An introduction to

Esther

Author, date, setting — and three ways to read the only book in the Bible that never mentions God

11,139 words

The sentence nobody quotes

Esther is one hundred sixty-seven verses long, and nine verses from the end it says this about itself:

Then Esther the queen, the daughter of Abihail, and Mordecai the Jew, wrote with all authority, to confirm this second letter of Purim. (Esther 9:29)

Second. The Hebrew is iggeret ha-Purim ha-zot ha-shenit — “this, the second Purim letter.” The book files itself as the second writing on its own subject, and takes the first one for granted.

That is a strange thing to invent. If Esther is late fiction, written to give a borrowed festival a Jewish pedigree, then its author declined to claim primacy and rested his authority on a document already in the hands of the people he needed to persuade. And if Esther is what it says it is, the same thing is simply true. Either way, the book’s standing rests on paper it did not produce.

It names that paper nine verses earlier — “And Mordecai wrote these things, and sent letters unto all the Jews” (9:20) — and both letters went out to the hundred twenty and seven provinces (9:30). Not into an archive. Into houses.

Hold onto that sentence. Every reading of the book below is finally an attempt to explain it.

The setting

The king. Hebrew Aḥashwerosh renders Old Persian Khshayārshā, which comes into Greek as Xerxes. He is Xerxes I, who reigned 486–465 BC — the king who bridged the Hellespont, burned Athens, and lost at Salamis. (The Greek Esther and Josephus call him Artaxerxes instead; that is a translator’s substitution, and it is one of the reasons ancient readers disagreed about which king this was.)

The dates the book gives. The narrator is unusually precise, and the story is anchored to the regnal year throughout:

So the book covers about a decade, and it dates its own events the way an administrator would.

The place. Shushan the palace — Susa, in what is now southwestern Iran, the Persian winter capital. Not Babylon, not Jerusalem. The empire is described as running “from India even unto Ethiopia, over an hundred and seven and twenty provinces” (1:1).

The people. These are not returned exiles. In 538 BC Cyrus permitted the deportees to go home, and some did; Ezra and Nehemiah are about them. Esther is about the ones who stayed — a settled, prospering, entirely diaspora community, deep inside the empire, with a man of theirs holding a post at the palace gate. Nobody in the book so much as mentions Jerusalem.

One note on the arithmetic. Esther 2:6 says of Mordecai’s line that he “had been carried away from Jerusalem with the captivity which had been carried away with Jeconiah king of Judah.” Jeconiah’s deportation was 597 BC. If that clause describes Mordecai himself, he is over a hundred and twenty years old and his cousin is a beauty contestant. Read as Hebrew usually reads — the relative clause attaching to the nearest named ancestor, Kish, at the end of the genealogy in the previous verse — the problem disappears. Critics take it the first way; defenders take it the second. It is a real ambiguity in the Hebrew and worth knowing about before somebody springs it on you.

The machinery. The world of the book is the Achaemenid administrative state, and the book is soaked in it: a chancery of scribes writing every decree in every province’s script and every people’s language; a courier relay running the Royal Road; a sealed signet authenticating royal writing; a register of the king’s benefactors; an archive of daily records that can be pulled and read aloud at two in the morning. Keep that machinery in view. It turns out to be the whole argument.

What happens

Briefly, for anyone who needs the spine. The king deposes his queen Vashti for refusing to appear at a banquet, and a decree goes out that every man should rule in his own house. A search is made for a new queen, and a Jewish orphan raised by her cousin Mordecai is chosen — she does not disclose her people. Mordecai uncovers an assassination plot and it is recorded in the royal annals, unrewarded. Haman the Agagite is promoted; Mordecai will not bow; Haman obtains a sealed, irrevocable decree for the extermination of every Jew in the empire on a single day. Esther risks death by approaching the king unbidden, exposes Haman at a banquet, and Haman is hanged on the gallows he built for Mordecai. Because the first decree cannot be revoked, a second is issued authorizing the Jews to defend themselves. They do, and win, and three times over the narrator stops to say they took no plunder. Mordecai and Esther write letters establishing the feast of Purim, and the book closes with a note in the royal annals about how great Mordecai became.

The backstory the book never explains

Twice in the first three chapters the narrator drops a piece of genealogy that a reader without the rest of the Old Testament will slide right past.

Now in Shushan the palace there was a certain Jew, whose name was Mordecai, the son of Jair, the son of Shimei, the son of Kish, a Benjamite (Esther 2:5)

After these things did king Ahasuerus promote Haman the son of Hammedatha the Agagite (Esther 3:1)

A Benjamite son of Kish, and an Agagite. Those are not decorations. They are the two sides of an unfinished war, and the reader is expected to know it.

The tribe’s charter. Jacob’s last words over his youngest son:

Benjamin shall ravin as a wolf: in the morning he shall devour the prey, and at night he shall divide the spoil. (Genesis 49:27)

A wolf, and spoil. Remember the spoil.

Now read the same verse in the Greek Bible, because it does not say the same thing:

Benjamin, as a ravening wolf, shall eat still in the morning, and at evening he gives food. (Genesis 49:27, Septuagint, Brenton’s translation)

Both plunder words are gone. The Hebrew carries booty twice — ‘ad, “prey,” in the morning, and shalal, “spoil,” at evening. The Greek has neither. In the morning the wolf simply eats; the Greek reads the rare Hebrew word for “prey” as the ordinary adverb “still.” And what he hands out at evening is trophē — food.

That is almost certainly a decision about a difficult word rather than a theological one. But it had consequences: the Greek-reading church inherited a Benjamin who ends his day feeding people, and roughly fifteen of the Fathers, beginning with Tertullian, read the two halves of this verse as the two halves of one man’s life — Saul ravaging the church in the morning, Paul distributing the gospel at evening. That argument belongs to the companion article. What matters here is which Bible you are standing in: the spoil thread that runs through the rest of this page exists only in the Hebrew. In the Greek there was never any plunder for Benjamin to divide, or for anyone to refuse.

The war. Israel’s first king was Saul, son of Kish, of Benjamin. He was ordered to destroy Amalek utterly (1 Samuel 15:3), and he failed in exactly two ways:

But Saul and the people spared Agag, and the best of the sheep, and of the oxen, and of the fatlings, and the lambs, and all that was good, and would not utterly destroy them (1 Samuel 15:9)

He kept the king alive, and he kept the stuff. The sentence for it:

And Samuel said unto him, The LORD hath rent the kingdom of Israel from thee this day, and hath given it to a neighbour of thine, that is better than thou. And also the Strength of Israel will not lie nor repent: for he is not a man, that he should repent. (1 Samuel 15:28–29)

Two clauses: the throne goes to a neighbour better than the one who lost it, and the word that says so cannot be taken back.

