Transcription
Welcome everybody to this ONS quarterly series, and our speaker today is Fabritzio Sini, who is senior research associate at the Austrian Academy of Sciences in Vienna. His primary project is the Silo Numorum Parthikorum, which aims to produce a definitive multi-volume catalog of Parthian coins.
Fabrizio earned his PhD from La Sapienza University in Rome and has a long-standing interest in what coins tell us about cultural history, such as religious beliefs and royal ideology in the pre-Islamic Iranian world.
First of all, I would like to start by thanking the Oriental Numismatic Society for the invitation to speak today. Our story begins in the 2nd century BC in northern Afghanistan, the lands north of the great mountain range of the Hindu Kush, the region known in ancient times as Bactria. While fully part of the ancient Iranian world due to its position, it is included together with other regions south of the Hindukush in the area conventionally defined in the Scythian literature as Indo-Iranian borderlands. Of course, as with all such definitions, one could object about this or that detail, but we are going to rely on it as a useful framework for the discussion.
As we know from textual sources, since the middle of the previous century, Bactria was ruled by a Greek Macedonian dynasty founded by Theodotus I, the former local Satrap, who had seceded from the Seleucid kingdom in the early 2nd century BC. The Greco-Bactrians had crossed the mountains southwards, entering historical India, and thus establishing a permanent Greek presence there, which ended only in the early 5th century AD. Indeed, from a certain moment onwards, these Greek rulers are called in the Greek literature somehow different from the Greco-Bactrians of the north. In addition to a few mentions in written sources, both are mainly documented by coins, which were struck following the Hellenistic canons, in this instance of Seleucid derivation, of portraying the royal bust in profile to the right on the obverse and a deity on the reverse, as you can see here with a Greek tetradrachm in the center and a Seleucid one on the left. The Greek legend identifying the king with his name and title is found on the reverse.
Once in India, a hybrid coinage based on a local standard was developed, in which the Greek legend was moved to the obverse, and its version in Indian language and script, that is Prakrit written in Kharosthi, appeared on the reverse. On the right, you can see one such coin.
Textual sources of different origin inform us about the end of the Greco-Bactrians due to the arrival in Bactria of a series of nomadic invaders. The most important of these sources is the Chinese Shiji, which includes the report of the expedition of a Chinese envoy of the Han court, Zhang Qian, thanks to which we have a terminus ante quem for the fall of Greek Bactria around 130 BC. After several troubles, Zhang Qian had reached the Yuezhi, at that time the dominant group among the nomads, who had the region north of the Oxus under their direct control, a control that extended in some form even to the south.
According to what is transmitted by Chinese sources on the origins of the Xiongnu Empire, the Yuezhi, as they were called by the Chinese, as we do not know their own name, had been forced to migrate from their original seats close to the Yellow River by the first of a series of defeats which kept on pushing them westwards in several stages. On the way, they met various other groups, some of which the Chinese texts call explicitly Saka, Said. At the end, the Yuezhi arrived in Bactria and, after a nearly two centuries long period during which they also crossed south of the Hindu Kush into India, one of their chieftains united them in a single polity, giving birth to the Kushan Empire. We know from coins that this personage was Kujula Kadphises, with whom the properly Kushan history starts.
The reconstruction of the Yuezhi phase in Bactria, including the beginnings on which we have some information, remains complex, currently a synthesis of data extracted by textual sources, western and Chinese ones, numismatics and archaeological evidence. The general idea is that the first wave to invade, to crash into Bactria in fact, was Saka around 140 BC, followed shortly after by the Yuezhi, who were there when Zhang Qian found them. The following decades are obscure, but sometime later, we find Saka in the former satrapy of Drangiana, to the southwest of Bactria, which later came to be known as Sakastan, the present-day name of Sistan being its direct descendant. Current research dates the first coins struck by Saka between Drangiana and their nearby Arachosia, on the border between the Iranian and the Indian world, just to the east, to the first half of the first century BC, during the reign of a certain Vonones. So the Saka would have been forced out of Bactria by the growing pressure of the Yuezhi. But instead of crossing the Hindu Kush, their main group slipped westward along its northern reaches to turn southward towards Drangiana. From there, they later expanded eastward, but in regions south of the Hindu Kush. There, in addition to the Indo-Greeks, who had not been harmed directly by the events north of the mountains, at a certain moment, they met the first Indo-Parthians, who crossed the mountain range through the Kabul Valley, something that we currently date around 70 BC. And with the progress of the first century BC, another actor arrived on the scene from the west, the Parthians, in particular the dynasty founded by Gondophares, the so-called Indo-Parthians, increasingly threatening the westernmost Saka territories.
