The Turkish Deep State

State Consolidation, Civil-Military Relations and Democracy

Mehtap Sooyler

• Publisher: Taylor & Francis
• Year: 2015
• Print ISBN: 9781138782334
• eBook ISBN: 9781317668794
• Edition: 1
• Topic: Politics & International Relations
• Subtopic: Politics
• Index: Politics & International Relations

The deep state ranks among the most critical issues in Turkish politics. This book traces its origins and offers an explanation of the emergence and trajectory of the deep state; the meaning and function of informal and authoritarian institutions in the formal security sector of a democratic regime; the involvement of the state in organized crime; armed conflict; corruption; and massive human rights violations.

This book applies an innovative methodological approach to concept formation and offers a mid-range theory of deep state that sheds light on the reciprocal relationship between the state and political regimes and elaborates on the conditions for the consolidation of democracy. It traces the path-dependent emergence and trajectory of the deep state from the Ottoman Empire to the current Turkish Republic and its impact on state-society relations. It reads state formation, consolidation, and breakdown from the perspective of this most resilient phenomenon of Turkish politics. The analysis also situates recent developments regarding AKP governments, including the EU accession process, civil-military relations, coup trials, the Kurdish question, and the Gülen Movement in their context within the deep state. Moreover, this case-study offers an analytical framework for cross-regional comparative analysis of the deep states.

Addressing the lacuna in academic scholarship on the deep state phenomenon in Turkey, this book is essential reading for students and scholars with an interest in democratization, politics and Middle East Studies.

1 Methodology

Sartori stresses the virtues of being a conscious thinker, who in contrast to an unconscious or an over-conscious thinker, “steers a middle course between crude logical mishandling on the one hand, and logical perfectionism (and paralysis) on the other hand”.¹ This chapter follows his guidance and falls into three sections. The first section discusses the characteristics of case studies and examines in depth the inferential potential of the case-study method. The second section addresses the limitations of case studies. The third section introduces five parameters to strengthen the causal inferences of this case study. Inferences are based on 1) a historical approach to causality, specifically, on 2) path-dependent explanations, which identify 3) causal mechanisms and justify hypotheses by applying 4) a process tracing method via theory-driven narrative. Finally, 5) concepts represent the main building blocks of ontology; two-level theory helps to identify the structural relationship between the main components of concepts.

Case-study method and its potential

This section addresses first the controversy among methodologists in defining central concepts by distinguishing between a case, case study, and comparative method. It is argued that “case” and “case study” should be differentiated; the N question does not inevitably apply to the distinction between comparative method and case-study method. Then, it elaborates on the potentials of the case-study method, which operates within the confines of the qualitative research tradition.

A controversy pertains to the distinction between a case, a case study, and the comparative method. For Lijphart,² in contrast to the statistical method, the comparative method covers a small number of cases. However, many methodologists consider that the N question does apply here only to some extent; case studies and the comparative method cannot be differentiated primarily through the number of cases.³ A case is defined as “an instance of a class of events”;⁴ put differently, “an in-depth study of a single unit (a relatively bounded phenomenon) where the scholar’s aim is to elucidate features of a larger class of similar phenomena”.⁵ Gerring distinguishes between case studies that aim at within-case variation and cross-case studies that aim at cross-case variation. Within-case methods may entail three types of variation: temporal variation, spatial variation, or spatial and temporal variation within-unit.⁶

Indeed, “a case study can employ cross-sectional, time series cross sectional, hierarchical, hierarchical time series, and even comparative-historical models”.⁷ To illustrate, an individual unit, such as a revolution, could cover multiple cases depending on the temporal variation, before, during, and after the revolution. Case studies can explore within-unit analysis in a single unit as well as a comparison of a single-unit.⁸ This book involves a within-case temporal variation of a single unit (and the dependent variable): the state. It explores the style of state consolidation under the seventeenth century patrimonial Ottoman Empire, the breakdown of the Ottoman state, the making of the Turkish state and its consolidation in the twentieth century. The deep state, an outcome of state consolidation in modern Turkey, is exposed to temporal variation to study its emergence, rise, and decline.

Another controversy arises on the question whether the case-study method can be considered to be a subset of qualitative methods, as George and Bennett argue,⁹ or whether it belongs to a different category. Yin regards qualitative and quantitative methods as mere types of evidence; case studies may have qualitative or quantitative evidence.¹⁰ Gerring joins him in not making a distinction of qualitative and quantitative methods.¹¹ Mahoney and Goertz¹² show that qualitative or quantitative research methods are traditions distinguished in the first place in their understanding of causality. The following lists characteristics and inferential potentials of this case-study by comparing how quantitative research tradition differs from the qualitative tradition in grasping causal complexity. Case-study method operates within the confines of the latter.

First, quantitative methods aim to explain “causes-of-effects”, while qualitative methods aim to explain “effects-of-causes”.¹³ Case-study methods prize causal mechanisms over causal effects. Types of causal mechanisms that are studied in this book are presented in the third section. Second, quantitative analyses apply correlational causes based on probability assumptions, while qualitative analyses apply necessary and sufficient causes. Third, quantitative methods find out the net effects of individual variables, while qualitative analyses are context-sensitive; they capture the effects of a combination of variables that are assumed as dichotomous or continuous.¹⁴ The third section shows how necessary and sufficient causes and combination of variables are applied in three-level concept generation and two-level theory formation.

