Sultanate of Rum

A branch that outlasted the trunk

Drafted from (eds), The Seljuks of Anatolia: Court and Society in the Medieval Middle East (2013)formation

After Manzikert in 1071 the Byzantine defence of Anatolia collapsed, and Turkmen groups moved onto the plateau faster than any authority could organise them. The sultanate that emerged was founded by Suleyman, son of Qutalmish — a member of a Seljuk line that had lost a succession war against the main house, and had therefore been pushed to the furthest frontier available.

That origin shaped everything. Rum was a Seljuk polity that the Great Seljuks did not control and periodically fought, established in territory taken from an empire rather than inherited from the caliphate, and governing a population that was Greek and Armenian Christian for most of its existence. Its rulers took Persian regnal names out of the Shahnama — Kayqubad, Kaykhusraw, Kaykawus — which is a small thing that says a great deal about what a Turkish dynasty on a Byzantine plateau thought it was.

The Great Seljuk empire — the trunk this branch had been broken off — ended in 1194. The Anatolian offshoot that its own family had exiled to the far frontier lasted another hundred and fourteen years after that, which is the joke buried in the arrangement: the losers of a succession war outlived the winners by more than a century, because the ground they had been pushed onto turned out to be worth more than the ground they had been pushed off.

Two seas

Drafted from (eds), The Seljuks of Anatolia: Court and Society in the Medieval Middle East (2013)expansion

Between roughly 1200 and 1240 the Sultanate of Rum was one of the better-run states in the eastern Mediterranean, and almost nothing in the popular picture of medieval Anatolia reflects it.

The sultanate had spent its first century as a frontier polity among rivals — Danishmendids to the east, Byzantines to the west, crusading armies passing through twice — and it emerged from that with Konya, most of the plateau, and under Kayqubad I from 1220 a coastline: Antalya on the Mediterranean and Sinop on the Black Sea, taken deliberately, because a landlocked plateau state that controls two ports controls the route between them.

The two ports are the whole of the strategy and they were taken in that order and for that reason. Sinop in 1214 gave the sultanate the Black Sea and the Crimean trade beyond it. Alanya on the Mediterranean, taken in 1221 and rebuilt with a dockyard and a tower still standing, gave it the other end. Antalya had come in a few years earlier. What lay between them was two hundred and fifty miles of plateau that the sultanate already held.

A landlocked state with two coasts is not landlocked; it is a bridge, and it can charge for the crossing. Everything the sultanate is remembered for materially — the roads, the caravanserais, the customs revenue, the treaties with Venice and with the Italian communes of the Crimea — follows from a decision to buy both ends of a line and then make the line worth using.

And he saw them coming

Kayqubad also did the one thing that mattered in the 1220s, which was to take the Mongols seriously before he had to.

He fortified. He took Erzurum in 1230, which put a garrison on the eastern approach. He fought and then negotiated with the Khwarazmian remnant under Jalal al-Din, who was the wreckage of the last state the Mongols had destroyed and was now on his frontier. And he sent an embassy east, which is the act of a ruler who has understood that the thing coming towards him is not a raid.

He died in 1237, four years after the Mongols reached Azerbaijan. The arrangements he had made were tested six years after that, by his son, and did not survive the test.

The dish

Drafted from Abu'l-Faraj), The Chronography of Gregory Abu'l Faraj, the Son of Aaron, the Hebrew Physician, Commonly Known as Bar Hebraeus (1932)expansion

The sultanate became the only Turkish power in Anatolia under Kilij Arslan II, who ruled for nearly forty years, absorbed the Danishmendids, beat the Byzantine emperor at Myriokephalon and outlived every rival. Bar Hebraeus, writing from Melitene, records one episode from the middle of that process, and it explains the method better than the battles do.

The neighbours had had enough of him. Kilij Arslan had attacked Melitene, failed to take it, and carried the country people off into slavery on his way back to Kayseri. So a coalition formed: Nur al-Din of Aleppo, the Artuqid lord of Mardin, the lord of Hisn Kayfa, the Armenians of Cilicia, and the Danishmendid lord of Sivas. They assembled at Sivas and marched to the gates of Kayseri.