Now, a fair part of all this is hypothesized based on numismatic evidence, since a crucial part of this evidence is represented by the imagery borne on coins. Let's have a look at Saka coins. Here in the center, we have a tetradrachm of the first Saka king who struck coins, Vonones, that we had just mentioned. The first thing that we can easily see is their depth towards the tradition established by the Indo-Greeks. A coin of Hermaneus is on the left here. The basic concept of having the royal image on the obverse and a deity on the reverse is the same, as are the local weight standard and the double legend in Greek and Prakrit with an analogous layout of the inscription.
At the same time, the differences are immediately apparent, as conveyed by the image of the ruler, who is no longer depicted in the Greek manner as a diademed bust with chlamys, but in the full-length image of a heavy cavalryman in battle attire with lance, armor, and battle helmet, in the style of the cataphracts that we know from Roman sources, on the obverse. There could be no clearer statement of how different the Saka felt themselves and wanted to be seen by others from the Greeks. Here we have to keep in mind that the last Indo-Greek kings reached up to the beginning of the 1st century AD. So, in the first half of the first century BC, when Vonones inaugurated his coinage, Saka and Greeks were both there and still had several decades of coexistence before them.
At the same time, the image of the Saka ruler is quite different even from that of their Iranian western neighbors, the Parthians. On the right here is a Parthian drachm of around 80 BC, with whom certainly there were contacts, as suggested by the fact that the Saka employed the title of "king of kings," not used by Greek kings and resumed in the late 2nd century BC by the Parthians, also with minimal variations, which have spurred a decades-long discussion on the attributions to different rulers. The image of the king as a cataphract remained in use by all the main rulers, from Vonones and his immediate successors, the so-called Group of Vonones, on the left here is a coin of Vonones, in fact, through Azes I, at the center here, who founded an era in 46 BC, starting with what we conventionally call the Azes dynasty, up to Azilises on the right, the last great king of the dynasty.
This emphasis by the Saka on the warlike aspect of their rule is hardly just superficial and highlighted by the comparison with the Parthians, whose military equipment was very similar, but who never depicted themselves in battle attire on their coins. Probably, it was connected to the nature of Saka power in that context, earned by military means. For example, there is the famous Pahlavi "People's Parchment," which mentions Saka mercenaries during the Indo-Parthian period, so the image of the cataphract king perfectly befits a ruler expressed by an elite of mounted warriors.
When we turn to the deities on the reverse, the interpretation is less straightforward, as there is no clear fracture comparable to that visible in the royal image. Indeed, a good number of the images are taken directly from the repertoire found on Greek coins. Using the terms associated to studies on languages with the idea of visual communication as a language, nothing especially new. The script, that is the Greek divine iconography, is exactly the same. The syntax, however, shows some difference. When we focus on silver series, we can easily see that while normally each Greek king used a very limited number of deities on his coins, often just one, the approach of the Saka was quite different, with several gods depicted on the series of the same king. It is true that, for what we have so far reconstructed, Saka coin production had a marked regional pattern, so that one could think of the employment of different gods in different regions. Yet, that structure largely followed Indo-Greek patterns, and there are historical examples that clearly show that a number of different regions not necessarily leads to the multiplication of types. The Indo-Parthians struck drachms in several regions across nearly five centuries, always with one and the same type, for example, although that did not depict a deity, as we know. The point is that the Saka wanted to have several gods appearing on their coins, and those that you see here are just a restricted selection. In fact, that is why I think that, with the Kushan example in mind and with the character of scholarly convention, we can speak of a Saka monetary pantheon.
