The Satrapy System

The title behind "satrap" is Old Persian xšaçapāvan, which joins xšaça-, "realm, kingship, sovereignty", to pā-, "to protect", with the agent-suffix -van: the office describes its holder as the protector of the realm.^[2] The name says what the office was. The word passed into Greek as satrapēs not directly from the Persian of the court but, the philologists judge, through a Median dialect form, and thence into every later language. Darius's own inscriptions do use the title of an official twice over (at Behistun, of Dādaršiš and Vivāna, the men who held Bactria and Arachosia for him against the rebels), so the office and its name are attested from the King's own hand and not only from the Greeks.^[3]

Yet the Old Persian royal texts do not use a word built on xšaçapāvan for the provinces themselves. For the administrative units that made up the empire they use a quite different word, dahyu (plural dahyāva), a term for a land or country. When the modern world speaks of "the satrapies" it is following the Greek usage; the King's own usage was to list his dahyāva, his lands. The gap between the two vocabularies is not pedantry: the Old Persian word had a breadth, running from ethnic groups to social strata to the regional scale of a satrapal province, that made such terminological overlap natural, and it lies at the root of the long dispute over how the empire was really organised.^[4]

Nor did "satrap" name one fixed rung of a ladder. The Greek sources apply it to figures of widely different weight: to a prince such as Cyrus the Younger, viceroy over half Anatolia, but equally to Mania, the widow who governed the little district of Aeolis, and who needed the approval of Pharnabazus, satrap of Hellespontine Phrygia, to hold it. Lycian and Greek inscriptions do the same, giving the title both to the governor of Lydia and to the Carian dynasts Maussollos and Pixodarus who ranked beneath him. Jacobs concludes that the word's semantic range matches the English "governor", so that "a satrap can be defined as the highest official of a particular administrative area, irrespective of its hierarchical level".^[5] The office was not even reserved for Iranians: Carian dynasts held it, and an Aramaic inscription suggests the same for Cilicia.^[5] "The satraps" of modern accounts, the twenty-odd great provincial viceroys, are a Greek-shaped simplification of a graded hierarchy of governors that ran from princes of the blood down to local dynasts and a governor's widow.

The empire grew with astonishing speed. Between the rise of Cyrus the Great about 559 and the death of Cambyses in 522 it swallowed the Median realm, the Lydian kingdom of Croesus, Babylon, and Egypt, each a great state with its own machinery of rule. The Persians did not, as a rule, tear that machinery down and build anew. They set a governor over each conquered kingdom, gave him the King's authority, and let the old provincial structures stand beneath him: Lydia was taken together with Cappadocia and the Ionian coast, Babylonia together with Assyria, Media together with Armenia, and so each great block entered the empire more or less whole; these conquered predecessor-kingdoms are what Klinkott calls the "true satrapies" of the early administration.^[6][4] The satrapal frame, on this reading, was less an invention than an adaptation, laid over the bones of the empires Persia had devoured, with the old provinces of each kingdom carried along as sub-provinces of the new complexes: Cappadocia under Lydia, Assyria under Babylonia, Libya and Nubia under Egypt.^[5]

The continuity is not merely inferred. Darius presents the twenty-three lands of the Behistun list as lands that came to him, an inheritance rather than a creation; the two men that inscription titles satrap, in Bactria and Arachosia, were in post when the crisis of 522 broke, and so must have been installed by Cambyses at the latest; and the administration visible in the Persepolis archive under Darius had antecedents, for Babylonian workers were already being sent to palace-building in the Persian heartland in Cambyses's reign.^[5] Some stabilising provincial system, the argument runs, is the only thing that can explain how the conquests of Cyrus held together at all through the quarter-century before Darius, and through the great crisis of 522 to 520 itself.^[5]

The tradition, following Herodotus, nonetheless credits Darius I with turning the inheritance into a system. After the revolts of 522 to 521, Herodotus reports,

he divided his dominions into twenty governments, called by the Persians satrapies; and doing so and appointing governors, he ordained that each several nation should pay him tribute.

