Darius I

Darius was born, on Herodotus's reckoning, about 550 BCE, the eldest son of Hystaspes (Vištāspa), and before his accession he served Cambyses as a spear-bearer in Egypt. He was not the obvious heir, and the opening of the Behistun inscription sets out to prove he was an heir at all:

My father was Hystaspes; the father of Hystaspes was Arsames; the father of Arsames was Ariaramnes. Eight of my line were kings before me; I am the ninth. In two lines have we been kings.

Darius I, DB §§3–4^[2]

The chain runs on past Ariaramnes to Teispes and to a founding ancestor, Achaemenes, from whom Darius claims both his own line and Cyrus's descend. On its face this is a statement of legitimate continuity. Almost every part of it now looks like a construction of 522.

The clearest check on it is Cyrus's own inscription. On the Cyrus Cylinder, written at Babylon in 539, Cyrus traces himself back as "son of Cambyses ... descendant of Teispes ... king of Anshan," and stops. He never names Achaemenes, and he calls himself and his forebears not "Achaemenid" but "king of Anshan," the old title of the kingdom of Fars. Achaemenes, as Kuhrt notes, "appears as founder of the Persian dynasty only from the reign of Darius I, whose genealogical statement is highly problematical."^[3] The ancestor meant to join the two houses is absent from the one house supposed to descend from him. Scholarship now marks the gap with a distinction the ancient texts never make: Cyrus's line is best called the Teispids, the kings of Anshan; the Achaemenids, named for Achaemenes, begin with Darius.^[4][5]

Darius, in Kuhrt's reading, supplied "the Achaemenids with an eponymous founder ... through whom he could link himself more closely to the founders of the empire," while being "almost certainly a member of the extended clan" but no close relative of Cyrus.^[3] Briant is blunter: Achaemenes as a founder-figure was "invented out of whole cloth," and the word Achaemenid was narrowed from a wide clan to Darius's own patriline, which conveniently shut his fellow conspirators out of the succession.^[6] This is not the old charge of usurpation, which Briant rejects: none of the six who seized power with Darius had any better dynastic claim than he did. It is something more deliberate, a re-founding of the monarchy that gave itself the ancestry it needed after the fact.^[6]

The supports for the old pedigree have gone one by one. The two gold tablets in which Darius's grandfather Arsames and great-grandfather Ariaramnes call themselves "great king, king of kings" are now taken as forgeries: their grammar is later than Darius's own, and a minor seventh-century lord of Fars would not have used the imperial title that Darius himself invented.^[7][5] The Pasargadae inscriptions that call Cyrus an "Achaemenid" are widely given to Darius instead, who is thought to have added the label to his predecessor's monuments to claim him for the new line; the case, with its minority view, is set out in the entry on Cyrus.^[6][5] Even the "in two lines" is unsafe: the Old Persian duvitaparanam is now more often read "in succession," and the Babylonian version drops the branches altogether for "nine kings of an eternal lineage."^[3]

None of this makes Darius an outsider. He belonged to the royal clan, married into Cyrus's family, and kept up the funerary offerings for the dead Cambyses rather than erasing him; the Persepolis tablets show him absorbing Cyrus's line, not blotting it out.^[8] The pedigree is not a simple lie. It is a real member of the clan claiming a royal descent he had largely built to fit, and it is the first move of a reign that would keep building its legitimacy in stone. Descent from Achaemenes stayed the central test of Persian kingship for two centuries, enough that it would later be held against even Alexander.^[9]

By Darius's account the throne had been stolen before he took it, and putting the thief down was what made him king. He tells the story at the head of Behistun. Cambyses, son of Cyrus and king before him, had a full brother, Bardiya (Smerdis to the Greeks), and killed him in secret before leaving for Egypt:

Afterwards, Cambyses slew that Smerdis. When Cambyses slew Smerdis, it did not become known to the people that Smerdis had been slain ... After that the Lie waxed great in the country, both in Persia and in Media and in the other provinces.

Darius I, DB §10^[2]

With the prince dead and unmissed, the way lay open for a fraud. While Cambyses was in Egypt, a magus named Gaumāta, an ordinary Iranian herdsman's name meaning "possessing cattle,"^[1] rose in Persis in March 522 and gave out that he was the murdered Bardiya. The people went over to him, he took the throne, and Cambyses died, "of natural causes" in the reading now preferred, where the Old Persian says only that Darius had no part in the death, not that the king took his own life.^[2][6] Gaumāta held power by terror, Darius says, killing anyone who might know the true Bardiya was already dead, and none dared oppose him "until I came." On 29 September 522 Darius and a small band ran him down:

I with a few men slew that Gaumata the Magian, and those who were his foremost followers. A fortress by name Sikayauvati, a district by name Nisaya, in Media, there I slew him. I took the kingdom from him. By the favor of Ahuramazda I became king.

Darius I, DB §13^[2]

Then the restorer's work: Darius puts back what the false king had torn down, the sanctuaries, the pastures and herds, the households, "as it was in the days of old."^[2]

One item in that list carries the story's only religious charge, and it resists every reading of the accession. The sanctuaries, the āyadanā that Gaumāta destroyed and Darius restored (DB §14), are a strange thing for a usurping magus to have attacked, and Heidemarie Koch has argued that the notice must rest on some fact: Darius set it down within months of the events, and no Near Eastern king had ever accused a predecessor of destroying, rather than merely neglecting, sanctuaries, so the charge is not a stock formula. But the fact will not resolve. Either this Bardiya swept away time-honoured cult places out of Zoroastrian zeal, or he was a defender of the old religion attacking new-fangled Zoroastrian fire altars; the evidence allows both, and, as Koch concedes, the questions must all stay open.^[10]

It is a clean and complete story, which is the trouble with it. Behistun is the only contemporary account of these events and, at the same time, the document that justifies Darius's seizure of a throne he had no clear right to. Kuhrt calls it "the primary source for what happened" and "profoundly and deliberately misleading" in one breath.^[3] The whole of it rests on a single claim that no one but Darius makes: that the real Bardiya had been dead for years, so that the king ruling in 522 could only be an impostor. Remove that claim and the story turns inside out. The man Darius killed in Media was then the true son of Cyrus and the rightful king, and Darius is not the restorer of the throne but the one who took it by force. Once a heresy, that reading is now the majority view.

The doubt is as old as the inscription, but the modern case against Darius rests on things he could not control. Kuhrt lays it out as a string of oddities the king's "best efforts fail to hide."^[3]

The first is a silence. Darius fixes almost every event of the crisis to the day, the risings, the battles, the killing, and leaves unspecified the single date a chronology of the crisis would need, Cambyses' death. As Briant observes, "Darius is careful not to specify the date of the king's death, even though other events are quite consistently fixed to the month and the day throughout his narrative."^[6] The blur lets him suggest that Gaumāta usurped the throne of a living king, rather than filled the vacancy left by a dead one.

The second is the Babylonian evidence, which Darius did not write. Business tablets from Babylon date by Cambyses until late April 522 and then by "Barziya," styled "king of Babylon, king of the lands," with no trace of a March revolt or a contested succession.^[6][3] To the scribes who kept the empire's accounts, Bardiya was simply the legitimate king who followed his brother. A magus impersonating a dead prince does not leave that record.

The third is the killing. Every later rebel in the inscription ends the same way, beaten, mutilated, and put on show, the proof of victory carved into the account. Gaumāta alone dies quietly, in a Median fortress, killed by a few men, with no body displayed. Kuhrt draws the inference: a king exhibits a defeated rival to prove he beat him, and hides the corpse only when showing it would reveal that he has killed the real prince.^[3] Nor did the imposture convince many at the time. The Persian nobles around Cambyses, Herodotus reports, "were convinced that it was Smerdis the son of Cyrus who was on the throne," and years afterward the rebel Vahyazdāta could still raise armies merely by claiming to be Bardiya, which he could hardly have done had Darius shown the name to be a lie.^[6][3]

On this evidence the balance has swung hard. Kuhrt states flatly that "with very few exceptions, most scholars do not believe Darius' account," and concludes that "it is, therefore, Darius himself who is the true rebel."^[3] Briant is more guarded, calling the deception "these days generally accepted" while insisting that "nothing has been established with certainty," so that the historian is "reduced to arguing for probabilities and choosing the option that appears the least uncertain"; once the secret prior death of Bardiya is doubted, in his phrase, "the entire structure collapses like a house of cards."^[6] The holdouts are real but few: Shahbazi, in the standard Encyclopaedia Iranica article, still defends Darius point by point, and an older tradition running back to Herzfeld read the inscription as plain fact.^[11][12]

The case is strong but not closed, and its shape should be stated honestly: it turns on one contemporary source written by the man it exonerates, and the counter-evidence, though it points consistently one way, is circumstantial. What can be said is that the founding story of the reign, the impostor, the rescue, the restoration by the grace of the Wise Lord, is most likely a screen built over a coup, and that Darius reached the throne by killing a king with a better right to it than his own. Why he did so is harder to recover. Kuhrt calls the motive "almost unanswerable," and the one concrete clue is Darius's own boast to have given back the "pastures, herds, households and houses" the usurper had seized, which may mean that Bardiya's brief reign had begun to bite into the property and standing of the Persian nobility, and that the nobles, Darius among them, killed him for it.^[3] The 1988 Cambridge Ancient History had already worked the clue out in detail. Herodotus reports that the false Smerdis won wide support by relieving the peoples of the empire of taxes and military service for three years; and in the Babylonian version of Behistun, where the Old Persian and Elamite say simply that Darius restored the "houses", the term is "bow-houses", a word that appears in legal texts with the particular meaning of fiefs assigned to state dependants in return for taxes and military service. Read together, the remission and the restoration become one policy reversed, and the struggle, in Cuyler Young's words, "pitted those who had been on top economically and socially, not all of whom were Persians, against those who under Gaumata would have reversed the existing order of society".^[13] The same volume aired a further reading now largely set aside: the Babylonian version labels Gaumāta a Mede, and his magianship and his Median refuge were read as perhaps "an attempted Median counter-coup to that of the Persians, under Cyrus", a last struggle for primacy between the two leading Iranian peoples. The inference from magus to Mede is itself contested and the conjecture has not worn well, but it shows how much political weight the affair could be made to carry.^[13]

Killing the impostor made Darius king; it did not make the empire his. Within weeks the provinces rose one after another, until almost the whole of Cyrus's realm was in revolt at once, Elam, Babylon, Media, Persis, Parthia, Margiana, Sagartia and more. Darius spent his first year putting them down, and he set the reconquest at the centre of his record:

This (is) what I have done, by the favour of Auramazda, in one and the same year, after I became king. I have fought nineteen battles. By the favour of Auramazda, I defeated them and took nine kings prisoner.

Darius I, DB §52^[2]

Each rebel, in Darius's telling, was a liar wearing a dead king's name. Nidintu-Bel of Babylon gave himself out as Nebuchadnezzar, son of Nabonidus; Fravartiš of Media claimed descent from the old Median king Cyaxares; Vahyazdāta, in Persis, claimed to be the very Bardiya that Darius had just killed; and Arakha, in a second Babylonian rising, took the name Nebuchadnezzar over again. Every one of them, in the inscription's language, followed the drauga, the Lie, and Darius, the servant of the Truth, put them down by the grace of Ahura Mazdā, their images roped together by the neck beneath his feet on the relief.

The two Babylonian risings can be dated closely, and set beside a famous Greek story they expose it. Nidintu-Bel was crushed by Darius in person in two battles, on 13 and 18 December 522, and executed; Behistun records no razing of the city walls, and Darius stayed on in Babylonia into 521 to secure his hold. Two months after he left, Arakha rose from Ur in the south, with texts dating by him there from late August 521 while Sippar still dated by Darius, until Persian forces under Intaphernes broke the revolt on 27 November 521 and Arakha was impaled at Babylon with his nobles, his supporters counted at 2,497. Herodotus's celebrated account of a long siege of Babylon, its recapture through the self-mutilating ruse of Zopyrus, and the destruction of its gates (3.150 to 159) cannot, Kuhrt observes, be connected with either rising on factual and chronological grounds, though his impaling of the leading citizens "to the number of 3,000" may be a dim reflection of Arakha's fate.^[13]

He did not fight everywhere himself. The war was run across fronts by his commanders: his father Hystaspes held Parthia, Dādarshi and Vaumisa fought five hard battles in Armenia, the satrap Vivāna held Arachosia in the far east, while Darius took the central prizes of Babylon, Media and Persis. One dated eastern episode shows what the reconquest meant on the ground. In Margiana a certain Frāda had been proclaimed king; another Dādarshi, a Persian and Darius's satrap in Bactria, a different man from the Armenian who fought the five battles, crushed him on 10 December 522. The record, in the Babylonian and Aramaic versions where the count survives, sets down 55,000 Margians killed and 8,500 taken captive, and the Cambridge Ancient History, which transmits the figures, judges them in its own notes "impossibly inflated", even while finding most of the Babylonian version's casualty and prisoner numbers plausible; what they attest is the scale Darius claimed rather than a count. It is on that claimed scale that some historians have read Margiana as a popular revolt rather than a dynastic quarrel.^[13] The beaten leaders met the fate the inscription lingers over, mutilation and public display, the nose and ears and tongue cut away, the ringleaders impaled where the people could see them, the exact opposite of the quiet, hidden killing of Gaumāta.