Fourth, in quantitative analyses, cases are selected on individual independent variables whose causal effects are generalized, whereas in qualitative analyses, cases refer to specific causal path(s) on the way to dependent variables. Therefore, unlike quantitative studies, qualitative analyses are able to take multiple causation, or “equifinality”,¹⁵ into consideration.¹⁶ The third section introduces a specific method to study causal path(s): path-dependent explanations. This single case study illustrates equifinality in the theoretical chapter concerning the development of the deep state phenomenon. Peter Hall suggests that studying a causal path allows us to detect omitted variables that are related to the dependent variable.¹⁷ Here, the theoretical chapter critically assesses the Barrington Moore research program on regime change and integrates the state and cleavage structures, omitted variables, into the analysis of regime change.

Fifth, according to Bennett and Elman,¹⁸ case studies offer more parsimony in analyzing interaction effects. Quantitative methods assume that causal variables are not affected by the dependent variable. Case studies capture interaction effects illustrated by reciprocal causation, i.e. endogeneity. The causal paths studied in this book reveal the endogeneity between the state and political regime. Sixth, in qualitative analyses observations are weighted; any diverging individual observation can disprove the theory. In quantitative analyses, on the other hand, observations are considered equal.¹⁹ The weighting of observations is explicitly elaborated in the theoretical chapter.

Finally, concepts are the building blocks of qualitative analyses and misfits or errors require a revision of the ontological basis of the research. Measurement errors are attached to problems related to the conceptual structure. In quantitative analyses, however, measurement errors are eliminated only if they result in a systematic error. Mahoney and Goertz²⁰ argue that clarification of concepts increases conceptual validity and makes the ontological basis of the research sounder. We can conclude that quantitative methods are more prone to “conceptual stretching”,²¹ or too vague and amorphous conceptualizations, than qualitative methods. As put succinctly by Sartori: “quantification enters the scene after, and only after, having formed the concept … the rules of concept formation are independent of, and cannot be derived from, the rules which govern the treatment of quantities and quantitative relations”.²²

Limitations of the case-study method

Limitations of case-study methods are related in the literature to the lack of hypothesis testing, case selection bias, identifying scope conditions and necessity, lack of representativeness, and degrees of freedom. The following discusses each limitation and their relevance to this book and asks whether they are inevitable. It is argued that some of these limitations can be traced back to divergent conceptions of observations, cases, theory testing, and validation.

The first limitation relates to the conventional view that case studies are useful for generating hypotheses rather than testing them.²³ Hall maintains that the low regard of case studies stems from the pervasive confusion about the definition of a case and theory testing, which can be traced back to Eckstein²⁴ and Lijphart,²⁵ who claimed that testing a theory is viable if it is based on observations on the dependent variable and a few independent variables. This assumption reduces a case to a single unit; consequently, it reduces observation to data drawn from that unit. More importantly, it reduces small-N comparison or comparative method in general to a less important version of statistical analysis.²⁶ Rueschemeyer traces this misunderstanding to the mistaken sharp distinction made between the “context of discovery” and the “context of validation”.²⁷ According to George and Bennett, this misjudgment is reflected in Karl Popper’s²⁸ proposition that there can be no logical method of discovering ideas.²⁹ However, his distinction overlooks that the process of making discoveries includes both explanation and testing. This book generates hypotheses, but at the same time tests hypotheses.

The second limitation relates to selection bias, which is according to Bennett and Elman³⁰ often misleading. A preconstituted population is not applicable to case studies. Furthermore, this limitation does not apply to within-case process tracing or causal process observations – applied in this book – since the process connects the causes to the outcome.³¹ Third, George and Bennett³² address the issue of identifying scope conditions and necessity: “Case studies remain much stronger at assessing whether and how a variable mattered to the outcome than at assessing how much it mattered”.³³ Therefore, we should identify precisely the unit(s) we refer to in our claims of causality with respect to necessary condition. Gerring³⁴ warns against possible caveats. We need to specify secondary units that are dealt with besides the primary unit under study. If not taken into account, the priority of the primary unit can be blurred. In this study, the state is the primary unit, while institutions, cleavages, and elites are secondary units. Gerring adds that, in comparing a single unit with other units, the boundaries of a single- and cross-unit analysis should not be conflated. Moreover, in order to identify scope conditions and necessity, scholars should study the unit, but relate it to a group of cases.³⁵ The theoretical chapter analyzes the state by relating it to Western European cases and confines the analysis to the deep state in a particular political regime, i.e. to the deep state in democracies which are the gray zone between autocracy and consolidated democracy.

Closely related...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright Page
• Table of Contents
• List of tables and figures
• Abbreviations
• Acknowledgments
• Introduction
• 1. Methodology
• 2. Theoretical framework
• 3. Banditry and the Ottoman State
• 4. The reorganization and restoration of the deep state
• 5. The rise, decline and restoration of the deep state
• 6. Conclusion
• Bibliography
• Index