We do not know how many Greeks were still living in Bactria in the 3rd century BC, and the Greek realm as ‘the land of a thousand cities’ is moonshine.130 The reported numbers, and another reported figure for cities (‘twelve’), are suggestive but often conflicting. Very many Greeks (3,000 or 23,000? – sources are confused) had been killed in an uprising after Alexander’s death. Relations with those Macedonians who remained behind can only be guessed at, and there must have been much intermarriage with the local population for those who sought a life in the east, just as there must have been numerous ‘camp-followers’ of various races who had travelled with Alexander’s army. Many of the Greeks had probably served with the Persian army, not Alexander’s. The Greeks’ status in what were still Macedonian/Seleucid satrapies must have been very ill-defined, but they still owed allegiance to the Macedonian king whose behaviour had become more and more oriental. Nevertheless, however much the Greek elements in the east may be denied by some students of Seleucid history,131 they were there with sufficient numbers and influence to create Greek states and urban cultures which were mainly non-Macedonian/Seleucid in appearance and behaviour, while Macedonians everywhere were more than ready, like Alexander, to ‘cash in’ on whatever service or even kudos Greekness might afford.132
Diodotos I, the Bactria/Sogdiana satrap, rebelled against Macedonian Antiochos II in the 250s, called himself ‘king’, and was succeeded by his son Diodotos II, who entered into some sort of pact with the Parthians. (Dates throughout this chapter will not be totally reliable but, within limits, are generally agreed by scholars.) ‘Kings’ were not much to Greek taste but Macedonian Seleucids had made them familiar. Diodotos II was killed by his satrap (perhaps of Margiana to the south) Euthydemos I in about 235 BC. Both Diodoti minted coins of purely Greek type.133 Euthydemos reigned until about 200 BC and extended his kingdom south into Aria and Margiana, and north into Ferghana. Around 208 BC he fought off Seleucid Antiochos III, having withstood a two-year siege in Bactra/Balkh. He moved even into Parthia. These were the kings who first established a Greco-Bactrian – becoming ‘Indo-Greek’ – realm in Asia, and they declared their Greekness with splendid coinage. We may continue to consider their political and imperial fortunes, and any continuing Macedonian presence and activity, before turning to their way of life, their cities and their arts.
Euthydemos sought to consolidate his northern frontiers against movement from north and east, while his son Demetrios I sought rather to expand to the south into India. He even occupied Pataliputra (Patna) for a time, far east along the Ganges, an event recorded also in Indian chronicles.134 The families and interrelationships of the Greco-Bactrian kings are ground for complicated scholarly discussion, and dependent a great deal on interpretation of their coinage, which we shall consider separately. Their family affairs are not pursued here in detail, but there was another king, Antimachos, who called himself god (theos), active in the south, who probably crossed the Hindu Kush and who even minted square coins of Indian type, while a Demetrios II minted bilingual coins. The latter moved farther southeast, into Gandhara (‘Gadara’ was a Persian satrapy), but was defeated by another northern satrap-turned-Greek king, Eukratides, whose rule began about 171 and lasted until around 155 BC. Other kings of about this time are the brothers Pantaleon and Agathokles, challenging Eukratides from the northeast. It was Agathokles who was the first Greek king to occupy the city of Taxila, whose fortunes are considered below. One thing that is quite clear from the confused record is that the Greek kings spent a major part of their resources and energy fighting or thwarting each other as much as consolidating a real Greek kingdom in the east. In this they were exhibiting that often self-defeating Greek pride and ambition which impeded them from ever becoming a ‘world power’ politically, as did the Persians, Macedonians and Romans, despite the model of Alexander the Great. And yet, within all this, Greek administrative practices could flourish and provide a secure social basis for life and trade – witness tax receipts, mute testimony to a tight administration.135
Their cities and arts
The Greeks, whether under Seleucid or new Greek rule, were described by a Chinese observer in the early 2nd century BC (see below) as living in walled towns, not especially warlike, but busy traders. That sounds quite authentic for Greeks as we know them. They, as had the Macedonians, profited from the extensive and good road system which had been established by the Persians throughout their empire – never a strong point for Greeks. Their towns could not quite be poleis in Greek terms, though they might strive to be. Greek buildings and practices could be readily supplied to enhance the Hellenicity – gymnasia, which were as much clubs as exercise grounds, libraries, theatres, stoas and appropriate temples. But oriental religions were strong and well-established, and they could accommodate a degree of Greekness more readily than the Greeks could the oriental. Moreover ‘Asian’ Greek traditions had had some time to develop for themselves. There were slaves, certainly, but many of other what we would regard as civilizing aspects of Greek life and government became apparent, as indeed they did also in the Seleucid satrapies farther west, for instance, and even in Babylon, where, however, Babylonian law was allowed to remain valid.
