Measures for Security in a Counterinsurgency

Abstract

Quantitative measures to gauge progress in a counterinsurgency in use today include the number of enemy, friendly, and civilian casualties, and the level of violence. While common, each of these has drawbacks. With this and the sparseness of the literature on this topic in mind, we argue for a series of improved measures. These include the ratio of who is initiating incidents, an historical analysis of incidents and related variables, and an analysis of insurgent target sets. These measures are presented using data for Al Anbar province, Iraq, along with a discussion of their advantages over more common metrics.

Key Words:

• Metrics
• Counterinsurgency
• Anbar

Acknowledgements

The author thanks Jonathan Geithner for having continued these analyses in support of MNF-W, along with Carter Malkasian and Jason Quirin for their helpful comments. The author also extends his sincere gratitude to the men and women of Multi-National Force-West, especially Major General Walter Gaskin, Major General John Allen, Brigadier General Charles Gurganus, Colonel John Holden, Colonel John Pollock, Lieutenant Colonel Joseph Paschall, and Major Edward Sullivan.

Notes

¹Scott S. Gartner, Strategic Assessment in War (Binghamton, NY: Vail-Ballou Press 1997), 163–77.

²Austin Long, On ‘Other War’: Lessons from Five Decades of RAND Counterinsurgency Research (Santa Monica, CA: RAND Corp. 2006), 39.

³Roger Trinquier, Modern Warfare: A French View of Counterinsurgency[1961] (Westport, CT: Praeger reprint 2006), 15, 27.

⁴David Galula, Counterinsurgency Warfare: Theory and Practice[1964] (Westport, CT: Praeger reprint 2006), 50.

⁵Lt. Gens. David Petraeus and James Amos, FM 3-24: Counterinsurgency (Washington DC: Dept. of the Army 2006), ix.

⁶James Clancy and Chuck Crossett, ‘Measuring Effectiveness in Irregular Warfare’, Parameters 37 (Summer 2007), 89.

⁷Long, On ‘Other War’; Clancy and Crossett, ‘Measuring Effectiveness in Irregular Warfare’; Andrew F. Krepinevich Jr, ‘How to Win in Iraq’, Foreign Affairs 84 (Sept./Oct. 2005), 1; ‘Measuring Success in Counterinsurgency Warfare’West Point Irregular Warfare Message of the Week, available at <www.usma.edu/dmi/IWmsgs/MeasuringSuccess.pdf>, accessed June 2008; Anthony Cordesman, ‘The Uncertain “Metrics” of Afghanistan (and Iraq)’, available at <www.csis.org/media/csis/pubs/070521_uncertainmetrics_afghan.pdf>, accessed June 2008; Richard L. Clutterbuck, The Long, Long War: Counterinsurgency in Malaya and Vietnam (New York: Frederick A. Praeger 1966); David Kilcullen, ‘Twenty-Eight Articles: Fundamentals of Company-level Counterinsurgency’, Military Review, special counterinsurgency edition (Oct. 2006), 1.

⁸Petraeus and Amos, FM 3-24: Counterinsurgency; Frederick W. Kagan, ‘Measuring Success’Armed Forces Journal (Jan. 2006), available at <www.afji.com/2006/01/1397777>, accessed June 2008.

⁹Craig Cohen, ‘Measuring Progress in Stabilization and Reconstruction’, United States Institute of Peace Stabilization and Reconstruction Series 1 (March 2006), 1.

¹⁰Jeffrey Race, War Comes to Long An: Revolutionary Conflict in a Vietnamese Province (Berkeley and Los Angeles: Univ. of California Press 1972); Thomas C. Thayer, War Without Fronts: The American Experience in Vietnam (Boulder, CO: Westview Press 1985).

¹¹Race, War Comes to Long An, 277.

¹²‘Charts to Accompany the Testimony of Gen. David H. Petraeus: 10–11 Sept. 2007’, available at <www.defenselink.mil/pubs/pdfs/Petraeus-Testimony-Slides20070910.pdf>, accessed June 2008; ‘Charts to Accompany the Testimony of Gen. David H. Petraeus: 8–9 April 2008’, available at <www.mnf-iraq.com/images/stories/Press_briefings/2008/april/080408_petraeus_handout.pdf>, accessed June 2008; Measuring Stability and Security in Iraq: March 2008 (Washington DC: US Dept. of Defense 2008) available at <www.defenselink.mil/home/features/Iraq_Reports>, accessed June 2008.

¹³See for example: Iraq Index: Tracking Variables of Reconstruction and Security in Post-Saddam Iraq (Washington DC: The Brookings Institution 2008), available at <www.brookings.edu/iraqindex>, accessed June 2008; ‘Measuring the Surge’ graphic, Washington Post, 9 Sept. 2007, available at <http://media.washingtonpost.com/wp-srv/world/pdf/surge_090907.pdf?hpid=topnews>, accessed June 2008.

