5 Methodology 


The following section describes the methodology applied in order to systematically answer the research question above. Before I delve into the details and peculiarities of empirical research, it is necessary to illustrate the ultimate purpose the empirical analysis serves in the context of this research. Empirically and analytically inspired scientists draw conclusions by applying two distinct models of research: Either scholars value theoretically derived generalizations and use empirical analysis merely in order to test theoretically generated hypothesis, thus they apply the classical hypothetico-deductive model of research, or they utilize the variety of observable phenomena in social reality as a means of stimulating theoretical generalizations, i.e. they apply an analytical inductive approach to theory development. Although both methods are regarded as equally capable of bringing about worthwhile findings, this analysis utilizes empirical analysis in the former and not in the latter way. Instead of inducing theoretical generalizations from observable facts, the purpose is to show that the model theoretically derived above illustrates social reality in a useful manner. This will be done by conducting a comparative examination of two cases.

Having said this, the research to be conducted needs to be carefully defined in order to arrive at valid conclusions. This chapter dwells on four distinct methodological questions that require clarification. First, the chapter investigates the peculiarities of comparative research in the social sciences. Second, it exposes the rationale behind the selection of the two country cases under investigation. It will be shown that the cases are selected in a manner that allows us to hold one possible alternative explanation – the geo-strategic location of the target country – constant, whereas another alternative explanation – the cultural and historic disposition of civil society – may be rejected by applying a “most-different systems” design (Przeworsk / Teune 1970). Third, one needs to address the question what exactly is observed. How can we identify observable indicators that are capable of determining the effects of civil society assistance and the “success” of civil society projects? What are the units of analysis, i.e. which donor and recipient organizations are the major objects under examination? And finally, the question arises what quantitative and qualitative means of measurement are employed in order to collect reliable information.

5.1 The Comparative Method and the Small N - Many Variables Problem 

Although research may well follow other purposes81, the ideal in social science is to draw generalizable conclusions from the cases under investigation and thus to “produce lawlike generalizations with empirical validity” (Scharpf 1997: 19). In practice, though, it is difficult to make causal inferences in the social sciences. One major problematic inherent to social research lies in the magnitude of possible alternative explanations to observable outcomes. Why should the development of a vibrant civil society, the establishment of numerous and pluralistic organizations and institutions of civil society, civic activity and mechanisms of participatory rule be the result of international assistance and not the consequence of an existing cultural basis, historically grown “social capital” (Putnam 1993), or simply the result of political liberalization and the granting of equal political rights and civil freedoms? Moreover, if a change in civic attitudes and institutions can be observed, how can this be attributed to bottom-up strategies to assist civil society in face of a variety of other powerful international factors as e.g. the prospect of EU membership? Alternative explanations are hard to rule out.


One possible method of “discovering empirical relationships among variables” is the comparative method (Lijphart 1971: 683).82 The comparative method draws largely on John Stuart Mill’s presentation of canons of experimental inquiry in A System of Logic: Ratiocinative and Inductive (1843) that provides several research strategies that allow for generalizations in social science. Of particular relevance for comparative analysis are two of Mill’s strategies: the “method of agreement” and the “indirect method of difference” (Ragin 1987: 36). In simple terms, the basic logic behind the two methods is to discard alternative explanations by comparing the differences and similarities among cases. While the “method of agreement” aims to single out alternative explanations by identifying the one and only causal factor that all observed cases depicting the outcome under investigation have in common, the “indirect method of difference” highlights cases that neither reveal the cause nor the effect and thus rejects competing explanations through paired comparisons (ibid: 36pp). The major task for scientists is thus to find cases that ideally agree in only one of the possible explanatory variables.

