They divided the sky: elite factionalism factors during contentious politics in GDR, 1989
Acquaintance with the main reasons for the behavior of the elite under the pressure of mass discontent during the Peaceful Revolution of 1989 in the GDR. General characteristics of the consequences of Gorbachev's refusal from the Brezhnev doctrine.
| Рубрика | История и исторические личности |
| Вид | статья |
| Язык | английский |
| Дата добавления | 08.07.2021 |
| Размер файла | 1,9 M |
Отправить свою хорошую работу в базу знаний просто. Используйте форму, расположенную ниже
Студенты, аспиранты, молодые ученые, использующие базу знаний в своей учебе и работе, будут вам очень благодарны.
As regards the military, they proved to be easier to discern: we coded an actor as “military” when s\he has access to power thanks to his or her military ranks or high positions in law enforcement agencies.
Once again, the sources mentioned in the previous section provided the data for these variables. Ministers and other officials from economic, industrial, infrastructural, and other technocratic circles, members of respective CC committees and People's Chamber as well as those of local party cells were coded as technocrats. Other party functionaries, leaders of the GDR's civil society, journalists, and propagandists were regarded as party functionaries. Those bearing a military rank and simultaneously holding an office in the Ministry for Defense, Ministry for State Security (Stasi) or the National People's Army of GDR were coded as the military. Table 2 indicates the frequency distribution of CC SED members by their office type.
Table 2. Frequency distribution of the CC SED members by their office type
|
Office type |
Frequency |
Relative frequency, % |
|
|
Military |
15 |
7.0 |
|
|
Party functionaries |
115 |
53.7 |
|
|
Technocrats |
84 |
39.3 |
|
|
Total |
214 |
100.0 |
A-3. Party hierarchy level position. There is no universal “rule of thumb” to apply for distinction between the inner and outer circle of winning coalition. To tackle the issue, we followed Meyer (1991) and Thieme (2015) in treating the CC SED members who simultaneously had a seat at Politburo, the Secretariat of CC SED, State Council or Council of Ministers as belonging to the inner circle. Those who do not meet this criterion were coded as belonging to the outer circle. Such a methodological move was taken because the members of these highest party and state organs had access not only to a wider range of material as well as non-material goods and privileges but also enjoyed noticeable leverage within the party and could exert additional influence over some of the important decisions. Contrary to Meyer (1991), we argue that the distinction between circles of power should not simply lie between those in the Politburo and the Secretariat of CC, on one side, and "ordinary" CC members and candidates, on the other. Some of the latter were ministers or high government officials, who, though subordinate to the party, still got their share from the common "pie" inaccessible to the rank-and-file CC members.
As for other variables, the information on the offices held by the CC members was retrieved from the directories. A simple dummy variable hierarchy_lvl taking value of 1 when an individual belongs to the inner circle of the winning coalition and that of 2 when s\he belongs to the outer circle was introduced. Table 3 shows the frequency distribution of CC SED members by their position within the coalition hierarchy. The size of the inner circle appears to be 2.5 times smaller than of the outer circle.
Table 3. Frequency distribution of CC SED members by their hierarchy level position
|
Hierarchy position |
Frequency |
Relative frequency, % |
|
|
Inner circle («1») |
60 |
28.0 |
|
|
Outer circle («0») |
154 |
72.0 |
|
|
Total |
214 |
100.0 |
A-4. Repression-related office. As discussed, for some of the winning coalition members support of the hard-line policies means facing a decision of repression or toleration of the contender, which might significantly influence their rationales for defection. The variable repress_relat_office can, thus, was designed as a dummy variable taking value of “1” if a person indeed holds such an office and of “2” if s\he does not. Five types of offices were coded as related to take a decision whether to repress the citizenry in case of contentious politics: (1) the General Secretary of CC SED; (2) Minister for Interior; (3) Minister for State Security; (4) First Secretary of the Bezirk Committee of SED; (5) First Secretary of a district committee. In addition to these five categories, there was one Chief of the Bezirk Committee of the Ministry for State Security in the database, Horst
Felber, whose sanction was most likely required by local party leaders to decide whether to repress contentious actors. For him, the variable repress_relat_office also took the value of `1'. All in all, as Table 5 indicates, only 24 out of 214 CC SED members were occupying offices that we deem repression-related.
Table 5. Frequency distribution of CC SED members by their repress_relat_office values
|
An individual holds a repression-relate office |
Frequency |
Relative frequency, % |
|
|
Yes (`1') |
24 |
11.2 |
|
|
No (`0') |
190 |
88.8 |
|
|
Total |
214 |
100.0 |
B) Dependent variables
B-1. Liberalization support and the day when it first went public. Liberalis_support is a dummy variable that takes the value of "1" should an individual either publicly express his or her support for regime opening or advocate soft-line policies during the time period in question. Conversely, the variable takes the value of "o" if an individual remained silence on this matter or acted as a hard-liner by calling for repressions. Meeting one of the three conditions is sufficient to for liberalis_support to take the value of 1: (1) a direct and verbal support of demonstrators or the reasons they went into the streets; (2) a call for the extension of or respect for human and civil rights by the state, party, and law enforcement agencies; (3) a call for lessening the role of SED in the GDR's political system. Any of these expressions not only indicates dissent within the general party line but also demonstrates to the contenders and bystanders the lack of homogeneity and, offering possible institutional allies for contentious actors.
The data for this variable was gathered from the three biggest East-German newspapers, Neues Deutschland, Berliner Zeitung, and Neue Zeit. They have been recently digitalized and given to the open-access by the Berlin State Library's project ZEFYS (ZEFYS Zeitungsinformationssystem). News messages from these newspapers were considered and looked through when mentioning a name of at least one of 214 winning coalition members registered in our database. Time frame settings allowed to include into a search engine only messages from between the 1st of October and the 17th of November of 1989 inclusively. In total, up to 4.000 messages have been processed. If a message met at least one of three conditions listed above, the variable liberalis_support for the individual mentioned there was coded as `1' and the variable firstlib_sup_date (from `the date of the first support of liberalization') took the value of the date of the message, e.g. `30.10.1989'. This feature is relevant for further discussion.
