Ma'munid Khwarazmshahs

The delta, and the line they displaced

formationDrafted from (ed.), The Cambridge History of Iran, Volume 4: From the Arab Invasion to the Saljuqs (1975)

Khwarazm is a delta. The Oxus, having run fifteen hundred miles from the Pamirs, spreads out before it reaches the Aral Sea into a fan of irrigated land surrounded on every side by desert — a green island a couple of hundred kilometres across with nothing around it for a week's travel in any direction.

That geography produced two things everything else here depends on. It made the province rich for its size, because irrigated alluvium is the most productive land there is and Khwarazm had been farming it with a maintained canal system since antiquity. And it made the province separate: distinct enough to have kept its own Iranian language, its own script, its own calendar and its own dynasty through the Achaemenids, the Sasanians and three centuries of Islam.

That dynasty was the Afrighids, and they had held the title Khwarazmshah since about the fourth century. The Ma'munids displaced them in 995 — a rival line from Gurganj on the western side of the delta, which took the eastern capital at Kath and with it the title.

Konye-Urgench, Turkmenistan · (ed.), The Cambridge History of Iran, Volume 4: From the Arab Invasion to the Saljuqs (1975) · Photograph: Tim Williams · CC BY-SA 3.0

Two things follow.

The first is the title. Khwarazmshah is not an Islamic office. It predates Islam by six hundred years, it was not in anybody's gift, and the Ma'munids took it because holding it was what made a ruler of the delta legitimate to the people of the delta. Seventy years later the same title would be handed out by a Seljuk sultan to a household slave, which is what had become of it.

The second is what the displacement destroyed. The last Afrighid was overthrown in 995, and al-Biruni — who was born in the old capital and was about twenty at the time — wrote afterwards that the conqueror had the Khwarazmian written records destroyed and the scholars who knew the old script killed or dispersed. He says this in a book about calendars, in the course of explaining why he cannot reconstruct the Khwarazmian one properly. Whether the destruction was as complete as he says is not checkable. What is on the record is that the man best placed to know said the evidence had been deliberately removed, and that he was describing the founding of the dynasty that then employed him.

The best court in the east, for about fifteen years

peakDrafted from (ed.), The Cambridge History of Iran, Volume 4: From the Arab Invasion to the Saljuqs (1975)

When Mahmud of Ghazna annexed Khwarazm in 1017 he took its scholars back to Ghazna along with everything else that could be moved. This is the court he took them from.

The Ma'munids displaced an older line of Khwarazmshahs at Gurganj in 995 and lasted twenty-two years. Khwarazm is the delta where the Oxus reaches the Aral Sea — irrigated, dense, rich for its size, physically separated from everywhere else by desert, and sitting on the route by which furs, amber and slaves came south out of the northern forests. It was a small state that could afford things.

What it bought was scholars. At Gurganj under Ma'mun II there were assembled, at the same time: al-Biruni, then in his thirties, already an astronomer of the first rank; Ibn Sina, the physician and philosopher whose Canon was taught in European medical faculties for five centuries; Abu Nasr Mansur, al-Biruni's teacher and a mathematician who did foundational work on spherical trigonometry; Abu Sahl al-Masihi, a Christian physician; Ibn al-Khammar, another; and al-Tha'alibi, the anthologist to whom a great deal of what survives of tenth-century Arabic poetry is owed.

Later tradition calls it an academy, which overstates the institution and understates the concentration. There is no evidence of a formal foundation with statutes and endowments. There is evidence of a ruler who paid well, a small capital where these men were in each other's company, and correspondence between them that survives — including an exchange of questions and answers between the young Ibn Sina and the young al-Biruni, in which they disagree sharply about Aristotle and neither gives ground.

Bukhara, Uzbekistan · (ed.), The Cambridge History of Iran, Volume 4: From the Arab Invasion to the Saljuqs (1975) · Photograph: Davide Mauro · CC BY-SA 4.0

Fifteen years, in a delta town, and two of the half-dozen most consequential minds of the medieval world were arguing with each other in it.

