Mongol dominion in Russian lands proved to be more abiding than that of the il-khans in Persia, as already indicated, partly because of cooperation with Russian princes. Mongol policies were not immutable, and, particularly during the long rule of Ozbeg (1313-1341) important changes came about in the Horde's relationship with Moscow. The favor Ozbeg showed the Muscovite prince Ivan I Kalita was in acknowledgment of not only his services as an efficient tax collector but also as a trusted ally. Thus, for instance in 1328, in cooperation with Mongol forces, Ivan occupied Tver, where a year earlier, an envoy of the khan had been killed in a local revolt. In 1331 Ozbeg granted Ivan the "grand principality over all the land of Rus'" and for the remainder of his life - he died in 1340 - relations between him and the khan remained harmonious. It has been argued that Ivan was but a willing tool in the hands of Ozbeg, a mere ????utor of his policies, which were formulated without regard to Russian interests. Yet, although Ozbeg's dominance cannot be denied, the collaboration in fact profited both. Ivan's pro-Tatar stance did not hamper him in dealing with internal affairs, while in matters of foreign policy the alliance was an effective bulwark against Lithuanian expansion, as unwelcome to the Russians as to the Tatars. The same policy served well also their respective successors, namely Janibeg and Simeon (1340-1353). Although during all this period no major decision - whether about war or peace, alliance or succession could be made without the consent of the Horde, through this very involvement in the internecine squabbles of the Russian principalities the Mongols descended, as it were, to the level of their subjects. Their power was neither all-pervasive nor absolute in the north-eastern parts of Russia, where, most of the time, it operated through the intermediary of Russian clients. The fact is that in those regions, so distant from the steppe, the balance of military forces no longer tilted in favor of the former conquerors. A coalition of the princes could have ended Mongol domination; that this did not come about shows that the "Tatar yoke" did not weigh too heavily on their shoulders. Early in his reign Ivan II (1353-1359) trod a path leading away from the overlordship of the Golden Horde; but, partly under pressure from his own boyars, he had to mend his ways, and at the death of Jarubeg he set off to Sarai to ask the new khan Berdi Beg (1357-1359) for the edict (yarlzq) confirming this position.

The assassination of Berdi Beg heralded a protracted period of internal strife resulting in the loss of the Golden Horde's dominant position in Eastern Europe. Though it was not the first occasion on which the Mongols were worsted in combat, yet the Battle of Kulikovo, fought on December 8, 1380, in which the Tatar forces of general Mamai suffered a crushing defeat at the hands of Dmitry Ivanovich Donskoy, grand prince of Moscow and Vladimir, has always been considered a turning point in Russo-Tatar relations: it finally shattered belief in Tatar invincibility. Though in 1382 a campaign led by Toktamish, de facto ruler of the Golden Horde (1380-1395), resulted in the occupation of Moscow and the reassertion of the Horde's supremacy, the stigma of Kulikovo could not be erased.

Toktamish was the last major figure in the history of the Golden Horde, a man of considerable political vision who, however, made the fatal mistake of antagonizing his former protector and mentor Tamerlane (Timur). It was with the help of Timur that in 1377 Toktamish, himself of uncertain origin, became head of the White (sometimes called Blue) Horde,59 formerly the apanage of Batu's older brother Orda, which occupied the steppes east of the Ural, now part of Kazakhstan. Taking advantage of the situation created at Kulikovo, Toktamish united the forces of the White and Golden Hordes. Intoxicated by these successes, he felt ready to take on Timur himself, and in 1387, and again in 1388, he invaded territories over which Timur claimed sovereignty. A long campaign led by Timur 60 brought him victory in June 1391, but did not put an end to Toktamish' restless aggressiveness. In 1394 he made a renewed attempt to penetrate south of the Caucasus, provoking thereby a retaliatory campaign by Timur which led to the destruction first of the prosperous Genoese trade settlements of the Crimea, then, at the end of 1395, of Sarai. Here, as in the case of Kiev, the horror of the massacre and devastation has been brought to life by excavation of the remains of that once flourishing city. Mutilated skeletons testify to the desperate and vain attempts of the inhabitants, men, women, and children, to escape a fate willed by Timur, the annihilation of the city and all its inhabitants. No one was left to bury the dead.61

