Ziyarid Dynasty

The man who wanted the crown back

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

Mardavij ibn Ziyar came out of Gilan, the Caspian country next to the Daylam that produced the Buyids, and in about four years he took Isfahan, Hamadan, Rayy and the Jibal, and stood a march away from Baghdad. What he intended to do there is the reason anybody remembers him.

He was not proposing to take the caliphate, or to hold it as a Shi'i amir with a caliph in his keeping, which is what his own officers would settle for a decade later. He proposed to abolish it. The sources report a golden throne made for him, a jewelled crown in the Sasanian manner, a plan to be crowned at Ctesiphon in the ruined palace of the Sasanian kings, and the public celebration of Sadeh, the old Iranian midwinter fire festival, at Isfahan. The programme was the restoration of the Persian monarchy, three centuries after it ended, by a Caspian highlander with an army.

It did not survive contact with his own troops. In 935, at Isfahan, his Turkish slave soldiers killed him in the bathhouse. The reported provocation is small — a slight, a withheld reward — and the structural cause is the commonest one in the military history of the region: a purchased guard with no stake in its master's project, at the point where the project stops paying.

What followed matters more than what he intended. His brother Vushmgir kept the Caspian provinces and lost the rest. His officers took the rest — and among those officers were Ali, Hasan and Ahmad, the sons of Buya, who had come down from the same mountains as mercenary captains and were now employers. Within ten years they had Baghdad.

So the most ambitious political idea in tenth-century Iran ended in a bathhouse and was inherited by three men who did not share it, and who took the city Mardavij had been marching on and left the caliph exactly where he was.

A mill in a village

Drafted from al-Rudhrawari, The Eclipse of the 'Abbasid Caliphate: Original Chronicles of the Fourth Islamic Century (1921)expansion

Mardavij began as somebody else's captain. The man he served was Asfar ibn Shirawayh, who had taken Rayy and then Qazwin, and who governed Qazwin by imposing an enormous levy on it, handing his Daylamites authority over the lives and property of the inhabitants, stripping the townspeople of any right they could appeal to, and torturing the local officials on a scale that Miskawayh says astonished Asfar himself. The people of Qazwin, men, women and children together, walked out to the oratory and spent an entire day praying to be delivered from him. When this was reported to Asfar he thought it funny.

The next day Mardavij revolted, attacked him and put him to flight. Asfar rode hard for a day and was not caught. On the second day he was hungry, and stopped at a mill in a village to ask for food; the miller gave him bread and milk. Mardavij, following, found that the hoofprints stopped, looked around, took hold of a labourer and asked about Asfar. The man said he did not know Asfar but had seen a rider go into the mill. Mardavij broke in, found Asfar eating, and cut off his head. Then he went back to Qazwin, calmed the town, redressed the grievances, stopped the extortion, made a number of handsome promises and left. Miskawayh's comment is one sentence: so their prayer was answered.

What he did next is the part that matters, because he did to Isfahan exactly what Asfar had done to Qazwin. He took Rayy and then Isfahan, and at Isfahan he seized property, had women taken, and governed as a tyrant. He sat on a throne of gold with a second throne of silver below it for whoever he chose to honour that day, and on audience days he had the army drawn up in ranks at a distance from his person. He told people that he was Solomon the son of David and his soldiers were the demons.

That is the same programme the first chapter describes in its dignified form — the crown, the Sasanian ceremonial, the coronation planned at Ctesiphon — seen from inside the province paying for it. The two accounts are not in conflict. A restoration of the Persian monarchy, attempted in four years by a man with no revenue base and an army to keep, had to be financed out of the towns he passed through, and the ceremonial that announced it was the same ceremonial that made his own soldiers stand at a distance and take orders from a man claiming to command demons.

Miskawayh adds the detail that finishes it. He humiliated the Turks, who became disaffected and started looking for their chance.

Vushmgir, between two powers

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

Mardavij's brother inherited the wreck of a project and held the Caspian for thirty years, which is the part of this dynasty's history that explains why there was a dynasty at all.

