Anshan

The region had as many names as it had neighbours. To the Elamites who first ruled it, the highland country and its chief city were Anzan, more usually written Anshan in modern scholarship; the same territory, roughly the modern province of Fars, was Parsa to the Persians who later settled it, Parsumash to the Assyrian scribes who first noticed them, and Persis to the Greeks.^[1] Anshan is first named in the third millennium BCE and recurs in the cuneiform record down to the Achaemenid period.^[2] It is not an Iranian word, and it belonged to the older, Elamite conception of the land; the highland peoples of southwestern Iran never called themselves Elamites, a label imposed from Mesopotamia, but knew themselves by the names of their several regions and dynasties, Anshanites among them.^[3] The name of the city and the name of the region were the same, and the ambiguity runs through the sources: 'king of Anshan' can mean king of the city, king of the highland country, or, in the mouths of Cyrus's Babylonian scribes, simply king of Fars.

For most of the modern study of Elam, Anshan was a name in the texts without a place on the map, and the texts pointed in two directions. The Book of Daniel and later tradition set Elam in the lowlands, around Susa in Khuzistan, and equated it with the district the Greeks called Susiana. Other sources kept Elam and Susa apart, and cuneiform titles that paired 'Anshan and Susa' clearly meant two different places. Where Anshan lay stayed open until the early 1970s, when inscribed bricks were excavated at the mound of Tall-e Malyan, near Shiraz in highland Fars, and read by the Assyriologist Erica Reiner. They named the site as the ancient city of Anshan.^[3][4] The identification is now secure: "The city of Anshan is definitively identified at Tal-e Malyan", at the northwest end of the Kur river basin.^[2]

The find settled more than a location. It forced a revision of what Elam itself had been. The French scholar François Vallat argued from the new evidence that the centre of Elam lay in the highlands at Anshan rather than at lowland Susa, since Elamite was the language of the highland inscriptions while the texts of Susa were overwhelmingly in Akkadian.^[3] Anshan, that is, was not a provincial dependency of Susa but the upland pole of a country that had always had two centres. The site had in fact been dug before it was named. Tall-e Malyan was systematically excavated from the early 1970s by an expedition of the University of Pennsylvania under William Sumner, whose survey of the Kur river basin and stratigraphy of the mound remain the basis for everything said about the city's size and history; it was in his excavations, in the seasons of 1972 to 1974, that an archive of Elamite administrative tablets was found.^[5][6]

Malyan is not one city but a sequence of them, separated by centuries of desertion. The mound was first a town of some size in the late fourth millennium BCE, in the Proto-Elamite (Banesh) horizon, when it was one of the earliest urban centres of the Iranian plateau, walled and administering its hinterland with the sealings and numerical tablets of the Proto-Elamite system. It was then abandoned around the middle of the third millennium and resettled about 2200 BCE.^[3] This rhythm of growth, desertion and re-foundation is the pattern of the whole site, and it is what makes Anshan so hard to read: the name persisted across the gaps far more steadily than the city did. Its greatest age was the six centuries that followed, the Kaftari period, from roughly 2200 to 1600 BCE. Sumner's survey of the Kur basin recorded ninety-four Kaftari settlements ranked in a clear hierarchy, and at the top of it sat Malyan, dwarfing everything around it.^[5] Beginning at about 40 hectares, the walled city rose to a "massive 130 ha of settled area during the Middle Kaftari period, when it was more than ten times the size of all but one other site" in the whole basin, the next largest being a mere 15 hectares.^[3] A massive city wall enclosed it in the Kaftari period.^[5] For the early second millennium this was one of the largest cities anywhere on the Iranian plateau, a highland capital commanding the agricultural plain of the Kur and the pastoral uplands beyond it.

The wealth of the place was agrarian and pastoral rather than monumental. The Kaftari economy at Malyan combined cereal farming, some of it irrigated, with the large-scale herding on which the surrounding highland population depended, and the city's glyptic and pottery show it in steady contact with lowland Susa across the four hundred kilometres of mountain that divided the two.^[3] This was the physical basis of the thing scholars now emphasise about Elam: that it was never simply a lowland civilisation but a dual one, "incorporating both lowland alluvial plains and a large expanse of the Zagros highlands", a state with a "distinctive dual highland-lowland cultural personality" whose two poles were Susa below and Anshan above.^[7]

