Here’s my first post about the book. In this post I’ll continue my coverage where I left off in my first post. A few of the chapters covered below I did not think very highly of, but other parts of the coverage are about as good as you could expect (given problems such as e.g. limited data etc.). Some of the stuff I found quite interesting. As people will note in the coverage below the book does address the religious dimension to some extent, though in my opinion far from to the extent that the variable deserves. An annoying aspect of the chapter on religion was to me that although the author of the chapter includes data which to me cannot but lead to some very obvious conclusions, the author seems to be very careful avoiding drawing those conclusions explicitly. It’s understandable, but still annoying. For related reasons I also got annoyed at him for presumably deliberately completely disregarding which seems in the context of his own coverage to be an actually very important component of Huntington’s thesis, that conflict at the micro level seems to very often be between muslims and ‘the rest’. Here’s a relevant quote from Clash…, p. 255:
“ethnic conflicts and fault line wars have not been evenly distributed among the world’s civilizations. Major fault line fighting has occurred between Serbs and Croats in the former Yugoslavia and between Buddhists and Hindus in Sri Lanka, while less violent conflicts took place between non-Muslim groups in a few other places. The overwhelming majority of fault line conflicts, however, have taken place along the boundary looping across Eurasia and Africa that separates Muslims from non-Muslims. While at the macro or global level of world politics the primary clash of civilizations is between the West and the rest, at the micro or local level it is between Islam and the others.”
This point, that conflict at the local level – which seems to be the type of conflict level you’re particularly interested in if you’re researching civil wars, as also argued in previous chapters in the coverage – according to Huntington seems to be very islam-centric, is completely overlooked (ignored?) in the handbook chapter, and if you haven’t read Huntington and your only exposure to him is through the chapter in question you’ll probably conclude that Huntington was wrong, because that seems to be the conclusion the author draws, arguing that other models are more convincing (I should add here that these other models do seem useful, at least in terms of providing (superficial) explanations; the point is just that I feel the author is misrepresenting Huntington and I dislike this). Although there are parts of the coverage in that chapter where I feel that it’s obvious the author and I do not agree, I should note that the fact that he talks about the data and the empirical research makes up for a lot of other stuff.
Anyway, on to the coverage – it’s perhaps worth noting, in light of the introductory remarks above, that the post has stuff on a lot of things besides religion, e.g. the role of natural resources, regime types, migration, and demographics.
“Elites seeking to end conflict must: (1) lead followers to endorse and support peaceful solutions; (2) contain spoilers and extremists and prevent them from derailing the process of peacemaking; and (3) forge coalitions with more moderate members of the rival ethnic group(s) […]. An important part of the two-level nature of the ethnic conflict is that each of the elites supporting the peace process be able to present themselves, and the resulting terms of the peace, as a “win” for their ethnic community. […] A strategy that a state may pursue to resolve ethnic conflict is to co-opt elites from the ethnic communities demanding change […]. By satisfying elites, it reduces the ability of the aggrieved ethnic community to mobilize. Such a process of co-option can also be used to strengthen ethnic moderates in order to undermine ethnic extremists. […] the co-opted elites need to be careful to be seen as still supporting ethnic demands or they may lose all credibility in their respective ethnic community. If this occurs, the likely outcome is that more extreme ethnic elites will be able to capture the ethnic community, possibly leading to greater violence.
It is important to note that “spoilers,” be they an individual or a small sub-group within an ethnic community, can potentially derail any peace process, even if the leaders and masses support peace (Stedman, 2001).”
