the Black Sea and Caucasus: Arrian, Phlegon and Flavian inscriptions.” Ancient Civilizations from Scythia to Siberia 9.3, 175–91. Bresson, A. and E. Fagan. 2022. The Greek inscription from Garni (Armenia) and King Tiridates the Great. Part I. Вестник древней истории n. 3, 597–622. Callieri, P. 2001. “La presunta via commerciale tra l’India e Roma attraverso l’Oxus e il Mar Caspio: nuovi dati di discussione.” Topoi. Orient-Occident 11.1, 537–546. Canali de Rossi, F. 2004. Iscrizioni dello estremo oriente greco: un repertorio. Habelt: Bonn. Canepa, M. P. 2018. Iranian expanse: Transforming royal identity through architecture, landscape, and the built environment, 550 BCE–642 CE. Oakland, CA: University of California Press. Chaumont, M.-L. 1975. “États vassaux dans l’Empire des premiers Sassanides.” Acta Iranica, no. 4, 89–156. –. 1976. “L’Arménie entre Rome et l’Iran, 1. De l’avènement d’Auguste à l’avènement de Dioclétien.” Aufstieg und Niedergang der römischen Welt. Part 2, vol. 9.1, 71–194. Berlin: De Gruyter. –. 1985. “A propos des premières interventions Parthes en Arménie et des circonstances de l’avènement de Tigrane le Grand.” Acta Antiqua Academiae Scientiarum Hungaricae 31, 13–25. –. 1986. “Armenia and Iran II. The pre-Islamic period.” Encyclopaedia Iranica. Vol. 2.4, 418–438. Clark, T. 2021. “Processing into dominance: Nero, the crowning of Tiridates I, and a new narrative of Rome’s supremacy in the east.” Journal of Ancient History 9.2, 269–296. Coert, J. and T. Schmitt. 2019. “Wer war Fl. Dades? Überlegungen zum Verständnis einer Inschrift aus dem kaukasischen Iberien.” In V. Cojocaru, L. Ruscu, T. Castelli, and A.-I. Pázsint (eds.), Advances in ancient Black Sea studies: Historiography, archaeology and religion, 347–390. Cluj-Napoca: Phoibos. Connors, C. 2017. “Babylonian stories and the ancient novel: Magi and the limits of empire in Iamblichus’ Babyloniaka.” In M. P. Futre Pinheiro, D. Konstan, and B. D. MacQueen (eds.), Cultural crossroads in the ancient novel, 39–50. Berlin: De Gruyter. Dąbrowa, E. 2018. “Arsacid dynastic marriages.” Electrum 25, 73–83. Dan, A. 2017. “The Sarmatians: Some thoughts on the historiographic invention of a west Iranian migration.” In F. Wiedemann, K. P. Hofmann, and H.-J. Gehrke (eds.), Vom Wandern der Völker: Migrationserzählungen in den Altertumswissenschaften, 97–134. Berlin: Edition Topoi. Dardeniz, G. 2015. “Was ancient Egypt the only supplier of natron? New research reveals major Anatolian deposits.” Anatolica 41, 191–202. 498 Lara Fabian De Jong, A. 2008. “Regional variation in Zoroastrianism: The case of the Parthians.” Bulletin of the Asia Institute 22, 17–27. Dignas, B. 2002. Economy of the sacred in Hellenistic and Roman Asia Minor. Oxford: Oxford University Press. Drachuk, V. S. 1975. Системы знаков Северного Причерноморья: тамгообразные знаки северопонтийской периферии антич. мира первых веков нашей эры [Systems of signs of the North Black Sea: Tamga-like symbols of the North Pontic periphery of the ancient world in the first centuries of our era]. Киев: Наук. думка. Duichev, I. 1953. “Имя Аспарух в новооткрытых надписях Грузии” [The name Asparukh on a newly discovered inscription of Georgia] Archiv Orientálni 21, 353–356. Dundua, G. F. and G. A. Lordkipanidze. 1983. Денежное обращение центральной Колхиды V–I вв. до н. э. (по материалам Ванского городища) [Monetary circulation of central Colchis 5th c.