Klaus Krajewski: Political integration and violent conflict
For the development of such a model, let us turn back to the contingency-table
(Fig. 1.) and look at the societies coded as having a high degree of political integration.
42 out of 90 show a high degree of internal warfare, which is not predicted by the hy-
pothesis. If we consider the almost tautological character of the hypothesis, the result
is all the more surprising. Integration, meaning the joining of many small parts into
one whole, excludes by definition desintegrating elements like violent conflict man-
agement.
In my opinion the seeming rejection of the hypothesis can be related to two general
problems of cross-cultural comparison? which have not been solved in the studies dis-
cussed.
(1) No adequate indicators to measure the theoretical variable âpolitical integrationâ
have been used. So far the existence or lack of political authorities above the level of the
local community was taken as a measure. To know about the mere existence of pol-
Itical authorities does not tell us anything about the degree of integration. It depends
on the extent and concentration of power in political office and how it is handled. Fur-
ther it depends on the basis of legitimation of power and political officials as well as the
degree of support by members of the community. These factors have to be included in
Order to measure the politically integrating forces in a society.
(2) The second problem relates to the definition of the appropiate unit of analysis.
Thus far âsocietyâ mostly being understood as the group of ethnic identity has been
used as the unit of analysis. The distinction between âinternalâ and âexternalâ is deter-
mined by the boundaries of the ethnic group. Only for a fraction of the cases used tes-
ting the hypothesis, the ethnic group is identical with the autonomous political units.
he frequency of violent conflict management has thus been measured for a unit larger
than the political units to which integration is attributed. On these grounds I conclude,
that the hypothesis has not been operationalized correctly. Let me illustrate my argu-
ment with the help of the following example.
ethnic group
chief- chief- chief-
dom A dom B dom C
local communities
Fig. 2, Example for levels of political authority
Here (Fig. 2.) an ethnic group ist depicted, containing three independent chiefdoms.
ere is no political authority above the chiefdoms, which could avert violent conflict
Management. The situation looks different for conflicts between local communities
Lan
2
E ;
OT an discussion of problems of cross-cultural comparison see Schweizer 1978.