Now open Esther at chapter one. Vashti is deposed, and the sentence reads:

let it be written among the laws of the Persians and the Medes, that it be not altered, That Vashti come no more before king Ahasuerus; and let the king give her royal estate unto another that is better than she. (Esther 1:19)

Same two clauses, in the same order: the crown to one better than the one who lost it, under a decree that cannot be taken back. And the KJV’s own margin at that verse gives the Hebrew away — against “unto another” it prints: “Heb. unto her companion.” The word is re’ut, the feminine of the re’a that the same translators rendered “neighbour” back in 1 Samuel 15:28. The English hides the echo. The margin restores it.

And then the ending. The Agagite hangs. His ten sons hang. And three separate times, in three consecutive scenes, the narrator interrupts a battle report to tell you what the Jews did not do:

but on the spoil laid they not their hand (9:10) but on the prey they laid not their hand (9:15) but they laid not their hands on the prey (9:16)

Kill the Agagite; take no plunder. Saul’s two failures, undone one at a time, by a man of Saul’s own house and tribe.

So the book opens with the language of Saul’s deposition and closes with the repair of Saul’s disobedience — and it never once says so. That reading of Esther as the rematch of 1 Samuel 15 is old and standard in Jewish tradition, not a modern invention. What we do with it afterward — where it goes in the New Testament, and why the wolf of Genesis 49 matters — is the subject of a companion article on this site, The Wolf That Became a Sheep.

The other name in the genealogy

Mordecai’s introduction carries four names, and so far this page has cashed only the last one. Read it again: “Mordecai, the son of Jair, the son of Shimei, the son of Kish.” There is a story behind the middle name too, and it is a panel of Benjamin’s arc that usually goes missing.

During Absalom’s coup, David abandons Jerusalem and flees east. At Bahurim a man comes out to meet the column:

behold, thence came out a man of the family of the house of Saul, whose name was Shimei, the son of Gera: he came forth, and cursed still as he came. And he cast stones at David (2 Samuel 16:5–6)

The curse is a house-of-Saul grievance, stated in so many words: “The LORD hath returned upon thee all the blood of the house of Saul, in whose stead thou hast reigned” (16:8). Abishai asks permission on the spot: “Why should this dead dog curse my lord the king? let me go over, I pray thee, and take off his head” (16:9). And David’s refusal names the tribe:

Behold, my son, which came forth of my bowels, seeketh my life: how much more now may this Benjamite do it? let him alone, and let him curse; for the LORD hath bidden him. It may be that the LORD will look on mine affliction, and that the LORD will requite me good for his cursing this day. (2 Samuel 16:11–12)

When the coup collapses, Shimei meets the returning king at the Jordan to beg for his life. Abishai wants him dead a second time, and David answers with an oath: “Thou shalt not die” (2 Samuel 19:23).

Then Esther 2:5: Mordecai, the son of Jair, the son of Shimei, the son of Kish. The Talmud does not treat that as a coincidence — it treats it as the reason Mordecai exists. In the same passage that puzzles over Mordecai’s genealogy, the rabbis have the tribes competing for credit for him, and Judah’s claim is exactly this wire: “The family of Judah would say: I caused the birth of Mordecai, as only because David did not kill Shimei, the son of Gera, when he cursed him (see II Samuel 16) was it possible for Mordecai to be born later from his descendants” (b. Megillah 12b). And with the Talmud’s usual honesty, the facing page runs the same wire with the charge reversed — Rava has the congregation complain that David’s mercy is also why there was a Mordecai for Haman to hate, just as Saul’s sparing of Agag is why there was a Haman at all (13a). Credit or blame, the identification is the same on both readings: the Shimei of Esther 2:5 is the Shimei of Bahurim.

Now count the panels. The tribe itself is spared at the brink of annihilation, and the remnant produces Israel’s first king (Judges 20–21). David spares a cursing Benjamite of Saul’s house, twice, once under oath — and the spared line produces the deliverer of the Jews. Mercy shown to Benjamin keeps returning as deliverance through Benjamin. And the return has an address: Haman’s decree ran the whole empire, and Judea was one of the hundred twenty-seven provinces — the house of David was under it like everyone else. David’s hope on the worst day of his reign, “it may be that the LORD will requite me good for his cursing this day,” reads, on the Talmud’s identification, as a prayer answered five centuries later in Susa: the curse he refused to answer came back as the man who saved his sons.

Two honesties, briefly. The plain text of Esther 2:5 can be read as naming Mordecai’s immediate ancestors, men of Saul’s house carrying its heirloom names — Hebrew genealogies telescope freely, and the identification with the Shimei and the Kish is the traditional reading, not a demonstrable one; though on either reading the narrator chose the two names that plant Mordecai in Saul’s house. And David’s mercy had a horizon: he left Shimei to Solomon’s judgment, and Solomon executed him years later when he broke his parole (1 Kings 2) — the sparing lasted exactly long enough to matter. One more thing worth noticing on the way past: David in that scene — leaving the city, crossing the brook, weeping up the ascent of Olivet, betrayed by his own counselor, cursed and refusing to answer — is walking a route the reader of the Gospels has seen before, and the refusal itself has its antitype named: “when he was reviled, reviled not again” (1 Peter 2:23).

The order to write it down

There is one more thing the Amalek war left unfinished, and it is the strangest of all. When Amalek first attacked, God gave Moses an order — not to fight, but to write:

And the LORD said unto Moses, Write this for a memorial in a book, and rehearse it in the ears of Joshua: for I will utterly put out the remembrance of Amalek from under heaven. (Exodus 17:14)

That is the only standing command in the Torah to write a particular thing in a book, and it is attached to this war. Deuteronomy repeats the charge — “thou shalt blot out the remembrance of Amalek from under heaven; thou shalt not forget it” (25:19) — and pairs remembering with blotting out, which is the tension the whole business runs on. You are ordered to erase a memory and to preserve the record of erasing it.

Now read how Esther ends. Not the battle — the paperwork.

And that these days should be remembered and kept throughout every generation, every family, every province, and every city; and that these days of Purim should not fail from among the Jews, nor the memorial of them perish from their seed. (9:28)

And the decree of Esther confirmed these matters of Purim; and it was written in the book. (9:32)

Two things are happening in those verses. The first is an inversion: Amalek’s remembrance is to be put out, and here is a remembrance that shall not perish. The second is lexical — the Hebrew of 9:32, ba-sefer, “in the book,” is the same phrase Exodus 17:14 used when it issued the order.