Physical evidence for the Greco-Bactrian kingdom(s) down to the later 2nd century BC, when the Greeks were being moved on south, is patchy (except for the coinage) but in places very rich.136 Their occupation of Bactra/Balkh itself is elusive on such a massive site, which is nevertheless often claimed as the source of various Greek objects. Some 20 miles (30 km) west is a major Persian site at Cheshm-e Shafa, where we might expect more Greek material. Far more important has been the excavation of a whole town founded originally under Seleucid rule but soon a major centre for the Greco-Bactrian ‘empire’, and lasting until the later 2nd century BC. It is to the east, on the Oxus, at Aï Khanoum (its medieval name, ‘Lady Moon’). The site was well excavated by the French in 1965–78. Much has been published though the site itself has been and is being much disturbed during the current troubles. It was very probably one of the new Alexandrias, just possibly the ‘Alexandria Oxiana’ mentioned in a text,137 and certainly an early Macedonian foundation, even if Seleucid rather than ‘Alexandrian’. There had been an Achaemenid Persian settlement not far away, as well as the major prehistoric site of Shortugai, which has yielded links with north India (Harappa), and was a market for lapis lazuli, the mineral for which this area of Afghanistan is the major source. There was also nearby a grand Hellenistic farmhouse complex, demonstrating a very busy market in corn. Aï Khanoum was an early foundation to be occupied by the Bactrian Greeks, and lasted until about 145 BC, when the Greeks were displaced by the ex-nomad Yuehzhi (on whom much more, below).138
The town lay in a good defensive position on the Oxus, at the point where it is joined by the Kokcha River from the southeast. There is a distinct flat-topped acropolis, 60 metres (200 ft) high, and a broad lower town along the Oxus, all substantially fortified and roughly triangular with mile-long sides. The character of the buildings and details of the architecture are essentially Greek but idiosyncratic in detail, the result of distance from contemporary Hellenistic architecture and town planning, and of a certain distinctiveness among the Bactrian Greeks, which appears too in other arts, here and elsewhere. The monumentality of the fortifications and palace attest the Macedonian taste, which had been carried east by Alexander and his subjects. The full excavation of Aï Khanoum and what it has taught us is an object lesson on how little we still know about the archaeological history of southern Central Asia, since a single site can so substantially adjust and expand our knowledge and expectations [40, PLS. XV, XVI].
The canonical buildings for a classical Greek polis are there. A temple, dedicated to the founder (Kineas) as it would have been in a Greek colonial foundation in the Mediterranean, a gymnasium, a theatre that could serve also as a place of assembly, a palace – since royal trappings are by now respectable among Greeks, at least so far away from home, although the palace plan owed most to eastern practice with series of courtyards. Monumental classical stone architecture with the usual orders (Doric, Ionic, Corinthian) was rather foreign to these parts but soon embraced. Neither house nor temple plans quite conform to the Greek model, though the courtyard remains an important element in the houses, even if not set centrally. The temples, although dedicated to Greek deities, follow a more eastern plan, with a central cult room and statue, with side chambers or niches, flat roofs and no exterior sculpture. The prominence of stepped platforms is decidedly Persian – indeed column bases taken from the nearby Achaemenid town were reused.139
