Scope conditions and case studies

Scope conditions: autocracy, independence, and statehood

The monarchic peace is set in the scope of independent authoritarian nationstate monarchies. The assumption underlying these scope conditions is that these monarchies have a distinct functional logic. The weight of regime type overrides the importance of state form in democratic regimes. Formal or “democratic” monarchies where royalty mostly entails a merely ceremonial or representative role have many more similarities with their republican democratic neighbors than with authoritarian monarchies in that the power is shifted away from unelected royalty toward democratic institutions, such as parliament, the prime ministry, parties, and courts. The Netherlands is usually best compared to Germany or Austria, not Saudi Arabia or Swaziland.

All MENA monarchies are unanimously considered authoritarian, thus obliterating potential definitional problems here. This region is best suited to a comparative analysis because it hosts the only remaining significant concentration of ruling and reigning monarchs in the world. The restriction to nation-state monarchies is essential for two reasons: it sets apart these monarchies from substate monarchies that persist today (mostly in Africa - e.g. in Ghana, Namibia, Togo, and Uganda) and it fills the timeframe cutoff at 1945 with meaning, as most nation-states outside the Western world consolidated after the Second World War.

Starting long-range analysis after World War II has become a default and sometimes arbitrary option for political scientists. Regarding an analysis of monarchic foreign policy, it also mows down the number of cases significantly and thus restricts the analysis. Before the world wars, monarchies constantly waged war against one another, negating the notion of any sort of monarchic peace. However, in this case, there are valid reasons for the cutoff that trump any potential confirmation bias. In fact, setting the cutoff at 1945 does not exclude many potential cases, because the monarchies and their context are too different to make meaningful comparisons for three main reasons, which follow.

First, before 1945 (and especially before 1914). monarchies tended to be empires to which a different functional logic applied. This fact has wide-ranging theoretical implications. Empires are defined by vague and fluid borders and thus an inlierent expansionary drive, making them prone to territorial irredentism via expansion and invasion by or of other territories. Territorial conflicts and wars are thus much more likely for empires and have, in fact, constituted the largest share of wars during the era of empires (see e.g. Wimmer and Min 2006).

Their relations between foreign and domestic politics is different from that of modern nation-states, which are based on a delimitated (even if contested) territory, and there are therefore far fewer territorial disputes. Military aggression for territory has become largely delegitimized in recent decades because the international law norm of territorial integrity has become more ingrained in state interactions; in fact, military disputes resulting in a major change in territory have sharply dropped since the second half of the 20th century (Zacher 2001). When this principle is encroached upon, an immediate and strong reaction follows. One obvious example is the reaction to Saddam Hussein's invasion of Kuwait, and another, more-recent one is the international outrage over the annexation of Crimea by the Russian Federation in 2014.

In the "age of nationalism”, territorial conflict resulting in militant revisionism is much less likely than it was in the "age of dynasties”, to follow John Vasquez’s terminology, because ethnolinguistic boundaries are more stable than dynastic (imperial) rule that hinges on a particular ruler’s persona, ambitions, and power from which an empire’s borders are derived. In "dynasty-era” monarchies, every succession left borders open to renegotiation, sometimes by violent means; the fact that the successor could also be an outsider if they produced a legitimate claim or could force their way on the throne was a further destabilizing factor (Vasquez 2009, 161-165). This is by now almost unconceivable. Even the succession pool of the traditionalist conservative Arab monarchies in existence today is, however indeterminate, not questioned by outsiders.22 By setting the limit on 1945, we thus eliminate empires and their inevitable territorial claims as accepted casus belli, allowing for a clearer analysis.

Second, this delimitation applies even more for the Middle East monarchies as most of them had not even come into existence before the 20th century, much less formed states, nation-states or otherwise. State building and institutional and territorial consolidation had not achieved for the Gulf monarchies and Jordan until well into the 20th century. Before, they were societies governed by tribal law and (semi-)nomadicism (Zahlan 2002, 24). The multi-centennial dynastic tradition of monarchies such as Egypt and Iran was affected heavily by external influence and control by colonialist powers, making them states, but often not sovereign. The nation-state condition also largely overlaps with the scope condition of independence in the region. Foreign policies of states that are controlled by foreign powers cannot be analyzed without addressing colonial dynamics first.23 The first part of the following case study on Saudi-Hashemite relations illustrates the importance of this scope condition.

