Abu Muqawama: Post
Abu Muqawama retains its autonomy and the views and beliefs expressed within the blog do not reflect those of CNAS. Abu Muqawama retains the right to delete comments that include words that incite violence; are predatory, hateful, or intended to intimidate or harass; or degrade people on the basis of gender, race, class, ethnicity, national origin, religion, sexual orientation, or disability. In summary, don't be a jerk.
Quants and COIN
So the quants, not content with mucking up the financial world
Anyway, you guys could probably care less why I never read the APSR. But based upon my limited personal experience in high-intensity and low-intensity conflict as well as my academic research -- to include field research in several active combat zones -- human or "moral" factors often explain war far better than number-crunching. (At its worst, the aforementioned number-crunching you see in scholarly journals is just qualitative judgements assigned numerical value, i.e. if 10="good" and 1="bad".) And all methods of analysis are inherently limited in their explanatory value.
That said, there is an article in Nature on the "unified model of insurgency". And Josh Foust and his gang of hired assassins have posted a critique of it on Registan.net worth checking out. Josh & Co. laud the authors of the Nature article for the way in which they have approached their subject. (And yeah, actually, they should be lauded, because honestly, God bless them for tackling a complicated issue with such methodological rigor.) What I get from the critique, though, is that the model the authors have constructed is -- surprise! -- too simple to reflect the realities of insurgencies and counterinsurgencies. And it reminds me of the way in which Quants on Wall Street discovered that all of their complicated computer models had failed to reflect the actual behavior of markets and indeed hastened the destruction of the very funds for which they had been constructed to generate income. (Honestly, didn't we learn our lesson with LTCM in 1998?)
I'm not trying to come off as one of those analytical dinosaurs the gang at Fire Joe Morgan used to poke fun at (you know, the guys who value "grit" and "hustle" over OPS), but we have to admit that in certain chaotic "systems" involving real live humans acting both rationally and irrationally -- such as international finance, or war -- the explanatory value of quantitative analysis might have its limits.
The eternal human need to
The eternal human need to make sense of the irrational.
And the quest for Utopia.
I certainly appreciate the
I certainly appreciate the length plug, but you forgot to call us a MERRY band of assassins.
Also, the link to the post is wrong.
Also also, Dirk might be onto something...
"The first people to dismiss
"The first people to dismiss quantitative analysis are those who have no idea how to do a quantitative analysis." Or something like that. I'm pretty sure an American PoliSci student on his 3rd drink said that to me.
I'm more open than that to criticisms of quantitative/empirical/bigN analysis, but it is - if good data can be acquired - a useful tool. Of course, there is some quantitative analysis out there that i just as bad as some of the bad qualitative analysis. Quantitative will never be able to predict the future of violent conflict consistently, but neither will qualitative analysis...
I didn't check out the links
I didn't check out the links you posted, but you have read Stephen Walt's "Rational Choice and Security Studies, yes?"
Furthermore, it's not like the "Nature" authors are the first to use quantitative models to understand insurgency. See, say, the work of Ethan (not Bruce) Bueno de Mesquita.
And it's true, the APSR does *not* publish a lot of non-quantitative work on conflict. "Security Studies" and "International Security" are the mainstay of qualitative research on the topic.
ADTS
PS - Do I really need quote marks in this post?
And AM, I would have thought
And AM, I would have thought you don't read the APSR (or BJPS) in part because you're in some hybrid discipline that isn't really political science per se (But since I've never been in a War Studies program, what do I know?).
ADTS
That Nature paper is very
That Nature paper is very different from the types of Nature papers I am used to reading ( studies on cancers and inflammatory diseases of the skin, etc). I wasn't sure how to read the paper to be honest; never really studied sociology or political science or anything like that. Goodness, it boggles the mind to jump fields, doesn't it?
So, my questions are going to be very basic and relate to the "Methods and Supplementary Information" section that has been linked in the above comment by :أبو مقاومة
1. In the first table: why were those nine countries (or conflicts) chosen for inclusion in the study? Were the inclusion criteria reasonable? I don't know the background literature, so I ask!
2. Why the particular start and end date for the data bases? Why not hold this data set rigid, say, cover a conflict for the same ten years in each country instead of varying the data sets temporally? Have the authors not introduced a huge uncontrolled variable by doing this? I don't know, so once again, I ask!
3. The supplementary states, "We rely on data collection by a wide range of organizations and scholars who all determine their own methods." And then goes on to say how they tried to deal with the differences (I think: again, not used to this type of literature). Okay, were those assumptions valid?
I could go on and on, but there are just too many random uncontrolled variables in the data sets they use. That old saw of "junk in, junk out". The points about assumptions made by some of the commenters above are well taken: many assumptions are made in collating a set of data points. If any one assumption is incorrect, the statistical analyses aren't really measuring what the authors think they are measuring.
Again, I mean no disrespect. I am not familiar with the area and likely am getting much wrong. I too laud the authors for the attempt at studying a difficult topic and am glad to see such studies. I just don't know what they mean.
Danger Room went after this
Danger Room went after this Nature piece as well.
Ah, that Danger Room post
Ah, that Danger Room post put things much better than I did:
"But there are a number of king-sized holes in Gourley’s argument. First, it assumes that the insurgent is the only one capable of making a first move in an irregular war. But, of course, the counterinsurgent can also strike first. What does that do to Gourley’s CNN-driven model of conflict?