That all of this happened "in one and the same year" is the inscription's claim, and not the historian's. The real fighting ran from late 522 well into 521, with further risings in 520 and 519 that Darius added to the monument later. The single-year compression is a borrowed flourish: the boast of nine kings beaten in one year repeats a formula from the inscriptions of Naram-Sin of Akkad, sixteen centuries older, and the "liar" epithet fixed on every rebel is itself an old Mesopotamian convention.^[4] There is also a plainer defence, entered in the 1988 Cambridge Ancient History: reckoned from Darius's first counter-move against rebellious Babylon in November or December 522, rather than from his accession, the claim is close to literal, and the formula and the calendar need not exclude each other.^[13] The importance of the year of revolts is not in its arithmetic. It is that the empire Cyrus had assembled could fly apart in a season, and that Darius's reconquest, and the tighter system he built once the fighting stopped, is what held it together for the next two hundred years.

The record of all this was cut where it could not be missed and could not be reached. Behistun is the last peak of a range that rises straight from the plain beside the royal road between Babylon and Ecbatana, a place with old sacred associations, and Darius set his relief and its text high on the rock face, out of reach of any hand that might deface them.^[14] Below and beside a sculptured panel run the columns of cuneiform.

The panel is a single frozen scene of victory. Darius stands with his left foot on the fallen Gaumāta, who lifts his hands in vain; the king raises his right hand to the winged figure of Ahura Mazdā in a ring above him and holds a bow in his left; two officers stand at his back, and before him the nine defeated liar-kings are roped together by the neck, their hands bound. The last of them, Skunkha the Saka in his tall pointed cap, was added years afterward, and the cap makes him the tallest figure on the rock, a head above the king.^[14]

The text is where the monument shows its seams. It was carved in stages, and in more than one language. The Elamite and Babylonian versions were cut first, flanking the relief; the Old Persian was added afterward, below, and the whole had to be recut when Skunkha and a fifth column were forced in.^[14][5] The order is telling: Elamite was the empire's established language of inscription and Old Persian was the newcomer. Darius says as much, in a passage long argued over, where he claims to have made the writing itself:

By the favor of Ahuramazda this is the inscription which I made. Besides, it was in Aryan, and on clay tablets and on parchment it was composed ... Afterwards this inscription I sent off everywhere among the provinces.

Darius I, DB §70^[2]

Most scholars take this as the announcement of a purpose-built script, the Old Persian cuneiform, the "Aryan" writing, created for this monument; the Elamite label beside the relief adds pointedly that it was made "as previously was not."^[3][5] A minority doubts that a king would boast of so technical a thing, and the passage is genuinely hard to construe; some philologists go further, holding with Pierre Lecoq that the made-for-Darius theory rests on a mistranslation of the Elamite word tuppime, "inscription, text," not "writing," and that the Old Persian script was already in use under Cyrus.^[5][15] But the Old Persian script does appear with Darius and scarcely at all before him, and ever after it served almost nothing except the words of the kings.

The final clause was no empty flourish. Darius says he sent the text "everywhere among the provinces," and he did. Copies of the words and of the image travelled the empire, and two survive: an Akkadian version cut on stone at Babylon, set on the palace wall facing the processional street, and an Aramaic version on papyrus from the Jewish garrison at Elephantine in Egypt.^[14][3] The Babylon copy was fitted to its audience, for where the rock at Behistun credits Ahura Mazdā, the Babylon stele puts the city's own god, Bel-Marduk, in his place.^[3] Darius's account of himself, in Kuhrt's phrase, "became 'the true history'" across the empire.^[3] The circulation carries a methodological sting, drawn already in the 1988 Cambridge Ancient History: since the king broadcast the text empire-wide, and "there may have been a version available in Halicarnassus, the birthplace of Herodotus", the Greek historian's agreement with Behistun is arguably the inscription echoing back rather than a second witness, so that "caution requires that we not consider true what Darius tells us at Bisitun on the grounds that the story is confirmed independently in Herodotus"; the historian is, in effect, dealing with a single source.^[13] It was the first great act of imperial communication in Persian history, one story and one image carried to every corner, each version pitched to the people who would read it, and, as the accession has already shown, a story built to be believed.

Cyrus and Cambyses conquered the empire; the question is how much of its government they left for Darius to build. The old answer was simple: Cyrus won the lands, and Darius organised them. On this view Darius was a second founder, the ruler who turned a run of conquests into a governed state with fixed provinces, assessed tribute, a coinage, roads and a chancery. The view has a long pedigree and it is not baseless. But every part of it has been qualified in the last forty years, and the qualifications matter more than the slogan.

The strong form of the older thesis is Dandamaev's. Reading the administrative terms that begin to surface in Babylonian tablets under Darius, the Iranian xšaçapāvan behind the scribes' ahšadrapannu, he concluded that the reforms remade the machinery from the ground up:

Although after the Persian conquest of Babylonia local administrative traditions were not interrupted, gradually significant changes took place both in the administrative system itself and in the terms used to designate officials. These changes were apparently connected with important administrative reforms of Darius I as a result of which an essentially new administrative system was created.

Dandamaev, Achaemenid History VIII, p. 230^[16]

That is the maximal claim: not adjustment but a new system. Even in stating it Dandamaev noted the inherited substrate, that the Achaemenid structures "most of all resembled the Neo-Assyrian structures," reaching the Persians "almost certainly through the Medes."^[16] The reform, on his own account, built on older foundations rather than clearing the ground.

Kuhrt states the moderate position, and it is the one the evidence best supports. A tribute system had to exist under Cyrus and Cambyses already, or the conquests could never have been supplied and manned; what Darius did was tighten it:

Although it is certain that a system of tribute must have existed in the empire's first decades (how else could the kings have raised the manpower and materials needed for their great conquests?), it seems very likely that Darius I introduced, or tightened up, measures for exacting and ensuring its proper and regular discharge.

Kuhrt, The Persian Empire, p. 669^[3]

Her verdict in a phrase: "Darius' act was a rationalisation to ensure regularity of income," and Herodotus's idea that Cyrus took only voluntary gifts is "certainly wrong."^[3] Rationalisation, not creation, is the word that fits the documents.

Briant sharpens the point at both ends. He grants the continuity without reservation: assessment "did not begin with Darius," and Cyrus, Cambyses and Bardiya had all "been careful not to neglect fiscal administration."^[6] Yet he refuses to let the concession collapse Darius's achievement into mere inheritance. The Achaemenid system, tribute and iconography alike, was more than borrowed parts:

it was not a simple juxtaposition of preexisting Near Eastern elements but a genuinely new elaboration ... his system went well beyond simple borrowing of accounting practices. The inverse thesis, which reduces the Achaemenid structure to borrowings from preceding kingdoms, is based, it seems to me, on a historical and methodological error.

Briant, From Cyrus to Alexander, p. 873^[6]

Darius's work, in Briant's reading, was "grafted both onto realities already observable in earlier times and onto the initial tinkerings carried out by Darius's predecessors," and still went beyond them by the organising force it required.^[6] The pure-continuity thesis and the pure-invention thesis are both errors; the truth is a real reform of an inherited thing.

The strongest form of the second-founder view is older than any of these positions, and it reads the revolts themselves as the evidence. Cuyler Young, in the 1988 Cambridge Ancient History, observed that the rebel kings of 522 do not claim the imperial title: "Phraortes declares himself king of Media, not a rival to Darius for imperial power", so that once Gaumāta was dead "the succession crisis is mostly over", and the fighting turned "not on the issue of who will be the next King of Kings" but on whether there was to be a King of Kings, an empire, at all. On that reading the polity of the Medes and of Cyrus had been something less than a real empire, "little more than the gluing together of several regional states through conquest"; when the centre did not hold, "the glue melted, and the structure came unstuck"; and Darius's victories and reorganisation "might perhaps be viewed as the actual creation, for the first time, of a real empire".^[13] The thesis concerns how securely the polity was joined, rather than which offices existed. The same author credits the provincial machinery Darius inherited mainly to Cyrus, and his own summary of the satrapal system, that it "must date at least to the reign of Cyrus" though Darius may have "extensively reorganized and tightened up an existing structure", anticipates Kuhrt's rationalisation almost word for word.^[13]

The maximal sceptic is Bruno Jacobs, and his case cuts against the transformation thesis itself. Jacobs points out that the oldest list of the empire's lands, in Behistun, is one Darius says came to him, not one he made:

The oldest dahyava-catalogue, that in the Bisotun inscription, is called by Darius I a list of countries which came to him (DB para.6), i.e. which he inherited from his predecessors. Consequently there is no room for an administrative reform of the sort that some have extracted from Herodotus 3.89. All apparent indications of change under Darius I by comparison with earlier times should therefore be attributed to a change in the nature of our sources ... rather than to anything having actually changed in the real world.

Jacobs, "Satrapal Administration," Companion, pp. 845–846^[17]

His conclusion is deliberately provocative: "the supposed reform can be removed from the history books."^[17] The argument turns on a fact about evidence rather than about government. Nothing comparable to the royal inscriptions or the Persepolis tablets survives from before Darius, so the empire looks organised under him and formless before him only because the documents begin with him. Jacobs adds a structural argument the sources support. The empire held together through the revolts of 522 to 520, which an unadministered federation could not have done, and Darius's own Behistun names two satraps, Dādarshi and Vivāna, who were in post before his accession and must have been installed by Cambyses at the latest.^[17] If there were satraps before Darius, Darius did not invent them.

Jacobs has since pressed the case further, on new ground. In the 2017 Verwaltung volume he argues that the royal lists are themselves administrative evidence after all: the lands they name can be brought almost completely into correspondence with the ethnonyms attested in the Fortification archive, which makes the dahyāva-lists a source of the first rank for the provincial administration, with the Behistun list already covering the whole empire; and the same list dissolves without remainder into the satrapy-list that can be reconstructed from Alexander's personnel decisions two centuries later, the only gaps being lands Alexander never conquered. For Jacobs that proves a continuity of the administration and its units from Cambyses to the empire's end: the system's character was not continual change and adjustment from the centre but continuity and stability.^[18]

The way through is to fix precisely what was inherited and what was genuinely new, and here the documents are decisive. The satrapy as an office predates Darius: Cyrus's appointee Oroetes held Sardis, and Cambyses' man Aryandes held Egypt, before Darius took the throne.^[4] The redistributive economy of the heartland was older still. The very archive that shows Darius's Persis running itself, the Fortification tablets, records an institution the Neo-Elamite kings had run a century and a half earlier, a central director of collection and redistribution whose office the Persian chief administrator simply took over.^[19] What is documented as new under Darius is narrower and more concrete than "the creation of provincial government." It is fiscal. In Babylonia the reform is visible in the record: land measured and registered, new taxation terms, a heavier and more systematic exaction of the empire's richest province. Kleber, who has read the Babylonian side closely, locates the genuine change there rather than in any founding of the satrapies:

the fact that Darius had local laws gathered in Egypt and reformed the tax system in Babylonia indicates that he aspired to a tighter control over and, perhaps, a more rigorous exploitation of these rich provinces. The tax reform is visible in new taxation terminology, and it is likely that the overall tax burden increased.

Kleber, "Babylonia," Companion, pp. 915–916^[20]

The working parts of that reform are visible in the Babylonian record, and they confirm its shape. Taxpayers were organised in "decuries" under a foreman, the rab ešerti, and paid through a "canal inspector," the gugallu, to the municipal governor; collection was never centralised at the satrapal level. New governors bearing the title pīhātu appear more frequently under Darius, perhaps a new layer of control, but, as Kleber notes, the new layer did not supersede the old: the institutions that collected the taxes and organised the manpower were still run by the same people as before.^[18] The reform intensified an inherited machine; it did not replace it. The earliest Babylonian cadastral texts date to Darius's third year, which fixes this fiscal intervention to about 519.^[7] Alongside it stand a codification of Egyptian law, a standardisation of weights and coinage on the reformed Babylonian shekel, and the perfected road network. That is the defensible shape of the transformation. Darius consolidated, intensified and documented an inherited administration; his real reform was fiscal and regional rather than the invention of the machinery of rule. The picture of Darius the systematiser is partly the reign's own doing, since the empire's records begin with the king who most wanted to be seen ordering it.

The newest synthesis lets the continuity and the transformation halves of this debate meet. Henkelman argues that what the early Achaemenids genuinely added was not the machine but the grasp of it as a machine: they were able to perceive "the cloud of administrative networks, functional hierarchies, logistic protocols and bureaucratic conventions as a system" that could be expanded, copied, and "applied to manage, monitor and develop other regions of their empire, notably in its eastern half," with "a conscious, determined, and systematic planning on the basis of well-understood and well-tested precedent," what he calls the imperial paradigm.^[18] Tested against Bactria and Arachosia, regions he chooses precisely because they lie too far from the centre for organic spread, the parallels of layout and procedure hold: the heartland administration was "applied ... as a model and a strategy" to lands that differed from it in almost every other respect.^[18] An inherited machine, consciously grasped as a system and copied outward: that formulation is stronger than either pole of the older debate, and it does not need the copying to be Darius's work alone.