Kilij Arslan did not come out. He asked for terms, and gave back the captives he had taken from Melitene. Then the allies made a further demand: he was holding four of his own nephews in prison, and they wanted them released.

What he did, Bar Hebraeus writes, was a most accursed thing. He killed one of the four, roasted him, arranged him on a dish, and sent him to the boy's father. And he swore that if they went on asking for the other three he would treat all of them in exactly the same way.

They left him and went their way.

It is worth being careful about what that story does and does not establish. Bar Hebraeus is a Syriac Christian bishop writing a century later from a city Kilij Arslan had raided, and the account is meant to horrify. But the shape of it is politically exact, and the shape is what matters: a coalition of five powers brought a sultan to the walls of his own capital, took back what he had stolen, and then discovered that there was nothing further they could make him do, because he was willing to do something none of them were. They went home.

That is how the Sultanate of Rum ended up owning Anatolia. Not by winning a decisive campaign against the coalition — he had just lost to it — but by outlasting the coalition and everybody in it. Nur al-Din died in 1174. The Danishmendid line at Sivas was absorbed within a few years and the last of it extinguished in 1178. The Byzantine army was destroyed at Myriokephalon in 1176. The Artuqids stayed small. Kilij Arslan lived until 1192.

He then divided his territories among eleven sons, which undid a good deal of it, and the sultanate spent a decade putting itself back together before the century that the caravanserais and Konya belong to. But by then there was no longer a second Turkish power in Anatolia, and the reason is the story just told: a ruler who could be beaten and could not be coerced, in a region where everybody else could be.

The forty years it worked

Drafted from (eds), The Seljuks of Anatolia: Court and Society in the Medieval Middle East (2013)peak

Between roughly 1200 and 1240 the Sultanate of Rum was one of the better-run states in the eastern Mediterranean, and almost nothing in the popular picture of medieval Anatolia reflects it.

The state that came out of the twelfth century had spent it as a frontier polity among rivals — Danishmendids to the east, Byzantium to the west, two crusading armies marching through. What it looked like by 1230 was a court.

A Persian state in a Greek country

The administration worked in Persian and the chancery style was Iranian. The sultans took their throne-names out of the Shahnama — Kayqubad, Kaykhusraw, Kaykavus, the names of legendary Iranian kings, borne by Turks ruling in Anatolia. They married Byzantine, Georgian and Armenian princesses, and several of them had Christian mothers.

None of that is decoration. A Turkish family adopting a Persian bureaucratic order is the standard move of the period — the Great Seljuks had done it in Iran two generations earlier — but here the order was carried into a country that had been Greek-speaking and Christian for a thousand years, and it did not displace what it found. Greeks and Armenians stayed in the administration. Christian officials held serious office. Byzantine formulae turn up on the coinage.

The result is a state whose ruling family were Turks, whose paperwork was Persian, whose taxpayers were Greek and Armenian, and whose sultans were named after Iranian heroes out of a poem written three centuries earlier and two thousand miles east.

Why it is not remembered

Partly because it did not last, and partly because of what came next.

The century after 1243 gave Anatolia a Mongol tributary regime, then a dozen squabbling beyliks, then the Ottomans, and the Ottoman account of its own origins had no particular use for a Seljuk state that had governed the peninsula competently and then been beaten in an afternoon. Rum survives in the popular picture mainly as the thing the crusaders marched past and the thing the Mongols knocked over.

What it actually was is on the ground: several dozen caravanserais, the mosques and medreses of Konya, Sivas and Divriği, a working coinage, treaties with Venice, and a court that could attract a Khurasani scholar's family fleeing the Mongols and give them somewhere to teach for fifty years.

A day's march apart

Drafted from (eds), The Seljuks of Anatolia: Court and Society in the Medieval Middle East (2013)aside?What aside means

Under Kayqubad I the sultanate took the Black Sea port of Sinop and the Mediterranean port of Alanya, which gave it both ends of a route across Anatolia, and then built the route: a chain of stone caravanserais spaced roughly a day's march apart, with free lodging, stabling, a bath and often a mosque and an infirmary, endowed as charitable foundations and open to merchants regardless of religion.

