Some help for a class

[Farnan]

Well everything is falling in place for me to be able to do something I really want to do, which is an independent study research project on counter-insurgency warfare. Basically it means I will independently research counter-insurgency warfare, write a paper on it, and a professor will serve as an adviser and grade the paper. Talking with one of my professors I'm going to do it using a case study, and this case study will be a comparison of the political and military strategy and its effects under the Security Transition Strategy (the initial strategy in Iraq) with the political and military strategy under the New Way Forward (known more popularly as the "Surge") and its effects with the goal of finding what worked and why, what didn't work and why and what lessons can be drawn.

Now the help I ask is does anybody have any good sources that can help me?

I know Parameters (the War College Quarterly), the Counter-Insurgency Field Manual (the doctrine applied in the New Way Forward) and probably GEN Tommy Frank's book may be a good start but any other advice?

[JP226]

How much math do you know? Can you do a linear regression?

[Farnan]

How much math do you know? Can you do a linear regression?

I did it before, I can review it and work with one of my profs to relearn it...

Why do you ask?

Won't start this class for two months, time to relearn some stat.

[Heinz Guderian]

Well everything is falling in place for me to be able to do something I really want to do, which is an independent study research project on counter-insurgency warfare. Basically it means I will independently research counter-insurgency warfare, write a paper on it, and a professor will serve as an adviser and grade the paper. Talking with one of my professors I'm going to do it using a case study, and this case study will be a comparison of the political and military strategy and its effects under the Security Transition Strategy (the initial strategy in Iraq) with the political and military strategy under the New Way Forward (known more popularly as the "Surge") and its effects with the goal of finding what worked and why, what didn't work and why and what lessons can be drawn.

Now the help I ask is does anybody have any good sources that can help me?

I know Parameters (the War College Quarterly), the Counter-Insurgency Field Manual (the doctrine applied in the New Way Forward) and probably GEN Tommy Frank's book may be a good start but any other advice?

will you have access to classified/national security documents? I applied for my dissertation through a well respected government connected professor, but was refused.

[Farnan]

will you have access to classified/national security documents? I applied for my dissertation through a well respected government connected professor, but was refused.

I doubt it, all my sources will have to be unclassified. Would be awesome to have that access, but my security clearance doesn't go high at all.

JP226: I planned on doing this using the statistics provided in sources. Due to a gap of about two years between the two strategies I'm not sure you'll be able to use linear regression. For statistics one method is to use a comparison of the number of attacks in a three month period at the beginning of each period and then the number of attacks in a three month period at the end of each period. The stability level index provided by the Foreign Affairs also as a point of comparison. Finally, you got the % of government income spent, level of corruption, change in number of civilian deaths per month, number of battalions in the ISF capable of action with limited coalition support, change in unemployment level, and a qualitative comparison of the ability of the ING to govern.

[JP226]

I did it before, I can review it and work with one of my profs to relearn it...

Why do you ask?

How do you plan on analyzing the data for the case study? I would suggest you regress it honestly that's usually a good rule of thumb. Your going to have to find some numbers, things like our losses as compared to their losses, whether the settings were urban or rural, whether the training was urban or not. Those kinds of things. And you can probably get that by pursuing the proper channels within the pentagon or the joint warfare analysis center down in Norfolk I believe.

Running the regression is easy, Microsoft excel does it and quite honestly it's as good as any of the 2 and 3k dollar software. It's just a matter of finding the data. The relationships themselves can be tricky, like whether one lement is linear or exponential. But if your prof. can help with that, that's no biggie.

It sounds like, you want to know what about the surge worked and why. As compared to what we were doing before. And the best way to approach that is to get yourself some numbers Like I laid out. The real trick is to find data about specific engagements when they were fought, what were the losses, what were the kills. From that you can setup a model that says kills/ losses are determined by who's in power (i'd do gates rummie Petreus and other), initial plan or surge, and I dunno some other variables. Maybe some training numbers that you could find, I know you can find numbers on when certain equipment were phased in like the truck thing that is taking some of the place of the humvee. You could also find numbers on reinforcements. This is all off the top of my head, and i'm no expert, but you could realistically find these things offer up some numbers and draw a conclusion.