Vushmgir — the name means "quail-catcher", which is what he had been doing when the news of his brother's murder reached him — could not hold what Mardavij had taken. Isfahan, Hamadan and the Jibal went to the officers who had killed or outlived his brother, and among them were the sons of Buya. What he kept was Tabaristan and Gurgan: the Caspian provinces the family had come from.

Then he spent three decades being squeezed. The Buyids came at him from the south and west, taking Rayy and pressing into the mountains; the Samanids, the great power of the north-east, came at him from the other side — and the useful thing about his career is that he survived by using each against the other. When the Buyids took his territory he acknowledged the Samanids and campaigned back with their troops. When the Samanids were distracted he governed on his own account. He lost Gurgan and recovered it more than once.

That is the posture the Ziyarids kept for the next hundred and fifty years, and it is harder than it sounds. The Saffarids took a province and told the caliph afterwards, and were destroyed. The Ma'munids of Khwarazm were asked for one sentence in the Friday sermon, and their own commanders killed their ruler rather than let him concede it. Vushmgir conceded whatever was asked, to whoever was strongest that decade, and kept the mountains.

He died in 967, reportedly in a hunting accident, while preparing to campaign alongside a Samanid army. The dynasty he left held the Caspian for another hundred and twenty years without ever again threatening anybody outside it, which was the correct assessment of what it could do.

A client among clients

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

After 935 the Ziyarids held Tabaristan and Gurgan — the strip of dense, rain-fed forest and rice country between the Elburz mountains and the Caspian, which is unlike anywhere else in Iran and hard to enter from the plateau. They held it for another hundred and fifty years, and for nearly all of that they held it from somebody.

The list of overlords is a summary of the age. Vushmgir manoeuvred between the Samanids at Bukhara and the Buyids to his south, taking Samanid support against his brother's former officers. His successors acknowledged the Buyids. When Ghaznavid power reached the Caspian the Ziyarids acknowledged Mahmud, and later Mas'ud. When the Seljuks came they acknowledged the Seljuks. The dynasty's method was to be too troublesome to occupy and too small to be worth the trouble.

Qabus ibn Vushmgir is the one the sources are interested in, and his career is the method's limit. He took the throne in 977, backed the wrong side against the Buyids within four years, and spent seventeen years in exile in Khurasan. He came back in 998 when Buyid power was failing, and ruled for fourteen more.

He was famous for his Arabic prose style — his letters were collected and studied — and for his patronage. Al-Biruni, who would later write the great book about India, spent years at Gurgan under Qabus, dedicated his Chronology of Ancient Nations to him around the year 1000, and observed lunar eclipses from the city. He was also, on the consistent report of the same sources that admired him, frightening. In 1012 his own commanders deposed him and he did not long survive it.

That is the second time the Ziyarid army disposed of a Ziyarid, by the same mechanism that had killed the founder in a bathhouse seventy-seven years earlier. What is unusual is not the mechanism but the recovery: after 1012 the dynasty simply went on, smaller, in the mountains, for another eighty years.

A tower and a book

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

Two objects survive this dynasty and they are both unusually good.

The first is standing. In 1006 Qabus had a tower built on a mound outside his capital: a shaft of unornamented fired brick, about fifty metres tall, ten flanges running its full height like a star in plan, with a conical cap and no decoration at all except a band of Kufic inscription. It is the earliest dated tomb tower in Iran and it is one of the great buildings of the Islamic world by almost universal agreement, and what it is remarkable for is severity. There is no tile, no colour, no carving. There is a geometry, executed in brick, at a scale that carries across empty country.

The inscription is the part that belongs in a history rather than in an architectural catalogue. It gives the date twice: once by the lunar hijri calendar, and once by the Yazdgirdi era — the solar reckoning counted from the accession of the last Sasanian king. Seventy years after Mardavij proposed to be crowned at Ctesiphon, his grand-nephew dated his own tomb by the Persian kings as well as by the Prophet's emigration, on a monument built to be seen from a long way off.

