Old And New Saray - Capital Of The Golden Horde
(a new look at sources)
By E. Y. Goncharov Translated by David Elliot

It’s amazing that all these NEW Sarays, Bulghars, etc. are up to now named only on coins, whereas no other archeological evidence of any real existence of these duplicate geographical points has been precisely ascertained by learned investigators. It turns out that the Fate was so merciless that it did not spare a single of the above mentioned geographical duplicates, which would have saved scholars from much vain work, searching for these duplicates. It appears to us that this search is nothing but chasing phantoms created by the scholars themselves.

(V. D. Smirnov, Crimean Khanate under the
Supremacy of the Ottoman Empire, part 1)

The archeological and historical literature of the last century dedicated to the problems of the history of the Golden Horde and the states and governments that had been coming into contact with it, usually indicates two capitals - Saray and New Saray. Moreover, there are marshalled the historical reconstructions based on the confident belief that there were two different cities.

The reasons calling forth this settled opinion are as follows: 1) the coins on which these mints are shown; 2) Saray and Saray Grande charted on Fra Mauro’s map; 3) an Arab biographer's indication of the death of Uzbek in New Saray; 4) ‘Arabshah’s notice on the construction of Saray, 63 years before its destruction in 1395; 5) the existence of two big cities, Tsarevskoe and Selitrennoe (in Lower Povolzhye, center of the Ulus of Juji) demanding attribution. All the remaining information presents itself as logical constructions by the researches of these problems. At the present moment, the existence of two Sarays appears to be a fact accepted by the majority of historians and archeologists.

In our opinion, the question is not to be considered closed, because the materials accumulated to date call for a different interpretation of these facts, which have been long known to Ballod, Yakubovsky, and other scholars.

In the excavated plots in Selitrennoe, the earliest buildings and construction periods are dated back to the first half and the middle of the 14th century with layers definitely attributable to the time of Toktamish also being able to be distinguished. Layers dated earlier than the 13th century are unknown; this is not because the archeologists are unable to divide the datable material, but because in these excavated plots constructions were made only in the second quarter of the 14th century. This is commonly explained by the “outburst” of city building, which has been noted long before for the rules of Uzbek and Janibek (1312-1357).

During archeological investigation of the aforesaid monument, attention was focused primarily on the excavations of sites lying north of the village of Selitrennoe. It is a well-known fact that cities and villages situated by rivers always have a natural growth directed upstream. It is known that the very oldest part sits lower and often converts into abandoned wasteland with ruins. As for Selitrennoe, the excavations rendering materials open to the study are located higher than the village along the Akhtuba river. But all available descriptions of the present state of the site witness that a certain part of it is now farmland. This area is situated lower (i.e. to the south) along the river and in all likelihood appears to be the older part of the city. Near the village the banks have been washed away with freshets, so that now there are exposed to view the remains of a rich and very large structure (masses of plinths, fragments of mosaics and other small finds) in the excavated trench. In archeological folklore they are called ‘remains of a palace,’ which can probably, but not precisely, be proved to be a real palace.

G. A. Fedorov-Davydov writes that “the city came into existence on a wilderness and empty place” [1]. For these reasons, all excavations point at A.D. 1350 plus or minus 20 years.

Here we abandon our little archeological excursus without any inferences and turn to the written sources.

The Arab writer Elomari (al-‘Umari) relating the affairs in the Ulus of Juji has two passages about Saray city. (3: 229, 241). “The capital of the local king is Saray. This modest city is between desert and river. Now lives there the sultan Uzbek, who constructed in it a school for learning.” The second passage has been repeatedly cited, therefore I will produce only a part of it, that which is important for this study. “It, that is Saray, is a great city, consisting of markets, baths, and establishments of piety (?); a place where one makes his way for all sorts of goods… The palace with a golden crescent above, encircled by a wall, turrets and houses, in which his emirs live.” What explains such differences in one and the same work? The author was for a long time the official secretary for an Egyptian sultan, collecting information about various countries from the persons who had visited them in person. Their names appear in different sections of al-‘Umari’s writings. The first of his informants had visited the White Horde and Saray in the first years of Uzbek Khan’s rule; the second one, judging by his remarks about the heir to the throne, had been there in the end of the same reign. Preparing the final redaction of the work, the sultan’s secretary, consistently, without critical observation, rewrote these and other witnesses' accounts. Meanwhile, the following events took place between those visits to Dasht-i-Qipchaq.