The empire was governed through satrapies, provinces each under a satrap, and the first difficulty is that the sources will not agree on how many there were. Herodotus gives twenty. Setting out the tribute reform he writes that Darius "proceeded to set up twenty provincial governorships (nomoi), called satrapies ... and assessed [each] for taxes" (3.89), then lists the twenty districts with the sum each owed.^[6] Darius's own inscriptions give a different figure and a different kind of list. Behistun names twenty-three lands, "Persia, Elam, Babylonia, Assyria, Arabia, Egypt ... altogether twenty-three countries," and the order and contents do not match Herodotus.^[21] Twelve of Behistun's countries answer to five of Herodotus's districts; Cappadocia and Arachosia stand in the inscription but not in Herodotus at all.^[7] The two will not be reconciled, and the attempt has a long record of failure: as one recent survey puts it, efforts to combine the Persian and Greek lists "have failed due to the incompatibility of the order and the question of the status of the 'lands' ... versus the 'satrapies'."^[21]

The resolution is that the lists are not the same kind of document, and were never meant to be counted against one another. The classic statement of the case is Cameron's, and it was made from the grammar: most of Behistun's twenty-three "lands" carry the Elamite personal plural marker, so that the royal title in the Elamite versions is "quite properly, 'King of Peoples'," and what the king's scribes were giving in their lists "were less the names of provinces or satrapies, rather the names of peoples whom they deemed worthy of specific recognition." Cameron closed by reviving a sentence of Eduard Meyer's from 1901 that the field had forgotten: the lists of subject peoples in Darius's inscriptions have nothing to do with the administrative division of the empire.^[22] Wiesehöfer's taxonomy is the clearest modern guide: the ethnic, fiscal and administrative structures of the empire were recorded in several distinct registers, and none is an administrative map.^[9] The royal dahyāva-lists of the inscriptions are ideological, the empire displayed as the peoples the king holds; Rollinger reads them as "a mythical worldview," speaking of tribute and vassals rather than taxes and governors, and so representing "not ... 'provinces' but countries/people that visualize the Persian-Achaemenid Empire and its place in the universe."^[23] Herodotus's twenty are a fiscal list, a grouping of neighbouring peoples for assessment; his own three words for them, satrapies, nomoi and districts, are used interchangeably, which reflects a Greek writer's loose terminology rather than an administrative distinction.^[5] The actual satrapy-lists that a modern map would need survive only from after the empire fell. The philologists put the mismatch in figures: only about half of Herodotus's districts and barely a third of the Old Persian lands correspond to the historically attested satrapies.^[5] The number problem is therefore not a puzzle to be solved but a category error to be avoided. Darius's tribute-list is real administrative data seen through a Greek lens, not an organigram, and the count of "satrapies" depends entirely on which list one reads. One caution belongs here: the "not an organigram" verdict is the position of Cameron, Rollinger and Wiesehöfer, and it was settled Cambridge doctrine between them, the 1988 Cambridge Ancient History ruling that the royal lists "are not lists of official administrative units" and that efforts to reconcile them with Herodotus, or to read their changes as ongoing provincial reorganisation, could be safely set aside; the same chapter added that the satrapies themselves "were often not precisely defined geographically, nor were they completely contiguous districts", the whole structure carrying "a kind of vague looseness about it which would be unacceptable in modern political geography".^[13] But the verdict now has a named, current dissent, since Jacobs's 2017 argument set out above rehabilitates the same lists as first-rank administrative evidence. Ideological presentation and administrative substance are both on the table, and the question is no longer closed by consensus.

One province lets the reform be tested against dated documents, and its answer is late. In Babylonia the governor Ushtani still bears the old combined title "governor of Babylon and Across-the-River" from the close of Darius's first regnal year into his sixth, that is to 516, so that for the empire's richest satrapy the reorganisation cannot have come before that date; Kuhrt placed it between 516 and 502 at the earliest. Her 1988 verdict was blunt: "if we leave Herodotus aside, the arguments for a satrapal reform under Darius I are still fragile", and the evidence already pointed forward, for "it has even been argued that the division of the satrapy belongs to the reign of Xerxes". That is where later scholarship, and the entry on Xerxes, now firmly place the splitting of the Babylon-and-Across-the-River mega-satrapy, so that the reform the Greeks credited to Darius reads better as begun under him, if at all, and completed under his son.^[13]

Whatever the count, the satrap himself is well documented, and the modern account of his office corrects an older one. The word is Old Persian for "protector of the realm," and at Behistun Darius does not yet speak of a satrapy as a territory. Naming Vivāna in Arachosia and Dādarshi in Bactria, he calls each a bandaka, a term for the personal bond between the king and a Persian noble. As Briant reads it, the satrap was a person before he was a post: "the duty of a satrap was not necessarily connected to a territory ... A satrap was first and foremost the personal representative of the king."^[6] That personal bond, in which the great servant acts as the king's instrument, defines the office here; territorial jurisdiction comes second.

The satrap's competences were wide and real. He collected the province's taxes, controlled the local officials and cities, judged cases, kept the roads safe, and maintained an army, and he had always to weigh present extraction against the province's capacity to yield again.^[24] Klinkott's study, the fullest modern treatment, reads the office as a genuine mediating role, the king's instrument of security and tribute who was also the official protector of the peoples in his charge; its most striking finding, the satrap's bounded power to conclude a strictly military, time-limited armistice on his own authority while a full peace remained the Great King's alone, is treated in the entry on the satrapy system.^[25] For Darius's reign, the telling feature is the calibration: the empire delegated substantial real power, including the power to treat with enemies, precisely because bounded autonomy made the system robust rather than brittle.

Substantial power in a hereditary office invites the province to drift from the crown, and the system answered that with a division of authority inside each satrapy. The satrap did not hold his province alone. Beneath him sat a documented apparatus, the chancery secretaries, the judges, the overseers and the informants that Greek writers called the king's eyes and ears.^[9] Two checks were deliberately kept out of his hand. The commander of the citadel garrison was appointed by the king, not the satrap. Xenophon says the king considered this a precaution against the satraps' revolts, and the arrangement is one of the clearest signs that the separation of powers was real and not a Greek theory, since the man who held the strongpoint in each province answered to the centre.^[6] The separation should not be over-generalised into doctrine, though: the 1988 Cambridge Ancient History, granting the fortress commanders their direct line to the king, cautioned that satraps in clear command of armed forces, within their provinces and beyond them, are common enough in the record "to suggest that it was not a regular policy so to divide authority"; the armed satrap Oroetes, below, is that caution in the flesh.^[13] So did the royal secretary lodged in the satrapal residence, through whom the king's written orders were read aloud; how that check could end an over-mighty satrap is told below, in the killing of Oroetes.^[6] The check on the great servant ran, in the end, through a scribe reading a sealed order.

The most famous of these controls, the king's eye, is largely a Greek construction. No Achaemenid document confirms a standing office of the kind the Greek writers describe: Briant finds that "not one item in the Achaemenid corpus confirms the existence of such an institution," Xenophon himself wrote that anyone who thinks the king "picked one to be his 'eye'" is "not thinking straight," the true saying being that the king has many ears and many eyes, and Kuhrt's editorial verdict is that "there is no supporting evidence for its existence."^[6][3] What the documents do show is the reality the legend schematised: travelling inspectors, an annual circuit of commissioners, and, in the Persepolis archive, a whole vocabulary of investigators, registrars, tablet-inspectors and road-inspectors.^[6][5] The oversight was real and it was busy; the single mythic Eye was not an institution but a Greek image of it.

Herodotus remembered Darius as the huckster king, the kápēlos whose fixing of the tribute reduced kingship to shopkeeping; the nickname, and what the Greek triad of huckster, master and father was built to do, are weighed with the rest of the reign's reputation below. What lies under the slur is a real reform, and Briant states its content precisely. Tribute itself was not invented; the conditions of its levy were changed:

From Cyrus to Darius, it was not the institution of tribute itself that was created from nothing; it was instead the conditions under which it was levied that were profoundly altered.

Briant, From Cyrus to Alexander, p. 69^[6]

Darius, he continues, "was the first to set a value on land and to establish fixed numbers in proportion to this valuation and in this way perhaps earned the qualification kapelos," though he flags his own reconstruction as conjecture.^[6] That conjecture has a documentary footing in Babylonia, where a property census, in place by about 500/499 and known by an Iranian loan-word, recorded the ownership of land to give a basis for assessing fiscal and military obligations, and still furnished the point of reference for property title a century later; Kuhrt set beside it Artaphernes's land-measurement of Ionia in 493 as the same policy seen from the Greek side.^[13] The huckster, in other words, is a Greek moralising of land-valuation and fixed assessment. Where Cyrus's levies were negotiated ad hoc by his satraps, Darius set the sums to a measured base, and the regularity is what a Greek pen turned into avarice.

The scale of it, as Herodotus records it, was enormous. His district list runs from Ionia to India: from Babylon and the rest of Assyria "a thousand talents of silver and five hundred boys for castration"; from India, the most populous of the peoples, "three hundred and sixty talents of gold dust."^[3] Herodotus then sums the whole:

Adding it all up, the total yearly tribute in Euboean talents for Darius was 14,560. I omit the bits less than a talent.

Herodotus 3.95, in Kuhrt, The Persian Empire, p. 675^[3]

(Translators construe the closing clause differently; Godley's Loeb gives "I take no account of figures less than ten.") The figure is staggering and should not be read as an audited total. Much of what it counts was rendered in kind, not in silver, so its numbers are at best estimated equivalents, and the reckoning is Herodotus's own arithmetic on data he reached through unknown channels.^[4] But the modern reassessment of the tribute goes deeper than doubting a sum. It overturns the picture that sum was long taken to prove.

That older picture was of an empire that bled its provinces white. In its fullest form, Olmstead's, Achaemenid taxation set up a one-way flow of wealth from the periphery to a hoarding centre, draining the subject lands and driving the stagnation that would deliver them to Alexander. Briant rejects it under a blunt heading, "Hoarding and Stagnation: Obvious but False."^[6] The rejection rests on how the treasure was actually held and used. Herodotus's own account of the storing of tribute, the metal melted and poured into earthenware jars from which, "when money is wanted, the necessary amount is coined for the occasion," describes not hoarding for its own sake but a working reserve.^[6] The bullion was minimally coined, drawn on as needed and, above all, spent on the gifts through which the king bound his Faithful to him. Strabo confirms that only a small part was struck, the rest worked into plate and furniture the Persians "consider ... as better adapted for presents ... and for depositing in storehouses."^[6] Nor did the wealth all flow to Persia. A large share stayed in the satrapal treasuries and was spent on the spot: of the five hundred talents levied in Cilicia, a hundred and forty "were spent on the cavalry who were garrisoned there."^[6] The system funded its own administration and defence in the provinces rather than draining them. The corrective is older than the modern synthesis, and it was first drawn from the Persian record: already in 1988 the Cambridge Ancient History inferred from the Persepolis treasury evidence that Herodotus's sums, "if his figures have any reality at all", "may represent only such payments as were due from the province to the central government", above a further layer of taxes paid wholly in kind "which stayed in the region where they were collected" and ran the satrapy's government, its garrisons and the king's decrees; the two-tier reading of Achaemenid taxation rests on the tablets rather than on Greek anecdote.^[13] That corrective has a counterweight of its own, best documented in the richest province. Jursa and Schmidl trace the strain of late-Darius taxation on even the wealthiest Babylonians, whose stable or falling real land prices are best explained by "increasing taxation and service obligations that were attached to landed wealth"; and a dossier of letters from the reign's last years shows a royal official stopping the priests' cult payments at Ezida because the temple's funds "had been requisitioned by the crown," a clear indication, they judge, of an economic and social crisis in northern Babylonia which Persian demands, if they did not cause it, "certainly exacerbated."^[18] The empire redistributed, but it also pressed, and in Babylonia the load was heavy and rising.

The Greek categories that framed the tribute break down under scrutiny too. Herodotus divides the peoples into those who paid tribute (phoros) and those on the fringes who gave gifts (dora), but Briant argues the distinction is ideological rather than economic: refusal to give gifts was punished as strictly as failure to pay tribute, and the difference between the Ethiopians' gift-gold and the Indians' tribute-gold "seems to reflect the politico-ideological domain rather than the purely financial domain," one system under two labels, the second a way for the king to honour a giver as a Benefactor while collecting all the same.^[6] The Persian side confirms it: Darius's own word for what the peoples brought him, Old Persian bāji, covers both.^[9]

The reassessment's positive content is the redistributive economy the Persepolis archive documents, and here Henkelman's work is the anchor. The tablets show not a hoard but a working institutional economy of receipt, storage and issue in kind:

Its purpose was to arrange, survey, record, and account for the streams of these commodities within what I will call the 'Persepolis economy.' The prime tasks involved were the receipt, taxation, storage, and transport of goods and their redistribution to gods, members of the nobility, officials, travellers on the royal roads, workers and livestock.

Henkelman, The Other Gods Who Are, p. 65^[19]

This is the real fiscal machine at the heart of the empire, of which Herodotus's silver totals are only the outward face. It ran on barley, wine and livestock rather than bullion, and it moved them out again to those the crown maintained. That the base of the whole system was payment in kind is not inference but record. The chief administrator himself, Parnakka, drew his own vast allowance in kind, two sheep, eighteen measures of flour and nine of wine or beer a day, and even the workers' silver, when they got it, was booked as a fraction of a wage otherwise paid in rations.^[7] The old Olmstead model, overtaxation feeding a sterile hoard, is displaced by an economy that taxed, stored and redistributed commodities across the heartland, with silver as an accounting equivalent more than a currency.

Against that in-kind reality stands Darius's coinage, and its importance is more political than economic. The royal coins were two, the gold daric and the silver siglos, both stamped with the king as an archer. Carradice's typology, the framework the field uses, tracks the archer obverse across four types; the daric held steady at about 8.35 grams, equal in weight to a Babylonian shekel and about 98 per cent pure.^[26][3] The archer image was not decoration. The earliest type shows only the upper half of a crowned figure rising from a crescent, a formula that recalls Assyrian images of a god and contemporary Achaemenid images of Ahura Mazdā, so that the crown makes him the king and the format makes him divine; it is "an image of divine kingship," and it stayed the royal type from its introduction to the empire's fall.^[5] The name of the gold coin is genuinely contested. The ancient derivation, from Pollux, takes daric from Darius himself, and the coincidence of Darius introducing it keeps that link alive; the philologically favoured view now derives it from Old Persian daranya, "gold," and Brosius, who hedged in 2021, states the gold derivation flatly in 2023.^[26][21] The two readings are not reconciled, and the honest position is that the etymology is open.