40 The site of Aï Khanoum at the junction of the Oxus and Kokcha rivers.
Monumental sculpture is found too, as well as all the luxury crafts of metalwork, and the ‘minor arts’ so conspicuous in all Greek towns – clay figurines, decorative revetments for buildings. The relative lack of fine stone, the preferred marble, for sculpture, meant often using the acrolith technique met also in other peripheral Greek sites in the west, with clothed parts of figures carved in wood and only the flesh parts in stone. Stucco is much in evidence, as it was in Parthia, for sculpture and reliefs. Pebble mosaics were made, something of a novelty even in the west at that date. In most respects the quality and style live up to the best of the homeland despite the overall foreign aspect of the town and its setting, a character revealed mainly in details in the arts. There could be no mistaking Aï Khanoum for a replica of Mediterranean life, but equally its Greekness was abundantly apparent, and not least in the inscriptions. There are papyrus fragments with texts from a philosophical dialogue and a play;140 dedications in the gymnasium;141 receipts and contents lists painted on jars, including records of payments in Indian coins and in olive oil142 and ‘cinnamon’143, the last recalling the old dedication on Samos (above, p. 7); and an inscribed stone in the heroon of Kineas, given by a Klearchos to remind the far-flung Greeks of the Delphic maxims:144
‘as a child be well-behaved;
as a youth be self-controlled;
in middle age be just;
in old age be wise;
at death grieve not.’
Eukratides was no doubt the last ruler of Aï Khanoum, and he was a busy warrior to the south. A relic, perhaps, is a strange Indian disc composed of shell, embellished with gold and glass, depicting an Indian scene, and perhaps story.145 And no less memorable but not wholly Greek, a gilt silver disc [PL. XIII], apparently wrenched from furniture (perhaps loot from another site),146 depicting the Greek goddess Kybele, who was not without eastern connections, in her lion chariot driven by Victory, attended by long-cloaked priests, holding an umbrella and making offerings on a high platform in a rocky landscape overlooked by the sun, moon and a star. Yet, however Greek parts of this look, it also bears the stamp of a craftsman far from home and of a non-Greek setting. More wholly Greek is a small ivory roundel with the relief figures of Aphrodite seated, flowers and a Victory, a relic salvaged from the now robbed site.147
Nothing has as yet been found to match this urban site in the rest of the Greco-Bactrian homeland, but there are tantalizing finds which do more than hint at a Greek presence, quite apart from what we shall find of their heritage here and to the south.
Far to the west along the Oxus is Takht-I Sangin, a temple site sometimes known as the Temple of the Oxus.148 This is because there was found there a small bronze statuette of a Greek satyr [PL. XVIII], playing pipes, on a base that carried a dedication in Greek by Atrosokes (a Persian name) to ‘the Oxus’, and another apparent inscription naming an Oxus deity. This is perhaps enough to declare the whole complex an ‘Oxus Temple’. The parallel with the Anatolian river named after a satyr (Marsyas) is not cogent since in that case the inspiration was a local name and myth.149 However, an Achaemenid finger ring from the Oxus Treasure (see below) carries an inscription which has been read as Wacsh (‘Oxus’).150 At any rate, these are poignant associations of Greek, Persian and locality. The building itself is odd, more like an eastern (Persian) fire-temple.151 A large courtyard leads to a broad entrance hall (iwan), into a four-columned sanctuary, with many smaller rooms around. Architectural details are certainly hellenizing but perhaps more like the Parthian versions of the classical than the Greek.