Saray, built as the Winter place of sojourn for the Mongol khans, was truly, actually not a great city until Khan Uzbek. The khans sojourned the majority of time at the headquarters of the Horde, followed with all the administration, which was not conducive to the great growth of a city. Before Uzbek’s time, Saray was rather a main commercial center and industrial trading center of the country than just a political center. The expression of Mongol camp mentality, following the Yasa (the Mongol customary law) and the absence of need for certain structures (appearing later) - all these things make Saray appear as Qaraqorum of Willem van Ruisbruck (Rubruquis, Rubruc). (In other words, the khans and their political administration wandered about their domain and did not set up a settled urban capital--translator's note.)

For successfully expanding Islam (this became one of the significant actions of Uzbek), it was necessary to build mosques and other “places of piety.” Baths were also required- an important element of Moslem culture. The entry of the Ulus of Juji into the system of states worshipping Allah and His prophet attracted merchants from Islamic countries. Economic bonds with Mamluk Egypt, the strongest Muslim government of the time and the strongest country of the Near and Middle East, were strengthened by successful commercial revolutions, which in turn required new markets, caravans, a “place where merchants find their way.” Growth of wealth and increasing demands of production always produced population growth, this did not elude Saray. Dwelling places of the region increased. To transform the village into a capital of a large Muslim government, giving it the appropriate aspect and status, Uzbek actually came to build a new city, which received the official name Saray al-Jadid (The New Saray). These reasons, it seems to me, explain the different characterization of the capital of the Golden Horde in the chronicle of al-‘Umari.

Because the sources of the 14th century and following time completely pass over in silence the existence of two cities with the same name and the transfer of the capital to another place, they testify that the territorial point, where there were Russian and Arab eyewitnesses, was not transferred. The town familiar to them had merely grown up, which had no special significance for political contacts (whereas the main goal of the visits were just those contacts), and the literary witnesses known to us did not reflect about it.

The date of the construction of Saray (1332) recounted from ‘Arabshah’s information agrees with the dating of the excavated plots of Selitrennoe. The previous text of ‘Arabshah apparently discourses about one city, and we have no reason to attribute it to two different towns. A cursory analysis of the information from al-‘Umari in the context of the development of the Muslim cities in general, permits us to make conclusions about the causes of the growth of Saray in the reign of Uzbek Khan (1312-1341). It also agrees with the data of ‘Arabshah and the archeological datings of Selitrennoe. Such circumstances allow me to consider the toponyms ‘Saray’ and ‘Saray al-Jadid’ to be relating to one city, situated in the aforementioned site (Selitrennoe). All the archeologists working on these ruins, investigated the remains of New Saray, because the location of the previous Saray has still not been discovered or exposed.

But, on the other hand, in the time of the rule of Toktamish and early in the 15th century, the dirham coins are issued both in Saray and Saray al-Jadid, which seems to be a witness about two cities with the same name. Concerning these affairs, we do not know the organization of the coinage in the Ulus of Juji nor the principles of distributing monetary regalia to cities; the rights by which cities were granted titles, etc.

The coinage is able to serve as an indirect witness that the New Saray was not a separate city. The Janibek dirhams were issued by three mints: Saray al-Jadid, Khorezm, and Gulistan. There is only one silver coin dated 749 AH with the mint Saray al-Mahrusa. The whereabouts of Gulistan will not be examined in detail here. Khorezm is not called into doubt. Only the location of the first mentioned place remains unclear. Under Toktamish, the majority of coins bears the mint names preceded with the Arabic word balad which in this period designates, primarily, ‘country; territory; region; city (or cities)’. We read balad Azaq, balad Qrym, balad Saray, balad Haji Tarkhan, but no balad Khorezm. Balad Saray al-Jadid is met only once, in a very rare issue of 791 AH. Only once on dirhams of 775 AH is found balad Ordu. Usually the word balad is translated as ‘a city’, but Orda was not a city itself. Khorezm is historically a region, the capital of which at various times was Kyat or Urgench. But why, having balad Saray, do we not also have balad Saray al-Jadid , even though Gulistan (Persian. “Rose Garden”) appears on the dirham coinage started under Janibek in 752 AH as balad Gulistan? The answer thrusts itself forward - Saray al-Jadid was not a balad.