Herodotus 3.89, trans. Godley^[7]

Under Cyrus and Cambyses, he adds, "there was no fixed tribute, but payment was made in gifts", and it was for the new assessment that the Persians coined their nicknames for the first three kings: "the Persians called Darius the huckster, Cambyses the master, and Cyrus the father; for Darius made petty profit out of everything, Cambyses was harsh and arrogant, Cyrus was merciful and ever wrought for their well-being."^[7]

The fiscal half of this picture has documentary support. Babylonian cadastral documents, land-registers listing fruit-trees, crops and arable extent, survive from the third year of Darius's reign, which puts a reorganisation of the taxes at about 519, just after the revolts: the land was measured and classed by its yield, and each province set to render a sum reckoned from its cultivated area and fertility.^[8][9] By about 503 the Babylonian archives begin to cite royal land-registries of a kind not attested earlier, which Stolper judges "very likely a consequence of Darius's administrative and fiscal reorganization".^[10] On this reading, followed by Dandamayev and in essentials by Briant, Darius the systematiser of the taxes is a defensible figure.

Jacobs presses the scepticism much further. Herodotus is the only source for a tax reform; the satrapal frame demonstrably predates Darius; and what Herodotus actually wrote is looser than the simplified rendering suggests. His conclusion is blunt: "the supposed reform can be removed from the history books".^[5] On the same view the catalogue of twenty is not administrative evidence at all: Armayor argued that the list is an intrusion from another literary genre, epic, its historical value "comparable to that of the catalogue of ships in the Iliad".^[5]

The two positions are further apart in rhetoric than in substance. Almost no one now defends Darius as the inventor of the satrapy, and almost no one denies that the Babylonian cadastres attest some real fiscal reorganisation early in his reign. What has gone is the received synthesis that fused the two, a single grand reform of 519 that carved the empire into twenty provinces and priced them: the provinces were older, the count of twenty is Herodotus's, and the reform, if the word still fits, was an overhaul of assessment inside an inherited frame. A provincial archive from before 522, which nothing now held provides, is what would settle how much older.

Whoever drew the map, the office at its head is clear enough. "A satrap was first and foremost the personal representative of the king", in Briant's formulation.^[11] He assessed and forwarded the tribute, the province's yearly render in silver and in kind. He was the supreme judge of his territory: civil and criminal causes alike could come before his chair. He was responsible for internal order and the safety of the roads, hunting down brigands and putting down risings. He raised and commanded the provincial levies when the King mustered for war. And he held a court in the image of the King's own.^[6][11][4] The working record puts flesh on each duty. Herodotus has Artaphernes, satrap of Sardis, survey the Ionian lands in parasangs and fix their assessments after the Ionian revolt; a Persepolis tablet shows the same Artaphernes despatching a party of Ionians from his province to the King, a transfer of workers only a satrap could authorise; and the building inscriptions of Susa, which name the peoples who quarried, carted and brick-laid the palace, presuppose satraps delivering labour as well as silver.^[5]

The men who held the greatest provinces were, as a rule, Persians of the highest birth, Achaemenid princes who had not reached the throne and members of the privileged houses; the office was a thing given by the royal court, not inherited. In the lesser units beneath them, by contrast, posts might pass within a family or rest with a local dynast, an arrangement Jacobs calls "regulated autonomy"; on his reconstruction it was above all the satraps of the great old capitals, Sardis, Babylon, Memphis, Dascylium, who were always freshly appointed from court, while lesser governorships were left to run in native or immigrant Iranian families.^[6][5]

Even at that height, the King's language held the satrap fast. Darius's inscriptions call his mightiest servants his bandaka, his "bondsmen", men bound to him by a freely accepted, oath-sealed subordination (the word is not that of a chattel slave but of the highest nobles in the King's proximity).^[3] The greatest viceroy in the empire was still, in the King's own word, bound. Briant's summary is that the satrap was no mere civil servant but depended personally on the King, and "had to behave as a faithful bandaka".^[11] The rest of that vocabulary calls for the same caution. Even fratama, "foremost", which at Behistun names the "principal supporters" of the Lie-Kings and has been read as a formal court title in a few Persepolis Treasury letters, is on close inspection probably no title at all, any more than the "foremost of the Persians" who people the Greek narratives held a fixed office.^[12]