Table 6. Frequency distribution of CC SED members by their liberalization support
|
Liberalization support |
Frequency |
Relative frequency, % |
|
|
Yes ("1") |
54 |
25.2 |
|
|
No ("0") |
160 |
74.8 |
|
|
Total |
214 |
100.0 |
As shown in Table 6, 54 out of 214 CC SED members at least once publicly supported the opening of the regime, which was documented and re-translated by the state media of the GDR.
Figure 4 represents the frequency distribution of the GDR's winning coalition members by the dates of their first defection. The peaks of defection, reaching up to 9 messages per day, occurred on October 18-20 and on November 10-11. Each peak corresponds to one of two turning points in the GDR collapse history. On October 18 the Central Committee convened to take several key cadre decisions (e.g. to remove Erich Honecker and install Egon Krenz as the General Secretary of SED) and to publicly correct the party line. On November 8-10 another Central Committee meeting took place where entire Politburo resigned, strict travel regulations were withdrawn, and the new Politburo was elected. Eventually, even more CC SED members expressed dissatisfaction with the atrocities of the old regime while appealing, for instance, to the mass gatherings in the streets or to the interests of the people they represented (Hertle and Gerd-Rьdiger (eds.) 1997: 183, 342). Newspapers published speeches and official comments on the Central Committee's decisions during a couple of days afterward, which explains why the number of messages skyrockets to 9.
Figure 4. Frequency distribution of liberalization support messages produced by CC SED members by date.
Figure 4
It is revealing that no permanent or gradual increase in the frequency of vocal liberalization support is observed. This contradicts the bandwagon effect hypothesis (Lyons 2016: 168), which stipulates that members of the winning coalition should exit the coalition with accelerating frequency. At least in the GDR case, no bandwagon effect occurred during the elite factionalism.
To check for robustness of this finding, instruments of survival analysis available in SPSS were applied. Common in medicine and economics, the survival analysis helps to assess the risks of exit, that is, the probability that an individual could modify his or her state, for instance drops out, stops using company's services or, as befitted our case, defects from the regime. We looked at the cumulative survival function that gives the probability that an individual survives (i.e. stays within a hard-line camp of the winning coalition) after a certain time point. The switch in the value of liberalis_support from "0" to "1" (within a timeline computed from firstlib_support_date) served as an exit option indicator. An auxiliary variable counting the days from the beginning of die Wende (declared by Egon Krenz at the CC SED Congress on the October 18th,1989) till each individual's liberalization support date was computed.
Figure 5. Cumulative survival function for the sample of winning coalition members
Since the graph of cumulative survival function in the first-order approximation is a straight line (see Figure 5), the probability of survival, as well as the number of hardliners among winning coalition members, decreased linearly. This, once again, contradicts the predictions of the bandwagon effect. Instead of rejecting the bandwagon effect outright, we suggest two explanations of this anomaly.
To begin with, single-party regimes and their institutional framework have impressive cohesive capacity that precludes elite defection more effectively than less institutionalized regimes. Accordingly, the very logic of single-party regime's political survival not only restricts the opportunities and speed of elite factionalism, what is partly accounted for by Geddes (1999) and Ulfelder (2005), but also constrains the impact of the bandwagon effect, making it impossible or at least slowing it down.
Alternatively, our data collection method might be responsible for the signalized anomaly. The data were extracted from the regime's official printed media whose task was to disseminate official party position enunciated by state and party leaders, especially by CC SED members, rather than grant opportunities to opposition leaders to express their visions. Thus, even if factionalism rates in fact increased in the course of time, it might have been kept underrepresented by the authoritarian media.
In any case, lack of evidence for bandwagon effect during the GDR regime collapse is an important finding that must be double-checked, explained, and become the subject of further research.
Findings and discussion
We used statistical techniques, foremost chi-square tests and binomial regression on the collected data to test our hypotheses.
To recapture, our Hypothesis 1 assumes that technocrats and military men who do not hold a repression-related office tend to defect less often than party functionaries.
The hypothesis is corroborated in several ways. To begin with, no individual coded as military in the database did support liberalization and, therefore, never defected. In other words, the variable liberalis_support for the military always takes the value "0".
In order to explore the relationship between variables type and liberalis_support, we perform two chi-square tests in jamovi software, one for the whole population (see results in Table 7), the other being controlled for the values of repress_relat_ojfice, which is, with all those individuals having the value of “1” the from the sample (results in Table 8). The removal is due to the fact that ignorance of the repress_relat_ojfice variable might significantly distort the results since is the model expects to have the strongest impact on a player's calculations. Additionally, as long as all those holding a repression-related office are party functionaries and not technocrats, party functionaries' preponderance among soft-liners might turn out to be explained not by their office type but by the responsible repression-related position they occupied.
Table 7. Contingency table and chi-square test results for the variables type and liberalis_support for the total population
|
liberalis_support |
||||
|
Type |
“0“ |
“r |
Total |
|
|
Military |
15 |
0 |
15 |
|
|
party functionary |
73 |
42 |
115 |
|
|
Technocrat |
72 |
12 |
84 |
|
|
Total |
160 |
54 |
214 |
|
|
Value |
df |
P |
||
|
X2 |
18.2 |
2 |
< .001 |
|
|
N |
214 |
|||
|
Contingency coefficient |
0.280 |
|||
|
Cramer's V |
0.291 |
Both chi-square and Cramer's V values are lower when the control variable's impact is accounted for, which suggests that there is indeed some hazard for result skewing. Therefore, only the results of the second test, (see Table 8) are to be considered.