That duality is written into the royal titulary. After the long era of the sukkalmah regents, the Middle Elamite period, from about 1500 to 1100 BCE, saw the old kingly style return, and with it the double title: 'king of Susa and Anshan' in the Akkadian inscriptions, 'king of Anshan and Susa' in the Elamite ones.^[8] The two toponyms name the two capitals, Susa in Khuzistan and Anshan in Fars, and the order in which they stand is not accidental. As Vallat showed, the Akkadian texts put Susa first and the Elamite texts Anshan first, each version leading with the seat that mattered to its audience, so that the same king could flatter his Akkadian-speaking and his Elamite-speaking subjects in turn.^[3] In the later Middle Elamite inscriptions, written increasingly in Elamite, the kings favoured the form that put Anshan first, following the older highland tradition.^[8]

The Shutrukid kings of the twelfth century BCE who bore this title were among the most powerful Elam ever produced. Shutruk-Nahhunte and his sons campaigned deep into Mesopotamia, overthrowing the Kassite dynasty of Babylon and carrying home the law-stele of Hammurabi, the victory monument of the Akkadian king Naram-Sin and the cult-statue of Marduk itself; George Cameron called the reign of Shilhak-Inshushinak the acme of Elamite civilisation.^[9] Shilhak-Inshushinak styled himself lord of both realms and left inscribed bricks and glazed wall-plaques at Malyan as at Susa, and the deities of the highlands and the lowlands were housed together, most spectacularly in the ziggurat that Untash-Napirisha raised at Choga Zanbil in Khuzistan.^[8] The double kingship, that is, was a real political fact of the Middle Elamite state, its two capitals genuinely paired. Yet the title outran the reality of the highland city. Excavation shows that after the end of the Kaftari period around 1600 BCE there was little substantial settlement in the Kur basin for two centuries, and the first major post-Kaftari building at Malyan dates only to the later fourteenth century.^[3] Potts draws the consequence: the Middle Elamite use of 'king of Susa and Anshan' was in part "ceremonial and anachronistic", much as later Babylonian kings called themselves 'king of Kish' or 'king of Ur' after those cities had ceased to be capitals. Anshan the name still carried prestige; Anshan the city was already diminished.^[3]

With the fall of the last Middle Elamite king to Nebuchadnezzar I of Babylon around 1100 BCE, the record goes dark. For more than three hundred years, from about 1100 to 760 BCE, written evidence for Elam is very thin, and these are the Neo-Elamite 'dark ages'; a handful of economic tablets from Malyan and scattered Mesopotamian notices are almost all that survive.^[10] In Fars the archaeological picture is one of decline. At Malyan itself only a small early Neo-Elamite occupation and a few burials are attested, and the great tell, the ancient city of Anshan, was largely given over, its "decline and abandonment in the early first millennium" usually linked to a shift toward pastoral nomadism and the arrival of Iranian populations.^[10]

Highland Fars did not fall out of the Elamite world altogether. Neo-Elamite figures were carved onto older reliefs at the highland sanctuaries of Kurangun and at Naqsh-e Rostam, the future Achaemenid royal necropolis, marking a continuity of cult and authority in the uplands into the first millennium.^[10] But direct control of Anshan by the kings at Susa seems to have lapsed. The name all but disappears from the Neo-Assyrian sources, and when it resurfaces around 691 BCE, among the contingents an Elamite king mustered against Assyria at the battle of Halulê, it appears beside 'Parsua' as if it were a separate power. Pierre Briant read the omission of Anshan from the Neo-Elamite titles as a sign that the highland region had been lost by the kings of Susa to an emerging power, the Persians, who become visible in the reign of Assurbanipal; Potts treats the suggestion as probable but stresses how speculative the detail remains.^[3] What is clear is that by the seventh century a highland kingdom of Anshan, whether already Persian or not, was no longer bound to Susa. Elamite kingship itself did not end abruptly. Reduced and local, Neo-Elamite kings held on in the lowlands into the sixth century, and one of the last of them, Tepti-Huban-Inshushinak II, is now dated so late by some scholars that he would have reigned as a vassal of the Teispid Cyrus himself.^[10] The old order shaded into the new rather than being swept away by it.

The crucial development of these obscure centuries was demographic. From about 1000 BCE, Iranian-speaking people had been settling in Fars and living alongside the Elamite population of the highlands, and the two slowly merged. Miroschedji named the process the ethnogénèse des Perses, the ethnogenesis of the Persians: not a conquest of Elamites by incoming Iranians but "a gradual melting together of the Iranians and the Elamites, creating a new people", out of which Achaemenid culture would be born.^[11][6] Assyrian pressure did the rest. When Assurbanipal sacked Susa around 646 BCE and shattered the Elamite state, the highland kingdom in Fars was left to grow; "at the same time Elam was collapsing, a nascent kingdom under Teispes and Cyrus I was rising".^[1] Its first king, on Cyrus's own reckoning in the Cylinder, was Teispes, and it took shape around 635 BCE.^[6]