The second is a book. Around 1082, Kay Ka'us ibn Iskandar — Qabus's great-grandson, ruling almost nothing — wrote the Qabus-nama for his son Gilanshah. It is one of the earliest substantial works of Persian prose and one of the most likeable books of its century. Formally it is a mirror for princes, the genre the Karakhanids and the Seljuks were both commissioning at the same time, and it is nothing like either of theirs. Alongside the chapters on kingship and justice there are chapters on how to buy a horse and how to buy a slave, on playing chess, on the etiquette of the bath, on how much to drink, on being a guest, on writing poetry without being tedious, and on what to do if you must take up a trade. He tells stories against himself. The advice assumes the reader will not be a great king.

He was right. Gilanshah is the last Ziyarid, and the dynasty ended in about 1090 when the Nizari Isma'ilis took its territory — the same Daylam and Elburz country the Ziyarids and the Buyids had both come out of, changing hands once more to somebody with a new idea.

A dynasty that opened by proposing to restore the Persian empire closed with a retired prince explaining to his son how to behave at dinner. Both documents are Persian, and the second one is the one people still read.

The signet ring

Drafted from Isfandiyar, An Abridged Translation of the History of Tabaristan (1905)contraction

Qabus was at Janashk, taking the air, when his own nobles and officers decided to be rid of him. He knew nothing about it until the night they surrounded his camp, attacked it and plundered his baggage; his retainers drove them off. They went instead to Gurgan, took the city, and sent to Tabaristan for his son Minuchihr, whom they told that unless he deposed his father and accepted the crown from their hands they would find someone else and be done with the family altogether. Minuchihr could not oppose them and agreed.

Qabus withdrew to Bistam with his household to see what would happen, and his son was made to march against him. Ibn Isfandiyar's account of the meeting is the best thing in the Ziyarid record. Minuchihr came into his father's presence with every mark of respect and in tears, said how the events had forced him into the appearance of opposing him, and offered — if Qabus wanted it — to declare openly for him instead. Qabus, who could see his own position clearly, gave his son his blessing, handed over his signet ring and the keys of his treasuries, and went back to Janashk to spend what was left of his life in prayer. The men who had deposed him could not feel safe while he was alive, and before long they found an opportunity to kill him quietly. He was buried under a dome outside Gurgan, on the road to Khurasan. Minuchihr mourned three days in the Daylamite fashion and took the throne, and Ibn Isfandiyar notes that his father was soon forgotten.

What the new ruler then had to do is the measure of how far the family had come down. The caliph al-Qadir sent condolences and a title, Falak al-Ma'ali. That was the easy part. The serious business was Mahmud of Ghazna, and Minuchihr bought him with presents and professions of loyalty until Mahmud recognised him in Gurgan, Tabaristan and Qumis and sent an ambassador with a robe of honour. In exchange Minuchihr paid fifty thousand dinars a year and supplied a thousand picked Daylamite soldiers for the campaign against Nardin. Then he sent the greatest noble in Gurgan to Ghazna to ask for one of Mahmud's daughters, and sent him a second time with the qadi of Gurgan to draw up the contract and bring the bride home.

Only after all of that did he dare avenge his father. He hunted down the men who had killed Qabus and put nearly all of them to death; one, the son of Khargash, got away into exile and was never heard of again. The commander of the old king's army evaded capture for a long time and finally took refuge with the sultan at Nishapur, who handed him over.

A ruler who has to marry into a foreign dynasty, pay it an annual tribute and lend it his infantry before he can execute his own father's murderers is not manoeuvring between two powers any more. He is a subject of one of them. That is the arrangement the Ziyarids lived under for the rest of their existence.

What the Caspian kept

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

The Ziyarids end around 1090, when the Nizari Isma'ilis took their territory in the Elburz and Tabaristan. The last of them, Gilanshah, is the son the Qabus-nama was written for.

It is worth ending on the region rather than on the family, because the Caspian is a system and the dynasties in it are variations.

Four houses come out of it or hold it. The Ziyarids and the Buyids are both Daylamite houses from the same mountains, one of which stayed and one of which took Baghdad. The Bavands held the Elburz through all of it and outlasted everyone. The Nizaris arrived at the end and took the ground from whoever still had it. Between them they cover 651 to 1273, and the pattern is consistent in a way the plateau's is not.