Table 8. Contingency table and chi-square test results for the variables type and liberalis_support with the accounting for the repress_relat_office "control" variable
|
liberalis_support |
||||
|
Type |
“0“ |
“r |
Total |
|
|
Military |
12 |
0 |
12 |
|
|
party functionary |
67 |
27 |
94 |
|
|
Technocrat |
72 |
12 |
84 |
|
|
Total |
151 |
39 |
190 |
|
|
Value |
Df |
P |
||
|
X2 |
8.98 |
2 |
0.011 |
|
|
N |
190 |
|||
|
Contingency coefficient |
0.212 |
|||
|
Cramer's V |
0.217 |
The association between the variables of interest is of moderate strength (Cramer's V equals 0.217) and statistically significant (p = 0.011). With some degree of certainty, we can now claim that technocrats and the military indeed tend to be hard-liners more often than party functionaries. The review of the GDR's printed media messages during October and November of 1989 gave us an impression that technocrats, despite die Wende proclaimed by then-new SED General Secretary Egon Krenz and the increasing intensity of contentious politics notwithstanding, behaved as if within the pre-Wende status quo.
Cramer's V value, though not high, cannot be ignored as long as our model presupposes co-causality. By co-causality we mean the simultaneous impact of several factors, such as an office type, holding a repression-related office, and age on the defection likelihood for each individual. If co-causality is indeed a matter, one predictor cannot explain all the cases, so that its share of influence (and, therefore, the association strength as expressed by Cramer's V) is never absolutely perfect and cannot reach the values of 0.9 or even of 0.75. We argue that even the Cramer's V of 0.217, thus, has to be taken seriously.
To check for exit trends, survival analysis techniques for firstlib_support_date variable were applied and cumulative survival function was again computed, now differentiating between the categories of winning coalition members (survival analysis was performed via Kaplan-Meier in SPSS). As Figure 6 illustrates, the probability of survival (i.e. the probability an individual stays in the coalition as opposed to defection) was higher for technocrats (the blue line on the graph marked as “2”) than for party functionaries (the green line marked as “3”), which means that the latter supported liberalization on average more often and earlier than the former. Besides, the mean survival time for the technocrats (approx. 18 days) is significantly longer than for party functionaries (11 days). As we conclude, this statistically corroborates the model prediction that party functionaries should support regime liberalization more vigorously than either technocrats or the military.
Still, the empirical results do not fully correspond with the model's predictions. 12 out of 84 technocrats turned out to be soft-liners without meeting the condition of contender's looming success. Apparently, the distinction between technocrats, the military, and the party functionaries does not allow to grasp more nuanced differences within these groups. If our assumption is correct and technocrats are less prone to liberalization support due to the uncertainty of their career paths under a new regime and because of the lack of motivation to change the already profitable status quo, what prompted every seventh of them to approve the regime's opening? One possible explanation is a different motivation structure that sets the technocratic defectors apart from the others. After all, both the workers of relevant (economic) ministries and "general directors" (CEOs) of state enterprises were categorized as technocrats in this study. The latter, however, might expect higher benefits after economic liberalization, which would create some additional incentives to hope for regime transformation. Przeworski also signalizes about the particular trajectory of "general directors" within the winning coalition (1991: 67-68). Another factor responsible for individual variations of motivation structure is a real-life probability that some technocrats might have had deeper and more dynamic links to the civil society, opposition groups, and their leaders. Network effects, however, were not accounted for due to the research design. All the model deficiencies notwithstanding, statistical results show a clear link between an individual's office type and his or her tendency to support regime liberalization, which corroborates Hypothesis 1.
Figure 6. Survival functions for winning coalition members differentiated by their office type
According to our Hypothesis 2, young party functionaries, unless they hold repression-related offices, tend to support regime liberalization more often than party veterans.
Since independent variables biological_age and institutional_age are interval ones and the dependent variable Hberalis_support is categorical, binomial regression techniques are required in order to test this hypothesis. Before the test, we excluded technocrats, the military, and those with the "1" value for repress_relat_ojfice from the population.
As Table 9 suggests, there is no strong positive impact of the variables in question on an individual's likelihood to support regime liberalization. First of all, a test with institutional_age does not yield any statistically significant results. The variable hardly predicts whether the elite member supports liberalization. The results of the test for the biological_age variable are also on the limits of statistical significance albeit indicating a weak negative association.
All in all, the proportion of cases when the biological_age variable predicted the value of Hberalis_support is extremely low (with R2 equals 0.111).
Table 9. Results of binomial regression for biological_age and institutional_age as independent variables and liberalis_support as the dependent variable.
|
Model Fit Measures |
|||||
|
Model |
Deviance |
AIC |
R2McF |
||
|
1 |
100 |
106 |
0.111 |
||
|
Model Coefficients |
|||||
|
Predictor |
Estimate |
SE |
Z |
P |
|
|
Intercept |
4.8254 |
1.8331 |
2.632 |
0.008 |
|
|
biological_age |
-0.0847 |
0.0528 |
-1.606 |
0.108 |
|
|
institutional_age |
-0.0147 |
0.0652 |
-0.225 |
0.822 |
Statistical results invited for a closer look at the data. Figure 7 plots the variable biological_age frequency distributions, first for those subjects whose Hberalis_support value is “0” (above), then for those where its value is “1” (below). Repres_relat_ojfice variable is controlled so that the objects with its value of “1” are excluded from the population. The distribution is skewed a little to the "younger" wing for soft-liners, while it is skewed toward its `older' wing for hard-liners. It is likely that the oldest winning coalition members tended to disapprove soft-line policies whereas the youngest coalition members had slighter chances to be hard-liners.