Many finds certainly go back to the Greco-Bactrian period, with the expected odd mix of Greek and local, but most were recovered from levels that are certainly post-Greek. An interesting combination of styles is often shown. Thus, an ivory scabbard [PL. XIX] copies a Median/Persian form that goes back to the 7th century BC.152 The creature curled at the chape terminal and the lion are related to the Mesopotamian and Persian but the border of tongue and cable, and the very naturalistic stag, are Greek. On a sword handle  the subject, in a Greek style, is Herakles beating down an elderly opponent;153 and on a part of a scabbard [PL. XXII] a Greek-looking fish-lady holds an oar and a fruit (?) but is also winged and has horses’ forelegs154 – but for the human forepart she would pass as a Greek hippocamp, and might do as well as a personification of the Oxus as the satyr,155 but we shall find another candidate for that honour. A smaller ivory [PL. XXI] presents a Herakles head in a form not unlike that used for Alexander on coinage.156Along with these luxury items are pieces of realistic clay statuettes, near-portraiture and other Greek and Asian trivia. While the architecture and function of Takht-I Sangin must be ‘native’, much of its wealth seems to have depended on access to luxury objects of Greek type, and there is remarkably a considerable hoard of Greek weaponry. I save for last another ivory, the mouthpiece of a rhyton drinking cup in the shape of a lion [PL. XX], close kin to many that are Greek and especially to the style that appears on metal rhyta with ‘Greco-Parthian’ associations, which we have already looked at.157
41 Ivory sword handle from Takht-I Sangin showing Herakles beating down an elderly opponent. 3rd cent. BC. (After Oxus)
We have already mentioned the other ivory rhyta which deserve notice here for their Greekness – the hoard found in the Parthian capital Nisa.158 The total Greekness of their style but also the determinedly Greek subtlety of their iconography seems to betoken Greek hands and it is perfectly possible that the whole find is booty from a Greco-Bactrian source. They are decidedly luxury items, enhanced with gold, glass and semi-precious stones . The modelled foreparts of the Nisa ivory rhyta include eastern animal subjects derived from Persian art – horned griffins, graecized lions, and, of other animal subjects, elephants – while others are Greek in subject as in style. The acanthus collar on the stems of the rhyta is Hellenistic Greek. The rims are decorated with Greek floral patterns, including egg-and-dart, but often also rows of frontal heads which are rather an eastern and Parthian architectural and decorative feature.159 The relief friezes below them are purely Greek, and the abstruse character of some of their subjects suggests that they were created in an environment that was fully aware of the minutiae of Greek iconography and not mere copyist. Moreover, they were likely to be for a clientele that understood them. They include mythological scenes, a fine parade of Greek poetesses [43–44], rusticity ,160 and scenes of sacrifice in the Greek manner. One carries a dedication to Hestia, in Greek.161
42 Ivory rhyton from Nisa, the neck decorated with facing heads and a festive scene, the forepart in the form of a horned griffin. 3rd/2nd century BC. (After Masson/Pugachenkova)
43–45 Scenes on the necks of ivory rhyta from Nisa, showing Greek poetesses and a rustic scene beneath a vine frieze. 3rd/2nd century BC. (After Masson/Pugachenkova)
Another western site about which less is known is Dilberzin, once a Persian site, but its niched temple walls carry paintings of the heroic Greek Dioskouroi horsemen , and in Greek style, not closely datable and not certainly of the main Indo-Greek period, yet fully expressive of Greek style and subject.162
Aï Khanoum, Takht-I Sangin and Nisa are all excavated sites, carefully recorded and for the most part published. We can be sure that more will be found – thus, the Hellenistic walls of Samarkand have been identified, along with fortresses, established no doubt by Alexander.163 In this area (Sogdiana) north of the Oxus, Alexander’s generals established control, and this was a zone for Greek settlement, just as it was south of the Oxus.164 As sources for evidence of Greek arts and life in Bactria the few excavated sites are essential, but they are not the only sources. It is characteristic of a country like Afghanistan, and indeed its neighbours, that a majority of finds is going to be made accidentally, and that, given the nature of the geography, they cannot readily be controlled by authority (especially not when ‘authority’, however defined, is hostile to the religious character of the finds, a common case where ‘national’ heritage is, as so often, meaningless). Moreover, in a land of massive sites and long occupation it is in the nature of ancient behaviour that much material is collected in hoards, hidden for security, but never recovered in antiquity, though now a major source for the antiquity-hunter – and scholar. Lack of information about provenience and even more lack of certainty about composition of groups, since a dealer may readily enhance a hoard by adding to it from other sources, bedevil attempts to deal in a completely scholarly way with hoard material, but it would be unscholarly to ignore it. Our area has had more than its share of hoard finds over the last century or two, and they are important for our subject.
