Shaddadid Emirate

Ganja

Drafted from Minorsky, Cambridge Oriental Series 6 (1953)formation

The Shaddadids were Kurds, and they appear in Arran — the country between the Kura and the Araxes, which is now western Azerbaijan — in the middle of the tenth century, when the Sallarid dynasty that had been holding it was failing.

Muhammad ibn Shaddad established himself at Dvin around 951. His sons took Ganja in 971, and Ganja was the seat for the following century. Fadl I, who ruled from 985 to 1031, is the substantial figure: he held Arran against the Byzantines and the Georgians and the Rus raiders coming up the Kura from the Caspian, and he built the bridge over the Araxes at Khudaferin, whose piers are still in the river.

Kurdish dynasties of this period are rarer in the histories than they were on the ground. The Marwanids in the Jazira were another and stayed Kurdish; the Ayyubids were Kurdish by origin and Arabised within a generation. All three came out of the same collapse in the same century — the Abbasid provincial system failing, and local men with local followings keeping what they were already holding — and none of them is usually described as part of one phenomenon, because the three are filed in three separate scholarly literatures.

The Shaddadid position was between things and always had been: Muslim rulers over a largely Armenian and Albanian Christian population, in country the Byzantines claimed, next to a Georgian kingdom that was rising, with the Shirvanshahs and Derbend to the east and Baghdad's authority to the south becoming theoretical. They managed it the way small dynasties in that position always do — by acknowledging whoever was nearest and strongest, and by marrying into whoever was next.

Then the Seljuks arrived, and the family got a second life in an unexpected place.

What the neighbours recorded

Drafted from Lastivertsi, History Regarding the Sufferings Occasioned by Foreign Peoples (1985)expansion

The Shaddadids left no chronicle of their own. What survives of their rise comes from the people they were rising at the expense of, and the Armenian who wrote the fullest contemporary account of the eleventh century mentions them exactly three times — which is itself the most informative thing about them.

The first mention is a relationship. Aristakes calls the Shaddadid ruler Apusuar — Abu'l-Aswar Shavur — and identifies him in a single clause as the man who held Dvin and Ganja and was the son-in-law of Ashot, king of Armenia. That is the whole of the Shaddadid method in eight words. A Kurdish Muslim family holding two cities in a Christian country had married into the Bagratid royal house, and the Armenian historian records it as an ordinary fact about who somebody was, not as a scandal.

The second is a war, and it is being fought against the Byzantines rather than against the Armenians. After Ani was surrendered to Constantinople in 1045, the emperor's lieutenant subdued the country and then marched on Dvin; Abu'l-Aswar fought him at the city gate, and Aristakes says an inestimable number of men were killed there — including Vahram, the great prince of Armenia, and his son, which caused the Armenians great mourning. Vahram had been one of the nobles who put the last Bagratid king on the throne and was ignored when it mattered. He died outside a Shaddadid city, in Byzantine service, fighting the emir who had married into his own royal family.

The third is a raid. In 1055 troops came into Armenia under the Seljuk sultan's name, though Aristakes notes that some say they were Abu'l-Aswar's. People fled out of the open country towards Ani and not all of them got inside, because night fell and the gates were shut. The troops came in the dark, seized the gates, put swords to work, and went home with booty and captives.

Set the three together and the shape of the expansion is visible without a single Shaddadid document. While the Armenian kingdom was being dismantled — by its own nobles selling their fortresses, by a catholicos selling the capital, by a Byzantine administration that garrisoned the cities and then quarrelled with the church — the family at Dvin and Ganja was doing none of those things. It was marrying into the royal house, holding its walls against the empire that had bought the kingdom, and taking what was outside the walls when it could. In 1045 Ani belonged to the Bagratids. In 1064 the Seljuks stormed it. By 1072 it was Shaddadid, and stayed Shaddadid for most of the next century and a quarter.

There is a reason a dynasty like this appears in other people's books only when it raids. A house that survives by marriage, by paying whoever is currently strongest and by not being worth a campaign does not generate the kind of event a chronicler records. It generates continuity, which is invisible until you notice that everyone who was writing about them is gone and they are still there.

A Kurdish emir in the Armenian capital

Drafted from Minorsky, Cambridge Oriental Series 6 (1953)peak

Alp Arslan took Ani in 1064 and sacked it. The Armenian nobility dispersed westward, and out of that dispersal came, eventually, the Armenian kingdom of Cilicia.