Although Hypothesis 2 can partly explain the bimodality of frequency distributions and a noticeable overrepresentation of younger party functionaries (biological age below 60) among soft-liners and of older party functionaries (biological age above 60) among hard-liners, we are inclined to conclude that our model lacks some explanatory value to cover all the cases and to provide meaningful differentiation between the individuals of the same age cohort opting for opposite choices. After all, the majority of CC SED members within hard-liners, as well as soft-liners, were of the average group age, which is around 60.
The statistical results offer only a partial and a dubiously robust empirical base to claim the veracity of Hypothesis 2. Young party functionaries indeed are somewhat more inclined to support liberalization than party veterans, but this tendency is only noticeable for individuals within the population with extreme values (i.e., for the oldest and for the youngest winning coalition members). If these particular findings are valid, further research on age-related effect pronouncement is required.
Figure 7. Frequency distribution by biological age for split samples of party functionaries
Our Hypothesis 3 suggests that those winning coalition members who hold repression-related offices tend to actively oppose to liberalization less frequently than those who do not.
Given the categorical nature of both variables, a chi-square test was performed and Cramer's V was computed in jamovi software (see Table 10).
Table 10. Contingency table and chi-square test results for variables repress_relat_office and liberalis_support for the whole population
|
liberalis_support |
||||
|
repress_relat_office |
“0“ |
“1“ |
Total |
|
|
“0“ |
151 |
39 |
190 |
|
|
“1“ |
9 |
15 |
24 |
|
|
Total |
160 |
54 |
214 |
|
|
Value |
Df |
P |
||
|
X2 |
19.9 |
1 |
< .001 |
|
|
N |
214 |
|||
|
Contingency coefficient |
0.292 |
|||
|
Cramer's V |
0.305 |
The association between the variables appears to be relatively strong (Cramer's V equals 0.305) while the chi-square value is statistically significant (p < .001). Since a co-
causality in liberalization support prediction transpires, the association between these two variables albeit not extremely strong is noteworthy. Moreover, if technocrats and the military are excluded from the sample, the association between repress_relat_office and liberalis_support becomes even stronger (see Table 11).
Table 11. Contingency table and chi-square test results for repress_related_office and liberalis_support for the sample including only party functionaries
|
liberalis_support |
||||
|
repress_relat_office |
“0“ |
“1“ |
Total |
|
|
“0“ |
67 |
27 |
94 |
|
|
“1“ |
6 |
15 |
21 |
|
|
Total |
73 |
42 |
115 |
|
|
Value |
Df |
P |
||
|
X2 |
13.5 |
1 |
< .001 |
|
|
N |
115 |
|||
|
Contingency coefficient |
0.324 |
|||
|
Cramer's V |
0.343 |
The conclusion that repression-related officeholders are more prone to support liberalization is corroborated by the comparison of survival functions for these two groups. The respective graphs are again drawn in SPSS based on non-parametric KaplanMeier techniques frequently used in survival analysis (see Figure 8). The green graph (the survival function for repression-related office holders) is located noticeably lower than the blue one for non-repression-related party functionaries. This trend, as well as a significant difference in mean survival time values (8.9 days for the latter and 14 days for the former), could serve as a statistical proof for the model's correctness.
The number of actual repression-related office holders must be taken into account. Only 24 out of 214 winning coalition members in the database were coded as such. Apart from that, inasmuch as the preponderance of them are also holding higher positions within the party and law enforcement hierarchy and have, therefore, disproportional access to the media, the messages containing liberalization support from the other group might have been simply underrepresented.
The association we spotted, though, is the strongest both according to the model's predictions and to the actual evidence. In other words, those winning coalition members who hold repression-related offices indeed tend to support liberalization more often than those who do not. This trend confirms our assumption that in their decision whether to support the regime opening winning coalition members take into consideration not only potential benefits but also possible risks related to further cooperation with regime, especially when this imposes responsibility for the violent repression of thousands of non-violent protesters. It is plausible that awareness of having potential responsibilities for repressive actions is cumulatively amplified by hopes for a political career after liberalization. As a result, actors who have to make these harsh choices tend to switch the sides and to defect even during the early phases of contentious politics. By definition, most repression-related officeholders were top leaders of the party's Bezirk or district committees. As some historical accounts witness (see, for instance, Bahr 2016), in a situation when several thousand people gathered to demonstrate somewhere in Rostock, Brandenburg or in Halle, it is no wonder they chose to tolerate the social movement and ordered that militia/police platoons hold back, often even without coordinating this decision with the central party leadership. Similar reasoning is suggested by Reuter and Szakonyi (2019) in their account of current elite defections in Russia. The essential question to study is: under what circumstances do the risks of being sentenced after opposition's victory outweighed more proximate risks of being punished for disobedience? This invites to further microanalysis of motivational structures, biographical factors, and network effects.
Figure 8. Survival functions for winning coalition members differentiated by their repression-related (green) or non-repression-related (blue) position
Our Hypothesis 4 premises that outer circle winning coalition members tend to support regime liberalization just as often as winning coalition members from the inner circle.
To test this hypothesis, we applied the classical null hypothesis approach so that confirmation requires merely proof that there is no significant association between the variables in question. Chi-square and Cramer's V values for hierarchy_lvl and liberalis_support is this case should approach zero. Their values are computed in jamovi after excluding from the sample those objects whose repress_relat_ojfice values equal “1” in order to avoid an unaccountable influence of this variable.
