DISINTEGRATION OF THE EARLY CHaSON 89units. In the case of Kyonggi Province, for example, there were five regiments
(yang) for the whole province, one for the capital and four for the outlying areas.
In three cases, a battalion consisted of a single district, in others, two or three
districts. Even though the Central Regiment at the capital included the newly
formed Military Training Agency and may have had an irregular complement
of troops, we may presume that the troop strength for this province (five regi-
ments) was equivalent to a single division. The whole system was to operate on
the basis of the classical militia system: military organization would correspond
to the hierarchy of villages and districts, unit commanders would be recruited
from local talented men, during planting and harvest seasons the men would
alternate between work and rest according to a fixed schedule, and during slack
periods the men would assemble for training at the commander's (Taejang) head-
quarters.^90
The organization of sog'o battalions for P'yong'an Province provides some
more details about the distribution of troops by type in 1596. Each platoon had
a regular complement of three squads identified by one of the three types of sol-
diers (samsu): salsl! ("'killers" or close-combat sword, pike, and spearnlen), saSl!
(archers), andposll (musketeers). Of the thirty-six squads, however, only eight
were musket squads, while five were close-combat swordsmen, and twenty-three
were bowmen, indicating the persistence of traditional weaponry, but even the
Japanese had mixed musketeers with archers and swordsmen in their units dur-
ing the invasion. Despite the royal command to eliminate status distinctions,
some attempt was obviously made to keep some units free of slaves entirely.
Six piatoons or eighteen squads - half the units of one entire battalion - had no
persons of base status at all. Finally, an attempt was made at least, to combine
sog'o organization with the restoration of the chin 'g,1'(1I1 system.Y1
In certain respects, however, the new system departed from the chin'gwan
model. Although company commanders were selected from the mylin and i (sub-
districts and administrative villages, respectively, units hclow thc level of the
central government's district magistracy), the platoon leaders and company com-
manders were not selected from local people. Variations from thc unit quotas
in the Chi-hsiao hsin-shu were also allowed.^92
Yu Songnyong realized full well that his attempt to combine the chin' gwan
and sogo systems was an unavoidable compromise with the farmer-soldier mili-
tia ideal of antiquity. In the typical fashion of practical reformers he argued that
after the fall from grace that marked the destruction of the sage institutions of
Chou China, the best that one could hope for was to make adaptations to the
real and imperfect world hecause in the "later age" (husc) after the fall of the
Chou dynasty, soldiers hecame distinct from peasants. Since that time, the task
of the soldier was to "put forth effort to guard the people," and the task of the
peasant was to "put forth [i.e., provide] grain to feed the soldier," a remark that
appears to justify the system of rotating service and support taxpayers in use at
the time and later advocated by Yu Hyongwon.Y3
During the lmjin War, however, it became impossible to provide rations for