The calendar he could not reconstruct

aside?What aside meansDrafted from (ed.), The Cambridge History of Iran, Volume 4: From the Arab Invasion to the Saljuqs (1975)

Around the year 1000, in exile at the Ziyarid court at Gurgan on the Caspian, al-Biruni finished a book called al-Athar al-Baqiya, the remaining traces of past centuries. It is a comparative study of calendars and eras: how the Persians, Greeks, Romans, Jews, Christians, Sabians, Zoroastrians, Arabs before Islam, and the people of his own homeland reckoned time, with the festivals attached to each and the intercalation rules that made them work.

It is one of the most original books of the Middle Ages. Nobody had written comparative chronology before, and doing it required a method he had to invent — establishing what each system's epoch was, converting between them, and treating each community's account of its own past as evidence to be weighed rather than accepted or dismissed.

The part that belongs here is Khwarazm. Al-Biruni was a Khwarazmian, writing about the Khwarazmian calendar, in the decade after his homeland's old dynasty had been overthrown — and he keeps having to say that he cannot recover things. He gives the month names and the festivals and the era, and he explains that the written material was destroyed and the men who could read the old script were gone, so parts of it are irrecoverable.

That is a scholar marking the limits of his own evidence, a thousand years ago, about his own country. He does not fill the gap with a plausible reconstruction. He says what is missing and why, and stops.

Bibliothèque nationale de France · (ed.), The Cambridge History of Iran, Volume 4: From the Arab Invasion to the Saljuqs (1975) · Unknown artist · Public domain

Two things it leaves. The first is that most of what is known about pre-Islamic Khwarazm — the calendar, the era, the script's existence — is known because one man who happened to be from there wrote it down while it was going, and the archaeology of the twentieth century has largely confirmed him.

The second is how rare that is. Archives get burned and replaced by travellers' tales; conquests get written up by men on the conqueror's payroll; traditions get compiled a century after the events by people with a case to make. What almost never happens is this: the record is destroyed, and somebody who knew what had been in it writes down what he can and marks the rest as lost.

A tongue of gold and silver

contractionDrafted from Barthold, Turkestan Down to the Mongol Invasion (1928)

The three years between the demand and the murder are recorded in unusual detail, because Bayhaqi copied them out of al-Biruni's history of Khwarazm, which is otherwise lost. The man who wrote down what could not be recovered about the Khwarazmian calendar also wrote down, from inside the palace, how the delta was talked out of its independence.

The demand was made deniably. In 1014 Mahmud's vizier, Ahmad ibn Hasan Maymandi, remarked to the Khwarazmshah's envoy — as though on his own initiative, and adding that the sultan knew nothing about it — that it would be well if the Friday sermon in Khwarazm named the sultan. Ma'mun understood perfectly that a vizier does not say such things on his own initiative, and hesitated anyway. The vizier repeated it, less politely.

Edinburgh University Library · Barthold, Turkestan Down to the Mongol Invasion (1928) · Unknown artist · Public domain

So Ma'mun called in his commanders and the leading men of the country, put the request in front of them, and said he intended to agree, because otherwise he and the country would be destroyed. They refused absolutely: they walked out of the palace, unfurled the standards, drew their swords and stood in the open cursing. To calm them the Khwarazmshah had to announce that no such request had ever been made and that he had been testing their loyalty. It then fell to al-Biruni — Barthold quotes the phrase, "with a tongue of gold and silver" — to talk the nobles into apologising to their sovereign for their insolence.