The coinage was Darius's and struck at Sardis, and it was in use by about 500, but the exact year within his reign is disputed, Briant placing its creation on the king's return from the European campaign, others between 519 and 512.^[6][26] One dating datum needs stating with care. Briant reports that a Fortification tablet of Darius's year 22, that is 500, "bears the imprint of a royal coin showing a kneeling archer-king," which would tie the coin type directly to a dated document.^[6] The numismatic literature does not corroborate a coin-impressed tablet, and the securely dated year-22 object from Persepolis is a seal, that of Parnakka, not a coin; the coinage's floor of about 500 rests rather on the absence of darics and sigloi from the datable Persepolis deposits.^[26] What is not in doubt is the coin's function. The gold daric was worth too much for daily exchange and served chiefly as a store of value and an instrument of royal favour, "doled out as a sign of royal favor," while everyday transactions, where coin was used at all, ran on silver.^[4] The siglos itself was regional, struck for the Greek economy of Asia Minor and not for the interior, which is why the empire's heartland ran on weighed metal and rations rather than on coin.^[26] The point holds even for Babylonia, the richest assessed province, where coinage was not in use in the Achaemenid period at all; Kuhrt's formulation is that perhaps it was "only the accounting of the tribute which was monetarized", the later obligation to render Babylonian taxes in silver being precisely what made possible the profitable business of firms like the house of Murashu.^[13] The ideology of the coinage is caught in Herodotus's story of Aryandes, the governor of Egypt who "aped" Darius by striking his own pure silver and was executed for it. Kuhrt cautions that the episode is undated and that no such Aryandic coins have ever been found, so it is best read as ideology, the purity of the royal gold as a royal prerogative, rather than as proof of a strict monopoly.^[3] The tale has a background that strengthens the reading, for Aryandes had been a suspect servant long before the coinage story: he seems to have been expelled in the crisis of 522, when a native dynast named Petubastis may have tried to seize power, and was reinstated only after Darius entered Egypt in person in the winter of 519 and put the revolt down "without much difficulty"; his expedition to Cyrenaica reads as an attempt to regain his master's confidence. Ray, surveying the varied reasons the sources give for the satrap's eventual fall, judges the coinage version "groundless": the fundamental cause was the fear of a declaration of independence by the viceroy of Egypt.^[13] Herodotus gives the reason the king wanted the coin at all: Darius "was anxious to leave a memorial of himself, the like no other king had achieved."^[3] The daric was that memorial, the king's image in gold, circulating.

Herodotus knew of one royal road, the line from Sardis to Susa, and the archive shows he saw only part of it: the road ran on to Persepolis and further east into India, "as is amply documented in the Fortification archive," a maintained network of routes reaching across the whole empire, not the paved highway of popular imagination, set with way-stations at intervals of a single day's travel and policed by the armed teams the tablets call road-inspectors.^[27] The developed system was Darius's: the archive shows it running smoothly between 509 and 493, and it was "probably during the reign of Darius that the system reached the perfection praised by Herodotus"; the network itself has its own entry at the Royal Road.^[27]

The road and the chancery were one apparatus, joined at the travel document. All official travel ran on a sealed authorisation, Elamite halmi, issued by a satrap or the king and written in Aramaic, which stipulated the travellers, their destination and the rations they were to draw, a countersealed receipt drafted at each station for each commodity issued; so the movement of people and the language of administration were one system seen from two sides.^[27] Urgent word moved faster, by the relay the Greeks called the angareion and the archive calls pirradaziš, the royal express, which could cover Sardis to Susa in about twelve days.^[27] Keeping the stations stocked was one of the main reasons the regional economies of the Persepolis kind existed at all, which is the point at which the roads, the chancery and the storehouse economy meet.^[27] The whole apparatus is caught working, at maximum range, in a single tablet:

Katukka, the grain supplier at the way station of Pirdatkaš, somewhere in the valleys of western Pārsa, who saw the approach of a group of almost three hundred men, with their camels and mules, on their way from Gandhāra to Susa (NN 0431) and who handed out flour rations on the basis of a travel authorisation issued more than two thousand kilometres away, would not have missed this point: indeed, the stylus of the Persian man had gone forth far.

Henkelman, in Jacobs, Henkelman & Stolper, Administration in the Achaemenid Empire, p. 186^[18]

The closing phrase is a deliberate echo of the tomb inscription's boast, quoted below, that the spear of a Persian man has gone forth far: the pen turned the spear's claim into a working fact. And the traffic ran inward as well as out. In Darius's reign the Babylonian administrators and business partners who handled tax assets were summoned to Susa in at least nine years between 510 and 486, meetings that were, as Kleber observes, "most likely also a stage where royal favors and loyalties were negotiated."^[18] The centre did not only send sealed orders down the roads; it pulled the provincial elite in along them, on a rhythm the tablets can date.

Aramaic was the empire's administrative language, and the qualification in that phrase is deliberate. It is better called administrative than official. Aramaic had served international communication in the Near East since the eighth century, under the Assyrians already, and the Persians adopted it "as the most generally familiar administrative imperial language," attested in use at Persepolis, in Babylonia, Egypt, the Levant and Asia Minor.^[3] But Kuhrt resists the stronger term: "This does not mean, however, that Aramaic was in some sense the official language of the empire," since local languages carried real administrative traffic beneath it.^[3] The mechanism was a layered one. The king and satraps commanded in Old Persian; the top level of administration worked in Aramaic, "the linguistic binder of the empire, the language used throughout its territory, between the Old Persian level and the local administrative level"; the local level continued in Egyptian, Elamite and the rest.^[5] The load-bearing figure was the bilingual scribe, the sepīru of the Akkadian texts, the hinge between the king's Iranian command and its execution in a local tongue; the pipeline can be caught working at both ends of the empire, in Egyptian governors' letters whose Demotic "smells of translation" and in a Bactrian correspondence between two Persian officials wholly dependent on their scribes for the Aramaic they wrote in.^[5][28] One point sharpens the whole account. Aramaic dominated administration but was excluded from monumental display, which used Old Persian, Elamite and Akkadian; so the channel of the chancery and the channel of the king's carved self-presentation were kept distinct.^[28]

Darius left one monument to the empire's connectivity that was also a working artery, or meant to be. In Egypt he cut, or reopened, the canal from the Nile to the Red Sea, and recorded it on stelae in his own voice:

I am a Persian; from Persia, I seized Egypt. I ordered this canal to be dug, from a river called Nile, which flows in Egypt, to the sea which goes to Persia. So this canal was dug as I had ordered, and ships went from Egypt through this canal to Persia, as was my desire.

Darius I, DZc §3, in Kuhrt, The Persian Empire, p. 486^[3]

A canal on the line had been begun by Necho II of Egypt and apparently left unfinished, so here too Darius completed an inherited project rather than conceiving one.^[3] The stelae let the undertaking be measured. They record that the king conceived the scheme while still at Pasargadae and later, perhaps in 510, summoned a conference of architects at Persepolis; reconnaissance ships reported about 84 kilometres of the old channel dry; the new cut was wide enough for two triremes to pass, roughly 45 metres, with an estimated twelve million cubic metres of earth moved, and the line was marked with at least a dozen stelae over three metres high in the three cuneiform languages and hieroglyphs. The best-preserved examples give the royal boast a cargo manifest: a flotilla of twenty-four ships, laden with Egyptian produce, sent to Persia.^[13] Whether it worked as a trade route is disputed. Strong arguments have been made against regular sailing between the Red Sea and the Persian Gulf at this date, which would leave the canal "primarily an ideological function," though its commercial purpose has been reasserted.^[3] The stele's own claim that "ships went from Egypt through this canal to Persia" is exactly the assertion the ideological reading discounts. The canal sits, either way, with the roads and the coinage as a statement of reach, the king joining his lands and saying so.

The surest counter-check on all of this is the archive that was never meant to be read as a statement of anything. The Persepolis Fortification tablets record administrative transfers of food in Darius's own reign, his thirteenth to twenty-eighth years, across a territory reaching from Elam down into Persis.^[29] Their value is exactly that they are not rhetoric. Where Behistun presents Darius and Herodotus reconstructs him, these are internal working documents with no case to make, "laconic and unpretentious," and for Henkelman their potential "stands unrivalled among the sources available to the historian of the Empire."^[19] They catch the empire running itself. They also, in passing, correct the propaganda: against Herodotus's claim that Persis paid no tax, the tablets show that it did.^[21]

The archive is bound to the dynasty at its head. Parnakka, the chief of the whole Fortification administration, was Darius's own paternal uncle, the brother of Hystaspes, identified as "son of Arsames" by the inscription on his seal; he was the king's highest representative in Persis, effectively its satrap though the tablets never give him the title, and that the top administrator is never once called "satrap" is itself a caution against how fixed the textbook system really was.^[21][5] Beneath him and his deputy Zissawis the machine ran on documented controls, a matrix of officials who countersealed each other's receipts so that only the highest sealed alone.^[5]

Two things the tablets show demand care in the telling. The workforce they record, the kurtaš, were dependent labourers subsisting on rations, and they were not simply slaves: Hallock, who edited the texts, declined the equation, noting that conscripts are called kurtaš and asking why slaves would be conscripted.^[29] These were the workers who built Persepolis, "hundreds of workers, men, women and children," of many trades and peoples from across the empire; and the women among them were rationed in their own right, more numerous than the men in the monthly ration texts, usually drawing less but in some cases more than any man present, presumably as supervisors or for special skill.^[21][29] This is the unpropagandised counterpart to the serene tribute-procession carved at Persepolis: not a frieze of grateful subjects but a paid, rationed, mixed and multi-ethnic labour force, its allowances entered in clay.

The archive settles the transformation question on the terms already argued: the machinery Darius is credited with was inherited in form, the Persepolis economy being "a large-scale and developed version of the Neo-Elamite redistributive palace economy," and what was genuinely new was the scale, the status of the dependent workers "redefined to include people from all over the empire."^[19][5] The Treasury tablets even catch the one change that was Darius's own moving on the ground, the slow shift from payment in kind toward payment in silver about 493; and Cameron, who edited them, is careful that it shows "no evidence ... of royal fiat," a gradual and partial transition rather than a decree.^[30] That is the reign in miniature. Darius did not conjure the empire's administration from nothing. He inherited an old machine, scaled it, tightened its fiscal grip, wrote it down, and left the records by which we can watch it work.

Behistun told the story of the accession once, on one cliff, against named enemies in a fixed year. The rest of Darius's reign told a different and larger story, without dates and without a single narrative, in the two capitals he raised from the ground. He began a new dynastic seat on a bare terrace at Persepolis and rebuilt the old Elamite city of Susa into a palace of the whole empire, and he furnished both with reliefs, inscriptions and buried records that say the same thing the accession inscription says, that the King of Kings holds a god-given, boundless authority over willing peoples, only now they say it as a permanent condition rather than a recent victory. Briant states the case at its strongest: from the late 520s the monarchy built a politico-religious ideology "of rare consistency," and "this is the program we see at work in the new residences in Susa and Persepolis as well as on the royal tomb at Naqs-i Rustam."^[6] The buildings are not a backdrop to the ideology. They are the argument in stone.

The centre of Persepolis is the great audience hall, the Apadana, the loftiest building on the terrace, and it was Darius who planned and began it. The proof is buried under it. In the north-east and south-east corners of the main hall Schmidt's excavators found neatly cut stone boxes, each holding two inscribed metal plaques, one gold and one silver, with small hoards of coins beneath; the excavation record is given with the building itself in the entry on Persepolis.^[31] The plaques carry the text known as DPh, which draws the empire by its edges, from the Sakā beyond Sogdiana to Ethiopia and from Sind to Sardis; the text is quoted below, with the frontier wars that fixed those edges.^[32] The deposit ties the inscribed corpus directly to the building's foundation ritual: the king's titulature and his god-given dominion were literally laid under the corners of the hall. The gold and silver plaques were, in the reading C. L. Nimchuk gives them, a message and a gift to Ahura Mazdā; the coins, Croeseid gold from Lydia together with Greek and Cypriot silver, add a further claim, control of the empire's minting, to the political and religious ones.^[33] Nimchuk's larger point is that Darius replaced the stock Mesopotamian boast of ruling "the four quarters" with his own named limits, so that the empire became spatially tangible, and the buried tablets "acted as anchors for the metaphysical corners of the empire," aligning the four quarters named in the text to the four corners of the building whose reliefs showed the peoples those quarters contained.^[33]

The coins are also the dating crux. Because the plaque text does not include Darius's trans-Danubian Scythian conquests of about 513, Schmidt inferred that the substructure was finished and the walls begun before that campaign, which places the founding of Persepolis early in the reign.^[31] That is Schmidt's inference from a geographic list, not a fixed date; but with the Lydian and Greek coins beneath the boxes it fixes the start of the terrace within Darius's reign on material grounds rather than on the later inscriptions, which are silent, since even Xerxes's own Apadana-stairway texts do not refer to his father's building at all.^[31]

The terrace as it survives is bare grey stone, and that is not how Darius's builders left it. The reliefs were painted: early excavators repeatedly saw colour on freshly exposed surfaces that then faded, a defined palette survives in traces and stray lumps of raw pigment, and instrumental analysis has confirmed original red ochre on fragments from the terrace; the palette and the laboratory record are given in the entry on Persepolis.^[34] The austere monochrome of the modern ruins is an accident of survival. Darius's monuments were polychrome, and the argument they carried was made in colour.

