Category Casualty estimation

Isolating the Guerilla

The Vietnam was significant in that it was third bloodiest war in U.S. military history (58,000 U.S. killed) and the U.S. Army choose to learn no lessons from it !!! This last point is discussed in my book America’s Modern Wars: Understanding Iraq, Afghanistan and Vietnam.

In 1965 Trevor Dupuy’s HERO (Historical Evaluation Research Organization) conducted a three-volume study called “Isolating the Guerilla.” It was an interesting survey of 19 insurgencies that included on its research team 26 experts. This included General Geoffrey Lord Bourne (British Army, ret.), Andrew C. Janos, Peter Paret, among others.

These guys:

https://en.wikipedia.org/wiki/Geoffrey_Bourne,_Baron_Bourne

http://www.nytimes.com/1975/04/26/archives/col-r-ernest-dupuy-88-dead-publicist-and-military-historian.html

https://en.wikipedia.org/wiki/Trevor_N._Dupuy

https://es.wikipedia.org/wiki/Andrew_C._Janos

https://www.goodreads.com/author/show/2793254.William_A_Nighswonger

https://en.wikipedia.org/wiki/Peter_Paret

http://www.legacy.com/obituaries/northjersey/obituary.aspx?pid=163090077

https://en.wikipedia.org/wiki/Theodore_Ropp

https://en.wikipedia.org/wiki/Gunther_E._Rothenberg

http://www.ur.umich.edu/9495/Oct03_94/29.htm

http://www.andersonfuneralhomeltd.com/home/index.cfm/obituaries/view/fh_id/12343/id/3994242

http://www.nytimes.com/1984/08/31/obituaries/frank-n-trager-78-an-expert-on-asia-dies.html

 

The first volume of the study, although developed from historical sources, was classified after it was completed. How does one classify a study that was developed from unclassified sources?

As such, the first volume of the study was in the classified safe at DMSI when I was there. I was aware of the study, but had never taken the time to look at it. DMSI went out of business in the early 1990s and all the classified material there was destroyed. The Dupuy Institute did not have a copy of this volume of the study.

In 2004 we did our casualty and duration estimate for Iraq. It was based upon a survey of 28 insurgencies. We then expanded that work to do an analysis based upon 89 insurgencies. This was done independently of our past work back in 1965, which I had never seen. This is detailed in my book America’s Modern Wars.

As this work was being completed I was contacted by a Lt. Colonel Michael F. Trevett in 2008. It turns out he had an unclassified copy of the study. He found it in the Ft. Huachuca library. It was declassified in 2004 and was also in DTIC. So, I finally got a copy of a study after we had almost completed our work on insurgencies. In retrospect, it would have been useful to have from the start. Again, another case of disappearing studies.

In 2011, Michael F. Trevett published the study as a book called Isolating the Guerrilla. The book is the study, with many of the appendices and supporting data removed at the request of the publisher. It was a self-publishing effort that was paid by Michael out of his personal/family funds. He has since left the Army. I did write the foreword to the book.

What can I say about this case? We did a study on insurgencies in 1965 that had some relevance to the wars we entered in Afghanistan in 2001 and Iraq in 2003. It remained classified and buried in a library in Ft. Huachuca, Arizona and at DTIC. It was de-classified in 2004 and came back to light in 2008. This was through the effort of a single motivated Lt. Colonel who was willing to take the time and his own personal money to make it happen.

The CRS Casualty Estimates

Let’s just outline the specifics of the casualty estimates for a war with North Korea in the latest Congressional Research Service (CRS) report dated 27 October 2017: https://fas.org/sgp/crs/nuke/R44994.pdf

On page 18:

Even if the DPRK uses only its conventional munitions, estimates range from between 30,000 and 300,000 dead in the first days of fighting, given that DPRK artillery is thought to be capable of firing 10,000 rounds per minute at Seoul. One observer states

Estimates are that hundreds of thousands of South Koreans would die in the first few hours of combat–from artillery, from rockets, from short range missiles–and if this war would escalate to the nuclear level, then you are looking at tens of millions of casualties and the destruction of the eleventh largest economy in the world.