Xenophon has Cyrus command his new satraps to model their establishments on the royal court, and the reality was close enough: the satrap's residence was a court-in-miniature, with its own guard, chancery, table and ceremonial, its rules copied from the central court's own.^[11] Jacobs puts it structurally: the sheer range of the satrap's responsibilities turned his seat into a branch of the royal court.^[5] The Greeks, who waited in its anterooms and begged at its gates, recorded the effect. Every satrapy kept at least one paradise, the walled park that advertised the ruler's command of land and water; Xenophon's description of Pharnabazus's seat at Dascylium catches the style:

That is where Pharnabazus maintained his residence, with handsome, large villages all round, abundantly provided with all the resources, and with game both in enclosed paradises and in the open spaces—magnificent game! Through the whole length flowed a river stocked with every kind of fish.

Xenophon, Hellenica 4.1.15–16, as given in Briant^[11]

The scale even of a minor governor's household can be measured from the book of Nehemiah, governor of little Yehud, whose daily table took an ox, six choice sheep and fowls, with wine every ten days, to feed a hundred and fifty regular companions.^[13] A great satrap's court dwarfed that. The point of the splendour was not comfort alone: the satrap was the King, locally, and had to look it.

The workings of a great satrap's world come into close view in one find: thirteen letters on leather, bought for the Bodleian Library in 1943–4, from the files of Aršāma, an Achaemenid prince who was satrap of Egypt for some four decades under Artaxerxes I and Darius II. Most were written while the satrap was away from his province, at Babylon or Susa, to his Egyptian estate-managers (pqyd), Psamšek and then Nakhtḥor, and they offer, in their editors' words, "a vivid snapshot of social, economic, cultural, organizational, and political aspects of the empire as lived by a member of the Achaemenid elite and his entourage".^[14] What they show first is that the satrap was also a great landlord, with estates scattered through Egypt and along the route to it: the celebrated travel warrant in the dossier feeds Nakhtḥor's party "from my estate which is in your provinces", province after province, the length of the journey (for the warrant and the road system it rode on, see the Royal Road).^[14]

They show, second, what the master demanded of the men who ran those estates. Writing after a revolt had swept Egypt, Aršāma holds up the conduct of a former manager: "when the Egyptians rebelled, then Samšek the previous official, our personnel and goods … [which] are in Egypt he guarded with force, so that there was not any loss from my estate".^[14] Other officials' men, he has heard, are doing the same in the present disturbances, and adding to their lords' estates; Nakhtḥor is not. The rebuke that follows is the very voice of the absentee grandee:

Guard [my] personnel and goods forcefully, so that there shall not be a[n]y loss from my estate. Also, from elsewhere, personnel of artisans … seek (in) sufficient (numbers), and bring (them) into my courtyard, and mark (them) with my brand, and make (them) over to my estate.

Aršāma to Nakhtḥor (TADAE A6.10), trans. Taylor^[14]

Fail, the letter ends, and "you will (all) be questioned forcefully, and a severe sentence will be produced".^[14] An empire's provincial government, seen from inside, is here indistinguishable from the management of a great private fortune: branded slaves, requisitioned craftsmen, rents pursued across a rebellion. The office and the estate fed each other, and the letters never trouble to separate them. Even the satrap's seal tells the story of inherited standing: the magnificent heirloom seal Aršāma used on these letters had served a namesake, a son of Darius I and Artystone, two generations before.^[14]

A second dossier shows the office from below. Three demotic papyri from Elephantine preserve an exchange, in 493 to 492, between the wab-priests of the ram-god Khnum and "Pherendates, to whom Egypt is entrusted, says to all the wab-priests (of) Khnum, lord of Elephantine" runs the address of his reply, the demotic scribe's rendering of the satrap's office.^[15] The priests had notified the satrap that they had chosen a new lesonis, the temple's chief administrator. Pherendates writes back rejecting earlier candidates (one had fled; one was another man's servant) and laying down the rule:

Now, the wab-priest whom it is suitable to make lesonis is (a) great man whom, it will happen, I will cause to carry out his functions, there being nothing which he has let fail, one who will be selected in accordance with that which Darius (the) Pharaoh has ordered.