The comparative method, however, is plagued by the problem that we usually encounter various possible explanatory variables, but only a limited number of cases for research (Lijphart 1971: 685). This predicament has become known as the small N, many variables problem. It is thus often not feasible to reject all alternative explanations by comparing differences and similarities for the simple fact that not enough cases can be found and that the number of possible causes augments the number of observable cases.83 This point is easily illustrated by the concrete case under investigation. The subsequent research encounters only a limited number of countries that undergo political transformation processes, that further receive significant Western aid to civil society, and whose investigation also does not exceed the resources available for this analysis. A selection of cases is thus inevitable and, as King et al (1994: 139) point out, in a comparative study faced with the small N, many variables problem, “selection must be done in an intentional fashion, consistent with our research objectives and strategy”. The usual advice of experienced scholars on how to best select cases capable of minimizing the many variables, small numbers problem, and deriving generalizable results are as follows: (1) increase the number of cases as much as possible; (2) reduce the property space of the analysis, i.e. combine two or more variables that express an essentially similar underlying characteristic into a single variable; and (3) focus the comparative analysis on “comparable” cases, i.e. cases that are similar in a large number of important characteristics which one wants to treat as constants, but dissimilar as far as those variables are concerned which one wants to relate to each other (Lijphart 1971: 686p). While pointing to comparable cases Lijphart refers to what has been labeled a “most similar systems” design, at that time described as “the currently predominant view among social scientists” (Przeworski / Teune 1970: 32). The “most-similar systems design” focuses on cases that are as similar as possible in a wide range of variables, which differ, however, concerning the outcome under investigation and the key variable. Such a design controls for possible explanatory factors, as those are held constant, i.e. they are similar in the analyzed cases. The differences among the cases are thus left as possible causes for the phenomenon that is to be explained. To this extent, the method follows the logic of Mill’s method of agreement that explains a constant outcome by another constant cause.

Przeworski / Teune (1970: 37p) state in criticism to the ”most similar systems” design:


“When the observed systems share characteristics X1, X2, …, Xk, the variations of the dependent variable Y (…) are associated with the variable Xk+1 (according to the hypothesis) or the alternative variables Xk+2, …, Xn (alternative hypothesis)….. The original hypothesis is (thus) confirmed, although alternative hypotheses are not rejected.”

They criticize this design for two reasons. Firstly, they point out that the findings resulting from a “most-similar systems” design are only valid for cases that share the characteristics common to the selected “most similar” cases. Surely enough, the design follows the logic that the findings can be tested and confirmed in other, different, cases and thus aims to remove the controlled variables one-by-one. Yet, as Teune / Przeworski point out, if the findings are not confirmed, “… we are back where we started” (ibid: 38). Secondly, and more important for their subsequent conclusion, with a “most similar systems” design experimental variables cannot be singled out. Przeworski / Teune stress the point that no systems are so similar that they diverge only in one factor. Other characteristics in which the two (or more) cases under investigation differ (or further characteristics the scholar is not aware of) may account for the observed outcomes. In other words, the dependent phenomenon is “overdetermined” (ibid: 34). The “most similar systems” research design thus strengthens the confidence in the explanatory power of a hypothesis but only to a certain degree. If the resulting findings cannot be confirmed in cases diverging in one or more factors held constant by the “most similar systems” design, a relationship has been determined that seems dependent on another unknown variable.


For these reasons Przeworski / Teune propose another design that focuses not on “most-similar” but on “most-different” systems. The “most different systems design” aims to trace similar outcomes in a set of cases as diverse as possible. The underlying logic is that systemic explanations can be rejected if similar processes of change can be identified. Rather than positively identifying relevant systemic factors, the “most different systems” design thus “centers on eliminating irrelevant systemic factors” (ibid: 35). The point is that if a certain outcome is observed in very different cases, the factors in which the cases differ cannot be responsible for the observed outcome.

Although Przeworski / Teune present the two designs in opposition to each other, it has been argued that the two strategies may be fruitfully combined. Collier (1991: 17) proposes a strategy that starts with a sample of cases that are roughly matched along a number of variables, controlling for their influence. As a result though, the differences between the cases are highlighted. In the same manner, my work will also take a middle position and partly “control” for alternative explanations by keeping them constant and partly reject alternative explanations through a most-dissimilar systems design. In other words, the dissertation will first follow the advice of Lijphart and focus on comparable cases and afterwards pursue Przeworski / Teune’s strategy and highlight differences.