The third reason is related to the main theoretical claim of this work regarding inter-monarchical behavior. It centers on the argument that the reason why monarchies don't fight each other is because of a strong ingroup identification and the existence of different codes of conduct inside and outside the ingroup. However, such an ingroup develops in a differential process of othering only in contrast to other groups consequently defined as outgroups. As illustrated earlier, this happens only when the ingroup is not the "default” category anymore and can thus be considered a salient category, as ingroup solidarity can take hold only then. A sense of common threat accelerates the process. Before 1945 and especially 1914, when most states were monarchies, monarchy was hardly a defining ingroup, because it was the "default” category. States tended to bond along different lines such as Protestantism or Catholicism during the age of the Wars of Religion, a cleavage embedded even in the names of the two blocs: the Protestant Union and the Catholic League. When in the 20th century monarchies started to be successively replaced by republics and the monarchs were robbed of their power or acquiesced into transferring it to a parliamentarian system, it started to consolidate into a meaningful category once again; only then were the conditions for a monarchic peace to occur met.24

The following analysis will focus on the regime or the ruling elite of the monarchies, not on the state per se nor on the level of the population. Government, regime, and state are not interchangeable. In the definition of Robert Fishman, a regime is

the formal and informal organization of the center of political power, and of its relations with the broader society. A regime determines who has access to power, and how those who are in power deal with those who are not. . . . Regimes are more permanent forms of political organization than specific governments, but they are typically less permanent than the state. The state, by contrast, is a (normally) more permanent structure of domination and coordination including a coercive apparatus and the means to administer a society and extract resources from it.

(Fishman 1990, 428)25

Here the relevant decision makers of the regime are identified as core members of the ruling family (usually including, but not limited to, the monarch, the heir apparent, and ruling family members in prime and foreign minister positions). Sometimes commoners can be included as well if they wield a significant amount of power and have decision-making capabilities inside the monarchic framework, especially in linchpin monarchies, where the ruling family is too small to control the regime on its own. An example is the iconic role played by Nuri al-Said, the long-time prime minister of Hashemite Iraq.

The definition of the "Middle East” is not consistent in everyday language or academic use. There are different levels of broadness of the concept with the broadest including even North Africa (thus making MENA a pleonasm). The relevant past and present (independent) monarchies that at some point have

Moving beyond democratic exceptionalism 49 constituted part of a regional system are Morocco since 1956; Libya 1951-1969; Egypt 1922-1952; Jordan since 1946; Iraq 1932-1958; Iran 1941—1979;26 Kuwait since 1961; Saudi Arabia since 1932; the United Arab Emirates. Bahrain, Qatar, and Oman since 1971; and North Yemen 1918-1962.27 Therefore, a broad definition of the Middle East will be used to cover the countries from Morocco to Iran.28

Multiple comparative case study approach

A case study approach is used for the following analysis for methodological and epistemological reasons: given that both monarchies and war are rare, medium-n or small-n approaches lend themselves more easily to the task than large-n studies. They can also serve to better solve the "too many variables, too few cases’’ problem that imposes itself in view of the many shared complementary factors laid out earlier. Therefore, we must find analytical units that are separable to untangle the complex web of shared traits, transnational diffusion, and interdependence, which is more difficult in a large-n-framework.

Furthermore, as shown earlier, some theoretical shortcomings are the consequence of an overreliance on large-n comparisons. A major reason for the preference of a case study framework is the prime motivation of the book - theory building and the laying out of the causal mechanism - to which such a framework is especially well suited.

For this purpose, a qualitative comparative-historical case study approach (Gerring 2007, 27-28) consisting of multiple, both synchronic and diachronic, case studies is chosen as the basis of the analysis, which, following George and Bennett, is also specified as a "building block” case study that serves to identify common patterns of particular types or subtypes of more-general phenomena (2005, 76) - here the monarchic peace as a subtype of the SPSP.

Since monarchies in the Middle East share so many features and therefore form similar political systems, monarchic foreign policy behavior in the Middle East is an especially likely case of an SPSP, rendering it in a good position for a plausibility check of the theory. To buttress the analysis and given data access problems associated with studies on the Middle East and authoritarian regimes in general (Ahram and Goode 2016), different case study designs and diversified data sources (statistical data, primary and secondary sources, and field interviews conducted in Qatar in April 2013 and Jordan in September 2015) are used for methodological triangulation.