Second, as others have already noted, using war reporting to make predictive models is a problem: Coverage, from the media or other sources, is notoriously inaccurate or incomplete. Comparing reportage from one conflict to another is even worse: In Afghanistan, for example, rural environs and rugged geography have led to relatively less coverage than the Iraq conflict. Gourley’s team tried to compensate by pulling reports from diverse sources, like media in Afghanistan and a Catholic NGO in Columbia. But that likely created more inconsistency, rather than less."
Nature editors: what's going on here? I kid, I kid....
Few things come to mind in
Few things come to mind in regards to this post ...some people make models, not all people like the models that people make, many people would rather be else where, statistics never lie, liers use statistics, and times change. Not all people think the same or explain things the same way. Numbers and relationships are all around us begging to be understood. Times have changed so have our tools. We sit in front of the number one destroyer of accounting and news paper jobs communicating words and numbers with each other.
Is it fair to compare todays methods to books written yesterday? If Clausewitz had a personal computer, then he may have included a chapter on quantitative analysis. If Clausewitz had available all the information that that we have available today at our finger tips, he may have saved Gourley a lot of time.
Gourley did nothing wrong. He used his knowledge to understand and communicate the world as he perceives it, much the same way as the rest of us (including Dirk Digalo who would rather do a Fourier Transform of the natural rhythm of heavenly bodies, just hope that it is Discrete. saves a lot of time and energy) are doing, in our own way. The quants on Wall Street did he same. Where the quants made a mistake is they turned their tool in to gospel, and they started to believe it. Then the people who benefited started to believe the models and it became a way of life. No one asked the model creators what the model could not predict, they believed because they were making money. That is the problem with people and models, no one wants to put the time into understanding the language. People make use of Negative Space. People just want to use models, like we use MicroSoft Software(then it crashes, Bill thanks us for his retirement). Models come and go as times change. Models help us understand the world and sometimes take us to places that we could not otherwise visit. We now are only getting to the point were we can prove some of the predictions that Einstein's Models/Theory predicted. Some models have more limitation than other models.
Coin is a model of warfare. You can only image what negative space exists in peoples minds in the Afghan-Iraq discussion (it is how we got there in the beginning and how people feel about were is going today....no one likes to loose).
It is amazing to me sometimes. I have seen people spend days crunching results using statistics only to walk up and see an obvious conclusion by short scan of the raw data. It has a lot to do with how our brains are wired and the experiences we have had. There are no surprises in Gourley's Results. The more organized your group is the more pain you can give. The more frequently you engage, the more pain you give. If your group is not organized, then the frequency of engagement and the pain level goes away. It is common sense really. What Gourley has done for us is organize the data and our thoughts. Maybe the model is simple. Maybe the model will never be complete or useful. It does what models do, it helps us go places that we may not have gone otherwise. That is all a model is good for, a guide. If a person says they have a model, then ask them what part of the world it predicts and what it does not. They should tell you. Do your home work and see if the model predicts your world. Then you have a guide as how to weight the model's results.
Here is were the rubber hits the road (Dirk, don't get excited). You have to consider the audience to which you are speaking to and what communicates your information in the best way. It is a left and right brain thing. Qualitative, quantitative and theoretical .....all that matters is the audience gets your message. Your audience has a mix of life experiences.
Here is the insulting part. People tend towards rewards. Academics at the PhD level have a good salary and they can get tenure and pension. Policy makers and lawyers can make a wicked bunch of money, more if you can put the two together. You have a great golden parachute if it does not work out, you can not lose. Academics teach people and can have a great impact on society. I have yet to figure out what Policy makers do, all I do is look at my tax bill and try to figure out were it all goes(....in their pockets?). Only other place you can do better is Pro-Sports or Wall Street(....yes it is screwed up, but it might explain what happened to political science....times have changed).
If you get the time, here is an interesting read...Discussion of the Method. No it is not going to help you understand COIN or math models. Billy Keon teaches at the University of Texas in Austin. The book has its origins in Human Engineering. There are two halves, discussing the same thing, one for each side of the brain. It is a mix of philosophy and engineering. It gives some insight to how we process information and make decisions. It is written so that you can skip parts and still walk away with the message. It is not an engineering text nor is it pure humanities, if you get to numbers, read past them to what interests you. Read it again and read a different book. Dirk, sorry there is nothing for you.....maybe there is.....think he talks about two newly weds standing at a distance from each other halving the distance as they move closer. ....they never reach each other, but they are close enough for all practical purposes......
As Koen concludes, it is all Heuristic. But then, that is only Koen's model. You'll have to read the book to see what is not included in the model.
John Robb over at
John Robb over at http://globalguerrillas.typepad.com/globalguerrillas/
For a different take on our economic problems.
He's so impressed by the insurgents he wants to organize his own economic one - to hide the $$ from our rapacious elites, not blow things up.
Although he's got a cool take on that stuff too, as well as Improvised Economic Darknets and nerdy energy weapons you can make in your garage.
It seems strange to attack
It seems strange to attack an entire approach to political science as outright silly and then offer an a-theoretical case alternative. Also, calling an approach 'irrelevant' is a worthless critique as that is merely a perceptual criticism and of no fault to those who do that sort of work.
I propose not to wait until
I propose not to wait until you earn enough cash to order all you need! You should take the loans or just financial loan and feel yourself comfortable
Add your comment