Even the hardened, by no means squeamish historian must pause here to ponder over the motives of Timur's expeditions to the north, which defy conventional explanation. What were they expected to achieve? According to the Arab historian al-`Umari, a contemporary of Ozbeg, the huge land of the Golden Horde was rich in steppes and poor in cities. Compared with that of the il-khans, so reports al-`Umari, the income of the khans of the Golden Horde was modest. Because of the lack of armaments and the poor quality of horses unsuitable for use on a mountainous terrain, the country was ill prepared for war." The campaigns against Toktamish involved strategic and logistic planning on a grand scale and months-long painful crossing of deserts and marshes which tested to their utmost even his hardened troops. What were the bonds of solidarity holding these men together? The sheer joy in fighting - so difficult to imagine by members of a society such as that of the U.S.A. where "counseling" is needed to those who witnessed an armed bank-robbery - cannot be experienced in the painful march towards a distant goal that promises no booty other than herds of poor quality that die for lack of adequate pasture or, at best, some gold or silver objects, difficult to distribute among the mass of warriors. But Sarai 63 was no Baghdad or Delhi with their accumulated treasures, and Timur's army could have found lands with booty more rewarding than that offered by the Kipchak steppe. There remained the women to be had but what they could offer was not particular to the place and could have been found at locations less distant.

Thus, once again the historian is faced with the recurrent phenomenon of conquest for conquest's sake and must take cognizance of the impalpable but pervasive power of a charismatic leader to rally his men towards a march into never-never land. Of all the military undertakings of the so-called Mongol era none seems to be as pointless as Timur's campaigns against Toktamish who himself escaped from Timur's grasp to lose his life in Siberia in 1406.

In objective terms the destruction of its capital was not a fatal blow to the Golden Horde, for Sarai had never played a role as important as had Baghdad or Delhi. Nevertheless, the event is an important station in the gradual weakening of the Golden Horde, bereft of the cohesive forces which, for a century and a half, had held together its motley population. The insignificant khans who followed on Toktamish - we barely know their names - could not establish a central authority strong enough either to set a course of policy or to command an army capable of its ????ution. Until his death in 1419, only Edigu represented a force to be reckoned with by the Russians, Lithuanians, or Poles; and his actions were mostly in response to events and reveal no central concept. By that time Tatar units were quite often mere mercenary forces in the service of alien rulers who knew, or pretended to know, the aims they were pursuing. The Golden Horde then disintegrated because it no longer had a raison d'etre, because it could no longer be cogently argued that it was in the best interest of the population to live in one state. The successor khanates of Kazan, Astrakhan and the Crimea seemed to respond better to local interests, or, at least, it was hoped that they would do so. Though they may have continued to exact tribute from the Russian principalities,64 in practical terms the khans no longer viewed them as parts of their own polities. Whatever their internal dissensions may have been, language and religion were powerful bonds between the Russian princes and set them, as well as their Slavic subjects, apart from their Tatar overlords whose power-base, in language as well as in religion, was more divided. Let us, briefly, examine these two factors.

While there can be no doubt that in the mid-13th century at the court of Batu the Mongol language was in general use, perhaps because of the pr????ent general illiteracy, almost no Mongol texts written on the territory of the Golden Horde have survived.65 Most of the edicts issued by the Golden Horde are extant only in a Russian translation. According to Grigor'ev, their originals were written in Mongol, then translated into Cuman. The final Russian version is but a rough translation made of the Turkic text.66 Clearly, depending not only on the period but also on the occasion, the Horde's chancellery used either Turkic or Mongol.67 The original of a decree issued on March 20, 1314 by Ozbeg, known only through a contemporary Latin translation was certainly written in Mongol.68 The existence of Arabic-Mongol and Persian-Mongol dictionaries dating from the middle of the 14th century and prepared for the use of the Mamluks in Egypt suggests that there was a practical need for such works in the chancelleries handling correspondence with the Golden Horde. It is thus reasonable to conclude that letters received by the Mamluks - if not also written by them - must have been in Mongol. There are no data available to show when the use of the Kipchak Turkic language, Cuman, came to be general in the administration. In 1393 Toktamish used Turkic to write to Ladislas II king of Poland. The language called "Persian" in the Venetian registers may have been Turkic as in the case of a treaty, now lost, concluded in 1333 between the Golden Horde and Venice.69

It is certain that Cuman was the vernacular most used; its ubiquity is vouchsafed by many travelers, and the Italian trader Pegolotti in his commercial handbook, usually referred to as La pratica della mercatura, written in the middle of the 14th century, urged his fellow merchants to engage servants who knew the Cuman language well.70

The major document of the Cuman language as used on the territory of the Golden Horde is the so-called Codex Cumanicus 71 a compilation of various texts originally written in the first decade of the 14th century by Italians and Germans. Its texts, comprising a Latin-Persian-Cuman and a Cuman-German dictionary, translations into Cuman of Latin hymns and passages of the Bible, short sermons, and riddles, were compiled in one of the Italian colonies on the northern shores of the Black Sea for essentially practical aims. And this is the crucial point that needs to be made here for our present purpose: while the Cuman language as reflected by the texts contains a number of Mongol loan words, the compilers thought that in the given circumstances. Cuman and even Persian were more useful to the users of the work than would have been Mongol. Though under Mongol suzerainty, the language of the rulers held no dominion in the Crimea. Thus, in fact, in the immense territory of the Golden Horde, three languages vied for primacy: Mongol, Russian, and Turkic Cuman, each of with its own ????. But the battle among them was fought with unequal means. The Slavic Russian and Turkic Cuman had the backing of an important literary corpus put down in the Cyrillic or Arabic ????s respectively. These, in their turn, constituted a link with one of the two great contending civilizations of Western Eurasia, namely the worlds of Christianity and Islam, both literate. The Mongol ????, recently borrowed from the Turkic Uighurs, had no hinterland to sustain it. The thin Mongol stratum had no literary and, hence, historico-political tradition on which to rely.

The divide between the growing number of Turkic-speaking Mongols, whom for the sake of distinction we may call Tatars, and Slavs ran not only along a linguistic line; difference in religion was an equally potent factor of division. Here again there was a three-way, perhaps even four-way conflict. With the passing of time, Mongol religious beliefs which in the mid-13th century were certainly shared by the leading stratum, received no sustenance from an organized clergy and had no sacred books to which they could turn for guidance. There was no Mongol counterpart to the Bible or the Koran.

Under Ozbeg and Janibeg (1342-1357), Islam, which among the 'Irks had deep roots going back into pre-Mongol times, gained general acceptance, though its adherents remained tolerant of other beliefs. No one has summed up better their attitude than Fr. Iohanca according to whom the Tatars could not care less to what religion someone belongs as long as he performs the required services, pays tributes and taxes and satisfies his military obligations according to their laws.72

The Russian Orthodox church had a twofold advantage over the Roman Catholics. Generally speaking, she was more accommodating towards the secular powers than was Rome, but, more importantly, she was already securely implanted in a large part of the territories which came under Mongol rule. An Orthodox bishopric was established in Sarai as early as 1261. On the other hand Rome had to send "missions" into these distant lands. A decree, issued probably by Mongke Temur, allowing the Franciscans to proselytize, was renewed in 1314 by Ozbeg, a convert to Islam.73 Under him, as under his predecessors, Catholic missionaries could work, with some ups and downs, unhampered.74 Some of them, such as the aforementioned Hungarian friar Johanca, "following the camps of the Mongols,"75 preached the Gospel to the far-away Bashkirs. This western presence further north in the forest zone - in partes infidelium aquilonares as the Franciscans called it - is worth noting because it seems to have disappeared without a trace.76 Yet we know that a certain Estokis, ruler of Bashkiria, together with his whole family, was baptized by a German Franciscan called Henry.77 In 1320 Johanca asked his superiors to send there other missionaries, preferably Englishmen, Germans or Hungarians, who were more gifted than the French or the Italians seemed to be in learning foreign languages. Religious clashes did occur, mostly caused by Christian provocation, as was the case in 1334 with the Hungarian Franciscan Stephen, who had apostatized to Islam but changed his mind, and in the mosque of Sarai solemnly professed his Christian faith and, for it, suffered martyrdom. Yet Christianity, in whatever form, could not resist the advance of Islam, a process of immense historical consequences. In Berthold Spuler's perceptive judgment: "The triumph of Islam among the Kipchak Mongols had effects which were the reverse of those brought about by its triumph among the Mongols of Persia; for whereas the latter identified themselves religiously with their subjects, the former set up between their [Slavic] subjects and themselves a definite religious barrier . ... By choosing . . . to guide their own nation to Islam, they precluded any possibility of its russification."78

The Mongols were the last Inner Asian people to deposit a new population layer on the fringes of Europe. Though they brought no lasting contribution to Western civilization, their historical role was considerable. History shows that the peoples of Central Eurasia were attracted by what they conceived to be the focus of whatever sedentary civilization lay within the purview of their interest. Hsi–ung-nu, Juan-juan and Turks thus looked towards China, Huns and Avars towards Byzantium. For the Mongols in the west, the Promised Land seems to have been Iran. To be sure, they campaigned in Poland or Hungary and sent haughty messages to the pope and to other western potentates; but their effort was directed, again and again, to reaching out south through the Caucasus into Anatolia and Iran. From Berke to Toktamish the rulers of the Golden Horde doggedly and vainly pursued the dream of bringing these southern lands under their domain. Their links with Mongolia, their own distant homeland, made them aware of the immensity of the territory which could be centrally controlled. The riches of Iran, Iraq, Mesopotamia and Asia Minor were more immediate and more tempting than those of little-known, fragmented Europe. Their eyes were turned towards the world of Islam which they joined at a time when it had entered a period of stagnation. They left their vision as a legacy to the rulers of Russia, who became more intent on expansion to the south and to the east than to the west. We cannot enter here into the ongoing debate about which, if any, contributions the Mongols made to later Russian civilization. Yet I venture to suggest that a balanced assessment of the relative importance of Europe and Asia, and, in conjunction with it, a racial tolerance greater than that of the Europeans, may qualify as traces of a Mongol world-view. To allow some raised eyebrows to be lowered to relax on reading this statement, let me cite a view expressed more than a century ago by George N. (later Lord) Curzon, the great antagonist of Russian imperialism in Central Asia, in a book dedicated "To the great army of Russophobes who mislead others, and Russophiles whom others mislead." It reads as follows:79

The Russian fraternizes in the true sense of the word. He is guiltless of that air of conscious superiority and gloomy hauteur which does more to inflame animosity than cruelty may have done to kindle it ... His own unconquerable carelessness renders it easy for him to adopt a laissez-faire attitude towards others, and the tolerance with which he has treated the religious practices, the social customs, and the local prejudices of his Asiatic fellow-subjects is less the outcome of diplomatic calculation than it is of ingrained nonchalance.

Although language and religion were important for defining and maintaining a national identity, as is so often the case, it was the fertility rate which decided the outcome of the struggle for national survival. Within a few generations the conquering Mongols were absorbed by the conquered Turkic populations. The process of assimilation was so fast that Al-'Umari 80 could already state in his time that Mongols and Kipchaks seemed to belong to the same race. He thus echoed the view ascribed to the Mongols themselves by Ibn al-Athir a century earlier.81 The Mongols eliminated through natural process, there remained but two contenders, Slavs and Turks, for the heritage of the Golden Horde. The contest continues.