It does not appear that CRS has done any independent analysis of this issues. Its sources in the footnotes are articles from Reuters, New York Times, CNN, NAPSNet Special Reports and GlobalSecurity.

And on page 3:

Should the DPRK use the nuclear, chemical or biological weapons in its arsenal, according to some estimates casualty figures could number in the millions.

 

Casualty Estimates for a War with North Korea

There are a few casualty estimates out there of the cost of a war with North Korea. A couple of these casualty estimates are summarized in this article: https://www.yahoo.com/news/u-must-invade-north-korea-091003273.html

They are:

1. As many as 2.1 million could die if nuclear detonations occurred over Seoul and Tokyo (source: website 38 North, October 2017).

2. As many as 300,000 could die in the first few days of a conflict between North Korea and the U.S. even without the use of nuclear weapons (source: Congressional Research Service, 27 October 2017: https://fas.org/sgp/crs/nuke/R44994.pdf)

The Dupuy Institute has not done any casualty estimates or analysis of a war with North Korea, nor are we planning to at this juncture. We have done a few casualty estimates in the past:

Predictions

 

 

Assessing the TNDA 1990-91 Gulf War Forecast

Assessing the 1990-1991 Gulf War Forecasts

Forecasting U.S. Casualties in Bosnia

https://dupuyinstitute.dreamhosters.com/2016/06/27/forecasting-the-iraqi-insurgency/

Osipov

Back in 1915, a Russian named M. Osipov published a paper in a Tsarist military journal that was Lanchester like: http://www.dtic.mil/dtic/tr/fulltext/u2/a241534.pdf

He actually tested his equations to historical data, which are presented in his paper. He ended up coming up with something similar to Lanchester equations but it did not have a square law, but got a similar effect by putting things to the 3/2nds power.

As far as we know, because of the time it was published (June-October 1915), it was not influenced or done with any awareness of work that the far more famous Frederick Lanchester had done (and Lanchester was famous for a lot more than just his modeling equations).  Lanchester first published his work in the fall of 1914 (after the Great War had already started). It is possible that Osipov was aware of it, but he does not mention Lanchester. He was probably not aware of Lanchester’s work. It appears to be the case of him independently coming up with the use of differential equations to describe combat attrition. This was also the case with Rear Admiral J. V. Chase, who wrote a classified staff paper for U.S. Navy in 1902 that was not revealed until 1972.

Osipov, after he had written his paper, may have served in World War I, which was already underway at the time it was published. Between the war, the Russian revolutions, the civil war afterwards, the subsequent repressions by Cheka and later Stalin, we do not know what happened to M. Osipov. At the time I was asked by CAA if our Russian research team knew about him. I passed the question to Col. Sverdlov and Col. Vainer and they were not aware of him. It is probably possible to chase him down, but would probably take some effort. Perhaps some industrious researcher will find out more about him.

It does not appear that Osipov had any influence on Soviet operations research or military analysis. It appears that he was ignored or forgotten. His article was re-published in the September 1988  of the Soviet Military-Historical Journal with the propaganda influenced statement that they also had their own “Lanchester.” Of course, this “Soviet Lanchester” was publishing in a Tsarist military journal, hardly a demonstration of the strength of the Soviet system.

 

Predictions

We do like to claim we have predicted the casualty rates correctly in three wars (operations): 1) The 1991 Gulf War, 2) the 1995 Bosnia intervention, and 3) the Iraq insurgency.  Furthermore, these were predictions make of three very different types of operations, a conventional war, an “operation other than war” (OOTW) and an insurgency.

The Gulf War prediction was made in public testimony by Trevor Dupuy to Congress and published in his book If War Comes: How to Defeat Saddam Hussein. It is discussed in my book America’s Modern Wars (AMW) pages 51-52 and in some blog posts here.