46 Painting from Dilberzin showing the Greek heroes, the Dioskouroi, with their horses. 1st century BC (?).
One in particular has been long known and impinges a little on our view of Greeks in Bactria. It is the so-called Oxus Treasure, now in the British Museum, until recently fully exhibited there, and said to be from a cache found buried in the banks of the Oxus in 1877. It was partially dispersed, then reassembled, an activity that is not reassuring if we are concerned about its unity. The objects are all precious and most are clearly Persian in origin, which does not mean that they come from Persia but that they were collected from a Persian site in Asia and deposited for safety.165 Most are earlier than Alexander, and one gold scabbard is probably Median of the 7th century BC and perhaps not from the main find at all but added to it.166 The treasure is likely to have been assembled from different sources.167 There are Greek pieces too, notably finger rings, which seem 4th-century in date, and therefore not strictly relevant to Greek Bactria, but they are mentioned here because some have thought the whole hoard to be later in date, they are not irrelevant to the record of Greeks and Greek art in the east, and they involve consideration of another hoard of mainly later objects, which are relevant at this point in our narrative. In the Oxus Treasure there are echoes of Greekness even in the Persian objects – asilver statuette of, it may be, a prince, gives him a Persian (Median) gilt hat and poses him with hands forward as if holding sacred objects, but he is stark naked, which is decidedly not a feature of any Achaemenid Persian representation, but normal in Greek art [PL. XII].168 Perhaps he once wore a dress – cloth of gold? The gold rings are 4th-century Greek (two knuckle-bone players) or Greco-Persian (addorsed bull-foreparts) in their types.169 At Takht-I Sangin there was a gold ring with a turquoise intaglio of Greco-Persian type.170
In 1993 another large hoard, parts of it closely comparable to the Oxus Treasure, came to light and is largely housed in the Miho Museum in Japan.171 Russian archaeologists, who had believed that the original Oxus Treasure represented material once at the temple at Takht-I Sangin, hailed it as part of the same treasure.172 Some scholars have expressed strong suspicions about the authenticity of many of the objects, but to the writer most seem genuine enough. But although many of them are decidedly ‘Persian’ in appearance, some are later than much in the old Oxus Treasure. Moreover, it has been suggested that the new treasure is simply part of a much larger find of precious objects from Mir-Zakah in western Gandhara, to the south, including thousands of coins and objects going as late perhaps as the 1st century AD, and not from the Oxus area at all. It may really be a matter of several ‘hoards’ created at various times since antiquity, mainly from Mir-Zakah, one batch of which at least somewhat resembles the content of the first Oxus Treasure, although perhaps not from so far north. Whatever the truth of the matter, many of the objects of the new treasure are relevant since they are Greek or graecizing products with that eastern flavour which we have learned to expect of the Greco-Bactrian kingdoms. Of the Greek material what seem to me the most significant are the gold rings and gems. But whether any of them need to be as late as the Greco-Bactrian kingdom is another matter, and it might well be that at least this part of the find (not the majority of the coins) represents the product of a crisis of the period of the invasions; a few seem ‘Hellenistic’ enough to leave discussion of them in this chapter rather than any earlier.173
The finds include more finger rings of Greek types, but also intaglios, Greek and Greco-Persian. The jewelry is mainly Persian or eastern and there are several Persian-type bracelets and cups (phialai). More substantial are silver-gilt rhyta of Persian type, but some with realistic animal protomes reflecting the Hellenistic versions of the Seleucid and Parthian world, which include remarkable essays of modelling and imagination. Other gilt silver vessels carry more Greek decoration, including one with high-relief figures in a Dionysiac scene with maenads, musicians and the god [PL. XIV; the silver corroded black]. A silver dish with a ketos (sea monster) reminds us of this monster, a Greek creation with a strong eastern history.174 Many gold plaques with figures of Persian worshippers, like those from the Oxus Treasure, are accompanied by others that are classical in dress and pose. The ill-defined history of the hoard, like so much else from the area, must not distract attention from the importance of the finds and their character, implying as they do a society well conversant with Greek manners, and indeed artists working in the pure tradition of the homeland, and apparently into the period after the fall of the Achaemenid Persian Empire.