“Three separate categories of international factors typically play into identity and ethnic conflict. The first is the presence of an ethnic community across state boundaries. Thus, a single community exists in more than one state and its demands become international. […] This division of an ethnic community can occur when a line is drawn geographically through a community […], when a line is drawn and a group moves into the new state […], or when a diaspora moves a large population from one state to another […] or when sub-groups of an ethnic community immigrate to the developed world […] When ethnic communities cross state boundaries, the potential for one state to support an ethnic community in the other state exists. […] There is also the potential for ethnic communities to send support to a conflict […] or to lobby their government to intervene […]. Ethnic groups may also form extra-state militias and cross international borders. Sometimes these rebel groups can be directly or indirectly sponsored by state governments, leading to a very complex situation […] A second set of possible international factors is non-ethnic international intervention. A powerful state may decide to intervene in an ethnic conflict for a variety of reasons, ranging from humanitarian support, to peacekeeping, to outright invasion […] The third and last factor is the commitment of non-governmental organizations (NGOs) or third-party mediators to a conflict. […] The record of international interventions in ethnic civil wars is quite mixed. There are many difficulties associated with international action [and] international groups cannot actually change the underlying root of the ethnic conflict (Lake and Rothchild, 1998; Kaufman, 1996).”
“A relatively simple way to think of conflict onset is to think that for a rebellion to occur two conditions need to be satisfactorily fulfilled: There must be a motivation and there must be an opportunity to rebel.3 First, the rebels need a motive. This can be negative – a grievance against the existing state of affairs – or positive – a desire to capture resource rents. Second, potential rebels need to be able to achieve their goal: The realization of their desires may be blocked by the lack of financial means. […] Work by Collier and Hoeffler (1998, 2004) was crucial in highlighting the economic motivation behind civil conflicts. […] Few conflicts, if any, can be characterized purely as “resource conflicts.” […] It is likely that few groups are solely motivated by resource looting, at least in the lower rank level. What is important is that valuable natural resources create opportunities for conflicts. To feed, clothe, and arm its members, a rebel group needs money. Unless the rebel leaders are able to raise sufficient funds, a conflict is unlikely to start no matter how severe the grievances […] As a consequence, feasibility of conflict – that is, valuable natural resources providing opportunity to engage in violent conflict – has emerged as a key to understanding the relation between valuable resources and conflict.”
“It is likely that some natural resources are more associated with conflict than others. Early studies on armed civil conflict used resource measures that aggregated different types of resources together. […] With regard to financing conflict start-up and warfare the most salient aspect is probably the ease with which a resource can be looted. Lootable resources can be extracted with simple methods by individuals or small groups, are easy to transport, and can be smuggled across borders with limited risks. Examples of this type of resources are alluvial gemstones and gold. By contrast, deep-shaft minerals, oil, and natural gas are less lootable and thus less likely sources of financing. […] Using comprehensive datasets on all armed civil conflicts in the world, natural resource production, and other relevant aspects such as political regime, economic performance, and ethnic composition, researchers have established that at least some high-value natural resources are related to higher risk of conflict onset. Especially salient in this respect seem to be oil and secondary diamonds […] The results regarding timber […] and cultivation of narcotics […] are inconclusive. […] [An] important conclusion is that natural resources should be considered individually and not lumped together. Diamonds provide an illustrative example: the geological form of the diamond deposit is related to its effect on conflict. Secondary diamonds – the more lootable form of two deposit types – makes conflict more likely, longer, and more severe. Primary diamonds on the other hand are generally not related to conflict.”
“Analysis on conflict duration and severity confirm that location is a salient factor: resources matter for duration and severity only when located in the region where the conflict is taking place […] That the location of natural resources matters has a clear and important implication for empirical conflict research: relying on country-level aggregates can lead to wrong conclusions about the role of natural resources in armed civil conflict. As a consequence of this, there has been effort to collect location-specific data on oil, gas, drug cultivation, and gemstones”.
“a number of prominent studies of ethnic conflict have suggested that when ethnic groups grow at different rates, this may lead to fears of an altered political balance, which in turn might cause political instability and violent conflict […]. There is ample anecdotal evidence for such a relationship [but unfortunately little quantitative research…]. The civil war in Lebanon, for example, has largely been attributed to a shift in the delicate ethnic balance in that state […]. Further, in the early 1990s, radical Serb leaders were agitating for the secession of “Serbian” areas in Bosnia-Herzegovina by instigating popular fears that Serbs would soon be outnumbered by a growing Muslim population heading for the establishment of a Shari’a state”.