Susa was the other half of the programme, and its foundation charter, the inscription known as DSf, is the most explicit statement the ancient world has left of a building raised by a whole empire. Composed early in the reign on the older view, though a strong recent argument puts it a decade later, it names, material by material and craft by craft, where everything came from and who worked it.

The cedar timber, this — a mountain by name Lebanon — from there was brought. The Assyrian people, it brought it to Babylon; from Babylon the Carians and the Ionians brought it to Susa. The gold was brought from Sardis and from Bactria. The precious stone lapis lazuli and carnelian was brought from Sogdiana. The stone-cutters who wrought the stone, those were Ionians and Sardians. The goldsmiths who wrought the gold, those were Medes and Egyptians. The men who wrought the baked brick, those were Babylonians. The men who adorned the wall, those were Medes and Egyptians.

Darius I, DSf^[35]

The palace itself embodies the same idea in plan. Darius set an Iranian columned audience hall, its seventy-two columns standing perhaps twenty metres high, beside a vast mud-brick residence in the Mesopotamian tradition, so that the building was, in J. Curtis's phrase, "a neat fusion of highland and lowland, new and old."^[33] Its walls were faced with glazed polychrome brick, friezes of striding archers, of lions, of winged human-headed lions beneath a winged disc, of which more than thirteen thousand bricks now survive in collections.^[33] The archers are traditionally called the "Immortals," the king's guard, but that identification is hedged and their identity is debated; Razmjou re-read the accompanying "servant" brick inscriptions at Susa as priests carrying offerings and animals to a sacrifice.^[36]

The DSf charter is usually read as a hymn to imperial cooperation, and it is, but the cooperation is not naive, and a close reader of the Susa material caught its silence. M. W. Stolper noticed that the one people missing from the list of labourers, transporters and craftsmen is the Persians.

Conspicuous by their absence as contributors, transporters, or workers are the Persians, the rulers themselves. This delicate omission and the nuances of this message stand in plain contrast to the starker terms of Darius's later inscriptions, both from Susa (DSe) and from Persepolis and nearby Naqsh-i Rustam in Persia proper. At the beginning of his reign, with his power newly imposed and delicately poised, Darius had his scribes write: "This palace which I built at Susa, its building materials were brought from far away."

Stolper, in Harper, Aruz & Tallon, The Royal City of Susa^[33]

The harmony of nations builds the king's house; the Persians command it. Stolper dates the charter to just after the accession, when Darius's power was "newly imposed and delicately poised," and reads its softer tone against the blunter later texts that tell the viewer to look at the peoples carrying the throne and know that the spear of the Persian has gone forth far.^[33] The dating that reading rests on is no longer secure. Jacobs has argued that the charter's own list gives a terminus post quem of about 512: DSf and its Elamite twin DSz name Indians and Nubians among the contributors, Hindush was taken only about 515, and Libya and Nubia enter the royal lists only after the African campaign of 513/12. Nor does the closing prayer's blessing on Hystaspes anchor an early date, since the Persepolis tablets show Darius's father alive to at least 501/500, his memorial appearing only in a year-23 text.^[18] If the charter is a decade later than the accession, the "delicately poised" tone was a choice of register rather than a mark of new and uncertain power; the reading is Stolper's, and it stands or falls with the older date. What does not move is the double truth of the programme: the image it projects is ecumenical, an empire of many peoples working as one, and the order it encodes is hierarchical, with one people at the top. Both are true at once.

The single object that gathers the whole method into one piece stood in the monumental gate east of the Palace of Darius, where it was found in 1972: a greater-than-life statue of Darius, its head lost, carved in hard stone. It is a Persian Great King in a Persian robe, and it is Egyptian in every other respect: greywacke from the Wadi Hammamat, Egyptian style and workmanship, an Egyptian dorsal pillar up the king's back. The robe is inscribed on one side in the three cuneiform scripts (the Old Persian text is DSab) and on the other in hieroglyphs, and the base carries twenty-four "fortress cartouches" naming the subject peoples, each surmounted by a kneeling figure in native dress with arms raised.^[33] It was almost certainly made in Egypt to commemorate Darius's canal, then shipped to Susa at the end of his reign or the start of Xerxes's, and its own inscription states its purpose as plainly as a monument can: it was made "so that whoever sees it in the future will understand that the Persian warrior governs Egypt."^[33] It too was painted, traces of brown-reddish paint surviving on the folds of the robe.^[36] A Persian king, rendered by Egyptian hands in Egyptian stone, quadrilingually inscribed, standing on the kneeling peoples of the empire and tied to a Persian feat of engineering on Egyptian soil, is the building programme's thesis compressed into a single figure.

On this reading, the monuments form a single designed argument: the grand buildings accumulate within a common programme. Briant put the point without qualification: the craftsmen "had no artistic freedom," they followed "the precise specifications provided by the Great King's counselors," the borrowings from the Assyro-Babylonian, Elamite and Egyptian repertoires were fused into "a royal art par excellence," and the whole reflects a programme that "leaves no room for improvisation."^[6] A recent architectural study by Soheil carries the same reading into the design of the terrace as a whole, treating Persepolis as an "intentional monument" in Alois Riegl's sense, planned from the outset to transmit a political message of imperial unity, and Achaemenid art not as a pastiche of borrowed elements but as "an original art created by the kings and court artists to express the political significance of the Empire."^[5] Darius's craftsmen came from everywhere, by his own boast. The message they were set to carry was uniform, and it was Persian.

The clearest image in the whole programme sits on the façade of Darius's tomb, and it is the image his own inscription tells the viewer to read. High on the cliff at Naqsh-e Rostam, west of Persepolis, the king stands on a three-stepped pedestal before a fire altar, the winged symbol hovering between and above him, and the whole platform is borne up by the representatives of the empire's peoples. The text of DNa, cut beside the relief, does not describe them so much as point at them:

Saith Darius the King: Ahuramazda, when he saw this earth in commotion, thereafter bestowed it upon me, made me king; I am king. By the favor of Ahuramazda I put it down in its place. 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, then shall it become known to thee: the spear of a Persian man has gone forth far; then shall it become known to thee: a Persian man has delivered battle far indeed from Persia.

Darius I, DNa §4^[37]

This is a monument captioning its own sculpture. To learn the extent of the empire the viewer is sent from the words to the image, and the image delivers the count exactly: Schmidt records thirty nations on the tomb relief, arranged in two tiers, physically supporting the king's dais, each throne-bearer identified by a trilingual cuneiform label.^[31] The labelled throne-bearers of the royal tombs became the reference set by which the unlabelled delegations elsewhere at Persepolis were later identified.^[33] Text, image and architecture cross-reference one another: the peoples-list of the inscription, the carved and named peoples of the relief, and the columns of the halls that Persian thought equated with the guardsmen who upheld the king are one system saying one thing. One caveat on the famous sentence was entered in the 1988 Cambridge Ancient History: the inscription sat on a rock face in Persis and "might be addressed specifically to Persians", and whether the same text would have stood "on public view in, say, Babylon" is another question, for to judge from the idiom of the Cyrus Cylinder "it well might not have"; the reign's own Babylon copy of Behistun, with Bel-Marduk in Ahura Mazdā's place, shows exactly that switching of register in practice.^[13]

What that thing is, and how the sculpture says it, is the subject of Margaret Cool Root's study, which reoriented the field. The obvious way to read a king raised above rows of foreign figures is as domination, the conquered holding up the conqueror, and that is exactly how the Near Eastern prototypes of the scene read. Root's argument is that the Achaemenid version inverts the formula, and her evidence is in the carving itself: each figure takes the weight of the dais "on only the tip of his thumb and the tip of his index finger," arms crossed and linked in a chain, so that the men hardly seem to bear any weight and the platform "appears to be virtually self-levitating."^[38] Against a tradition whose prototypes "consistently stress the subjugation of the peoples by the king," she reads the reworking as "a cooperative effort of voluntary support of the king by the subject peoples," designed "to convey the aura of a sacral covenant between king and subjects," an inherited image of domination taken up "precisely because of the iconographical impact it still conveyed" and bent to mean its opposite.^[38] And she anchors the reading to the king's own words on the same tomb: "The man who cooperates, him according to his cooperative action, him thus do I reward."^[38] The peoples uphold the king willingly, and the text tells you the king rewards cooperation. The relief is the Pax Persica made visible. Root's thesis, and Briant's counter-reading of the same stones, are staged in full with the Apadana reliefs in the entry on Persepolis; the same repertoire travelled, adapted on the Harpy Monument at Xanthos and on sealings from Daskyleion.^[33]

Above the king on the tomb, and above the enthroned figure of the audience scene carved for the Apadana, hovers the winged symbol, and what it represents is a live dispute. The traditional identification, used by Schmidt in his excavation report, is that it is "the winged symbol of Ahuramazda," the god named by Darius.^[31] Boyce rejects this. She observes that the figure in the winged circle is always male, whereas the fravashi to which it was once assigned was thought of as female, and that its appearance varies, notably in its crown, and that "sometimes, in both gesture and apparel, and in the objects which it holds, it is the mirror-image of the king beneath," which she takes to make it "wholly improbable that it should represent the supreme God."^[39] Her preferred reading is that it is the royal khvarnah, the divine glory: the winged disk represents the divine grace "as it is accessible to all men, while the figure in the circle is the royal khvarenah, which accompanies each ruler and attaches itself to a whole dynasty through the sacred power of royal blood."^[39] On the tomb, Boyce reads the scene accordingly: the king salutes his own hearth-fire, and the winged figure returns his gesture and mirrors him, "as would be fitting for the representation of his own khvarenah."^[39] Root's reading is different again: for her the god in the ring is neither an object of worship nor the king's glory but the other party to a covenant, delivering power, the abstract power of the ring and the concrete power of the bound rebels, into the king's hands.^[38] Three readings, three theologies of the same carved figure: the supreme god, the royal glory, the covenant-partner. The neutral "winged figure" or "winged symbol" is now the common scholarly practice, the identification standing as an open question, with Boyce the fullest statement of the khvarnah case. The dispute's three-way shape is older than its current literature: the 1988 Cambridge Ancient History already noted the argument that the winged symbol, "usually understood as an image of the god Ahura Mazda", is really the king's khvarnah, and judged it flatly unlikely, even as its own account of the royal image reached for the same concept from the other side, finding in the reliefs no portrait of Darius, Xerxes or Artaxerxes but "a dynastic image", almost perhaps of the khvarnah, the glory of kingship, "rather than an image of a king".^[13]

The programme's fullest statement of what a king should be is not on any relief but in the second inscription of the tomb, DNb, a first-person self-portrait that Briant aptly called the prince in his own mirrors. It defines the king across two spheres, the battlefield and the seat of judgement, and it begins with justice, with an even hand held over strong and weak alike:

Saith Darius the King: By the favor of Ahuramazda I am of such a sort that I am a friend to right, I am not a friend to wrong. It is not my desire that the weak man should have wrong done to him by the mighty; nor is that my desire, that the mighty man should have wrong done to him by the weak. What is right, that is my desire. I am not a friend to the man who is a Lie-follower. I am not hot-tempered. What things develop in my anger, I hold firmly under control by my thinking power. I am firmly ruling over my own (impulses).

Darius I, DNb §8a–b^[37]

The virtues are exact and connected: even justice between weak and mighty, no friendship with the Lie-follower, and the self-mastery that makes even justice possible, the holding of anger under the control of thought. The text then sets out how the king judges and how he requites, and the standard is desert on the one hand and due process on the other:

The man who cooperates, him according to his cooperative action, him thus do I reward. Who does harm, him according to the damage thus I punish. What a man says against a man, that does not convince me, until he satisfies the Ordinance of Good Regulations.

Darius I, DNb §8c–d^[37]

The king rewards and punishes by what is deserved, and he will not believe a charge on one man's word until it satisfies the law; he hears both sides before he condemns. Only after the judge does the inscription give the warrior, and it gives him as a trained body, not a boast of conquests:

Trained am I both with hands and with feet. As a horseman I am a good horseman. As a bowman I am a good bowman both afoot and on horseback. As a spearman I am a good spearman both afoot and on horseback.

Darius I, DNb §8h^[37]

And it closes with an address to whoever will come after, a demand that posterity credit the record:

O menial, vigorously make thou known of what sort I am, and of what sort my skillfulnesses, and of what sort my superiority. Let not that seem false to thee, which has been heard by thy ears. That do thou hear, which is communicated to thee.