How deeply Elamite this new Persian kingdom was is shown by more than its geography. The name Cyrus itself, in its earliest Elamite form Kuraš, is most plausibly explained as an Elamite word rather than an Iranian one, as is the name of his ancestor Teispes; the administration the dynasty set up in the highlands, the great archive later kept at Persepolis, was conducted in Elamite, not Old Persian.^[6][12] The earliest Achaemenid royal inscriptions are accompanied by Elamite versions, and the first draft of Darius's Behistun text was, almost certainly, in Elamite before it was in Old Persian.^[12] So thorough was the convergence that Wilhelm Henkelman, whose study of the Persepolis religious texts is the fullest modern treatment, concludes that "Persia is the heir to Elam", the Elamite state being the logical predecessor of the Persian one, and the Persian pantheon of the Fortification tablets a fusion of Elamite and Iranian gods; the formula, he stresses, "remains an interpretative model" rather than a settled description.^[9] Walter Hinz put it more strongly still: the Persians inherited Elam's art and civilisation, and had profited from an Elamite education long before they ruled.^[13] To a lowland Elamite, Miroschedji suggested, the arrival of Cyrus may have looked less like a foreign conquest than the restoration of the old kingdom of Anshan and Susa.^[11][3]

How live this fusion still was in Cyrus's own century is shown by the revolts that broke out in Elam at the accession of Darius. Two of the Elamite rebels named at Behistun bore Iranian names and patronymics, and one, though Darius calls him a Persian based in Fars, proclaimed himself an Elamite king under an Elamite throne-name.^[12] Henkelman reads such cases as evidence that the acculturation, the Persian ethnogenesis, was an active and wide-reaching process even around 520 BCE, still at work in territory that was then reckoned Elamite.^[14] The making of the Persians out of Elam was not finished when Cyrus took Babylon; it was still going on under his heirs.

It is against this background that the Teispid kings called themselves 'king of Anshan'. The title has few precedents, and most scholars read it, in Matt Waters's words, as "a conscious modification of the traditional Elamite title" 'king of Anshan and Susa', keeping the highland half as the seat of Cyrus's family's power. It was, on this view, the dynasty's original style, borne by Cyrus and his forebears, and evidence of the Elamite-Persian fusion at the root of the empire.^[15]

One small object anchors the claim outside the royal monuments. A cylinder seal inscribed in Elamite for "Cyrus the Anshanite, son of Teispes", taken to be Cyrus's grandfather Cyrus I, survives in impressions on the Persepolis Fortification tablets, where it was still being used to seal documents in the reign of Darius, more than a century after it was cut.^[16][1] The image on it, a mounted warrior riding down a broken foe, is worked in a Late Elamite style; the seal is an heirloom of the founding house, a prestige piece kept in use long after its owner's death.^[1] Such a relic of Cyrus's line was still in service under Darius, whose own accession came at the expense of Cyrus's sons. Whether the Cyrus of the seal is the same as the Cyrus of Parsumash who paid homage to Assurbanipal remains disputed. According to the Assyrian record, at the close of the last Elamite war a king of Parsumash named Kurash, hearing of Assurbanipal's crushing victory over Elam, sent his eldest son Arukku with tribute to Nineveh to do obeisance, around 639 BCE. This is the earliest appearance of a Persian ruler in a dated source, and its Kurash has conventionally been identified with Cyrus I; but the long reigns that the Cylinder genealogy would then require sit awkwardly with an Assyrian date in the 640s, and Miroschedji and others have judged the equation chronologically strained, so that the identification is not settled.^[3][15]

Cyrus took that style into the Babylonian record himself. The Nabonidus Chronicle, recording his war with the Median king Astyages, names him ruler of the highland seat:

(Astyages) mustered (his army) and marched against Cyrus, king of Anshan, for conquest … The army rebelled against Astyages (Akk. Ishtumegu) and he was taken prisoner. They handed him over to Cyrus.

the Nabonidus Chronicle ii.1-2, trans. Kuhrt^[17]

And the Cyrus Cylinder, drawn up at Babylon in 539 BCE, sets his pedigree out in full, four generations deep, every one of them a king of Anshan:

I, Cyrus, king of the universe, mighty king, king of Babylon, king of Sumer and Akkad, king of the four quarters, son of Cambyses, great king, king of Anshan, grandson of Cyrus, great king, king of Anshan, descendant of Teispes, great king, king of Anshan, eternal seed of kingship.