Every city, where silver coinage was permitted, was a center of a province. Is it possible, in general, in the Golden Horde, as in other Muslim states, that coinage customarily cited the name of the place where taxes and other revenues were gathered in silver from which the coins were made? A precedent to this is found in the Crimea at the beginning of the 14th century when copper coins were issued with the mint name of Solkhat, which according to historical writings was the administrative center of the Crimea, but the dirham coins bore the name Qrym (applying the regional name for a city was established later). A second precedent, already mentioned, was the Khorezm regency with the capital at Urgench (Gurganj). Consequently, the numerous balad’s discussed in the previous paragraph can be interpreted as ‘[coinage of] the district of Azaq, Gulistan, Qrym, Ordu, Saray…’ Saray al-Jadid is able to appear as a mint of the capital, which in all countries has a special status and its own administration and financial organization. The existence of two mints in one city is not an unusual situation. There is a well known example of minting double dirhams in the same year at Shirvan and Shirvan Bazar at the time of the short rule of Birdibek, but this has not become the basis to conclude the existence of a second Shirvan. In the middle of the 12th century, a dirham was issued with the inscription balad Binkath, the capital of the region named al-Shash (6; 167). Even earlier, from the middle of the 10th century until the 13th in Akhsikath, the money was minted with the regional name Ferghana, and at the same time it was minted from the name of the city. Translating the word balad as a term denoting a certain administrative division will not agree with the existence of Saray al-Jadid, and therefore it was not applied to it . If the Arabic word balad is translated ‘a city’ then it becomes obscure why it was not applied to Saray al-Jadid.

The last “strong pillar” of the interpretation of “two capitals of the Golden Horde” appears on the map of Fra Mauro. It has already been the subject of investigation of Golden Horde geography and toponymy (5; 177). The map has the following objects written: ‘Saray Grando, 1 ordo of Saray, Saray, f. Cara Saray’. The latter name being a river (Lat. flux), we will not consider it. ‘Saray Grando’ is written with capital letters encircled by contour lines, as are also written Asia, Lithuania, Organca (Urgench), Parthia, Rossia, Scythia, Tartaria, etc., which means that this type was employed for signifying larger zones, rather geographical than administrative. The second toponym is written with middle sized shaded letters. The similar script is used for ‘Edil, 1 ordo de Cagatai, 1 ordo de Organca, Ruenia, Rossia rossa, Rossia negra, Siroan (Shirvan), Tataria’ (the last name being cited for the second time but on the opposite bank of the river Tanais, between it and the Northern Caucauses), etc. This script would have designated administrative regions, for example ‘Rossia’ being divided into ‘Rossia rosa’ and ‘Rossia negra’, and ‘ordo de Organca’ being included into Organca (Urgench), just as ‘1 ordo de Saray’, ‘Edil’ and perhaps the second Tataria as part of ‘Saray Grando’. The smallest letters in a different script signify certain objects, to wit: cities, rivers, pictured things and buildings; so are written down Astrakhan, Tana, Riga, etc. Saray is also named this way, that is, as a city. There are no more denotations of the town of Saray on the map. The existence of a second, new Saray would have been marked, as is marked the existence of Organca nuova (New Urgench) near the south-eastern end of the Caspian Sea. In our view, the data of Fra Mauro’s map underscore that there was only one Saray located in the region of the same name.

In this view, the existence of two capitals in the Golden Horde is not confirmed by the facts at out disposal. The capital city of the Ulus of Juji was the urban conglomerate Saray - Saray al-Jadid, the administrative center of the Saray region, situated near the modern village of Selitrennoe in the Astrakhan oblast.