–1st c. BCE (based on material from Vani settlement)]. Тбилиси: Издательство Тбилиского Университетаю. Eraslan, S. 2015. “Tethys and Thalassa in mosaic art.” Art-Sanat Dergisi 4, 1–13. Eremian, S. T. 1953. “Развитие городов и городской жизни в древней Армении.” [The development of cities and urban life in ancient Armenia]. Вестник древней истории, no. 3, 11–31. Fabian, L. 2019. “The meanings of coins in the ancient Caucasus.” Historische Anthropologie 27.1, 32–51. –. 2020. “Albania in Greek and Latin texts: The use and utility of ‘views from the west’.” In R. Hoyland (ed.), Caucasian Albania, 13–30. Piscataway, NJ: Gorgias Press. –. 2021. “Bridging the divide: Marriage politics across the Caucasus.” Electrum 28, 221–244. Fishman, S. G. 2016. “The space between: Ceramic production and exchange in Roman-Parthian period Oğlanqala, Azerbaijan.” Journal of Archaeological Science: Reports 9, 12–24. Foss, C. 1986. “The coinage of Tigranes the Great: Problems, suggestions and a new find.” The Numismatic Chronicle 146, 19–66. Franklin, K. 2021. Everyday cosmopolitanisms: Living the Silk Road in medieval Armenia. Oakland, CA: University of California Press. Furtwängler, A. E., I. Gagoshidze, H. Löhr, and N. Ludwig, eds. 2008. Iberia and Rome: The excavations of the palace at Dedoplis Gora and the Roman influence in the Caucasian kingdom of Iberia. Langenweissbach: Beier & Beran. Gagoshidze, I. 1992. “The temples at Dedoplis Mindori.” East & West 42.1, 27–48. –. 2001. “A royal palace in first-century Iberia, according to the archaeological material from Dedoplis Gora, Georgia.” In I. Nielsen (ed.), The royal palace institution in the first millennium BC: Regional development and cultural interchange between East and West, 259–279. Aarhus: Aarhus University Press. –. 2008. “Kartli in Hellenistic and Roman times. General aspects.” In Furtwängler, Gagoshidze, Löhr, and Ludwig (2008), 1–40. Garsoïan, N. G. 1976. “Prolegomena to a study of the Iranian aspects of Arsacid Armenia.” Handês Amsorya 90, cols. 177–234. –. 1984–1985. “The early-medieval Armenian city: An alien element?” Journal of the Ancient Near Eastern Society 16–17, 67–83. –. 1989. The epic histories. Attributed to P’awstos Buzand (Buzandaran Patmut’iwnk). Cambridge, MA: Harvard University Press. –. 1997. The emergence of Armenia. R. G. Hovannisian (ed.). New York, NY: St. Martin’s Press. Gazzano, F. 2016. “Croesus’ story in the history of Armenia of Movsēs Xorenac‘i.” In Gazzano, Pagani, and Traina (2016), 83–114. Gazzano, F., L. Pagani, and G. Traina, eds. 2016. Greek texts and Armenian traditions: An interdisciplinary approach. Berlin: De Gruyter. Gyulamiryan, H., S. Muradyan, M. Zardaryan, A. Lichtenberger, and T. Schreiber. 2021. “The ArmenianGerman archaeological project: Results from the excavations in Artaxata 2018–2021.” Ɍʂʖɸʆɸʍ ɸʎʄɸʖʏʙʀʌʏʙʍʍɼʗ 24, 5–18. Multifaceted Highlands: The Economy of Armenia and its South Caucasus Context 499 Halfmann, H. 1986. “Die Alanen und die römische Ostpolitik unter Vespasian.” Epigraphica Anatolica 8, 39–50. Hammer, E. L. 2014. “Highland fortress-polities and their settlement systems in the southern Caucasus.” Antiquity 88.341, 757–774. Hammer, E. L. and B. S. Arbuckle. 2017. “10,000 years of pastoralism in Anatolia: A review of evidence for variability in pastoral lifeways.” Nomadic Peoples 21.2, 214–267. Hewsen, R. H. 1986. “Armavir.” Encyclopaedia Iranica. Vol. 2.4, 415–416 –. 1988–1989. “Introduction to Armenian historical geography IV: The Vitaxates of Arsacid Armenia. A reexamination of the territorial aspects of the institution (part one).” Revue des Études Arméniennes 21, 271–319. –. 