The Kushan case is worth a momentary and, in fact, only apparent detour. The Kushans started their own coinage in the first century AD, in line with what was produced till then in the framework of the local tradition of Greek derivation, although employing in some instances, such as here on the left, an original image of the ruler, on which we are going to return in a short time, with Vima Kadphises in the early 2nd century AD. A gold coinage was introduced, as you can see here on the right, and the processes of reorganization of coin production started previously brought farther forward. The royal image was subject to a deep transformation. The earlier patterns were largely abandoned in favor of Parthian ones, adopted in view of the new imperial scale reached by the Kushan kings and adapted to his needs. Still, Vima Kadphises only used a single deity on the reverses of his coinage, the one that you see here on the bottom right, with his son and successor, Kanishka, in the second quarter of the second century. The concept governing the selection of images for the reverse changed visibly, and a number of deities were introduced, the Kushan monetary pantheon, accompanied by an epigraphic caption identifying each one of them. The same approach was followed by his successor, Huvishka, after which the Kushan monetary pantheon shrank dramatically, de facto disappearing as a production pattern.
Before comparing the Saka and Kushan monetary pantheons, it is necessary to focus on our understanding of their respective cultural backgrounds. The reason is that while both Saka and Yuezhi belong to what we define as the Central Asian nomads, the traditional scholarly reconstruction posits differences. The definition is not mine, and that this difference can be clearly detected in how they were depicted in their own visual communication. Needless to say, here is where royal imagery plays a crucial role, for it shows how the rulers saw themselves and wanted to be seen. Unsurprisingly, coins are in this regard fundamental, although the evidence is not limited to the numismatic one.
A watershed in the relevant discussion is indeed represented by the interpretation of the sculptural program of a Kushan-time building excavated in the 60s at the site of Halchaian in southern Uzbekistan. It is due to this evidence that Halchaian stands as one of the most important findings of Soviet archaeology in Central Asia. What we see here, known to every student of Kushan civilization, is the graphic reconstruction of the three friezes on the western wall of the building's main hall, published by the excavator of the site, Galina Pugachenkova. As you can see at the bottom, there is a scene with horsemen, one of which is equipped as a cataphract, whereas the other ones wear the lighter rider's costume of jacket and trousers, and two scenes with a number of dignitaries wearing the same outfit. Pugachenkova saw the connection between the largest part of these dignitaries and the image of the ruler on the so-called Heraios coin series from Bactra, the earliest local coins on which the word Kushan appears, as found in the portrait with thin mustache and thrown-back hair cut square at the end, visible in both instances. The few fully bearded portraits were connected to the presence of members of a second ethnic group, which Pugachenkova identified with the local Bactrians, thus explaining the scene with horsemen as a joint parade of the two components of the new elite: local Bactrians and formerly foreign Kushans.
What is crucial to bear in mind, however, is that the clay sculptures were found in fragments, so what we have is a hypothetical reconstruction which should be critically assessed. That is precisely what another great scholar, called Bernard, did in a seminal article published in 1987. Bernard recognized in one of the small fragments a horse hoof in contact with a piece of armor, from which he rightly inferred that it could only belong to a scene of battle. Hence, he reinterpreted the frieze with horsemen as a battle, and the two groups as enemies. The victorious one was clearly identified as that which those who commissioned the frieze belonged to, which we have already seen could only be the Yuezhi Kushan. Those in the battle, the Yuezhi Kushan, are represented by the light horse archers with thin mustache. The others, fully bearded and in heavy armor, were no longer identified with Bactrian allies, but with Saka enemies, a link established by the armor style identical to that found on Saka coins, those that we have seen before. Thus, the battle would celebrate the triumph over the Saka, which gave the former control over Bactria. The characterization of the facial features of the members of the two groups, the Yuezhi attributed with Mongolian features and the Saka with Europoid ones, completed the picture of two well-distinct groups separated by political, military rivalries, as well as culture and ethnicity.