5.2 The Selection of Cases

The comparable set of cases for this study are the countries of Central and Eastern Europe which are located close to the European Union, have recently joined the European Union, and that all received civil society assistance from the West in the period between 1990-2000. Of course, we also witness transformation processes enhanced by Western donors including civil society assistance further to the East. However, those countries are so dissimilar regarding geo-strategic as well as cultural and historic conditions that a comparison would only have limited significance. Demonstrating that civil society in Poland is more stable and vibrant than that in Belarus hardly comes as a surprise. The stabilizing influence of the prospects of EU membership, which requires stable democratic institutions guaranteeing civic participation and free association on the one hand, and the cultural and historical traits ranking from early experiences with associative forms to an encompassing dissident movement on the other hand, are both strong explanatory factors that account for these differences. In the face of such obvious explanations, the large Western attention civil society organizations received in Poland that far exceeded the assistance granted to civil society organizations in Belarus will be regarded as a rather minor causal factor. The selection of cases is thus based on a design that is partly most similar and partly most different in order to deal with two prominent alternative explanations: the stabilizing influence of the European Union (EU) on the one hand, and pre-existing cultural and historical strands of civil society on the other hand. Whereas the first alternative explanation is held constant, the second is chosen in a way that it is most dissimilar.


For these reasons, the dissertation project focuses on civil society assistance in Poland and Slovakia. Being geographically close to the EU, the two countries Poland and Slovakia have been chosen because it is believed that the prospect of EU membership is an important factor, which should be held constant. The interest of the dissertation is not so much to assess the impact of possible EU membership – requiring vibrant civil societies and democratic conditions (Copenhagen Criteria) – on civil society in democratizing countries, but rather to analyze the mechanisms and outcome of civil society assistance in the form of financial and technical assistance directly granted to domestic actors. As the multilateral framework of transnational civil society activities is relatively similar in Poland and Slovakia due to a geographic closeness to the EU, its impact will be held constant. However, among the countries geographically close to the EU, Poland and Slovakia are the most dissimilar. As the largest country in CEE, Poland is culturally homogenous and has a firm national identity and no major ethical or cultural cleavages. Slovakia, in contrast, is a small and ethnically heterogeneous country, characterized by late independence and a late and feeble national awakening. Poland is the country in the former communist block with the most encompassing dissident movement and a forerunner with regard to democratic reforms with no major drawbacks on its way to becoming a consolidated democracy. The Slovak lands, on the other hand, showed only weak traces of dissident culture during communism and were governed by a repressive communism system that left little room for experiments with liberalization. Furthermore, Slovakia faced an authoritarian reverse wave between 1994 and 1998.84 One has to note, however, that both countries belong to the most Catholic countries in Europe and that both were at least partly part of the Habsburg Empire. Nonetheless, a comparison of both countries is capable of revealing the effects of foreign assistance in two different domestic settings that are additionally shaped by thoroughly different cultural and historical developments structuring civil society development. The project consequently aims to examine the extent to which two countries with quite different legacies shaping the revival of civil society reacted differently to civil society assistance.

Before I go on to outline how the processes of change in the two countries under investigation will be traced, a caveat is in order as to what the chosen design implies for the comparability of the resulting findings. In other words, if causality between civil society assistance and civil society development can be demonstrated in the two cases under investigation by methods described below, to what extent can the empirical findings claim validity beyond the two cases? As we learned from the discussion of the work of Przeworski / Teune, a most-similar system design cannot reject alternative explanations. Consequently, the resulting findings are only valid in cases that depict the same characteristics than the cases examined. For this study this implies that the findings on the impact of transnational influences on civil society are only valid for cases that depict multilateral and bilateral pressures impacting on the domestic system in a way similar to the EU integration process. In other words, the findings on the impact of transnational civil society assistance are only valid in countries exposed to similar international adaptation and integration pressures. If we want to test to what extent the findings are valid beyond the range of countries facing a similar international environment, additional investigation, e.g. in the former Soviet Union, is necessary.85 However, this conclusion does not hold for the other alternative variable: cultural and historical preconditions of civil society. If an impact of civil society assistance on civil society development is demonstrated for both cases under investigation, i.e. for cases that are most different concerning their cultural and historical legacies, the “cultural approach” that stresses the importance of institutional legacies, informal rules and culturally defined cognitive scripts can be rejected.