1990–1991. “Introduction to Armenian historical geography IV: The vitaxates of Arsacid Armenia. A reexamination of the territorial aspects of the institution (part two).” Revue des Études Arméniennes 22, 147–183. Honeychurch, W. 2014. “Alternative complexities: The archaeology of pastoral nomadic states.” Journal of Archaeological Research 22.4, 277–326. Iatsenko, S. A. 2001. Знаки-тамги ираноязычных народов древности и раннего средневековья [Symbol-tamagas of the Iranian-speaking peoples of antiquity and the early middle ages]. Москва: Восточная литература. Invernizzi, A., ed. 1998. Ai piedi dell’Ararat: Artaxata e l’Armenia ellenistico-romana. Florence: Le lettere. Jacobs, B. 2000. “Achaimenidenherrschaft in der Kaukasus-Region und in Cis-Kaukasien.” Archäologische Mitteilungen aus Iran und Turan 32, 93–102. Jones, H. L. 1917–1932. Strabo’s Geography. Cambridge, MA: Harvard University Press. Jong, A. de. 2013. “Hatra and the Parthian commonwealth.” In L. Dirven (ed.), Hatra. Politics, culture and religion between Parthia and Rome, 143–160. Stuttgart: Franz Steiner. –. 2015. “Armenian and Georgian Zoroastrianism.” In M. Stausberg and Y. S.-D. Vevaina (eds.), The Wiley Blackwell companion to Zoroastrianism, 119–129. Malden, MA: Wiley-Blackwell. Kéfélian, A. 2016. “The Roman army and the transmission of Latin loan words in Old Armenian.” In Gazzano, Pagani, and Traina (2016), 143–162. Khachatryan [Khachatrian], Z. 1996. “The archives of sealings found at Artashat (Artaxata).” In M.-F. Boussac and A. Invernizzi (eds.), Archives et sceaux du monde hellénistique. BCH Supplement, Vol. 29, 365–370. –. 2005. “Archaeological research in Artaxata. Preliminary report 2003–2004.” Parthica 7, 19–28. –. 2007. “ɭɸʉɸʛɸʋɸʌʗ Ɋʗʖɸʎɸʖɿ ʑɸʖʋɸʇʍɸɺʂʖɸʆɸʍ ʇɼʖɸɽʏʖʏʙʀʌʏʙʍʍɼʗʂ ʃʏʙʌʔʂ ʍɼʗʛʏ” [City of Artašat in the light of historical and archaeological research]. əɸʋɸʗ, no. 2, 3–36. –. 2010. “əʂʍ əɸʌɸʔʖɸʍʂ ʋɸʌʗɸʛɸʉɸʛ Ɋʗʖɸʎɸʖɿ ʎɸʗʏʙʍɸʆʏʙʋ ɾ ɸʍɸʆʍʆɸʃʍɼʗ ʋɸʖʏʙʘɼʃ” [Artašat, the ancient capital of Armenia, continues to offer surprises]. ɐʒʋʂɸʅʂʍ QR ɓɋ, 36–53. Khatchadourian, L. 2007. “Unforgettable landscapes: Attachments to the past in Hellenistic Armenia.” In N. Yoffee (ed.), Negotiating the past in the past: Identity, memory, and landscape in archaeological research, 43–75. Tucson, AZ: University of Arizona Press. –. 2008. “Making nations from the ground up: Traditions of classical archaeology in the South Caucasus.” American Journal of Archaeology 112.2, 247–278. –. 2016. Imperial matter: Ancient Persia and the archaeology of empires. Oakland, CA: University of California Press. K’imšiašvili, K. and G. Narimanišvili. 1995. “A group of Iberian fire-temples (4th cent. BC–2nd cent. AD).” Archäologische Mitteilungen aus Iran 28, 309–318. Knauß, F. 2000. “Der Palast von Gumbati und die Rolle der Achaimeniden im transkaukasischen Iberien.” Archäologische Mitteilungen aus Iran und Turan 32, 119–130. –. 2006. “Ancient Persia and the Caucasus.” Iranica antiqua 41, 79–118. Knauß, F., I. Gagoshidze, and I. A. Babaev. 2013. “Karačamirli: Ein persisches Paradies.” ARTA: Achaemenid Research on Texts and Archaeology 4, 1–29. 500 Lara Fabian Kovacs, F. L. 2016. Armenian coinage in the classical period. Lancaster, PA: Classical Numismatic Group. Kozlovskaya, V. and S. I. Ilyashenko. 