Table 12. Chi-square and Cramer's V values for hierarchy_lvl and liberalis_support (repress_relat_office being kept at 0).
|
liberalis_support |
||||
|
hierarchy_lvl |
“0“ |
“1“ |
Total |
|
|
Inner |
41 |
10 |
51 |
|
|
Outer |
110 |
29 |
39 |
|
|
Total |
151 |
39 |
190 |
|
|
Value |
Df |
P |
||
|
X2 |
0.0361 |
1 |
0.849 |
|
|
N |
190 |
|||
|
Contingency coefficient |
0.0138 |
|||
|
Cramer's V |
0.0138 |
The chi-square test yields neither statistically significant nor positive results. Therefore, there is no clear association between an individual's position within the party hierarchy (his or her belonging to the outer or to the inner circle) and his or her tendency to support regime liberalization, which corroborates Hypothesis 4. Still, graphs of survival functions which were modeled in SPSS based on the values computed with the help of Kaplan-Meier do somewhat differ for the members of the outer and inner circle.
Figure 9. Survival functions for winning coalition members differentiated by their hierarchy level
The graph of survival function for the inner circle members (see the green “2” line in Figure 9) lies somewhere slightly below (i.e. has on average somewhat lower survival rate values) than the blue graph (marked as “1”) of the survival function for the outer circle coalition members. Besides, the difference in mean survival time values between these categories is around 1.2 days with the mean survival time for inner circle members equaling 11.7 days and for outer circle members, 12.9 days. Despite a slight and hardly noticeable discrepancy, it looks like in our sample inner circle winning coalition members supported liberalization more often than outer circle members even though, as stipulated by the classical selectorate theory, the regime's existence is more beneficial to them than to the outer circle members.
The anomaly, once again, might be accounted for by the data: the state- or party- owned newspapers under autocracy in crisis are more likely to publish someone's call for regime liberalization if s\he turns out to be the head of a parliamentary committee or a general director of a state enterprise rather than if s\he is the head of a local party cell, though such cases occurred as our study of media messages revealed. The differences in support frequencies in relation to the hierarchy level should, thus, be taken with caution and must be double-checked for other, perhaps bigger, winning coalition samples with a higher degree of inner differentiation.
Conclusions
This paper aimed to disclose some patterns in the downfall of one of the most durable dictatorship types, namely single-party regimes. In particular, we studied the factionalism within the GDR winning coalition under the strain of mass contentious politics. The bulk of evidence we examined sheds light on noteworthy regularities in ways members of the winning coalition chose either to support or to obstruct the liberalization demands voiced by citizens in the streets as well as the timing of their decision. Such a choice provoked splintering alongside cracking lines untraceable prior to contentious politics.
As it turned out, the most robust predictor of eventual defection was what we dubbed in this study a repression-related office. This position obliges a person facing a challenge of popular contention to adjudicate whether repression or toleration is the most appropriate tactics. Contrary to common assumptions, it appears that individuals who hold repression-related offices are less eager to apply force indiscriminately and tend to support regime liberalization more often than those who do not. One can speculate whether such individuals are reluctant to provoke bloodshed, or they fear eventual criminal responsibility for violent crackdown after the regime change. Exact rationale notwithstanding (and, arguably, varying from person to person), it makes strategic sense for contenders to seek geographical diffusion of non-violent contentious politics. Participants of mass demonstration, if organized in large numbers, by the sheer fact of their presence in the streets put local elites in an unenviable position between the hammer of further support for the repressive regime and the anvil of granting concessions to the contenders. As tensions mount and carnage looms ominously, some winning coalition members must thoroughly consider all possible outcomes for their present-time decisions. Under strain, the number of fractures within the winning coalition increases and the gap between individual actors widens. The coalition, therefore, loses its original cohesion, which makes factionalism and defection more likely. We find additional evidence to this model prediction in the meeting transcripts of CC SED Congresses which took place in October and November of 1989 (see Hertle & Gerd-Rьdiger 1997: 183).
Another remarkable finding is the absence of substantial evidence for any causal links between a person's position within the party hierarchy and his or her eventual support for liberalization. Although the standard selectorate theory stipulates that inner circle members who hold a privileged position within the winning coalition and thus enjoy broader access to both material and non-material benefits should support the regime status quo more vigorously, we find no evidence that these happy few indeed advocated hardline politics more eagerly in any statistically significant instances. In other words, even though coalition members from the inner and the outer circles were granted different amounts of perks, the difference gave no impetus for privileged members to support the regime to the dire end. However, our study offers a vindication for the modified selectorate theory. Elaborated by Cao & Ward, it stipulates that “if state capacity is low, the size of winning coalition should not affect the level of provision of public goods” (2015: 266). We surmise that due to the drastic drop in state capacity observed in the GDR after die Wende, the regime, which since the Honecker era had been gradually shifting to public goods distribution simultaneously shrinking the allocation of private goods, suddenly faced the lack of loyalty from populace, selectorate, and individual members of the winning coalition. Therefore, it is arguable that in fat years and with longtime horizons, the single-party regime in East Germany managed to build-in a “safety- lock” against possible party splits along the line separating the Fьhrungskern (the ruling core) and rank-and-file CC SED members. However, the very same strategy proved to be fatal during lean years and rapidly shrinking time-horizon.
Next, the absence of a linear relation between a winning coalition member's age (either biological or institutional) and his or her support for liberalization suggests that age on its own without considering the type or importance of the office a person holds can hardly indicate anything consistent about his or her inclinations to support regime opening and reforms. Only the oldest winning coalition members were never inclined to defect, for, presumably, they lacked any further political career prospects or even the very understanding of reforms' necessity due to the prolonged ideological indoctrination. Concurrently, the youngest CC SED members were overrepresented among regime liberalization supporters, which reveals their eagerness to remain in politics after regime changes. In addition, they are likely to have cultivated closer ties with civil society. These connections might be partially responsible for youngest CC SED members' awareness that the citizenry has valid concerns and yearn for change. For a better understanding of both the oldest and the youngest generation, a deeper look into biographies, memoirs, and correspondence of winning coalition members is advisable. Most probably, the age anomaly we signalized about is simultaneously conditioned by a group of factors which includes both the biographical and social network ones (e.g. if an individual is in a close relationship with actors who are more mistrustful of the regime). Learning how the state and party apparatus interacted with the civil society in the GDR's capital city as well as in other localities seems to be a promising research direction in a more general effort to explain elite factionalism in terms of the social network approach.