“[One] part of the demography-conflict literature has explored the role of population movements. Most of this literature […] treats migration and refugee flows as a consequence of conflict rather than a potential cause. Some scholars, however, have noted that migration, and refugee migration in particular, can spur the spread of conflict both between and within states […]. Existing work suggests that environmentally induced migration can lead to conflict in receiving areas due to competition for scarce resources and economic opportunities, ethnic tensions when migrants are from different ethnic groups, and exacerbation of socioeconomic “fault lines” […] Salehyan and Gleditsch (2006) point to spill-over effects, in the sense that mass refugee migration might spur tensions in neighboring or receiving states by imposing an economic burden and causing political stability [sic]. […] Based on a statistical analysis of refugees from neighboring countries and civil war onset during the period 1951–2001, they find that countries that experience an influx of refugees from neighboring states are significantly more likely to experience wars themselves. […] While the youth bulge hypothesis [large groups of young males => higher risk of violence/war/etc.] in general is supported by empirical evidence, indicating that countries and areas with large youth cohorts are generally at a greater risk of low-intensity conflict, the causal pathways relating youth bulges to increased conflict propensity remain largely unexplored quantitatively. When it comes to the demographic factors which have so far received less attention in terms of systematic testing – skewed sex ratios, differential ethnic growth, migration, and urbanization – the evidence is somewhat mixed […] a clear challenge with regard to the study of demography and conflict pertains to data availability and reliability. […] Countries that are undergoing armed conflict are precisely those for which we need data, but also those in which census-taking is hampered by violence.”
“Most research on the duration of civil war find that civil wars in democracies tend to be longer than other civil wars […] Research on conflict severity finds some evidence that democracies tend to see fewer battledeaths and are less likely to target civilians, suggesting that democratic institutions may induce some important forms of restraints in armed conflict […] Many researchers have found that democratization often precedes an increase in the risk of the onset of armed conflict. Hegre et al. (2001), for example, find that the risk of civil war onset is almost twice as high a year after a regime change as before, controlling for the initial level of democracy […] Many argue that democratic reforms come about when actors are unable to rule unilaterally and are forced to make concessions to an opposition […] The actual reforms to the political system we observe as democratization often do not suffice to reestablish an equilibrium between actors and the institutions that regulate their interactions; and in its absence, a violent power struggle can follow. Initial democratic reforms are often only partial, and may fail to satisfy the full demands of civil society and not suffice to reduce the relevant actors’ motivation to resort to violence […] However, there is clear evidence that the sequence matters and that the effect [the increased risk of civil war after democratization, US] is limited to the first election. […] civil wars […] tend to be settled more easily in states with prior experience of democracy […] By our count, […] 75 percent of all annual observations of countries with minor or major armed conflicts occur in non-democracies […] Democracies have an incidence of major armed conflict of only 1 percent, whereas nondemocracies have a frequency of 5.6 percent.”
“Since the Iranian revolution in the late 1970s, religious conflicts and the rise of international terror organizations have made it difficult to ignore the facts that religious factors can contribute to conflict and that religious actors can cause or participate in domestic conflicts. Despite this, comprehensive studies of religion and domestic conflict remain relatively rare. While the reasons for this rarity are complex there are two that stand out. First, for much of the twentieth century the dominant theory in the field was secularization theory, which predicted that religion would become irrelevant and perhaps extinct in modern times. While not everyone agreed with this extreme viewpoint, there was a consensus that religious influences on politics and conflict were a waning concern. […] This theory was dominant in sociology for much of the twentieth century and effectively dominated political science, under the title of modernization theory, for the same period. […] Today supporters of secularization theory are clearly in the minority. However, one of their legacies has been that research on religion and conflict is a relatively new field. […] Second, as recently as 2006, Brian Grim and Roger Finke lamented that “religion receives little attention in international quantitative studies. Including religion in cross-national studies requires data, and high-quality data are in short supply” […] availability of the necessary data to engage in quantitative research on religion and civil wars is a relatively recent development.”