5.3 What to Observe

The aim of this section is to identify general questions to be asked in each case in order to allow for a “structured, focused comparison”, i.e. a comparison that focuses only on a specific aspect of a case, here civil society development, and that employs general questions to guide the data collection and analysis (George 1979: 61p).86 In other words, the following will clarify what should be observed in each case. Additionally, the section specifies the objects of research. It reveals what type of recipient and donor organizations and actors will be investigated.

5.3.1 Leading Research Questions 


The analytical framework and the derived working hypothesis outlined above suggest four major questions each case study should clarify: (1) What is to be transferred, or what is the content of civil society assistance? (2) What is the outcome of the assistance granted? (3) Does the outcome positively contribute to civil society development in the sense of the concept of civil society outlined in chapter 2? And finally (4) is a “successful” transfer contingent on the applied donor strategy or on the interaction between donor and recipient or on the domestic setting characterized by domestic actors and their interactions? In other words, can we identify facilitating factors on which a positive outcome of civil society assistance depends?

The most troubling questions are without doubt the determination of the “outcome” of civil society assistance and, by the same token, the question how to measure “successful” transfer. The typical answer to the question how to define success is to determine whether stated objectives of the proponents of transfer have been achieved or not (Jacoby 2000: 11). Related to this work, one can thus state that if the major goals of donors engaging in civil society assistance are realized one can speak of ‘success stories’. However, as also Jacoby observed for his study on the transfer of institutions in post-war Germany and after German unification “since the actors provide no clear and stable standards against which to measure outcomes, analysis cannot center on the question of success” (ibid). By the same token, as has been shown in chapters 3.3. and 3.4., donors that engage in civil society assistance provide only limited accounts of their goals and aspirations. Donors usually define their goals rather vaguely and often quite unrealistically and point more to the functions civil society is assumed to deliver than to observable and clear indicators (Quigley 2000).87 As a result, donor objectives cannot serve as points of reference. Moreover, an analysis that exclusively focuses on realized goals of donors is incapable of detecting unforeseen side - effects of the assistance granted. Those may, however, be more valuable for recipients and more capable of triggering domestic changes than the originally intended goals (Carothers 1996: 96). In response to this problematic, Jacoby avoids focusing on “success” and thus avoids struggling with the troublesome question what “success” might be. Instead he concentrates on the “performance and persistence of transferred institutions” (Jacoby 2000: 11). Performance refers to the benefits to the proponents of transfer, but also includes unintended consequences and outcomes. Persistence covers the rooting of institutions, their reproduction over time and the ways they gain legitimacy in the new society. Moreover, he regards effectiveness as decisive. It occurs,


“…when the transferred institution acquires a legal framework, when it performs in the new society in ways broadly consistent with the aims (promoting efficiency or justice) that led to the transfer attempt, and when it persists by being reproduced over time….”

In short, his measure of institutional transfer encompasses the outcomes of “legality, performance, and persistence” (ibid).

What does this method suggest for the subsequent analysis? Civil society assistance needs to be evaluated by assessing the performance, persistence and effectiveness of transferred structures. We thus need to clarify what structures are transferred by civil society assistance. Chapter 3.4.1. points out that donors more often than not are referring to NGOs if they talk of civil society. Civil society assistance is thus largely restricted to the support and assistance to local NGOs. We may thus conclude that the transferred structures of civil society assistance, are nothing more than supported non-governmental organizations, i.e. major beneficiaries of assistance.


Assessing civil society assistance thus translates into an assessment of major beneficiaries of aid. Who are they? What have their main activities and achievements been? Do they manage to sustain themselves once assistance comes to an end? In brief, the research needs to focus on recipients not on donors. Evaluating donor strategies, projects and activities is a task identified by many as a major goal of research88. However, it accounts for nothing more than an assessment of the output rather than the outcome of assistance. The outcome of assistance is, in contrast, eminent in recipient organizations and especially in their activities and achievements. Moreover, a focus on recipients and their activities allows us to comprise the unintended outcomes and consequences of civil society assistance.