Pherendates to the priests of Khnum, 492 BCE^[15]

The satrap, that is, vetted the head of an Egyptian temple's administration, by criteria the King himself had fixed; Darius appears throughout as "Darius (the) Pharaoh", the conqueror's dynasty wearing the local crown. And the letter closes exactly as Aršāma's Aramaic letters close, naming the Persian officer who "knows this order" and the scribe who wrote it. That formula is no coincidence. The same administrative endorsements recur in the Aramaic of Egypt and Bactria, the demotic of Elephantine and the Elamite of Persepolis, and in Tavernier's judgement they reveal "a coherent system, imposed by the Achaemenid kings and used throughout the existence of their empire"; the notes appear only on letters from a satrap or his office, never on letters to him.^[16] Beneath the babel of local languages ran one chancery discipline, and the satrapy was its unit.

Tribute was the empire's great harvest, and gathering it was the satrap's first duty. The Greeks distinguished two renders. Phoros, assessed tribute, was paid by the fully subject peoples: a fixed yearly sum, largely in weighed silver, set for each province from its measured land. Dora, gifts, were brought by peoples on the empire's fringes who kept a measure of autonomy, the render marking a lighter, more honourable bond; the gifts, in Klinkott's analysis, were at once additional honours to the King and a tax-like render from peoples outside the regular administrative structure.^[9][17] Herodotus's ledger keeps both registers, and adds a third party: "The Persian country is the only one which I have not recorded as tributary; for the Persians dwell free from all taxes."^[7] The ruling people rendered in kind and in men, never in assessed silver. The gift-bringers he itemises with an ethnographer's relish: the Ethiopians beyond Egypt brought every other year "a gift of two choenixes of unrefined gold, two hundred blocks of ebony, five Ethiopian boys, and twenty great elephants' tusks".^[7]

The assessed register was heavier. "Babylon and the rest of Assyria rendered to Darius a thousand talents of silver and five hundred boys to be eunuchs"; India, the richest province, paid in gold-dust.^[7] The whole account then closes in a single audited-sounding sum:

Therefore it is seen by adding all together that Darius collected a yearly tribute of fourteen thousand five hundred and sixty talents; I take no account of figures less than ten.

Herodotus 3.95, trans. Godley^[7]

The figure is enormous, and tempting, and it must be handled with care. What the administrative record confirms is not Herodotus's arithmetic but the texture of the thing: much of the empire's wealth moved not as coined silver but as goods. The Persepolis Fortification tablets, the crown's own accounts, are full of renders and rations in grain, wine and livestock; the Elamite documents record small cattle received as state tax in Darius's own reign.^[9] Strabo preserves the render of a province reckoned wholly in beasts:

Cappadocia paid the Persians yearly, in addition to the silver tax, fifteen hundred horses, two thousand mules, and fifty thousand sheep, whereas Media paid almost twice as much as this.

Strabo, Geography 11.13.7, trans. Jones^[18]

The tribute did not, for the most part, circulate. It flowed inward to the treasuries of the capitals and was largely stored, bullion melted into jars rather than issued as coin, to be drawn on for the King's gifts, wars and works.^[11] The satrapy system was, among other things, a machine for gathering the produce of a continent to the centre.

Herodotus's thousand talents cannot be audited, but the Babylonian archives show the machinery that stood behind any such number, and it is thicker and stranger than a tribute-line suggests. Babylonia entered the empire whole: its first satrap, Gobryas, governed "Babylon and Across-the-River", a domain nominally coextensive with the old Babylonian empire, and only later, perhaps by 503 and certainly by 480, was the great province split in two.^[10] Within it, the crown planted a system of military land-tenure: allotments called "bow land", "horse land" and "chariot land", each owing the service its name declares, held by families of colonists and natives grouped into organisations (haṭru) under a superintendent who allocated the fields and collected the dues; the service owed could be, and increasingly was, commuted into an annual tax, the ilku, paid mainly in silver.^[10] The Murašû archive from Nippur, the file of a family firm that leased such lands, advanced the silver and moved the crops, shows the whole pyramid at work in the later fifth century: control ran upward from the firm's tenants through canal-managers and estate-stewards to members of the royal house, and not to the satrap alone.^[10] Stolper enters a caution here: "There are no extant royal or satrapal archives from Achaemenid Babylonia", so the view is from the paying side, not the collecting one.^[10] But it is enough to correct the picture of a province simply weighing out silver once a year. Tribute rested on tenure, service, water-rights and credit, administered continuously; the terminology of the bow-land system, first attested in the earliest years of Persian rule, was still in use under the Seleucids.^[10]