Finally, statistical tests revealed a significant association between a type of the office a person holds and his or her decision to support or not to support regime opening. The statistically significant finding that party functionaries are more inclined to be softliners than technocrats or the military does seem to resonate with a thesis advanced by Przeworski (1991: 67-68) that hard-liners tend to be concentrated predominantly in the law enforcement institutions and within the propagandistic apparatus while soft-liners mostly happen to be regime's "politicians". Besides, Przeworski speculates that "socialist managers" should be overrepresented among liberalization supporters (ibid.). Although we don't address this line of inquiry directly, we conjecture that the winning coalition members who expect to instrumentalize their economic, political or symbolic capital to build political career after liberalization are the most likely supporters of regime opening as regime change offers them tangible benefits (for more elaborated arguments in this direction, see again Reuter & Szakonyi 2019). Further studies of this presumption require a more differentiating format of the office type variable so that individuals could be classified and ranked according to the strength of their incentive to instrumentalize regime change for eventual rent augmentation.
The military officials, for their part, turned out to be rather reluctant to support liberalization, which concurs other scholars' conclusions. Still, one must not ignore that rationale to oppose liberalization might be contingent upon both strategic and value judgments. Axiologically, the military ethos, namely the cult of order, the mistrust in street politics, and the apprehension to act against the state interests often perceived as the regime interests, makes the military an unlikely institutional ally for a contender. Strategically, officers and rank-and-file military men might be unwilling to disclose and declassify information about all the atrocities and repressive actions they had been implicated in prior to contentious politics. These stimuli differ significantly in their effects, for if there are no dirty secrets, military men have fewer reasons to actively hinder the regime change. Therefore, a more nuanced approach with more qualitative data must be deployed here to explore the variations within the military's motivation structure.
All in all, not only does this study help to elucidate the critical weeks in the GDR history when the winning coalition factionalized irreparably but also it contributes to democratization theory, social movements theory, and selectorate theory. In particular, we suggest that selectorate and winning coalition members are not merely “interchangeable”, for they differ in expertise, expectations, and values. These factors have a considerable impact on their inclination to defect. In addition, we emphasize that even under highly resilient single-party regimes the citizenry does have leverages upon the rulers when it manages to discover and to play on internal cracks between the elites through sustained contention.
Bibliography
1. Acemoglu, Daron & Robinson, James (2012). Economic Origins of Dictatorship and Democracy. New York: Cambridge University Press.
2. Bahr, Andrea (2016). Parteiherrschaft vor Ort: Die SED-Kreisleitung Brandenburg 1961-1989. Brandenburg: Ch. Links.
3. Bauernkдmper, Arnd (2005). Die Sozialgeschichte der DDR. Mьnchen: Oldenbourg.
4. Boix, Carles (2013). Democracy and Redistribution. Cambridge: Cambridge University Press.
5. Boix, Carles & Svolik, Milan (2013). The Foundations of Limited Authoritarian Government: Institutions, Commitment, and Power-Sharing in Dictatorships. The Journal of Politics 75(2): 300-316.
6. Brooker, Paul (2004). Non-Democratic Regimes: Theory, Government, and Politics. New York: St. Martin's Press. Buch, Gьnther (1987). Namen und Daten: wichtiger Personen der DDR. Berlin: J.H.W. Dietz.
7. Bueno de Mesquita, Bruce & Smith, Alastair & Siverson, Randolph & Morrow, James (2005). The Logic of Political Survival, Cambridge: MIT Press.
8. Cao, Xia & Ward, Hugh (2015). Winning Coalition Size, State Capacity, and Time Horizons: An Application of Modified Selectorate Theory to Environmental Public Goods Provision. International Studies Quarterly 59: 264-279.
9. Collier, Ruth (1999). Paths toward Democracy: The Working Class and Elites in Western Europe and South America. Cambridge: Cambridge University Press.
10. Der Bundesbeauftragte fьr die Unterlagen des Staatssicherheitsdienstes der ehemaligen Deutschen Demokratischen Republik. (2019). Official Webpage, https://www.bstu.de/archiv/ (accessed 10 December 2019).
11. De Sierra, Sebastian Garrido (2011). Eroded Unity and Clientele Migration: An Alternative Explanation of Mexico's Democratic Transition. Annual Meetings of the Midwest Political Science Association, Chicago, April 2011, https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2012655.
12. Die Volkskammer der Deutschen Demokratischen Republik (8. Wahlperiode) 1982. Berlin: Staatsverlag der DDR. Dix, Robert (1982). The Breakdown of Authoritarian Regimes. Political Research Quarterly 35(4): 554-573. Eklцf, Stefan (2003). Power and Political Culture in Suharto's Indonesia: The Indonesian Democratic Party (PDI) and Decline of the New Order (1986-98). Copenhagen: NIAS Press.
13. Enquete-Kommission. Aufarbeitung von Geschichte und Folgen der SED-Diktatur in Deutschland (1995). Vol. VIII. Frankfurt am Main: Suhrkamp.
14. Fulbrook, Mary. (2015). A History of Germany 1918-2014: The Divided Nation. Oxford: Wiley-Blackwell. Gallagher, Mary & Hanson, Jonathan (2015). Power Tool or Dull Blade? Selectorate Theory for Autocracies. Annual Review of Political Science 18: 367-385.
15. Gandhi, Jennifer & Przeworski, Adam (2007). Authoritarian Institutions and the Survival of Autocrats. Comparative Political Studies 40(11): 1279-1301.