Incorporating the suggestions of Jacoby into this analysis and drawing on the findings of the previous chapters, the general questions guiding the two case studies thus will be:

1. Who are the major beneficiaries of civil society assistance? Are “main recipients” ident i fiable?


This question targets the content of transfer as well as the selectivity of external assistance to civil society (see chapter 3.4.1). Has civil society assistance been distributed equally over the population at large and over the various organizations in existence? Or did the external assistance favor a small segment of society and focus primarily on “some favored cliques” as has been critically put forward by some observers (Wedel 1998). And if this is the case, how can these cliques be characterized?

Main recipients are thereby defined as NGOs that


2. Do civil society organizations that received assistance still exist, and are they able to sustain the m selves?

This question refers to the sustainability of organizations after external allocations come to an end.

3. To what extent are major beneficiaries of external assistance rooted in society?


The question is, whether major recipients of assistance are perceived by the domestic population and by the political elite as legitimate in the sense that they act and perform for the good of the people and the nation or whether they are perceived as “bridgeheads of foreign influence”. This question focuses on the legitimacy of the recipient organization. It allows us to clarify whether recipient organizations that greatly rely on foreign funds and that are internationally well connected are embedded in society or whether they are closer to Western organizations than to the people of their country.

4. Do major recipients of assistance perform and persist in a way that corresponds with the ident i fied characteristics of civil society?

Do main recipients provide important services to other civil society organizations that help raise the quantity and plurality of civil society organizations? Do they facilitate the establishment of networks and horizontal ties among civil society organizations? Do they represent the interests of non-governmental organizations toward government? Do they raise public awareness of democratic rules and procedures and stipulate civic participation, and do they fulfill a watchdog function? In brief, do major recipients of civil society assistance fulfill the democratic functions attributed to civil society, and thus act as carriers of civil society?


In this regard, the analysis will examine the effectiveness of recipients in strengthening the two dimensions of civil society identified in chapter two: the structure of civil society embodied by the quantity and plurality of existing organizations, and the cultural dimension of civil society, portrayed on the one hand by the horizontal relationships among different groups of civil society, and on the other hand in the horizontal relationship with the state. The subsequent analysis thus aims to assess to what extent recipients live up to the normative ideal of civil society outlined in chapter two. This will be done firstly by focusing on the services main NGOs provide to local NGOs thus contributing to NGO development. Secondly, the case studies will focus on examples of common action (e.g. campaigns) and on processes to establish an appropriate NGO legislation. While the former depits the type of relationship among NGOs, the latter illuminates the relations between NGOs and state authorities (see chapter 2.2.3, table 2, p. 44).

5. How benefit recipients from external assistance?

Finally, the analysis will inquire to what extent civil society organizations in Poland and Slovakia benefit from external assistance in fulfilling their various tasks and in acting as carriers of civil society?


Recipient benefits in this regard may come in three different forms of assistance (see also chapter 4.4.2):

All three forms of assistance may translate into domestic bargaining power: Financial resources may raise the capabilities of recipients; knowledge and ideas may trigger learning providing the recipient with a wider range of strategic options; moral support may increase the internal integrity (belief in stated principles and goals) and the external standing (reputation in population, political acceptance) of recipient organizations. Whether this will be the case depends, however, on the given actor constellations in a domestic setting and on recipient perceptions (see chapter 4.4.1). In other words, the question is less what kinds of ”goods“ have been transferred by the donors and more on whether the recipients valued the transferred merits.


Whereas the first two questions that focus on main recipients and their sustainability or “persistence” refer to the output of assistance granted, the questions on recipient’s legitimacy and on recipient’s effectiveness as carriers of civil society are inevitable in order to uncover the actual domestic outcome of civil society assistance. Only a focus upon the rooting of recipient organizations in society and a clarification of the consistency of recipient’s achievements and performance with the normative ideal underlying the concept of civil society allows us to reveal the effect of external assistance that will always be a combined result of external activities and domestic responses. Even if determining the outcome of civil society assistance is the most troubling task the case studies face, it is not the end of the story. The analysis additionally needs to clarify which conditions facilitate a “successful transfer” in the above sense of the word. The question thus is whether recipients’ achievements can be plausibly attributed to donor programs and projects, to donor strategies, to the type of interaction between donor and recipients or to the attractiveness of transferred formal and informal structures to key domestic actors. The last research question aimed at determining the benefits of assistance for recipients therefore aims to illuminate whether the observed outcome is actually a result contingent on donors’ actions. In other words, it clarifies whether the outcome is a result of accident or skill.