Several dozen survive. They are the most substantial physical inheritance of the sultanate, and they represent a specific economic strategy — a landlocked Anatolian state making itself the cheapest and safest transit between two seas, and insuring merchants against loss on its roads.

The population being governed was largely Greek and Armenian Christian throughout, and the sultanate's practice was accommodation rather than conversion: Christian officials in the administration, dynastic marriages with Byzantine and Georgian houses, and a court culture that was Persian in language and literature while the countryside was not. Jalal al-Din Rumi wrote in Persian at Konya under Mongol overlordship, which is why the world calls him after this sultanate.

That mixture is the ground the Ottomans grew from. What made the Byzantine frontier productive for a small Turkish principality in 1300 was two centuries of Turkish, Greek, Armenian and Persian coexistence in Anatolia that this sultanate had organised, and that nobody had to invent again.

Kösedağ

Drafted from (eds), The Seljuks of Anatolia: Court and Society in the Medieval Middle East (2013)contraction

On 26 June 1243, at Kösedağ in eastern Anatolia, a Mongol army under Baiju met the forces of Kaykhusraw II and destroyed them in a few hours. The sultan fled before the fighting had finished. There was no siege, no long campaign, and no second battle.

What followed is the strange part, and it is why the sultanate has two possible end dates sixty-five years apart.

The Mongols did not occupy Anatolia and did not depose the dynasty. They imposed tribute and took the sultanate as a vassal, and the Seljuk house went on sitting at Konya for another two generations while Ilkhanid governors collected the revenue and decided, increasingly openly, who would sit there. Rival sultans were installed by rival Mongol factions. The country was partitioned between brothers by Mongol arbitration. A rebellion in the 1270s was put down by a Mongol army.

What a defeat like that actually costs

Not territory, at first. Authority.

A sultanate whose sultan is chosen at Tabriz cannot discipline anybody, because everybody it might discipline can appeal over its head. That is the mechanism, and it worked downwards through the whole structure. Provincial commanders stopped remitting revenue. The Turkmen of the marches — who had always been the least governable element and had been held in place by a combination of patronage and force — found that neither was available any more.

The sultanate could no longer restrain them and the Ilkhanate, whose interest was the revenue of the settled districts, did not care to. So the frontier filled with independent beyliks, a dozen or more of them, each with a warband and a stretch of country, and among them one in the north-west on the Byzantine march that was too poor to be worth anybody's attention.

Two battles

Manzikert opened Anatolia to the Turks and this sultanate organised what came in. Kösedağ broke the organisation and left the country full of small Turkish states.

The two battles are a hundred and seventy-two years apart, and the second one is what made the first one's consequences permanent. Without Kösedağ there is a Seljuk sultanate governing Anatolia from Konya, with a frontier it polices; after it there is a peninsula of competing warbands with a Mongol tax farm behind them, which is a much better environment for a small aggressive principality with nothing to lose.

1308

Drafted from (eds), The Seljuks of Anatolia: Court and Society in the Medieval Middle East (2013)end

The last Seljuk sultan of Rum disappears from the record around 1308, and nobody recorded how.

There is no deposition, no execution anybody bothered to write down, and no successor. The line simply stops being mentioned, in a country where the mentions had been getting thinner for two generations, and the Ilkhanid governor went on governing exactly as before.

Sixty-five years of a state that was not one

That is the interval between the battle and the last name, and it is the reason the ending here is a slow absorption rather than a conquest.

The Mongols had every opportunity to abolish the sultanate in 1243 and chose not to. A dynasty with a name people obeyed was a cheaper instrument than a garrison: it collected, it legitimised, and it could be replaced when it stopped being convenient, which it repeatedly was. What that produced was a throne occupied by men who had been put there from Tabriz, competing with brothers who had been put somewhere else from Tabriz, in a country whose actual revenue was being assessed by Persian officials working for a khan in Azerbaijan.