16. Geddes, Barbara (1999). Authoritarian Breakdown: Empirical Test of a Game Theoretical Argument. Annual Meeting of the American Political Science Association, Atlanta, September 1999: 1-55.
17. Geddes, Barbara & Wright, Joseph & Frantz, Erica (2014). Authoritarian Breakdown and Regime Transitions: A New Data Set. Perspectives on Politics 12(2): 313-331.
18. Geddes, Barbara & Wright, Joseph & Frantz, Erica (2018). How Dictatorships Work. New York: Cambridge University Press.
19. Glaeser, Andreas (2010). Political Epistemics: The Secret Police, the Opposition, and the End of East German Socialism. Chicago: University of Chicago Press.
20. Greene, Kenneth (2007). Why Dominant Parties Lose. New York: Cambridge University Press.
21. Grix, Jonathan (2000). The Role of the Masses in the Collapse of the GDR. New York: St. Martin's Press. Hadenius, Axel & Teorell, Jan (2007). Pathways from Authoritarianism. Journal of Democracy 18(1): 143156.
22. Haggard, Stephan & Kaufmann, Robert (2016). Dictators and Democrats: Masses, Elites, and Regime Change. Princeton: Princeton University Press.
23. Hauslohner, Peter (1981). Prefects as Senators: Soviet Regional Politicians Look to Foreign Policy. World Politics 33(2): 197-233.
24. Hertle, Hermann & Gerd-Rьdiger, Stephan (1997). Das Ende der SED: Die letzten Tage des Zentralkomitees. Berlin: Links.
25. Hinnebusch, Raymond (2006). Authoritarian Persistence, Democratization Theory and the Middle East: An Overview and Critique. Democratization 13(3): 373-395.
26. Hodnett, Gary (1975). Succession Contingencies in the Soviet Union. Problems of Communism XXIV: 1-21. Kaiser, Monika (1997). Machtwechsel von Ulbricht zu Honecker. Funktionsmechanismen der SED-Diktatur in Konfliktsituationen 1962-1972. Berlin: Akademie Verlag.
27. Kitschelt, Herbert & Wilkinson, Steven (2007). Patrons, Clients, and Policies: Patterns of Democratic Accountability and Political Competition. New York: Cambridge University Press.
28. Kopstein, Jeffrey (1997). The Politics of Economic Decline in East Germany, 1945-1989. Chapel Hill and London: University of North Carolina Press.
29. Kricheli, Ruth & Livne, Yair (2012). Mass Revolutions vs. Elite Coups. The Annual Meeting of the American Political Science Association, Toronto.
30. Kuntz, Philipp & Thompson, Mark (2009). More than Just the Final Straw. Stolen Elections as Revolutionary Triggers. Comparative Politics 41(3): 253-272.
31. Kuran, Timur (1997). Private Truths, Public Lies: The Social Consequences of Preference Falsification. Cambridge, Mass: Harvard University Press.
32. Kaelble, Hartmut & Kocka, Jьrgen, & Zwahr, Hartmut (1994). Sozialgeschichte der DDR. Stuttgart: Klett-Cotta Verlag.
33. Lindenberger, Thomas (1999). Herrschaft und Eigen-Sinn in der Diktatur-Studien zu Gesellschaftsgeschichte der DDR. Wien, Kцln, Weimar: Bцhlau Verlag.
34. Lyons, Terrence (2016). The Importance of Winning: Victorious Insurgent Groups and Authoritarian Politics. Comparative Politics 2(48): 167-184.
35. Madarasz, Jeannette (2003). Conflict and Compromise in East Germany, 1971-1989: A Precarious Stability. Basingstoke: Palgrave Macmillan.
36. Major, Patrick (2010). Behind the Berlin Wall: East Germany and the Frontiers of Power. Oxford: Oxford University Press.
37. Malesky, Edmund & Abrami, Regina & Zheng, Yu (2011). Institutions and Inequality in Single-Party Regimes: A Comparative Analysis of Vietnam and China. Comparative Politics 43(4): 401-419. Meyer, Gerd (1991). Die DDR-Machtelite in der Дra Honecker. Tьbingen: Francke.
38. Miller, Michael (2012). Electoral authoritarianism and democracy: A formal model of regime transitions.
39. Journal of Theoretical Politics 25(2): 153-181.
40. Mitglieder des Zentralkomitees der SED (1986). Neues Deutschland, April 22: 4.
41. Morency-Laflamme, Julien (2015). A Missing Link? Elite Factionalism and Democratization in Africa.
42. Canadian Journal of African Studies / Revue canadienne des йtudes africaines 49(3): 459-477. Mьller-Enbergs, Helmut & Wielgohs, Jan & Hoffman, Dieter (2003). Wer War Wer in der DDR? Ein Biographisches Lexikon. Augsburg: Weltbild.
43. Neugebauer, Gero (1974). Die Volkskammer der DDR. Zeitschrift fьr Parlamentsfragen 5: 386-411. O'Donnell, Guillermo & Schmitter, Philippe (1986). Transitions from Authoritarian Rule: Tentative Conclusions. Baltimore: John Hopkins University Press.
44. Opp, Karl-Dieter & Gern, Christiane (1993). Dissident Groups, Personal Networks, and Spontaneous Cooperation: The East German Revolution of 1989. American Sociological Review 58(5): 659-680. Osborne, Martin (2003). An Introduction to Game Theory. Oxford: Oxford University Press.
45. Paterson, Peter (1967). The Selectorate: The Case for Primary Elections in Britain. London: Macgibbon & Kee. Pepinsky, Thomas (2014). The Institutional Turn in Comparative Authoritarianism. British Journal of Political Science 44(3): 631-653.
46. Pfaff, Steven (2006). Exit-Voice Dynamics and the Collapse of East Germany: The Crisis of Leninism and the Revolution of 1989. Durham: Duke University Press.