The comparative case study consists of two main parts. The first entails an overview over Middle East monarchies and a "most likely” case that serves to plausibilize the mechanism of ingroup identification leading to ingroup favoritism under exceptionally favorable conditions, when monarchies face a common threat and contextualize the monarchic peace in a regional context. During periods of common threat to monarchies, ingroup identification is expected to be especially high and war therefore highly unlikely. Instead, secondary implications of ingroup favoritism are seen clearly in patterns of alliance and monarchic solidarity. The case study serves to illuminate the theorized mechanism of a strong coalescenteffect of external threats on ingroup formation and the ingroup favoritism that follows it. The cases chosen are two periods when monarchies faced a strong common threat not as individual states but as monarchies, which promises a particularly strong effect on monarchic cohesion: first, during the Arab Cold War in the 1950s and 1960s, and second, following the wave of rebellions and instability sweeping the region from 2011 to 2015.

The second part consists of two-by-two case studies that serve to trace the mechanism in less-favorable conditions for ingroup identification, namely during intragroup conflict between dyads (a triad in the Saudi-Hashemite case). All four are set in a context of intra-monarchic conflict. The study combines cross-case and within-case inference. While the four case studies form a most dissimilar case design to enhance generalizability (cross-case comparison), each case will be based on theory-testing process tracing (Beach and Pedersen 2011), to identify the causal mechanism responsible amid other possible pathways (within-case comparison). Following the advice of Goertz, both pairs start with a typical case where the causal mechanism can be expected to work best to clarify the causal process. The "subsequent case studies then explore the limits of the causal mechanism: how general is it?” (Goertz 2017, 71). The last two of the case studies further try to disentangle the effect of the monarchic political system from other factors by juxtaposing two periods of dyadic relations in which one monarchy broke down.

The four cases were also chosen to cover a set that is as diverse as possible to enhance the generalizability of the findings. They feature at least eight countries (Bahrain. Qatar, Saudi Arabia, Jordan, Iraq, Kuwait, Iran, and UAE) that span over two subregions (Persian Gulf and Levante), cover a wide time period covering different stages of the development of the regional system (from the 1940s to the 2000s), have different issues at their core (territory vs. national power and regional hegemony), and include different types of monarchies: rentier and non-rentier states, dynastic and linchpin monarchies, major regional powers, and small states in different arrangements (from small-small Bahrain-Qatar to large-large Saudi-Hashemites). They also include Iran as a non-Arab, non-Sunni state to refine the concept of similarity and isolate the shared monarchism as the relevant feature.

Two additional episodes of intra-monarchic conflict will be examined in a shorter fashion: the Saudi-Yemeni War of 1934 and the Qatar Crisis since 2017.

“Near misses conflicts that failed to escalate into wars

Because there are no cases of inter-monarchic war since the consolidation of the Middle East state system and because proving a negative is demanding, other approximations to war and the monarchic effect on it must be found to further analytically separate political system effects from other factors.

Therefore, the first pair of case studies will be "near misses” of inter-monarchic war in Chapter 4, cases where monarchies have de-escalated major conflicts before they erupted into wars. This approach has been fruitfully applied in the

Moving beyond democratic exceptionalism 51 examination of the democratic peace theory, which also lacks instances of contradictory cases (inter-democratic war) (see Layne 1994). The chosen cases must be from among cases of monarchic dyads that experienced "close calls” of conflicts severe enough that war or at least militarized dispute can be seen as a realistic possible outcome. They show how important the institutional similarity was vis-à-vis other factors for decisions not to escalate.

“Quasi-experiments monarchic breakdown leading to escalation

Despite the attempts to entangle monarchic effects from other possible confounding factors in the previous case studies, there is still a degree of overlap that needs to be further separated. Two case studies of Chapter 5 therefore look at two "quasi-experiments” in a diachronic study : pairs of monarchies where one experiences regime change. A regime change by revolution or coup would hardly affect oil wealth, geographic location, and capabilities in terms of balance-of-power approaches; however, it is likely to change alliance patterns and possibly the militarism of the new regime. From a perfectly controlled environment, though, it is a valuable method to gauge the effect of the political system. Large-n studies with a similar basis have been successfully conducted for the DPT (Hensel, Goertz, and Diehl 2000).

< Prev   CONTENTS   Source   Next >