The “Internal” Ethnic Communities in Khazaria 231
In their descriptions of the Khazar state, historians usually highlight the
semi-autonomous status of the subjugated regions, to some of which the cen-
tral authorities sent out their own representatives (tuduns). Also often cited
are the accounts from Eastern sources of the ninth and tenth centuries that tell
of the Khazar khagan’s harem, which consisted of 25 women, all of whom were
daughters of rulers of various subjugated to Khazaria regions and states. There
is also talk of some sort of “inner Khazaria”, ruled directly by the khagan, that
was actually his “domain”.29
This issue also concerns the interpretation of the account of the Byzantine
emperor Constantine Porphyrogenitus regarding Khazaria’s nine “climates”.
According to A. Novosel’tsev, the “climates” correspond to the administra-
tive-territorial units of Khazaria, which were equivalent to the regions and
tributary peoples, named by Joseph, all of whom were governed by tuduns.30
Firstly, let me point out that these “climates” have no connection to the tribes,
named by Joseph, but rather refer to a specific geographical region, probably
in the area of the Taman Peninsula, the Crimea and/or the Don Valley (i.e. the
western half of Khazaria). These lands are the ones where the Khazar ruler
does not name any tributary tribes or peoples, but only urban (settlement)
centers. On the other hand, it should always be borne in mind that “many
states and peoples were dependent on the khaganate; they were in different
environmental conditions and stood at different stages of development: from
kingdoms and tribes in the North Caucasus area to the distant Mari on the
Viatka, from the Eastern Slavs along the Dnieper to the Alans near Khwarezm.
Naturally, the level of subjugation of the vassals varied”.31 It cannot be assumed
29 See for instance Artamonov 1962, 189–191, 272, 382, 406, and 408–409; Dunlop 1967, 109
and 224 (see Ibn Fadlan’s text on p. 118); Magomedov 1994, 103; Noonan 2001, 76–77, 81,
and 85; Novosel’tsev 1990, 108–109 and 143–144; Pletneva 1976, 22–23, 34, 48, and 57–59;
Pletneva 2002, 117, where “the domain” of the Khazar khagan is with an area of 650 000 sq.
km! Pritsak 1981b, no. 11, 263–265; Romashov 2002–2003, 83 and 86–90; Romashov 2004,
218; Zakhoder 1962, 144–145.
30 Novosel’tsev 1990, 108–109.
31 Romashov 2002–2003, 83. According to Howard-Johnston 2007, 192, “the khaganate had
several management strategies to hand. Two, direct rule and investiture of a client-ruler
belonging to a detachment of the ruling Khazar stratum of the empire, are revealed by
the DAI (De administrando imperio, Constantine Porphyrogenitus—Author’s note) [.. .]
Thus a rather looser form of management is suggested for the Volga Bulgars by... Ibn
Fadlan. Other systems, including the appointment of Khazar governors and recognition
of local tribal chiefs, may be envisaged for the wide range of sedentary subject peoples.
Arrangements are likely to have varied according to the size of a subject people, their level
of institutional development and the geographical disposition of their territories”.