2018. “Tamgas and tamga-like signs from Tanais.” In W. E. Keil, S. Kiyanrad, C. Theis, and L. Willer (eds.), Zeichentragende Artefakte im sakralen Raum. Zwischen Präsenz und UnSichtbarkeit, 167–198. Berlin: De Gruyter. Kunze, R., A. Bobokhyan, K. Meliksetian, E. Pernicka, and D. Wolf. 2011. “Archäologische Untersuchungen zur Umgebung der Goldgruben in Armenien mit Schwerpunkt Sotk, Provinz Gegharkunik.” In H. Meller and P. Avetissyan (eds.), Archaeology in Armenia, Ergebnise der Kooperationsprojekte 2010 – Ein Vorbericht, 17–49. Halle: Landesamt für Denkmalpflege und Archäologie Sachsen-Anhalt. Kuznetsov, V. D. 2007. “Тамга Савромата II из Фанагории” [A tamga of Savromat II from Phanagoria]. Древности Боспора 11, 225–234. Kvavadze, E. and I. Gagoshidze. 2008. “Fibres of silk, cotton and flax in a weaving workshop from the first century A.D. palace of Dedoplis Gora, Georgia.” Vegetation History and Archaeobotany 17, 211– 215. Lang, D. M. 1983. “Iran, Armenia and Georgia.” In E. Yarshater (ed.), The Cambridge history of Iran: Vol. 3.1, 505–536. Cambridge: Cambridge University Press. Latyshev, V. V., ed. 1885–1901. Inscriptiones antiquae orae septentrionalis Ponti Euxini graecae et latinae. 4 vols. Saint Petersburg: Tipografia Imperatorskoi akademii nauk. Lichtenberger, A., M. H. Zardaryan, and T. Schreiber. 2021. “Failed Roman imperialism. An unfinished Roman aqueduct at Artaxata in Armenia.” Archäologischer Anzeiger no. 1, https://doi.org/10.34780/ 8f82-fyw2. Linderski, J. 2007. “How did King Flavius Dades and Pitiaxes Publicius Agrippa acquire their Roman names?” In J. Lindersky (ed.), Roman Questions II, 262–276. Stuttgart: Franz Steiner. Lindsay, I. and A. Greene. 2013. “Sovereignty, mobility, and political cartographies in Late Bronze Age southern Caucasia.” Journal of Anthropological Archaeology 32.4, 691–712. Luther, A. 2018. “Zu den Dokumenten aus Avroman.” Gymnasium 125.2, 155–177. Luttwak, E. 1976. The grand strategy of the Roman Empire from the first century AD to the third. Baltimore, MA: Johns Hopkins University Press. Mahé, J.-P. 1996. “Le site arménien d’Armawir: d’Ourartou à l’époque hellénistique.” Comptes rendus des séances de l’Académie des Inscriptions et Belles-Lettres 140.4, 1279–1314. Manandyan [Manandian], H. A. 1954. О торговле и городах Армении в связи с мировой торговлей древних времен (в V. до н. э.–ХV в. н. э.) [The trade and cities of Armenia in relation to ancient world trade (from the 5th c. BCE to the 15th c. CE)]. Ереван: Издательство Ереванского университета. –. 1963. Tigrane II et Rome. H. Thorossian (trans.). Lisbon: Livraria Bertrand. –. 1965. The trade and cities of Armenia in relation to ancient world trade. N. G. Garsoïan (trans.). Lisbon: Livraria Bertrand. Manassero, N. 2013. “Tamgas and literacy among the ancient Iranians.” Re:marks, Journal of Signum 1, 43–61. Manoukian, H. 1996. “Les empreintes d’Artachate (Antique Artaxata).” In M.-F. Boussac and A. Invernizzi (eds.), Archives et sceaux du monde hellénistique. BCH Supplement, Vol. 29, 371–373. Maranci, C. 2018. The art of Armenia: An introduction. Oxford: Oxford University Press. Marciak, M. 2017. “The site of Arzan.” Ancient West & East 16, 305–324. Marciak, M. and R. S. Wójcikowski. 