Jopinions

To understand why some civil wars last longer than others it is imperative to establish a definition for civil wars. James Fearon writes that “a civil war is a violent conflict within a country fought by organized groups that aim to take power at the center or in a region, or to change government policies”. Fearon also refers to the commonly used death threshold of “1,000 killed over the course of a conflict” to define when violence becomes a civil war (2007). Research in this field is extremely important as findings and discoveries are important in informing policymakers and governments in their attempts to prevent or end civil wars. In this paper, a theory will be developed in which two hypotheses will be established. In this theory concepts and mechanisms will be explored. A method will then be created in order to effectively test the two hypotheses. Indicators are essential to understanding how the variables in this research will be used. Empirical data and evidence are then needed to carry out the method and discover if either hypothesis is proven correct.

Theory 

This research essay will attempt to explore a relationship between the presence of high-value natural resources and the duration of civil wars.

Natural resources are extremely important for groups to achieve and to maintain power, especially in lesser developed countries. Whether it is the government/state in control of natural resources, rebel groups, or even corporations, their value equates to power and influence. It is sound thinking to presume that the presence of valuable natural resources would result in more economic and military capacity for the state or possibly rebel groups. This could lead to peace negotiations being difficult to achieve and perhaps fighting to become more intense or prolonged. The hypothesis for this essay is that more valuable natural resources in a region should lead to longer civil wars. 

One of the key concepts that relates to this hypothesis is the relative power of the parties involved in a civil war. Differing levels of natural resources alter these power balances and therefore change the likelihood of civil war termination. Phillip Hultquist’s excellent analysis of this concept (2013) allows a further understanding of how these mechanisms determine civil war duration. Hultquist remarks that “power parity provides a condition favorable to the peaceful resolution” and therefore termination. If one party is superior in power to the other, Hultquist predicts that peace is far less probable and therefore civil war duration is longer. If power balance is extremely important in civil war duration, and it is expected that power can be gained from natural resources, it is fair to presume that high levels of natural resources will lead to longer civil wars. 

While power balances in civil wars have been studied extensively in the current literature regarding civil wars, the effect natural resources have on these power dynamics is less known. I can speculate that high levels of natural resources lead to longer civil wars when the resources are controlled by the state rather than the rebel groups. Many scholars find that shorter civil wars occur when the rebels are powerful and pose significant threats to the state. “A state is likely to resist intermediaries until the cost of mediation is outweighed by the anticipated price of continued conflict … Mediation should then be less likely when a state is faced with a relatively weak rebel group” (Clayton, 2013). It seems apparent that states should generally be in control of their natural resources, however, other possibilities cannot be ignored. If states are mostly in control of natural resources in the region and it is true that natural resources provide power, it is likely the hypothesis is proved correct.

The ways in which natural resources can provide parties with power are highly numbered. Perhaps the strongest way this occurs is through export/sale of high valued natural resources such as oil. Fossil fuel exports can fund military personnel and technology that can be used to fight opposition. The revenues from these natural resources make military success far more viable and probable.

The idea that the degree of power parity in civil wars can predict the outcome has been explored well in the literature surrounding civil war termination. Measuring the power that rebel groups hold is a troubling and difficult task. In order to effectively test my hypothesis, careful research behind measures of power is important. It may also be true that the more powerful a rebel group becomes, the more natural resources it can acquire, changing the balance of power even further. Although I speculate that the state most likely controls natural resources, if rebel groups obtain valuable resources this may lengthen a civil war. “Natural resource financing creates longer wars by enabling rebel groups to keep fighting” (Humphreys, 2005). 

To come to a reliable and well researched result, the mechanism of this research will test if high natural resources lead to less power parity and therefore longer civil wars. If this mechanism is correct, the hypothesis will be confirmed.

Natural resources may increase civil war duration; however, the opposite may be true. The second hypothesis that will be tested in this essay is that the absence of valuable natural resources in a region should lead to longer civil wars.

This second hypothesis may be proven correct because states with fewer natural resources may be less economically advanced than those that have an abundance. States in this scenario would be less able to defeat rebel groups, meaning rebel groups can fight for longer periods of time. If rebels see some success in a weak state, it can be said that they may be further motivated to continue fighting, especially if casualties are low. If rebels fight for long periods of time, civil war duration will be long. 