“[Some] studies [have] found that conflicts involving actors making religious demands – such as demanding a religious state or a significant increase in religious legislation – were less likely to be resolved with negotiated settlements; a negotiated settlement is possible if the settlement focused on the non-religious aspects of the conflict […] One study of terrorism found that terror groups which espouse religious ideologies tend to be more violent (Henne, 2012). […] The clear majority of quantitative studies of religious conflict focus solely on inter-religious conflicts. Most of them find religious identity to influence the extent of conflict […] but there are some studies which dissent from this finding”.
“Terror is most often selected by groups that (1) have failed to achieve their goals through peaceful means, (2) are willing to use violence to achieve their goals, and (3) do not have the means for higher levels of violence.”
“the PITF dataset provides an accounting of the number of domestic conflicts that occurred in any given year between 1960 and 2009. […] Between 1960 and 2009 the modified dataset includes 817 years of ethnic war, 266 years of genocides/politicides, and 477 years of revolutionary wars. […] Cases were identified as religious or not religious based on the following categorization:
1 Not Religious.
2 Religious Identity Conflict: The two groups involved in the conflict belong to different religions or different denominations of the same religion.
3 Religious Wars: The two sides of the conflict belong to the same religion but the description of the conflict provided by the PITF project identifies religion as being an issue in the conflict. This typically includes challenges by religious fundamentalists to more secular states. […]
The results show that both numerically and as a proportion of all conflict, religious state failures (which include both religious identity conflicts and religious wars) began increasing in the mid-1970s. […] As a proportion of all conflict, religious state failures continued to increase and became a majority of all state failures in 2002. From 2002 onward, religious state failures were between 55 percent and 62 percent of all state failures in any given year.”
“Between 2002 and 2009, eight of 12 new state failures were religious. All but one of the new religious state failures were ongoing as of 2009. These include:
• 2002: A rebellion in the Muslim north of the Ivory Coast (ended in 2007)
• 2003: The beginning of the Sunni–Shia violent conflict in Iraq (ongoing)
• 2003: The resumption of the ethnic war in the Sudan [97% muslims, US] (ongoing)
• 2004: Muslim militants challenged Pakistan’s government in South and North Waziristan. This has been followed by many similar attacks (ongoing)
• 2004: Outbreak of violence by Muslims in southern Thailand (ongoing)
• 2004: In Yemen [99% muslims, US], followers of dissident cleric Husain Badr al-Din al-Huthi create a stronghold in Saada. Al-Huthi was killed in September 2004, but serious fighting begins again in early 2005 (ongoing)
• 2007: Ethiopia’s invasion of southern Somalia causes a backlash in the Muslim (ethnic- Somali) Ogaden region (ongoing)
• 2008: Islamist militants in the eastern Trans-Caucasus region of Russia bordering on Georgia (Chechnya, Dagestan, and Ingushetia) reignited their violent conflict against Russia (ongoing)” [my bold]
“There are few additional studies which engage in this type of longitudinal analysis. Perhaps the most comprehensive of such studies is presented in Toft et al.’s (2011) book God’s Century based on data collected by Toft. They found that religious conflicts – defined as conflicts with a religious content – rose from 19 percent of all civil wars in the 1940s to about half of civil wars during the first decade of the twenty-first century. Of these religious conflicts, 82 percent involved Muslims. This analysis includes only 135 civil wars during this period. The lower number is due to a more restrictive definition of civil war which includes at least 1,000 battle deaths. This demonstrates that the findings presented above also hold when looking at the most violent of civil wars.” [my bold]