The same adviser's next suggestion was to find allies. Two Karakhanid princes were fighting each other near Uzgand; Ma'mun offered to mediate, made peace between them in 1016, and concluded an alliance. Mahmud sent an embassy from Balkh to say he was displeased. The Karakhanids answered blandly that they regarded the Khwarazmshah as the sultan's friend and brother-in-law, that their treaty with him was merely a supplement to their treaty with the sultan, and that if there were any misunderstanding they would be glad to mediate that too. No answer came. They told Ma'mun about the embassy; he proposed that they and he send detachments into Khurasan from different directions to skirmish, on the understanding that civilians were not to be touched and the whole thing was to be understood as a means of restoring peace. They declined to provide troops and repeated the offer to mediate, which he accepted. In the winter of 1016 Mahmud received their envoys at Balkh and sent them home with the assurance that there was no serious disagreement between himself and the Khwarazmshah, and that what little there had been was now entirely removed.

The letter arrived immediately afterwards.

It is worth having at length, because it is the clearest statement anyone made in that century of what a sultan thought a client ruler was. In the matter of the sermon, it says, the Khwarazmshah submitted to our will, knowing how it might otherwise turn out for him, but his people did not leave him a free hand. I do not use the words guards and subjects, because men who are in a position to tell a king do this and do not do that cannot be called guards or subjects. This is plain weakness and an incapacity to rule; and such it was. Angered by these people, I have stayed a long time here at Balkh and have collected a hundred thousand horse and foot and five hundred elephants, to punish the traitors who set themselves against their sovereign's will and put them back on the right road. At the same time we shall rouse the amir, our brother and brother-in-law, and show him how a kingdom is ruled; a weak amir is not fit for the work.

Three terms followed, and although the letter offered them as alternatives, the Khwarazmshah read them — correctly — as a list. He was to introduce the sermon in the sultan's name, willingly and in full obedience. He was to send money and presents worthy of a sultan, which would afterwards be secretly returned, since the sultan had no need of surplus money and his lands were already tottering under the weight of gold and silver. And he was to send his nobles, imams and jurists to petition for mercy.

Ma'mun complied with all of it. The sermon was changed first at Nasa and Farawa in his Khurasanian territory, then in the other towns, and finally everywhere except the two capitals, Kath and Gurganj. He sent eighty thousand dinars, three thousand horses, and a party of shaykhs, qadis and notables.

His army was at Hazarasp at the time, under the chief chamberlain, watching the hundred thousand men at Balkh. It did not march against them.

One sentence in the Friday sermon

endDrafted from Bosworth, The Ghaznavids: Their Empire in Afghanistan and Eastern Iran, 994-1040 (1963)

What Mahmud had asked for was a sentence. No tribute, no garrison, no territory, no change of administration. The khutba is where a polity states publicly whose authority it holds under. Tahir had omitted the caliph's name at Merv in 822; the Saffarids and the Samanids spent two decades competing over whose name Baghdad would authorise. It is the cheapest possible demand and the most total one, and it was conceded.

The army decided that the concession was the offence. On Wednesday 20 March 1017, at Hazarasp, the Khwarazmian forces turned against their own sovereign under the chief chamberlain Alptigin of Bukhara — no relation to the Alptigin who had walked off with Ghazna fifty-five years earlier, though the repetition of the name is worth a moment. The vizier and several of Ma'mun's adherents were killed and the rest ran. The Khwarazmshah shut himself in his castle; the rebels set fire to it and killed him while it burned. They put his seventeen-year-old nephew on the throne and ran the country themselves, plundering the rich and settling private scores, with a vizier of their own choosing.

Mahmud then had what he needed, which was not a khutba but a pretext: the murder of his sister's husband. The Ghaznavid army came north, beat the Khwarazmian forces, took Gurganj, enslaved a large part of the population, executed the commanders, and annexed the delta. The nephew lasted a few months.

And the scholars were taken to Ghazna. Al-Biruni went, and spent the rest of his life in Ghaznavid service, and wrote the book about India there. Abu Sahl al-Masihi died on the road. Ibn Sina refused, left before the army arrived, and spent the next two decades moving between the courts of western Iran — Rayy, Hamadan, and finally Isfahan under the Kakuyids — writing the Canon and the Book of Healing while working as a vizier and, more than once, in hiding.