A long section devoted to coinage might seem out of proportion but in our area and period coinage is important for more than study of mercenary matters. It was in everyone’s hands. It demonstrated graphically who was in command, and more subtly, from the usually divine subjects of the reverses, it could comment on matters beyond the religious. Its inscriptions in the east are not all Greek and they tell us about how non-Greeks could use this gift from the west. Its art in the east could be sublime and match anything of the Greek homeland. Moreover, the evidence in the east has become plentiful, largely from hoards. Separate consideration here of the coinage is justified for two different reasons: it derives from western practices, providing a continuum which deserves to be looked at in more than passing historical comments, and it sometimes involves matters that go beyond the immediate concerns of a person or place.175
In the Persian period there had already been some local production of imitations of Athenian ‘owl’ coins, and some punch-marked oval and discoid concave coins punch-marked with various patterns, not inscriptions, barely coins at all .176
A Seleucid satrap called Sophytos, probably an Indian by birth, had started minting coins of Greek type in the early 3rd century BC,177 and we have seen that Greek coinage, even locally produced, was no novelty for the east. But Diodotos begins the major Greco-Bactrian series, one of the most distinguished for its quality and originality in the whole Greek world.
The coinage of the Bactrian and Indo-Greek kings provides us with important, if sometimes equivocal evidence for the history of their rule. In itself it is remarkable as one of the few Greek arts to flourish for so long and in its native form so far east, in a form often purer than that of the Greek or Greek-inspired artists and architects working in the same regions. It has other things to teach, however, about symbolism and Greek iconography in the service of power politics outside the Greek and Seleucid worlds, but also, since so many of the coins were struck for a non-Greek population, about the juxtaposition of Greek and oriental subjects and language. That the kings took coinage seriously may be judged from the fact that some minted imposing coins of medallion-like size and weight, possibly mindful of their role almost as jewelry as much as for exchange, often in the hands of the non-Greek; many homeland Greeks might have thought them hybristic. The size and magnificence of much of the coinage is sometimes marred by mistakes in the inscriptions. This might suggest that non-Greek artisans played a role in the creation of the coin dies,178 but a degree of illiteracy among Greek artists even at home is well attested (on painted vases, for example).
Punch-marked silver bars were being used in Paropamisadae (south of Bactria) even while the Achaemenid satraps still ruled there, and a few Persian coins had also passed east, although they were more plentifully current in the western empire. The punched bars give rise to the square coins of the Indian Mauryan kings from before 300 BC on, and although these were no longer made after the fall of the empire in the later 2nd century, they continued to be used for centuries. At Aï Khanoum a hoard of 677 such punch-marked coins was found with just six of the Greek King Agathokles; but the latter are bilinguals, Greek and Indian, and of Indian shape.179 Such are the models, in form alone, for many other coins minted by the Greek kings. But we start in Bactria.
47 Punch-marked silver coin from north India. Late 4th century BC. (London, British Museum)
The obverses of the coins usually carry the royal portrait. Symbols of Macedonian kingship are not eschewed: the diadem band and, from the 180s on, the flat-cap kausia. Demetrios I adopts the elephant headdress [PL. XXIII],180 which had been worn by Alexander on Ptolemaic coins and by successor rulers, including Seleukos I; and Eukratides I wears a fine cavalry helmet decorated with horn and ear [PL. XXIV], which might recall Alexander’s ‘horned’ horse Bucephalus, and had appeared, on a different helmet type, for Seleukos I. Heliokles I, Menander and Amyntas dared show their busts wearing an aegis, which is more familiar to us as Athena’s magic goatskin wrap, a very explicit bid for divinity, as is the theos (‘god’) in legends of coins of Agathokles and Antimachos I. The bust, three-quarter back, with the aegis, is almost a novelty even for its prime wearer, Zeus.181 The aegis will occupy us further.
Of other epithets soter (‘saviour’) is common, even dikaios (‘just’), while Eukratides I is ‘the Great’ (megas), or at least ‘Overlord’. In their Indian legends in the Kharoshti or Brahmi scripts, each king is declared a true maharaja. There are a few commemorative issues. Agathokles compliments Diodotos I by repeating his head and reverse type (Zeus), where his own name appears; Antimachos I does the same for him, and for Euthydemos I; Eukratides celebrates his parents Heliokles and Laodike.