This distinction is at the base of the discussion on the attribution of other visual evidence, such as the magnificent embroidered fabrics found in some of the barrows at Noin-Ula in the early 2000s in Mongolia, reputed to have reached there as a result of long-range exchanges. The vast majority of the numerous figures exhibits the same appearance of the Yuezhi Kushan of Halchaian and of the early Kushan coins, thus leading some scholars to ascribe the fabrics to that context, whereas the excavator, Natalia Posmach, had thought of a Saka origin in light of the figure of a dismounted warrior wearing the same armor of the Saka kings on their coins.
Now, my impression is that returning on our evidence can sensibly modify this general picture. When we look at late Saka coins, on which the ruler is not clad in armor and helmet, such as those of Vasudeva and the Satavahanas, who ruled in the early century AD, both at the center here, we see exactly the same basic iconographic type of the so-called Heraios coins of Kujula and of the Yuezhi Kushan at Halchaian and Noin-Ula. That we can no longer rely on the opposition: heavy armored bearded warriors versus lightly equipped figures with mustache only, is confirmed by the dismounted cataphract at Noin-Ula that we have seen before, as well as by the armored warrior with just a thin mustache and no beard on the obverse of coins of Tanis. In other words, there is no visual distinction at all between Saka and Yuezhi Kushan. When we add that some of the linguistic elements surviving in Kushan royal onomastics might in theory be connected to a Saka background, and above all, when we properly consider the care that both had in handling their official visual communication, that is, how keen they were to appear as they did on coins, the most natural conclusion is simply that there was no ethnocultural division between them.
Here it is worth stressing that the Kushan version of the royal imagery of Saka derivation is not a mere imitation of the patterns established by Vonones and his successors. Rather, the contrary, the so-called Heraios series of Kujula use the same language but spelled in quite a different way. And to see a Kushan king in armor, we have to wait for the late 2nd century AD. In my view, this excludes that the Kushans were simply using Saka iconography without having anything else in common with Azes and his successors in a context so sensible to the language of images. That would simply make no sense.
All this means that, as a first wrong step in our reconstructions, we have been misled in interpreting the Chinese sources. The fact that they named a tribal group explicitly as Saka and another as Yuezhi in no way precludes that even the Yuezhi were themselves Saka. This, for example, removes the main point of friction in the attribution of the Noin-Ula fabrics, for example.
This said, I am perfectly aware that a non-negligible part of the broader problem is what it means to be a Saka, especially considering that such names were often generic even in antiquity. Here we should never forget that the idea that such groups were in any way homogeneous is misleading. To be even clearer, I would find myself quite at unease in defining Saka ethnicity. We have the scientific tool offered by linguistic classification, but in this instance, we know next to nothing. Not to say that historical problems pertain to other different levels. However, this does not mean that, with the due awareness that identity is a construct, we cannot address problems of identity, even if a subjective one.
Here is where discussing about religion can be indeed meaningful. And here is where we reconnect with what we have only quickly seen before, that is, the images of deities on coins. In the light of what we have said so far, the first thing that naturally springs to mind is checking if the Saka pantheon corresponds to the Kushan one, as we know it from the coinages of Kanishka and Huvishka. Bearing in mind that a set of coin images, in this instance what we call a monetary pantheon, must be compatible with the religious leanings of the issuers, but it was not conceived as an illustration of a religious system. This means that once its organization is critically assessed, the Kushan monetary pantheon, for example, is crucial to discuss Kushan religion. But the religion of the Kushans was a broader phenomenon, whose nuances cannot find the full expression on coins.
