Civil War Onset and the 'Third Debate'

Scientific Essay, 2010

17 Pages, Grade: 90%




Research Approaches

IR Theoretical Assumptions and Parallels

Ontology, Epistemology and Methodology

Agency and Structure

Missing Variables, Effectiveness, Contributions and Further Research

One Methodology’s Weakness is another Methodology’s Strength

Implications for Policy



According to Yosef Lapid, International Relations, by the end of the 1980’s was “in the midst of a third discipline-defining debate”[1] (Lapid 1989: 236) between positivism and post-positivism (Smith 1996: 11-13). Scholars studying the phenomenon of civil war and its causes seem to have been largely exempt from this debate. There are two discernable reasons for this. First, the study of civil war has long been marginalised by the dominant theoretical paradigm of Realism within security studies, which does not concern itself with war, within the intrastate system. This is surprising, as since the start of the millennium, intrastate conflict has been far more prevalent than interstate conflict (Mack 2002: 516). Second, within the academic field of civil war onset, which this paper focuses on specifically, the theoretical approach is primarily positivist. Most literature on the subject of civil war, focuses around the so called ‘greed-grievance debate’ (Murshed & Tadjoeddin 2009: 87-89), and though the latter does usually focus on identity, it remains essentially positivist (Aspinall 2007: 953). This however, does not mean that the ‘Third Debate’ does not apply to the academic field of civil war onset as this paper will show.

This paper analyses two journal articles, one positivist: Greed and Grievance in Civil War (2004) by Paul Collier and Anke Hoeffler, and one post-positivist: The Construction of Grievance: Natural Resources and Identity in a Separatist Conflict (2007) by Edward Aspinall. Using these articles, this paper will compare and evaluate their research approaches, their worth and effectiveness in addressing the subject and research question and their contribution to knowledge. Finally this paper will give some recommendations as to future areas of inquiry.

These two articles are compatible for analysis, primarily because they ask comparatively similar research questions. The Collier and Hoeffler (hereafter referred to as C&H) article develops an “econometric model which predicts the outbreak of civil conflict” (Collier & Hoeffler 2004: 563), and are thus primarily concerned with the question: what causes civil war? Edward Aspinall is more explicit and begins his article with the research question: “when and under what circumstances does natural resource extraction give rise to violent conflict?” (Aspinall 2007: 950). The fact that Aspinall focuses on natural resources in relation to civil war onset specifically, does not make the articles incompatible, instead it results from the dominant discourse regarding civil war onset (including the C&H article) focussing on the importance natural resources (ibid: 950-951).

Research Approaches

This paragraph compares and evaluates the two articles mentioned above with regards to their different research approaches. This is accomplished by focussing on the following factors; IR theoretical assumptions and parallels; ontology, epistemology and methodology; agency and structure. Each of these are discussed in the following sub-sections starting with the C&H article and subsequently compared to Aspinall’s article. Interpretations of the definition of (post-)positivism are primarily given in the sub-section ontology, epistemology and methodology.

IR Theoretical Assumptions and Parallels

The Collier and Hoeffler article’s central aim is to build a predictive, econometric model of civil war onset. Their quantitative study, using 98 countries, employs a logit regression analysis in which they proxy for opportunity (greed) and objective grievances. C&H argue that greed variables hold more explanatory power (Collier & Hoeffler 2004: 587-588). These variables are greatly influenced by economic theory literature (ibid: 564-565), resulting in an overall theory which “juxtaposes the opportunities of rebellion against the constraints” (ibid: 564). They identify the following factors as having an effect on the probability of civil war onset; finance (availability of lootable resources, diaspora funding), cost of rebellion in the form earnings foregone in rebellion (male secondary school enrolment, GDP per capita), military advantage (dispersed population, mountainous terrain), grievance (ethnic dominance was the only grievance proxy that was found to have an effect) and finally population size and time since previous civil war (degradation of conflict-specific capital).