The reverses, usually showing deities, can be more informative, and our enquiry becomes as much iconographic as numismatic. Those who determined the devices for the coins were not mindless in their choices, simply copying established Hellenistic or earlier types. Several are significant variations on what we see elsewhere in the Greek world, and we should remark on them even if we cannot always explain them.
There were Bactrian gold and silver issues for the Seleucid Antiochos II, with his portrait and an Apollo reverse. Then Diodotos I mints with his own portrait, but on the reverse a Zeus, though he still names Antiochos, whom he then comes to replace with his own name. The Zeus is interesting [PL. XXV], a slim archaic or early classical figure flourishing a thunderbolt, long familiar. But here he has his aegis, with gorgoneion, slung from his outstretched arm like a shield, and this is odd. The aegis belonged to him before being worn by Athena, but he is seldom seen wearing it. Alexander as Zeus could adopt it, but its addition to this Zeus type seems novel, though it appears shortly afterwards in Greece (Epirus).182 The other major striding, striking deity for Greek art is Athena with her spear, a Promachos, and the figure, on tiptoe, is a Hellenistic type, commonly related to an Athena Alkidemos which stood at Pella in Macedonia and which appears on many Hellenistic coins. For us she appears first with Menander and often thereafter , but always with the thunderbolt, and without a spear, which is the usual western type.183 Moreover, Menander has some coins of her with the aegis on her outstretched arm as on Diodotos’ Zeus, and not her usual shield. This she had not done for many years in Greek art although the archaizing type may have been known in the Hellenistic period,184 when the aegis becomes more fashionable for divinity, as on the Great Altar at Pergamum; but our coins are special cases, especially in the east. Such iconographic details are significant. The aegis can tell a lot about coin symbolism. Archebios has a blitzing Zeus, frontal, and Heliokles I and others have him standing with the bolt.
48 Silver coin of Menander, the reverse featuring Athena wielding a thunderbolt. Mid-2nd century BC.
Seated Zeus is at his best in frontal/three-quarters view from Antialkidas on, and on some coins of Amyntas and, later, Hermaios, his head is radiate. Zeus holding a Hekate on his hand [PL. XXVI], for Agathokles and Pantaleon, or an Athena for Amyntas, are unexpected combinations, probably reflecting various local cult associations. The relevance of the deity to the ruler is seldom particularly apparent; in landlocked Bactria a Poseidon could have more to do with earthquakes than the sea, but there are other aquatic themes. The bolt also appears alone on coins of Indian shape.185 In favouring these attributes our Indo-Greek moneyers must have had ideas of their significance, riverine if not marine.
These figures derive, as does much else in Hellenistic coin iconography, from classical sculptural models, such as the Pheidian seated Zeus at Olympia. Other figures on the Bactrian and Indo-Greek coins have the same source but often without obvious precedent in coinage farther west. The Artemis with her torch on Diodotos’ coins may also have her dog, as in the sculpture groups; for Demetrios I and others she draws an arrow in another familiar type. Given the prominence accorded to Dionysos in Greek stories about the east it is perhaps surprising to see so little of anything Dionysiac on the coins, except for his bust on coins of Agathokles and Pantaleon. The weary but triumphant Herakles seated on rocks is a traditional figure deriving from the sculptor Lysippos, but when he appears first for us, with Euthydemos I [PL. XXVII], his club is propped on a rock pillar, which seems an intimation of a different though related statuary group.186 Crowning himself, for Demetrios I and others, he is a Hellenistic type, first here on coins.187 His role as the hero/god paradigm for rulers is of considerable antiquity in Greek art and thought, reinforced by Alexander’s use of his image. Of other heroes, the galloping Dioskouroi introduced by Eukratides I [PL. XXIV] have antecedents but are nowhere better composed for coins, and with original variations (the turned horse’s head). The pair appear also standing and symbolized by their two pilos hats, and they have other eastern connotations (cf. ills. 46, 148). The same may be said for the goddess with cornucopia, probably Tyche, seated (for Amyntas on his big silver) , and standing on coins of Indian shape for Philoxenos. The debt of the iconography of the Indian goddess Hariti to such figures is generally acknowledged (see also below). The unusual type of Skylla or Triton, with fishy legs and unique in this type on coins, was adopted in the late period by Telephos ,188 and on Hippostratos’ Indian coins. It too will contribute to the iconography of Indian ‘Triton’ figures and, holding oar and dolphin, as he does for Hippostratos, is recalled in the later arts of Yuehzhi Bactria, as we shall see. Helios standing, with Selene for Telephos, or in his chariot, is seen on coins of Platon; he will serve as model for the Indian sun god, Surya. Poseidon will serve as model for figures of Shiva. However, these radiate figures, Apollo, Helios or Zeus-Helios, may also be interpreted as an allusion to Mithra, the Persian deity long venerated far to the east. This is most convincing where the Zeus wears a ‘Mithraic cap’, conical with a nodding crown (coins of Hermaios; and later worn by some Kushan kings). The only problem here is that Artemis too may appear radiate on a type with an otherwise impeccable Hellenistic sculptural pedigree.189 This radiance seems another eastern characteristic.