In 1072 the Seljuks handed Ani to a Shaddadid. Manuchihr ibn Shavur took the city and his descendants held it, with interruptions, until 1199 — a hundred and twenty-six years of a Kurdish Muslim dynasty governing the royal capital of Christian Armenia.

The arrangement worked better than the sentence suggests. The Shaddadids of Ani governed an overwhelmingly Armenian city; they used the walls the Bagratids had built and added to them, minted coins, maintained the churches, and appear in Armenian sources as rulers to be complained about rather than as an occupation to be resisted. Manuchihr built the mosque that still stands on the edge of the gorge with his name on it, beside a cathedral that was fifty years older than his family's arrival.

The Ganja half of the house did not last as long. The Seljuks took Ganja itself in 1075 and the senior line ended there. So the dynasty's story begins in Arran and finishes in Armenia, because its centre of gravity moved when its own capital was taken and it was handed somebody else's.

The mechanism behind that is the ordinary one for a family that holds territory as property: divide it among the members, let the branches diverge, and end up with what are effectively separate polities. Usually this is how a dynasty destroys itself. Here it is what saved it. The branch that was given the ruined city outlived the branch that kept the good one by a hundred and twenty-four years.

Where Saladin's family came from

Drafted from Minorsky, Cambridge Oriental Series 6 (1953)aside?What aside means

Minorsky's volume has three essays. Two are on the Shaddadids of Ganja and of Ani. The third is called "Prehistory of Saladin".

The Ayyubid story is usually taken to begin in Egypt, with a Kurdish officer in the service of Nur al-Din of Damascus, sent south on his uncle's staff. The histories say the family was Kurdish and move on.

It began here. Saladin's grandfather Shadhi came from Dvin, in the Armenian borderlands of Shaddadid Arran — the town Muhammad ibn Shaddad had established himself in around 951 — and the family was of the Rawadiya, a Kurdish group associated with that country. Shadhi took his sons out of the Caucasus to Iraq in the early twelfth century; one of them, Ayyub, ended up commanding a fortress at Tikrit for the Seljuk governor, and the other, Shirkuh, took service with Zangi. Their descendants took Egypt.

Minorsky's essay works out what can be established about that migration from the Arabic and Persian sources, and the answer is: a Kurdish military family from a frontier zone, in the service of a local dynasty, moving south when the local dynasty stopped being able to employ them, and finding better employment with the Turkish military households that were then assembling in Iraq and Syria.

It is a completely ordinary career, and it is worth having because of where it ends. The man who took Jerusalem in 1187, and whose family ruled Egypt, Syria, the Jazira and Yemen for the next eighty years, came from a Kurdish town in the Armenian borderlands of the Caucasus, two thousand kilometres north-east of anywhere his dynasty is normally discussed.

Being descended from a place is a genealogy and not a succession, and Shadhi's departure from Dvin did not transfer anything from one state to another. It is worth knowing anyway, because the alternative is a reader who finishes an account of the Ayyubids without ever learning where they were from.

Tamar takes Ani

Drafted from Minorsky, Cambridge Oriental Series 6 (1953)end

In 1199 Georgian forces under Tamar's generals, the Zakarid brothers — Armenian nobles in Georgian service, who governed the Armenian territories the kingdom took — captured Ani from the last Shaddadid emir.

That ending is worth pausing on, because it is almost the only time in a thousand years that one Caucasian state ends another. Everything else in the isthmus is ended from outside it: the Bagratid kingdom by Byzantium, Derbend by the Seljuks, the Shirvanshahs by the Safavids, Iberia and Armenia and Albania by the Sasanians. The Caucasian polities are neighbours who intermarry, raid, shelter each other's exiles and quarrel about passes, and hardly ever finish one another. What finishes them arrives.

The city changed hands afterwards more or less continuously — Georgian, Mongol, Turkmen, Ottoman — and was finished by earthquakes and by the shifting of the trade routes rather than by a conquest. What is on the site today is a ruined city on a closed border in a military zone, with a cathedral, a mosque, several churches and a great deal of wall.

The dynasty invented no script, propagated no religion, and left nobody claiming to be its heir. What it has instead is a position: a Kurdish emirate governing an Armenian capital under a Turkish sultanate, ended by a Georgian queen, and a family that left it in the twelfth century and took Egypt.

That is a fair description of the Caucasus. Nothing in it inherits from anything else in it. Everything in it connects to somewhere else.