The tribute flowed in; the King's bounty flowed out, and the two together were the empire's circulation of loyalty. The satrap and the great servant were bound to the King not by pay alone but by gift: robes of honour, gold necklaces and armlets, decorated weapons, horses with golden bridles, grants of land and whole cities, and a place at the King's table. Plutarch preserves the shape of it in the cities the King gave an honoured suppliant, "Magnesia, Myus, and Lampsacus, to maintain him in bread, meat, and wine", one city for each: favour at this court was measured in the revenue of provinces.^[19] To render faithfully and be rewarded richly was the satrap's proper course. Xenophon, idealising, sets out the whole accounting a governor answered to:

These officers the king delights to honour, and showers gifts upon them largely. But as to those officers whom he finds either to have neglected their garrisons, or to have made private gain of their position, these he heavily chastises, deposing them from office.

Xenophon, Oeconomicus 4, trans. Dakyns^[20]

A governor who kept his province full and guarded was raised; one who let it slip, or lined his own purse from it, was cast down. That exchange was very nearly the office's whole rule-book, and the same author supplies the ideal it aimed at, in his portrait of Cyrus the Younger's Anatolia: "Throughout the satrapy of Cyrus any one, Hellene or barbarian, provided he were innocent, might fearlessly travel wherever he pleased, and take with him whatever he felt disposed."^[21] Safe roads and full fields were what a satrapy was for.

Against the danger built into so powerful an office, the crown was believed to set a system of checks, a rough separation of powers meant to keep the satrap from becoming a king. The garrison commander of the province's fortress, and its senior financial officers, could answer directly to the court over the satrap's head, so that the man who held the province's civil power did not automatically command its soldiers or control its treasury. A royal secretary, posted into the satrap's own writing-office, was the King's ear in the chancery. And royal inspectors toured the provinces, the officers the Greeks made famous as "the King's Eye" and "the King's Ears", who came unannounced to see how a province was kept and carried word back to the court.

How real, and how systematic, all this was is genuinely uncertain. The neat "separation of powers" comes largely from Xenophon, a Greek writing an idealised picture, and in hard practice the great western satraps often held civil and military power together.^[20] The "King's Eye" in particular may owe as much to the Greek imagination, which loved to picture the King's realm as a place of watchers and informers, as to any single Persian office; scholars dispute whether it was one high post, a class of travelling inspectors, or partly a literary motif.^[11] What is not in doubt is that some real oversight existed, and the texture of it is better documented than the spies. A satrap could take no military or diplomatic initiative without royal leave: when Aristagoras of Miletus brought Artaphernes a scheme against Naxos, the satrap approved it in principle and then, in Herodotus's telling, added the operative sentence, "The only other thing we need is to get the king's approval."^[11] Treasurers and storehouse-keepers released nothing without central instruction; coin was not the satrap's to strike on his own authority; satraps were routinely denounced by their neighbours and summoned to court to answer.^[11] The letter Darius (if the Greek copy is genuine) sent the official Gadatas, threatening him for taxing the gardeners of a sanctuary of Apollo, shows the same reach downward into a province's conduct: "thou shalt know the anger of a king wronged, unless thou changest thy course."^[22] The sealed authorisations that the Persepolis tablets required for travel and disbursement are the machinery of a state that watched its servants, and they catch a great satrap at the paperwork from the other end: the same satrap's writ, seen once already in the Persepolis record, turns up again in the ration accounts, where parties of workers moving between Sardis and the heartland drew their food on documents authorised by Artaphernes, the provincial seal answering to the central ledger.^[23] The Royal Road and its relay carried the whole system's correspondence, and a summons to court, at the empire's fastest speed. The glamour of the spy-network is Greek; the paperwork of control is Persian and real.