The research questions will structure the analysis of the output and outcome of civil society assistance in the two case studies (see chapter 7.4. and 8.4. respectively).

5.3.2 Objects of Research

Summarizing the contemplation above it can be stated that recipients not donors are the major objects of this research. A comparative analysis of donors is not the purpose of the case studies, rather the aim is to analyze how recipients respond to civil society assistance in different contexts. Rather than focusing on a small number of donors and comparing their strategies, which is the prevalent practice in studies focusing on civil society assistance,89 this analysis will center on a small number of recipients, so-called “main recipients” that may be and often are, supported by a wide range of different donor organizations. Having said this, we still cannot neglect the donor side. Firstly, pinpointing donor programs and activities is inevitable in order to determine to what recipients actually react. Secondly, the identification of “main recipients” is hardly possible without an analysis of donor programs and projects. Only an analysis of project lists will reveal which organizations heavily benefited from foreign funds.


A descriptive approach to donors, their programs and projects is thus necessary. Therefore the study aims to draw a broad picture of civil society assistance in both countries. It goes without saying that such an approach impedes a detailed analysis of each donor. It would exceed the scope of this work to focus in detail on every donor active in civil society assistance in both countries. In order to solve this dilemma, the analysis concentrates on four major donors that have been of special importance in the region and also serve as illustrative examples of different types of donors and donor practices. The chosen examples are the European Union as the largest supranational actor, the United States as the largest non-European national donor, and Germany as the largest European national donor active in CEE. Finally, the Soros foundation provides an example of private commitment. In the same manner, in this study it is not possible to cover every single project that benefited non-governmental organizations in one way or another. More often than not, projects in a variety of issue areas, be it environmental protection, decentralization or support to small and mid-sized enterprises (SMEs), are carried out by respective non-state actors such as environmental NGOs, foundations devoted to the goal of local democracy or economic think tanks. For a variety of reasons that have been outlined above (chapter 3.2.) donors are keen on relying on these actors while pursuing other purposes than civil society development. Sure enough, the mere fact that donors finance non-state actors to implement their projects builds up the capacity and institutional strength of the organization in question. In such cases civil society development is, however, not the stated objective of the donor but a (welcome) byproduct. Donors thus often support NGOs indirectly. Although the author is well aware of the effect this donor practice may have on civil society, it is impossible to investigate in a systematic fashion in such projects. Instead, only programs and projects are covered that are explicitly designed by donors with the goal of supporting civil society. In consequence, the study focuses automatically on projects that aim to support what has been called ’the Third Sector‘ (chapter 2.1.6.), as a summary of the major programs focusing on civil society assistance again proves that donors refer to NGOs when mentioning civil society (see also chapter 3.4.1. on this point).

On the side of recipients, the research concentrates on ‘infrastructural organizations’ that aim to strengthen civil society by providing necessary infrastructure and support. The main reason for this lies first in the heavy emphasis placed on those organizations by donors. As will be shown (see chapters 7.4.1. and 8.4.1), infrastructural organizations are among the main beneficiaries of external assistance and are thus to be seen as “main recipients”. Secondly, the number of these major receivers of assistance is surprisingly small and thus does not exceed the scope of this analysis. And finally, infrastructural organizations are regarded as fruitful units of analysis as donors consider those organizations to be important catalysts of civil society development. The assistance to “infrastructural NGOs” is grounded in the belief that the services those organizations provide, ranking from lobbying the government to informing NGOs about funding possibilities to the provision of training, are the missing blocks needed for civil society development. These organizations are thus taken as intermediaries between donors and NGOs on the ground, and are regarded as multipliers of assistance.