2016. “Images of kings of Adiabene: Numismatic and sculptural evidence.” Iraq 78, 79–101. Mari, F. 2016. “Cyrus the Great in Movsēs Xorenac‘i, Patmut‘iwn Hayoc‘: Telescoping the king.” In Gazzano, Pagani, and Traina (2016), 115–142. Masson, V. M. 1997. “Кавказский путь к цивилизации: вопросы социокультурной интерпретации” [The Caucasian road to civilization: Issues of socio-cultural interpretation]. In Древние общества Кавказа в эпоху палеометалла (ранние комплексные общества и вопросы культурной трансформации), 124–134. Санкт-Петербур: Институт истории материальной культуры РАН. Multifaceted Highlands: The Economy of Armenia and its South Caucasus Context 501 Meißner, B. 2000. “A belated nation: Sources on ancient Iberia and Iberian kingship.” Archäologische Mitteilungen aus Iran und Turan 32, 177–206. Melikishvili, G. A. 1959. К истории древней Грузии [On the history of ancient Georgia]. Тбилиси: Издательство Академии наук Грузинской ССР. Mordvintseva, V. 2013a. “Исторические сарматы и сарматская археологическая культура в Северном Причерноморье” [The history of the Sarmatians and Sarmatian archaeological culture in the North Black Sea]. In I. N. Khapunov (ed.), Крым в сарматскую эпоху (II в. до н. э.–IV в. н. э.). Vol. 1, 14–43. Симферополь: Бахчисарай. –. 2013b. “The Sarmatians: The creation of archaeological evidence.” Oxford Journal of Archaeology 32.2, 203–219. –. 2015. “Сарматы, Сарматия и Северное Причерноморье” [Sarmatians, Sarmatia and the North Black Sea]. Вестник древней истории, no. 1, 109–135. Mousheghian, K., A. Mousheghian and G. Depeyrot. 2000. History and coin finds in Armenia: Antiquity. Wetteren: Moneta. Muratov, M. 2017. “The language(s) of images in the arts of the Bosporan Kingdom.” In V. Kozlovskaya (ed.), The Northern Black Sea in antiquity: Networks, connectivity, and cultural interactions, 169–191. Cambridge: Cambridge University Press. Nachmias, A., G. Bar-Oz, D. Nadel, I. Petrosyan, and B. Gasparyan. 2021. “A monumental horse burial in the Armenian highlands.” Archaeology, Ethnology & Anthropology of Eurasia 49.3, 41–50. Nercessian, Y. T. 2006. Silver coinage of the Artaxiad dynasty of Armenia. Los Angeles, CA: Armenian Numismatic Society. Nurpetlian, J. 2010. “Ancient Armenian coins: The Artaxiad Dynasty (189 BC–AD 6).” Berytus: Archaeological Studies 51, 117–167. Odišeli, M. D. 1995. Spätantike und frühchristliche Mosaike in Georgien. Vienna: Verlag der Österreichischen Akademie der Wissenschaften. Olbrycht, M. J. 2021. “The Battle of Tigranokerta, Lucullus, and Cataphracts.” Mnemosyne, 75.6, 946–981. Pakhomov, E. A. 1926. Монетные клады Азербайджана и Закавказья [Coin Hoards of Azerbaijan and Transcaucasia]. Vol. 1. Ваку: Издание Овщества Обследования и Изучения Азербайджана. –. 1938. Монетные Клады Азербайджана и других республик и краев Кавказа [Coin Hoards of Azerbaijan and other republics and regions of the Caucasus]. Vol. 2. Ваку: Издательство Азербайджанского филиала Акедемии наук СССР. Patterson, L. E. 2013. “Caracalla’s Armenia.” Syllecta Classica 24.1, 173–199. –. 2015. “Antony and Armenia.” Tapa, 77–105. Payaslian, S. 2007. The history of Armenia. New York, NY: Macmillan. Perevalov, S. M. 2003. “Бакур-алан из древней Иберии” [Bakur-Alan from ancient Iberia]. Nartamongæ 2.1–2, 47–56. Perihanân, A. G. 1971. “Les inscriptions Araméennes du roi Artachès:(A propos d’une récente trouvaille épigraphique en Arménie).” Revue des Études Arméniennes 8, 169–174. Petrosyan, H. 2021. “Politics, ideology and landscape: Early Christian Tigranakert in Artsakh.” Electrum 28, 163–187. Plontke-Lüning, A. 2009. “Iranische Feuertempel: Genese, Entwicklung und Funktion.” Ancient West & East 8, 203–221. Preud’homme, N. J. 2019. “La stèle des victoires du piṭaḥš Śargas et la réaffirmation de la domination royale en Ibérie du Caucase.” Camenulae 22, 1–20. Qauxč‛išvili, S., ed. 1955. ʽʨʸʯʲʰʹ ˂ˆʵʭʸʬʩʨ [K‘art‘lis c‘xovreba]. ʯʩʰʲʰʹʰ: ʹʨʩˁʵʯʨ ʹʨʽʨʸʯʭʬʲʵ. Qauxč‛išvili, T. 1999–2000. ʹʨʽʨʸʯʭʬʲʵʹ ʩʬʸ˃ʴʻʲʰ ˄ʨʸ˄ʬʸʬʩʰʹ ʱʵʸʶʻʹʰ [Corpus of Greek Inscriptions]. 