Darius I, DNb §9a^[37]

That closing note, "let not that seem false to thee," is the same anxious insistence on being believed that runs through Behistun, and the same address to a reader down the years that DNa makes when it tells the viewer he will come to know the empire's reach. Briant reads DNb as defining the king across the two complementary spheres, the good infantryman, horseman, archer and lancer who is also "a good commander-in-chief," with "intelligence and a spirit of analysis and decisiveness that free him from the emotion of panic," and the master of justice who "tries to reconcile the interests of the powerful and the poor" and is "guarantor and restorer of civil peace."^[6] The text was durable enough to become a template: Xerxes repeated it in summary in his own inscription XPl.^[6]

The most striking thing about this self-image is that outsiders who had no reason to flatter Darius reported the same ethic. Herodotus gives the Persian schooling in three things, to ride, to draw the bow, and to tell the truth, the very pairing of the warrior's training and the horror of the Lie that DNb makes, and reports the custom that not even the king may put a man to death for a single charge, a Persian weighing a man's offences against his services before he acts.^[6] Briant triangulates DNb with exactly these passages and with Plutarch's Persian king as arbiter of good and evil, and the convergence is the point: the DNb ideal is not merely the king's own advertisement, but a genuine Persian conception of kingship that independent Greek observers recognised and recorded.^[6]

Behind the even hand and the enmity to the Lie-follower stands the pair of terms that organises the whole of Darius's thought, arta, the Truth or right order, and drauga, the Lie. In the accession inscription every rebel is a follower of the Lie; in a Persepolis prayer Darius asks Ahura Mazdā to keep the land of Persia from three evils together, and the third stands beside war and hunger as a threat of the same order:

May Ahuramazda bear me aid, with the gods of the royal house; and may Ahuramazda protect this country from a (hostile) army, from famine, from the Lie! Upon this country may there not come an army, nor famine, nor the Lie.

Darius I, DPd §3^[32]

The Lie set alongside the invading army and the failed harvest, as one of the three things a king prays to keep out, shows how much weight the term carries. What the term means, though, divides its readers, and the division is the interpretive crux of the whole ideology. Boyce reads arta and drauga religiously, as the asha and druj of the Zoroastrian moral cosmos, and holds that the whole Achaemenid dynasty, Darius included, was Zoroastrian, while candidly granting that the inscriptions never once name Zoroaster and that Herodotus's account of Persian religion describes practices that do not accord with Zoroastrian priestly custom.^[39] Wiesehöfer reads the same pair politically, and sharply:

Unlike the followers of Zarathustra, for whom "truth" (OP rta) and "lie" (drauga) were moral concepts according to which they tried to organize their lives, Darius indeed considered drauga as anything directed against his own god-given and dynastically legitimized reign — that is, any kind of rebellion or usurpation. To him even rta was ultimately what he himself proclaimed as truth.

Wiesehöfer, Ancient Persia, p.49^[9]

On this reading the Truth is loyalty to Darius and the Lie is rebellion against him, and the moral-cosmic vocabulary is an instrument of legitimation, which fits the accession inscription exactly, where every man with a claim against Darius is by definition a Lie-follower. A third position resists the framing that both readings can invite. Rossi, in the most recent synthesis, treats the whole corpus as the deliberate work of an "efficient and influential royal chancellery" that chose the lexicon and organised its diffusion across the empire's languages, and both he and Jacobs argue explicitly against importing the modern word "propaganda" into the Achaemenid material, preferring to speak of a planned "ethical and political message" addressed to the king's successors and to his subjects.^[5]

These readings are not exclusive, and nothing forces a choice between them. The concept was at once religious and political, and Darius's use of it worked because the two senses overlapped. Arta was a real term of Iranian religious thought, with a cosmic and moral weight that Boyce is right to insist on; and Darius deployed it so that loyalty to his own god-given kingship fell on the side of the Truth and every challenge to it on the side of the Lie, which is what Wiesehöfer is right to insist on. The overlap is not a confusion to be resolved but the very thing that made the vocabulary effective: a king who casts his enemies as enemies of cosmic order borrows the authority of the one claim for the other. Whether to call the result propaganda is partly a question of temper, and Rossi's caution is worth keeping, since the inscriptions were also, demonstrably, an ethical charge laid on those who would rule after him. What is not in doubt is that the ideal of DNb, the even-handed judge who masters his anger, hears both sides and hates the Lie, and the political use of the Truth against rebellion, are the same king speaking, and that the monuments in stone and the ideal in words were built to say, permanently and everywhere, what the accession inscription had claimed once: that Darius ruled by the favour of the Wise Lord, as the friend of arta and the enemy of the drauga.

The reconquest secured the throne; it did not end the fighting. Once the year of the Lie was over, Darius turned to the edges of the empire, and the first thing the edges taught him was that a king's own governors could be as dangerous as any rebel. The lesson came from Sardis. Oroetes had held the satrapy of Lydia since Cyrus's day, and during the crisis of 522 he did nothing to help the new king. He had also been settling scores: he contrived the deaths of Mitrobates, the governor of neighbouring Dascylium, and Mitrobates' son, killed one of Darius's own couriers, and, most famously, lured Polycrates, the tyrant of Samos, to the mainland and crucified him.^[6] The satrap's charge was to keep order and to extend Persian power, and Oroetes had done the second while making himself unaccountable in the first.^[6] What made him a problem was not disloyalty in arms but a servant grown too large to command.

Darius did not march on him. According to Herodotus he sent a single agent, Bagaeus, to Sardis with a series of sealed orders, which Bagaeus handed one by one to the royal secretary to read aloud before Oroetes' own guard. When the guardsmen showed that they would obey the king's writing rather than the satrap, the last order was produced:

“King Darius orders the Persians of Sardis to kill Oroetes.”^[6]

They killed him on the spot, and his property passed to the king.^[6] Briant reads the episode not as a Lydian revolt but as a demonstration of the bond that tied the Persian ruling class to a legitimate centre: the guard obeyed the seal because the monarchy, and not the local grandee, was the source of their standing, and the affair shows the king defending the aristocratic order that Cyrus and Cambyses had built.^[6] It is the first recorded case of a satrap defying the crown, and it set the pattern for the whole history of the empire. The over-mighty servant, recalled and destroyed by the reach of the king's word rather than by the king's army, would recur down the centuries, and the same lever, a written order read to a man's own soldiers, would end others as it ended Oroetes.

The frontier proper lay further out. In his third regnal year, about 519, Darius led an army against the Saka, the horse peoples the Greeks called Scythians, in Central Asia. The campaign is recorded on the fifth column of Behistun, a block of text added to the monument some years after the original was finished, and it survives in Darius's own words and nowhere else. There is no Greek narrative of it at all, so everything turns on reading the royal account:

Afterwards with an army I went off to Scythia, after the Scythians who wear the pointed cap. These Scythians went from me. When I arrived at the sea, beyond it then with all my army I crossed. Afterwards, I smote the Scythians exceedingly; another (leader) I took captive; this one was led bound to me, and I slew him. The chief of them, by name Skunkha—him they seized and led to me. Then I made another their chief, as was my desire. After that, the province became mine.

Darius I, DB §74 (Kent)^[35]

Read straight, the passage says Darius crossed water with his whole army, beat the pointed-cap Saka, killed one captured leader, took their chief Skunkha prisoner, and installed a client of his own. The pointed-cap Saka are the Sakā tigraxaudā of the inscriptions, and the campaign against them is distinct from the European expedition of a few years later, though Darius describes both in the same idiom of crossing the sea to reach an enemy beyond the natural boundary.^[35][3] The victory mattered to him out of all proportion to a distant frontier war. His image of the captured Skunkha, in his tall cap, was cut into the Behistun relief years after the other figures, and fitting it in forced the destruction of the original Elamite version of the inscription, which had to be carved afresh to one side.^[9] The Companion draws the inference that Darius went to this trouble because Skunkha's people had resisted even Cyrus, who died fighting Central Asian nomads, so that beating them let Darius claim to have succeeded where the empire's founder had failed.^[5] Manning adds the rhetorical point carried by the water: the sea-crossing let Darius say he had gone farther than the kings before him, reaching the ocean-edge of the ordered world.^[40]

The old reading of the passage is now under revision, and on the monument's own evidence. Kent and Briant took Skunkha to be the captured rebel king, punished and replaced.^[35][6] Bruno Jacobs, in the Companion, points out that this is not what the text says. Darius reports that he did not get hold of the actual leader of the rising, who is neither named nor said to be punished; Skunkha led a different Saka group, one never accused of rebellion, and he was deposed but, in Jacobs's words, "neither humilated nor executed."^[5] That is why, alone among the captives roped beneath the king on the relief, Skunkha keeps his hat: he was not one of the liar-kings.^[5] Jacobs identifies his people with the Dahā, west of the Aral. The revision also unsettles the confident identification of the figure by his headgear, since, as the same study notes, the types of Scythian cap on Persian reliefs "do not neatly fall into three mutually exclusive and distinguishable categories."^[5] The column that records the campaign also introduces something the accession narrative had lacked: a religious charge. The Saka, like the Elamites named beside them, are damned as faithless, people "by [whom] Ahuramazda was not worshipped," and the frontier war is reframed as the extension of true worship.^[35][6] Briant calls the politico-religious implications of this addition enormous.^[6]

The other Scythian campaign, the one the Greeks did report, went west. About 513 Darius crossed into Europe against the Scythians north of the Danube. Herodotus gives it at length, though his real interest is Scythian ethnography and the campaign reads, in Briant's phrase, "more as a pretext than as the true subject"; his figures for the army, 700,000 men and 600 ships, are the usual rhetorical inflation and cannot be used.^[6] The army crossed the Thracian Bosporus on a bridge of boats built by Mandrocles of Samos, and at the crossing Darius set up two commemorative stelae, inscribed in cuneiform and in Greek, listing the peoples of his host; the Ionian tyrants of his fleet were left holding the Danube bridge while the king marched north.^[3] The Scythians refused a pitched battle, fell back, and burned the country as they went, and Darius, unable to bring them to fight, at last turned back on the advice of Gobryas and recovered the crossing.^[6]

Told as Herodotus tells it, this is a failure: a great king marched into Russia and came out with nothing. The modern reading from the Persian side is different, and both of the standard treatments reach it independently. Briant judges the outcome "far from negative," because the campaign's lasting result was the conquest of Thrace and the reduction of Macedonia to a tributary protectorate.^[6] Megabazus, left behind with the army, subdued Thrace and its coast; Macedonia under Amyntas gave earth and water; and Persia held the region for a generation.^[6][3] Thrace was worth holding, with the silver of Pangaeum, timber for ships, and manpower for the levy, and no Achaemenid army was ever again risked beyond the Danube, which the kings took as their northern frontier.^[6] Kuhrt reaches the same conclusion by a sharper route. Darius never used his large fleet to support the land march, and that, she argues, is the tell: the expedition "was limited to fixing the Danube as the effective northern boundary," not to conquering the south Russian steppe.^[3] The historicity of the campaign is not in doubt; what is wrong is Herodotus's frame, the literary motif of a king who oversteps a natural boundary and is beaten by a free people, a motif shaped to prefigure Xerxes.^[3] Kuhrt also cuts a link that older textbooks took for granted. The Scythian expedition was once made the humiliation that provoked the Ionian Revolt, but more than ten years separate the two, and several of the tyrants who served on the Danube were rewarded, not disgraced, so "there can certainly be no direct causal link."^[3]

The campaign has recently acquired physical proof. In the summer of 2016 excavators on the acropolis of Phanagoria, on the Taman peninsula at the far north-eastern corner of the Black Sea, found a fragment of an Old Persian royal inscription bearing the name of Darius, the first Achaemenid royal text ever recovered from the north shore of the Black Sea.^[41] Designated DFa, it is a small piece of a grey marble slab, all that survives of what was once a large royal stela. Shavarebi, who published it in 2019, attributes it to Darius and reads it in the setting of the 513 expedition: the best-preserved verb is plausibly "(I) crossed [the sea]," the same idiom of going across the water that Darius uses at Behistun for the eastern Saka, and Herodotus records that Darius set up stelae at the Bosporus on his way into Europe.^[41][3] Shavarebi suggests the stone was first raised near the western Black Sea shore on the king's route and carried to Phanagoria later, perhaps as ballast in a ship. The reading is not unopposed: the excavator, Kuznetsov, argued for a date under Xerxes and a claim of Persian dominance over the northern Black Sea, and an early press report of the word "Miletus" in the text is dismissed by Shavarebi as a misreading of a form of martiya-, "man."^[41] With that caution, the fragment is the first object to answer Herodotus's account of the crossing from the Persian side.