Having outlined the major research questions and objects of research, one limitation is in place. A focus on recipients, their sustainability, legitimacy, activities and achievements may translate into a focus on success stories that neglect failures. The present study analyses the output and outcome of civil society assistance, as can be observed in a particular timeframe, namely ten years after the start of foreign assistance. This approach implies that failures in assistance that happened at an earlier period in time are no longer observable. This research does not deal with wasted funds, examples of corruption, or ineffective donor activities. A critical assessment of donor strategies is the price to be paid for determining the effects of assistance in single domestic contexts.

5.4 Measurement Technique


Now that I have clarified the key questions that will guide the subsequent analysis and determined the main units of analysis, the question arises how I will measure recipients’ benefits, their sustainability, legitimacy and effectiveness. The study will rely on different sources of information: available opinion polls, surveys on the NGO sector in Poland and Slovakia, in particular a survey on donor-recipient relationships conducted among Polish and Slovak NGOs, materials of donors and recipient organizations, expert interviews and secondary literature.

The case studies rely heavily on qualitative methods of measurement. However, this does not imply that quantitative methods of measurement are neglected. In line with the recommendation of King et al (1994: 44) ”when we are able to find valid quantitative measures of what we want to know, we should use them“, the study draws on statistical material when available. For example, surveys on Polish NGOs conducted by the Polish non-profit organization KLON/JAWOR in 1993, 1994/1995 and 1997 with the support of the Phare Civic Dialogue Program (BORDO 1998A) were a avaluable source. Moreover, the database of KLON/JAWOR on NGOs in Poland was an important source. This database, which has been in existence since 1990, covers app. 20,000 NGOs operating in Poland. Further important data on Polish and the Slovak non-profit organizations is provided by the Johns Hopkins Comparative Nonprofit Sector Project (see Anheier/Salamon 1999). This project investigates the non-profit sector in 22 countries in comparative perspective. Moreover, the study highly benefits from a survey conducted among Polish and Slovak NGOs in the context of a research project that focused on “democracy promotion and protection in Central and Eastern Europe, the Middle East and North Africa” (DPP).90 The survey in which 72 Polish and 93 Slovak non-governmental organizations participated and which took place in 2002/03 inquires into characteristics of NGOs in Poland and in their relationships with foreign donors. The survey covers three main areas:


The main results and a brief interpretation thereof as well as methodological remarks are to be found in appendix 8. Throughout this analysis, the author will refer to this survey as “DPP survey”.

However, collecting quantitative data is only partly suitable to shed light on the key questions the case studies should answer. The specification of recipients’ benefits, their continuing existence and the degree to which recipients are accepted by the population are variables that are publicly available. It will be more difficult, however, to depict the effectiveness of specific recipient organizations in acting as carriers of civil society, especially as regards what has been called the “cultural dimension” of civil society. This is even more so, if the period under investigation is rather short at least in terms of cultural change. A horizontal relationship inside civil society that is based on tolerance and trust and that is capable of resolving major conflicts in society is as difficult to pinpoint as a change in the vertical relationship between civil society organizations and the state. These indicators are not easily open to empirical analysis. What is needed instead is a focus on unique processes and careful interpretation. One needs to invest in interpretative qualitative case studies, in short narratives that allow us to trace complex processes and produce “understanding via richness, texture, and detail” (McDonald 1996: 10 cit. in Bates et al 1998: 10).

As a result, the case studies will only partly rely on quantifiable indicators and in addition focus on the role recipients played in relevant decision-making processes, namely in legislative processes and public campaigns. The ability of recipients to act in concert as well as the feelings toward the state and the degree to which civil society organizations rely on external assistance, resources and advice is most evident in unique processes that have been identified as relevant for civil society development. These are in the first place public campaigns that demonstrate the ability and confidence of civil society activists to raise their voice and to strife for increasing citizen participation, and secondly, lobbying processes that aim to enhance the legislative environment under which civil society organizations act. Unfortunately the author does not possess the resources and capacity, and in particular language skills, to engage in what Clifford Geertz (1973) calls “thick description”. The analysis therefore makes use of the extensive secondary literature on civil society development in Poland and Slovakia and studies focusing on NGO campaigns and civil society achievements in both countries. In addition, materials of donors and recipients proved to be valuable sources of information on donor and recipients’ strategies, objectives, underlying concepts and activities. And last but not least, the study relies heavily on qualitative expert interviews conducted by the author in both countries under investigation. In order to get a broad as possible picture, experts have been chosen from four different groups: (1) representatives of donor organizations in the home country (if possible) and especially in the recipient country; (2) representatives of recipient organizations; (3) local scholars working on civil society in their country; (4) politicians and representatives of NGOs in the country under investigation (see list of interview partners in appendix 9). A snowballing process has thereby identified valuable interview partners, i.e., all interview partners have been asked whether they can indicate further persons that may be of value for the research.