This theory relies on the assumption that the state is weakened to a similar level to that of the rebel groups. In a situation where there is a lack of natural resources, there may be cases where rebels can compete with the state. For example, external support could allow rebels to have similar military power to the state causing neither side to have a clear advantage and fighting to be prolonged. 

Method

In this research, the level of natural resources is the independent variable. It is essential in this study that the relevant natural resources are measured in order to properly test the hypothesis. 

An article from Ross (2004) identifies natural resources can be involved in civil wars. He notes that “both quantitative and qualitative studies suggest that the production of oil is associated with the onset of conflict, particularly separatist conflict”. Although Ross describes the relationship oil has with onset rather than civil war duration, it still proves the strong correlation oil has with conflict which is useful to include in this research. Oyefusi also finds truth in this correlation. They find that “oil-dependent countries have a particularly high risk of experiencing secessionist civil wars” (2002). Because many scholars believe that oil plays a tremendous role in civil wars, this will be one of the natural resources that this research includes in testing. 

Along with oil, this research will include what is referred to as “lootable” resources (Le Billon, 2001). Ross comment’s that “between 1990 and 200, civil wars occurred in five diamond-producing states, three opium-exporting states, three major cannabis-exporting states, and two leading coca-exporting states”. He refers to this group of “lootable” resources as “gemstones, drugs [and] timber” (2004). 

For the independent variable, I am going to use oil and gemstones/precious metals export revenues from the decade prior to the onset of civil war. The reason export revenues will be used is that my research presumes that states that export lots of oil or gemstones/precious have access to lots of oil and gemstones/precious. If I were to instead use production numbers, this may be misleading as states that have high natural resources may not produce them internally. Also, these measurements will be taken from as close to the civil war dates as possible to provide accurate findings. In the case that data is missing or incomplete, assumptions may have to be made. These assumptions would be based on other data and evidence in order to maintain as much accuracy as possible.

The dependent variable in this research essay is civil war duration. To measure civil war duration it is imperative to define what a civil war is. If a definition is not established, it is impossible to decide when a civil war begins and ends. Fearon, who provides the population for this study, defines civil wars in the following ways: “(1) They involved fighting between agents of (or claimants to) a state and organized, nonstate groups who sought either to take control of a government, to take power in a region, or to use violence to change government policies. (2) The conflict killed at least 1,000 over its course, with a yearly average of at least 100. (3) At least 100 were killed on both sides (including civilians attacked by rebels)” (2003). 

In order to avoid misinterpretation and false results a control variable will be used. This variable will be low drug exports and revenues. This is because this is another valuable resource that can be used to achieve success in civil wars. By making sure that no actors in this research are profiting enormously from the drug trade, it ensures that the other natural resources that are being tested are not distorted. 

As previously mentioned, the population for this research is provided by Fearon in his work “Additional tables for “Ethnicity, Insurgency, and Civil War”” (2003). From the definition that has already been established, Fearon identifies 127 civil wars spanning six different regions. These regions are, Western Europe, Eastern Europe, Asia, North Africa/Middle East, Sub-Saharan Africa, and Latin America and the Caribbean. This data includes all civil wars from 1945 up to 2003 that fit the definition. 

Two cases will be selected from the data set, both in compliance with the control variable. I believe that choosing cases from within the same region will provide the most reliable results as other factors may be similar, for example, language, culture, ethnicity, religion and other factors that may be hard to control for. It is also important for this research that both cases occur in the same time period or as close together as possible. This is because the value of natural resources and the state of the international economy is always changing. By selecting cases in the same time frame, this problem is eliminated. 

The first case to be used in this research is the Lebanese civil war which started in 1975 and ended in 1990. The second case will be the civil war that occurred in the Yemen Arab Republic from 1962 until 1969.

Empirical part: Case study

Lebanon

The civil war in Lebanon which began in 1975 stemmed from years of political uncertainty. “At the time of independence in 1943, there was sharp disagreement among the Lebanese on how the country should be structured and oriented”. This uncertainty continued for decades leading to the civil war. “In the start of 1975, the Lebanese political scene became extremely tense due largely to the Lebanese government’s non-response to Israeli attacks on the large Palestinian refugee camp of Nabatiya” (Chamie, 1976). There was also an event in which “11 demonstrators, and five soldiers were shot and killed” during a strike. Numerous other incidents related to these issues also contributed to the beginning of this civil war. Due to the ethnically arranged political system, this civil war was fought mainly between Christian and Muslim groups. The main actors were, “the Lebanese Front” (Christian), “the Lebanese National Movement” (Sunni Muslims), “the Lebanese Resistance Detachments” (Shi’i populist Muslims) and “the Palestine Liberation Organization”. “Other participants in the war included Syria, Israel, and splintered contingents of the Lebanese Army” (Ochsenwald, 2020). 