49 Silver coin of Amyntas. On the reverse, an enthroned Tyche holds a cornucopia. 1st cent. BC.
50 Silver coin of Euergetes. On the obverse, a Triton; on the reverse, Helios and Selene (Sun and Moon) and a Kharoshti inscription. 1st cent. BC.
The range of devices on the coins of Indian shape (square) and other bilinguals makes few concessions to the experience of those expected to handle them. Many of these coins are of copper (or nickel). Most carry the royal portraits and a range of the other Greek devices seen on the principal Greek-inscribed silver issues. Only Agathokles and Pantaleon, with the earliest of these and in their Taxila mint, make a point of showing Indian deities, as well as symbols (the stylized mountain – Meru? – and the sacred tree in an enclosure) which appear on native Indian coins;190 while Menander shows the eastern chakra, wheel of destiny. Appropriate animal devices include the elephant, zebu bull, and standing lion, the last more Persian or Indian than Greek, although it appears in Greek style with a raised paw for Menander II. Otherwise Greek subjects are recruited – club, wreath, aegis, thunderbolt, dolphin, cornucopia, tripod, bucranium, boar’s head, etc. – although not all of them used on the larger silver. Apollodotos II resurrects the old seated Apollo testing his arrow who had served as model for many Persian and Hellenistic devices, including Seleucid.
Stylistically the Bactrian and Indo-Greek series must be judged as being as well composed and executed as most in the Hellenistic world, while their iconographic range, even apart from the Indian devices, is more varied than most. Only with Hermaios do we begin to detect a retreat into pattern for the treatment of heads and bodies rather than even summary treatment of classical and at least semi-realistic forms.
And their legacy? ‘Hermaios’ coins go on being minted after his death, in an increasingly debased form, even until the first of the Kushan kings.191 The coins of the Scythian kings take their lead from the Greek, but without the portrait heads: instead, the king is seated cross-legged, or is a horseman, sometimes in armour. The other devices are simply copied from the Greek – Zeus, Tyche, Athena, Nike, the Dioskouroi, the rest – but even here we may find innovations as yet not traced to their sources. Thus, we have for King Azes a Poseidon in the usual classical pose but accompanied by a reclining river god,192 while Maues has him thus also holding a thunderbolt. These are unique combinations for the god in or outside the Greek world, and it is as though the Indo-Greek Poseidon had become obliged to share the limelight with a major river (the Indus?) and had usurped Zeus’ thunderbolt for the occasion.193 These images could only have been composed, and executed, by a Greek. Imitation of anything remotely Roman is vestigial or highly dubious.194
Parthian coins minted for India follow suit, but some carry Parthian portrait heads and figures. Parthian coinage remotely reflects Greek precedent, with the royal heads, but debased versions styled more directly on the Greek appear; one, for example, with a garbled version of a Parthian royal head on one side , and a version of the seated king with his bow, which had graced Persian  and Seleucid issues.195
The earliest Kushan coins roughly follow suit, but the range of mainly standing deities on the later Kushan coins of India has many sources, among which the Greek are still prominent beside some Roman, but already subject to many local transformations and in an Indian style that remains quite distinct from the classical western.196
51 Silver Parthian coin from Bokhara. The reverse has a debased version of the older Greek motif of a king seated testing his arrow. (After DCAA)