The Bosnia intervention prediction is discussed in Appendix II of AMW and the Iraq casualty estimate is Chapter 1 and Appendix I.

We like to claim that we are three for three on these predictions. What does that really mean? If the odds of making a correct prediction are 50/50 (the same as a coin toss), then the odds of getting three correct predictions in a row is 12.5%. We may not be particularly clever, just a little lucky.

On the other hand, some might argue that these predictions were not that hard to make, and knowledgeable experts would certainly predict correctly at least two-thirds of the time. In that case the odds of getting three correct predictions in a row is more like 30%.

Still, one notes that there was a lot of predictions concerning the Gulf War that were higher than Trevor Dupuy’s. In the case of Bosnia, the Joint Staff was informed by a senior OR (Operations Research) office in the Army that there was no methodology for predicting losses in an “operation other than war” (AMW, page 309). In the case of the Iraq casualty estimate, we were informed by a director of an OR organization that our estimate was too high, and that the U.S. would suffer less than 2,000 killed and be withdrawn in a couple of years (Shawn was at that meeting). I think I left that out of my book in its more neutered final draft….my first draft was more detailed and maybe a little too “angry”. So maybe, predicting casualties in military operations is a little tricky. If the odds of a correct prediction was only one-in-three, then the odds of getting three correct predictions in a row is only 4%. For marketing purposes, we like this argument better 😉

Hard to say what are the odds of making a correct prediction are. The only war that had multiple public predictions (and of course, several private and classified ones) was the 1991 Gulf War. There were a number of predictions made and we believe most were pretty high. There was no other predictions we are aware of for Bosnia in 1995, other than the “it could turn into another Vietnam” ones. There are no other predictions we are aware of for Iraq in 2004, although lots of people were expressing opinions on the subject. So, it is hard to say how difficult it is to make a correct prediction in these cases.

P.S.: Yes, this post was inspired by my previous post on the Stanley Cup play-offs.

 

Economics of Warfare 3

Examining the third lecture from Professor Michael Spagat’s Economics of Warfare course that he gives at Royal Holloway University. It is posted on his blog Wars, Numbers and Human Losses at: https://mikespagat.wordpress.com/

The link to the lecture is here: http://personal.rhul.ac.uk/uhte/014/Economics%20of%20Warfare/Lecture%203.pdf

This one starts with the war in Kosovo (1998-1999), which was actually a successful invention although very poorly done. It does pick on a constant theme of Dr. Spagat’s, which is how to get the correct counts of actual people killed in the conflicts, including civilians. For those of us who actually try to do things like quantitative analysis of insurgencies (for example America’s Modern Wars)….this is very useful. A lot of other people don’t particularly care, sometimes because a particularly high or low number serves their political agenda (or cosmology).

Starting on slide 11, Dr. Spagat discusses Iraq casualty estimates. This, along with Colombia, were the two areas we discussed with him when we were working on our Iraq and insurgency material (2004-2010). He was one of the few people out there doing work similar to ours. He points out that there were two estimates of deaths in Iraq, one of 150,000 and one of 600,000. Needless to say, the lower one was closer to correct. The higher number got heavily broadcast. This whole section is worth reviewing and remembering for any future conflicts. I like the picture on slide 14.

Sorry about this abstract look at some very sad and gruesome statistics.

P.S. Merry Christmas

So, is it 25,000?

I hate to keep harping on this…..but I do believe in reasonable estimates of opposing forces strength and losses. I am not sure we are doing that.

