tion"--with the two distinguished mainly by the strategies' respective targets. The target group for a party that follows a restrictive strategy is either another party or some social group that threatens to become an opposition party or force, with the implication that the target group is "outside" the government. The target for a party that follows a strategy of subverting the political system is probably the government, with the obvious implication that the party in this case is itself outside the government.
Because of the apparent bipolarity in these conceptualizations of party strategy, it might seem more parsimonious to treat the two concepts as one, perhaps incorporating them in a concept of "nonelectoral orientations," as an alternate to the strategy of "open competition in the electoral process." In such a combined concept, the target group of the strategy would be fixed simply by determining the governmental status of the party; a party which is high in governmental status but which follows a nonelectoral strategy would be assumed to be restricting competition from opposition parties, and one low in governmental status would be assumed to be engaging in subverting the political system. However, by making these assumptions, given our present state of knowledge, we may miss identifying some interesting deviant cases. Therefore, we treat "restricting party competition" and "subverting the political system" as separate strategies, and we stand ready for education otherwise by data analysis.
Operational Definition. A party's reliance on a strategy of subverting the political system is scored in accordance with the following weights. This scoring is done in conjunction with variables 6.00 and 6.10 to ensure that the sum of the orientation scores equals the arbitrary value of 4.
Coding Results. Tables 7.3a and 7.3b contain the results of our coding for BV620, "subverting the political system." With all the parties coded, we can report that approximately 80 percent of the parties during the 1950s shunned a strategy of subversion completely, with the aversion somewhat greater after 1956. However, this still leaves some 20 percent of the parties oriented in some degree toward stimulating system instability, although less than 5 percent concentrated on a subversive strategy. In many cases of these "antisystem" parties, impartial discussions of their goal orientations were absent, requiring coders to estimate scores for BV620 on the basis of scanty or suspect documentation. Hence, we find a significant -.29 correlation between BV620 and AC620, indicating that the higher variable codes drew lower AC codes.
These variables embrace specific party activities that come to mind as direct tactics under a strategy of subverting the political system. They are
While variables 6.11 through 6.16 and variables 6.21 through 6.26 refer to nonelectoral activities, the two sets differ in two respects. First, they differ with respect to the implied targets, with the former set of activities presumably aimed at nongovernmental groups and the latter at governmental groups or the government itself. Second, they differ with respect to the scope of the activities, with the former more narrowly aimed at political activities and the latter spilling over into various aspects of social life. This second source of distinction in nonelectoral activities may be due largely to who controls the state's capacity for repression, permitting it to be rather selective in exerting force on its opponents. A party outside the government usually operates with an inferior capacity for exerting force and may not be able to discriminate among sectors of life it chooses to dis-