226 CHAPTER 5
system, which was typical for the Turkic khaganates. In accordance with it,
the main lands were divided into lots, which were managed by members of
the same family (the sons, brothers or other relatives of the khagan). These
lots were not held for life and had a hierarchical status. In case of the supreme
ruler’s death, he was succeeded by the next in rank (the eldest brother, if there
was one, if not—by the oldest one of the ruler’s sons or nephews). The transi-
tion of power in the separate land lots was done in the same way.13 In Eastern
Europe after the ninth century, such a system existed in Kievan Rus’.14
With regard to Khazaria, the Khazar ruler Joseph was adamant that the
power passed down from father to son. This is also confirmed by John Exarch,15
who adds: “and from brother to brother”. What he probably meant was that in
the absence of a direct heir (son), the power remained in the family. Such an
assumption is also possible in light of the dynastic line, described by Joseph,
which contained one instance of a throne inheritance by a brother. But those
accounts are from the ninth and tenth centuries.
Although it is highly unlikely, it nevertheless could be presumed that an
appanage system existed during the pagan period of the Khazar state. This is
also backed up by the available information on the structure of Kubrat’s Great
Bulgaria, where various regions were governed by the sons of the Bulgar ruler.16
During the next period, however, neither Danube Bulgaria nor Volga Bulgaria
used this system.
When comparing the state structure of Danube Bulgaria with that of
Khazaria, both with common roots in the steppe empires tradition, one major
difference should be highlighted, since it had an impact on both the ideologi-
13 Gumilev 2004a, 65–66; Golden 1980, 40–41; a similar system was used by the Huns: in their
state, power was originally passed down from father to son (Kradin 2001a, 56–57, 145, and
227–229; see also Khazanov 1975, 195–199). The transition of power from father to son and
from uncle to nephew was not compatible in the steppe empires. Therefore, the switch
from one practice to the other was not an uncommon phenomenon. Both practices were
based on the belief that sovereignty belonged to the whole ruling family, and not only to
one of its branches (Khazanov 1975, 195–196).
14 See for instance Shepard and Franklin 2000, 223–226, 268–282, and 355–360; Golden 2003,
no. 1, 39.
15 Ioan Ekzarkh. Shestodnev, in Kochev 2000, 159; for the so-called patrimonial state, see
Stepanov 1999a, 63–64. In front of St. Cyril the Philosopher the Khazars explicitly stated
their custom to place on the throne rulers from one family (Prostranno zhitie na Kiril, in
Dinekov 1963, 27).
16 On this issue, see Iordanov 1996b and 1997. I am inclined to think, however, that the his-
torian’s theory that this system was also used in Danube Bulgaria is not quite acceptable.
See also Tortika 2006a, 49–50.