[Axeman]

I suggest you read up on the South African border wars of the 1980s. It is a perfect example of how mobile warfare *thank you britan for those SAS you lent us to train* with support is one of several solutions , but not the only one. There are tons of sources I could list , but I suggest , as I always do , look in history. The Soviet Army in Afghanistan for example learned quickly that Mechanized Waves will not effectively pull the root out of Mujhadeen , and Soviet efficiency skyrocketed when they started the independent airborne brigade attacks. Just look in history my friend , as for specific sources , the South African Military College comes to mind.

[JP226]

JP226: I planned on doing this using the statistics provided in sources. Due to a gap of about two years between the two strategies I'm not sure you'll be able to use linear regression. For statistics one method is to use a comparison of the number of attacks in a three month period at the beginning of each period and then the number of attacks in a three month period at the end of each period. The stability level index provided by the Foreign Affairs also as a point of comparison. Finally, you got the % of government income spent, level of corruption, change in number of civilian deaths per month, number of battalions in the ISF capable of action with limited coalition support, change in unemployment level, and a qualitative comparison of the ability of the ING to govern.

There is a gap in data where nothing was collected for two years? I would use the stability index as your Y. A regression is a snapshot for a given point of time, just like a poll. What you could do is a right censored linear regression. Which is a regression over time. Infact the only difference is the setup of the data. If it's points collected over time versus observations then it's automatically right censored. It's not like you need MA and AR processes like they use for price forecasting. If you can do that stability variable, do stability = intercept + income + corruption + battlions + ING ability. I think with civ deaths and employment you are going to introduce may have too much collinearity. They should be included between corruption, ing ability and income numbers. Heh, if you were using stata software I could give you some tips on how to find collinearity, i'm not sure how to do it with excel so you may have to go wiht your gut. However I'd also setup a date variable as to when the surge was implemented. When you record the numbers everything before the "surge date" gets a zero and everything after get's a one. And then plug that into your regression as well.

With those date numbers you will setup a flat out number as to which is better, the idea today or the idea yesturday in pertaining to stability. Which in and of itself could be some serious conclusions, much less an A.

[Bovril]

Won't it be really difficult to do a study of an event with absolutely no historical distance? You can hardly assess the long term success of the strategies employed.

[Farnan]

Won't it be really difficult to do a study of an event with absolutely no historical distance? You can hardly assess the long term success of the strategies employed.

Only problem is that events in historical distance are not necessarily applicable to current situations.

If for example, the US was to employ some successful strategies even from Malaya it would be called a major war crime (talking about the resettling of Chinese to "New Villages").

[hellheaven1987]

Only problem is that events in historical distance are not necessarily applicable to current situations.

If for example, the US was to employ some successful strategies even from Malaya it would be called a major war crime (talking about the resettling of Chinese to "New Villages").

Yes, but that is Emergency so government usually can take more harsh measure.

The reason why American could not/cannot apply this strategy mainly due to most gureilla conflicts it fight are not under its territory; therefore, British might able to do that and escape war crime easily since that was part og its empire, but American could not/cannot do such thing in Vietnam and Iraq because it was/is not part of US.

Edit: However, that does not mean historical cases should not be studied; in fact, several counter-insurgency strategy has been used for thousands years and never really stopped using (like build castles/fort to control surrounding area).

[Farnan]

Well I refined the class and got the professor to officially sign off on it, now I just need the department head's signature and I should be good.

Well the study will be applying Counter-Insurgency theory to examine the different periods that I study along with analyzing each effort on its own:

Current Reading list I am considering:

US Army and Marine Corps Counterinsurgency Field Manual
On Point II: The US Army in Operation Iraqi Freedom from May 2003 - January 2005 by the Combat Studies Institute.
On Guerrilla Warfare by Mao Ze Dong
Guerrilla Warfare by Che Guervera
Counterinsurgency Warfare: Theory and Practice - David Guala
Learning how to eat soup with a knife - John Nagl

[mrmouth]

The COIN center blog has some interesting tidbits every once and awhile.