By about 1280 the sultanate was a formality. By 1308 it was not even that, and the fact that nobody can say what happened to the last of them is the most accurate possible statement of how much the office was worth by the end.

What replaced it

Not one thing. A dozen.

The beyliks that had formed in the marches while the centre could not restrain them became, in the fourteenth century, the political map of Anatolia: Karaman at Konya itself, claiming to be the heirs of Rum and asserting it for two hundred years; Germiyan, Menteshe, Aydin, Saruhan and Hamid across the west and the Aegean coast; Candar on the Black Sea; Eretna and then the Ottomans.

Karaman's claim is the one worth pausing on, because it shows what the Seljuk name was still worth. In 1277 Karaman took Konya while the sultanate was an Ilkhanid tributary and put a man it said was a Seljuk on the throne. The attempt collapsed within weeks. The claim to be the heirs of Rum was maintained for the next two centuries anyway, by a dynasty that had briefly held the capital for long enough to say so — and it was maintained specifically against the Ottomans, who had no such claim and did not need one.

What was written at Konya

Drafted from Rumi, Selected Poems from the Divani Shamsi Tabriz (1898)aside?What aside means

On 28 November 1244, a man in coarse black felt walked into Konya.

Nicholson gives the date precisely and then admits that almost nothing else about Shams of Tabriz is fixed: his parentage is disputed, his birthplace is disputed, and the traditions about him contradict each other on nearly every point. He calls him a weird figure who flits across the stage for a moment and disappears tragically enough. What Shams did in Konya was to take a forty-something professor of religious law, with a chair and a following and a reputation, and comprehensively unmake him.

The sultanate had lost its army at Kösedağ seventeen months earlier.

The professor

Jalal al-Din's family had left Balkh in Khurasan ahead of the Mongol advance and travelled west, arriving in Anatolia in the 1220s while Kayqubad was recruiting scholars. He taught at Konya for most of his life and died there in 1273. He is, on most estimates, the most widely read Persian poet who ever lived, and the name the world knows him by — Rumi, the man of Rum — is the name of the sultanate.

After Shams, and after Shams disappeared, he wrote. The lyrics were published under Shams's name rather than his own, which is either the most extravagant act of attribution in Persian literature or a statement about who was speaking. One couplet Nicholson quotes puts the break at a specific age:

Forty years did Reason plunge me in care; At three score and two I was made a prey.

What he wrote in a country of four languages

The best-known of the poems is a list of everything he says he is not:

I am neither Christian, nor Jew, nor Gabr, nor Moslem. I am not of the East, nor of the West, nor of the land, nor of the sea; … I am not of India, nor of China, nor of Bulgaria, nor of Saqsin; I am not of the kingdom of 'Iraqain, nor of the country of Khorasan. … My place is the Placeless, my trace is the Traceless.

It is worth reading that where it was written rather than as a devotional generality. The man setting down that he is not of the East nor of the West was a refugee from Khurasan, living in a Greek-speaking city, in a Turkish sultanate, whose administration ran in Persian, under the protection of Mongols who had not yet become Muslim, in a household where the audience for a public lecture would include Turks, Persians, Greeks and Armenians. The Masnavi has lines of Greek and Turkish in it because the people in the room spoke them.

A poem announcing that its author belongs to no nation and no confession was composed in about the only place in the thirteenth-century world where that would have sounded like a description rather than a provocation.

The inversion

The other thing worth putting on the record is the timing, because it is the opposite of what one expects.

The decades that produced Rumi, the historian Ibn Bibi, the caravanserais, the Konya and Sivas architecture, and the Mevlevi order that grew out of Rumi's funeral are the decades in which the sultanate was a Mongol tributary being administered from Tabriz. The political collapse and the cultural peak are the same period. Whatever relation there is between a state's strength and what gets made under it, this is not an example of it.

The order that formed around his memory became an institutional fixture of Anatolia and then of the Ottoman empire for six hundred years — a longer run than the sultanate that housed him managed, by a factor of four.