3 vols. ʯʩʰʲʰʹʰ: ʲʵʪʵʹʰ. Ramishvili, R. M. and V. A. Dzhorbenadze. 1976. “Археологические исследования в зоне строительства Жинвальского гидротехнического комплекса” [Archaeological research in the construction zone 502 Lara Fabian of the Zhinvali hydraulic complex]. In Археологические исследования на новостройках Грузинской ССР, 30–44. Тбилиси: Мецниереба. Rapp, S. H. 2009. “The Iranian heritage of Georgia: Breathing new life into the pre-Bagratid historiographical tradition.” Iranica Antiqua 44, 645–692. –. 2014. The Sasanian world through Georgian eyes: Caucasia and the Iranian commonwealth in Late Antique Georgian literature. Farnham: Ashgate. Rckavandi, H. O., E. W. Sauer, T. Wilkinson, G. A. Abbasi, S. Priestman, E. S. Tamak, R. Ainslie, M. Mahmoudi, N. Galiatsatos, and K. Roustai. 2008. “Sasanian walls, hinterland fortresses and abandoned ancient irrigated landscapes: The 2007 season on the great wall of Gorgan and the wall of Tammishe.” Iran 46.1, 151–178. Rollinger, R. and A. Kellner. 2019. “Once more the Nabonidus Chronicle (BM 35382) and Cyrus’ campaign in 547 BC.” Ancient West & East 18, 153–176. Rtveladze, E. V. 2010. “The great Indian road: India-Central Asia-Transcaucasia.” Anabasis. Studia Classica et Orientalia 1, 80–96. Russell, J. R. 1987. Zoroastrism in Armenia. Cambridge, MA: Harvard University Press. Santrot, J. and R. S. Badalian, eds. 1996. Arménie: trésors de l’Arménie ancienne: des origines au IVe siècle. Paris: Somogy and Musée Dobrée. Saprykin, S. Ju. 2007. “The unification of Pontos. The bronze coins of Mithridates VI Eupator as evidence for commerce in the Euxine.” In V. Gabrielsen and J. Lund (eds.), The Black Sea in antiquity. Regional and interregional economic exchanges, 195–208. Aarhus: Aarhus University Press. Scarborough, J. 2015. “Fish glue (Gr. ΙΧΘΥΟΚΟΛΛΑ) in Hellenistic and Roman medicine and pharmacology.” Classical Philology 110.1, 54–65. Schleicher, F. 2021. Iberia Caucasica. Ein Kleinkönigsreich im Spannungsfeld großer Imperien. Stuttgart: Kohlhammer. Schmitt, R. 1985. “Asparuch und Konsorten im Lichte der Iranischen Onomastik.” Linguistique balkanique 28, 13–38. Schmitt, R., M. L. Chaumont, J. R. Russell, M. V Esbroeck, and H. Papazian. 1986. “Armenia and Iran.” Encyclopaedia Iranica. Vol. 2.4–5, 417–478. Schneider, P. 2017. “From India to the Black Sea: An overlooked trade route?” hal-01376630v2 f. Schottky, M. 2012. “Vorarbeiten zu einer Königsliste Kaukasisch-Iberiens. 1. Anfänge der Pharnabaziden.” Anabasis. Studia Classica et Orientalia 3, 239–250. –. 2016. “Vorarbeiten zu einer Königsliste Kaukasisch-Iberiens. 5. Im Schatten Schapurs II.” Anabasis. Studia Classica et Orientalia 7, 204–221. Schreiber, T. 2021. “The archives of Artaxata: Archival practice in the capital of ancient Armenia.” Electrum 28, 277–310. Sherozia, M. 2008. “Monetary circulation in Iberia in the 1st century BC–1st Century AD” In Furtwängler, Gagoshidze, Löhr, and Ludwig (2008), 235–252. Sherwin-White, A. N. 1984. Roman foreign policy in the East, 168 BC to AD 1. London: Duckworth. Shnirelman, V. 2006. “The politics of a name: Between consolidation and separation in the northern Caucasus.” Acta Slavica Iaponica 23, 37–73. Shortland, A. J., P. Degryse, M. Walton, M. Geer, V. Lauwers, and L. Salou. 