Surge refers to the increase in US troops in Iraq JanMay 2007 using - PDF document

‘Surge’ refers to the increase in US troops in Iraq Jan­May 2007 using COIN doctrine; build­up phase of the surge completed by end of May 07 followed by ‘surge’ of operations that began in Jun 07 (“Arrowhead” series) (Accidental Guerrilla p 144) SIGACTs are ‘significant activities’ ­ attacks against Coalition and Iraqi government forces, and civilians and infrastructure, as observed by Coalition Forces. 0

Sidebar: ‘Purge before the Surge’ – By the time Bush announced the Surge on 10 Jan ‘06, his entire national security team was shaken apart; specifically, Rumsfeld was out, Casey was replaced by Petraeus, Abizaid chose retirement over job as D Nat Intel, and Khalilzad was made ambassador to UN (MR Gordon & Gen BE Trainor, The Endgame, Pantheon, NY, 2012, p 308). 1

Afghanistan: 33,000 troop surge applied in late 2009 and formation of Afghan ‘local’ police (ALP) was designed to replicate COIN tactics from Iraq and effects of SOI. “COIN is the comprehensive civilian and military effort taken to defeat an insurgency and to address any core grievances.” (S. Melton) COIN also refers to ‘clear, build, and hold’. Simplistic differentiation between COIN and CW is their CoG – population vs. military, respectively. Politicization of the Surge (Senator McCain, Feb 2, 2013) “I’m very aware of the history of the surge and the Anbar Awakening, and I also am aware that any casual observer will know that the surge was the fundamental factor, led by two great leaders…”. 2

Source: War Within (p 200 & 262); B Woodward (2008) Awakening began in Ramadi (Anbar) in Oct 06, which led the way for SOI standup. Situational Awareness Council of Colonels presented six major negative trends including “current strategy is not working” and “ethnic and sectarian conflict is increasing” (03 Nov 2006) However, “(Commanders in Baghdad) thought that they had enough troops” (ISG member and incoming SecDec Gates, 05 Dec 2006); Bob Gates was member of the Iraq Study Group 3

SIGACTs: “intensity of insurgent activity measured as attacks per capita against Coalition and Iraqi government forces … they capture violence against civilians and between nonstate actors only when U.S. forces are present and so dramatically undercount sectarian violence” (p 790 of E Berman, JN Shapiro, J Felter (2011) Can Hearts and Minds Be Bought? The Economics of Counterinsurgency in Iraq, J Pol Economy 119(4): 766 – 819). Limitations: SIGACTs –vary from disabled IED with no casualties to an incident (eg., suicide bomber) with dozens of casualties AOs – from a neighbourhood to a village to a district with thousands of population to a village 4

Trend lines were lower, but not necessarily significantly. Source: Biddle S, Friedman JA, Shapiro JN (2012). Testing the Surge: Why Did Violence Decline in Iraq in 2007. International Security, 37(1), pp. 7–40. pairwise t­test for dependent samples. The t­test assesses whether the means of two Sidebar: Steve Biddle is quoted (Dec 2006) as supporting the Surge “noting that American forces were limiting the scale of the violence (and) if the mission shifted to focus on training, violence would increase” (The Endgame p 303) 5 groups are statistically different from each other.

Optional Timeline analysis by Province (see back slides) 6

Note: Al­‘Askariyya (al­Askari; one of the holiest in Shi’a Islam) shrine at Samarra (~ 125 km N Baghdad) purposely bombed 22 Feb ‘06 (month 25) by AQI to incite sectarian war GWB file indicated 1200 weekly attacks in Iraq Dec 2006 (War Within p 305) more than half occurred in Baghdad (p 126); also, ~ 50% of the war during 2006 took place within Baghdad (Accidental Guerrilla p 130) dramatically in Aug 07; (early) 2007 was the deadliest for US troops (331 dead Apr – Jun) (Accidental Guerrilla p 148), but not for civilians, which peaked in late 2006 7 • civilian deaths peaked Sep 06 – Jan 07 (2700 – 3800/month; 125/day in Dec 06), and • violence began to decline in early 07, although unevenly, before decreasing