Unfortunately, data on oil export and revenues for Lebanon and the surrounding region before 1988 is extremely limited. However, the data that is available is still very useful. Worldbank.org has data on oil rents as a percentage of GDP for Lebanon from 1988 up until 2020. Their data shows that Lebanon’s oil rents as a percentage of their GDP is 0% for every single year, meaning they are an oil importer. This suggests that Lebanon had little to no oil production capabilities at the time of the civil war. Furthermore, Barlow (1982) finds that in Lebanon between 1950 and 1972, “no oil was discovered”.

In a dataset that measures gold and diamond revenues as far back as 1997, Lebanon appears to be a state with few valuable natural resources. The data provided by Gaulier and Zignago (2010), shows that Lebanon exported just $2,000,000 more in diamonds than it imported. Their total diamond export value was $60.4 million. This amount of diamonds is extremely low compared to states who are rich in diamonds such as Botswana which exported $2.71 billion worth of diamonds in 2000. Furthermore, the difference between export and import values suggests that Lebanon simply export previously imported diamonds, rather than their region being naturally plentiful. It could be said that Lebanon’s small size (area) is a contributing factor to its export number being significantly lower than other states, yet states such as Sierra Leone disprove this notion. Sierra Leone exported $129 million worth of diamonds while only importing $14.6 million worth. This proves that Lebanon was not rich in diamonds in this time period. The same data also shows that Lebanon exported just $13.3 million worth of gold in 1997.

While this data is representative of Lebanon seven years after the end of its civil war, I believe it is extremely relevant as you would expect revenue to rise after civil war. This means I expect the export revenues during the civil war period to be lower, meaning they still were a state with low amounts of valuable natural resources.

This Lebanese civil war is the joint longest civil war that Fearon (2003) identifies in his data set under the North Africa/Middle East section. However, this civil war fails to confirm the first hypothesis. Lebanon was a state with low natural resources, yet its civil war duration was extremely high. If the mechanism that was explored is reviewed it reveals that high-value natural resources would lead to less power parity and therefore longer civil war. What happened was that power parity was present, yet civil war duration was long. Gaub writes, “In total, Lebanese militias received some $30 billion in outside assistance … this support was not enough to tip the scales so that any one party could claim victory. The provision of arms and financial assistance by outsiders, therefore, helped to prolong the conflict” (2015). The evidence that has been studied confirms the second hypothesis. External support negated the need for natural resources and caused a high-duration civil war.

Yemen Arab Republic

“The 1962 coup in Yemen culminated decades of indigenous political and social movements driven by their desire to overthrow the autocratic imamate”. It was a “battle between tribal and modern national identities” (Orkaby, 2017). The conflict was based upon the fear of Egyptian intervention. Saudi Arabia was particularly opposed to “President Nasser’s involvement” and saw it as “an attempt to undermine the House of Saud”. There was also the concern that Nasser “had the potential to undermine the fidelity of tribal groupings … upon whom the political stability of the Federation rested” (Jones, 2004). 

Barlow (2010) finds that the Yemen Arab Republic had not “become important exporters of oil by the end of the period under study” which is the period between 1950 and 1972. Barlow recognises that the civil war contributed to “stagnation” and the slow development of industries such as oil production. Data from worldbank.org shows that Yemen enjoyed high oil rents as a percentage of GDP between 1990 and 2010 (peaking at 42.3% of GDP), however, oil rents slowed in the 2010s (as low as 0.6% of GDP) due to another civil war. Because there is no qualitative data for Yemen’s oil revenue in the 1960s, this gives a good indication that they were low due to the civil war’s disruption. 

Similarly, data is sparse for diamond and gold export and revenue values for Yemen in this time period. The closest date the study from Gaulier and Zignago (2010) tracks is 1996 when Yemen’s gold exports were extremely negligible at $42.8 thousand. The data on diamonds is extremely inconsistent yet points to the idea that Yemen has effectively no trade in diamonds. Overall there is a significant lack of data, however, I think it is fair to assume that Yemen was not rich in natural resources during the civil war from 1962 to 1969.