2011. “The evaporitic deposits of Lake Fazda (Wadi Natrun, Egypt) and their use in Roman glass production.” Archaeometry 53.5, 916–929. Shortland, A. J. and H. Schroeder. 2009. “Analysis of first millennium BC glass from vessels and beads from the Pichvnari necropolis, Georgia.” Archaeometry 51.6, 947–965. Simpson, S. J. 2015. “Sasanian glassware from Mesopotamia, Gilan, and the Caucasus.” Journal of Glass Studies, 77–96. Sinclair, T. A. 1995. “The site of Tigranocerta I.” Revue des Études Arméniennes 25, 183–253. –. 1996. “The site of Tigranocerta. II.” Revue des Études Arméniennes 26, 51–118. Solomonik, E. I. 1959. Сарматские знаки Северного Причерноморья [Sarmatian symbols of the northern Black Sea]. Киев: Издательство Академии Наук Украинской ССР. Multifaceted Highlands: The Economy of Armenia and its South Caucasus Context 503 Speidel, M. 2021. “Provincia Armenia in the light of the epigraphic evidence.” Electrum 28, 135–150. Sullivan, R. D. 1990. Near Eastern royalty and Rome, 100–30 BC. Toronto: University of Toronto Press. Sundermann, W. 1989. “Bidaxš.” Encyclopaedia Iranica 4.3, 242–244. Tarn, W. W. 1901. “Patrocles and the Oxo-Caspian trade route.” The Journal of Hellenic Studies 21, 10–29. Ter-Martirossov, F. 2001. “The typology of the columnar structures of Armenia in the Achaemenid period.” In I. Nielsen (ed.), The royal palace institution in the first millennium BC: Regional development and cultural interchange between East and West, 155–163. Aarhus: Aarhus University Press. Thompson, M., O. Mørkholm, and C. M. Kraay. 1973. An inventory of Greek coin hoards. New York, NY: American Numismatic Society. Thomson, R. W. 1978. Moses Khorenats’i: History of the Armenians. Cambridge, MA: Harvard University Press. –. 1996. Rewriting Caucasian history. The medieval Armenian adaptation of the Georgian chronicles. The original Georgian chronicles and the Armenian adaptation. Oxford: Oxford University Press. –. 2004. “Armenian ideology and the Persians.” In G. Gnoli (ed.), Convegno internazionale La Persia e Bisanzio: (Roma, 14–18 ottobre 2002), 373–389. Rome: Accademia Nazionale dei Lincei. –. 2014. “The major works of Armenian historiography (classical and medieval).” In V. Calzolari und M. E. Stone (eds.), Armenian philology in the modern era. From manuscript to digital text, 303–320. Leiden: Brill. Tiratsyan, G. A. 1979. “Armenian towns of Hellenistic times in the light of archaeological research.” In R. Vardanyan (ed.), S. Mardanyan, G. Muradyan, and A. T’op’chyan (trans.), From Urartu to Armenia: Florilegium Gevork A. Tiratsyan in memoriam, 127–138. Neuchâtel: Recherches et publications. Tonikyan, A. V. 1992. “The layout of Artashat and its historical development.” Mesopotamia 27, 161–187. Toumanoff, C. 1963. Studies in Christian Caucasian history. Washington, D.C.: Georgetown University Press. –. 1969. “Chronology of the early kings of Iberia.” Traditio 25, 1–33. Traina, G. 2016. “Traditions on Armenia in submerged Greek literature: Preliminary considerations.” In G. Colesanti and L. Lulli (eds.), Submerged literature in ancient Greek culture. Vol. 2, 111–123. Berlin: De Gruyter. –. 2017. “Strabo and the history of Armenia.” In D. Dueck (ed.), The Routledge companion to Strabo, 93–101. London: Routledge. –. 2019. “Dynastic connections in Armenia and Iberia: Further reflections on the Greek inscriptions from Bagineti (SEG 52, no. 1509–1510).” In U. Hartmann, T. Stickler, and F. Schleicher (eds.), Iberien zwischen Rom und Iran: Beiträge zur Geschichte und Kultur Transkaukasiens in der Antike, 123–128. Stuttgart: Franz Steiner. –. 