They said, iconographic analysis has different levels. One is that of the images' immediate morphology. So, in this instance, which gods are depicted. Another is, for example, what these deities are supposed to do altogether as a group. When we focus on the Kushans, the formal level is made explicit by the names that appear beside every deity from Kanishka onwards. This is a crucial feature that unmistakably identifies each deity, be it on gold coins of Huvishka, gold coins of Kanishka, host nearly a dozen deities, or copper. Thanks to that, we can see that the vast majority of these gods are indeed Zoroastrian deities. A connotation that appears even more clearly when the copper coinage is examined, as the wide selection of deities found on gold is restricted to a much smaller number. All of Zoroastrian background, Mithra, Nana (here with the Greek version of her name, Nanaia), Atar, and Vayu. The Buddhist images at the end of Kanishka's coinage do not alter the picture, and recently, Manjusri, one of the Bodhisattvas, has been added to deities on Kanishka's scope. Indeed, even the indications coming from the inscriptions of Rabatak inscriptions, sorry, of Rabatak go in the same direction, that is, the local regional variant of Zoroastrianism, meaning that the Kushans had undergone through a deep process of Bactrianization. Accompanying the reverse deities with an epigraphic caption was a complete novelty. And unfortunately, none of the gods on Saka coins is provided with a name indicating its identity. Thus, the deities that we find on Saka coins are conventionally referred to by scholars as Zeus, Athena, Heracles, and so on, based on their iconography. Indeed, often the images are taken directly from the Indo-Greek series, as we can see here with the Greek versions on top of some images of Zeus, Athena, and the Dioscuri, and their Saka direct descendants at the bottom, two coins of Azes and one of Azilises. For the Saka, in the Greek coinage, already shows that some of these iconographies were employed to depict local Indian deities. That is the case with the image of Zeus accompanied by an elephant on this coin of Antimachus, which is a clear attempt at portraying, by means of the image of the Greek king of the gods, the Indian king of the gods, Indra, whose associated animal is precisely the elephant. A striking confirmation of these processes comes from glyptic evidence on the seal on the right here. The same scene of the anointment of Gajalakshmi found on coins of Azilises on the left here, with elephants and lotus flowers, is visible with the goddess depicted in the guise of a Greek deity, leaving no possible doubt that the image of this Greek deity was employed to portray the Indian goddess, regardless of the figure's appearance. While interpreting these images, when not accompanied by any secondary feature, remains problematic, it is in any case clear that phenomena of iconographic transfer based on the adoption of Greek iconographies to depict non-Greek deities were already on the way in the Indo-Greek context. This must be stressed, as the subject impacts a vast area of the ancient Orient, from the Near East to India across Iran, and even today, not all scholars agree on this basic development. On the contrary, there are those who, in absence of features that might visibly qualify these divine images as non-Greek, would in every respect consider them to portray fully Greek gods.
This said, Saka reverses host also deities that have no antecedent in the Greek repertoire. Leaving aside those of clear Indian derivation, which are found mainly on the coppers (bottom row here, with the exception of the Azilises coin that we have seen before), some of these deities appear as Greek-looking figures, as the style and general concept of their images clearly owe to Greek divine iconography. However, the combination of their iconographic attributes finds no correspondence in images of deities known from Greek religion. This is the case, for example, of the two images in the top row here: a female deity conventionally defined by scholars as "lamp goddess" and a couple formed by two deities depicted as Zeus and Tyche. So, reconstructing the Saka monetary pantheon against the background of a Greek religious system seems difficult on a double level: that of the whole system, which differs from the Greek approach of one god for each king, and that of some of the individual images, which remain without a precise identification. An Indian interpretation is made difficult by the fact that clear Indian features, which, as we have seen, are at times visible on some series, especially the copper ones, are normally lacking. Actually, here there is no surprise, as one would struggle to understand why the Saka should have employed exclusively Greek or Indian gods for their coins.
With the Kushans, the picture appears to be different, despite the fact that they came, if what exposed so far is accepted, from a similar Saka background. Indeed, when one compares the images of the Kushan monetary pantheon with those of the Saka coinage, the distance is immediately clear. Kushan coins are dominated by Mithra and Nana, Mah, Vayu, Atar, and Vayu. Athena is very rare and identified with the quintessential Zoroastrian Ashtat. Zeus is completely absent, and the same holds true for Poseidon. On Saka coins, the recurrent divine figures are precisely those of Zeus and Athena, in addition to Tyche, not to mention the lamp goddess that we have just seen. The image of Poseidon, for example, is well attested after having appeared on the Greek series only rarely. Thus, the religious figures on the coins of the Saka show that their beliefs differed radically from the Zoroastrian ones of Kushan Bactria.