You look fairly covered as can be on your end without a clearance.

Honestly, why not email the COIN director over at the CAC, and maybe he can point you down the road less traveled...

http://usacac.army.mil/BLOG/blogs/coin/default.aspx

http://usacac.army.mil/CAC2/COIN/director.asp

[JP226]

Won't it be really difficult to do a study of an event with absolutely no historical distance? You can hardly assess the long term success of the strategies employed.

Long term success is irrelevent if you are comparing two relatively close episodes. Essentially it is which one is better and why today. Then maybe we can talk out of our asses about which WILL ultimately be better, but that's the opinion realm.

[Bovril]

Long term success is irrelevent if you are comparing two relatively close episodes.

That is, surely, one of the problems with comparing two relitively close episodes.

Essentially it is which one is better and why today. Then maybe we can talk out of our asses about which WILL ultimately be better, but that's the opinion realm.

Well, it depends what your interested in. If your interested in what effects certain actions have in the first couple of years or less of their implmentation, fine, but that's not terribly interesting. Is it? Besides, comparitive studies using only two examples are notoriously tricky anyway, so I'm slightly surprised that this project is even being undertaken. A straigtforward study of the surge would make more sense to me, but whatever..I'm just some guy on teh interwebs

[kev-o]

@ Bovril

He's doing a comparison between one event to the surge in Iraq.

@ Farnan

To help out, I think a really good case you can use in comparison with the surge is Charlemagne's campaign against the Saxons. I recently finished a paper on the socio-economic aspects of Charlemagne's campaign in Saxony and there is quite a few parallels that can be drawn from this and Iraq. Charlemagne sought to spread Chrisitianity in Saxony, Bush sought to spread democracy in Iraq. Charlemagne went in with few troops, Bush went in with few troops. Charlemagne had to change strategy, so did Bush. It was a long one (over 30 yrs), and he used different tactics to pacify Saxony (forced conversion, slaughter, raids, rebuilding of the saxon forts, displacement of the captured, etc.) Some good primary documents to look at are Charlemagne's "Three capitularies detailing military affairs in the Carolingian Empire," as well as Einhard's "Life of Charlemagne."

If you need any more info about the sources PM me.

[hellheaven1987]

Na, the thing is that the topic he is looking for is too huge to do research - a paper with 200 pages would not even enough to include everything.

Anyway, I would suggest to pick up several cases study based on geography, culture, finance and strategy/tactic both sides use. A few good cases are:

1. Soviet gureillas against German during WWII.

2. Chinese Communist against Japanese/Nationalist (a must study for gureilla warfare).

3. Tito's Partisan in Balkan.

4. First War of Indochina and Second War of Indochina.

5. Korean War (not for gureilla warfare but it would suggest what happened when gureilla turned into regular unit).

6. British experience in Middle East.

7. Arab rebellion against Ottoman.

8. Malaya Emergency (I highly recommand a study of this case).

9. Several African gureillas (too many).

10. First/Second Chechen War (a more modern case of gureilla warfare).

[JP226]

That is, surely, one of the problems with comparing two relitively close episodes.

It's war though, not stock derivatives. And there are relatively few outcomes. No need to overcomplicate the situation.

[kev-o]

Thats fine and dandy, but he is looking in how counter-insurgency took place, not just the insurgency.

[hellheaven1987]

Thats fine and dandy, but he is looking in how counter-insurgency took place, not just the insurgency.

The thing is that you need to both study counter-insurgency and insurgency in same time, so you can compare why some insurgency success, but others fail.

My case list contains successful cases for both sides; it also contains different geography, culture, finance and strategy.

Oh, and add Algerian War as another case study; it is extremly good case to compare what French learned during First Indochina War and applied on Algerian War.