************

The only written mention of Saray al-Jadid is a notice by al-Asadi about the death of Khan Uzbek in New Saray (3, 447; 445 Arabic). The place of his burial has been counted as unknown until this time. Still we can, in our view, substantiate its location. Near the aul (village) of Lapas, in the Astrakhan oblast is the well known grandiose burial complex from Golden Horde times. Writing of it in his book, V. L. Egorov (2, 117) counts four burial structures, but in the excavation of 1996 there were unearthed not less than 16 burial structures. The four largest ones look like earthen platforms with remnants of mausoleum structures. The very largest of them is a two-tiered platform, bearing extensive ruins with the usual rubble of decorated bricks, colored plinths, and mosaics. Local Nogai residents call this burial mound (kurgan) “Devlet-khan”. On the shore of a channel of Akhtuba, Great Ashuluk, a little ancient settlement is located. The removed material and remains of construction witness without doubt that here lived and worked the builders of those four mausoleums. The processed numismatic material collected at Lapas city and the burial mound includes 189 bronze coins and 3 silver coins. Its division by mint and year is as follows:

Silver	1) Mokhshi	712 AH	1 coin
	2) Saray	717	1
	3) -«-	720-s	1
Bronze:	1) Saray	no date (leopard)	2 coins
	2) Saray	721 (star)	14
	3) -«-	726 (falcon)	14
	4) -«-	731 (16 for a dang)	54
	5) Saray al-Mahrusa	731	7
	6) Saray	730-s (leopard and Sun)	36
	7) Azak	-«-	1
	8) Saray al-Jadid	740-s (2-headed eagle)	13
	9) -«- -«-	750-s (flower)	3
	10) period of revolt, unepigraphic		1
	11) uncleaned and unattributable		56
	Total		204

 

These statistical data testify to the fact that Lapass (a town) existed in the period when puls of Uzbek predominated in monetary circulation and Janibek’s coins had only just taken root in the market place, that is towards the end of Uzbek's rule and the beginning of Janibek's rule. Judging from the character of the remains - which were the dwelling places of local potters, house-builders, and other craftsmen working on the erection of these burial monuments, we conclude that in the mausolea there were buried four Chinghizid Muslims, namely, Uzbek, Tinibek, Khizrbek and some fourth, whose name and death are unknown to us from extant historical sources. The very largest mausoleum, to which, judging by the form of its remains and the mosque or chapel (zawiye) attached to it, belongs to Uzbek - a khan who converted the Golden Horde to Islam, thus bringing the Golden Horde to the highest international level. There is no other person worthy of such expense on the constructions of so fashionable a resting place who lived around the 1330’-1340’s that we know. If our version is true, then the fact that the location of Lapas complex is near Selitrennoe (about 40 kilometers) and not Tsarevo, corroborates the conclusion that it is namely Selitrennoe that appears to be the remains of balad Saray or Saray al-Jadid.

1. G. A. Fedorov-Davydov. Golden Horde Cities of the Volga Region, Moscow, 1994.

2. V. L. Egorov. Historical Geography of the Golden Horde, Moscow , 1983.

3. V. G. Tizengauzen. Collected Information Related to the Golden Horde, Moscow, 1883.

4. I. V. Evstratov. On the Golden Horde Cities Located near Selitrennoe and Tsarevo Villages <…>, in: Bronze Epoch and Early Iron Age in the History of the Ancient Tribes of South Russian Steppes. Saratov, 1997.

5. Y. E. Varvarovsky, I. V. Evstratov. Concering the Identity of F.F. Chekalin’s Translation of the 1459 Map by Fra Mauro, in: The Antiquities of the Volga-Don Steppes, Volgograd, 1998.

6. E. A. Davidovich. Monetary Economy of Middle Asia in 13th Century, Moscow, 1972.

7. A. P. Grigoriev. The Golden Horde City of Orda, in: Orienalistics, Leningrad, 1990.

8. Essays on the History of Arab Culture, 5-15th Centuries, Moscow, 1982.

A note on the translation: Dr. Vladimir Nastich has greatly improved my translation, but I am still responsible for it. He asked that I follow his spelling of names and places, better than mine, but not consistent with Steve Album and Michener that I usually followed. I have included Fra. Mauro’s map as a necessary part of the discussion. If there are other articles or monographs that might aid the collector of Golden Horde coins that need to be translated from Russian (or French), I would be happy to do so with the understanding that I am just a collector and claim no scholarship in these areas. I just want to make enough references available to make the hobby interesting. David Elliott