By Laia Balcells, Duke University and Stathis Kalyvas, Yale University
*This memo was prepared for “The Political Science of Syria’s War” conference on November 8th, 2013. A version of this piece was previously published on the Middle East Channel on ForeignPolicy.com.
Will the civil war in Syria prove to be an endless Afghan-style quagmire? We argue that the Syrian civil war resembles the war in Libya more than it does Afghanistan’s conflict. NATO’s intervention, following U.N. Resolution 1973, made all the difference in Libya: By strengthening the rebels’ hand and severely weakening Muammar al-Qaddafi’s forces, it turned what appeared to have been a sudden military defeat into rapid victory, against all forecasts of protracted war. The reason why the Libyan conflict was headed toward a relatively quick resolution is fairly straightforward: It was a conventional rather than an irregular war. Let us explain.
The analysis of warfare in civil wars has been plagued by an imprecise use of terminology. Afghanistan, Libya, and Syria are all described as “insurgencies,” a term frequently used as synonymous of civil war. This is a problem because not all civil wars are guerrilla wars or irregular wars or insurgencies. A guerrilla (or irregular) war is a type of military contest characterized by a steep military asymmetry between the rival sides, whereby the weak side has no alternative but to fight a war of evasion and ambush against the strong side. The objective of rebel combatants in guerrilla wars is typically to win through attrition. This produces long wars that frustrate the ability of conventional armies to translate their military superiority into victory. Counterinsurgency is notoriously hard, like “eating soup with a knife,” to use a well-known metaphor. No wonder that Vietnam and Afghanistan turned into military quagmires.
However, the civil wars in Libya and Syria are no guerrilla wars; despite the initial military superiority of the regime forces, these conflicts look more like conventional than guerrilla wars. Unlike guerrilla wars fought in mountains or jungles by elusive bands of fighters, conventional wars entail pitched battles, artillery contests, and urban sieges across clearly defined frontlines. In the news from Syria, we can read examples of clear-cut victories in battles indicating that the war is being fought conventionally. Also, conflict maps show large, contiguous areas that are militarily held by the rebels or the regime of President Bashar al-Assad. Many cities are being fought over, with frontlines bisecting them.
Conventional civil wars go back a long way: just think of classic conflicts such as the U.S. and Spanish civil wars. More recently, conventional civil wars were fought in Bosnia and Azerbaijan. In our research (Kalyvas and Balcells 2010) we find that conventional civil wars are much more common than generally thought; they represent 34 percent of all major civil wars (i.e., those causing over 1,000 battlefield fatalities per year) fought between 1944 and 2004. More significant is the fact that conventional civil wars have increased in proportion after the end of the Cold War: They account for 48 percent of all civil wars fought between 1991 and 2004. In contrast, guerrilla wars have declined from 66 percent during the Cold War to just 26 percent after its end.
In a recent paper (Balcells and Kalyvas 2012), we compare conventional with irregular wars and a third type we call symmetric, non-conventional (SNC) — basically wars in failed states — and find that the former tend to be less bloody on the battlefield, causing on average 62,000 fatalities, as compared to 84,000 for guerrilla wars. However, once we control for war duration, we find that conventional wars are much more intense, causing on average 3,000 deaths per month compared to 1,250 for irregular wars. Another striking difference between these two types of war is their duration: Conventional wars are shorter, lasting an average of three years, whereas irregular wars last an average of nine years. We also find that irregular wars produce significantly more violence against civilians as compared to conventional and SNC wars. This result is consistent with the idea that civilians constitute much more of a valuable resource in insurgencies than in other conflicts and they are targeted because of it (Kalyvas 2006). Lastly, we find that 66 percent of insurgencies end with a government victory, compared to just 38 percent for conventional wars and merely 19 percent of SNC wars, which tend to end in draws.[i] In a multivariate analysis on civil war outcomes, we find that civil wars that have conventional features are significantly more likely to end in rebel victories (the marginal effect dy/dx of conventional civil war on rebel victory — as opposed to incumbent victory — is .29). That is the case even when controlling for external support received by rebels and incumbents. Quite intuitively, external support for rebels has a positive impact on rebel victory (the marginal effect dy/dx is in this case 0.18).
In short, we find that conventional civil wars are more intense, shorter, and less likely to end in regime victories than irregular civil wars. How about Syria then? As we said above, with its pitched battles, visible frontlines, and urban fighting, this conflict resembles the Libyan war. Thus, if past record can serve as a guide, the Syrian civil war may well turn out to be shorter than generally anticipated and result in the regime’s defeat.
As of November, the war looks like it is headed toward some kind of stalemate. The reason is that both sides are receiving substantial amounts of external assistance. The rebels are supported by the Arab Gulf states, Turkey, and to some measure the West, whereas the Assad regime is backed by Russia, Iran, Iraq, and Lebanon’s Hezbollah faction. Furthermore, substantial numbers of foreign fighters are currently operating in Syria. The extent of foreign assistance and that it appears to be quite balanced between the two camps helps explain the current stalemate. However, let us not forget that this war is still (unfortunately) young. Already processes of centralization are taking place within the rebel camp and major shifts in foreign assistance on either side may help tip the balance and produce a decisive military outcome.
Let us close by recalling an intriguing historical analogy. Just before World War II, the Spanish Civil War became a focal conflict in Europe, the ideological and military battleground where fascist and anti-fascist forces clashed while the entire world stared. Today, Syria has become the key battleground of Sunni and Shiite ideologues and activists. The stark ideological dimension of the Spanish Civil War was expressed in extensive external support by foreign powers and massive participation in combat by foreign volunteers. The decisive assistance provided to the Republican camp by the USSR — which turned out to be much more substantial compared to the limited Anglo-French assistance — led to the centralization of the highly fragmented Republican camp, but also its eventual domination by the communists; however, this came too late to counterbalance the massive assistance offered to the Nationalists, on both ground and air, by Nazi Germany and Fascist Italy — and which eventually helped tip the balance in favor of the Nationalist rebels. The ideological dimension of the Spanish conflict had a clear geopolitical stake, the domination of Europe, very much like the ideological dimension of the Syrian civil war overlaps with a geopolitical conflict between Saudi Arabia and Iran over the domination of the Arab world. Lastly, the Spanish Civil War witnessed intense violence on the battlefield, widespread atrocities against the civilian population, and mass displacement — all related to extensive political polarization (Balcells 2010). Unlike Spain, where this polarization was associated to ideological identities, the violence in Syria is connected to sectarian identities. Yet, even a cursory examination of the patterns of violence in Spain (e.g., Preston 2012) suggests how ascriptive and non-ascriptive identities alike can produce high levels of violence in the context of a civil war.
In the spring of 1938, the Spanish Civil War had gone on for about two years and many observers thought that it was headed toward a stalemate. A year later, it was over.
Laia Balcells is an assistant professor of political science at Duke University. She is the author of “Continuation of Politics by Two Means: Direct and Indirect Violence in Civil Wars” in the Journal of Conflict Resolution and the forthcoming Rivalry and Revenge: The Politics of Violence in Civil War based on her dissertation research. Stathis Kaylvas is Arnold Wolfers professor of political science and director of the Program on Order, Conflict and Violence at Yale University. He is the author of The Politics of Violence in Civil War (2006). Balcells and Kalyvas’s research focuses on warfare dynamics during conflict and determinants of political violence. They have co-authored “International System and Technologies of Rebellion: How the End of the Cold War Shaped Internal Conflict” in the American Political Science Review, and “Endgame in Syria?” on the Middle East Channel.
[i] There are few cases whose technology of rebellion changes over time, but here we use the coding in the last year of the war.