Since the overthrow of Saddam Hussein's regime following the 2003 Anglo-American invasion, Iraq has become a theatre of a asymmetric warfare conducted by a multifaceted Sunni insurgency (Baram 2005; Chehad 2005; Hashim 2006; Kohlmann 2007; Robinson 2007). Particularly striking is the wave of mass casualty attacks against Iraqi civilians, frequently involving the use of suicide bombers. We define suicide attacks as violent acts whose execution requires the death of at least one individual, namely one or more attackers (Bloom 2005; Gambetta 2006; Hafez 2007; Moghadam 2008; Pape 2006; Pedahzur 2005; Reuter 2004). The number of suicide attacks perpetrated by armed groups has increased over the last three decades, with an estimated 2,406 incidents associated with several campaigns between December 1981 and the end of December 2008, responsible for between 24,383 and 26,408 deaths. (1)
Of those, 1,336 attacks have been perpetrated by Iraqi insurgents since the military invasion of Iraq in March 2003, causing between 12,447 and 13,403 deaths, and accounting for about 56% of incidents and 51% of fatalities, respectively, caused by all suicide attack campaigns since the 1980s. A high percentage of the targets of those suicide bombers have been the new Iraqi forces and defenceless civilians, in particular Shiites (Hafez  estimates 44% and 23%, respectively, compared with 15% made up of coalition forces). Data from Worldwide Incidents Tracking System (http://wits.nctc.gov) show that, between January 2004 and December 2008, civilians were more than 70% of those killed (11,604) by suicide attacks in Iraq.
A recurrent argument concerning suicide attacks is that armed groups resorting to this tactic are generally engaged in a campaign for the liberation or independence of a territory, that is, for nationalist purposes (Pape 2006; see also Brym 2008). To some extent, our data support this hypothesis, which accounts for the suicide attack waves of Hezbollah's campaign in Lebanon in the 1980s against the US, France, and Israel (Kramer 1998; Ranstorp 1997; Reuter 2004); the Liberation Tigers of Tamil Eelam's struggle for Tamil independence since 1987 (Hopgood 2006); Hamas and other Palestinian organizations against Israel (Levitt 2006; Ricolfi 2006); the Kurdistan Workers' Party's campaign for the independence of Kurdistan; the Chechen insurgency for secession from Russia; and the struggle by the Kashmiri separatists such as Jaish-e-Muhammad and Lashkar-e-Toiba against India (for an overview on all these campaigns, see Hassan 2008). Such a nationalist theory can also account to some extent for Al-Qaeda and its affiliates (Gunaratna 2002), notably some Iraqi Sunni insurgents such as Al-Qaeda in Iraq (AQI) and Ansar Al-Sunnah, and its allies such as the Taliban in Afghanistan, insofar as their immediate goal has been the liberation of Muslim countries from the US-led occupation (Cordesman 2008; Goodwin 2006a; 2006b; Hafez 2007; Hashim 2006; Karam 2007; Moghadam 2008; Nance 2007; Pape 2006; Riedel 2008).
That said, three questions related to the decision-making of certain organizations' elite members within the Iraqi insurgency remain unanswered: (1) why only certain armed groups--in particular AQI and its allies--within the Iraqi insurgency have mounted (even if not exclusively) a specific terrorist (2) campaign against Iraqi civilians (instead of attacking occupying forces), in which (2) they have targeted the Shiite population, (3) by resorting to suicide attacks instead of other tactics. These questions certainly require identifying a complex set of conditions for organizing an armed struggle such as that of the Iraqi insurgency, the most important being, on one hand, the Sunni insurgents' common capacity for exploiting contingent political and social opportunities after the overthrow of Saddam Hussein ("Sunni Insurgents' Political Opportunities"), and, on the other hand, a regular flow of militants, including volunteers to recruit as suicide bombers ("Suicide Bombers in Iraq: Profiles and Motivations"). …