Cyrus, the Cyrus Cylinder §§20-22, trans. Kuhrt^[18]

Four generations stand in that pedigree, Teispes and then Cyrus I, Cambyses I and Cyrus II, and not one of them is Achaemenes. That silence is the crux of the dynastic question. Only from the reign of Darius does Achaemenes appear as the founder of the line, on a genealogy widely read as Darius's own construction (see the Achaemenid Dynasty).^[12] What matters here is that the one title demonstrably Cyrus's own, and his fathers' before him, was neither Persian nor Achaemenid but the archaic Elamite style of Anshan.

Why an Iranian dynasty ruling in Fars should present itself under an Elamite name has been read three ways, and the debate turns on how Elamite Cyrus's people really were.

The strongest form of the Elamite reading is Daniel Potts's. Taking the title at face value, Potts argued that Cyrus was in origin king of a polity, Anshan, that was "culturally Elamite, not Persian", whose people held an Anshanite identity;^[19][14] the Achaemenid empire, on this account, "however 'Persian' it may have been, in one sense evolved from the Neo-Elamite social, cultural, linguistic and perhaps even political milieu".^[3] At the other extreme, Antigoni Zournatzi has suggested that the title carried no such weight, and that 'king of the city of Anshan' was simply the form Cyrus's scribes chose to make his kingship legible to a Babylonian audience, for whom Anshan was the familiar old name of the region.^[14]

Henkelman occupies the ground between them, and his is now the most influential position. He points out that the Anshanite style was not confined to Cyrus's Babylonian texts, where a Mesopotamian audience might explain it, but appears already on the Elamite seal of Cyrus I, so the claim to rule Anshan was a fact of the Teispid line long before the conquest of Babylon.^[14] Since 'Persian' had by then become the dominant name for the people, while 'Anshanite' is nowhere attested as a living identity, the reference to Anshan cannot be a plain geographical label. It was rather "a conscious strategic, ideological choice, a 'deliberate archaism'", reaching back past the recent past to the prestige of the old Elamite kingship of Anshan and Susa, a title that still resonated in Elamite Fars in the seventh century.^[14][9] On this reading Darius's later insistence that he was Persian and Aryan is not a break with Cyrus's Elamite self-presentation but the maturing of an ethnic identity that was still, in Cyrus's day, half-formed.^[14] What would settle the question is what the highland kingdom looked like from the inside before 550, and of that the sources say almost nothing; the debate rests on a seal, a handful of titles and the archaeology of ethnogenesis. The balance of the field now lies with the ethnogenesis model, for which the Anshan title is a central piece of evidence.

As a living city, Anshan had little part in the empire its title helped found. Cyrus did not restore it. In his highland homeland the old capital "lay uninhabited, an unoccupied ruin throughout the Neo-Elamite period", so that when he built a new seat he built it fresh, at Pasargadae on the Murghab plain, less than forty kilometres northeast of the abandoned mound of Malyan.^[3] Under Darius the centre of the highland moved again, to Persepolis at the eastern edge of the same Kur basin. Anshan the city was supplanted, first by Pasargadae and then by Persepolis, as the capital of a region that now took the Persian name Persis.^[2]

The archive that ran the heartland confirms the eclipse. Anshan is barely mentioned in the Persepolis Fortification tablets or in the Neo-Elamite tablets from the Susa acropolis, which is why Henkelman concludes that "Anšan cannot have been a major, central town at the time".^[14] The tens of thousands of tablets from Persepolis, written in Elamite and administering the very region that had been the kingdom of Anshan, are the fullest proof of how deep the Elamite inheritance ran in the working empire; yet the old highland capital that gave the dynasty its name barely figures in them. The living substance of Elam had migrated into the institutions, the language and the gods of Persia. The title, meanwhile, faded almost as soon as it had served its purpose. Cyrus used 'king of Anshan' beside the Babylonian world-styles he took over from Nabonidus; the same Nabonidus Chronicle that opens by calling him king of Anshan later writes 'king of Parsu', the two toponyms treated as one, and after Cyrus the name Anshan drops almost entirely out of royal use.^[12][1] Miroschedji suggested that its inclusion in the first-millennium titles was a way of evoking a golden age of Elam's past; once the empire had grander claims to make, the golden age could be let go.^[12]

What can be seen at Malyan today is the eroded record of the highland city, not of the empire. Sumner's excavations of the 1970s recovered the Banesh and Kaftari town, its city wall and its Middle Elamite building, along with glazed architectural ornament and the administrative tablets that first fixed the site's identity; the great tell is the largest pre-Achaemenid city yet dug in Fars.^[3][5] Gorris and Wicks stress the persistence of the royal style, "the continued employment of this royal title after the tell's abandonment".^[10] The city was gone; the title outlived it by centuries, long enough to become the first style of the kings of Persia.