The case of Yemen neither confirms nor denies either hypothesis put forward in this research. Its length of 7 years is not short yet not particularly long either. The fact that the state was low in natural resources means for the first hypothesis to be correct the civil war must be high in duration, meaning the second hypothesis is mildly proven correct.

Conclusion

In conclusion, these results do not strongly corroborate either hypothesis. The case of Lebanon completely contradicts the first hypothesis but somewhat confirms the second. It doesn’t completely confirm it through the initial mechanism discussed, yet mostly it does. This research can’t prove that the fact the state had low natural resources caused the power parity, yet with more research, this connection could possibly be made. The case of Yemen is unfortunately under investigated. The data and assumptions that could be made infer that the civil war doesn’t confirm either hypothesis. The mechanism introduced in this research did not work as predicted as states with low natural resources experienced long to moderately long civil wars.

Word Count: 2,971

Bibliography

Barlow, R. (1982). Economic Growth in the Middle East, 1950-1972. International Journal of Middle East Studies, 14(2), pp.142–143.

Chamie, J. (1976). An Investigation into the Causes. World Affairs, 137, pp.175–176.

Clayton, G. (2013). Relative rebel strength and the onset and outcome of civil war mediation. Journal of Peace Research, 50(5), p.610.

data.worldbank.org. (n.d.). Oil rents (% of GDP) – Lebanon, Yemen, Rep., Jordan | Data. [online] Available at: https://data.worldbank.org/indicator/NY.GDP.PETR.RT.ZS?end=2020&locations=LB-YE-JO&start=1988&view=chart [Accessed 12 Jan. 2023].

Fearon, J. (2003a). ADDITIONAL TABLES FOR ‘ETHNICITY, INSURGENCY, AND CIVIL WAR’. American Political Science Review, 97(1).

Fearon, J. (2003b). Ethnicity, Insurgency, and Civil War. American Political Science Review, 97(1).

Fearon, J. (2007). Iraq’s civil war. Foreign Aff, 86, p.4.

Gaub, F. (2015). Lebanon’s civil war: seven lessons forty years on. [online] European Union Institute for Security Studies. Available at: https://www.iss.europa.eu/sites/default/files/EUISSFiles/Alert_21_Lebanon_civil_war.pdf . DOI 10.2815/475463.

Gaulier, G. and Zignago, S. (2010). CEPII – Documents de travail – BACI: International Trade Database at the Product-Level. The 1994-2007 Version. [online] http://www.cepii.fr. Available at: http://www.cepii.fr/CEPII/fr/publications/wp/abstract.asp?NoDoc=2726 [Accessed 9 Jan. 2023].

Hultquist, P. (2013). Power parity and peace? The role of relative power in civil war settlement. Journal of Peace Research, 50(5), p.624.

Humphreys, M. (2005). Natural Resources, Conflict, and Conflict Resolution: Uncovering the Mechanisms. Journal of Conflict Resolution, 49(4), p.514.

Jones, C. (2004). ‘“Among Ministers, Mavericks and Mandarins”: Britain, Covert Action and the Yemen Civil War, 1962-64.’ Eastern Studies, 40(1), pp.99–100.

Le Billon, P (2002) Controlling resource flows to civil wars: a review and analysis of current policies and legal instruments. Background Paper for the International Peace Academy “Economic Agendas in Civil Wars” Project Conference Policies and Practices for Regulating Resource Flows to Armed Conflicts. Rockfeller Foundation Study and Conference Center.

Ochsenwald, W.L. (2020). Lebanese Civil War. [online] Encyclopedia Britannica. Available at: https://www.britannica.com/event/Lebanese-Civil-War [Accessed 10 Jan. 2023].

Orkaby. A. (2017). Beyond the Arab Cold war : the international history of the Yemen civil war, 1962-68. Cambridge ; New York, New York: Oxford University Press, pp.4–5.

Oyefusi, A. (2007). Oil and the Propensity to Armed Struggle in the Niger Delta Region of Nigeria. [online] p.3. Available at: https://openknowledge.worldbank.org/bitstream/handle/10986/7002/wps4194.pdf?sequence=1&isAllowed=y [Accessed 12 Jan. 2023].

Ross, Michael.L. (2004). What Do We Know about Natural Resources and Civil War? Journal of Peace Research, 41(3), p.342.

‌