¹⁴Krepinevich, ‘How to Win in Iraq’, 8.

¹⁵Petraeus and Amos, FM 3-24: Counterinsurgency, 1–29.

¹⁶Krepinevich, ‘How to Win in Iraq’, 7.

¹⁷‘Measuring Success in Counterinsurgency Warfare’, West Point Irregular Warfare Message of the Week.

¹⁸‘Charts to Accompany the Testimony of Gen. David H. Petraeus: 10–11 Sept. 2007’; ‘Charts to Accompany the Testimony of Gen. David H. Petraeus: 8–9 April 2008’.

¹⁹Petraeus and Amos, FM 3-24: Counterinsurgency, 1–29.

²⁰Galula, Counterinsurgency Warfare, 50.

²¹Neil Munro, ‘Counting Corpses’, National Journal, 5 Jan. 2008, available at <http://news.nationaljournal.com/articles/databomb/sidebar.htm>, accessed June 2008.

²²See <www.iraqbodycount.org>. Number cited as of June 2008.

²³ Iraq Index: Tracking Variables of Reconstruction and Security in Post-Saddam Iraq, 4.

²⁴L. Roberts, R. Lafta, R. Garfield, J. Khudhairi, and G. Burnham, ‘Mortality Before and After the 2003 Invasion of Iraq: Cluster Sample Survey’, Lancet 364, 20 Nov. 2004, 9448; G. Burnham, R. Lafta, S. Doocy, and L. Roberts, ‘Mortality after the 2003 Invasion of Iraq: A Cross-Sectional Cluster Sample Survey’Lancet 368, 21 Oct. 2006, 9545.

²⁵<www.iraqbodycount.org>, accessed June 2008.

²⁶Roberts, et al., ‘Mortality Before and After the 2003 Invasion of Iraq: Cluster Sample Survey’, 9448; Burnham, et al., ‘Mortality after the 2003 Invasion of Iraq: A Cross-Sectional Cluster Sample Survey’, 9545.

²⁷Munro, ‘Counting Corpses’, <www.iraqbodycount.org>, accessed June 2008.

²⁸Hannah Fischer, ‘Iraqi Civilian Deaths Estimates’Congressional Research Service Report RS22537, 22 Nov. 2006, 1.

²⁹‘Charts to Accompany the Testimony of Gen. David H. Petraeus: 8–9 April 2008’; Measuring Stability and Security in Iraq: August 2006 (Washington DC: US Dept. of Defense 2006), 32.

³⁰Galula, Counterinsurgency Warfare, 75–94.

³¹While a formal definition of the term ‘incident’ is lacking, in this context it typically refers to enemy actions. These actions include attacks (direct fire, indirect fire, improvised explosive devices (IEDs), mines) and also IED and mine finds (i.e. prior to explosion). We will use the term ‘incident’ in this context as well.

³²‘Charts to Accompany the Testimony of Gen. David H. Petraeus: 10–11 Sept. 2007’; ‘Charts to Accompany the Testimony of Gen. David H. Petraeus: 8–9 April 2008’.

³³Galula, Counterinsurgency Warfare, 75.

³⁴Government Accounting Office, Iraqi Government Has Not Met Most Legislative, Security, and Economic Benchmarks (GAO-07-1195) (Washington DC: Government Accounting Office 2007); David M. Walker, ‘Securing, Stabilizing, and Rebuilding Iraq: Iraqi Government Has Not Met Most Legislative, Security, and Economic Benchmarks (GAO-07-1220T)’ (4 Sept. 2007), Testimony to Congress available at <www.gao.gov/new.items/d071220t.pdf>, accessed June 2008.

³⁵See for example ‘Measuring the Surge’, Washington Post , 9 Sept. 2007.

³⁶Race, War Comes to Long An, 144.

³⁷Ibid., 231.

³⁸These categories correspond to the types of incidents in the MNF-W Significant Events database, which was the repository for reporting on security incidents. For clarity, we have included surface-to-air fire incidents under the direct fire or rocket categories, depending on the event. Entries in the database such as defensive actions in which no shots were fired, escalation of force incidents, accidents, and unexplained explosions were considered to be initiated by neither side (i.e. effectively ignored).

³⁹Niel Smith and Sean MacFarland, ‘Anbar Awakens: The Tipping Point’, Military Review (March–April 2008), 41.

⁴⁰Our data set extends only to August because that is as far as the author could calculate during his time in Iraq. A later calculation for the month of November 2007 showed that the fraction of incidents initiated by friendly forces had continued to increase, up to 75 percent of the total.

⁴¹Cordesman, ‘The Uncertain “Metrics” of Afghanistan (and Iraq)’.