The checks existed because the danger was real, and the reign of Darius furnishes its textbook case, told at length by Herodotus. Oroetes, satrap of Sardis since Cyrus's day, is the over-mighty viceroy entire. He lured Polycrates, tyrant of Samos, to the mainland and crucified him. Through the chaos that followed Cambyses's death he helped no one and settled scores: "he slew two notable Persians, Mitrobates, the governor from Dascyleium, who had taunted him concerning Polycrates, and Mitrobates' son, Cranaspes"; and "when a messenger from Darius came with a message which displeased him, he set an ambush by the way and killed that messenger on his journey homewards, and made away with the man's body and horse."^[24] A satrap who murders the King's courier has, in everything but name, seceded.

Darius, newly on the throne, dared not move openly: Oroetes kept a thousand Persian spearmen and held Phrygia, Lydia and Ionia together. So the King put the problem to his council in terms that define the office's pathology:

Who is there among you, men of Persia, that will undertake and achieve a thing for me not with force and numbers, but by cunning? Force has no place where cunning is needful.

Herodotus 3.127, trans. Godley^[24]

Thirty volunteered; the lot fell on Bagaeus. His device turned the crown's paper power against the satrap's armed one. Coming into Oroetes's presence at Sardis, he handed a series of letters sealed with Darius's seal, one by one, to the royal scribe who attended the satrap, watching the guards' reverence for the documents grow. Then, "Persians! King Darius forbids you to be Oroetes' guard", and the spearmen lowered their spears; then the last roll: "King Darius charges the Persians in Sardis to kill Oroetes." "Hearing this the spearmen drew their scimitars and killed Oroetes forthwith."^[24] The story is Greek, and shaped as a moral tale ("Thus was Oroetes the Persian overtaken by the powers that avenged Polycrates of Samos"), but its mechanics are exact: every check the system possessed is in the scene, and the satrap's own bodyguard obeyed the seal over the man, so that the King recovered a province without an army.

The danger did not end with Darius, because the office tended, against the crown's design, to slide toward the hereditary. In Hellespontine Phrygia the descendants of Pharnaces held Dascylium for four generations; Briant, weighing the evidence, judges the Pharnacids a highly exceptional case and insists that appointment and dismissal remained the King's prerogative throughout, but the exception shows the pull.^[11] In Caria the native Hecatomnid dynasts simply were the satraps, father to son. The fourth century then supplied the institution's supposed collapse: the great "satraps' revolt" of the 360s, when, in Diodorus's telling, the western governors rose together and the empire tottered. Here the sources deceive. Weiskopf's study argued that the single coordinated revolt is a construct of Diodorus and never took place, dissolving on inspection into separate, local insubordinations, and Jacobs endorses the demolition; Briant, more cautiously, holds that Diodorus's fresco of an empire in flames arouses the greatest reservations.^[5][11] Real satraps did rebel, singly and dangerously. What the evidence will not support is the standard image of a provincial system rotting into general revolt; the machine, crisis by crisis, kept working, and its crises are in fact where its automatic responses come into view, the denunciations, the summonses, the sealed letters, doing what they were designed to do.^[5]

The tidy answer to how many satrapies the empire held, twenty, the number every handbook repeats, is precisely the figure the field trusts least, and the account most books still give needs correction here.

The trouble is that the sources do not agree, and are not even counting the same kind of thing. Herodotus's tribute-list gives twenty districts. The Behistun inscription of about 520 names, at most, twenty-three lands (dahyāva). Xerxes's later inscription XPh lists thirty-two; Darius's tomb-inscription at Naqsh-e Rostam names about thirty. These lists do not match one another, and they are different kinds of list: some are fiscal, some are the ideological roll of "the lands I hold" carved for display, where each people appears as a throne-bearer beneath the King. Darius himself, on his tomb, tells the reader to look at the picture rather than trust a number:

If now thou shalt think that "How many are the countries which King Darius held?" look at the sculptures (of those) who bear the throne, then shalt thou know.