And this is not our reading. It is one of the best-attested traditional readings of the book, and it is old. The tannaitic midrash on Exodus — the Mekhilta de-Rabbi Ishmael, at this very verse — breaks Exodus 17:14 into three clauses and assigns each to a different Amalek writing: “Write this,” what stands in the Torah; “a memorial,” what stands in the Prophets, meaning 1 Samuel 15; “in the book,” what is written in the Megillah. The Babylonian Talmud carries the same derivation, naming Rabbi Elazar HaModa’i as its tradent, and preserves the dissent of Rabbi Yehoshua, who assigns the third clause to the Prophets and leaves Esther out (Megillah 7a). A second passage, at Megillah 18a, makes the link a formal verbal analogy between Esther 9:28 and Exodus 17:14.

What that means is worth stating plainly: on the Talmud’s own account, the question of whether the book of Esther was allowed to exist was a question about the Amalek writing order. The recorded objection to canonizing it is that Amalek is allotted three written mentions and not four. Esther is admitted as the third — supplied by the words “in the book.”

And the practice follows the theory. On Purim morning the synagogue reads Exodus 17:8–16, the Amalek war and the order to write, and then reads the Megillah.

One more thing, because it bears directly on the sentence this page opened with. In the Talmud’s telling, Esther petitions the sages — “write me for the generations” — and they resist, on the grounds that publicizing a Jewish victory will draw the anger of the nations. Her answer is not theological. It is archival:

I am already written in the chronicles of the kings of Media and Persia. (b. Megillah 7a)

That is Esther 10:2, in Esther’s own mouth, used as the argument for the book’s existence. And it is the same move the Jerusalem Talmud makes independently, as we will see further down. Whatever else the rabbis thought about this book, when they had to defend it they reached for the Persian archive.

Author and date

The book names no author. What it does instead is describe, repeatedly, documents being written:

No book of the Bible talks about its own paperwork this much.

Tradition. Josephus takes Mordecai for the author. The Talmud (Bava Batra 15a) assigns the writing of Esther to the Men of the Great Assembly, and a related tradition has Esther and Mordecai compose it and petition the sages to include it. These are reception, not signatures.

Modern dating runs from the late Persian period (late fifth to fourth century BC) to the Hellenistic (second century BC), and the linguistic evidence is genuinely two-edged. The Hebrew is Late Biblical Hebrew, which pushes late. But the loanword profile is remarkable: Esther carries more Persian vocabulary per verse than any other book in the Bible — roughly thirty Persian and Elamite personal names and a substantial stock of Old Persian administrative terms, including dat, “law,” straight from Old Persian dāta — and not one Greek loanword. For a book supposedly composed under Greek-speaking administrations after 331 BC, that silence is worth explaining.

The earliest outside mention of the festival is 2 Maccabees 15:36, from the late second century BC, which calls the fourteenth of Adar “Mordecai’s day.”

Three different Esthers survive. This matters more than most introductions admit:

  1. The Hebrew (Masoretic) text — 167 verses, no God.
  2. The Old Greek (Septuagint) — the Hebrew plus six large Additions, which supply prayers by Mordecai and Esther, God named throughout, a dream and its interpretation, and the full texts of both royal decrees. This is the Esther the early church actually read, and it answers the “godless book” objection before it can be raised, which is a large part of why the West kept the book while the East hesitated.
  3. The Alpha-Text — a third, shorter Greek form, differing substantially from both.

The Greek text closes with a colophon claiming that “Dositheus, who said he was a priest and Levite, and Ptolemy his son” brought the letter of Purim to Egypt, and that Lysimachus son of Ptolemy, in Jerusalem, had translated it — dated to the fourth year of a Ptolemy and Cleopatra, which puts it either at 114 BC or 78/77 BC. Whatever its worth, it is an ancient claim about how the document traveled.

Canon history. Esther is the only book of the Hebrew Bible with no fragment among the Dead Sea Scrolls. It is never quoted in the New Testament. The Talmud preserves a live argument over whether it “defiles the hands” — the technical test for canonical status (Megillah 7a). Melito of Sardis leaves it off his list around AD 170; Athanasius in 367 puts it outside the canon proper. Luther wished it gone. A book this contested is a book worth introducing carefully.

The three readings

Below are the two standard positions, each written as strongly as its own advocates would write it, followed by the reading this site argues for. Read all three before deciding. Every scripture quotation is checkable in any Bible you own; every external claim is sourced at the bottom of the page.

Case One: it happened

The historical reading, stated at its strongest.

Start with the fingerprint. The author of Esther knows the Persian court from the inside, and knows it in ways that a later storyteller had no way of learning. Not the famous things — the obscure, procedural things.

The register of benefactors. The entire hinge of the plot is that Mordecai’s service to the king was written down in a register, went unrewarded, and surfaced years later when the record was read aloud (2:23; 6:1–3). That is not a novelist’s contrivance. Herodotus reports that the Persians kept a formal list of the king’s benefactors — orosangai — with their services recorded and their rewards owed (8.85). Esther’s plot turns on a genuine, and genuinely obscure, institution of Achaemenid government.

The courier relay. Both decrees go out “by posts,” urgent, riding royal horses (3:13; 8:10, 14). Herodotus describes exactly this: the angareion, a relay of riders and stations, the fastest thing in the world (8.98).

The chancery. Every edict is issued in the script of each province and the language of each people (1:22; 3:12; 8:9), written in the king’s name, sealed with the king’s ring. That is documented Achaemenid multilingual practice, not invention.

The court. Seven counselors who “saw the king’s face” (1:14) — Herodotus knows the seven noble houses with right of access (3.84). The golden scepter, the death rule for unbidden approach, the king’s gate as a place where an official sits, the Old Persian dat for law: it is all in place.

The palace. Jean Perrot, who directed excavations at the palace of Darius and Xerxes at Susa for over a decade, wrote that one “rereads with a renewed interest the book of Esther, whose detailed description of the interior disposition of the palace of Xerxes is now in excellent accord with archaeological reality.”

The chronology fits the king we independently know. Esther’s great assembly of “the power of Persia and Media, the nobles and princes of the provinces” falls in Xerxes’ third year (1:3) — precisely when Herodotus has Xerxes convene his nobles and generals to plan the invasion of Greece (7.8). Then the book goes quiet for four years. Xerxes is in Greece. He comes home beaten in 479, and Herodotus’s account of his return is a squalid business of harem intrigue (9.108–113) — which is exactly the moment, the seventh year (2:16), when Esther has him collecting a new queen.