Sunni – AQI split began when AQI demanded upholding Quran religious customs while Sunnis held fast to tribal customs, (e.g., refusal to yield to AQI’s demands for inter­ marriage, which was one strategy for cementing ties) (Accidental Guerrilla p 159­60) 8

Most SOI standups occurred shortly after full strength of the surge troops were deployed and became operational. SOI standups in Diyala and Tamim occurred long after the surge and the violence has decreased in these provinces, which was clearly a reactive development. Peak SIGACT in Baghdad occurred long before full strength of surge; hence, the latter is not connected to decrease in violence. SOI standup in Baghdad, Salahaddin, and Babil is a result of the insurgency losing steam/defeated ­ too short a period elapsed from full surge strength to be defeated but additional influence was likely. All peak SIGACT occurred before full surge strength and SOI standup. The SOI standup in Anbar is not connected with the surge. 9

obs 3 = average of SIGACT observed during the first 3 months after SOI standup pred 6­3 = average of SIGACT predicted during the first 3 months after SOI standup using the linear regression of SIGACT 6 months prior to SOI standup Interpretation: Observed & Predicted post­SOI SIGACT are not different based on 6 months of pre­SOI SIGACT, but they are if based on 9 months or more. 10

Note: Decrease of US Forces from peak of 160,000 (Dec 05) to 126,900 (Jun 06); decrease of total coalition force (CF) from peak of 183,000 (Dec 05) to 146,900 (Jun 06) – suggests that the ‘Surge’ simply restored US troop numbers, certainly it did not add capacity beyond 2005 levels. Despite increased tempo, ops Together Forward I & II (summer 06) were considered dismal failures to “clear, hold, and build”; instead considered by Kilcullen as the “Kiss of Death” that describes the killing of tribal leaders/members that supported ISF & CF 11 troops after they left an area � clearly contradicts the notion of SOF effectiveness.

Salah ad­Din) where the Surge occurred was 10,442,320 in 1997 (total Iraq pop 22,046,244); hence, the ‘local’ troop density was considerably higher, by about a factor of 3. Quinlivan’s 20:1,000; COIN FM 3­24 20­25:1,000 However, British lost Cyprus campaign with force ratio of 110:1 while Indonesians defeated Dar’ul Islam with a ratio < 3:1 – difference is attributed to level of local support (Accidental Guerrilla p 181) Also, insurgencies are lengthy – normal duration between 15 and 40 years (Accidental Guerrilla p 184) While effective thresholds are debatable and conflict­specific, the important point is that both growth and readiness were increasing throughout 2006. can lead, level_2 can follow, and level_3 is in training. (# refers to number of units) 12 Note: Combined pop of the 6 provinces (Al­Anbar, At­Ta'min, Babil, Baghdad, Diyala, and Effective Troop Density Ratios: McGrath ratio (13:1,000) ; RAND ratio 13.5:1,000; Readiness Levels: ISF readiness = (3*#level_1 + 2*#level_2 + #level_3)/6, where level_1

Key question is whether the Surge was necessary or whether a change in operational remains unknown is the optimal mix of the two and the quantification of alternative commanders; e.g., Gen. Pace believed that there was no military answer to the situation by Gen. Odierno (The Endgame p 300), which was not appreciated by other high level strongholds; hence the need for a Surge to deal with the ‘local’ insurgency, as advocated Baghdad and surrounding belts was intense in late 2006, especially with Sunni and AQ troops until ISF strengthened further in 2007 were difficult to make. The situation in War”). Thus, it is understandable why decisions to bridge the ISF with a surge of US 2006, which would not have been visible to military planners and the NSC (“Fog of Hindsight (and this analysis) reveals that violence was on the cusp of declining in late violence. causes, which collectively were primarily responsible for initiating the decline in While it is apparent that COIN and CW both contributed to the decline in violence, what doctrine (i.e., adoption of COIN) would have sufficed with the 140K troops in Iraq – winning strategy. “Fog of War” was especially dense in late 2006, making it very difficult to identify a growth of ISF increased tempo of SOF early adoption of COIN decreased tempo of JAM Much earlier shift of loyalty of Sunni insurgents Alternative hypotheses/causes: have worked, albeit at a slower pace. considering that violence had started to decline prior to the Surge, the latter would likely 13