Darius I, DNa §4, trans. Kent^[25]

Modern scholarship has largely turned against Herodotus's list here, and hard: in the current Companion's summary, the list "should not be taken at face value for the structural history of the Persian Empire", its districts called nomoi rather than satrapies, its order Greek and Babylonian rather than Persian, and its contents matching neither the army-list nor the royal dahyāva rolls.^[26] Klinkott's monograph on the office reaches the same verdict from the German side: the tax-list is scarcely usable as history for the Achaemenid administration, its nomoi cannot be equated with the satrapies as administrative units, and it fails precisely as a register of satrapies.^[4] Jacobs, whose is the authoritative reference treatment, holds that the list is not usable as historical testimony at all: its arrangement is chaotic and conflicts with geography, it wildly over-weights the western regions (a sign Herodotus had no authentic Persian source), and its data conflict with every other source, Greek, Latin and Persian alike.^[6]

The Old Persian dahyāva lists, by contrast, Jacobs treats as the primary evidence, and argues they were long misread. The older idea that they were lists of peoples (so that the empire was imagined as a sum of nations, not a territory) is, he holds, now refuted, for dahyu denotes a territorial unit, a country in the administrative sense; and the roster of names is not decorative. Set beside the Persepolis Fortification archive, the lists turn out to match, item for item and with explicable exceptions, the destinations and starting-points of official travel and the ethnic labels of the archive's workforce: the King's display-lists and the administration's working paper drew on the same register of provinces.^[5] He reads the swelling numbers in the later lists not as carelessness but as hierarchy and real change: the Behistun list gives the standard administrative units of its day (his "Main Satrapies"), grouped under a handful of "Great Satrapies" and subdivided into "Minor" ones, while the later inscriptions add genuine new conquests (the lower Indus; Thrace, Libya and Nubia) alongside the flattering fiction that the King was forever enlarging his lands.^[6] The registers of the Alexander historians close the frame from the far end: the provinces where Alexander confirmed or replaced governors map, with few discrepancies, onto the old dahyāva rolls, a congruence that in Jacobs's words proves the satrapal administration "a coherent system and one that remained robust over a period of more than 200 years".^[5]

Jacobs's full reconstruction is his own and not universally followed, but its governing negative lesson is broad consensus: there is no single clean canonical map of "the twenty satrapies". The number was not fixed, it changed over time and by the kind of unit counted, and the neat scheme is a schematisation, Greek or modern, laid over a messier and more interesting reality.^[4]

The satrapy outlived the dynasty that made it. Alexander conquered the empire satrapy by satrapy and governed it the same way, confirming sitting satraps or appointing his own, Persian grandees among them; the lists of provinces his successors divided at Babylon and Triparadisus are, in effect, the last edition of the Achaemenid provincial register, which is why the Alexander historians are a primary source for the system they inherited.^[5] Briant's verdict on the deeper continuity is that, from the standpoint of Near Eastern imperial geopolitics, Alexander was "the last of the Achaemenids".^[11] The texture persisted with the title. Peucestas, Alexander's satrap of Persis, learned Persian and adopted Persian dress to govern Persians; in Babylonia the fiscal vocabulary of the bow-land system ran on into the Seleucid era; and in Cappadocia and Armenia, where royal control had frayed earliest, the local satrapal houses simply became kings, still parading their Achaemenid connections, one line styling itself in a Greek and Aramaic inscription with the nostalgic title of "legitimate satraps".^[11][10] The Greek word satrapēs went on to a long career under the Seleucids and Parthians, and survives in modern languages as a byword for a provincial potentate, which is, in its way, the office's afterlife in one word.

Seen whole, the satrapy was the empire made governable: a frame that let one King tax, judge and defend a quarter of the known world, and hold its many peoples in a single order. It was also the empire's standing hazard, for the power that held a province could turn against the throne that granted it. The institution was two-edged to the end, the protector of the realm who was also its greatest danger, the man set to hold a province for the King being always the man best placed to seize it.