47. Pollack, Dieter & Rink, Dieter (eds.) (1997). Zwischen Verweigerung und Opposition-Politischer Protest in der DDR 1970-1989. Frankfurt am Main: Campus Verlag.
48. Przeworski, Adam (1991). Democracy and the Market. Princeton: Princeton University Press.
49. Reuter, Ora & Szakonyi, David. (2019). Elite Defection under Autocracy: Evidence from Russia. American Political Science Review 113(2): 552-568.
50. Roeder, Philippe. (1993). Red Sunset: The Failure of Soviet Politics. Princeton: Princeton University Press. Ross, Corey (2002). The East German Dictatorship: Problems and Perspectives in the Interpretation of the GDR. London: Arnold.
51. Rueschemeyer, Dietrich & Huber, Evelyne & Stephens, John (1992). Capitalist Development and Democracy. Chicago: University of Chicago Press.
52. Schroeder, Karl (2013). Der SED Staat: Geschichte und Strukturen der DDR 1949-1990. Kцln: Bцhlau Verlag.
53. Smith, Benjamin (2005). Life of the Party: The Origins of Regime Breakdown and Persistence under SingleParty Rule. World Politics 57: 421-451.
54. Stepan, Alfred (1988). Rethinking Military Politics: Brazil and the Southern Cone. Princeton: Princeton University Press.
55. Stern, Geoffrey (1978). Brezhnev and the Future: Leadership and the Problems of Succession in the Soviet Union. The Commonwealth Journal of International Affairs 68(272): 340-347.
56. Sutter, David (2000). The Transition from Authoritarian Rule: A Game Theoretic Approach. Journal of Theoretical Politics 1(2): 67-89.
57. Svolik, Milan (2012). The Politics of Authoritarian Rule. Cambridge: Cambridge University Press.
58. Tanaka, Seiki (2016). Retirement Planning for Dictators: What Happens to Outgoing Dictators? International Relations of the Asia-Pacific 16: 1-25.
59. Thieme, Tom (2015). Eliten und Systemwechsel: Die Rolle der sozialistischen Parteifьhrungen im Demokratisierungsprozess. Baden-Baden: Nomos Verlaggesellschaft.
60. Thomson, Henry (2016). Regime Structures, Intra-Elite Factions and the Problem of Authoritarian Control, East European Politics and Societies, https://ora.ox.ac.uk/objects/uuid:a566ded9-8e89-48e9-9322- 5ab5e185e8ba.
61. Ulfelder, Jay (2005). Contentious Collective Action and the Breakdown of Authoritarian Regimes. Regimes and Regime Change 26(3): 311-334.
62. Wintrobe, Ronald (1998). The Political Economy of Dictatorship. Cambridge: Cambridge University Press.
63. Wright, Joseph (2008). Do Authoritarian Institutions Constrain? How Legislatures Affect Economic Growth and Investment. American Journal of Political Science 52(2): 322-343.
64. Verfassung der DDR. Verfassung der DDR vom 6. April 1968 in der Fassung des Gesetzes zur Ergдnzung und Дnderung der Verfassung der DDR vom 7. Oktober 1974. Berlin: Staatsverlag der DDR.
65. ZEFYS Zeitungsinformationssystem (2019). Online database, Staatsbibliothek zu Berlin, http://zefys.staatsbibliothek-berlin.de (accessed 18 August 2019).
66. Размещено на Allbest.ru
Подобные документы
The Effects Of The Industrial Revolution. Change in Urban Society. The Industrial Revolution presented mankind with a miracle that changed the fabric of human behavior and social interaction. Economic growth. Economic specialization.
реферат [23,8 K], добавлен 11.12.2006Особенности деятельности исследуемой церкви на белорусских землях в 1989–2014 гг., ее исследование и анализ. Назначение и структура основных приходов, история их развития. Направления и оценка эффективности работы белорусских греко-католиков за рубежом.
курсовая работа [75,3 K], добавлен 20.10.2015Russia Empire in the XX century entered into a complex economic and political environment. Consequences of defeat of autocracy in war with Japan. Reasons of growing revolutionary motion in Grodno. Events of revolution of a 1905 year in Byelorussia.
реферат [9,4 K], добавлен 14.10.2009Features of the socio-political situation of the Kazakh people after the October Revolution of 1917. The creation of KazASSR in 1920, its internal structure of the state system, main stages of development and the economic and industrial achievements.
презентация [1,2 M], добавлен 01.03.2016Bourgeoisie and proletariat as two massive flows in France, which prepare and made revolution. French Revolution as an impact on the appearing the entire political events in the European countries. Democratic actions in Switzerland after revolution.
доклад [10,7 K], добавлен 14.04.2010Gordon Wood is Professor of History at Brown University. He is one of the leading scholars researching issues of the American Revolution in the country. Problems researching revolutionary nature of the American Revolution.
реферат [21,4 K], добавлен 27.09.2006Revolts and revolutions often occur in the course of history, however, revolutions are considered to be a more recent development. The Frondes and a revolt. The French revolution. The comparison of a revolution and a revolt.
реферат [8,9 K], добавлен 09.12.2004The Industrial Revolution was a period in history when mankind found innovative and efficient ways of producing goods, manufacturing services and creating new methods of transportation.
реферат [15,7 K], добавлен 28.04.2002The Spanish Empire as one of the largest empires in world history and the first of global extent. Seaborne trade. Broken Spain and England's relations. The main reasons of war. Some main facts about the Spanish Armada. The first colony of England.
творческая работа [8,9 M], добавлен 13.01.2016History is Philosophy teaching by examples. Renaissance, French Revolution and the First World War are important events in the development of the world history. French Revolution is freedom of speech. The First World War is show of the chemical weapons.
реферат [21,6 K], добавлен 14.12.2011