Before I analyze civil society assistance in Poland and Slovakia while applying the methodology outlined above, a brief description of the major donors under investigation will be given. As all four donors chosen as representative examples for this study are active in both countries they often have the same programs in both countries. For this reason, a summary of the major activities of our four donors in Central and Eastern Europe will serve to introduce the two case studies.

Footnotes and Endnotes

81 Other purposes are e.g. portraying correlations; determining typologies, providing descriptive case studies or historical research. See e.g. Beyme (1988: 52), King et al (1994).

82 For the comparative method, see e.g.: Ragin (1987), Beyme (1988), Collier (1991); Przeworski / Teune (1970).

83 King et al (1994: 119) point out that a determinate research design requires at least one more observation than inferences. Applying the example of successful joint collaboration between countries in building a high-technology weapons system, they come to the conclusion that: “With seven causal variables and only three observations, the research design cannot determine which of the hypotheses, if any, is correct” (ibid: 120). The small N, many variables problem thus hampers causal inference.

84 For a detailed analysis of the cultural preconditions of civil society in Poland and Slovakia see chapter 7.1. and chapter 8.1. respectively.

85 In this regard the author will settle for what King et al view as an “indeterminate research design“ that is only to a limited extent capable of producing generalizations. A determinate research design capable of producing lawlike results is thus sacrificed for a more detailed insight into the unique particularities of civil society assistance in a smaller range of cases. In order to justify this “small N” selection I want to point to Satori’s suggestion that concepts that are applied to a broader range of cases can lead to conceptual “overstretch”. In other words, large N comparisons capable of causal inference deal more often than not with rather uninteresting research questions (cit. in Collier 1991: 14). Or in the words of Charles Tilly: “… with the multiplication of cases and the standardization of categories for comparison the theoretical return declines more rapidly than the empirical return rises” (1984: 144).

86 The method of structured, focused comparison has been developed by George (1979) in order to combine contributions from historians and political scientists for the purpose of developing policy-relevant theory that is grounded in systematic examination of historical experience. George stresses the need to collect information in a systemized way along theoretically identified general questions in order to be able to compare cases in a systemized fashion. A special focus on deviant cases then allows for the cumulative development of theory (George 1979). The method builds the basis of what George and McKeown (1985) then call the congruence method (cit. in King et al 1994: 45).

87 One simply needs to remember how representatives of German political foundations refer to the democratization of Poland as “after we did this” or to the changes in Slovakia as “their biggest succes”’ to understand this point (interview with the author).

88 See e.g. Carothers (1997), Crawford (1996), Robinson (1996).

89 See e.g. Crawford (1996), Robinson (1996), Quigley (1997), Diamond (1997).

90 The project was a combined effort of Professor Claus Offe at the Humboldt-University, Berlin and Professor Philippe C. Schmitter at the European University Institute in Florence and their respective research staff conducted in 1999-2002. It was inspired by the question whether “democratization from without” was at all feasible and aimed to clarify the actual outcome of external involvement in processes of democratization. A part of the project focused on external activities to support civil society assistance.

© Die inhaltliche Zusammenstellung und Aufmachung dieser Publikation sowie die elektronische Verarbeitung sind urheberrechtlich geschützt. Jede Verwertung, die nicht ausdrücklich vom Urheberrechtsgesetz zugelassen ist, bedarf der vorherigen Zustimmung. Das gilt insbesondere für die Vervielfältigung, die Bearbeitung und Einspeicherung und Verarbeitung in elektronische Systeme.
DiML DTD Version 4.0Zertifizierter Dokumentenserver
der Humboldt-Universität zu Berlin
HTML generated: