Before examining these relations in their earliest stages, it remains to give a brief account of the third people in the 'second Bulgarian Empire,' the Cumans, always mentioned by the sources with terror as providing the main strength of Ioannitsa's armies, but apparently never accorded any political status within the state or mentioned in the title of its rulers. Yet, as we have seen, the Cumans had been closely associated with the Vlachs ever since the days of the uprising of 1185, when Isaac Angelus had driven Peter and Asen across the Danube, whence they returned with their Cuman auxiliaries. 
A people whose language was clearly Turkish, the Cumans are none the less distinguished from other Turkish tribes by the accounts of their appearance given by the sources. Strikingly handsome physically, they had blond hair and blue eyes: indeed, it is now generally agreed that the names Polovtsy, given them by the Russians, and Falven, sometimes given them by the Germans, come from words meaning 'yellow.' It is therefore conjectured that ethnically the Cumans may not originally have been Turks. Scholars explain the multiplicity of names by which they have been called at various times — including Kipchak and its variants — as the result of the amalgamation of various tribal groups into the mass of the Cumans during the early Middle Ages in northern central Asia. This was their original home, and it was there in the tenth century that the tribe of the Kipchaks won supreme power over the other tribes. The expansion of the Kitai state forced the Cumans south in the early eleventh century, and thereafter west; so that between 1050 and 1080 Cuman tribes became supreme over the plains of South Russia and Rumania, and as far west as the Carpathians, the Danube, and the Balkan mountains. In the great area between these western
boundaries and Lake Balkhach, the Tian-Shan range, and the Altai and upper and middle courses of the Irtysh in the east — bounded by the forest zone in the north and by the north coast of the Black Sea in the south — there was no central political state formation.
Rasovskii, however, distinguishes five separate independent Cuman groups: the central Asiatic, the Volga-Yayik (or Ural), the Donets-Don (between the Volga and the Dnieper), the lower course of the Dnieper, and the Danube. It is with the last only that we have to do. The Cumans remained nomads until the Mongol invasion of the mid-thirteenth century. They seem never to have attempted the establishment of a territorial state or to have taken the other steps which might have led them to adopt a sedentary life. They assisted the Byzantine Empire, in the late eleventh century, in the destruction of their fellow-Turks, the Pechenegs; they assisted the Hungarians in various campaigns against Byzantium; and they played a part in the internal struggle of Kievan Russia in the eleventh century as well as in the wars of the Kings of Georgia against Persians, Armenians, and Seljuks. Throughout they remained a steppe-people, without cities or towns, living in felt tents on milk, cheese, and meat, as Robert of Clari says.
Indeed, Robert's description of them is perhaps the most valuable we possess, giving as it does a complete picture, even to such aspects of their religion as impressed a western observer. It deserves quotation here:
When he (Ioannitsa) was lord over them (the high men of Vlachia), he went to the Cumans and he wrought so with one and with another that he became their friend and they were all in his service and he was just like their lord. Now Cumania is a land bordering upon Vlachia, and I will tell you what kind of people the Cumans are. They are a savage people, who neither plow nor sow, and they have neither huts nor houses, but they have heavy tents made of felt in which they shelter themselves, and they live on milk and cheese and flesh. In the summer there are so many flies and gnats that they scarcely dare come out of their tents and sally forth from their country when they want to make a raid. Now we will tell you what they do. Each one has at least ten or twelve horses, and they have them so well-trained that they follow them wherever they want to take them, and they mount first on one and then on another. When they are on a raid, each horse has a bag hung on his nose, in which his fodder is put, and he feeds as he follows his master, and they do not stop going by night or by day. And they ride so hard that they cover in one day and one night fully six days' journey or seven or eight. And while they are on the way they will not seize anything or carry it along, before their return, but when they are returning, then they seize plunder and make captives and take anything they can get. Nor do they go armed, except that they wear a garment of sheepskin and carry bows and arrows. They do not worship anything except the first animal encountered in the morning, and the one who encounters it worships it all day, whatever animal it may be. Now John the Vlach had these Cumans in his service, and he used to come every year to raid the Emperor's lands even up to Constantinople, and the Emperor was not strong enough to defend himself against him. 201
More valuable data on the Cumans as they appeared to the Latins of Constantinople after they had got to know them better is supplied by Joinville, famous biographer of St Louis. While St Louis was fortifying Caesarea in Palestine in the year 1250, Philippe de Toucy, who had been bailli in Constantinople, visited the camp, and told St Louis about the Cumans, with whom the Latin Emperor Baldwin II was in alliance. This alliance, Philippe reported, had been sealed by a blood-mingling ceremony: the Latin Emperor and the Cuman king and their leading followers were all bled into a bowl of silver; wine and water were added, and both parties drank of the mixture, making them brothers. Then a dog was made to pass between the two parties, and the Cumans thereafter cut him to pieces; the Latins followed suit with another dog. This was to symbolize the fate of anyone who should betray the alliance. Philippe de Toucy also told St Louis about Cuman burial customs: they had buried one of their great men fully clothed and seated in a chair, and had put his horse and his 'best serjeant' alive into the grave with him. The serjeant was given gold and silver by the other Cumans to take into the other world, and to keep safe for them; and also a letter to be delivered in the other world to the first king of the Cumans, in which the present king testified to the Serjeant's good character. 
Although Rasovskii, the leading authority on the Cumans, objects to
the comment, reiterated by the Byzantine sources, that these people were
uncivilized, and although it is true that nomad civilization is not to
be belittled, at least with regard to the discipline and cohesion which
permitted the conquest of so vast a territory, it must be said that the
Cumans represent a degree of political maturity far less advanced even
than that attained by the Vlachs and Bulgarians. It is perhaps not to be
wondered at, then, that we find them as auxiliaries in the armies of the
Asen brothers, but not as participants in the political life of the country.
Their importance as an element in the future struggle between Ioannitsa
and the Latins is very great; but it is almost exclusively military.
[Back to Main Page]
85. The splendid series of articles by D. A. Rasovskii has unfortunately, so far as I know, been carried down only to the year 1170; so that it contains little more than a mention of the activities of the Cumans after they had crossed the Danube at the invitation of Peter and Asen. I depend upon it, however, for the brief sketch of the Cuman background in the text. D. A. Rasovskii, 'Polovtsy,' Seminarium Kondakovianum, VII (1935), pp. 245-262; VIII (1936), 161-182; IX (1937), 71-85; X (1938), 155-178 (see 156-160 for the Cumans and the Asen family); XI (Belgrade, 1940), 86-128. So far as I am aware, the only significant work which Rasovskii's full bibliography omits is the very scarce book of the Rumanian Uniat priest, Ioan Ferentx Cumanii si Episcopia Lor (Blaj: Tipografia Seminarului Teologii Gr.-Catolic, n.d., but preface dated 1931), pp. 152, of which I possess a copy. The sections of interest here — not superseded because not yet reached by Rasovskii — are pp. 46-56, 'Cumanii si Infiintarea imparatiei romano-bulgare' and 'Cumanii si consolidarea imperiului romano-bulgar.' Ferent necessarily relies on Nicetas throughout, and on the treatment given Nicetas' passages by G. Murnu, Vlahia Mare (Bucharest, 1913), not accessible to me. G. Moravcsik, op. cit., note 2 above, supplies (p. 48) references to an article inaccessible to me: D. Rasovskii, 'Rol' Polovtsev v voinach Asyenei s Vizantiiskoi i Latinskoi Imperiami v 1186-1207 godach,' Spisanie na Bulgarskata Akademiya na Naukite, LVIII (1939), 203-211; see also G. Ostrogorsky and S. Radoychich in Byzantinoslavica, IX (1947), 140. A. Bruce Boswell, 'The Kipchak Turks,' The Slavonic Review, VI (1927), 68-85, has some errors, but is not a bad introduction to the subject. Before Rasovskii the standard work was J. Marquardt, 'Uber das Volkstum der Kumanen,' Abhandlungen der K. Gesellschaft der Wissenschaften zu Gottingen, Phil. -Hist.- Klasse Neue Folge, XIII (1914), 25-236, an immensely learned work, concerning itself chiefly with the philological evidence for the period of the origin of the Cumans. P. Pelliot, 'A propos des Coumains,' Journal Asiatique, 11 serie, XV (1920), 125-185, pointed out that Marquardt did not have access to many of the most valuable Chinese sources, and so reached some conclusions which should be modified. The article of S. Salaville, 'Les Comans,' Échos d'Orient, XVII (1914), 193-209, contains some episodes from Assyrian history, with which the Cumans are, perhaps somewhat fantastically connected; but for the period of the Middle Ages it is interesting, since it includes some pages on the bishopric of Milcov, the short-lived thirteenth-century Roman Catholic Cuman bishopric, with which Ferent's work is particularly concerned. Salaville (p. 204, note 7) cites C. Auner, 'Episcopia Milcoviei,' Revista Catolica, I (1912), 533-551, and in (1914), 60-80, not accessible to me, but which is presumably superseded by Ferent. Reference may also be made to Rasovskii's interesting general article, 'Les Comans et Byzance,' Izvestiya na bulgarskiya Archeologicheski Institut IX (1935), 346-354. The chapters in F. Uspenskii, Obrazovanie vtorago Bolgarskago tsarstva (Odessa, 1879), pp. 75-88, and in R. Roesler, Romanische Studien (Leipzig, 1871), pp. 328 ff., have been superseded by Rasovskii. For the Cuman language the most famous monument is the so-called Codex Cumanicus, which belonged to Petrarch, and is now in the library of St Mark at Venice (ed. G. Kuun, Budapest, 1888) and more recently K. Grønbech, Monumenta Linguarum Asiae Minoris, I (Copenhagen, 1936).
86. Robert de Clari, La Conquête de Constantinople (ed. P. Lauer, Paris, 1924), pp. 63-64; tr., E. H. McNeal (New York, 1936), pp. 87-88. The story that the Cumans worship the first animal they see on any given day is also told of the 'Mordwit Tartar' in the sixteenth century by the celebrated English envoy to Russia, Giles Fletcher, Of the Russe Common Wealth (ed. E. S. Bond, London: Hakluyt Society, 1856), p. 96. Cf. Eustathius’ description cited at the head of this article.
87. J. de Joinville, Histoire de Saint Louis (ed. N. de Wailly, Paris, 1874), pp. 270-274. For the Latin-Cuman alliance see also Alberic of Trois Fontaines, Chronica, MGH, SS, XXIII, 947, 949, where more details about Cuman burial customs are found. For the practice among Turkic peoples of killing dogs to solemnize treaties, see V. Grumel, 'Sur les coutumes des anciens Bulgares dans la conclusion des traites,' Izvestiya na Istoricheskoto Druzheslvo v Sofiya, XIV-XV (1937), 82-92, where an alleged instance of this practice by the Bulgarians after their conversion, is shown by Grumel to be attributable rather to the Pechenegs. This is challenged by Yu. Trifonov, 'Kum vuprosa za Vizantiisko-Biilgarski dogovori s ezicheski obredi,' Izvestiya na Bulgarshiya Archeologicheski Institut XI, 2 (1937), 263-292.