The pattern is this. The Caspian provinces are separated from the Iranian plateau by a mountain wall and from everything else by sea, forest and rain. They were the last part of Iran to become Muslim — two centuries after the plains — and they went on producing infantry when the rest of the region had gone over to cavalry, which is why Daylamite soldiers turn up in the pay of half the powers of the Islamic world for a century. Nobody on the plateau could hold the mountains, and nobody in the mountains could hold the plateau for long: Mardavij tried and was killed, and the Buyids succeeded and stopped being a Caspian power the moment they did.

So the Caspian dynasties survive by paying and outlasting, and the durable ones are the ones that understood that early. The Bavands, who were ended three times and came back twice, are the extreme case. The Ziyarids, after Mardavij, are the ordinary one.

Saying that the Nizaris ended the Ziyarids is accurate and slightly misleading. What ended in 1090 was a family holding a province. The province went on doing exactly what it had been doing — resisting the plateau, producing infantry, sheltering whoever could not currently be dislodged — for another two hundred years, under different names, on the same terms.

The book that was quarried

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

The dynasty produced two things worth having, and only one of them travelled. The tower stayed where it was. The book went into general circulation, and what happened to it there is the Ziyarid afterlife.

What Qabus was

Take the measure of the man first, because it makes the ending funnier and sadder.

Qabus b. Vushmgir, styled Shams al-Ma'ali, came from a house that traced itself — by al-Biruni's reckoning, and al-Biruni lived at his court — to the Sasanian king Kavadh, through the Qarinvands, one of the seven great families of Sasanian Persia. Ibn Khallikan says that the number of poets who came to him and praised him in splendid qasidas exceeded that of any other court. He wrote Arabic well enough for his letters to be collected as a model of style, and a treatise on the astrolabe that a competent judge approved; he corresponded with the Sahib Ibn Abbad; Ibn Isfandiyar praises his eloquence, his horsemanship, and his knowledge of philosophy, astronomy and astrology.

And then, Browne says, having listed all that: unfortunately, with all these gifts of mind, birth and character, he was stern, harsh, suspicious and at times bloodthirsty. He executed a chamberlain on suspicion of embezzlement, and it was that execution which drove his nobles into revolt, deposed him, and had him put to death.

The book

Seventy years later his great-grandson Kay Ka'us, ruling almost nothing, wrote the Qabus-nama for his son — and put the account of Qabus's deposition and murder in it, along with everything else he thought a young man ought to know.

It is a strange and generous book. Browne, who catalogues its anecdotes one by one, notes that it holds stories about a rich pilgrim's brutal answer to a poor one, about how a caliph's favourite slave was saved from drowning, about Plato's distress at being praised by a fool, and about the physician al-Razi being alarmed because a madman smiled at him. It is a mirror for princes written by a man with no principality, for a son who was going to lose what was left.

And then everybody stole from it

Here is what became of it.

Dawlatshah, compiling his celebrated biographies of the poets four centuries later, mentions the Qabus-nama exactly once. Browne demonstrates that he is in fact indebted to it throughout, and gives the instances: Dawlatshah's account of the deposition and murder of Qabus b. Vushmgir is taken from it; so is his story of how the Sayyida, mother of Majd al-Dawla, talked Mahmud of Ghazna out of attacking Rayy.

Some of the borrowing goes further than borrowing. The Qabus-nama has a rule that nobody might strike a man the king had already struck, out of respect for the royal chastisement, and attributes it to the Greeks; Dawlatshah attributes the same practice to the court of Mahmud of Ghazna. The book has the celebrated story of a threatening letter answered with the three letters A.L.M.; in the Qabus-nama the answer comes from the caliph and the riddle is solved by one Abu Bakr Kuhistani, who is promoted for it, and Firdawsi is not in the story at all. By Dawlatshah's time it is a Firdawsi story.

What kind of survival that is

So the last Ziyarid's book outlived the dynasty by being taken apart for material by writers who did not credit it, whose versions of its stories became the famous ones, and who attached its anecdotes to more famous men.

That is not a bad description of what happened to the family as well. A house that had proposed to restore the Persian monarchy ended as a small Caspian principality; its greatest figure was murdered by his own officers; and the one piece of it that went on circulating did so under other people's names.

The tower, meanwhile, is still standing in the same field.