The push east ran the other way as well. About 515 Darius sent an expedition to the lower Indus, the land the inscriptions call Hindush, and the Carian captain Scylax of Caryanda sailed down the Indus and around Arabia to Egypt, a voyage of reconnaissance and a test of whether the empire's far extremities joined up.^[6] The Indus valley became a satrapy, and its wealth was proverbial: in Herodotus's tribute list India is the twentieth and richest district, paying 360 talents of gold dust.^[3] The reach of these conquests is what Darius's own inscriptions use to define the empire. On the gold and silver foundation tablets from Persepolis he draws its bounds by its edges, and the edges are the frontiers of these wars: "from the Scythians who are beyond Sogdiana, thence unto Ethiopia; from Sind, thence unto Sardis, which Ahuramazda the greatest of the gods bestowed upon me."^[35] The campaigns were a single centrifugal push, north-east against the Saka, south-east to the Indus, west into Thrace, and the king measured his kingdom by how far in each direction it had gone.^[4]

The western frontier broke where the Greek cities of the Ionian coast met the empire, and it broke for reasons that had little to do with Greek freedom and much to do with the working of Persian power. The revolt began in 499 out of a failed imperial venture. Aristagoras, the acting tyrant of Miletus, had proposed to the satrap Artaphernes an expedition to take the island of Naxos and restore some exiled aristocrats, and Darius had authorised it and doubled the fleet.^[42] The expedition failed, and Aristagoras, fearing he would answer for it, changed course: he gave up his own tyranny, proclaimed equal rights in Miletus, and set about deposing the pro-Persian tyrants of the other cities.^[42] The move arose from factional politics within the cities. Miletus was not rich enough to pursue a policy of its own; the very Naxos expedition had needed Artaphernes' backing.^[6] What modern scholarship recovers under Herodotus's tale of a poor-spirited Aristagoras is not a people rising for liberty but the internal rivalries of the cities, and Briant and Kuhrt agree that the mainspring of the revolt was this factionalism, not any "national consciousness," which did not exist.^[6][3] The old economic explanation, that Persian rule had strangled Ionian trade, has not survived the evidence: the coin hoards and tribute records show the cities prospering under Persia, and the most careful modern account traces the outbreak to the logistical shock of the abnormally large Naxos fleet, the painful and unrecoverable requisitions it demanded, falling on grievances that had been simmering unseen.^[42]

In 498 the rebels, joined by twenty ships from Athens and five from Eretria, marched inland and burned Sardis, the satrapal capital, and in the fire the temple of the Lydian goddess Cybele was destroyed.^[42] The Athenian and Eretrian help was slight and soon withdrawn, but it gave Darius the grievance that would shape the next decade: the burning of Sardis, and above all of its temple, became the wrong that had to be repaid.^[42] The revolt spread to the Hellespont, to Caria, and to Cyprus, and the Persian reconquest, traced in the entry on the Ionian Revolt, was methodical and hard.^[42] The war was settled at sea. In 494 the Ionian fleet, some 350 ships, met a larger Persian fleet off the islet of Lade, and the battle was lost when the Samian and other contingents, bought beforehand, deserted at the opening.^[42] With the fleet gone, Miletus was stormed. Most of the men were killed, the women and children enslaved, the oracle-sanctuary of Didyma plundered and burned, and the survivors deported to the mouth of the Tigris; Herodotus called it the third enslavement of Ionia.^[6][42]

What followed the storm is the part the Greek narrative slights. In 493 Artaphernes summoned envoys from the Ionian cities and imposed a settlement. He had their lands surveyed and measured in parasangs, reassessed the tribute on that basis, and bound the cities by oath to submit their disputes to arbitration rather than make war on one another.^[6] The reassessment was not a punishment: the new total, as Herodotus himself records, came to much the same as before, and it stayed in force for the rest of the century.^[42] Even the hostile witness concedes that the measures were "conducive to peace," and modern scholars read the settlement as a piece of statecraft, imposing arbitration precisely because inter-city war was the endemic Greek disease.^[6] It was also a mechanism of control. By treating the revolt's violence as a quarrel among the Ionians rather than a rising against the king, the survey and the arbitration undercut any memory of united resistance, and the same careful measurement refilled the satrapal treasury.^[42] Leniency and control were the same act. The reform outlived the empire that made it: a century later Artaxerxes II and his satrap were still settling a Milesian border dispute the same way, and the Athenians, when they took the Ionian cities into the Delian League, inherited the Persian survey as the basis of their own tribute.^[6][24] Persia governed the reconquered coast by administration as well as by fire.

The reprisal against the mainland Greeks who had helped the rebels came in two campaigns. In 492 Mardonius led an army and fleet along the north Aegean coast. Herodotus treats it as a failed march on Greece, but the objective was the reconquest of the Thracian and Macedonian outposts loosened by the revolt, not Athens, which the distances and the logistics put out of reach in a single season.^[42] The fleet was wrecked in a storm off the promontory of Athos, and the losses, though the reported figures of 300 ships and 20,000 men are conventional and inflated, were real enough to cost a year, but the campaign restored Persian power in the north and made Macedon a tributary subject.^[42] In 491 Darius sent heralds through the Aegean and mainland to demand earth and water, the tokens of submission, and most cities gave them, Aegina among them; the refusers, Athens and Sparta, killed the heralds, and their names went onto the list of targets.^[42] In 490 the punitive expedition sailed under Datis, a Mede, and Artaphernes the younger, the satrap's son. It went by the islands: Naxos, unpunished since 499, was taken and burned, while Delos, the shrine of Apollo, was deliberately spared and honoured with a great offering of frankincense, the calculated benevolence set against the calculated punishment.^[42] The Lindian Chronicle preserves a second act of the same diplomacy, probably from this crossing: when a sudden rain broke his siege of Lindos on Rhodes, Datis made a treaty of friendship with the town and, saying the gods protected it, "dedicated to Athena the clothing he was wearing, including his jewelry, as well as a sword and his covered carriage."^[43] Eretria, one of the two cities that had aided the burning of Sardis, was besieged, betrayed on the seventh day, and sacked; its temples were burned "in revenge for those of Sardis," and its people deported to a village in Elam near Susa, where Herodotus later saw their Greek-speaking descendants.^[44] Then the fleet crossed to the bay of Marathon, chosen for its cavalry ground and as a beachhead away from the city, on the advice of the exiled Athenian tyrant Hippias, who expected to be restored.^[42]

The battle that followed has carried a weight in the Western imagination that it did not carry in the empire that lost it. Marathon was a limited punitive expedition rather than an invasion aimed at annexing Greece. The force was too small to hold the country, and it never aimed to make Greece a province with Persian governors; its purpose was to punish Naxos, Eretria and Athens, to display royal vengeance, and to restore Hippias.^[42] Waters calls it "a minor setback at the end of an otherwise successful campaign"; Dandamaev notes that only a small part of the Persian army was engaged, so that Darius "could justifiably say that he had lost a battle, but not the war," and could even count the campaign a net gain in prestige, so many cities having given earth and water, his own inscription still numbering the peoples "beyond the sea" among his subjects.^[4][7] Herodotus's account is short, Athenocentric, and forgetful, far thinner than his set-piece narratives of Xerxes' battles; his battle narrative even omits the Persian cavalry, a silence that has been explained in several ways. The old hypothesis that the horses had been re-embarked before the battle "really rests on an entry in the Suda," a Byzantine lexicon a millennium and a half late, and Krentz, the battle's most careful modern historian, sets it aside: his own reconstruction keeps the cavalry at Marathon but out of the fight, camped by the lake behind a bottleneck at the Makaria spring and too slow to deploy, so that Miltiades "planned to get inside his enemy's decision cycle" and "the Greeks ran to cross the plain before the Persian cavalry could reach them."^[42][43] The numbers are contested in both directions. Herodotus's 600 triremes is a stock figure; Krentz takes an old water-supply estimate of about 16,000 soldiers at Marathon as "as good a figure as any," with the cavalry in the few hundreds.^[43] The Athenian dead, 192, are known from the public casualty lists and are reliable. The Persian figure of 6,400 divides scholars: Hyland holds the real loss was probably in the hundreds or low thousands, while Krentz notes the count had a mechanism, since Kallimachos's vow of a goat to Artemis for every enemy killed "meant that the bodies were counted, at least approximately."^[42][43] The generals suffered no disgrace. Datis's sons went on to command Xerxes' cavalry, and Artaphernes the younger succeeded his father at Sardis, which is not how a court treats the authors of a catastrophe.^[42]

The Persian side of the battle has a documentary anchor that the Greek legend lacks. The commander Datis is not only a figure of Herodotus: he is the "Datiya" of the Persepolis Fortification tablets, recorded receiving rations while carrying a sealed royal document from Sardis to Persepolis in 494, in the years of the Ionian war; and the ration itself, seventy quarts of beer, gives his rank, for in the whole archive only Parnakka, the king's uncle who ran the Persepolis economy, and Gobryas, Mardonius's father, drew more.^[3][43] The Marathon commander steps out of Greek narrative into the administrative record of the empire he served. That anchor matters, because the Persian sources are otherwise silent about Marathon, and the temptation is to read the silence as a suppressed defeat. It is not: the royal inscriptions are equally silent about the empire's resounding successes, Cambyses' conquest of Egypt among them, for after Behistun they do not narrate particular campaigns at all, but present the empire as a finished, divinely ordered whole. As the Companion puts it, "the Persian view of the events and their importance is entirely disproportional compared with the Greek and western perspective."^[5] Read from the Persian side, the wars against the Greeks were the punishment of rebellion within a world-ruling order, not a contest of political systems. Athens, by giving earth and water in 507 and then repudiating it, had in the king's eyes become a vassal in revolt; the campaigns answered oath-breaking, and the antithesis of East against West, the reading of Marathon as the birth cry of Europe, is a later construction laid over the ancient facts.^[5] Persia was pragmatic about forms of government: it removed tyrants and set up democracies in the Ionian cities when that served its purpose.^[5] The dissent from this consensus is worth stating: Cawkwell argued that the expedition of 490, like that of 480, did aim at the incorporation of Greece, reaching it by the island-hopping route, and that Herodotus's whole account rests on a serious misunderstanding of how the empire worked.^[45] Krentz supplies a middle reading between the punitive raid and the invasion: Datis's island-by-island route, "beginning with Naxos, perhaps the primary target of the campaign," carried out the very plan Aristagoras had put to Artaphernes before the Ionian Revolt, for "Darius tempered aggression with caution," consolidating his hold on the Aegean islands before any reckoning with the mainland.^[43] The current majority holds to the limited, punitive reading, but it does not claim Persia had no designs on the Aegean, only that annexation was not this expedition's task.

Marathon did not end the western war; it interrupted a plan. The Persian court registered the defeat as a slight to be answered, and Darius prepared a far larger expedition, to be led in person, against Greece.^[9] It was the preparation for that expedition, and not Marathon itself, that broke the reign's last years. The exactions raised across Asia for the Greek war, the levies of men, grain and ships, bore hardest on Egypt, the empire's rich western core, and in 486 Egypt revolted.^[46] The rising, led by a native claimant now identified with a pharaoh Psammetichus IV, broke out before Darius died.^[5] A demotic letter from Elephantine dated 5 October 486 shows rebels moving through the southern border country in the reign's last weeks.^[7] Ruzicka's reading makes the interlock exact. The strategy that reached for the far west required squeezing Egypt, and the squeeze endangered Egypt itself; the army already mustered for the Aegean was diverted to reconquer the Nile, and the Greek expedition passed intact to Xerxes, delayed four years because Darius's forces had been spent in Egypt.^[46] Darius died in the middle of these preparations, after a reign of thirty-six years, with the Egyptian revolt unsuppressed and the reckoning against Greece unbegun. Both passed whole to Xerxes, who crushed Egypt first and then turned west.

The end of the reign was a check rather than a collapse. The troubles at the edges were real, but they were troubles at the edges: the Babylonian date-formulae record a smooth transfer of the throne, and the Persepolis archive shows no break in the empire's building and administration.^[3] Briant rejects the picture of a general insurrection that would have crippled the European expedition, and gives the reign's close its truest measure. The revolts in Egypt and Babylonia, he writes, "indicate that Darius's imperial policy had discovered its natural limits."^[6] The Babylonian record underwrites the judgement: in the empire's best-documented province the reign's last years show real land prices flat or falling under a rising fiscal load, and the letters in which a royal official stopped the priests' cult payments at Ezida, their funds requisitioned by the crown, run from late Darius into the first years of Xerxes.^[18] The wars of Darius had carried the empire from the Indus to Thrace and from the Central Asian steppe to the cataracts of the Nile, and where they stopped was not where his armies were beaten but where the cost of holding what he took began to outrun the taking. That is the point the reign ended on: not a broken empire, but an empire that had reached its edge.

Darius names one god above all others, and names him constantly. Ahura Mazdā (Ahura Mazdā, Old Persian Auramazdā) made him king, gave him the empire, and steadied his hand against the Lie; the phrase "by the favour of Ahura Mazdā" runs through Behistun like a refrain, and the tomb inscription at Naqsh-e Rostam opens by setting the god at the head of creation itself, maker of earth and sky and man, "who made Darius king, one king of many, one lord of many."^[35] Does that make Darius a Zoroastrian? He worships the god the Iranian prophet Zarathustra set at the centre of his teaching, and he opposes drauga, the Lie, to what is right; yet he never names the prophet, never quotes the Gāthās, and uses none of the technical vocabulary of the hymns. The century-long debate that gap has fed, from Boyce's confident affirmative, through Kellens's argument that the question itself is unanswerable, to Henkelman's tablets-first scepticism and the middle positions of de Jong and Skjærvø, is staged in full in the entry on Ahura Mazdā.^{[39][47][19][48][49]} What belongs here is its shape, and the parts of the evidence that are Darius's own.

Boyce read the inscriptions as Darius's public profession of Zoroaster's faith, but the props of the maximal case have weakened one by one: the Letter to Gadatas, in which Darius appeared to protect a Greek cult and invoke "the god who told the Persians the true course of events," is now judged a probable forgery of the Roman period, a letter the 1988 Cambridge Ancient History was still citing as genuine evidence of royal policy, which dates the collapse.^[39][3][13] Nor was Boyce's the only maximal voice: that same 1988 volume, testing the inscriptions against explicit marks of belief rather than Boyce's continuity argument, ruled that "in their theology and basic beliefs, therefore, Darius and Xerxes were Zoroastrians", while conceding in the same pages the evidence the sceptics would later build on, that the haoma cult practised at Persepolis "would not have pleased the prophet" and that the crowd of other deities funded there "is shocking by Zoroastrian standards". A note in the same chapters even drew a speculation from Behistun's seams that chimes with the tablets' picture: only the rebels of the late-added postscript are castigated for failing to worship Ahura Mazdā, which made its author wonder whether the god was recognised as the imperial deity only in the composing of the inscription, Darius's personal religion having "come into focus only as a result of the events which marked the first year of his reign".^[13] Kellens answered the whole question by dissolving it: the Achaemenids had no terminology of religious affiliation, could say "Ahura Mazdā is a great god" but not "I am a Zoroastrian," and the modern yardstick is more obscure than the religion being measured against it.^[47] The Persepolis archive pushed the field the same way, and its testimony is the most Darius-specific evidence there is. Ahura Mazdā appears in the Fortification tablets infrequently, usually beside other gods, and with allocations no larger than theirs: in one much-cited text, PF 0339, the god is granted seven quarts of wine while the originally Elamite Humban receives twenty. A king who ran a whole pantheon through his bureaucracy, funding Ahura Mazdā and Humban from the same accounts, is not easily read as the head of a Zoroastrian church.^[19]

Yet the sceptics affirm the connection they deny the label, and here too the evidence is Darius's own. His phrase for "the other gods who are," tayaiy hantiy, is a Gāthic idiom, and the royal names themselves are quotations from the hymns, Darius's own name, Dārayavau-, being a Gāthic phrase meaning that the best thinking sustains.^[47] The material is saturated with Avestan and Gāthic language. What is contested is not whether Darius belonged to the Iranian religious world of Ahura Mazdā and the Gāthās, which he plainly did, but whether the modern category "Zoroastrian" is the right instrument for describing it. The most defensible position now runs this way: Darius was a Mazdā-worshipper steeped in the Avestan tradition, with deep continuity between his religion and the Avesta, but "Zoroastrian" is probably the wrong tool, resting as it does on a definition the specialists no longer share.