And there is a name. The Persepolis texts record a court official named Marduka at Susa in the late years of Darius I or the early years of Xerxes — the same name, the same city, the same reign, the same kind of post.

Now the objections, answered.

“Herodotus says twenty satrapies; Esther says a hundred twenty-seven provinces.” Different words for different things. The Hebrew medinah is a smaller administrative district, not a satrapy; large satrapies contained many. Herodotus is counting tax districts.

“Persian kings could only marry from seven noble families.” That is one sentence in Herodotus (3.84), and Herodotus is not a modern historian — he records custom, hearsay, and legend alongside fact. If it was a custom rather than a law, a strong-willed king could ignore it; and the book takes conspicuous pains to say twice that Esther did not disclose her origin, which is precisely how such a rule gets circumvented.

“Herodotus names Amestris as Xerxes’ queen in those years.” The book never says Esther was the only royal wife; she waits thirty days without being called (4:11), which is not how a sole consort lives. And “Esther” may itself be a shortened form of “Amestris” — foreign names get clipped constantly.

The strongest form of this case is the simplest. The book expects to be checked. It dates events to the regnal year, names officials, quotes decrees, and finishes by pointing the reader at a public archive. Fiction does not send its readers to the record office. That is the behavior of a document that believes it is true and expects its first readers to be able to verify it.

Case Two: it is the charter for a festival

The literary-fiction reading, stated at its strongest.

Begin with what the Persians and Greeks do not know. Xerxes is one of the best-documented monarchs of the ancient world. His queen was Amestris, daughter of Otanes, and Herodotus knows her well enough to record her cruelties. She is queen in precisely the years Esther is supposed to be queen. And no Persian, Babylonian, Egyptian, or Greek source knows anything about a Jewish queen of Persia, a Jewish grand vizier named Mordecai, an empire-wide decree of extermination, or the seventy-five thousand dead the book reports. An event of that scale leaves marks. This one left none.

Then look at the scale of everything. A banquet lasting a hundred and eighty days. A year of cosmetic preparation for a single night (2:12). A gallows fifty cubits high — seventy-five feet, taller than the palace. A decree of annihilation published eleven months in advance, giving every intended victim most of a year to leave. These are festival numbers. They are the proportions of a story told to a hall, not a report filed with a government.

Now look at what is not in the book. That God is never named is the one thing everybody knows about Esther, and by itself it proves nothing. Keep going, though, and the hole gets much larger than a name. There is no Jerusalem. No temple. No covenant. No law of Moses. No prayer, no sacrifice, no priest, no prophet, no blessing, no thanksgiving. A book about the deliverance of the Jewish people contains no religious content of any kind.

The economical explanation is that it was not a religious text. It was a diaspora court tale — a good story about clever Jews outmaneuvering a wicked official at a foreign court, of a type well known in the period — and it was pressed into service as scripture when the festival needed a warrant.

And the best evidence for that is what happened the moment somebody tried to use it as scripture: they put the religion back in. The Greek translation carries six large Additions supplying prayers by Mordecai and by Esther, God named throughout, a prophetic dream and its interpretation, and pious speeches at the throne. The translators felt the hole and filled it. Nobody has to patch a book that was religious to begin with.

Look at the genre. Adele Berlin’s reading is hard to shake once you have seen it: Esther is farce and burlesque — pompous officials, a king who cannot decide anything and is managed entirely through his own decrees, a villain marched through the streets leading his enemy’s horse, a wife problem solved by an empire-wide memorandum on husbandly authority. The comedy is not incidental; it is the mode. Nobody asks whether a farce happened.

Now the names. The hero is Mordecai, and the Babylonian chief god is Marduk. The heroine is Esther, and the Babylonian goddess is Ishtar. The festival is Purim, from pur — and the book has to stop and translate that word for its own readers: “they cast Pur, that is, the lot” (3:7). A Jewish festival bearing a foreign name that Jews had to have explained to them, celebrated by two figures named after Babylon’s chief god and goddess, is not what a native Israelite institution looks like. It is what an adopted one looks like.

And that is the argument. Purim is not in the law of Moses. It has no sacrifice, no temple connection, no priestly role, no Palestinian setting — the only Jewish festival with none of these. What it has is a book explaining where it came from. Charter myths work exactly this way: the practice comes first, the story that authorizes it comes after.

The silences line up. No fragment of Esther at Qumran, the one book of the Hebrew Bible with none, in a community that copied everything and would have had strong opinions about a festival not in the Torah. No quotation in the New Testament. No mention in Ben Sira’s long roll call of Israel’s famous men. The earliest external notice of the festival is 2 Maccabees, centuries after the events.

And here is the card that wins the hand. We do not have to reconstruct the objection to Purim — Jewish tradition recorded it. The Jerusalem Talmud (Megillah 1:5) preserves the fight in the open:

Rebbi Samuel bar Naḥman in the name of Rebbi Jonathan: 85 Elders and among them more than 30 prophets were sorry about this matter. They said, it is written: These are the commandments which the Eternal commanded to Moses. … So did Moses say to us: No other prophet will reveal to you anything afterwards. And Mordecai and Esther want to introduce something new?

The same passage records the community pushing back on the letters themselves: “are not the troubles which come upon us enough, that you want to add to ours the trouble of Haman?” A festival that had to be argued into the calendar against the charge that it was an innovation, over the objection of elders and prophets, is exactly what an adopted foreign festival looks like from the inside — and the book of Esther is the argument that won.

Case Three: it is the king’s paperwork

The reading this site argues for.

Both cases above are strong, and they fight over the wrong question. They both take Esther for a work of Jewish religious literature — and then one side defends its accuracy while the other explains its purpose.

Start with the hole Case Two just put its finger on, because Case Two is right that it is the most important fact about the book, and right that “the author was being tactful” is not an explanation. No God. No Jerusalem, no temple, no covenant, no law of Moses, no prayer, no sacrifice, no priest, no prophet. When Mordecai wants to tell Esther that deliverance is certain, the strongest thing he can say is that relief will arise “from another place” (4:14). The book is not coy in one verse. It has no religious vocabulary from the first verse to the last.

Case Two has the fact and the wrong explanation. That absence is not a defect, and it is not a patch job waiting to be discovered. It is a genre — and, at the end of this section, it turns out to be a clue about what kind of feast Purim is.

Somebody noticed this a long time ago. Robert Gordis, one of the major Jewish biblical scholars of the twentieth century, made the case in the Journal of Biblical Literature in 1981, and his opening sentence is the whole thesis:

A Jewish author undertook to write his book in the form of a chronicle of the Persian court, written by a Gentile scribe.