⁴²Clutterbuck, The Long, Long War, 95–131.

⁴³Carter Malkasian, ‘A Thin Blue Line in the Sand’, Democracy: A Journal of Ideas 5 (Summer 2007), 48.

⁴⁴ Iraq Index: Tracking Variables of Reconstruction and Security in Post-Saddam Iraq, 50.

⁴⁵Krepinevich, ‘How to Win in Iraq’, 8.

⁴⁶Ibid., 8; Cordesman, ‘The Uncertain ‘Metrics’ of Afghanistan (and Iraq)’; Kagan, ‘Measuring Success’; Michael P. Noonan, ‘Iraq and the ‘Metrics’ System’, Foreign Policy Research Institute E-Note (Sept. 2007), available at <www.fpri.org/enotes/200709.noonan.iraqmetricssystem.html>, accessed June 2008.

⁴⁷‘Charts to Accompany the Testimony of Gen. David H. Petraeus: 10–11 Sept. 2007’; ‘Charts to Accompany the Testimony of Gen. David H. Petraeus: 8–9 April 2008’. The ORA is an assessment tool used by Multi-National Corps-Iraq (MNC-I) and Multi-National Force-Iraq (MNF-I) to gauge the development progress of Iraqi Security Forces.

⁴⁸Race, War Comes to Long An, xv.

⁴⁹Noonan, ‘Iraq and the “Metrics” System’.

⁵⁰Our contention is that although an IED find is a more positive result than an IED attack, it is analogous to an errant shot fired at friendly forces. The fact that the shot missed does not remove the hostile intent behind it. The same is true for IEDs found before they could explode. Therefore, we assigned IED finds as incidents initiated by insurgents, as long as the IED had been emplaced. IEDs found before emplacement were treated as weapons caches. As a check on the plausibility of these assignments, we performed additional calculations in which IED finds were counted as incidents initiated by friendly forces. The results of these calculations were entirely inconsistent with intelligence assessments of the security situation in Al Anbar.

⁵¹A common suggestion has been to use casualties as a weighting criterion. However, since the vast majority of incidents do not cause any casualties, the net effects of casualty-weighted counting schemes turn out to be minor and not worth the additional complication of the metric.

⁵²Trinquier, Modern Warfare; Galula, Counterinsurgency Warfare:; Petraeus and Amos, FM 3-24: Counterinsurgency; Clutterbuck, The Long, Long War.

⁵³Temperature data were only available back to early 2005, but given the regularity of the data we projected them backwards to match the data set for incidents (dotted lines).

⁵⁴In comparing the numbers of reports for cache finds that actually mentioned civilian tips to anecdotal reporting on the fraction of caches found via tips, it was clear that the tally of the former considerably underestimated the total number of civilian tips being received. As such, the numbers of tips shown in Figure 6 should be viewed as minimal, as it is almost certain that the actual numbers of civilian tips for cache finds were much greater. However, detailed analysis of the data we did have failed to discover any significant biases (e.g. with respect to geographic location or reporting by unit) that would account for the increase in reported tips in 2007. Therefore, while the numbers of civilian tips as shown in Figure 6 should not be taken as absolute, the overall increasing trend is valid, and was corroborated by anecdotal reporting.

⁵⁵Long, On ‘Other War’, 21–34.

⁵⁶Galula, Counterinsurgency Warfare, 52.

⁵⁷Race, War Comes to Long An, 280.

⁵⁸The data have been averaged over 90 days to smooth the curves. Also, to avoid confusion, numbers on the vertical axis in Figure 8 have been omitted. Instead, the data for each target type are plotted on a scale normalized to match the peak intensity of the curve for Coalition Forces. For the sake of this discussion, the overall trend is more important than the actual numbers. The actual scaling factors used were: Iraqi Army incidents (x6.1); Iraqi Police incidents (x14.2); civilian casualties (x4.3); civilian incidents (x27); infrastructure incidents (x55).

⁵⁹We used civilian casualties here, vice incidents against civilians, to emphasize the fact that the attacks against civilians in the mid-March to mid-June timeframe were of the ‘spectacular’ variety, for example involving Suicide Vehicle-Borne Improvised Explosive Devices (SVBIEDs). We realize the use of civilian casualties has the inherent drawbacks discussed earlier in the paper, but given that we are more concerned here with where the peak due to the shift in insurgent tactics appeared in time than to absolute numbers of casualties, the drawbacks should not detract significantly from our conclusions.

⁶⁰Jason Campbell and Michael E. O'Hanlon, ‘Measuring Progress in Iraq’, Wall Street Journal, 13 July 2007.

⁶¹The last official Iraqi census was conducted in 1997, and population movements since the invasion dramatically altered the number of people in some areas. These numbers represented the best estimates of those we spoke with in Al Anbar.