  1. The U.S. claimed in August that there were 19,000 – 25,000 ISIL:Islamic State strength estimates
    1. I don’t believe I have seen strength estimates from before this time that is more than 25,000: Global Security
    2. About half were in Iraq.
  2. The U.S. now claims there are 12,000 to 15,000 ISIL: white-house-isis-down-to-12000-15000-battle-ready-fighters
    1. This includes 3,000 to 5,000 in Mosul: War updates
  3. The U.S. claims were have killed 50,000 in the last two years: Over 50000 killed
    1. Including 25,000 in the last 11 months: pentagon-counts-isil-dead-refuses-discuss-them
    2. 25,000 – 50,000 = 15,000…meaning they recruited 40,000 new fighters in the last two years?
  4. The U.S. claims were have attrited 75% of ISIL: US-Officials-Say-ISIS-Has-12-000
    1. 15,00 times 4 = 60,000…meaning working backwards this was the ISIL strength….or this includes the 40,000 new recruits added to a strength originally equal to 25,000? This math actually works, if you accept the figure of 40,000 new recruits.
    2. Of course, this is exterminating and replacing the entire ISIL force each year for two years in a row. Can’t really recall the historical president for this.
  5. The UK estimates we have killed 25,000 ISIL: uk-us-number-isis-fighters-killed
    1. This seems more reasonable.

Of course, part of the problem is that ISIL and every other insurgency movement out there does not consists only of full-time fighters. For example (see America’s Modern Wars: Chapter 11: “Estimating Insurgent Force Size”), looking at some the insurgent strength estimates from Vietnam, Cabanas (Mexico) and Shining Path (Peru), we came up with some rules of thumb for determining the mix of insurgents (see page 120):

  1. Full-time insurgents make up 10 to 20% of the force.
  2. Full-time and part-time insurgents are 20 to 50% of the force.
  3. Casual insurgents are 40 to 80% of the force.

This was based on a very limited selection of marginal data and of course, may not be relevant to ISIL, as they have developed a more conventional-like force structure. But, it does bring up the issue that most insurgencies are not only full-time committed fighters, but lots of people that only sometimes active (often regional), and a large collection of people that are only occasionally active (and may become inactive if things are not going well). Also, many of the insurgents are “support personnel” vice fighters. It is an issue that I don’t believe I have seen anyone else attempt to deal with analytically. Our efforts were only preliminary.

Over 50,000 killed

OK…latest estimate from DOD is that they have killed over 50,000 ISIL fighters over the last two years: Body count = 50K+

Here is my post on the subject in August when the count was a mere 45,000: some-back-of-the-envelope-calculations

I don’t think I have much more to add to this without getting very sarcastic. Note that they refer to it as a “conservative estimate.”  Something does not add up somewhere (either their loss estimates are way too high or their force size estimates have been way too low).

War Updates

We are not a news site or current affairs site (because it takes too much time)……but…a few things of note in the middle east:

  1. Mosul has still not fallen. Offensive started 17 October….we were on the outskirts of Mosul by the beginning of November, and now in a grind that some said would take 6 to 8 weeks. So far, it appears to be taking longer than that. We are claiming that 2,000 ISIL fighters have been killed or wounded. 2000-Islamic-State-militants-killed-or-injured-in-Mosul-offensive and us-says-2-000-is-fighters-killed-gravely-wounded-in-mosul
  2. The Syrians appear to be pushing hard to take all of Aleppo before the new U.S. president arrives in office. I suspect this is an attempt to get a negotiating advantage in light of what they perceive to be Trump’s attitudes towards Russia and Syria. According to this report, they have at least 85% of the city: thousands-flee-heavy-aleppo-fighting.
  3. ISIL has retaken Palmyra. latest-syria-says-98-percent-east-aleppo-retaken and recaptures-palmyra

The odd claim is the one made in the second article of this link, where Russia Foreign Minister Lavrov claims about the taking of Palmyra by ISIL that: “…it has been staged to give a respite to bandits in eastern Aleppo.” Of course, one person’s freedom fighters is another person’s bandits.

To take a quote from the first article from Lt. General Townsend: “‘At the start of the campaign, we estimated somewhere between at the low end 3,500, at the high end, about 6,000. By our calculations we think we have killed or badly wounded over 2,000. So if you do the math, that’s still 3,000-5,000’ militants remaining in Mosul, Townsend said.”