Gordis then lists the evidence, and the list is impressive:

  • The narrator makes no reference to the God of Israel, to Jewish practice, to Jewish history, or to Jewish concerns — beyond resistance to the genocide itself.
  • Mordecai keeps being tagged “the Jew” (5:13; 6:10; 8:7; 9:29, 31; 10:3) — an ethnic label an insider would not need and an outsider would supply automatically.
  • The Jews are always “they,” never “we.”
  • The king’s name appears twenty-one times and never once without his title. “A chronicler at court would be careful to observe the dignity of the king’s estate.”
  • The counselors are listed by name (1:10, 14), an intrusion into the story and a necessity of court protocol.
  • Haman’s ten sons are catalogued by name at their execution — a list with no narrative function whatever, and exactly what an official record does.
  • The decrees are quoted verbatim at promulgation and at countermand (3:13; 8:10–11), in the bureaucratic manner: “to each province in its script and to each people in its language,” “to destroy, kill and annihilate.”
  • The narrative is all event and no motive. We are never told why the king held two banquets, why Vashti refused, what became of her, why the eunuchs conspired, or even why Mordecai forbade Esther to reveal her people. Gordis: these are “precisely what one would expect in an official court chronicle, which would naturally record events rather than expatiate on motives.”

His conclusion: the book of Esther “represents a unique genre in the Hebrew Bible — ostensibly a royal chronicle by a Gentile scribe at the Persian court.”

Now look hard at the word doing all the work in that sentence. Ostensibly. Every item on Gordis’s list is evidence that the text behaves like a Persian court record. None of it is evidence that a Jew was imitating one. The imitation is not a finding; it is the assumption the finding was fitted into.

Drop the word and ask what happens.

The book claims a chancery origin for itself, step by step. This is not inference. It is the narrative:

  • 3:12 — “Then were the king’s scribes called … in the name of king Ahasuerus was it written, and sealed with the king’s ring.”
  • 8:8 — “for the writing which is written in the king’s name, and sealed with the king’s ring, may no man reverse.”
  • 8:9 — “Then were the king’s scribes called at that time … and it was written according to all that Mordecai commanded.”
  • 8:10 — “And he wrote in the king Ahasuerus’ name, and sealed it with the king’s ring, and sent letters by posts.”
  • 9:20 — “And Mordecai wrote these things, and sent letters unto all the Jews.”
  • 9:32 — “And the decree of Esther confirmed these matters of Purim; and it was written in the book.”
  • 10:2 — and the whole account of Mordecai’s greatness is in the royal annals, where you may go and read it.

By the book’s own account, the Jewish vizier had the king’s scribes, the king’s name, and the king’s seal at his disposal, and used them. There is no step missing between that and a court record about Mordecai existing in the Persian archives. The book says the machine that would produce such a record was in Mordecai’s hands.

“But it’s in Hebrew, and the Persian chancery wrote Aramaic.” True, and the book answers it in the same breath it raises it. Esther 8:9 describes the chancery issuing that very decree —

unto every province according to the writing thereof, and unto every people after their language, and to the Jews according to their writing, and according to their language.

A Hebrew-script, Hebrew-language edition of an imperial document, produced by the king’s own scribes, is not a problem for this reading. It is the reading’s own paperwork, described in the text. (Honest limit: 8:9 establishes that the chancery produced Jewish-language documents. It does not by itself prove the narrative is one.)

The book has a seam, exactly where this reading predicts one. Read straight through and the texture changes twice. Chapters 1 through 9:19 are event, date, decree, and register — the court record. Then 9:20–32 stops being narrative and becomes legislation: Mordecai writes, the Jews undertake, Esther confirms, “and it was written in the book.” Then 10:1–3 is a colophon in the exact annalistic formula that closes the reigns in the book of Kings — tribute, acts of power, and a pointer to the chronicles.

Gordis saw the seam and had to explain it: the establishment of a Jewish holiday, he wrote, “is strictly speaking beyond the purview of a Persian chronicler. Hence, the establishment of Purim is ordained in two appendices.” Which is true, and is a problem for a literary imitation — the imitator broke his own device. On this reading it isn’t a device at all. It is a document boundary: the imperial record, plus the Jewish community’s own festival legislation attached to it, plus the archival citation that tells you where the first part came from.

“But no Persian royal chronicle has ever been found.” This is the objection that sounds decisive and isn’t, and the reason is physical. The Achaemenid chancery wrote Imperial Aramaic on leather and papyrus. Clay tablets survive by the tens of thousands — the Persepolis Fortification archive alone runs to some thirty thousand — because clay is nearly indestructible. Leather is not. The one substantial body of Achaemenid chancery correspondence we possess, the Aršāma letters in the Bodleian, survives because sealed leather documents happened to sit in dry Egyptian ground in a leather bag. There is no surviving Achaemenid royal chronicle for the same reason there are no surviving Achaemenid account ledgers on parchment: the medium is gone.

Gordis conceded this in his own footnote — that no historical chronicles of any kind survive from the Achaemenid period — and then noted that M. N. Dhalla, in Ancient Persian Literature, cites Esther itself as the evidence that the Persian kings kept such chronicles. That is the situation in one sentence: the standard scholarly treatment already uses Esther as a witness to Persian record-keeping, while declining to consider that it might be a specimen of it.

What we do have are the neighbors of such a document, and they all fit.

The Bible already contains Persian archival practice. When the rebuilding of the temple is challenged, Darius orders a search of the imperial archive, and the file is retrieved:

Then Darius the king made a decree, and search was made in the house of the rolls, where the treasures were laid up in Babylon. And there was found at Achmetha, in the palace that is in the province of the Medes, a roll, and therein was a record thus written (Ezra 6:1–2)

Ezra also quotes Persian chancery documents verbatim, in Aramaic, at length. The idea that a Persian imperial document could end up preserved inside Hebrew scripture is not a hypothesis. It is Ezra.

A Jewish festival regulated by imperial letter is documented. In 419 BC — sixty years after Xerxes — a letter now called the Passover Papyrus reached the Jewish garrison at Elephantine in Egypt, telling them, on authority passed down from Darius II through the satrap, how to keep the festival of unleavened bread. A Jewish community’s festival observance, fixed by a letter sent through Persian imperial channels. That is Esther 9:20–32 attested as ordinary practice in the same empire, within living memory of the same events.

And sealed satrapal letters carried by couriers are in our hands. The Aršāma correspondence is precisely the kind of object the book keeps describing.