A different sultan in every town

Drafted from Battuta, The Travels of Ibn Batuta (1829)afterlife

In about 1332 Ibn Battuta came ashore on the southern coast and walked across what had been the Sultanate of Rum. His itinerary is the best surviving description of what replaced it, and the striking thing about it is that he records who ruled every single place, because it was different every time.

Alanya: Turkmen, under Yusuf Beg son of Karaman. Milas: Orhan Beg son of Menteshe, whom he calls a very excellent prince. Birgi: Muhammad ibn Aydin, who showed him a meteorite. Konya and Larande: Badr al-Din ibn Karaman. Aksaray, Niğde, Kayseri, Sivas, Amasya, Erzincan, Erzurum: all of them, he says, subject to the king of Iraq — meaning the Ilkhan, eight hundred miles away at Tabriz.

Not one sultan of Rum. A generation after the last Seljuk stopped being mentioned, a traveller crossing the peninsula met half a dozen rulers and a foreign tax administration, and the country's name was still Rum.

Konya

He found it large and handsome, abounding in water and gardens, and belonging to Karaman — though the king of Iraq, he notes, has occasionally had the rule of it too.

He lodged with the qadi, a member of the brotherhood of youths, in a beautiful cell with a great many disciples who traced the authority for their legal opinions back to Ali. And he went to the tomb of the Shaykh Jalal al-Din, better known by the title of Mawlana, who was, he says, very highly esteemed.

Then he tells the story he was told there, which is not the story anybody tells now.

The sweetmeat seller

The Shaykh, Ibn Battuta was informed, had at first been an ordinary lecturing doctor with a large number of pupils. One day a stranger came into the lecture-room with a basket of sweetmeats on his head, which he was selling. The Shaykh said: bring your basket here. The man took out a piece and gave it to him, and he ate it.

Then the man went out — nobody else having tasted the sweetmeat — and the Shaykh became agitated and went out after him, abandoning his reading and leaving his pupils sitting there expecting him back. They eventually went looking and could not find where he had gone.

Some years later he returned with his mind deranged, speaking nothing but Persian verses. His pupils followed him about writing down what he said, and published it under the title of the Mathnawi, a book, Ibn Battuta adds, highly esteemed in these parts.

That is Shams of Tabriz sixty years on, remembered in Konya as a man with a tray of confectionery. It is a folk version and it is wrong in every detail, and it has the shape of the thing exactly right: a stranger came in, the teacher ate something and followed him out, and the man who came back was not the man who left.

What was actually holding the country together

Not a sultanate, and not the Ilkhanate either.

In every Turkmen town, Ibn Battuta says, there was a brotherhood of young men — the akhis — and no people on earth were more courteous to strangers, more ready to supply them with food, or more opposed to oppressors. A tradesman around whom the young men of his craft and any destitute foreigners collected would be made their president. He built a cell and put a horse and a saddle in it. In the evening the members brought in everything they had earned that day, and it was sold to buy food and fruit for the house. A traveller who arrived was taken in and kept until he left the country.

Ibn Battuta was fed by them from one end of Anatolia to the other, which is why his account of the beyliks is so detailed: he was passed along a network that existed to receive people like him.

The carpets

Sixty years before him, Marco Polo had crossed the same country going the other way, and his chapter on it is one page long.

There are three sorts of people in Turcomania, he says. The Turkomans, who keep cattle in the mountains and downs and breed excellent horses and very valuable mules. And the Armenians and the Greeks, who live mixed among them in the towns and villages, occupying themselves with trade and handicrafts, and who weave the finest and handsomest carpets in the world, and quantities of rich silks. The chief cities are Konya, Sivas and Kayseri. The people are subject to the Tartar of the Levant.

He was writing while a Seljuk sultan was still nominally on the throne at Konya, and he does not mention that there is a sultan. What he mentions is the suzerain, the towns, and what was for sale.

Between the two travellers the whole afterlife is there. The state disappeared without anybody outside noticing much. The country went on: three populations living mixed in the same towns, a craft network feeding strangers, a shrine at Konya, and the best carpets in the world.