Darius is often named as the heir to a Persian policy of religious toleration, a state that let its subject peoples keep their own gods and even helped them rebuild their sanctuaries. The reputation is real, but its documentary basis is far narrower than the celebratory account allows, and most of what once supported it has been taken away.

The most famous prop, the Cyrus Cylinder, belongs to Cyrus and does not carry the weight long put on it. Modern reading treats it as a conventional Babylonian restoration inscription in the old Assyrian mould, not a charter of tolerance, and its handling of the "human rights" myth is set out in the entries on the Cylinder and on Cyrus rather than here.^[3] For Darius, the question is what remains once the Cylinder is read as Babylonian court convention: the case for a proclaimed policy narrows sharply. Amélie Kuhrt states the position bluntly: much of the argument has been circular, assuming the exceptional toleration policy the decree is supposed to prove, and "this novel Achaemenid policy is based entirely on the evidence of Ezra – nothing else supports it." The full demolition is staged in the entry on Cyrus.^[3]

The one document that seemed to show Darius himself protecting a foreign cult was the Letter to Gadatas, a royal letter in which the king rebukes a subordinate for taxing the "sacred gardeners" of Apollo and disregarding the crown's care for the god. For a long time it was read as a genuine Achaemenid letter re-engraved in Roman times, and it did double duty as the personal, Darian counterpart to the biblical evidence. That support has collapsed. Briant's study of the text now judges it very probably a Roman-period forgery, and Kuhrt, following him, records that the old comparison "can no longer stand," so that the classic bundle of "authentic royal letters" has come apart as a group.^[3] What is left is the Book of Ezra, and even there the detail cuts against the standard story: the actual rebuilding of the Jerusalem temple is dated not to Cyrus but to Darius's own second year, 520, whatever the date assigned to the decree that authorised it.^[3]

Stripped of the myth, a real practice remains, and it is worth stating precisely. Darius accommodated local cults, but he did so in their own idiom and on the crown's terms, without proclaiming any universal principle. In Egypt he took up the pharaonic role, ruled in the register of Egyptian kingship, and ordered a codification of Egyptian law: the order reached Egypt in his third or fourth year (519/518), instructing the satrap to assemble "the wise men among the warriors, priests, and all the scribes of Egypt" to set down the law as it had stood in the last year of Amasis, and the commission sat until his nineteenth year, its findings copied at Susa onto papyrus in Aramaic and Egyptian demotic.^[13][10] In the Fortification tablets, likewise, his administration underwrites Elamite priests and Elamite gods alongside Ahura Mazdā. Where the crown confirmed a sanctuary's privileges, as with the immunity later granted the temple personnel at Jerusalem, the arrangement was transactional: a community that had submitted kept its standing, and the king protected it partly to bind it and partly for the credit.^[6] This is pragmatic management of a plural empire, and no declared charter of freedom of worship. The two should not be run together, and the modern myth of a proclaimed toleration policy has to be distinguished from the workaday accommodation the evidence actually shows.

Between the demolished proclamation and the ad hoc accommodation there is also a genuine scholarly middle. Peter Frei argued from the Xanthos Trilingual that the empire possessed a standing legal mechanism, which he named Reichsautorisation, imperial authorisation: a local community's norms could be ratified by the king or his satrap and thereby become, in his pointed phrase, locally valid imperial law, a process he judged at full strength "a fixed institution." His chief exhibits include Darius's Egyptian law-collection, ordered by the king himself, alongside Ezra and the Passover letter, and his own hedges deserve keeping: whether authorisation was obligatory the material cannot say, the Trilingual's wording suggesting it was not, and almost all the surviving cases concern religious matters.^[10] The thesis has been contested since, the standard criticism reading the cases as satrapal pragmatism rather than an institution, but it marks a real position between charter and improvisation: an empire with a procedure for adopting local law as its own, and Darius's codification as its chief early exhibit.

It should also be said that the tone did not stay even across the dynasty. The harsher note, the inscription against the daivas and the language of stamping out cults the crown would not have, belongs to Xerxes, not to Darius, and is treated in his entry. Darius's own dealings with the gods of his subjects are, on the surviving record, those of a ruler who left them in place.

The Greeks handed down two portraits of Darius, and neither is a photograph. One is the shopkeeper who reduced kingship to bookkeeping; the other is the wise, god-favoured father-king and lawgiver. Both are constructions, and reading them is largely a matter of seeing what they were built to do.

The first was, on Herodotus's account, the Persians' own. When Darius set the empire's tribute on a fixed and regular footing, dividing the realm into tax-paying provinces and assessing each, the Persians are said to have nicknamed him kápēlos, the huckster or retail trader: "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."^[50] The fiscal setting of the slur is Herodotus's own:

Having done these things in Persia, he set up twenty provinces, which they call satrapies. He set up the provinces and appointed governors, and set the tribute they should pay according to the peoples ... In Cyrus' and Cambyses' reigns there was no set tribute - instead gifts were brought.

Herodotus 3.89, trans. Kuhrt^[50][3]

The nickname memorialises the birth of a fiscal state, though not quite as Herodotus frames it. His claim that Cyrus and Cambyses took only gifts, never tribute, is wrong: a system of payments had existed before Darius, who regularised it rather than inventing it.^[3] The huckster tag therefore marks the shock of fixity and regularity, the moment tribute became an assessment rather than a present. How far it reflects a genuine Persian memory is disputed. The standard commentary allows that, given the fixed tribute, the spread of coinage and the hoarding of precious metal, the nickname "may be authentic," while noting that Herodotus does not otherwise paint Darius as greedy, and setting it against a competing tradition of the generous king who halved the tribute his officials had set.^[51] Briant is more sceptical still, finding it "quite difficult to find any factual justification for it" and treating the label as Herodotus's own schematising of the three founders rather than a fact of Persian folklore; others read the triad as a Greek parable, sharpened for an Athenian audience weighing itself against Persia, though it may keep a dim memory of how the reigns were felt by their subjects.^[6][5] The fiscal reputation was two-sided from the start.

The idealised Darius is older in Greek literature and no less made. Aeschylus, writing within a generation of the wars, built the Persae on a contrast between the dead Darius and his son: the father faultless and god-favoured, "a king without peer," "the all-powerful, benevolent, invincible, equal to the gods," the son rash and ruinous, denounced by his own father's shade for the hubris of bridging the sea.^[6] Plato drew Darius the lawgiver in the same key: in the Laws he is one of the seven who divided the empire and fixed its tribute, a moderate organiser not reared soft among the palace women and eunuchs, who "added to the territory Cyrus had bequeathed at least as much again."^[6] Diodorus completes the set, listing Darius among the great lawgivers of Egypt, so honoured for repairing Cambyses' sacrileges that "he alone of all kings was addressed as god in his lifetime."^[3] None of these survives as primary text in a form that can be quoted verbatim here; each reaches us through the modern editors and historians who carry the translation.^[6][3]

The most consequential Greek use of Darius, though, is neither portrait but a speech. In the Constitutional Debate (Herodotus 3.80 to 82), the seven conspirators pause before seizing the throne to argue the best form of government: Otanes speaks for isonomia, equality before the law, Megabyzus for oligarchy, and Darius for monarchy, and Darius wins. It is the founding set piece of comparative constitutional theory, and Greek political thought chose Darius as monarchy's spokesman. Tourraix, in the fullest recent treatment, weighs its authenticity the way most now do: Persian accession stories circulated in the empire and Herodotus probably knew them, but nothing suggests that the Greek constitutional vocabulary had Iranian equivalents some seventy years before his day, and the debate's truth is conceptual rather than factual, for in telling history Herodotus is thinking the political. The function-reading of Giulia Sissa, which Tourraix endorses, gives the purpose: in the course of the narrative "democracy will prove to be a unique form of government: the winning one," and "the debate offers an explanation to historical events, in terms of political theory."^[15] Herodotus himself insisted the scene was no invention, pointing to Mardonius installing democracies in Ionia a generation later; the consensus reads the speech as Greek thought in Persian dress. The huckster and the lawgiver were not, then, the Greeks' only Dariuses. They also made him the voice of monarchy itself.

The modern correction strips the ideology out of the picture. Briant's method is to detach the Greek portrait from the history, and he warns that it did not stay in its own literature. The Hellenocentric reading, he wrote, "has infiltrated Iranian studies," so that a passage of Aeschylus gets cited to prove an empire in irreversible decline and a weakling on the throne after Darius.^[6] The idealised father-king and the huckster alike tell us more about Greek preoccupations than about the man. The decisive point is that the wise-father-and-ruined-son antithesis has no counterpart in the Achaemenids' own record. Kuhrt gathers the royal inscriptions to show that Xerxes presents himself in exactly Darius's terms, the same great god who "created happiness for man" and "made Xerxes king, one king of many," the formulae of father and son standing in mirror image; she files them as "two sides of the same coin."^[3] The very formula Boyce read as Darius's Zoroastrian profession is used word for word by the son the Greeks cast as the impious tyrant. The lawgiver too shrinks under scrutiny: the "royal law-giver" is a Greek category, imported by Hecataeus into an Egypt that had no legal code and no notion of a king who legislated, kings there ruling by maat, even though Darius did order a genuine act of law-gathering, the codification of Egyptian law recorded on the verso of the Demotic Chronicle, ordered in his third or fourth year and completed only in his nineteenth.^[3][6][13] The Babylonian evidence sometimes cited to widen that Egyptian act into an imperial law will not carry the weight, as the 1988 Cambridge Ancient History showed from the loanwords themselves: the Old Persian dātu in the Babylonian texts is best rendered "decree, royal command" rather than law code, the official called dātabara appears in the tablets functioning "more as witnesses than as judges", and, Egypt apart, "convincing evidence for major legal reforms and codification under the Achaemenids, and particularly in the reign of Darius, remains elusive". The same chapter tied the two Greek portraits into one administrative fact, reading Darius's law and his taxation as aspects of a single reorganisation of the empire's business.^[13] The Greek Dariuses are Greek work; even the name reached the Greeks wrong, for Herodotus, glossing the royal names at 6.98, renders Dārayavauš as "Doer" or "Restrainer," a folk etymology far from the Old Persian "holding firm what is good," which is why Tavernier calls the first student of Old Persian names "a weak linguist in that matter."^[1] The Persian record, meanwhile, shows one self-identical royal ideology running unbroken from the father to the son the tragedians opposed to him.

Alone among the kings, Darius speaks past his own age to the people who will one day read his monuments, and what he asks of them is belief. At the end of the Behistun inscription he turns from the record to the reader:

Thou who shalt behold this inscription hereafter, let it not seem false to thee; believe it to be the truth.

Darius I, DB, col. IV^[2]

The demand is the most revealing sentence in the whole text. A record that has to insist on its own truth is a record that expects to be doubted, and the king was right to expect it, for the doubt he tried to close is exactly the one that the study of the reign has never been able to close. On his tomb the appeal to the future hardens into a command, binding right conduct and obedience into one:

O man, that which is the command of Ahuramazda, let this not seem repugnant to you; do not leave the right path; do not rise in rebellion.

Darius I, DNb^[37]

It is the most personal thing the Achaemenids left behind, and the most calculated: the same voice that built the pedigree and told the story of the accession, asking those who come after to take both on trust.

Darius died in October 486, with the revolt in Egypt still unbroken, and was laid in the rock-cut tomb he had prepared at Naqsh-e Rostam, its façade carved with the king at prayer before the fire, raised up on the platform of thirty subject peoples. He had already named his heir. The heir was not his firstborn but Xerxes, his eldest son by Atossa, the daughter of Cyrus. Herodotus tells the succession as a court drama: an elder half-brother pressed the custom "that the eldest should rule," and the matter was settled by a queen he calls all-powerful.^[50] The kingmaking Atossa, though, is now read as Herodotus's frame rather than Persian fact, a Greek trope about harem power; what decided the matter, on the Persian evidence, was the argument from pedigree, and the demolition of the kingmaker story is set out in the entry on Xerxes. The pedigree point stands either way: through his mother the throne passed back into the line of Cyrus, so that the union of the two houses which Darius's genealogy had only claimed was made real, at last, in his son.

The empire held in the shape Darius had given it. Xerxes carried his father's creed forward almost unchanged, and the system of satrapies, tribute, roads and coinage that Darius built in the decade after 522 governed the largest state the world had yet seen until it fell, a century and a half later, to Alexander. Whatever the truth of how he took the throne, the empire that is remembered as Cyrus's was, in almost all its working parts, the empire Darius made.