One more piece, and it comes from the other side’s own witness. Recall the Jerusalem Talmud passage that Case Two uses so well — the elders objecting that Mordecai and Esther “want to introduce something new.” Read the sages’ answer, which sits a few lines earlier in the same text:

They told them, if this is your fear, already it is written and deposited in Archives: are these not written in the annals of the kings of Media and Persia?

Their defense against the charge of innovation was that the record was already lodged in the imperial archive — and they cited Esther 10:2 to prove it. That is a rabbinic argument, and it is offered here as reception, not proof. But it settles one thing: the earliest Jewish readers we can hear did not take 10:2 as a literary flourish. They took it as a real citation of a real document in a real archive, and they leaned their case on it.

If this book is fiction, here is where it lies to you

There is a difference between a story being untrue and a story making a false claim, and Esther is unusually full of the second kind.

A novel invents. It can furnish a palace, name officials, count casualties, and describe court procedure, and none of that is deceit — it is furniture, and every reader understands the contract. But a book that stops mid-sentence to tell you where the paperwork is filed has stepped outside the contract. It is no longer asking you to imagine something. It is asking you to believe something, about the world you actually live in, that you could in principle go and check.

Esther does that constantly. Read the whole book with a pencil and fifty-five of its one hundred sixty-seven verses — one in three — contain a statement that does no narrative work at all. They sort into ten kinds:

Count What it is Where Advances the plot?
7 Archive citations — a document is named and located 2:23, 3:14, 4:8, 6:1, 8:13, 9:32, 10:2 2:23 and 6:1 only
3 Chancery and language notes — issued in every province’s script, every people’s language 1:22, 3:12, 8:9 no
4 Glosses of foreign words for the reader’s benefit 2:7, 3:7, 9:24, 9:26 no
6 Calendar equations — ordinal month given a Babylonian name 2:16, 3:7, 3:13, 8:9, 8:12, 9:1 no
6 Parenthetical identifications of persons, offices, and custom 1:1, 1:13, 1:14, 2:5–6, 2:15, 9:29 no
7 Persian law and procedure explained 1:8, 1:19, 2:12, 2:14, 4:2, 4:11, 8:8 4:11 and 8:8 only
3 Name lists — twenty-four people who never act 1:10, 1:14, 9:7–9 no
9 Numbers, measurements, and décor 1:4, 1:5, 1:6–7, 3:9, 5:14, 7:9, 9:6, 9:15, 9:16 no
4 Ongoing custom explained to the reader’s present 9:19, 9:26–28, 9:31, 9:32 no
2 Annalistic coda 10:1, 10:3 no

Most of that is furniture, and we are not going to pretend otherwise. Twenty-four named Persians who never speak, a pavement of four colors of marble, a gallows measured in cubits — a capable author invents all of it, and Gordis is right that this is what a deliberate imitation of a court record would look like.

But four items on that list are not furniture. They are claims. If the book is fiction, these are the places where it does not merely invent — it asserts.

One. It cites archives that would have to exist. Not vaguely. Twice it says a specific entry was filed (2:23, 9:32), once it has that filing retrieved and read aloud years later (6:1), and once, in its closing sentence, it sends the reader to the file itself: “are they not written in the book of the chronicles of the kings of Media and Persia?” (10:2). If there was no such archive, or no such entry, the author did not imagine a document. He invented a citation and pointed his readers at it.

Two. It describes imperial paperwork that would have to have been produced. The book says the king’s scribes were convened on a stated day of a stated month, that the text was written at Haman’s dictation and later at Mordecai’s, in the king’s name, sealed with the king’s ring, and issued to a hundred and twenty-seven provinces “unto every province according to the writing thereof, and unto every people after their language” — including, explicitly, an edition “to the Jews according to their writing, and according to their language” (1:22; 3:12; 8:8–10; 8:9). It then says Mordecai and Esther wrote again, with authority, to all 127 provinces (9:20, 9:29–30). That is a claim that a specific, dated, multilingual body of correspondence went out through a real bureaucracy and landed in real communities. Either it did or the author manufactured it.

Three. It explains why its readers keep the days they keep. This is the boldest one, because the evidence is the reader himself. The book does not simply narrate a victory; it stops and accounts for a calendar irregularity in its audience’s own practice — why Jews in unwalled towns keep the fourteenth of Adar while Shushan keeps the fifteenth (9:19), why the days are called Purim (9:26), why fasting attaches to the feast (9:31). If those explanations are inventions, the author is telling people a false origin for something they do every year with their own hands.

Four, and smallest, but of a piece with the first. “And the king Ahasuerus laid a tribute upon the land, and upon the isles of the sea” (10:1). The story ended a verse earlier. This is a bare fiscal note about a real king’s administration, sitting exactly where the summary of a reign goes.

A note on what is not on this list, because the distinction matters. The book makes other checkable historical claims — that the empire held 127 provinces (1:1), that a sealed royal writing could not be reversed (8:8) — and both have been contested. But being wrong about an administrative detail is error. Fabricating a citation is a different act.

The seam, and what it costs each reading

Now put the fourth column of that table together with something structural.

Every statement in Esther that explains a custom to people already keeping it — all four of them — sits inside a single fourteen-verse block: 9:19 through 9:32. There are none before it. Not one. Across the preceding eight and a half chapters the narrator explains Persian things constantly — the drinking rule, the twelve-month harem regime, who may not enter the gate in sackcloth, the death law of the inner court, what a sealed ring means — and never once explains a Jewish practice to a Jewish reader.

Then at 9:19 the texture changes and does not change back. Mordecai writes; the Jews undertake; the days are named; Esther confirms; “and it was written in the book.” Then 10:1–3 closes in the annalistic voice of the opening — tribute, acts of power, and the archive citation.

That boundary is a problem for every reading, and it should be stated as one.

For Case Two, the boundary is the whole point: the tail is the purpose and the story in front of it is the justification. That is coherent, and it is the strongest form of the charter argument.

For a literary imitation, the boundary is a broken device. If a Jewish author adopted the persona of a Persian court scribe, he abandoned the persona for fourteen verses at the one place where it mattered most. Gordis, who proposed the imitation, had to concede exactly this: establishing a Jewish holiday, he wrote, “is strictly speaking beyond the purview of a Persian chronicler. Hence, the establishment of Purim is ordained in two appendices.” He saw the seam. He had to call it a technique.

For the reading argued here, it is not a technique and not tampering. It is a document boundary — an imperial record, with the community’s own festival legislation attached to the end of it, and the archival citation last. Which is what a document looks like when a community receives it and adds to it.