At this point, one is left to wonder about another little-explored possibility, which is that these images might refer to a Saka religion, that is, another Iranian religion, but different from the Zoroastrianism spread amongst Iranian sedentary peoples such as the Bactrians. Remarkably enough, we have a religious system of this kind described by Herodotus in the Scythian Excurses in his Book IV. There, Herodotus spoke of the religion of the European Scythians. "The only gods whom they propitiate are these: Hestia in particular, and secondly Zeus and Earth, whom they believe to be the wife of Zeus. After these, Apollo and the Heavenly Aphrodite and Heracles and Ares. All the Scythians worship this as gods. The Scythians called royal sacrifice to Poseidon." Also, in the most crucial passage, he even added the deities' original names: "Hestia is called Tabiti; Zeus, in my judgment, most correctly so-called, Papaeus; Earth is Apollo; the Heavenly Aphrodite, Argimpasa." There is no need to elaborate on the relation between the terms Saka and Scythian to see the potential importance of these lines for our discussion. Indeed, in Herodotus, we find exactly the gods visible on Saka coins: Zeus, Poseidon, Athena, a variety of other figures being in addition partly preserved in the folklore of the Udmurts, the modern descendants of the Finno-Ugric peoples. Most significantly, the combination of the images with the names provided by Herodotus allows identifying even figures otherwise uninterpreted, such as the lamp goddess. This deity is so defined by scholars due to the lamp with flames that she holds in her right hand. This iconographic attribute is shared with the later image of the Kushan Faroh, a god notoriously associated to light and warmth. The connection of the goddess image with fire is clear, and this provides an easy link with Hestia, the deity of the hearth, as well as with the Scythian Tabiti, whose name corresponds to an Indian deity with an analogous function called Tapatī, known as she who brings warmth. So the lamp goddess may well be the depiction of Herodotus' Scythian Hestia, that is, Tabiti. Similarly, the couple Zeus and Tyche of Saka coins could well portray the Scythian-Saka couple Papaeus-Aphrodite, in an image that would otherwise remain with no explanation. The list of identifications includes also other deities, that is, Tagimasa, being another notable example, but also minor figures. All in all, the Saka monetary pantheon finds a remarkable correspondence in the Scythian religion described by Herodotus. While on copper, there is a role for Indian figures coming from the local religion, the basic coordinates of the coin system strongly suggest that the Saka kings were essentially giving visual form to their own Saka gods.
What does this tell us? First of all, it attests the existence of a Saka religion among the Saka of the Indo-Iranian borderlands in the first century BC. That should not be a surprise, at least in theory, as it would be only natural to expect that the Saka had a Saka religion. Still, they had been so far left in the dark on this point. And if this reconstruction is accepted, we have concrete evidence which confirms Herodotus' description of the Scythian religion of four centuries before. A significant point worth stressing is that, in returning to the question we posed above, we would have a piece of evidence helping us in concretely defining, with the caution expressed before, a Saka identity. That this might be a rewarding interpretative path is indicated by the specular phenomenon pertaining to the Iranians, who are reputed to have mainly defined their identity as Aryans, Iranians on the base of a common religious tradition, as a western one. The Scythian religion would thus define a religious culture that we associate to groups that scientifically can be classified as Iranian based on the language that they are supposed to have spoken, but that in all likelihood did not define themselves as such. Here the comparison between the two groups of Saka that we have discussed is illuminating in showing the dynamics of the possible alternative cultural trajectories. Differently from their southern relatives, during the two centuries in Bactria, the Saka that later became the Kushan completely Iranized themselves in the meaning already associated to the term in antiquity, as eminently shown by the term Arian used by Kanishka at Rabatak, in addition to the bulk of the gods that appear on their coins. With regard to the wider history of religions in the Iranian world, the Scytho-Saka case provides us with a religious system that was never involved in the development of the strand of religious tradition that led to what we describe with the definition of Zoroastrianism. In extreme synthesis, the latter, as far as the scholarly deconstruction goes, was the result of the reform of the earlier religion of distant Indo-Iranian background. The Daeva inscription of Xerxes shows that in the early fifth century BC, worshippers of the old gods rejected by Zoroastrianism were still active, faced by the new imperial religion of Ahura Mazda backed by a royal father. The Scythian religion never belonged to this horizon, on neither of the two extremes, so to say, and does not show any sign to have been ever in dialogue with this religious world.
Thank you so much.