Hamadan, Iran · Bosworth, The Ghaznavids: Their Empire in Afghanistan and Eastern Iran, 994-1040 (1963) · Photograph: Nick Taylor · CC BY 2.0

That is what an annexation looks like at the scale of individuals. The campaigns and the square miles are the part that gets counted; the part that mattered most about Khwarazm in 1017 was six or seven men and where each of them went next.

The portrait

afterlifeDrafted from Browne, A Literary History of Persia, Volume II: From Firdawsi to Sa'di (1906)

The dynasty lasted twenty-two years and its afterlife is a scattering of men. The best account of how they scattered is in the Chahar Maqala, written a century and a half later, and Browne translates it.

The letter

Shortly before the annexation, Mahmud of Ghazna sent Ma'mun II a letter by the hand of one of his nobles. Browne gives its text, and its politeness is the whole of its menace:

I have heard that there are in attendance on the Khwarazmshah several men of learning, each unrivalled in his science, such as So-and-so and So-and-so. You must send them to my Court, so that they may have the honour of being presented thereat. We rely on being enabled to profit by their knowledge and skill, and request this favour on the part of the Prince of Khwarazm.

It was a command and Ma'mun understood it as one. He called the men in and put it to them straight: the Sultan is strong, he has a large army out of Khurasan and India, and he wants this country. I cannot refuse to obey his order.

What say ye on this matter?

Who went and who ran

Three of them — al-Biruni, Ibn al-Khammar and one other — had heard about Mahmud's generosity and were willing to go.

Ibn Sina and Abu Sahl al-Masihi were not, and with Ma'mun's connivance they got out privately. A dust-storm caught them in the desert and al-Masihi died in it. Ibn Sina, after what Browne calls terrible hardships, reached Abiward, and made his way to Tus, then Nishapur, and finally to Gurgan on the Caspian, where Qabus b. Vushmgir was ruling.

Now the detail that makes the episode. Of all the scholars he had demanded, the one Mahmud actually wanted was Ibn Sina; so on learning that he had escaped, the sultan had a portrait of him copied and circulated through the lands.

At Gurgan, Ibn Sina restored a favourite kinsman of Qabus to health and was summoned before the prince — who recognised him immediately from the portrait, and did not hand him over. He kept him honourably in his service instead, until the philosopher moved on to Rayy and eventually became a minister at Isfahan.

What happened to the ones who went

Al-Biruni spent the rest of his life at Ghazna and wrote the great book about India there, so the arrangement was not a disaster. It was not comfortable either.

The Chahar Maqala has Mahmud, sitting in a four-doored summer-house in the Garden of a Thousand Trees, asking al-Biruni to predict by the stars which door he would leave by. Al-Biruni wrote his answer and put it under a quilt. The sultan had a hole knocked in a wall and went out through that instead, then called for the paper. It read: the King will go out by none of these four doors, but an opening will be made in the eastern wall by which he will leave the building.

Mahmud, who had wanted to make him look foolish, had him thrown off the roof. A mosquito-curtain broke the fall. Asked whether he had foreseen that, al-Biruni produced a notebook in which he had written, under the day's date, that he would be cast down from a high place and reach the ground unhurt. He was imprisoned in the citadel for six months.

When the prime minister eventually interceded — pointing out that the man had made two exact predictions and been given fetters for them — Mahmud's answer is the one worth keeping:

Know, my lord, that this man is said to have no equal in the world save Avicenna; but both his predictions were opposed to my will, and Kings are like little children: in order to receive rewards from them, one should speak in accordance with their opinion.

He then ordered a horse with gold harness, a robe, a satin turban, a thousand dinars, a slave and a handmaiden.

The measure of a small court

Set that beside what Gurganj had been. A prince who paid these men, let them argue with each other, and connived at the escape of the two who would not go.

He was killed by his own commanders for temporising over one sentence in the Friday prayer, and the men he had assembled were divided between a sultan who threw one of them off a roof and a road across the desert that killed another. Twenty-two years, and the only thing the delta could not keep was the reason anybody remembers it.