Two major observations can be made at this point about the C&H thesis. First, their approach is highly empirical, based on ‘quantifiable’ facts for which they give very precise definitions in the appendix (ibid: 594-595). Thus, for example, they argue that ethnic and religious hatreds have long been perceived as a cause of civil war, but ‘hatred‘ cannot be statistically measured. As a result they argue that hatreds “can evidently only occur in societies that are multi-ethnic or multi-religious and so our proxies measure various dimensions of diversity” (ibid: 571). Second, their thesis is heavily influenced by economic theory, which in turn, at least according to John Searle, is almost exclusively guided by ‘Western Rationalistic Tradition’ (Krasner 1996: 108) and according to Wendt, is behavioural (Wendt 1992: 181). Although C&H do not specifically mention rational choice, reference to causal mechanisms such as ‘earnings foregone’ (Collier & Hoeffler: 574, 581, 588), suggests that the C&H thesis is based on the individual’s trade off to, or preference for, either production (not rebel) or appropriation (rebel). This trade off is based on the perceived opportunity to rebel, which in turn leads to the aggregate concept of civil war (Sambanis 2004: 261, 263). Thus taking into account independent variables such as GDP per capita is important, because it proxies for the cost to the individual in terms of “income forgone by enlisting as a rebel” (Collier & Hoeffler 2004: 569). It is important to bear in mind that the main purpose of the C&H thesis is to explain the phenomenon of civil war at the aggregate or macro level. This is supported by the fact that they give several causal mechanisms for each proxy. Thus, according to C&H, the time since previous civil war variable can either be interpreted as “the gradual deprecation of rebellion-specific capital, and hence an increasing cost of rebellion [greed/opportunity], or the gradual erosion of hatred [grievance]” (Collier & Hoeffler 2004: 589). This, in addition to the identification of ‘identifiable rebel organisations’ as the main actors, will be discussed in greater detail in the sub-section agency and structure.

From the analysis of the C&H thesis above, one could argue that it shares the main assumptions and approach of neo-realism, or the ‘neo-neo synthesis’ depending on how incommensurable one finds these paradigms (Waever 1996: 166, Brown & Ainley 2005: 45-48). The main assumptions by both ‘neo’ schools of thought are apparent in the C&H thesis: anarchy (discussed in the sub-section agency and structure) and rational choice (Brown & Ainley 2005: 46). If we were to distinguish between which IR theory (neo-realism or neo-liberal institutionalism) the C&H thesis parallels most it would be neo-realism. A distinctive parallel can be drawn with regards to neo-realism and neo-liberalism, focussing on absolute and relative gain (Brown & Ainley 2005: 46-47). As discussed above, individuals or the rebel group as a whole make a trade off between production or appropriation, depending on which is more lucrative. Thus, one could argue that faced with the choice of sub-optimal absolute gains from production (refraining from rebellion) or optimal gains from appropriation, the relevant actor within the C&H thesis will always favour rebellion, which is consistent with neo-realist assumptions. In addition, with regard to approaches, a number of authors stress influence of (micro-)economic theory in neo-realism, in particular with regards to methodological naturalism (Guzzini 2000: 157, Jackson & Sørensen 2007: 75, Wendt 1992: 181). As mentioned above, C&H take an economic approach with regards to their model and causal mechanisms. Remaining parallel assumptions and approaches of neo-realism, such as the focus on structure are discussed in the following sub-sections, other aspects such as state-centrism and distribution of capabilities do not apply to the study of civil war.


[1] Ole Waever (1996), refers to this as one of the two ‘Fourth sub-Debates’ (Waever 1996: 161-170)

Excerpt out of 17 pages


Civil War Onset and the 'Third Debate'
Dublin City University
International Relations
Catalog Number
ISBN (eBook)
ISBN (Book)
File size
383 KB
First Class Honours
Civil War, Third Debate, Collier and Hoeffler, Aspinall, Uganda, Positivism, Post-Positivism, Social Constructivism, Ontology, Epistemology, Methodology, Agency, Structure, Rationalism, neo-Realism, neo-neo Synthesis, Paradigm, Aceh, Identity, GAM, Quantitative, Explaining, Understanding, Foundationalist, Essentialist
Quote paper
Bachelor International Relations (Hons) Ralph Myers (Author), 2010, Civil War Onset and the 'Third Debate', Munich, GRIN Verlag,


  • No comments yet.
Read the ebook
Title: Civil War Onset and the 'Third Debate'

Upload papers

Your term paper / thesis:

- Publication as eBook and book
- High royalties for the sales
- Completely free - with ISBN
- It only takes five minutes
- Every paper finds readers

Publish now - it's free