We assume he is talking 2,000 killed or seriously enough injured to no be able to return to action in the next couple of months….so, maybe 1,000 killed and 1,000 seriously wounded (kind of grabbing numbers out of thin air here). So, total losses are 4,000 – 6,000 if you count all wounded? That is kind of the entire opposing force.

Then there is the estimates that coalition has lost 14,000 killed and wounded since the start of the offensive (see our post at: Casualties in Iraq (November)). Does that mean that trained conventional counterinsurgency forces are losing something like 2.3-to-1 fighting the insurgents. That would be significant if that was the case.

Something if off somewhere in these various numbers. I not sure which number not to believe (although the estimate of insurgent strength has traditionally always been way too low).

We probably need to keep talking about Afghanistan

image description

Shawn posted a very nice summary a couple of days ago. It is worth reading if you have not already: Meanwhile in Afghanistan

Recent article reports the same trends: Afghan-government-lost-2-percent-territory

A couple of things get my attention in all this:

  1. They are talking about control of territory. I believe control of population is a lot better metric.
    1. One notes in Shawn’s write up that control of population is 68.5% (vice 61.3% of area).
  2. The insurgent level of activity is very high:
    1. 5,523 Afghan Army and police killed (15,000 casualties)
    2. 22,733 incidents from 8/1/2015 to 8/15/2016

Based on Chapter 11 (Estimating Insurgent Force Size) of my book America’s Modern Wars, working backward from this incident data would mean that there are something like 60,000 – 80,000 full-time and part-time insurgents operating. There is a lot comparing apples to oranges to get there: for example, how are they counting incidents in Afghanistan vice how were they counting incidents in the past cases we use for this estimate, what is the mix of full-time and part-time insurgents, how active and motivated are the insurgents, and so forth; but that level of activity is similar to the level of activity in Iraq at its worse (26,033 incidents in 2005, 45,330 in 2006 and 19,125 in 2007 according to one count). We had over 180,000 U.S. and coalition troops there to deal with that. The Afghani’s have 170,000 Army and Air Force or around 320,000 if you count police (and we did not count police in our database unless there were actively involved in counterinsurgent work). The 5,500+ Afghan Army and police killed a year indicates a pretty active insurgency. We lost less than 5,500 for the entire time we were in Iraq. The low wounded-to-killed ratio in the current Afghan data may well be influenced by who they choose to report as wounded and how they address lightly wounded (as discussed in Chapter 15, Casualties, in my upcoming book War by Numbers). Don’t know what current U.S. Army estimates are of Afghan insurgent strength.

Now, in Chapter Six of my book America’s Modern Wars, we developed a force ratio model based upon 83 historical cases (see chart at top of this post). It was very dependent on the cause of the insurgency, whether it was based on a central idea (like nationalism) or was regional or factional, or whether it was based on a overarching idea (like communism). I don’t still don’t really know the nature of the Afghan insurgency, and we were never funded to study this insurgency (we were only funded for Iraq work). So, I have not done the in-depth analysis of the Afghan insurgency that I did for Iraq. But…….nothing here looks particularly positive.

We never did an analysis of stalemated insurgencies. It could be done, although there are not that many cases of these. One could certainly examine any insurgency that lasts more than 15 years for this purpose. Does a long stalemated insurgency mean that the government (or counterinsurgents) eventually win? Or does a long stalemated insurgency mean that the insurgents eventually win? I don’t know. I would have to go back through our database of 100+ cases, update the data, sort out the cases and then I could make some predictions. That takes time and effort, and right now my effort is focused elsewhere. Is anyone inside DOD doing this type of analysis? I doubt it. Apparently a stalemate means that you can now pass the problem onto the next administration. While it solves the immediate political problem, is really does not answer the question of whether we are winning or losing. Is what we are doing good enough that this will revolve in our favor in the next ten years, or do we need to do more? I think this is the question that needs to be addressed.