2021a. “Ancient Armenia: Evidence and models.” Electrum 28, 13–20. –. 2021b. “Armenia and the ‘Orontid connection.’ Some remarks on Strabo, Geography 11, 14, 15.” In M. Blömer, S. Riedel, M. J. Versluys, and E. Winter (eds.), Common dwelling place of all the gods, 345–356. Stuttgart: Franz Steiner. Trever, K. V. 1953. Очерки по истории культуры древней Армении (II в. до н. э.–IV в. н. э.) [Studies on the history of the culture of ancient Armenia: 2nd century BCE–4th century CE]. Москва: Издательство Академии наук СССР. –. 1959. Очерки по истории и культуре Кавказской Албании IV в. до н. э.–VII в. н. э. [Studies on the history and culture of Caucasian Albania: 4th century BCE–7th century CE]. Москва: Издательство Академии наук СССР. Tsetskhladze, G. R. 1989. “Монеты Диоскуриады из Херсонеса таврического” [Coins of Diosurias from Tauric Chersonesos]. Вестник древней истории, no. 4, 91–95. –. 1992. “Greek colonization of the Eastern Black Sea littoral (Colchis).” Dialogues d’histoire ancienne 18.2, 223–258. –. 1993. “On the numismatics of Colchis: the classical archaeologist’s perspective.” Dialogues d’histoire ancienne 19.1, 233–256. 504 Lara Fabian Tuallagov, A. A. 2014. Аланы Придарьялья и закавказские походы I–II вв. [The Alan pre-Darial and Transcaspian campaigns of the 1st and 2nd centuries]. Владикавказ: ИПЦ СОИГСИ ВНЦ РАН и РСОА, 2014. Vardumian, G. D. 1991. “Дохристианские культы армян” [The pre-Christian cults of the Armenians]. Армянская этнография и фольклор (материалы и исследования) 18, 60–146. Vargyas, P. 2008. “The silver hoard from Nush-i Jan revisited.” Iranica antiqua 43, 167–183. Versluys, M. J. 2017. Visual style and constructing identity in the Hellenistic world: Nemrud Dağ and Commagene under Antiochos I. Cambridge: Cambridge University Press. Vinogradov, J. G. 1990. “Côte septentrionale du Pont, Caucase, Asie centrale.” Bulletin epigraphique, Revue des Études Grecques, 103, 531–560. Voroniatov, S. V. 2009. “О функции сарматских тамг на сосудах” [On the function of Sarmatian tamagas on vessels]. In Гунны, готы и сарматы между Волгой и Дунаем, 80–98. СанктПетербург. Wages, S. M. 1986. “A note on the Dumbarton Oak ‘Tethys mosaic’.” Dumbarton Oaks Papers 40, 119–128. Weaverdyck, E. J. S. and M. Dwivedi. Forthcoming. “Introduction.” In L. Fabian, K. Leese-Messing, E. J. S. Weaverdyck, L. Morris, and M. Dwivedi (eds.), Economies of the edge: Frontier-zone processes at regional, imperial, and global scales (300 BCE–300 CE). Heidelberg: HeiUP. Wheeler, E. L. 1977. “Flavius Arrianus: A political and military biography.” PhD diss., Duke University. Wilkinson, R. D. 1982. “A fresh look at the Ionic building at Garni.” Revue des Études Arméniennes 16, 221–244. Wolf, D. and R. Kunze. 2013. “Gegharkunik–Neue Quellen für altes Gold aus Südkaukasien.” In H. Meller, R. Risch, and E. Pernicka (eds.), Metalle der Macht – Frühes Gold und Silber, 111–140. Halle: Landesamt für Denkmalpflege und Archäologie Sachsen-Anhalt. Yao, A. 2012. “Sarmatian mirrors and Han ingots (100 BC–AD 100): How the foreign became local and vice versa.” Cambridge Archaeological Journal 22.1, 57–70. Ziegler, K.-H. 1964. Die Beziehungen zwischen Rom und dem Partherreich: Ein Beitrag zur Geschichte des Völkerrechts. Stuttgart: Franz Steiner. Zimansky, P. E. 1985. Ecology and empire: The structure of the Urartian state. Chicago, IL: University of Chicago Press. Map 1: Armenia and the South Caucasus. © Peter Palm.