And there is a reason that block would exist at all, which the earlier section on the Amalek war already supplied. The appendix is not only festival legislation. By the oldest Jewish reading of it, it is the discharge of a standing written order — Exodus 17:14’s “write this for a memorial in a book,” the one command in the Torah to commit a particular thing to writing, issued at the outbreak of the Amalek war and left open for a thousand years. That is why the block ends the way it does: “and it was written in the book” (9:32), in the same words the order used. A community that believed it was finally executing a Torah command would attach exactly this, in exactly this place, to exactly this document.

One further detail belongs here, and it is the one place where something inside that block looks genuinely inserted. Esther 9:29 reads “Then Esther the queen, the daughter of Abihail, and Mordecai the Jew, wrote with all authority.” Gordis argued that “and Mordecai the Jew” is a later addition, and his first reason is grammatical: the verb is feminine singular — built for Esther alone — against the masculine form used of Mordecai nine verses earlier. He adds that Esther receives a full patronymic here while Mordecai appears with no genealogy at all, and that the “fastings and their cry” of 9:31 recalls Esther’s own fast, which suits a letter issued in her name. That is checkable by anyone with a Hebrew text.

Where this is going

One more piece belongs here, and it is posted as a signpost rather than argued, because it is the direction this series is heading and a reader deserves to know it early.

Go back to the missing furniture — no temple, no priest, no sacrifice, no law, no Jerusalem. Both of the standard cases treat that as something to be accounted for: an import that never got naturalized, or a lapse in a pious author. Both assume the book ought to have looked Mosaic.

Suppose it never was Mosaic, and suppose that is the point. Leviticus 23 lists the appointed feasts of the LORD — Passover, Unleavened Bread, Firstfruits, Weeks, Trumpets, Atonement, Tabernacles. Purim is not on that list, and it never got added to it. It has no sacrifice, no priesthood, no sanctuary, no pilgrimage, no Palestinian setting. That is usually filed as an embarrassment. It is at least as reasonable to read it as a statement: this feast does not belong to the Mosaic order because what it commemorates does not belong to the Mosaic order.

What this site will argue over the coming weeks is that Purim is a type of the war of Gog and Magog — the nations massed against the covenant people and broken in a single day. And the categories in that war are conspicuously not Sinai’s. They are older. Ezekiel’s roll call of the invading coalition is read straight out of the table of the nations in Genesis 10: Gog of the land of Magog, chief prince of Meshech and Tubal, with Gomer and the house of Togarmah (Ezekiel 38:2, 6) — every one of them a son or grandson of Japheth in Genesis 10:2–3 — joined by Cush and Put, sons of Ham in Genesis 10:6. That is a war described in the vocabulary of the generations after the flood, not the vocabulary of the covenant at Sinai. And it lands, at the end of the Bible, under exactly those two names: “Gog and Magog, to gather them together to battle” (Revelation 20:8).

A feast memorializing a conflict framed in pre-Sinai terms has no reason to carry Sinai’s furniture. Which is precisely the book we have.

Attestation, up front, because this vein is not empty. The connection between Esther and the battle of Gog and Magog belongs to James B. Jordan, who has argued since the 1990s that “the book of Esther describes the fulfillment of the battle of Gog and Magog” — on grounds including the wordplay between Haman and hamon (“multitude”), the scale of the slaughter, the 127 provinces against Ezekiel’s list, and the Japhethite-Hamite coalition against the Shemite remnant. Peter Leithart follows him. Anybody working here builds on Jordan and should say so out loud.

Two things are offered as ours rather than his. The first is the direction of the arrow: Jordan’s primary claim is that Ezekiel’s prophecy was fulfilled in Esther; the claim here is that Purim typifies a battle that is still ahead of us, with Esther as the type rather than the antitype. The second is the argument from the calendar — Purim’s exclusion from Leviticus 23 read as a feature rather than a defect, and the book’s missing Mosaic furniture read as the same signal from the other side.

And one problem, stated now rather than buried, because a flaw in a type is usually where the seam is: Ezekiel’s Gog war ends with Israel taking the plunder — “they shall spoil those that spoiled them, and rob those that robbed them” (Ezekiel 39:10) — while Esther stops three separate times to say the Jews took none. Those do not obviously line up. Whether that is a fatal mismatch or the exact seam where type parts from fulfillment is one of the things these pages have to work out, in the open, rather than assume.

One more signpost, and this one costs nothing to check. The law of the inner court is impartial about death: “whosoever, whether man or woman, shall come unto the king into the inner court, who is not called, there is one law of his to put him to death” (4:11) — the queen included, and she has not been called for thirty days. And the decree she means to plead against includes her; Mordecai says so to her face: “Think not with thyself that thou shalt escape in the king’s house, more than all the Jews” (4:13). So the only mediator this book has is a figure standing in both columns of the ledger at once. She goes in as royalty — the Hebrew of 5:1 says she “put on royalty,” and the King James’s own italics at “royal apparel” concede the noun is supplied — and she pleads as one of the condemned: “let my life be given me at my petition, and my people at my request: for we are sold, I and my people” (7:3–4). Royal enough to stand in the throne room; kin enough to say my people; under a death sentence from both directions while she does it — the court’s law if the scepter stays down, Haman’s decree if it doesn’t matter. A mediator, in other words, has to belong fully to the throne and fully to the people being pleaded for, and the one this book supplies is both at once. The New Testament has a name for that shape: “one mediator between God and men, the man Christ Jesus” (1 Timothy 2:5). And what Esther did at that throne — approached uncalled, under law of death, and “obtained favour” (5:2) — is what the church is told it may now do at another one: “Let us therefore come boldly unto the throne of grace, that we may obtain mercy, and find grace” (Hebrews 4:16). This page only flags it; a later page will press it.

Finally, this joins a pattern that keeps turning up in this project and will get its own treatment: the latecomer. Benjamin, the last-born son. Esther, the last book argued into the canon. Purim, the feast added after all the others were fixed. Esther herself, raised to a throne that a queen commissioned in the ordinary way had already held and lost. Paul, of Benjamin, calling himself one “born out of due time.” None of that is argued on this page. It is flagged, so you know what the next pages are for.

What this reading buys.

  • The missing God is genre, not embarrassment. A Persian court record does not invoke the God of the Jews. There is nothing to apologize for and no theology of hiddenness required to explain the silence — though the silence, once it is there, does other work. (The Greek Additions later supply what the Hebrew lacks, precisely because later readers felt the lack.)
  • The outsider’s-eye vocabulary is simply literal. “Mordecai the Jew,” the Jews as “they,” the king never without his title.
  • The seam is a seam.
  • And it explains why this book, of all books, survived in Jewish hands. The Jews kept the copy because the counter-decree that saved their lives was in it, and because the legal basis of Purim rested on it. You keep the paper that says you are allowed to be alive.

What this reading costs. Stated plainly, because a case that only lists its strengths isn’t worth reading.

  1. The private scenes. A chancery clerk does not file Haman’s evening at home with Zeresh and his friends (5:14), or the moment his wife tells him he is finished (6:13). The best answer available is that the record was compiled under Mordecai’s authority — the book says the scribes wrote what Mordecai commanded — and that a fallen vizier’s household was not a place where secrets kept. But that is an inference, and the text does not make it. This is the strongest objection to the reading and it should be left standing.
  2. The sympathies are Jewish. So they would be, in a record compiled under a Jewish vizier. Worth noticing, though, that the sympathy never once becomes religious: it is political and legal from beginning to end, which is what a chancery document under Mordecai would look like and not at all what a Jewish devotional author would produce.
  3. The comedy is real. Case Two’s farce reading is not manufactured; the book has genuinely funny proportions. A court record composed to flatter a king can be pompous, and pomposity reads as comedy at a distance of centuries — but this is a partial answer, not a full one.
  4. Nobody can produce the original. Every argument here is from the internal behavior of the text. That is a hypothesis about a document’s origin, not a discovery of the document.

Where the credit goes. The observation that Esther behaves like a Persian court chronicle is Gordis’s, published in 1981, and anyone building on it should say so. Morteza Arabzadeh Sarbanani argued in 2023 that the book’s Old Persian and Aramaic vocabulary with no Greek vocabulary indicates an author working from Persian sources rather than Greek ones. The step taken here — dropping ostensibly, and asking whether the document the text behaves like is the document it is — is ours, and it is offered as a hypothesis with its costs listed above.

What all three readings agree on

It is worth ending on the common ground, because it is larger than the dispute.

All three agree the author knew Persia from the inside. The historicist, the literary critic, and the archivist all concede the loanwords, the court procedure, the palace, the chancery formulae. Nobody thinks this book was written by someone guessing.

All three agree the surface of the book is deliberately godless. They disagree about why — providence hidden behind events, comic irony, or the conventions of an imperial record. But nobody thinks the author simply forgot.

All three agree the book is about a real festival. Purim was being kept by the second century BC and has been kept ever since. Something happened, or something was believed to have happened, early enough to fix a date in Adar and keep it there for over two thousand years.

And a last word on why this page does not try to settle it. The reading of Esther that this site is mainly interested in — Benjamin, Agag, the neighbour better than she, the spoil untouched — does not depend on which case wins. Those are features of the text as it stands. They are there whether the text is a court record, a chronicle imitation, or a festival scroll, and they are checkable by anyone who owns a Bible and is willing to turn to 1 Samuel 15 with a finger in Esther 1.

Which is the whole method here anyway. Nothing above asks to be believed on anyone’s authority. Bring a Bible; check everything.

Sources

Primary texts. All scripture quoted from the King James Version. Jerusalem Talmud, Megillah 1:5, in the Guggenheimer translation via Sefaria. Babylonian Talmud Megillah 7a (canonical status) and Bava Batra 15a (authorship). Herodotus, Histories 3.84 (the seven families), 7.8 (the war council), 8.85 (the orosangai register), 8.98 (the courier relay), 9.108–113 (the return from Greece).

On the Persian-chronicle genre. Robert Gordis, “Religion, Wisdom and History in the Book of Esther — A New Solution to an Ancient Crux,” Journal of Biblical Literature 100 (1981), 359–388 — full text; the section “Esther As a Persian Chronicle” begins at p. 375. Also Gordis, “Studies in the Esther Narrative,” JBL 95 (1976), 43–58 — full text.

On Persian sources behind the book. Morteza Arabzadeh Sarbanani, “The Book of Esther as a Source for Achaemenian History,” ANE Today (ASOR), June 2023.

On Esther as the discharge of Exodus 17:14. The derivation is tannaitic: Mekhilta de-Rabbi Ishmael, Tractate Amalek 2, at the lemma of Exodus 17:14, where “in the book” is assigned to the Megillah. Carried in the Babylonian Talmud at Megillah 7a — R. Elazar HaModa’i for it, R. Yehoshua against — with Esther’s own archival reply, “I am already written in the chronicles of the kings of Media and Persia,” at Megillah 7a:3, and the derivation itself at 7a:5. The verbal analogy between Esther 9:28 and Exodus 17:14 is formalized at Megillah 18a. Mekhilta de-Rabbi Shimon bar Yochai on Exodus 17:14 preserves the same pairing. Note for anyone checking the Mekhilta in English: Silverstein’s rendering truncates the Esther 9:28 proof-text, dropping the very clause — “and their remembrance shall not perish from their seed” — that the Hebrew quotes.

On David’s sparing of Shimei and Mordecai’s descent. Babylonian Talmud Megillah 12b — the tribes compete for credit for Mordecai, Judah’s claim resting on David’s sparing of Shimei son of Gera — with Rava’s reversed-charge version of the same wire at Megillah 13a. Quotations follow the Steinsaltz translation via Sefaria.

On Esther and the battle of Gog and Magog. James B. Jordan, “The Battle of Gog and Magog,” Biblical Horizons No. 2, and his “Esther: Historical and Chronological Comments” (Biblical Horizons, 1996; republished by Theopolis). Peter Leithart, “Esther and Jerusalem” (2004), which explicitly follows Jordan’s reading of Ezekiel 38–39. Note that Jordan and Leithart are one stream, not two independent witnesses.

On the comic genre. Adele Berlin, The JPS Bible Commentary: Esther (Jewish Publication Society, 2001); see also her “The Book of Esther as Comedy”.

On the documents. The Passover Papyrus (Elephantine, 419/418 BC): TheTorah.com and the Center for Online Judaic Studies. The Aršāma letters on leather: Christopher J. Tuplin and John Ma, eds., Aršāma and his World: The Bodleian Letters in Context (Oxford, 2020) — Bodleian commentary volume.

On the archaeology and the Persian setting. Encyclopaedia Iranica, “Esther, Book of”; Encyclopaedia Iranica, “Courts and Courtiers i” on the orosangai. Jean Perrot on the Susa excavations, quoted in “Top Ten Discoveries Related to the Book of Esther,” Bible Archaeology Report, which also covers the Marduka texts.

On the Greek versions and the colophon. Additions to Esther, Early Jewish Writings; on the dating of the colophon, E. Bickerman’s 78–77 BC against the alternative 114–113 BC.

On the absence at Qumran. Michael Kruger, “The Dead Sea Scrolls, the Book of Esther, and the Argument from Silence”.